notebooks/clustering-techniques/clustering-techniques.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clustering Techniques\n",
    "\n",
    "Se aplicarán técnicas de agrupación como *k-means* al set de datos proporcionado y se determinará si los datos del conjunto pueden ser agrupados y cuáles son los grupos óptimos.\n",
    "\n",
    "### Tabla de contenidos\n",
    "\n",
    "- [Carga del dataset](#Carga-del-dataset)\n",
    "- [Análisis descriptivo de los datos](#Análisis-descriptivo-de-los-datos)\n",
    "- [Tratamiento de datos](#Tratamiento-de-datos)\n",
    "  - [Missing Values](#Missing-Values)\n",
    "  - [Normalización](#Normalización)\n",
    "- [Conjuntos de entrenamiento y validación](#Conjuntos-de-entrenamiento-y-validación)\n",
    "- [Agrupamiento *k-medias*](#Agrupamiento-k-medias)\n",
    "- [Análisis de los resultados](#Análisis-de-los-resultados)\n",
    "  - [Estudio de centros geométricos de los clusters resultantes](#Estudio-de-centros-geométricos-de-los-clusters-resultantes)\n",
    "  - [Representación visual de resultados](#Representación-visual-de-resultados)\n",
    "- [Conclusiones](#Conclusiones)\n",
    "- [Apéndice](#Apéndice)\n",
    "  - [Clustering Jerárquico](#Clustering-Jerárquico)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Imports y funciones auxiliares\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import datetime\n",
    "import errno\n",
    "import math\n",
    "import os\n",
    "import sys\n",
    "\n",
    "# Función para validar que el fichero existe\n",
    "def assert_fichero(nombre):\n",
    "    if (not os.path.isfile(nombre)):\n",
    "        raise FileNotFoundError(errno.ENOENT, os.strerror(errno.ENOENT), nombre)\n",
    "        \n",
    "# Función para imprimir mensajes por la salida de error\n",
    "def eprint(*args, **kwargs):\n",
    "    print(*args, file=sys.stderr, **kwargs)\n",
    "    \n",
    "# Función para representar histograma por grupos con las categorías agregadas\n",
    "def histogramas_con_categorias_agregadas(df_agrupado, nombre_algoritmo, nombre_archivo=None):\n",
    "    \n",
    "    # Preparación de los datos para representación\n",
    "    ## Se eliminan las columnas de Channel y Region y se agrupan los registros por Group\n",
    "    df_agregado = df_agrupado.drop(['Channel', 'Region'], axis=1).groupby('Group')\n",
    "    \n",
    "    ## Se obtienen los números de muestras por grupo\n",
    "    n_muestras_grupo = df_agregado['Group'].count()\n",
    "    \n",
    "    ## Se suman todos los registros y se dividen por el número de muestras de cada grupo\n",
    "    df_agregado = df_agregado.sum().divide(n_muestras_grupo, axis='index').T\n",
    "    \n",
    "    n_grupos = len(df_agregado.columns)\n",
    "    n_categorias = len(df_agregado.index)\n",
    "    \n",
    "    grupos = np.arange(n_grupos)\n",
    "    offset_y = np.zeros(n_grupos)\n",
    "    \n",
    "    # Elementos de estilo\n",
    "    ancho_barras = 0.6\n",
    "    colors = ['xkcd:goldenrod', 'xkcd:coral', 'xkcd:turquoise',\n",
    "              'xkcd:sea blue', 'xkcd:lavender', 'xkcd:indigo']\n",
    "\n",
    "    f = plt.figure()\n",
    "    for categoria in np.arange(0, n_categorias):\n",
    "        plt.bar(grupos, df_agregado.iloc[categoria], ancho_barras, \n",
    "                bottom=offset_y, color=colors[categoria])\n",
    "        \n",
    "        offset_y += df_agregado.iloc[categoria]\n",
    "    \n",
    "    _ = plt.ylabel('Gastos totales normalizados por categoría\\ny número de muestras en grupo')\n",
    "    _ = plt.xlabel('Grupos')\n",
    "    _ = plt.title('Agrupación con {:s}'.format(nombre_algoritmo))\n",
    "    _ = plt.xticks(grupos, ['G{:d}'.format(x) for x in grupos])\n",
    "    _ = plt.legend(df_agregado.index, title='Categorías',\n",
    "                   loc='center left', bbox_to_anchor=(1, 0.5))\n",
    "    \n",
    "    if nombre_archivo:\n",
    "        f.savefig(\"{:s}.pdf\".format(nombre_archivo), bbox_inches='tight')\n",
    "        \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Carga del dataset\n",
    "\n",
    "El conjunto de datos proporcinado se encuentra disponible en el fichero `Wholesale customers data.csv` que ha sido descargado desde la web: [Machine Learning Repository - Wholesale customers Data Set](https://archive.ics.uci.edu/ml/datasets/Wholesale+customers)\n",
    "\n",
    "```sh\n",
    "curl -O https://archive.ics.uci.edu/ml/machine-learning-databases/00292/Wholesale%20customers%20data.csv\n",
    "```\n",
    "\n",
    "En el enlace anterior puede encontrarse más información acerca del contenido del fichero."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Channel</th>\n",
       "      <th>Region</th>\n",
       "      <th>Fresh</th>\n",
       "      <th>Milk</th>\n",
       "      <th>Grocery</th>\n",
       "      <th>Frozen</th>\n",
       "      <th>Detergents_Paper</th>\n",
       "      <th>Delicassen</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.322727</td>\n",
       "      <td>2.543182</td>\n",
       "      <td>12000.297727</td>\n",
       "      <td>5796.265909</td>\n",
       "      <td>7951.277273</td>\n",
       "      <td>3071.931818</td>\n",
       "      <td>2881.493182</td>\n",
       "      <td>1524.870455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.468052</td>\n",
       "      <td>0.774272</td>\n",
       "      <td>12647.328865</td>\n",
       "      <td>7380.377175</td>\n",
       "      <td>9503.162829</td>\n",
       "      <td>4854.673333</td>\n",
       "      <td>4767.854448</td>\n",
       "      <td>2820.105937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3127.750000</td>\n",
       "      <td>1533.000000</td>\n",
       "      <td>2153.000000</td>\n",
       "      <td>742.250000</td>\n",
       "      <td>256.750000</td>\n",
       "      <td>408.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>8504.000000</td>\n",
       "      <td>3627.000000</td>\n",
       "      <td>4755.500000</td>\n",
       "      <td>1526.000000</td>\n",
       "      <td>816.500000</td>\n",
       "      <td>965.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>16933.750000</td>\n",
       "      <td>7190.250000</td>\n",
       "      <td>10655.750000</td>\n",
       "      <td>3554.250000</td>\n",
       "      <td>3922.000000</td>\n",
       "      <td>1820.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>112151.000000</td>\n",
       "      <td>73498.000000</td>\n",
       "      <td>92780.000000</td>\n",
       "      <td>60869.000000</td>\n",
       "      <td>40827.000000</td>\n",
       "      <td>47943.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Channel      Region          Fresh          Milk       Grocery  \\\n",
       "count  440.000000  440.000000     440.000000    440.000000    440.000000   \n",
       "mean     1.322727    2.543182   12000.297727   5796.265909   7951.277273   \n",
       "std      0.468052    0.774272   12647.328865   7380.377175   9503.162829   \n",
       "min      1.000000    1.000000       3.000000     55.000000      3.000000   \n",
       "25%      1.000000    2.000000    3127.750000   1533.000000   2153.000000   \n",
       "50%      1.000000    3.000000    8504.000000   3627.000000   4755.500000   \n",
       "75%      2.000000    3.000000   16933.750000   7190.250000  10655.750000   \n",
       "max      2.000000    3.000000  112151.000000  73498.000000  92780.000000   \n",
       "\n",
       "             Frozen  Detergents_Paper    Delicassen  \n",
       "count    440.000000        440.000000    440.000000  \n",
       "mean    3071.931818       2881.493182   1524.870455  \n",
       "std     4854.673333       4767.854448   2820.105937  \n",
       "min       25.000000          3.000000      3.000000  \n",
       "25%      742.250000        256.750000    408.250000  \n",
       "50%     1526.000000        816.500000    965.500000  \n",
       "75%     3554.250000       3922.000000   1820.250000  \n",
       "max    60869.000000      40827.000000  47943.000000  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Descomentar la siguiente línea para descargar el dataset\n",
    "#!curl -O https://archive.ics.uci.edu/ml/machine-learning-databases/00292/Wholesale%20customers%20data.csv\n",
    "\n",
    "# Comprobación de que el fichero de datos está disponible\n",
    "fichero_datos = 'Wholesale customers data.csv'\n",
    "assert_fichero(fichero_datos)\n",
    "\n",
    "# Carga del archivo como un DataFrame de Pantas con separador de columnas ','\n",
    "df_raw = pd.read_csv(fichero_datos, sep=',')\n",
    "\n",
    "# Muestra una descripción estadística del conjunto de datos\n",
    "df_raw.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Análisis descriptivo de los datos\n",
    "\n",
    "En base a la información disponible en la [web anterior](https://archive.ics.uci.edu/ml/datasets/Wholesale+customers), se conoce que las variables `Fresh`, `Milk`, `Grocery`, `Frozen`, `Detergents_Paper` y `Delicassen` son variables continuas que reflejan el gasto anual en cada uno de los productos asociados.\n",
    "\n",
    "Mientras que `Channel` y `Region` hacen referencia a variables categóricas:\n",
    "\n",
    "- `Channel` puede tomar los valores `Horeca` (*Hotel/Restaurante/Cafeteria*) (`298` registros) o `Retail` (`142` registros).\n",
    "- `Region` puede tomar los valores `Lisboa` (`77` registros), `Oporto` (`47` registros) y `Resto de regiones` (`316` registros).\n",
    "\n",
    "Aunque en la información aportada en la web no se da una relación directa entre las categorías y los valores numéricos asociados, sí se da información acerca del número de entradas disponibles en el conjunto de datos para cada una de estas categorías."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = df_raw[['Channel', 'Region']].hist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "De los histogramas puede extraerse la información necesaria para relacionar el valor numérico que se ha asociado a cada una de las categorías con su valor real.\n",
    "\n",
    "Así pues:\n",
    "\n",
    "- **Channel**\n",
    "  - *Horeca* = `1`\n",
    "  - *Retail* = `2`\n",
    "\n",
    "- **Region**\n",
    "  - *Lisboa* = `1`\n",
    "  - *Oporto* = `2`\n",
    "  - *Otras regiones* = `3`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tratamiento de datos\n",
    "\n",
    "Antes de generar los clusters en base a los datos aportados, es necesario hacer un pre-tratamiento de estos para que el algoritmo de clustering pueda trabajar correctamente con ellos.\n",
    "\n",
    "#### Missing Values\n",
    "\n",
    "En este caso, puede verse que el conjunto de datos no dispone de ningún registro incompleto, por lo que no debe hacerse ningún tratamiento al respecto."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No se han encontrado registros que tengan elementos nulos. No hay que hacer limpieza del dataset.\n"
     ]
    }
   ],
   "source": [
    "# Se buscan los registros que tengan algún elemento nulo\n",
    "numero_registros_con_elementos_nulos = df_raw[df_raw.isnull().any(axis=1)].shape[0]\n",
    "\n",
    "if numero_registros_con_elementos_nulos > 0:\n",
    "    eprint(\"Se han encotrado registros que tienen algún elemento nulo. Hay que tratar el dataset antes de continuar\")\n",
    "else:\n",
    "    print(\"No se han encontrado registros que tengan elementos nulos. No hay que hacer limpieza del dataset.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Normalización\n",
    "\n",
    "Para que los algoritmos de clustering puedan agrupar los registros deben poder calcular correctamente las distancias entre los mismos, y para que todas las variables aporten el mismo peso las distancias deben ser del mismo orden, lo que implica que los datos deben estar normalizados.\n",
    "\n",
    "En este caso se hará uso de la clase [`sklearn.preprocessing.MaxAbsScaler`](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MaxAbsScaler.html) que normaliza cada variable dividiendo todos sus registros por el máximo absoluto de la misma."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Channel</th>\n",
       "      <th>Region</th>\n",
       "      <th>Fresh</th>\n",
       "      <th>Milk</th>\n",
       "      <th>Grocery</th>\n",
       "      <th>Frozen</th>\n",
       "      <th>Detergents_Paper</th>\n",
       "      <th>Delicassen</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "      <td>440.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.661364</td>\n",
       "      <td>0.847727</td>\n",
       "      <td>0.107001</td>\n",
       "      <td>0.078863</td>\n",
       "      <td>0.085700</td>\n",
       "      <td>0.050468</td>\n",
       "      <td>0.070578</td>\n",
       "      <td>0.031806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.234026</td>\n",
       "      <td>0.258091</td>\n",
       "      <td>0.112771</td>\n",
       "      <td>0.100416</td>\n",
       "      <td>0.102427</td>\n",
       "      <td>0.079756</td>\n",
       "      <td>0.116782</td>\n",
       "      <td>0.058822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.000748</td>\n",
       "      <td>0.000032</td>\n",
       "      <td>0.000411</td>\n",
       "      <td>0.000073</td>\n",
       "      <td>0.000063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.027889</td>\n",
       "      <td>0.020858</td>\n",
       "      <td>0.023205</td>\n",
       "      <td>0.012194</td>\n",
       "      <td>0.006289</td>\n",
       "      <td>0.008515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.075826</td>\n",
       "      <td>0.049348</td>\n",
       "      <td>0.051256</td>\n",
       "      <td>0.025070</td>\n",
       "      <td>0.019999</td>\n",
       "      <td>0.020138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.150991</td>\n",
       "      <td>0.097829</td>\n",
       "      <td>0.114850</td>\n",
       "      <td>0.058392</td>\n",
       "      <td>0.096064</td>\n",
       "      <td>0.037967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Channel      Region       Fresh        Milk     Grocery      Frozen  \\\n",
       "count  440.000000  440.000000  440.000000  440.000000  440.000000  440.000000   \n",
       "mean     0.661364    0.847727    0.107001    0.078863    0.085700    0.050468   \n",
       "std      0.234026    0.258091    0.112771    0.100416    0.102427    0.079756   \n",
       "min      0.500000    0.333333    0.000027    0.000748    0.000032    0.000411   \n",
       "25%      0.500000    0.666667    0.027889    0.020858    0.023205    0.012194   \n",
       "50%      0.500000    1.000000    0.075826    0.049348    0.051256    0.025070   \n",
       "75%      1.000000    1.000000    0.150991    0.097829    0.114850    0.058392   \n",
       "max      1.000000    1.000000    1.000000    1.000000    1.000000    1.000000   \n",
       "\n",
       "       Detergents_Paper  Delicassen  \n",
       "count        440.000000  440.000000  \n",
       "mean           0.070578    0.031806  \n",
       "std            0.116782    0.058822  \n",
       "min            0.000073    0.000063  \n",
       "25%            0.006289    0.008515  \n",
       "50%            0.019999    0.020138  \n",
       "75%            0.096064    0.037967  \n",
       "max            1.000000    1.000000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import MaxAbsScaler\n",
    "\n",
    "transformer = MaxAbsScaler()\n",
    "df = pd.DataFrame(transformer.fit_transform(df_raw), columns=df_raw.columns)\n",
    "\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conjuntos de entrenamiento y validación\n",
    "\n",
    "En el caso de los algoritmos de agrupación no supervisados, al no disponerse de una variable objetivo, no tiene sentido hablar de conjuntos de entrenamiento y validación dado que no hay una respuesta correcta contra la que poder comparar el algoritmo entrenado.\n",
    "\n",
    "Por tanto, se proveerá con el `100%` de los datos al algoritmo de clustering para que realice el agrupamiento con todos ellos."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Agrupamiento *k-medias*\n",
    "\n",
    "Para realizar el *clustering* de los datos se usará la clase [`sklearn.cluster.KMeans`](https://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html)\n",
    "\n",
    "Este algoritmo requiere únicamente el número de agrupaciones que se quieren obtener o parámetro `k`, pero como no se dispone de ninguna restricción sobre este parámetro de entrada, se hará uso de la función [`sklearn.metrics.silhouette_score`](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.silhouette_score.html) para calcular como métrica la media del coeficiente de silueta de los clusters generados que se calcula con la media de las distancias de los puntos del cluster y la distancia media al cluster más cercano de cada una de las muestras del cluster.\n",
    "\n",
    "Los valores obtenidos para esta métrica se encuentran dentro del rango `[-1, 1]`.\n",
