Relatorio-Estatistica.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Análise Estatistica\n",
    "### Amanda Luna, Renato Henriques e Sheila Silva\n",
    "### 12 de junho de 2019\n",
    "\n",
    "## Apresentação do problema\n",
    "\n",
    "O Conjunto de Dados de Habitação de Boston consiste no preço das casas em vários locais em Boston, esse conjunto de dados é do ano de 1970. O conjunto de dados (dataset) nos disponibiliza quatorze variavéis distintas que tentaremos avaliar a relação entre elas para predizer os valores das casas através de regressão linear múltipla.\n",
    "\n",
    "## Descrição do Dataset\n",
    "\n",
    "O dataset escolhido consiste no preço das casas em vários lugares em Boston nos anos 70. Além disso, é composto por 506 observações, com 13 atributos . Juntamente com o preço, o conjunto de dados também fornece informações como:\n",
    "\n",
    "1. CRIM - taxa de criminalidade per capita por cidade\n",
    "2. ZN - proporção de terrenos residenciais zoneados para lotes com mais de 25.000 pés² (cerca de 2.322 m²)\n",
    "3. INDUS - proporção de negócios não-varejistas por acre (4046m²) por cidade\n",
    "4. CHAS - Variável de Charles River que mede quantidade de bairros próximos\n",
    "5. NOX - concentração de óxidos nitricos\n",
    "6. RM - número médio de quartos por habitação\n",
    "7. AGE - proporção de unidades ocupadas construídas antes de 1940\n",
    "9. RAD - índice de acessibilidade a rodovias radiais \n",
    "10. TAX - taxa de imposto predial de valor integral 10,000 dólares\n",
    "11. PTRATIO - relação aluno-professor por cidade\n",
    "12. B - 1000(Bk - 0.63)^2 onde Bk é a quantidade de negro por cidade\n",
    "13. LSTAT - porcentagem de pessoas de classe baixa por cidade\n",
    "14. PRICE - valor médio de residências\n",
    "\n",
    "## Importando as bibliotecas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline \n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import scipy.stats as stats\n",
    "import matplotlib.pyplot as plt\n",
    "import sklearn\n",
    "import statsmodels.api as sm\n",
    "from sklearn import metrics\n",
    "from sklearn.datasets import load_boston\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from scipy import stats\n",
    "import seaborn as sns\n",
    "sns.set_style(\"whitegrid\")\n",
    "sns.set_context(\"poster\")\n",
    "from matplotlib import rcParams\n",
    "from sklearn.datasets import load_boston\n",
    "boston = load_boston()\n",
    "bos = pd.DataFrame(boston.data, columns=boston.feature_names)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Carregando o dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      CRIM    ZN  INDUS  CHAS    NOX     RM   AGE     DIS  RAD    TAX  \\\n",
      "0  0.00632  18.0   2.31   0.0  0.538  6.575  65.2  4.0900  1.0  296.0   \n",
      "1  0.02731   0.0   7.07   0.0  0.469  6.421  78.9  4.9671  2.0  242.0   \n",
      "2  0.02729   0.0   7.07   0.0  0.469  7.185  61.1  4.9671  2.0  242.0   \n",
      "3  0.03237   0.0   2.18   0.0  0.458  6.998  45.8  6.0622  3.0  222.0   \n",
      "4  0.06905   0.0   2.18   0.0  0.458  7.147  54.2  6.0622  3.0  222.0   \n",
      "\n",
      "   PTRATIO       B  LSTAT  PRICE  \n",
      "0     15.3  396.90   4.98   24.0  \n",
      "1     17.8  396.90   9.14   21.6  \n",
      "2     17.8  392.83   4.03   34.7  \n",
      "3     18.7  394.63   2.94   33.4  \n",
      "4     18.7  396.90   5.33   36.2  \n"
     ]
    }
   ],
   "source": [
    "bos['PRICE'] = boston.target\n",
    "print(bos.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tratamento de dados do dataset\n",
    "\n",
    "Após carregar o dataset e adicionar a váriavel PRICE, referente ao preço das casas, fazemos o tratamento dos dados e obtivemos os seguintes resultados. Que são referentes a quantidade (count), média(mean), desvio padrão(std) e os valores minimos até os valores máximos para cada atributo, como podemos ver na tabela abaixo:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             CRIM          ZN       INDUS        CHAS         NOX          RM  \\\n",
      "count  506.000000  506.000000  506.000000  506.000000  506.000000  506.000000   \n",
      "mean     3.613524   11.363636   11.136779    0.069170    0.554695    6.284634   \n",
      "std      8.601545   23.322453    6.860353    0.253994    0.115878    0.702617   \n",
      "min      0.006320    0.000000    0.460000    0.000000    0.385000    3.561000   \n",
      "25%      0.082045    0.000000    5.190000    0.000000    0.449000    5.885500   \n",
      "50%      0.256510    0.000000    9.690000    0.000000    0.538000    6.208500   \n",
      "75%      3.677083   12.500000   18.100000    0.000000    0.624000    6.623500   \n",
      "max     88.976200  100.000000   27.740000    1.000000    0.871000    8.780000   \n",
      "\n",
      "              AGE         DIS         RAD         TAX     PTRATIO           B  \\\n",
      "count  506.000000  506.000000  506.000000  506.000000  506.000000  506.000000   \n",
      "mean    68.574901    3.795043    9.549407  408.237154   18.455534  356.674032   \n",
      "std     28.148861    2.105710    8.707259  168.537116    2.164946   91.294864   \n",
      "min      2.900000    1.129600    1.000000  187.000000   12.600000    0.320000   \n",
      "25%     45.025000    2.100175    4.000000  279.000000   17.400000  375.377500   \n",
      "50%     77.500000    3.207450    5.000000  330.000000   19.050000  391.440000   \n",
      "75%     94.075000    5.188425   24.000000  666.000000   20.200000  396.225000   \n",
      "max    100.000000   12.126500   24.000000  711.000000   22.000000  396.900000   \n",
      "\n",
      "            LSTAT       PRICE  \n",
      "count  506.000000  506.000000  \n",
      "mean    12.653063   22.532806  \n",
      "std      7.141062    9.197104  \n",
      "min      1.730000    5.000000  \n",
      "25%      6.950000   17.025000  \n",
      "50%     11.360000   21.200000  \n",
