mnist_tutorial_DBN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the libraries\n",
    "from DBN import DBN\n",
    "import torch \n",
    "import torchvision\n",
    "from torchvision import datasets,transforms\n",
    "from torch.utils.data import Dataset,DataLoader\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import math\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loading MNIST dataset\n",
    "mnist_data = datasets.MNIST('../data', train=True, download=True,\n",
    "                    transform=transforms.Compose(\n",
    "                    [transforms.ToTensor(),transforms.Normalize((0.1307,), (0.3081,))]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "mnist_data.train_data = (mnist_data.train_data.type(torch.FloatTensor)/255).bernoulli()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number shown is the number: 2\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Lets us visualize a number from the data set\n",
    "idx = 5\n",
    "img = mnist_data.train_data[idx]\n",
    "print(\"The number shown is the number: {}\".format(mnist_data.train_labels[idx]) )\n",
    "plt.imshow(img , cmap = 'gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# I have have set these hyper parameters although you can experiment with them to find better hyperparameters.\n",
    "dbn_mnist = DBN(visible_units=28*28 ,\n",
    "                hidden_units=[23*23 ,18*18] ,\n",
    "                k = 5,\n",
    "                learning_rate = 0.01,\n",
    "                learning_rate_decay = True,\n",
    "                xavier_init = True,\n",
    "                increase_to_cd_k = False,\n",
    "                use_gpu = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------\n",
      "Training the 1 st rbm layer\n",
      "RBM fitting: 6000it [00:54, 110.06it/s]\n",
      "Epoch:1 ,avg_cost = 0.5687551498413086 ,std_cost = 0.11034592241048813 ,avg_grad = 12053.50390625 ,std_grad = 1995.814208984375\n",
      "--------------------\n",
      "Training the 2 st rbm layer\n",
      "RBM fitting: 889it [00:04, 220.36it/s]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-6-3b4f1def2c7a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m10\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m \u001b[0mdbn_mnist\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_static\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmnist_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_data\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mmnist_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_labels\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mnum_epochs\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/machine_learning/Deep Belief Network/Deep-Belief-Network-pytorch/DBN.py\u001b[0m in \u001b[0;36mtrain_static\u001b[0;34m(self, train_data, train_labels, num_epochs, batch_size)\u001b[0m\n\u001b[1;32m    105\u001b[0m             \u001b[0m_dataloader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDataLoader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_dataset\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mdrop_last\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# create your dataloader\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    106\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 107\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrbm_layers\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_dataloader\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0mnum_epochs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    108\u001b[0m             \u001b[0;31m# print(train_data.shape)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    109\u001b[0m             \u001b[0mv\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtmp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mview\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtmp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mFloatTensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;31m#flatten\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/machine_learning/Deep Belief Network/Deep-Belief-Network-pytorch/RBM.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, train_dataloader, num_epochs, batch_size)\u001b[0m\n\u001b[1;32m    224\u001b[0m                 \u001b[0;32mif\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muse_gpu\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    225\u001b[0m                     \u001b[0mbatch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcuda\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 226\u001b[0;31m                 \u001b[0mcost_\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mgrad_\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mnum_epochs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    227\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    228\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/machine_learning/Deep Belief Network/Deep-Belief-Network-pytorch/RBM.py\u001b[0m in \u001b[0;36mstep\u001b[0;34m(self, input_data, epoch, num_epochs)\u001b[0m\n\u001b[1;32m    197\u001b[0m             \u001b[0mlr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlearning_rate\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    198\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 199\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcontrastive_divergence\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_data\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mn_gibbs_sampling_steps\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mlr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    200\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    201\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/machine_learning/Deep Belief Network/Deep-Belief-Network-pytorch/RBM.py\u001b[0m in \u001b[0;36mcontrastive_divergence\u001b[0;34m(self, input_data, training, n_gibbs_sampling_steps, lr)\u001b[0m\n\u001b[1;32m    135\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_gibbs_sampling_steps\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    136\u001b[0m             \u001b[0mvisible_probabilities\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_visible\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhidden_activations\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 137\u001b[0;31m             \u001b[0mhidden_probabilities\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mhidden_activations\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_hidden\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvisible_probabilities\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    138\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    139\u001b[0m         \u001b[0mnegative_visible_probabilities\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mvisible_probabilities\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/machine_learning/Deep Belief Network/Deep-Belief-Network-pytorch/RBM.py\u001b[0m in \u001b[0;36mto_hidden\u001b[0;34m(self, X)\u001b[0m\n\u001b[1;32m     70\u001b[0m         \u001b[0mX_prob\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmatmul\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mW\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     71\u001b[0m         \u001b[0;31m# print(X_prob.shape)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 72\u001b[0;31m         \u001b[0mX_prob\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_prob\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mh_bias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;31m#W.x + c\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     73\u001b[0m         \u001b[0mX_prob\u001b[0m  \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msigmoid\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_prob\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     74\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "num_epochs = 1\n",