    "\n",
    "- Generalmente, valores negativos indican que se han asignado erróneamente muestras a un cluster dado.\n",
    "- `0` indica que hay solape entre clusters.\n",
    "- `1` es el mejor valor que se puede obtener.\n",
    "\n",
    "En este caso, se probarán distintos valores de `k`, comprendidos entre `2` y `10` para determinar con la métrica anterior cuál es la mejor agrupación posible para el set de datos tratado."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor configuración de agrupación obtenida:\n",
      "  - k = 5\n",
      "  - Media de silhouette_score: 0.59\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.metrics import silhouette_score\n",
    "\n",
    "best_model = None\n",
    "best_labels = None\n",
    "best_k = None\n",
    "best_silhouette_avg = None\n",
    "\n",
    "for n_clusters in range(2, 11):\n",
    "    model = KMeans(n_clusters=n_clusters,\n",
    "                   random_state=10, # Se fija la semilla para facilitar la reproducibilidad de los resultados\n",
    "                   n_jobs=-1        # Se establece a -1 el número de trabajos para usar todos los hilos de ejecución\n",
    "                  )\n",
    "    \n",
    "    cluster_labels = model.fit_predict(df)\n",
    "    silhouette_avg = silhouette_score(df, cluster_labels)\n",
    "    if not best_silhouette_avg or best_silhouette_avg < silhouette_avg:\n",
    "        best_silhouette_avg = silhouette_avg\n",
    "        best_model = model\n",
    "        best_labels = cluster_labels\n",
    "        best_k = len(np.unique(best_labels))\n",
    "\n",
    "df_agrupado = pd.concat([df, pd.Series(best_labels, name='Group')],\n",
    "                        axis=1, sort=False)\n",
    "        \n",
    "print(\"Mejor configuración de agrupación obtenida:\")\n",
    "print(\"  - k = {:d}\".format(best_k))\n",
    "print(\"  - Media de silhouette_score: {:.2f}\".format(best_silhouette_avg))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Análisis de los resultados\n",
    "\n",
    "Dado que el conjunto de datos dispone de más de dos (y de tres) variables para la agrupación, la representación visual de los *clusters* resultantes no es directa.\n",
    "\n",
    "Por lo que las conclusiones sobre las agrupaciones deberán extraerse a partir de los centros de los *clusters* obtenidos y de representar gráficamente la información agrupada pero separadamente.\n",
    "\n",
    "#### Estudio de centros geométricos de los clusters resultantes\n",
    "\n",
    "Para que el estudio de los centros geométricos de los clusters sea más intuitivo, se aplicarán las transformaciones inversas a la normalización empleada en [*Tratamiento de datos - Normalización*](#Normalización)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col4 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col6 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col8 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col0 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col1 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col3 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col5 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col7 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col2 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col0 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col1 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col5 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col2 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col3 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col4 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col6 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col7 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col8 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }</style><table id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aa\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Channel</th>        <th class=\"col_heading level0 col1\" >Region</th>        <th class=\"col_heading level0 col2\" >Fresh</th>        <th class=\"col_heading level0 col3\" >Milk</th>        <th class=\"col_heading level0 col4\" >Grocery</th>        <th class=\"col_heading level0 col5\" >Frozen</th>        <th class=\"col_heading level0 col6\" >Detergents_Paper</th>        <th class=\"col_heading level0 col7\" >Delicassen</th>        <th class=\"col_heading level0 col8\" >Count</th>    </tr>    <tr>        <th class=\"index_name level0\" >Group</th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aalevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col0\" class=\"data row0 col0\" >1</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col1\" class=\"data row0 col1\" >3</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col2\" class=\"data row0 col2\" >13878.1</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col3\" class=\"data row0 col3\" >3486.98</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col4\" class=\"data row0 col4\" >3886.73</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col5\" class=\"data row0 col5\" >3656.9</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col6\" class=\"data row0 col6\" >786.682</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col7\" class=\"data row0 col7\" >1518.28</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow0_col8\" class=\"data row0 col8\" >211</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aalevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col0\" class=\"data row1 col0\" >1</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col1\" class=\"data row1 col1\" >1.32184</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col2\" class=\"data row1 col2\" >12499.4</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col3\" class=\"data row1 col3\" >3366.22</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col4\" class=\"data row1 col4\" >4145.01</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col5\" class=\"data row1 col5\" >3969.8</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col6\" class=\"data row1 col6\" >799.966</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col7\" class=\"data row1 col7\" >1167.78</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow1_col8\" class=\"data row1 col8\" >87</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aalevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col0\" class=\"data row2 col0\" >2</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col1\" class=\"data row2 col1\" >1.5</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col2\" class=\"data row2 col2\" >6209.47</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col3\" class=\"data row2 col3\" >10104.4</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col4\" class=\"data row2 col4\" >15983.3</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col5\" class=\"data row2 col5\" >2101.5</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col6\" class=\"data row2 col6\" >7493</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col7\" class=\"data row2 col7\" >1556.14</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow2_col8\" class=\"data row2 col8\" >36</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aalevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col0\" class=\"data row3 col0\" >2</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col1\" class=\"data row3 col1\" >3</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col2\" class=\"data row3 col2\" >9178.34</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col3\" class=\"data row3 col3\" >8799.59</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col4\" class=\"data row3 col4\" >13575.1</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col5\" class=\"data row3 col5\" >1428.09</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col6\" class=\"data row3 col6\" >5630.35</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col7\" class=\"data row3 col7\" >1750.31</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow3_col8\" class=\"data row3 col8\" >98</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aalevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col0\" class=\"data row4 col0\" >2</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col1\" class=\"data row4 col1\" >2.875</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col2\" class=\"data row4 col2\" >17674.5</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col3\" class=\"data row4 col3\" >36953.2</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col4\" class=\"data row4 col4\" >51511.1</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col5\" class=\"data row4 col5\" >2383</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col6\" class=\"data row4 col6\" >26343.5</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col7\" class=\"data row4 col7\" >2679.62</td>\n",
       "                        <td id=\"T_693e433e_942a_11e9_8da2_003ee1bf34aarow4_col8\" class=\"data row4 col8\" >8</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x13e629e48>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_centros = pd.DataFrame(transformer.inverse_transform(best_model.cluster_centers_),\n",
    "                          columns=df.columns)\n",
    "\n",
    "df_centros['Group'] = pd.Series(range(0,5))\n",
    "df_centros['Count'] = df_agrupado.groupby('Group')['Group'].count()\n",
    "df_centros.set_index('Group', inplace=True)\n",
    "\n",
    "df_centros.style.highlight_max(color='turquoise').highlight_min(color='coral')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A partir de la tabla pueden sacarse las siguientes conclusiones:\n",
    "\n",
    "- A partir de los máximos (*turquesa*) puede deducirse que el grupo `4` es el más adinerado, dado que es el grupo que de media más gasta y además es un grupo bastante reducido, apenas 8 muestras de las 440.\n",
    "- A partir de los mínimos (*coral*) no queda claro si el grupo `0` o el `1` son los menos pudientes. Ambos grupos tienen gastos similares, sin embargo, el grupo `0` está deslocalizado geográficamente dado que pertenece a la región 3 mientras que el grupo `1` se podría decir que está principalmente localizado en *Lisboa* dado que la región asignada al centro está próxima a 1. No obstante, el hecho de el grupo `1` tenga un máximo en cuanto al gasto en congelados, puede dar a entender que tiene mayor poder adquisitivo frente al grupo `0`.\n",
    "- Las diferencias entre los grupos `2` y `3` parecen ser más bien puramente geográficas dado que a nivel de gastos ambos grupos son similares."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Representación visual de resultados\n",
    "\n",
    "Como ya se ha mencionado, la representación visual de los datos cuando se trata de más de 3 variables involucradas en el estudio se complica, dado que no se dispone de una representación espacial intuitiva.\n",
    "\n",
    "Para intentar corroborar los resultados extráidos a partir del estudio de los centros geométricos de los clusters, se ha representado individualmente cada variable separando cada uno de los grupos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3sAAAFBCAYAAADOs7YBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3Xd8VfX9x/HX99zsBQQSRgKEvRXZqOAq7lUXYLV111arto4uf52/Vmu1P3HXVmtd4K4TJypKQTayRCCMJIyEMAIJSe695/v744YMSMglyc1Jwvv5eOQB53vO/Z73TbjkfM74fo21FhEREREREWlbHK8DiIiIiIiISNNTsSciIiIiItIGqdgTERERERFpg1TsiYiIiIiItEEq9kRERERERNogFXsiIiIiIiJtkIo9ERERERGRNkjFnoiIiIiISBukYk9ERERERKQNivI6wJHq1KmTzcrK8jqGiIiIiIiIJxYtWrTDWptW33atrtjLyspi4cKFXscQERERERHxhDFmUzjb6TZOERERERGRNkjFnoiIiIiISBukYk9ERERERKQNanXP7ImIiIiIiDSW3+8nNzeX0tJSr6PUKS4ujszMTKKjoxv0ehV7IiIiIiJy1MnNzSU5OZmsrCyMMV7HOYS1lsLCQnJzc+nVq1eD+tBtnCIiIiIictQpLS2lY8eOLbLQAzDG0LFjx0ZdeVSxJyIiIiIiR6WWWugd0Nh8KvZERERERETaIBV7IiIiIiIi1Wzfvp3LL7+c3r17M3LkSMaPH88bb7zhdawjpmJPRERERESkgrWWCy+8kIkTJ5Kdnc2iRYuYMWMGubm5NbYLBAIeJQxfxEbjNMY8DZwL5Ftrh9ay3gDTgLOBEuAqa+3iSOVpbj9cNZ9nCnMAiMPwx97D2BPwMyW9O/3ikwlayysFOSwv3sPpHbpwUvs0AL7cs4P3d25jUEIymTPjiCY0zKqDQ8axsUTFGNL6xhAV6xAocylYV06g3NKpTwzxKT6sa9mxwc/+XUG2bC7jrX9+yaKvvqVTejt+/NcJLJuTQ5+h6XxnyhB8PoectYV88OIKktrHcc4PjiW5fRxFu/bz7r+XUVJUxpBxmbzyyALKSvxc/4eTOPb4Ho3+3vi/6AAUVy77+j2BLc3GdPwuTvJIANydM7F7ZmOSx2A6XogxBlv8NW7BqxDTDXf9LTX6jJ5Q3uhcDTHvmd1VC6nQrVssSelRpPYI/dyKCwPs3OQnJtEhrU8MTpShvCT0cwNY+2URr0//nILtuxl9/ECyxiWQn7uHUy8ZTP/hXQCY/eYaVszLZdjx3ZlwXn8A1izeyqevr6ZLj/YEXZf3n19Oz/4d+dlDZ5KQFBPx920De3C3PweBQpy0yZiEgRHfp0hDVf+8nHnlMOLiGzZ8tURGIODy8YwVZK8sYPxZfTluYk+vIzUZu3kjdt4cSO2ImXAKJjbW60jSRi3+bCPzPlhf4xjPa9Ytxea/eMgxXmswa9YsYmJiuPHGGyvbevbsyU9+8hOeeeYZXn/9dfbt20cwGOSzzz7jrrvuYubMmRhjuPvuu5k8eTKfffYZ999/P++88w4AN998M6NGjeKqq64iKyuLyy67jJkzZxIfH8+LL75I3759I/JeIjn1wjPAI8Czdaw/C+hX8TUWeLziz1Yve9euykIPoBTLndlfA/C33G/5cvgpPJy3jn9t3wjA/bnf8vd+I4lxHK5eswCAD9efSAwxGEIPZVosecvKANi+ppyh5yax4r19lO5xAdi6soyh5ySzdWVZZSHx5INvsXr5JgCKdhdzxwUvVWZa+OlGptw2luvGP01piR+AN/+xmH/89xquP+FfbF5TeMj7WjRxI4/N+n4T/CIurrEUXFvxQcq5H4b8B1u8HHfjryvXO5l3YlLPJrj8dLD+Ru676dQo9AB2wpadZUAZmcPjSO7sY/WHxWBDqws3+Ol3UgLL39qLvzTU+L+/epbivfsB2Lh+GzwX2vb5v85l2vvfY/HnG3n6j19U7uL635/E0HGZ/OzsFwkGbY3dr5iby7wP1vFO3s8i8n4PsDZAYNkpULICADf3AaKOnY1JGh7R/Yo0xPyPs2t8Xj5+ZSWPfHSlx6mkuj9f9zYznwv9jnzuvv/ym2cu4Kwrj/E4VePZ1Stx7/ktBIOh5Tmf4/vtPR6nkrbonX8t5U/XvV25vPDTjfz6n+d5mCgkuOpS7K4PQgsVx3hO6pnehgrTypUrGTFiRJ3rFy9ezNdff01qaiqvvfYaS5cuZdmyZezYsYPRo0czceLEevfRrl07li9fzrPPPsttt91WWRQ2tYiV/dba2cDOw2xyAfCsDZkHtDfGdI1UnuY0aMWsOteVuEEe27KeZ7dvqtH+cN5aHs5bW7lcvdADavy9tMhly/KyykIPwA3Atm/KKFhfdYXrmxWb68wx89llvProgspCD2Dj6h0895c5tRZ6AFh48jef1dlnOPxfHO6qk4u75VHcvIdrtm55DHfLY4ct9A7fb/PbuqqMbavLKws9gKJtAbasLKss9Nas3FxZ6B0sGHB59dEFvPzQ/BrtL02bz6uPLDik0DugcFsxS2ZvqnVdU7G7Z1UWegC4pbhb/xnRfYo01MGfl0WzNrJu+XYPE0l1ewpL+OCF5TXaXnroK4/SNC370czKQg+Ab7/BZq/zLpC0WQd/ZmY+u4y9u72dJNyWrKkq9IADx3it1U033cSxxx7L6NGjAZg0aRKpqakAfPnll0ydOhWfz0fnzp056aSTWLBgQb19Tp06tfLPuXPnRiy7l9d4M4Ccasu5FW2HMMbcYIxZaIxZWFBQ0CzhGqO+AVLjHB++g4ZRjXV8xDq+ymVL7QfzB1TbtGq/DlTv9nAjtTo+h5i4Qy/sxsUf/mJvdGwtO25KTgw4B93m4kSDExfZ/TYxxwl9HdJe7dsXHVP/9zo6pub3OybWV+/PILaen2GjmVpuQzr4ZybSQtT2eYmJjfBnRMLm+BwcX81fVtFt5ecTVcv7qK1NpJEO/szU9rlqdk4tJ+Fra2uhhgwZwuLFVU+XPfroo3zyySccqEMSExPr7SMqKgrXrbowc/BcedWnVIjk9A/e39AbBmvtk9baUdbaUWlpaV7HqVfphIvrXNcpKoabuvXhloyq+3J9GH7efQB3Zg4gquKHXUppjYKv+t+T0nx0GxpLUlrVQUxUnKHr4Fi6DK466B45vu7nqCbfOobJt4yhfaeEyrZhx2dy+Z3HM2RsrTU3jmO45a+T6uwzHId9ts6Jw8m4HV+PX9Vs7v5LfBm3gC+pYf16IOPYOLoOjcWp9v9vas9oug6JIy459LHr3a8bHdNSan19fGI0U386jqvvnlCj/epfT+Dyn40nLqH2Z4569E9l8Ojaf35NxbSbgEmplisqFafbjXW/QMRDB39eTrt0MD36d/QwkVSX3D6OS28eU7nsi3L4/s9P8DBR0zFnnw9x1U5UjhyD6ZHlWR5pu37wixPwVSvuJt86hsRkb0/CmrhemPTLqxoqjvFai1NPPZXS0lIef/zxyraSkpJat50wYQIvvfQSwWCQgoICZs+ezZgxY+jZsyerVq2irKyM3bt388knn9R43UsvvVT55/jx4yP2Xoy1h7+C1KjOjckC3qljgJa/A59Za6dXLK8BTrbWbj1cn6NGjbILFy6MQNqmNWtXHmetmAfAoNhE/qfXUHYFyrmwUwadokMfwNm7C1hevIfTOqQzMCF00P9tyV4+2rWdwYkplP9nP8kkA+BEOfQaE09UtKFDj2gcn8ENWnbl+AmUW1J7RBMdFyoi9mz1s3+3y8b8/cx/fhWzP1pGZlYaP3l0Aos/3USfoemMODkLgN07Svj8jW9Iah/HxAsGEB3jo7wswOw311C8p4wRp/Tk5Wnz2V/s59rfTqRrz/aN/t74v3gCqBpgxTfkLWzpBpzUszFxoecB7b6luHu+xCSPwUkJHQjYsi24hW9jYrsRXFW9oL6E6AkvNjpXQ1R/bm/ouUnszQ+SnO4jqVOoyisrdtmV4yc20aF9RhTGMQT9lp2bQrekrvlsH7NmLmJr7g5OmDSMrHExbM8pYsJ5/UnPDP2bWLVgCyvm5XLM8d0ZODJ0p/P2nD18+c5auvRoR1xiNK8/sZB+x3Tm+784Aae2S4pNzLp+7M63wb8T0/F8TEx6xPcp0lDVPy/jz+qL47TsCXSPRos+3Uj2ynzGTOpNzwGdvI7TZOzOQuziBZjUjjB8BKa223JEmsDGb3aw4OPsGsd4XrPWxe768JBjvJZk9erVDBo0qNZ1W7du5ac//SlfffUVaWlpJCYmcuONN7J//34WLlzII488AoRG7qxtgBaAu+66izfeeINevXqRlJTE+eefXzlAy+TJk5k5cyaxsbFMnz79sAO01JbTGLPIWjuqvvfoZbF3DnAzodE4xwIPWWvHHLzdwVpLsSciIiIiIi3X4Yq9SMrKymLhwoV06hTeya3GFHuRnHphOnAy0MkYkwv8FkLzCFhrnwDeI1TorSM09cLVkcoiIiIiIiJytIlYsWetnVrPegvcFKn9i4iIiIiItDQbN25stn21igFaRERERERE5Mio2BMREREREWmDVOyJiIiIiIi0QSr2RERERERE2iAVeyIiIiIiIodhrYub/yL+JePwz8vEv2Qcbv6LWOs2qt/333+fAQMG0LdvX+69994mSltFxZ6IiIiIiEgdrHUJrrqM4NqbYN9i8OfDvsUE195EcPXkBhd8wWCQm266iZkzZ7Jq1SqmT5/OqlWrmjS7ij0REREREZE62IIZ2N2fgFtcc4VbjN31MbbgpQb1O3/+fPr27Uvv3r2JiYlhypQpvPnmm02QuIqKPRERERGRVmx7zh7e/fcy1ize6nWUNimY99Chhd4BbjHBvGkN6jcvL4/u3btXLmdmZpKXl9egvuoSsUnVRUREREQksuZ9sJ67LnwJf3kQgOt/fxLX3D3R41RtTFlu49Z7SFf2RERERERaqaf+OLuy0AP49z1z2F9c7mGiNig2s3Hr65CRkUFOTk7lcm5uLhkZGQ3qqy4q9kREREREWqmSvWU1lgPlQcrLgnVsLQ3hy7gFnMTaVzqJ+DJubVC/o0ePZu3atWzYsIHy8nJmzJjB+eef34iktcRr0t5ERERERKTZfPeHI2ssn3bpYNqlxnuUpm0yaVMw7U87tOBzEjEdvoNJm9ygfqOionjkkUc444wzGDRoEJdddhlDhgxpgsTV9tGkvYmIiIiISLO55Mej6dy9HfM+WE+foemcd81wryO1OcY4+Aa/jC14KTQYS1kuxGbiy7gVkzYZYxp+/ezss8/m7LPPbsK0NanYExERERFpxSac158J5/X3OkabZoyDSZ+Kkz7V6yhHRLdxioiIiIiItEEq9kRERERERNogFXsiIiIiIiJtkIo9ERERERGRNkjFnoiIiIiISBuk0ThFREREREQOw7oudu4X2JlvQ+EO6NgJc9Z5mPETME7Dr59dc801vPPOO6Snp7NixYomTByiK3siIiIiIiJ1sK6L++BfsE89DhvWQ9Ee2LAe+9TjuNPuw7pug/u+6qqreP/995swbU0q9kREREREROpg534BK5ZBWVnNFWVlsHwpdu6XDe574sSJpKamNjJh3VTsiYiIiIiI1MHOfPvQQu+AsjLszLeaN9ARULEnIiIiIiJSl8IdjVvvIRV7IiIiIiIidenYqXHrPaRiT0REREREpA7mrPMgNrb2lbGxmLPOb95AR0DFnoiIiIiISB3M+Akw9NhDC77YWBg2HDP+xAb3PXXqVMaPH8+aNWvIzMzkqaeeamTamjTPnoiIiIiISB2M4+Dc9nPs3C9Dg7FUzrN3Pmb8iY2aZ2/69OlNmPRQKvZEREREREQOwzgO5oSJcMJEr6McEd3