      "75%     16.955000   25.000000  \n",
      "max     37.970000   50.000000  \n"
     ]
    }
   ],
   "source": [
    "print(bos.describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Análise exploratória dos dados\n",
    "\n",
    "\n",
    "Analisando a váriavel PRICE referente a o preço das casas, podemos observar que ela segue uma distribuição normal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 842.4x595.44 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(rc={'figure.figsize':(11.7,8.27)})\n",
    "sns.distplot(bos['PRICE'], bins=30)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Separando o dataset de treino e de teste\n",
    "\n",
    "Basicamente, antes de dividir os dados para o conjunto de dados de teste e de treinamento, precisaríamos dividir o conjunto de dados em dois: valor alvo e valores de previsão. Vamos chamar valor alvo de Y e os valores de predição de X.\n",
    "Portmanto, temos:\n",
    "\n",
    "Y = Preço das casas em Boston\n",
    "\n",
    "X = Todos os outros atributos\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "quantidade de dados para treino de X/quantidade de atributos (339, 13)\n",
      "quantidade de dados para teste de X/quantidade de atributos (167, 13)\n",
      "quantidade de dados para treino de Y (339,)\n",
      "quantidade de dados para teste de Y (167,)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X = bos.drop('PRICE', axis = 1)\n",
    "Y = bos['PRICE']\n",
    "X_train, X_test, Y_train, Y_test = sklearn.model_selection.train_test_split(X, Y, test_size = 0.33, random_state = 5)\n",
    "print(\"quantidade de dados para treino de X/quantidade de atributos\",X_train.shape)\n",
    "print(\"quantidade de dados para teste de X/quantidade de atributos\",X_test.shape)\n",
    "print(\"quantidade de dados para treino de Y\" , Y_train.shape)\n",
    "print(\"quantidade de dados para teste de Y\" , Y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se também verificarmos a forma de cada variável, podemos descobrir que agora já recebemos nosso trem e testamos conjuntos de dados com a proporção de 66,66%(339) para dados de treino e 33,33%(167) para dados de teste.\n",
    "\n",
    "## Regressão Linear Múltipla\n",
    "\n",
    "A separação dos dados acima irá se encaixar em um modelo baseado em X_train e Y_train. Agora nós já temos o modelo linear, tentamos prever isso para o X_test e agora temos os valores de predição que são armazenados em Y_pred. Para visualizar as diferenças entre os preços reais e os valores previstos, também criamos um gráfico de dispersão que pode ser visualizado a seguir."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Preços vs Preços preditos')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "lm = LinearRegression()\n",
    "lm.fit(X_train, Y_train)\n",
    "\n",
    "Y_pred = lm.predict(X_test)\n",
    "\n",
    "plt.scatter(Y_test, Y_pred, s=10, alpha=0.5)\n",
    "plt.xlabel(\"Preços originais\")\n",
    "plt.ylabel(\"Preços preditos\")\n",
    "plt.title(\"Preços vs Preços preditos\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O gráfico de dispersão deveria criar um gráfico linear, entretanto o modelo não se ajusta a 100% por isso não temos uma obtemos essa linearidade.\n",
    "\n",
    "## Erro quadrático médio\n",
    "\n",
    "Para verificar o nível de erro de um modelo, podemos calcular o erro quadrático médio. É um dos procedimentos para medir a média dos quadrados de erro. Basicamente, ele irá verificar a diferença entre o valor real e o valor previsto. Para usá-lo, podemos usar a função de erro quadrática média do scikit-learn executando o trecho de código a seguir:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Original    Predito\n",
      "226      37.6  37.467236\n",
      "292      27.9  31.391547\n",
      "90       22.6  27.120196\n",
      "373      13.8   6.468433\n",
      "273      35.2  33.629667\n",
      "417      10.4   5.670680\n",
      "503      23.9  27.039467\n",
      "234      29.0  29.927047\n",
      "111      22.8  26.356613\n",
      "472      23.2  22.452460\n",
      "99       33.2  32.205044\n",
      "339      19.0  21.786417\n",
      "309      20.3  23.411384\n",
      "304      36.1  33.608944\n",
      "272      24.4  28.286195\n"
     ]
    }
   ],
   "source": [
    "diff = pd.DataFrame({'Original': Y_test, 'Predito': Y_pred})\n",
    "diff1 = diff.head(20)\n",
    "print(diff.head(15))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>Original</th>\n",
       "      <th>Predito</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>167.000000</td>\n",
       "      <td>167.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>22.523952</td>\n",
       "      <td>21.847663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.711260</td>\n",
       "      <td>8.394229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>7.000000</td>\n",
       "      <td>-6.249525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>16.400000</td>\n",
       "      <td>17.425063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>20.700000</td>\n",
       "      <td>21.725754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>25.050000</td>\n",
       "      <td>27.079831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>42.756775</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Original     Predito\n",
       "count  167.000000  167.000000\n",
       "mean    22.523952   21.847663\n",
       "std      9.711260    8.394229\n",
       "min      7.000000   -6.249525\n",
       "25%     16.400000   17.425063\n",
       "50%     20.700000   21.725754\n",
       "75%     25.050000   27.079831\n",
       "max     50.000000   42.756775"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diff.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Temos que os valores da média dos preços originais e preditos são 22.523952 e 21.847663 respectivamente.\n",