    "batch_size = 10\n",
    "\n",
    "dbn_mnist.train_static(mnist_data.train_data,mnist_data.train_labels,num_epochs , batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlEAAAJCCAYAAADgPpLrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzsvWmsZUd19v/UvZ7dxnObdrs9YLeH9oznAQMxDmaSIxJBUPLKf0JCPpAokYgS8r7K1yjKh4gvbxL8KsiRgkggIUzBgDG2scFT43nAbo94bNsY4xlw3/3/0HdX/+rpW9XnnnvO6evj9UitrnN3ndpVtVbV3mc9a61KXdcpEAgEAoFAILA4zGzvDgQCgUAgEAi8EREvUYFAIBAIBAJDIF6iAoFAIBAIBIZAvEQFAoFAIBAIDIF4iQoEAoFAIBAYAvESFQgEAoFAIDAE4iUqEAgEAoFAYAgs6SUqpXRhSunelNL9KaXPjKpTge2DkOf0IGQ5XQh5Tg9CltOFNGyyzZTSrKT7JF0g6TFJN0n6WNd1d4+ue4FJIeQ5PQhZThdCntODkOX0YYclfPd0Sfd3XfegJKWU/l3SRZKqyrDzzjt3u+++uyRpbm6u2vDMzGAGMr4AppSq9fxebJ/XWm3wXv7iucMOOyx4bdOmTUU9ts8+eHutPtX62Kr33HPPPdt13f4LfnELFiXP3Xbbrdtrr70W7P/s7Gwut+Zt0LmvfUeqz2NLh1iPfZVKmbV0dND+sv2WPrDs88TPGzduHLksJWnXXXft9txzT0mlPkvSr3/961zecccdq+PhWDl3Po+tdcvP/J7LqYZB9wHXj5ae1ur5PPHa66+/nss1HXvhhRf06quvDqJIi5Lnrrvu2u2xxx4LNjSo3i7U34W+z88tfeC1UaxNtjHovu31Bn1+1PrgfXz66afHsjZbe22tLz4ermHWc1mwnut3bT/0PtWecw7OJe/VmmOuK+9f61rtuVzr3/PPP6+XX355IKVYykvUakmP4vNjks5ofWH33XfXe97zHknSL3/5y+IaBbTLLrts6WBjojgZ3NylctJeffXV4tpuu+224LWddtqp2ieWf/WrXxX1egWXSiV8+eWXi3pUhhUrVuTya6+9VtTj3Pi4agvAlYHz9sUvfvERbRuLkudee+2lP/qjP5K09Xy85S1vyWXKiHMjleOuPYClclG63uy8884Lfm/XXXettsF+9C/1PX7xi1/kso+L4Py2XoAo5xdeeKHaRuvlhJ///u//fuSylKQ999xTv//7vy9JWrlyZXHtqaeeymVe8/H0L2GS9NJLL+Wyrz/Kydc39YCypk5J5TyzPeqDX6O+cQ+QSp1g2XWRsth3332r93r22Wdz2XWs3xe+8IUvaEAsSp577LGHfud3fkfS1vtC7QdO6yXixRdfXPD7UjnfXDtSuS/yms8955Rl6pO3wTn1PlEOrZcyPkt8n609aJ9//vmi3t57753Ln/3sZ8eyNvfaay998pOf3KrPDq4Rl/szzzyTy5w77k+StHHjxuK+BNdjbT/1e/Nevjdy/9hnn31yuaVHXFe+/p577rkF25NKebL92nPin/7pnzQoxu5YnlL6ZEppfUppvT8AA28sUJavvPLK9u5OYIkIeU4PKEt/aQ288RBr842DpViiHpe0Bp8Pmv9bga7rLpF0iSTtvffeXf8m7b9G+OuBb7b+5s1fDPwV5NYc1vO3Tf4iYRtueaBJnIpMS9lCfVzo+1L5Js72/Fcg++S/njhOjssXWs2c38A25UlZHnTQQV0/Dz5v/LXGXyL+Ek0doNXOrQ685nNPS0atLNV/Hfkvy1obLgc+qDgOlwPnxn8dUe78pei/iofAotfm6tWrM2XAX3RS2W+Oxy1mXLe8tv/+JcPBX5O+NilfWkD8XrQa0+rllijKl7+6f/aznxX1atZk1yO277+saxZkXx8HHHDAVnW2gUWtzZUrV3a9vraoEfartadxDvwFjd/zfYx7VYvWoQ6wDZdRzdLi1kLqL79DPZFKffB5qu3pLZeEAbHotXnQQQdlNxi/P8fEa24lptxbzzLujc7MUF8pJ1qvJGm//fYbqE/+DrDQfaRST/lc4/7gcGsWdZj7me///fgXQ3svxRJ1k6S1KaXDUko7SfpdSV9fQnuB7YuQ5/QgZDldCHlOD0KWU4ahLVFd172eUvoTSd+RNCvp813X3TWyngUmipDn9CBkOV0IeU4PQpbTh6XQeeq67luSvjWivgS2M0Ke04OQ5XQh5Dk9CFlOF5b0ErVYdF2X/Q2cc65x0O4fQ56dfHQr4sP9WWq+AC0fD/pMOI/KNlh2/wHyw+Se3Z+LbXjkQy3azeepxRePGq2AAfozuZ8W5UBfE/dd4Nhac0Ue2yNLej8Uqe07dd999+Uy59c5cn6PvgOuhx6hSbBN8vSur8OEpS8Wc3Nz2VeiFRXV8nWiLwrruU9Qy0+i5g/ofhLUb8rC1ybHQl+Qt771rUU96jD9bVx+bM/1vhZG7n4sfb1hc/RtC3Nzc1vNaw/2n3rrMqIcONcezdXyZeR8c6/2vtWiaj2qkf3gPuC+hoyYa0X91vYf73ttrUtb69u40Pfd94ZapLnvGYxk47PW5d7yP+WaZhuuEzV/ulY066OPbglYdP856gTn3/0pW6A82T9/vvY65u8CLcSxL4FAIBAIBAJDIF6iAoFAIBAIBIbAROm8lFI2yzt9V6Ns3LTH75GycfqqleGYZvhWuOfPf/7zXH788S1RqG5mroVlOy1CE2srzJ59crMiv0dTpFNg46IKeszNzW1FrfWgOb2WHFWqZ6ltpTFomdPf9ra35bLL303OPVxvOPeksZyKJL1EObcS/zm9yz7VEgQu9L1xoV9rTLcg1U3yTmdxDbcSxroMCeoBKVhHLdzaU0wwOSipC5d7zXxPXZbK/chlzc/UHac9+2tDhMcPhJRSpqd8/6Qucf/wcXLva6V1oCx9P6iFurve1NIaOF1Duphrx2XJPefAAw+s9p1opbLgHHq9SeU+7HXF78exttYc9bFGcUvlM9Vp0loCXd8bSTG2TvCg/nNfcT3iGGvuPFKpEy5ryrB1msAwCEtUIBAIBAKBwBCIl6hAIBAIBAKBITBxOq83k7oZjVEBNL21zuWhGbeVAd1Nm2yT7XmkAs1+pJjcpMrIAprumZlZqkcWeNQCTcZusqRJlX13imQx0QXDgLJsZfvlNTfP1ygDj955+umnc9nN6aRraMIlleD1CNebJ598MpdbZ7J5dFcPp1WpNy5nmti5HlxvfMzjACNnHbWDTZ1GoXy55py655w7FXDooYfmMqnrgw46qKj3k5/8JJdvu+22XH7ssceKepQ7x+FzyvYpM+qoVMrMqbLaWWCe0bn/3rgod65Njy6sHSbsFEqNznLdJIXilDn3RcqydSC1R/gRpKE4p+5eQd0jBbhq1aqiXk3fpVIHqDfueuGU1zgwNzeXn00tCqsVMcfvcR6PPPLIoh5p3Z/+9KfFNT7bPPqZ4PzXzpOUpDvvvDOXuW9622vWbEnwXpOtVFKT/jzk3sTv+bvBMK4TYYkKBAKBQCAQGALxEhUIBAKBQCAwBOIlKhAIBAKBQGAITDxjec+9Ox9dy+Dtvgvk8emb5Lw4v+f8MLlTcrHOx9fCV93PgDz59ddfn8uHHHJIUe/000/PZfqFeJgpuXv3OyHXS47Z/blap6aPAsxw3TopnqBv00Lt9Wj52rjPETls1jviiCOKeuT66fd00003FfUYEk2+3Llz9peh+J4pmD4TLmfKiD4eHhLvKQfGgdnZ2byGXHc4Vq4J912o+TJ6Pa4f98W54447cvm8887LZffdoK8T/Xf4fUm69dZbc5lr5/DDDy/q0ReLeup+fLX0KH6Ne477040rtcFCcL+rWooU3+u4f9IPzH1GKEvWk8o54B7pfkVMa3DMMcfksvsd0ueM69vXB9e6+zwSrVQy3E+pv96e+8yNA0wN5HLifsj59j2YzxQ+51y/6aPofoicL+5X7utZO7WDcpZK3yT6Mro/K/dXrlNPhUO4byTXPuXpz6TeX2wxqQ/CEhUIBAKBQCAwBOIlKhAIBAKBQGAILJsUBzT70QTttB9NwTRFehjx6tWrc/nBBx8srtH8yDacxqAJlCZip3YefvjhXKbp0akK3ovhv8y07fBQTVIBNN86nTfusPiZmZls2vZxso+kD5zq4vdoqnc6jzL3uT///PNz+ZxzzsllDz2+5pprcvlb39pygPoNN9xQ1Dv33HNz+Z3vfGcuH3bYYUU9zu9DDz2Uyy1z+6A0mbfhdNA40HVdpjScbqplU3fd5Gea2j3FASkgngQglRTCVVddlct33XVXUY9hz6RUnJK/+uqrc/lDH/pQLpM2kko5cbxOQXA/8nkincV+OAVUOxx4VNi0aVOeE6drqINcj74fU2bUP2+P8+P6wPVNys1D0zkfpFecOuR+zHu5uwa/d+yxx+ayu3zwueC0XO0gZJedp+gYF3pd8/txjvkM8Az0tflySuyBBx7IZT6jHNyjfB1w/km7klqXStrusssuy2Wn7km/kcZ3feNe4lQw54m0pOtO3+ZiDn0PS1QgEAgEAoHAEIiXqEAgEAgEAoEhMFE6b9OmTdkc6ofz0sRP8yCpMqk0M7MNpzzuv//+XKbpXyrpIpoU3YRHUx/NvaeeempRj6bNdevW5TIzskrS3Xffncs0t3p0CU2WTl+xH60oxnEfjNl1Xb5Hi3KkadejGjkWfscpGZqwnRriuFn2bLtf+cpXcvnaa6/NZTdZH3fccQv216M9alnZPRrqvvvuy2XX0Zr8PFJo3Nnnpc263/fB55jypSm8RdOR1vYImKeeeiqXna7fsGFDLlPuTiORuiAt43Jau3btgn3ytX700Ufn8uc+97lc9rXJ/cKjzKjDpBacWun3lXFF0O6www6Zzm5FzHFvcUqe1ygjp5O4z/ipALzG9t294uCDD87lm2++OZedQqfevOc978nl1n5cOxRbKt0ovE+kkEjhuczGfdC7tFmverrZaVeuTcrMnwd85vE7zPwvlW4xTzzxRHGNMiTl9va3v72ox/X+7W9/O5d9LZHq+8hHPrJgX6WSciR97GNkPd/XuYey7HRer2O+t7UQlqhAIBAIBAKBIRAvUYFAIBAIBAJDIF6iAoFAIBAIBIbARH2iZmdnM0ft/DR5WnKiztXze+RH6e8glXyp88jkTumz4j4O7Ad9AZjxWip5cvpfeYgu/Uno4+EZX8mze1Zk+hFxLtzHYxJh8b0PiPsFsC/klp3rrnH49H2QSp8ED1OmzJiuwGXETNacm4suuqioRx+SG2+8MZc9i/hJJ52Uywxhv/fee4t69BPysFvOUy3zsF8bF7quy74vrrf06aHe+njYz9ZpApSZzyvXO/1SmG5CKn05mH2cPjXeBk8QYCoLSfr+97+fy/Tp8NQK9NPxfYVyo957CHgv98WEUS8G9D31bPGUC/3K3A+Ra5r7pY+Ffqnu+0W/FO5xvqfRD+7444/PZc+YTd8pprDxFBI8GaKVlZx7gvs6+XzU0DqFYZTo++dj4FqlPH3N0a+N/sKe2oHryq+dddZZuXzUUUflss8V/Zi5l/u+xs/MRO4pMLhOON7bb7+9qEdZe+oG7jPU4Zpf1WL8FcMSFQgEAoFAIDAE4iUqEAgEAoFAYAhMlM4jPIRw0ENOGYJJOsjNdzT3Ou1AEz3bYzi0t0lT6T333FPUozmZ33EzJ6kQmh6devOwZMJTHvRw86ObzEcNZrh2WoLmWM6Hp7UgrUMztdNXPNSZGcql0iRMmXsmbNIEvK9ni+f8MsOum+3POOOMXCZl5zQ19dyp6Vq6AKdMPLx/HOi6rkrPUvcpT6dMKV+mGvC1SarB55XUDmXtGflJr//lX/5lLpOqkKR3v/vduUy6yWXxyCOP5DLDsn2for7xsGqp1IPaAcnSFlp7XKkrZmZmss44NVs7kLdFIXMsvq/wmuspXRtIjXtKghNPPDGXSev4qQOkpEjh+p5422235TL3RbpuSOXcOK3DFCnUG1+bTpuNA62TPjjnlK2vFz7nOG6etiCV69Z1gnI6+eSTc9npN7o0kNrzOWZaEWaW9z2BVDtdNjwFA+lf1wnqLfcOT9lRO4S9hbBEBQKBQCAQCAyBbb5EpZQ+n1J6OqV0J/62T0rp8pTShvn/9261EVg+CHlOD0KW04WQ5/QgZPnmwSCWqEslXWh/+4ykK7quWyvpivnPgTcGLlXIc1pwqUKW04RLFfKcFlyqkOWbAtv0ieq67gcppUPtzxdJetd8+V8lXSXprwa5Yc8pO2dLvwlyp87hsx55T/d7og+M88j0aeIxDgyBlkounKGazu3SF4SpGjzNPX0jGHLpKerJ07q/EcdcC1eWtubue4xKnjwmxI9z4XzQv8vlwHHS38F9wuij4z4I9K9gG35kxPr163OZOuT1eDwM5ee+eTztnDLx/tH/wHn2mq+Jy652BMEo1+bMzEzWz9bxR7zm64DXuA5cnlzTP/zhD4tr9KO54IILcpl+EVK5ljiPnuqE+wB1xf09eCQTx+V+OZSF+wdxTVMX/Wib3hfKfc9GKc9eFj5OpgtZtWpVLrfSwHDMPh/cFxn2LpU6QN9D90OkjFrHJDFNwnXXXZfLnlqBPo/0jXQ9ZP/oT+lgKh33/6n5RI1Sljxiy33cKF/6+fkeQl3l/LhvIMdDv0+plBP3zSuvvLKoR9nw2eBy51joy+g+x5dffvmC4/B9kSkZfF+n7xif+Z52p9Z2C8P6RB3QdV3vVfqUpANqFVNKn0wprU8prR/3eW6BoTGQPCnLcTuuB4bGUGvTX4QDywaLXpv+oA0sGwy1NmOvXd5YsmN5t/nnVPUUxq7rLum67tSu606dRALIwNLQkidlWYvyCSwfLGZtTiLKKLA0DLo2JxHNGVgaFrM