GKSIiIiIi0gap2BMREREREWmDVOyJiIiIiIi0QSr2RERERERE2iAVeyIiIiIiIm2Qij0REREREZHDcK1lev5mxi/5hO7z3mH8kk+Ynr8Z19oG95mTk8Mpp5zC4MGDGTJkCNOmTWvCxCGaekFERERERKQOrrVctmous3bnU+wGAcj3l3HT2sW8viOPlwaNwzHmiPuNiorigQceYMSIEezdu5eRI0cyadIkBg8e3GTZdWVPRERERESkDi8V5NQo9A4odoN8sms7LxfkNKjfrl27MmLECACSk5MZNGgQeXl5jc5bnYo9ERERERGROjyUt/aQQu+AYjfItLy1jd7Hxo0bWbJkCWPHjm10X9Wp2BMREREREalDbtn+Rq2vz759+7j44ot58MEHSUlJaVRfB4tosWeMOdMYs8YYs84Y84ta1vcwxnxqjFlijPnaGHN2JPOIiIiIiIgciczY+EatPxy/38/FF1/M9773PS666KIG91OXiBV7xhgf8ChwFjAYmGqMOfhpw7uBl621xwFTgMcilUdERERERORI3ZLRj0THV+u6RMfHrRn9GtSvtZZrr72WQYMG8bOf/awxEesUySt7Y4B11tpsa205MAO44KBtLHDgWmU7YEsE84iIiIiIiByRyWndObV9+iEFX6Lj47QOnbksrXuD+p0zZw7PPfccs2bNYvjw4QwfPpz33nuvKSJXiuTUCxlA9aFpcoGDnzj8HfChMeYnQCLwndo6MsbcANwA0KNHjyYPKiIiIiIiUhvHGF4ePJ6XC3KYlreW3LL9ZMbGc2tGPy5L696gaRcATjzxRGwj5ukLh9fz7E0FnrHWPmCMGQ88Z4wZaq11q29krX0SeBJg1KhRkf2OiIiIiIiIVOMYw5T0HkxJb10XniJ5G2ceUP2aZmZFW3XXAi8DWGvnAnFApwhmEhEREREROSpEsthbAPQzxvQyxsQQGoDlrYO22QycBmCMGUSo2CuIYCYREREREZGjQsSKPWttALgZ+ABYTWjUzZXGmD8YY86v2Ox24HpjzDJgOnCVjfSNqyIiIiIiIkeBiD6zZ619D3jvoLbfVPv7KuCESGYQERERERE5GkV0UnURERERERHxhtejcYqIiIiIiLRo1loKs/1sXVVGebFLTKJD18GxdOwdjWng1AulpaVMnDiRsrIyAoEAl1xyCb///e+bNLeKPRERERERkTpYa/l2VjF7tgZwA6E2f2mQ7LklFG6Kov8piQ0q+GJjY5k1axZJSUn4/X5OPPFEzjrrLMaNG9dk2XUbp4iIiIiISB0Ks/01Cr0D3ADs2RKgcIO/Qf0aY0hKSgLA7/fj9/sbfJWwLir2RERERERE6rB1Vdkhhd4BbgC2rixrcN/BYJDhw4eTnp7OpEmTGDt2bIP7qo2KPRERERERkTqUF7uNWn84Pp+PpUuXkpuby/z581mxYkWD+6qNij0REWnzctYWsm9PqdcxRESkFYpJPHzJVN/6cLRv355TTjmF999/v9F9VadiT0RE2qzCbfv4wah/cNnAxzin2//xyqMLvI4kIiKtTNfBsTh1DGvpREHXIbEN6regoIDdu3cDsH//fj766CMGDhzY0Ji1UrEnIiJt1tP/+wXfLtkGQHlpgIdu/5Cd2/d5nEpERFqTjr2jadc16pCCz4mCdt2i6NgrukH9bt26lVNOOYVjjjmG0aNHM2nSJM4999wmSFxFUy+IiEiblfNtYY3lgN9l68Y9pHZO8iiR1MZaS9n+AHEJDTtgEhGJJGMM/U9NpHCDn60rq82zNySWjr0aPs/eMcccw5IlS5o4bU0q9kREpM2acMEAFnyyoXK5c/cUBozo4mEiOdjizzfx5+veJi97F8ed1JM/vPBdOnVN9jqWiEgNxhg69Y6hU+8Yr6McERV7IiLSZl3y41EEyoN89NJKuvRoxw1/OJmoaJ/XsaRCIODy2yveYMeWvQAs+XwTD93xEX944SKPk4mItA0q9kREpM0yxjD1p+OY+tNxXkeRWhRu3VtZ6B2wZvFWj9KIiLQ9GqBFREREPJGemUL3fqk12kae0sujNCIibY+KPREREfGEMYZ7X7uUESf3pH2nBM668hhuuvc0r2OJiLQZ9d7GaYy5D/hfYD/wPnAM8FNr7fMRziYiIiJtXO8h6Tz6yfe9jiEi0iaFc2XvdGttEXAusBHoC9wZyVAiIiIiIiIthetaPnhhOVeP+Sdnd/0bV4/5Jx+8sBzXtY3uOxgMctxxxzX5HHsQ3gAtB7Y5B3jFWrunoXNJiIiIiIiItCaua/nlJS+z4OMN7C/2A7Arv5h7f/Qus15fzT2vXIrjNLw+mjZtGoMGDaKoqKipIlcK58reO8aYb4CRwCfGmDSgtMmTiIiIiIiItDAfTV9Ro9A7oLTYz/yPsvloxooG952bm8u7777Ldddd19iYtaq32LPW/gI4HhhlrfUDxcAFEUkjIiIiIiLSgsyY9tUhhd4BpcV+Zjz4VYP7vu2227jvvvtwnMiMm1lvr8aYaOAK4CVjzKvAtUBhRNKISKtibePvUxcRERFpybbnHP72yvrW1+Wdd94hPT2dkSNHNuj14QinhHyc0C2cj1V8jahoE5GjlLv9efzzehCYk0Jw7U1YG/A6koiIiEhEdO6e0qj1dZkzZw5vvfUWWVlZTJkyhVmzZnHFFVc0qK+6hFPsjbbW/sBaO6vi62pgdJOmEJFWw5ZuIvjt9eDfBrYMd9s/cLc+6XUsERERkYiYcutY4hKja10XlxjNlNvGNqjfe+65h9zcXDZu3MiMGTM49dRTef75pp3dLpxiL2iM6XNgwRjTGwg2aQoRaTXsvkUc/F+A3TvfmzAiIiIiETZp6lDGfKfXIQVfXGI0Yyb1ZtKUoR4lq184Uy/cCXxqjMkGDNATuDqiqUTqEbziIqj+vFi/gWBdnLPPx4w9AZu7Gfel56EgHzP2eMwFF8P+UuyMZ7HfrILyMhIvmoStmEakQzBIZruOZMTE88esIRyT1J43d+Txf3lrca3ltox+XJSWycriPdy9cQWbS0so9Jex1V8GgA/YNeZcYmNjm/R97tm5n0d//jEr5uVyzPHd+fG9p5HSIb5J93GkTPJYMNFgqx5UdlJO9DCRiIiISOQ4juGeVy/joxkrmPHgV2zPKaJz9xSm3DaWSVOGNmrahQNOPvlkTj755MaHPUi9xZ619hNjTD9gQEXTGmttWZMnEQlT8Gc/qlnoAaz9BgD34QcwHTpiH74fdobGEbI5myAmFrLXYed9CUC766Zgq416tCsqil3Fe1hevIcl+3bx+pDjmbJ6Hm7F+su/+YrPY+KZ8s08tpQfOvNIEOi04F32nnhRk77XP1/7FrPf+haADat2ULRzP39+5dIm3ceRMrEZ+AY+R3Dj3eDfidP5KkyXazzNJCIiIhJJjmM44/JhnHH5MK+jHJE6iz1jTF1HrX2NMVhrX49QJpHD276t7nXWwuxZlYVeZfOShZC9rnLZ7/PV3b2/jGe2baws9AAs8O/8jbUWegeUR2Bkyjnvrau5/O7aJt9HQzidLsLp1LSFrYiIiIg0rcM9s3dexde1wFPA5cD3gH8COo0v3jlMoQZA3/4QHVOjyWRkQkZmWN1HG8PYlNRD2scmdyTGRGYOlLpkDepUc3lwWrPuX0RERFq2QMDlmT9/wQ8nPsN9P36PnfnFXkdqVVr6NFKNzVfnkau19uqKkTejgcHW2kustRcDQyraRDzhe/bV2lcYBzPxVMzEUzBXXQ9xFc+29emPuWgyzjU3QsdQsfT3j7+ouhXUWnBD1/ESHR8P9D6WK9J7cl2XXvgwOMBVnbO4snNPpvUdTpKv9gvij/cZ3pRvE4BfPHEO6Zmh4Xw7d0/hrsfObvJ9iIiISOv15G8+5e//8xlfz8nhjb8v4leXvuJ1pFYjLi6OwsLCFlvwWWspLCwkLi6uwX2Y+t6cMWa1tXZQtWUHWFm9rTmNGjXKLly40ItdSwsTfPIRKCzE98vfYvcWgbWYlHaV621ZGewtwnSquhpm3SDs2AEdO+Fu2MBNc95lwtiT+N7A4eSWldAhKobEasVcob8M10JaTNXAK8XBADsD5XSPTeDdgi18snsbf+s3ImLvMxBwyc/ZQ+ce7fD5mvfKooiIiLRslw18lJy1O2u0vbv1Z6SmJ3qUqPXw+/3k5uZSWlr3Yzpei4uLIzMzk+jomtfajDGLrLWj6nt9OMXeI0A/YHpF02RgnbX2Jw2L3Dgq9kRERNqOb5du4683zyR7RT7Hn9WXux4/h+T2DT+LLXK0ue2sF/nqw/WVyymp8bydexsxseEMui+tVbjFXjijcd5cMVjLhIqmJ621bzQ2oIiIiBzdXNfyq0tfJS97FwAfv7yKhORYfvnkuR4nE2k9bv7LaWSvzKcgby9xCdHc/tCZKvSkUlj/EipG3tTomyIiItJkCvKKKgu9A5Z+scmjNCKtU99jOvP6+p+wfkU+Gb07kNROV8alSr0PABljxhljFhhj9hljyo0xQWNMUXOEExERaayFszZw34/f49/3fsm+PS33uYyjUaduyZWDUB0weHSGR2lEWq+oaB8DjuuqQk8OEc6VvUeAKcArwCjg+0D/SIYSERFpCl+8tYafX/Ry5eC7s99cw1Nzr/U2lFTy+Rz+OP0i7rnhHTZ9s4Mxk/rwk/sneR1LRKTNCPc2znXGGJ+1Ngj8yxizBPhlZKOJiIg0zn/+sYTq45Ctmr+Fb5duo//wLt6FkhqOOb4701f8iGDQ1YjDIiJNLJxir8QYEwMsNcbcB2wljNs/RaRts8XLsf6dmHYnYIweBJeWKald7CFtiSmHton3VOiJiDS9cP5nvbJiu5uBYqA7cHEkQ4lIyxZYcy2BxSMJLp9EYNGx2PLtXkcSqdWVdx1fYxj/C647jozeHTxMJCIi0nzCOR0/EfiPtbYI+D2AMeZcYF0kg4lIy+TuXYjNf66qYf9a3C2P4Mv6o3ehROrQ95jOvLruZr76YD1derZn2PhMryOJiIg0m3Cu7D0MfGGMGVSt7Q/hdG6MOdMYs8YYs84Y84s6trnMGLPKGLPSGPNiOP2KiIfKtx7SZMu3eRBEJDwpHeKZNGWoCj0RETnqhFPsbQCuAV41xlxa0Wbqe5Exxgc8CpwFDAamGmMGH7RNP0IDvZxgrR0C3HYE2UXEA6b9aRDTtXoLTvrlnuWR2llrsfuWYWunoIGuAAAgAElEQVQpzkVamvkfZzNj2ldsXF3gdZRKbtFcAqsuI7DyItxds7yOIyLSIOHcxmmttYuNMScB040xYwFfGK8bA6yz1mYDGGNmABcAq6ptcz3wqLV2V8WO8o8ovYg0O+NLIOqYT3HzHsT6C3G6/ACn/Slex5JqbPk2AsvPhpIVgA+nx6/w9fwfr2OJ1OrBn33IS9O+AuARn+HPr1zKxAsGeJrJ7s8muPwMcEPzMgZ3vY85bh4m8RhPc4kA+MuDFG7bR5ce7byOIq1AOFf2tgJYa3cAZwAWGBrG6zKAnGrLuRVt1fUH+htj5hhj5hljzgyjXxHxmInvja/vQ0QNegGnw+lex5GDuDn3VxR6AEHczf+LLd3gaSaR2uzdXcqrjy6oXA4GLc/dN8fDRCHuzncqCz0AbAB3x3+8CyRSYc67azm/+4N8t9dDXD7scXLW7SQQcHnmz19w40nP8Neb3mNnfrHXMaUFqffKnrX2nGp/d4E7K76aav/9gJOBTGC2MWaYtXZ39Y2MMTcANwD06NGjiXYtItI22bKDCzuLLd2MievlSR6Ruriuxbq2RlvA73qUpoqJ7XloW1xW8wcRqSbgD/Kn695m944SADas2sEjd35E1uA0nr03dJJk2Zc5ZK8s4PHPfuBlVGlB6ryyZ4x5sOLPt40xbx38FUbfeYSmaTggs6KtulzgLWut31q7AfiWUPFXg7X2SWvtKGvtqLS0tDB2LSJy9HI6XVSzISYDkzLOmzAih9EuNZ6zrqx5a+TkW8Z4lKaK6XguptrnyHQ4HZM22cNEIrB7Rwm7Drpql71qB5++trpG29IvNuvqnlQ63JW9A2Or39/AvhcA/YwxvQgVeVOAg0dx+A8wFfiXMaYTods6sxu4PxERAZz074Hrx82fDrFd8XX/FcbRROLSMv3yH+cyZlJvNqwqYNyZfTn2hO71vyjCjPERNWgGtuRbsAFM4uD6XyQSYZ26JtP/uC58u6RqBOzjz+7Lpm8KyVm7s7ItJTWepHb6P19C6iz2rLWLKv78vCEdW2sDxpibgQ8IDejytLV2pTHmD8BCa+1bFetON8asAoLAndbawobsT0REqjhdrsLpcpXXMUTq5fM5nD41nKEAmpct/ppg3qNgAzjdfoSTPMrrSCLc+9qlPPrzT1i/Ip9xZ/Thh/97Crlrd5K9Mp+CvL3EJURz+0NnEhMbzhiMcjQw1traVxiznNBgLLWy1noyJNWoUaPswoULvdi1iIiIHAVsWS6BRcdAcF+owcQSNWIBJmGgt8FE6hDwB1m/Ip+M3h1IahfndRxpBsaYRdbaes9CHa7sP7cJ84iIiIi0Cu6ON6sKPQBbhlvwKr6ed3sXSuQwoqJ9DDiua/0bylHncLdxbmrOICIiIiItgYnpXEtbFw+SiIg0Tr3z7BljxhljFhhj9hljyo0xQWNMUXOEExEREWlupuMFmPanVi0nj8WkHzzGnIhIyxfO05uPEBpJ8xVgFPB9QqNmioiIiLQ5xokmatj7uEXzQ6NxpozHGON1LBGRI1bvlT0Aa+06wGetDVpr/wWcGdlYIiIiIt5yUsbgtDtehZ6ItFrhXNkrMcbEAEuNMfcBWwmzSBQRERERERFvhFO0XUlonrybgWKgO3BxJEOJiIiIiIhI49R7Za/aqJz7gd9HNo6IiIiIiIg0hXBG4zzXGLPEGLPTGFNkjNmr0ThFRERERERatnCe2XsQuAhYbq21Ec4jIiIiIiIiTSCcZ/ZygBUq9ERERERERFqPcK7s3QW8Z4z5HCg70Git/VvEUomIiDQBay3P//W/fPzyKrp0T+GH/3sKvYekex1LRESkWYRT7P0J2AfEATGRjSMiItJ0XnlkAY/9chYA3y7ZxupFW3lt3U+IjvF5nExERCTywin2ullrh0Y8iYiISBP78u1vaywX5O1lzeKtDB2X6VEiERGR5hPOM3vvGWNOj3gSERGRJtZjQMcay9ExPrr17uBRGhERkeYVTrH3I+B9Y8x+Tb0gIiKtyTV3T2DwmG4AxCdGc9v/nUFqeqLHqURERJrHYW/jNMYYYIi1dnMz5WkS/i+qP1rYDqfn7ZiEAZiOF2CMgy3bglvwMvgScdImY6JSsMESbMHL2EAh7oZfwhNnAgZwibr9N9jNGzHDjsX06Q+A3ZiNXboIk9EdRo7BOA52107s3C8gLh771ONVEdLS8T349+b8Fog0GbdoHu62f2KcBJyMn2Di+2GDxdj8GdhAIU7apZi4Xkfcry3cgf3vFxAXhznhJExCQgTSy9EutXMST829li0bdtGuUwKJybFeRxIREWk2pr4ZFYwxy621w5opT71GjRplFy5cWOf6moVeTabzD/D1+DWBJeMgsDPUmDAI3/D/4n49Cbsv1K954mwMppYODObGWzCxsbjT7gfrhppP/g7mgotx774TivfVvvOLp+K76LLw3qRIC2H3LSGw9ESw/lBDVEd8I7/GXXkedt/iUJuTSNTwLzCJ4T/aa7dvDX1eSopDDd0ycf70ACZGY0BJTcGAZdfm0L+/Dj2i8UXV8n+ztAmBgEtUVDg3HImIiDFmkbV2VH3bhfO/6mJjzOgmyOQ5u/05gnmPVBV6ACWrsTn3VxZ6fNTlMB1Y7Ltv4r77ZmWhB2A/n4X7wbt1F3oAr01vZHqR5ufmz6gq9AAChbi5f6sq9ADcYtytTx5Rv/bTj6sKPYAtubBsUSPTSlsTKLcsf2sv62aXsG52Ccvf2kugXFO+tjXL5uQwefBjTIj9E7ee8QI784vrf5GI1LB+RT7T/28e8z/K9jqKtDDhFHtjgbnGmPXGmK+NMcuNMV9HOliz0olikdpFpx3SZKJrG9ziCM/GO7VdOdcZfampcEM5pUVVJ9ZKi1wKN5R7mEiaWjDo8pvLX2fzmkIA5n+czcN3fORxKpHW5dPXVvP9457koTs+4tYzX+ChO/UZkirhHF2dAfQBTgXOA86t+LPVMZ2/jy/jZoiqNjpbwiBM5h2Y5IqLl5O2HaYDgzn3QpxzLqxxYGpOPg3njHMhKanu1148tZHpRZqf0+VaSBhSuWxSz8Zk/AyTPLZqI18KTrcbj6hfc8rpkJxS1dAjC4aPaGRaaWvcYC1tAV3Za0t2bNlLfm7NMd9WLcjzKI1I6/TcfXNw3ar/G195eD7FRWUeJpKWpN559qy1m4wxxwITKpq+sNYui2yshoueUH7oAC1Zd2DiB2I6nocxDlEjl+AWvIxxEjFpl2F8iZhhH2N3vIL1F+Le+HNs9QFa7vhtxQAtwzG9+wLg/Ol+7NLFmIxMGDEa4zg4907DzvsyNEDLPx+ripCWruf1pFUy0R2IGrEQu2c2OIk4KWNC7cd8hN3xGta/E6fTdzGxRzZnmUlLx/nLNOy8ORAXjxl3AiYqOhJvQVqxjr2iyfu6lEBp6CAmKtbQsZee62xL0jJSyOjTgbz1uyrbjpvY08NEIq1PIODWWLaurVH8ydEtnAFabgWuB16vaPou8KS19uEIZ6tVfQO0iEjrYN0gfLMa4uMxvfp4HUdaqLJil4J15WAhrV8MsYm63betWbtsG3+9eSbZKwo4/qy+3PX42SS1i/M6lki9VhcX8XJBDmkxsVyZ3pNkj05avvfsMv549VuVyxdcP4JfPHGOJ1mk+YQ7QEs4xd7XwHhrbXHFciIw11p7TJMkPUIq9kRaP1u8D/cPv4bcilldRo3Fue3nhGZ7ERERadkW7t3Jqcs+p6xiwL7jktrz3+Gn4nj0e2zZnBzmfbCOPkPTOfWSwTi1PRsvbUq4xV69t3ESupex+pMTQTSkiYg0gv3kg6pCD2DhV7ByOQz15BySiIjIEXlya3ZloQewZN9uvtyzg4ntDx3YrDkce0J3jj2huyf7lpYtnGLvX8BXxpg3KpYvBJ6KXCQRafP27D6kye7ZpbNIIiLSKkTXMoJ0jOPdbeYLZ21g7vvr6TM0jdMvH6Y5K6VSOAO0/M0Y8zlwQkXT1dbaJZGNJSJtmTl+IvbD98CtOCualIwZPtLbUCIiImG6OaMvrxTksicYmov2tPbpjEvpWM+rIuOtp5Zwzw3vVC4v/mwTdz99vidZpOWp95m9lkbP7Im0DfabVdhZH4YGaDnzXEzXDK8jiYiIhG1beSn/3JrNf4sKaeeL5pquvZjUoXOz57jyuL+z7uv8ymWfzzAz/w6S22ugo7as0c/sGWP2AgcqQVPt71FAjLU2nFtARURqZQYOxgwc7HUMERGRBkl0fPx9azb5/tCcdm8U5vHhsInN/txeTFzNQ3JflINPt3FKhTr/JVhrk621KRVfyUA34E/ANmBacwUUEREREWlpPt6dX1noQeiqyIyCnGbPcdUvT8Tnq3rqfcpPx5GQpDlJJaTeq3PGmPbAbcD3gReB0dbawkgHExERERFpqbrGHHqbZLda2iJtwvkDmL7yR8z/eAN9hqYzfEKPZs8gLdfhbuPsBNwOTAaeBo6z1u5prmAi0nJZa7FFX4K/ENPhdIwvwetIIiIizWpcSke+l96DF/JDUwkNTkjhxm59PMnSvV9HuvfzZoAYadkOd2VvE1BAaOqFEuDa6hMeW2v/FtloItJSBVdfgi18O7QQk0HU8NmYWM3vIyIiR5enB4zm9sz+7Ar4GZ/SEZ9Hk6p/NGNFaOqFYelc/ONRxMVHe5JDWp7DFXt/pWpQluRmyCIirYBbNLeq0AMoz8Pd8ii+Xvd6F0pEGsy1FsejA1SRtmBIYjtP9z/9/+bx0B0fVS4v/28O9752mYeJpCWps9iz1v6uGXOISGsR2HVIkw0cOkm6iLRsQWu5M3sZT23bQJITxe+yhnB9195exxKRI/T20zWnv5795hr2FJbQrqMesZDDjMYpIlIb0/40iOtVrSEKp/MPjqgPay3uqy8SvPEHBG/7Ie6XnzVtSBGp1/PbN/HolvWUui47AuX8ZN0SVpcUeR1LRI5QSmp8jeXY+GhidRunVFCxJyJHxDixRB37OU73X+J0/SG+Yz7FSRl/RH3YeV9i33gF9hZBQT72iYexW/MilFhEajN/784ayxZYtPfQK/ciUrfXCnIZvOB9us19m7s3rMBaW/+Lmtj1vzuZ2Piqm/Wu+91JxCWo2JMQTYwuIkfMxHTBl/X7hnfwzaqay9bFfvsNpmtG44KJSNgmtOvEP7dtqFx2gPEpGs1PJFybSov5/pr5BCoKvL/mrqF/QhLf75zVrDlGnpLF69m3sHT2JnoPTSdrYKdm3b+0bPVe2TPG3GqMSTEhTxljFhtjTm+OcCLSRvXtf0iT6XNom4hEzuS07vyq+0DSomPpHZfIMwPG0Cc+yetYIq3GV0U7Kwu9A77cs8OTLKnpiZx6yWAVenKIcG7jvMZaWwScDnQArgQ07J7IUczd8Tr+BYPxz+1MMPsXWOse0evNCSdhzjoPYmIgJQVzzY2YTE3d0NSs6/fklqKWJOAP8vBdH3Nh1jRuPOkZVn6l24UPMMbw26wh5I47l9Wjz2Ryuj6DIkdiZHKHQw6kRyenepJFpC7hFHsHxmM+G3jOWruyWtvhX2jMmcaYNcaYdcaYXxxmu4uNMdYYMyqcfkXEO7Ysj+A3V0LpOgjsws37G3bb00fUh3EcnCuuwXl6Br7H/41z2hkRSnt0ssH9BL65ksCcFALze+EWvOJ1JM+88MBcXnxgLttzilj2ZQ53nD+D0v1+r2OJSBvQJz6Jx/uNJD06lljj8MOuvbmmS6/6XyjSjMJ5Zm+RMeZDoBfwS2NMMlDvaXxjjA94FJgE5AILjDFvWWtXHbRdMnAr8NWRhheR5mf3fgW25sGyW/QlTtfrjrgv0wbm9vKXB5n95hqKi8o46bsDaXfQqGhecPP+hi14KbRQvoXgmmsw7U7GxKR5G8wDCz7ZUGN5944S1i7dzrDxmR4lEpG25KouWVzVJUvzVUqLFc6VvWuBXwCjrbUlQAxwdRivGwOss9ZmW2vLgRnABbVs90fgL0BpeJFFxEsmaSTgq9mWPMabMB4LBFx+fMq/uXvKa9xzwztMHfI42zbv8ToWdu+CgxrKsMXLvQnjsf7Du9RYjo2PoudADUJywFdFhfxo7SLuyv6aTaXFXscRabVU6ElLFU6xZ4HBwC0Vy4lAXBivywByqi3nVrRVMsaMALpba989XEfGmBuMMQuNMQsLCgrC2LWIRIqJ64mv/z8gpiuYWJwuN+B0vcHrWJ6Y/+F6VsyregZsV34xb/x9kYeJQky7iTUbfEmY5JHehPHY1b+ewPgz+wLQIS2Bu586n5QO3l99bQkW793FaV9/ztPbNjItby0Tl37G3oBucRURaUvCuY3zMUK3bZ4K/AHYC7wGjG7Mjo0xDvA34Kr6trXWPgk8CTBq1Kije7QBkRbA6XwFTucrsNZW3oppg/uxha9jA/swyaNxc+/HlqzGST0Lp+dvMU6sx6mbXnlZ8JA2fy1tzc3JuAXK8nDzp0NsV3y9/oKJatfsOTaXlnBPzmo2lZZwaVomV3vwLEty+zj+9u5UiovKiE2IJipK08se8Hz+JvzVBvDZ5i/l/V3buDRNA7WIiLQV4RR7Y621I4wxSwCstbuMMTFhvC4PqP4bI7Oi7YBkYCjwWcXBYhfgLWPM+dbahWGlF5GIc3d9hJv7N7AuTsYtOB3PqVxXWei55QSWnQTFSyvW+IBQ0eOWrAQsvl73NG/wZjD+rL5075dKztrQ5NTxidGcf+1wj1OBMVH4+jyAr88DnmUIWstZK75g3f59AHyyOx/Ak4IPIDGl7Z1saKyOUYd+T1Kjwvn1LiKNFfRbyotd4to5beL5dWm5win2/BWDrVgAY0waYQzQAiwA+hljehEq8qYAlx9Yaa3dA1ROBmKM+Qy4Q4WeSMthS1YTXHkB2AAAwT2fY46bF/r75nsgsBOny1XgS65W6MGBQu8At/DdNlnsxcZF8Y//XsO7zyyluKiMM783jO799DwYwNJ9uysLvQNeLsj1rNiTQ93QtTcvFmyu/DmdldqFU9unH3E/btBSvDNIXLJDdJyunIrUp3BjOdlzSgj6ITbZYeBpicS399X/QpEGCKfYewh4A0g3xvwJuAS4u74XWWsDxpibgQ8IneZ/2lq70hjzB2ChtfatRuQWkWbg7pxZWehVtOAWvIa77UkI7AJCBaCTecdh+zEJgyKY0lvtUuO5/GfjvY7R4nSLjcOHIUjVbYI9YxOaPYe/PMjDd37Exy+vokuPFG594HSOPbFHs+doidJiYlkyYhKzdueT4ovi+HZHPhlzcWGAbz4uxr/fYhzoNT6e9H66iipSFzdo2TB3P8GKx2PL9rpsWrifgd9J8jaYtFn1FnvW2heMMYuA0wjNr3ehtXZ1OJ1ba98D3juo7Td1bHtyOH2KSPMx8X0PabNuWWWhd4Bblgfx/WH/t6EGJx5MDAT3QMKgNnlVTw6va0w8v8sazO82riKIpXdcIr/oMbDZczz/1//yyiOh0Ul35Rdz54Uv8dbm24hLiG72LC1RjONwZmqX+jesw+ZFpfj3hwp668Km+fvp1CsGJ0q3pYnUJlBmCZTVHH6itCicG+ZEGuawxV7F7ZsrrbUDgW+aJ5KItBQm9VxM+hXY/BcAi+l0KU7nKQS3PFhjOye+H07fabjbX4DgXpz0qRDTGcq3YuJazm17gTKXPVsDxKX4SEzVLTORdlf3gVyR3pO88v2MSOqAz4PnUhZ9trHG8t5dpaxdpnn2qtsXDBBlDHHOkX8myoprHqQG/RAot8So2BOpVUyCQ2InH8U7qh53aNctiu3flOGLMaT2iNbJEmlShy32rLVBY8waY0wPa+3m5golIi2DMQ6+nr/DTR6PSToGJ2UsADbzdtzcB4EgJnkcTsbNmKj2+DJuqtlBCyr09u0IsPqDfZW3znQbFkuPkRqCP9K6xcbTLda77/OgkV1ZNGtj5XJcQjRZg478dsW2yO+63Lh2EdPzc4j3+fhVj4HcnjngiPromBVN3rKyyuXkzj5iEvTcnsjhDDg1kZzFpZTsCpKU5mNHdjnB8tC6xE4+hp6dhHFU8EnTCOeZvQ7ASmPMfKByxlVr7fkRSyUiLYKbP53gmmuAIJgoGPAMTtpl+HrdExreP7C71TyPl7estLLQA9i6ooyug2OJijUEyizR8TpAbYuu+vUEctbu5Iu3viW1SxK3P3Qmye3DmSq27fvX9o08nx86j7svGOBXG1ZweocuDEsMf5qOzGPj8EUbducFSGjvkHGsvrci9YlJcOhzYugZ5s0L91cWegDFO4Ls3hKgQ6ZuNZemEU6x9z8RTyEiLVJww6+pHFnTBghuuBsn7TIATEzX0MTqFWzZFojqgPG1zKtlBz8jYS3syvWTs6QUf4klsaOPficnEJes2ztbkv27g2xfE7py1HlA7BGPWJeYHMu9r11GWWmAmFifhjiv5ut9uw9pW7Zv9xEVe8YxdBsaR7ehTZlM5Ohha5k92uoRPmlC9Z7KttZ+Tuh5veSKr9UVbSLS1gV2HrRciHX92OLl2GAJALY8n8DSiQTmZxH4qgfu9uc8CFq/9P41RwhM7uwjZ3Go0AMoLgyyeWGpF9GkDmXFLive3cu21eVsW13Oinf3HvKMWLhi46JU6B3kOx0611iONoaJ7dIivt+cdTv5ZtFW7EFHuf7yYB2vEGm5Zu8uYNLXnzN68cc8vmX9Eb8+vV8MTrVLL/HtHNpnhHMtRiQ89f5rMsZcBvwV+IzQaJwPG2PutNa+GuFsIuIxp8vVuFserVw2qecQWNAPyrdAVHt8A57B7pyJ3Ruae4/gHoLrbg4N7BLdwaPUtUvrG0NUrGHnJj9xKQ6pPaNZ9sbeGtuU7NTBZiQErEuUOfLbZAs3lNe49TboD7V1G6pbBZvChZ0y+EuvYfxj2wZSfFHc3WMwPeIiOz3Gn69/m7efDs3JOWh0Nx764HusX57Pn69/m83f7mT0ab343XMXktpZw9BLy5dfXsr5K+ew3w397rht/VK6xsRxYaeMsPuIb+/jmAuS2ZHtxxdtSOsTjePTiamWxF/qsnH+fvblB0lK95E1Jr5VzSkazqmDXwOjrbX5UDmp+seAij2RNs7pfT8kDMIWzcWkHB8abbN8S2hlYDfBtTdDXFbNF7n7oXQDtLBiD6BD92g6dK96DiKhg0PJrqorRe10NrVJvVO4hZ+uX0ZuWQnf7ZTBk/1HkeQL/3scFXvoAU9tbdJwt2X257bM/s2yr6//m1NZ6AGsXrCFVx9byBtPLCI/twiABZ9s4KE7PuJ3z323WTKJNMbnewoqC70DZu7cdkTFHkBcso/MY/UIQUuVPaeEXTmhOYfL9rkEy22rmhcxnLLUOVDoVSgM83UiERW88QcEr54MgC3ag91T8/kTW1qKLciv2eYGsdu3YoNBgm++wg1P3MMzX3wIwObSEvYG/DW2LygvI7+85q19+4IBNpWGxir6T34ut3y7uEnfV0tijA9f1xuIGvAvfF2vh9KDblEpz8Vpf1rNtphMSBzWfCEbof8pibTPjCI2yaHzgBiNztmEigJ+frBmAZvLSnCB13bkcc/msKZordSxVwyJHasOgBI7+uiYFdPESaW5bN+855C2Td/sqCz0Dli9cEtzRRJplMEJKYe0DUpIrnN761pyl5by9ZtFrJlVzP49upukNdidF6ixvOeg5ZYunFOs7xtjPgCmVyxP5qCJ0kWaW/B73z3078ZgTjwJc8PN2NmfYp97CkpLoXdfnNt/BTsLcR/8CxTu4F/9srjp1BMgtT//tkX88PNXwHFIcHzc02sYN3TtzU3rFvPMto0AXNG5J0/0G8kz2zZyZ/Yyig86k/f37Rt4vM9wrunWp7m+BRFlg/uxO98FE41JPQvjhA6wnU4X4m79e+V2psPpOD1+AQRwd7yOieuFL+uPGKd1jCIWl+JrVWfnWpPVJUXsC9b8hbhw764j6sMXZRh6ThJ7tob6adc1SsORt2Jjz+hDcvs49u4OnUAzBs656liWz80lb33Vv43jTsryKKHIkRmS2I7f9xzCPTmrKXVdzuvYlR92rfs4YMuKMnKXhv79l+xyKdkZ5NjvJpGzuIz8b8vwRRu6j4gnra9OarUkCR18FBdWHfcltLJ5es3BD0jXupExFwMnVCx+Ya19I6KpDmPUqFF24cKFXu1eWoDqhV5tzHU/xv77H+CvukpnvnMmNnsdZK8DIOGH3wsdadQi2hge7zeC675dVKP9yX4juXndEsoPM0xW2YSLw30bLZb17ySw9EQoDX2vSDqOqGM/xzhxWLcUd9MfsXs+g5ju2LI8KNuE0+kCnN5/BX8hwbU/xO7+DJM8El+/J1rN1AzStEqCAXrNf4/d1a6W/0+PQdzdc7CHqeT/2Tvv8Diqe3+/Z2b7rnpvlmTLvRsbA6ZjCJ1AIHAJIZ0bEpKbdvNLCBeSAAklPTeNEHJTCAkJvUMA040b7k22eu9t+8yc3x8j72q1km0Zy5LMvM/j5/GcOXP2rHZn53zOtx0JwR4df5dOSp4Np/f9Ofbs397GX+95m4HeEJd9bhmnXjyLfVtbufem56je0c7JF1Tw37+6AF+aFZdpMTWQUvJcVzMd0QhX5hTjOYir+vZn+hloT9wsLlripHFzvFYlApZcnoIrdWoJiuMZf6dG5ZoAoX4DV4rCzDM9eLMmPuxDCLFRSrn8UP0Oa6ZSykeAR973rCwsjgX7KxOEHoBsbIDGhsO6PCol7/Z1JbW/2995UKF3vGC0/SUu9AAG3kN2PA5ppyGD+1Cm3QLiu/FELWBa+2xpSP8OZLfpFiv73kHb/Qnsy9ZNxNs4phiGxDAkNpvl4X4Aj2rjn3NP5htVW6gNB7gyu5j/LhlbwW6Liad5Z5jadUEAhGK6Pg+Nex0rMxbkctufP5zQVrEoj9+9/sn3M00LiwlBSsmHd7zN890tAHy/bievLz6LQufIIQHudDVB7Ck2CPcPW9Nl4qUAACAASURBVFdI6G/XLbE3ifBm2Vh8RQrRkMTuElMus/OoKxMhRL8Qom+Ef/1CiL7RrrOwmFCEgNPPhqzsxOalyxHL4psfdn10P/k8u5NP55cn3BwK8Mm8Mooco8d0OafYzT8q+kBSk9H1DNq6CvRt56Ktr0B2Ph5P1DKI7HkV2fdO4oX+zUg9OJ6znXAe/e1GLsz/Meek3sW9Nz2Hrh8fGwKyuwv9xz9A/9x16Hd/H9nRPuYxTk/PYd2y1bSefCm/mrkMp2ItXqYShiFpeC9+/0oD6t+zypNYWBzgjd6OmNADqA8H+W3z6OUXSpa4YnHIqh3KT/KQmj/M7iIgJcf6rZxsCCFwuJUpJ/TgIJY9KeXoEaYWFhOI+uBjya6cs+eCYaBccCli1hzkN2/FePiv0N6GOPEUxAWXQCgIHi9yz056//Aw3s9dgxy8aTNDYYqz8yh2uvle6XwW+dL5x7yT+WnDXgwp+WrxLE5MzeKpBau4tWYHteEAnZEQTVHT9UIF2ldcdIz/EuODknstRsNPQR/c07HnIjueJFZcPdqO3vInUNNAjydcEN6lYM9Bdg0J6fUunrRF1o8GNbs7+NFNz8aK4j76mw3MXpLPpZ9dOrETOwoY9/8aNg+6Mm99D+O3P0e95Y4xjdEVjfDrpn3UhgN8JLuY8zPzx2GmFuOFNGBY2CV65NChHwfQIhLVhhVnaXHc0qNHk9q6teS2Azi8CgsvSSHUr2N3K6g2gZSSYI+RELNnWfUsjiaHFbMHIITIBWJO9FLKuvGa1MGwYvYsLMYfGazEaPm/wQQtF6JvOTWxg3chaun30ffdBJFGRPpq1Dl/ASOIvvdzgzF7y1Bn/h7hPX5jtJ77y1a+/8knEtou+9wyvvXbqS/89U9