{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Uczenie Maszynowe: Naiwny klasyfikator bayesowski\n",
    "\n",
    "## Zadanie 1\n",
    "Zadanie polega na implementacji klasyfikatora naiwnego Bayesa dla zmiennych ciągłych, gdzie za rozkłady cechy przyjmij rozkłady normalne.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.style.use(\"ggplot\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Do testowania twojego rozwiązania użyj trzech generatorów danych sztucznych `generate1`, `generate2` oraz `generate3` (funkcje te przyjmują jako argument liczbę elementów do wygenerowania z każdej klasy - domyślnie $N=100$). Sposób ich wywołania jest przedstawiony poniżej:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1c6d9aae1d0>"
      ]
     },
     "execution_count": 10,
     "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 Bayes_helpers import generate1, generate2, generate3\n",
    "\n",
    "X, y = generate1()  #możesz dodać argument (10) oraz print X,y żeby zobaczyć, jak te tablice dokładnie wyglądają\n",
    "plt.scatter(X[:,0], X[:,1], c = y)\n",
    "#print(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "W implementacji będzie przydatna klasa `norm` z pakietu `scipy`, która zwraca wartości funkcji gęstości prawdopodobieństwa dla zmiennych ciągłych."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.4867195147342979e-06"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from scipy.stats import norm\n",
    "\n",
    "# (X, mean, std)\n",
    "norm.pdf(5, 0, 1) #gęstość prawd. rozkładu standardowego dla wartości 5\n",
    "#norm.logpdf(5, 0, 1) #logarytm gęstości prawd. rozkładu standardowego dla wartości 5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Zaimplementuj klasyfikator naiwnego Bayesa dla zmiennych ciągłych. Pamiętaj o zabezpieczniu się przed problemem wynikającym z mnożenia wielu małych liczb (prawdopodobieństw)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "class GaussianNaiveBayes():\n",
    "    def __init__(self):\n",
    "        self.means = {} \n",
    "        # Słownik, który docelowo powinien zawierać tablicę/wektor warunkowych średnich dla każdego atrybutu \n",
    "        # Każda tablica/wektor powinna być typu np.array\n",
    "        # np. 1) means[1] powinno zawierać wektor średnich wartości atrybutów dla klasy o indeksie 1\n",
    "        #     2) means[0][1] powinno zawierać średnią 1-go atrybutu dla klasy o indeksie 0\n",
    "        # (Możesz spróbować zaimplementować efektywniejszą implementację używając macierzy)\n",
    "        self.stds = {} \n",
    "        # Analogiczna struktura dla odchyleń standardowych\n",
    "        self.class_log_prob = None \n",
    "        # Wektor zawierający logarytmy prawdopodobieństwa dla każdej z klas \n",
    "        # np. class_log_prob[1] zawiera logarytm prawdopodobieństwa, że klasa jest równa 1, log P(C=1)\n",
    "        \n",
    "    def fit(self, X, y):\n",
    "        # TWÓJ KOD TUTAJ - proces uczenia czyli uzupełniania struktur zainicjaliowanych w init()\n",
    "        # X jest macierzą gdzie każdy wiersz zawiera kolejną obserwację (obie typu np.array) \n",
    "        # y jest wektorem wartości indeksu klasy (0 lub 1). Jego wartości odpowiadają kolejnym wierszom X\n",
    "        unique, counts = np.unique(y, return_counts=True)\n",
    "        self.class_log_prob = np.log(counts/np.sum(counts))\n",
    "        for clazz in unique:\n",
    "            self.means[clazz] = np.mean(X[y == clazz], axis=0) # \"y == clazz\" daje tablicę Fałszów i Prawd, którą to tablicą wybieramy podzbiór X tworząc nową tablicę\n",
    "            self.stds[clazz] = np.std(X[y == clazz], axis=0)\n",
    "        \n",
    "    def predict_proba(self, X):\n",
    "        # TWÓJ KOD TUTAJ - predykcja - zwrócenie prawdopodobieństwa dla każdej klasy i każdej obserwacji\n",
    "        # Funkcja powinna zwrócić macierz o dwóch kolumnach (dwie klasy) w której kolejne wiersze \n",
    "        # zawierają prawdopodobieństwa P(c|x) przynależności dla klas dla kolejnych obserwacji w macierzy X\n",
    "        results = np.tile(self.class_log_prob, (X.shape[0],1)) #zainicjuj \"results\" log-prawdopodobieństwami bezwarunkowymi klasy (a priori, po ang. \"priors\")\n",
    "        for i in range(len(self.class_log_prob)): #dla każdej klasy...\n",
    "            for j in range(X.shape[1]): #dla każdego atrybutu...\n",
    "                results [:,i] += norm.logpdf(X[:,j], loc=self.means[i][j], scale=self.stds[i][j]) #dodaj log-prawdopodobieństwa warunkowe P(x|c)\n",
    "        probabilities = np.exp(results) #wracamy z log-prawdopodobieństwa do prawdopodobieństwa\n",
    "        probabilities /= np.sum(probabilities, axis=1, keepdims=True) #teraz to są naprawdę prawdopodobieństwa, bo znormalizowaliśmy\n",
    "        return probabilities\n",
    "    \n",
    "    def predict(self, X):\n",
    "        # Gotowa funkcja wybierająca klasę z największym prawdopodobieństwem\n",
    "        prob = self.predict_proba(X)\n",
    "        return np.argmax(prob, axis=1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Przetestuj twój klasyfikator na wygenerowanych wcześniej danych."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "X, y = generate2(100)\n",
    "\n",
    "gnb = GaussianNaiveBayes()\n",
    "gnb.fit(X,y)\n",
    "#Trafność na zbiorze uczącym\n",
    "np.mean(gnb.predict(X) == y) #argumentem mean() jest tablica zawierająca Fałsze (0) i Prawdy (1), mean() liczy udział Prawd w całości tej tablicy - czyli trafność\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0YAAAFACAYAAACY4mwsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvDW2N/gAAIABJREFUeJzsnXd4FOX2x78zs303vRASIARCk6rcS1F6tV/L/VFELyDFLlhRsKCicrkC14ZXiogUC9JsqAQRUEB6LwkttEASQur2nff3R8juO7O7yRISAuR8noeHnXnfeefM7Gxmzpzzfo/AGGMgCIIgCIIgCIKoxYg1bQBBEARBEARBEERNQ44RQRAEQRAEQRC1HnKMCIIgCIIgCIKo9ZBjRBAEQRAEQRBErYccI4IgCIIgCIIgaj3kGBEEQRAEQRAEUeshx+ga4PPPP4dGo6lpMyqFIAhYsGBBTZsRkIkTJyI1NbWmzVDQsGFDTJo06bLGOH78OARBwB9//FFuP/V3o953jx49MHLkyMuyhSCI2gXdr64efv/9dwiCgFOnTtW0KTh58iR69+4Ns9kMQRBq2pxKUV3n82p8FqnNkGNUhQwbNgyCIPj9s1gsIW1/6tQpCIKA33//XbF+4MCBOH36dDVY7E+fPn0wbNiwK7IvombJysrCP//5z6DtS5cuxbRp07zLdG0QxPUD3a8qh8fjwYwZM9C5c2dERETAbDajRYsWePjhh7F169YraktF3HzzzcjKykJiYmJNm4J33nkH2dnZ2LlzJ7KysmraHIIIyrX5WucqpmvXrvjmm28U60Tx8vxPo9EIo9F4WWMQNYcsy2CMQZKkmjZFQUJCQrnt0dHRV8gSgiBqArpfXRoulwt33XUXNmzYgAkTJmD69OmoU6cOMjMz8ccff+DZZ5/FunXratpMLzqdrsK/81eKjIwMdOjQAU2aNLmscZxOJ3Q6XRVZRRD+UMSoiin7Q8T/i4+P97b/8ccfuOWWWxAWFoawsDC0bdsWv/zyCwCgfv36AICePXtCEAQ0bNgQgH9qQtnymjVr0Lp1axiNRnTv3h1nzpzBunXrcOONN8JsNqNPnz6KN3fHjh3Dfffdh8TERJhMJrRu3Rrz58/3tg8bNgyrV6/GvHnzvG8Py94Gnjt3DsOGDUNcXBzCwsJwyy23+N0A1qxZgzZt2sBgMKBNmzZYs2ZNSOcsLS0NXbt2hclkQkREBLp3744jR44AABhjeO+999CoUSPodDo0btwY//3vfxXbN2zYEK+99hrGjBmD6Oho1KlTB88//zw8Ho+3j8PhwGOPPYaIiAhERUXhscceg8PhCMk+nrKQ96JFi9CoUSMYDAb06dMHx44d8+vz9ddfo3nz5tDpdDhw4EBIxwIANpsNI0eORHh4OGJjYzFu3DjIsuxtX7RoETp27IiIiAjExsbijjvuQHp6ut84x44dQ+/evWE0GpGSkoKFCxcq2itKG+FT6YJdG927d8fo0aMV2zHG0LhxY0ycODGkc0oQRM1A96tLu1998MEH+PXXX7Fq1SqMGzcOnTp1QkpKCnr06IFXXnkFa9euDdl+IHC68qRJk7znEgD27duH/v37IzIy0hud4seZPXs2WrRoAYPBgJiYGHTr1s2b6qVO/WKMYdSoUWjcuDGMRiMaNWqE8ePHK+6FZfevFStWoHnz5jCbzejZs6f3ngwAhYWFGD58OBISEqDX61G/fn08++yzQc+bIAhYvXo1PvvsMwiC4I3yZWVlYdCgQYiMjITRaESPHj0UUbcy+3/88Ud06dIFBoMBM2fODLqfDz/8EM2bN4fBYECTJk3w9ttvw+12e9tDuXdmZ2dj+PDhqFOnDgwGA5o1a4bPPvtM0efAgQPo1q0bTCYTbrjhBu9vojzKe85RE8q1U95vEyiN0DVq1Ah6vR5xcXHo378/bDZbhXYSABhRZQwdOpT17t07aLvb7WZRUVHsmWeeYenp6Sw9PZ0tXbqUrVu3jjHG2Pbt2xkAtmTJEpaVlcWys7MZY4zNnTuXSZLkHWfu3LlMEATWvXt3tmnTJrZt2zaWmprKunTpwrp37842btzItm/fzpo1a8YGDBjg3W737t3so48+Yrt27WKHDx9mH3zwAZMkif3222+MMcby8/NZ165d2YABA1hWVhbLyspiDoeDWa1W1qJFC3bfffexLVu2sIyMDDZp0iSm0+nY/v37GWOMnT59mplMJjZs2DC2b98+9uuvv7LWrVszAGz+/PlBz8mqVauYKIpszJgxbOfOnezAgQNs9uzZ7MCBA4wxxj766CNmMBjYp59+ytLT09knn3zC9Ho9mz17tneM5ORkFhkZyd59912Wnp7OvvrqKyZJEvvss8+8fcaOHcvi4uLY8uXL2YEDB9hzzz3HwsLCWOPGjUP+fhlj7PXXX2cmk4ndcsstbPPmzWzz5s2sQ4cOrE2bNkyWZW8fo9HIunXrxjZu3MgOHTrECgsLQz6WsLAw9uqrr7KDBw+yL774gplMJjZ16lRvn88++4x9//337PDhw2z79u3srrvuYqmpqczhcDDGGDt27BgDwOrWrcsWLFjADh48yCZMmMAEQWBbtmzxjqP+bpKTk9lbb73lXe7evTsbMWIEYyz4tbFo0SJmsVhYUVGRd7u0tDQmiiLLzMy8pHNLEMSVg+5Xl36/atu2Levbt29I57ci+xlT/o0t46233mLJycne5datW7PBgwezffv2sSNHjrCffvqJff/994wxxrZu3cokSWLz5s1jx48fZ7t372azZs1iJ0+eZIwxtmbNGgbAu+zxeNiECRPYpk2b2LFjx9iKFStYQkICe+2117z7K7vH9e/fn23dupXt3LmTtWvXjnXr1s3b56mnnmJt2rRhmzZtYpmZmezPP/9kM2fODHousrKyWOfOndkDDzzAsrKyWH5+PpNlmXXo0IG1bduWrV+/nu3evZsNGDCARUZGspycHIX9zZo1YytWrGBHjx71Houa119/nTVo0IAtXbqUHT16lP3444+sfv367JVXXvH2qejeabVaWfPmzdmNN97IVq1axY4cOcJ++eUX9uWXXyrsadOmDVu5ciVLT09nDz30EIuIiGAXLlwIevwVPee8/vrrimeRiq6din6bS5YsYWFhYey7775jmZmZbMeOHWz69OnMarUGtZHwQY5RFTJ06FAmSRIzm82Kf3feeSdjjLG8vDwGgK1Zsybg9idPngzYHuhGA4Dt2LHDu27KlCkMANu6dat33bRp01hMTEy5Nt99991s5MiR3uXevXuzoUOH+u0/KSmJuVwuxfqePXuyMWPGMMYYmzBhAmvQoIGiz/fff1/hjaZLly7sjjvuCNper1499sILLyjWjR07lqWkpHiXk5OT2V133aXo079/fzZo0CDGGGPFxcVMr9f7/eFu3759pRwjACwjI8O77tChQwwAW7VqlbePIAh+jkGox9KlSxdFn5dffpklJSUFten8+fMMAPvjjz8YYz7HiL8hMMZY586d2ZAhQ7zLl+IYMRb42nA4HCw2NpbNmjXLu27QoEHs9ttvD2ovQRA1D92vLv1+ZTQa2dNPP61Y9+KLLyrOX3kvhNT2h+IYhYeHs7lz5wYcb+nSpSw8PJwVFBQEbFc7RoGYNm0aS01N9S6//vrrTJIkr6PLGGNffvklEwSB2Ww273Goz3tFqI81LS2NAWD79u3zrrPb7SwhIYG98cYbCvu/+OKLcscuKSlhRqORrVy5UrF+3rx5LCIiIuh26nvn7NmzmV6vD3q+yuxZsmSJd11WVhYDwH7++eeg+6noOUftGAWCv3Yq+m1OmzaNNWnShDmdznLHJAJDqXRVTMeOHbFz507Fv08//RQAEBUVhZEjR6J///647bbbMHnyZBw6dKhS+xEEAa1bt/Yul+URt2nTRrHu/Pnz3pQyq9WKl156CS1btkR0dDQsFgt++uknZGZmlruvLVu24OzZs4iMjITFYvH+W79+PTIyMgAA+/fvR4cOHRQpFF26dKnwOLZt24Z+/foFbCssLMSpU6fQrVs3xfru3bvj+PHjsFqt3nXt2rVT9ElKSsK5c+cAAEeOHIHD4cDNN9+s6BOKfYGIi4tTKMg0bdoUsbGx2L9/v3ddnTp10KBBg0odS+fOnRV9brnlFpw+fRqFhYUAgJ07d+Lee+9FSkoKwsLCvPtRf4+BxuFtrAp0Oh2GDRuGWbNmAQDOnz+PZcuWYdSoUVW6H4Igqh66X13a/SoQL7zwAnbu3Ik5c+agpKTEm/ZcWfvVPP/88xg5ciR69OiBiRMnYvv27d62vn37olGjRkhJScGgQYMwc+ZM5ObmljverFmz0LFjR9SpUwcWiwUvv/yyn02JiYmIi4vzLiclJYExhuzsbADA448/jm+//RatWrXCmDFjsHLlSkW6dyjs27cPMTExuOGGG7zr9Ho9OnbsiH379in6dujQocKxbDYb7r//fsV3/sgjj6CgoAA5OTkAKr53btu2DTfccAPq1atX7v74542EhARIkuR93ghEec85gajo2qnotzlgwAC4XC4kJydj2LBhmD9/PoqKikLef22HHKMqxmg0IjU1VfGPV4SZNWsWtm3bhr59+2Lt2rVo1aqV90Z0KYiiqJjMXyZ/qdVq/dYxxgCU/gFfsGABXnvtNaxZswY7d+7E7bffDqfTWe6+ZFlGixYt/G6gBw4c8D4QM8b8JDhDleSsqJ+6vex4eNSTMQVB8P6hLutfnRKhapvMZnPAfqEcS3ljW61W9OvXD4Ig4LPPPsPmzZuxZcsWCIJQ4fcYyr4qwyOPPIItW7Zg9+7dmD9/PqKjo3HnnXdWy74Igqg66H7lI5T7Q9OmTf1eLsXGxiI1NRVJSUmK9aHYL4qi399ll8ulWH711VeRnp6OAQMGYO/evejUqRNeeeUVAIDFYsHWrVuxbNkyNG3aFP/73/+QmpqKbdu2BbR/8eLFeOKJJzBw4ED89NNP2LFjB1577TW/fQa6nwLw3lP79++PEydOYMKECbDb7XjwwQfRq1cvxbzeUAh0zgN9N8Hup2WU2bV48WLFd75nzx5kZGQgOjo65HtnKNdBIPGHihzDS3n+COXaKe+3mZSUhIMHD+Kzzz5DfHw83nrrLTRr1gwnT54M2YbaDDlGNUCrVq3w7LPPYuXKlRgxYoR3MmHZj+1S/7iEyrp16zBkyBAMHDgQbdu2RaNGjfwmHup0Or/9/+1vf8PRo0cRHh4e9CbasmVL/PXXX4ptK6qjAwDt27cPOnExPDwc9erVU0xoLTuOlJQUmEymkI47NTUVOp0Of/75p2L9hg0bQtpeTU5OjmLSZHp6Os6fP48WLVoE3eZSjmXTpk2KPhs3bkRiYiLCw8Nx4MAB5OTk4O2330bPnj3RokULXLhwIaDTE2ic8mysiEDXBlB6fnv16oVZs2Zh9uzZGD58+DVbx4QgCCV0v/Lx4IMPYvXq1di4cWOV2B8fH48zZ84o1vERoTIaNWrkjdK8+eab+OSTT7xtkiShW7duePPNN7Ft2zbUrVsXixYtCmrTjTfeiGeffRbt27dHkyZNcPz48QqPJRDR0dEYPHgwPv30U/z4449Yu3btJWUktGzZErm5uYptHA4HNm/ejJYtW16SLS1btoTBYMDRo0f9vvPU1FRIkhTSvbN9+/bYt29fldcpKu85JxChXDtA8N8mUBp9u/XWWzFlyhTs2bMHVqsVy5cvr5Ljud4hx6iKcTqdOHv2rN8/xhgOHz6McePG4Y8//kBmZiY2btyI9evXe0PJsbGxsFgs+PXXX3H27FlcuHChSm1r1qwZVqxYgc2bN2P//v0YPXq03x/llJQUbNu2DUeOHEFubi5cLheGDBmClJQU3HHHHfj1119x/Phx/PXXX3j33Xe9P7THHnsMOTk5GD16NA4cOIDVq1djwoQJFdr06quvYuXKlRg7dix2796NQ4cO4fPPP/eGhV9++WV8+OGHmDVrFjIyMvDpp5/ik08+wfjx40M+brPZjEcffRSvvPIKvvvuOxw6dAgvvvgiDh48qOi3bNkyNG/evMIaHCaTCcOHD8e2bduwdetWDB06FK1bt0afPn3K3S7UY9m5cycmTpyI9PR0LFq0CO+//z6eeeYZAEBycjL0ej0+/PBDHDlyBKtXr8aYMWMCvo2aM2cOFi1ahPT0dLz22mvYuHEjxo4dG8opC0iga6OMRx55BDNnzsT+/fupKCxBXCPQ/erS7ldjxoxB79690a9fP0yePBl//fUXMjMzsWHDBu9DaVlkLBT7+/Tpg7S0NHzzzTc4fPgwJk+ejPXr13vbi4uL8cQTT+C3337DsWPHsGPHDvz888/e72DFihWYPn06tm3bhhMnTmD58uU4efKkIj1NfU737NmDFStW4MiRI3j//fexdOnSEL8RHxMmTMDSpUtx6NAhZGRkYOHChbBYLIr08Yro1asXOnTogAceeAB//vkn9u7di3/961+w2+147LHHLskei8WC8ePHY/z48fjoo49w6NAh7Nu3D1999RXGjRsHILR75+DBg5GcnIy7774baWlpOHbsGFavXo2vv/76kuxRU9FzjpqKrp2Kfptz5szBrFmzsGvXLmRmZmLhwoUoKioKel0QKq74rKbrmKFDhzIAAf/l5OSwM2fOsHvvvZclJSUxnU7H6taty0aOHMny8/O9Y8ybN481bNiQaTQa7wTMQJNZ+WXGGJs/fz5Tf51ffvklA+CdYHrixAnWr18/ZjKZvEo0Dz/8MOvevbt3myNHjrCuXbsys9msmNyXm5vLHn30UZaYmMi0Wi1LTExk99xzD9u+fbt327S0NNaqVSum0+lYy5Yt2erVqyuczMoYYz///DPr1KkTMxgMLDw8nPXo0YMdOXKEMcaYLMtsypQp3nOSkpLCpk+frtheLRrAGGMjRoxQHJfVamWjR49m4eHhLDw8nI0aNYq99NJLigmPZZOEjx07FtTWskmS8+fPZ8nJyUyn07GePXuyw4cP+/VRE+qxjB8/ng0bNoyFhYWxqKgo9vzzzzO32+3ts3jxYpaamsr0ej1r164d+/3335kkSd4JumXiC1988QXr3r070+v1LDk52W8Cq/q7qUh8Idi1wRhjTqeTxcXFsX79+gU9dwRBXD3Q/apy9yuXy8U++OAD1qFDB2axWJhWq2UNGjRgQ4YM8U7iD9V+p9PJxowZw+Li4lhERAR7/PHH2auvvuo9lzabjQ0ePJg1bNiQ6fV6FhcXxwYMGMBOnDjBGGNs7dq1rGfPniw2Npbp9XqWmprK3n33Xa9Cqlp8wel0stGjR7OoqCgWFhbGBg8ezD788EPFdxHo/rV+/XrFvfHNN99kLVu2ZGazmYWHh7Nu3bqx9evXl3veAglNnDlzhg0cOJBFREQwg8HAunXrplBODUU8gmf27Nmsbdu2TK/Xs8jISNahQwc2Y8YMb3tF907GSsUUHnroIRYTE8P0ej1r1qyZtz2YPeoxAlHec476nFd07VT021yyZAnr3Lkzi4yMZEajkbVs2VKhfkuUj8BYNU08IIjrkIkTJ2LBggU4fPhwTZtyVZGXl4ekpCQsWLAA999/f02bQxAEQRAEccnQRACCICqNy+XCuXPn8NZbbyExMRH33HNPTZtEEARBEARRKcgxIgii0vz555/o2bMnUlJS8MUXXyiUpwiCIAiCIK4lKJWOIAiCIAiCIIhaD6nSEQRBEARBEARR6yHHiCAIgiAIgiCIWg85RgRBEARBEARB1HouWXzh1sgRpR/Uk6x1Wu9HQasN2sYMOkWTbPS1ebjPbotyDJfF58M5LUp/zhkmcP24bcKV06fcYbJvIcylaDNaHN7PkWaboi3OWOL7bCj2fdYVKfrFan1t0ZpiRVukZPXtWvSNbxacin4m0WeXQVBW9NaCcW2KJkhckTItfJ9FVeFPiW9T+cVSgCKhwfr61gffJlQkgfxzonYiJmTUtAm1HnXRzcoSGxuL3NzcKhnrSkD2Vi9kb/VC9lYv16O9iYmJIY1FT6QEQRAEQRAEQdR6LjliVCZiJzBZ2eDhohuiyt/ycMtuZRREcPsiT6LbN6boVI4vOXyRCUnHVG2+z7LO1092KKMZfJtHp4x4Obkol03rVrSVaH1RLpPb99ko6RX9TJIv+qOXlW0GwRcJ0gpu7rMqKsR8yxJUx8n1Vca7AHDigiJ/2GrNQcUpUX2HzPc9qaNHMteXjx7Jqh1UJoLkUV1LFEEiCIIgCIIgrjT0BEoQBEEQBEEQRK3n0gu8yhff7qvno3ARCyYrIwCCx7csqCJGcPmWBS0XPXIpxxBdXDTDqYqkOLlIEDdlR3QoukHUc9Ekp9In9Dh9+3a4lKfF6vJFk4o1vkiQUVLGbYq4Zb2obDMIvu34KJEWyvOhE4NHjERuWVRFe/goER8Lq2y5TQ9TR6tCiwTJChsvf/4RQRAEQRAEQVwJKGJEXBIeN7Dplwj8OC8WF7Iv3a8mCIIgiOsCjwf6tDSY5s6FmJVV09YQBFEF0JMtETJnT+jwzohUnM4wwuUS8c2HdXHbQzkYNOZsTZtGEARBEFcMMSsL0cOGQZOeDtHphOfDD2EdOBBF48bVtGkEQVwGl+4YXUyxYqpUK4EXX1CnXfFiDLIquYtLrROCpNUBytQ60aWSoOZS64IJMQBKMQZ1m0fLyYHrlFLhVo3PLoPGJ5xg0iilx41uX7qcSVTLcPuW7cw3vgHKfnybKCjT5bRc+pxKNkGx7AkmxAAoxRj8Mt0CCyyU8dELDXF8v9m7fD5Lhx8+i0OPe/OQ0MDp17+ywgy8GAMJMRAEQRBXG5Evvgjd3r3eZencOZgWLIDtnnvgbtasBi0jCOJyoKdOIiQ8biDruMFvfX6uDj8vjK0BiwiCIAiiZtAcOeK3TsrLg2nevBqwhiCIqoIcIyIkBBHQaNVxqlLCIjwB1xMEQRDE9QjT6QKvj4y8wpYQBFGVVJ0qHb+sSrPjVep4hTrFeAAEvo6RS/mwzTgVOVGlKCdpuVS6IAp1gFKljleoA5QqdbxCHQA4dL7TFEyhDlCq1Bk9qjpGIl/HiFOlE5Q1k7ScKp2WKe1wKVLd1AWKuLYgCnWlsIAfAaXynOyneiei6Y0lyDpmAJ+Dl5BsR78Hs739A6XgXQ5U44ggCIK42nB26ADN4cMQuOcdd716KBkxogatIgjiciHxBSJkHp9yDG6XgPQdFjjtAmITnRj+yimYwyliRBAEQdQeCiZNgmC1Qrd1KwSbDZ6EBBS+9BLkmJiaNo0giMuAHCMiZHQGhuc/OQJHiQRbsYToOi6/wCFBEARBXPfodMj/6CMIJSUQioog16njn0lDEMQ1xyU7RmVqdII6XY5bFmRVjha/7FamjgmSLzWKSbxCnTJlilepk5zK9CpeYS6YQp26n2xXpdJpuRQ8nXLfTj6VjlOi4xXqAKVKnVFSptKZJF9en0nmFOoEpQIevyyp0sj4FDx1Kh1fDNal+G6U/XTcH251mp2H6yuVoyCnN3ugN3sgV9BPTVUUfyXFOoIgCOJqgZnNYGZzxR0JgrgmoCdLgiAIgiAIgiBqPZVPpZNVIgp8CNmjmnPCqwEwlS/GizFw2wludR0jXxtTiS/wYgzBhBgAQOYiSB6VoIzI9ZUdyvE9Op8tTq3vlNlcymgPL8Zg4IQYAMAo+tr0nOCCWnxBx0WF+M+AssaRVE7ESOK2U+vIVVeNI1+vUPtdfvSIIAiCIAiCIKoKihgRBEEQBEEQBFHrIceIIAiCIAiCIIhaTyXqGF1MgZJU6U9cihZTCTPwYgxMVcdIkPg6Rly6nKRMI+PFGHghBkApxhBMiAFQ1jWSHKo0O16YQRs8lc7FCzFolal0Bk58oVgqp8YR99kgB693xKfOAYCW+dLuXCrpBEmRmsZ/Vtcj4rdRU4kaR6rUyHJrIVGNI4IgqpEffvgBv/32GwRBQP369fH4449DF6QQJ0EQBEGooSdJgiAI4ponLy8PK1euxOTJkzF16lTIsowNGzbUtFkEQRDENQQ5RgRBEMR1gSzLcDqd8Hg8cDqdiIqKqmmTCIIgiGuISqvS+afLcWlN6iJnvEqdWgrNw/lmIvdZpWzHq9TxCnWAUqVO4hTlZK1Kuc3Bp8spzZB1gT8DSpU6D5fG59AoBynR+OzSS+oaR0FS6USlep2eS63TqlTpeAU7tSqdKPBqcCxoP37Zo/oOUQU1jvgxJdV1EEyxTlYfC9U4IgjiEomOjsZdd92Fxx57DDqdDm3btkXbtm39+qWlpSEtLQ0AMHnyZMTGxlbJ/jUaTZWNdSUge6sXsrd6IXurl9psb+XlugmCIAjiKqG4uBhbtmzBxx9/DJPJhGnTpmHdunXo1q2bol+fPn3Qp08f73Jubm6V7D82NrbKxroSkL3VC9lbvZC91cv1aG9iYmJIY126Y1T2Vl5WvZHnxRhUkQg+usQLMQBKMYZgQgyAUoyBF2IAlGIMIifEoBZY4GscyU5FE2RFNEkV6eACQ26db99upzKuYuPEGEo0yrCTkatxxEeM9KIysqTnIkgGQRlNKhF8Y0ii8jzywgxSuREj7rz61TGqTI2jqhdYqIoaRxQ9IojaxZ49exAfH4/w8HAAQMeOHZGenu7nGBEEQRBEMOiJkSAIgrjmiY2NRUZGBhwOBxhj2LNnD5KSkmraLIIgCOIaglLpCIIgiGueJk2aoFOnThg3bhwkSULDhg0VKXMEQRAEURFV5hgp0+WU6VUKMQZV7RnIXGpXMCEGQCHGwAsxAEoxBqYRufWqGjtcXSNZlWbn4dLlRJX4gujg23hxB6UdTg1X40gtzKD1DWpw+1Li1Kl0BkUqnbKNF1/QqYUZQkylq/IaR36Zbpde46i89LuqEGagekcEUTsYMGAABgwYUNNmEARBENco9IRIEARBEARBEESthxwjgiAIgiAIgiBqPZecSscuqsoJojoljvOxJFW6E5dmx6vQAYDApTUFU6gDlCp1vEIdoFSpE/lUOqcqlUvDK8+p1No4lTp1mh2vUscr1DFVKp1b67PZoVOe2mKnL5WOr3HEK9QBgJ5Ps7uUVDquja9/pE6X03LLLlVaGp9251QpC+rKqXFUEYwBF7K1MIcaOxcGAAAgAElEQVR7oDeW7j9YWl2pzaGl1pFiHUEQBHEtIhQXQ7DZIMfG+td+JAiixiDxBaJa2fhTJBa/n4S8c1oYzTJadS7Eo/8+DomuPIIgCKKWIdhsiBwzBtpduyA4HPAkJaFg4kS4/v73mjaNIAiQY0RUIzmndZjzejLOZ5VGwQpyS9cZLR6MeuNUDVtHEARBEFeWiGefhfHHH73LUk4Oop55Bjm//gpmMtWgZQRBANXkGDFVGpZCpU4dMuYVw4Ip1AFKlTqPKpWOU6kTOIU6dSqdyKXESU5VGpkjeJqdrOPayk2589nh0CpV6axcaqBB40t7M0h6RT+dGDzNziD7+mr9Uul8y1quiKukSnm0q1TeePjir6oat/AEK/6qHo5rW/ZJHa9T5B3HLWLPhnDFeJLqmghVsY4gCIIgrhmcTuh27fJbLR0/DuPixbAOHVoDRhEEwUNPnUS1YS0OPBvJ7aJ8aoIgCKJ2IbjdgMvlv54xiBcu1IBFBEGoqbqIkSLyo/K3eDEGVTRJIbgQRIgBUIox8EIMAMC4KIvAiy+4VfVrnL5lWasSZtAGr3HECy7ImsBCDADAuDE9KmEGB1fjiK9pxAsxAICBW1bXOFJEhcoVX+BqGqlr+FRBjSPFd6j2cbimGzoXYP2yWMiq6F9iYxuqAqpxRBAEQVwrMFEE3G6/9Z74eFip/hZBXBXQHCOiWsjP1mLJB/VUThFDXH0HHv33sRqziyAIgiBqgqhHH4WUna1YJ+v1KB4xAnJiYg1ZRRAED70eJ6qFFR8nITvTqForIDreiYgY/zdmBEEQBHG9Ip0+Dd2OHf5JFiYT7HffXSM2EQThT6UjRmX1jMoQxOBpTLwYg0KIAVCKMQQTYgCUYgyi0p9TpNZx4gt8Wh0AiJJvWdIq7QhWqwgAPJzggsiJNkjl1DtSp9K5tL5TbeOEGYpV4gvlpdLxywZRVf+IW7Zy504SVHWMyqlxJJaTZhcsfc5/FlFpv/NndX4tAGC3SfCAAfy+mSqtkWocEQRBENcR4pkzEPPz/dcXFEA6cwaeBg1qwCqCINTQkx9RLbTskg9RVaQXAGLqOmrAGoIgCIKoOdw33ABPvXr+6+vXh6tVqxqwiCCIQJBjRFQLPQZlo0WnQkVkrm4jG4ZOPF5zRhEEQRBEDcDMZpQMGQJPdLR3nScqCtbBg8Eslhq0jCAInktPpStLNSovzUil9KVQqZPUdYx86U/BFOr821RzVLgUOXBpdYJLlXLH9VPXOJIUanPKBDGJC3IoFOpUKXeSIh1POb6ixhGnUGeVlIMUaXypdTqVYp0+RMU6Ha9Qp0pJLK/GUXmKdRI3Pr+VR6UyWJYaKWoZxi/Yj3VL4rB7bSTqJNtx9yNnYYnyn1+kTpfjU+sqW+OoqtPqSm2h9wgEQRBE5Sh5/HE4uneHefZsgDGUjBgBd+vWNW0WQRAcpEpHVBuSBug5MAc9B+YAAESVk8MYcHCrBQe3hKFlp0I0vamkJswkCIIgiCuCu2VLFEyfXm4f6dQpGH78EZ569WDv3x/Q0KMaQVwp6NdG1Aguh4B3H26G9G0W2Io1MIW50fzvRRg3Jx06bcXbXw4eDyAFrj1LEARBEDVG2Ntvw7R4MaScHDCdDq7UVOR9/jnkpKQrZ0SZSJZIWRJE7aPKHCNepS5UhTpApVLHRxQ8KlU6bkwmK8fnVekE/ofsVj79Ci4uDculUjvj1eacyvE1nLqdMpVOFQHh0/FUZ9bNpda5ubc/do3SDquWS5fzU6zzKc/pVKl0vEqdXvZtp1aeq4rir4pkOj/tUS6FrZzir1+/n4Rd6yIAVtrJWqTBjt8jsXxGXQwYc9bbT52qVxnFurK0ul8WxeCnL+JQmKdBRLQbdwzNRf/B5wNuEwhSrCMIgiCqC83+/TAtWgTponqd4HRCt38/Il98EXkLF1b7/oX8fEQ+9xy0Bw4AAFwtWiB/6lSwyMhq3zdBXC1QxIioEQ5tCfM6RWUwWcDejeEKx6iq2P57GOa9k4TCC6Webc5pPea+o0N8PSdu7FpU5fsjCIIgiEvBNH++1yni0Ry7MkXRo0aNgmHDBt9+MzMhFBUh75tvrsj+CeJqoPKOkVpgIVQxBlnVjxdj4IUY/CJL3LKqhpIiusTXO3Ipoyp8NKlc8QWHuo2LpDiC1zvilyWHuo2LOml8kSynVvkVWDW++j9alTiCTvQdp15URtR4MQZeiIH/DJRf44iPEolCaDWORNX3pIgS+QkzcPvS+Et5A4BGx8oVWOAjSJcizPDdnHivU1RGYZ4WK2bHVcoxqih65HQI+PL9Oti/xQytXsbdw3PRoTc5YARBEERgZE6xjofpAtcFrFLS06Hdv99vtXb/fkhHj8LTqFG17l46fhxhU6ZAysmBJzERhS++eGXTBwniIhQxImqEXoOycXhHGGzFvkvQHO5G34cqHy0qLpDww5x4nDulR89789C2S7E3O9NpD+y4O2xVnxLHGPDqQ42w8w8LyrzBA9vM+NeLWfjH8NBT9wiCIIjag3XECJiWLIHm5EnvOiaKcHbsWCXja7dsgemrryDHx6Nk1CiFIybk5kIsLvbbRiwqgpiXV62OkXTkCGIefBCaEye863TbtiF38WLIdetW234JIhA0UYKoETrffR7/eOIU6jW1IiLOifrNrLj36VP4W1//NIJQOLrPiOfuuAGLptbD6q/jMOnhVPznyYbeoFVqW2vA7Zq2C7z+cti+1oKD203gQ2TFBRqsXBjjF0QjCIIgCKA0YpQ/ZQqcbdrAEx8PV3IybPffj4JJky577IiXXkLMQw/B/NVXCPvgA8TecQd0f/3lbWc33QR3crLfdu4GDeCqZknx8ClTFE4RUJo+GDZlSrXulyACUS0RI6ZKdQtVjCGoEAOgTJcT/eemeDfj6x15VOla/BjqGkfcsuRUbseLKmi4tDq/VDqF+ILSRr5cEePS6jwqGU675Nu3TqNMgyvh0uwMbqUwg1YMnD6nTqWTuHQzvzaRT5ELUZhBUNcg4tt8H3elRWHVrCRYizSIquPE4PHHcNeTp3H7o6dhy9fBEumCpLmYJak4dcEFFvi0utkT6yPrmMG7bLdK2LwqArs2mNDuFiseeO4MDm4zI2OXCS6HBK3eg6btrBj4zJmANY8cNgGnjxoQW9eJ8GiVEIgKdb2j3ZsssFv9Ze+KLmhgt4owmgOnERIEQRC1BI8HYdOmQff77xAYg7NdOxS++iqc3boht2tXiHl5kE0mwGi87F1pDh6E4YcfIBb50rk1J04g/O23kfvdd6UrDAYUP/YYwqZOhSYrCwDgrlsXxU88Aej1gYYFAIjZ2RBzcuBu0gSoZMqfmJ0d2O6MDOg2bIBgt5c+W1z8x/R6CPXqQeN2Q7ZYIMfFAdpqlrMlag2USkdUO1t+jMHsMU3hdpY6C5l7gVOHTJi4fDei6rgQEetT1CvM0yD7uBF1G9kRFqAYbDByTvn/4baXaLB2WSza3XICBiPDu9+mY+NPkdj7lwWtOxeh820FAdVIF38cj1+/jMX5c1pYIj1oc3MRnpmWGbLEd6sOxTCYPH7OUXiUG3ojOUUEQRC1neihQ6Ffs8b7HlC7axc0x48jb9EiQBAgx8Qo+kuZmRDz8uBq2fKSHRDj8uWQLlzwWy+eOQPBZgO76HzZBg+Go2dPmD//HBAElAwdCjkhIfCgdjuinngCuh07IBQVwZOUhOLRo2F74IGQbBIKC6Hdvx/avXshnQ2cQq/fsQP6//u/oGPEX/yfCQLkuDh4EhPhSUyEu2lTuJo3h7tFC7gbNqQ6UMQlQVcLUe3MH9/Y6xSVkXvKgOUf1Mfwt48CKH0RNGd8Cnb8Fo38bC2i6jjRvs8FjJqU6Rc8DITBHCiqw5DQ0O5dkiSgy1356HJX8HS9fZvN+HZGAoryS38adquEtcujEFfXiaEvZVVsCIC/9SxG85usijlGlgg3+g/Oo7IQBEEQtRzp6FHo161TJEcIAHQbN0Jz4ADcLVr41hcUIGrUKGgPHIBQXAxPgwYofvpp2O6/P+T9uRs2BJMkCKoyKMxk8hN2kBMSUPTSSxWOGTFhAgw//+w9BjEjA+H/+Q+cN98MT8OGys6MQTpyBPrNm6HbtAm6rVuhycz07TMiAkyng+B0etd5YmJQ+NJL8CQngxkMpVlEF/8JDgciRBFFZ85ALCqCdO4cxKwsSGfOQHvwYKldFzOQZIMBrptugrNjRzjbt4ezfXuw8PCQzx1R+6g6x4hPJwpVoQ5QqtQFUagrXeRS7tQ1jvgnZ/7J06Oyw83vS/mgLjp9YzJJuZ2o5dPsAivUAYCsSLNTpftp+TQ7Lh1MqzxOt9Znl1WjDA1rJC5FTqVKx9c14j/z9Y0AdZqdMiLDq9RpmbLNzl0q6hpHyjF848sACnO0sBUFvswy95suJsoxrPoiAWsXx8PlKD3+3NMG/PZ1PJJbWtHnAT7MHlh57u9985F11ACnw3f+khrbccfQ7JDrHQHAd3PjvE5RGR63iJ1/hGEoQnOMZMiYOO8wvnw/AQe3WqA1yLhraC469SVVOoIgiNqO5eOP/Z9jAIhOJzTp6QrHKPK552D4809fn8OHETZ5Mhxdu0KOj/cbIxC2e++FZdYsaA8e9K5jGg0c3bpVutq5bscOvzKGUnY2zDNnovCddwCbDfr162H45RcYVq+GlJMDoNThcXbsCOvgwXC1bAlXy5aQ69SBdOwYwv7zH0jZ2aWqdOPGlatKx2JjYc/NDdxot0Nz+DC0Bw5Au2cPdH/9Bcv770OQZTBJgrNDB9h794ajTx+4U1P9p24QtRqKGBHVirUo+B/dqATf26GtP8d4naIyXHYJm36IUTlGgXnghdOQBGDzqkjYrRLi6zsw6o0TMIVdWuqaJ0j2nrqocEXoDAxDx2VRIViCIAhCCRcZ4Sl7aPciywEltDVnzsD0xRcofv750Pan1yNv7lxEvPwyNJmZYAYDHN26oXDChMpY77UtEJrjxxE1ahT0a9ZAtNkgh4XB3qsXnF26wNGhAzyNGwd0RDwpKcifMaPy9vAYDHC3agV3q1awXUzFE4qLod2xA/o//4QhLQ0RkyYBkybB1bgxbAMGwPrPfwZPGyRqFVfEMeLFGColxAAof0jqHxVf/0gOLr4guHkBB9UTMBetEjWqOkZcZIhxz+4ajTpiFGKNI+6s80IMACBzjU5JGZmxcsIMekn5tqlY8s2x0XKRH73qOMuLGOm4NlEVZeGFGeyKiJFyDFHRJiMq2Y7Y+jZkHzMr+mm0Hjzw+tFSAQVB8CtNVUbGdgtmjkvBv17LhEElWqCIBAnAoBdOY9ALp0vtreQboD4Dz2PH2gjYSngnjaHpTf4ypqGgFmYgR4kgCKJ2Yx08GMYffoCocpA8CQlKeeoywYEAmBctgpSVhcKJE8HCwircp6dBA+QtXHhZdvO4WrSANiNDsY4JAgxr18ITEwPbgAGw33orHJ06VVqUoSphFgucXbvC2bUril56CdLp09CnpcG4fDnC330XYf/+Nxzdu8P6wAOw9+9f6Ugace1DT2lEtSKKwD9fO4aoRJt3nd7swv0vZCKyjgt/fR+LdwfdgOxMPQTR/w2UrViDtIUJePdfza+I1HWHPoXoNzgXsXVLb1iWSBf+1rsAI149Xf07JwiCIK57XJ07w3bffZDNpS8MGQBXUhLOL10KoFRoIeqRRxBz330QrIFLSkjnzsH81VeIHjIkaPSmOin497/hbN0a7KKwAQPgTklB3qxZOLd1Kwreeac0Ve8qcIoC4UlKgnXoUJxftgzn1q9H8ZNPQnPoEKJHjUJ8164wz5kDIUBdJ+L6p8oiRnqjB/c/koPUVlacPWnAVx8moPACZeoRQKte+Zjw805s/CoBTpuIroPPIT7BiVVz62LZ1AawFvrCaZJGhsctQKXZjaN7LNj7ZzjadancPJ0zR/VI+yYWUfEu9B+UB4Mp+I3krhHZ+HuffFgLJTRoZke9VMfFFspDJojaREmJgP/9z4z9+7VITnbjySeLER1NxciIy0QQUDB1KkoefhjGFSvgTkmB7b77AL0e4tmziB4yBNpjx7zdZa0W0Ggg2Gx+dyHtvn3Qp6XB0a/f5dlUUADzp59CLCyEdfBgeOrVC9pVt3EjwqZPh27PHshGI+z9+qHw+efhadbs8myoITyNGqFo3DgUPfccDL/8AsvMmYh47TWEvfceSkaMQPHo0STYUIu4dM+l7LU9l6pkNHsw5ZvDSG1ju6h9UIgOvQow/oFGyD6tD12MIZgQA79fKFPuAJUYQ3kpd/yySmBBcPvCpoJLmaYmcKl1Yjn1jiQuLU52KJoUtZCYxAsxqPv5xnBrlKFcByc5ydc0AgAtl1qnEfmUu/JS6TxB20RVfSIdC9ymTrkTg9Q70kfI6PuIL+riAbDumzoKpwgAIAA6owdOm/LEOKwSju41o3WXQm58VSojbwvznceFk+th1ZdxKDivhSAw/Dg3Hi98fASN29gUQgxFFyS8O7oRjh8ywmkTEV/PiX+NO+11jALVO7pU+NQ6SqsjiKuXwkIBAwbEYM8eLcpeiqxebcCiReeRlESy+8Tl427ZEkUtWyrWhU2bpnCKAEB0ueBo2xaao0ch5eUp2+x26HbvvizHSLdhA7TjxkF3tFQl1rRwIYpHjEDJU08p+mnS0xE+aRIMq1eDSRJkjQbMYgFk2V+J7lpEo4H9jjtgv+MOaLdvh+WTTxA2fTpMn3+O4qefRsnQoeXWdCKuD6rkyWzgU+fQtJ1NIQhXP9WBka+cqYrhiesQjxsoyff3yz0uETq9/xtZS5QLbbv712GoiKxjeqz6KhYF50sdMMYEnDlqxMQHm8GlUhWcNrYhdm8IR+F5LexWCSfSjZj1Rj0UF1CuMUFcC5SUlGDq1KkYO3YsnnnmGaSnp1d6rPfft2DPHh34SPHhw1q8/Ta9OSaqDykrsPqpIMvwJCb6rZfDw2Hv3bvyO2QM4W++CeGiUwQAUk4OwqZOhe7XXwEAYk4OIl58EXG9e0O3aRNkkwmCxwPR7YaUkwPDL78g4vXXK2/DVYjrpptwYdYs5KxcCVfr1oh44w3Ed+sGw88/17RpRDVTJY5Roxa2gOvjklwB1xOEpAHCY/2vD53Rg8735CAs2jcpVavzoE23fDRoHvg6K4+1S2NQkOuf41xwXoOpT6V4l512AZmH/CuMnzthQNrXMX7rCYK4+pg7dy7atWuH//73v/jPf/6DpHLkfivi0CFtwPWnTlGKOFF9uBs3DrjeEx+PktGj4YmN9a5jOh0cXbrAdeONld6fePZsQGdMdLkQNXYszLNmIb5HD5i+/holw4fD1q8fRNW8J4ExaLdtq7QNVzOuNm2Q9+WXyP3yS7CwMESPGIGohx+GeJrmHV+vVMlf+As5gW8g1kL/N+1MJT8WTKXOL12On1xYXoocp0THVOlgAqfqplCoA8A49TZBlcYncml3fKqbqFKl0/B1jPwU63yfJe50yaqoBZ9y51Gl0rm4VDpeoQ4IXuNIp6p3pCknlY5PkZNU505SpJFxbSrXWgqSSgcA4MdkDL1HnsY3Exuh6Hyp4yKIDI3bF+KBN46i8/3Z+GVmElwOEX+/7Ty63p8DDwDFkKrLQJ1aBwBRCS5AYABTtwnI2GVGcTFgssjwMDGouEPpnCclVZ1Wp4bS7Aji0rBarThw4ACeeOIJAIBGo4HmMirex8QE/n2Gh1MaHVF9FI0ZA9369dBxNYfcDRqg8OWX4UlNhatZM1hmzIBgs8Herx9sAwde1v6YxQIWJD1MKihAxMSJcHTsiIIpU+BOTUXE2LEB+wpXQh2pBnF264aclSthnjMHYe+9h/hevVD46quwDhlCdZCuM6rEMVowrQ7a3lyMusm+t/x55zT4ekZoxceI2snf785FbH070mYmwWGV0LRjAfqPOgNBABq1LcZjHx+C9jL/4PS4LxdfvZeE/Bz/qJHTJsFaKEF2CziRbkSdeg7knFLeIOKSHOgzKHAROWuRiHUroiFJDN3uzofRTA9MBFFTZGdnIzw8HDNmzEBmZiYaNWqEYcOGwWAwVGq8Z58twqZNOkWEKD7eg6eeIqUqovpgUVE4/+23CJs2DZqjR+GJj0fR8897i526W7Wquno/AFhYGFzt2kETJALibNsW57/9FoLVCt2WLbDfcQcMq1ZBys9X9mvVqtz9SKdOwfDTT3AnJ8PRp09gOWynE5pjxyCdOAExLw9Sbi7E8+dLlfnKpMtlGcxkgpiYCLNOBzkyEu769eFp3BhyVFT1OilaLUoefRT2229H5PPPI3LcOOh/+w35U6eCRUVV336JK4rA1KGZCugr/t/FLZUXX0oLKx5+OQtRcW4UF0pYPCMe29ZezMUu5+23ImLE91NFkgSFcILyByXwy5Ki0JCyn44L1WiVbYxrYwZlBEzmlj0m32e3SWmH2+Sz32lWHrPL7LPfbeHXK7rBzT1ceywqYQOzL/XMYFLWPwg32b2fowy+lLNovTLkHa0r8X6O0ZYo2iI0vr6RknK7MNHOffaNbxCV6XBmwWeXwU/cwXc8Wi7iovWL/PDbCGAMWDmrLv76IRZOu4T4BnYMf+sYCrJ12LcxHM3/VoSmfyuGhrtGeFGF3X9Y8NZDzeF2Kb+Txq1L0KpzITb8FI3zWVqERbkhewS4nCIcNhF1kx0Y+HQWeg9QTnYFgD9/iMDcd+rhbKYBAEOdBk40b18MUQA69cvHLXcU4EK2FutWRCI6wYWbbyuAVhf4p+awCVj/fRRsVhE9/nEBYVEeihjVEsSEjIo7ESFx5MgRTJgwAW+99RaaNGmCuXPnwmg0YtCgQYp+aWlpSEtLAwBMnjwZziDFNgFgxw4Br78u4dw5ICoKeOEFD3r3Dvw71mg0cLuDVIi+CiF7q5crZu/Jk9A88wxw/DhgNEL+v/+D/PTTELZuhbB2LVjnzmCdO5fvMNhs0LZoAUGVUscMBri/+grC7t2Q5s4FTpwAYmLAIiMBux3iiRNgUVFg7dvD/c03gNkccHjppZcgLFoE8dw5MJ0OrHlzyLfdBmHrVsBkgiAIEA4dAjIy/GpYMqMRsFhKa3+IYulxlJRAKCjw2w+LigJr2hTsb38D69QJcseOQIMG1eMsMQbxww8hjR8PJCTAvWABWKdOQbvT9Vu9hGKvLkTp+Mo7Rn4jhXjhVcZJAhSOkqB+08AtC7zanNox4pe1qvQ/zlFiBuXJk/W+vrLR189jVI7PO0oulWPk5B0j7rOfY2RhXD/lVyObOeU5s9IhMZl8MngRRp8TE21QOjgxep8zxDtJABDFOUa8kwQonaFwKbCTBAAGgXPeVAVkDYo0vsBOUmmb77NOELDig3r4fkY9OLiiqwazG4II2Io0MFrcaNq+CC9+dgjai8INfFqdCBFfTUvET5/XQVFe6XdZp74d3e/LxXez6sJu9Y0rSjL6DsrFzXfmodXfrdAb/X8eLoeAp/q1wKnD/nOSAECr9yAuyQlbsYQL2TpIGhn1GjswftZR1G+slCvcv8WE6c81xJljejBZQHw9BwaPOYv+D5z39iEn6fqFHKOqIz8/HxMmTMDHH38MADhw4ACWL1+Ol19+udztzpypGpGg2NhY5OYGji5fjZC91cuVsFew2RB7553Qcml3stEIT1ISpOxsiIWF8FgscLVvj7y5c8tVVIs9ehSavn0h2kvv7x6zGY7evVE8ejRiHnxQESFiAEoGDID9H/+AXK8e3KmpQcfV7N6NmEGDIKkcGQZfNjzT6yGbzRCKiiC6XJBNJjgvih8Eksk2fvMNIqdPh3DiBBgAOSEBtltvhSDL0KSnQ7tzp+84EhJg79sX9v794bj55ipXldPu3o2oRx+FdOYMCt56C9aHHgrYj67f6iUUexMDiJcEgp64iKuev36MUThFAGAvkWArKnVMbcUa7FoXiW+nB6+7MOjZM3h3+X7cOTILg547hSk/7cex/WaFUwQAskfEiXQjbuxWFNApAoBDO8w4cyz4H1eXQ8KZowZcyC51sD1uEZmHjPjoxQaKfowBn7xSH6ePGMDk0ltE9ik9vv6wDqxF9NMkiEshMjISMTExXkdnz549qFdOLRaCuNYxLlwIjUp5UbTZoDl8GGJhaWkLqbgY+nXrEDZ9etBxNHv3Qjt4MIBSh6d46FDkzZ+P/BkzYJk71y9tTgCg278fzh49ynWKAMA8e7afU1Q2hvezwwExLw+iq/TFqmi1Qr9xI0wLF/pvZ7Ui7L//hXDihHcc6exZ6LZsQcE77+D8kiU4e/Agcn7+Gflvvw3nTTfBuGQJYh58EAlt2iDyySeh27gRVVUx3tWmDXJ++gmOLl0Q+dJLCJ84EfB4KtyOuHq5NuV11FWeg9Uu8qhCslyUQlBFMxSpey5lmyDyggucEINKAIEXY5BU+WF8sIpxn9UiDbz4gqwKjJVX48jOKToohBgk5QO8RvQE/AwoxRj8xRd8yzq+3pGqjhHvaqvFF5QCDpxIg6CKjHGfHTKDtTDQZaqucyXg4NYwyBf/2J3MMGDv+kiktC7BDX8vhiAAdRvZMHxipiLNrjxkvxpNpduZLB7oDQy2kkBbBbEPwNkTOjjtAnSGUhvPndQh57R/aPfsCT22/BaO7v8ovRmpRRoogkQQgXn44YfxwQcfwO12Iz4+Ho8//nhNm0QQ1Yb20CG/1DPA/+4jMAbd1q2+FYxBu3kztHv3Qg4PR+SECUBUFHJXrIC7grlC/Bjl2rZrF8xz5sC4bFlIw/nZ7PFAv349Sh57TLFet24dpItOEY/m1ClIR4/C07gxoNXC1bo1XK1bwzpsGGC3Q//HHzD88guMP/4I07JlcKWmIn/aNLjatw/NvpISsCCpgiwyEnnz5iH8jTdgmTUL0unTuPDRR1Tz6Brl2nSMiFqDIABRCQ6cP1XxBGqNVoYsAx+PaYJdayNRfEELg8WN1N0msEQAACAASURBVDYlGD/vEPRG5Q2k14Ac7NkQDjsXjZI0MtpyRWQDkdLShvpNbUjfYSm3n599OgaJUy40mGRoAsw70upkWCLojRNBXCoNGzbE5MmTa9oMgrgi2G+/HcZlyyDaKi5lwS5OHxCsVkT/61/Q7N4NqaQEDBcFGNauhTuAUEnJ8OHQr17tl0rnDCIRrtu0CWH//jf0mzdDNpthHTAA+t9+gyY7+5KPjwWYE8LCw8F0OggOh1/fYI4LDAY4+vSBo08fFL75JgzffQfzokXw1K1bvgGyDE1GBkwLFkCw2eBJSkLxM88E7itJKHzzTXjq10fExIkQRo5E3syZgDFwyj1x9UKvnomrnntfyER0km9ekyh5IGmUTo7B4kb3ATlY+20cNv8UjeILpTcBe7EGezdEYP7b9f3G7XRbPvoPyUZcPQcEkSG6jhO33HkBA58pf86BIAAvf3oUrTsXIibBAXOYG6Kkio5p1G/xGJrdaIXEvYqIjHWjYYDaTPUaO9Cua1G5NhAEQRC1G0ePHnB07QqZmzPtiYmBbDIp+nksFlgHDAAAhL35JvQbN0IqKU15EADA4YCwdm3AfbhuvBHFo0fDlZwMJorwxMbC3rcvCt94Q9FPs38/oh96CLH33w/NiRMomDgR57ZuRcHUqSh47z04W7eGJy4O7vh4yCqHxxMeDo/KKZMjIlAyfLifPc5OnQKm77maNYOckBD4RHEwoxG2gQORu2IF5PLmnHg8MPz6K4zLl4OFh6Pw9deh37ABhh9/LHf8klGjkD9lCvRr1iBm2DAIITitxNVF1UWM+LBqeUIM6totXFoQX+NIEFX9ZK6fyp3j9fMZl9vpVyNJ9rWVzenw9uXS7tQ1jsDVOAJX40hU1TuSnL5lpq5jxIlC8OlzsuobCDXNTl3jyM09iNs1vj+SJap0v/JqHPGCCKKgFkQIXP9IVKfccd+vOpVO0bfcKDwnzCAATTsXYtyS3Vg5ox5KzmvR/tbzuHBWhz8W10HxBS3MkS50/kcuOt+Tg/eGN4fL6S8DeniXBR7vTn3jD3vtJO5/6gxOZhiR0MCB6ITSHGcPAyTVdcyn1sUlufDutxkozpcAAVi9OBprlsbAXiIitq4LPe7LwU9f1EHOaR10ehlNb7Ti6feO+9k1bsZxvPd0Qxw/aIDHLaBushNPv5cZUMm0DD61jtLqCIIgaimCgAuzZ8OwYgWMv/wCT2wsip94Asbvv4fpq68gXrgAOTIStnvugf3eewEAuj17/IYRnU5g2TKgf/+AuykZMwbWhx+G9sABeOrWhae+70WjUFiI8EmTYFq0qNSBmDABxcOHKyIljt694ejVC2JuLpjFAv2qVTDPmVNqX0wMip56Crpdu2BcsQJifj7k6GhYBw+Gs3t3f2NEERc+/RRx48aBHT4MJklwN2uGCx98cJknU4nm6FHotmyBs21b2O++GwBg79ULckREhdtahwwB0+sROXYsokaNKhW+IK4ZKJWOuGrxuIGNi+tg3+9RCItx4bYnTyEhyRc+v21kForytLBEuaHRljo9Zap0aiRtcE8sLMqDGzpUrjaJJbLUSfzHyBz8Y2SOoq3PgHwUnNdAb5RhMJU5M0pnKyzSgze+OAJrkQi3S0B4NKXQEQRBEOWjW7sW5i+/BJMklIwciQuffuptK3nkEZSMGAExL6+0tg+vwhtEnj5YkVdve1gYnB06KNbp161DxHPPQTp7FiUjR6Jo7NhSKe9ACALkuDgAgP3uu73OhtesXr1Q/OSTpY5RVJSfqjCPJyUF7t9+w/n0dECjCahcd1m4XLC8/z5cN97otdP0+ecwLVmCvNtuC2kI2z//CcHhQOSLLyJy3Dhg3ryqtZGoNsgxIq5KZBmYMfwGpG+IhMddGhXZ83sUHvlvOpp1vKi2owEi45Wy5beOOIP9GyK8qXQAoDN40OG286gsjAHWIgkGk0eRChcKETGh1QEwhVFxWIIgCKJiwt98E6aFCyEWl77Q069bh+Inn0TJI4/4Omk0kOPjFdtJJ09Cc1Himn9F54mOhvz00yHvXygpQfibb8K8YAFcqanIXbECrptuAhiDUFgIZjKV69gERaOB4HBAv349pKwsSFlZELOySo9Tlr3/mMUCqUEDmCwWyPHxcDdpAlfTplUmdiA4nYBGA3u/fgCAsPfeg3bXLuTNnAlPw4Yhj2MdMgRSVhbCpk+HJzUVIDGYa4LrwzHiVVk4BTmmUqUT+LQjtZyiyMupqdKT+PQ8N1czyaUcQ+S280uzU6TP+aIXauU5Pn1OnU7Fp9LxCnWl4/i+Sgc330VSpdJpuBRFjahuC02xjk+J81ev4/YtqlLp+DRKrt6RqM6rY8Cun6OR8VeE1ykCgLzTBnz7nwZ48du9F/upthMENPl7Ee579gR+W5CAglwtzOEetO+bh9tGZMGbqan4agIrzwGAhzGsWx6NFf+ri/xcDcxhHnS6NR8PjjsNmU87vISperJCme/yi86RYh1BEETtQczJgeG777xOEQBIeXkwLVyIkqFDgQACCgAAmw3RDz8MSBJs998P3fbtEAoLIcfFoeShh2Dq2BEIoW6NlJmJ6OHDoUlPR/Gjj6Lw+ecBoxGG77+H5eOPIebkgFkssPfvj6KXXy5/aoUsQ7tjB/Tr1kG3Ywe0O3dCOu97ickuRpnksLDSB6KLBV6FwkKIP/yACE6AgWk0pQ5S69aw9+gBR48eYCGkvQWCmc1wdOuGqCefhCs1FdK5c8ifNk3haEqnTwNOJ+SICLDo6KBjFT33HMSsLJjffhuG+vVhv+uuStlEXDmuD8eIuO7YmxYNt8N/ok1+dsVvhPoOO4s+D55Ffo4OYVEuGCsWtAMAOGwCdDqfQ3r8gAFz32iA/JzSiaJ5Z0vrDIVHu/GP0edCPhaCIAiCqAq0W7dCk5Xlt17KyoLm6FG4b7gh4HYRr74K7f79OD9vHhx9+gB2e2mqXXw8oNH8P3vnHR9Fmf/x95Rt2ZRNg0hC7wiIqKB4YsHeDvEsYEfOU+4UFQs2BPV+KHgH6okKNgT0FLsoKoKKCp5SPKR3Qgspm7LZbJ2Z3x+bbGZnSxKlnc6b17yYnXmeZ56ZzO7M9/l+n8+XtIS1YrF+/z3Zo0YhqCoVb7xB8JRTIsfesoWs8eORdMpz0u7dqC4XXqOXRFWx/vAD9k8+wfHxx0glJWiCQLhrVwJDhhA85hjCvXqhtGmD0qoVJFCmA8jLzaVi+3ak/fuRN2zAsmYNlrVrsX/+OWlvvYUmywQHDMB/7rnUDRuGlp3djDNsxDdsGMEBAxCqqwkffXRM/zMmT0bw+RArKxErK6l87jm09CQqtYJA9aRJOHbswHXnnZT36EG4a9cW9cXk0HJwDKMEI/nJy+q9CImFGCK7dOXU2FFxvRhDjKa/8bh6L5FBmEFTk4svCLp2NJ0Qg2DwCok6r5NkFE4I6PMf6cQi4rxCunUpufhCXI4jnbdK0XmP/HFeIZ2wgRR7nlbduVkMnqBk4gv6dYjNa2T0Jokxnr3GVaNIAwK06l4HggZa7DVwpIdREnp+iLnvJFkg+6hILHXcrB394QRYuyyD1ye1x11ixZ6mcOxpVVw3vpj3n28TNYoaCPgkvvs4O8YwSpbvCGDbOjvbfk6j14Ba2nQM4vOKlO+1kl8UwO44sN4jiPcgNWB6kkxMTEz+91E6dULNykI0JE1VXa6kqmyON9/E+cYbeG67LWIUAdjtqVXZDKTNnEnWxIkgSSitWuF87TVC/fqhZWSQ/uyzMUYRgOj3Y1+woNEw8vtJmzcP58yZWLZuRbPb8Z96Kv777sN/5pmRuUmKgvWbbxB37ybUvXvEKAqFkHfuRMnLi52/JAhomZmEMzMJd+3a6IlRFCwrV2L/4gvsCxeSNX48mf/3f/guvpjaUaNijZwmUIqKkIJBHG+9he/yyxFqa8kaPz6imHf99SgdO5L56KM4Z82i9q9/Td6Q1Up47lykgQPJHjWK8k8+SS4tbnLYMT1GJgcd92YLa+dmkZYeov8NFTjzmhYYGHR1Kcvfzmf/psYfD3tGiIHDWp4LIRk1bpm3/9mWr99qRcDXaGnu32lHtmr4vYmNiVCwaSMmFBT4+6hObFieTm21TFZuiMycEMGAiKdSxpUX5pSL3VxzT/zIn4mJiYnJbxi/n7Q338SyejX+IUMInHtubDh/CsLduxM89lhsX30VHU7TZJngiSeiJgjpknbtIuuhhwicdBKeu+5qcVfF0lKyR4/GtmxZfQfCyHv3Iu/di+Dz4Z4zJyasT48QDCL4fDhfegnnzJlI5eUE+/al8qmn8J93XoxxIO3cSfaoUchbtiAGg4TbtiVw3HFY161DKilBrRd/qJo2LfX8JUkidMIJhE44Ac999yGvX49z1iwc77xD2ltv4R88mJoHH2y2gaR06kTIH0kXYp8/P5LL6OabI30PBFCdzqQ5nWIoKqJy+nRyr7iCjMcfp+bRR5t1fJNDj2kYmRxUfnw6mzVzs/C7I7fa+o+yOP2hffQ5pzJlPWuayqhZG5g/sT3lxXZsaQoDLi3jtBElB6Rfe7bYmTKyJyXb45OvhUMiP33l4qKbSlixOIuwQf67bbem8xLMmdyGFYuz0Oo9XtUVFqorZBrcXHUemQ9fbE2Hnj5OuagqRUsmJiYmJr8VxIoKcoYPx7JhA4Ki4Hj/fYInnIB79uxY9bgUuF98kcwJE7D+978gigQGDcIzblx8QVXFNXYsQMSgSJUHIgHy6tXkjhiBVJn4eW1Ztw6xtBTfhRdiW7QI0Zh0NS2N/FNPRd6zB/9pp1E5ejTBQYMSRhG5xo7Fum5d47F37ULasycaBSTW1CB9+CFqXh41EyY0+xzCPXtS/fjj1Iwbh/P113FOn07+OedQN2IEnrvvjirlpWyjPjxR3rmTwB/+EDXo7F99hVxcjP/iixHdbmxLluAbOjRpO8GTT8Y7ciTpL72E/8ILCQ4c2OzzMDl0HBrDqLk5jprdXvIcR/o8Q8aQPk33WTCKL+j7ZeyjbiQnJswuJCYvF4zdJ+pEEPS/TaohH5EqJ16HJsLs9KF0OnGHsEFGzSclF0dorjCDPsdRvPhC4+dghcC6eY1GEYBnr5Xvnm5NlyG10chJyZizikhIn7NA4brnNwOwb5ODrd9nsuW/6bQ/xlt/0sY8SboPKe85jTmPdUhoFDUQ8IsMHlbGii+y+GmJi7oaGYtNoePRdYyaWIyiD9tLkO9o3XJn1CjSdSTmk88rMe9fBQR8IiedW4UzUy/ocGBC6xowRRpMTExMDj+ZDz+Mde3a6GfR78e2dClps2ZRN2pU8xpxOKh54gkAhLo6bAsWYP/oI/znnhuTPyjt9dexffcdVZMnoxQVtbivrttvT2oUAZF5NtXV+C+6CP9nn2H/6ivE6upIwlm7HdsPPxDq2ZPyZ55JaQQItbVIO3bEb1cNAlqKgnXJEtLmzEHo0wf69m32O6XmclE7ejTeESPImDYN5yuv4PjgA2ruv5+6a69tuh1NQ9q9G8uGDaguF9ZVq7B/+im1o0dHEs6GQtgWLULatYvaW29N2oxn3DjsCxfiGjuW0kWLDpiSnsmBw3w7Mjlo7P7KTl1JvO3tLZGp2dO8kTFVhVl/7cqzlx/Nu+M78exVvZl+XU9CgV9nOLj3JZ7Q2UBBBz+iCGOf38LDr6/jsjt2MfbZrUx6fz3OrKZDASU5ed4kPdvWpDHtjo7cdk4vFr6ZXNnGxMTExOR/H3n79rhtgqJg+/bbFrdl+/RT8s46i+wxY8i+7TZanX021m++ibRZU0PGE08QOOkk6kaMaHHblpUrsWzalLKM0qYN4c6dQRSpmj6d8n//G99FFyFIEqgq1Y8+StmnnzbpGdEkqdny3pYtW3Ddey/ypZeSd+GFiBUtS8WhuVzUTJhA6eLFBPv3x3X//WTfcgtCknDAKIJA9cSJWP/zHxwffohYXk7N+PERD1gwCBYLVc88E1HYq/8bJDx+WhrVjz+OvH076S+91KK+mxwaTMPI5KDhbKMg2eLFAGSHhq0ZxgXAd7Nbs+azHOqqIoaUzyOz/utsPvlnu1/VN3t6kuMLGu16eBn1f9uim7r0q+PKsXsYeG5Vc8PAOfHcSqx24zESGUsRA29/sY1/TzsKb03LvpLu/TJzphQwZ0oB7lIzMtbExMTkSEZzJI5UaLG0dDBI5qRJWHbsQNA0BFVF3raNrAkTQFEigghuNzXjx7c4Ukfct4+cUaNShvaF27WjZty4xkgZRcHx8cc4PvqIUM+elC1ahHfkyOYZPA4HoQRqelqCfjdE+wg+H9affiLr7rubd1IGlE6d8N54I4ETTsA+fz75552HvH59yjpaTg5VTz2F5777qL39dsLdukEgQPqLL2JdsiRSJi0tkgcpBYFTT8V/5pmkT5uGWHJgpgeYHDiOrDepJAp1kV26MDgx+Zc8JlzO4IZNGS6ny3mkGcLDBEGXpFN/7HBsHwXdW7NoyDOkD33Tq8tJgdhjqbqQuHjlueSqdFJMmJ2uDTG2YFAXSucz5jjSqdQZQ+Rk3WdLzLpBlU63L3OAj8yuCpVr9NdCI++YIGRI+Osl+aQkSmoAaxbnxOQyijQhsHVFBiFjuFlM+FyS7QCCwBnX7GPPZge1lY2eo7TMEJfevpuzry3BYtNQ4sLZEucuUgztS4LARTeWsr/Yxo+LXNRWRsQXRFmjqtyCx21BELS4ULv9xTa+eDOXbWvTKNtjJSsvzFV37aVd58Q/sgvfzGH2lKMo3xdxxX/+Vi7X3L2Psy53JyyvJ5l6XaT/5niJiYmJycGg7oorkNeuRaqpiW4LFxTgSRF+lQjrihXICcLPpOJirF9/TfqMGdRdeimhvn1b1kFFIXv0aITaWmpGjyb95Zdj+qpkZ+MdORLvyJFRlTjB5yP7lluwL1yI96qrqH700RaHiFU99RTCqFFYVq9G8PsjyV7D4Yg3JgW2JUtIf/JJpD17kHfvRrPb8d5wA4EzzkheKRQi59prsf74I6LPhyrLSMXF5F10EZUzZqSsq9XniUqbM4dQz56EjjsO/+DB5Nx4I6E+fQj17UtgyJAmz7d64kRanX46GVOmUP2PfzRZ3uTQcWQZRia/KQQRBj1XxYrxGXi2ysiySqtjA5w0oekX9waShaRJv/LOHTS0AjSBRbML8NXK5BQEGP7ADjp09/+6husRBPjzI7u4Ztwe3CUW8guDWGwau7faWLMsndefbEtVWexonCRpvPNca9z7Gx8oG1c6mTBrCx16xE5qDfoF3vpX66hRBFC+18a8fxVw2h8rsdiaF8pnYmJiYnLo8F12GYLHQ9pbbyFWV6O0aoVnzBiULl1a1I7mdKJZrRHjQY/FQtq8eaBpeO69t8X9S3/hBWw//EDlU0/h+9OfUAsLSXvzTUSvl3BhITUPP4zSqVO0vFBXR84112D9z3+oeuwx6m64ob6DWrM9VfKaNWQ99BDW5csRVBU1M5NQjx6Eu3RBKShArKjAsnYttuXL4+oKqkrm1Kmxl+A//8Fz//3UXX99wuM5X3wR27ffNgo71F9DNTOTnBtvxP3ii8mNm/pzUlq1InPKFNwzZ6IWFBDq35+au+5C6dwZwevFOWMGan4+gcGDUdrFR7goHTrgvfpqnK+9Ru2YMQnLmBweTMPI5KBiz9M4eXpktClTbFrNzcjAq/ez48cM/DWNt6ogqXQe2LSSm6bBj/Nz+fGjfCxWlXP+vJdOxzTGEQ+6pJw/XGLM9H1gRQ/saSptOjUaNUWdAxR1DrDskxxWfe2KKetIV2OMIogklJ09uZCHXt4Ws33zz2nsK44fkSsptrLl5zR6Hu89gGdhYmJiYnKgqBs5krqRI39VG6E+fQh36YJ19eqY7YrdjmPBAupGjEApLGyyHdvixaS98QaSw4H9pJPImDIF3/nn47v0UgB8I0bgSzZHyecj57rrsP7wA5XPPov/j39s3NeC8D01OxvR66X29tvxn302od694+qL+/aRd/HFyHv3Rrdpooj/5JOxLV2K6G8c1JS8XjIffRTN4cB3ySVxSWJt330XH1FEJDxQUFVyRo1KbRwBgbPPRtq/n+xbbkHNyiLUrRtqdjbWb78lbd48NJsNtVUrsm+6iYr33kvYRu3o0TjnzCH9X/+ievLkZl0rk4PPoTeMmpv81Rjqkyq8R19Wp1CnGaoIumNrhn4IMW3EhodpamMfBV3InTERrBajSmdM/qr7rA+rMyaaTZX8VdKvG+sJicsZQvpUnSReQIz1WIg6dThJSKVYp1eoM6qd6dowzKmRSBy+ZQzb09PpVC8n/6WEr585irBfBAQ0ReT7NwvoPNBDl5M8jYX17Wgas8Z2Zfn8PEL+yDmv/dbF0LE7OfOa/QT9Akv+XUDx2nSOHeKm/znu+lBpXZ/rV0MBgdce6cCWVRkAdD/ew/UPFWOxRgqIhql6qRTrGrjr2W3842+d2LnBQTgs0LooiChpbFiREVe2slxGNVzLjOwQaU6V2urYY9udCpk5hhHEFqIPszPD6kxMTEyOQASByhdeIPvaa7Fs3Rp90Zf37wdBwJMq4Wg9mY88gmPOHCRvZCAt+/330Ww2qp94AgQBoboa58svI+3ZQ93llxM64YTGdzZNw3XPPViXLaPq6acjRpGm4br1VuyLF6Pa7YS7dqVq0iRUnYcpEWphIWWff566zFFHUT1hAhnTpiGVlCBkZODv3x8lPx/H4sXxlyccJvvOO8mYNo2qJ54gOHhwdF+y5Kqa04l7xgxyhw8n+y9/ofz99wn37p20T3XXXEPo6KNRs7NROnbE8vPPpL37LoFBg/BdcQVAJDSwshLatk14TnVXXknav/+NZ9y4hHmoTA495luPyRGPMzuMGhLQe3Oq99n47Kn4H5oG9m+38/Pi7KhRBFDrtvLla22oLLHw6NBjeH1iR5a82Zrpt3VjytW9CIcSGzHTbu7GwtcK2P5zOtt/TufTVwuY+teWhT0YychWmDB3M08tXMe0T9fxz4830rFXYo+aKy8Ut62oS4AOPevitnfq5aOwUyBuu4mJiYnJbwulXTvIyorxfggAgoB11aqUdcWKCuwffRQ1iiAibKA5HGg2G/Lq1eRfcAGZTz6J8403yL32WrLqcyIBOF94gbR338Vz9934hg0DIPuGG3B88AFidTXy/v3Yv/2W/EsuwdqQHPZXErjgAso/+4zyBQsIL19O1b/+RfDkk1ETCFqEunWjYtYsNIuFvOHDybrvPoT6c/XccgtKXl5MeSUzE9+IEWguF+7XXkPNzibnhhsQy8pS9inUvz9Kx46IFRVk3X03gZNOihpF6dOnI2/ciJqdnbS+97rrEIJBHPPmtfRymBwk/icNI03VYpZmo6pJF01pXFC12EVRGpdwWLcosYuunBBWY5eQbgk2LqJhkQL6RTMsRBfRsOj3SX6hcfGJMYsQkKJL2LD4/ZboUheMXWqDtuhSE3REF0/YHrNUhx3RxaPYYxe1cfGqtqSLX7PELFu+z0JV4m/VmjIrIU1MuKxZkk2tO16Su7rMwqwHurBrXXq0zZBfYt3SLL6Y3RoVoouCxr5iK5tWZhATYqcJbFyRQeleCwoaquGfHkXTYhYjmTlhco8KoaJy5Z17KOwcaxy1Kgpw9d17Et7OD7y4jYHnVNK6nZ+C9n5OPKeSB2bGS8H+GhRNTbqYmJiYmBxeEr24C6qK7csvU9aTf/45JiwtWtfjQd6yhayJE2OkxUWPB8eCBVhWrcL6zTdk/v3v+M4/n9rbbosUCIWwrlwZn3uoqorMRx5B2pP4OdZiRDGSkykzE4DAGWcQ6tcvJqZCcbmou+oqAmeeSdlnn1F7002kzZ5N/pAhWFasINyvH9UTJxLs25dwYSHBnj2pvfPOaHJWNT8f9yuvILrdZN9yS+Q9sQmE2lpCxxyD77LLAHA+9xzSrl1UTZ0ayS2V4PkPEO7Rg+Bxx5E2d27SMiaHFnOOkckRT343P5G4tliPjiMzechYYU8vNmeYgDf2FndkKLhL4g0mVRFZ962Lc0fGSmdW7LHhqYyXLPW4ZSr2Wcltk1oxpyVktwrz6Jsbmf1EEeW7G1Xp2nZJ7AHKyFZ46OVtNOQqlqQDnxjWxMTExOTIRa03EPRookjomGNS1lM6dEDJyUFyx4ohqbm5KEcdldCQEWtqcLzxBvavviLcqRNV06ZFQ+uE2tqEL/ZiOAyahmXNmmbNeWoxokjF7NlkTJuGdeVKNJstokrXMD/I4aDm4Yfxn3cerjFjyLv8cipeeQX/0KH4hw6FUCgiKW4Iew/37k313/9O9tixpM2ZE0kCmwKlXTssGzaQPnUq1p9+ItyuHd6rrmoUVUgx56ruiitw3XMP8tq1KUP3TA4NpmFkcsRz/LUVrPvQRdmmRne5wxXixKv3J63T6QQPbfvUsuX7RoED2abQ+3Q3e9Ynji9Oy4o3tNr18pJf5KdsV6yrPq8wQLse8aFsRpZ9nM2it/JQFYGB51Ry3jXlKeek5h0V4o5p2+PmLaUiUCcy79kCdm5w0KpNkOF3lJCd/+vmGZmYmJiYHPl4hw9HLi5GrK6Obgv16kVdfThXMpQOHQgedxz2hQujw2maJBE88UTUVq0S5lvSBAF52zbkPXsof++9mLk6msuFUlgYb2g5HCAIBE45pemTUZSIYtySJWgWC3XXX0/gtNOarudw4LnvvpRFggMGUD5/PrlXXEHuDTdQ8corBAcPRtq3j/SnnkKsqCB4/PF4R42Ceklu3xVXkPbuu2ROmoT/3HNRW7VKfgBBoGL2bGzLlqEUFUU9R83Bf955aPffj+PDD/GYhtFhR5owYcKEllSYPfEgxkGmVDHRGhdjubhdDRuEmH36wXQh0bE0DTQtUqx+HU0jNqun0FhOFGPKCYKAoGoIDeF9+nA9QYzuE0RdOUFAULToov+MJCAqRBcEQbcOvGH96gAAIABJREFUYrhx0UQhuo6oq6NbFxXQBBDCAkJYiAg/hMXoookRz4mqiGiiQFiRoosgQkiRCNWvB1WZoCojClp0PajKSIJGSJUJNezT5OgiCkTXJUEjpMmEtEi5hvWQJiOhEdYkwpqEKGhoFpHOZ3uoLZWxORRyOgcYfOd++lxYSRgxuohoqAioCAgC9D3fjddtQRQ1sgsDDBq+nwvv2YUkwKbvMwkHGucfuQoCXD95M+nZ4cbbRRCw2DW8VRLF65yE6svLVoVTLyvl2CFVEfd9gttMQ2PuE4XMebwtuzY6Kdlh56dvMtm3w8aAcypjymuAaLgfNd0/IYUXyFsjcv9l3fh2fg57ttnZ9F8n33+WRf8zqsh0HdxwN30fjf03aR5C+m2Huwu/ezweT9OFmkFaWhp1dU0PlhwpmP09uByq/ob79SNcWIi0aROS243/jDOofPFFtIx4IR8j/vPPx7JhA5atW1E7dcI7YgQ1jz0Gooi8a1dEOEBpFFsKt2uHdcMG6q68slGWux7nSy8RGDQIy6pViPVzeTRBINyuHZWvvIJqmNOTiOxRo3C+9hqWbduwbNuGfdEiNIuF0HHHxZX9JddXS0vDf+GF2L/4gvRXXyWcm0v22LHYly3Dsm0btu++w7ZsWWTOlCSBIBDs35/0V15B2rcP//nnpz6AzUa4SxfCRx/dov5qDgfWlSuxffst3j//ucUJeQ8Gv8XvW0YzvhPwPzrHyOT3hzNf4fype7nu3a2MmL2d7mfXNFnHlqZy+aRt3P3Rasa+9zPn/G0P679yseydVlgdCvb0MK6CAF2Or+aGyZsp6JQ4h9GwsbvpdkINYn1C3HBQYumH+fz4afIJlf46ke8+ysWvC+ULByVWfZVF2Z74UL5fyltPH8XWNbEesL3b7bz696IDdgwTExMTkyMX/9ChhPv2RcnNxT1rVvPVzaxWsNlQsrMJ//e/Ea+LppHx2GNYly5FzcpCcbkIt2uH/5RTCHfvjmaz4bn//thmlizB+v33+P/0J0L9+6PKkeeeoGnI+/fjnDmzya7Iq1dj+/77mLxMYnU1aW+8EZnTfYBQc3KoeOstlIICsh96CFkXMihoGtYVK0h7883oNqVzZ2pvvBHH++8jbduWqMkDgu+CC5D37kXesOGgHcOkeZiGkcnvhk1LM5l7dxc2LXNRvd+Ov1ZGkjRunLqJ3qclz4u0e5ODLasyYwQg3PtsvPt026RzJffvtFFdHj83qarMyuaf4kP5KkosLP3Exe6tibOFKwrUVktxx9u50Z6wfEVJ/LFNTExMTH6DqCq2r74icOqphgiX1AgeD/bPP8d/0UXRXD+u228nfcYMrKtXI5WVIVZVEe7SheonnsC+eDF111wTZ3jZvvuOussvJ23OHKzLlkUTpgKItbU45s9vUnzBtmxZTDhgtP6+faRPn47z+edJf+YZ0v/5Txxvvw07dyLt2oX9k0+Qdu9O3nAwiGDwCqs5ObhnzozMLzJeE0XBunRpzDbvTTeB1Ur69Okpz+HXEKiXE7d9/fVBO4ZJ8/j9zjEyqozofkzichwpsfmKouXCse5OfXheXKieqBvx0P1uiYY8RjE5jgKxP3D6HEdqM3McxedJ0q8bchyJjTuDUuw1qNPnONLnNDLkIBJ1+Y+MOY5i9pGiXJJ8R/U7m0n8CNPC54uoKYs1PCr22PlgWjuu/+fmyAb95aq/D5Z/lkOtO97QqCy1UF0l4co23B8C5BQFyMgJ46uN/Ypl5oRo36sWFTUS/KfBjAfa88PnLtwlNjKyQ3Q/rpb7Zm6N5kl6fWprvvkwB2+NhCs3zEWjShhyeSSOO6dV/A87gDNTict9pOdAizTs3SXz9vRW1Hkkzry8gn5/qI1GA5i5kExMTEwOHvK6dUhud/Pm4+iwf/EFgt+Pb9gwLEQkvK3/+U/MO48AWFauJOMf/wBJovamm+La8V94IZYVK7Bs2pTwOFJpKbYvvqDuuuuS9iV47LEo6elItbUx2yWPh8wnnoivMGYMrSQJQVFQcnIInHYaVU891fguFwqRNW4ctu+/B78ftbCQ6oceiuRiAsK9eqG0a4e8c2dc0+GuXWM+q/n5eIcPxzlnTiTfUDPCAluK2qYNoW7dsC1ZgpKbi/3LL1Hat6f2ppvQUsh9mxx4zDcWk98NPk/icQBPRWrvSusOdSDEGxlWu8rWVekUr0+L25eWoZCWGSImYSwanfrWclTHRpW5ZR9n8+W8PNwlEYPNU2lhxSIXr02KhMJ9OiePd587il2b0nCX2Ni21smrj7Vl06rIMYffuY/W7WJV63JaBbns1n0pz+lA8t0nWdw1tBsfz2rFl+/m8tioTjx7f/IcUyYmJiYmBw7rypVARGCgJdi++QbV5SLYvz8Q8c6IVfHRE1JVFfaFC/GfeSZqQUHc/nCXLoiVldi++AKxJj7MXbXbUbKysH35JeK+xM+mUL9+IIqGJyZ4hw6lZPVq9m3YwN5t29i7cyeeP/8ZTRSjBpzkduP44APSdCF7WQ8+SNpbbyHv2IFcUoJ1xQpct9+OoOufZ+xYNCl2lDnUrVtEgMFA3TXXIIRC2D/6KGH/DwTB/v2xffcd2XffTdoHH5Dx9NPkXXwx0tatB+2YJvEcWR4jvacm1eQzYw4V3Yi0Pq+RIBrKqbpyBpMwRnvfeGz9Z/0LstGTpHdhG9vQeWeEcOO6GIxtQ+/FkUTDPlG/bvQmNR5P1Xl74jxGot6bhGFf4w+EatgXEHUeHt26YDAYhBiPUew+SfdZ0nmFJOEXigQY+qj3PMV4neoPm9UmACvjJ98d1bOOUIMHJeYejPy3c3068c4XDfc+K1Ou64UjQ6FdTy93vriB9OwwaFC8wUHZbjuxLigBVdH9nVD5+r0cAr7YH2ZNE1j/YzqKpvH1e7n4vbH7qyusvPdCAfc+v43cwgAPvbqJWZOKqCy1kJGlcNmtJfQ50UsqVDQ2rnKw9j8ZHH2Cl57H/bJJlpoG/366gIp9jfOmfLUy3813MXRUKUWdAylzHpneJBMTE5Nfh/Wnn1ByciL5fZqLpmH99lsCgwZFhAaAcOfOKAUFiIa5NA2y3r5LLknYlFBdTe3NNxPs35/cq6+OL6AouCZMQCorQ2ndGv+pp1L9z3/GvCdl/+1viDU1cbEMaps2qLm5Mdss69fH50tSFDKeeorABRegFBZiXbYsroxlxw7SZs/G+9e/AuC79FKkzZvJfOYZwkVFhPr1o/rhh9ESSKCHu3cn1LMnae+/HxWeEPfswf7JJ6hFRfjPOisi+92ApiH4/QnV/ZISCsXMsQKwbNtG5oQJVM6e3fx2TH4VR5ZhZGJyELngvl3s25BG2ZYGD49G2z5ezvlbbHyyZhA+3PJjJsYcSiAQDkYeJj6PzMYfsph5T2fumLkRgEVzW+OtivdEley0x7QvSnFFAHDvl3n4yu5sXxfvjQII+BoNinbd/Tz06pZIe81wAodD8NiNnVn/YzreGhlnRpiex3t56OVt0fC95lJdIePeH3+e1W4Lyz7L4rLRpS1qz8TExMSkZcjr1hHq27dFambivn3Ie/bgvfnmxo0OB3XDh5M+fTpSZSUAakYGSps2CH4//jPOaCxbn5vI+fLLqJmZSCUlyBs3xhkjAEIoFE1EK+3fj+P99wn16xcNrUt79VUcH38cX49Gb1hs5xM/50SPh1aDBxPs3Rtpf+J0HsaEuLX33ot9yRIEv5/KF15IWKcB34UXkjllCmJFBc4XXiBt3jyk0lJUiwWlWzcqXnkFtT5XU/6QIQQHDqR60qSUbeqRSkoSbk+UiNfk4GEO15r8bnC1CTJ63jpO/8se+pxTwXljixnz1s/Y0xVUBd6Z2IEJQ47lwT8czz+v7M3ujRGjRJSbZyzs2uCMOpwycxOr6FisWsyz6+xrSkgzJqoVNMr32lj9bRZ1NYnGLjSqK2TK91pYtsDFfZd2Z8xZvXj4qq5sW9v06NQ701uz4sssvPVtez0yK77K5M2nWjfnNGNIS1dwOOMfhBabQtsuiVX+TExMTEwOEJqGvGMH4Y4dW1TNsjEyiBfq1Stmu3f0aNyvvUbd0KH4LroI94wZCJpGqH9/sNuxLl5M3sUX0+rEE8kZPhw8HjxjxlA9cSLKUUclPJbRXBODQeyffQaA/eOPcT3wAEqSeTuaNV7F1XfRRWhpsYOGqtWKarMhBALYVqxANMxVAtBkGXnbNqTiYggGyRw/nrzzz0csLcWycSPymjUJ+9BAcNCgSJ/ff5+0119HKo0M/ImhEJa1a3Hde29jfzIzW6wwF+rcOeF21Zk496LJweEI8hg1vHz+gknh+lCdVKE5+nKGWDF9WJlgEF/QDBMRoxjD1HQjJXFu3nBjG1oKkQZRFxJnbF/Sh8FJsfuShc/FhcslEWkwllUMwgyK2HirBJoZLicbQhnlmFC35hkbkkGIoaH90lVW1v6rNb5KCUeOwglj3LTvpysYk3qqUaBAdmmcO67RQyQKENJE3nukA9/NaY1aH+ZYsdvO9Jt7cP/8nxhwSSnbVmUQ8Kb+ugiihiaAgsZZ1+/l67fzKd3piNnfc2A1Sv25Swgcc4qHi/8SKVu534LdqeKpklFCxvtYn4hLYPOqDMZe0ItwSMBTWe+xWQd7t9mZ9M4G8tqEknqPVi/NQFMNf3tNYM0P6XGCDU2JNFjtGv1O8VC620pY1+cOPfwMOLNpSXUzzM7k94DRC21icqAQ3W5EjwelffvoNsuPP5IxdSqi242am0vNXXcRPvbYmHpyg2HUrVtcm6H+/amqn3ckeDzI69dTO2YM0vbtuO65B1k3T8i+eDGBjz/Gd801VD/5JPJ55yFXVET3G1JINiIICJWVZI0bR7BfPzxjxpB9++0xynRKenrCRLW+4cPJ3LUL7f33EcvLUfPzEXy+qKGiPzaiGH0fE8JhHIsWYdm0iXDnztiWLInu04Ds0aMpW7IkUW8BCPbti2a1kvbmm1GPmh5527bolz3coweODz9s0Zffe+edpM+eHfMOqmRmNpms1+TActgNozYd/Nz6+B4K2gcIBURWL01n+kOFqGpznyIaHXv6ycwJs2GFk0DAfJn6reNeL7Pk9lzqShpv38otVi55qZi8bsEWt6dpsOnbrKhR1MD+bQ6+faM1Q27cR9kOBys/yaO2UiYtM0xNuRW/QXGufe/GeT1Ol8Kfn9zCvx/rgLvEitWu0mNADSMf2xF3/MvG7OXim0rYt93OD5+5ePMf7eLKiJKKqsTG3VWWxo+k7S+28+a0Nvx1crzSTgNyEq0J2dKyMLoGRv99F84MhZVfZxIOCRR19vO3x3e1RDXWxOQ3yaxZDt54w0llpUhenspf/lLLxRe33JPqdousWSPTvr1C+/aJVVJNfp+I9eFXSps2AMhr1pB9yy0xxou8aRPu2bMJ9+jRuK24GNXlQmsi55G8ZQuCqhLs25f0p5+OaRdADARInzkzMrcnN5fq8ePJHjcOzW5HS09HycvDsnYtYrDx2aympeEbOhTHBx8gud24X3+dUJ8+eMaOJe311xErK1GzsqgbNgz/0KHxnRIElClTqBg9GmnXLpR27cg/66z4YkA4Jwe5vDz2nHbtQtq1yzADOHJNpOJilHbxz2AA7HZC3bohJPBGAWgWS9QICnfujFhdjVBV1WxVOTU/H6VdO4TaWtScHNSMDOouvxzfVVc1q77JgeGwGkYWq8rDL++gQ49GVa3CjgEkWeOpe5PcmDqycsI8/NI2OvTwY3eq7NtpZd5zrfn0jfyD2W2Tw8zqZ7NijCKA2n0Wvn8mnwufSZ0rIRGaCkFfgrd4TaBidyRP0MV3FXPxbbuoKbeSlR9k+Ud5zH+2iIo9dhxOhfa9axk1eUtM9V4n1vD3+T9TWyljdShY7RpikpEjm0OlQ686gn6Bj2aEqTMo6MkWCDbzfchdmlpl79yryli/3BlzjLT0MGddWZ6iVnJEEW64fy833G/GQZuYNPDJJzaeeCKL6urIb8vu3fDQQ1m0axemX7/mJ6x89NFMPvrIzp49Mrm5CscdF+SFFypJEGFk8jukQUVOrX/5zpg2Lc54kffuJWPq1Jg5NGJZGUp+0+9KDTmClKKihIpzEJH5zrrrLsJduyKWl1N33XV4R46MGF5paWTefz/2xYuj3h3/Oefg+9OfyB02jFCPHoT69AHAe+ONeK+/HqG6OiKAIKd+RdUyMgjXhwKqOTlQXBy7X5Iihl95/LMt4ZM4FELavTu5YVR/HeTNmwm3bYu8a1fjsQSB4MCBjeVatQJAKisj3AK5baWgAESRirffbnYdkwPLYTWMTv1jFUVdYqWGZQv0PakWSVJR9HmCErxQjp26k6MHNKppFXUKctUdJfywyIW71BKjUAcgGHMGJaOZOY7i8hsJKfqbJHzOqC4nhBr3iUblOV3/RTm2fVkfIqerF5/vSB+OZ9gnJu+XolOiC+nC6gTREHqVIswuWY6jlPmOEijW1VUlvm3rKmX8WsQoEPV/Q6PNowutQ4vsdxUGqdwTmyzVkRXi+EtLo4p1FqtGZpsAGnDipWUcf1E5ezamkZUbIrdNZDRM0TD87TUc2ZHjKQ3Hi/Yj/hw696+l+wkefvrKFQ13c6SHyS8MUrwxsRBDLBrd+kU8V2qSfFADz6vk8m12Fr2di8ctk5Ed5rRLKjjl4sgDtq5W5Lv5LiQZBl1Qid0R6fSBzn3UFGaYnckvQVVVxo0bR05ODuPGjTts/XjtNWfUKGqgvFxi+vQMZsyID8NJxJdf2pgzx0FtbeTHuqJCYuFCO489lskjjzQdqmry2ydqGGVlxXyOK2cI/RIrKuLU3hLRkJhVKSwkcPrpkdxHBuW0wODBeO65B7m4mGCfPhElNp0aW82kSXiqqpCLiwm3b4+WlYVYVobthx+ouesuwwGlJr1YifDeeCPSww8jud3RbWp2NqE+fZLmVzIiAOF68YRkKIWF2L75BvfMmWROmoRUUoLmcBA84QSqH3ssWk6rnxdknzcP7513Nvs8VJcL2WDgmRxaDqth1LoomHBAwOZQsdpUfOEkkl2AJGu06xofktCqTYjzripj7tQ2CesVdvJx00O7yTsqhM8r8dm/81j4tulhOhKp/S9sftpBoExEdmq0Pi9En1F+nG0Vyn+ML5/VPnGy0+Zw0UM7ef22LpRvj0hs2zPD9L+gnLa9k8tYy1aN9n28JL9Lfxl3v7iR954pZP1/MpEtGkOuKKNj7zoeuao7pbvsKWpqtO/hY+hfEqvx6PnTX/cz9Kb91LhlMnPC0fC67+Zn8cr/FVGy04YgwL+fKmD0/xXT75TEoQMmJkcan3zyCYWFhfh8vsPaD78/8UBCbW3zBxhefz0tahQ1oGkCK1Y07S5atszK1KkZuN0iLpfKrbd6OPXUlocamxx5iKWluO6+G2nrVoT6+7whH0+4XTtsy5bF1TGKMwiBQMTL0gTR9jMyqLvySuyffYZ12TJEnw9Nlgl3707No4+i5uVFPC1S4iei5nIRcrkaz6FeNS7cs2czzrhpfMOGobRujXPGDESvF3nTJkJ9+lBz772R5LM7dqSsr9XPRTLO+44rl56O4PMRPPVUygcPRiwvR0tPj5Hltn/0EZkTJgCQ8dxzpH36Kdpzz0Hv3k2fiNUKQfN7ejg5rIbRl++7uOj6crLzYz0vpXut+LypXzc1LfKASESMp0lHhivMxJe2UtS50UvVvqsPQYTP327Vwt6bHEyCZbD1PpHg3sa/Zd1OEYctRK/bvbhXW/Bsabx9XV2DDLijIlFTzaKwdx23frCWH+bmU7XPygl/KqXzMYfHGJCtGleO3QNERuoaRBTGz93IA8N6Ul2e+KXInq7wwMubsTmalxdKtkBO68aRv6Bf4LUnCinZGTG+NA32bnMwc0Jbnlm43pwzZHLEU1FRwcqVKxk2bBjz588/rH3p0SPEjz/aYrYJgsbAgc1/6Un2nWtqLve6dTJ/+5uLEl3I8datMjNnujn++F8+gGRyBKCq5Fx7Ldaff47ZnDl5MpUvv4znvvuwrloV4yUJde+O55574tppDtHIGEEAWcb9wgtkPfAAtiVL0DIz8V59NZo98sywff01ms1G8OSTm243EHkPS6Q690sJnnxy9Nh5F10EioJaWIj7jTfIvvZaLJs3J417CHXvjnX9egg3EebaIOZQL6qgGsMRfT4yJk9Grp/7JWga8rZtqGPHwoIFyb/U9WiiGCcAZnJoOayG0d7tdha9k81Zl1eSlRP58u3ZbmXmIwm8PYbkr6oisH29nTYdYh8yJbssLJiTG5k4Ygi3ufSmkhijCCDDpXDOFWV8/lbjzd3s5K/Gp5PSWE4zhIAJgqL/oFs1tKEP9zN8gfSKdVIgeeJWWRfeZjwXfVhZXPJX3UdjmJ2+L4qujZAhlM6nD4MzqNJJYvNV6fa/FCa4N7aMUiew82MHra+C/rN87H1JxLNDJqNTmG6jfKjZFrwNh9Cft/H3P4linZihMfjmxjwCofgboXFddz/G6YTo9gX9EptXppPuUshpE8BbYaF1Bx+STExYnXHukRJzfSLHPaqTj0ffXcuUUd3ZtcmBPhZPEDT6/aGG/HaBSDgfIKV4e9KH2TUYXmt/dLJvR7xHqnSXjR3rHXQ6OvkIvBlmZ3Ik8Oqrr3L11Ven9BZ98cUXfPHFFwA8/vjj5CWRCW4psizHtPWPf8CGDSo//SQQCgnY7RoDB2o8+KAdmy2V57eRMWMEvv1Wo6pK99svaQwZIqXs93PPSZSUxP6Il5ZKzJiRy7nnhhP290jH7G8EYcEC5M2b47bb1qwhz26Hnj3RlixBmTwZYfNmtO7d0e6+mxydt6ahf1gs0T4m669UL4udl5sLooj4r38h2GwoO3cifPIJGe+/T+bll6NddBHKiBFQVAQ2W1w7cedRn0Q10+lE0x+3vBxh1Sq0Ll0gOxtKSqBTJ4yT6pq6vrIggChGyuTlwY8/ov3xj7BsGYLBI6O1bYt43XUwbhzZWVmR8kkwXo+48/r8c+QE3imhuJj8sjK0o49O2jaAZLMhNvT7MPJ7/r4ddlW6mY8UsvCtHM4bUYG7zML8WXl4a5oXnPSPO9vhcO6gYy8fdofK/t1W5v6zgGp34tMqaBtIuN2ZriBbVAZf6KZNez9fzc9j97YWZCs2OeCEk+gAKL76+T5Z0Puu5GFuRwJL383nw6fbUrLdgSBqiKKG1aaRWxjg/Jv2cMblLRc7KOwUYPInP7PozXwWzm5NKChital0P66WGyf+urjktHQVi00l4Iv9/slWFYfTVMIyObJZsWIFWVlZdOrUibVr1yYtd+aZZ3LmmWdGP5cnmJj9S8jLy4tr6+234b337KxYYWXw4CDnnuvH4wGPp3lt9uoFo0al8/bbaezbJ5KdrXLiiUFuvbUq0XzyKKWluZAgyLeiIkR5eQVbt0p8/nkuuble/vhHX3PeYw87ia7vkczB6m/azz/j8sdPI9C8Xtzbt6M25BIaO7ZxZzgcJ0CQ63CA201F/fZk/XXa7WQB7s2bUXNzSQsGETp1wlteDgMGwIABiPv3kz59Ot6qKpSsrGbd4EKbNhRYrQQWLqSmXrQgc+JE7PPnI+/di2q1RgwPiwWloADvtddSN3JktH6q6ytUVlLw00/U/vWvePRl5s7F9uWXpL32WtR4UfPzqbnnHmxLl5IJVDgcaCn+blm7duFwuSjXzWXSY1FVcq1WBOPfyGrFHQyiNHFP5FRUINrth/1e/y1+39q0STzFxshhN4wAdmxw8Nz4ohbX89bI3De8C62LAmS4FLZvcCQNowNYuSSTk8+rwmLw3NZUyTzz0VradvEjW+DCa8v4+qMcnns4khcgWY4jzSC+kDLHkc47K+hHGYwCDjrJaMEozKDzAIgpchzFeoUMAg568QWDtyd2X/IcRzHiDkLsw1fvQfLFCTPo15MLLEiChnxKCL72gyHqw9pGoFaJjLZKxpw7QhIvlHFQR02xzyjMkIwk3qPIPqitlHlnSnsq6gUdNEVAUcAXgt0bZd56oj09BtTQukOgGceK/Wi1w3nXlXLedaUJCus8e7p+Ncd71KVfLUVd/Gz9OTaZXNuufo7qkDr8x5j/KLZHR443SY/pWfptsXHjRpYvX86qVasIBoP4fD6efvppbrvttsPWJ1mGyy7zc9llvzzZ8R131HLTTV62bJFp00YhP7/p+7tr1xDffhtv7XTsGGbChEzeeceB2y0hSS6efz6dl19206GDOfjxv4D/nHNQnnkmLmeP6nKhFhQ0ux3V5YrmMkpFg7qaWFKCmptLYPBgsu6/H7GiguDAgYT69EFt3RrLhg3IO3bE5FNKheZ0EhwwANuXX8KDD2JbuBDH3LlI3oh4UFTe2+9H9HjImDqV4PHHE+7bt8m27V99haAo+I0S3oJA4IwzCJxxRlwd59y5KAUFcYljjYilpSitkydDD/XvT7hr17hQR61HD5S2bZvsu+DxoGVkNFnO5ODxm3gz2L/bxpY1aSmNIoCFb+ey5od0wrp33707rEiSSsee/ugEdFdumDOGVtC1rznh/HDhOFPGNkBC0D3brR2h/T3/Gw/vr14riBpFiagqtfHxjNTqN4caQYB7Z2zh6IE1ZLcOklMQ4Jg/1HDfjG2Hu2smJk0yYsQInn/+eZ599lluv/12evfufViNogOJ06lxzDGhZhlFAHff7aF37yCNIy4aPXsGueSSOubNixhFAIoisHGjhXHjsg5Ox00OOGphIXV/+hOKITTOf/75LcoirObnR4yrJuazKPUqbQ1KaUqHDlQ+/zxK+/ZYV64kc+JEskeOJFxURODUU1t0Lv6zz8ayYQPWb7/F+dxzUaMoEZLbTfrMmU03Gg7jfOEFlIICQv36NV2+HsuaNYS7dGmynOjxRK9J4gIilS+8QLhe8jvcqhX+U04h/MYbzeqHWFmJavjbmhxajggWBN+1AAAgAElEQVSP0aFCVQQeuLorZ19eTv/BHsr3WXnruQKefDt+1CTDpXDe8DI2r04/DD01ESSBnCcdSN/X4VmkYe0EOZeJ2DOP/InDyz/O5YsXmzZ6QkdgMuKCdkEmvbuR6goZQdBw5ZiTQE1M/tfIytJ4770KXnzRyZo1Fnr0CHHTTV4mTcqgqio+xG7nTrlhLrnJ/wCeBx7Ad/HFOGfPRrXZyHj55WZJb+sJd+qEWFMTzS2UjFDPnmiyjOWnn/Cfdx4AWmYmdVdcgVhaiqAoiGVlhJvpKdLjHTEC58yZ5IwcGR96lgAh1PTz3/nSS1h//hn38883KXTQgLRrF5b166keP77JshVvvolQlzqMX2nfnsCgQQi1tZQtXoyWnR2Z/9JUaJqmIe3Z02ID0+TAcsQZRm06BGjXzc+m/6bh3p8kUWVc6FKSX3NjSI0gooQFFryez4LXG38IQoHE9SvLExw/Jj+OIUytmTmONJ3qiVF8QQg31tOM+/ShdJIx1E0fBte4XTKExKmSXpjB2EbyfSQN1Yt9yOqFCIyXVdCFugnJwt4MZJ8EGSdF1r2A2IRXsFk0U5hBxeidSqJWowurU0ICH/yjHd7q1ElWna4Qp12zl1D9PWMUcIi5s4yXKuUliBdVgNiwOiOJwuyycsOG1mLbawnGMLtDHVqXDFPA4bfL0UcfzdFNTHI+HGzaJLNjh8Rxx4XIzW2e9+eXkpamcdttsVEPyTxONptmGkX/Y4T79KF68mQAnG+/HZNstFn1670j8ubNBFMlerXbCfXqhXXVqtjtkhSdz6QUtXwqBAAOB/4zz8T5yitNPhUUp5O6K65IWcby449kTJmC/6yz8F94YbO7Yf/8c4D40LtECEI0R1HKvmzZQrh7d7QWJHcVKisR6+p++fU0OSAcMU9/SdYY/9J2pn64mYmv7uCZTzbxt0m7ST354sDw03cZGAci9hXb+ODl5HGkhwtRVLFam5813eTQsme9k4rdiULoNIR6Vb7sowKcevU+Oh2TPGzAxMTkt4PPJzBiRA5Dh+Zxww25nHtuHn//+6GfR3D99V7aG/K9SZLGoEGJhYmONDQtci1NNeNYwp06JVSqS0WoPn+QZfXqJssGBw7Eunw5Qu2Bn14g7d6d0CjSBAGtPtGl0ro1viuuIHD66YkbCQTImDSJvGHDUHNzqfr735vv/lQU0mbNIti7N0qnTr/sJIyEQshr10avcXOxbN0KQLhDhwPTD5NfxBHjMbrunn2ceFZNRMYYyDsqzFmXuVm9NJ0lHx3ceMsZjxRhsWkcM8iD3aFSXmLl5ceLqKlMPep/KBEEjZtuWUn/40qw28OUVTh5+dVjWbvezL90JOHICmO1q4T8BmU3u8q1j29GEqHXH6rJzEseErDtv07mP1uEr1ai/dFeLrt9N3bnwR1dbgmaBj9+kcnXH+SQWxBi2M37ceWZxrqJSTLGj8/k668bB0z27pWZPdvJaaf5OfnkQxce7HJpPPVUFY8+mkVJiYzVqjJoUIBHHqk5ZH34pcyYIfD883lUVkYS1l51VR3XX39kK5MeKkJHH41j/nxaEg+ptmpFuGNHbN9/j/fmm1OW9Z9/PukzZ2JbtAj/H/94ILocRZ8YVU+oRw+qpk6NeLQGDUoqLCGvXUv2mDFY1q8n1LUrSl4emY89hufOO1G6dm3y+PYFC7Bs3Yr7ued+1XnosaxZg+jzERwwIHVBTcO2eDGO995DadUq6l0KNScRrMlB44gxjHoP8EaNogbsaRpnDKs8cIaRPnRGFyqjqgL/eqA9kqxhtam65LL15dXGsjEKdcYEac3McZS0juGzPnTuhtFruODCbVgskaGyVq18jB2zlDvuOJuaGlusSp0+bM8Qcifr39cN4ULJwvGApPmPjAN3il6xzhBmFxAbDU39aRvV5GIV61KEgBni4JrKjZQQw3mqql5xL9ULS2JDwNUuyFE9vGz5PvaeVUMCnz5fxOUP7CA9LxTteVmxjY+mtaXWbaHjMbUUdvMy5+HOVJdGVCfWfJPNhv9kMv7tNcjW+vNrYf6jRKQKs0umYKeiomnw5OhO/LDQFZX1/u5jF3c9vZ2eJzTvJSWZgt2REmIHzVe202OG35kkY/Xq+EE2j0dk7lwnJ59cdUj7csIJIT78sBybLY/a2nIsR874X1K+/NLGww/LuN2R34i9e2HyZJnCQoWzzvrf8HYdTEJ9+uCcOxdp506UFngbAieeiOOTT5pMaho8/niU1q1xfPDBATeMam+5BdvSpUiG+TeW9evJvfJKam++GXnXLtSqKjSLBWQZ+wcfIL/zDgX79yPU1qLm5hJq2xbL5s1Y6j1n1uXLcb/wAuH+/ZMfPBwmY9o0wh074r/gggN2TrZly4CIpy0pmoZr9GjsCxci1uddU51O1IyMFqkLmhx4jpgnuaoleRk7hCJkSljQGUVHFv0HlEWNogbatPEydOiGw9Qjk2SMemEjvYe4ke2NDxtVEdm7IZ25D3Smrj5P1/ZVTqZc2Yfv3i7gv4tzeX9qO168q1vUKGpg+8/pfPXmkeEZXPt9OssXZ8XkOtpfbOfVSWZMtIlJMpLNAZcO4+MmI4P/CaMI4JVX0qJGUQPV1SKzZjU91+P3QINnwrZ0aYvqBc44A7G6GmtT9USRussuw75wIVKC5KWJkLZuRWxGHpxw377UjB9PsHdvVN0XQgCkqioyH3+cvKFDaTVkCK0HD6b1oEFkPfEE4pYtiB4PmiiiZGdjMcyxkvfuJfPJJ1MeO/2FF7CsX0/Nvfce0C+jfcGCyPm0Sv7ctvzwA7bFi6NGEYDo9aJJkqmCcpg5Ygyj5V9mEDSIktRWi8yf3TKlld8qNmtiCzEn55fnxzA5ODhdYW58fkPCl6GyYjtL3478WL77ZAfDfCSBQALDXA2LbPwh8yD1tmV882EOvtp4R3PZXqsZ929ikoSBA4NIUuwXJCdHYdQoMyVEc/D7E7+q+P3mCyRAuFs3lIICbEuWtKie//TTUZ1OHB991GRZ78iRIEnNkswWvF7y/vhHXHfc0aQcOIDv0kupmjw5Jk9kA5rVStXEibiff57KqVMJG4wNUVGiyVqNiBUVSY9pWbWKjClT8J1/fkqhhrRZs7CsXNnkOTQg7dmDdeXKJsUfHB9+iJRozpamxYp8mRxyjphQujefaUVB2yDHDvaQla1QUSrz5bvZrPy6GS+E+i/TL7S0NVWnmGZUZNOjK4chTI1fkPxVM4TYCUI4Ybl9e9IobBc7Wb+2VubzjzsghJTYkCpdPSleGq7x2MYEr2Li9chnXT19k8brrbMGFGM0oV6xLqaJ2H7oP/+i8Lhfwy8YKlAM6nUKAqU7bYQSPswFvDUiIQ2qy6wJ98dtETU69avVBcbprkkzE8NKhnbVJOp1kFzBThIEijr7EQQNzeDhdTgVNCFxkFxz1exSJYmNb/PIeyH6JeF3R8zIlMlB5YEHaigtFfnxRytVVSJHHaVw1VV1HHOMOTevORx3XJDvvrMS+/uo0b9/6sTTvxsEgcApp2BfuDASFic389XO4cB/zjk4Pv6YcIocQgBq69b4hg3D8e9/Uzt6dMpcPprTiWfsWFwPPkjGE0/gGTeuya7YFyxIaBAIoRCh444jdOyxSDt2IATj/+ZiEhlvNYkinLRzJzmjRqG0bk3V5MlJ3xszH3wQZJm6q6+O3ZFiLpfj3XcB8DVhGIU7d0YTxbgpGWpOTrNlxk0ODkeMYaRpAtPubkuGK0x+mxB7tltjwnV+7zw7pTcPT1lOUXsPsgw1NRa+XVLE+nV5h7trJgnYuSoDTU3w4yZolO908Oz1vagpS/z1s9jDhPyN+9r18nL6VSUHq6st4uyryvl0bj67NjVmB7faFU48t/Iw9srE5MjGYoHp06uoqBApKRHp1ClMkjnnJgkYM8bDf//r5IcfNHw+EYdDpV+/EHfc4TncXTti8A8ZQtq8eViXLyd44onNrld37bWkvfsu4pw5cOmlKct67rwT+wcfkDlxIpUzZqRu9/rrsaxfT8Yzz6C6XE0KPMjFxYmHu2QZafdu0p96ChQFzRo/oKhmZETC6eqT0AKEW7emNkGCZ2nPHnIvvxzB76fi7beTymlnPvII9kWLKF22jIwpU9BkGcHvx3PffRGjKJFxFArhfPVVAqecgtKxY8rzrRs+HOfs2Vg2bYpu0wThgM51MvllHDGGUQOeKhlP1a/oVnO9RwlyHDXuMngwxMSiDZphZF1oIseRxaIy+IIyCooCfPlxK/YVp2FE0yW00ec0Kt1lZ8zIUzj9nN20KfLy1Zft2bEtiwYRAL1Qg6DzZIlxuYp03iSjV0hILDIRaTRxG8Zy+jY1w2R0fa4efToiYzBgc3McpRJtaC7J5rYBKMbJ9C0YxMnv4cOWHiZgCDsTRY3v39FNrBS0GBdcQRcv599WzIJn2hL0S3Q9oYZrJmxDsqkoDaen+1vH5wjSod/VzNxHRowiDbJd4f5XNjHzwQ7s32XD7lA44Zwqht++L0XrzWu/JaTyLh2J3iQTE4DcXPWg5y86Elm1SmbRIgdHHx3k7LMDLZ7OYbfDZ5+FmTfPw9KlVk48MciQIQFzKoaOwOmno1mt2D/7rEWGUfD44wkeeyyWp5+GSy5J6a1QioqovfVWMqdMoW7JEgKDBydvWBConjQJsaaGrEcfRVBVam+5Jel7WbhbNzTiH1Wq04nrzjsR65OqqpKEJgjRsDsNCPbvT9X/s3fe4VFUbR++Z2Z7yaZCSOi9CFKk2UXBAr6WT8UCiL4WFFFRwF4odhQBESsKIiqWVxQVEBsqilhROiEBAgTSs9m+M/P9sSGZLVmSEBBxb6692J1yzpnJlPOc5zm/55lnSB43Dl1eHsHmzal44IEo4QVxzx7SLrsMsaKC4sWLCcaS064yeJx33olhzRqa9u6Ne/hwfGeeSdKjj5L0wANUTJ0a8zhMn36KVFBA2RNP1H5eDmA2UzJ/Po777kPavh1dXh7+/v3r5F1LcHgR1Mje/UEYLF56uNrS+NTnqRlHUSostE67XYTREZasNeLhktXGzwNzNtK8jRudPpQ89quPm/DKMx3DK9O4wIVId7i+5reqj1hnCM2iPWNQHsMuziHJ5qOs3MQ7Hx3Hjz+3rN5MNuk038PbGDSLmu9CxLqa3wFLzfdghG0nm2sup2DEiKhi0QSDmWuMPq1IAYDZUhNoZzWGu82TDDXrHEZP2DqHvsbEStJ5NN/DTS+7VPPbIvpqXWeKUKWzarbVrovcTi+Ejm3eqI5s+9bBgUe9ICmoskCtVoqgcupVe9j4bSqFO0yAQGqWlyHX7qZ9n0oy23mwJQdrVY2DcMNIGz4XqV5HLdtFl1fHzOEN7KE01DCKX+Y/p7ekb5bzdzfhX8+ePXsapZz09HSK6jDZ/GjhSLRXUeCmm5JZtcpERYWI0ajQtWuQN98sxuGo30BWXdu7davEAw84yM/XYTYrnHWWj0mTnEfciPo7rofUkSPRb9zIvjVr6iUmYProI1JvuonSOXPwXHhh/I29XpqceSb4/RQtXx4K/YqH30/KuHGYly7FM2QI5U8/HXMfobKS9PPPD/OgKA4HKiCVl4dte8CAUgnNQdr/2WekTpyIbv16RJ8PxeHAe/rpOO+4A9HpJHDccZiWLSP57rshEKD4rbcI9OkT3YaKCuyzZuE57zwCvXsjFhZiWbyYyrFjq4895bbbKJs+HdUekYdMlskYPBiCQQq//jqqDxjverDOm4fjgQfY/8UXBDt3jn8+jxDH4vMsKyurTmUddR6jY5Wxk7fRumONnHFKeoCzLtzHyo+zyNtqa5Q6+pywl+tv+p3k5JAxkZ1dyW2ZP1JcYmHr9kTI3ZFm5EtbWf5Yc3atsyFKKl6nxL4tcVSUVIFVC7PQGk4le0wsfqwNqgCpTf30HFzC1VNyEyOlCRIkOOp5/30Ty5ebCARCnUSfT+S33ww8+GASM2eWH2Tv+uNyCVx3XSrbttXI7eXk6FAUuOeeYz/szn3ZZaSOGYNx1arak6HGwDtsGMrcudiffBLPeedBjHC1akwmSufOJf3CC0keO5aShQvjG2EGA6UvvID/1VdJeuQRDIMHUzprFv6TTgrbTLXZKF60iKQpU9Dt2IFiteLv1w/7s89GFSlo/hf8fjKGDkXy1gxqiuXlmD/6COOKFQiyDEYjotOJv2dPSmfORG7fPmZTjatXY/7gA4SyMtySROD440NergOH8uuvoSS3MV7A5nffRb95MyUvvljvOUKWt9/G3737UWMU/dtJGEb1RRuCFzGHpPYcRyrNmkerxyWlBDn74t28+JjGa6QVbahjjiMIxaZecOHmaqPoAGmpXi6/4E8eeeLU6DIjtSPihMiFr4sXrqgVaYgYEdRsqw1TCwrhD1WvUPtDuSHhcpEodfQoyHFEOLTr5IjyTFR5kIxw7uT86uWfT8+ObxgBsbxJihw6V8W7TXzzZlOSMvwMGxcqN+p1FPb3qL9IQ6hMbahe4+ZCiuRIhtn9kzxJCRIcCyxbZq42irRs2XJ4dMIXLrSwbVt4t8bnE1m50vSvMIy8Q4Ygp6ZiefPNehlGiCLy1KnoL7gA68KFIQW6OAR69KB82jSSJ07E/uSToXk38RAEXNddh2/AAFJuvpn0yy7DfeGFVN56K8FOnao3U5o1o0yTaFXcvTuUn6mwMG7xWqOoukpVRToghe33o1gslE2ZUqtRBOA79VQ8F19MoFs3LG+8QWV6erXIhHHFCuzPPkvFvfei2sIHswW3m6SnnsLfq1e95wjp/voL/fr1lD3ySL32S3D4SEhf1ANRUukxsIKeJ1Wg09cvTjzgj90pKysyxlzeEEym2JLeFvORy6yeoHZOvqGAJu0PLVN7MCCxZEZLnr6qG15X4vZNkCDB4WPPHpFly4zs2NEwISSrNfYghcFw6ANcscjNlYg1uORyCf+OdAJGI+4rrsC0fDlSTv3CdNWzz8Z3yinYn3gCaffug27vvvJKXCNGYH/uOawvvFCnOoLHHUfRsmU4b7kF04oVNBk0iJRrrsHw008xZb2V7Gx8AwaEcvscIqLbTfrw4aSOGgUxDCkAVRSRduxAlST8fftimz0b6yuvIO7ejVhaSvm0afhPPjlqP/vjjyMVFFD+0EP1Vka2vfgiisVy8BDGBEeMRM+qFpo29zHu0Z3cNTuX3qeU06F7JXM+28C0BVuZOn8rcz7bSI+BFXUsTWDdGgfBCPtk9w4TS9+uW8xjXdi5I7a0+eatiVxQRwPmJJnr39hIz/OLyO7uDAkvNAAlKLLh2xTevL9dI7cwmsoyiYI8I/IRTLScIEGCI0tOjsTEiQ7GjUvmhx/0qCrccYeDYcMy+O9/0zj//HRuvDG53s+Bm2+upEmT8J3MZoWhQw9P/r2LL/Zgt0cPWjZvLv9rwo9d11+PajBgnz27fjsKQki6WlFwTJpUp/xD5Y8+imfYMBxTp2J5/fU6VaNaLDjvuYd9a9bgvOMOjD/9RPpFF9H0hBNw3HtvKBeTRpK77LnnqBw3Dl/fvgSysw8pgYfo8WD84guSHn4YAMMPP2CbMQPjN9+EEteaTHiGDSPYuTO+M87A8MsvWBYvRklNxTN8OIEIMQcAw08/YZ03j8prriHQt2/sij0e2LoVwRM+L1rauRPzkiW4R4xATU4+hCNL0JhIDz9cdYXUkTcmv3uYmnIEiPtkVKs/J55bxv0v5NLr5EradPEy8OwyTr+glMwWASRdKJw2OS1Ih+4uPluUXhVdVzUdUK35quWX71JJz/Rjscr4vAI7tlqY81Bb9u6whB5ABz4INd8FIWydIBC+naJWfwRBYOO6FI7vU0RSshdJgkBAYOPGdGbPPAHZD4KsIAgCgqwiVEmcHfh+4LcYVBGDofLEINUfNL9VzXcEwrcTBUQZRBkEBMRgzSe0bdV3zTpVECAoVn8UBJSgiBIUUQUIBqXqj4KAP6jDH9QhiOBXpOoPCPhkHT5ZBwL4FB0+RYeKiFfRV39EQcWv6PArOlRBwKfqqz8I4Fd1+FUdIhBQddUfgIAqEVAlRCCoSgRVCQQIIlV/oOa3iIqCWP0x2hW6nlvGCVcU0bSNm01fJaPEE2SIg6wInD6qQHPlhl/JgiDEXB75ESJ+I4DfJ/DsLe1Z9GRLPpvflO+WpJOUGqRFx5oHu6r5J0S0v7a6VOILQYSXUfu/yPrqVl7dPw0pvyFI9tuOSD0JasfpbJwQK4vFgtt9aB7hI8mB9i5ebObWW1P48UcjGzfq+ewzM999Z+CLL0xUVISeZx6PyPbtoWfgwIF1zx2Unq6QmSmza5eEKKo0ayYzfLiHW2+trLehUpfzm5WlsHWrjp07xeoQvtatAzz5ZBmZmUdWDfDvuh5UqxWxuBjLW2/hueCCWuWoI7FYLLj0epSkJGzz5iGnpxPo2TP+TqKI9+yz0W/YEEr8Ksv4Bw6sm9fEbMZ/4om4rr6aYLt2CG43pqVLsb7zDrbnn8f84YcYV63CsH49SkoKwY4dCfTrh1hYiFgVWteQp7QAoKoE27XD/tRTBHv0QNqxA/PHHxPo1g1BELA//TTG777D36cP3kGDkDt0iDnvSnA6SRs5EtVmo/SVV2JuY5s+Hcf996OfMwfz4sWIBQX4q9T8kh55BP2GDZTOnRst5vA38099nsXDXsdzfGyr0kVSpyexynOfbaFDj3DLXlGi59MF/HDvVR34c02VpyaeYl2VK1jSKRhM4HFJYcur0f4OU6VTad3JgywL5OdZQRcRo12lUqfTy5w1bDdduhXzx28ZfPltWxTNXCjVWLOfbAyvWzbV/JbN4eu0inUBjUJdpPJcUKNYJ5si19Vcalr1OtkS8cLShATqI8IATaaa33ZThKKcRrEuyVAzImnXhW+XrK+5eWwR62waVTq7GD6qqVWws4o1nQOTGN5RqItiHYABhWAAnj23O4U50dLtNcQSMYXszpVMWfF72DLtX007zyeeazhSlU4UBOZOaMtXbzcJqzcty8fjH/1FamZ0aGZ95gM1VMGuofU1rPwjYxglVOn+fv7NqnT79hUxZEgGmzaFv08MBgW/P/oe69/fxwcfFNe7LlUFp1PAalXrLdV9YP/i4nSKisro1CkYPxOHCqtX63nvPQvZ2TLXXeciOfnIx9H9ndeDuG8fTU45Bd+pp4Y67HWgur2KQurIkRhXr6bovfdiqrdFEQjguPturG+/jfessyidPRs1KXYESzwEjwfDt99iWLsW3fbtoU+MpK6qIIAohoQV4qCV9dbi79GDyttvR//77zjvugs8Hizvv49x1SrKHn8c+/PPE2jfHs/ll9deuKKQcv31mD7/nOLFi2NKpJuWLsVx551IlZU1u1kslE+dSqB7dzLOOQfXNddQMWXKQc7Mkeef+DxLqNI1hDrkOEpKkUltEqPjF6MfpigCwUAdO1BVYgyyHzxBkQPj+FG5kMLEHUI3fdvOTm6fvInMFh5UVWB3noXpD/Rgb75mIn+VIRaU4bMlrfhsSavQcr2KgObhoTmQKG0BTf4jVQpfGZafSHsyogQiaj/Hai3CD3XNdwThOY/i+v/i5CfSItejg63dVtEKD0S0RCssEZkLySTWXFsKMuhg5BvbWHpPS4q2m5CDAp5yCb87dGvqzTItj3eye70Vr1PbeVFp3ctJIOKZrz13WkGEyL6I1jjxe+GPr1OQJJUep5VhNKhs+cVO5Bku3mPko5czGfnATqKpu4hCpFBDrDYdjMMh2hBefv07UglxhwT/NPLzJQoKou+XWEYR1FtsqxpBgKSkhhknubkSt9ySQl6eHllOp1WrIE89VUaPHsGY2wsCnHRSgJNOanzVu38KStOmVN58M0lPPYVr7Vr8tYV4xUIUKZ09m4yhQ0m94QYKP/0UpWnT+Pvo9ZRPn06ge3ccDz1E+rBhlL70Ur1V1lSzGd+QIfiGDKlZKMuhEDRZDn2XZRSHg3S3G2XMGHR5eah6PWJFBWJpKYIsowoCgS5dcI0cSdIjj4QZJqoo4u/XDyU1FcHvRygrQ01Oxj18OGJJCaaVK3GOH49qjS+UZJs5E/OyZZQ//DD+AQOQcnIw/Por/t69kduFwtwtb78dVjeE5jqZ//c/LO+9h+Jw4LzjjnqdowSHn3+XYVQH3JUinhiT2hUZxIjeZX6OiU2/NY7Udm2Iosqdj2ykTUdX9bLOPSqYMO1P7hzdn4Y5kxMcTTgyA/z39c3VdntlkY5VLzXDWain1wVFdD+jlG/nZ/Lt/ExK9xixOIK07lnJZQ9vP+S6//gqmYUPt2FfnhlBUMls4+GGJ7ejxO5z8MsXKQyfkI/B/O9LUpkgwbFGaqqC3a5SVha+3GgM3d8+nxi27OyzwyMpDjeqCrfcksLvvx8IURL46y8Dd9yRwvLlhQ3yPv1bcN1wA7otW1Aa4LlRU1MpefVV0v/zH1JHjaJ48WJUhyP+ToKAe/Rogp07k3LjjWSccw6VN9xA5e23o1riRUQcBEkKU4GrNq+zsih5++3QoLMgQCCA5a23MPz0E4Fu3XCPHo1qsSA6nVgXLULcswc1JQVfv35U3H8/KArmt9/GumABlTfcACYTwQ4dQuILBzGKzG+/TdL06bj/7/9wXXMNKTfcgOGHH5BKSpBTU/GdeCJlzz9PbZPyDL/8gujxUPbII4m5RUchCcMogmBA5JdvkmiSXYRBEwqWt8WEzy2Q1caPIELBDgPPTGxdZ89EQ+naq4zsVtFxk1ktXWS1dLNn58HknxP8UzjgMLFnBDn//hqvjACcOrqAAZfvp2CLmfSmPpIz6x7nD6EORsAnojcq1fUE/QKLprShYHvopaUisGebldfub0tWew97c6NfZnu2mXlmTAfunr+5QceYIEGCowe7XWXAAB979kjIcs27rEePAP36+Vm61ERhoURamsIZZ/i47rojO+cgJ0ciNze6m5KTI46Na9UAACAASURBVPHzzwb696/fc/DfhGqxhDrnDSTYtSulL79M6jXXkDZiBMVvvRUlUx0L/4ABFH75JUnTpmGfMwfz//5HxUMPhWSsD4cCxgE3psGA++qrcV99ddhq19ixuK+5BssbbxBs0QL/6aeDPhR9UXHffTgeeADbCy8Q7NIF07JlBFu3jl1PMAiyjPnTT0meMAHvaadR9tRTWF94AdPy5QjB0GiiVFKCedkyAi+9hPf00zGuXl29rrrJHg+qToe/LmGKCY44CcMoBnMfzMblFDnhdCc6g0r+NhPP3deC8hI9Wa29iCLkbzciRMYVNCTHUWS4mea7AIjIiGJ0CIIgqIiCjFpVlhCUNeu0uYQi5jrFyYUkijVlRIU1hYXB1bQ/MgxODaubiHWxy4uyLQVtyFo42sdLQ8culbqG2UUkc1KkusWRKGFzzSJX1rJdBHpt+GNVcwUTNOvhQY+Cv6ptUoSynaIJU9NX7ffFa8347q2muMr12FMDnDFqL2dcsY9NP9spyIuYCAYU7DAy4uEcinYb2bHBSrhXUmDrbzb25hlo2sqnXRyHhuVCike8sLt4YXa11d0YNCT8LkGCv5vp08tJSVFZvdpAMCjQuXOARx8tx+FQufXWSnbskGjeXMbhOPLXt6LEltlWVSGhlHkE8J1xBqVz55Jy442kjh5Nyeuv18k4UtLSKJsxA/eVV+K4915Sb7wxNK9n7Fi855wTMX/68JM0ZQri/v1IBQXY5s2j5OWXUVNSUFNTqZg8GcP336P/80+CbdtSeVuEII7HQ/LEiRh++w2hogKxpIRAr16UvvoqGI0YV62KMnyEYBDj119TsmgRhl9/xbx8OUIgELWNfc4cSusodZ7gyJEwjGKgqgLzn8xi/pNVCzQd2D0xOpKHk/W/Oti9w0KriPw3BfkWduUmvEUJ4vPrZ6l89HRL3BWhEbLSvUbef7w1TVt5MJgUJJ0aFTan06mkNA0w+pHtTL3kuCrVvBqcpXqK9xjDDaMECRL8I9Hp4KGHYqeesNlUunWrJa72CNChQ5BWrYL8+We42lebNkH69Ut4iw4FweUKzRc2m+Nu5z33XMpmzSL51ltJv+giiufPR6njJHZ/374UfvYZlsWLsc2ZQ+qNNxJs0QLX6NF4LrgApVmzxjiUuIiFhQhuN6Xz5gFgf+wxku+9l9IZM8BkQklNxXvBBdQmIJ98xx2YP/oofOzP5UI9YNzVFs8pSSBJlL34Irpzz8Wwbl3UJkL5v3ce3NFMIo/RYSY5PUDrTu56J4Q9gBwUeX5qe3ZstRDwC/h8Its3WXn24a4k5hclOBirFmVWG0UHqCzVs+KVbNr2rKRZu2i/W7N2Hpq19dKyq4uMFtHGT5MWXlof54paniBBggQABQUimzbpDtmrIwjw9NNldOnix2hU0etVOnYMebSOsNPh2CEQwPDddyTfcQcp48Zh/OKLg+7iufBCSubPR9q5k4xhw9D/+Wfd69PpcF95JftXraLklVeQs7JwTJ1K5gknkP6f/2B94QWk7dvrlDepISgZGQiVlVgWLADAec89KMnJpIwfD4D1tdeQdu6MWb/gcmH49deonpY+NxfTihUAeP7v/1Ai5lApVivuSy6p/h3o0iVm22pbnuDv5d/7aIm8CeLqf2pD3zTS10qEopxG5k1vCHLXrFw696zEbFMoKjDw/ouZrHg3IzysLt6bo6pNf65xcMtFvejSu5JgUGDzOgeqKEFVuFVSip/ufZ3s3WUmd2tS2LFEHpU21C0yn4w2+i8q10yYilzNCIkuMsxPqOW75niit6u9rkjbXWteRo5j1hZaV9fQuVD5NdvG208bZhcZYifHkW5SRG2YYLT6YXUZmnbIhF8j2jbqI8LG5LBwLgW/N3ZbfF4RWYTrZm/i9Ts7UrjDBCJktvEwZvZmAqqKZJY5/YoCPnmhOZWlIePKmhzgtMv2Y7QHw1sVcRlorx85bohZw9Tl4oXd1VXd7u8KuUuQ4FilslJgzJgU1q/X43YLZGfLTJhQwXnnNdy73K1bkBUrisjJyaCoqJy+ff1RRtHu3RK//66na9cAbdokYuxqJRjE/MknGNaswX3ppShNm2J75plQ7qGD4DvjDIo+/JDUq68m7aKLKJsxA+/559e9bknCe+65eM89F922bZg++QTTp5/imDoVx9SpyGlpBHr2xN+rF8EuXZCzspAzM1HS06PlEEP670i5uUj796PbuBH9hg34TjstNJcpAteYMRi/+ALjl1/iGzSI8sceI3n8eKS8PDznnIOSnR2zyeKePUj79kUtFwIBxL17AfBccgm6LVswf/IJYkkJSloanqFD8V58cfX2lePHY1myBLzeUN5AQSDQvTuVVcZZgqOLf69hdJi5afJOTjy7rPp+ttq9jJqwmz9+TGLf7vqH48lBkb9+iU7UNuqWHAadX0CTZj4qKyRyNicxdWJfPO7EnzYBtOxeydY1Eao3gkqHfqHQmWbtvdy3ZB1F+UZESSW1mT9M2nvYzXvodnIZK17LAgWGXFNAu57h8qMJEiRIAHDnncl89VXN+23zZpHJkx0MGFBEamrDlSxFEQYOVCkqishno8L48Q6+/vqASIRM//5+5s4tPaIeJVUFvx88HgGfT6ge6ysrEzAYwGxWD4vuQH3R//knuk2b8A4ahO+ss6qXC5V1e6YHu3ShaOlSUq+9ltQxY3CvWEH51Kn1VlYLtm9P5W23UXnbbUg7dmD8+msMv/+O/rffsH/5ZVjuIVWvR7VaUSUpdCGIIkJ5OaLXi1ZEXElOJti2bcz6/D17IuXnY1y1CrGoiMBxx6Hbvh3VZgsZXjEwfvMNjgkTQqILke3PzAwzCp333kvl+PGIu3ejZGejasMTZRnHAw+A34/7kksQfT78xx+Pa/Tog4YxJvh7SPSeDxNderuiBjnSMwP833V7eX5ym0apo2P3coYO343dEbpxbUkyx/ct5aZJf/HMwwfJWJ3gX8H5E3aS94ednetsBHwSBrNM294VnHNTfth26c3DR3SDAYG/VjlQgxI9TivjxhnbEn6TBAkS1Iosw19/6aOW5+freOMNC7fd1vgDKgsXWliyxFydd6m4WGL5chOzZtm4447GqU9VYd8+kZwcHTk5OnJzdRQUSOzfL7J/v0RRkYjLJYSp+tUQmkOj16skJSk4HCrp6TLZ2TWfNm2CdOkSJD398KdAMC9ZgpKejm/wYAAckyahNG2K0qRJzcEexIJTmjSh6H//wzZ7NvaZMzGuXk3ZU0/hGzSoQW2SW7UKU5MTnE5027cjFRQg7t2LtGcPossVkuWWZVBVVJsNU6tWOC0WlCZNCHToEJr3VFvbDQY8w4YRbNcO29y5GFevxnnrrTGNIqGigqQpU7C+9RaB9u0pf/JJbHPmIOXlIQByRgauUaOicjupZjNy+/YRJ0sh5cYbMX3+OcEnnqB8xIgGnaMER5aEYXSAOiR/DW0XO6wutKqmjNqiqfSGg8TRakdK5AOhcgFEUaW8LHx0Ydhl+dVGkZa27StqRjkiVekCNdtHtkSrUhfVfE0xUm0hcRF71lWVLpp46+qvWBc/kCsinDBO+JyiDZ/TxQmzCwuDiwiz06qwaRPBRmTbNcUpQ9EEsUWG2ek1ioGogAnGvrOeTV8mk/NzEh0HltPltDIQ1OrksPqIQ8n93cq8OztSkGtGVQSatPJw+QO59DqrNGy78JC1iLMc56Q3RphdWHl1TCAbSSLkLkGCxkNVa03bgs/XcHfJ/v0ier1KrIH9ZctMUcloZVng+++NDTKMVDWUUPa33wz89ZeeP//Us2GDnvLymjpMJoWsLIUmTWS6dQuQkSFjtapYLCpms4rBoKKqYDbbqKhw4fcLVFQIlJeLlJeLFBaK/PKLgY8/lghqMpinp8t06BCke/cAffr46d3bT1ZW4xlLQkUFYmEhFXffDYD1hRcQS0spnTWrZiOvF7GyErG0lGDHjrUXptdTeccd+M46i+Tbbydt5EjcF12E8667kFu0OKR2qnY7geOPJ3D88XG3M6Sn4ykqqnvBBgOB44+ndNasUBLYSG+NomD6+GMcVQp2zrFjQ8lXTSY8//kP5rfeQiotxX3FFXU6RmnnTtIuughdQQEqIM2di1lV8YwcWfc2J/hbSBhGh4ncTWZadQzXOSkr1rHk9YNkkNbgSPVz14yttGjrQRBVCnaZeeaejuzdGbqhg/7YLxvl8MxhTPAP4Kd3Mli7OAO/WyI128uFD+WR3sJH98GldB9cetD9VRUW3NuePVtrFA/35Vp4Z1objjulDL0xcXElSJAgGp0O2rcPsmtXeLeiadMgI0bUX6xl0yaJSZOS2blThySpHHecyIwZAqmpNc+g2gTB4kzzjKKwUOSrr4x8/33os3dvqFCTSaVLlwDnn++hS5cAbdsGaddOplkzuU7lp6dbKCqq/bhlOeSJ2rZNx6ZNerZsCf0/f76Vl14KSWJnZsqceqqPQYO8nHqq75Ak09WkJOSsLFLHjMF7+ulY3nuPkvnza8K5tm7F/tJLiKWl6LZswXPxxbgP4uEI9OhB4aefYp85E9tLL2H+5BNco0ZRedttKKmpDW7rYcVgCB+OU1WMK1eS9NRT6NevR7FYkFu1Qr9xI9KOHcidOqHabLivv75e1aRdcQW6ggKgaqg3Lw/7rFn4zj231vC9BEcHCcPoUFAjRnM0HoCZd7UkNSNA225ubEkKBbsMLH87nbxNJtDknolIlROW4+jeWVvo0b9GRjWtSYB7Zmzk3muOIy0zyNK3Mul/RjEp6TWT+OUgbPg1GVWuym8k1C61GpnHKMyLE9GuMDGGeB6jMJGGOG+PsO3iCT3UzXsEtQsz1CbDCcRN0KtE/HFkQ+xtoz06WgGHiHWa+rReIjniRGq9SSYhXKShtvxHaxels/LJLLxVKnR7NlrZl2dm7PsbSHKElxEp0gAgBwR+/jidfbnRcc/780ysX+2g+xllNQvjeFm1OX2iWlvLuz1S8COeN0nSnuPDIOAQVlcje5YSJDhWmTGjjGuuSWXzZh1ut0DLljI33lhZb89HMAhjx6ayaVNNaF5BAdx8cwoLF5aQk6MjJUVh+HAXP/6ox+WqsZBMJoVzz42f5W7vXpHPPjPz6acmfvzRgKoKpKbKnHiin5NO8tGvn5/27YOHdZ6SJEFWlkJWlp9TT62ZO+X3w/r1en791cBPPxlYvtzE4sUWJEmlXz8/w4Z5OP98L2lp9X/eOO+7D8uiRchNm1L67LPIVaIDgtuN7v77cffvj+v66xGcTpImT8Zz/vmoDkf8Qk0mnHfdhWvUKOxPP4113jws77yDa/RoXFdffUQkuRuEomD86ivsM2Zg+O035IwMFKsV0eVCzM1Fl5uLtH07xf/7X02oYR0QystJevhhdHl5Uet0e/ZgfvddXDfd1IgHkqCxSRhGhwm3U2LipR1p29VFk+YB1q+1U1le99PdJNtHy/bRWcZbd3LxwtJfMZoVSosMbPgtiRZt3aQ39eNy6tjwu4OXp8dxgSc4Zvnt3bRqo+gAhdssfP9aU869Pb+WvUKs/yqZ96a1pTDXFJW3CEDUqehNDev4qyrs3W5Gp1do0jKR+yhBgmOVjAyFjz8u4vff9RQWipx4oh+brf5ejm++MZCTE/2+XLdOz6BBGezfL2GxqBx/vJ+rr3bz6admCgtD3qQzz/RyzTXR706PR+CTT0y8846F1auNAHTqFGD8+ErOPttD167BenmaDhcGA/TqFaBXrwD//a+LYBB++83AF18YWb7cxH33JfPggyqnnOLjwgs9DBvmqdccfveVV4IsY1q+HIDgccdhfeEF1C5dcF9xBUgSuoICgu3bh5K5ejwIioJqjZ83UWnWjPLp03HdcAP2p57C9txz2ObOxXvuubiuvRZ/374Hnb90JBCLi7G8/TaWhQvR7dxJMCuLsqeewrx4Mca1a8O21eflYXvuOSqmTKlT2bann8b28suITicqMVSBATVC2rtBx7BvXyjcsUOH2t2mCRrMUWcYGc0KQ0cU0aqTl+8+dbD2yyT+afl6mmT7uOzmfViTZL58P5UfVzgQ6vnENVlkdProF4okQWqT0Oi/1e4hKSXAI7d1oaLCRGmxgYpSQ/3iCBL8o1EU2PuXBVUBrzP2A7Jwe/y3ZsAr8O6UtuzfXvsDO7Odh479YieBjEfuOivz7m7H/p0mJEmlWTsvt83ZQlqzRHLGBP8+PB546y0L69frGTzYy5AhvmPica2q8M03Rt5914zNpjB2rItevWpPR3AwKitFAjF2dzpFysulqu+wYoUZm83N558XsmOHRPPmMnZ7+HtzyxYdr7xi5aOPzDidIq1aBZkwoYJhw7x06PD3Ja+tKzod9O3rp29fP3fd5WTjRh1LlphZssTM7benMHlyElde6WbUKDfNm9dRqlySQoaKLIMsI/j9KJddFqpMVTF+802oHyFJ6Ddtwvb887iHDw9TsquNYMeOlL78MtKOHVjnz8fy9tuYP/6YYIsWeM87D8/QoQR69Tqi/RSxqAjjypWYVqzA9NVXCH4/voEDcY4bh9y8OXK7dlhefz3mvtKePXWqQ79mDbbZsxGrLtxYvdZg69Z4/u//GngUobxKKTfdhP6vvxBcLuTsbCruuAPfsGENLjNBNEeVYZSR5WPqG7m07OhDkuC0C8r443s7k69tjaIcQeOoPjmOIhgwpIxbpu0iIyt0cwwcXM6qT5KZMVGjRKcNwVPCHw4HIq92bjWxb5cRW7fwka/IpiQlBznvsr08MbFr1ZLwB6Macd60oXpRM2U1oV1CoPYwO23IUz1SBIWLVcTdr3GFGSJffV5No+OH0kWGh8UOkQvEC5eLzHFUi6iCEhmOpzlXihgZqhe6tgo3G1gxMYuS7QZUBcQYdpFkkGl3enlUG0VNOOfGr5NDeYwiEVTM9iBN2ngZ8eRWZBFkza2hCBDwCXy9oBlb1ySRlBFg2LhdpGaFJL/lILw4vgO7t9SMNDpLDMwc25EHPgglCAxrVX0GluNcBnUNs9NyJEPuEvw72b9f5MorU9m8WY+iCHz4oZn+/f0sWFDyj09WOmmSgw8/NON2h+6jlStNTJ5czrBhDfMQn3mmj1atZHbsCD8xsfoBf/yhx2JR6dIl/En/008G5syxsXKlCZNJYdgwL5df7qZ/f/8/1hgVBOjaNUjXrk7uvtvJ6tUGXn/dyty5NubOtXH22V5uvbWSHj0ObpQqGRmhMisrQ9LV6eng9WJetgzjjz9SNn06AIFevfD370/yxIns/+qrOktzy61aUfHggzgnTMD08ceYly7FOm8ethdfRM7MxHfiifj798c/YADBdu0a1ZskVFRg+OUXDD//jOH77zH8/DOCqhLMysI1ahTuESMwL16MfeZMpPx8lPT0mGH7qiji69v3oPWJe/eSet111UZRWBmSFPK8tW5N+b33hr7Xgn71amwLFoCi4L7ySnynnx623jFhAiZNQl5x82YcU6ZQNGBAYt5SI3JUPY5vmrqHNl1qHqRmq0qv0yo4/cJSvvzgyE/kEwSVE85w0qqTlzWfJ7Er52D+apWrbttbbRQBmG0KAwaX06qjmx1b6u5CVVWBuVNacdujuWS18iKIIQllQ4zJ7zpdYkL8vw1VhRWTsti/oeaalCE0f63KMBMlhVZ9K+nxn5K4ZYlSKMdG5FVkTQ4w5pUNtO1TGfOdFQwIPDuyG1t/clQb4H99k8xNczbRvpeLjd8nU5AbbXDtzTFRtNtAenbCa5Sg8SgqKmLOnDmUlZUhCAJnnXUW55133t/drGoefNDBxo2G6t9er8h33xlZtMjCqFHRoV9HGlWF1asN/P67nlNO8dGjR928Kdu2SSxbZqo2igAKCnQ895ydoUN9Derv2mwq48c7eeYZe7X4QlKSSmlptEUjy0KYyvSaNQYee8zO2rVGUlNl7ryzgtGj3YeUR+loRBDgpJP8nHSSn927JRYssPDGG1Y++8zMmWd6GT/eWSevnWqz4e/dG/u115LUti0YjTgnTKg2nPB4MC1dinPSpHrnK4JQ6Jhn+HA8w4cjlJdj+vxzTJ9/jvG777B88AEAckoKwU6dCLZvT7BjR4KtWqGkpYU+6enR4Wc+HxQUoMvNRSwpQdqxA11uLrrt29Ft24Zu82YEVUUVRQLdulE5fjyes88m2K0bCALGr77CumABYlX+JqmoCBVQdDrEKkVfVRTx9+mDe/TouMdn/PJLkm+9FdHpjLk+2L49Ja+8QkrfvviLi2stxzZjRigMr7w8VO433+AaNQrnffdVnUgVw7p1Ufvpdu/G8vrrVE6YELedCerOUWUYZbaM7igZjTBwSMURN4xsjiDTFubSposHk0Xl0pv3s+bzJJ65syW1DVdbk+TqMDctjlSZk84pq5dhBLB+bRJjh3Zn4JAS9AaVLn1cnDt8X9hol6dSJGejmUfn/YEtKUhZiYEFs9uwbcNBJkwm+EdTss1A2U5D9Ioqo0hvkTllzB5OHbMvpidJS+dTy8ho7WXftvDrs2l7D+1OqF3y9of3mpDzc1KYV7I438z7T7bmrrfWEwgIUR5LCI36yv5/6JBtgqMWSZIYOXIkbdu2xePxcPfdd9OjRw+aN2/+dzcNgLy86BtRlgW++cb4txtGHo/AyJGprFunx+UScTgUTjrJxwsvlB50CsPy5SZKSqI32rdPoqREbJBIAMCll3o4+2wvn35qwmpV6d/fxpAhEoWF4XV16hRgxQojs2bZycnR4XSKpKXJTJtWxuWXezCbj/2Bw+xsmXvucXLLLZXMmxdStRs2LIOzzvLy4IPltGsXP8TONWYMljPPxCWKKE2bhnk1Um65hUCPHqH5R1VY3ngD7+DBKJmZ9Wqn6nDgueQSPJdcAqqKtH07xjVr0P/6K/qtWzF//HG1YRC1ryCEwu8EAaHKeNFKIqiShNyiBcG2bfEMG4b/hBMI9O4dc26UZdGiaqPoAAIgBIOheUB2O86bb8Z1442hTmgsAgHsTz6J/fnnCXTpQtn06SRPnIhUUjMQqQoCvv79kdu2jesRE1wuLO++G3bsYmUl5iVLqLzpJtTU1NDIhRL7XhJ8ibm7jclRZRj5PLE7S6VFR76ZN0/bTZc+NS+r5DSZU88v4/tlDtZ8HnvUxOsScLskIoO3fD7YscVUnedI0OasiVS204TWqSL4fQLffJwGwKqlaWRkeuncs5KkFJmiAgObfrcx9Mp9pFUbZC5atHFz73+7s3enJepm1L4i4gjKxc1xpEWMKqT2MLs6hxoJtf6It2Fky6q/RT5KApqDiwyl0/6OCqWrZV30dmKt6wJqzUtdkcSY+wDIGus3KheSIOJRoxMphtXjlsj/005AV0xADc99BKDX/Jb0Khc9lsuSB1pTvMuIpFNp0tbLlTO24de0SxLCr4r13zuQg9H3bOl+AwGg08llZLTysi9i7lJGSy9prTyhsDzNNaFEXHVxTad4fZ06XmZ/V8hdgsNDSkoKKSkpAJjNZrKzsykpKTlqDKPaOujJyX+/J+PRR+388ENNB7C8XOTzz0288YaF0aPjG20dOgQxGhV8vvB7w2pVsdkO7diSklQuvzykMJeebuOGGyqZP99Kfr4Oq1Wmc2eZs87yctNNKWH1m80q55zj/VcYRVrsdpXbbqvkv/918dprVp57zsagQU34739djB/vjJp/pUUdOBC5qAi8XvTr1hHo0QPbjBkIXm+Y+IDl9dexz5yJ+aOPcN5+O/6TTmpYYwUBuV073O3awZVXVjVCRSwsRNq1C7G4OOQNKi5G8HhqDANFQbVYsGRnU6HXoyYnE2zeHLllS9DHfy9q6651FaEks9L+/bUaRfp160ieMAH9+vW4Royg/OGHwWymcudOLG+8gbR7N2pyMv7eval4+OGDNke3dStilbR32PLdu9GvX4//lFNAFAl26IBu586wbeQmTaqT4yZoHI4qw+jrD5Np1cmL1V7zMC3YpWfxnLpLJTYWrTpFizybrSpnXlxaq2EkBwV++cpO0+Y+jJoIoh2bzaxeVn8XdCQBv8hD/+1Mi3ZuWnbysfG3JG57ZJvGKArRNNvHlTft5Ol7Oh9ynQmOTpLbBXC0DlD4Z+3DuRV7Y3iUaqHNCZXc+slf7NtsxmSUadIunsh5iKZtYkviWuyhgQG9SeWy+3N595E27M81IUoqme08XPvE1qNBnCjBMcz+/fvJzc2lfWQm+r+RSy91s3GjDqdGJKVZM5lx4+qfiLSxWbcuukMZCAh8+aXpoIbRmWf66Nw5yB9/1Dxv9HqVU0/11jrY3lBuvtnFVVe5Wb3aSGamzN69EjffnEIgYk5sfr6OmTPtPP54bO/DsY7NpjJuXCWXX+7mySftvPSSlQ8/NDNlSjnDhh3k2W40YvzuOxyTJiF4PBR+801ouSwjlpRgnz2b8scew9evH8l33YVqsYTEFBoDQUBp0qRO8tim9HS89UnwqsE1YgTGVatqDX8D0G3bFt08lwv7k09inTcPJT2dkldfxXvOOTXl3nAD7pEj0W3Zgty0aZ09anJ2NkpKSpRxJKemhgy+KsqeeYbUa65Bt2kTottNsGVLXNdei3yUDP4cKxxVhtGSeRkYjCqnXViK1S5TVKDn9cebUVSPDl5jUVvyVI8rflzBi5OzcTkl+g6qQG9U2Z1jZPa9zcMmjaqaDKxClMvl4OzKsbArN+Tqttpjx4E7Uv2hTK9ynBG7OHmMBCFCxEG7TrM8MvdMOBGXV9ixajxjEUWET4KM8l3FqU/ridMWESmOUPM9GOkx0n6PqFqtRbQh0isUjOsx8sVcF0+kQetlgpr8R/0fL+OHSSmUbNajxvDc6GwqbiXUMwlE5LMyUWNMSweOWoCUzgH0goy3qk69UHP9KBEn5JT/FvDLJxlhIXiW5AAnXVlQLdLQ7cxSupxSxqbvHehNCh37VyBKVNeuKZBJEAAAIABJREFU9bJE3Vlxr604I8F1HSQ+jJ6lSOJ5mhI0Ll6vl6effprRo0djiSGNu3LlSlauXAnA448/TnojTVrW6XRxy7r11tC81TffVCgrE2jWTOX++xVOOCGlUeqvL9r2WiyxuwJ2u6FO5+fTT2HcOJktWwT0ehg8WGXyZAOi2HgTwg+0Nz0dLBa4/XYdH34oYjarMRXsSkvNpKfX0YNwGDjY9XAkSE+H116Dm28OMm6cxI03pnLppTLPPisT2bSw9j74IBIgfPQR6XZ7yBMjitC0KcrChSRPm4bcujXq/Pk4VBUOluvoMHBI5/fii1H//BPlrbcQ8vIQYnj79U2a1JSvKIiLFyPdfz/Crl3I11+PPG0atuRkYkoptGhRv/ampyOccgrqBx8gVIliqYKAcNJJpPTpE7Ydq1cjr12LXFCAevrpmJOSqIdae505Gq7f+tCY7RVUtX7xH4PFSxul4qOdEXcUcNkt+8I8P8UFOiZd2p787TEUvA5GLUk5owwj7XYR68LC2apCrSY9vYUzLoie0Pfeq9nMe6pttMa9pgMuREohaX/rwvdT9Zp1mu+qIUI1yKAJFTOGr5PNNetkQ007guYIw8KkCSOLONVBk8aYMEeu09RlUmN+B1DMms6tMbyjK5lqDAijMfxta9H8thpq5sNZ9OFz42z6GuPHrguP/bVqftuk2N8BLJrfkQleLWLNOqMQYM9qE9/clY5nf00nwJQaZPAje2k/ODQarY80jIQYhlEVeo1RrDWMIrcTUSneZWDJtDaU7jZgtCqcctVe+l5QFGYGSHEMkLDtItbFC72MZ2ZIdbR44hv19S8vbl21tNiWteOQy05QQzAY5IknnuD4449nWB0lbPfUUY73YKSnp1PUwBHsvwNtexcssDB1alKYgEJKiswrr5QwYEDDZbcbkwPt/fJLI+PGpeDxCNxxh5Off9bz+eeRXUOVCROcjB9/eL1xHg/s3StRUSHidAq4XCJut4AkqaSk2HG7nRiNKmlpCmlpMmlpCqYGdB8ag2AQ5syxMWOGneRkhZkzyzjttJp3SazrV792bUgAISMD1W6vXm78/HN0u3bhuvbaI9b+SBrjfhOcTowffYTjySeRNGUFMzMpWbiQYOfOGL/5Bvtjj2H46y8C3bpR/sgjIanzxm5vMIj9iScw/vADqCr+Pn2ouP/+UGKrv4F/8vOsNrKysupU1lHlMTqaWPhMU2zJQfqe4cRiUyjZr+PduRkNM4oOI6880Yq2Xd206hAKa1IU2L7RyttzWx5kzwTHAoIA2Sd5Ofvlffz8dCqe/SIGu8LxV5dWG0WHk7QWfq59cTMG4e+fJ5Hg342qqrzwwgtkZ2fX2ShKEGLUKDcFBSKffGKmoiIkXjBihPuoMYoglF3iiSfszJplp2vXAHPnltC+vUxurkROjp7t20PdGUFQ6dkzwJgxrkaru6JC4M8/9fz5p56//tKTl6cjPz9aCCKaaNGolBSZ9u2D1Z9u3QL07h3Aaj2886F0OrjttkrOOsvLuHEpXHllGrfc4mTSJGetAhuBvn1BVbHNnIlYUlI918jyzjsEO3U6rO09Eqh2O96rrkJp1Qr7rFmIpaXIaWk4x49H2r0bxwMPYPzhB4ItWlA6cyaeiy8+fPmXdDqc991H7cF9CY4UCY/RQdAbFCx2mYoSXXTOm/pMlIjyGKm07eLBnhxk4282/F6xajOh9n20eYY0ddscQS4ds5esVh62b7Tyv9ez8bpDTzohymOk+R0RvhXmQYryJtXsF+Y9MoSHKqh6jcfIFOFNMmo8RmHfw9uh9QppvUcQ7kHSbgfhHiRZE9sejJiAq/UgKaaIDr3GgySZwsMJDRqPkclQ891iCO882AwaT5A+3BNk1/y2SjWeJmuEZ8ki+lFlKFou4FwN9q4yzS8NIJlC66rbEeFNMomBmOtMYrhX64D4gqpAyQYDgqjStKsXQQj3Lmm9R4aIMDJtLqRor1M8T1MNDfUmaTlaPEsNKT8pa1eDyk4QzaZNm3jwwQdp2bJl9fPxiiuuoHfv3nH3S3iMaggEQsILKSnKQdXoGorfD7/+asBsVunRI1Cn12hhocjttzfh669FrrjCxdSp5Zg1z/viYpHZs23k50v06uXn2mtdYevri8cj8MMPBr7+2siqVUa2bq15zzVrFjJsmjcP0ry5THa2THKygt0eEpswm1VUVcBuT6GwsBSPR6SkRKSoKPTZvVsiJ0fHtm06iopCJ1mSVLp3D9C3r5/TTvNx4om+Rp+fFX58Ifn4RYusnHqqlzlzSunYMa3269frxXH//Yjl5ShJSUiFhZTMn9+ouYfqS7z7TSwsxPryy4hFRbivuCJk4NUBwe3G/O672F55Bd327ciZmVSOHYvrqqtqV6cjlMdIl59PoFMn1KSkerf3aORYbG9dPUYJw+hQaKBhlJwe4KFXttOqoweTVWFPnpG3ZmXyxftpDTKMQuu0yVM12yUMo3+kYWQK+tlwkw7nOgHVL4CoYuug0PtlD8nptYfZ1ccw2rvOxOf3ZlGaZ0AQIbW1j2HTd5HduWakNWEYJQyjY52EYXTkWLnSyNSpSeTl6dDrVdq3D/L886W0bVu7nPTWrTquuiqV4mKJRx8tY/jw2KIvh4rHAytWmPjgAwvffmvE5xMwmVT69/fRv7+fHj0C9OgRqLP8eF3Ob0mJwB9/GPjpp9Dnt98M+HwCNpvCoEE+zjnHw5AhvsOmrrdokYX77nOQlSWzdKlCSkr89po++QTF4cDfr1+tIV76desItmkTFnp3OKjt/Bq++47kO+9El58PgJKUhPuii6h49NHYBakq+l9+wbJ4MeaPPkJ0OvH37Inr+uvxDB0aX+kuECB53DiMa9YgFhcjZ2XhueginHfdVef2Hq0ci+1NGEZHggYaRo8s3MoJp4c7TPfl6xl3Xmcqygwx9wGOmGHUvmsZo8flkJLmw+XS8cl7rfjmi5rQvIRhdPgNo5LXg+ycJUUpU2Se76ffEzXXTkMNI0mWeeM/7SjaHB4a2rSbmxs+2lR9CSUMo4RhdKyTMIyODB6PwODB6eTmhr8zTjjBx5IlsRNfrl1rYPToVHQ6lY8+UmjVqrDR2/XLL3refNPKJ5+YqKwUycyUGTbMwxln+Ojc2c+UKQ42b9aj16ucfLKPe+6pPfRMS0POr9cL335rZMUKEytWmCgqknA4FC65xM2IEW46dqxb0t36sHatnmuuSUUURV57rYg+feoYPqnNqnsAWabJwIGIJSX4zjwT75AheAcNQk1pfIGRmOdXVUkfNgzD77+HNys1laIPP0Ru1y60QFHQ//knxi+/xPzhh+i3bUMxmfAOHYp7xIjQHKI6PPft06Zhe/FFBE1+IcVup+Sll/CfeurB23sUcyy2NzHH6EgQaVPGu5Gq8hUZTArN20Un42raPMCQ4cW892KzqH2qichxpEUSZM66uJDeJ5ezb4+J91/JoqJUT6Tde0DxJGYTBYX0pl7ueWIdmdk1kp7NW7oIBkS+/7JZ1WFqcxVFCERov0esq3tkrnbLyNE57bo4inVqLcsBbToeOUqxTiPuEFG8r5b8R1G5kDTfI8sPKjVv1IDGiAxEbFfyiyemupozV0+lxjoMRGRv1eY8kjXnX444B8W/6ynJjR7xK91hJH+zjYxOvqjyInMhiZq/TaTRpK1PaxhVFuv4+oUsSnYbaT+gnFOvKEBnUKvKI6KMGiLzxIZtG29sp6HKdg3YLNKAkussj5cgwbHPypVGduyI7nLk5enYu1ekWbPwZ8iyZSbGjk2hWTOZN98spk+fFOL1exQFli418dlnJjIyFG6+uZLMzNjeHUWBzz83MXeulbVrjdhsCkOHern4YjcDB/qRpNBj5cIL0/j555pRto0b9ZSViUyffnhkwE0mGDzYx+DBPh5/vJw1awy8+aaFBQusvPqqjQEDfIwdW8kZZ/gaLYqtb98AH39cxKhRTRg+PI1XXy0NE2WolVgNEEVKX3oJy9tvY1q+HPPSpaiShL9vX3wnn4x/4ED8vXrFDUs7FITKypj5gKSSEiwLFhDs0QPjt99i/OorpKIiVEHA368fpTfdhHfo0Hp7uYw//hhmFAGITifWhQujDKPDhW7dOmwvvwyyjOuqqwg0NK9UgmoShtERRlWptaMlBxv2pBMllWmvb+a4vhXoq/q6Jw4pYfKYTuTn1e9Gv/z67WFGEYAjJcDQS3ZUG0YJDj9SLX828XDocmo4nDlKC7aZmHddZ4ryQgfxx/JUfl+azq1v/lVtHCVIkODYRJJi96UFIXo++2efmbjxxhR69AiwYEEJqanxw9dUFa67LoWvvzZWJ3pdscLE7Nml9O0bCNtu2TITTzxhZ+tWPS1aBJkypZzLL3dHiR98/72BDRvCvVvBoMDq1UbcbgGL5eDPLFWFVauM7NsnUlAgcfzxgboZHYTO14kn+jnxRD+TJ1eweLGF11+3MHJkGn36+Jk4sYKTT/Y3ioHUpo3MV18FOPdcgWuvTWXBgmJOOsl/8B0jEQQCPXtS3rMn5Y8+in7dOkwrVmBauRL7008jqCqKyUSgTx8C3bsT6NaNQNeuBNu1q3ty1jioJhNqjMllKmB/5RUAlORkvKefjm/QIHynn46SlnbI9UZXeGTeZ9aXX8Y2axZSSQkApi+/xDVyJM777jsi9R+rHHOGkSiqdO4dmiOx6VdrWP6go4GATyRvs4nMluEPnb07DKxY3LAbdPAl++nevyJsWlDzNl5uuGcHD954XL3KSkmL/TC0WBrfhZ+gdjKuBdcvENxfs0xKUmlyaeOovzU93kty6wAlW8JH7lJa+0nvWLcXd335+JHW1UYRgCqL5Pxi59tFmZwxeu9hqTNBgmMRrxd++cWAw6HQrVvwH5EwedAgL23aBNm2LbwD3LZtkKZNa55rX3xh5KabUujZM8CbbxZjtx+8k/n110a++sqI319jYe3apeOxx5L44INQmN5ff+l4+GEHP/xgpGPHAHPmlDJsmCdqOu0BcnJ0YfLlB6isFCgqEtHpVLxeIWp+1LhxyVgsKi+/HDL6rroqlYkTnbRpE6RZs9ojNuKRnh7ygF13XSXvvGNh5kw7l1+ezsCBPqZMKadr10N/P2dmwjvvFHPJJWmMHp3KokXFYUZlvRFFAj17EujZE+ekSQilpRjXrMGwejWGn37C+tprCL7Qu0aVJORmzZCbN0fOykLOzERNSkJJSkJ1OFCNxtDJFIRQfh+vF1EQsBYUIFRWIu3bh7R3L9Leveh2745qimq3U3HnnfhOPJFg587RaUwaiK9/f/R//BEVSuceMaJRyo+L14vljTeqjSIIeavMH36Ia8yYw2Pw/Us4pgyjjj1djJ+eT/O2oZstf7uRGROas+V369/csnCm396S+1/KpXUnL0azyr58A/OfakZluQ6tOykqx5E2tE4TVjd8zN6YD/f0TF8oZkBbRNXQ3NmX7OXc4fuwJQUpKzawaG5LfluTwYbfHPQ7rSjqubF/rymklwrhc5gi6tS+wiJfKUqcdbVTeylCVGZYal8Xtl1tIXfhv5WIxKraMDu/drsI41vWhuMpkaF0mvxNauzvANYWEo77gjjn+VFLFMQkSL5QQjfIQLnGs2iRIttYU442zC6ghl8gJjHAgEdKWX1/Ks48HYKo4mgT5NTHi3CrxurzYNDMMVIi5jNpQ+kiw+z0mt9KVWFlBdGhe6ossvmHZAZevR8x4o8haWIeI5PLhgVbxgmzi5dAVtviePOUFE274l23ckQbGyP/UYIEkXz4oYmnn7aTl6fDZFLp1CnISy+VkJV1dEvmm0zw+OPlPPCAgx07JPR6aN8+wOzZpdXb/PCDgeuvT6VLlwALF9bNKAJ4+OGkMKPoAKEcQwLTpiWxaJGFlBSFxx4r48or3eh0sH+/yKRJDnJydOj1MGCAj8mTKxAEaNlSJilJpqIi/MnRpInMzJk29uyR0OngrrsqOO64kGGye7dIWZmI16tSXhVt16aNTLNmMiaTSosWh2bAGAwwcqSbSy91s2iRlRkzbJxzTgbXXedi4kTnIYs0pKYqvP12MRdfnM7IkWm8/34R3bo1zqCompKC95xz8J5zTmhBIIAuJwf9hg3otm5Fys9H2r0bw5o1SIWFCP6De6wOpJWVU1JQmjVDbtYMf8+e6HJzkfLzUQUBuV07yh5/HCU7u1GOQ4vz7rvR7dqF4aefQuIL2dl4LrgA32mnNXpdkehycpD2Rg8o6vbsQb92Lb4D5zlBvTlmDCNBUBk/PZ+2XWvCwNp29TL+qXxuHtIxWmr7b8RZpuOuyzr8P3vnGSVFtbXhp6o6T08OhCFnRJIomDCACVGMCCKKIBjgYsLL5YoBjJhzwAgqCIKgKEau8qGCYEBByTnPMLmnc1fV96OZ6erq7qFnJFPPWrPWdMVTHU7VPvs976ZRCy89Ly1HklS2rq1bfaSsvADpOfE7Lr83/mPcGRfuZdg9W0jLDD8a5jfzcdcj67j3ZgfzpufT45wi2nYqr85sb9vk4K1nj/6aBUcb9h4m7D1MMcYMB4qcTkEu/aQA198ioqSSfcKBkWUkwpoSf7Q0o+HBuT4Dg2ONsjKBSZPS2L49fOv2eASWL7dw110ZzJxZsp+9Dz+nnRbgm2/28ttvZhYutGKzRQxQN2+WGD48iyZNQkybVkxaWnIP+Xv3ihQUJBqyUDn//Fx275YYPtzNXXe5SE8PH1eW4YYbsli5MjJgs369CY9HoGfPAF9+aSMrS8HrFQkGwx1jXp7MWWf5MZngmWfKWbjQyqxZDjIzK8nPV3jnHScXXuhj3ToTu3ebyMuDM8/08+uvFlJTVRYutDFsmJtWrf5ZsGGzwbBhbq64wsPjj6cxebKTr7+28eyzZfToUQcJnIa8PIWZM4vo1y+XIUOymT9/b1RG74BhNhNq1y6cwYmHz4dYUYFYURFOkQKCqoKqolqtZDRuTLHfj+p0HhAZXp0wmyl94w3EnTvDdt3t2qGmp+9/vwOAUr8+Sno6oscTvTw9HblFi0PShmOVYyYwatXRS8NmsQ9YDZv7adXRy/oVjoPfCO2I8X6eMPPy/Yx/YytNWvkwmeHSG4tY+Ekmrz2ocX9TdMYJ+gwS0LydhxRnbKelqvDbj2mRLI+GvgP3VAdFVeTUDzBw+BaeHteOe4d15PyrCul0Sik7tziY+35j3G4rVWPsNV2Zdl2MNUKC90S/vOYkd12KqyXOLOld36KyTvpMkDZhp0R+OkHddqrmtaLLGGkzSNr/Q/rMklmT+dEbOKjaTFD0uxXUZJAc2u10Jg3a/ewnBgGBStUGarSbnaJxudMbOJhVbTZJd51xzBe6XVdEwXoH3vLIe5fd1MtZt+4mqIpRLncQnSXSGxkkm02KOl6SmaVwmxNQq8jRmDdlcGCZM8deHRRp2bTJhMslJJ1hOZwsXWrhnnsy2LJFAgSmTHEwdKib2bPD9+gpU0rIykr+OlavNuFyxfvFqmzfbqJJE5m5c2Pd1r77zsratTrnVEVg2TIru3aZmDixnPbtQ1x0UQ5paQpNm8rcemsln3ziID8/HNj06OFn/nwb27aZqFcvgNms0rChzLZtEqWlkJcHjz1WXt1tjB6dwZIlFlq1CsU1dastmZkqTz5ZTr9+Xv797wyuvjqbu+92cfvtlf9ILZafrzB1ajGXX57D0KFZfPxx8UGzDE+IzYZis6Hk5cVfn5ODeoS4pin5+QQOQkaqxnNmZxPo3h3p88+jTLUCXboQatPmkLblWOOYCYzCAwkC+ocRVRH0arIjgtuf2EGL9pHsVka2TO+rSlnwcTbrVyQv/du2wU5xoYnsvOgRqIpSiU/eqR93H5s9/sh9Slr4GKGgyJez8vlyluaHfpCKPRscP3S6rDQcsE/PwV9pIq1+gL7jtpGaG/7eFWywseitBighgdOu30Pzzgeucr2BwbFA+GFXJd7w1NEwz0hVYeLEcB2jKvbsMfHcc6mEQgLTpxfTrFnt5uG0bBkiN1dm7159JCDQr5+HJ58sx+mMfajfulWKK79zuQRatw7i8YTf0O7dA7RqFeKGG8Ij86KoUl4e3s9uB6dTxeMR+N//bLRtG6JXr3CGqKAA2raN/lzy82V27JCq34sD9ZmdeWaABQv2Mm5cOk8/ncavv1p47bXSpLNu8TjxxHCNqWHDsvj3v9N56aWyo+I7djxR9sILKHl5WJYuBUUh2KkTFQ89FNnA58PxwQdYfv+dYOfOeG64Ia45hUE0x0xgtGGlnZ2bLbQ6MdpRbcdmCxv/OvK+CA2bxma3UjNkLh5UxAsJA6OqTi7SO+3dZWXlkjTOuLikOpvs9wn89HUWrjJz3I5s5xY77btWRi2TZfj710OTAjY4vITcsGGuDf9ekSZX+rG1+AcTbGtJ58tLOfmKyChfVTZpyQd5LHghn8qisKzl7wWZ9Byyh4vu2nHI2mZgcKRz5ZVeJk92xthet24divvwf6Sxa5fErl2xqQyfT+Tyyz01OqHtU1HF3NPy8xVOP93P/Pl2QtXzL1VOP93Pq6+WJTxenz5+XnlFprAwdg5RRoZaHRg1bChTUBDZplkzmTVrTPh84Tk/TqdKKCSwc6fEBx84WLjQyq+/WvjrL5GsLBNOp7ovUJIoKREZMiQ84KN34vunpKSovPhiWEo3fnw6/frlMHVqCU2b1s3wAeCCC/yMGePi6afTOPtsP/37H5ziukcdPh+O2bORtmzBe8UVhDp0ODztMJupmDAh7irB5SJ7wADMK1ciKArqvHnY586leMYM1IyMQ9vOo4xjJjACgSdHN+Ge57fTsLkfVNi1xcpTdzSmZvHXQWI/NY7ijVQBlBbFamVtDpk7n9xG644eBFFl6zo7z49tRkVp+ON78o5mDFhvo+uZFagKLP0ukzlvx88Woaq89XhTmrVx07ytB8kEfr/AmuWpfDK1IVTL92qod6TRJAmhxFrpGOlcguGmWHuFyNeyNrI6IdFzQYzBgqb9+o8pqo6RbqVG0ibXUMdIK61TdOu0Rg2y9n/ddtr6R9raR/rXQVP0Or9W4qeV1e373/W3wJpxZnxbBUBg8xw7Ta7w0/7uyIBCQIwcI6CR3GmLx0K0MUOQ6O+BWbOfdjuIldmF/AI/TqlfHRQBeMvMLJudy6lDCknJDNVZZpdoH6g5CaqV3dUkuYs6r+51TY8jNZk9GBgkIjVV5aGHynnssTS2b5ewWsPmCy+9lDgAOJJwOBQscaz5BUHl8ss9cfaAQADuvTed33834fXm0aJFiCefLCM/PyIFeemlMpxOlZkzw3K866938/DDFTW2JT9f5sorvXz4oaM6A5SfH+Luu11s3Wpm40YTPXsGyMxUWL06cl/u0iXA+vUmvvzSTna2jMsl0KlTgNatBU44IUhamkLbtjZSUhyceGKIefNsLFpkpU2bEP36eWs0M3jxRSeNGslceqm3TtNmBAEGD/bQvHmIm2/Ool+/HKZNK642h6gLt99eyQ8/WBk/Pp2TTw7QvHndA61jAWnLFrKGDsW0cSOCLJMyfTrevn0pf+qpw920KFKffhrLn39WvxZUFcvKlaROmkTFpEmHsWVHPoKqrwC6H84X+x+sthwgVJq2DT/khQ0NjpAHEN2D0K0Td3DJDUXVdYcAdm2xcMel7aoDnioemrqBHr2jO/m/f0lhzFVtqb4+7RwPzVykquBElFRsdhmPxwwIWKwyFw0opNWJbn7/IYNFX+SgCqa4xwCi7S01FniCfthLa49njr4OVfM6+n+ds5pNs86im7tilTT/69dF2hyyaf636oIOzTpZ53khW7XbqUmtU606raY1cuOQrNE3EYslcoOyWSKBht0SHXQ4zJHXKaboUVSH5rV+nV2KvHZKkaykY9/yP24w4/oj+v22ZCucMc1FSuPwdVg1AZBNiByvpsDILOgCI01YoA+MtNtKqOxZbeft/m0JeGLD4BveWEuH88tiAiMpyikuucBIIvnASEoQGNVEbST9VYFR/fxdtdjL4GCwa9eB+QwOZaX4UChcbDQ1Vam19KyKw1XZ/sYbs/j22+iOt02bAAsWxLqhQnhezty59qgBqBNOCPLll3urbzc//2xh8OAs6tVT+OCD4oQP74EABINCVM2iP/808cEHKaSlKdx8s5t69RS++87KtGkO/vWvSr75xkaLFiHatw+ycaOJyy7zsX69iUmTwsXmLr7Yx1VXRWdSFAXy8mr3/qoqXHhhLn//bSY/P8Rtt1UycKC3znN7Nmwwce21WXg8IrNnF9G+fc3BUU3fh507RS64II/WrYPMmVN8wLNddeFwfX+zrrsO28KFUcuU1FSK33uPYPfuCfc71O3NHjAA648/xiz3n3oqxR9/vN/9D9f7W1eSaW/Dhg2TOtYxlDGqQmDr2iNPOqfnjYn5mC0qXc50YbUpFO228NajDWKCouz6AVqfGDuS1qydlxYneNi0an/zkVSG/Wc7PXqXYk+RKd1r4cNXG/HzgizmvR/9JREOjLW/wRGK7AH/7tiBgkCxyI7PLLQd6Yuz18HHmRPEmirHBEa2tBBZjQ3HOgMDPSYTdOx46CSwB5JXXinlzjsz+OknC+XlEm3bBnnnndK4QZHHI/DLL5aYrPz69SYWLLBy0UV+/vzTzA03ZJGfLzNrVjF5ebGTigMBGDs2nV9+seL3h7NFEyeW06VLiM6dQ3TuXB61/Rln+CkrE3nlFSdduwa58kovRUUiWVnhgaLWrUO8/XZpzHmqpH51CRwEAb76ai/ff2/lpZec3HdfBs89l8qYMS4GD/bU2kyhVasQH39czBVX5HDttdnMnVtU52xPfr7Cgw+Wc9ddmUyb5uD66+Nn944HTFu3xiwTXS4cM2dSXkNgdKhRUuNXiU+03CDCMRgYHaHoHOsUReCl/zZGMqlYrAped1Wvp+nUBZH0zBC2lNiO3p6ikFM/yKZ48DXyAAAgAElEQVRVVcePX+No4O276DdkD9Z9GY7cBkFGPriJLaut7NkZHVSpGmcToSZhkCb7FeM8p3W60DniJZu7i6pRo/uK1nhvUBPcjWLkctp1yTvWCdqPRrOdvlaRErUu+hj+KMe6xPWOanKsC2ikdPr6R9G1kSLb+RUTKiqqOc7oo6Qi5Iq4ZBtqCHa9a6d4qQlRgib9vDS+LPwwEK8WUhVmXVRdG5mdKQfyu3pY87U56j1v0MFDZtsgPtUUVTMJorM/sZmg+OuSldxBtOwu2cyS/ldam5pHBgbHCykpKhMmVHD22bn07u3jvfcS24y73QI+X+ydIxgU2L5dYuNGicGDs8jKUpg5M35QBDBuXDqzZzuqA6zdu02MHp3J118X4XDE/hat1vB8riuvjGSCkrGs/qcKWVGE3r399O7tZ+lSC089lcq992YwfbqDJ54op0uX2gXDTZrI++oSZTNgQDaffFJU53pX/ft7mT3bwWOPpXHBBb6DY+F9FKDa4pdWUerVi7wIBnG+9BLWJUtQTSa8/fvDzTcfohaGcf3rX1h++w2pMFIlXs7NpfK22w5pO45GjoCE6PGNHBI0QVEsW9fbKdwZWxhzz3YLfy3bf+Tfo3dZdVBURW6DIP1vjS0MFh8Vm0NGqOEh0uDoQDALmHNil5tSoF7f8E3u73vMbHjVSukyE8VLTKx4KIVVzx38DGy/57Zz0qAi8tp5yG3r5cTLS7h28saDfl4DA4NDz+OPp6Kq8Oij5TVul5Oj0KBBbJYjL0/m3HP93HRTFgDTpxdTv378B3VZhmXLYrNOmzaZeOYZZ63aHQhUl9Q56PToEWDWrGJefbWEoiKJfv1ymDQplWAtE4WtW4eYPr2E8nKR4cOz8NcxCS8IMGlSGV6vwHPPHb9Zh2Dz5jEDworDgXv48OrXmSNGkPrcc1gXL8a2aBHp48YhTZx4SNsZ6tKF0meewX/qqQRbt8bfvTtlTzxBsEePQ9qOoxEjY3QkoyrIQZj+fD2G37eLvPxwj1hcYOLTt3NxV4hUpTiiahxpskdmS/ybRUqqTIyPuSb/r8oy515ayBXDdpORFcRdaWLh/Fw+mtwkuqCPvsma/5Otd1QT+shdVZPVE2j2rDFjlHhdTB2jRNkk3dshaDI8ipzYfCGQIHsE0RmjoBz9LgQ1c7L0maaAHPlJ+zXzj4ImCVVWCVTESiAUQaC40I7shbJfiUoOyl6Rnd/YyL9ZJehIXAvJrJtHVOtskhnOnVgYtV1IMBFSq1ZHH1/SvOn6jJF2W0U7F0kX3Ac1L/XH0GaXks0s6Uk202RgcDyxYoWZuXMdjB7tonHjmqVdggD33lvB2LEZbNsW7tuysmSuucbDM8+ksXGjienTi2nRIvFxQiEIBOL94ATeesvJiBHuhEFVFZWVAnfdlcFff5mRZWjePMSzz0YbQBwMBAEuu8zHOef4mTAhnZdeSuXXXy28/nopOTnJn7tjxyDPP1/G8OFZPPBAOk88UXNAmogWLWQGDfIwfbqDkSMradLkODNiUFVMO3fGPr8IAuL27ShZWZj+/hvrsmVR6hnR7UadPRtuuilcnfcQEejVi+JevQ7Z+Y4VjMDoKOD/5mWyYomTy4YVYbaozJuSS8EO6373y67nDwdAOnwekYXzsmrct1UHFyPu3UpmTvgBN5cA1zTcTlmxhW/mRuobWW0yKakhSou0I3IqJ51WzLl9CygutDF3WnMqymOzXgaHFrlQRS6OXa6UQ8VPIJhAjmPkFCwFf6GAo9lBb+JBIeAV+eT+ZuxY6QBVIL+Dmysf3Yw1jkTVwMDg4PLYY2lkZcmMHFm5/42Bnj0DzJ+/l1mzctm508fgwR4WL7bw8supjBtXQc+eiS2+AX76yYrbHS8wUvnvfyv2GxQBjBqVyYIFkQfanTtN3HRTFl98URQ1n0hVoaBAJD09dsBk+XIT772XQkqKym23ucnPTz6oSE9Xee65Ms4808/YsRn06ZPD+++X0K5d8m5zffr4GDnSxauvptK9eyDGMCJZRo92MWOGg+efT+XZZ48ON8QDheDzIRYWxiwX3W7s8+fj6twZy6+/IpbHCTyLi5GKipAbNToELT2EhEKkPfwwlp9+QggGCbVqRfkTT6DkxJGnHCUYgdFRQuleM1OeTL6ycla9AJNmrKd+4+ibRmWFyNL/ZbLk28wa9dBXD99VHRRV4XAq9OpXyDdz8xFFlX89uJZO3cuxOUKU7rUy481m/PR9ff7z+EpOOWsvDkf4hnPmeXt48r7OrF+XnfwFGxxwxHQB0Skgl+vm5NjB2gxM2SClg6zr082ZYM07eqWU0/7VijXfZVa/LljvoLLYzIj31xzGVhkYHH+sWGHmhx+s3H9/ea2Kj2ZlqYwfr1BUVMH27RKPPJLGuef6GDWq5uDqnXccPP10WrUdd1VxXFFUufZaD7feuv8i0sXFIn/9FfuotG6diZ9+slQHZl98YeX551MpKJBISVE491yRiRPDRhmPPprKtGkp1e348ksbDz5YQb9+tdPlXXWVlzZtQgwZksVVV+UwfXoxnTsnr637z39c/PKLhQceSOfss/21yjpV0aCBwuDBbqZOTWHMmIqDnjU7klCtVlRnrPxSNZsJtmkDQOCUU5AzMpDKdEFjTg5ybu6haOYhJf0//8ExaxbCvjnl5g0bkHbvpujzzw98sa5DhBEYHQ50RgzJ7xdtzBC9StWsEhhyzy4at4wVE6/5PYWn7mwR3keMvjFpU79WR/zRLLNZBllm2JhN9O63p9puPCsnyC1j1yGICiefUVQdFAE0bOxh2OjV/Pe20+Ies66qopifnLZ/1r4fMaYMNcnsNFbnim5l1DrNPjEyOG079FI6zeE0MrtgjOROI8eLOUb8ekoQbcwQ878JhK4yFPjR+iGYWwuEOluRBQHryX48C5VqOZ3oUMm4ECpNdoKh6MIafo1czqqz8t6fzG7PUit/v+NE9gpktQnQ7c5SnGl6y+/Ia73xg97yO+o90Fh0eHZKbP8z9ka2fWUKuzfZyWke/o0oMarxCInMHCDW0EFLXSV4BgbHKm++mYLTqTBoUN1czVQ1bKQgivDEE2U1PncpCkyf7tAERRC+26g89FAZQ4cmlzGpqIhvAOH3ixQWhoOm5cstPPpoKi5XuO8pKpKYOlXl119zOOMMPzNmRLdjzx4TL77o5JJLfLV+duzYMcicOUUMHJjNNddk89FHyQdHJhM89VQ555+fy4QJabz8ct0yPrfcEg6M3njDycSJNdeLOqYQRXznn4+0bRuiZrJWsE0bfJddBkDohBMInHoqtgULqus8yk4n6oABYVePg4T1m29Iee89BJ+PQNeuuO6+G+wHd36w4PViXby4OiiqwrRmDdbvv8ffu/dBPf/BwgiMjipULhlSxJl9yhFE+P2HVGa9Wj/mwRmgftP4nb4jVaFn32KK9lhYvTyVRGHJb/+Xwck9yzDpisxtXhsuoNfp1LKoGkwAOfUDDLplS1z5Xm59H1WjdQaHD9s9TvypIurvftQQmFqI5I0zVde7yn3EgneGn8qfVQQJ8i5TybrgwLZhyxd2lj2cia8k/BCxZ6mdwj+sXDljB9IBVly6Cs34KmLnpXnLTVQUWKoDIwMDg4NLYaHIvHl2brzRXatskZaXXkph4UIbTZsGeeqpVMaPd5GbGz9jUVkpsHNnvDmpAnv2mPjySxspKQpnnBGo0Qq7aVOZ/HyZsrLojRwOhfvvz9AFXhFUVWDVKjN//22OMX4A2L1bYscOqU7zdJo1k5kzp4grr8xh8OAsvvqqKGlpXuvWIUaPruTZZ1MZONDDmWfWLEWMR36+zOWXe5k+3cGYMa46f55HI65770W128OBj89HqEULyh97DG1F3tLJk0mZPBnrokVgNuMZOBDnjTfCQaoL5HjnHdKefrpawmdZsgTL8uUUz5r1z60Sa0AoL0fwxA5yiH4/0qZNcJQGRkdnnus4ZeTDOxlx3y669qykyxmVXH/3bsa+uCXutvnN4nd2LU5wM/7VTTz2/jqem7OK9Oz4I01fzKjPr4sycLvCXxG/V2DV707efqIZAOZ4ts+A3y/pXboB8HpMGEHR4UeQBGwjU8ie4iTnAycZDzmQUoWo9dk3SDR91USTl0wHPCgCWDU1tTooqqJ4tZU1s9MO+Lnqt/eQ0Sj2t5DZ2E9+x/3LaAwMDA4M8+bZCYUErruubtmi//1P4Omnw33E1q1mZs1KYcCAbMrL499XPB4Bjyd2nSCofPSRnREjMrnhhiwuvDCHdesSR0aiCOPHV9CoUXRGW1EEevf28corpZx3XvyByJYtQ0yZUoxVXwQccLlEPvzQQUlJ3e6LDRsqTJtWTDAocMstmbVymxs1ykXDhiEmTUqjrsnrG2904/GIfP75kV838oAiCFTefTdFX3zB3u++o/Stt1Dy8qK3MZlwjxpFycyZlHzwAb5LLjl47VFVHB9+GDWvSQDMf/yB9bvvDt55CVuUy/XrxyyXc3Px9+lzUM99MDEyRocbfa+UILpPSQvRvXc5Nk3NBbMFupzuol6+l4Kdkc5JkuQYN7QqbPs2s6cotO/q5s7HNzNxeOtwUzRhshKEiSPacOIpLrqdXcb6laksWZAVHvmSVLZvstO0dfQNrrJCYurzzbj1vxtp0jKyzucVWfJdDmoocmM5EI51ehJF+TGKJq2SUSezq7nGkaYmjjZLp7/nKYnleLKslchF1ulVe9rj+3WfpfYY+hpHIY1jXVDW/K9E3/gDSuSnHzAFEq+TIv/bxejt/GJknUM35BoUE0vpvKWxDyFqSGD7byk0HhA5h9a9Tn8MrdOdJER/AGY1sq1kVehyQzFLXsnDUxQe0XNkBek6qATVIeLb9/nH1ELSHrMG9zqpjq53R2dpTgODuvPJJ3Y6dAjSpk3yhgFa7r5bQtb1hWvXmnn1VSf//a8rZvuPPnIQCsXeFVQVCgvDfVcwKLB6tYUxYzL47LM4zjRAebnAsmVWysrCx2rTJsidd7q4+GJfdZKgYcMQy5ZZqaiIPl+rViF69w7QoUOI33+PpMMFQSU7W2HaNAe33JKcCUU8WraUee65sNvcxInpPPZYcm5zNhvceWclY8dm8O23Vi64oPaZ8y5dgrRqFWTWLHudpZEG/xzB60UsjS02LHq9WJYuPbhyNkHAddddpN9/P6ZduwBQ0tLwXnbZUW0yYQRGRwn1GgfIyIlNxWTmhWjWzhcVGCmKQCiYXHjRpJWXxBI3gb9+SeOvX9JiJtG9/GArchv4adrGg82mULLXzA9f5fDHkmwm/svObfeup16+H79PZMn3uXw4udl+qrMaHM3s/VZk90wJ2SNgy1dpNzaErV78oUhrpoJbVzxckFRyux0cWVu3G0podmYlv72dAyqcMrSInNaGhM7A4FCxdavE8uUW7r+/bjbRqgqbN8e/p23cGP8xxmZL3hRg2zYTe/aIMQ51K1eaGTo0i927Jfr29XLXXS7at48N7Lp3D9K/v4d58+zs3SthtSp07AhPPFGOIMDUqcX85z8ZrFtnwmyGU0/18+CDFQQCAikp/0yG1qePj1tvreT1152cd56PXr2S69uuucbDq686efHF1DoFRoIAV1/tZdKkNHbuFI8rE4aDiVBZSdrEiZhXr0a1WPD27Ytn2LCEg+aq3Y6SlQW7o2tTKnY7gdPiz+s+kPgvuoiiTp1wTp6MUFGBZ/Bggt26HfTzHkyMwOgoYc9WK6V7TdhTokfti/eY2LQqOpWtyAIb/nLQMIGcTossC2TkhBj18Bbym/sJ+EWWfJvBR6/lU1PuprzUzF3XdObknmU0bO7l5/9lU7jTBhLs3ubggVs7g5Qoh6NyTp+dnNFrDwGfxMczWrJpbcZ+22pwZFL8Oex4xkyoLPx9qfwLlm+0c8o0D6aU2O1b3ejlz4cl/MWRSDnzxCAtrjh4o47ZLQL0eXTnQTu+gYFBYr7+Omx1ffHFdauOunSpBb8/vuV2u3bx86+XXurjkUfUmCxTPCQpPEVk9WoTjzySxp49EsFgOGDKzZWZP38vXbrUnOd96KEKRoxw89VXVlq2lLn66lRKSsLBQlaWyptvxo7qayXpqgpffmllzhwHJhMMHVpJjx7J5ZbHjq3g22+t3H9/OmecUZjUHH+zGYYOdfPgg+n89ZeJc85J6lRRXHihj0mT0vj+exuDBxtZo3+MqpI1eDDWX36pXmReuRKpqAjXf/4Tfx9BwH3ddUhPPlnthKcCwZNOwl+XD7UOKA0bUnGIC9geTIzA6EgjgWOdp1Lix/npXDKkGIcz3Nn6/fDb/6Wyd5clJoZ5dkxTbHaF1p3dpGfKiHGyNaoKFpvMc3NX0aBJJIhq3taDPUVhylONI02JKgar0qZzJS3aefhzWQa/LMqsXq53J4nHvx//mzPOK8RiDV9rl+7FTHm5Dd9+1pimLSu48IqdlJVYmT+nGW6vrhia5v0RdDJEbQtFrTatpqKwMTK72jvW6V3ptCosuaZ12o9aV8RVK4XUyyK1BV8V3X5amV3AFLluveROK7MLyDqZnWY/v0ZKp/0fwC6Fb9q7PwoQKot+s9wbRNa/l0LDEWpUsVcAZ2+ZDtkBtr1jRvGopLWTaT3Kh8dkj/oQbYK2SGz0KK1FI6UTdVI6rQRP1Dg5WnRyvICaWI4nqVqJnLaYrG47rZQuRlMZf7t4rw0MjmW+/tpG+/bBOhcE/eADB06nitOpsGdP5Hd7wgnBhJbb77+fgiwLNGsWorBQxOcT9hkVxQZKoRA884yT//3Pxo4dkX5OklRGjqzcb1C0caPE4sVWTjwxyIgR4QBBFFNrdY333pvG7NkOPJ5wX/3DDxZGjark1lvdCAJ8952VBQusdOoU5KqrvFg0RjVWKzz8cAWDBmXz7rspSdmQA1x9tYfHH0/j/fdT6hQYtW4dIj8/xPffW43A6ABg/f57zH/9FbVM9HiwffFF2GXObI67n3fIEJSmTXG88w6iz0fglFNw3X77QTVeOJYxAqOjiLcfy2fj33Z6X12KKKos+TqDz9+LXxvI65a4f0grJs1YR9czozXMqhr+E0Wo3ygYM83Jalc57bxSpj7dKMZNx2qTefDNdbTu6MaZJlNaZOLXRVk8+9/WJDM7qFnrSrqdXlwdFAFkZAXoN3ArjZpVcv6lO0nPDN+Ezr9kO8891pVVK4/eQmHHA4or3kO+gHdj4of/jC4KGS/6Y2y+DQwMji28XoFff7XUeS6N1yvw5Zc2brhBYeDAIp57Lo2yMoFWrULcc48LpzO2n3G5BN56K4VLLvHy6qulXHBBDmvW6C0vVSQJZBmKiyWmTo219ZdlgXnz7AwdGv+hX1XhjjsyWLjQSnGxRGqqQteuAd59t6RW17h7t8hXX9mrgyKAsjKJt99OITdX5t13naxda8LnEzGZVKZOTWH69GKysiLXfvbZfs4+28drrzkZOtSdVNYoI0Pl4ou9fP65nVCo9nO/BCF83vnz7SjKUVu25ojBvGIFojfWyEN0uRBcLtSsrIT7+s8555BliI51jMDoSCaOMcPCT7NY+GmcH0eCGkfp2bGdnSBEDyTEG1SwO2UsZhm/Lzw6V5VIGTZuG13PiNQtyMwJ0bPPXn7/IY2Fn+Wiaibhx8sedTttL2mZsW3KyvFxXt8dpGvWNWzs4aZRfzNmxJmRY2ovWXcMbZ+s1pQl0hKTFdL8r8/2aLJJQnQCLXq7JGscCZpMkP6til4XfQxVc8yQvv6R5nXIrKl3pMsYabNCQUv0Or+sMVwwRQIXny5j5Dfte50Zgi26C5DA1M1CWVCKCX6sYkjzf2QETJ9Z8gtmzXbR67RmDBZdNslH/CxRUJcx0maJtJkl/X7adTGZpaQzRon3MzA4lvnjDzOhkED37rW3hQZYtMiKzydy+eVB2raVef31WEmano8+cuB2i9x2WyWBAAkstQVkef8lJCorEz/tz5pl57PPbAQC4W1cLpFFi6w89lgar76632ZWs2yZhcLC2HtWUZHElCkprFgRsfwOhQRWrrQwcWI6L7wQXYfollvcDBqUzeef27nqquTqNF14oY85cxz8/LNAu3bJt7mKk04KMn16Cps3S7RsWbeMoEEY37nn4nzjjSiHOQA5Jwc1w5hucKgw4vtjHG9l3RwPSgrM1UGRltZxLI5tdpWb79vCU9P/4tbxG7ElKA4LsHFNKl5P7NdOECAjThCXk+fDkWJkFY5kHDfbEOtpFghg6yiQdonRvRgYHO/8+ms4U9OtW90Co2++sZKWpnDWWckNJqgqvPtuCt26BejSJYjVCqmpsfsKQnJ19Ro2TJxJmT8/EhRpjsxnn9np3dvEwIFZzJ27fzvrNm1CpKfHDqw0aCCTkaHErYO0YUPsuHbPnn5atgzy7rtxJncm4Oyz/ZhMKl98Ubf+ukuX8Oe6fPkBLkJ3HBLq3BnfmWeianSScnY27ptuMtJxhxAjY3QU0KqjhyFj95CRE6Ki1MS05+qx6tfYtH88vpieQ+PWPtIyNCPnAWKKs6pqJHNUuMvMBy80jHs8ORT/RpKZHSIz28WJ3V20PrGSf1/XKW5n/seSTDaudnJit0jWyecVWfNXGt1OK8ak+0b6/SIBv2FndyRj6Wwm9xU7rneDKGUqlk4SudeBYDL0zQYGxztr1pho3DhEZmbdsqQ//mjFYlE56ywTTmcWo0dXcvrpiYOs1atNbN5sYuTIcDZFFOGCC3xs2ybh82nUFOlKTNHWMJGAqUWLEA88UBFnmzCWBLFAYaFIYaEA2PjjDzM7doiMHp143k/79iG6dg2wcKG1+tySpNKhQ4BEwZvNFvt+iiIMGODlscfS2L1bpEGD/TvFpaaqdOkSZMmSuj0Otm4dwmxWWb/eeJw8EJS9/jr+GTOwffMN2O24brmFUJcuh7tZxxXGN/kIJ7+Fjwfe2kK9xpGsSdM2Pibe1Iz1KxKMCmmkPwtmhecg9bm2CIdTpmi3heICE72uKMVqj3SsOzdb2PS3g4oyE7Neqx9l/w1UF9pZuiCd1h3dWON0ylW06lDJBVfu5qQzy2nYzIffJ7LoixzmvR/2tX/w5g6MGLeZ5m3dhEICi7/L5bPpjXhx5q80ax25ecgh+Ov3LEI+FZCj2gGxtwtti7TrYsbz1Pj/h19rTRt0NY4SSOR06ipCNdQx0u6nldWhl+1pkm7642sNFlRLYimdGlUzSVfvyJS4FpLVpAmiNcYMfl3Uqq13ZKtnQhoXdmRXgQoxVF2sRyuXg4hpQ3hd5H9tXSQAsxhph1Ux69ZFRnFjaxzFr3+kl9xp1+mlbj4hvhxPL7nTSusknfRVK63TS/AMDI4XNm400bJl3WoXrVplqjZDKCoCsLF2rZnXXy+he/f4SoLvvw8b9vTqFXHAGzfOhdOp8PXXdnw+gRYtQowe7WLEiCy2b4/0OyaTyuDBboqKJBo3DjFypJusrMS/3WHDKlm82KILsKIzUS6XxJw5DkaMcDNpUhpLl1pQ1XCmZcKECmz7/IXefruERx5JY/lyC6GQit8vYrGEjSGsVgW/P9JPp6bKXHNN/HlPvXr5eOyxNBYutHHttckZInTuHODDD83IctihrzaYTNCkSYhNm4zHyQOCKOIdNAjvoEGHuyXHLcY3+Qjn+jF7ooIigNyGQQbdUcDEm1okdYwFs7KrA6QwKru3WjntwnLMFpVt6228en9jKis0D58JBvtnvd6ArPohup9ThjM9iMOpxGSfLFaVIXdvi5LGNWvtxu5QmDm5CV63iRcnaMTM+2y9H7qjI/+6fy0NGnsJBkRW/pbJ5Gc6JHWNBgYGBgZHFqoKmzaZ6N69bo5ljzySFrOsoEDi5ZedvPde/LlGixZZad8+GFWTSBBg9Gh3TNbmzTdLmDQpjTVrzOzZIzFmTAW3356coxvAaacFueOOSj780MHevSKqStwsVHm5yKhRmXzzjW2fM164RtKuXSbeey9s1GCzwSOPRGenvN6wZfhZZ/n46KMUCgtF0tMVrrrKy4AB8ecQtWsXon59mf/7P2vSgVHHjkHefltg0yYTrVvXPoht1SqUsJ6UgcHRhvFNPsJJy4o/X8eZ8U8mOQrMfKUBM19pEL00KQmrwOSHmjLlyUZkZIe477X1tO4Y3fkGAuBMj+5cHU6Fnn32MnNyk4RH3rPDwX23dNXUPxIQTIau1sDAwOBoxOMRcLtjC6cmS0FB/PRFfAlcmLVrTfTunVy9pI4dQ0ybVsKQIVnYbGqtgqIqbr7ZzdChbvbskdiyRWL48EwqdXN709Jkli+3VAdFYQSWLzezebNE8+aaLH0QfvjBSq9efsrKRH75xUJRkcjw4ZX07euLkZvrEYRwBmjNmuQf75o1C9+vt22T6hQYNWigsHSpIXk3ODYwAqMjnF2brXQ7O9bmtGB7khMdddKfJKOf/e7n90kU7JSY+049RozfTmZOuDOVZSgvNpPbIFbmkOKUEQQFRRGSqncEsc5zgvauEIzuwLUyODXB8vBKjZNYzDpN516DYx2JHOp0r0MxdYzi1z/SqcESutfpt1V0c77kKCmdRran207RONHJoegbWsgSeV+17nUBc/R2/pDGvc687/9A2Ave59C4ze2Tzqmyiv9ND8rvIVDA1FYi5247gi3cNqukq1UU5V6nk8EpkTdB72aXSEoXK7mLHFPUaSq1sjtflCQuejvtMWJqFdXgemdwcPjjjz949913URSF3r17c/nllx/uJh3XFBeH+5ns7LoN5KWkxP/d5OfHf3gvLxfYu1eiVavkH+6DQViyxMKVVybn4hYPsxkaN5Zp1EimR48gCxeK1X1xRobMeef5eeedWOl7SYnE1q3RgVFRkcjLLzvp1cvPhAnpNGok06VLgB9+sPLbbxYeeijxnKcqWrUK8d13NkIh9htIAeTnh8+/a1fdgpucHJmyMpFAIPG8K4NaoqoIbjeq3eUpyA0AACAASURBVJ5Q3yju3En6gw8ibd+O6nDg6d/fkOAdAIzA6Ahn6lP16dDdTYsTIiNgW9dZeefxBjXsdej47pNc9my3cdXwPVgdMqt+S2XtH07Gv7K+uhBtFcWF+hGzw0+D+hWc2GEvGzdlsGlz/JpQBkngUhCeKIZNQVAh1ERC+m8GQnakQ/c9UUnoG3/1dDF5rUxRgUzuC/t3bTIw2B+KovD2229z3333kZ2dzX//+19OPvlkGjVqdLibdtxSWhoOjDIz6zYw0KpViD/+sEQN+LRsGWT8+PjBwfbt4f6madPkA7G1a0243SKnneavUxu1CEJ4rtCbb6awdGkqoujnppvcdOsW5Kuv7GzeHD3A2KhRiM6dowd2gkGBBg1kCgpEJEnl/vvD13rSSQHGjUvOsrlZM5lgUGDPHolGjfb/XtSrF/584lmGQ7iA7bJlFrp0CdK+fZxyG/vmYZWVieTlGYNA/xTr11+T+sILiAUFqE4nvvPPxzV+fFRtFcHjIfv66zGvXVu9zLR6NYLPh2fYsMPR7GMGIzA6wnGVmhhzRSuuGVVA0zZ+dm2xMOOlerjK4ldAPhT0ubaQ864swmaX2bPDymsTmvLwba0BEAQBUPnjp3S6nVOGdV8h18JdFqa93PiwtTkWlTFjltKt224yM/24XGbWrM1hwtO9CIYMSUBtER4qQvg98mChFsqE7ivF9Go42FQrFOTfg9VBURWBVTKB9TKW1sZ7bvDP2LBhA/Xr16devbB3/Omnn84vv/xiBEaHkeC+Z35zHW9XqiqQl6fQv7+HTZtSqF/fw+23V5KdHf/h2+sNBx4pKck74FXVOMrJqdsD/YIFVl5/3Ul5uUhurszYsRWMHOnmgQfsFBVF5kFde62b115zUloa7uvS0hQuv9wb49aXmqrQuXOQefPsiCK8/76Ds8/2s3Chlfr1kwv4qjJtXm9yA5GSBGazSkBn9qcoMGpUBj/+aKWkRCI9XeGUU/y89VZp1GdadZ8PGpU1/jHS1q2kjx+PaffuyLLt21GdTirvvLN6mWPKFEzr1kXv63LhmDXLCIz+IUZgdBTgcUlMmaS3z44t/poUCQrBhldpZF6i7nj79rtsaAHXj9mFMy38umUHLyedWc4bjzbhyw/rIQgKFw7YiyAqbPzbgd8rsnubjY9ea0jB7hTYJ0tSNZ78ycrqQCeRq8HXP6oQrE4uJwDnX7iZnj23YbWG25OaGuSkrrsZOuA33nq3W+zJ0EnyoiRxunbUILOLutQouZzuXJoRUkVXwVw7eipYE7veKSGNK53OvU7VrAtZohuZqEhsSI6+zuA+9zqxMIh9YzDWJXBzCNdKkUAHO+wJIrjU2G1cULbRhLmpDZtOSqeV1ll0UjqttM6rKQTr26RQPktBtEO968BRryYpXeJ1ouaDq8m9Lmqd7sMWa5Dg6Y9j8M8pKSkhOzuS9c3Ozmb9+vUx2y1YsIAFCxYAMGnSJHJycg7I+U0m0wE71qHgULTX6Qz/4rOy0sjJqb1dtyRJ2GwiTz1lxWSCUMgCxClwvg/bPllubm7y55Ok8D6NGqXXuo3ffy/w73+b9llzA5hZssTK1VcrvP66QE5ODooC770nsGKFRJcuKpKkkJmpMmKEwtlnW4HoDj4nB0aNgjffFBEEkbfftjNzpsrVVytMnqxgsez/M8vJCbfH4chM+posFjCZHOTkRNrz2msiX34pEQyGj1deLvLddzZefz2PiRMjfVhGRvjekJqaxaH6CRyrvzfpwQeRNEERgOj34/z+e2yPPBLZbsuW2GkCgNnjOSDvS0x73W7EyZMR1qxBufpq1PPPT/658xBwIL8PRmBkUCt6X1VSHRRVYU9Rue2BbaRlhmh9ooce50VGk8qKJf73cS4FO2xhL2fg4mt30+uyIqw2hd3bbbz+aEtK9+qe/g8wZovMLaNX0LZ9CQgCqamB6qCoCkmCdm2LDmo7jkVEtxyeW6THr0KFAkjQ0ATZEnh088JyRaROByb7WfCmQskMkMvCN+mSr1Sa/Vsg+/y61U8xOLrQD4BAVQY7mvPOO4/zzjuv+nVR0YH5zefk5BywYx0KDkV7KyosQA6lpRUUFdVeqhYIZBAMWigqKkqqvZWV4fMVF5dTVJRcQVmPxwpkU1CQ/D5VPP54FoWF0f1XICAwfbrEtm0KM2YUMXJkBl98YSO0bzAqI0Nm3LgKOnTwor2c99+38/HHDrxegSZNZMaOLaVfPwWHQ6229K7Y//QiAEpLbUAW5eWlFBUlN99KlhuwenWAnj1FyspEsrMVPB6ZYDD6MVFRBBYulCkqKq5eVl5uBzIpKyulqOifGEMlz7H6e8ssKCCeuFx2u6P2t/XsScbs2Yj+6N9VIDeX4gPwvmjbK23bRtaQIUjr1yOoKsKsWfjPOovSN988YgrPJvP+NmwYvz6nHiMwOlbQPhTUJXsEURkkVV9/Z18GKcUZv5O12FQuumYv2fUDUSn2jGyZa2/fwYK52QgyXHHTLq67fScpqeFzt+rgJr+5l7v6d64u5FpnY4ZQ/LYJwLhHltLjjILq33CcZyggbGYgBMLH0RszqFpjhqiaRokbJsSYL8SvfyTHGCxEDqLo12liSFl/fM1bIFi0y/XmCxqjCt3xlaAm61SDSUNVxog8M6k5FsTt0R20XN+Mq00aPl/4XNYLM7DNLEF07csa2gXkMx240pwQAF+M+YKmBlEN66xiCLlEoWSOF7kscl3BAoGtr5sQzjIhiEKsgYM2YyQmziaJmg/UrDuG1nChpqyTPkOkN3sw+OdkZ2dTXBx5WCsuLiYzM/MwtsjAuW+eqdtdt5HljAylWuqWDLm54fPt3Zu8NLfKhW3dOhPdu9cuMHK5Erft998F3nvPztdfR4IiCDvqvfyyk0GDvNVz6t98M4Wnn3ZWu9n99RcsXmzhtNMCBIMCeXkyF13ko3fv5ILLqrlCyc738XrB5xP44QcrHk+krfrBwyokKbr/qqwM7+N0Gpnwf4rnssuwLliA6I02Awm1bBn12te3L4Hp07EuXlz97BPKz8c1duwBb1Paffdh1sj2RI8H6/ffY/viC3yXXHLAz3e4OTJCPYOjhuLCxJYzGdkhLHESP/XyA6Skhn+4vS4rrg6Kqmja0sOF/fdUv+4zcBfPfPgHr3z6GxNeXUGDJnWrgVFFbj0P7U4ojRrYEISwflqL3y+yeHH+PzrXcYkoUHZdHqF6kYg4lGPCc3U2WCNvuv+KTCofyid0TgqhM1Pw3ZuHOiqxLKY2+JbKyIWxwUaoEII7D8gpDI5wWrZsye7duyksLCQUCrF48WJOPvnkw92s4xqnM/ybrKysW2CUmangcolJz13JywsPThQUJP9ok58v43AotbK3rqJRo8TZGI9H4LPP7FGFWavYvdvE3r2R5R9/bI+x+C4vF+na1c8rr5QyaJCHefPsDB+eySWX5NC7dy7DhmWya1f86ywoEDGZ1BqL02qpMsnQBkUQvieaTNH9qt2ucOml0XboLlf48636vA3qjv+ii/D16YOSng6AarEQ6NiR8scei95QFCl5/33KH30Ub58+uAcPpnjuXALdux/wNpl27IhZJvr92L744oCf60jAyBgZ1IopT+Qz/vUNZNeLzeoEg0LcFDCAza7iqaQ6QNIimaBFu3D9iEuu28WQO7fiTAsfv3lbDw0ar+TuQSfhdtVNcpWd48OZFv/OWrDHjt0RwlVh4ZefG/DpJ23qdI7jHf8paRSckELKd6UQVAlekIaaEdu9yG1s+MfV0yypfc2MeEj1BLABuvIloh2k2BqRBscgkiQxbNgwHn30URRF4dxzz6Vx4yPJ8OX4o8rQoDYZHC0NG4bvAzt2SDRIwog1NVUlN1dm9erk7xWiCKeeGuDLL+1MmFCRlL11FfffX8GqVWY2bIg9X3q6SlaCOoSKEh0sVlTEBjiqKrBxoxmn00PXrkGeeUbkxx8tBIPhbdesMbN1q8T8+UXVUrsqVq0y07x5KGmV0/btiS+6USMZSVIpKpLIzFTo08fL4MHRg5UFBRIZGYph1X0gEATKXnoJ04oV2L/8kmC7dvj69o3vu24y4Rk8GM/gwQe1Sao9/pOdkpd3UM97uDACo2ORusjqICljhr9/SeGOS9vz9Oy15DUKVne8ZcUS7gqBtDjKFUWGSpeEqqqU7LXQoGm0XMHnEVn8TRYoChdcUVAdFFXRuIWXy67bzvRXmtbcfM3/Wlnd5jUOCnfbyddlngp2ObjztrNwOkOUFNvw+0wIkrZIkP5CIv+KUeYL0Td97YRIWWfMIGgMDIQa5HhayZ0s62SNmibGSuQ0/4eEuMvDx4y/XXjb+LWQlGD0tShmncxOBO954Q5UMsmwTwlgNkd/nn4pcrF+TWdvMenkcprPwqKXukWtC6F2UBGbh1BW66RuJ5hwORwQBKv+GJrXphqMGUyixohBid5OSmDSANESPL0xg35bgwPDSSedxEknnXS4m2GwD4dDJTNTZseOugVGVfWI1q83ccop+99eEKBbtwC//Va7J/RBgzwMH57Fd99ZueCC5OdCNWyo8OmnRYwalcFPP1mrgxZJUunZU2HnzviPWJKkRrng1asns3Vr9LYmk0KDBjKLF1vYtk3il1/M1cevYt06MzNmOLjxxsi9TVHg998tXHxx8nWZ1q9P/Ch48cVexoxxsXu3RP36CnZ7bFZo+3aJxo0PzCCXQZhQp064OnU63M0AwHvJJZjWrUP0RL5nocaNqRw58jC26uBhSOkMak3RHivDzu7IGw81Ytl3afwwP4OHbm6JzRE/ja4iENwnJ5j6TD6FOzWSqwCsWJbGLwszGPnARpq1jS+by29a9+J7fp+Jrz5pTHlp5LwVZWa++bwJrnIbu3c68fuMMYKjGUEQsD2RjnSGBbGRiNhUxNLXQvqDRo0kA4PDSaNGcnV9odpSNf+nNhmgk08OsGWLiT17kn+8Oe88H3l5Mm+84Uw4/zQRGRkq06aVMnNmMRdf7KVXLx8PPFDOlCmJr1sUBRya++W//+2KKlorSSpNm8rs3Gli6tQUXnjBidsdez2KIsRIANeuNVFWJtKtW/LzpdatM2GzqbRrF6BqiFEUVbp0CXDHHZXYbNC8uRw3KALYutVE48bGYM+xivvWW3GNHk2gQwdCTZviP/VUSl5+2cgYGRhoUWSBT96pxyfvhGVRoqRiscXvNOWggNWu4HVLrPg5nX8POIEBo3aRlhHil4UZLPi0PlfetIsL+hditsQeI+ATWLowC6tNxh8UIcb0ef/M+aAly5flcNmALQiiwLzZzdi4NtP4BRxDCBkitsfTozJNgmiMYhoYHE7atAnx4491cx1NS1Np1y7I0qXJZ4DOP9/PI4/AvHl2br7ZndQ+ZjPccYeL8eMzmDvXzpVX1n4grkePID16ROoWiaIDvz/+vcpsVqmoEKuzRqefHmDOnGJeeslJUZFI794+rrnGi8kEN9yQxbZt8QNDh0Phkkt8eL3hazCZ4JNP7EiSmrRRA8DSpRZOPVXljTeKefvtFFavNnHSSQGuv94TI9PT4/PBli0S/frVffDS4AhHEHDffjvu228/3C05JBiPhcc6+uGvA+xYV+VWp8gC2zfYOKFbbManYKcVr0uozk8W7LTy4r3NNU1SOPXcEqxxAitFDmeoBt66lWH3bKas2MLstxrx49e5cdqscVoTIyN1VVe8ebWT5yecSLUVEHJsiKUkHi7USuS0DnUxI4yqVmYXvVIri9PWP4pxr9PWKtK7xkWtUxOu08rlYuokRcnsdMfQrtPI5RTdvTlKZhfSyexMid3sJFOkMUFTZJ3fFL1dlFzOpJfBKXG3C6/TutlFS+60aCVy+nVmjfTNpDmeWSeJ067TO89pj2+40hkcr3TsGOTjjx3s3StWu8bVhjPO8DNtmgO/P7lBjlatQnTuHGD2bEfSgRHA9dd7mD3bwYQJaZxzjo+srH/2G01JgZYtQ/zxR2zWKD9fjjFGaNRI5oknyqOWud2xGaEqJEmha9cATz2Vyq5dEjabyimnBPjxRytnneVP+r0uLRX4+28z998vk5KicvvtlUleYZj1680oikD79kZ1V4NjA0NKZ3DAeGhEKypKxahAwe0SmfFKA+JlecwWhVYnusmuF0AQ49+EigvNZOT4adbGR17DAG06VnLL+I00aZX8Da8mMrN8jHngd556/UceevZnOp9ceECOa2BgYGAAnTqFH5iXL6+bec6ZZ/rx+UT+7/+SVwr07+/h77/NLFuWfKZJkuDJJ8soLxe57bYsPJ5/XrzyjTdKyciQ0c6AdToVbr65MqExgqLAnDk2Zs2y89NPFny++NtdfrmX3bslfv3Vyq5dJjZtMvPRRw527jRx9dXJZ28WLbKiqgLnnJM4ECwvFxg7Np3LLstmwIAsPv88kgFcsSL8uZ5wghEYGRwbGBmj441/aMzQupObYffuJrteEE+lxFczsvlqejh7U1poYkCXzlx/1y66nuWicKeVD1+sz5a1DkABRSQv30+/Gwto3t5LgyZ+snKDuF0SPo+ILGuSOYDfK+D3SuQ2iO5ws/OCXD18O8+Oa6drY/zrUXUDjVWZH3tKiEeeX0azVpERshaty3n5yY78vKhBTCinLSAZVXFalzISFW12TTdamMBwQYgxaYi/HYSzaJHtolsZlZHSXLfWUAFA0azT11ASNW+3rHmu0KvStGYM+qyTakrStMEUubigFN0dBTRZIr9JZ16gyQSZpOh1WhOHmjJLWkMEi26dKYH5gt4EQowyX6ghm6TLEOnNHgwMjlW6dAlgtyv88EPtjA2qOPtsP+npCh9+KJKsr8bAgV5eeCGVp55KZdas4v3vsI8TTgjxzDNl3HVXBoMGZTF1agnp6YkDho0bJR54IJ0dOyRSUlQuucTLbbe5q2+t+fkKK1YU8MwzThYtspKfrzBqlItOnaI709WrTXzwgYOKCoElS6zVLn6pqQqhUOx9rVGjEG3ahJgzxxG1XFUFTCaViy5KPjCaO9dB/foyp52mUlISuz4YhGuvzebPPyM3g5UrLRQVVXDjjR6WLLGQlyfTvLnRpxkcGxiBkUHSZNULMv71rVGuco1a+JCDAt/OygFAVUXee7YR7z0bu3+fawu57s5d5NSPDnRsDgVFgb27LdgcMumZMiWFZn5ZmE6zdvHNGBwp0Z1wWkYQBYHKiuRHJS+/bktUUASQmR3g8oGb+XlREt6wBgYGBgY1YrWG7bAXLarbPCOrFS65xMvcuQ4mTBBISdm/xM1uV/nXvyp58MF0fvjBQs+eyRsRXH21F7tdZdSoTPr3z2HKlGIaNoyVpblcAkOHZrFxY+Ses369CVmG0aMjigZJgrFjKxk7Nr5E7fHHU5k2zUFpaazkrrRU4qST/AQCYbmbqkLTpjKjR7vYssWEGmcw0OlUYsY83W4Bt1sgNzd6XUmJwPffWxk+3I0oxs+uffKJjVWrou+r5eUiM2Y4GDLEw5IlVk47zV+rcVYDgyMZQ0pnkDQD/lUQY7WdmqFw0bX7H5GzWBWuHLEnJiiqQhTBUylxT//2TLq9JXdd2YHnx7Vk+wZHzLYBv8DP34ULgzZo4uXJD/7k5U9/55W5v/HQ5JWkZSSX0m/SIr4cz5lau+rnBgYGBgaJOfdcPxs2mNmwoW5jsf37e/B4BGbNSt5lcvBgN40ahbj33gy8tfQF6NvXx5QpJWzeLHHBBbl8+qktZi7pO++ksHFj9PV4PCKff558G3fuFPnoo/hBURXbtpmYPLmUE08M8vTTZXz77V6uvdbLNdd44tZJats2hHVfDOrzwW23ZdC7dy4XXJBL3745LF4cCYCmTUshFBK46qrERdR//91CMBgb9ZSWivz1l4k9eyTOPNO4ZxocOxgZI4MoRFGlftMArlIJV1n01yO7XvyAw+6IjKbFM2YAaHmih/qNa5ZRiILK9o0Otm+sCoZUJj/UlMYtvDRr78FqVamskFixJI0Fs3NACHHv86toeUKkU89tEGDcM6u496bOCc9T1cI1f6ZxRq8CdAouykssEJJjjBiizRcSyOoAZM1NTjfQmKj+kd6kIcqYoUbzBZ0MTmu+oDVR0Jk0aNcJeomcZuAwyqRBl4wTNK9FfS0kTbX0GNMG7TElrbmDro0aA4eQTkoX0JhrmHTrfJoPtCbJnXadXuoWtZ9GIieJOtmeRi4nxkjpNAYOgmG+YHD80revlwcfTOPTT+2MGeOq9f4nnxykRw+FN95wcv31nijJdSJsNnj66TIGDszhiSfSmDCholbnPOssP2+9VcLDD6czcmQWb7wR4N57KzjjjHAQsHmziXhzZysrkx9v/uILG4WFNV+M2y1w662ZDB3qpn9/b/XcpJYtZQYO9DBzpoPiYglQadRI5tFHIwYO//lPBvPm2avbuXevxD33ZPDVV3sxm1XeeiuFc8/1ccIJiY0tzjrLz0cfOfD5oq8rN1fh88/DDngXXZRgIpSBwVGIkTEyqKZn31Je+XotL3y2jle/Xsv41zdjtkYe6Jb/kEooTmxUsDP8JO1MD9H3+r2c378Iqy36QbCsyITHVfMNYOOq2OxQZYWJu6/uwNNjWjLj1QY8MLwdD49qh6oKnHhKOY1axA4FNmnlIbve/rXs8z9qxIbVaVHL9u6xMm1y6/3ua2BgYGCQHPXrK3TvHmD+/NjMSzIIAtx9d7gI6uef78c/WkPPngGGDHHz1lspLFyYvJTv++8tXHRRLiNHZlJSIvw/e2ceZ1P9//Hnufvc2TczZhjDGCmiRKQi0Z4SFaFo0VcoKYq2b1Lp+237fokvURIpvokQLX6S4ltZImt2xlhnX+6+/P64zJx77jJ3NtfyeT4ePZp7zzmf+77HzDnn/Xm/Pq83rVrZOHFCzf33J/HAAwksXhxBt24WIiJ8JXZpaZWTJbm5aqZPN7J8uYEzPccLCyVmzzaybJmBxo2daLXB3eOyshysWJFH375mH8OGF18sZfhwj0Sve3crq1ef4tJLPR/kcsHGjVqUyduhQ571TPPnG8nLU1ccH4hbbrFy1VV2JNlkTnKyk7/9rYxlyyLo3Nnm47AnEJzPiIqRAIDEVBuP//0oDdI9mU9MvJOktGJMZUd4f3QGAN9+nsh1txfRumMZutP3mEN79Ex/NZ3uffJ56NljpGbYcLlg+Os5/PxNPFNezsBqVnPskIGDuyNoe43vRbi0SM3e7ZF88HJTn23gaWL3y/JEflmeiPzOEBntRKP1vctqtS4MEVUvBLXb1LwwtD33P3KArJYllBZr+fyjbHIPR1V5rEAgEAhCp1cvM+PGxbF5s5Yrr6y+g1nPnm4uvdTOm2/GcPPNFiJCVKy9+GIJv/+u4/HH4/nyy/wKl7xAFBVJvPBCHIcPVz4enTgBd9xhol07Ox9+GMXw4QZUKjcxMS5sNhdOp+e+1KiRnRdfLMbthrFj1cyZk0Renqeao9VCUpKTvDw1drtEr14mJk0qomlTJ7t3yzOe0+ZAEW6aN3cwfXphwPU7v/+u4623YujY0cpHHxWglVXnXS78SuAAjh9XsXixkU6drHTqFFwGp1LBnDn5zJgRxf/+pyMqys2wYaWUlKg4eFDD009XvwIoEJzLiMToYkY2dXfPo6cqkqIzqFRwabvKdThOh8SLA7O4oVchHbqVcGS/nkUzGuB0wICnPUnRmeMiIt3cdF8BTVqYGXPvJTRvU05yQxtut2f2z2GHw3sMzHk/jdwDRg7vOXOXc4NMgud2KyVmsgrW2liO5xhIz/Qu4588qid3v4FK0VzgJMlcKjH737KETK0GPDNuUiA/VQLL6jwxBpbZefU/ksvnFO518v5EklbhWCeX2SllcAF6F7kUp8DLsU45RgD5nM9+9lBldsrjZK59siuQ2650tvPvXgfeUj2nQiKnUldus8m2qRX7yV9rFBI5uexOLp/z2U+l2M/hht02MEhomukqnB+VEjwhpRNcbNxzj5nXXovhs8+MXHllcdUHKFCrYfz4Yu6/P4lp06IYNSq0fjuRkW7mzs3nrruSePDBBL7+Oo/MzMD3hHnzjF5J0Rm2b9cxdWoRQ4aUs2WLllWrDPzf/+n5889K7fGRI1r69EnGZvNM6FUiYbfDsWNqGjRwMXNmPu3aOZg6NZKCgjPXOU/ydMMNFgYNKicuzs0VV9gDJkV796p5+OEEGjVyMnOmd1IEnmavmZkOjhzx/i5JSU5KS1UUFKh49dWSkEwT9HoYMaKMESMq33v00XgSEpz07CkauwrCgMuF9s8/QZKwX355nQ4tEiMBOoOL7Db+F1+qNd4PcC6nxKqFCaxamFDx3jW3FFUkRXIkCbLbmLlnyHGuu62YtMzKfTRaMEa72PprNGUlofeakGM1q/lsUiMefDqHhhlWnA44etjAfyY0xZ/2OxjGKAdPjN1N9uUlxCfYMJWr2flnHJ9MvoRTx0NfTCu4eJH+Z0I1owiOOkArQRMN/D0RGojLrEAQHe2mVy8zixdH8MILJTVqoHrttTbuuMPMBx9EcffdZpo1C80iOjXVxbx5BfTqlUjfvol8/nm+32Pz8lT8/LN/yZ3D4anC6HTQrp2ddu3sjB5dSlGRRE6OhpwcNTk5ak6eVLN2rY6tW/3d1yROnVLx2296MjJczJoVebqi5Nlmt3vWAXXrFryKs2WLlsGDE9BoPElfoHP59tvFPPqoij17tNjtEqmpDm66ycLnn0fSv7+Jyy+vXuXO7YYZMyL58ksD27frMBjc9O2byJ13mnnsMZNwphOcFTR//EHcmDFoDhwAwNGsGcyeDWlpdTN+nYwiOG/JamVi7NTDpDf1vybn0J6q9dxlxWrsNgl1hO/FWaWC1leXk5zmewFObWyj001FrFzYoPqBn+bHJcn8tiqeLnfkY7Oq+OXbRGxWNVIIi3MrcTP+gy20alc5ixkd6yAl7QTZl5by/ONXU5hXfH2k/QAAIABJREFUM6tZwUWC2YXqP0VIR0+Xxxxu2GWHtwrgvZr/fgsEFxJDhpTzxRdGPvwwirFjaybBGj++mLVrG/DUU/EsWpTnUykJRPPmDubNK2DAgATuuSeJmTML6dChMgGZMyeCSZOiOXpUg0dt4P2Un5XlQOcn14mLcxMXZ/dKMt5+OypAYuTpNfTHH56k4tgx30ewo0fV5OWpSEryv27n228NDB8eR1KSizlzCmjSJHBymJHhZMWKPH74wcCxYypuvNHCQw8lkpTkYuzY6plRAEyaFMWUKVGUl3uqXBaLxIYNerZs0bJnj5Z//rP6lUCBoFo4ncSPGYN2586Kt3Q7duAaMgSWLatef84AiMToImf4m7lkZPsmRWaTxMGdEfx7TEaVY2z7LYojew00v9x/Sf14jo4mLXxda2xWKM7XVDSPrcAVWMLmlm06I6szlaj49vNk2fojN26n0/9BnLndubmkTSnJaVbsNhVNL/F/k05vYqL/kD188MZlSA6ZJkwmifNxpXMFkdnJX8u+p0opuXPK5HJKmZ1cqqeRFNv8u9m5lO518gavSkc5h8xRzktyp9ivhjI7leyqI9/mVivHl51Hxfd0yyqZboXkUS6zk2SyOqWznUpVuU0VRGYn308piTvzWv9jMbpjvs5OrhwnZSe1qOK941e64AkEFwOXXOLgzjstfPxxJI8/XlajqlHDhi7eequIDRt0uKq55r9NGzuLFuUzaFAC992XyPjxxTz0kInSUompU88kRSBPigwGF9nZDt55pyjkz3n0UROLF0dz8KD/h7SMDAeJiS40GrdPA1edzk2En0lGtxumT4/k9ddjuOIKO7NmFZCcXPUJ0Gjgtts899/nn49l/34NX3wRuMokx+mEtWt1mEwSXbvaWLLEUJEUybHbVaxeraegQCWMGAT1inbLFtT79vm8L+3di2bHDhytWtX6M0RidBETFesgpZH/kv338xOY+lIjQpGkud0Sr/+tCX//+AAZ2VYvK9Uj+/XMeTeNpFS7zxqmnL0RbPwptjZfoUZEx9r5+9RtZLYoxxjpoqRQjTEy8MW8QUNhRSoQCAR1wdNPl7JsmYF//zua8eOrX7UA6NnTQs+eNbsuN2/u4JtvTvHkk/G88EIca9bo6dbN4nddkdHoZOLEYnr3tvg4wgUjIcHFlCkOhg3z9CGSN2LNyrIzfHg5UVEusrIc/PWX98xR27Z2nya2Bw+qGTculjVrDNx+u5lJk4r8Jk/BWLAggrlzIxk2rDSkvkM7d2oYMSKeffvUOJ0SGRkOSkoCn4RTp1QcOqQWiZGg/vFXFZKkOqkWgUiMLmrsNhV2m++FzuGAvzYZT3sXnL74VvELd+yQgaHdL+W6Owq5a3AehggXece1zJiQRnGemreGN2H0vw7TvLUZtcbN0YN63h+TgdMBkjIEeQVJuVFejVFskhszeN3FFNOKT766m8uurKwQxcQ7cbkIeOM7dlgPDgduVYAqkTuwbs/XfEFuzBC435Ekr4i4lBUpWRxOpTGD7GdZlcinz5A2cC8kL/MF2c9Oh3ccKlmeKx8PwCWvCimrSbLqj9cYiquR1zZ1kIqR4jivbWr5z4rKktp/ZQnAIasYyX8FVYqKkXS6mmS6MomIlAK0x70fOGxpekr1RlQWxXEIBBcnLVs66N/fxKxZkTzwgImWLQP30Kkv4uLczJ5dwIcfRvKPf8Swdq0ejcaFw+F9jYiOhq5dbdVKis7Qo4ebtWvz+OorA198YaS8XEVampNXXimpSB6mTy9g7Ng4Dh1So9O5advWznvvVcrR7HaYNi2Kf/0rGq3WzRtvFPHQQ6Zqx/Pzzzqeey6O666z8txzVUsY3W4YPTqOXbsqk7aDB7U+PePkpKS4aNr07P9bCi4u7FdcgSMrC92OHV7vu5s3x3HppXXyGSIxuoixmlX8tTmChk28H+Zy9+lZszSuRmP+8k08v3wT7/O+xaTm9b9lodW7UKvdWEzVWgRUpzTJ9jWaUKk8Tnkahezr8H4jn/3Hv424QHAGt0FF4SMNifv0OJoTNtBK2BvpKXwqPdyhCQTnHGPHlvDNNxH8/e+xfPFFfp0v2rfZwOGQMBoDV1VUKhg6tJyuXa28+GIsv/3mu470ssvsIcnVAiFJ0KePhT59/Fe3srOdLFyYT3m5hEbjRn86BKcTVqww8N570fz1l5bbbzfz2mvFNGxY/Vi2bdMwZEgCzZo5+PBDX/c6fxw96jGSUOJwSEiS26sCBh654c03W4iLExJhQT2jUlH0zjvEjR6N5uBB4LT5wsyZomIkqBveeyYD3IdpcYUZrc7FiSM6Jo9t5LeSVBfYrSqq38EidCTJjT7ChcUiEWheXmldfYbfV8ej1kk0bmqiKF/Dwb3RzJ3SjJLCmrnmCS4urO2iOdE2Ct1+My6DCleGrs4u1ALBhURCgpsxY0p48cU4FiyIoG/furN8ttlgwoQYjh/3WGP//e/Ffk0TznDppQ4WLsxn9mwjr70Wi9UqodW6yc628/bboa8rqg1npHNms8T8+RHMmBHFwYMaMjMdfPxxAbfcUjPZ4JYtWvr3TyQqys2cOfnExrpPW4mDIYivkkbj9pLEVyJx//3laLWese12j+Nfr15m7r1X2HYLzg6Otm3J++47NKerRo5WrUhKToa8vDoZXyRGFzlWs4qJwzLR6l1otW5MZQEqOXJJWE0f9gJI5JRSMel0H6OISCcPPHWE9KYWDuwy8tWMFExl8hX/3sfd9chxbrn/FNGxDkqKtCyb24Bvv0jx0cjt2hxN4yzvLuJWi8SGNQl8+1+Z3WPFncG3H5L8k4OeDZXiuwXofyQ5lQYUcpmdd/wqWY8jt7LHkUxap5L97NIE3s+nF5L8FMuSSJ9+RJogMju5lM6ulPHJx/B/DHjL53y2yfsfKaV08j5YXrI6xX4BJHeeMeT6ucofnWrl76rM3OG0/M7ayPMUJsmeZSRJeZzQ4Qsubh56yMSSJRG8+mos119vJS2t9n8TBQUq5s0zYrVKfPhhIS+8EMuOHVquuCL4dJwkweDBJvr3NzFnjpE5cyLZsUPHDTc0oHdvM717e5q7+k8WaofNBr/8omfp0gi+/dZASYmKK6+08cILBdx6q8XvZ9rt8PHHRn77TU9qqpMRI8p8zt/69VoefDCRuDgXCxbkEx3t5pFH4tm5U4vLBc2aOfjXv4pISfE97ykpLpo3d3DypPeHq9VurrvOSu/eYt2tIMyoVDhat66XoUViJABOV3L8O3aHhbsHn+CRcbnoDG4kCa69tZjrbi1k3MCWFJ7y1QJ0vqWAgSNziY7zPMknp9kZPDqHY4f0bPktwWvfKa82IzLGQfsuhehPj683uHn0+QMkN7QxZ1Lm2fiKAoFAcNGiUsF77xXRo0cyzz4bx9y5BbVKPNxuz1oak0lixIgyJAkMBjdr1+qrTIzOoNN5HOUeecTEhg065swxMn++kU8/jSQuzkWXLla6dLFy5ZU2mjXzb99dFTYbbN+u5Y8/dGzcqGX1agNFRSpiYlzccouFfv1MdOxoCzj/uHSpntGj4ykrq1RFrFxpYPLkIjp29MjilywxMGpUHKmpLhYsyCM93cUDDySwZk1lmejIEQ2PPprA0qV5fj9r6tRCnngino0btdhsns9yOiXGjo3jr7/KGTeuZnbrAsG5jkiMBOccTVuaePj5XPQK153MlhYeGZvDu8828znm9gGnKpKiM8QmOOn18AmfxMhuU/HRPzK44poiDLLPiIxycf1tp1jwYWOslvCtgRIIBIKLgcxMJ+PHl/Dcc3G89140Y8bU/GHb7YY9e7RkZzvIyHBy8qSK33/X8fHHBV777dypwWyWaNcucLIkSdChg40OHWxMmFDMTz/pWbXKwJo1epYs8TT81mjcZGU5yM52kJLiJD7eRVyci5gYN04n6PUqCgs9pgu5uWqOHFFz9Kia/fs1WK2eTCQ11Un37hZ69jTTpYu1Yo1RII4fVzFuXBxlZd4KgNxcDf/4RzQLFuTzxhsxfPhhFFddZePjjwtISnKRm6tixw7fCcW//tLwxx9av+ciOdnF228XccMNych1EeXlKpYtM/DUU2U+7nkCwYWASIwE1aceZXUAvYecwBDggtsww1J5rOw4nc6/DEOnd/m40hmi3Lw2YxdRMb7HxCXaSWpo4ch+I5JTsRjJ7f+7upV9mOQSOaV9UKD+R4r9ArrXgZeDnY/rndy1Ty41dCqkXLLXPr2Q5HI8uUNdEOe5YDI7eV8hAJfd/35KSVwwuZxcZucjpZOrEDX+3weF5E4ppVPjf5vPGIH7KckleEpJpc9rgeAipX9/E5s2afnXv6Jp08Ze4/U0paUSq1bpefJJT3L1wQdRdOlirZCKlZVJzJ1rZN8+DUePqsnIcPLSSyVERLiDurzFxrq56y4Ld91lwe2GvXs1bN+uZdcuDbt2adm2TctPP+kpLfU3iMfEKC7ORVqak0aNnHTtaqVdOxtXXmmrtnxwxoxICgv9T9rl5qrp1y+R//1Pz8MPl/HKKyUVFa2SEhVms+/9y2SSyM/3/+WPHFFz551JPk59AHl5KnJy1GFxFBQI6huRGAnOOXT6wDeLQGugdv8ZSZtryrzeczph2/pon32HvHCQxln+b76GCCfPv7uLNcuTSWxgJ7WxmSP7jcyf3piyMmHCIBAIBHWJJMEbbxSzc6eWp56KY+HCPFq3rv4Dt9Mp0bKlg59+0hMV5eaXX/QsWVK5GHvGjEisVolRo0pp0MDF6NFxOJ2eOamyMomoqKonKyQJsrM9VSIldjsUFakoKZHQaCAlJZ6ysgIiItx1VllRVorkHDumJi9PxaRJhfTp422E0KKFg8aNneza5X18kyZOOnf239Pob3+Lo6jI//3WYpF47LF42re38fDD5cyYEUVZmcS111oZPNgUkvOdQHCuIhIjQe1QVCxqVEFSVFy+X5DI1d2LiIj03s1qlvjv1FS/x336Thot2pZz2VVlaLSeIpHFpGLfdoNX9QWgWcvygKFotND8MhNNLzlUoXfv2K2QdtcVMnbQ5ZQUeq74Xt/SpxGTPMTA5gteZYkQTRpAYdTgUlaTvJwCKn5UKypGLnnFSNHjSG7uIK/MSIqZQ5UmSDVJVmVxaZSVJv/VpKAVI3XgapVPQU1eyVKFWFlSVoLkFSPZGPgUAAP31fJ6rVb+nSAQCE5jMMDMmQXcfXcSAwcmsnhxHpmZAexDA5CQ4GLQoHLeeSeaa6+1Mm5cSUWyk5OjZtMmHePHF5OW5qKgQEVcnIucHDWpqS5efz2GjAwHQ4eWERFRs++g1XrkZ8nJntdJSZCXV7cmK4MGlbNsmcFvwpKa6mTu3AJatPBN2tRqeO65El56KZajRz0XQpXKjV7vwmqVfBK33bs1/Pln4IlAh0PFgQMqDhzQsGRJBFar52K3apWBVasMzJtXIAw5Bect9ePJLLjoaZxt5uUZ+3nnqz28OO0AaZmhyyM2rI7hh/8mUXS6xO90Qlmxin88mcnW330rQAA2q4p92yMqLsYqFURGuxg2PoeEBt4zYg571Vds5SLgppeYGPjkoZC/g0AgEAhCJy3Nxbx5BdjtEgMGJHLqVPUfT9q2tTNnTgFDh5bTsKGT+fM9Wc6GDTratbORkeFJtnbu1JCbqyY720FioouxY0swmST+9rcEjhypn/WlhYUSY8bEcs89iQwcmMDPP1dfgXDZZQ4GDy4nMfFM8uMG3Nx8s5lffz3pNyk6w803W4mLq0zUXC6J3bt1DB3q3bMwN1dN//6JISY2UkVSBJ6q3fr1OlaurGKxlEBwDiMSI0Gd07i5hdfn7Oe620u4vGM5XXoW88Zn+2mQHqrtncSUlzIYcdtlvPtsBs/0akGfVm1Z+61v41g5bTuXoVZUBVIa2ej92HGv99Z9l4DFVP3prBvvOkmfR45U+ziBQCAQVE12toM5c/I5cULF/fcncuJEzR9RWrd2cNVVnkmxtDQndrtH4nbkiJqvv46gRw8LOh2UlEikpLh48cVSbDbJ71qc2mI2Q79+icybF8nvv+v58UcDw4fHs3RpkGZCfigullCpKpORa6+1smHDCWbNKgy6Tgpg0yYt+/f7ioR27dKSm+tJBg8eVNOnTyLl5RI33WT2aTMQCmazimeeiWPy5EgfQYlAcD4gpHSCusXtZtCYY6Q29na5Scu0MWjMcd5+ukkIY3hmtU7lavh+fpJiU+ArrUbjX7YQHWP36nn01UcpxCXZuOamIozRDtwuifgku09SpSQy2sUDww7jsMHXcxtVvB/05qGQyHm9qolJA3gZNfjK7GT9ieQyNWUvJLksTtHjyCU7V5Ja3hdJIceT9VDykeN5yeW8w/eSz6mD7KcObOCg8jJYkAJuC2TEoNwml9wpt8klcb4GDoG3yaed3IrpV6XsTiAQeGjXzs6nnxYwaFACvXsnMX9+Po0aVU9Wd4bmzT3H6fVufv1Vx6uvxqBWQ3q6s6Kp7JdfGtmwQUvjxk6aNPGsHzp+XMWSJRFs3qylf38T113nfx1OqHz6aSTbt3svvMnPVzNzZhQ9e1atpti7V8OcOUYWLDBSUqLittvMjBpVSqtWoa/FKipSnbbd9sZmkygrk9i5E/r0ScJikZg/P5+WLe2MHg3r1+uwWCSSk50cP64iP7/qx8aCAjWTJ0ejVsOwYYGl6wLBuYi4PQvqnNhE/xfr+JTQeknUlNwDvrNvpcVqln+erHhX4uN/ZPDEba0ZfsflPNy1DZvXxVRUkex2sPu5gYAnObqhZ910VxYIBAKBL5072/j883wKClTcc08ie/fWbg73iivszJxZQFqak3vuMTFyZKVRzyOPlJOS4qKwUMVrrxVjMklMmBBDeblE375m5syJDOjcFipbt2px+3E1LSoKXJ0qLZX46qsI7r03ka5dGzB7diRdu1r57ruTzJxZWK2kCDzntEkT32MaNXJSXi7Ro4cWpxMWLsyjTRs7Oh1MmlTEqlWn+OGHU3z3XR6PPmoiMbEySY