2Y69d3hg2xcHGlNKqruueTCmtkjRQ2tau67KZzKkNUiLcBJwyoMmcJuczzzyzqEdzpofM81402/kp5OwHzZ48fdrrMey7lQWXJlU36zP0uTVPnJtWaP0AWLQ85+bmsjnWKQP2nzQYzed9Gz0YWuvhqccdd1wuM1O4VNK4nF/PHM7wV5r1GcoslSHWzz77bC6vW7euqMcQa96XpmeppEl8/LUw2iVmnx9qbfK+/kJF/WSfXcdIxXAMTmuTUvL0BzS133TTTbnsVBHv9b73vS+XPds9KTyGWzOlgVSu4VNOOSWXXbepm05tUU6ktD2VQa8vtXQlhqHWZr9fuVWK46T10UO9WY/r1NcLw9TdpYD6wfZ8P2bmcK5Tb48h7Zy7H/3oR0U9tk/d85Qz1LWW9a5GU0tb77vbwFBrc3Z2dit3gh58jlCeTs/y+cV6vl64nzr9ST3m2vTM5qRG+Uz10wnYHmVzyy23FPXuvDP75hdt+D5FHfN0OpQvZVhzl5lEioOvS7p4vnyxpK8N2U5geSDkOT0IWU4XQp7Tg5DlFGKQFAdflHSdpKNSSo+llD4h6e8kXZBS2iDpPfOfA28AhDynByHL6ULIc3oQsnzzYJDovI9VLp1f+XsVrcyrpAxa0U78fMIJJ+SyU3GEZzZlNAdNe04Z0DxP2sGjQTgWZrl2mo7RMDTPMpusVJpePdqIJm46HPq9aubIUcqzv4c7PpIaaB0Ezc/HHHNMLntkCc3KLiNGy/DwTDdT08TPax6dQZqOZmmnETlmRgC1MkS76ZjXmKXXaZ7aQbWjlOWmTZvyXLQyjNN036K6SM/6GqYZn5mKvU3ey6OieBgt5en+QKQ7rrnmmlz2g4qZuZlRSJ7RmXSC01Jc01yPTsf2693p0FHJc2ZmJsvQdYlzSv1rRX3xmq9N0ntOdXHc3NN8H2AbnF/PXE29IR3LdS+VcuEa9gzz3JtcR0kDUqd8jDWn71GuzZYbDGkw9tNPbOCcc024LPg9pz+ZLfzII4/MZV8HH/jAB3KZMnPKmDQd70vaXSqjMtlfn3vqpl9jG63I2f5zLbp9IUTG8kAgEAgEAoEhEC9RgUAgEAgEAkMgXqICgUAgEAgEhsCwKQ6GwqZNm7J/i/sJkJ8mt+s8KsNSyXOyLJUcvPOjDOOkf5PzzeS/6RPjPkzMaEtfEA9L5ffYB88ayzF7EkT6WvCa+804nz1q0O/C/VDo60H+nekDpDKcmd/xkHBy5x4uz7kjr870BFKpH/QF8cy29M/4+Mc/nsuuQ5xvyt+z3nNcrqPsYyvc3X0OxoGZmZms7z5W+sRwvbRC2ilb97GiDxP1Qyrnlf5IXo/7AP1oWj5AzIrsfTrttNNymfrn/o+Ur/shcj1yL6ntdTVft6UipZRl2cr+38pwTVmy7O3xmo+H/m5cI+6HQz9EwvMjcU+jf5uHs1NGfK6wD17P11jNh8hT7kw6J5frN/dDrlOXBeVG3Xe5sz36PXn71HX6Jvu9OI/uQ0l/pK9+9avVvnPf5J7jsqB++Lio97yvr81hEtWGJSoQCAQCgUBgCMRLVCAQCAQCgcAQmCidNzs7m016btKtZX71EESaM9mGm2pJlTgFxAN/adL2kGNmO6bJmBm0pfIwWobbuumVJkVSBh5mT3Orm7TZR9bzNhaZsXzRIJ3HDNTeL5pmnS5lxmcePOtjIdXi10jRMAyemZSlcu4pP35fKmlWUrNuOqY5u0W30YTtpmNSUmzPabJxU7M9evl4ZmFm4Sd8zdLszpBxp85IgTBbtVTOOQ8ydfM8w6gps5NPPrmoRzqPNPmGDRuKeuedd14uM3ze1x9l41QO6X/K2uXZ6/BiwqgXg67r8j1bqU/YX19X3O+4l3jaAa451xtmEqdczz777KIe6eNaWhmp3McpF9/ruB9RD/2QesrFZcT2KSe/17j3Wak87N33Wu6v7HNL7uyzj5v7sD97a6kueHi3X6Ne/eAHPyjq1bLTsw9SeXAx16bv3byXU4d0reE+4PLrddj3+xbCEhUIBAKBQCAwBOIlKhAIBAKBQGAITJTOk7aYGd0Dn2Y1RsR4xm5SGzQveoQb6RE3QfPerWzHpP1oCnZz4z333JPLzLztkT2M/HrwwQdz2U2KNG97tADNtKQz3fQ9TJTBYjA3N5f76dF0/EzKzk2stTk9//wyqe/111+fy7fddltxjXNKCs+jOt///vfnMk3HHjH4H//xH7nMLPiuQ5QRzddOvZFO8MimGg3osnST+zhAeboZv0b7+HqhzlFP/bBf0j6eRZpzRP1w6pC6v3bt2lwm9SuVNP8//uM/5vK5555b1KNusm2XJyO1nFqpUWW+PnoadzGUwWLQdV2Woe9BpMTo8uB94R5JPXXKjtf8IFrOFV0bPMM49YjlM844o6jHaE3SiB7BzD2H+4DrEJ8ZvuaoDzXKzNsYJ3odcjlxXlv0MNcm1zddCqRSvn6N+ytpc9exL3/5y7l8xRVXVPtHPeBJFBdeeGFRj89HRmqzLJX7tWegp75wTfte1+uB06EthCUqEAgEAoFAYAjES1QgEAgEAoHAEIiXqEAgEAgEAoEhMFGfqLm5ucxLO5fM0NZadlGp9EWhr4H7mzjHTbBN+muQS5dKHpXZsK+99tqi3r333rtg/zyklp/JDz/xxBNFPfrAeOqG2unqzuG6H8ao0XVdvmfLJ4pycD+lQcH5+d73vldco1zo3/aud72rqHfZZZfl8vHHH5/Lnq7ife97Xy5TN1wO1Ff6ZHioOPWylaWf68FTQUwiK/Ls7Gz2FXDfFq4L+pV46gP6L3IduF8jw43pxyBJt9xySy7Tj83TlNB3g/4P7ne3atWqXKYeffOb3yzqfehDH1qwbZ8L6sHq1auLaxwz9cD9vnqfq3FlLJ+Zmclz4vsi01BwbPTh8mtcw76v0h+G/okO7mnuG8N9gakVPISd+zHL9EmVpHPOOWfBe7nvaSvDNftb288W+jwOdF2X++A+WVyP3E/8mcIxcK/x5zDTDrROmKDcuQcv9LnHmWeeWXzmuuUe6j5W3Ne5rry9hx9+OJdbaWJ4X3929T5+i0k/EpaoQCAQCAQCgSEQL1GBQCAQCAQCQ2DiKQ56qqIVqtkKMSbVQdO6m99INXhILVE70Nj7QVN1K1Mqw609zJJmSpry/RBcZnF2aqdG03l2dDf7jhrMousHgLKPlIsfCsx5ZD03B3MefT5IFfGazz1pO97LdYNzT7kya65U0qcM5faDcvnZ++4UVQ+nPccVCk90XZfnpZVioTXHnBPqt1PNlLvrPs36XMMens61T9l4WPyVV16pheCHLJOSJ23hMuK6al2jXjql1s/bYsKoFwOmq/D9k2uT1KfrGFMXHHzwwbnsaR2477iMSAeRknZqnJQS5ed00je+8Y1c/tKXvpTLHopPcFy+/7Sy6nP8vOb0rh8yP274/sL7c99spV6gWwGpMklav359Lvth6uvWrVuwfd9DuYb5PPyN3/iNoh5TFLSe10wbxLXeek64PLkOuIfVThXxvbqFsEQFAoFAIBAIDIF4iQoEAoFAIBAYAhOl82ZmZrJZ182NjJhrRXHRZE4z+V133VXU8yip2r1IRbl5nqZTRnZ4BmaaGNmem8gZpdM6aLN2WKRUmsw5T+M6zLSFnrJxaoRjoxz80Gke3EyTvEex0eTKQ6Gl0hxNcy4jNaQy4zjn0LPPU37sr9NO9913Xy4zEsbl0IoOqmW19vn0eRsHuq7L8+IUECOhatSeVOoj17fT31x/bk4nlbZx48YFvyOVpwlwnf7t3/5tUY96QCrKKWj2ndS9U0XUZ4+uqx1m7rTduKLyeL9elh6lxPnm/ukRURx36/QD6qrPKeee8+v7IqMhL7jgglxuHY7L/h5++OFFPbplUOakJaWSYmwd4E499zE6vTcOzM3NZYrMI3X5zOIc+3hIT3Jdcb1JJT3oex73a+5d3/72t4t6nK8Pf/jDuexUJCl00ogeWUhdpK54ND33SXfN4XtDLSpQGm5thiUqEAgEAoFAYAjES1QgEAgEAoHAEIiXqEAgEAgEAoEhMPGM5T1X674jDLcl7+scJXly8sEebk3+1kNb6cNEDth9ceinwj55BmZyseR53beL/gj0n/AwU97XfUZ4L86Fn67uYdWjRtd9P8C1AAAgAElEQVR1eQzu48DP5JzdD46+J5TRueeeW9Tjqd4e9sxTwumv4OHWnA/6/LjMyc0zi7WnjKB/G7/j/kycC/ctoW6z7+5DQ5+kcaJfk/SfcLDP7rNCPxKuF/c1ZKZv1+/zzz8/l+m78W//9m9FPc4J7+WZ6j/3uc/lMteO+4JwXNQd9kEqZeNZvglPBUD0OjFO36jeL8X3IPrw0KfE/ds4B1wvfgoDfW98jdCnhiHsvg+wj/QP9Uzv1Df2yfd3PluoG77+2F8fP32uqDfuQzMJf0Vpi674muMYeM2fUfQppNzdx+ukk07KZd9rKfuvfvWr1XvxWc5n2Y9//OOi3oYNG3KZMvvABz5Q1GNqBK57T5VBP0nXRe453N/8OTnM6RDbtESllNaklK5MKd2dUrorpfRn83/fJ6V0eUppw/z/e2+rrcD2RchyuhDynB6ELKcLIc83Dwah816X9Omu69ZJOlPSp1JK6yR9RtIVXdetlXTF/OfA8kbIcroQ8pwehCynCyHPNwm2Sed1XfekpCfnyy+mlO6RtFrSRZLeNV/tXyVdJemvWm3xYEwPWaZZlCZzp2V4jVSch0/WzNZSaeqjidfNxwytpKnQwydpTuY1NzM/9thjuUxTph/kShOom/z52UPRiYWohlHKcr4NSW0alOZvnw8eHNoaF03rRx11VHGNenT11Vfnsh+GSn0g5eOh3dQpmoudYqtlvXUap0ZVSKX5nRSHz2ctI/Mo5ckDpV3nSEmyL05dUhYcz7HHHlvU47wyC7JUzuXdd9+dy0wpIUmnn356LlMn/EBjpsTgGvbDS0mps41B9x+p3MO4r/h+0e9VXCej3md7WqKVXoFUhtN0lB/XsLdHeTn9wzngnsvM45L0kY98JJepU625J3XstE4tnN3ba6UfYd+5hn1tOr2Gvo5MnrOzs7kPTidShzhudxHh3nPEEUfksuvm2WefncuesZxpYvhM5bryfrB9d/vgXL7zne/MZVKKfi/u4y4LUpP+DKE+U4/crWiYUwQW5VieUjpU0smSbpB0wLyiSNJTkg6ofC2wDBGynC6EPKcHIcvpQshzujHwS1RKaYWk/5L0513XFT+3u82vbwu+wqWUPplSWp9SWu+/BALbB6OQpVshAtsPo5BnK6FiYHIIWU4XRiFPtz4FlhcGis5LKe2ozYrwha7rvjL/540ppVVd1z2ZUlol6emFvtt13SWSLpGkffbZp+vNyW6KI/1EM7Ob10i50TzoER+kaZw6pAmP7Z144olFPSrvgw8+mMsevcRIH1J2buq9+eabF+yvt8f+OT1EMyXb8DE6ddRjVLJctWpV11NabgqnXGjabR2SzMMoPSKKL2w0RUuljFhmhIgknXbaablM2sGz47KNhx56KJc9CocHFZNKduqNsnSKlbpNU7RHCrVeWEclz9WrV3f9nHm2Y8qTOu1rk7pKnfBx81BSn1fKg/qxatWqoh5pDFJRnqme0bInnHCCaqAMKQun7KhXTgWwDaeJiX4PXICqH4ks99tvv652AC33BY7F9w/KnLJ0Sp57F/c3qVzvXNMnn3xyUe/666/PZbo2OP3DNUe5+H5R67vvpbX1J5UUHtef6+sk1uaBBx7Y9ffxfZ39bI2H8iX96f3nNX8uUTZ8znkkHNcF15/vK6Rxua97pD3lSxcQ11k+h929p6YTNXp6MZGzg0TnJUn/Iumeruv+AZe+Luni+fLFkr428F0D2wUhy+lCyHN6ELKcLoQ83zwYxBJ1jqT/JemOlNKt83/735L+TtKXUkqfkPSIpI9Uvh9YPghZThdCntODkOV0IeT5JsEg0XnXSqrZts6v/D2wDBGynC6EPKcHIcvpQsjzzYOJZizn6eLuZE5ul1y6c5bkQcnZehoD8ud+ijzbpP+Jh0WSV2UYp4fy3nrrrblMPpc+NVLpg8Dx1/yXFgJ9AZhd1f2v3K9mnHCfKILcsvvB8TN9ojyLLGXr2btrodgOhrozY737RLF9+t14GDV1iPrq/j+85j4e9CWgH4HLzr83LvSy8izX1Gn2zX0SOP9cS63UDo888khxjSkPbrzxxlymH5VUzgnTWbhPFP0wbr/99lz2zOacf47LfaKom+7jwczN1Et39O73gcX4XSwGDIl3GVHnqPuuY+w//Ws8+zxTtbh/Ddc3Q9PdJ4o+bdQ1n3uOhXuzh+lTf7nnej3uWy4LPltqJ1d4f8eJft21ThOgv1YrAz3l6X59lK8/U7i3nXPOOblMf2GpfoKApy7gXltLvSGVqU74He8715k/a6h/nBv3Het12NtuIc7OCwQCgUAgEBgC8RIVCAQCgUAgMAQmTufVqB+aG2mWdNNerZ6bLw8//PBcdtMezY3M7OrUIekJmgq9T8xwTDO4h2WzPZr+W/mzvO+pcgCsmx89ncA40M+X0zWUC8OKPTyY5nVmoHbKrkWrHXzwwbnM+V6zZk1Rj+GvDD//8pe/rBrY91YWb5YXY94nPUF5tUz248KmTZsyReK6w3ng+Dy0uUb7OcXJw7J9rAx3Z/Zkp+mYTZn9dYqNVDD1iBSBVOow9bQV0u7UDqmQFgXt8zZqvP766/mevn8QHLNTcZQZ580pHu5jpEulcl3wYGmuWamcD65TP8mBfTr66KNz2fd+rtUW/cxrPk/sO3W0dYjxuMCTPnyfZz9bKSu45ngwtI+H7bt7BNvg3Lmc2I9rrrkml31foV5x3fuBxnwuk5ZzXeT+47Qz781nNPsgbVnvQecFAoFAIBAIjBnxEhUIBAKBQCAwBOIlKhAIBAKBQGAITNwnqucgBz3R3kEOmDyn86NMQ+C8Nfl/+jcxlb1UcqwM2fYTz8mr0heCp177NZadl+ZY3J+E/lPkpZ3Ddb+tUYM8vfPvDD+mnN13oeYT5PU4Fh8X/RV4L/qpSdKPf/zjXGZIrsucPnKc09WrVxf1yNPTN8aPxWAIrfvQ0D+QMvdw88Xw88NidnY2h5S73yJ1jvrY0luuEZ8T+hm5fwzXXMu/kP2gHnmqE17jum/5IXIc3nce69PyD+L33K+q3y+GOTF+EKSUsg752uT80tfN/VC4Bmu+XlLpo3jeeecV11j3ox/9aC4//vjjRT0e+cEjPtz/kWuJ4/I9gX2nf04rBYNf455e0zWp/awaFbquy7rWSsXA/c99vNhP7s/eHr933XXXFdeY5od+pX6vmr8U/Zmk0m+VfXrggQeKesccc0wuU3dazwlPZ0Fdbx3J5H0cBGGJCgQCgUAgEBgC8RIVCAQCgUAgMAQmSuellLIJzk3ZNJnS9Oi0Xy1jsIc2kzpx2oHUXOuk+FqaAKdXaOJltlzvO03ppBN8Lmj+93HRdMrvOQVDM/Y4MDc3l+