    "Com isso, temos que o valor da raiz quadrada do erro médio é maior do que 10% do valores das médias (2.2523952 e 2.1847663). Portanto, o nosso modelo não possui uma boa acurácia, porém, consegue realizar predições razoáveis.\n",
    "\n",
    "Alguns motivos para isto ter acontecido são:\n",
    "1. Precisariamos de mais dados\n",
    "2. Os atributos que utilizamos talvez não possuam muita correlação com o valor que tentamos predizer\n",
    "3. Assumimos que os dados possuíam um relacionamento linear, o que pode não ser o caso \n",
    "\n",
    "O gráfico de barras abaixo consiste em uma amostra dos resultados comparativos dos preços originais e preditos pela regressão linear múltipla.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnAAAAIBCAYAAADJbYLHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3XtclGX+//E3ICAjliIgahaaOp5Cw3PlITXNWnO3Lc0UDy1qeepbnrC2ojRLs1Iy7WBrpWZmq5mrpZWhRoZ5NjBMFMP1hIImB0GY+f3hj0kW1IGZAe/h9Xw8evTouq/7/lz3MOK7677v6/awWq1WAQAAwDA8K3oAAAAAKB0CHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMJgqFT2AskpMTFRubq68vLzk6+tb0cMBAAC4otzcXBUUFMjX11fNmzd3+HiGDXC5ubmyWCyyWCy6ePFiRQ8HAADgmnJzc51yHMMGOC8vL1ksFnl6espkMpVq35SzKQqtEeqagVGPegapRT3qUa/y1HPnczNKvezsbFksFnl5eTllDIYNcL6+vrp48aJMJpPMZnOp9l0Wu0y9O/R20cioRz1j1KIe9ahXeeq587kZpV5SUpIyMzOddtsXDzEAAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwhl0HDgCAK7FYLEpPT9f58+eVm5srq9V6zX06mzpr//795TC68q/nzudW3vU8PDx0e9Xbdfr0aQUEBMjTs2LmwghwAAC3YrFYlJqaquzs7FLtF2QKctGIKr6eO59bedezWq26yXST0tLSlJWVpfr161dIiCPAAQDcSnp6urKzs1WlShWFhISoWrVqdv0Fe+z8MdWtXrccRlj+9dz53Mq7nsVi0ZG0I8r/I1/Z2dlKT09XYGBgudS+HPfAAQDcyvnz5yVJISEhql69eoVd4oJ78vT0lK/JVyEhIZL+/L6V+zgqpCoAAC6Sm5srSapWrVoFjwTurPD7Vfh9K28EOACAWyl8YIGZN7iSh4eHJNn1gIwr8O0GAAAopcIAV1EIcAAAAAZDgAMAADCYShngortFV/QQAADXmfJc9qIi6sG9VMoABwAApIKCAn3xxRcaNWqU7rrrLrVs2VJt27bVI488ooULF5ZpiQyz2aweHXo4NK6IiAiZzWZt377doeOUxltvvSWz2az58+eXW01HuPVCvqFRa4u1pbx6fwWMBABwPSrp74nrkSv+7kpNTdXo0aN14MABeXl56bbbblN4eLjOnDmjxMRE7dq1Sx9++KFiYmIUHh7u9PpwjFsHOAAAUNzJkyf1yCOP6PTp07rvvvsUFRWl2rVr27afP39eCxcu1LvvvqshQ4bogw8+UIcOHew69rp165SWnebQ+GbOnKmcnBzVq1fPoeO4s8ob4KJvLKHtXPmPAwCAcjZlyhSdPn1aAwcOVHR0dLHt1atX11NPPaXg4GC99NJLmjp1qr788kv5+/tf89i33nqr/M77OTS+unW5P/BauAcOAIBKZM+ePdq6dasCAgI0derUq/YdNGiQwsPD9d///lerV6+W9Oe9Yl9//bWmTp2q1q1bq0OHDlqwYIGkK98Dd+DAAT355JO666671Lp1aw0cOFA//PCD5s+fL7PZrPj4eFvfku6BK2wrnB3s06ePbrvtNnXu3FlzZs5Renp6sZpnz57VnDlz9Ne//lXh4eFq2bKlunTpokmTJunQoUNl+vyuFwQ4AAAqkW+//VaSdO+998rX1/ea/R988EFJ0po1a4q0v/nmm/rqq690xx13KDAwUI0bN77iMbZv364BAwbo66+/VnBwsLp06aKjR49qxIgR2rhxY6nGHxUVpddff101atRQly5dlJ2drdUrVisyMrJIv9OnT+vvf/+7FixYoOzsbN1xxx3q0KGDcnNz9eWXX6p///46fvx4qWpfTyrvJVQAACqhPXv2SJJatmxpV//27dtLkvbv31+kPTU1VZ9//rmaN28uSbJYLCXun5eXp6lTpyo7O1vTpk1T//79JV16h+ikSZO0fv36Uo0/Pj5en376qVq1aiXp0v18/f7aTwkJCdq+fbvatm0rSXr77bd19OhRDRs2TFFRUbY3J2RmZmrEiBHauXOnvvjiCz3xxBOlqn+9YAYOAIBKpPBSY2BgoF39C/tduHBB5879ea94mzZtbOFNuvK7Zzdt2qTff/9dPXr0sIU3SfL19dWMGTN0ww03lGr8gwYNsoU3Sapdu7bu6naXJGnv3r229po1a6pz584aN25ckdde+fv76y9/+YskMQMHAACMoXCmrEoV+yLA5f0uf3F706ZN7dr/xx9/lCTdc889xbb5+/urc+fOWrvW/uVcLg9vhQJqBUiSsrOzbW3jx48v1i89PV1JSUm2e+suXrxod93rDQEOAIBKpHbt2kpOTlZGRoZd/dPSLi0JUrVqVdWoUcPWfuONJazmUILCWa46deqUuL20S4WUNGPn5eUlqWjAlKTff/9dS5Ys0c6dO3X48GFlZmZK+vNF9P/b30i4hAoAQCXSrFkzSdLu3bvt6l94WdJsNhdpv9Il0/9VOMt1pbDkqhC1Zs0a3Xvvvfroo490/vx5denSRePGjdN7772nF1980SU1yxMBDgCASqRv376SpLVr1yonJ+ea/T///HNJ0gMPPFCmeiEhIZKkY8eOlbjdFfehZWVl6YUXXpCnp6fee+89rV+/Xm+++abGjh2rrl27Ki8vz+k1yxsBDgCASqRZs2bq2bOn0tPTNW3atKv2XbVqleLi4hQcHFzmAFf4BofY2Nhi23Jzc233yDlTcnKysrKy1KJFC3Xt2rXY9ri4OElXfnLWCAhwAABUMi+99JJCQkL073//W0899ZTtPrdCubm5WrBggZ599ll5enpq5syZpX5atFCvXr0UEhKiDRs26IsvvrC15+fn68UXX7Q9FXv5k6KOKpz1O3DggFJTU23tBQUFWrBggb7//ntJl87TqHiIAQBQabniJfFGUKtWLa1YsULjxo3TunXrtGHDBoWFhSk4OFjnz5/X7t27lZWVpaCgIM2ePVsdO3Ysc62qVavq1Vdf1YgRIzRlyhQtWbJE9erV0759+3Ty5EnVrVtXx44ds/upWHsEBwfrvvvu07p169S3b1+1b99eVapU0d69e5WWlqZGjRrp4MGDOn36tNNqljdm4AAAqISCg4P1ySef6LXXXtOdd96pI0eO6Ntvv9Vvv/2mFi1a6Pnnn9dXX33lUHgr1KlTJ3366ae6++67deTIEX3//fcKCQnRokWL1KJFC0mX3r/qTDNmzNDYsWMVEhKin376SXv37tVNN92k559/XqtWrdKNN96o3bt3l/gKLiNgBg4AcF2K7hZdrvWOnT+mutXL7yXq5V2vJF5eXnrggQdKdX/buHHjNG7cuCtuT0pK0rHzfz6wcPr0aZ07d06NGjXSO++8U6z/9OnT5enpWWQ5kcWLFxfrV1JboYjICE15akqRNj8/v6uOddu2bUX++1rndb1hBg4AALhMYmKi7rvvPv3jH/8o9vTnihUrlJSUpDvvvFMmk6mCRmhMzMABACpUaFTxVfgr671p7qhTp05q0aKFtm/frq5du6pVq1by9vZWcnKykpOTFRQUpBdeeKGih2k4zMABAACX8fb21scff6yJEycqJCREO3fu1ObNm1VQUKDHHntMX375perXr1/RwzQcZuAAAIBL+fv7a8SIERoxYkRFD8VtMAMHAABgMHbPwBUUFGjp0qX6/PPPdfjwYfn5+ally5YaMmSIunXrVqTv8ePHi7VdLjw8XMuWLSvrmAEAACo1uwPc1KlTtXr1avn7+6tTp066ePGitm3bpri4OI0fP15jxoyx9U1MTJR06cW3TZo0KXasBg0aOGHoAAAAlZNdAW7dunVavXq1GjRooCVLligwMFCS9Ntvv2ngwIGaN2+e7r//foWGhkqS9u/fL0mKjIws87vTAAAAUDK77oH78ssvJUkTJ060hTdJaty4sfr27SuLxWJ7Maz05wxc4erKAAAAcB67ZuBiYmKUkpJim2G7XFZWlqRLqzkX2r9/v0wmE5dK4RLlvTo7AADXG7sCnI+PT4n3sn3//ff6+uuvZTKZ1LNnT0nS2bNndezYMbVo0UKLFi3S6tWrdeTIEVWvXl133323xo4dq9q1azv3LAAAACqRUq8Dd+HCBU2ePFkHDx5UcnKy6tatq1mzZtkurRbe/5aQkKADBw6oXbt2CgkJ0b59+/TZZ5/p+++/18cff6yGDRs690zgdlidHQCAkpU6wB07dkzr168v0paUlKR27dpJ+vP+t8aNG2vBggW21ZWzs7P13HPP6T//+Y8mTpyolStXOjp2SVLK2RQtiy26JImjl9iiYx3b/3/FpsQ6/ZjuXM+en5+Rz+96qUU96lV0PVf9We9s6qwgU1CRF6rb43zueR1T6fZxRHnWc+dzq8h6FwsuKi07Tctjl19zn05+nRRYJfCa/ezlYbVaraXZITs7W7m5ufL09NSPP/6ol19+WWlpaZowYYJGjhypgoICHTt2TNWqVVNAQECRfXNyctS7d2+dPHlSy5cvV+vWrcs88KSkJGVmZsrf319ms7nEPledwYm+sfgO0efKPJ6riY6NLtf7ttyl3vUyA1een6e7/OyoR73ScPaf9cIrQc2aNSvzMdzNsfPHVLd6XbeuV8e/jjw8PMqtXt3qdUv1XbMnt5RGqd/EYDKZVLNmTd14443q06eP5s2bJw8PD7377rvKzc2Vl5eX6tevXyy8SZKfn586duwo6dIlVgAAUP7i4+NlNptL/Kdly5bq2LGjIiIitHz5chUUFJT7+E6cOCGz2azu3bsXaTebzWrevHmRtlOnTunl517Wzz//XJ5DrHAOvwu1devWuvnmm3XkyBGlpqaqUaNGV+1feK9cTk6Oo6UBAHBMSVdjrkcuukJkMpnUo0ePIm35+flKT0/Xjh07bAv2x8TEuKS+M0RFRSkuLk5DBg6p6KGUq2sGOKvVqtdee03Hjx/Xa6+9pipViu/i4+Mj6dIPfd68eTpw4IDGjBlT4hTh0aNHJUkhISGOjh0AADigZs2amj17donb9u/fr8GDB2v9+vX65ptvdM8995Tz6Ipbt25dscukFoulgkZTsa55CdXDw0Pfffed1q1bV2Sx3kKpqak6fPiwbd23pKQkrV+/Xl999VWxvmfOnFFcXJy8vb3VoUMH55wBAABwumbNmql///6SpA0bNlTwaC659dZbWcXi/7PrHrjCH+D06dN14sQJW/vJkyf19NNPKz8/X48++qh8fX01YMAASdKiRYu0Y8cOW9+srCw988wzyszM1EMPPaSgoCBnngcAAHCym266SZKUnp4uSerevbs6dOig/fv3q1+/fmrZsqV69OihX3/91bbP5s2bNWnMJLVr105hYWHq27evPvjgA+Xl5ZVYY82aNXr44Yd1++2366677tIrr7xie0nA/7r8HrijR4/KbDZr69atkqQhQ4bIbDbbrvRJUkZGhmbNmqXevXurZcuWat++vf7xj39oy5Ytjn84Fcyue+CGDBmi+Ph4bdq0SX369FF4eLgKCgq0Z88eZWdnq2vXrnryySclSXfddZeGDx+uRYsWafDgwQoPD1fNmjW1fft2ZWRkqG3btpoyZYpLTwoAADju4MGDkqQ6derY2vLy8jRy5EiZTCZ16dJFycnJuvXWWyVJ8+fP19y5c+Xt7a2wsDAFBARox44dmjVrljZt2qSFCxfabruSpNdff13vvfeefH191bFjRxUUFGjp0qX64Ycfrjk2k8mkvn376qefflJaWpruuOMO1apVSyaTSZL0+++/a/DgwTp58qRCQkLUvXt3ZWRkaOvWrfrhhx80btw4jR071pkfV7myK8B5e3trwYIF+uSTT7Ry5Ur9/PPP8vT0VJMmTfTggw+qf//+8vT8czIvKipKrVq10pIlS5SYmCiLxaKbb75ZkZGRGjp0qLy9vV12QgAAwHHbtm3TihUrJEn33//nsi7Z2dlq3Lixli5dKm9vb1ksFtvSYnPnzlXdunU1/c3purP1nbb+EyZM0MaNGzVv3jw9/fTTkqS9e/fq/fffV2BgoBYvXmy7NHrgwAENHTr0muMLCAjQ7NmzNWzYMKWlpenxxx+33Z5ltVr1f//3fzp58qQGDRqkqVOn2rLH3r17NWLECL311lsKCwtTly5dnPehlSO7n0L18vJSRESEIiIi7Orfp08f9enTp8wDAwAArpWRkaGJEycWacvLy9ORI0dsl0UHDx6sTp06FekzcOBAWyAqnMD54IMPJEn//Oc/1eDWP9+FbjKZ9PLLL+vuu+/W0qVLNXbsWPn4+Gj58uWyWq0aP358kfvamjRpovHjxys6OrrM5/Xzzz8rISFBt956q5599tki72sPCwtTVFSUoqKi9MEHH7h/gAMAAO4lOztba9asKdLm4+OjgIAA3X333XrwwQfVq1evYvs1bdq0yH8XFBRo+/btkqQOHTroD+sfRbYHBASoefPm2rlzpxITE9W6dWvbum0lBagePXo4HOAkqVevXkXCW6F7771Xzz77rHbu3KmCgoIS+1zvCHAAAFRS9erV08aNG0u93403Fl0/7+zZs7pw4YIkqU2bNlfd9/jx42rdurVOnTolSapdu3axPsHBwQ7dblV47Hr16pW43c/PTwEBAUpLS9O5c+dKfPnA9Y4ABwAASuXy+94l2d7W4Ofnp549eyonP0d+VfxK3LdwFYrC9dyu9EbPktadtVfhMa/2aq3C9eMuf6jCSAhwAADAITVq1JC3t7fy8/M1c+ZMncw+ec13oQYHByslJUXHjh1T/fr1i2w7f/68Q29sCg4OlnRprdqSZGZmKiMjQ1WrVpW/v3+Z61SkUr8LFQAA4HI+Pj5q1aqVLl68aFuX7XJ5eXl68MEH9eijj9rWabvjjjskSd99912x/ps3b7a7dkmzbO3atZMkffPNNyW+y3X9+vWyWCxq37693XWuNwQ4AADgsMKlP1544QUdPnjY1p6fn69p06YpISFB2dnZtsWBH330UXl7e+vtt9/Wvn37bP1TU1P12muv2V3X19dX0qVZu0Lt27dX8+bNlZycrJdfflkXL160bfvll180a9YsSdKgQYPKcKbXBy6hwphKegG1i172DKACufrPOr83nKZXr14aOnSoPvroI40cPFK33XabAgMD9csvv+j48eMKCAjQG2+8YevfuHFjTZkyRS+//LIeeeQRdezYUd7e3tq6dasaN26skydP2lX3lltukSS9+OKL+vLLLzVhwgTdcssteuONNzR06FAtXbpUGzduVFhYmDIyMrRjxw4VFBRo9OjR6tatmys+inLBDBwAAHCKZ555Rm+//bZahbdScnKyNm/erKpVqyoiIkJffPFFsfeYRkRE6L333lN4eLh27dqlXbt26b777tPChQuv+gDC5UaNGqWOd3XU+fPnFRcXp5SUFElSgwYNtGrVKg0fPlze3t7auHGjkpOT1aVLF3344Ye2N0gZFTNwAABIOnb+2DVvvDdyvct16NBBSUlJpd7PniVHevbsqeYdmtt9bl26dClxLbjExMRibSWNOSAgQK/MeaXEerVq1bIt2utumIEDACeJ7hZd0UMAUEkQ4AAAAAyGS6gAUEqhUWuLtaW8en8JPQHANZiBAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAFBKVqu1QusT4AAAbqVwAViLxVLBI4E7Kwxw9i447GwEOACAWyl8N2ZWVlYFjwTurPD7Vfh9K28sIwIAcCvVq1fXhQsXdOLECUlStWrV5OHhUWEzJXAfVqtVVqtVudm5yvrjUoCrXr16hYyFAAcAcCsBAQHKyspSdna2jh49avd+Fwsu6pxX+b3cvjzrufO5VUS9rItZquZdTSaTSQEBAeVW93JcQgUAuBVPT0/Vr19fQUFBqlq1qt0zb2nZaS4eWcXVc+dzK+96Hh4e+m/2fxUUFKT69evL07NiohQzcAAAt+Pp6anAwEAFBgbavc/y2OWKbh/tukFVYD13PreKqPdp7KfqF9iv3OqVhAAHAM4UfWMJbeV3aQdA5cAlVAAAAIMhwAEAABgMAQ4AAMBgCHAAAAAGQ4ADAAAwGAIcAACAwRDgykF0t+iKHgIAAHAjBDgAAACDYSHf8sLingAAwEmYgQMAADAYAhwAAIDBEOAAAAAMhgAHAABgMAQ4AAAAgyHAAQAAGIzdy4gUFBRo6dKl+vzzz3X48GH5+fmpZcuWGjJkiLp161as/+HDh/XWW29px44dOnv2rG6++WYNGDBAjz76qDw9yY0AAABlZXeAmzp1qlavXi1/f3916tRJFy9e1LZt2xQXF6fx48drzJgxtr6//vqrBg0apMzMTIWHh+u2225TfHy8pk2bpt27d2v27NkuORkAAIDKwK4At27dOq1evVoNGjTQkiVLFBgYKEn67bffNHDgQM2bN0/333+/QkNDZbVaNXnyZGVmZmrWrFnq16+fJCk9PV3Dhg3TmjVrdM8996h3796uOysAAAA3Zte1zC+//FKSNHHiRFt4k6TGjRurb9++slgsiouLkyTFxcUpKSlJ7du3t4U3SQoICNALL7wgSVq8eLHTTgAAAKCysWsGLiYmRikpKQoNDS22LSsrS5Lk5eUlSdqyZYskqWfPnsX6tmnTRrVq1dKOHTuUmZkpf3//so4bAACg0rJrBs7Hx0dNmjSRj49Pkfbvv/9eX3/9tUwmky2wHTx4UJLUpEmTEo/VoEEDWSwWJScnOzJuAACASqvUL7O/cOGCJk+erIMHDyo5OVl169bVrFmzbJdWT506JUkKCgoqcf/C9tOnT5d1zAAAAJVaqQPcsWPHtH79+iJtSUlJateunSQpJydHklS1atUS9y9sz87OLm3pEqWcTdGy2GVF2qK7RTt0zOhYx/Yvciw7xuLMepIUmxLr9GOWZ73r6ecnle/nWZ61ortFX/OzNvJn6Yp6lfm76Yp6fJ7u+buFeiXr5NdJgVUCr93RTqUOcCEhIfrpp5/k6empH3/8US+//LKmTZum7OxsjRw50rbGm4eHR4n7W63WIv92VGiNUPXu4NwnWh39pVLR9aJjr/0Xs5HrXYuRP08+S+fi83QuPk/ncuffLdQrLikpSZmZmU4bQ6kDnMlkkslkkiT16dNHderU0SOPPKJ3331XQ4cOtW27cOFCifvn5ubajgPgOhB9Ywlt58p/HAAAuzn8SoTWrVvr5ptvVmZmplJTUxUcHCzpyve4paWlSbryPXIAAAC4umsGOKvVqlmzZumpp55Sfn5+iX0Kn07Nz89X48aNJf35NOr/HuvQoUPy8vLSrbfe6si4AQAAKq1rBjgPDw999913WrdunW2x3sulpqbq8OHDMplMatCggTp37ixJ+u6774r13blzp9LT09WmTRvWgAMAACgjuy6h9u/fX5I0ffp0nThxwtZ+8uRJPf3008rPz9ejjz4qX19ftW/fXo0bN1ZcXJw+++wzW9/09HS9+OKLkqThw4c78xwAAAAqFbseYhgyZIji4+O1adMm9enTR+Hh4SooKNCePXuUnZ2trl276sknn5QkeXp6asaMGRo6dKiee+45ff755woODta2bdt07tw59e/fX927d3fpSQEAALgzuwKct7e3FixYoE8++UQrV67Uzz//LE9PTzVp0kQPPvig+vfvb1s+RJLCwsK0YsUKxcTEKD4+Xr/99ptuueUWPf3003r44YdddjIAAACVgd3LiHh5eSkiIkIRERF29W/UqJFiYmLKPDAAAACUzOFlRAAAAFC+CHAAAAAGU+o3MQAwntCotcXaUl69vwJGAgBwBmbgAAAADIYABwAAYDAEOAAAAIMhwAEAABgMAQ4AAMBgCHAAAEiK7hZd0UMA7EaAAwAAMBjWgQMAoFD0jSW0nSv/cQDXwAwcAACAwRDgAAAADIYABwAAYDDcA+dEvG8SAACUB2bgAAAADIYABwAAYDAEOAAAAIMhwAEAABgMAQ4AAMBgCHAAAAAGQ4BzQ7yQGQAA90aAA64jhG8AgD1YyNdd8UJmAADcFgEOuN4QvgEA18AlVAAAAIMhwAEAABgMAQ4AAMBgCHAAAAAGQ4ADUG5YJgUAnIMABwAAYDAsIwKgfLFMCgA4jBk4AAAAgyHAAQAAGAwBDgAAwGAIcAAAAAZDgAMAADAYAhwAAIDB2L2MSEFBgZYtW6ZVq1bp0KFDKigoUP369XXfffcpMjJSvr6+tr7bt2/XoEGDrnisvn37avbs2Y6NHAAAoJKyK8AVFBRo9OjRio2NlclkUqtWrVSlShXt2bNHMTEx2rRpkz766CP5+flJkhITEyVJt99+u2666aZixwsPD3fiKQAAAFQudgW4FStWKDY2VmazWe+//75q164tSUpPT9fo0aO1a9cuzZ8/XxMmTJAk7d+/X5I0adIktWnTxkVDBwAAqJzsugdu1apVkqRnnnnGFt4kKSAgQNHR0ZKktWvX2toTExPl6empZs2aOXGoAAAAkOwMcDVr1lTDhg0VFhZWbFtoaKgk6dSpU5KkvLw8JScnq2HDhjKZTM4bKQAAACTZeQn1nXfeueK2ffv2SZJCQkIkSb/99psuXryoevXq6c0339SGDRv03//+V4GBgerdu7eeeOIJ3XDDDU4YOgAAQOXk0DIiVqtVMTExkqRevXpJ+vMBhk2bNunjjz9W/fr11aZNG/3xxx/617/+pYcffljp6ekODhsAAKDysnsZkZK88cYb2rZtmwIDAxUZGSnpzwcY2rdvr7lz5yogIEDSpQcenn76aW3dulUvvPCC3nrrLQeHfknK2RQti11WpC26W7RDx4yOLd3+5V3vqseyYyzOrCdJsSmx5X4OV93/Oj+/K3HFz+56+iz5bl7/51fe9a6nz9Mdvp/XSy3qlayTXycFVgl02hjKHODmzp2r9957Tz4+PpozZ44tqE2dOlXjnGU4AAAgAElEQVQREREKCgqSv7+/rX9AQIBmzpype++9V998841OnTql4OBgh08gtEaoenfo7fBxLufoL5XKVi86Nrrcz+Fq3Pn8jP5dKe9619PPTjL++bn751ne9crz83T374oR6iUlJSkzM9NpYyj1JdT8/Hw9//zzmj9/vnx9fTVv3jy1a9fOtt3b21sNGjQoEt4K1a5dW82bN5fVarVdagUAAEDplGoGLisrS08++aS2bNmiG264QfPnzy8S3uwRGHhp+jAnJ6dU+wEAAOASu2fgzp07p4iICG3ZskV16tTR0qVLSwxv06dP15gxY3TmzJkSj3P06FFJfz61CgAAgNKxK8Dl5eVp5MiRSkhIUKNGjfTpp5+qSZMmJfbduXOnvv32W23cuLHYtgMHDmj//v2qUaOGWrRo4djIAQAAKim7AlxMTIx2796tOnXqaPHixVedPRswYIAk6c0331RycrKtPT09XVOnTlVBQYEiIyPl4+Pj4NABAAAqp2veA3f27FktXrxY0qUnSWfMmHHFvrNnz9bDDz+suLg4rV+/Xv369VPbtm3l5+en+Ph4ZWVlqU+fPnrsscecdwYAAACVzDUD3N69e3XhwgVJUkJCghISEq7Yd/bs2fL09NTcuXO1fPlyrVixQrt27ZKnp6caNWqk/v3766GHHpKHh4fzzgAAgFIIjVpbrC3l1fsrYCRA2V0zwHXp0kVJSUmlOqiHh4ceeeQRPfLII2UeGAAAAErm0Ku0AAAAUP4IcAAAAAZDgAMAADAYh15mD6BsuIkaAOAIZuAAAAAMhgAHAABgMAQ4AAAAgyHAAQDsEt0tuqKHAOD/I8ABAAAYDE+hAnBbzBi5QPSNJbSdK/9xAJUcM3AAAAAGwwwcAMNjXT0AlQ0zcAAAAAZDgAMAADAYAhwAAIDBcA8cAKfjnjQAcC1m4AAAAAyGAAcAAGAwXEI1MC5TAQBQORHgALg33hwAwA1xCRUAAMBgCHAAAAAGQ4ADAAAwGAIcAACAwRDgAAAADIYABwAGFd0tuqKHAKCCEOAAAAAMhnXgAMDIWOcOqJSYgQMAADAYAhwAAIDBEOAAAAAMhgAHAABgMAQ4AAAAgyHAAQAAGAwBDgAAwGAIcAAAAAZDgAMAADAYAhwAAIDB2P0qrYKCAi1btkyrVq3SoUOHVFBQoPr16+u+++5TZGSkfH19i/Tft2+f3n77be3bt0/Z2dlq1KiRhgwZor59+zr9JAAAACoTuwJcQUGBRo8erdjYWJlMJrVq1UpVqlTRnj17FBMTo02bNumjjz6Sn5+fJCkuLk6jRo2SxWJRu3bt5Ofnp61bt2rixIk6ePCgnnrqKZeeFAAAgDuzK8CtWLFCsbGxMpvNev/991W7dm1JUnp6ukaPHq1du3Zp/vz5mjBhgi5cuKBJkyZJkv71r3+pY8eOkqTff/9dEREReuedd3TPPfeoZcuWLjolAAAA92bXPXCrVq2SJD3zzDO28CZJAQEBio6OliStXbtWkrR69WqdOXNGffv2tYU3Sbr55ps1YcIESdLixYudMngAAIDKyK4AV7NmTTVs2FBhYWHFtoWGhkqSTp06JUnasmWLJKlHjx7F+nbv3l1eXl7avHlzWccLAABQ6dl1CfWdd9654rZ9+/ZJkkJCQiRJv/32mySpSZMmxfr6+/srODhYx48f1+nTpxUYGFjqAQMAXCs0am2xtpRX76+AkQC4EoeWEbFarYqJiZEk9erVS5KUlpYmSQoKCipxn8L206dPO1IaAACg0rJ7GZGSvPHGG9q2bZsCAwMVGRkpScrJyZEkVa1atcR9Ctuzs7MdKW2TcjZFy2KXFWmL7hbt0DGjY0u3v7vXu5bYlFinHtOdz8/dvyvUc269qx7LjrFc7+fn7vWuxdm/O6+XWtQrWSe/Tgqs4rwrj2UOcHPnztV7770nHx8fzZkzRwEBAZIkLy8vWa1WeXh4lLif1Wot8m9HhdYIVe8OvZ1yrEKO/iGvbPWiY6PL/Ryuxp3Pz+jfFepRj3qXHa8cf7eU9+8x6hWXlJSkzMxMp42h1JdQ8/Pz9fzzz2v+/Pny9fXVvHnz1K5dO9t2Pz8/Wa1W5ebmlrh/YbvJZCrjkAEAACq3UgW4rKwsPf7441q+fLluuOEGffDBB+ratWuRPsHBwZL+vBfuf13rHjkAAIDSuF6ukpQnuwPcuXPnFBERoS1btqhOnTpaunRpkZm3Qo0bN5YkJScnF9uWmZmpU6dOKSAggCdQAQBwU5UxUJU3u+6By8vL08iRI5WQkKBGjRrpgw8+sC0b8r86d+6sr776St9++22x2bmNGzeqoKCgWDsAAMC1sMTNn+wKcDExMdq9e7fq1KmjxYsX2x5YKEnv3r31+uuva9WqVerZs6ctrKWmpur111+Xh4eHhg0b5pTBAwCAikOgqjjXDHBnz561vfoqICBAM2bMuGLf2bNny9/fX9OmTdP48eM1atQotWvXTtWqVdNPP/2knJwcPfXUU2ratKnzzgAAAKCSuWaA27t3ry5cuCBJSkhIUEJCwhX7zp49W9Kl12gtXrxYb7/9tvbs2SOr1Sqz2axhw4apT58+Tho6AABA5XTNANelSxclJSWV+sDh4eH64IMPyjQoAAAAXJlDr9ICAABA+SPAAQAAGIxD70IFALgeT/oB+F/MwAEAABgMAQ4AAMBguIQKAACcL/rGEtrOlf843BQzcAAAAAZDgAMAADAYAhwAAIDBEOAAAAAMhgAHAABgMAQ4AAAAgyHAAQDg5qK7RVf0EOBkBDgAAACDYSFfAADcBO/NrTyYgQMAADAYAhwAAIDBEOAAAAAMhgAHAABgMAQ4AAAAgyHAAQBQAVibDY4gwAEAABgM68ABAOBCrM0GV2AGDgAAwGAIcAAAAAZDgAMAADAYAhwAAIDBEOAAAAAMhgAHAABgMAQ4AAAAgyHAAQAAGAwBDgAAwGAIcAAAAAZDgAMAADAYAhwAAIDBEOAAAAAMhgAHAABgMAQ4AAAAgylzgFu5cqXMZrO2b99ebNvx48dlNpuv+M/AgQMdGjQAAEBlVqUsO+3atUvTpk274vbExERJktlsVpMmTYptb9CgQVnKAgAAQGUIcBs2bFBUVJSys7Ov2Gf//v2SpMjISD3wwANlHx0AAACKsTvAnThxQm+88YZWr14tPz8/BQYG6vTp0yX2LZyBa9GihXNGCQAAABu774GbM2eOVq9erZYtW2r58uVq2LDhFfvu379fJpOJS6UAAAAuYPcMXMOGDTVz5kw98MAD8vS8cu47e/asjh07phYtWmjRokVavXq1jhw5ourVq+vuu+/W2LFjVbt2bacMHgAAoDKyO8CNHDnSrn6F978lJCTowIEDateunUJCQrRv3z599tln+v777/Xxxx9fdQYPAADgehXdLbqih1C2p1CvpvD+t8aNG2vBggWqX7++JCk7O1vPPfec/vOf/2jixIlauXKlU+qlnE3RsthlRdoc/WCjY0u3v7vXu5bYlFinHtOdz8/dvyvUox71Slkv+sYS2s6Vup49tYz8e/p6PL/S1uvk10mBVQLLNqASOD3ADRs2TL169VK1atUUEBBgazeZTJo+fbp+/vlnJSQkaPfu3WrdurXD9UJrhKp3h94OH+dy5Z2sjV4vOjb6uvi/kULufH5G/65Qj3rUq7h618vvMck1Y3HFMUOj1hZrS3n1/jLVS0pKUmZmpjOGJckFb2Lw8vJS/fr1i4S3Qn5+furYsaOkS5dYAQAAUHrl/iqtwMBL04c5OTnlXRoAAMAtOD3AzZs3T+PHj1dSUlKJ248ePSpJCgkJcXZpAACASsHp98AlJSVpw4YNatiwocxmc5FtZ86cUVxcnLy9vdWhQwdnlwYAAKgUnD4DN2DAAEnSokWLtGPHDlt7VlaWnnnmGWVmZuqhhx5SUFCQs0sDAABUCk6fgbvrrrs0fPhwLVq0SIMHD1Z4eLhq1qyp7du3KyMjQ23bttWUKVOcXRYAAFRmV1mSxR05PcBJUlRUlFq1aqUlS5YoMTFRFotFN998syIjIzV06FB5e3u7oiwAAEClUOYAt3jx4qtu79Onj/r06VPWwwMAAOAKyn0ZEQAAADiGAAcAAGAwBDgAAACDIcDBYdfT+/UAAKgMCHAAAAAG45JlROCeQqPWFmtLefX+ChgJAACVGzNwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAymSkUPAAAAwHCibyyh7Vy5lWcGDgAAwGAIcAAAAAZDgAMAADCYMge4lStXymw2a/v27SVuP3z4sJ5++ml17dpVrVq1Ut++fbVkyRJZLJYyDxYAAABlDHC7du3StGnTrrj9119/1UMPPaS1a9eqbt266ty5s06cOKFp06Zp8uTJZR4sAAAAyvAU6oYNGxQVFaXs7OwSt1utVk2ePFmZmZmaNWuW+vXrJ0lKT0/XsGHDtGbNGt1zzz3q3bu3YyMHykl0t+iKHgIAAEXYPQN34sQJTZ48WePGjZPFYlFgYGCJ/eLi4pSUlKT27dvbwpskBQQE6IUXXpAkLV682MFhAwAAVF52z8DNmTNHq1evVsuWLTVjxgxNnz5dp0+fLtZvy5YtkqSePXsW29amTRvVqlVLO3bsUGZmpvz9/R0YOuBcoVFri7WlvHp/BYwEAICrs3sGrmHDhpo5c6ZWrFghs9l8xX4HDx6UJDVp0qTE7Q0aNJDFYlFycnIphwoAAACpFDNwI0eOtKvfqVOnJElBQUElbi9sL2n2DgAAANfm9Fdp5eTkSJKqVq1a4vbC9is9BFFaKWdTtCx2WZE2R286j44t3f7UK/0xHa131f2v4/O7Hs+NetSjnvHr8Xv6+q/Xya+TAquU/PxAWTg9wHl6Xroq6+HhUeJ2q9Va5N+OCq0Rqt4dnPtEa3k/degO9a6nJzXd+fzc4btCPepRr2LqXS+/xyTjf5ZlqZeUlKTMzEyn1XD6mxhMJpMk6cKFCyVuz83NLdIPAAAApeP0ABccHCzpyve4paWlSbryPXIAAAC4OqcHuMaNG0v682nUy1mtVh06dEheXl669dZbnV0aAACgUnB6gOvcubMk6bvvviu2befOnUpPT1ebNm1YAw4AAKCMnB7g2rdvr8aNGysuLk6fffaZrT09PV0vvviiJGn48OHOLgsAAFBpuOQp1BkzZmjo0KF67rnn9Pnnnys4OFjbtm3TuXPn1L9/f3Xv3t3ZZQEAACoNpwc4SQoLC9OKFSsUExOj+Ph4/fbbb7rlllv09NNP6+GHH3ZFSQAAgEqjzAHuWi+kb9SokWJiYsp6eAAAAFyB0++BAwAAgGsR4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYKpU9ADgJqJvLKHtXPmPAwCASoAZOAAAAIMhwAEAABgMAQ4AAMBgCHAAAAAGQ4ADAAAwGAIcAACAwRDgAAAADMZl68B98cUXmjJlyhW3P/7443rqqadcVR4AAMBtuSzA7d+/X5J05513KiAgoNj2Zs2auao04HwsVAwAuI64LMAlJiZKkl555RXVrl3bVWUAAAAqHZfdA/frr78qMDCQ8AYAAOBkLglwqamp+uOPP9SiRQtXHB4AAKBSc8kl1ML732rVqqVp06Zp8+bNOnHihOrWrasHHnhAkZGR8vX1dUVpAAAAt+eSGbjC+99WrlypNWvWqFGjRmrVqpVOnjypmJgYDR06VBcuXHBFaQAAALfn0hm4Pn36aMaMGTKZTJKko0ePasyYMdq1a5fmzJmjqKgoh2ulnE3RsthlRdqiu0U7dMzo2NLtTz3qXY+1qEc96lWeevbUKu34Ha131f2v48/SVfU6+XVSYJVAh457OZcEuJiYGKWmpurmm2+Wj4+Prf2mm27Sq6++qr/97W9avny5JkyYIG9vb4dqhdYIVe8OvR0dchGO/uCoRz13qEU96lHPveqV9/ivxuifZVnqJSUlKTMz02k1XHIJ1dfXV40aNSoS3go1a9ZMISEhys7OVkpKiivKAwAAuLUKeZVWYOClKcScnJyKKA8AAGBoTg9wmZmZeu655zR+/Hjl5+eX2Ofo0aOSxBpxAAAAZeD0AFetWjV98803Wr9+vX7++edi2zdv3qyMjAw1adKEAAcAAFAGTg9wHh4e6t+/vyRp2rRpOnnypG3b77//rhdffFGS9MQTTzi7NAAAQKXgkqdQR48ere3bt2vHjh2699571aZNG0lSfHy88vLyNHz4cN13332uKA0AAOD2XBLgqlatqg8//FAffvih1qxZo/j4ePn4+Kh169aKiIhQr169XFEWAACgUnBJgJMkHx8fjRw5UiNHjnRVCQAAgEqpQpYRAQAAQNkR4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYAhwAAAABkOAAwAAMBgCHAAAgMEQ4AAAAAyGAAcAAGAwBDgAAACDIcABAAAYDAEOAADAYFwa4H788UcNGTJEHTp0UHh4uCIiIrRlyxZXlgQAAHB7LgtwK1eu1PDhw7Vr1y6FhYXp9ttv165duxQZGanly5e7qiwAAIDbq+KKg546dUovvPCCqlevrk8++URNmjSRJO3du1fDhw/Xyy+/rG7duql27dquKA8AAODWXDIDt2TJEuXl5WnYsGG28CZJYWFhioyMVG5uLrNwAAAAZeSSAFd4n1vPnj2LbbvnnnskSZs3b3ZFaQAAALfn9EuoVqtVBw8elKenpxo2bFhse2hoqDw9PXXw4EFZrVZ5eHg4ewgAAOB/Rd9YQtu58h8HnMLpM3Dnzp1TXl6eatSoIR8fn2Lbq1Spopo1ayonJ0dZWVnOLg8AAOD2PKxWq9WZBzx+/Li6deumevXqaePGjSX26d69u/773/9q8+bNZX6QYe/evbp48aI8PT1lMpkcGTIAAIBLZWdny2KxyNvbW2FhYQ4fz+mXUD09rz2p54zMWFBQIEmyWCzKzMx0+HgAAACuVphfHOX0AFc4G5abm3vFPoXb/Pz8ylzH19dXubm58vLykq+vb5mPAwAA4Gq5ubkqKChwWmZxeoDz9/eXyWRSRkaG8vPzVaVK0RL5+fnKyMiQr6+vbrjhhjLXad68uaNDBQAAMCSnP8Tg4eGhRo0aqaCgQCkpKcW2Hz58WBaLpcj6cAAAALCfS9aB69y5syTp22+/LbatsK1r166uKA0AAOD2XBLgHnzwQfn6+ur999/XL7/8Ymvft2+fFi5cqKpVq+rRRx91RWkAAAC35/RlRAotXbpUL730kry9vdWxY0dZrVbFx8crPz9fM2fOVL9+/VxRFgAAwO25LMBJ0vfff6+FCxcqMTFRPj4+MpvNeuKJJ9SpUydXlQQAAHB7Lg1wAAAAcD6X3AMHAAAA1yHAAQAAGAwBDgAAwGAIcAAAAAZDgAMAADAYAhwAAIDBEOAAAAAMhgAHAABgMAQ4AAAAg6lS0QNwlYyMDNWsWbNI25EjR7Rt2zZlZGQoICBAbdu2VWhoqMO1CgoKtGvXLp06dUq1a9fW7bffLk/PK2fjHTt2KDU1VX/9618dqpudna2DBw/qzJkzys7OltVqlZ+fn4KCgtSoUSOZTCaHjn8t58+f17Jly7R//37l5+erZcuWeuihh1SrVi2n17JYLPr666+VmJio/Px8tWjRQr1795aPj0+ZjjdkyBC1aNFCEydOlJeXl5NHWzZ79uwpcn7h4eEur/nNN98oKSlJY8eOdXktZzt79qxiY2N16tQphYSEqEuXLqpRo8YV+3/77bf69ddfHT5Xi8WiX375RcnJyUpPT1deXp6qVKmiG264QXXq1FHLli0VEBDgUI3L/frrr9qyZYsOHTp0xXp33XWX/P39nVIvIyNDW7duvWq92267zSm1rsWZ38/y/j2dnp6uX375RV26dLG15eXlafXq1UX+HmrXrp369Onj8M+vov48lIfy/iwl6cKFC9q2bZsOHDigM2fOKCsrS5Jsf8c2adJEbdu2dfnfs1fjdq/S+uWXX/TPf/5TOTk5Wr9+vaRLQef555/X2rVri/W/88479dJLL6lu3bplqrd9+3ZNnjxZx48ft7XVrl1bEyZMUN++fUvcZ9KkSfrPf/6j/fv3l6nmN998o48//lg7d+6UxWIpsY+Xl5fatGmjxx57TF27di1THUlKS0vT7NmztWXLFnl5eemvf/2r/u///k/Hjh3T4MGDderUKRV+hTw8PFS9enXNnTu3TO+7zc7O1sKFC4vUeuSRR5SRkaFhw4bpwIEDRWrVq1dP8+bNU9OmTUtdq2nTpvLw8FCrVq302muvqX79+qU+RmlZLBatWrXKdn79+vVTly5dlJubq3HjxmnLli1F+rdq1Upz5sxRSEiIy8bk6Hexonz11Vd67rnnbL9UJalq1aoaNWqUHn/88RL3cfRcrVarPvnkE7399tvKyMgots3Dw0PSpe9m27ZtNX78eLVt27ZMtaRLf2m9+OKL2rBhg61G4fEv/7Xt4eEhf39/DR8+XKNGjSrz/5BcuHBBMTExWrx4sfLz84tsu7y2JNWvX19jxoxRv379ylTLXs76fpb37+mlS5fqlVdeUe3atfXdd99JkpKTk/XEE08oNTW12M+vVq1aeuWVV9S5c+dS15Iq5s9DeSnvzzIzM1Nz5szRv//9b124cEGS9L8xqfDPgclk0oABAzRu3Dj5+fmVqZ4j3GoG7pdfftGgQYOUm5urFi1a2NrHjBmjrVu3qlq1aurevbvq1q2r06dPKz4+Xj/88IMGDx6sFStWlHrmKDk5WSNGjFBOTo6aN2+uW265RQkJCfr99981efJk7dq1S88//7zTzs9qtWrKlClas2aNrFaratWqpQYNGigoKEhVq1aV1WpVbm6u0tLSdOjQIcXHx2vbtm3q37+/oqOjbV86e507d04DBgzQsWPHbG0LFy7U2bNndfbsWZ08eVJdu3bVX/7yF/n4+GjTpk1atWqVxowZo1WrVumWW26xu1Zubq4GDx6s/fv32/6w7NmzRydPntSJEyeUlJQks9msPn36yMfHR5s3b9ZPP/2kyMhIrVq1SkFBQaU6N+nSL7jdu3frgQce0BNPPKFhw4aVeUbvWiwWi0aNGqUffvjBdn7r1q3Tc889p8OHD2vz5s0KCgpSt27d5OPjo61bt2r37t0aOnSo/v3vf5f6/yi/+OILu/qlpqaW2N/R2WFX2rNnjyZNmqSCggL16NFDoaGh2rt3r7Zt26a5c+dqz549mjt3rtN/ls8884y++OIL1axZU507d5aXl5cSEhJ06tQp/e1vf1NoaGiRP3cRERGaNGmSHnvssVLXysnJUWRkpBITE9W5c2d16NBBVapU0fbt2/Xtt99q0KBB+stf/qJDhw5p69at+u677/TWW29p//79mjt37lVnlkpSUFCgMWPGKC4uTo0bN1bbtm3l5eWlXbt2KSEhQQMGDFC7du1s9Xbv3q2oqCjt27dP//znP0t9fuX5/Szv39Pr1q3TtGnT5OPjo7vvvlvSpZ/niBEjdOzYMYWFhenhhx9WvXr1lJaWps2bN2vdunUaO3asPv30UzVr1qxU9Srqz0N5KO/PMjMzUwMHDtTBgwd1ww03qGvXrmrYsGGJf8cmJydr69atWrRokeLj4/XRRx85bRbcblY38vjjj1vNZrP1X//6l63tm2++sZrNZmu/fv2saWlpRfrn5eVZp02bZjWbzdZp06aVut6ECROsZrPZ+u6779raLBaLdcWKFdbbb7/d2rRpU+uzzz5bbL+JEydamzZtWup6S5YssZrNZusDDzxg3bFjxzX7b9++3dq3b19r06ZNrZ999lmp682cOdNqNputzz//vPXMmTPW33//3Tpy5Ehrs2bNrC1atLBOmTKl2D6fffaZ1Ww2W5955plS1Xr77betZrPZOmrUKOvevXutP/74o/XBBx+0tmzZ0hoWFmYdMWKE9eLFi0X2mTt3rtVsNlunT59e6nMzm83WSZMmWT///HNr69atrU2bNrX26NHD+umnn1pzc3NLfbxr+fjjj61ms9n697//3frVV19ZV6xYYe3Ro4f19ttvt7Zp08bav39/6x9//GHrf/HiReuzzz5rNZvN1jlz5pS6ntlstjZt2rTM/1zPRo8ebW3atOn/a+/c42LO/j/+mikpt6xrvutamBHSdnXJrRSStuzKbq6hdVlrI1p8XVIui6UQhb6iKMolJaFY0SoSTUtSqRRFqpXu0+X8/ug3n0djZtJnqk+Znefj4Q/nc3n1OZ/zPp8z57zP+01CQ0OFyu/fv0/GjRtHuFyu2PYird0RQsjVq1cJh8Mhjo6OpLS0lCqvrKwkW7ZsITo6OkXFRzcAACAASURBVCQrK4sqv337Nhk/fjzhcrkkJiaGtt6xY8cIh8Mhfn5+IscuXrxIhg0bRmJjY6my/Px8smDBAsLlcklQUBBtPX9/f8LhcMjevXtJTU2N0LHDhw+TESNGkKSkJKosOTmZWFhYEC6XS27cuEFbj8n2yXQ/PWfOHDJ8+HChPjogIIBwOBzi4OAg9pqwsDDC4XDIL7/8QluvNeyBKZiuy99//51wOBzi5OQkZOeSKC0tJU5OToTD4ZD9+/fT1msqMrWEamhoiD59+gj9WtuxYwfOnj0Lf39/fPPNNyLXEEIwdepUVFdX4/bt27T0jIyM0LlzZ4SHh4sce/r0KZYsWYKPHz/Czs4OTk5O1DFpp64tLS3x7t07XL9+XcS/TxKFhYWYPn06vv76a1y6dImWnpmZGVgsFq5fv07N3pWUlGD8+PGoqKjAtWvXMGjQIJHrzM3NUVZWhjt37jRay8LCAsXFxYiIiKB+Kb5//x7Gxsaorq7G5cuXRZZKa2trYWpqChaLhcjISFrPxuVyYWlpib179yIzMxO7d+9GVFQUNQW/YMECzJw5E3369KF1X0nMmjULubm5iIiIoH6lZWZmwtzcHIQQBAQEQFtbW+gaPp8PY2NjdOnSBdeuXaOlt3XrVgQGBlLLxIMHDxZ73sOHD5GdnY3vvvtOqHzHjh209JhkzJgxUFNTw+XLl0WOvX79GgsWLEBubi7Mzc2xf/9+6lhTloxsbW3x+vVrREZGisxkVFVVwdTUFAYGBti7dy9Vnp6eDktLS4wdOxbHjx+npSdY1gsNDZV4vFevXvjf//5HlZWUlMDExAQDBgxAYGAgLb1Zs2ahuLgYN2/eFDtTP3XqVAwZMgQeHh5U2fv372FmZoaRI0fC19eXlh6T7ZPpflpbWxtaWlpCdbJ161YEBQUhPDxcot/17NmzkZOTg7/++ouWXmvYA1MwXZfGxsZQUFDAjRs3Gj2LXVNTg2nTpoHNZlNuW0whU0uo5eXl6N+/v1CZwJeDw+GIvYbFYmHo0KG4e/cubb0PHz6IHRQCwIgRI3DixAksWrQIPj4+6NOnD+bPn09boz5ZWVmYMGFCowdvANCtWzcYGBjQbsgA8PbtW0yaNEmoQ+/UqRO0tbURGxuLvn37ir1OXV2d1uANqFsqGT9+vNDHsWfPntDW1sajR4+grq4ucg2bzQaXy5Xq3dVn4MCBOHbsGO7du4fDhw8jMTERbm5ucHNzg7a2NkxNTaGjowNNTU2plyEyMjIwZswYoSn2gQMHQktLCzweT+xUv5KSEkaOHCnVu3NxccGUKVPw3//+F0lJSTA2Noa9vb3Ix3n9+vXIzs5utgGbJH+bxsBiseDp6fnZ84qLiyX6lvXt2xenTp3Cjz/+iGvXrqF3795CH2VpSU1Nhb6+vtj3365dO2hqaoq8J3V1dRgaGiIpKYm2XnZ2NsaNGyfx+IABAxAbGytU1qlTJ+jp6eHhw4e09TIyMjB27FiJbhZDhgxBXFycUFnPnj1haGiIhIQE2npMtk+m+2k2m42OHTsKlSkrKwMAevToIfG6nj17IjU1lbZea9gDAGzZskXqa1ksFlxcXD57HtN1WVBQgMmTJ9NyQVBQUICmpibtb15zIFMDuL59+yIxMRHV1dVQVKx7NIEvXFpaGrS0tESuqa2tRVJSklQ+VKqqqnjz5o3E41paWnBzc8PKlSvx+++/o2fPnpg2bRptHQFdu3ZFQUEB7evy8vKkGnh06tQJWVlZIuXW1tbo1q0bioqKxBpRRkZGg7ufxKGsrIz8/HyRcmNjY1RVVeHDhw/o1auXyPHc3FwRA5eW8ePHY/z48Xj8+DFOnz6NqKgoPHnyhPpAKSgooFevXlBTU4O/vz+te7PZbJSUlIiU6+np4fXr1/jw4QN69+4tcvzDhw9o3769VM8zYcIEXL16Fc7Ozjhw4ABu376NPXv20PJNpEtaWhpev34t4mjfGBrro9m9e3dkZGRIPN6/f394enpi4cKF8PHxQe/evbFw4UJaf8un1NbWCjnAf0phYaHY96uoqIji4mLaeh06dKD8QcXVS3p6utj6LSoqkrixqSGUlJSQnp4u8XhOTg74fL5IeVVVFSorK2nrAcy1T6