2JcRIR4X9CsrxcxdKlNVywJRCEEVExEtQ5+cf9W9KcPFI3rm6S5Oba24vofHMRRw7oWfxRCqZSNZNfaEJSwz00aWFGp4fCUxpWL4nnrz/8r0ty2FUU5XlueC8NvoSOPYro2K2IvdsiObQngnGT95LoJ5kzGIPPXra/Pp8+D+dgjHJQcErPx+83I2d/VO2/uEAgEFwktG9v57//zWPAgETuuiuJL75w0qZNzcdLSHDz+OOe6oXJJDF2bCw33WShZ08LWVkO9u7VYLdLLF9uICrKzahRnuTpgw+iOHRITWJizY0UOnSwsXRpBA5FZT0hwXvMkydVrFxpYMUKA7/8osdmk8jIcDBuXAl9+5pITq46BrsdvvwygnXr9LRqZeehh0wYjW4iIjzf6e23o8nJ0SBJbjIzHdx4o4V7700iNRXmzs2jWTPv+1tEhOdYgJEjy7jzTjOzZkUSGenm0UfL+eknPa+/HkNenu/aLFMIknWTSeLVV2PYulWLRuPmxhutjBxZVqU0UCCoL0RiJKhzZk1MpUVbE+nNKuUHh/fomP3PhtUfTOFYp1JLjP9kH22vKa1oANvt7kJefSSL3AMGnrqjJR1vKqJRMxu/rIjn+GE9Xo1/wNu57czHAL9+F8evP1SuY/rwjQye/ed+dAZvGdfJXL1XbyR5/5qrrivg6Ql7SEg+k1CV0bhZOc892JaCU3ovSR8ymVrI7nXgJc8LKrOTxSgpbNckr/0UMjvZa7c8Rqf3fpKj8rNVSuc5mQxOpZTqefUu8u+A59mv8mcf57kgMjt3ICmdOohcLkSJnM9+MomcjyudfF8fxzohvhcIqqJ1awfLluUxaFACd96p4eWXI3nssfJaO54ZjW4efLCcDz+MYuFCIzExLnr29Ejr1q/XcccdHnnbL7/oSEx0BW0GGwoPPGDiv/818scflZODyclO7rrLxLx5Rtav17F+vY4DBzwXvYwMj8nCbbdZaN/eFnKSYDZD//6JbNqkw+GQWLTIzZdfGvnii3ySklz06WOme3cLixZFoNXCvn0apkyJ5pprrHz5pQRULVnMynLy+uslFa/vu89MVJSLESPisVi8A01PDz6e2w2DBiWwbl2lWcO2bTpOnVLz5pvFoX1pgaCOEYmRoM45cUTPc/c146ExJ0hKtXPiiI5P306lMK/2zQ1u6FXAFdeWoJOZ3mRkW/nb34/wyuDmuFwS//suHklVuzunwehk2+/R/LE2hradSzBEuHG54Mj+CKa/nhnwuN6P5MqSIg9pGRYeGHaIKeNb1ComgUDgnzlz5rBx40Y0Gg0pKSkMGzaMyMjIqg8UnPM0buzk66/zeO65FF59NZY//tDyj38UEx1du8mFDh3sdOhQyIYNWuLjXWRlOVm3Tse+fRpuuMFjFPTttwa6dbPgcIBG8bT09NNx6HRuGjVyEhfnIjraTVSUC6PRc6+IjJQoKNBjsUjk5ano3NmKzQZHj6or3nvpJc9EXEKCk/btbfTvb6JrVwuXXeaoUfI3bVoUv/+u40w/ALdbYudOLePHxzB5chEAcXFueva0MGpUHKtWGRg4sJwJE4pJSkoirwYq8bw8FddcY6NrVytr1ugxmz3JUbNmdsaPD57cbNyo5c8/vU+szSaxerUOk0nCaBQTSIKzj0iMBPVCXKKT5HQbSal2omKddOlZxNezGtR63E49ir2SojM0aFS7xbFn0GhdPPPP/Vzargy9wUXeCS3/ndaQhk2s5B40suTTVExlmoB9aCKj/c+QJTWwcXW3PBIb2Pnf/yVTlC+axQoEdUWbNm3o378/arWauXPnsmjRIgYOHBjusAR1RHS0m/nzHYwfb+Htt6PZvFnHBx8U1rqSAx7J3hkSE100b+5ZhzN7tpHSUhVdulh9kiK3Gw4c0LB/v5qCgmD23okVP0mSG63W0+pPr3dzySU2Hn20nDZt7GRlOeuk78+mTZVJkZyDByu/wHffGRgzJpayMhVvvlnEoEGmGn3Wtm0aXnghlpwcDVqtm0svdfDee0WsXGkgM9PBY4+VExMTPLHZtUtLmZ+m7SUlKk6cULF/v4YjR9TccouF1NS67QklEARCJEaCOicuyc6LHx4iLbMyWUnPsuJywdJPThshSBLgpk3nMm68p5Djh3Us+SQZU2nwHhKnjvmvOpnLvUv4cmmaT/VILs9TNGcdPv4gXXsWVPQxSmjgIDrWyVN3t6Ks+Ewyo7hAy+Rs+Sd843O5ILt1CVdeW4hO56bf44dYNq8h//24aeVOIbrXeUIOYoWm9i+zU7rXybUZklJaKGsMK6llkkGFlC6ozM5rmyJ+h1xKJ3O9UzRB9ZLcKWV2av/Odp5t+N3mUowf1G0uQIPXoJK7YA1efbaJ7od1Sdu2bSt+btGiBb/++msYoxHUByoVPPlkGZ06WRk+PJ67705i6NAynnmmtMZNWZW0aOEgPt5F796JNG/uYPjwMho29H0glyT4+mtPecVsliguligrU1FaKmEySajVkJgYS1lZEXq9m8REF48/Hs+GDZ5ZPatV4s8/daxc6aR3b19XuiNH1MyYYcTplHjkkXKfdT+BiI31nzwYjS5KSz1reb74IpJWrexMnpzPJZdUz8DhDFYrjBgRz549lfe73FzPuqVPPikMeZzrr7eSlOT0WZ8UH+9i6NB4du/WYLOpmDTJQd++Jp57rizASAJB3SESI0Gdc9/wk15JEUBUjIsb+xRVJka4eX7yITreVEJktOdi3uO+AiYMacqhvwLf5f47tQGdbyn2Gr+sRMWqhQl1ELmbVu3LfJq7NsywcdegE8yb1LjKET56qwlNmptJb+q52blcnqQtIbnyBpTc0MbdDx3lpxUNOXmsen0sBAJBcFatWkXnzp3DHYagnujQwc7KlaeYMCGGqVOjWb48grfeKuL662uvGpAkGDeulPJyCa3WjU7nmZcKVs2pNCfwTkqSktzk5XmqUb//rmXHDu9JM5dLYsMGHUVFEnFxlZM28+ZF8M470Zw44Xk8W7YsguHDyxgypGrb65EjS/n1Vx3HjlU+2sXFOWnd2s6NNyZz/LiaJ58s5ZlnStHVQrTw7bcGvz2RduzQUlIiVVkpOkOTJk5uusnC4sURFRK8hAQnGo2bbdsqAzx+XMOcOZHcfbelxsmcQBAqIjES1DnJqf7lDREyN7c215R6JUUAjbNsDB2fy7h+zQOOXZSn5Y2/ZfLYS0dJamjHVKZi1cJ4ls5OCihv8zE2kM/We5k7uNFoA/VCcvhWXc6MJzM5OHpAz7N9W9Nv2BEapFnZuyOSW+47SWS09007IdnOLb1zmTMp8/RHK4IPYNKg/D4+5gtO2TiyDM+nKhTEfEFyys6BRpYlKis/oVaTHMpqkn/zBWUVxasXkqIqpAowhnKcYJUleUXHx5jBq5rk9vu+737KbbIxlUVL4bhUbSZMmEBRUZHP+/369aNDhw4AfPXVV6jVaq6//vqA46xcuZKVK1cC8NZbb5GUlBRw3+qg0WjqbKyzwfkcb1ISzJoFAwfaGTFCQ79+Sdx7r5PXXnOSlVX7z6qL0yKPt7BQwmTy/aM3mdS4XIkVn2e1wsyZGq/GtqdOqfn00xhGjIgg2r/BqlfcH3/s5o03XJw86Um4br3VxWuvRdOmjYvPP3fQqZMe8NWjV+f3welU4XT63nAdDhVRUYnVOn+zZkGvXk7mzQODwc3IkS4GDvRVXhQUqFmwIJF//9tZ7XjPBUS89UtdxisSI0Gd8+vKGK651Xct0AmZXXf33oVeSdEZQlkrtHebkbFBkqeaI5F7wOBT7SrK1/DNZ8peSIEpLtAy/XWPTE6lge69Tvns43JBabH48xMIQuXll18Oun316tVs3LiRV155BSnIFH+PHj3o0aNHxeu8mqw494Nn8fr50+PsQoi3bVv4/nuYMiWaadMi+fprLQ9n2YDVAAAgAElEQVQ9VM7TT5f5WGGfbeTxXnmlioYNk7wqOQApKQ5iY/MqTA/+/FPLoUOJyqHYvx9++KGELl28701Hj6pYt07PqVMq7r3XTHKyizZtYP587+NbttRz3XVW1GoCGixU5/fhhhtUpKcnkZvr/X3S0+2oVPnVNnG47jrPf2dQqZIB3+RIrzeRl1da7XjPBUS89Uso8aalpYU0lpi3FNQ5qxfFs/mXaKzmyoeTw3v0TP975S/l8Rz/dXyrbFZNb3Dx5MTDTF7+F/9eupuHxx5FVc82x/8em8nebUZsVk/secc1LP8smZy9xhqN53JK7P7Td5rvWI6B775MrVWsAoHAw+bNm/n66695/vnn0ev9uLMILlgiImD06FJ++eUk999vYtasSDp1asCbb0Z7VV7CSYMGLu6910xsbKVqIiXFwbBh3tLtxEQnUVG+97iYGDepqa6KLhFuN/Tvn8C116YwcWI0q1frAyoaALp2tfpIxGtDQoKLYcPKSE/3yNrUajfZ2XYmTqwbi+0OHWxIkvf3adTIwaOPVi0nFAhqi+QO9tfkh5tU99VXLIILCE8T1mI631pM7n49i2Z6GysYo51MWr6HxlnWivcsZomF05L59G1Pv6M35u2jXZfKRm92uyfpemdUkxACCO2G6M/WW6Vyc3X3IlLSPb2Q8k/ofMcMsnjea7ZapUJvcPLsP/fSok0ZOr2LU8f0zJjYhG0b4rz2CziG8o7mJdFSxCGXzwUdI8h3kcvs5DI+ZTMNmczOrdwml7r5HCcfX25soJDjyU0PNEqpm7yPkbI/keR3P7fSWUq+TRGiS+1/m6+ULkgfoyDHnZmS2jT9GQS158knn8ThcBAV5WmknJ2dzeOPPx7SsUePHq2TGC7EGdZziVDj3bNHwzvvRLN8uQGNBvr0MTF0aHmF29zZwl+8W7ZomD07kogIN0OHltO4sa+pwsCBCfz4o/fa0yZNPPbd995rYtSoMiZOjGb69Ejs9soLS5s2NpYty6txAlST34eCAomlSyOIi3Nx222WWq1bkmOzwahRcWzc6LHtbtjQyXPPldK9e+XzwoX6+3uucCHGG2rFSCRGgrCR2dLM0PG5NGhkx2JS8esPMXz6z1RAosklZt5dtJfoOG85RP4JDccO6YhPdmAqU7FmaTwLpjTAZxFHLRKjJi1M3Novn8I8Dd/MTaa8ROM7ZjUSozMYoxxERDpPJ1qSd1IjEiORGAnCgkiMzg+qG++BA2o+/DCKBQuMWCwS115rpX9/E7feasZwFjxvqhvvGZOHDRu0vPhiLKdOqYiKctOsmYPbb7fgdEp89pmRxYvz6N49mb17vaVmkuRmwIAytm/XUVCgJinJxdChpdx+uzXAJ9YuXrsdvvoqgi1btHTvbuXGG611Yjkup6xMoqREomFDl8/YF/rvb7i5EOMNNTESixwEYePgrgjG9vW/VqhRlsUnKQKIT3aQmFI585fe9AROByycnlInMT067gg3980jNsEzk3dL3zzeGdWUnZuiaj22qUzj6YEkEAgEgnqlaVMnEycWM3p0KXPmGJk/38jw4fHExcVyzz0m7rzTQocOtjqVmNUGSYLcXDXvvRfNrbdaOHZMzdVX27j3XnPFPlOnRnHsmIqyMt8MxO2WmD+/sop06BC88EIcDRvmc+WVdVstKyiQGDAgkR07tDgcEgsWGOnY0cYnnxSgrX0f9wqiotx+pYUCQX0intIE5yQ71kdx6qiG5DTvC7qy+GCMctHlziLfxMgdZOGtrPIjd3hLzbBw032VSRFAelMbj798hFH3tPQeU+nyJv9o2SbJ5SIi0skDI47Q7DITZcUaPv8gnUO7jd5fxuVSjCHrM+RUyC3kDnbKypi8d5FK7fd9z2FBKkGqAK53kiJGWcyST8VIFXCblwue/HuqledAVo1Rut7JtqkcQapJXhWpwPsF6zPk1U/JZz/5+QnibBesx5FAIKg3EhNdPP10GU89VcbatTo+/9zIvHmRzJoVRWKik5tvtnDzzRY6dbKFbDNdG8xmicWLI9i7V8Mjj5STnl55ff/tNx1dulgZOrScHTs0TJsWRfv2NjIzPfs0a+bgjz90JCe7OH7ce1yVyu0lrQOPo92UKdHMnBl6b6FQePXVGP78s1I3Zzar+PlnPZ9+auTRR2vWMFZOaanE++9HsXOnlrg4N08/XSpsugVnDXF7FpyTFJ7S8ss3cZSXVv6KnjFEUGIw1o370Kh/HiIuyVfznZRmwxgdWoM9JTq9i4lzd3Lf345z1fUldL2zgAmzdnF5x7pZpCoQCASCqlGp4PrrbUydWsSffx5n2rQCrrvOytKlETz8cCKtWqVyxx1JTJwYzY8/6ikoqJ8mzF9+GcHmzVpatLAzblwsu3dXzk8fPKgmNtaTnF12mYOEBBe//lqZgFx3nZW1a/UMH15GcnLlPUmjcaHR+E/qysvr9nscOKBm+XLfXoNOp8SaNbU3PjGbJe6/P5Hp06NZs8bAkiURDBiQ4HUeBIL6RFSMBOcs0/6ezvofo7mlXwFOh8SBnQYGPnMCfYT3DeD4kWpejOWVn9MVl8val3JJO/8zXVazhFW5yc8YFciqULf3P07zVt5OOskN7Qx4MpexA2Mr31RWXGRVIrdSXC0rNyide7yqSSH2QkLlnVh6VXgCVHegqmpSkONc/qtJyoqO1zanIvkNcpy82uO1hizIfsoeSvIxVfZgFSN59U7RT0lesAtSTRIIBGeXqCg3PXta6NnTgtUKGzfqWLdOz9q1OqZNi+KDDzx/rxkZDtq0sdOypZ1mzRxkZjpp2tRRq8rSTz/pGTy4nOuus7F1q5ZVq/S0aOGphjRp4mTz5kotWqtWdg4erHxMu+MOM/fdl8Stt1p4//1C5s0z8uuvetq3t+FwwKpVvglL27b++wrWlKefjqtoxqokOrr2FbdZs4xs3eqtxzt2TMN770WzYEF+rccXCKpCJEaCc5qNq2PYuDoG8CQBl7YzcWWXUiIi3bhckLNXz39eDm1BXTB6Dsojwuh7UXe7Ydtv0TjsqlD9HLy45Ipy1H7+ymIT6/ZmJRAIBILqo9dD5842One2MXq0Z8H/li1atmzRnf6/lmXLvBOOyEgXyckukpJcJCU5iYx0ExHhxmDw/Ne6tZ2HH/b9LJsNrrjCzqlTnlmTm2+28tlnRsAzedapk43Vq/Vs3arl8svtFBaqKtz01q/X8uGHUTid8MMPBoYOLWPGjMqGxyUlEv36qdm2TYvTKaHTuWnb1sbIkWV1dq6OHlV5JWpyYmKcjBhR+8/askWHW9nwHMjPF7NJgrODSIwE5w1ut8T4RzNp362ELncWk7PPwNLZSVhMtV89G2hJUnGBmknjMmo87oFdEXS5w7cgVFok/vQEAoHgXCMqys2119q49trKZqpms8TBg2oOHNBw8KCG48dV5OerOHVKzcGDGkwmCbPZ85/VKtG7t9lvYqTRQEyMq6K/0lVX2Xj99ZiK7enpTnr0sDB3rpG4OBebN+t4+21P8tO4sZMJE4pJTfV/s4qJcbN4cR4LFhjZuFFL585W7rnHgqYObzXK+1glbgYOLKdly9qvA7rkErtPIgoQ58eMSSCoD8TTmeA8Q2LDj7Fs+FEmQ+N0pacmXqGnM6LFHyfRvluJl/GC0wmrv07AeXpxf0IDK3cMyMPhkFj+WTLFBVqvMfzx9ccpdLmjgKzLKp2FCk5q+HJ6qldDPskV5KKvvBvJ9lVaYXsp2uQSOeUSKblpgGJ2LpDMzsdgIahczhVwm5fMTpLL3gIbOPgaRFR+IR9zB7lkLkTJna9xQgDTBh9VYxBzB8n/fv7GEQgE5y4REW4uvdTBpZdW58E/yecdlcojj/vkk0jMZonISDfJyU62b9dQWqqiYUMnd99twWh0c+yYmvvuKyIjw4nbTcCESI5OBwMHmhg4sBphVoPUVBdZWQ7y8rwnI5s3dzB6dGW1yGyW+OKLCHJy1PTpY6ZVq9DP25Ah5XzzjYFduyrXFCUnO3niibqrfAkEwRCJkUAA7N4cycLpydzSr4AG6TZKCjXs2BDJjAnpANw+4BT9Rx4nOc0jgbv1gTxmvtGIn79JCDquxaTi03fSeOylI2h1LnL3R/DFlIZs/TVGWJ8IBALBRcZVV9lZssTFG29Ec+iQhk6dPK5zO3ZIFdbUN93k3XuorvsD1YZJkwoZOjSeffu02O2edVETJhSjP73Ud88eNUOGJLB3r+a0hbiRnj0tvPVWaIZDkZFunniinPffl3A4PEnXk0+W06mTreqDBYI6QCRGAsFp5n+QytezUshsaeHUUS35xz0zVgajkwefPUZCg8pZr5RGdh4afZS138bjcga+az3zzgGuvaWQyBjPbJ/TrsJcdo40zhAIBALBWefZZ0tZv16H222lSxcrOh106HB+rDtt1MjF0qX57N2rwWKRuOwyu1cvqJEj49mzp9I8oahIzcKFEfTta+LKK4N/R6cTBg9OYN06HRaLCklyYzRC48bCqltw9hCJkeDCwR24p0yoWMoldm08rW8+PcS9Q48Tn+x7YU7PtHJZu1K2/hbt9f4ZJ7SW7crofHNlUgSQ3tTK4y8f5rm+l3q5xgVzKfOR2cmlY4oeR17uZ6G61yldJQLI7LwkdhBcZid/rXCskwJI6Xwkd84QZXZSkPMTquQuiJSOIO518uN8tnn1MVLK7M6hKWCBQHBWiYlx0727teodz1EkCbKzfe+Jx46p2LbNt8OryaRi4sRoFiwoCDruwoUR/PyzHvtpJ1C3W2L3bi2vvBLLRx/VbS8mgSAQQswjEFRBh27FfvMslRq63RP4Qn/j3QVExfrqwpPTbFSsixIIBAKB4AJg8uQonAEUFPv3Vz0Pv3KloSIpknP4sJjDF5w9RGIkEFRJ4Nn9+KTAJf5jh/X481SwmlVBxxQIBAKB4Hzj5MnAMvFQ3PESE/03Uo+IEI50grOHSMMFFyZ1IKs74zZ3IkdPyyvNfnc5+JfBz2Gez17+WSK3DThFRvNKyYTNCutXxfo62bkCz1H4yOxk8jlJ+d0CyOxq3CRWLrNTBd4vqMzO57NlN0/5NqXkLphcziF3g1M44snHDFly571J8pLIyRvSotxRtp/y30K2TXkORF4sEAguMDp3trJihQF/F7iUlKqTmxEjyvi//zOQm1v5aBoV5eSee/zffwWC+kBUjASCKpg+Po2jB3110/t3GPjvf1IDHmcxqXltSBab1kSTe0DH/h0GvpqZwsdvpddnuAKBQCAQnHUGDjSddo/zniiLj3fyxBOlVR6fnu5i0qQiOna0kpnpoHVrG6NGlfHww6Z6ilgg8EVUjC5CJMlN17sLuf6OEkxlKhZMSSZnr29DtQsGd5D1PCFUk/KP6xh+S0seeu4Y19xcjNUksXldNJ++3RBTqRR09j9nj4Fx/bP9BVVh0uB5qZhNk5cmfKoxskqE4rsF7IcUrBeST2gBqknKbuShVpOUZRZ5/yavipFi/EAmDeBVCfL5zvJ9Q60s+VTe/Bs/+FR+vEIMYswQ5DiBQBAe9u5VM2VKFKWlKm65xUyfPpYgTUwFVaHTwRdf5DNvnpFZsyKxWqFlSwcjR5ZyxRWhOct16mTjq6/y6zlSgSAwIjG6CBk75RCdby1BZ/A8oF7VtZT/vJLOz8viwhzZuYupTM20Vxox7ZVG4Q5FIBAIBLVkxQo9L74Yy4kTnseglSv1fPedlZkzhftZbdBqYdAgE4MGiSqP4PxEzI1cZDS/3MRVXUsrkiKAxFQH9w8/iXBKEwgEAsGFjtsNkydHVyRFAHa7il9+0fP7776yaYFAcPEgKkYXGdfcXEx0vK/cKr6BHWO0C1PpRdZ8VCmzq4lRg1IG5zVe5dyDVuciItJFSaEakLwMCyTlwv1gY1bHqOHM+IEkdhC6zM6tlKKFKLNDaXoQQGLm008piIGD3BJWrfid9eqNJJPEBZPcBetjJNsWTC5HkB5HyuNEHyOBIHxYLHDypO/FsrRUxfffG7j66vOj2eqFjt3u+TeJi3MJiaPgrCESo4uMfdsjsFkkr4oRgKlUjcUkrjz1gSS5eWJ8Dld1LSUiyknBCS2fT05l7Yr4cIcmEAgEFx16PURHuzl2zPt9nc5Nq1YiKQo3bje8+WY0331noLRURVKSk8ceK6dvX+FOJ6h/xJPwRcav38eyb4e30YLZJLHu21hcARqzCWrHQ2OOctuAPBplWUlMcZDdxswTrx0hpfH52/lcIBAIzldUKrjjDjNGo3cVuWVLOz17WsIUleAM06dH8vHHkezbp+XkSTU7duh4880Ytm0Tc/mC+kf8ll1kuFwSLw9syhOv5ZLRwordJvHL8lgWTksOd2jnBnXR/8hrPBftu5ag03u/ndzQzv1PnGDyCxmndwu8vqsuZHaBJHZQDZmdYr+QZXbK+OU9/GrqbCeXyIUqh1RqMYJI6aQAUjqfsYP1MQry++MjyRMIBGeVZ58tIzbWzdKlBqxWiebNHUyYUBxSI1JB/bJihQGLxft6nZenZtq0KD74oChMUQkuFsQl4CKktEjDP59qEu4wLho0Ov9JjzHaf5dvgUAgENQvkgRDhpQzZEh5uEMRKLDb/U8cmc1iQklQ/wgpnUBQzxzZq/d5z1SmYuWXCWGIRiAQCASCc5dLL/Vd56XTubjtNrHGSFD/iIqRQBCIWjaGPcPkcY1JaWwj8xIL+gg3RXlq/vd9LBtXR1Hh2KaUisnDUMjsfKR13jvLd6z8OYhU7+zL7OTxy6pmdeFspzxOfq4UMQZv8BqguWwweVw1pHR1ItMUCASCC5BXXy1h714N27drMZtVxMU56dLFSu/eYv2XoP4RiZFAUM+UFGoYeWcLru5RTEa2lXXfxpK73xDusAQCgUAgOOeIjnazeHE+q1fr2b5dQ7duVlq1coQ7LMFFgkiMBOctHbqXcP+wk8QkOCjJ1/D5pAZsWhNzdj68miYNbrfEbz/E8dsPgXZQ9tgJvYLkfZgk3zFwQGezmuT0XkvlVeHx2k9RGZOCVOzk1SUfYwbZcfIQfXoJBflywXooeQcpOybYyapGNUkgEAhqycaNGv7xj1hOnVIRE+Ni8OBy7rnn/Km4SBJ062alWzfh3io4u4jESHBe0vKqcka9k0NiyplZJCtpzXJ45cGm7NtuDGtsAoFAIBCEi5wcNU88kUBubuUj3r59GgwGN7fdJhINgSAYwnxBcF7Sb8QJWVLkISnVwQMjT4YponMTtcbN4OeO8O7CXbzz5S76jzwavBIjEAgEgvOaf/0ryispAigsVPPxx5Fhiuj8Ze9eNY88Es+ddyYxYEACv/+uC3dIgnpGVIwE5yXGaP8yrsiYMFhgh9pHp1pjhiiD8znMO5axk/fT+ZYi1Kf/0i+5opyGGVbefbZpaOOfTZmdUuoW6Bjwkt35JHpKE4eKHZUBy2L06RUlM3cIOoY8xsDfWfQtEggEZ4v8fP/XKZNJzIVXh6NHVTz0UAKHDmkr3tu5U8O0aUVcfbUtjJEJ6hPxVyI4L8nxY4ENcPAvYWpwhpTGVlpfXVaRFAFoddC2cymxCb52qAKBQCA4/2nf3u5XGdCokeidVx3efz/aKykCOHFCw6RJUWGKSHA2EImR4Lzkk7casneroaLg4HTC7i0RzHknNbyBnUOkN7UQn+zr5BOb4KBBIzHbJRAIBBcijz1WRvv2NjSayuQoO9vOK68UhzGq84/jx9V+3y8uFo/OFzJCSic4Lykt0vBMr+bc9XAezVub+WuLkWWzk7BZzoELVh31Pwo8foBeRQr2bTdyMldLg3Tv6lD+CS05e/QVsjuf/juBZHzKz6oPmZ2cIM528vPoE0WADw/Z5c7nwND6KSk3BY0xmCRPUGOWLFnC3LlzmTlzJjExZ8mhUiA4xzAYYMGCfD7/3Mivv+rIynLw+OPlxMSI9aXVITvbzqpVviqUlBRRebuQEYmR4LzFalbz36kp4Q7jnKU4X8vab+O4pW8+xijPU3t5iYofv07AXO5/JkwgOF/Jy8tj69atJCUlhTsUgSDs6HQwaJCJQYNM4Q7lvGXkyDJ+/lnPjh2VhguZmXZeeKEkjFEJ6huRGAkEZ5NA1aSaVpL8VHcMRidXdS2hrFTL9FfT2fZbFN375ON2SayYl8T6H2MVQ9RxXySocTXJa8hQDRyUBDB08IkoaA8l+b9HkNnBkCtNSkKsmglCZvbs2QwYMIC333473KEIBILzFKsV1qzRI0nQpYuVhQvzmTQpit27tTRs6GTUqFJSU8X1+0JGJEYCwQXETffl88DIEzRsYsNuk8jZa+CNoU35ZXnzcIcmENQbGzZsICEhgczMzCr3XblyJStXrgTgrbfeqrMKk0ajOa+qVSLe+kXEW7/UR7w//igxcqSavXs9E17Z2W6mTHHyr3+dmThTAQk1Gluc3/qlLuMViZFAcIEQFetg4LPHSW3sWVOkN7hp3trMM+8cYvS9l4Q5OoGgdkyYMIGioiKf9/v168eiRYt46aWXQhqnR48e9OjRo+J1Xl5encSXlJRUZ2OdDUS89YuIt36p63gdDnjqqWR2766s9O/aJTFihJPvv88LKlIIhYv9/NY3ocSblpYW0lgiMRIIzgXqoBfSDXcXViRFcho2sRIdZ6O0UPbnHuLi/zqR2SlxhfjZwXaTyeWC9giqgeQO/MjuAo3nDHJ+RCPdOuXll1/2+/7hw4c5efIkY8aMASA/P5/nn3+eiRMnEhcXdzZDFAgE5ymbN2s5cMB37e3Bg2q2bdPSpo1ocXGxIBIjgeACwWJW4XSCWnFtdzolnA7RYFRwYZKRkcHMmTMrXg8fPpyJEycKVzqBQBAyej1otWBX5D8aDeh0YpLrYkJ4xgoEFwg/L43n6EHfxreHdhswlQoXOoFAIBAI/NG6tZ2sLN++f1lZDi65xPd9wYWLqBgJBOciNeiFZLWoeP/ZxgybkEuDRh7zhYO7DPzzySZ+xq+Gw1ygEEOV2YX62dVwtvMari4kd0oCSfCC9FPyiasm4wtqzZQpU8IdgkAgOM+QJJgypZCnn47j4EENkgSZmQ4mTy6qk/aDgvMHkRgJBBcQ29dHMfzWFqQ1tWEpV1FwUhvukAQCgUAgOOfJynKydGk+OTlqJAkaNRKNXC9GRGIkEFxwSBw94CupEwgEAoFAEJzGjUVCdDEjEiOB4HyjBjK76o0fotQt6BA1kNlVx9lOHktdSO6UwwdrLhuI6sjjlJI8gUAgEAgEYUcI3QUCgUAgEAgEAsFFj6gYCQQXEvJqUl2vGK0DwwbPMKFVeGpk4OA7SOXPIX4uVK+6VIGiClQtsweBQCAQCARhR1SMBAKBQCAQCAQCwUWPSIwEAoFAIBAIBALBRY/kdgdbyS0QCAQCgUAgEAgEFz6iYiQQCAQCQS0ZO3ZsuEOoFiLe+kXEW7+IeOuXizlekRgJBAKBQCAQCASCix6RGAkEAoFAIBAIBIKLHvWrr776ariDEAgEAoHgfKdZs2bhDqFaiHjrFxFv/SLirV8u1niF+YJAIBAIBAKBQCC46BFSOoFAIBAIBAKBQHDRowl3AAKBQCAQXCisWLGCb7/9FrVaTbt27Rg4cGC4Q6qSJUuWMHfuXGbOnElMTEy4wwnInDlz2LhxIxqNhpSUFIYNG0ZkZGS4w/Jh8+bNzJo1C5fLRffu3enVq1e4QwpIXl4eU6ZMoaioCEmS6NGjB7fffnu4w6oSl8vF2LFjSUhIOOcd1MrLy5k2bRo5OTlIksQTTzxBixYtwh1WQJYtW8aqVauQJInGjRszbNgwdDpduMOqYOrUqWzatInY2FjeffddAMrKynj//fc5deoUycnJjBo1iqioqBqNLxIjgUAgEAjqgG3btrFhwwbeeecdtFotxcXF4Q6pSvLy8ti6dStJSUnhDqVK2rRpQ//+/VGr1cydO5dFixadc4mny+Xio48+4qWXXiIxMZFx48bRvn17GjVqFO7Q/KJWq3nwwQdp1qwZZrOZsWPH0qZNm3M23jMsX76c9PR0zGZzuEOpklmzZnHFFVfw7LPP4nA4sFqt4Q4pIAUFBaxYsYL3338fnU7He++9x7p167jhhhvCHVoFN9xwA7feeitTpkypeG/x4sVcfvnl9OrVi8WLF7N48eIaXxuElE4gEAgEgjrg+++/5+6770ar1QIQGxsb5oiqZvbs2QwYMABJksIdSpW0bdsWtVoNQIsWLSgoKAhzRL7s3buX1NRUUlJS0Gg0dO7cmfXr14c7rIDEx8dXLFqPiIggPT39nDyvcvLz89m0aRPdu3cPdyhVYjKZ2LlzJzfeeCMAGo3mnKxyynG5XNhsNpxOJzabjfj4+HCH5MVll13mUw1av349Xbt2BaBr1661+psTFSOBQCAQCOqAY8eOsWvXLr744gu0Wi0PPvggzZs3D3dYAdmwYQMJCQlkZmaGO5Rqs2rVKjp37hzuMHwoKCggMTGx4nViYiJ79uwJY0Shc/LkSQ4cOHBO/84CfPLJJwwcOPC8qBadPHmSmJgYpk6dyqFDh2jWrBmDBw/GYDCEOzS/JCQk0LNnT5544gl0Oh1t27albdu24Q6rSoqLiysSuPj4eEpKSmo8lkiMBAKBQCAIkQkTJlBUVOTzfr9+/XC5XJSVlfHGG2+wb98+3n//fT744IOwVmOCxbto0SJeeumlMEQVmGDxdujQAYCvvvoKtVrN9ddff7bDqxJ/Rr/nQzXOYrHw7rvvMnjwYIxGY7jDCcjGjRuJjY2lWbNmbN++PdzhVInT6eTAgQM88sgjZGdnM2vWLBYvXky/fv3CHZpfysrKWL9+PVOmTMFoNPLee++xZs0aunTpEu7QzhoiMRIIBAKBIERefvnlgNu+//57OnbsiCRJNG/eHJVKRWlpaVgNDQLFe/jwYU6ePMmYMWMAjzzp+eefZ+LEicTFxZ3NEL0Idn4BVq9ezcaNG4eh/CYAAAJMSURBVHnllVfOyYQjMTGR/Pz8itf5+fnnnBRJicPh4N133+X666+nY8eO4Q4nKH/99RcbNmzgjz/+wGazYTabmTRpEk899VS4Q/NLYmIiiYmJZGdnA9CpUycWL14c5qgCs3XrVho0aFBxzerYsSO7d+8+5xOj2NhYCgsLiY+Pp7CwsFbXXJEYCQQCgUBQB3To0IFt27bRqlUrjh49isPhIDo6Otxh+SUjI4OZM2dWvB4+fDgTJ048p13pNm/ezNdff8348ePR6/XhDscvWVlZHDt2jJMnT5KQkMC6devO2Yd28FS4pk2bRnp6OnfeeWe4w6mS/v37079/fwC2b9/O0qVLz+nzGxcXR2JiIkePHiUtLY2tW7ee08YWSUlJ7NmzB6vVik6nY+vWrWRlZYU7rCpp3749P/30E7169eKnn36qqC7XBNHgVSAQCASCOsDhcFSsJdBoNDz44IO0bt063GGFxPmQGD355JM4HI6KhdfZ2dk8/vjjYY7Kl02bNjF79mxcLhfdunWjd+/e4Q4pILt27eKVV14hIyOjogL3wAMP0K5duzBHVjVnEqNz3a774MGDTJs2DYfDQYMGDRg2bFiNraTPBgsWLGDdunWo1WoyMzMZOnTo/7dzx1QAg0AQBc8BDVawghCMgDpU0GEgXQSkyns3I2G73+x7KPMHa63Ye8e9N0op0XuP1lrMOeOcE7XWGGN83lgYAQAA6bnrBgAA0hNGAABAesIIAABITxgBAADpCSMAACA9YQQAAKQnjAAAgPSEEQAAkN4DIeCAJEiGU/0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from Bayes_helpers import plotGaussianBayes\n",
    "plotGaussianBayes(X, y, gnb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Użyj funkcji do generowania danych, aby wygenerować zbiór testowy oraz sprawdź na nim trafność klasyfikacji metody."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8135"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test, y_test = generate3(1000)\n",
    "np.mean(gnb.predict(X_test) == y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Ćwiczenia**\n",
    " - Pamiętaj o przetestowaniu Twojego algorytmu dla wszystkich trzech generatorów danych. W których ze zbiorów założenie o warunkowej niezależności zmiennych jest spełnione? Jak brak spełnienia tego założenia wpływa na działanie klasyfikatora?\n",
    " - Z pliku `Bayes_helpers` zaimportuj klasę `GaussianBayes` (identyczna obsługa jak tej zaimplementowanej przez Ciebie). Klasa implementuje algorytm Bayesa bez założenia o niezależności zmiennych (ale z założeniem o normalności rozkładów). Porównaj wyniki - szczególnie dla zbiorów dla których założenie o warunkowej niezależności zmiennych nie jest spełnione.\n",
    " - Klasyfikatora `GaussianBayes` nie można wytrenować na zbiorach które mają mniej niż 3 przykłady dla każdej z klas. Jak myślisz dlaczego? Jak ten problem będzie się zmieniał dla zbiorów o wysokiej liczbie cech?\n",
    " - Nawet używając klasyfikatora `GaussianBayes`, który zakłada kompletny model zależności i prawidłowy rozkład danych (nasze dane są generowane z rozkładów normalnych) - często nie jest w stanie uzyskać 100% trafności nawet na zbiorze uczącym. Jak myślisz, dlaczego? \n",
    " - Czy gdyby przepisać do klasyfikatora prawdziwe wartości średnich i macierz wariancji-kowariancji cech (z generatora) - uzyskalibyśmy 100% trafność? Co możemy powiedzieć o takim klasyfikatorze? Czy jest możliwe uzyskanie klasyfikatora bardziej trafnego niż taki? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Generatory prezentują 3 sytuacje 1) zmienne niezależne 2) zmienne zależne ale klasyfikator dobrze sobie radzi \n",
    "3) Zmienne zależne powodujące znaczną utratę trafności. \n",
    "Zależności można zidentyfikować po prostu na podstawie obejrzenia wykresów wygenerowanych danych lub (jeśli chcesz bardziej bezpośrednio) poprzez zajrzenie do źródła generatora i obejrzenie użytych macierzy wariancji-kowariancji. Brak znacznego wpływu zależności na trafność w sytuacji (2) można wytłumaczyć poprzez narysowanie \"krzyża\" o długości około 3 wariancji w każdej osi, a następnie poprzez analizę nakładania się ich patrząc na każdą z osi z osobna. Sytuacja 2 wyjaśnia użyteczność NB w praktyce - zmienne prawie zawsze są zależne i estymacja prawdopodobieństwa jest błędna (co widać na wykresach), ale niekoniecznie musi to negatywnie wpływać na jakość predykcji (co również widać) - bowiem jest różnica pomiędzy prawidłową klasyfikacją (zgadza się to, które prawdopodobieństwo jest największe) a prawidłową estymacją prawdopodobieństwa (zgadza się jego wartość).  W analizie ćwiczenia warto też zwrócić uwagę, że Gaussian NB jest klasyfikatorem liniowym: https://stats.stackexchange.com/questions/142215/how-is-naive-bayes-a-linear-classifier \n",
    "2. Ładnie widać dużo lepsze dopasowanie się do rozkładu danych, a w sytuacji (3) dużo lepszą jakość predykcji.\n",
    "3. Macierz wariancji-kowariancji jest osobliwa w takiej sytuacji. Im więcej cech tym więcej przykładów dla każdej z klas jest potrzebnych - dokładnie (n+1). Dodatkowo należy wspomnieć o jakości estymat - szczególnie o traceniu stopni swobody.\n",
    "4. 100% trafność klasyfikacji jest możliwa do uzyskania tylko jeżeli warunkowe prawdopodobieństwa P(y|x) = 1 lub 0 dla każdego x. Jeżeli nie jest to prawda to nie jest możliwe skonstruowanie klasyfikatora ze 100% trafnością.\n",
    "5. Nie, ale jest to klasyfikator teoretycznie \"idealny\" tj. o najwyższej możliwej trafności predykcji - klasyfikator Bayesa. Zgodnie z prawem wielkich liczb wraz z rozmiarem próbki GaussianBayes w takiej sytuacji jest zbieżny do tego klasyfikatora. Warto podkreślić brak powiązania NB z rozkładem normalnym - można modelować zmienne bardziej odpowiednimi dla nich rozkładami."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Zadanie 2\n",
    "Zaimplementuj algorytm naiwnego Bayesa dla binarnych cech. Możesz założyć binarność, choć implementacja dla cech nominalnych nie byłaby dużo bardziej skomplikowana.\n",
    "\n",
    "*Wskazówka:* W zależności od Twojej implementacji funkcja `np.nan_to_num` może być przydatna do zabezpieczenia się przed sytuacją mnożenia zerowego prawdopodobienstwa (logarytm 0 na komputerze to $-\\infty$) przez zero."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import norm\n",
    "class NaiveBayes():\n",
    "    def __init__(self):\n",
    "        self.prob = {}\n",
    "        # Słownik, który docelowo powinien zawierać tablicę/wektor warunkowych prawdopodobieństw dla każdego atrybutu \n",
    "        # Każda tablica/wektor powinna być typu np.array\n",
    "        # np. 1) prob[2] powinno zawierać wektor prawdopodobieństw o długości równej liczbie atrybutów. Każda kolejna wartość wektora to prawdopodobieństwo, że dla klasy o indeksie 2 kolejny atrybut przyjmie wartość 1.\n",
    "        #     2) prob[0][6] powinno zawierać prawdopodobieństwo, że szósty atrybut równa się 1 dla klasy o indeksie 0.\n",
    "        # (Możesz spróbować zaimplementować efektywniejszą implementację używając macierzy)\n",
    "        self.class_log_prob = None\n",
    "        # Wektor zawierający logarytmy prawdopodobieństwa dla każdej z klas \n",
    "        # np. class_log_prob[1] zawiera logarytm prawdopodobieństwa, że klasa jest równa 1, log P(C=1)\n",
    "        \n",
    "    def fit(self, X, y):\n",
    "        # TWÓJ KOD TUTAJ\n",
    "        unique, counts = np.unique(y, return_counts=True)\n",
    "        self.class_log_prob = np.log(counts/np.sum(counts))\n",
    "        for clazz in unique:\n",
    "            class_data = X[y == clazz] #wybieramy z X przypadki z klasy clazz\n",
    "            self.prob[clazz] = (np.sum(class_data, axis=0)) / (class_data.shape[0]) #ile razy wartość każdego atrybutu jest równa 1 / liczba przypadków z klasy clazz\n",
    "        \n",
    "    def predict_proba(self, X):\n",
    "        # TWÓJ KOD TUTAJ\n",
    "        results = np.tile(self.class_log_prob, (X.shape[0], 1))\n",
    "        for i in range(len(self.class_log_prob)):\n",
    "            results [:,i] += np.nan_to_num(X@np.log(self.prob[i].T))\n",