fA+FwQpvy10cPe1dTSskpm19h00A8S+5Dc1ZyRU7xUZ2nxfiy7w0/Hfvji1dXmoIekaTk2Shf6cbuTnYQ0sIGla8F6G3SsLkEZSvdZJdbCScsjj+Cus7ijS9SGw4AZsz/msVnsjwlM6GfFjboadCwuwWpBbYjsg5pgxstNoe1eWJhctRi9oQQlwI/BgqBNqAU2AXMf7+TtLCwmJwI90zU8jtjx0bqqci+N+PnU0/DaL4PHIUoxV9HLbopds628LljOteJZPGpJaiqQNfjm2ZLT582gTM6ipTPgN07E47HskDZ6u9NEHoAf2ytscTeFGP6Kg++7Aj+Lp2Bdo1gj+mm3FUTReqSitO91LwbpKcxiiddpWylm9a9EXqbTJOgFpJUvRkgvdCGzfnBimmVhoF84hHk269DZhbK1dchpldM9LQsjhK6lLhVldeWnMlvm6ro1iJcn1fK8pRM2iNhsuwOFCEI9urseHYALWw+JzJL7cw6y3vYryOlpObdIG17IgDkzXFQeqJ7SroTWkwMh5Og5XbgJODfUsqlQoizgOvGd1oWFhaTCXXevzAa7kEGdiPSzsSovR2MfgCMgfUIZxFK9gevCHJheQbf/evl/P62NQz0hrn8P0/gQ9dOjRqDh0L53BcxfvVTM4PttDKU//zSmK5Pt9kRwFDfkSzbkVt39KhEsWEtcI4xiiLIn+tESsm7f0qsk9fbpFG7IUj7PnMR2hvU2POqH7s78TMydAj1GfhyPmBi79/PI//1N/OgqQGjpgrlF79HOJ0TOzGL90VLJMSn9qznlZ425npSuG/mcr5XZto/KoP9LN/0b7b5e5nm9PDH2Sso3OmJCT2Artoo/k7tsLM5dtdrtO6OxF9/V4S0Qvv7SpQ0HshoF0bbg2AEUXKvRTitjb3JwuF806JSyk4hhCKEUKSUrwohfjbuM7OwsJg0CHsmavldABgdj8eE3gGMzqc+kGIPYPVH57P6o5PP0UGGGzE6n0I4ChBZFyHE2NJEi/xC1NvvRUYiCMfYRVqpy8uNhTP4dZOZrCDL5uCbJXPGPE7Eb1D5mp/+Nh1nisKMVSMkNLAYd4QQeDJVAl1Dak1lqfQ1Jwb1hXoNMkqc9LfE+9lcYkrUpZLhelBTELb0ozPelk2JDQP9ZrboeQuOyvgWE8N/V23hlZ42AHYF+rlu97vsXnE+ihB8df8WtvnNTZG6cIDP7F3P09HTk8YYIdQvhr9To3Z9iFC/TlaZA3UETRfo1ieV2JP6ANrmUyBUBYDR8GNsS9cho+0YVd9EhqpRsq9AKf9hrG6vxbHjcJ6YPUIIH/A68KAQog3wj++0LCwsJivCPeuw2saC1HWIRBDu4zeZy7FEDmxG23p2LLOqyLwQ2/zHj2isIxF6B/jpjCV8Iq+M2pCfs9JzSbWNfXFSsz5If5spHML9Bvte97P0ylQr6ccEMGOVm8rXA4R6DdzpCtNP8dC4OUSoL56B1u4RFC9xgoTOmghOn0LpCjeKOnk/L6n1oe/6KLLnFRBOlNJbUEv+36Cl4iFAQ8m5BuHIG9O4ongacvOQeq2qCgWFR3fyFsec4aWZasMB2qJh8h0utg0kWr9rQgF8M1Q6q6MxNwd3mkJHdZja9UFS820UL3Gh2s37w9Alu//tJxo0OzfvCJM7K/k3OK1wcm14yc4nY0IPAK0bveUBZMsDEG0FwGj6JdjSUUv/Z4Jm+cHlcL4tlwEh4KvAx4A04PvjOSkLC4vJi/DOQ5l2C0b93SCjiLSzUAq/eMTjGW+/gfzL/dDfD0uXo9z4FYTHcxRn/MFDb/xlQgkN2fUscmAzwrfkmM9liS+dJb7Ds5RIKanbEKKtMoLNKZh2ggt/Z2KZlEhAEg1JHJ7JKx6mEs2RII+0N5Ki2rjqEAWhvVk2llyeihY2YvF3pSvcRAKGaXn1KUxf5QYpcPoUssodZJXZ8WUffKmhaQYvPbSdqh1tnHzBTJadUXpU3+OhMBp/YQo9ABnGqLkVkX4u+q5rIFxj9qn/MbZl6xCOw88oKy79CLJ6P+zYCm4P4tpPIDIyD32hxaTmtLRsatviOQpnun3Uhfxs8/dyTkYuDw45d2JKJgVFLlIutNFRFcHuVuhridK2xzTt+Tt1tLBkxqnmMy/Yo8eE3gGCPToVp3to2h5GCCiY7zzkPXXMGcFzRES7kINC7wCy5xWwxN4xZ9RvixDiV8DfpJRvDWn+0/hPycLCYjIjtR6Uoi+bAk/vR7jKjnys/j7kfb+E6KBPy6b1yCcfQVzz8aMz2Q8qMpLcZESY7PKobW+E5h1mBlA9Itn3WoDMMntC0WFXmpIUE3Yo6is7ee2JPeSXpHHmFXOw2Se/S+GxoCo4wKrNr9Klmd+XXzXt460lZ2NXRo+tk4bE0EA6pFl3yqsw/8KUhJjKHc/2x6yxzTvCzD3XS1rh6FbdOz/zJM//dRsAf733Hf7nj5dy4fWLj+I7PTgyuGd4C7L1zzGhB0C0BaPtQdTirx/2uMLrRb35e8jeHnC7EQ4rVu944N7piwkYOi91t7LAm4ZHUTltyxoAypwers4p5p2+Lhb70vjJdHODzZdjw5djQxqS+k2JtSq7aiPMONWDFpE43AqKDYwh3tGeTJXs6Q6yp09e90eReQl45kNgh9lgz0cUfQnaH0zYeJyIDUeLg1v29gI/EkIUAA8DD0kp3zs207KwsJhsSGlg7PsSRssDIARK/g0oM346xjEkcs2/kVs2IYqnwcw5caF3oE9t9dGc9kEJ9urUvBsk2KOTXmSn9ER3zJ1mKqMU3Ije8ShI828rUk9GpKw45vPQpMHjHU3Uhv1cklnILM/By7f2tybGf0kJaUU2hAK9jRruDJXylWPLQrflzTq+dO5fiUZM8XHaJbO45/Grx/5mjkP+0FIdE3oAW/y9PN/dwiVZI7sa9rdpVL7mJ+KXuFIUZp7lxZupIqUk4jdweBVCfVpM6AEgoWV3ZFSx19sZ4MW/bU9oe/iX646p2FMyL0Jv/0e8QU2DEV3TjyzBjEg7OjGAFpODTLuDh+aeBMBbvR2cvfW12LmacIBrXdP485yVI14rFIEzRUncwEpVqXzNT2dNFEWFjGl2eps0tJAktcB085zsCNWNbcmbyI5HkUYIJfsKhD0LZv0Bfd9/QbQFkXEeyjTLqjcRHKyo+s+BnwshSoFrgAeEEG7gIUzht/cYzdHCwuIYo9f/yPSvF3bUkm+jFHwG2fkERsvvzQ4SjOZfIzLPQ2ReOOIYRv96COxCpJ8dy8oln3oU+Y+/mv9fvxYWLAKvD/xDdv4WLBrfNzeEva/6Y6nk2yojoMD0k6e+C6mSdgpi6bsY7f8CZyFK7scmJIvlNbvW8lRnMwDfrdnJcwtP49S07FH7+3JtdFQNEf8CogGDgXYd1SHILrfjTh+bVe7vP3s3JvQA3nhqLzW72imbmzO2N3McMlKVXeMgtXer3goQ8ZvnQ/0G1e8EmLHKw+6X/YT7DVQ7Iy5MlYNsK6s2BUUVGEb8de3OY+uipuReg9S6MFr+hHDkoJTehnDPNn8DQ2aCIRxFKLkfw+hbh9H4M5AaSuEXUdLPOKZztZhcdETDSW1tI7QNZfrJbipfC6CFpVl3L98W82gwNOisirLwEh9OnzKlypUI1YvIS/TKUbIvR2RdCnoAYTv4Zp/F+HHIb5GUslZKebeUcinwH8CHMevsWVhYHIcYXc9j1NwMkWYI16HvuxE58B4ysDOpr/TvSG6TOnrNreibV6Hv/Sza+rkY3S+b595Yk9h5+1bE578MM2dDZhYsX4ns7kKuf2c83loCkYARE3oH6GvSRuk99RDeBahl30UtuAGhHn5Np6PF7kBfTOgBRKTBLxorD3pN3iwHeXMcCNVM9FG0yEn9pjChPoNQn0HV20H628b2GckRxMtB9MwHik/nl5M2JNXfPE8qF2QWjNhXGpJgb+L9EuzRqd0QjFkp9Cg0bg2TWRofU7FB4fzR3Rd9aS6u+tKJsWPVpvCJb606ovdzpEgjihzYDIFtyP6NMLAFYUvFtnQtSultKCU3Y1u2AYwg+rZzkR3/QnY+jr79AqR/2zGdq8XkYnVGHoWO+AaHiuC6vIPHnKYV2ll2VSqLLkth6VWp6NHkH6RgjzGlhN7BEEK1hN4EczhF1W3ABZjWvXOANcB3x3VWFhYWE4bsfSOpzeh9E5HxIaj9PnF7gIKScW68T9vf0Ku+BdEOYMiiUIYx6n6AknEOpKdDU0P8nMuFmL8IZdkKjMf/ifynWZPKeO4pxGUfQfno+JX0tLsEdo8gGog/aKdCevipwkh66lAaSyiC8pM8lJ9kWlcbt4aS+vQ2a6TkHr7l56NfXslbz1SiRc3v5CkXVFA+74Nr1euKRkiz2VGFoMLtY+MJq3m4vZ5U1c41OSU4RonXE4ogrdAWK5YOkFZkTyjFAKCFJaUnusid6SDsN8gosePwHHzR+uV7z+XUi2ayf0c7K8+dzrRZWe//jY4Bo+X3yNb/Mw+0TvR9X4S0MzBqb0V2PAKAHNiISD8HjGD8QqlhdDyG6j0+amtajB2vamPN4jP5ZeM+erQon8gv5eTUQ39/FZvAk2E+b9IKbbTtjbtTCwWrvIzFUeVgCVrOxbTkXQisA/4O3CCltMouWFgcz6QsT2qSfW9DpAVlxk8wWv8MUiK88zE6/oUi7GBLR9/7WZAjW11kVwDj5RcQK1cha6ogEAChIC68DIIBcDqRLzydeM0LzyCvGj/3Q6EIKk7zsP9N0zXNl6NSeqJV+uFoMdeTygWZ+TzX1QKAXQhuKqwY0xjeEcS3N2tsgnzZGaX8ZfN/suax3RRMS+Psq+aN6frjhYZwgGt3vcu7/V0UOdz8dtYyzsvI5x9t9dzXXEWqzU6ew8Wlo8TrAcw4zUPt+iAD7TqpeTZKV7ho2hamaXvcbc2breL0mv/GwrIzy1h2ZtmRvr33hezfOLwFo+X+mNADkN0vINzJ31/hnDbOs7OY7JS6vPxoxpHHmGaVOYisMGjdE0F1CEqWuHB4jw+rnsXkQIzk4gIghHgF+BvwiJSy+5jO6iAsX75cbtiwYaKnYWFx3GK0/R19z/Ujn3QUoC7dhL7tbAgMenMLJ0r5HRhV/z3yNfXZKM+fDPqgtW/1BYhFS5DPPAF7dpqi77wLzBi+rs74db4U1N/9+ei9sVGQUqJH5HHjMjOZiBgGj3Y0UBsKcGl2IXM9qWMeo25jkJadppjIn+tk2nJLkB8J1+16l392xK3quXYnP5q+iOv3rI+12YRg5/IPUeo6fLdfw5A0bgnRXa/hyVCYdoL7kJa8yYbR+mdzs+oAwo4o+Ray7vaEfqLkZghsN2uKASLtLNQFTyCUyZ9Aw8LC4vhDCLFRSpm8Qz+MgyVoOfvoTsnCwmIqIAPbRz8ZacZo+FFc6AHIMLJ/CwgnyPgOv8i+CmHPhpc10Jvi/V99EZmZaQo9AGkgX3gGcfHlyKcfi19/+UeP1ls6KEIIbM6pn4FzMuJQFK7JfX+Wj2knuGNJPyZzYe7JzlZ/YrHntmiYZ7qaE9o0KXmjtyNJ7B3YFB7Jyq4FJX3NGoEuHT0iCfbqU07sidyPo4SqMVr+aBZ9Lrsd4ZmPVn9P/DdN2FBzrkB4v2vGKksN4Tu0NUe2NCHXvQMZWYiTViHso5egsLCwsBgPLKdgCwuLBET6eVB/z+jnbRlJsVfCnoky9yH06m9DtB0l7xOQvhqCe0EbVsPKMKC1JXngomKUO3+C3LsLUTELMX1sLn8WcaTUMWpuwWh7COEoQCm/a0pnDbRE3vvnnIxc9gT7Y8ez3D5OS83mH+0NCf2W+BLLBPykYS931+9Gk5IvFVbw3bL5Cedr1wdjpRbCAwb73giw7MpUhDJ1PjMhBErBDWDLQCheRPpZCNWHuvAFjMafgoyiFH4J4TUzBQvv/EOMaCIr92Dc+T+x8jLyzTWo3/7ueL0NCwsLixGxxJ6FhUUCSvrpMOt+9MZfAAIiLRBtBUCknYEo/hqi52Vk76vmBbZslMIbQZgJW6TWjQzVIndcZJ6fXoJSHS+nIE48GXHK6RivvRx/UacTsWgpIj0DUVZ+bN7ocYzR9L8YDT8GQEaa0HdegTixGmEbuxulxfHBnWUL0KTkua5m5npSuXf6IircPjYM9PDXtlo8isrl2UV8o2oLTqHy9eJZ2BWFb1fHs03+sH43y1MyuHhIXJ9/WIKWaEASCUqc3qkj9mSoGu29k0HrMhuafolt6bsI3xKU7CtMK17KiQcfZKRxX3gmsY7o9i3ImipE2fSjM3GLSU9DOIBHsZFpn7wF0S2OfyyxZ2FhkYSSdz1Knhm3Z4SbkfV3gZqCKPk2imJDLHwGo+NJiLai5HwUFCfahvlmuYbhzKxHZpej9H0MiooRp52JsNkRN30N+e/nweVGuexKRHrGMX6XcXqaovQ2aXgzVbLK7RNSk+5oInteTWzQ+5H9GxAZB/fOlwP95gK1sx2xchVi8bJxnKXFscSj2vhlxVJgaUL772adwE9nLGaLv4ezt7wWy6P7ak8bXymemTTOuv6uBLGXWmAj1BfPJOhKU3B4ptb9Y7Q8EBd6AIGdGB2PYtTfFXdZd92Bbck7CHvmxEzSYlKwy9/Hk51NlLo8fCS7GPso2WuDus61u9fybFcLdiH4StEs7ihfcIxna2FhYok9CwuLUZGRZvTNJ0NkMOau+wXEkjcxmn+PUXsb6API3tcR2ZePLPQOUOJHWfqphCbl5NPg5NOOeG5+XeMbVVt4oauVOZ4UfjJjMXOOIAFIy+4wNWvj6dT7Wx2UT/HC6sK3DNn17JAGxyFdz6SUGD/8LtRUmcevvYLylW8iVpw8jjO1OJZEDYM1ve14FJVTUrNimxoe1cazXS1DC6YQlgZRQ6JKwcV9hcwKp7DR3c2padkJY5YudyN16GmM4klXKVvpTtosCXTrdFRFsLsEORWOqZEMqW9tYmxyqBqj7a+oRV8+7CHE+RcjN7wL0UExvGAxllVv6vJaTzsXbX+D6GAM6z/a63ls/sg1Ie9vqeLZwUzEUSm5t2EPl2cXcUJKBr1NUZp3hhEC8uc5SSuw4jgtxhdL7FlYWCQhtT5k5+MYnc/GhR6AfwtG8/0YVV/nQNU02fEI2A9et0wU3nTU5/id6u080FIDQGMkyFU732HrCecd0iqnRyU9jVFUhyCtwBbL9HiAtsoI05a7Ue1TyzoxFKX460j/DmTnE2DPQp1+D8KRd/CLaqpiQu8AxpqXUS2xd1zQFY1w1pY17B6M2zs3I48n5q9CHbxfpo+QgfOsjBzO2JNLeodZFP3i/iKKqp0waITXwgaNW8NEAgaFC5zkz3UmxeoNdGjseHYAOagk2/dFWHhJyqSK6VPyP43R/Pu4dc8zD+mZndzRCCe3HQRRMQvl7p8h162FzCzESutemsr8sqkyJvQAnu1qYXegD11K1vZ1cVJqJvO9aQDsCfQnXb8n2M9cLZXdL/k5MExPo8aiy1Jwp1k1Xi3GD0vsWVhYJCAj7WibT4Fw7cjnQ1UklcfW+xDZH4nXpRJOaHMiavKQaQEoHnmsg86joR65aR0iLx+Wn4RQEx+GL/e0JhzvDQ5QHw4yzTW6VS7sN9jxTD+RwULqaYU2lGHPWCHMf1MZoXqxzfsHUg+A4kSIw1hIeH3mGx+ymBG+lHGcpcWx5IGW6pjQA3ipu5UXu1tQENzfUo1XUTk9LZvXezsQwPV5pZzty2VDa1/COM07w5QsM8tfVL4WiBVZ723S0CKSkqWJpTHa9kRiQg8g0G3Q16KRVjh5rBnCVY5t2SaM9ocRqg+RczXIiJmN88Bmly0LJffasY+dV4C45PKjPGOLiUCQ/GB4tKOR79XujB3/umIZnyko56KsAn7fUh1rdysqZ6fn0r0rOvQnFmlAV10UQ4vQdqDO3jIXWWVWjJ/F0cMSexYWFgkYbX8ZVejhKkcp/BJ6y/1gxF0fhWsmMrgX0lcjUk9CvnEf4vnlCGk+HmXdq/C9mw97DnLbZox77wBdRwLipFMRX/p6Qp8lvnT2Bgdix3l2JwWOg9e7at0djgk9MBeoRYudBHrCMf1asMCJYpviag+QkTbTldNRABnnIsTBXedEbh5i9fnIl54zG1JSkB4v+l3fM7OjXnoFwuEc8zyCuk6PHqHAMfb6eMEenY7qCDaHIGemE5tj6n4uUutBdr8MrjKUlBOO+et3RJOtUu/0dXJP/Z7Y1k2Gzc7aJWeT7XBS4vQQCRlJ1xiD+ViiISMm9A7QWRUle7qDxi0hIgFJ9nQ7WmSEWr6T8GMUzkLU4q8ktNmWrsVo/bOZjTPveoSzaIJmZzEZ+HJRBc93tRAZ3L24NKuQ3zcnekPcXreTzxSUk25z8Km8MjYP9JBpd/DtaXPId7joSIkkjRsJGLTuMtujIUnlawG8WSquFMvaZ3F0sMSehYVFIiO4KonMSxFpqxCppyAUFXXeoxi1tyGjXSjZH8ZouBeMAGAmBxFbTaEXu36fD9neZh50dcCMWQib+fMjDQP5xCPItW9Cdg7K1ddhPPsk6PEsf3Ltm8hrPo7IyY213VW+iPpwkHf6OilxuvndzBNGDZY/gD7CwtObZWPxZQ56mzU8mSqpeVP/Z1H6t6NtOQt0s7aayPowtnkPH/I65ZM3IM84BzraMTathxefMcfbthnaWhFf+MohRkjkz601fH3/Fvp0jVNTs/nHvJPIth+eYPR3amx/dgA5+DVoq4yw6NLJ5f53uEj/NrStq0HrNo8LvoBa8bNjOodrc6fxq6b9sYVqjt1JWySUYKPv1qLsCfazNMX007Q7BapToIfjvTwZ5j2m2gWqQyTcU3aPYOcLA0QHN1T6WjS8Wcn3pEzWkJMS4chHLflmUrs0QmaGTtU3AbOymChOS8th47LVPNXVRJnTy2XZhRS+83RCH7+u8bX9m/lV034AvIrKcxWnsTI1C4DMMjsZ1Ta6682NksxSO4Y27Lkkob9Nt8SexVFjCkRJW1hYHEtE2lkkbr2riLLbkV3Po285DW1dBbLjMWxL3sS+Yic4imJCz0QHJXE1J5EYLzyD8dUbMb7/HYyvfwHZZgavyxeeQf7rb9BQB5s3Ytz9/QRXwtEocrpZs/hMWk++hL0rLuCcjEPEpAE5Mx0MNXA5fYJAl0bDlhCqQxwXQg9Ab/x5TOgByM7Hkf6th3WtKJ+BWHESrF+b0C7Xvhkrrn04dEbD3FT5Hn26uah5s6+DH9TtOsRVcVr3RmJCDyDYY9DbrI1+wSRGr787JvQAjObfmOVJpIHUeo7JHBb50nll8Rl8Or+MmworeH3xmVS4k910i50eGsIBvlm1lU/v3UD/ijC2QYO56oBgr8Gmh3vprIlSusIVu59sTkFWuT0m9A6ghZO/M5PVQiuN6CG/43rtHWjv5KO9k4tWeSNS6gftb3F8UeL0UOLwoElJxDC4sXBGwvmP55Xym0GhB+A3dO6u34NhSFp2hal+O0hmmYPFl/tYckUKs87y4stJfu74si2hZ3H0OD5WNhYWFkcNs37e0AWPjqy7I15XD4nRch8i9xqUtFMRzsLkMU7ogsZshGGuBMU8D7zwdHxLv6Md+fg/ETd8Cbl5Y+LFPd2Iiz+M3LktZt0TK09JsOoNJd12+LENvmwbCy7y0b7PjI3ob9No2GxaMjuro0SDBoULDu4KOiXQA0lNUg+OzXsuMwsah4yTkTmmkhT7g37Cw0w4O/x9Sf2kIRno0LG7RcJO9khG2pE8UYM9OoYB3sxJvDjSeoc1SIyeNRh1d0K4BuE7AXXu3xCu8a0xuSIlkxUpmTSGgzRHgnwmv5xHOxrZOGAK0etyp7EiJYOFG16kNmx+9kY33BgySzDogx5okYBk/5sBllyRwrKrUgn2GdjdAjOdZyjhNVPz7fQ0RokGzd+UzFL7iIvbiUQaIfS9NyDb/wn2HNTp96DkXoP0b0Vv/BVIzawlKnWMuu/Hr2v5AzL1FETexydw9hbHin4tyqmbX43Fvs50+3hz8Vks9KbxSHsDZS4PHxu0oA9lQNeofjtI+z7zBmrfF6F49T9grAAAIABJREFUqYvixeazJrfCQaBTp60ygmo3Y/aGJ2wJ9et010VxeBUyp9mnpIeDxcQxuX5xLSwsJh5lBLEzxCpxABncj6H1YIRqEGlnInvXACBST0MWLEO67kfW5kJ6GGXBXbDzH4nXd3aa/YuKkdu3xE/YbIhVZyAWLkFuXA95+aal6SjhzbLhzbIRDRk0bkl0WW3fFzkuxJ5ScAN6x6PAoFj2LR9zUWjl45/G+NndEAqB04ly/WfHdP0SXzr5dhct0fji//zM/IQ+kYDBzucHCPWZorBgvpPSFWZsX95cJx1V0ZhlyJejEujWiQYkmaV2hGImCOmqNYtWp+bbmLPaOynjLZX8T6F3vxBv8K3AqLsjFhsrBzai7/8atvmPjftc7qjdyZ11uzCAWW4fzy44ldZomBTVzmxPCi91t8aEnk0KPtU9SqkACQPtOr4clZq1AQLdBjaXad3rqjGTUHgyFArmOehvixIdDPG1uyff52M0/AzZ/nfzINqCvvcz4JmNvvUc0M24YL39YZSRXDr925LaLI5P/tnRkJDkqDI4wD/a63mxu5Wnu8zSQ490NHJqahZv9nXG+n0ut5yO9xJj9dr2hskstVOzNkigRye9yMayj6ZicwoCXTpN20N4MlXSC+30t2nsfD6e0TZjmp3ZZydnz7WwGA1L7FlYWCSg5H0co+nX8SQtnvmIoq8ge14mZvFTvBhdz0Dn4+axcKBU/A7Ftxg8s9DWFkF6ENLNbGRS+ROicAY0NcZeR1TMxLj/10ihQPkMqN4PLjfiuk8hUtMgNQ1RPG383qdNoNjAGOIZaHNOvoXokaCkn4FY8hZG+8PgLETJ++SYC8WLhUtQfvkHqK2GaaUI79jikxyKwlMLVvGdmu3UhgJclVPMl4sSi3Q3bQ/HhB5A844wubMcuNNU3Kkqiy9PoasuitQldZtCDKwzxat3l0rRImdM6IEZH9a+P0Le7LEnkRlvlOwrYMHTGO2PIFyliJxr0DfMSegjB7aMcvXRoy4UiAk9MDPY3tuwl3umL+LRjkZe6m5l9hDXTtUQOOUo0R7CFOC160MEus0RtZCkr1lj8ZUp6GFT7NWuCxLqi3sKtO6OkDvLOakssXJgw7CGKEbzAzGhZ7aFkVo3oHJgEwVApJ8z8phNjcjtWxDF0xDzrGLaxwNBPdlld2egLyb0AGrDAa7MLuay7CL2Bwe4LKuIs9Jy2GDvS4hvtTkEe1/xx37/OvZHUVRBar6Nfa/HPSoKFjgJ9xsJca7ddVGCPTru9MlzD1lMbsZV7Akhzgd+jvnreL+U8q5h578GfBbQgHbg01LKsedot7CwOGoIeza2ZZvMOK/BJ4zQe1Hm/AXZ8kdQvSi5H0ffdVX8IhlBdj2BKPgUMtKckKkTQGptqN/+K/KpR5HtbYgZM5GP/RMG47lIS0fc/QtETi7CeXQX68FeHS0s8WWrCa4vqk1QstRF7XrT8qTYiKWUH0okaDDQpuHJUHGlJj9c1zy6i/tuW4O/L8Ll/7mMT9585IXijyYiZRlqyrL3N4bHA3MPXoz9ALqUPNPVTHXIz0WZBVS4fSzypfPUglMT+q3r6+Ku+t0M6Brf6p6Lh8QU/BG/EXNhsrsU8mY5qV0fxIjrOvwdelImyAPXTlaUjPNQMs6LHeveJeDfHDsWGWeP+xwawgGG/4XqQgHO2/o67/abNebSVDsfzirk8c4mFCFGTDdv9wjyZjtxehWCPYkL4GhIoiBwpguQjBhn2d+qTSqxJ9JOR3Y+GW9QPIiUE5Atif0UzzyY+zf0uh+AEUQp/CJK5oeSxpMb15lWccMwswlffDnKf1w/vm/CYty5KqeYH9bvpn0ws22WzcFSb3pSvz49yg+KFia0lSxzUbPWfC4KxSymXvVW4nOyr0VjoD3xfmnZGSatKHmpPobwaQuL8RN7wizs9CvgXKABWC+EeFJKuXNIt/eA5VLKgBDiRuAe4OrxmpOFhcXhIWwpkHkR+uZVEBqMP/AswLbkTYTqQQb3JV8kNeTAZozm+8BRApVhRLVZZ0+56FpITUVccAkiJw/54B/jQg+gtwfqahDFJUf1fVSvDdC623SfcacpzD3fh8Mdt1QUzHeRUWIn0GOQmqcmFVLvaYyy5xV/LFFI2Ulu8ufExWhzTQ+3/Mej6Jq5hP7d/6yheEYmq68+PIF0PPHJPet4uL0BgFuqt/PMglM5PT0noU9rJMT5217HP5i/3zDgDhbFztvdgq7aCO37I+TOdJKafyBj60hZVNUEy6wQZjzYVME29yH0/V9D+rcgMlajTv/xuL/mipRMSp2emJsmwDxPKj9u3Bs77tWjFDndrFt6Dm3RMEojGEM80BxegRaWNLwXor0yQkqummCd9WQotFVGaNpubqKMZC2fbK6cSuEXINyA0fYgOPJRy+8yXdPb/47seQUAkbISkXstQvWgZB+8bp7xxL/ML/cg8vmnkJddaW6eWExZch0u1i49mz+11CKRXJ9XRrbdwR31u6gPm8LNJgTX55UlXZs/x4k3U2GgQ0dRISVXxe4RCQmNvFmmq/pw8mY56W3UYta9tCIbnozJs1liMfkZT8veicA+KWUVgBDi78BlQEzsSSlfHdJ/LXDdOM7HwsJiDBhtD8aFHkBgu1k0Pfc6EDZE5oVmHTcAVETW5WhbzjCtenuLUF5ZAQzm9Wyqx+j9LPT1QV4BYkmyxeloF/D2d+kxoQdmFsGWnWGmnZBovXOlqkTDkh3PDRDsMUgrtDHjVA8Oj0L9plBCRsj6TSFyZzlQBi2Em9+oiwm9A2x4teYDJ/aqggMxoQcQkQY/a6wk1Wbn5upt1IYDXJVdzAy3Lyb0AN7wtfNGRiuX+ouwOQXdDVFa95gmvI6qKPPP95GSZyNvtpO2ykhM2LnTFbJnOPBkqDTvCGPokD/XgTdrCkUmqCngLERoHQhH4cixskcZu6LwwqLTuatuNw3hANfkTiPf4UoQewCGhMW+dPSoZH0kMblMxB9fnIYHDNxpgvx5TnoaongyVDKm2dj/RtxiERmWVl51QHrR5BLlQthQp9+DOv2ehHbbwucx+taBjJplZw7XFVobZs3UjalTb8LioBQ7PXyndG5C25rFZ/Kbpv10a1E+kVfKiamZCecjAYPKNX7624Y8TAQULnDSWRMl3G+QkqeSN8eBv0OPeZsA5M5ykFFiZ+ElKXTVRnH6FLLKJ9f9YzH5Gc8nYxFQP+S4AVh5kP6fAZ4bx/lYWFiMBWOEjI7hevRNSyGwE9QMRMEXEfYMlKxLMTqfjrlvil3DLHT1Q7yzW5uRdbVQUAjNTWbboqWwYBHS0GH7NoiEYdFShOPwM20OJxJIXlxF/IaZvVGXMWEgpWTfawHCA2b/3iaN2vVBZp7hJTqsqLQelab4GzQOzjmhIOk15ixLbjve0Um2vEWkwaXb36R10OXph/W7+XxBcrIPR4nC7CKfadHbN8RXU0L7/gi6LmneEcabpeJMUfBl28iebgpuX7aNmWdMIYE3BH33tcje1wGQ/etB96POGH/rXrnLy+9mnUBHNMzD7fX0B6Is9KSyLWBmSvUqKjcMfk6KalrmRiqfcICwXzLzTBfpRTbcaSod1clFozNL7ehRid0lKFzkSrKgT2aU1LElNgIQH7oYed8v48ennTnmmNehbPf38lZvByelZrHYl+w2aDGxFDs93Fm+cNTz9ZtCiUIPzN+3ygjLrk4lGjKfQTuf84Mwa/G50xS8mSoZ00xhp4Ul/e0avc0SxQZZZUf+bLT44DEpnpJCiOuA5cAZo5y/AbgBYNq08UvYYGFhYSIDe5ChOtPaYAzuMtrzzV3uwKBxXu9Gtj+EurIWoTihb0hdNmc0edChtDYjvnUb8vF/gceLctV/gGFg/OA22DM4fl4ByvfuQqSkjjpMrxblrb4O5rhTmO5OXEyl5duwu0Us5TtAaMBgy+NmNrWUPJU55/rQIzIm9A4w0G4+mB1uhYg//pC2u0TCQrV8Xg5f+/mH+P1trxH0R7jok0u4+NNLDv7ej0NmulO4KLOAZwYTFagIPpSex0vdrQn9dgf6+WLhDH7TtB8DODs9l8/mm+UG7K6Rai3Anpf8sfiUgTadwvmuSVun7XCR0e6Y0DuA0fnEMRF7AO2RMCvfe5nGiLk5U+Jwc+/0RYQMnatySih3mZn+hCIoPdFN1VsBpAGqHexuJcFt05etsunhXvQoICB/bnLMbf7cuEvuBwHljLORObkYWzZCawty7270H9yKcvV1iBmzxjTWAy3V3Fi5KXb88xlL+Pyw2m4WE0NDOEBV0M+JqZm4lNHdKv2dI9cH1QYTtnTsj9DXMthHQldNlAUX+/Blm/dM2G+w66WBmJdJf2sA+/nKB+qesnh/jOc3pREYur1fPNiWgBBiNfAd4AwpZXj4eQAp5X3AfQDLly+3wlItLMYRGaxC23wK6IMpphUPSuFNKIVfQNt2fmJnrQsZqkb2vYPUA+BZYLp7LtsHjVmIqLkrKX0GYmDIYn72XOSt3zTT+gPGts2Ij34sLvTAtACueRlxycjxMe/2dXLx9jfp0zUEcFf5Qr5SHF9IKTbB/At8NG0PEw1JPBlKQqmF/ladjn0Rcmc7cKUmLmBT8swHd9g/LPlEUKJHZYLgu+qmE7nixuXomoHDObkevjLaBarXFOPjRNQwMJD8fe5JPNxeT3XIz6VZhZQ4PdxSu53QkNilBd40fjxjMf9dMhu/rlPh9tEUDrKmp50TMzLJLDPT9gO4UhRsDpGQiEBK6K6PTv14FVsK2PMgGhfDwl3xvoaUvT1mAqTWFsTylShnjJwlEuBv7XUxoQdQHwniUVS+WFjBzkAfvVqUNJt57+bMcJBWaCPYo+PLtqFHzXi9QI9ORomdnqaoKfQApJlSvvREFy07I0gpKVzgIjXfhqFLhMKYs8JOVcS8BYiGOuTTgxmLW5ow7rkd5Rf3jykJ1R21uxKP63ZZYm8S8PPGSr5dtQ0dSb7dxbMLT2W+N23EviPVBwXImelACJHw7DlAqM/Al216nnTXRRLCCQA6qyOW2LM4bMbzm7IemCmEKMcUedcA1w7tIIRYCvwOOF9K2TaOc7GwsDhMjPaH4kIPwAhgtP0NnEUoWRdiNOyOn/MsQN91TdzaZ8tGlN6O5H+QH3sV2ZANaX7wgbLvO8jaarN+HsSEHgCtzbB7R/JkAv5R53lb7Q76BpO8SOB7tTv5bMF0fGr8Z82VqjL9FDMpQuvu5L2ksN9goEMnq9xOT0OUYK9BepGdvLlOWnaHsbsUosH4g9juFoy0gauqCqo6yhN9ApBaH/ru65Ddz4MtHbX8LpT8Tx/117m7bjf3NOwhahh8tqCcH01fjDJkMf/rimV8rWoLPVqUM9JyuHmaWW6gwOHGkJI/Nldz0/730KTEKRQeWngSZyzIQYtIUvNtdNclW4hdKZPn73ykCGFDrfhfs56b3gfOEtTyuxP6yEgrRsfjCFsaIvvDiEPE9Bn33A41Vea1m9ZjRKMoq88/6DVDaY6EmLv+eWrDATyKyv9WLOVjeaWAaeE+kNhItQumr4onGumoSnTbNDTILndQMM+cr6FL9r3hp7MqiuoQTFvuInfm5CuPMR7ILZsSGwYGYH8ljKEUQ9BIXOWHjOQEHhbHlh4twq0122Pu6y3REN+r3cnD804esX80lGyjSMlXsTnNYumZpXba9sbvI8UGaYU22vdFqN0QRBvh+khwasSASq0Po/l3EKpGZF+OknHuRE/pA8m4PTWllBpwE/ACsAt4WEq5QwjxfSHEpYPd7gV8wD+FEJuFEE+OMpyFhcWxYqjQO0CkAWP/f4FnMUrxN8AzD5F9JUr+5+JCD0DrQEYHC7C7olDRDDl9kGJH+cJXUe/+Bcp1n0Y4R1i4zpwD6RnxY4cDceqInt2A6Yo2lIChM6CP7C4DkFFiRxm6vSUgGjLY8cwAjVvCBLoNZp7pJafCwc5nBsxit91G7BrVDuUnuRPKN0xWjIYfmUIPQOtB33eTWRLjKLK2r5Nba3cwoGuEpcGvmvYzf8ML/LGlOtZnvjeNe8oXsWHZOby46HSy7OYi/6XuVmaue5bP79uENGBhMI3ssJObq7fhy7aRXmhHUQSZ0+xklsWTEWSW2ZMybgb7dCpf97PjuX7a9yfHi01WlOzLsK2sxbZsE+oJ25F6PzJoJkSSwSq0jYsx9n8Jfc/16FtXI4dv7Q9BNtbHhF6s7e3XR+kN1+ZMo9ARvwdLnR429HfFsnQGDJ2v7t9CYMj91LIrzObH+tj2VD9dQ0R49vTE2KG0Qhv2IRlvW3eH6dhvFlnXwpKqt4NJbtOTGRlpQa/8AtqWs9Dr7z3o5zAcUTIs7ES1QWHRmF7/i8OseDdaVr0JpzMaSfBYAGgMB0fpPXJunv4WnaatEbY92Y8rVaHidA+p+TbSi23M+5APQ4f9bwVGFHoALt/U2PTSd1yKUfMdjJb70bdfhNH+r4me0geScbUBSymfBZ4d1nbrkP+vHs/Xt7CwGDvyIFnjZM9LqMVfNf1S7LmgJtelE4oDiR0YYpVxlif2Ofs85KsvmSUXAKZXIFaegpg9F/nv5yEagUXLkK/+GxkOIc5cjZiRWJD7urxSvlW9LXa8Oj2XfMfo1g+HV2He+b5Y9sbcmQ72vhq3HEoDGreEBv8G8esMHead78OblVyaYbIi/duGNWhI/y6E4+glj9k00J3UVhXy8/nKTeTYnWwa6OHOOtMFzaUoPDF/FWem5xI2dD65ex0dWoRszcnPmpZREjUtRU8PNJrR24MIRTDrTC+hfnOB7UpJNKsaumT7U/0xN8L+1gBSQm7F1EheIFQvUk1F37QEQtWAQCn+uvll1Lpi/WT/OmT3vxEj1HQDIDXNFBJDxJlIzxy5L5DjcLJu6Wr+0V6PKgRX55SweutrCX169Sgd0QjTVBvdDVFq3o0vZitf9bP4ihRcKSqFC53YXIKeBg1PhkLBfPMeHOjQ6NgfSa6zJ80YJqdv4j8jGe1Adj4F9mxE5gUIkbwk0ndcESu6LvveAiOMWnrLYY0vLvkIsroKtm8Bjwdx7ScRQze0DoNbSuexwJvGW30dnJSSxRXZYxOLxxNy4D3kwHuItNMQ7pmHvmCcmOH2sSIlg/X98d/Aq3NKiBoGdiVZhCl2YBQtqEehfV+EtEI7Eb9BqN/A0ELkVNgZIe8VYD5+c6aAdVz6dyL73k5oM1r+gJJz5QTN6IOL5fBrYWExjIMIGjUF7b2TQQ5aUDwLwD0LgoOp222ZiIxzkQ13JV5X1YL+79tBi6Kcd5EZU3TPL5Dr14LLjVi+EmGzQXYO4pqPIwN+jK9/EfrMtO/ytVdQvn8PoiwuGr9aPIscu5PnulqY50nly0XJMU/RkEHbngjRkEH2DEdC9kYtbCQVpjU0aRZsG4YnQ5kSQk9G2kEo5mfQ9Uz8hC0dcQRZBQ/GGWk5CEZejzzW0ZhQiiFkGNxZt4sz03OpDwfp0Mzvz9U902JCD+Di7iICPRqe9MRH03CRFx4waN4Ror9di8eLDdK0LTQpxZ4M1WI0/wb0EEr+pxC+xQAYDfcOCj0AidHwY0TuCAW4pTbsUEP++X7km2sgLR1x0irk22+YQjE9A/GReMnapnCQLLsD5xAf5ByHk2Knmz+21LC+v4tVqdnsGMzICXCCL4NpLvOz6W1M/CPLwWLprhQVLSzpqY/S06AR7lfIKLEjDdj53MCIhZ+FAim5E7/0kIG9aFtOj4lqkb4a28JnE/uE6mJC7wBGx6MJYk+GmyDSAr4lsHMHxhP/hFAIsfp8lNPPRv32d5F9vebv3BFmF/5wdhEf/gCLPAC94ScY1d8aPFJQ5/wFkXUJRvUtGF3PIDyzUcvvQXjGlgDnSHls3irubdjD3mA/Gaqdu+p3c3PNNj6eV8ovK5ZiGxKodyhjsKLAvtf8RAZr7vW1aKgOMxvuSF670jA3uiY9thQY/pSwjRzXOFUYHrM/VZj4X1wLC4tJhQzuHrFdZJyP1PviQg8gsB1l3qPQvw4Z3IdS+AUMdZhFod+NeHIB6Gb8irFzO8ptP4CSUnC6zGQFg7uhsmqfWZC4sSEm9ADQNYwXn0G94aaEoa/LK+W6wbii4RiGWTsv1GtaKlt3R5h3gS+20LQ5zXpFnVXxhWzeHCdCgaq34tuw2dPt2JyT22VGSh197w3ItgcBgcj/DKLkFmT73xGOApTyOxDqkad+H4n53jT+b/YKbqvZQU04sUxHmctLZJiFuG+w9liZ00OZy0NNKEBpJLnIdHtlhNIV8UdT6+4wTdtNl92CBU5yZzrY8Vx/Qr23oUzGcmYy2o22eRVEzdB0o/UBbEvWIrzzkOHhecskIv1MZOcjoA+YTZ55iGGxLvKFp5Evv2AetLUiu7sQd9yLCAagYjbCbqcpHOSqne+wYaCbTJuD/61YykdyigF4vquFq3fFM+hmqXa+UzKHV3rbmeNO4dbSeebrSEmwN3nF6R1MklO7Pkh3vfnZBroN9r7qJ63QliT0FBs4PAqlJ7oT3DwnCqPp14nW055/Y7Q/hpJjJoSSwX1IxWfWQxzi2i7c8Q0nvfYOjLo7AR20eSj/NzNWY0/ur0RmZCIWLkGkTu0F7kQjpYZR94MhLQZ63R0oAxsxmn5h9gntRwvuw3bCtmOSBCjH4eSe6YuoDPazcMOLMTnzQEsNi73pCUl0IoHRhZnDK0gvtlP/XmJYQrDbYNZZXuo2Bgn1GUmir+rtAIsuHT1T9WRAOEvMNUHTr8wGNRW15P9N7KSOkECPzr7X/AS6DTwZChWne6dUojBL7FlYWCQS6UpuU1IR3nmg9SVZcmTPa8imnwOgdzwKJd9O7FCXg9CH/ChKifHW67BpPXR2mOPNnov48jeRP7gNgsn1/QAzicsY6GvRYkJv8GVp2xtJsCrMONVDan6EYLdOepGd9GIzHszpU+hp1PCkq6QVqnTVRfFkKEkWpsmC7HgE2faX+HHLfajzn0Apu/UgVx3B60gJ+/aCqiKmV3BN7jSuyZ3GbTU7+HHDHqJScl5GHl8rnsVWfy9PdjbFrl2Zmsmsdc/REA5yTnouJQ4PnV3hJPemA+nIAfpbNarXxjvUrA1i6HJUoQeQN3sSWvW6nooJPQCMEEbbg6jld6LkXoPe9XT8nGs6Su7VKKknY7T/HWzpKLnXIZTE9yX3DtuUiUYRvT2IxctiTbfV7mDDoLttlxbh85UbOT8zH69q45GOhoTLO/Uoi33p3Fo2P6F98yN9hAeG/L0FFC1y4ssx76MDZUoOEPGPbB03NDPDYFtlhIziiS8KLaMtSW163e2I1JVoOy4D/xZQPIjsK5Ed/zRriDqnoZbdYV4fqsOouwMY/I3Z15dUTF1u2gC+FOTatyAjE3HmOQhXsuu7xSGQerwE0AH0AEbXi4ltwb2mldydXM9zvHivvyfpmZjk4j7Cz1VmuQ0janqTdDdEkzNC55vPnWCvMeIGVqB7Eu5qjYA646eI3GvNBC3pqxH20d3LJzNVbwVif/NAt8H+twIsvDhlgmd1+Fhiz8LCIhHVm9xm9GE0/ASRfSXYMkAzH2YifTWy7c9DO0Lbg2AIxIaZUJ0PjhFq7nW0Q2dH/HjPLnjuydGFHkD52FLTq7bkBacWNuhr1UjJVRFCoCiC3AoHWlgmWBu0sKSnIUrHvghaRJoPW2EmaMmbPfliJWQg2RorA7sh84Kj9xrhMMYPvwuVg6+1aCnK129G2Gx8r2w+/1U0E7+hUeI0rXV/mXMi9zdXsyvQxxlpOXy+ciP+we3pF3ta+X8ls7lpXkWCFRUgNS/+WOprTU64E+4/+CLHmz0JBblthELYwo5WeSMMbEFkXQpSQzinoRR/w4wdc09HnXbzqEOKWXOQG96NN9jtUJ6YvGOHvy/huE/XqAsHmOtJjX1OQyl1Jd770ZCeKPQAJDRtC6OFJGUr3aTk2gj1xa39Tp9C0UInPfXRhBqXB+iujdLXqiV8zhPCSPsFge1otd8zhR6A8f/ZO+8wO6r6/7/OzK27927vfTfJpveE0ELvVYoI2EERFUUR+MoXUX6CNEFFFEVFmqCg0ntNIyEhvffN9t5ubzPn98fcvXdn791gEJLod9/Pk+fJnDnn3Jmddt6f8v4EkN1Po85dj9A86JEetF3fQsYGETknkSB6ADm+1PmEQP/xjRAX85ArlqLcetf/mfITnxSEYkcp+aqh6hiHUnY10rcBGdic7GjJA3vZQT22o7LzsQhBbJgr+7jsQsK6xsu97QR0jfFkIUakRwT6Y4TiKeueDo38OitWpyA4oJNTbsGZpdK0ZgTBHY7/oFtIcc8H9/xDfRj/Fvy92n63D3eMkb0xjGEMZiije0bk4GLUmUvRG38CgKj8IXL9seZOmh9WT0CsTSbQSzWG0K2Ge232PCguSZ07Iw3JjMObm8vds6eweuNijssu5MbKiab8o3RIt57qb47R3+zDmWOItfh7Nfa8HyAakGTmq0w4IQMk7FocSF0MSmhaE6Jwgg3lMFPkFHlnQtPPSB60apCJbZcibGUoFdch7BX/1m/I9xcniR7AxnWwZhUsOBqAPKuNPIx7Z7N/kCt3rGa9f4CFWQWcmlOcIHpDWOXtY6Anlcw1rg7S8EGQgjobuZWpn6jcSqvhpd2RXnmza3uE7JJD7zkaDpF3FiL7eORgXATFOQF94D3wrjC2faCUfecji6rLjnaw2xG5eYjTz4aO9njOXi7K57+SEi54Rl4Ja4Z5GcpsDv7W2cSEDDdXldTySm876/0DCODq0nHMchmkdKWnl1+27MIaEnybianHoUPnjggZeSpV8xzEwrrhCc9VqT3KicOtUn9SJl07wvQ1RdFGXKrBtuihJ3vWNEIpjlpEcLf50ZdhhOYBWwn6huMTHiYTyQAoGUAeEUGscYKmwbwFyL6eBNEDDK/43t0w7tCJi/ynQhn3K4T5+X3yAAAgAElEQVR7flyg5QSUgvOR4Ra04C6kbw1Yi1An/O4jS5R80qi0Z/DUpAX8eN8WBmJRriyt5bMFFZywYRFrfQabe4mFZJH8rkokoQHzN8TTHmPu55LP7+4lo5cdAiibfvhFMPw3I6vEwmBb8nuVXfqfRZ/+s452DGMYw6eP6MDo+xzj0LZeAMFdAMi+VxEFn0F2P53oIoq/CNs2mIYJzQI33IxSVoEoKka2NKG/+yZE4qvA/EKU089GegeRr79irCbrJyEWngi6zjeKMni+zwg7WzLYQ2ckxM9qp/Pd3et4vb+DyRlZ/HrcLGa4kh6U4eGAIxEc0OnYFqZrZyThffD3ajStDpFTbhlVBU2LSiPZfkTKUcuePvyeMBNnf3JqlwcCxT0fJj2J3no/CBWRdSz63usA41T0vtewzNuUVm0QQDbsQX/yEejqRBxxNOLSL4BqQS59z/C6jq+Hvt7UcQP9aQ3MX96+is1xsY+lnh5siiBbtTI4TE1lYVYh4X2p1tFY3JjdtTOCI0uhfKad9i3xnL2pdnLKrThzVPSYJDig4e81e/oCA4efxVUIC+r0N5ADi4xwQNc8tFWVpj567wujkj0ZDqP/4k5D1VEoiFNOR/nKVciJk2HdGvAMILduhrnzEcOMIDdVTiKsa7zY24ZbtbLW189dLTsAuCC/nA9mn8T/a9zKC72tbA14WDTQRZ3Dxemblibqu31JqcOtpyfPvh7NuA6DOhm5KhWzHbgKLPQ0RNi9JI3BJI7cykNPxtXSq4h1PArE7xdhQ5n4JHiWIT1Lkx3tNZA5A9n9TGoo4UicYkO58hEjpDY3D/3396f+CT6mSMv/dQihIoq/BMVJ8SJhr8AyewUy0gGWfITy6dxXv2vbw2/bduNQVE7JKWKTf5Asi5UbKiYyx51LvdPNXHcuA7Eos105vNbfkSB6ALeUbuK+9tlY4h+OlTk9LAwXmTzftgyF5rVBehqi2DJEWhEjYQEkFNXbqJqd6pn/b4aMxQCJsByad8e4YzNoWBHA26XhLlKpPeo/6+8/RvbGMIYxjMAoi2VbGUrBhegNNybb9CBYClDrH0b3fojInI1wz0OXG1OGi0wXBPzI3Tth3ASU2+5FLnkXnE7EiaciHA7EF65Anv0ZQ73O64Wl7xEJh3hxrjmE8589rejA37qbAVjh6eWSbR+wdd7piaLeWaUWFIuRK5QOYa+eEmYW6NMonzF6mGZ+jTVFieuOr7/ES39eD8DUBeX86rXLcWUfXOsygFJ4cULSOrb1s+adoT1IzwpE9sKUcTIaRf/57YkyGPK1FyEjA6JR5Iv/NDotehuOOQ4slmReksOBmLcgZb6AFksQvSGs8w3y9JQjuWHvRppCAS4urOD6ynq2rExT03EYvJ0xcsqtWOzG39zqVNBjki2vji7QcrjWcBNCQeSeBBiCE1iLIdqZ3O807nEZbkXb9U2kZxnCvQB1woPIRVsMogcgdeRbr6FPnoZ86IFkiOCbr0BFJeLkZHkGq6Lws9rp/Kx2OqdvXGIiHs/1tvJ4ZyN3Ng95a72s3NLLTZWTTIW8L6tazl96jiQvaE8RiVBUaPwwSYB2vutn1kVZRgmT/YgFZuQeBgItfW9ietfJCPqGhWAtQRReBsEdYMlBCgfa9i8gctOUvVCzQEve66LgIoQrmccjzjwP+eFKCMVDlecfhahMLyg1ho8PYUuNFPlXIRv2oP/xt9DUCNNnolz9XUR20mj4Zn8H39uzPrG9yZ8UDnurv5NVs0/mpI2L6YsrDL/S184NFWZv+PqMAS6tXs78QB7N1gCXTq6kJuZk97IAUgPVJsgsUGndaBi1wl4I9GuUTrXTvTuCxS6omusgr/r/pqFA/8dfka++ALpEnH42ymVp1Io/ZdgyFCae/MmKnB1MjJG9MYxhDGbYKoEPzW0ZMxC5pxj5eiMgVCc4xiEbf4ps/z04JiBq7bAtuaiR1hg89wxyg6HIyYRJKDfdinL5l5F7d0N7GzIrG6EoyA1rkU/8ObFAsgBl9aW0uJNhntWODJYOdpuOoyHkpyUcTMjFB/q0UYkeAoom2gn0a6ZE9+xyC5n5FqrmOWjdEEKLGlLxUjcWtvk1ZqvihvebE0QPYMvKVp57aA1fvPGYUX744EDYSlPW2sJanL5zS1Oy3mEccvMGaG029/vwA8SPboO3XjcEWk4/B5FfkDJdhmphjivHZNk+NruAE3OKWD3HXFpVG+36xGF1KikCLVp0/wIth6Uap4wh+98xvK45JxqevvG/Qdt5pUEW7JWotXcDoO26GtlvqGzKgXfQdlyJ0nl56pxbN5lDBMEIsz05fS0+dURcswAWDXSZ2kK6Tl/UHHO50F9Itt+GLuPlNR0CqRvehUjQ/PtSN4SRPuoatG+NUD794BtEhkPv+HOaVg2ircieZ1FnLkbbeCLoQSQge19EKf0WeuefQQ8j8s5GjhQICZkL24vqWpRfPIhc+yEiLx+mz/pYx/pKbzvLPT0cmZXPufkHNyftvxlSSvQH7kuKf21ch3z8YcR3fpDo805/1yijwavF+E3r7gTRG0JHJEi1PYPGuEpxpqJyelkx7ZEQV+TX8M3Scfh7NXIrLMQikvLpjoTi8BC0CORWWamebwj6+Hpi7HzPj65JiifZDwuRo4MBuWUj8rlnktsvP4esn4SY+8mWEjpYiEY0gv4IWbkHV6hpjOyNYQxjMEHJPgK971lzY2AjMrDRkCJ3zYOh2lPWIkTJVWibz4Rwo9EW2oU8SoWw1RBocQeQU7yID9Ym59u1HbnkXfQNa2FdfK7qWsS1NyIfecikaieAXy5bxVdPPwGfIsiz2Li3biZ/7mhgZzApilBuc1JmTy4gY+FUQmB3K2TmqxTV28gqtlB/Yib7VgUTSfFVc530N0fp2hFB18DiSIYV6hrsWxkkt8qaEFjobBpM+Y2OxtS2gw2l4gfofa8lrolS9p3R608Vl4LdAeGkh0ZU1yL7+8A3THTCmYEyYRJMmJR2moaQn4fbG9CQ/LRmKvc072CNt585rlzOzS/Fp8VwqeZPji1TEBwRbmt1CmIRSWGdDVtGapBo6CMEWtTDzPgtNR+xDSeA3/B2C/cRqDPeRik4H5F7ClrTPcjB99Ca7kCtvgU5uMQ83rMM5v0K3hxWN9FuRxx/MvLdN82Er36yaaw3FuXbu9fxYm8bJTaHSUjiy8U1zMjM5qluM6m/pLCS9miIZ7pbUKXg2t6JCGlcB6mDxSESku8d28N07zILMGXmqZRMtpuKsKf+UQ6HGmH7CfeVYYMM6sPOQUbBmo/lyDbQQ0j/JrOKKvFQ3OofI9ta0B95CJr2IabNQlzxDSOy4WPgrqbt/KRxS2L75qrJibIY/yqCHo2WdSEifp38Whslkw8/kalDAs9gisqz3LXDtD3LlUZcaRimuVLLaoxzurijdjqPdTYS0GJ8obiacc7k9Q95Nba+7ksYI72dfgrqzORNKODMNjzgkYBu6j/QGmPqsDJC/82QDXvSth0o2Vs80M0z3c1U2J1cXTqOXOvB/1C89sRGfnXdm3j6giw4bRy3/fVC3DkHx+j133+njGEMYzggiKyFpBRCHYLuQ+Sejii8BNn/OmTMBGIQGvFC9mZAWz5CKkhPJnSmiqnIHVuTRA+gsQH5+ksp8uUAZze2ssensuu445mWkY1TVal3uuiIhHhnoIsJThe/nzDXVMjW6kwlCu4ilfELkx7CaEiCBFuGwFVkQeqSXYv8iXC12IgUnUhAokXBEv9OLDh9HK5sO75BwyorBJz82QNbiH0aEI5qLPO2ID3LEbYSREZ6gib9W0EoKN/4DvqjfzAWP9NmIi66FOonIR/8lUEmhEBc8vlRf68jEuKYde/SG7dwP9zewOo5p/Dr1l080LabpZ4ebtq7ibdnHM/kzGRtKEVNvc9chSq5lVYKx9viCfFmi3d+tRUtIunbl0blleS1OVygdz2VIHoA0rsK2fMsouhyo2RGy51GOxDzLEe45plyxoRrHsrUGcjvXo/+9uvgcKKcdyGibjz6t76HfPov4PMhTjgFcYLZc3pr41aejpO5hpCffIuNGysnMjHDzRm5JYSlzqLBbl7sbcMmFH5QUU+tM5NZrhxyVRunZ5Xg2Gt+dod7wvOqLHg6rPQ1RlFUIxdv+9s+YmFJTqUFZ5ZCX3OUsMd8jR3Zh4FiqtgP4VGciKwFyI4/mNttFejtDyG7/4FMt3wKG+Uc9Pt/bnjMAfnBMrBaEFdfm/an5L4G5Duvg6IgTj0LUWHO5XygdVfK9oGQPV2XbHvDl/CGe7uCCIUDVhUOh2K8//JOhBAcc84EbPb/guVjVjaUlkN7stalmGg2mFxSWMn7gz082rkPq1AosNppCgcQwNdKarmipJbV3n4e7mgAYLYrJ0Embqg0wjlf7+vgwi3L6Q1FuLC4nBvDk0xRJ1I3ohiyywwREItdUD3fidVhfM8GWqLmKBUJfY3R/xNkT0yZnhqlMnX6Ac3xal87F25Znpjn+Z42Pph90r+lijvYHqVhRZCQVyevykrdMRlYbKPPN9AT4M6rXiYaMRYXK9/cw2N3LuOau08Zdcwnif/+O2UMYxjDAcEIIRvd8i50H/q+nxvF1QfeRet9FtxHgHdVss+aCYiQsZgQCGhwmL1HqgXKK1MnV1TIyYWBEXWK6saTfeLpzM9IJkUX2Ry8On0hUV3HqqTmAKUL9RvuUIgGdba9Ocy62hWgco4jJS9pONzFqumFnp3n5HeLvswT9yzH7wlz/tfnMOeEmtEn+JSgtfzSkCVXMlCrf4xS8BmEYkPknJC2v9QjaFsvNgg7IPLPRXngKUQkhoirooqjFiLH1SN3bUeMm4AoGT187J/dLQmiBzCoRXmofQ+/adudaOuJRbinZQePTExKcAcHUr10/U0x+ptiBAd0quc7zQItU+x4O2MEB4wkeatT0NdoNg7EwoeXUirR1LqVMmrc33rvCyP6diJq7wIZRXo/QLjmoE58GACx4BjUBebwYOWohXCUOQ9TX74UuWo5oqiY5fXma9Ybi3BcdgF1ThdCCBxC5e9TjqIjEsKhKDgUlblr32Z33GP+Wks7T3D0iIM3wt8aVgTp2hVBAIXjrJROt7PpeV/iGRtojpGzwInNqRH2mB+qoRzMQwo1jcCCkmHU0qu7G5F7KrLnWWSf4VEV2ScgY4PIfTeljhuCsCO9ngTRG4Lcuhmp67BnJ7izEs+S9rcn4KVkFIV8fwnKPb82Qj7jsI14t9nEgeU7+nu0lHdhX2P0gMie3xPma0f/mX3bjHI5dVML+ePyK8hwHVzLitSjyK4nkYFtiLwzUHJO/LfmE0KgfPd69D89mMjZE1+60tRHFYLfTJjD3XUzUIXAoahs9XvIVNVEqZIHJ8zhBxX1DMSizHHlmEhEfzTC1ZvWcHP7VKaHcuhriLCispdKzJ5eu0shq8RCyVQ72SWWuCHMQLqi7HbXoc973R+kHgJhRYh/z7Aj6sYjrrrGyB/XdcTZn0FMnnZAczzSsc+0olnvH2C1r5/57o9X80/XJLsWBRLRQ32NUWwZQWoWjC7a0ri9J0H0hrB7Q+covT95jJG9MYxhDCbIyH5eQI5xSM1nEL0hhBtRKm9E2kqQvvWIrGMh0GAaJhAwYxbC4URGo4jJU6G4FJnpAn88VFC1oCw8EY47Ef3pvxjWVlWF7BzEuRciMtK/SNMRPTCkki12YQrnzKtKhsoMtsVScvqCgxqKionwuYpU9Kg0ZObnpsbZj59RzP/7ywVpj+FgQO95Ab3hfxLb2rbLEPO2INIUFtY9q5DdzyCjnQmiByB7X4L+lxGFF5n6i6JiRNEouX7DkJVGIU1BpJgMeqJmL51qEeix9IaFrp1hquc7qZztpGKmEerSuilMy7qkuzWdF+8A18KfOpSiz6G33ANa/D635ICzDn1wOThG1o5UEJnTESVXIsu+Y4R67qcUykjoi95G/vG3gGGuOUI/mbXVSYXYTEXlpA2LCUudSwor+WP9PGyKQonN+Pu+0NOaIHoAAZne8tHfFKVrZyTxO927o1idSkp0prc7RsiTOoevM3bIy2OI/HORfrNqMK75CGc1wjnByKsc90s0ayHoMdSqH6Lt+e7+5yw4G1xuKCqGrmHv0Yoq9Ju+By2Gl1Wccgbi6ONMRA+AYAC5agXijHMSTT+snMR3hwmE3FRl9jx9FOwuBSHMhi6H+8Aekjf/ujlB9AD2bunm7ae3cN6Vsw9onn8X2s6vGqqoAK2/hPqHUYq/aOojo73ojT81SjTknIhS9b8IZXRiK6pqUH96z0f+dmY8BH3ZYA97Qz5OyzWLwgwP0xyOtb5+ruuYxPSQEQ6ap9nI2mdJePHAiGZo2xwi7DUuUm6VlfoTMxBC4OuJ0bI+VQU2XeTK4QCph9B2XoXs/jtYclBr70Ip+fK/Nady/Mlw/Mkfe3xOmu9Ttnpg75/+lij9TVEcbgV3iSUlTcTbvX8V6IlzS8nOdzLYmwwNX3D6uP2M+GQxRvbGMIYxmCD9qUqaZB2LWvZNRN456M13pvH7qSAsCHsF0l4B9UsRHSOsZls3I379B+QdPzFCmwAmTYWyMkQkijjpNERNLQDKN69Fv+6bEAgY4i07tiNvuT0lxGZ/UK2Cyae7aN0QIhLQUSzQuiHEYFuMyjkOHNmpC57MPJX8Ghv7VgWJBHQK6mzUHuk0WVkPF+j9byEH3kH6Rl4vDX1gEfS/hex/A5E5HaXiB0jfGrRNZzFarpIcyrn8GLiooJzftO5mvd8QZZmc4eb6inre7O9MtAF8pbjGNK5ooo3WDWYCOATVJujYFqZzRxjVJqiY5WCgxRy6GUtTas9ddBiECA6DcNRimbkUveNPSATStw59y/nGzsxZ4JoDvrWG9H/5tWibz4GoEQ4o245AnbkIISzIXTuQ770JNjvi9LMRpeVGn+4uCPiNPMtli02//ZN3ltB1zdW8GByg0u6kIRRI7PtbdzNHZ+VzdHYBj3fuI0u1MmHEgjUm0hPxQBqPrJSkkAp3kYW+ptRw20iafNqDjmh/aptnMdIDsf53UWcuQVt3NMQMkhPrewGRf+6IASqobtAGIO9s1Jo7DG/Rt76P/tADhsFq4hQoLoYNyZxl+fbrhrJtOoyolfiNsnHMd+fxflyg5UC9EbYMhcq5DprXhpA6OHMUymceWJ5Q0J/6oKVr+zQhI10GgRgGve3BFLKnbf8ScuAtY4z3A9A8qON++dHzNzci33gFNM0g42lqIV6zay1/jIdrZioqr08/jiOyUq/Hi71t3NG0jaCm8cXiahYEC037LSiUT7dRNdeBlDDQYiZ0/U1RPB0xskut9DenD1f3dsbIrznMYtYBvfUBZPffjI1YL9quqxG5JyHsaSJ5/kXIbVvQX30BdB3lzHMR02Ye0Pjryut5pbc9EX3yleIa6jPcHzEqiZ49EXYvTb473cUqVocw0kCG2j4ipNbhtHLfS5fx25veobNpkFMvncol3z14IjNjZG8MYxiDGbHUemqgoBQacv5K6dXoXU9COG6lzj4Rfd+PIRZXLfMshykgV0xCRId9jJxO5JL3YM+wHJTtW1DOOg8x9wjkhyvQfnozWCyImjqD6A1B6siV7x8Q2RuOkFcnFn8xB/ojRPw6k051UTotHiIoDZIQ8uvEwpIpZ7iwZx5mLqJh0NoeRN/zvVH3S/9aZLuRbyT7Xkb61oGayaiiFMKOkn/eR/6u9PuQTz6K3LEVMX4iIl7IO0O1sGzWibzR34EmJWfklWBXVF6dfiy/ad1NYzjARQUVnJ1vrkPo6xnteAwv7HCRjx1v+8mrtuIbZkEVKox0PoU8h48cp971FFrzvQCoFdch1Ey0tt8kO/jXo4x7AGXiY2AtQG/9dYLoQTy/r+8N8E1Bv+1HCflSuWIZyr2/QT73DPLNVw2GVTceRqijZms615TWURP20hsJ0xAyE/pFg93c2LCRUFzkpdLm5Ah3Hqu8Ruhp/ijW7xRDiYCCWiuufJWmNSGiISOPZbAtikyjuBoLHQbXSN9P0epIC3rrLxNEDwDNC7ZyhGsO0rcWFCdKxS3IlftgVxdiwnxkVj/yg5fAZke55XZwZiBstvT19nLTkLaKKsT8I1OaaxyZ9Mci1DkyU8f8Cyib5qBwvI1oUOLMUQ44V+mUz03lsTvfx9NnPI/Z+U5OuWTqxzqWjw3FZhSak8PIj+JE73wcvfdlRMZEROk3EkRvCHrPcx9J9mRfL/qtNyUUoOWKpSh3/AJRVpHo0xwO8KeOZMSKX9f4ecsOnp58JIsGugnoMU7NLaY5HOTSrR+gxa/4zfs28x4npfzm7mUB8qpsVM5x0r07lTgPeY6co+S3ZuQfXkatIUjf2hEtGtK38WOTPdnRhn7XrYlcfn3TepTb70NU/eslTCZnZrFt/hm83d9Jud3JkVn5Hz1oGDp3mg2S3k6N8cc5adsUJuTRya22Ujn7ow0oUxeU8+C7B79sBIyRvTGMYQwj4RgPwZ2mJuGaC4De+Rf0jochYzJKydcR7vlIKZFbzkqZRh69HRZPRyCQQqJcfLkpET7Rb6Afdu0wRA3ibgG5bTMAEUWh2ZVBrceHml+YMnZ/0KKSTa96IJa6sBlojaHrkup5TsqmGSUYdrzjx9tlMIeunRFmXuBGKIId7/rx9cRwuBXqT8rE4Tr0H1m99dcjWoTBfBQnStX/psjKy/7XEYWXpk4kBcQsiPJLEnXeErt0Dfn2G7B9K0yoR5x6lpHbsmqFsb+jHen1oN54C2CE054zQhY+32rnJzWjLwoHO1KZQG6lSvURGbRtMn9gpQ7uQpWQV8ffo6HaBDnlKr0N5jnS5bccCujeNWg7vspQ/qu28wpE+XWpHaMd4Kgx+ulpvJx6ELl8qblOhc+LfPNVwxMxhL27YXw9ZOcYpTSE4LlLLuELe9YkiIYCDKdZYV1LED2A5kiQn9VOw67UMxiLcoajmIa9qce0Z2mAyrn2RL6k3aXQsCKIM1dl6lkuLHbB+n96Rr0WI0s2HAooJVeidT1pJg8JCCNKYWSrvRx19gfI4G6kpQD9dxcgVsbVGrcuR760AvT4O+zdN1Duvt8Yd9RC5NJFyYmycxCnngVWG/KFf0A0Akcei3LF1SlE7NW+di7ftpKgruFQFB6feATnF5Qf8PlaHQrWjyn8V1SexaMffo0XH14HAs67cjb5JQe35piw5KCUfQe99RfxBitkTELb+TUAZC8wuBxspRBJKmwKR2o4+0jI1cNqIYJRY3TF+4iLPpdoCmhayv3g1aKcsWkpi+NlgMY7XVxVUsvInhFVx6GZvxsRP3Rsi6BFjQiHzh2RRKq81SmwZijsXR5A1yWZBQr+HvMzEwsf+mcIhsrKvAlaAJF3FiLnRGTPP5MdFCdYC9H730RkHYtIlyu7v/nXfmgWbdM05JqVB0T2wMh1dVssZB1g+CaQKrwiDFVid7GFnHJB8WQ7qlXQtTNMX2MUu1uhfIYDW0aqwXjLqla6mj0ccWodmVkHTxV3jOyNYQxjMCPUkNIk/ZvR+15H23lFok33LMcybwdC96OnU++c3IzMH0SuHY+If+jEkcciX34+uXB1ZiDmzEe+9ao5/kvXeWfaFK6YO4Vup4Nqf4B/zFvAjAM4jQ939UEsPTGzuxXCHp2QTyer2IK301yTLxqS9DVGad8WJhhXHwz06Wx60cv8y/cvxX1QEBsZgqagLmhFqJkIxYbsew0ZHOZBVbNQKr6P1v86xOJhlU2FiCXTET4nsmIH2jUvoVYmw9TkU48hX3vJ2Fj5PrS2mELRANi4Dr2jDZGdi3Q42B30UWnPwKnunxDL+LWOahpWzH1Vu8DhVsnITZ3DmatSPl3B4hC48i30t0ZTyJ5ymJSfkgNvk1boSHUbXiIAYUOG24itMAwZovBSUDKTXidHHWROA9fm1HnSFCkU4TDcehdsXg9TZvBQ107kYNI7pQMzM7MJ6zpfK62lPxbllb4O0xwhXedPHQ2s9PRxQkYh/0uqGILUIOKTTD/HTdOaYIKY+3o0gv0a1Uc490u6nWlCqA82lKwFyHEPojffAZYsI6cytAdQUCr/B6X0amT300lPhXMSSlFckVYPom39DGLzCPakDzvnnm7kmg/B5TLI+vwjIRxG5OUjS8rQr/82BIOIk09HXPrFUb1tN+7dmChyH9J1bti78WORvX8XpTU5fOO2f08Q5eNAagEI7QbnJNS6uxD55yADW1FyTiG2/TJzZ89SlPG/Q2+4Me6JLUWtu9c8XygEg/2I4mSUgcjOTn1Ss83htBMz3JyQXciiOLETwDxXLj9vSRpGdwd97A2leoxt2uj3e39zlLqjnVTPczDYFsORpZBXa2P7MPGwdOjcHqFs6sGt1TYSUo+ibTzZCJcFcNSizliCUtmM3vkXsBYinBPRNhxr7LcWYZn+JiLzX1eTFYVFqdemsOiAjnN30MepGxfTFjFCZa8pG8994/YfChoJ6nRuDxMLS3KrrHg6Ymhxu1BBnZUdbwcS9US790Qom2an8cNkKK63M8b089ym5/quq1/hhT8a75PsfCe/X/xlaiYfmBH742KM7I1hDGMwQ0sT3hRuQPa+OKKfDznwDiLnBMPi2v47w0runAzBbQCIVZMQLcbLTP7+fvjCFSg/ug359mvIYBAsFuSrLxoKnMMggW/PrKfbaSymGjMzuG77h/xtwWksHuhmUoabqZmp9Y22BTz8tnU3USmp6HByKqkKkhaHwFWgsuF5b2K7cHxq7oNqFQmilzjlCIR82qH37qV8/TSQYYRi/B2VmtvRNp8L2qCRS1lwEfreG8A1D+FeAL598EgXImyct2gphGdfQX7pGAiHECVlyMXvmn9y2SKoqjGH4VosyB98m61FhVx8wRk0CEmuxcof6+eNWvz5tsat3N+6CyHhj+oRlGjJBYtE4iow/rZF9SUcbfgAACAASURBVDY8nTH69hmy/nk1Vra/7U+EbdYc6USLplq3tejh4dkTmammCSX7WETx59FafwMyinDNRt+bLOAsOx9FGf8ghFtAD6H3voC2Zga4i1HKToO2uLLnzDlw1vnwzhumWohSi8H13wZNg6oa7JeYQ3MV4MzcEqa7cvhMQRnd0TCPd+6jOWx4NY7PLuTh9r2s9BnGhLd8ndzEVENgaeS5WIy2vkazZ8zXY4gcjczfG46KmYd2kQpxYaPdXzc2woBwos5cirCXGWw21oM6c6nhtZAxRN4ZCMWO1CPENp8LkTZwHgv+/ZxLcyP6cBGWsgr40hVww7UMrRTly89BRSViYXoi1R4J7Xf7vxl63xtoO75oGKisxahT/omSfSxkG+RBWEvMr0LFiVL4WZSiz0FwD2RMQQyz/uhL30M++gcIhaCqBuX6mxH5BTB3AUydAVvi+c/jJiCOPSEx7u/dzTzYtgeLEHyjpA5VEVyQX05TOMBIuFQLt1RN5t6WnUR0ncuKKpF7JKR5hsAQy9n0so9An/Fic8dUbJmp4mEjoUUO/XtO9r2aJHoAoQZk12OoNbeh1tyGjLQTWznMsxrtQmu+E8ukJ5JzRPvR992M7vkAJftYlJrbEJZh3/a5RyAWHINc+b6xPWc+4kizKvFH4b6WHQmiB/Dbtt18u2wcdaOI6uiaZMurPsLxmq5CQP3JmWhRicOt0NsQTRA9gGhQGp7ZYQj06wQH9ITRsnl3X4LoAQz2BnninuXc8sj5B3QuHxdjZG8MYxiDGXqaYsih3UiRWg9G738LueOrgAaueajj70fbFa8n1ZuZIHpDkEveRTnzXLDZkD/5H4jFjI91do7xwV2zEoRCYPIUmtzmF/GmWIj6Va/hj1u5f1w1hZurkzl8HZEQx69fxGDc/HZGoCSF7EmnZNpZbtY/6020xUKSkEfDkaUk8r0yC1Ryq9K7iMQoohUHFfYyCJi9e9q2LyCmPoe26xvI3ufBXoNS9UOjNtiOpIiB9KxAsX8XwgPmObf2o3/n68YidMo0cLshMIz4u9woV1yN/qt7oLsT7A7eLcjhvtlTWVeQx0D879Ifi/LtXWs5M6/EVPcQjHpTtzdtS2yHRuQQCgSeTp3CcZLG1UECfRq5VRaq5jnZvThgys9rXhNMq0inHVzdiFEhcs9AKb8Wve13gETknY3W+xKKvQK19k6ENQ+t+e7UgdEu1JpbiW0+P1m/Uu1Ev+BN1KwXEfYMxASjfpe8+Xbki/9E+n2I2fOQj/8pybCa9nH97iaWlGYRHrYyuavFKBp9fn4Zz0w5iuemHM0dTdvJsVj5f9VTqVyVDA0tijjSEj2LXVA8yTAUDH9uhvY5c1Sq5jtpWhNMyakEI0z6UBM+be+IkFoZRB9YAr5VyHhJDFH0edT6hxHD7mPpW2cQPUAu2A5vzEPEVKTQwaohIvH3Rt14ZKu5YD1tLbBsCaaVIsDO7TAK2bussDIhCgJweVHVxzjb/zxIKdF2X5OMRIh2ou/9Psqs9xN91OqfEPOuhFgfoKBU34qwxOt4umYm5sG/Hhl1IB95CMLxsOSmfci/PwXnXWjUGG1qhOmzEGecg5g5J+GRWTbYwxe2J8sKfeDp4/vlE1jj6+fsvFKKrXY64yrDDkXh80XVTMnM4qvFNfyhYy97gn6iSNKZB22ZAnephfZhIeveTi1tVEPK2DQhggcdmjdNmyf5/2gPKXnikS5z911fM9SgAT2wGRntxTL5ycR+oaiI716P7Pg8SD0hTHUg6IyMSAkAuqPhUcneYFssQfTAeKX2N0WpO9oIQR1oTWXiNqcgNOzUhWJWTPX2pxpp0rV9Whgje2MYwxjMSPcCB1AyETknIwfeAVRE0eeRXY8n9/tWo3c/C/7V0JGDeCFVaGBIaU4uW2SOwx8cQBx5DOLLXwNVxbVyOce37mNxeVLeOkPTaFOTH7i7mrfz7fJx5MT191/sbUsQPQC3nkrWHBbFCMUYwde0KMw4z01/S5SwT2ewPcrW132pAiAC7JmHwWtTT5Nn5FmCtud7yJ5/GNuh3ejNP0fkjygLofvRLcsRmSpiuFfCN4wlbd1sLD57e4zrpKqIy7+CqKlD3Pdb2LqJva88xwVHTiWaJmSzMxpm2WAPdzRtpyHk