XkdAXNrOyj06qkrVqUFUNtPeyWlApN8jQ9S9JNN92UywceeGAuuzmeMqMOtegfZlb2kFmO3+eJbVJe3oafPj9uuC6RwiJ94PQkdZ9y8bmjTlMuUmmG57r1e/EzqQDPts5T5Dn/Tsuwj5x/hnVL5fhdnpw3b5/odWxcpwrMzMxkfXKqi3Nfm0OppMMpk9a8+VyxDdLmngaE65YUnq91Uk/su+tGbV59P+dYWqdVtPZq/9644WPlGqFsnWKrpVBx+qqW/sfBNeJzUktD5JRpjQr15zBlwX3S1x+fIf7c5LOG+6nrSt/+YtZmWKICgUAgEAgEhkC8RAUCgUAgEAgMgYnSebOzs9n86aa8WnSWR8DQZM7vuPmSJjs3C9MUyfY8Eq5mnndzNM3MjGRxMyevMdKklaHa6QmOq3VQc+tQ4FGAh0n7IY6cb86V94kmVsqZ2aildsZgzj1pOo9wq0XJuTm+djC0R7FQR2mmdlqSJudBKWyfp3FFcRHMcu0RWIw0Y2SL0yOMrKLZ3dcLv+e6Uzuk2033NNfzQGmnPhl1VstkLZVyYxuub4MeTkz4fPY0SYvyWwrm5ubyGFxvqdPcB31+ORbqqdfjyRA+H8w+Tp12/eYh1OyTy5LrrPWMYH+5D7g7Adv3PZh6Tj309e33Hgfm5ua2osJ61E5OaB1AXPu+w+VJuXEP9ecm9/9WBD3lwWu+r/Aa90Jfm3Q7cL2vvTc47dc/D4LOCwQCgUAgEBgz4iUqEAgEAoFAYAjES1QgEAgEAoHAEJioTxTD4p33JI9KrtT5VnLSrfBx8rnud1Hjh52rJy9KXwL3UamFwnvfOS624X5P9JVw3zGOk/31sPhxgxnLna8nz8458Pkgp8258fYY8up+J5w7ysszTbNPDN11nyiGc1PXXIc43/SncD8CcvPO07uvXu1eHtY8Duywww65765LnGPKzOVE/yOGJbeyJ/taqn3P/R/YJ2ag93vV2nY/JeoV17r7cdBXxk+Kp+xrIdXS1nowaqSU8v7a2j+55nx+KSPqn++zvR/dQm0wWzXnzfdZTwXTw+eNfW/5qNZOcvD7cvyuNzV/MZ/PSWQs52kC7pNVS3HgewjnqHXSBcfne1nNJ811gs/2lk8s67VSElEWvK/7uFEnfG/lPsNUHN6nYU4TCEtUIBAIBAKBwBCIl6hAIBAIBAKBIZAmET6db5bSM5JelvTstupOAPtp+/djUn04pOu6/bddbXDMy/IRvbnmcVuYRD9GLksp1uZ26sM4ZRlrs8QbXZ6xNifbh4FlOdGXKElKKa3vuu7Uid50mfZjOfRhqVgOY1gOfVhO/RgWy6X/y6Efy6EPS8VyGMNy6MNy6sewWC79Xw79WA59IILOCwQCgUAgEBgC8RIVCAQCgUAgMAS2x0vUJdvhngthOfRjOfRhqVgOY1gOfZCWTz+GxXLp/3Lox3Low1KxHMawHPogLZ9+DIvl0v/l0I/l0IeMiftEBQKBQCAQCEwDgs4LBAKBQCAQGAITfYlKKV2YUro3pXR/SukzE7zv51NKT6eU7sTf9kkpXZ5S2jD//96tNkbQhzUppStTSnenlO5KKf3Z9ujHqBCynB5ZSiHPaZJnyHJ6ZCmFPJe7PCf2EpVSmpX0fyW9T9I6SR9LKa2b0O0vlXSh/e0zkq7oum6tpCvmP48Tr0v6dNd16ySdKelT8+OfdD+WjJDl9MhSCnlqiuQZspweWUohT70R5Nl13UT+STpL0nfw+a8l/fUE73+opDvx+V5Jq+bLqyTdO6m+zN/za5Iu2N79CFm+uWUZ8pwueYYsp0eWIc83hjwnSeetlvQoPj82/7fthQO6rutPyHxK0gGtyqNESulQSSdLumF79mMJCFnOYwpkKYU8M6ZAniHLeUyBLKWQZ8ZylWc4lkvqNr/OTiRMMaW0QtJ/SfrzruuKY8on2Y9pRchyuhDynB6ELKcLIc/NmORL1OOS1uDzQfN/217YmFJaJUnz/z897humlHbUZkX4Qtd1X9le/RgBQpbTI0sp5DlN8gxZTo8spZDnspfnJF+ibpK0NqV0WEppJ0m/K+nrE7y/4+uSLp4vX6zNXOvYkFJKkv5F0j1d1/3D9urHiBCynB5ZSiHPaZJnyHJ6ZCmFPJe/PCfsFPZ+SfdJekDS/5ngfb8o6UlJv9ZmTvkTkvbVZq/+DZK+J2mfMffhXG02Od4u6db5f++fdD9CliHLkOd0yzNkOT2yDHkuf3lGxvJAIBAIBAKBIRCO5YFAIBAIBAJDYEkvUdsrk2pgPAh5Tg9CltOFkOf0IGQ5XRiazpvPpHqfNie+ekybHeA+1nXd3aPrXmBSCHlOD0KW04WQ5/QgZDl92GEJ3z1d0v1d1z0oSSmlf5d0kaSqMuyyyy7d7rvvLkmanZ2tNjw3N5fLMzOlsWzTpk3VawSv8TvzfV3wXjvsUE4HXzBZ5nekciysx/s4Bq03KLwN9vHZZ599tuu6/bfRxKLkudtuu3V77rmnpK3l4PNT62Ptmn+fc9WSeQu1Pnl7g8piUD1s/Ujh92q65vWeeOKJkctSklasWNHtvffCx0/VxudzyrkbVJ4+37V58P2iNV+1vg+65njt9ddfL65xj/Br/F7tvsRzzz2nl19+eRCFW5Q8d9lll26PPfZYsKFaX1r77LD7U23deh8G7VNN5i0dIrwe22/p8qD7zzPPPDOWtbnrrrt2b3nLWxbsJzGoQaQ17lo9b3/Q52br77U9orX/DapHrefhIH166aWX9Nprrw2k+Et5iVook+oZXiml9ElJn5Sk3XffXR/4wAckSStWrCjqcXJeeeWVXN5tt92Ker/4xS9yeZdddslln4z+ZU2SXnihyM1VbMi/+tWvctkfIr/+9a9zmZvKq6++WtTjWPgd9s/BejvuuGO1nsM37h477bRT8fmXv/xlLn/uc597ZICmtylPyvItb3mLPv7xj0vaepyUHxeYKzY/s/8vv/xyUY8LwPWhtrH6BuBt9th1112LzzvvvHMuc669vRdffHHB7zjYd1/IHEtN16Ryfv/mb/5mJLKUSnnuvffe+vSnPy1pax3jWiJ8HVCGnK/XXnutqNfSfd6ba9NfCtgG5e5zx7XJ9ny91B4Kzz33XFFvv/32y+Vnn322uMbvUbbep/5en/3sZzUgFrU2V6xYoQ9/+MPFvRbqC8suY+o398tBXz6lcr7ZPvcm7wdl6WuT7bX2Wdbj+L0e1y33LK/L/rb2n3/+538ey9rcY4899NGPfnSrvkjl3LVeiAiO29sjfKysy/Jee+1V1KutR+8f1wvn3/cf9oPr1uu1dIL7P/vha7Pfj77+9cGzSCzlJWogdF13iaRLJGm//fbr+oXmm9jzzz+fy5wATsx8G7nMhe6T8bOf/SyX/QHIjZsLlS9o0tYb90L9k0phsm1/eFC5+l8W0tYLmEre+uXDh5g/yFsP9mFBWR544IFdP9aXXnqpqMc5bW3AtV/8vnlyHn2uag8rHz838ZaVhHPaeojz4cwNxdvjWPwa9Zd647/sWhvdUkB5HnLIIV0/Z77mqLf8QeL9pK5ybN4eZdb61bnPPvvksm+YNb1yufMliPelbKVSh/fff4sxwR8QfHHye3EslJm/oPR6OgoLNO6dZbnvvvt2/fz7Syr7Qn30+WDfWuuK7fv+SVnyIeb7pz+se/j+yXuzDe8Tx9KyKnIvcT2k/Pg93+tqP2qXCn9u9vdtWcy41/h4KMPWfkWZufWXcuLa9x9TXO98zrvhhGB7rou1FzH/YdwzI9LWcuG9OWe+N/XrZjFuTktxLF9umVQDS0PIc3oQspwuhDynByHLKcNSXqKWWybVwNIQ8pwehCynCyHP6UHIcsowNJ3Xdd3rKaU/kfQdSbOSPt913V0j61lgogh5Tg9CltOFkOf0IGQ5fViST1TXdd+S9K1B68/MzGTfkp///OfFNfqckKt2zpIc66BOg+7PQi6V7beiPFjP+VZy97yvc7vse8vPh/W8T+TkOX73H3AfsUGwGHmmlLI/hHPnnF+OxX0X2GeOs8WJuy8dOXzOjcuIvhvUB5/fJ554YsF+0GdPKsfsfSLYD/chYhuUpftADePftti1+frrr+c12Yqm43jcl4V6TP8M18Waf4bfi/rhvj219djqO/2jXD9WrVqVy0899VQuu27TJ9NBnyB+z3Wnx2LW6GLXZr8fev9re4avuZqvoc8vZe57MOvSz8z9TWr+gD4/NYdxB3WPvnmt9deKwmz5rg3qzE0M89zs9wAfA8dH+LriWLn/udypw76vUR7c4w844ICiHr9H2baerxyH+51x3a5cuTKX3XeW68z9r6inHAd9k6UtwWU+zy1ExvJAIBAIBAKBIRAvUYFAIBAIBAJDYOwpDoiUUjaTeU6mGiU2KFXC8EapNPd7Phfem2HUbo7ed999c5nm0UceKdOB8Br73srRQvMi7+PteZ/4mSZVp3ycLh0HeirNTco04XLMbqatmUzdjN8yyddCUT3c+qCDDlqwfaeJOL8M3fVUEzSP815uYiYt0qKkaH4eV9h0C7Ozs9nM76HDNMMz5N9paKJF2XGOXZ6cI86/t8G5oyx8X2HfKQumQHFwPbZy27RSonBunFIbJOHwUtB1Xd5PndZgigrqtO8ftTQgHs7Oei3q8Jlnnsnlln5zv/c9nTrg1wi2z3H5dzgXTrnWaEWX+SSwadOmvCZ9vVDn2E+XBeeE427lVXvyySeLa9xra64pUkmRHXLIIT6cjDvvvDOXN27cmMu+5ih3yslTh9T2BP/MPdr73q+JSaU4CAQCgUAgEHjTIl6iAoFAIBAIBIbAROm8TZs2ZVOi01Q0sZH2cXM0zXI0zztlRzOu0xOkJGjSXrNmTVGP5kJSSh5lwIid1jEftSzL3h7H5eZzmko5RjdH185BGyV6E3Hr7LzWGWc0TbcyYfN7fo1mas7NW9/61qIePzPaw83ZHAu/s2HDhqIe55c0gUdv0ezdygT+9NNP57JHk9aiu0aJubm5rGsup9qRNC1qgWZy10WufV/fnMtaRnepnC/uHa6Lxx57bC5z/dWOspHK8Ttl+fDDD+fyoMeNOGXQz9soM5YTjJx16ox7XysTN+eUa7PlNuBtMCKPbfg64N7N/vmaO+yww3KZa51UvVTK7MADD8xlUopSqQO+5mr9bUWkTgJOU1G3uDZ9XVHWfKa4fpA683tR9tznTj311KIe1zv3UH9G33fffbncOjmEY+F+6hQ/rz3+eJm/lFF91FnfB/o2FrM2wxIVCAQCgUAgMATiJSoQCAQCgUBgCMRLVCAQCAQCgcAQmKhP1MzMTDWTLv0+yE+73wX5UvogOFdP/ts5+Jo/kvvH3HjjjbnM/jq3fvfdd+fyunXrctl56bVr1+byTTfdlMvuM9HyVSCf3fIFod/POEAfGueP6b9SS/8glePkd1w3OE7PMMu65N+PPvrooh79KSi/W265paj36KOPLtg/59gZIk++3dNV0MfD+16Tkf+95b8zKtCPppVWhHB/CsqXa87TPnB9u35TD7gn+JzU/CY9ezL1j3sJM5RL9Uzsjz32WFGPKVE8tUftJATqgLTFn2tcqSy6rst7mftlUkY1f0Kp9F9pZV/nXLnvG33LuEYc9INjG/Q/k6Q77rgjl++6a8tJKRdccEFR793vfncuUx/oRyeVqWrc547z1sri7T5S48Ds7Gw1pUMtrN99tWpZyn2/4rry5xJ9ALnn+9xxvX/nO9/J5QcffLCox3llOgXfL3z99PB3A+7Rrs/cr6ljfq++T4s5TSAsUYFAIBAIBAJDIF6iAoFAIBAIBIbAROk8aQvt5Gb82mGEbpakmY3maA+Hpdna6SHWJcW2fv36oh5NxjRlMnRXKs2SNP976O29996by6TfPLs422uFUXP8rVQQ4wCzzzudVwslb4XnkgpxEytN6J5CgLQMTd7HHXdcUY/zfemll+ayZ5+n2ZdhzsyoK5VjoYwOPvjgoh776/RNLduw16sdMjpKkM7zlCCugz2cYuAYSDO4aZ3tO63G+ffDUQnuHyx7qPpPf/rTXF69enUuO7VcoxaczmvR0zWKg3okbVkTTsGPCjMzM3k/8P2Tc0UqysPKuQbZhqe/4HyQ6vTPhx56aC4fccQRRT3ui6TcfB3QlYHh8Z5Zm+Mi5ePPHK7p1vrm3to6HH1cmJuby2vG6VlSU5ShU2CcE8qW60MqaVy6NkilvnNPYOZxqXy+3nDDDbns1CEpXu7x7r7AvZz7yv3331/UYxsuJ7bJ9Vg7ZDkylgcCgUAgEAiMGfESFQgEAoFAIDAEJkrndV2XTbRudie1U6NKpNJk6abNWr0zzjijuMYIOlINnh2c2W4ZHUTTtCS97W1vy+VWplOaQ2lmvvrqq4t6rWgszhvH79lbnWoYNVJK2RTq8zZoFl/WI/3hES80pzs1RBPx8ccfn8seKXT55Zfn8k9+8pNcJpUglRTEOeeck8sexUc6gbSDUws0WXvkGHWF+uWm6BqdNkqklPK68/nneqSs3eTNayw7jUKa26lDmtrZxlNPPVXUO+aYY3KZkZe+r3BeSRUyakiqH9DqkbiUjc8TqQGub1+L485Y3kLtkGSPdKLMSVn53nT44YfnslNIpGrpssCoZ6mk8Hhfp38oyxNPPDGXnZK57rrrcpn7xRNPPFHUo055JBrHQj13XXY3inGAe63v8+w3XQd8HdROGvDIOrbn80r6nm34/s+oSu6n5513XlGP+wDvxX1cKvcI0uutLPueHZ1rnzrm89TL3ddDC2GJCgQCgUAgEBgC8RIVCAQCgUAgMATiJSoQCAQCgUBgCEzcJ6rnXN3Pg9xyLXu5f6a/gvOj9Jk466yzimvklcnVO7fL0EqGgjovTd8I+kx4lmvy+Az99Ey69PFwvwD6V9R8MKTxh8XTv81909gv+pe4PwW5afLePr8Ma/WQeNZlaO0DDzxQ1Pve976Xy/SXO+qoo4p673jHO3KZvhCe9Z7joi+Cp2CgX10rrQf133033G9o3PBwd+ogw/Jbfnj8jsuMfhIeWs40BNQP9ys6//zzc5m677rIa1/4whdy2VMX+En0PTz9CH1LWj5RrVQIiwmfHgYMiff90+v1cN8YjpP9pz+pVM6v+0Q99NBDucz93lM+0NeGPoW33XZbUY/7JK+dfvrpRT2uabb33e9+t6hH3aPeSeVY6Gvj8zQp9PJo+Smxz60M7JwTzyLe8pPjKRCcY/dXpB8on9GedobX6B/oKYS4BqmXvi9yTXv6EPaJ1/w9pN/nI8VBIBAIBAKBwJgRL1GBQCAQCAQCQ2DiBxD3JnA3xdH8SNrHze40xdEESApMktasWZPLbgIl1UPq0OkW1qMZ2zORM3yS43LzJakd0kPeHk2PbpalyZamSD+gddzh0wyJd3BsNWpPKmkjmqU9U3ErczgpWMrS+8Y+Ub+OPfbYoh7TVzBUt0XdUA6ua9QNNzGz76R/PAvyJMKoSc86hUwTP6md1hpmaLmnpeBY77nnnuo1yszXEueV/XXa