    "            results [:,i] += np.nan_to_num((1-X)@np.log(1-self.prob[i].T))\n",
    "        probabilities = np.exp(results)\n",
    "        probabilities /= np.sum(probabilities, axis=1, keepdims=True)\n",
    "        return probabilities\n",
    "    \n",
    "    def predict(self,X):\n",
    "        prob = self.predict_proba(X)\n",
    "        return np.argmax(prob, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Przetestuj algorytm dla podanych danych. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/student/.local/lib/python3.6/site-packages/ipykernel_launcher.py:26: RuntimeWarning: divide by zero encountered in log\n",
      "/home/student/.local/lib/python3.6/site-packages/ipykernel_launcher.py:27: RuntimeWarning: divide by zero encountered in log\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0. , 1. ],\n",
       "       [0.5, 0.5],\n",
       "       [0.5, 0.5],\n",
       "       [1. , 0. ]])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nb = NaiveBayes()\n",
    "X = np.array([[1,1,1], [0,1,1], [0,0,1], [0,0,0]])\n",
    "y = np.array([1,1,0,0])\n",
    "nb.fit(X,y)\n",
    "nb.predict_proba(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podejrzyjmy wyestymowane wartości prawdopodobieństw:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: array([0. , 0. , 0.5]), 1: array([0.5, 1. , 1. ])}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nb.prob"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Spójrzmy na analogiczną listę estymat dla gotowej implementacji algorytmu `FullBayes` (czyli wersji algorytmu bez założenia o niezależności)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{1: {(0, 0, 0): 0,\n",
       "  (0, 0, 1): 0,\n",
       "  (0, 1, 0): 0,\n",
       "  (0, 1, 1): 1,\n",
       "  (1, 0, 0): 0,\n",
       "  (1, 0, 1): 0,\n",
       "  (1, 1, 0): 0,\n",
       "  (1, 1, 1): 1,\n",
       "  'suma': 2},\n",
       " 0: {(0, 0, 0): 1,\n",
       "  (0, 0, 1): 1,\n",
       "  (0, 1, 0): 0,\n",
       "  (0, 1, 1): 0,\n",
       "  (1, 0, 0): 0,\n",
       "  (1, 0, 1): 0,\n",
       "  (1, 1, 0): 0,\n",
       "  (1, 1, 1): 0,\n",
       "  'suma': 2}}"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from Bayes_helpers import FullBayes\n",
    "fb = FullBayes()\n",
    "fb.fit(X,y)\n",
    "fb.prob"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Czy coś cię niepokoi w wyświetlonych estymacjach? Jak ten problem będzie się zmieniał w zależności od rozmiaru zbioru i rozmiaru wymiarowości?\n",
    "Odpowiedź: dużo zer, za mało danych by stworzyć poprawne estymacje. Rozmiar wymiarowości: odpowiedź znajdziesz poniżej w tekście, rozmiar zbioru - im większy zbiór, tym ten problem będzie mniej widoczny."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Zadanie 2b\n",
    "Rozszerz Twój kod o estymowanie prawdopodobieństwa estymatą Laplace'a."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "class SmoothNaiveBayes(NaiveBayes):     \n",
    "    def fit(self, X, y):\n",
    "        # TWÓJ KOD TUTAJ\n",
    "        unique, counts = np.unique(y, return_counts=True)\n",
    "        self.class_log_prob = np.log((counts+1)/(np.sum(counts)+len(unique)))\n",
    "        for clazz in unique:\n",
    "            class_data = X[y == clazz]\n",
    "            self.prob[clazz] = (np.sum(class_data, axis=0)+1)/(class_data.shape[0]+2) # \"+2\" bo są dwie wartości atrybutu (cechy są binarne)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Również przetestuj i porównaj uzyskane estymaty z wersją algorytmu bez rozmywania."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.1 , 0.9 ],\n",
       "       [0.25, 0.75],\n",
       "       [0.75, 0.25],\n",
       "       [0.9 , 0.1 ]])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "snb = SmoothNaiveBayes()\n",
    "snb.fit(X,y)\n",
    "snb.predict_proba(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pora na test działania zaimplementowanych metod na większych zbiorach danych. W tym celu użyjemy funkcji `generate_binary`, która generuje sztuczne binarne dane. Jej pierwszym argumentem jest liczba przykładów uczących, a drugim argumentem jest liczba cech $k$. Poniższy kod nie tylko generuje dane, ale także dzieli je na cześć uczącą i testową."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from Bayes_helpers import generate_binary\n",
    "Xb, yb = generate_binary(5500, k = 10)\n",
    "X_train, y_train = Xb[:5000], yb[:5000]\n",
    "X_test, y_test = Xb[5000:], yb[5000:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wytrenuj na tych większych danych zaimplementowane przez ciebie algorytmy `NaiveBayes` i `SmoothNaiveBayes`,  także gotowe metody `FullBayes` i `SmoothFullBayes` z `Bayes_helpers`, które są pozbawione założenia o niezależności zmiennych. Dla każdego klasyfikatora zmierz trafność klasyfikacji i podejrzyj zawartości estymat `print(classifier.prob)`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: {(0, 0, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 1, 1): 1,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  ...},\n",
       " 1: {(0, 0, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 0, 0, 1): 2,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 0, 1, 1): 2,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 0, 0): 1,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 0, 1): 1,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 1, 0): 1,\n",
       "  (0, 0, 0, 0, 0, 0, 0, 1, 1, 1): 5,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 0, 1, 1): 1,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 0, 1, 1): 2,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 0, 1): 1,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 1, 0): 1,\n",
       "  (0, 0, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 0, 1): 4,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 1, 0): 2,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 0, 1, 1): 6,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 0, 1): 1,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 1, 0): 2,\n",
       "  (0, 0, 0, 0, 1, 0, 0, 1, 1, 1): 6,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 0, 1): 1,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 0, 1, 1): 1,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 0, 1, 1, 1, 1): 1,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 0, 1): 2,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 0, 1, 1): 1,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 0, 1): 1,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 0, 1, 1, 1): 2,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 0, 0): 1,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 0, 1): 11,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 1, 0): 6,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 0, 1, 1): 33,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 0, 0): 2,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 0, 1): 6,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 1, 0): 2,\n",
       "  (0, 0, 0, 1, 0, 0, 0, 1, 1, 1): 14,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 0, 1, 1): 2,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 0, 1): 1,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 0, 0): 1,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 0, 1): 2,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 1, 0): 1,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 0, 1, 1): 6,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 0, 1): 2,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 0, 1, 1, 1): 8,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 0, 1, 1): 2,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 0, 1, 1, 1, 1, 1): 1,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 0, 0): 4,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 0, 1): 11,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 1, 0): 1,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 0, 1, 1): 31,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 0, 0): 3,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 0, 1): 5,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 1, 0): 4,\n",
       "  (0, 0, 0, 1, 1, 0, 0, 1, 1, 1): 12,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 0, 1): 1,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 0, 1, 1): 1,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 1, 0): 1,\n",
       "  (0, 0, 0, 1, 1, 0, 1, 1, 1, 1): 2,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 0, 0): 1,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 0, 1): 3,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 1, 0): 2,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 0, 1, 1): 7,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 0, 1): 3,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 1, 0): 1,\n",
       "  (0, 0, 0, 1, 1, 1, 0, 1, 1, 1): 7,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 0, 1): 1,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 0, 0): 12,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 0, 1): 44,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 1, 0): 9,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 0, 1, 1): 96,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 0, 0): 1,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 0, 1): 20,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 1, 0): 8,\n",
       "  (0, 0, 1, 0, 0, 0, 0, 1, 1, 1): 49,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 0, 0): 1,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 0, 1): 2,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 1, 0): 3,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 0, 1, 1): 5,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 0, 0): 1,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 0, 1, 1, 1, 1): 1,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 0, 0): 2,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 0, 1): 9,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 1, 0): 6,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 0, 1, 1): 20,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 0, 0): 2,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 0, 1): 5,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 1, 0): 2,\n",
       "  (0, 0, 1, 0, 0, 1, 0, 1, 1, 1): 14,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 0, 0): 3,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 0, 1): 1,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 0, 1, 1): 4,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 0, 0): 1,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 0, 1, 1, 1, 1, 1): 2,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 0, 0): 6,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 0, 1): 40,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 1, 0): 15,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 0, 1, 1): 84,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 0, 0): 5,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 0, 1): 9,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 1, 0): 13,\n",
       "  (0, 0, 1, 0, 1, 0, 0, 1, 1, 1): 48,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 0, 0): 1,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 0, 1): 4,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 0, 1, 1): 4,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 1, 0): 1,\n",
       "  (0, 0, 1, 0, 1, 0, 1, 1, 1, 1): 4,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 0, 0): 1,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 0, 1): 7,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 1, 0): 2,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 0, 1, 1): 18,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 0, 1): 4,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 1, 0): 3,\n",
       "  (0, 0, 1, 0, 1, 1, 0, 1, 1, 1): 19,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 0, 0): 1,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 0, 1): 1,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 1, 0): 1,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 0, 1, 1): 3,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 1, 0): 1,\n",
       "  (0, 0, 1, 0, 1, 1, 1, 1, 1, 1): 2,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 0, 0): 30,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 0, 1): 171,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 1, 0): 65,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 0, 1, 1): 403,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 0, 0): 23,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 0, 1): 91,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 1, 0): 53,\n",
       "  (0, 0, 1, 1, 0, 0, 0, 1, 1, 1): 226,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 0, 0): 3,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 0, 1): 7,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 1, 0): 5,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 0, 1, 1): 35,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 0, 1): 6,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 1, 0): 5,\n",
       "  (0, 0, 1, 1, 0, 0, 1, 1, 1, 1): 22,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 0, 0): 10,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 0, 1): 70,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 1, 0): 20,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 0, 1, 1): 109,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 0, 0): 5,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 0, 1): 26,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 1, 0): 14,\n",
       "  (0, 0, 1, 1, 0, 1, 0, 1, 1, 1): 57,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 0, 0): 1,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 0, 1): 4,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 1, 0): 2,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 0, 1, 1): 10,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 0, 1): 5,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 0, 1, 1, 0, 1, 1, 1, 1, 1): 6,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 0, 0): 36,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 0, 1): 153,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 1, 0): 74,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 0, 1, 1): 372,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 0, 0): 24,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 0, 1): 109,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 1, 0): 32,\n",
       "  (0, 0, 1, 1, 1, 0, 0, 1, 1, 1): 225,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 0, 0): 2,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 0, 1): 12,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 1, 0): 7,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 0, 1, 1): 25,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 0, 0): 1,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 0, 1): 10,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 1, 0): 3,\n",
       "  (0, 0, 1, 1, 1, 0, 1, 1, 1, 1): 20,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 0, 0): 13,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 0, 1): 37,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 1, 0): 18,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 0, 1, 1): 93,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 0, 0): 3,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 0, 1): 29,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 1, 0): 11,\n",
       "  (0, 0, 1, 1, 1, 1, 0, 1, 1, 1): 59,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 0, 0): 2,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 0, 1): 3,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 1, 0): 2,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 0, 1, 1): 15,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 0, 1): 3,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 1, 0): 1,\n",
       "  (0, 0, 1, 1, 1, 1, 1, 1, 1, 1): 5,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 0, 1, 1): 1,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 0, 1): 2,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 0, 1, 1, 1): 1,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 0, 1, 1, 1): 1,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 0, 1, 1): 5,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 1, 0): 1,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 0, 1, 1): 1,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 0, 1, 1, 1): 2,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 0, 1, 1, 1, 1, 1, 1): 1,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 0, 0): 1,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 0, 1): 4,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 0, 1, 1): 3,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 0, 1): 2,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 1, 0): 1,\n",
       "  (0, 1, 1, 1, 0, 0, 0, 1, 1, 1): 5,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 1, 0): 1,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 0, 1, 1, 1, 1): 1,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 0, 1, 1): 2,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 0, 1, 1, 1): 1,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 0, 0): 1,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 0, 1): 2,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 1, 0): 1,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 0, 1, 1): 9,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 0, 1): 1,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 1, 0): 2,\n",
       "  (0, 1, 1, 1, 1, 0, 0, 1, 1, 1): 3,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 0, 1, 1): 1,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 0, 1): 1,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 0, 1, 1): 4,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 0, 1): 1,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (0, 1, 1, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 0, 1): 2,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 0, 1, 1): 1,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 0, 1): 2,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 0, 0): 1,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 0, 1): 1,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 0, 1, 1): 4,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 0, 1): 1,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 0, 1, 1): 1,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 0, 0): 3,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 0, 1): 4,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 1, 0): 3,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 0, 1, 1): 6,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 0, 1): 1,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 0, 1, 1, 1): 6,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 0, 1, 1): 4,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 0, 1, 1, 1, 1): 1,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 1, 0): 1,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 0, 1, 1): 2,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 0, 0): 1,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 0, 1): 6,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 1, 0): 3,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 0, 1, 1): 12,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 0, 0): 1,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 0, 1): 3,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 1, 0): 1,\n",
       "  (1, 0, 0, 1, 1, 0, 0, 1, 1, 1): 7,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 0, 1, 1): 2,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 0, 0): 1,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 0, 1, 1): 4,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 0, 0): 2,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 0, 1): 9,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 1, 0): 2,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 0, 1, 1): 19,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 0, 0): 3,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 0, 1): 3,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 0, 1, 1, 1): 19,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 0, 1): 2,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 0, 1, 1): 3,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 0, 0): 1,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 0, 1): 7,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 1, 0): 5,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 0, 1, 1): 9,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 0, 1): 1,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 0, 1, 1, 1): 5,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 0, 0): 5,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 0, 1): 10,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 1, 0): 6,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 0, 1, 1): 23,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 0, 0): 1,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 0, 1): 5,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 1, 0): 3,\n",
       "  (1, 0, 1, 0, 1, 0, 0, 1, 1, 1): 17,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 0, 1): 2,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 1, 0): 1,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 0, 1, 1): 3,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 0, 1): 1,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 0, 1, 1, 1, 1): 2,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 0, 1): 3,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 0, 1, 1): 8,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 0, 1): 3,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 0, 1, 1, 1): 4,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 0, 1, 1): 1,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 0, 1): 1,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 0, 1, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 0, 0): 9,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 0, 1): 53,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 1, 0): 21,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 0, 1, 1): 128,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 0, 0): 5,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 0, 1): 31,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 1, 0): 15,\n",