54KCcm6vmZYQ/hhCpkwd63AJ9q0MJkQLQh6dWDiQInWtxcBhTw0fVg59SqUB3Q/2Gij9JsJWGq/PJtEx6utZZn+IyE5d4It47TCTxRxA60TUlyNsyRAmUVUN37oWoemwc3siPHYIC5vb2HTuxbzR38lTXY2s8CT//i/0tvF6Xzuf374KXzwkdKmnh7muXNbEPXthNb17wZGt0L4lTMUsB1XznAQH/IR9huJtzZHGPRUN6NgzBSFPqnHE0xGDAxPU++QR6Ulpkn2vgHd5crvrSWTBRUgZRW++BymjKMVXgLAZ5Wcqe5CffwfZmQv5HpAOlN47oKQMceKpyMf+mOqEnzQl1e0ZJ+/p8Itxs6hxZLLc08vMzGz6Y1FO2bCYk3KLuL5iYkodPoBAv0bjh8mCz5VzHSjK4acovF/IKIRH1CsM7jFtC/ccLEfsRnpWGOUyHDUjpuhH23wW0rcG+two4ePM+9takff/PFkXcdN6yM1DzJqb6PNKX7tpTETq3B03mNzfuosXph7D37qb2ewfZI4rl3yrDSklF25dkVAi/hb1KadXe6SDono7rRtTc2IdbgV3sYq30/j4jCwhBOy3HuxBg2tOSpMUDvTeVxA5J0HGNENx2J8sHaIUfx4pdWTvS8jgDmTvK+bxceI3EqKkNG37v4IvFlebruMkp3u/qrbpiqMLFRo+CODt0sjMU7C7FMI+gxBmlVqoWeBg+1t+o6algMo5DqyO5LM5eV4pE2YVs2t9siTLuVfM+tjndKA4DFYtYxjDGA4rKBmgpVqcFPdslLo7kcG9oLqQ3g/QhpM9QEZaQXEhFs1AjFzIOxxGft6OrZCZalETLhcivwC5azty0zoe27aZHx05h/WFeRzf2sHSiRNoG9Y/InUaQwEeHdiHIkRKLZ3tdg8jYXMaNacy81X8vcmvZeF4G4pFkF1qYd0/PGjp1a5BQiwi034MDipi3WkaBdK/aUS/PoQlK3XBGetGntECy6fAgAtyvYg2s5KjsFrhez80CnTPmY8y/0hkTzfy7p9CWwtvTpuYlugBLHDn8eXtH9IRNe6j+1t3ka1aOTbL/Bu5emqCetin4Wk3ez68nRrlM8wLo/xqK+5iFX+3+V61HvoIQaSUxNbMhng5i5S/v38j0rcaYStBlH0H2fcaCIFa8QOEtRit/Q+gp7H6CvMnW3/nDeQzf4Fg0CiunpMLA0mPrzjiKKodmVxVWscbI3LzAJ7tbk0QPYBdQR+/rJuJKgQfevuYlZ0bP3rz/e7r0vB1aQT6Naae6WbyGZn07IniLlbJLrHStDpI2+b0JTUA7O7DgHgIK8iRUQypxEkOLkNv/RVD0jb63u8hyq9Htv/GuEbOKNR0QUs+4vX5yNgrhllfURDnXIBctzpxTcQpZyAKi+DMc408vlA8Z29YyOBI2BSF6ysNMnjO5mW81W8sFpd6euiPRfh5nZk1S12y/W1fopB6+5YwqtUoXfKfBKHYEHlnIvteTbbln4eMtCNsw/LtVBci99S0c+jtvzOIHkCuF5nlR3iSiqZi6nSjWPcwyK3md+gE++iCIu2REEsHe1jh6WWlt4+3B7r4U0cDv5swx1RyJt3dHglJPJ0x8msstG1KkjfVCvm1Nkqm2PF1afj7NVo2pEbbHA4iR3hWpDTJpp8avjxbBZaZi7BMf9WIYAjsRCn+IkrhxcR2XoXsfDT9nP9mEfZ0uKCgnH9MOYpnupupsGdwbfkElP0o0rqLLeRUWBhoMd6N1gxByKszGN8eijipPsKJajVq+gohmHVRFv5uDbtLwTZCzVsIwQNvfZF//PZDupo9nPK5qcw/ufYTP9fRMEb2xjCGMZhhLQKtb0RbIVJ1I/Vwslh39vFgyTWJhaiFF6PbysC7NXXeDBfy8T8ZC9+JU6C4FDrj1rYZs2HaTOTaD9F/cSdISRHwh0Xxj4nVysMnnsF3epPS8SdkF3Le5vcThKLC5mBhVgFLPYbF3lIoyJCCQEdyqe3t0tj4gpfiiTayyy2EBjWkDj17I8TCEluGGJ3oYYTd7FjXSjikMWth1aGzlqeJjRPFX0JY8tGHWVGxFCAqb0T4N8Y9sgqi5GsGYS/cijw/7j0KWuGvpyGGOfekxQL33m78f8m76Jd/GVqbjVA0oL5/MOUYKu1OTsgu4nNFFZyz+X3TvncGOvnfqklcX1HPE52NWIWSdhEU8uo4c1Wi7UkSYncplM+y48hSGWyLkpFnCLj0NaYSimiaKOSDDb33hQTRG7VP35vIptsBDZRM1KnPg4wQWzN9FIVI0FET4WCys90ISxvyEC1bBOdfjOjpQvb1IY49HuXYEwzjSWsL19RV80Z/B9F4/7PzSimxpxKAGmcmS2cZ3sVoSGfNxlSjyRC8nRo9DWH2LA0mIhPLZ9rpb9nPQwSUTz8MiIeaAfqIc8ucEffsxU9GWJHCktyOQyhWZP1jsOlyxPtToLkQoipiSBBK6sinn0CccDLKL34HWzdBbj5yzUr0668xrllWFsqtdxkh7X9/CimEUWu0IL1ggzcWTRC9ITzb3ZpC9gIDeoLoDWGgNUpGrkLz2hCxiKSo3kbFLMcBl2A42FAnPoq+71akbzU4apB9bxDrfBQypmKZ8gxYS9B2X4PsexWRMRF13K/BNRt9303obQ+Z35MC5NmrENu+DP0qYt4COO9CWLEUuoeFFtaOQzY1ItetRpSUctne3bwR7uH5cdUoUqKP+JvtDvlYOSxioS8W4dUR3sAoOrYRgZyt68O0EsaRpTDx5Ex69kWJ+HQychV0TSKEwkBbdNQ6pOIwiOJkhIKzCZEW9LYHEBlTkR0Pg+ZDj3ZCxkRk5+Ojjxvl3ffv4tz8slHzyNNh4smZeDpixMKSnHIrq/9q/t4NtMSYeJLZaK0ohkLnaMjOc3LlLceNuv/TxBjZG8MYxmBGNI3XKNqNvuV8dHsVlhnvgupC7/gzIu8cI39FRhH5F6K3P4zsfw2hnAojpKbpGxY2tWMr4rs3QH8vctsWKC2HgX70t15Lr+oQjXJlTKVo8pG80tfORKcbXcqEMhpASyTEjZWT+GnNVCJS57jsQtAkLRtC9DREifiS83bujDD74iz2LAvgiZOKwdYYxRNHL1Kr6zq/+/nLbF5jLOLrZ5fw4LtfOqjyycmDSfX6SO96cM1BFFxshKPZq1HyzwHvatRpryCDe9Cb7kR2/pmU4mfOKPKrdYjVBRAIIE46DfnUY+b5X/gnVCbzhdyR1OS475dP4NvlE/DEomQqaiK/EoxaYbPWvMX2oJdMReXecTOR21Kvtb9HMvEkGxG/TsijY3UK8mut9DXEyKu2Ujjexo53fDStTp/v8FHCBgcDcmDxfveL4q8gW39FIp9F96PvuwUUu3mxM5CB2FEJio6c3ISItII1Ll6wb2/KsyL6elG+9f3Etv7Xxw0BEOA4q5VV37+Bl3LdVDsyuKiggvZIiEc69tEVzzma787ltNzixPjhYUjpoNoEHVvDphS0tk1hciqsBNMUXh9CcFDH8a/XNP50YCs21TQEIDaAKP0GBLaC4kCpuA6kjtZyj6mbyJyG8H4Ab8xNyUtOIBiCaMyIN545B7weo8zC0DXzeND/9gTs3JaoKSrfexPlrvsR2UnFX78W4wNPL3UOlyk/DIxnaiQcbgXFYn4O7G6FXYuT6oGtG8I43GpaYarDCcKSgzr+V0ipE/twAkTjJCqwBW3PDxCOWmT3XwGjJmVs2yUo1beit/wi/YT5DsS1tyIiFoTLuAHlt76P/tCvoaMdJk5GzD0C/ebrQNeRgM3u4KlwiNblqwmqKuedczL7soyx1fYMjsnK56H2vaafcakWvlpcwyOd+wCIColtlFTvkEfH0xEjGtAZbIsx2AYd2yNMOiVzv95xq+PQE3Ul53hk6dXo7X9gpEEEQIbb0NsfSnyvpG8tevN9pFUpjkPknPwpHe2BQQhBdmkyWsiZrRAYJtjmyDoc2Pa/jjGyN4YxjMGMdFWQhxBuQmv5BXLgXQhuN9qEHcvMxWhtv0b2v2a05fqhK7mQkMhUoYfGBuQrzyeLpb6/GGr2UxPpzw/xmV/9ns/EZcd/3borpYtFKBydbYQKSinZ9k6SzJnPEUKDWsq+wY4Y5TPstG02FrBCSWopbN3YmCB6ADvXdfDyo+sprsjiDz9ZhN8T4TNXzeGrNy8c/Rw+KaSzfgY2IAMbELlnYpn1PrENJ6K33Ast9yLyP4NSdBmy+y+jTinKwqjX3ZTY1v5irtWHpiHmLUBuN7y2HxaZQzIB7m3ZybfLJ5BlsfKn+nlcu2c9XdEwJ+YUEtRibA8a+aB+XeMHu9fzMsenPZawXzLzAjfBAZ0d7/oS0v6OLIWaBU76mw8DRrcfiPxzkO2/NTeqRYjsBaC6UMq+idY5gkxHuxDWYX/TwQzEPxciovHP9NYqWDisMHH9ZLBYzLmvU5JlEqTfh3xtmIJuNMrEF59nyi0/A+Dl3jae6W7h4oIKqhwZVNidnJdfZhLVie4nVEyoUHOEk/YtZtItdSifYSfk0UYlfL37IuRWHFqiIdxHI/0bzI09TxnLUOcELLNXIdRMpH+zkZ/c/XeQGqLoMnT/dqR/G6J1P8WZZ8xCv/1HBikvLEJcdJmhkjocrc0JogeAx4NcuRxxmlG3dJ2vn7M3LaM3FkEBPl9UzbM9Lfh1jSKrnbvrpieGSl0iMVSExx2bQcOKILGwxF2s4i620LvX/M7wdMQOe7KXQGwAwmaxG+nfjBzRRrgpHsEwAqobLPko2nXI629E9vbAuAko370BUT8J9b4HkeEwwm5Hu+PHMKz2JGHj/i73GyEDK15bwnM33QyKwsWFFWQoKnc372BLwPASZyoqXy+pY3JmFl8tqaElHCRz7/6X2oEBPREyCIZDsmN7OEXHZzgU9dCTPQB1/K9Rqm5CxrxoW86D0BDxVRA5JyB7/m4eEG5CFH9xmHdPQNPRiE0CsnJQLrvlgI9BtjSh/+Nv0N+LOOZ4lNNS6/4ChHQNm1BMIZwdcZXOnJCV5nWhRJ5r2XS7yfNde2QGO9/zEw1JrA5B7VEHVi/wUGOM7I1hDGMww1oM4dQQvQQCm5NED0CG0TseRnqGCUoMpFqcTQvTTBdycMC8UO3rRZx8BnLLxqRi2nBEwmzxD/KDvRvYGfBxUk4h5TYnrRHjI1zryOTiwopEd29XKpkbQma+iqtQRbViCtu0ZyhUznFSMsWOFpVsftlLLH4oPk+qzHbTzl4euOFttJjxVf7DjxdRMS6XUy9NrU32iUJxpFdNBWT/68Ss+UbZhaG23ueRthG1iQYzECsnwmAm1HQizzWHpIizzkP+NRluI848F3GGUYdPrlzBMSUVjERbJMQ6Xz8/3LuJXUEf5+WX8aOqyZTanRyzzlzKISB1BpQI+XpqSF9GnooQAm9XjLA3aQUOeXT6mz6dMJ9PFCMU5wDQupB9hviA1vcqIvdUQ9Y/DqXoCwhnHdqOuPLflqok0QNE0IH83++h2ZyIk05FOet8lGtvRH/6L+D3I048BbHwRGQkgnx/MbKlOZVc7N0NwHM9rVy6Lfm8Ts5ws3bOqSl5LEJNb4Gfdo4Lu0sQi0Bh2GaqL5VXY8VVYGHmZ7IIeTXW/zNV8Cni3c8q9iBBxlIFWhII7kLveRbZ8xyy72WjLWsh6uQn0bdcgOyKG03EGSYnhcTIkxR145FbNxtED6C7C/mPp6B2HDQMExkpLjWHEAL4k6U0bt23ld6Y4UHXgX/0tLBl7ml0RMNMy8zCHlcjat0Yom1TCF2HglorIa+OrktyKi2MPzYjRdwDDMXh/wTISCfSvxEyZ4N/XaJd5J1mhMsGtiQ7W0sQOachu540T6J5IepDPvkiwh+PxNizC/0vj6Bc833konegscFIJ0gHlxt8XrDayLnsy1xZPs60e9HME3iyq4mBWITPFVYmVB4XZOWzAPiAgTSTDp0I5JRbUt5rQggKx9kSQlUpf5fDQBR6CMJWavyb+R566wPIaLfxPss+Br35HlNIu8g/G6X8+8i8s5GB7YjGGuSrz8T3SuQ996Pf9nOE1wNVNYiPqtkaiaDfeWsiL1bu3olus6GckFQP92kxvrZjNS/0tlJsc3Bf3UwuKCjn6l1reLyzESHhubaF5IQM44e/R0MIKBsWbu4utjD7s1mEvTp2t3LYkO1/FWNkbwxjGIMZMpXUJCEQeWchB5eYm9UMRNYxyCGrXsQ6chScdhYiEACrFXHa2cjlS1KCOcT4CYh7f4v+2ot0vfsGP14wm3WFeZzQ2smPMvK4aOsKGuJFa5/oauKq0lqmZ+agIri4sILsYSItnvZUUiAE5FZbqZ7vQLUqVM930vCBkW9ksQsq5zqIRSStG0J4u7QE0QOYOqsW59/tBANGo8WqUFiWlSB6Q1izqPHTJ3v66OE92KvTCuyQOUw5TYJ4aQHCF7dO9mbDzq3on3sL5URD7EA55wJkVS1yx1bEuAkway5y5XLweFA+exkzx9cz8Z2/s8OdtHDOyMzmc1s/oDFef+pPHQ24VAt3183g3PwyVvuS+Z0TnW5yFFtK9E9GvkLIo+HMVtIqztldArtbMUljD4dtRHigrkseu3MZbz+zhZKqHL55x4mMn16cduwnBX1IFGI0aIOQdRxK1jFI/waUnFMRJVcYlmR7JXrvS8iWNHP0GrlB8slH0fMKEEcciaIoSJ8XMWc+Qgi0++6AzUmP1YDNypKyYsYPepkSz7N8otOcT7gt4GW1t58jsswKdXos/YJGapItrwUIDeqoViidakfXDBW6iF9nz/sByqbaDZGCVMFUymYeBjl7vvX73S29a5JED8CzFNn+x6TgB4AlBpFhi1EhDW+REGhvJIVFAOjpRtx0Kyx6B7o6YMIkgxCO/N1Q8tltj5gNOkFdQxdQZnck/qTerhjNa5Njuncn33sDzTH2rQwy/rhMao9y0rw2hBaVFI634S5U2fyql0CfRnaZlbqjnR8ZtnuwoXf9DW3nlUYkg3BA1jEQ6UDJORml7i7QQ2iRDmT/G+Achzr+9yg5x0FwhxE+qHkN1VSAoC1J9IbQ1ID8w2+Q7xvfM/num3DSqbBtS8K7J445HvG1bxlksLQsEf45HFkWK98sMxPAoKax0T/AxAw3qoWUXHB3sYotQ6Fogo3sMit9+6IMxo2TQoHCCTakJnHmKPS3RPF2mF+GmXmHB1mXMQ9699OgB1EKL0GtvcO03zLtJbR9tyBDDSgFF6KUfw8hFESBoRKtPX+beULPIPK6byI1DQoKUf7nJ4gyc7kFKSXyn38zSLrFYhKlApCrV8Iwsvfz5h0819sKGKI6V+z8kJCu8Vj8PVgTyUwQvSH0NUVNZA8Mb6ozZ/S/u687RtvmMJn5KqXT7IeVAu4Y2RvDGMZgRjoiobgQ+eeiFH8BkXMKsv+tZLiMtRBR8jXwb4FwM9KzHBQd9ORLUSIRaz9EllWgfPZy4+V9yhlGCYYhy/b0Wci8fFj0NgwO8tVTjuW9CkN1bVNBHm0D/gTRG8LywV4eGD+HyP9n7zwD46jutf87M9slrXq3ZMmS3Hs3LmAwxmCDDZhgQm8JJQFuCGmEkBtCQgIB0iC0kJAQktB7B5vi3nuTLKv3vto6c94PI+1qtCsZcjH4vVcPH7DOFI12ds6c51+eR9ejJMi9ndFMQUpoKQ/S1RBi3NIEUgtteNt1uppDpPZkJA5+6IkyigZIcLu4+Ycr2Vu+l2BA47zrphHntvPIHR+a9hsz7T+XiP7sGCAzYklBLfkjCBta84v09oSJ+OkoWVciVAda1YNQeSRC9HogOlXk4w+h22yIOfNgy0ZkQz1i3smI7Fz0vzyKfNco05WvPA/ZubzQ1c6Vp81lY1Y6E5pa+EnRNFaWmknKB20NNAb8uFUL56bmUu7zMMqVwEhnPNpuovynupt1yj71otp8jDzVhdUhCPqMpa3FIUgrtpNWbKdyi5f22lCUGEWwX6ziX7/bwKM/WQ1A2e5GDm6r5YWym7DajuNi6TOE3QUgOzcgu3aiqwmo6ReAxY2In4Ys+y7CMrisqNy5Dfnhu2FiJxOTENfdZCJ6W9JTWLZsEe12YyHz/a27+RmQYTUvegWQZjUvdjxaiB3+NkClr5SORFKzx4+v3fgOakFoOOhn5MI49r3rCRO71oogeVPsUUTP6oKknBOgfHAgixkARzHYozPXMtSGib32D2pJgd7YiBIKwviJ8HGfuaGwCCUrB1Zdiv7kI8j+ZdK98EYI3tcz8tlxJKIOOTshhUv3b2RdRzNJFiv3j5jEqc1Zsc4SRkddiNbKINU7fYT8kpQCC8NnONn9eme4zLa1Iki5CiUnx6jI+IogpUQr+16kZF36QOvAOiPi04kaj2X8K0gZQvRRqlULfopa8FOCW2dAb6lunB+Z0oFocUeOHz0O+ZF5/mb/XpS77zcEWrJzYPpMhKKG7TE8WojfVx9ml6edxcmZXJY5PEroZl1HM+ftWUtLKIBLUXku5yTijpq/8531GiNPdeBKUTmyvptQUCc5z0JcqoojUeXwao9BEAXkjLNGkb2k3K9++S41L6HtJ4H3IAB6xS+xTNmAcER6u4VrNJaxzw50CiRrEPQj0L0VCU2N6P9+GvWW75mP+ehD5Iv/ZiCILLMQy5YuMxn06TprO5rDPzda/OhI0zznbTPsS3In2rHYYwdBpJR4mjQsDkHzkQCVW421U8vRIHX7/Uz7WsQcvra8jdLdDUw4KY/ElC9fMvqr/7YMYQhDOLEgYkTd9S4QCiJpIXrVr5F6AJGyFJG8GJF4ck+tfo/xea57iQAAIABJREFUb+JScHaBJ1IWKBBGA3xdLXrpIZQHH0Hu3GaUMWXlIOYvhJxhyNtvhUAAj0Xlw2suMl3CRw4LaRYbTaFIWcsoVwJn7vqYD9oaKHHG80jJNOb29Owl5VppLo1dxhnoltTt9eNt18K9Ep11GnqQQcsEM3NSWHrrWabJ/zu/W8JjP1mN1xNg6RWTWXrll+edY8LoZ7GknmFIynt2GqIsre+APQeReAp6ze8Q9jwsU9YSOlIw4GnkxnWwZRNyg6GmKf/9NOI7P0R+8K55x7oaCqVk9UtvE1AUbLqOP7WIVKstXHoGUOKMZ+rWd8MiIONdbr47bCQztr3Ps2oK6VrsLI8WkDSXBhm3NJ7KbT6QhneRzalwcLWHlvLY96m/cMHaNw6bfm6q7eLA1lrGz45ezH9hOJbZX9xE9KZ/g2cnALL+r2gILCMfNXwSu7Yhi4dDQ1JMU3MAYbMj+xA72tuQ680KqL+cNjFM9AB+M2Uci9uauD57BG+11lHb069yTVYhWbbIffjhkV38ofowIalzo7uE8zvywtdxJL2LuM5E0+/RghjlZn2IXcgv6Y7Rs6f5QQtKVOtXHPXWYlQwJJ2JknomSsYqCLURqrgrUi4trKhZV6Ej0Wv+YIyp0uQZLZGI738bPRCAnFyYdRIcOgDpGTBrLrKzA5qbkO+9NeBllY4s4R/le4hXLVyeWUCixcobLXWMdiVQH/CFsxFtoSA3Ht7G/pFLBv0zXSkqh9Z4woItLeUhbC5vVD9lZ/0J1gcrQxDsV+LqPYRW9j1E0mkoKWcYu3VtQ295A+Ec1ZMtUtArfoZe+1hUhYNcvAWx+zqoaUJMmgpfuxi5/hNz24DThcgfbnhYxsBl+zfyWo/a5nNNVTQE/dyWZ/ZJvK1sBy09c2C3rnEP+7hLTIoKfDQe8lO/j3BGz4OOPV6h8VAgkgmUULs3eq7ratJJG6TF/cuA3vRimOgBEGpGq/4DlqJfG56fWgfCkjjwCQA5eh2UzkI0JSKFRMh+80JjPfo7byBfeQGkjjhrOVSWm3dx2Ll5wSw+yslkSpeX381bQl+d0FMS001KtomqlSsyC3iyrhwNSUbQTn9taC1o2JZ4WjTGnhFPyG8I6TiTVJyJKgGvzr63uyLPUb/LDnZL2mqDJGVbef7hzdx/01vousQZZ+W+Vy9i6smxv1/HC0NkbwhDGIIZqgtirKNlwzNoalJYeEICaB7wHY0QPYD210GNLbxhbG9Df+Ff8OoLfc5dB+MmQI/CoyukMazTQ1VCJNI8sr2T7087lRsObaUq4GW+Ow1dGpkjMDzCLt2/kYMzl2ARCh2ZAZpVP6labLXMgMfcFA/GgtXhVvC2x86cOZOUqCjfBTfO4LzrpqFr8vhmi44FXwX4awjtXgq+UhA2lMJfIRKmo+2YF1ZEEynLIMWDtIZMPWFhxMUj1/QROQiFkG+/DqpiWthisUDQ+KLYekqeHOMm8GehccOhrVQHvCxITGOY3RUmegC7uzv4S/1RJPB+fD2r2gd+6YWCkoMfeMIqaN42jeEznAMSPYAR88wZy8IxaWx+P/L9tNpUcouSBzz+i4Bwjo6hNydQRtyHcIyA+OloG/NMW2Xre8b/Q0appfDao4mexQKajph1kvG8vGsuFRRWK8w6ySi3BVoc5u9+SAhO27WGeNXCgyMmk2K18WhtGU/UHeHvDRX8MG800xOSub8qsoD7W3I5GSEH432JbHW2MH66m+RqK92tkXvqSlZ6ypvM9yU5z0pTacBUwqZrRrYpOc+cFfvSEcvHUGoo6SvRSm9BdqwD9zyE6gIZQmrdhHYvQ7hnoY5/Ha3yXhA6UbnpXpXammpE0UiYMw/5+stwYB/6v59GnHfhgJd0MDmReXTQVWlkIh6rK2Pr1NO5Msvw41q0w6zy6tU1ahxeiubFUb3Dh65LUvKttNWE8LXrJGSqZBTbouY5T4uGw63g64jMc84klcbDARIyVRwJKg1VHRzeWc/4OcNwJ8fORFQcbGbnp5WMn51LwZgBVEn/UwjV8JWUfXrWdJ/heVj9ILLotwhbNtq+VfRWOojMy1CSTkOvuDv2OZO6EdfOQukpIQTQV1yA/FdPD6bFgjLI/WkJBsJErxdP1ZfzjewR/K3+qNGzl5FHhc8cSDi3Pi+mAKViEbRWmp+ZlqNBLHbzcx9LqMXyFYhA94dsfTt6rPkF9IyvoR24HLyHIG4yljFPI5wl6LWPoze/gnAWo+R9HywphhL0gp2wqxAZ54WKTFP2VRSNRP71scj5//EXxKIzTb/zlvkzeWmEkU1832Hniup9fJIeqbC5ObeEuoCPfzdWkmt38qvCicxwp/DcuDk8UHWQfDmw2EpHbYjWqiCH13jC81jeVKPdwxQwiXF/fR0a/mTBwz/6AF03dvB6gjxyx4c88tEVA/7O44EhsjeEIQzBDBm7IRyhGJL+fXdtXw3WGC/5WCSiF4oCpQfNY/V1kB8xGBVAQUeniewNV6w0hwLcVTCOhUkZZNudTNhsftlUB7zU+H3kO1w8XFdGni2OVK/xVtSQqL2LZwFpJTZaq4ImiXKrU2HYZAcHV3sI+aJnb3+XjpQyqmxHVRWO0Ud+/FHze7TurQbRA5AB9CM/QCQvNi1sZctrYHEgpx2A9WMQKBG11OGFiIWLzWQPQAjEWcuRL/WU4wgFlq+k8fUXuWfqePamJHGGT+OWlDSWCMEDRZN4s6WOaQnJUQsfgGJnHBYh2Glv40LyY2avhDD681rKIy/U7ladlmMItPg7dOhTSXvlj+ezf2stu9ZW4UqwcfN9i0lOP77latJXEWsUveYhLNP3AtLorewrXBA/EQAl7Xz0iruQVWnRn0qcE1IzYcIkxJRpyOwcqK0xtlmtiFMWIfOHI/ILkOVlXL7vMOuzop/PLi3Ed4/s4Md5Y3ir1bAf8OoaPzm6h5tzS0z73lM3iTF+Izq/yJONe7uKmgMZo2x0NWo4kxSGTXTg79JxJinhBVBiroW26hB2t6C72fws2RNOgN4wxQZ6v7kuUI128BuRXj3fEUg5yyAe7Ua5n2x6AT3UAVoXaObPtv/3WFZVGL1evfB5DQGqxCRojxbt+NvIQrr6eMOV+7p5o6WWFam5lPk8nJacEfYRBci1OZkQl4jPreNKUZG6JCnPSsEsF7omUVTRk0U194wlZllJGmah7NNuult1HIlKj+x/CASU1R7ioZ+8jaYZmYh7X17FtIUFpmt946kd3H31q+i6RAj4/sNLWX7tVL4wSC02y+mBXvPHHvXayD6y/u/og0lYAtjMJX7KOecjJ01FVhxFjB2PSI1WGe6FU1FJUC10apGXRprVzik7VrO3R43zN1UHWZ6awzONEaXQMcHo7JZqE+RMsBu2C97I82FPUEjJt5pEj/oTc4Du1hgNzV8yZLAlejDYjnbgKoPoAXi2ox3+NiJlGXrZd4zjWkHv+BQ1+zqoS0K8PAchjdyadHtg4elQW42YNhMZitGOEedCLDgV+ekacDp5N998Tzd1ttKthXCpxjrEqijcVzSJ+4rMnpTpVjvjXIkcoJNuEcIlo9ctqk1Qu9tnen6qdvhIyj12sCqzxE5Xh5/uTnNrTGuDZ4Ajjh+GyN4QhjAEM/yxFqqgZH8T2b0P6S+PDFpSEJmXGaVnvaEtS2q0x14vbDbEhZdCdaVZnEAoiMVnIvfsgO5uvKrK2myzeuS/0tz848AmAEY7E/h0yqnMS0zjoDeiXldgd3Ff5X7WtDeR73Gy0hvJGqkItDidrBwH6UU23FkW8qY6ObrJCxJUK+RNcZCQaWHqBW58HTo7Xzb39eghCPkkVueJ03gdhtYFvnLzmPSDHqNcTZOIDQbRA2ORKlO7UL97OyIlFW3aTNjSowqpqihLliEmTEaOn4g8egRGjUU+8gcuWrwgfJ/WAN2Ht5Me5+bmUkP84sn6cpalZJNksdIWMt6WJc54Ls8sIMfmhHeUqAVyWpEVq1MhrdBKR110aZk9TmCLE1G9er2o2u4jc1Qk7J2cHsejH19JQ1UH7hQnDteXkFHqK+zRF75SCNYhbNmoo/5sLIr8RyF+CmrRAwBIXyki7QJoPxJ9fHsntHciyw4j4hNQ7vwl8oV/ITvaEcvORRQWIR/5PfIjQ/n0ciDJH+DVwjzeycumyRXJ0LSFglG9LAApFisKxhLaqathoteLjlqNjloNBIw8xYU9XmHPW11GcERAzgQbcakqhz/y0l4duzRQC5wIUoKx+gZ1k0IqgGx52/A/7DvW9j44RoFI66fGabaYEWMnIPuqbwL4fSh3/Bz56ovI7VtMpM8RY2Fb4eumZNOb1AZ8ZFhsXJKRz6bOVgocLn5ROAG9G/a+1RXmRa2VISacHU9cqrG8U62CkQvjKN/oJeDRSS20kTPBjqIKJi53o+s6m/8RMZfXghpP/OJDNC2SiXj49g94fO1Vput67M414WyFlPDonau/ULInFCtY3BCKQSgAFCdRtXOAiJuCZGCLGdRogRUxvBAxvDDGzmY4VZWfFYzj1tId6Bh+emelZPHj8ogiaLeukWix8tPhY1nd1siU+CQSalRC/cSTS05x4UpSKZztpPSTbrQgWJ2CgplO4lItWOwKrZVBQ4zKo0WRvUD3CfAMWTKix4Qwq3UDsmu7UQXUF13b0Ns/RawfjZCR4I/oiINZc1AnGMqoct+eqA51ZeQYxORpyKuuQ9+2CdlpXrMoumRPdzvpVkdML0ow+ioX7VxDqKe/+gx7NrN9PbZNPc+xUHrsZfaayZrUIDmvn4JqDCGqgFeSkORg/jmj+OjlA+Hxsy43k84vA0NkbwhDGEI/xIqMWlBG3AeenYR2LzfMiBUnatEDKClnIMa/jl7/N4Q1HV04wbYxSryAc85HOfs8hMuFbGpE7tkF9bUG0Zu7AD79CBYuRrjd2CqOkuIP0OSM9BFpfbJp+72dPN9YxT2FE2kK+vmorYmxcW5SLTYeqTMWyae0RL+ILD6FormRko3ssXaS8yx01odoqQxS+mk3iTkW8qc5B267OgGSEjGRcgYiYSqyo0/flnM0St4P0No/CgsdiKTTkFt3mV6wADTHoX/7GigqQdz0XZg5x1A1kxi+VLoGhUXId96Av/+FGqedtdnzTKd4rqoU1W0mB2+01LJpyiLeaq3DpaosT83lxsPbeLGpmpeD0Z6Efo9OwUwnFruC1alQtcMfJgeqTZA2wk7aCDu1+/y0VQXwtpnfsLFk5gEyhrljjh8XxJIRBXCMMKxNAOGeg8i4DNnyKsI1HoQVvf5vhvogIBwnQ2DgWi25dSNsXGdEtwF5tBz9+pvCRK8Xy8srWV5eyZ0zJ3Hv1Igv27T4ZM5NyzVlIGxC4ZLM4RQ7E7i38gCW4CBBDQnVu/xYbCKSBZdQvz9AaqFtUI8wLXgCLFT1WPYyCiJugllxM24CQnEgO9ebxhBO6N9fBDBzDjTWQ3qmkXkdNQYORERFxMLTEdm5iG98C+3nd5jI3pX7DvPnSeOotRlLsxkJyfyzoSLcW9kQCrClq5Wd0xeHj6nY4o36rGt2+yk5ObK8S8yxMmlFdJAj6NPxNIVM1Q3BoBZWHO5FS72HD57byz/uX48QcPGtc+juMmdFuzsHqAj5DyF1/8BEDxU1/0cgLGh71xMWosq6EiX3BvCVotc9YfT9YX4W9cZnUOLuij5l/99fX4vcuxtRMAKcTuTunYjhBdxQMpqzUrLZ4+ngJHeqSWG4Fw5F5Yf5Y/hh/hgAtmxqpz8T2P+OB5tLMGKeiykXuCnf4KXlaJADH3jIm+okvdiGr1OjZldsvz1X8lddSgJ0rY8e070I9xyjDLoHImkhIKFv7FTYESnLkC2PRZ2C7Vuhh+yJMeMQqy41evZ0o2dPTJ5m7NfaAocPkZCs4LFFvt+6gHnbVyOA63OKeKBoMjV+L3+tLyeg61yaOZy/1pWHiV5G0B4memAEPxUrTDnfjdWhoAUl5RsibD0xx0JGiR3VImgsDWB1KjSWBqLJXreGPU7hv/9+Lv/87Qb2bapBSklNWSub3j/CjNOOHWD4ojBE9oYwhCH0gwr0j8iHCK3PRsm/A8vMw+hV96PXPoZ2+FvIzs0oI36NGj8V7cBl0PoOctwIxIYx5lOs+xhOPxPt9/fB/j1QWIS48GLweJBPPBzeTRYWoSan8It1W7nh5NmEVAWrphHsVydZFfBy2s417PK0k2axceuwkVx/aGt4u0fEKP/QjPKX8g1evO0ayXlWhs9wUn8gQFejsb+vI4DUByYNsV68B7bV4ukIMGleHqr6FbFBDWSoCyX/TmT7h6C6kT2lg0rxwxCoRPprkU0vgj064xLOSJQeghefBXcibDJe5nLzeqgoB4cDNhov8WR/ALc/QEcfAZB8odKmml8rdkWl0BnHd+MNEYMflO3kHw1GJLZbhLBL833trNPY+UonE89JwOZSGHdWHI2HgiAgc6SN7jaNis1eAl6JPSFGOPVESLpaYhNLkbYS2fIGpJyJfvQuZOUvAZCe7YQ8u0z7yuJqxNZRsU7T8ztsyDV9hD5qqw0p8gHw4807sSel8Pak8Yx1ublz+Dhy7U7uHzGJx+rKcKtWbs8fwzC7i5XprrBn5bbKdvxdAz8Lfo/5gdCCRpXvQHC4FdzZJ8LSI8bfZE1HpJ9vqG76SsFegEjukXAPtRhiFI5ilNQV6M3vxiZ7ZyyDx/9oPCcb10FKKuLcr0FrC2LG7MhCFcBpZFqPJsRxJCGeWfVNbC1t4PXl5xCvWliakk3GuldMpz/Y3WX6ua0quqy5+UiQtqo2csY7yJ3kwNMcomKLL5zZy51kp7UiyKGPupH9pkmH08aEqUXs3BLJSM5YVMiPVz0fFpm9/cLnWXblJF55ImJfce510/hCIWJ9R1REzvUoKWeh9N6XKRuR9U9C/GTD200oKPk/QNpzkVW/h1Ad+C3gt4KmwJYONO0XKKcsMsoEg0FD1fboEcTEKYhZJyE3rUf/3b2g68a3RAiQhmm9OO9CCs5fFc4YnZqUwRx3Kut61B1TLTa+mW1WThnoXRLolpR+3E3eFAdNpcZ9DIQkpZ90o1qhesfAFjuDPWNfGmKJHMkgJC1BKPFIzy5E0skoOTciPbsMAhioBVTDakZrQ4oYn82GdWjNTSjnrTLEcs5YhsjIRNd0xLSZAOhP/wX55isgJddPGceds/p4JPYEhiXwUE0pK1JzufzAxnDQ5I81pZyXFrFzOKUzOjCsBSXtNSFajgZxJCgUznHSWR9CC0n0kKR8Qzc5ExykFtpoKgvQeCg62NFRFyIh3YrDZeXyH8zlihmPc3CbUTb/2pPb+dWLFzL/7JGf7bP+H+JEmHGHMIQhnEhQHIb6Zn+E2tDLbkW4RqNX/Kwnagp6ze/BNQo8uyMlUJmtUSVNdHWhP/Ew7Owxxj2wDxkKQUK/hfGRUsjM5pKDZSysrmV3SjLDurpYsnIZzT1EKsvqYEtnC7s8RnS+KRTgmwe3MMblDouB7Hd00B9ChQPve/B3GQvUhoMBEISJXi/aqoKotgFUEPu8nKSU3HHRC7z/7F4ARoxL56EPLyMxdeCG7+OGlqeRLSAtqajjXkLbtRh0LxKjT0+d8Cb6rqVGaWcuSPRwGWd/yKoKqKsxj334LoyMEHinpnHv2s3cNH8WfotKlqebu05aSl0owMq9awn2rAxXpuVy6X6jJPSm3GKT5PUBeyezvdHZq4BH0lQWJOjTqd1jRLYzR9tRbXDww4iyYKy+yv5qnF8NYkfdZdWv0QCRshTZv+TWsx3i+pT3lEdL6kt6uOy4CTB6DLxvVnUUuo4cPRb274061qpLfvTBx9xx+U2m8Rtzi7kxtzhq/16MWhRP6acePM06VjsE+5SjZY+14+/Sqdoe6S9yZ1nImeCgpSJIsKfULC5VJT5dxepQyBxtO0H8p2JEbTpWo3esBtcklMkb0fdegKy6z9hmzUKduhOt9DvG/AfAmfSNLggEfLom0kcJ0NIMcfGIxWchX3sR/YN3EDNmo8xfCMB9k8fx05mT0BWFjG4vr9d2cmlmpPx8WUo2zzZVRX5ONVu7hAbIkmpBqNzmw5WqUvZJd9i+pGq7D9UKtXv9UUSvFxdffTpzVuRxcFsdMxaNoLXBY3IT0XVJzohk/vvv57LjkwrGzx7GGRdPiH2y/xBSxioB1pA1f0Cr+SMy/3aUrKvR9l/cUzaogr8KJetaQttm9ZAKYNsIxOaRCE1FKjpCrwAq0LduQvnBncg1HyDXfWz8