9bLLLstlHmDrIdCnnHJKtb8Ex+9yqR2I7ik/eqqglaJlKZidnXMkwAAAACAASURBVM00nofuU/eZBsTTanB+mJrDZU6d9nFy/zz77LNz2WVZSznCv0vlnk4697TTTivqcS/8wQ9+kMtOXZFm9nVL2dbWujS+rPNESinft3U/ytDpKLp3cG/00yF4zWk1PqdIf3rqFq45urR4dnTOK9ez94kHy9MVwGVRO6hYKvWWa8D1vt+T4wDiQCAQCAQCgTFjmy9RKaXPp5SeTindib/tk1K6PKW0Yf7/vVttBJYPQp7Tg5DldCHkOT0IWb55MIgl6lJJF9rfPiPpiq7r1kq6Yv5z4I2BSxXynBZcqpDlNOFShTynBZcqZPmmwDZ9orqu+0FK6VD780WS3jVf/ldJV0n6q221NTc3l3lz52wZikufGK9XOx6EfkRS6X/kodjk2hkq68e5MHSWKRM8HJahvPSncK749ttvz2X6WPnp5/RV8GMn6PPBeXIO1314eoxKnnNzc9nfwmXE0Fj2y+vVjhNxvwvOh/utMfyVPib0cZFK7psy4pE9UqmH5Ow97cAhhxySyxyv+6DQn8t9vViX43CevnYEwSjXZtd1WVaeHoNzwjBz9x1iP+mX4vKkzwP9HaTSN4I+c76GuS7oi+N9r51sf8EFFxT16O9BnwkP36bOep84H7zm/ol9G/73Ucmz67q8FtxvhPekXFzn6HtDmbd8VN0nhbLlWvI1wiOVWPY9nXsw08X4GCkz6pqnWGn5vVCXa/uUt0+Mcm2mlLLcWmuJz0b3peUaoZ9Sy8eqNV9Md+C+Zu9+97tzmbLwtDMEn6n0XZRKWbRSp1BOjz76aHGNY6Eu+jy5fAfBsD5RB3Rd1yebeErSAa3KgWWPkOf0IGQ5XQh5Tg9CllOIJTuWd5vNC9XMVCmlT6aU1qeU1vsvkMDyQ0uelOWghwwHth8WszYnndAzsHgMujZjn13+WMzaHMY6Epgchk1xsDGltKrruidTSqskPV2r2HXdJZIukaSVK1d2Pc3kZsTawm+ZFJkd2qkFtn/VVVcV12hiZCirhxwzbQKpGE8nwNBbmpn9wcQQT5oR3axP86WPi/emKdepMqcatoGB5ElZHnjggV0/Bs/ITFmyX04xUkZMT+AmVn7P54M0GE3CToPWMlI7FcCx0OztIdD8ntMOBEPF3fRPqo+0k2eZXuQL61Brc/Xq1V1PlXo/qZ9cLzWaSirXEmUrlevPU1EcfPDBuUx5Hn744UU9muu5zlwXSQGxDafuub6pO66LpL18D6OecsxO1/eU44BZkRe9Nvfdd9+u1y1fLxwn9xmnmmvpF1zmnA+mBJFKdwjK1dNLMIM215Wn2qDM6NbgLxl0ryAV6c+SI488Mpd9DXMPY5/8OeUpNbaBodbmfvvt1/X7g6+XGh3XygrP+XId5PfcHaXmmuHphThHpNz8eVhzqzjjjDOKetx7uRf6c5j9c3qa3zviiCNy2eep71Mtfc9CGNYS9XVJF8+XL5b0tSHbCSwPhDynByHL6ULIc3oQspxCDJLi4IuSrpN0VErpsZTSJyT9naQLUkobJL1n/nPgDYCQ5/QgZDldCHlOD0KWbx4MEp33scql8yt/HwgeZUWTN02FHm3DqDuanP3g10ceeWTBelJp2uR93/72txf1aO5uRSWRnqDJmVFl3gb7VzsE0b8j1TO2uzm+Zo4cpTz7eXWTMD+TDmnRUrWM9ZJ0wgkn5LJHS5FCYYZdvxepIUbaOQVBXWEbLiPSRDRfuxwINx3XqBVHLYpolLJkBnqnxCgb6p/X45om7ef6Qf32jPFcM5xzrhepHn3qc8zs2Fz3Tnfze6QCnIJo6TDHyblxaqF2yOmo5DkzM5PXkM8HqZHWPkt9ZBtej/rgUcvck6nrHjlFKocUoK916tf3v//9XHbqnn2nvJwS5n7hNGXt4OYWhU2M+rnZ36dFz1I2TsmTMqXcGQ0rlVTfzTffXFzjfLlsCMqJ3/GTOfjs5TWXE/e/H/3oR7nsY+RpFk7Dc944fqdDhzlFIDKWBwKBQCAQCAyBeIkKBAKBQCAQGALxEhUIBAKBQCAwBIZNcTAUUkrZV8B9QMij0l/B6zGskb4zzu22srK+4x3vyGX6WPmp7AxTJo/Kk8GlMnyXaQzcH4G+ES2unj4DztGyH5wn9yOaRN6fXjbuX0KuunZ6uFRmOKYPQsvv4Mwzzyyuefh8j3e9613F52uvvTaXWxnQORb6OrUyAFNHvT+85qHYtZPiPXR3wFD4JYEZy1t5hthn90ng3HG+PJM1fWDopySVvhFcS34vhslzzt1nh+tn7dq1uex+Zpxz6oT7fVFPvQ2uOd7X57P3FWr5wS0FmzZtymPwNcc9iHuG16N+U//cr4X1PA0IM1lv2LAhl+knI5X737333pvLfprAd77znVymL6PrF3WUsvQ1THn5uDhP1Cn3b3MfpXEgpZSfYb7Xtp4BBK9xH/I0Erzm2el50gf9mXwOmC6i5WPEtDP0k3T9YD9aPqd8lvsJE5Qb08n4+m5lcK8hLFGBQCAQCAQCQyBeogKBQCAQCASGwETpPGmLCdzNkjRttzLE8lBLmuBJ30ml6dapALZJ069n0mUYO0Os3XzMazQ9OrVw11135TLpJg+bboX78xpNlG5mbh2uOQrMzc1lc7ibc2smfjfFsh5pEtIuUkn5OE3J9qk3bkbmvJEubR2KTF3zkHrKluZrNwfzs9PFtUN6vQ03OY8DMzMzeYyehZl0GcfqOlfL2M2M/t6+zyvN8JTFrbfeWtTj3LHslOmNN96Yy6R2PEM110trvqmnTsdRl2placs+06ImlgKmq/C1ybmnznkKD64XuiX4XkU66Mc//nFx7aSTTspl6opTSAxNp6uF69cf/MEf5DKzkl9zzTVFPe4JRx11VC57ChvK3OlMuhRw/D6f7g4wDqSUsjyckqQ82RdfV9xDW2lnqN/+TGX79913Xy77WuJzlKkQfE++7LLLcpkUnqedqe3rPB1EKiljlwuf0ZRhLQO8u5S0EJaoQCAQCAQCgSEQL1GBQCAQCAQCQ2CidN7c3Fw2HzrF1soiStAEzgMSaRKWSpPuHXfcUVyrmUB/67d+q1rvkEMOyWU3W9O0zPYeeOCBoh6jBxh55GbOWsZdqaSYatnLpXakxiiQUsr9dOqwFk3mMuf3/JDWWj038fMAYsqLGXqlUj9I+XjGbM43Tb0e7UEqhONyMzqpIZczv0e9cf1fzGGYw6Lrujx/rWy/nDunJ3mN68X1g3S6rxGa7lsUG+XLzNhej3PMDMxOKbENRot5e+yTr03KkBFArve9ro8rOm92djbLrJVhnWvO69UOeuce5t/zg2gZzUUqzfWZB8JSrn4Q7fnnb0n2zQhp7/sNN9yQy6ecckouux4yqrqVgZt7qbsTDJPhehj0uuI0VU2HnHbkGEjZ+Ukf3F+5t0rlvnDPPfcs+B2ppNdvu+22XPZIXMqJri6kdKXSfeass87KZV+bHLNf4zxx/6m5XyxGrmGJCgQCgUAgEBgC8RIVCAQCgUAgMATiJSoQCAQCgUBgCEzUJ6rruuzT5KHI5DPp2+JZcOlzQq7UQ3QZ7ujZaBlCSR8rZtiVpPe85z25TM7ceeTvfe97uczQbg/VpD8TeWQPKWcIvvPIbJ++AMzCLW2dLX2SIP/uYyNq8nPd4Fy5T0At07SHMzPUmfPrekN/GPq6Oei7Qf8XT91RC/uXynG1UlKMy3eGmJmZqfrRsZ/0CXHfFvpMsOy+MlwX7q9IHyne1+eOa5D+V+6fwVBs+md4NmyuTcrC557pGphdWyr3Ko7fZTvuFAdzc3PZ78N9PuivyP3D/Yo4j1ynnk2a+7b7S3EP4hx42ppvfvObC45j9erVxWeub/bJ/duYPobrkb5AUv0ECe875eS+Mr5/jANzc3M5pYrr0qA+hJQTdZ2h/5J03HHH5bLrJ/1umcbA69155525zH3X54735vr+7//+72qfuJfQp00qfTT9GUK95704F9KW59ViTokIS1QgEAgEAoHAEIiXqEAgEAgEAoEhMFE6j6G3bnanqY8UUMscTROl0140rX/rW98qrjF0kyZomvul0iRImsEz5DKzOakiNzfS3NqiipgN20P/2SeabFu05zhAatbvRbM7w7s9dJ7mV5raPdyY6SDcnM32zznnnFw++uiji3o01/Owaj/sl+0x+7wfxkmTNakhN+/TXOw6Sj2kLJ0SHHf2ealMceA0He/vdA7B75FuYQi7VIZHO+3M9CG85pQg1wj3AVJ7UiknyobrVCpTXZCydJmRivK5oDsAdd11ou/7YrIiDwunJdgv0tAOzimpUx8L6Rp3h/jud7+by1zrflg891bOiZ8MQf3ievaM+KSI+fzwdBqtNDC1tCJO+w1zYO1iwYzlPv98HtClwylO7q+tPZn0t1PoXAukWp3GrZ2+4Hsyaddnnnkml9evX1/U4/OF6761/pxi5PsF59DfQ/q5iYzlgUAgEAgEAmNGvEQFAoFAIBAIDIGJ0nmbNm2qHthICoNmSa9P8xtpn3Xr1hX1SAl6VmpGFtDc7SZctsH7rlq1qqjHgxDZXusgU9ICHuVDU6nTTTSd0uQ+aCbbUWF2dnarjNU9+Heajt08T1My58NN5jRZ33///cU1RuHx4GKnN3mALbMdewQQTf6kHZ5++umiHmlWmph93qlrHpXEcXKe3JTsJudxoRaRwjFRFk47HnrooblMCsjb5WenSe++++5c5rhdnqRQmRnbTxMgTcB5dWrhhBNOyGXSPC5PytppZ9JjlK1TJv34FxMBtBiQam9lu+c4/bB06ip10+lN6jezTkslhUJZ+sGxbJ9Ur9NJPBCXe6vPI58fpIRdltyPXA9JYdeoIGlrvRwHuq7L/fG9gHRZLXJdKuk9ypr0nX/2yGo+Y3laiB9UTNlwnTl9Sh1j//x5XXMn8PXXinatHTbt89TLOui8QCAQCAQCgTEjXqICgUAgEAgEhkC8RAUCgUAgEAgMgYmnOOg5dQ/bJl9Nzta5SfqptDhQ+iQ4t04ulny6h6vefvvtuUwO2DOvMjUC+XjPbE4Ol9y6h2+zvx4+yu8xlHRSfjNEP1/uT+Gn1vdw3xDOD31tPOydvibu/7Bx48Zc/tGPfpTL7qvwpS99KZfpJ+O6wX5wXKeffnpRj9w87+V6TT10XeY1+h+478YkUhxIW+bW1wHl0TrdnHLiPHr/qQfus3PmmWfmMteL+67xXvQH9PZOPfXUXG5l0l+5cmUuc/49DQD1zdvgGuQ69X2q15dxhsf3cvJ7sM/0aXOfSq4znv7ga5364Oue6T0oP/eN4fqhD43LkrKg7yIzWkulvnG83C+l0p+rtcboe+r7RcvnbFTgaQI+BsqXY3C9pQ8n9x3vP/dkP3GD4PPQwVQDN9xwQy67Tyj9GrlO3WeOz1euJU+VwZQxnrG8ljZpFHtrWKICgUAgEAgEhsA2X6JSSmtSSlemlO5OKd2VUvqz+b/vk1K6PKW0Yf7/vbfVVmD7ImQ5XQh5Tg9CltOFkOebB4PQea9L+nTXdTenlPaQ9OOU0uWS/j9JV3Rd93cppc9I+oykv2o1NDc3l81sbmamCZ2UmIe78/Pxxx+fy26Opon4tNNOK67RJE0T4BNPPFHUY7jnWWedlcs8mFiSrr322lxmdu0PfvCDRT2OmaHcHjZMGsMpMNIENMU6ZeDh5301jUiWDLt1kzDN9eyH95GUGCkjPxyWIbM+HzTb/ud//ueCf5dKGfG+fijpSSedlMvUIacRaQZmn9zcTn318GjqPO/VOrTZMDJ5Ej5WzhHXWSsrPqktrjGpzPDv9AjnkhmNvR5Tf/BePEBaKtNtUC+d7iXV2joAnTSJpzAhtVU7WUHaQicbNToyWc7MzGRqzfeBWmoAXwe1er5Hcq92mo6Z6knt+XxwHTDVAN0ppDLLPA+V9zQwDJG/8cYbc9npdK5h7zvXMZ8XrXQdhpHJM6WUnx1Op1NupMtc7nRhIN3Gv0vlocB+iDvnhPX89A2mPKgdGu2fSffycHepdHfhs9H3Se61pGqlMoUC92F3nRjmUPBtWqK6rnuy67qb58svSrpH0mpJF0n61/lq/yrptxZ998BEEbKcLoQ8pwchy+lCyPPNg0X5RKWUDpV0sqQbJB3QdV3/yviUpAU9zVJKn0wprU8prXfHssD2w1Jl6RbCwPbFUuXpFrTA9kPss9OFpcrTreqB5YWBo/NSSisk/ZekP++67gWaRruu61JKC9o1u667RNIlkrT//vt3vbnMIzlICdEE6FFn3Ozp+c+Mww7fVC666KJcpimYmY8l6dxzz12w/VbEDikOz3LNCCBSe05tMiLPo2HYPs2ZboZsbaSjkOXq1au7/ntOeZC+4SHMbmJmH5nt+sgjj6zW+/a3v11co7mec+Dzdsopp+QyI4D++I//uKhXe5nwQzEJmpg9spBz69dINZHGqh2KWcMo5HnQQQd1vT45Tce5pK76XHGzZ+Zpp1FI8/DQaKmkfdieryXOJdeVH3ZMSp70gR/szVMDSO04LUD6w+k8rkfSPIt5QR2FLPfdd9+uXzM+9+wj6azWQeek3zxLNOmQk08+ubhG2pwUiu8DXFv/8z//k8t0jZBKuoZrh/Sw95Ey9x9+3C/cTaAWQer7yrao91E9N/s5835yTJxvXy+kYTk/vtcw6q6VMZ6RcVx/UnmiB/XI9YPPObbhFOkVV1yRyxyX08JcZ61D1LmH+XOzb2MxUXsDWaJSSjtqsyJ8oeu6r8z/eWNKadX89VWSnq59P7B8ELKcLoQ8pwchy+lCyPPNgUGi85Kkf5F0T9d1/4BLX5d08Xz5YklfG333AqNEyHK6EPKcHoQspwshzzcPBqHzzpH0vyTdkVLqT3H935L+TtKXUkqfkPSIpI+Mp4uBESJkOV0IeU4PQpbThZDnmwTbfInquu5aSbUjjc9fzM1ap4uTayY/75wlfRLIt1511VVFPfop+YntvBd9MOiXI5V+UAypvf7664t65KkZcuqcPjlm+g253wXb87BQcuD0FRrE+XCUspybm8t98fBggv5tnsqB80F98Gy49Kv6vd/7veIaQ6yZsdxPZSdPz9B5z4TN1BPMiuyhxZxvcvjui0b+3f1O6FNTSUmx1b2Ica1N992qpWLwOWEIOuE+XUxh4X4NDKvmGvbM/bU5cd3hvbkefb7pA8F6HgLO8bvPCLM9c99yv5Pen4e6MUpZEu7fRr+Rmp+IVPr6UM5+ugLb8wzXXPucU8pVKn2iOL/cc6Vy/6D/j++f1BX6/7R8Mt0Pp+YH5c+tWkj8qNdm33f3caOvGeXkc8zxUaddZq0Qf96L69ZP5uCpA+xTK40G9xz6Xklbr7MenjqFc0P9kMpnKsc4iiCMyFgeCAQCgUAgMATiJSoQCAQCgUBgCEz0AGJmuW4dvkkToJsXSYnQ7OcUBE2APMDQ61555ZW5TDOkVGZEJ83jBx8yJJ+hpJ5Jl+ZzmoidDuPcOJ1A+oqmZaf9xn1obUopm2Od1iGFQCrDw4NpXidN6e1xLAxtlkq6hvLycOY1a9YsOI4f/vCHxWeG5pMKcFM0Tduklty0XevrYjCpA4h7uD5SbjTBO01FapR6y7UjlWv67LPPLq5RvqTTnSrkZ1Lynk6AIdGkNLw9rltSWS53UghORdbk63rfz1sj2/WSkFLKe4PLqEYhu45xDVO/fX45V067XHfddQve12kYrltmll67dm1R74477sjlQw45JJc97J/0HvdIf+bUstQ72IbLeFJrs++r742k2vlcc93i9zgep2dvvfXWXPax0iWCz2HKTJLWrVuXy5wfz0TOe3G/vv/++4t6pCa51l3unItWpn7uMT5PtdQgLYQlKhAIBAKBQGAIxEtUIBAIBAKBwBCIl6hAIBAIBAKBITBRn6gddtgh+4w4L05+sxV6Sz6XHKhz0/QR8nBPpkMgt3vLLbcU9cjT0k/E/bRqp1Z7PYZv05/CQzrJ2dL3Rip9PDg3nj7Aj6YZNRh26zIi/85+uW8IeWrWcz8Oys9Pkec46Qvh6QTIpdNfpxX2TP1yOXCMHIcfBdIKo+Y1cvDukzTMyeKLxQ477LDVnPFaj1ZIMOtRv92/gOknvva1Mtcg0x9wDTPdhFT6sNDn0X0DuZbos+Prg3PMvruvJa+5T1QNrdQe40DXdVk/W/6K3N9aPiS85se+0G/G9bt2hIinnOFRPTzCxf3WuAa5J1DvvB59uHxfYX+97/we5ey+U5NYmzMzM3lMfL5IpXw57laaEs63++2ed955uexjO+6443KZ8+B743e/+91c5tr3Zz73a67TVlof6qWn73BdJ/jewHcF14m+XvhEBQKBQCAQCIwZ8RIVCAQCgUAgMAQmSucxy7VnfqVZmJSBm9tIddDsx0y3UkkxOd102mmn5TKz27ZClmn2dPMlzdY0KbpZkqZw1vNQTV5zs2wtO7pncPZ7jxozMzPZBOt9pGmWplOnDGoZdp0yomydhtmwYUMu0wTv+kA6iGZgp4E5vxyHm46pU+xTi7J02oH34tw4teDfGweYsdznmKZt6pXTUlwvXKcud17z9X3zzTfncis7uve9h8uTa45rxPWD7ZMy8RQH7K+Pn/NEubs8+3keZ4qDXj99nDWa3+VAWVJ+PD1AKqkW3z85Ps69U2LcT5lB+8EHHyzqMbUM9dBlzs8chz9zuM84DU/UTieQtp63caGXp6d44X7D9De+Z3D/op56ShbOg8tz5cqVucy5dJqObVDWrotMU8H91J9l1DHqSisth8Nl38NdJ/pnWWu/cYQlKhAIBAKBQGAIxEtUIBAIBAKBwBCYKJ1HuGmP5jaaAFtmZpo23SxHz/+TTz65uEZTJyOS3HxJcyHNx24Sp4mRUWBuIuaYabJ0uoPttQ73pcnRs8t6VvVRg9nn3cRMmbXMw6TfWlFfvOYZ3NkGaUU37Xofa6hFL3mkSi2yzmXO9pxWZl224VmhJ4GUUqYXPTKFekv6wKkurlvqt69hXnPaoRYV5pQ370XZuJy4bmtZm739ls5Sx3xcnI+afkhb5nMxEUCLwczMTNY71yXek3JtZfNuUZMcs88HZcQ2fK0zCo/64LIk9cQ2fK2zDUYzO6gD7pLgLhs9WlTkONHrrlNdnNcWBUX58rnm4+ZzhPSgVD5TuA/43sp7sU9ej+ubZZcnaUT2t0Vtuj7XXD1cZ3vdiei8QCAQCAQCgTEjXqICgUAgEAgEhkC8RAUCgUAgEAgMgYn6RHVdlzll94kin1kLUZVKPxXyw86P0pfB/VTYJnl3D6nltVaYOf2WWr4tNZ+Mlj9Xi2/mvdxXwbMkjxoppXx/93Hg2Mh1ux8K+9ziswkPyaXPC/XBeXXOFXl11y+Gt1Pmzp3Th6aWDdfbcx8i9qPl+zYu3xmCPm4tH5NWag76i7T8xPjZ26jpdCtNQss/hv6LvK9nfuZ+xLXjPijUA9dFhnpTD3wN9/ceV4qDubm53G9fS7XUL76/sf+Uifspsd6g/kF+L84b2/f9gjKqnSwg1f1f+Ozw7/n6Y/vcI9zvyNf7ONB1XdY7X0vUT86/y53rgPVavqKeMd77tFAfpHKNcH239jGub1+btTXs/lyE6yJ1onUSRX+vxazNsEQFAoFAIBAIDIF4iQoEAoFAIBAYAmlcJuUFb5bSM5JelvTstupOAPtp+/djUn04pOu6/UfZ4LwsH9Gbax63hUn0Y+SylGJtbqc+jFOWsTZLvNHlGWtzsn0YWJYTfYmSpJTS+q7rTp3oTZdpP5ZDH5aK5TCG5dCH5dSPYbFc+r8c+rEc+rBULIcxLIc+LKd+DIvl0v/l0I/l0Aci6LxAIBAIBAKBIRAvUYFAIBAIBAJDYHu8RF2yHe65EJZDP5ZDH5aK5TCG5dAHafn0Y1gsl/4vh34shz4sFcthDMuhD9Ly6cewWC79Xw79WA59yJi4T1QgEAgEAoHANCDovEAgEAgEAoEhEC9RgUAgEAgEAkNgoi9RKaULU0r3ppTuTyl9ZoL3/XxK6emU0p342z4ppctTShvm/997zH1Yk1K6MqV0d0rprpTSn22PfowKIcvpkaUU8pwmeYYsp0eWUshzuctzYi9RKaVZSf9X0vskrZP0sZTSugnd/lJJF9rfPiPpiq7r1kq6Yv7zOPG6pE93XbdO0pmSPjU//kn3Y8kIWU6PLKWQp6ZIniHL6ZGlFPLUG0GeXddN5J+ksyR9B5//WtJfT/D+h0q6E5/vlbRqvrxK0r2T6sv8Pb8m6YLt3Y+Q5ZtbliHP6ZJnyHJ6ZBnyfGPIc5J03mpJPBb6sfm/bS8c0HXdk/PlpyQdMKkbp5QOlXSypBu2Zz+WgJDlPKZAllLIM2MK5BmynMcUyFIKeWYsV3mGY7mkbvPr7ERyPaSUVkj6L0l/3nXdC9urH9OKkOV0IeQ5PQhZThdCnpsxyZeoxyWtweeD5v+2vbAxpbRKkub/f3rcN0wp7ajNivCFruu+sr36MQKELKdHllLIc5rkGbKcHllKIc9lL89JvkTdJGltSumwlNJOkn5X0tcneH/H1yVdPF++WJu51rEhpZQk/Yuke7qu+4ft1Y8RIWQ5PbKUQp7TJM+Q5fTIUgp5Ln95Ttgp7P2S7pP0gKT/M8H7flHSk5J+rc2c8ick7avNXv0bJH1P0j5j7sO52mxyvF3SrfP/3j/pfoQsQ5Yhz+mWZ8hyemQZ8lz+8oxjXwKBQCAQCASGwJLovO2VBCwwHoQ8pwchy+lCyHN6ELKcLgxtiZpPAnafNudseEybuduPdV139+i6F5gUQp7Tg5DldCHkOT0IWU4fdljCd0+XdH/XdQ9KUkrp3yVdJKmqDDvvvHO3YsUKSdLMTGkE2+w/thl8sZubm6t2YHZ2tlqv1p5fI7wNts/vvP7669V6xKZNm4rPHDP7VPv+QvdiG+yT34ttPvPMM892Xbd/9SabsSh5rlixott7xhd7vAAAIABJREFU74Uz7dfkUpv3bYFz5XrDcfOa16u15zKvfa+lh4SPsaWHvBfl7G2w3iOPPDJyWUrS7rvv3u21116S2uuAcJ1jv9mGz2lLngTnxPtQ0zHvOz/vsMOW7c7vW+uvy8zHXGtjp512qrbR13v++ef1yiuvDLIoFiXPXXbZJe+zrb1vUN1srZdae632B9WvQX/gt/pOeXn/Wnsw+9iqx/afe+65sazNXXbZpdt9990XvH9tjlpy57VB9cPB+Wn1aZg9v7UnD7oPOwZ93vb1XnzxRb322msDdX4pL1ELJQE7o/WFFStW6L3vfW8uEzvvvHMuv/rqq7n8y1/+streHnvsseB3pHIT8zZqm+lLL71U1Nt3331zmcrw7LPPFvX6h4+399xzz1X7+6tf/WrBv/u9fvaznxXX+sUklcrgfef8/tM//dMj2jYWJc+9995bn/70pyVtrdicD/aLMpbKDYgy8RfHX//617nsevPCC1tShuyyyy65vNtuu9W6Xtz3xRdfLK695S1vWbAfr7zySlGvttlwHN4njkMqdZS64m1Q5n/4h384cllKm2X2qU99SlI5p5K055575jL1++c//3lRb8cdd8xlrkf2XyrnweXEzY7tsw9SuWYom5dffrmox7W/zz77VO/LNigz1+3nn39+wb76vVav3pIPkWtdkl577TVJ0v/7f/9PA2JR8lyxYoU++MEPLnhvrkHqn+sm54frwHWDc8X2pPpLta8lyrL2w8I/s57v72yP8qJ++ve4Z0lbZCS1959f/OIXufzFL35xLGtz99131/vf//4F+8m+UR9dnhw7x+1zzL2ntV9zfvz5VXvBav1waT3Ldt1111zmvuIvaK0fbpwPzqH/KOrrffWrX9WgGHuKg5TSJ1NK61NK6znxgTceKEt/WAXeeAh5Tg9in50uUJ4tQ0Jg+2MplqiBkoB1XXeJpEskab/99uv6Xyv8BSOVb5F8a/Zf5fxlxbL/CmJ7/vbO7/GXhf+y5NsrLRTeHsfCtt/61rcW9fjWy3u5FY192n//0jpMywl/0fmvgRbtUME25UlZrlmzpuvn2C0NXPScGx9nzazqVgL+MmlZSfir0/WG801d8V9b/GVJmTvYBnXNfwFSDv6rj23Q6slxSFv/MhsAi16bBxxwQNdbWF1v+auOD2j/JUg94DpwXaR83RJIrFy5Mpf9Je+ZZ57JZVp9HIcccsiC39m4cWNRj2Oh/rYsi36Na5Djch2ruTQ0sKi1uXLlyq7vi88bdY5ryftIC7hbXwjKkmtHKvdxln19s4+cE7cc8XstC3eNVm1RQb7m+D3qte9h/iwYAItem5Sn7381K43PCfWR8+jPYa5vX7e1Z7Tva7Vnj6917t2Up3+fzzney/dnfs8tsNz/qds+T329xfiKL8UStdySgAWWhpDn9CBkOV0IeU4PQpZThqEtUV3XvZ5S+hNJ35E0K+nzXdfdNbKeBSaKkOf0IGQ5XQh5Tg9CltOHpdB56rruW5K+NaK+BLYzQp7Tg5DldCHkOT0IWU4XlvQStRS4nwe52ZavD/0VyIk6V03O1n0P9ttvv1wmz3vAAQcU9ciXsh9PP12edUi/JXLrzgEzOoi8rHO79BFohd6yf34vn7dRY3Z2Vn2KA+ef2Wdy7J4Sgfx+K0qSuuEcPuebHHsrSov13K+HelTzi5CkQw89dME+eeQm56YVdch++Dx5FNw4MDMzk/XJ56TmqOx6S3nSj4Z+G96+6zf9Ndie+x8ddNBBuUy/J78X55/rhf5RUrl+Wv1jFKX79lDHqM8+T75exoHep8PXC9cFZeQRc5xv9tfHXIuOlco9mPfyepQF57CVumDQ6E+Ow/1mea2153DOWvo1TvRz4T48nGPOg8+df6/291ZUc22Pdh8m99uqtU0fOu6N/iyjnLh3t6I3fd3yeUgdc/ktwk9xy3cW/Y1AIBAIBAKBQLxEBQKBQCAQCAyDidJ5s7Oz2RznSSRr9Jub22iyo0nXUwHQ3Oj0CNvkvdxUS/MjaSOnyn7yk5/kciu1AGkRUopu7qeJ1c3MDKml2drDdz08eBzoTcYtGdEs7iHQpE1qIbh+rZXYjWZkp6Q4xy39ormY9OuRRx5Z1OM1yuvAAw8s6j3++JboZaf62EdSBm6Kb4WYjwpzc3PZzN1KxUAq28O7mRqhZqqXyrXv8/XEE0/kMmXtOrFq1apcpnne5c57PfbYYwuOQyrXLcfr6R7Ypxa1U9NLqU2ZjAJd1+U5dzcHgjSJ75HUR/bX9zfqgFOCBx98cC63sk5zTpmuwueN64Lr2ek87rO1dS+VuuF6UztBwGki39PGhV5WrazrlKe7M1BXKScfD/cyp115Lz5fnH6jDLk2vU+UO2XWSnvy6KNb8pQyLYxUzo1Tt0TrueluA4MgLFGBQCAQCAQCQyBeogKBQCAQCASGwETpvK7rqtEM/LubZwlSGzQBOu1VO/tJKk2WbM9N2rWM2g8//HDx+cknn1ywvHbt2qIesyS3DjklPAM2zY2tyBj/3jjQm0/93i1zLEEZ0dTrsiSF4JQY546mdY++Yh8ZYfXAAw8U9U488cRcPv3003PZdYG60jofr5X9mnRCLaPuQvceB3bcccdMkbWoAJrqndrhvNJU79nBSbH4WA8//PBcZlQiI/CkkqY66qijcpnUuiR985vfXPBevg+RrmXZ944WSPGSMvFjO3pZD3uY6rYwNzeXZehnDvKetchWqdwjW/rHter6zfVNmTs9zb2QexrpV6nc74877rhcdh067LDDcply8Plme36N46f++t5Wi3obJShPf1ZwzrkmWi4i1FOPSG9RYpQTn9G+vrlfUD98Xyc1x3HQ1UUqqXfqh7uscCyuY2yf69Hp7r7vi4nSC0tUIBAIBAKBwBCIl6hAIBAIBAKBIRAvUYFAIBAIBAJDYOIZy3vO1cOjyUm3fAXoo0B+2L9D/xjn9BkaSd79vvvuK+rRb4JcrN+LflDkXq+++uqi3rHHHpvL5JudU6bfifedPjdsw7n5VqqFUaG/R41XlkoZOZ9Pnwxy7M6dc8xPPfVUcY112Z77JrEe2/C+MyybflV+31rGZPcjIDfvmccZPs++u4/ZuHxniJRS9gNwXeL96U/gvj6UO/XWfRw4VqY0kMp1S78XXwf0N6QvFlNKOFjP+/SOd7wjl+kT5X45lDXHIZW+ZJS1h/738HD1UWFmZib3032/KDPug61TB7iWfF+h75u3Qb2h/5GvTeob581D2On7xGvuy0o5s78ur9oYpVIP6UPka7GVQmJUSCnlvrZOsOA8eCoY7ks8bcF9f/hc9rFSNlzr9EmUyn2ePop33HFHUY97KlMUuW8dn4fUgVZ6Bpc1x0J5ehvD+J+GJSoQCAQCgUBgCMRLVCAQCAQCgcAQ2G4pDloHXrLsh5/SDF/LKitJd999dy57ZmFSO2zD6zGsukYjep9uueWWXHZzNMdcCzmVStOyp3ugqZpUkZujhzlIcTFIKWU5eRh4LSOz01Q1GsYpMdZzfSAVSvOwm6IZmvzII4/k8tvf/vai3m/+5m/mMk3M7INU6tcxxxyz4HekUi9JFUqleZzX7rnnnqKe02bjADOWt7I3t7KDc11xjp26v/XWW3PZQ+tXrlyZyzTj88BhqdQRrh+n6ahza9asWfA7UqmzHOMpp5xS1Fu/fn0uuy6SAhrkgPFWapOlopdZ64BnUh4tCpdy9TVMFwjfB+jmwPv6+mbaEoazO61DWVCnPP3FunXrcvn73/9+Lnt2ce6ZPi72kdecJhqnDHuklHI4v6dY4DOrlmVeKl0HKGuuMamkJ/2ZwtMFbr/99lw+4ogjinp89vCZ588kzh2fZb6vHH300Qv2wfd49tdpVl5rpfbo53cxLhRhiQoEAoFAIBAYAvESFQgEAoFAIDAEJkrn0SzpJl03q/VoRd2RYuGBpFJp7qUJXiqpo9/+7d/OZTfP12i1D37wg0U9mh9p+v3GN75R1PuLv/iLXKZZ8owzzijqMSLIo5c4b5wLNyv7AaujRtd1+Z5OjbAvNJ16FlnOL