b76SFDhuDBgwcoKCig/KoNDQ3h6+uL+/fvw8zMTOSaoqIixMXFSfXsrWEPQN1muqKiIur/dOy9sQM4putSTU0NKSkptK6pqalBUlJSi0Rf+BwyFQdu2rRpePv2LTZs2ICqqioAwIwZM9CnTx+4urqKdLC1tbXYuXMncnNzYWxsTFtPW1sbz58/b3DpdeLEidiyZQtqamqwfv162kth9dHR0cHjx49x9erVRl8TFBQEHo8HQ0ND2nqGhoZ48eIFzpw5I1RuaWmJ/fv3ix28HT9+HC9fvqS9C1VbWxs8Ho/aZSTAzs4O586dEzt4u3r1KpKSkqCrq0tL63Po6Ojg4MGDiI2NxaFDhzBr1iwMHjwYhBDk5OTgyZMntO85fPhwxMfHIzExUah83bp1iI6OFjt4e/jwIRISEpoUskFVVRVubm74448/kJGRASsrK/j5+Ul9v89x9epVmJiYgBCC/v3749ixYzhx4kSj/jV2mdHAwAAvX75scBCtpaWFP/74A2w2G3v27MGJEydoDyjro6mpieTkZLFL9YmJiUhMTISGhoZQeVZWFh4+fChS3hj09fWRm5sLd3d3kWNBQUFIT08XWXK/efMm4uPjRcobw6hRo5CZmSm2TqOiopCUlCQyS8zj8fDgwQPajuL1YaJ9Mt1P29jYoKysDIsXL6YGVsbGxtDX14eLi4vIjOWbN2+wfPlylJSUSLWrtzXsAajbYKCrqwtCCPr164fdu3c3+t+uXbsapcF0XU6ZMgXp6enYtm0bysrKPnt+RUUFNm/ejNevX2PKlCm09ZqKTPnAlZeXY86cOUhJScGAAQNga2uLkSNHorCwEJs2bYKSkhIsLCzw9ddfIy8vDxEREXj16hV69OiBK1eu0I7dlJCQgLlz5wKoG9QYGhpK3B3l4eEBDw8PsFgsdOnSBR8/fqTte5CZmYnvvvsOZWVlMDQ0hImJCYYMGYKePXtCRUUFLBYLFRUVyM/PR1paGiIjI3H//n106NABgYGBtD8kL1++xPfff4+KigoMGzYMBw8elBhuIywsDL6+vkhMTISysjIuXLhAS+/JkyeYN28eCCEwMTHB5s2bxQ5qACA2Nhbnz5/H9evXoaCggLNnz2LUqFG0nq2+D1xjKS8vR05ODv755x/a4SGioqKwbNkyqKioYM6cOfjpp58ktrfU1FRcuXIFvr6+qKqqwvHjx6XeEl+fvLw8bNq0CdHR0TAwMMDu3bvh7u7e7H4wNTU1+Omnn3D//n2sW7cOS5YsabZ7A3XtctasWeDz+dDT08P48ePx008/iT334sWL2LJlCwghUFRURHV1tVTPGhkZiVWrVkFJSQlLlizB5MmT0bFjRzx69Ahubm4oKiqCq6srvv/+e7x79w7nzp2Dv78/Pn78iF27dsHa2pqWXnJyMmxsbFBVVQVdXV1MnDgRHTt2RHx8PK5duwY2m43Tp09DT08PycnJcHV1xePHj6GgoAA/Pz+JS4aSiIuLw8KFC0EIwcyZMzFp0iRK79SpU6iqqsLhw4cxZcoUvHr1CkeOHEF4eDiqq6vh6emJSZMm0dITR0u1T6b7aaBu6Tc0NBSKioqYOHEiRo4cCWVlZRw9ehQlJSXgcrnUd+jp06eoqamBrq4uTp8+Ta0eNZbWsAcBFRUVWLBgAf7++2+4uLhg9uzZUt9LEkzWZXFxMebOnYuUlBR07twZo0ePFvrGAhDahXr//n0UFRVBXV0d586dQ5cuXZr9+RtCpgZwQN306a5duxASEiJyrP5yhOCxuVwu9u/fL9WvZAAICQmBs7MzysrKoKmp2eBGAT8/P+zZswfV1dVgsVhSGU5ycjKcnJyQkpLy2SUnQgiGDBmCnTt3il0+bqyeo6Mj0tPTER0dLXGaePPmzbhw4QJ69OiBffv2SRUHLjo6Go6Ojvj48SPu3LkjcQC3YcMGBAcHQ1lZGc7OzlKFvJBmANdULly4AGdnZ9TU1CAiIkKiD+G6detw9epVsFgsODg4YNmyZc36d/j7+2Pfvn1gs9no3r07srOzm92R+cOHD5g5cyaKi4tx48YNie9SWh48eIANGzYgNzcXI0eORFBQkMRzIyMjsWnTJnz8+FFquwOAAwcO4Pjx4yJ2RwjB7Nmz4erqCgC4f/8+Fi9eDEVFRaxZs0bqAezt27fh5OSEkpISoX6rffv22L59O9Xu4+LiMH/+fAwaNAi7du2iPXgTEBAQgJ07d1L9k0CPxWLh119/pfwbY2NjsWjRInTp0gXbt2/H9OnTpdKTREu0T6b7aaBusOTl5UWFQpHkVtC+fXt8//33cHR0lDoobGvYg4Dc3FxYWVmBEIKbN2/Sdp9pDEzWpSAO3IULF6g4cOJsHqhz/bGyssKaNWugqqoqlV5TkLkBnIDc3FzcuHEDz58/x6tXr1BSUoLKykqoqKigW7duGDp0KCZMmICxY8fSjpn0Kf/88w9u3ryJ2tpa/Pjjjw2em5KSAk9PT8TFxSE6OlpqzejoaNy7dw9paWl4//49ysvLwWaz0aFDB/Tq1QtDhgzB+PHjYWBgQDv+mziSkpKgqakp8XhsbCzy8/NhYmLSpICGfD4fUVFRMDU1lXjOzZs3kZGRgW+//VbqILceHh7gcDgN6rQE7969Q3h4OBYtWiTxnMDAQDx9+hSzZ89usYj3ghhYCQkJzdKJiyM4OBgbNmzAnDlzsH379ma/f3V1NWJjY1FRUfHZ5Yv8/HycOnUKcXFxOH/+vNSaMTExCAwMRGpqKmprazFo0CBYW1sL6b979w537tzBhAkTmryLuaioCOHh4UhJSaH0zM3NhXx2S0pK8P79e7E7wumSkZGBS5cuCelZWVkJ2X5hYSGSkpKgr68vtX/m52iJ9tka/TRQ9yNYEHeuuLhY5Ds0evToZokf1hr2IODMmTPYsWMH7Ozs8NtvvzX5fpJgqi6Butm4uLg4pKWlIS8vT+w3Vl9fv1UC+AqQ2QHcvwGBT5Yg3U27du3QuXNn9OnTh9qpI0eOJGpraxEREYGysjLaS3yNJT09He3atWMk0wWTyG2v5WGifcppHmpqahAfHw8VFRXG0qzJkeEBXFVVFYqKivDVV199NrXMhw8fUFZWJnU6LQEFBQUS8yOqq6s3y0wYALx48QIeHh6Ijo6mpnjrw2KxoKGhgalTp2LRokVS/yLx8PDAsGHDYGJi0tQ/+bMEBwdj8ODBGDFiRItrtRUKCwuRn58vlMe2V69ezZpHsyHevXuHFy9eULlXm3uZs6Vgsl1+SmNsT11dHdOmTWuS7TWGlsydy3Se5ZbUa6t9S3PlxP63wHQO4i8BmRvAFRYWYufOnYiMjASfz0f79u1hbGyMX375ReISw/r16xEWFiZVvCagzkH94MGDDU7zd+zYEaampli5cmWTZiMCAwPh4uKC6upqqKiogM1mU/nvRo8eDQUFBaSnp1MhD1RVVeHh4QF9fX3aWoJ8oXZ2dli7di1th1C6Wu3atcOGDRsoh2NZJCkpCX5+frh79y4KCwvFntOtWzdMnDgRixYtwtChQ5ukFxsbi3v37kFRURHm5ubgcDiora2Fq6srgoKCUFNTQ51rbm6O7du3M58OhiZMtsv6MGl7QOvkzmUyzzJTem21b/kSAum2BQjDOYi/JGRqAPfx40fMnj0bWVlZYLFY+Oqrr6i4aSoqKti1a5dYh9umGNLBgwfh5eUFBQUFDBgwAAoKCsjMzASfz4eZmRm6du2K9PR0JCYmUr8Y9u7dC3Nzc9paPB4Ptra2GDx4MFxdXamNCcnJydi6dStKS0sRFBSEDh06ICsrCwEBAfDz84OSkhIuXbokMWq1JASZDwTBjrdv3y5ViAK6WkZGRnB2dsbXX3/dIlqthbu7O44dOwZCCBVTrmfPnpQfkWB3U15eHmpqasBms+Hg4CBxR9nn2LJlCy5cuEA53CoqKsLNzY2aRVJRUcE333wDJSUlJCQk4MOHD9DS0sKZM2fadN5EJtulAKZtT1zuXBaLReXO9fPzE8mdm56ejgEDBkiVO5fQzLNcWFgIFosldZ5lJvWY7ls+DXgsCU9PT8TExMDPz0/IIV/aAX9bIzExEU+fPkVFRQXU1NRgaGgoVay0jRs3UjmIR4wY0WAO4rdv34LFYkmdg/hLQ6YGcHv37sXJkydhamqKnTt3okuXLsjNzYWbmxtCQkLAZrOxY8cOzJo1S+g6aQdw9+7dg729PUaPHo09e/ZQS1CCsCXPnj2jwpOUl5fj4sWLcHNzQ2VlJc6fP08FGW4sP//8M2JjY6mI2vXJz8/H9OnTYWtrizVr1lDlgvAH1tbW2L17Ny09LpeLKVOmoGPHjrhy5QpYLBbMzc2xcuVKqXftNqQ1btw4VFdX48GDB2jfvj0WLVoEOzu7FtnV5OXl1aTr6WYcuHr1KtatWwc1NTU4Ojpi8uTJEj+yJSUluH37Nvbv34+8vDwcPHhQbMDKhrhy5Qp+++03/Oc//8HcuXNRUlICf39/sNlssNlsdOnSBSdOnKB2whYXF2P9+vWIiopq850fk+1SANO25+fnh507d2LEiBFYunQpSkpK4OXlhcLCQrDZbGhoaMDb2xudO3cGUOfA7uzsjAsXLmDFihX49ddfaemdPXsWrq6u4HA42LZt22eXY+Pj47F9+3akpqZKFT6CST2m+xbBDLE0sFgs2itBTPdlAubOnQsjIyOsWLFCqDwzMxPr1q3Ds2fPhMoVFRVhY2MDJyenRm9+CQsLg6OjIywsLODi4kItpfP5fOzYsQNhYWEIDg6mVrX+/PNPbNu2De/fv4ePjw9Gjx5N65laqy6lRaYGcFOnTkVpaSkiIyNFHInPnz+P7du3g8Viwc3NTeiDKO0AbsmSJUhKSsLNmzepjlRASUkJzMzMMG3aNGzdupUqf/LkCebOnQtTU1McPHiQlp6RkRG4XC68vb3FHl+6dCkyMzNFgo3a2NigoKBAJEju56gfaiM0NBR79uxBfn4+2Gw2jI2NsXTp0mab+RBo7dmzB8eOHYOXlxcqKyuhrKwMGxsb2NnZNWlp6FO0tLSoYM+A8FR8QwjOo9tWbGxskJqaipCQkEYvoWdnZ8PS0hJcLhcBAQG09GxtbfHixQuEh4dTQZB5PB7mzJkDFosFb29vkVRNxcXFmDJlikg+4bYGk+1SANO2x3TuXKbzLDOpx3TfsnTpUkRHR4PFYmHw4MESUz6lpqaioKBAZJDh4+NDS4/pvkyAuFBMeXl5mDVrFvLz89G7d2+MHj0aPXr0wOvXr/HXX3+htLQUenp6OHXq1Gd90wHmcxC3Vl1Ki0yl0srJyYGRkZHYXWCCD9fWrVuxfv16dOvWrcnr5M+ePYO2trbI4A2oS0OlpaWFW7duCQ3gvvnmG+jq6oLH49HWKy4uFus4LaC2thZ5eXki5b1790ZycjJtvfrMnDkTJiYm8PT0xJkzZxAZGYlbt26hb9++MDc3h6mpKTQ1NZsckoXFYmH58uWwsrLC3r17ER4eDl9fX5w5cwYGBgYwNzeHiYlJk51Vg4ODsWrVKqSnp6Nfv35SRe2mQ2pqKkaPHk3L/7Ffv34YPXo0Hj16RFvvxYsX0NPTE8pgMWrUKGhqauL58+diY4V17twZo0aNkiqXpoATJ05IfS0A2Nvb0zqfqXbJtO0xnTuX6TzLTOsBzPUt3t7eVCy7vLw8LF++HDNmzBA5TzBxQHfA9ilM92UN4eHhgfz8fFhYWGDHjh1C3+LCwkJs3LgRd+/eha+vL+zs7D57P6ZzELelumwMMjWAa9++vdh8hAJsbGyQl5cHDw8P/Pzzz/D392/SkktlZaWIU2V9SktLxR7v3LmzRAf2hhg4cCASExPx8uVLkb87Ly8PCQkJIss75eXlSEhIEJuKii4dOnSAo6Mjli5dCn9/f/j7+yM7OxvHjx/H8ePHoaysDE1NTfTr1w9qampwcHCQWktNTQ0HDhzAzz//DB8fH4SGhiImJgaxsbHYunUrBgwYAC0tLUqL7hKOuro6zpw5gx9//BFZWVnUslxL0bFjxwYHAJIoLS2VajmmtrYW1dXVIuUcDgfPnz9HeXm52J19fD6/SbulDx06JKRL9xcs3QEcwEy7ZNr2mM6dy3SeZab16tPSfQtQN3NkZGQEJycnrFu3DhEREXB2dm6RJVum+7KG+PPPP9G9e3eRwRtQNwB3c3ODiYkJgoODGzWAYzoHcVuqy8YgU7lQORwOEhMTG0zsK/BJKSoqwpIlS/Dq1Sup9QYPHoynT5/i8ePHIscyMjLA4/FEZlwKCwsRHx8v1U7Ub7/9Fnw+H8uXL0dMTIyQ1qpVq1BeXk790quurkZMTAzs7OyoX0TNhaqqKlasWIGoqCicPn0atra26Nu3L8rLyxEfH4/g4GAcO3asWbQ0NDSwY8cO3L17F7t27cLEiROhpKSEzMxMhISE4MiRI0IznHTo1q0bDh8+DGVlZbi4uKC8vLxZ/mZxjBgxAnFxcY12cAbqdjfHxcXRThMG1CWBfvToERW5XMDu3bvB4/HERg1PS0vDo0ePmpTb8vLlyxgwYAAIIejbty+WL1+OZcuWffaf4Lym0JLtkmnbYzp3LtN5lpnWE0dL9i1AXRL5gIAAODg44NatW7CwsGgwH2tTYLIva4ji4mIq1ZU4OnToAD09PWRlZTXqfkznIAbaTl02CiJDXLlyhXA4HDJp0iRy+fJlkpGRIfa86upqsnTpUsLhcIienh6ZPn064XK5tPWCgoIIh8MhBgYGJCgoiOTn55Py8nJy7949YmpqSrhcLvH29iaEEFJUVESCgoLIlClTCJfLJSdPnqStV1VVRebNm0c4HA7hcrlEW1ubjBkzhnC5XMLhcIi1tTUpLy8nhBBy584dwuFwCIfDIcuWLSOVlZW09TgcDlm/fn2jz//nn3/I3bt3ib+/P/Hw8GgxLT6fT/7++28SEBBA9u3bRzZu3EhL61MOHz5MOBwOOXLkSJPu0xCJiYlk+PDhZMSIEWTTpk3k1q1bJCsri1RUVFDnVFZWkjdv3pCoqCiyZcsWMmLECDJ8+HDy5MkT2nrBwcGEw+GQsWPHEl9fX1JcXCzx3IKCAuLv708MDQ0Jl8slISEhUj2jgPz8fKr937p1q0n3EgeT7VIA07YnuIe2tjbZvXs3KSgokHhuSkoK2bdvHxk5ciThcrnk7t27tPUyMjKIjo4O4XK5ZOHChcTX15fExMSQtLQ08ubNG5KTk0PS09PJw4cPib+/P1m8eDHhcrlER0eHpKWltWm91uxbBDx//pzMnDmTcLlc8ttvv5Hi4mKybt06qb47DcFEXyZAXL1aWlqSxYsXN3jd3LlziY6OTqM0IiIiCIfDISNHjiTu7u6Ex+ORtLQ0cu7cOaq/CgoKIoQQ8vbtW+Lu7k4MDAwIl8slly5dku7B/h8m61JaZGoTAwBs27YN58+fpxxIQ0NDxZ7H5/Ph6OiIiIgIAJDaAdHR0RFhYWFic6VNmDABXl5eYLPZ1I5VoG56fcuWLVItVfH5fBw9ehQXLlxAfn4+gLqZB0tLSzg4OKBjx44A6mZTPD09YWJiIlXIEoDZfKGtkZtUQEVFBcLDw9G1a1dMnjy5xXT++usvrF+/ngqJUJ9Pc/uR/w+t4OrqCmNjY6n03N3d4eXlBRaLhfDwcImhLNasWYPr16+DEAIbGxu4uLhIpVef5ORk/PDDD1BVVcX169ebNd1Ma7UVJm0PYD53bmvkWWZCrzX7lvpUVVXB3d0dPj4+6NWrF7p27YoXL140q+M7U30ZUFevXC4X9vb20NTUxKBBg3Dy5Em4ubnhxo0bYgPjP378GPPmzYOuri78/PwapdNQDuL6/VVz5SAWwGRdSovMDeCAuvAely9fRqdOnT77MQoICMDRo0eRn58vtSEFBQUhKChIKD+ilZUV5s+fT+20yc7ORlBQEIyNjZtth1xhYSEIIejWrVuzZXmoz/z582FkZNTsydTFYWxsDDMzM2zYsKHFtVqTyspKBAcHIzo6GqmpqQ3msTUzM2tyUF0ej4fQ0FBs2rRJoiP/8ePH8eDBA9jY2GDq1KlN0qvP4cOHceTIETg4ODTr9nom26UkWtr2BLRG7tzo6OhGtc/myrPc0nptrW+Jj4/Hhg0bkJ2d3So7F5uL+vH1gLrl0YEDB+LZs2cYOnQofH19KZ+/1NRUhIeHw8fHBxUVFXB3d6fV18TExCAoKEgoR29L5iD+UpDJARxdCCHIyMiAurp6a/8pXxx8Ph/Z2dmorq7GoEGD2nQAWGlorjRrn6O6uhqZmZmoqKhA7969hZKVf6lUVFQgNDQUX331Vas4AjP17loLpm1P1vWYpLy8HCdPnkRycjIOHDiAdu3atbhmc9vD27dvkZSUhKSkJDx//hzPnz9HTk4OgLpBXWRkJBUwefXq1bh58yYAtEjCe6Ztva30LTI3gKusrMTRo0cRFhaGvLw8qKmpYerUqVi8eLHELetNSaX1qV6fPn1gZmbWYnqtQVZWFqKjo6GoqIgpU6ZQ2+y9vb1x7NgxaheQiooK5s2bh9WrV0uV3ojpd9cYmuv+Bw8exLBhw0QC8vL5fLi5uSEwMBBlZWVUOYfDwYoVK5o0I9YW65NJZOFZGmt7HTp0wNy5c6W2Pbp6TbV1pvVawxaYrsvPwYQ9FBUVUQM6Ozs7anbOw8MDKSkpmDNnjkj8yeaAaVtvK32LTIUR4fP5WLhwIXg8HuVPlJWVBW9vbwQHB8Pd3R26urpir5VmHCtO79WrVy2m1xocPXoUR44cofIU7t27F15eXnjx4gX++OMPsFgs9O/fH0pKSsjIyMCJ0StK9wAABt1JREFUEyeQkpJCO6I1n8/HggULkJiYyMi7o0Nz3N/T0xOWlpZCAzhB+0lISABQF95AEPQyOTkZDg4OsLOzg5OTE209pm2hrfIlP4unpyc8PDxa3PZkXa81+ham67KxtLQ9qKqqYsyYMRgzZoxQ+apVq1pUF2De1ttC3yJTAzhvb28kJCRAW1sbzs7OGDRoEBITE3Ho0CHExcVh8eLFOHLkCIyMjL5IPaaJiorCoUOH0LlzZ8ycORMlJSW4ceMGNaDo2bMnPDw8qDAX2dnZWLt2LaKiohAUFEQrfpK3tzd4PB5jdUk3VEb986VJdyMOb29vPHnyBFwuF7///jvlUwIAd+/exfbt2+Hj4wMdHR3aS5Cy3DbbwrtraaKionDw4EFGbE/W9ZjuW5iuS1m2B6af7UurS5mKA3ft2jWoqqrCy8sLXC4X7du3h76+Pvz8/GBvb4/KykqsWrUK8fHxX6Qe0/j6+kJJSQnnzp3D1q1bsXfvXri7uyMnJwe5ubnYvHmzUIyyfv36UUnSL168SEuL6brs3r07CCEghFD5QcX9E1C/rLmc1kNCQqCiooKjR48KDd4AYMKECfDx8UG7du0avVurPrLcNtvCu2tpmLQ9Wddj2haYrktZtgemn+1Lq0uZmoHLzs6GoaGh2GjXjo6OUFRUhKenJ1auXNnkLAytocc0T58+hb6+vtDfPXnyZGhoaCA9PR1jx44VuaZ3794YNWoUtSzYWJiuy7CwMDg7OyM8PByjRo3Cnj17xAZXXrduHcLCwkQSMzcHOTk5GDVqlERH2P79+8PQ0JB2XQKt1zbFBbWmw+eSmgNt4921NEzanqzrMW0LTNdla9mDLNr6l9a3yNQAjs1mi00fJODXX39FQUEBAgMDYW9vj3PnzjUpxRTTekxTXl4udneUhoYGXr58Sfl3fIqCggJt/wCm61JVVZVK6+Lq6gpLS0usX78etra2Que15K8qNTU1dOnS5bPn8fl82vdurbZpa2srdZ01dgmiLby7loZJ25N1PaZtgem6bC17kEVb/9L6FplaQtXQ0ACPx8P79+8lnrNt2zYYGRkhJycHS5YsaTCXaVvTY5qBAwciPj5eJG/rvn37EB4eLjbn4rt37xAfH087JEtr1aWFhQVCQkKgq6sLFxcXLF68GG/fvm3yfcVRUFAg9HwTJkzAo0ePJOZIzcvLQ3x8PAYMGEBbq7Xqc9euXVBRUaECEevp6TX6nyRHckkw+e6Yhknbk3U9pm2B6boUwLQ9yLKtfzF9S3OndmhN/Pz8CIfDIVZWVuThw4dUaptPKS0tJdbW1oTD4ZDJkycTa2trqVKaMK3HNKdPnyYcDoeYm5uTW7duET6fL/HcmpoaEhUVRczMzAiXyyVnzpyhpdUW6vLs2bNEW1ub6OrqkosXLxJCSLOluxGkYOJyuWTcuHFk2bJlZNOmTYTD4ZC1a9eSqqoq6tyamhoSExNDZsyYQbhcLjlx4gRtvdasz8ePHxMdHR2ira1NUlJSmnSvxtKS7641YNL2ZF2PaVtgui7FwZQ9/BtsvS33LTIVB662tpZKZs1isaChoSExWfLHjx9hb28PHo9HTYfSjYjNtB7T1NbWYsOGDQgJCQGLxUJoaCgGDx4s9ty1a9ciPDwchBBMmjQJnp6etKaZ20pdZmVlwcnJCTweDxMnTkRVVRXu37/f5PtfuXIFz58/R1JSEpKTk/Hx40fqGIvFwtWrVymfmdWrVyMiIgKEEOjp6eHUqVO040O1dn3evHkTq1evhpaWFgIDA5t0r8bSUu+uNWDS9mRdrzW+C0zWpSSYsod/g6231b5FwdnZ2blV/4JmhMViYcaMGejZsyc+fPiAXr16wcLCQuy57du3h5WVFaqqqpCUlISamhrasWqY1mMaFosFU1NTqKurg8/nw9bWVmI6Jh6Ph8LCQixbtgwbN26kUojR0WoLdamqqopZs2ZBSUkJFy9eRGZmJlgsVpPvz+VyYWRkBGtra9jb28PKygr6+voYOHAgOnbsiB9++IGqs/v37yMnJwcLFy7Ejh07pIpA39r1qaGhgaysLNy7dw/9+vUT2WXbErTUu2sNmLQ9Wddrje8Ck3UpCabs4d9g6221b5GpGThpKSoqQmJiIsaPHy+TerJMS9ZlcnIy3NzcUFZWJlUoD2kpLS2lEqMzTXPW57t37+Du7g51dXXY29s3w1/XeFrr3cmRHWSpn25pe/g32Xpb6lvkAzg5cuTIkSNHjpwvDJnahSpHjhw5cuTIkfNvQD6AkyNHjhw5cuTI+cKQD+DkyJEjR44cOXK+MOQDODly5MiRI0eOnC8M+QBOjhw5cuTIkSPnC+P/AE+BsJIiow1HAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "diff1.plot(kind='bar',figsize=(10,8))\n",
    "plt.grid(which='major', linestyle='-', linewidth='0.5', color='green')\n",
    "plt.grid(which='minor', linestyle=':', linewidth='0.5', color='black')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Análise de Resíduos pelo teste de Shapiro-Wilk\n",
    "\n",
    "Analisando os resíduos desse modelo temos que pelo teste de normalidade de Shapiro-Wilk, obtivemos o resultado de 0.99047 com p-valor de 0.32761 que é maior que 0.05. Então temos que nossos dados seguem uma distribuição normal visto que não houve violação na normalidade dos resíduos. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.9904677271842957, 0.3276057243347168)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats.shapiro(Y_pred)"
   ]
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}