ztXvIS6+Avnx6rDPXs/FRP752ovIZeci7Ma8pQrBOxMW8EpzDa2hACtSc0m3mec0e4IStirpj6DXIBXmPx7aBiiB7oUr8QTI7IlYlQc6suJOsOVimboZveb3aDsWGJssaagj/4zevgZZ+7AxFybNhvp+52lthk3N6Af2IX71O+Tdd0CVESDUs3MQ192CfOPl8O7f3baHw+543s/LQZGSqoR40+meb6oKEz2Adi1IosVCts1BbcBHSii2FczhjyJkNi5VJaPExpH1xgTYXqPRURei8CSnab++6NPayb7NtWGiB8ZX6tUntg2RvSEMYQhfEWx54Ns34Ga9+eUw0euF7FgL3kgkWGwvjhbdmD0X+knDU3oI5phLAQEYPRbWf0KOx0tCIIQ7GGSt18bTY0eiCsFlmcNZsutj0yFNoQA/zB9NzeFtlPo8jAxF92ZIlTDR60VHXRCbS5h6IJxJKvFpKtVt0ZHVvhV62z6qCBM9gLI9jbz4yBau+FF0eeKXhlAzeuU9Ecl4ABlEr/urQfQAmhIYLAUmJk9DrmmHrj6k3+lCjBlnkvu/9EAZZ5VXUZqYwOSmVuyTT2NyVg57pp/BmvZG4hQLlx/YGCZ+77TWsSo9jw2dRnnW08lHme5NwRKDdPo6NOr2RaKldXv9qBZMJWexkDHyBPBw60/k+kG2vI5IPMWcW1ITUfK+j77/UkAzelf6H5fgQP3Zg4iMTKTPh0xKjhgKC2F4IhYWI1e/h9y7K9J32YtQ9Ie3y9POU/XlJKhWrs0uJNtmVl50JalMWGoEZKSUNJcH8bZqJOZacWdakNIQAGmrDuFKVsO+VJPPddNaaSgLJuZYokSNTmh070BWPwiBPubzwTr0+j9D+3t9duwnyIJEJMSQmtdC6L++ywhkAXLLRnS/n+aqo9y1/DT0Ho/QBpeTXzg6+GefQx8qmUqa1c7ajiZmuVO5q2Cc6dTOJEFggMwrGFYywX5BkdbKoEkUpD9sdqtpDlv9QvT7YMTYdOafM4rFF40f8Dz/E0h9MCEmiV7xS0MIKdwfpqEf/ZmRbu4leu0uxIbR4XeR0M3zjP7x6qh3kvzg3YHLsMF4hnTzO8SmKOFMeCyULHCx582umHOXM0khMddCcz+F4fQiK02HAzEvJWmYhfTiE2Cek4Pco0A1es1D6JX3RsZCTYb/a3NE6E3agwO/iTraka+9ECZ6ANTWGP68fXDZonk8X1wQ8xRpFhuT45OixrNtTnZPP4O3W+qokt3mEtMY8DRrNNCvdLlVD2dkY8HSp3I6ISmaGCckG20qRw80YbGqZOa5efnxbezbXMO0UwpYcsmEL2zeHCJ7QxjCEMwQg71EFJT0r6HVPkHfXgiRMBvs+cjODcZAZ7RUt1i2wuj96s3sARSMQFlxAfqu7RFiMWc+7NnJpoxULlq8gJr4OEa1tvN8fS23n7k8fOiSlCwOVEdm6PEuN6clZ7Jn+hkc9Xfz0kfVUdcQQsfhUE0Zofg0C2lFNko/MqLf9gSFhAwVZ2LsrJdqjYy31EVnQJtqY2RFv2xYU6OGRMI0ZMurEGqDtrho5UAkIicPpkyD/ALEkqXIp/9iLGyEQHztYsS8kw0/t3WfQKeROU31B0htMLJ1VTu28FtPI7UBL6sy8tnf3REmegBBKSl0xLEybRgvNVfjSwkRGqmRW+ugqSwi0W9ziSj5cTBKa4RiLqVNLbTSUR8k6IWMYhs542P79n2pCDUfcxeRfS3SXwW+w6DGI3KuR9iyUKduRba9j9RjeLEF/YgMo+ePlibErT+CNe+DxwMnLUC2tyN270AsWoLMGx5N9vphl6eduds+wN/zgf6t/ig7pp0eVrGLumYhSCu0QaF5LHucg2wzB0G1CtJGnAAL0v8UWqzn+DMsvOafAh9/YGT0ANxuKB4Jzzxl2k1+uoZmixJVnr7Tbc5KuC1WHiyOeHfqUrK1s5UMm51hdhe7uzvIw3xMXyQNs9B4OGB6ZlwpFuzxhtXCZ8GCFaNZfs0UXnvSKNs8++opzF12fDMSsn9PcRQ0pK+s35iO1Pqs2luj5zkT0tLAZgN/HxsOVxxizlzk35+MfYzdYZSzHwP1AR8pFhtWRSEu1cL0VYl0NWlYnYLavX7aqoK4klWGz3TiSFDobtVpOOhHtQrypjpIyLAyalEc1Tv8aEFJxkgbyXlWdE3iSDgBsnoAoabBt2tdRGXQA42mMZPJfX8IBRwxbD+SkiE51cgAAi8U9bPvkZLFyVmk2+zcNmwUBY44Hq4pZWdPJVCBw8Uwu5Nz93xKQNeZ0xX9voy6FAG2OIGnz9QuFIhPU6mPsb/fF+ClJ3cSxMfC88cw8aQ8ll8zhZcfN9Y/ialOLrx5Ft9Z+gzr3jK8YHOLkqkuNYJ3r/9lB9VlrVxz5yA2Vp8DQ2RvCEMYghne3TEGBTiLUfN/gpK4AEY9iVZ+OwSbEakr0Lu2QajVaLhuXxOzHUa+9TrKVdehP/I7o8wsbziMHY/cvxfxiwcQpYeQ3d3IvzwCwSDLrvwanT39YAeSEzk1XqOij+3B93qMbN9sqWWsy82lmcP5Sflu8u0uLs4YTqnSxWzMtfgWXSFzjI3GwwECHkl8morDrSI1mHxBAp4mjQPve6jaPnCvRJ9iOmYvKSIpzUVbk1HGoSiCM77+xZYzfW7Y8xGZV0GwOWyVIRJmI0OdKNldpPTrAAAgAElEQVTXo7e+C12xX9LKN76F/qufGb5gQsA55yFy84y+yk3rYe8uxNnnIS65Cv2ylaZjQ0KwxOqltMZ4cT3fVM3NOdHlgaNcbm4fPtY8WADDpjhoKjUWn2lFNgIePeo+JOdbiU+3cHSTl6BPklZoZcRJLhTLiZY1GtweQiTMRElbiZJ2PrJrK9qh65GV96BV3oNwz0ed8Caa/k70cT6JXlGOfOyPUHYYVAvi/AsRKy9Cv/P7UF9nPHojiuGU0495lX+tKw8TPYAKfzdvttRx/iBZiv8TUFyI/NuR3buhl1DYclCyrkLv3o9sfTPmYQIBWzeh/Pw+5AfvIA8dgNQ0CGmgWsx1Xe5E1MbaqHPUOB3UBrxk25zUB3x81N7IOFciY+Pc1AV8nLnrY/Z2d6AAt+WNYlZbtFWARGJzKuSMt5OSb6NwtuToJi9aEBIyVbLH2VBtAkeCQkd9kPaawWX8FUXwg0eWcf0vTgX4UsrUZax+sL6Im4SSeTl6xyeRMftwlGHfRWt+2bhvg00L+QUoZ56DdMUhe4m4xYKychVi4hRkRhZy907kxx+At0+VhLcbOjuoc9hJslhx9FPyqvF7uXDfejZ2tpBmsfFQyVSW9/SHSSlRLAoj5kR/fgUznRTMNBObxGwridlGekjXJW2VQXQNrA4F1XoCzHnSO/A2ayZK3q3oHR9DVyTAq2Zfi25JjjxD3gECQhYLYuVFiJNPQ3/3TejoEVSKi0eZfwrMmYt85QXk4QPRyw0h2N7Zwj1FkxgT52ZjRwsT4hJJt9o5LTmD05MyOWn7B+FA5OTu2L6rFkdEfCpjlA1XskpXk2aUpwsj0NjdqpE0zBLlZfnog69Sdsgo5/737zZy78sX8oNHlnH21VOor2hn1uIiPnhuX5joAWGi14uXH982RPaGMIQhHC/EfvFbp+9BO3oXwXVZhhJn/o8R6SsJbRoTjvBJFEj+GsSXQVu/Msp334C581F//HP0+lrkj2+DN14xJup33kDc/Rt45PcQDNJis4WJXi8arCq/qTrIstRsLtu/kR2edsa53DwzZjb1AR9n7/4UrWfaf7axipGeGBHDkNH0nl5iJSXfysEPu+lqMv7etCKrobx1jGC3FpBYei4tPtHBo59cwTMPbMDT4Wf5NVOYMOcrWignnYviHo9e9xf0naeAJQ2l+CGEmoB26JvQud6gqe75SC32QkH/99PGYgaMpoI3X4Ubb0U+9KAxBMgdWxGXfyPq2A2ZaZRazQufMp+Hc1JzeKXZeOktT81hRVpO1LFgLGBcySpNZUHq9vrJGmunaL6L6p0+pCbJGGkPL3xSC61IHRT1BFjwxETsWlMl7/vIQAOybY0heJR1BdgLwRMpjZUdH6M3vUhMWXlAPvsPg+gBaCHks8+Az2t4Vfai7DDER5cx94fbEq3QmGD5P7wsEA5E+oVIfzn69jngKETk3oKwZqG3vo22ZYJRbpv1DfSml2Of419/Qy441VCyLTeIolzzPmLBQuTqD4y0dGISDC8kdGiv8Zz1KdXyWlSu2L+JH+aPZsWetXh76vh+XjCehqAv7OemA7+uPMDbenbUJQhgygVuKrf62PZ8B44EhdGL47G5BLV7Amx/oRMhIHeig4I5dnY8fwxi1YO+JO/I3kYObK1l8oLhZOUn4vUEePvp3TTXd3HaBWMpGD2wX91nQ+xnSKSuAMWB9JaiV9yNSD7T8BG1piADDWibx4J7Jkr2dehHfx11vLQHUG9/AL2yAv03vzD6KW/4L2NbW6vRVx4KIZacjXL5NWjvmol9g8PORWVbWdfZQpLFygNFk/l6Rn54+x3lu9nYU6beFApw7cEtzCON8vd8RjmtgOHTHWSPO3Z2sHKbl/r9AYQqURQRFkqyxQkmLEvA6jwRVFqioRTegwy1GYb3zpGIhDnGGkFKtCM/BMWJyLgUYUlCaoejT1BQhPj2d+DVF9F/+VOYPBWRlIIM+KG8DP3WG2BEMco1NyJfeYFUn59mp/nzbNCCXHNwMxlWO+ftXUegJ6i1ubOVkC5NFScuPTpTKhDYXIKieU6ajwSp328sDCwOKF7gpLNRo35f7MVCTWVTmOiBQdRf+NMWpIT7b36LhqoOFl04jrTswedod/IXV6Xyf3hWH8IQhvDZIdHqn0aviEhWa4evR9Ha+pVy6IYtQ2vsyK88Wo4oGQ0fvgfdfXx3qith++bwoqfLErtM5e8NR3mjpZYdPeUYe7o7uPbgZnJszjDRA1jd3ogvTuPi1oKY52k8HMTv0U2lTU2lQdKKju3B1v8Fm1eSyvceOuuYxx13aF3IttUQ6JHSDzWhV96DSFxo8tqTHR+DOyNaQAego5/RcyCAXP+JeayrCw5EC4Bken3RYzYH8xLTKLS7ODU5gyUp0QvTXrRWBTnwXuQ70VIRZMI58XTUGWVoldt8+Dp1RpzkRAiBOEEqmT4PhHs++t4VEXGjqvsRKWdH7Sc71sY8XiIRwX5ZQ6kjW2KZGx+7RO/a7EKeqi+n0m9E6E9OTGdRUuYxj/tfCxlCKHajOgHAV4ps+AdkXhY2VUdrR69/EkTsbACaBts2hYleeMznR3ngYWiog5LRyLdfpyrOZSJ6vVjd3oi/XAsTPYC7K/ZxWpK5UkECQXTsUYJAguodPmp3G5lxf6eOt91D/nQndT2eYRKo3OpD8vl9J//1uw08+F9G5tliVbj7X+fz1K/WsmeDUTr/1C8/4eE1VzB2RuzAzmdCINp0HkBkXoZ+8GoIGVkQ6S9HGXEvsvlV6Ojp42591wg8Horx+/1WZGMjPPbH8JDcvwfxnR/BQw9Exv75FHpGZpT88t0zJrKuh8y1hYLccGgrZ6VkkdQTAdztMQuDtWtByrd5CfZOjxIqNvtoqwkSn2ohe7wDi03g7dBoKg1gsQnSi22014T6qXFG3m8Bj6ThUIDciSdAyXos6F5k5T09P6xDxk1CzfsB2v6LwrvIhr+jjHsNyR+jj6+tQv7tz7C9xwKlohzmnmysGXrFpw7sQ3/4QWioY6FlIs+NGhF1Gh14vO5ImOiBcT8ag+Z31WFHJ8SId3S36IZYWJ++vJDP8BptLh94frXaoqmV1aZyx0XP4/UY53rr77tYfs0UVFWEPS1Vi4Ku6UhpPFffvGvhgL/j8+LEDAsMYQhD+OqgxjBKBUMyvv+Yvy56x4TJ4IlB9oRAjJ9o/DuWGp9QEEuWgtVGfrc3qgkeINPqYGuXeRGwrasNuxI9lZ1hiVYzNCGGOFpKgfWYJOKE1ZkI1iO9B8xj/gpiRsjrUmL2soiTTzMPTJiMyMmN2o8Zs6OGits7ubEx4l2Wb3dxyNvJFQc28duaw6zcu441bY0DXn5/6Wpvm07tLr8xLgFp7GMysv3/DLLtw2hxI0SPQXQPVDcifgqxMnsSgZjbr6wnMQlx9gpzb0tSsuHvdgxk25zsmLaYf4yexavj5/LmhPkoJ+wX/MtACL2377gXwQbo7Nf7KINmAaS+SE2DYfnR43Y7Ij0DMW4iwmZDzJ5LozP2Yn2kM54Ozfw98ekaS/spcebanFgHqFVsrTQ/JwGPjGly39U4iCFiDISCGo/duabPzzq/u+3dMNEDCPg1nn9o0+c6bxTssYMOsvn1MNHrhd7yLrLdLNgl29dAa3Qvo0BECXzg80H/MUBui/4b9iWbxT68usYhbyevNtfwt/qjLEg0ZzQLHXHYfOaXipTQXq1RvdPPodUeuts0dr3SSfUOP0c3+dj9ehftdYMrUWmhE8CrcgDoTS+ZBzw70Fte67eXRDYPkB0PabBjq3nvrZtMfpWAUcEQCPD8yNh+dRYhmBgXLZg0PzGd67OLUHvegjW2gctRY9lmBH06YhBF4fTMJKafNDr8syvexuwlxWGi14u6o+3c+8oq5iwpZsE5I/nj+5fyr3038LOnz+X5w99mwfJBrHc+J4bI3hCGMAQzlFiRXgWRujRqVM38OiLj65EB13hIPjX2ecdOgOYmpJSIhYvNZWb5BTB5KqSkwaIlkJhIcbs5QqrqOv9dMI5TktJN4ycnpnNL7khcfXonVqYNI6N54Khn5iibUUbTZ75OzrOQkmdj4tkJ5E60M3xG7ON9nYP3t3xlsOYjUpeZhkTSqai5N5nIhHDPhe4B+iS6OhHnXYiYNRdx/irE1ddBTq5xf8BgujNmw7YtMQ+/d8cBdkw7nbcmzOf5sXNY0x7J+gal5A81MUp2ehBLkCUY40XrHUDC/MTCALYdmZfS/7WrJC/CMmk1IutqlOzrsEz+GJF+IbEaXwUgX34OFp5ukLmJU6BkFPLZZxBfvxxx9rlw6mJISoKXnvtMVxqnWjg/fRiLk7NQ/08TPQNKYj91YFsOIuUc85glGRhAiGnUOERHO0yfFRmLi0ecFTmH3Lkd/anHqYiPIT4hJXcOH8tVWeYF7DmpOVyVVciTo2ZwenImF6bn8V/DSmIq2UokcalmgqFYjDmuP9w5n++eh4I6vm7zotXniSYmFuv/MPXuORJzWElbESUipsRPQsSbff5E/HSwRWdfJNLoae2PsdGqomLS1KixxRVmS5ocm4MflO1i5d51XHNwM/9sqOCqzAKKHHEsSc7ixXEnkVY0sFBRe02I2r1+k1Knr0MftERdsXBiqHEOAOHol2VTExCJ0QrVIn1l1BhgqKFm9gvW5uRCST/yUzACdBlLIoBhNidPjZrJTbklTOmjxnl6ciZLU7N5sHgyFbOXsmHKaczujl1y7EhUyBprw5lkfsbSi23kToxlOxHB169axB1/vIDbHz+bZw/eyKKvjcUVb75nE+flMWdJMfe/fhG/evFCJs3LJ68kldNXjSdj2CDCNf8BjmsZpxBiCfBbDNOhx6WU9/TbbgeeAqYBzcCFUsry43lNQxjCEI6BYCzxDhtChhC5tyLrHgXFjkhZit6xAbXwHhj2XWSgEb1rM/LovQa50PuRxj070ffsRMw/BeW6mxF3/Bz5wr/B5TLM1b1e9B9/F9pa8VhUDidHyyV/r2wns92pWFIE6ztbGO1MYEZCCg1BHzumnc7rLbWsaWvk/dZ6xoeiJ0sdScYImzGBuy2MXxpPzW4/6JA13piIHYkKtjhlQJ+j/pm/hqoOnntoE56OAGdfOZnR0wYuVTyu6FwPrnMQWd9Edq5D2HLAkore/BrqxA+QbR8g/UeR7etAiSZMAoF8+XkQCsptt4Omod92k1EOqKqIVZdCUjLykT8gYznLA0gY7XIz2uVmf3cMn8NBLj97vN0kE59eYiO92G7YL/S+zQUk5f7/0H3gAjxRo/rhGxCZlyFb3jIEjZJPR+peEBYsJQ8bPUcHr0a2voNQF4FmXhwIgNpqqK9F3HUf8lc/hQ7jc5ZbNiK+fyf89VGzz9sQPhek4kKknmt4hjoKEe65SK0TkXOTUd1gTUM4hiMb/xX7BGvXoK9dA8tWoHz/TvRPPoSaavRnn0FZuQr8fsOGQerkFg+PeYqL929kclwSfyiewoaOZlpDQTQpub/qIN/KKWaMK4HFOz/iX42VvM8pMc+hWARxaSqeJg2LXZCUZ6WzQSNrjI2WiiCKKkjIVGmvMxO3UFDjhWc+4sf/dYjMPDc337+YmYtGUHOklVee2IYQgpNXjOKD5yJZlq/dNJNN7x9h8/sGQXMl2Ljg2zM+/4ffF96jMYdl916UvB8a/q6hFqMHWU1AZF5uePP1+lUmzoW6bTHPQSgIJaPh0H5QVZg5B9raEEuWIT/60Ci7nTQF9kaLld2yYy+eiZN5MTWBQkccy1NzueFwJAvVEAqQbLWxd8aSyEFjDX2eloog3a0aAU+EnigWY1t/uLNVhGKnfr8f1SLIGmtHStBDkvRiG073iVvHLuImIbu2gv+o0Z+XeRky1GHco8Z/GwJImVcguwa4PxKYMh0+/cgQZklONYLFDofRn3xwvxGAHJZvLpfug6qAl19U7GOUK4G1k0/lzvI9rG5vJM/upMrvJc/u4ntlO/lHQwW3B0dHHS+RxKeqdDZojDkjnro9fjwtGjaXwOpUSBluIz7DQsPBAAGPHuWVKIRg1IRhFM2NZJd/8exKHvzOO9RXtHP6qvFccttJ/+lH/LkhpDw+qWAhhAocBE4HqoBNwEVSyr199rkBmCilvE4IsQo4V0p54WDnnT59uty8efNxueYhDGEIEPx4kIhh3ETUSR+j7z4L2dHjT2RJwjLpE7SjP0E2vQCAeOo0RPfAmTXx458jf3dvRGGrYARiwanIpx4HoFtVSbtm1YA1k7fkljDHncpF+9aHqzG/kT2CafHJfPOQkXV6+Og0xoaiCSMYL9gJyxKo2+8PN14DFC9w4e/Sqdwa3X/WiwnL44lLNt7OXk+AVWMfpqHKWHBbbSqPfXolo6YeX8I32D0SKWeh5NyAtvtswizJNR515J/Rts8yxnYPQ/lk0oDnYNJUaKg3iEUvUlKhqAQ2rR/4uPNXoZ4XmcJX7l3Lq82G4qBNKLw5YT7zEgcWbtCCkvbaEDaXID7N+IxbKoLU7jbuR/Y4OynDT9yIdi+CawtAG5hwqaOfQe/ahqzqFZBQUEc/hd78OrLxGWNoWyHKhrEDnkOcfiayn3gEJ82HtR/HPgBAVVGf+mwZv//tGHSec41FnfAu2rYZEOi5j5ZU1Cmb0PYuB88uAMSfzhpY2t9mg0uugj//KTKWlGxYy7z5CgAb0lNZeP6ZA17GVVkF2ITCn2rLTGOtwSAvNhvP5oelC1EGKNJSbYJxZ8ZRtrY7Uq4pYPRpcbTXhcI9fX3xzqubePOlyDPuSrDxxIar+cbcJ+lsNZ5Dd4qDS247ierSVqafWsiiC8cRCmqseekALQ0eTl4xiozc/1lmIrjvv6Hp7gG2CpSRT4BiQ99/Ob2iYkrWNxC5N6HtONmwP/nTEpSofsYe2O2Im79nlAe+12NzIgRcf4tBKPret/6YuwC1R9TlteYazt+7zrT5+uwik11GX3Q1htj3rgctYIi1FMx0kpxnZddrnWHlx/h0lXFnxg9aKngiYNBnyDYMddRf0Gr+AM09ZZ3ChjL+dWT3AWTpt4yhwZ6hkWMQl12N/O2vobHBGEtJRfz8PuQv7gz77yVcexGaGvs+T4lP4ts5xVx1MMIbChwufl4wnkv2G+XZP68cz/zAwL3KIxe68HskRzdGyj2Hz3RidQoOrxlY3CghE8adGXsN8kVBCLFFSjn9WPsdzxDpTOCwlLKs54L+CSwH+nb2Lwd+2vPv54A/CCGEPF4MdAhDGMJngCCmdwKAZyey6jcRogcQakOr/i2y6cXIWEYblA/cMyc/XRMhemBE50oi0TW7ruMOBOmwx36ZvNRczdqOZlPb3RO1R2hPj0SpBxNq1ENQt89Pw0FzmU/tHj/6MXohepupAda+cThM9ACCAY03ntp53MneYJAtb6IJG6Z72L3bMITuHRtVg/xk4sAvWasV2s19MXS0g33w0hWlX8/fP8fM5uWmGsr9Hs5OyWGka3D1MdUqSMk3Z4RT8q1RYyc8UpdAw58H3Kw3vWB4HkZG0Cp/BX3l5icfQTYmQZnxHIn+C/qSUdCf7GVmQ0amQdRjYU50KdUQYqB7L3rVfRGiBxBqRq9+IEz0AFBDoA3y3ezvc9jWitAjUlLjW9uwB4P4rbHPcaC7M+wN1oun6ys4KfHYvmBgKAc3HAqY+/Ik1B/w0zFAT1jpQbM/aXdngGd/vzFM9AA6WnzYHBZ+8EikbNxiVTntgoGDE58bg1bLS0PcyJJk2lGvexyhOPv4XOowENnz+5Gb18NHH/Y5rYTXXoQYPeAmpEb62k9PzqTEGc8hr1HW61AUrsoqGPDQ+HQLUy9w09kYwpmoYo8zftekFQm0VgRRrYLkfOsJT/SOiUAVsvW9CNEDkAFk9W+Rnp2f7RwH9yE3rosQPYCWZuSrL5qM1vO6PJQnxg4ubOtq49+Nlaaxcl83H/bpH/eqg7dmNBwK4Gky71O900d86uDZVXECqUUfz569XKDvJ1zVMxZzHyllCGgHomYxIcQ3hBCbhRCbGxsHbvAfwhCG8AUgcXB/LiljqVBJUCPlCnLmAaTT17PF+C+MUWMQ8TFMgItKjLp8QJWSX6/fjtIr0tIv/lPkiCdWpf7UPrX5jycfiblPGAPMfseSs05IjcTIEmJII8ca+8LhmjLwNkcRIpapelyfTJ5VR56yA6mEjLuj9inlstpQzlqOmGdWAhNzT0ZZugJcA3hsnX4WItWctbMIhfPTh3HrsFHHJHr/m6AW3jX4Do4RIPstMPQgInFu5GcBcvE25LVvI695CzmuHCwWw/T5kqtQ5p6MOGVRZP+84ShnLEX51q1GiZMQUDQShuUZdWLTZ6FcdvUX9Sf+r4ewxFDbtCTTlzzISz5ExnuQ6OZnCBBnLIsWN1IUOP1MGG2QoriQxl/W7yC928gY2HXzfLUsNYdMm3k+ybQ5uLpPP98+1UwGoy7ZFr3gVCxiQJ+2giJzkM7hsjKsKCVqP2fccc6wl3x38O0yRDQjlJiUtyYP3CMMGM9F/1eElOA8ho/gOeeG/2lXVNZMOoW7C8Zz27BRrJ98GhPjB8/mqFZBUo41TPTAsJ7JGGkntdB2AlvK9EeMntO+iLVWkCGTSJVMa47epy9iEW+H3XA078H3tu0Z8PCTE9PJd8SZT4nRA9t7hj+lHR50rWCxif5LEKRu+PANhvwZx9+P8rPi/4fmB6SUjwKPglHG+RVfzhCG8L8alvHPEdpQ2Cc62gfOEpRht6G1vAG90TnFiZr9TXTXOPSy7wASUrzIiz9E1idDQggl76+InXWQlm4sUNtake+9HfF0y85BzJqDmDnHiLbqksunzWDRwX3s8nXSmJfHrUf30KmFyLM7+dWIiRzs7uTi/RvCU/TlWcP5dm4J5b5unqw/QkVKN93BEHGt0VFzi12QPdZuRLn7lHFmj7djj1PY/15PmU0/FMx2hE3dAaafWsisxUVseKfUOL4gifOumxZ13BcNy6T30PZ/HdmxHmzZEGyBUANY0lBL/oiw5RJqfgWCRnBMZFyCmn01eA+h1/wBkDBrJHLCLqNHr+BelNoCZEM9Ysp0RHoGFI+EjAzD9L54FOLMZQiLFeU3DyG3b0GkpCFHFCP27oKMTESviMsQUGzpyNFPox++BUId4CwE7yFAQ7hPQs27FV3rQK99OHyMmnszIu1cNK0L2fI2OEsg2GxYaagpKDc8gIifC4qC6BEjUq69Ebl0BXi6oKgEoSiQ4Eb95QOGENKQ4MqAEFN3I7dGi3IAiIxLUXK/hd7wD/DuNwbtBag516Mj0St+DkhwBpGXrI4cmPA0ylGBKBiBmDQV2dGOPLDPUA20WhEXfB0lOxfuuBt59AjYbJyXncs5UqcjFKIx6OfO8j0c8Xk4Ly2XW3JLKHbGc8m+Dfiljk0o/KpwAuel/7/27jxMrqpO4/j3re6ks5KQBcyCBDAoBCOySRQUBFkFRVE2nQedGYQRGRdGXBhEHxcEEcQNURYFBMEJEDAPGAKBsJMASQiykyAIJCQdSHe2Xn7zxzlFKp3qTndIpyuV9/M89fStc5c69/76Lueec88dzaDaXtzw+svUv3MV3FlmJYC+gwuMGNeHFW+28vrzqTBa0wtG7lLHslG9eO7eZWsVdg44dHeWNS/l/ilPM3zUQL5+4cHstt8Ybr7sUV54Ij3Pvf0uwznwmHHrv/E7oVddfzrqd7dm9NegZgta3nyA4kpoq+OoGXUqzQuuhubFsPfzxKwdUKSC6RqvmxkwgMIhHyck4u+re5rWYUeiYVvReu7T5V9fst0O1LQpDA7tVcfp22y4nhM3GXs+CQ+Xf+6UXsMpjP5vYvmTJT1511AYdSrROJfWF85ISUc9SEz8ICwaxFtxLN5Q2W1PCod9gtbp02BRrugZvCWFjx1GRO6sKoITn3uRXRYt4e5RW7PtqhauP/wQ7l2+lD0HDuFXY9ON0TuWLODZ5Q0UgO+8cycOGfIOrtt5Ar946RkkeH3lCoYt6LNWa5faOjHyvX3oO7hpjcc7RoyrY+iYXrzxr2aalq99rdB/mBiwZeW0SOnOZ/YmAGdHxMH5+7cBIuInJdPclqe5X1It8CowvKNmnH5mz2zjiMY5qdezmn60LrgGCv0pbH1CehFq81JaF1wFTYsobHUM6js2zbP8OaJxDhq0D9QMhKYFqG6b8stfuIC45y7o2wftuz/qX6a2r0RDSzMvrGhkp34Dqc139R5ZWs9t9a+yc78tOGLoyLe6jS+90G1a0UrTyqBP/wKLX2yieVUwZNte9O5bICJY8lIzy+pbGDyqlv651q6lKdZoZrNqWSu9+qhs05qI4JFp82l8cyUfOHgH6vps/HtoEc2pMNFne1RITS2j+U2ifgrUjaKwxepXJURLA0Sg2s2npq0SxKqF0LJk9b4SrcTrE4nGWWjwgRQGf2QdS7ANLSJoWXI36j2CQt0IYsnt0HskhS1ST5rR0kgsuhGiBQ39JKpNTcVi+fPEyvloiwnEihdobXySmmFHonbe2xKvvAxbDFrnMa49C1etZGZDPe8fMHitmr7ierz5ajOtLcEWW9e+1VnE4NG9KNQojX+lmZWNrQwenY59ACvebGHpwhYGDq+hd/8CKxpa6DuwBhVEa2t6kXfRyhXN3Pe3dGH8wcPH0rtu4xznmuYcB0tuga3+k5qh+xANs9GWB77Va2rr0oeJxZNR33ej4Ucj1RIr/0Xrwuugpj+tvd8FU/4Dlm5F4fib0ewnobEB7bk3GjSYiIAZDxIvzkfj35feAwtE/WLiiTlo5GgYNZrWe++C8e+nZujwjrK72WlZ8Qqtjx8GLQ0w9hIKy2dDa1O6Vug9gmhdRSy8llgxj8LQI9GA9Cxja/3txBvT0cA9aVl4CyyaCMOPp3bomamzqSFDYbc9UKGGaFhK3Hs3tLaiD+6LBqWa03jtFXh9Iey4E9QvhiX16bKvK70AAA9jSURBVKZXmef3WiJ4pKGeEb37MLqufI1b04pWVjW20m/LGhoXt7CyoZVBI3u9VTte/1ITDQuaGbBVLVuOTgW51ua07/UeUKBuoKif10z/oTX0HbxxOtDp7DN73VnYqyV10HIA8DKpg5bjI2JuyTRfBt5b0kHLpyLisx0t14U9MzMzMzPbnPV4By0R0SzpVOA2UiP3yyJirqQfADMiYhJwKXClpGeBxcCx3ZUfMzMzMzOzzUm31sNHxGRgcpu0s0qGVwCf6c48mJmZmZmZbY66szdOMzMzMzMz6yEu7JmZmZmZmVUhF/bMzMzMzMyqkAt7ZmZmZmZmVciFPTMzMzMzsyrkwp6ZmZmZmVkVcmHPzMzMzMysCrmwZ2ZmZmZmVoVc2DMzMzMzM6tCioiezkOXSFoIzO/pfGxAw4DXezoT1iHHqPI5RpXN8al8jlHlc4wqm+NT+aotRttGxPB1TbTJFfaqjaQZEbFHT+fD2ucYVT7HqLI5PpXPMap8jlFlc3wq3+YaIzfjNDMzMzMzq0Iu7JmZmZmZmVUhF/Z63iU9nQFbJ8eo8jlGlc3xqXyOUeVzjCqb41P5NssY+Zk9MzMzMzOzKuSaPTMzMzMzsyrkwp6ZmZmZmVkVcmFvPUh6h6RrJT0naaakyZJOknRLD+ZpmqTNrjvZzpDUIukxSY9LulnS4LexrB9IOnBD5s/WVBKv4mfMBljmPEnD3n7urC1JIemqku+1khYWj4eSjpT0rTx8tqTT87CPWRuApK0l/VnS8/l8dL+ko3o6X7a27ji2VbOS7TVX0ixJ35DU4XWrpDGSjt9YedzQeWhzvXK9pH4bOn+V6G3E+vE8vIekizZObjc9Lux1kSQBNwDTImKHiNgd+Dawdc/mzDqwPCJ2jYhdgMXAl9d3QRFxVkTcvuGyZmUU41X8zCsdKam2h/Jl5TUCu0jqm79/DHi5ODIiJkXEOT2SsyqXz0c3AndHxPb5fHQsMLrNdN2yz3hf7DIf27qmuL3GkY4rhwLfW8c8Y4AuFbS6Ybt3OQ8lSq9XVgEnb7BctVFh/2/rE+u3RMSMiDit23K3iXNhr+v2B5oi4uJiQkTMAqYDAyT9VdKTkq7OJ2IknSXp4Xyn5pKS9GmSfirpIUlPS9o3p58oaaKkWyU9I+nc4m9JOijfuX0k3/UZsFHXftN3PzCq+EXS/+TYzJb0/ZL0/5X0lKR7JF1TUhtxhaSj8/ABkh6VNEfSZZLqcvo8Sd/PMZoj6T0beR2rTt4nJkm6A5ia09aKnaT+kv6W7ww+LumYksV8xTHpNpOBw/PwccA1xRE5dr9qb0ZJhbxf/bCb81iNPgqsanM+mh8Rv2y7zyg5L+8Xc0r3DUln5LRZks7JaTvkc9BMSdOL+0yO1cWSHgTOzeeo4XlcQdKzxe+2bp2Nk1KrkmKN4MuSLs/pn8vXEI9J+p2kmpzeIOlHOaYPSNqkb0hHxALgJODUvI1q8nYqngO+lCc9B9g3b4+vtTedpP3y//Uk4Imc1t55v6N94SJJ9ynVrB/dTh7GlcRotqSxnVzt6cC78m/dmH9/rqSTihPkOF+Q06eW7Iud2n/XNx7dqQuxfkuOZ7E1yQBJl+f9Z7akT+f030qakbdV6fXeOZKeyNP+LKd9Ju+DsyTdndM6+l+apjLX/xUjIvzpwgc4DbigTPp+wBukO6oFUqFinzxuSMl0VwJH5OFpwPl5+DDg9jx8IvA8MAjoA8wHtgGGAXcD/fN0ZwBnlSxrj57ePpX4ARry3xrgeuCQ/P0gUje8yjG7BfgwsCfwWN72A4FngNPzPFcAR+dx/wR2zOl/Ar6ah+cBX8nD/wX8oae3wab0AVry9n8MuCGnnQi8VNyXOojdp4HflyxrkGPS7fFqAMYDf837xWP5eHhLSex+lYfPLtmXpgF7kwqG3+3p9dgUP+2dj0q2e+k+82lgSj4Obg28CIwg3UG/D+iXpytOPxUYm4c/ANyRh6/I+1tN/v69kmPfQcD/9fR2qdRPJ49tZeNUsozBwBxgd2An4GagVx73G+Df8nCw+lrjXODMnl7/9dheDWXSluTtclJxnYA6YAawXemxJ4/raLpGYLs8rqPzfkf7wvWkc9DOwLM5vW0efgmckId7A33Xtc5ALXATcEr+Xvz/6As8DgwtiXNx2Wex+ljbqf23Uj7rGesxwONttznwU+DCkuVs2WYb1pDOP+OBocBTrH47weD8dw4wqk1aR/9LZa//K+VTSVW41eChiHgJQNJjpH/Ee4D9JX0T6AcMAeaSDtAAE/PfmXn6oqkR8UZe1hPAtqSD/M7AvfmmQW/SP5V1rG+OxyjgH6QTKaQLk4OAR/P3AcBY0oH+pohYAayQdDNrezfwQkQ8nb//kdQ89ML8vTSun9qA67I5WB4Ru5ZJnxIRi/Nwe7GbDpwv6aekA//0kvkdk24SEbOVnj86jlTL11m/A66LiB91R742N5J+DexDav71a9bcZ/YBromIFuA1SXeRLnA/AlweEcsAImKxUouRDwLXl9ygriv5qevzcgAuI12UXgh8Ebi8u9avCnTm2NZenCbl2oKrgJ9HxExJp5IKfQ/nOPUFFuTlrCJd1EM65n2sW9ao5xwEjC+pTRtEOges6sJ0D0XECzn9Q5Q573diX7gxIlqBJzqoPb0f+K6k0cDEiHimg/UqXq9AOp9dmodP0+pncbfJ67AIaAX+ktOvAiZ2cf/dFLQXw6fbmf5AUnN2ACKiPg9+NteK1pJudO1MqtVdAVyaawaL+8y9wBWSrmP1tcO6/pfKXf9XBBf2um4uqWannJUlwy1AraQ+pLtte0TEPyWdTbpz1HaeFtaMx1rLItViTImI49Y/+5ul5RGxq9KDzreRCmUXkbbnTyLid6UTS/rqBvjN9uJq66+xZLhs7AAk7UaqKf+hpKkR8YM8yjHpXpOAn5Hucg7t5Dz3kW6GnZ8vsqxr5pJqggCIiC8rdUQ0Iyc1lp1r3QrAknYKJmssN5/XXpP0UWAv4IT1/M3NWWfjdDbwUkQUC9QC/hgR3y4zbVPkKgiq5JgnaXvSuiwgrftXIuK2NtPs13a2DqbrzHZf175Qeq1WtuleRPw5N5s8HJgs6UsRcUc7y1vrhkDO64HAhIhYJmkaa15HrvFzncjz+h4XNppOxnpMF5a3HXA6sGdE1Eu6AugTEc2S9gIOIF3bnwp8NCJOlvQBUsxmStq9g3zsR/lr9orhZ/a67g6grk2b6fHAvu1MX9whX893W9orKHbGA8CHJBXbcPeXtOPbWN5mJd+5Pg34htKDybcBX8xxQdIoSVuR7ugcIalPHvfxMot7ChhTjAXweeCubl8JKyobO0kjgWURcRVwHrBbT2ZyM3MZ8P2ImNOFeS4l1QRep8rqLGBTcQfQR9IpJWnt9d43HTgmP3cynNTs+SFSS4cv5JthSBoSEW8CL0j6TE6TpPd1kI8/kGoVNrUag0pUNk6SjiBd8Jd2QjEVODqft5A0RNK2Gz3HG0HeFheTmikG6RxwiqReefyOkvoDS0mtc4ram66tsuf99dgXaJuHXHB5PiIuItWCj+/i6g8C6nNB7z2k5u9FBVZfVx4P3LOeea4YXYh1e6ZQ0hGfpC2BLUiF3DdyDeyhedwA0uMek4GvAe/L6TtExIMRcRawkFSb2tV8VAyfXLsoIiJXpV8o6QxS9e88Uo9o5aZfIun3pDbWrwIPv43fXijpROAa5c5AgDNpvyrb2oiIRyXNBo6LiCsl7QTcn5s6NACfi4iHlR7ang28Rmq7/Uab5ayQ9AVSM4laUlwvxjaKiPh7udiRHmY/T1Ir0ASc0v5SbEPKTVi63PV1RPxc0iDgSkkn5CZR1gn5fPRJ4AKlRwUWki5oziA16St1AzABmEW6+//NiHgVuFXSrsAMSatIhe/vkGrofivpTKAXcG2et5xJpOabbsL59pWNk6Svkx5FeCgf8yZFxFk5Pn9X6qa+iXSRO79nsr7BFZs09gKaSX0e/DyP+wOpqdwjShtkIfBJ0nm7RdIs0vNpv2hnujWs47zflX2BMnmoAz4vqYl0HfjjLm6HW4GTJf2DdKP5gZJxjcBeOW8LgGLHS13Nc09bn1i354fAr5Vey9BCugk5UdKjwJOk/hbuzdMOBG5SaoUn4Os5/TyljnREuqkyixTXruSjYhQfSDSzEpIGRERDvtt9N3BSRDzS0/kyM6s0Su9LvCAi2mvhYlbxNsXzvqSGiHCv7NYh1+yZlXeJpJ1JzXD/WOkHfDOzniDpW6QadD+rZ5s6n/etKrlmz8zMzMxsI5A0lPy+2DYOiIhFGzs/Vv1c2DMzMzMzM6tC7o3TzMzMzMysCrmwZ2ZmZmZmVoVc2DMzs6oj6U5JB7dJ+6qk33ZhGZMlDV7HNA3tpF8h6e28V9XMzOxtc2HPzMyq0TXAsW3Sjs3pHcovIS5ExGERsaRbcmdmZrYRuLBnZmbV6K/A4ZJ6A0gaA4wEHpU0VdIjkuZI+kRxvKSnJP0JeBzYRtI8ScPy+BslzZQ0V9JJpT8k6YKcPlXS8LYZkbS7pLvy/LdJGpHTT5P0hKTZkq7txm1hZmabKRf2zMys6kTEYuAh4NCcdCxwHbAcOCoidgP2B86XpDzNWOA3ETEuIua3WeQXI2J3YA/gtNx9OkB/YEZEjAPuAr5XOpOkXsAvgaPz/JcBP8qjvwW8PyLGAydviPU2MzMr5Zeqm5lZtSo25bwp//13QMCPJX0YaAVGAVvn6edHxAPtLOs0SUfl4W1IBcNFeRl/yelXARPbzPduYBdgSi5T1gCv5HGzgasl3QjcuJ7raGZm1i4X9szMrFrdBFwgaTegX0TMlHQiMBzYPSKaJM0D+uTpG8stRNJ+wIHAhIhYJmlayTxttX15rYC5ETGhzLSHAx8GjgC+K+m9EdHc2ZUzMzNbFzfjNDOzqhQRDcCdpKaTxY5ZBgELckFvf2DbTixqEFCfC3rvAfYuGVcAir1uHg/c02bep4DhkiZAatYpaZykArBNRNwJnJF/Y0CXV9LMzKwDrtkzM7Nqdg1wA6t75rwauFnSHGAG8GQnlnErcLKkf5AKb6VNPRuBvSSdCSwAjimdMSJW5VcwXCRpEOm8eyHwNHBVThNwkXv+NDOzDU0RbVucmJmZmZmZ2abOzTjNzMzMzMyqkAt7ZmZmZmZmVciFPTMzMzMzsyrkwp6ZmZmZmVkVcmHPzMzMzMysCrmwZ2ZmZmZmVoVc2DMzMzMzM6tCLuyZmZmZmZlVof8HH0AB4ZdrWzkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(15, 5))\n",