       "  (1, 0, 1, 1, 0, 0, 0, 1, 1, 1): 72,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 0, 1): 4,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 1, 0): 1,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 0, 1, 1): 7,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 0, 0): 1,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 0, 1): 1,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 1, 0): 1,\n",
       "  (1, 0, 1, 1, 0, 0, 1, 1, 1, 1): 7,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 0, 0): 4,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 0, 1): 15,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 1, 0): 4,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 0, 1, 1): 36,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 0, 0): 1,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 0, 1): 9,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 1, 0): 6,\n",
       "  (1, 0, 1, 1, 0, 1, 0, 1, 1, 1): 18,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 0, 1): 2,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 0, 1, 1): 6,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 0, 1): 1,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 1, 0): 3,\n",
       "  (1, 0, 1, 1, 0, 1, 1, 1, 1, 1): 6,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 0, 0): 10,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 0, 1): 58,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 1, 0): 19,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 0, 1, 1): 119,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 0, 0): 3,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 0, 1): 27,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 1, 0): 13,\n",
       "  (1, 0, 1, 1, 1, 0, 0, 1, 1, 1): 50,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 0, 1): 4,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 1, 0): 1,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 0, 1, 1): 9,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 0, 0): 2,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 0, 1): 3,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 1, 0): 1,\n",
       "  (1, 0, 1, 1, 1, 0, 1, 1, 1, 1): 4,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 0, 0): 5,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 0, 1): 19,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 1, 0): 3,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 0, 1, 1): 34,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 0, 0): 1,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 0, 1): 8,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 1, 0): 1,\n",
       "  (1, 0, 1, 1, 1, 1, 0, 1, 1, 1): 18,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 0, 0): 1,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 0, 1): 2,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 0, 1, 1): 4,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 0, 1): 1,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 1, 0): 1,\n",
       "  (1, 0, 1, 1, 1, 1, 1, 1, 1, 1): 3,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 0, 1): 1,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 0, 1, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 0, 1, 1): 1,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 1, 0): 1,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 0, 1, 1): 1,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 0, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 0, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 0, 1, 1, 1): 1,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 0, 1, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 1, 0): 1,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 0, 1, 1): 4,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 0, 0): 1,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 0, 1, 1, 1): 2,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 0, 1): 1,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 0, 1, 1, 1, 1): 1,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 0, 1, 1, 1): 1,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 0, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 0, 1, 1, 1, 1, 1): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 0, 1): 1,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 0, 1, 1): 2,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 0, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 0, 1): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 1, 0): 0,\n",
       "  (1, 1, 1, 1, 1, 0, 0, 1, 1, 1): 1,\n",
       "  ...}}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from Bayes_helpers import SmoothFullBayes, FullBayes\n",
    "fnb = FullBayes()\n",
    "fnb.fit(X_train, y_train)\n",
    "fnb.prob"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ile estymat zawiera klasyfikator \"pełny\", a ile klasyfikator naiwny? Jak będzie się to zmieniać wraz z rosnącą liczbą wymiarów?\n",
    "Odpowiedź: Pełny (2^k - 1), naiwny (k-1), plus każdy z nich ma dodatkowo estymatę P(c). Te wzory pokazują zmiany liczby estymat wraz ze wzrostem liczby wymiarów.",
    "\n",
    "# Zadanie 2c\n",
    "Zbadajmy zależność pomiędzy trafnością klasyfikacji zaimplementowanych metod przy zmieniającym się rozmiarze zbioru danych. Poniższy kod jest gotowy i początkowo nie musisz w nim nic modyfikować."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/student/.local/lib/python3.6/site-packages/ipykernel_launcher.py:26: RuntimeWarning: divide by zero encountered in log\n",
      "/home/student/.local/lib/python3.6/site-packages/ipykernel_launcher.py:27: RuntimeWarning: divide by zero encountered in log\n",
      "/home/student/bayes/Bayes_helpers.py:63: RuntimeWarning: divide by zero encountered in log\n",
      "  results [i,j] += np.nan_to_num(np.log(self.prob[j][self._hash_example(X[i])] /self.prob[j][SUM] ))\n",
      "/home/student/bayes/Bayes_helpers.py:65: RuntimeWarning: invalid value encountered in true_divide\n",
      "  probabilities /= np.sum(probabilities, axis=1, keepdims=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from Bayes_helpers import SmoothFullBayes, FullBayes, plotAccuracyIterationsPlot\n",
    "from collections import defaultdict\n",
    "\n",
    "#Generowanie danych\n",
    "Xb, yb = generate_binary(5500, k = 50)  # <- Tu kontrolujesz liczbę cech\n",
    "X_train, y_train = Xb[:5000], yb[:5000]\n",
    "X_test, y_test = Xb[5000:], yb[5000:]\n",
    "\n",
    "N = X_train.shape[0]\n",
    "iterations = range(50, N, 50)  # <- Kontrola ewaluowanych punktów (od 50 do N co 50)\n",
    "\n",
    "results = defaultdict(list)\n",
    "results_train = defaultdict(list)\n",
    "\n",
    "for i in iterations:\n",
    "    for classifier in [NaiveBayes(), SmoothNaiveBayes(), FullBayes(), SmoothFullBayes()]:\n",
    "        classifier.fit(X_train[:i],y_train[:i])\n",
    "        results_train[type(classifier).__name__].append(np.mean(classifier.predict(X_train[:i]) == y_train[:i]))\n",
    "        results[type(classifier).__name__].append(np.mean(classifier.predict(X_test) == y_test))\n",
    "\n",
    "# Tę ostatnią linijkę możesz chcieć wywoływać w osobnych komórkach, \n",
    "# tak aby na koniec porównać wykresy dla kliku ustawień\n",
    "plotAccuracyIterationsPlot(iterations, results, results_train)\n",
    "del results_train[type(NaiveBayes()).__name__]\n",
    "del results[type(NaiveBayes()).__name__]\n",
    "plotAccuracyIterationsPlot(iterations, results, results_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wykonaj ekspermenty dla różnej liczby cech: 2, 10, 100, 500."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from Bayes_helpers import plotAccuracyIterationsPlot\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "**Ćwiczenia**\n",
    " - Czy rozmywanie ma pozytywne skutki dla klasyfikatora naiwnego? a dla klasyfikatora `FullBayes`?\n",
    " - Który klasyfikator z testowanych jest najlepszy w jakich sytuacjach?\n",
    " - Jaką trafność  ma klasyfikator `FullBayes` na dostatecznie dużym zbiorze uczączym? W jakich sytuacjach `NaiveBayes` mógłby również osiągnąć 100% na uczącym?\n",
    " - Czy jest możliwe, żeby klasyfikator `FullBayes` nie będzie miał 100% trafności (nawet zakładając nieskończenie wielki zbiór danych)?\n",
    " - Przeanalizuj wygenerowane wcześniej wykresy określając czy mówimy o przeuczeniu czy niedouczeniu czy ...\n",
    " - Czy można w jednym klasyfikatorze Naiwnego Bayesa łączyć cechy ciągłe z dyskretnymi? Dlaczego?\n",
    " - W jaki sposób można obejść powyższy problem? Podaj co najmniej dwa sposoby.\n",
    " - Klasyfikator naiwny robi założenie o warunkowej niezależności cech. Co mógłbyś zrobić jeżeli spodziewasz się, że pewne pary/grupy cech są zależne i może to mieć duży wpływ na jakość predykcji? Podaj przykład takiej sytuacji."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Tak. Dla Naiwnego dla niektórych danych widać dziwne skoki w jakości działania - niektóre cechy są bardzo rzadkie i pojawienie się ich wraz z daną klasą w zbiorze mocno wpływa na odpowiedź klasyfikatora. Warto też zauważyć, że czasami wylosowany zbiór potrafi być trywialny (ze względu na trafność) z powodu class imbalance rzędu 1:99 - wtedy lepiej powtórzyć eksperyment żeby zobaczyć różnice. \n",
    "2. FullBayes/SmoothFullBayes będzie lepszy dla niskiej wymiarowości, dla wysokiej wymiarowości lub/i małego zbioru danych lepszy będzie SmoothNaive. Warto zwrócić uwagę, że w obecnej implementacji podejść Full, mnożenie przez zerowe prawdopodobieństwo jest dość sprawnie zamiatane pod dywan i uzyskuje się odpowiedź ZeroRule.\n",
    "3. W podanych eksperymentach tak, jest to możliwe ponieważ generowane zbiory są liniowo separowalne.\n",
    "4. Nie, rozumowanie analogiczne do ćwiczenia dla ciągłych.\n",
    "5. W zależności od wykresów ;) \n",
    "6. Bezpośrednie użycie reguły Bayesa do klasyfikacji danych mieszanych nie jest możliwe, ponieważ prawdopodobieństwo uzyskania przez zmienną ciągłą konkretnej wartości równa się 0. \n",
    "7. a) Dyskretyzacja ciągłych i użycie Bayesa dla dyskretnych b) mnożenia gęstości prawdopodobieństw z samymi prawdopodobieństwami. Rozumowanie motywujące takie postępowanie opiera się na odczytywaniu prawdopodobieństwa zmiennej ciągłej w jakimś małym przedziale o długości $\\epsilon$. W takiej sytuacji prawd. można przybliżyć jako $\\epsilon \\cdot f(x)$. Przy $\\epsilon \\rightarrow 0$ zbiega to do prawidłowego prawd. = 0. Jednak, ponieważ w regule Bayesa to prawdopodobieństwo $\\epsilon \\cdot f(x)$ pojawia się w liczniku i mianowniku, $\\epsilon$ można skrócić. c) ew. mogą się pojawić propozycje zespołu dwóch klasyfikatorów. \n",
    "8. Po prostu modelować zależności tylko tych wybranych >małych< podgrup cech. Mnóstwo przykładów: np. analiza wydźwięku i cechy \"nie\" \"dobry\". Inny sposób to redukcja skorelowanych cech."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Zadanie 3\n",
    "Klasyfikator naiwnego Bayesa często jest używany do klasyfikacji tekstów. Przetestuj działanie algorytmów na podanym rzeczywistym zbiorze danych: \n",
    "> The 20 newsgroups dataset comprises around 18000 newsgroups posts on 20 topics split in two subsets: one for training (or development) and the other one for testing (or for performance evaluation). The split between the train and test set is based upon a messages posted before and after a specific date.\n",
    "\n",
    "Podany zbiór jest wieloklasowy, więc poniższy kod wybiera z niego podzbiór postów tylko z dwóch tematów."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import fetch_20newsgroups\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "categories = [  'comp.graphics', 'sci.space']\n",
    "newsgroups_train = fetch_20newsgroups(subset='train', categories=categories)\n",
    "newsgroups_test = fetch_20newsgroups(subset='test', categories=categories)\n",
    "\n",
    "vectorizer = TfidfVectorizer(binary=True) # Przekształcenie tekstu na cechy binarne\n",
    "vectors = vectorizer.fit_transform(newsgroups_train.data)\n",
    "vectors_test = vectorizer.transform(newsgroups_test.data)\n",
    "vectors = vectors.toarray()\n",
    "vectors_test = vectors_test.toarray()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dokumenty w zbiorze można wyświetlić w następujący sposób."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['From: ab@nova.cc.purdue.edu (Allen B)\\nSubject: Re: thining algorithm\\nOrganization: Purdue University\\nLines: 15\\n\\nIn article <1q7615INNmi@shelley.u.washington.edu> kshin@stein.u.washington.edu  \\n(Kevin Shin) writes:\\n> I am trying obtain program to preprocess handwriting characters.\\n> Like thining algorithm, graph alogrithm.\\n> Do anyone know where I can obtain those?\\n\\nI usually use \"Algorithms for graphics and image processing\" by\\nTheodosios Pavlidis, but other people here got them same idea and now\\n3 of 4 copies in the libraries have been stolen!\\n\\nAnother reference is \"Digital Image Processing\" by Gonzalez and\\nWintz/Wood, which is widely available but a little expensive ($55\\nhere- I just checked today).\\n\\nab\\n',\n",
       " \"From: stephens@geod.emr.ca (Dave Stephenson)\\nSubject: Re: Clementine Science Team Selected\\nNntp-Posting-Host: ngis.geod.emr.ca\\nOrganization: Dept. of Energy, Mines, and Resources, Ottawa\\nLines: 32\\n\\nnickh@CS.CMU.EDU (Nick Haines) writes:\\n\\n>In article <stephens.734792933@ngis> stephens@geod.emr.ca (Dave Stephenson) writes:\\n\\n>   Remember the first government scientist in the British Empire was\\n>   the Astronomer Royal, who was paid [...] from the Department\\n>   of Ordinance Budget (i.e. the military). Flamsteed House (the original\\n>   RGO) was built out of Army Surplus Scrap ( A gate house at the Tower of\\n>   London ?), and paid for by the sale of time expired gunpowder [...]\\n\\n>At the time, astronomy was vital to the military, in that navigation\\n>and cartography were of primary impoortance to the military, and good\\n>cartography was impossible without good astronomy.\\n\\n>The relevance these daysis somewhat less obvious.\\n\\n>Nick\\n\\nIt still applies, except the astronomy these days is Very Long Baseline\\nRadio Astronomy coupled to GPS and Satellite Laser Ranging. The data\\nfrom NASA's and the Naval Observatory's (among others) is a vital \\nsource of data for studies into crustal dynamics, Earth rotation, and\\npurturbations. Every time there is a leap second added to the New Year,\\nremember the military and science are still co-habiting nicely. The\\nsame VLBI was used to track Gallileo as it passed the Earth, and used\\nso little fuel that it can afford to observe Ida. \\n \\n--\\nDave Stephenson\\nGeodetic Survey of Canada\\nOttawa, Ontario, Canada\\nInternet: stephens@geod.emr.ca\\n\"]"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "newsgroups_train.data[0:2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Analogicznie możemy uzyskać dostęp do informacji o klasach:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1], dtype=int64)"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "newsgroups_train.target[0:2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "...i do \"zbinaryzowanego\" tekstu:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectors[0:2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wytrenuj klasyfikator i sprawdż jego trafność na zbiorze uczącym i testowym."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9745114698385726\n",
      "0.9272030651340997\n"
     ]
    }
   ],
   "source": [
    "nb = SmoothNaiveBayes()\n",
    "nb.fit(vectors, newsgroups_train.target)\n",
    "print(np.mean(nb.predict(vectors) == newsgroups_train.target))\n",
    "print(np.mean(nb.predict(vectors_test) == newsgroups_test.target))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.02679464, 0.02677336, 0.02677336, 0.02651596, 0.02539572,\n",
       "       0.02480149, 0.02467892, 0.02451428, 0.02360392, 0.02324052])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sort(nb.prob[0])[::-1][0:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.02464983, 0.02444116, 0.02297182, 0.02288026, 0.0228593 ,\n",
       "       0.02260498, 0.02260498, 0.02258878, 0.02256857, 0.0225666 ])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sort(nb.prob[1])[::-1][0:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([13687, 10240, 20707, 16054, 10053, 21418, 21642, 15836,  3796,\n",
       "       22360])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.argsort(nb.prob[0])[::-1][0:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['lines', 'from', 'subject', 'organization', 'for', 'the', 'to',\n",
       "       'of', 'and', 'university', 'thanks', 'graphics', 'any', 'in',\n",
       "       'posting', 'host', 'nntp', 'is', 'it', 'edu', 'have', 'can', 'or',\n",
       "       'this', 'on', 'that', 'me', 'you', 'anyone', 'if'], dtype='<U80')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(vectorizer.get_feature_names())[np.argsort(nb.prob[0])[::-1][0:30]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['re', 'space', 'that', 'the', 'in', 'from', 'subject',\n",
       "       'organization', 'lines', 'of', 'to', 'writes', 'edu', 'article',\n",
       "       'is', 'and', 'it', 'be', 'on', 'was', 'not', 'for', 'posting',\n",
       "       'this', 'you', 'at', 'nntp', 'host', 'com', 'but'], dtype='<U80')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(vectorizer.get_feature_names())[np.argsort(nb.prob[1])[::-1][0:30]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Ćwiczenia**\n",
    " - Dlaczego klasyfikator Naiwnego Bayesa dość dobrze sprawdza się do powyższego zadania i analogicznych?\n",
    " - Przeanalizuj wartości estymat prawdopodobieństw. Które cechy/słowa są najlepszymi wskaźnikami dla podanych klas? Jakie słowa bardzo słabo wskazują na którąkolwiek z klas?\n",
    " - Czy byłoby możliwe wytrenowanie równie skutecznego klasyfikatora z podobną liczbą cech? W jaki sposób można by to uzyskać?\n",
    " - Analizowany zbiór jest oryginalnie wieloklasowy z tego powodu możemy go wykorzystać do wielu testów wybierając różne pary klas. Pełna lista tematów: 'alt.atheism',\n",
    " 'comp.graphics',\n",
    " 'comp.os.ms-windows.misc',\n",
    " 'comp.sys.ibm.pc.hardware',\n",
    " 'comp.sys.mac.hardware',\n",
    " 'comp.windows.x',\n",
    " 'misc.forsale',\n",
    " 'rec.autos',\n",
    " 'rec.motorcycles',\n",
    " 'rec.sport.baseball',\n",
    " 'rec.sport.hockey',\n",
    " 'sci.crypt',\n",
    " 'sci.electronics',\n",
    " 'sci.med',\n",
    " 'sci.space',\n",
    " 'soc.religion.christian',\n",
    " 'talk.politics.guns',\n",
    " 'talk.politics.mideast',\n",
    " 'talk.politics.misc',\n",
    " 'talk.religion.misc'\n",
    " - Czy są pary tematów dla których ten klasyfikator działa znacząco gorzej?\n",
    " - Jakie są zalety stosowania klasyfikatora Bayesa dla tego problemy (i w ogólności)? Czy do tego problemu sprawdziłyby się reguły lub drzewa decyzyjne? Dlaczego?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Chociaż słowa w języku naturalnym są silnie zależne od siebie, w zadaniach polegających na wykrycie tematów dokumentów dobrym predyktorem jest po prostu istnienie słów=kluczy. Np. dokument o samochodach będzie zawierał słowa \"silnik\", \"kierownica\", itd. Jednocześnie te słowa są w miarę jednoznacznymi wskaźnikami tej kategorii i raczej nie będą się pojawiały w innych (zwykle nie jest konieczny kontekst wypowiedzi czy sposób użycia w zdaniu). Czyli wynika to ze specyfiki zadania. Klasyfikacja tekstów niezorientowana na tematy np. analiza wydźwięku jest już duża trudniejsza pod tym względem: \"dobry\", \"nie ... dobry\", opisy cech bez explicite używania pozytywnych/negatywnych przymiotników....\n",
    "2. Najlepszymi wskaźnikami klas są pewne słowa kluczowe \"lines, graphics\" i \"space\". Jednak dużo słów jest zwykłym szumem - stopwords i podobne. P(słowa|klasa) jest wysokie dla stopwords ;( Odfiltrowanie ich poprzez dodanie w TfidfVectorizer `stop_words=\"english\"` trochę poprawia te estymaty, ale nadal widzimy niestandardowe stopwords dla tego zbioru - jak nagłówki tesktów \"subject\", \"organization\". Możesz dalej je odfiltrowywać np. poprzez porównywanie estymat słów dla każdej z klasy.\n",
    "3. Tak, jak wyżej.\n",
    "4. ...\n",
    "5. Jak najbardziej - patrz odpowiedź do 1. Np. para \"talk.religion.misc\" i \"soc.religion.christian\".\n",
    "6. Przede wszystkim szybkość działania. Problem jest bardzo wysokowymiarowy a cały trening można zrealizować w jednym przejściu po zbiorze danych. Z tego się nota bene wzięła popularność w tekstach - o big data mówimy kiedy mamy duże N albo duże p (wymiarowość). W tekstach big data było w tym drugim kontekście od zarania dziejów...\n",
    " Inna zaleta to bardzo prosta modyfikacja dla strumieni danych a także pewna interpretowalność. Pewnego rodzaju minusem natomiast jest potrzeba ręcznego modelowania/wybierania rozkładów i oczywiście naiwność. \n",
    " Drzewa decyzyjne raczej radziłyby sobie gorzej na tym problemie, cechy są binarne i jest ich bardzo dużo więc kolejne splity często nie niosą żadnych informacji. Pruning ograniczyły wykorzystywanie (i identyfikację) wszystkich słów kluczowych, co więcej może się zdarzyć że przy ograniczonej głębokości drzewa żaden z warunków podziałów nie zostanie \"zaktywowany\" (dane rzadkie, ograniczona liczba podziałów, które pytają się o jedną cechę, a jej brak zwykle nie niesie wiele informacji np. z powodu wieloznaczności słów). Z drugiej strony potencjalnie drzewo mogłby się nauczyć pytac o jakieś frazy \"Czy jest słowo machine?\" \"Czy jest słowo learning?\" - jednak użyteczność tego jest ograniczona poprzez reprezentację bag-of-words."
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}