    "\n",
    "etiqueta_x = \"Variables\"\n",
    "etiqueta_y = \"Valores Normalizados\"\n",
    "agrupar_por = \"Group\"\n",
    "\n",
    "df_representar = pd.melt(df_agrupado, agrupar_por,\n",
    "                         var_name=etiqueta_x, value_name=etiqueta_y)\n",
    "\n",
    "paleta_colores = sns.xkcd_palette(colors=['goldenrod', 'coral',\n",
    "                                          'turquoise', 'lavender', 'indigo'])\n",
    "\n",
    "fig = sns.swarmplot(data=df_representar, x=etiqueta_x, y=etiqueta_y,\n",
    "                    hue=agrupar_por, palette=paleta_colores, dodge=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La primera conclusión que se puede extraer de la gráfica es que el grupo `0` es el que más población tiene; seguido de los grupos `1` y `3`; después por el grupo `2` y finalmente el grupo `4`.\n",
    "\n",
    "En la gráfica también llama la atención que el grupo `4` tiene las muestras de mayor gasto en *Milk*, *Grocery* y *Detergents & Paper*.\n",
    "\n",
    "Otro punto a tener en cuenta es que la variable *Channel* tiene mucho peso a la hora de generar los *clusters* dado que todas las muestras de un mismo cluster pertenecen al mismo *Channel*. No así en el caso de la variable *Region* donde hay clusters con muestras en dos categorías distintas como los grupos `1` y `2` y en menor medida el grupo `4`.\n",
    "\n",
    "Un posible dato interesante es que, descontando el grupo `4` por la baja cantidad de muestras, el grupo `2` es el grupo con mayor dispersión en las muestras lo que puede indicar que se trata de un grupo con mayor inhomogeneidad que el resto."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "histogramas_con_categorias_agregadas(df_agrupado, nombre_algoritmo='KMeans')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Otra manera de representar gráficamente los resultados es la gráfica superior, que muestra los gastos totales normalizados y segementados por categoría de cada grupo y dividos por el número de muestras de cada grupo.\n",
    "\n",
    "Al estar divididos los datos totales por el número de muestras de cada grupo pueden compararse y ver en proporción qué grupo gasta más en cada una de las categorías.\n",
    "\n",
    "Como ya se había visto a los largo del análisis de los resultados, el grupo `4` es el grupo que más gasta mientras que los grupos `0` y `1` los que menos.\n",
    "\n",
    "Además es interesante ver cómo los grupos `0` y `1` son muy parecidos entre ellos, al igual que los grupos `2` y `3`, lo que podría llevar a pensar que los dos primeros grupos podrían haber sido uno solo y los otros dos otro único grupo. Incluso podría llegar a pensarse que los 4 grupos podrían ser uno.\n",
    "\n",
    "Sin embargo, he aquí la *trampa*, en esta gráfica no se representan las variables *Channel* y *Region*  que si se analizan en la tabla para los grupos citados, puede comprobarse que efectivamente estas variables hacen que los grupos sean distintos.\n",
    "\n",
    "Los grupos `0` y `1` comparten igual *Channel* pero tienen distinto valor de *Region*. La situación es análoga para los grupos `2` y `3`. Además el valor de *Channel* entre los dos primeros grupos y los dos segundos tampoco coincide."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusiones\n",
    "\n",
    "El método *K-Medias* ha sido capaz de agrupar un conjunto de datos de 8 variables, y además se ha podido hacer uso de una métrica que ayudase a determinar cuál era el mejor número de clusters.\n",
    "\n",
    "Un punto negativo del estudio es que debido a la gran cantidad de variables, su representación visual no es sencilla ni intuitiva lo que complica poder obtener conclusiones acerca de los *clusters* generados a partir de una representación gráfica de los datos.\n",
    "\n",
    "En este caso, los centros geométricos de los datos ha sido la fuente de información que más ha ayudado a extraer conclusiones sobre los grupos obtenidos.\n",
    "\n",
    "Aunque de nuevo este segundo método se complica si el número de variables es demasiado grande.\n",
    "\n",
    "Esto demuestra que aunque los métodos de agrupación son eficaces y sencillos de utilizar, extraer conclusiones a partir de los resultados no es trivial."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Apéndice\n",
    "\n",
    "#### Clustering Jerárquico\n",
    "\n",
    "Por completitud y para tener otro método de agrupamiento con el que poder comparar los resultados, se ha probado también la clase [`sklearn.cluster.AgglomerativeClustering`](https://scikit-learn.org/stable/modules/generated/sklearn.cluster.AgglomerativeClustering.html) que implementa un algoritmo de *agrupamiento jerárquico*, comenzando con *n-grupos* y agrupándolos de dos en dos hasta que sólo quede uno.\n",
    "\n",
    "Igual que en el [agrupamiento k-medias](#Agrupamiento-k-medias), se ha usado la métrica `silhoutte_score` con el fin de determinar el número de clusters óptimo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mejor configuración de agrupación obtenida:\n",
      "  - Número de clusters: 5\n",
      "  - Media de silhouette_score: 0.59\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cluster import AgglomerativeClustering\n",
    "\n",
    "best_model_h = None\n",
    "best_labels_h = None\n",
    "best_n_clusters_h = None\n",
    "best_silhouette_avg_h = None\n",
    "\n",
    "for n_clusters in range(2, 11):\n",
    "    model = AgglomerativeClustering(n_clusters=n_clusters)\n",
    "\n",
    "    cluster_labels = model.fit_predict(df)\n",
    "    silhouette_avg = silhouette_score(df, cluster_labels)\n",
    "    if not best_silhouette_avg_h or best_silhouette_avg_h < silhouette_avg:\n",
    "        best_silhouette_avg_h = silhouette_avg\n",
    "        best_model_h = model\n",
    "        best_labels_h = cluster_labels\n",
    "        best_n_clusters_h = len(np.unique(best_labels_h))\n",
    "\n",
    "df_agrupado_h = pd.concat([df, pd.Series(best_labels_h, name='Group')],\n",
    "                          axis=1, sort=False)\n",
    "        \n",
    "print(\"Mejor configuración de agrupación obtenida:\")\n",
    "print(\"  - Número de clusters: {:d}\".format(best_n_clusters_h))\n",
    "print(\"  - Media de silhouette_score: {:.2f}\".format(best_silhouette_avg_h))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col0 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col4 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col5 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col6 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col8 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col0 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col1 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col5 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col0 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col1 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col3 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col7 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col0 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col2 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col0 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col2 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col3 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col4 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col6 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col7 {\n",
       "            background-color:  turquoise;\n",
       "            : ;\n",
       "        }    #T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col8 {\n",
       "            : ;\n",
       "            background-color:  coral;\n",
       "        }</style><table id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aa\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Channel</th>        <th class=\"col_heading level0 col1\" >Region</th>        <th class=\"col_heading level0 col2\" >Fresh</th>        <th class=\"col_heading level0 col3\" >Milk</th>        <th class=\"col_heading level0 col4\" >Grocery</th>        <th class=\"col_heading level0 col5\" >Frozen</th>        <th class=\"col_heading level0 col6\" >Detergents_Paper</th>        <th class=\"col_heading level0 col7\" >Delicassen</th>        <th class=\"col_heading level0 col8\" >Count</th>    </tr>    <tr>        <th class=\"index_name level0\" >Group</th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>        <th class=\"blank\" ></th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aalevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col0\" class=\"data row0 col0\" >1</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col1\" class=\"data row0 col1\" >2.99528</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col2\" class=\"data row0 col2\" >13966.9</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col3\" class=\"data row0 col3\" >3549.7</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col4\" class=\"data row0 col4\" >3932.67</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col5\" class=\"data row0 col5\" >3926.77</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col6\" class=\"data row0 col6\" >788.972</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col7\" class=\"data row0 col7\" >1537.58</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow0_col8\" class=\"data row0 col8\" >212</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aalevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col0\" class=\"data row1 col0\" >2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col1\" class=\"data row1 col1\" >3</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col2\" class=\"data row1 col2\" >9038.2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col3\" class=\"data row1 col3\" >9313.69</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col4\" class=\"data row1 col4\" >14208.8</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col5\" class=\"data row1 col5\" >1443.93</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col6\" class=\"data row1 col6\" >6065.85</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col7\" class=\"data row1 col7\" >1776.47</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow1_col8\" class=\"data row1 col8\" >101</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aalevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col0\" class=\"data row2 col0\" >1</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col1\" class=\"data row2 col1\" >1.31395</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col2\" class=\"data row2 col2\" >12264.3</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col3\" class=\"data row2 col3\" >3210.2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col4\" class=\"data row2 col4\" >4034.77</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col5\" class=\"data row2 col5\" >3308.19</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col6\" class=\"data row2 col6\" >794.477</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col7\" class=\"data row2 col7\" >1116.14</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow2_col8\" class=\"data row2 col8\" >86</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aalevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col0\" class=\"data row3 col0\" >2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col1\" class=\"data row3 col1\" >1.5</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col2\" class=\"data row3 col2\" >6209.47</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col3\" class=\"data row3 col3\" >10104.4</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col4\" class=\"data row3 col4\" >15983.2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col5\" class=\"data row3 col5\" >2101.5</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col6\" class=\"data row3 col6\" >7493</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col7\" class=\"data row3 col7\" >1556.14</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow3_col8\" class=\"data row3 col8\" >36</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aalevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col0\" class=\"data row4 col0\" >2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col1\" class=\"data row4 col1\" >2.8</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col2\" class=\"data row4 col2\" >25603</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col3\" class=\"data row4 col3\" >43460.6</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col4\" class=\"data row4 col4\" >61472.2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col5\" class=\"data row4 col5\" >2636</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col6\" class=\"data row4 col6\" >29974.2</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col7\" class=\"data row4 col7\" >2708.8</td>\n",
       "                        <td id=\"T_6ffd25f0_942a_11e9_8da2_003ee1bf34aarow4_col8\" class=\"data row4 col8\" >5</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x141b69b38>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Cálculo de los 'centros' de los clusters jerárquicos\n",
    "df_centros_h__ = pd.concat([df_raw, pd.Series(best_labels_h, name='Group')],\n",
    "                            axis=1, sort=False)\n",
    "\n",
    "df_centros_h = df_centros_h__.groupby('Group').mean()\n",
    "df_centros_h['Count'] = df_centros_h__.groupby('Group')['Group'].count()\n",
    "\n",
    "df_centros_h.style.highlight_max(color='turquoise').highlight_min(color='coral')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Representación visual de los grupos por agrupamiento jerárquico\n",
    "fig = plt.figure(figsize=(15, 5))\n",
    "\n",
    "df_representar_h = pd.melt(df_agrupado_h, agrupar_por,\n",
    "                           var_name=etiqueta_x, value_name=etiqueta_y)\n",
    "\n",
    "fig = sns.swarmplot(data=df_representar_h, x=etiqueta_x, y=etiqueta_y,\n",
    "                    hue=agrupar_por, palette=paleta_colores, dodge=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Representación de los grupos por gastos totales por númeor de muestras en grupo\n",
    "histogramas_con_categorias_agregadas(df_agrupado_h, nombre_algoritmo='Clustering Jerárquico')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En este caso el número de grupos óptimo también ha resultado ser `5` y aunque la numeración difiere con respecto a los grupos obtenidos en el caso de *k-medias* los grupos son muy parecidos, salvo por alguna fluctuación de muestras.\n",
    "\n",
    "Sin embargo, pese a que los resultados no difieren mucho entre uno y otro método de agrupamiento, sí pueden apreciarse diferencias sutiles. Por ejemplo, ahora el grupo más numeroso es el grupo con mayor gasto en *Frozen*.\n",
    "\n",
    "Comparando los dos gráficos y dejando al margen que los grupos no coinciden en numeración puede verse que cualitativamente los grupos son muy similares entre los dos algoritmos de clustering."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}