+fQzcg0CTsNzDmgCdujqhhpd+qpp+byn/7pnxb13vve9+YyqUJSUFKZLZdmdNdXRnUyk7JUmtWpXx5Z43o5DnRdl03gLk/OHcfnkVq1A1D9INlaBmKpjLqjSd7nhPcidXj88ccX9dhf0ks333xzUY/r7Prrr89lZq2XSrl7xKa7A/Twva6n2MZFuaeU8j5Uy5YulXuVU3GkjdjGQw89VG3Px0kak5F7nhGf9R588MFcvvDCC4t6lO29996byy2Kjc8Vp8IoP3e9IM1FHWpFMf7/7L1prGVHdfb/1L02YAx4wm63x/bQHhqPuG08gWcwhgCCKHoRifwnJEgkSEEhUpz3/ZQPkZAS+MQrxZaC7A8RL5ESwDhgbNo4HgDb7Xlst2e3aU/MM/je/f/Qd5d/9fSt6ntPn3P6+PR6JMt1+tTZu6rWWrX3Xc9aq0aJfk4e3kEdIpXtB3tzjRl+4ONnBvmtt95afMcwg3vvvTe3WyccMCPPdeyss87KbT4bnLrnGM8888zc9mcBx+dj4l7C9wYPNeh1eDmZs+GJCgQCgUAgEBgA8RIVCAQCgUAgMADiJSoQCAQCgUBgAOywiuUeT0HenanfzrOT62TshsdxME7C4xrIzZ544om5fcYZZxT9HnjggdxmTIZXaGW6O8fu8RmsYP7Zz3520d9I5dp4qia5bXLCzm17DMmw0XVdjl1ybpr8NmNIvDI7Y58Yu8AYHKnk/VslL7huTJuWpJNOOim3Tz311Ny+6KKLin48CZzcPuNppDKehPy5x12wn6dYc204do+r8orMo8Ds7Gy+j8fRMNaMMRkbN24s+lFOnDfjXKRyvRiDIZUlJ2jTzzzzTNHvqaeeym3GOJxwwgnVMXHsPsd//ud/zm2Wm/BT7hnH5mnxtfg/1/te10eVHv/KK6/kEgs+Ru591FXfZzhP2pLHNVKnPV2eekM789IFt99+e24zFf+II44o+rGUAdf62muvLfpRV6hDHsPGsfs60Taprx4rM47TBObm5raK8enB/Y/PBl9jfkebYAyiVJasuPHGG4vvvve97+U2Y658r6WdMXbNn0mXXHJJbvOUAN8vGIfGZ6WvybPPPpvbrs+1Ugver9eDiIkKBAKBQCAQGDHiJSoQCAQCgUBgAIy9YnnvTm5VyK2lxkql+5J0i6dP8kBRLxPAirZMmaS70n9HF7+7+ugyJiVIV6b3o9uUB0JKJc3lNF3t0F6nkUZ1uGmPrusyJeKucFY45hjdxUy3Kn/jVAtd654WTjcw5eVlI0iXsSq5l9YgTUS3slOWTNOnTJxupN44hU06p0XhOqU9Cvz+97/PZSG82jHTgKlXnkZNPPbYY9V+tL93vetdxXdcL+qL2/DVV1+d26Tc3Oa4X3AeXlme1A71zQ/DJrwUR60qstNc/Z42KhtNKeVrt05roP75WLj23IOckmZpGdcbptLzOy+TcOedd+Z2qyQBaVGm2POUCKl8lrBKvZ8YQIredbRWVsSfR7WyFsMET4fwNSa9yn3CQwD47GGFcd+7SdF7GRDqC+3WQ2lIuXEc1AepLG/CskMsMeKgTrg8qcP+DOH7Bt8V/Pna24fv1S2EJyoQCAQCgUBgAGzzJSql9KWU0osppQfwb3unlK5PKW1c+P/oI18DQ0HIc3oQspwuhDynByHLnQdL8URdKeli+7fLJK3rum61pHULnwOvDVypkOe04EqFLKcJVyrkOS24UiHLnQLbjInquu6mlNIq++cPSjp3oX2VpBsl/f22rjUzM5M5XY81IGdJnrmVQsrYKZ7+LpVxJc53k0dmGQPngAnGVviRKoyjId/v6dG8BrlipmhLZal8PwGeaaLkh/3oA4/X6DFMefYcufPqlBnH4bE9jDMiT+39yE/7sQe8PtPbPeaFqbyMa6DspPKIoD5N3McglXNm/ESrtITrfO0aHvfl8X49hm2bvQ26zdVSvLk+Uhnbd/jhh+e2p0DTHn1ujFviOK655pqiH4/1Yer6+973vqJfLWbu3HPPLfoxjqMVx0b9YxkVqYylYrtW2sJLHAxLnrOzszlOxfcWgnuLp4szVoQ67fsMY2h8r+Jn7lW33XZb0Y9rzzX1uEbaD9fU0/R5X+65HofL54zHX3H+7Of2XYuZG7Zt9nuFx5XSfmhnHhPF/YWxSCwvIUnXXXddbjNOSSrXmbrrR65Rhnwus3yFVNrZDTfckNt8/knlcS4f/vCHF52Hw59J3KsYV9Y6FmmpGDQmakXXdX1U3vOSVtQ6ppQ+mVJan1Ja7wGzgYnBkuRJWbYCbgM7FAPZpgfOByYGy7bN2GcnFvHcnEJsd2B5t+XVrfr61nXdFV3Xre26bu04MhkC24eWPClLL7oXmDwsxzY96ycweViqbcY+O/mI5+b0YNASBy+klFZ2Xbc5pbRS0ovb/IXKEgdOGdBtTjed0zL83Eqxfvzxx3PbKRv+jm4/p1HobSFN56madAHS5el0Hquj083+/e9/v+hH17JTARw7XctOO5BOWAKWLc+UUnaPu1eK60G3uNN0dE1T5q4bfMD7mp588sm5TVkyjVcqXcKEV93mOEhxeCpszT3ulAbpKXcxk1aknD3dvOW2XgQD2eb8/Hx2bTsVUKNdPbWZa+LrQLRsjrj55ptzm/RdP94ef/Znf5bbrmOsXk3az6tXk/ZhmRLXbdICLk+uG2kCT5dfZmmDgeUpbZ0Sz72Fc3NalbbJ3/jpCtzHWFZGKvdCVpPmPiiVdBnlt88++xT9SAdRRtQTSTruuONym3bres25ONVHnaf9tSrALwEDyVJ6dU/1EgsE186fG9Q5lh/xcgKk97wUBdeczxfXb+ocy3t4v/vuuy+3SfF6WA2vR0rebZ2y9v2a+zB12+XXOwecNm1hUE/U1ZIuXWhfKunrA14nMBkIeU4PQpbThZDn9CBkOYVYSomDL0v6vqSjU0qbUkqfkPQ5SRellDZKunDhc+A1gJDn9CBkOV0IeU4PQpY7D5aSnffRylcXLPdmc3Nz2f3rtAxdfXSZusuOh9PSFdfKmvCgWd6LrkI/DJUuUbrnnSrjAcS8nrutOeda9ofDswU4Z7p23VVdO7ByWPKcn5/PGZBOT1B+rcqv/I5Zdu5i5UGkzPqSSjc8KRnPuiANQze+007Ur1pFdal08ZNmcDcy18bXgvPk9Z3OrFFjw7TN2dnZXC26Vb2a43SqiFRzzQUvlXZF2l0q50q78rVj5tbpp5+e226bX/jCF3L7nnvuyW23zY985COLjs/pXsq3RWNQ31x+ve64nIcpz153W5mz3Hd8v6Bu8nQFPxycNuf6wL37vPPOy20/TJq0MG3OK73T5h555JHc9qwv6huzKT00hHrutDL3VuqUB3nXwiaGKcv5+fn8DHPakfKlbF23HnzwwdzmXJmNLJVV3D3ulRXjqRO1Q3z9Xr7GvDfX25+HxxxzjBaDh/OQWvY9rHaKhvfrqeuoWB4IBAKBQCAwYsRLVCAQCAQCgcAAiJeoQCAQCAQCgQEwaImDwW62yy65gin5Vank0xl34PFMjC8g5+7pmK0YE/L/vJ5z5oxhYhVW5/5raacep8XqyfwNeWip5Nm98ngtLdn55lal92GA1eddRuSTyW97hWvGe5Gn9vguVsR1GXGerKbslZUZe3PooYf6dDJYRZeV6b16L/l48uqels218PRwxt4w1sFl5/Fdo8Ds7GyOgXA5UYasSu5rzLIB1H23D6bM+3pt2LAhtxkL5zbH0has1Oxp9gcddFBuM4bpwgsvLPox7uI73/lObtcq/0vteCPasKdL9+nWHj83LLCUjJfH4D5BvXUZMb6E33l8F2PTmEYulWvHeDSPc2WcI+OZ1q1bV/Tj7/j88BhK6hRjuPy0A+q1x8BQf6k3HkPTKtExLHRdl9fSx0nd6mMapXqVfP/O9zXK2uOvGAdFvWLJBKlcZ8rC9wten/v6e97znqIf7apW3d6v53rP+XP/cdvsbXI5ZUjCExUIBAKBQCAwAOIlKhAIBAKBQGAAjJXOm5uby242p2XoYmNqZavirrssCbpxvbIyDx2m+/iwww4r+h1//PGL3uuJJ54o+tGly9946i1Te1ll11OlSY95+iipKLoi/aDRUZ+31HVdtcQB3eR06bt7mHQGXfJOt5Em8BRrXpP0IFN6pbKKMelBX6c77rgjt0lPOVVBvWFpDKcM6FZuuf5J2bmL2emPUYAUkNtLjSZ1vaXcaLdup3TP33jjjcV3dOVTp93mqHMsV/C3f/u3Rb9vfetbuU3b/KM/+qOiX20v8Ur3TLsn3SiVlAH1xSnBnn5cZrXrgeC6xHVr2SbtghSSnwRAXfF7kcL73ve+l9uewk4aitSQH0DMcgq33nprbntoCKtun3/++bnt6839x2VUKxHjJQ1GHTYhbdk3+n2ldXIC6eXWwbrUAa69VIawMJxBKtePFei9/MvBBx+c29QP31cY3sIQDqf4KV/uz/5uwLVonaZAytLl3q/vcg4iDk9UIBAIBAKBwACIl6hAIBAIBAKBATBWOk961U1GWk4qaRC68VtVgelydRcg3c5+AHGtyqy7cPm5zyqUts5koVuf7ugbbrih6Ee3ZO1AXKleXVUqKSG6kt0tO2oKiAcQu+uTn7lWPkbSBJSr6wbn4vrANeV3XhGfdAWpPj+IllmepCecuuF3pAc9U7FGWTo4PnfFjyqLi5ibm8uZL+4K5/pTnj5Ouv85ZtIwUpn998lPfrL4jn1ZLZz0gSRdfvnli/7GqR3SqZdddlluuy5SbpyHZ4ryXk6hc48g7ewZbf3+s8yDiJeMruuynFoHC1NGvqfxO2Y4si2VFK7T8F/84hdzm/Jz6pAZVx//+Mdz2yubU5ZHH310brN6uVSuPSlFX2/arWfO1qqZ7wg6j2Pw/W+p2YFcE+qAZ5jy85o1a4rv+Ewlrc2MZqmkWpnl6Hsj15LUnj9PeA3qkcuMsnCqj3bA+3oWX782kZ0XCAQCgUAgMGLES1QgEAgEAoHAAIiXqEAgEAgEAoEBMNaYqNnZ2Rwf0KpyzbgD54AZT8FreAr6sccem9vkz6XydHjGzjAFWirT2MmpetVYpkcznuS6664r+pGf52+8tAK5XfL2Uj3GyLndVvmHYWB+fj7LxlOWOf5WxW6On3L18heMU3IOn9WqqQOt6sSMUWG5C6mMg+M4XL/4mRy7x0iwArfHL/AajC/yuIvlnCg+KLquy2P3mIRavJqf8s41P+WUU3L73nvvLfq14ilWr16d29wHvBQFx8j4Qo+nYywI43y8VAbj5Lg/eD/qn5cwqVVCdrn36zSqatdd12XbdBkx1oPjbVXiphxOOumkot/b3/723Pb1YOkP7nF+kgPjSBkPeu655xb9GKvGkgkey8rTBKgnHi/Hvd9j37ivUK/9Xl7WYRToui7vD74XcNyUp8edMa6Rzwo/LYMxb64TrCz/7ne/O7fvv//+oh9j4/gc4qkAUhnzyH2Tz26pHjvbOqXD91Du/5SZl2fon/NR4iAQCAQCgUBgxIiXqEAgEAgEAoEBMFY6j5SBp0fTTe5UH1FLnXYXMV2Wnp7Oa5BacNc3XZt0OXtqM/v967/+a263SitwDE49keZytyTXiS5HpztGlT7dY3Z2No/bKRTOjW5kT53nXCgHp/1IEzidR1cv9cbLApAKuPvuu3OblK1U6gDdxe7Gp4xYTdt1lzSwU4KkDEjbuit51NXn+3v2Y3WKonbQtVf75drxO6b+S6VOkDqTSsqA31FmUumG//CHP5zbrGgslXQT5XTbbbcV/f77v/87t0mZuyw49lYaNSkl1zGnv4eNlFK2Ew+HICgXHtAtlXKm/flees455+S272N/9Vd/ldvXXnttbreocVYp9zIttHXuAzyMWiqr6pNO8rIWhJeCoC6TGnOdd9psFJiZmck64+PkXsl19JIV3F/5rPTnJtfL91rKg/18z+N3vL7voawsz/YFF1xQ9HvXu96V26TXnYLl88XDW2qlFvyZ369blDgIBAKBQCAQGDHiJSoQCAQCgUBgAIz9AOLe1eo0FSkhfuf9SJ3xN55lRTe+0xPMHKLLkvSKJG3atCm36Q70ath33XVXbtNF7gc40v3IzBV3M3OO7palm51jdzevV3AfBfqxuNudY2TbXaR0hfM7p+K4pp5NQTc1qQuvRF6DU03M3CO15CClS9dxq9q6U9ikdUgbeb9ahf1hYtddd82UotOH1C26zN1emB3DjB23F1Y0vuSSS4rvvv/97+c2qRO/BrP6SDd5dXBWWea8XI+4X/DQYs+cpQ17dhDtsZapJ42eak8p5bH5vsAQCOqmU2zUOVIeXrGcttqitpjV5zbCe5PW8XWjrZLi8QwzZoN6CAHB6/v8SflwLTzUwPVoVOjH6vKsUZROU1E2nI+fJkDK9NFHHy2+I13L563vV8yCpF21qqNz73Yb5jOQeuSZs5SnP/94feqfj723iaDzAoFAIBAIBEaMeIkKBAKBQCAQGADxEhUIBAKBQCAwAMYaE5VSyrylx72QpyVX6vEgjDNiirGnGzMm5oQTTii+Y2zE+vXrc5tpyVKZkrl27drc/upXv1r0Y9wEY3Q8nun000/PbcZ4OH9Nbtdjezj/1uniHq8ybMzPz+d19JRtypZjdBlRzqxY3opj8BgApo+zwrifKM8YJsrF141yYYwPS1xIZRwGU3o9dodxJ17NnDEGXDOPHWilqQ8L8/PzOb7D9ZaxSfzOYxcYG8a5Pv3000U/6gur+EtlLArLPjBtXSpPF2C84sc//vGiH+2AsRCsjC6V+wDn6+nbHIfrDuXEObp9j1qec3Nzee+qpXBL7VRvxsYcddRRue3xR1w3L1fBe1E3vOo25cJK5B4TxVIW1I3jjjuu6Md4Fu5k+gDfAAAgAElEQVSXrte8fsvm/FlFjOM0gbm5ubzHuM3xWcbYQI/Xq5XVaMWnebwo4784by/7cOqppy46jlWrVhX9uEezn5cQYnwlTx/x5wRt08fE2Cd+53tyP0f/9xa26YlKKR2cUvpuSumhlNKDKaW/Wfj3vVNK16eUNi78f69tXSuwYxGynC6EPKcHIcvpQshz58FS6LxXJH2267o1kk6X9NcppTWSLpO0ruu61ZLWLXwOTDZCltOFkOf0IGQ5XQh57iTYJp3Xdd1mSZsX2r9IKT0s6UBJH5R07kK3qyTdKOnvt3Gt7Erz1EK6TOlm9sNo6YJ95plnctsP6j3xxBNz2yulMjWShyB6mQSmuJOScLck015bKeD8ji5yd9FyTO6WJEiBtSru9himLGdmZjId5dVha1Wdvfoz3bYu59r1HJT7wQcfnNtOgzI1/bTTTsttlxF1g+vr6bTUKdJyXh6AbnSv6EzZkoJw2bmt9BimPBeusdW4fGzUYV9jUiAsQeClPqgv3/zmN4vvSLlcfPHFuU0qXCopPI7JdYXrT6rV0/F5ADHl5JQdKRN3+fNeHJ/TE/0cSYkMU5YppSwzp7BqB7H6PKmD1D+X5Z133pnbjzzySPEdZcF7OYXEKvMsoXDmmWcW/ViugiEalJ1U2iD3VqfeuBa0danUedJEXgrB974ew7bNHq19nvrolchrlcN9T2ZoilOCvDfbbgeU4fvf//7c9j2UdkE9cmqVY+f6O83KfYvliaRSvrR9t49e10d2AHFKaZWkkyXdJmnFgqJI0vOSVlR+FphAhCynCyHP6UHIcroQ8pxuLPklKqX0Jkn/KekzXdcVr9/dlte2RV/dUkqfTCmtTymtH0eAbGDbGIYsx1VkLrBtDEOerfMqA+ND7LPThWHIs+aJDkwGlpSdl1LaVVsU4d+7rvuvhX9+IaW0suu6zSmllZJeXOy3XdddIekKSdpnn3263kXoblG6B7kJPPfcc3693OaD3A8jpMuylVnFh4dH+9MFeN111+W2u3DpzmS2kbvIST9y/s8++2zRj9leTkVyjHSLu2vT3dM9hiXLww47rOtd3k4/cU3pWneaiOtG17Fn5VBe7mali59Ug9ODdOVTfi23L+XicyRNwHm4HPg7/66W2ePr5FmNxLDkecABB+SF9Wq9vD/pT7dhyo2ycIqNNKBne/Ez7fvmm28u+pHGXbNmTW6T0pXKTCHSN9dcc03Rr1ZZ3224dhizVOoB9d7lXKuGPCxZ7rfffl0vG7cXrgH3Fqc1qKvUdddN/s6z87iPkUJ3KpXZdcya9LAJ3ptV5X0/5trzXi4Hro3rqFcm79EK0XAM87nZ65bTeZw7aS8fP9eE9uw6TJmdffbZxXe0Rx7m7Sc78Pq1DE2pXDtWIvdnHmXI9XfbpH64PEnRe6Yv0ev9UCuWpy1X+zdJD3dd9wV8dbWkSxfal0r6+pLvGtghCFlOF0Ke04OQ5XQh5LnzYCmeqLMk/Zmk+1NKvZvlf0v6nKT/SCl9QtLTkv5kNEMMDBEhy+lCyHN6ELKcLoQ8dxIsJTvvFkk139YFwx1OYJQIWU4XQp7Tg5DldCHkufNgrBXLZ2dnc4yIB7KSw2zxkeRsyY96ajnjWTw+hlw402M9LZQnULMytqd0Mo7jySefzG1WOZfKFGty1h5bQk7ZU4p5QnYt5VRavMTBMPHKK69kbtljmMg/8zvnonnSNvlyj89gnJLHeNx99925zbg4Vo73cZDP9/Ro8vHUSefYeS/Gj7gc+DsPEOU60R48TsFLI4wCc3NzOebB4zw4P66Py5PjpE67XfHkeJYikUp50KYZTyiV5Uhoc5/61KeKfhwj+3maPftx//HyI1wLj/GoxRE5evv2mI5hgacJeAwT9axW7oBj9O88ru+II47I7fe+973Fd7z3+eefn9u+L1599dW5zb3aZX7XXXflNuPqfE+onX7hek0d8GdOLXbM0+/d3keBmZmZvJ+39hA+K/xkB+oqn5s+H8qdpXak0m5ZwsTXgPs645E9/pF7KJ/XblccE59rntzUqkTO5y3X0PXe996lIM7OCwQCgUAgEBgA8RIVCAQCgUAgMADGSufNz89nd7K7zehOZdvTHWsHlLqbs3bgoH93yimn5La7JekSve+++xYdg1RSjEyx9notpAx4uKNTC3RPO7VFioPfuavafzdszMzMZDerpyxzvVslDri+vIYflkm5PPXUU8V3rHpL97PTeVxTpjq7zElD8Tc+Jl6D+tqqSu70Da/B+fuYxnHIKeF6S5c35+qUAWVBt7uvHSkbp72Y1k7b9xIHtMEPfOADuc0DSqWSbiL16/pBG27R5DVqQSr3NFZn9vn3Y1/OIafLAU8T4LykUn6Ua+tQYIYykKqRSr095phjiu+YIk+5uiy5pt/97ndz29eHtk+a3MM1aGekfLxiNvXcDw7n/kkdcDrT6aBRgBXonXZs6SpB2o526vbH75zqI4XKdfA9mWVFqDuuY1x/ysZLJvC0EM6xVZbCn9G8N5//Tgn2Nj3UEgeBQCAQCAQCga0RL1GBQCAQCAQCAyBeogKBQCAQCAQGwFhjopiq6XFA5CzJxfLUdKlMrWS8iZcCILfrXDHjVsi7e6l88qpM7/SU2uOPPz63PYaLYMxE63yrVro/16ZWXn9b1x8Guq7L9/B71Y5B8RgSHrnSOj6Bc/O1Z4wG2wcccEDRj/rBmAmPcSDPzvgo5/PJq7Of8/SMCfNU/9pRR60jjEaFmZmZrFseC0F50h79CAzaFdsuM8qTMUtSuc48QunYY48t+lGejOvwGBXGPHDsraNYWEbE49hapUmoS5R1rezFcuIuloO5ubkcC9U69oVr47rJWCrGwnh8F9fbbZj7M9eA8aV+L+7bHhPF8R500EG57bFO1K9WrCFt1fdP7sEch+vDOGyz67q8lr6Hco2p017yh+A1/Mgx6ovrp8e01u717W9/O7f53PR4Oq45n2U+Juomx+DHYVEXXdaUE5/DvtcNEq8YnqhAIBAIBAKBARAvUYFAIBAIBAIDYIeVOPBUcLoRWQrA01fpYqQ779BDDy360S3s1CGpE7r2PM2Vbk9ew12ApF/osvT70sXIOXqJALoSW6dscxyjpu8WQz82X4/a6fBehoK/Yz93kXM9/F4EqQZ39ZJqYbulX5Qr9UkqdYMuZne3t9zjpCdqJT6k0VW2Jubn57MOOdXF+ZHidgqErntSta6btIsWrc9qx+5e/+M//uPcZlkRpxxI2XAdnb6izpEKZrV8RyvdnON1negpwVGVOODYfP+gLGhLvm61UhYuc8qZJwFI5RpT10kPOmibTh2yTALX1J8l3Ftpf63yI66HpIhbZVrGVeKgl5VT6FwH6p/TuJQnKWkvU0L7832nRh36GvBeXDsPl6Eu8V6+/9Qq0LvMKF9/1tRKI9QqwC+ncnl4ogKBQCAQCAQGQLxEBQKBQCAQCAyAsWfn9dSVHzJIVxzdbU4F1DLrnAKii5GZHFJJSTCzwzO66JYkTccKqj5Gug3d9UqKiS5KP8iV7kx3QdPdze+cvhpH1ki//u465rharvBaZWG/HuXMSu9SuR6tQ3xZ3Z739cwSXo/yb1VRb2WicRytKv10U3u1Xad7RwEeDu46V6vm7W58riWpcae9aoeI+725Xp5tU6ui7XbATB+O77nnniv61Q4gblUs972J9AR13auG9zTXqLLzZmdn8zq6LpG+oF65LGkHXDenP0jZeWYqbaGVpcv9jhmPvj7sx/V1OdSylr0fKTuXM/cjjtcpLq94PQp0XZfn4eEM/EwZ+r5GWXM+Ldt0UNbs1zqJonaQtVTKoxWyQZ1jWIXLopXZSf3jeF1nIzsvEAgEAoFAYEyIl6hAIBAIBAKBARAvUYFAIBAIBAIDYKwxUSmlzM0690rek7EQzrcuNZW/FR/DeCymZXscEe9N7ti52Fq1ded2eT1yrs7L1lKlpZLb9rR7olUKYBiYnZ3N3LenmpI7b8W1MM6I8RO+vry+8+rsy3X0mDvKiJy9X49jpLw8Lolz5NhdXryvf8d78/puG6NMhe/B0wQ87oVlIBhD6HLiHFgawOPJaMNeMb4W5+enzfOaGzZsyG2PseK9WmU0OA6WtvB4xVpFe6mUNW3T08j7vWSUFct7m2mVx2A8j8fBEfyuVU3fyx9wrRi35mUjanbQilfkbzwOjuNgFXWPD2uVJ6B+8b7+/GmdtDAssDSQl+Zg2YClpuVzX3PbbJ3mQD2oVeeX6jbn60394PPKnxOUG6/h+yLn72OizrVsov+dz72F8EQFAoFAIBAIDIB4iQoEAoFAIBAYAMlTskd6s5RekvQrSS9vq+8Y8Fbt+HGMawyHdl2377a7LR0LsnxaO9c6bgvjGMfQZSmFbe6gMYxSlmGbJV7r8gzbHO8YlizLsb5ESVJKaX3XdWvHetMJHcckjGF7MQlzmIQxTNI4BsWkjH8SxjEJY9heTMIcJmEMkzSOQTEp45+EcUzCGIig8wKBQCAQCAQGQLxEBQKBQCAQCAyAHfESdcUOuOdimIRxTMIYtheTMIdJGIM0OeMYFJMy/kkYxySMYXsxCXOYhDFIkzOOQTEp45+EcUzCGDLGHhMVCAQCgUAgMA0IOi8QCAQCgUBgAIz1JSqldHFKaUNK6bGU0mVjvO+XUkovppQewL/tnVK6PqW0ceH/e7WuMYQxHJxS+m5K6aGU0oMppb/ZEeMYFkKW0yNLKeQ5TfIMWU6PLKWQ56TLc2wvUSmlWUn/V9J7Ja2R9NGU0pox3f5KSRfbv10maV3XdaslrVv4PEq8IumzXdetkXS6pL9emP+4x7HdCFlOjyylkKemSJ4hy+mRpRTy1GtBnl3XjeU/SWdI+jY+/4Okfxjj/VdJegCfN0haudBeKWnDuMaycM+vS7poR48jZLlzyzLkOV3yDFlOjyxDnq8NeY6TzjtQ0rP4vGnh33YUVnRdt3mh/bykFeO6cUpplaSTJd22I8exHQhZLmAKZCmFPDOmQJ4hywVMgSylkGfGpMozAssldVteZ8eSpphSepOk/5T0ma7rfr6jxjGtCFlOF0Ke04OQ5XQh5LkF43yJek7Swfh80MK/7Si8kFJaKUkL/39x1DdMKe2qLYrw713X/deOGscQELKcHllKIc9pkmfIcnpkKYU8J16e43yJukPS6pTSYSml10n6X5KuHuP9HVdLunShfam2cK0jQ0opSfo3SQ93XfeFHTWOISFkOT2ylEKe0yTPkOX0yFIKeU6+PMccFHaJpEclPS7p/4zxvl+WtFnSH7SFU/6EpH20Jap/o6TvSNp7xGM4W1tcjvdJumfhv0vGPY6QZcgy5Dnd8gxZTo8sQ56TL8+oWB4IBAKBQCAwACKwPBAIBAKBQGAAbNdL1I6qpBoYDUKe04OQ5XQh5Dk9CFlOFwam8xYqqT6qLYWvNmlLANxHu657aHjDC4wLIc/pQchyuhDynB6ELKcPu2zHb0+T9FjXdU9IUkrp/0n6oKSqMuy+++7d3nvvLUmanZ0tvuPL3Pz8fG7PzMxU+xFbgvi33c+vyX6t3/j1a9/xGq3fLPW+/h2vyXlwzfx3mzZternrun2rN9mCZcnzjW98Y7fnnntu45IlXnnlleLzLru8qn4tmc/NzeW26w1/xzl7v9Y1auBvWvrF6/E3i/2O8HnWwDm+8MILQ5eltMU2e3n6mDk/ypDya425JU+/V+t3RM3m3F74uTVe4g9/+ENut3TF78XxLsX2f/zjH+uXv/xlveOrWJY8d9ttt26PPfZYdIyD7HdL3cda37VspLZufu2l6gb7LXUejqU+Z3ivl19+eWS2uddeix8NV1uv1vosVQeWeo2l2nDrWVb7fWscy3luLuV6xDJsc7teoharpPoO75RS+qSkT0rSXnvtpc985jPq28Tvf//73P71r3+d229605uKfr/97W9zm4vhG+Tvfvc7jqH4bvfdd89tPhR4bf/dG97whkV/I0m77rprbnMDft3rXle9HjcSjlUqFYDr4uPYbbfdcvs3v/lN0Y/X/Lu/+7untW1sU56U5R577KG/+Iu/2Gq8Ut1wfvKTnxT9+BLG8VM+kvTTn/40t11vfvWrX+U21/TNb35z0e8Xv/hFbrtOERzvz3/+ak03rrtUyvktb3nLomOVSr30BzL1o/Ww/uUvf5nbn//854ciS2lreX7qU5+SVOqzVK7lj3/849zu/yDqQblTnr52P/vZz6rfUU7Ub18frivt0W2TcqLutP44e/7553PbdYU27PfiXLiGvv/0Y/qXf/kXLRHLss03v/nN+tM//dNFx8jPXBt/saFucq19n+We2Xphow7RrqT6uvnDlLrh9k3QXjgP1+vWSwfXhnA9pJ5fccUVI7HNPffcU5/+9KcXHSfnx2fFG9/4xuoAqAP+fOGa+D5cey75utb2cl/T2suMP4d5DY7BdZvwedEGuTau9z0+//nPV6/t2J6XqCWh67orJF0hSYccckj3+te/XtLWD1QuDjdnbrjSls2+hxtZ7Xq+oAQfwty0pfJFhMbD30h1QfhLFF8O+Z3PkcrFB7T3pbL55tbaZAYFZXnAAQd0/Zr4S2DN69PLvkfNk+bry2v4yyLXh2vjukFDp/FRJlKpAzQ2vy/HTv1yefGlyh+mNa+i92vp7/bA5dnLzedAeXBd+ULl2HffV/8gd/vgi4k/lDhX9nOdqMnadYzfcRw+JtoP7+t6RH3hXiSVc2mt03I9uEsBZbnffvt1/d7gexD1jN/5erQekgTn7P1oZ9wjfG+q/WHsD12uG+Xlf7jQfmov21Kd/fBxcC18n3V9GxYoz4MPPrjr7+vjrL3s+b5GWde89/5d7dpSuQ6tl9OWR4g6x73W9wTOhff1fZEvW67PvKavDeHvAEvB9gSWT1ol1cD2IeQ5PQhZThdCntODkOWUYXteoiatkmpg+xDynB6ELKcLIc/pQchyyjAwndd13SsppU9L+rakWUlf6rruwaGNLDBWhDynByHL6ULIc3oQspw+bFdMVNd135T0zaX2n5mZydyn85mML6hxpVIZNEjeupUV5Zwt+WFyoG9961uLfuRRycc7Z0sulvd9+eWXi34vvfRSbpNvbsUItILiGWfg3LPHkCwFy5Fn13V5Dh4nUIsf8/gMxjLwO5c55czgUqkeo9IKLGRshOvNc8+96lmnvFw3OEbGELkc+B1118fY4uKXmk1ILNc2d9111xzH5DF6NXv0uVKGXDvXj1ZCBGMyuF6uz7SDVrwir9EKkF5qfB7n74Gx7Eubrq3TMrPFlizPlNJWttaDewv7eH/KnL/xOLBWxlwtM87jeqgPK1euzO1WNhe/c9uhzLnPtsbn4BgZI+dBz0vN+CSWa5td11XjbhmTRTtYamajr0krDrS2X/ley+9qMXg+xlrSlY+JSQkuP8rdn6mcC9fMdXE5NtkjKpYHAoFAIBAIDIB4iQoEAoFAIBAYACMvcVCDu/boJm1RBnTTtWoy0WXnNArpPV7P3Yj77LPPov3cpc0UZlJFXsahVlPFSwTwXk4n0FXK+Xs6qtffGTbm5+fz2Fru11ZxxloZBk8bpov14IMPLr6jLHl9lxFpwMceeyy33T1PWfDapBkcrYJylEOrllmrdtI48Ic//CFTcPvvv3/xHdeI6+hyp6xJLbgs6JJv1Rai3Tq1ULMDp3s5xhdffLE6JuoY5+t1yXg9T5VmX9q+09P99VslWrYHXdfle7gu1VLEW6n7tRp+UruUA6/BfczXnutGubp+UTd4PaekSKHzOx87r+/7bK2Wkj9nvObVKDAzM5Pp61YpBs7HyzlQFlwHX2Ne33WnVs7CbZP2SL3yfpRnq1wK59iqv0ba1fcB6hhl7c+afozLsc3wRAUCgUAgEAgMgHiJCgQCgUAgEBgAY6Xz5ubmsvvT3cd0k9Ld6JkXdMUxY8rpLLr9jj322OK7H/3oR7nNDLoDDjigOiZSHJs2bSr63XnnnblNN77PsUY7ePYVXY9OeZH2oTvUMxBb1WaHgZmZmewKdRnVqsU7dcN5c+19Luzn1MiBBx6Y29QHr7p9//3353YtQ0kqqQa6s/fbb7+iH13MlNHTT5cnP9Ct7q7jVoYL4XT0KLDrrrvmObprnfNrZbb4mvdwymDVqlW57XrKjM0TTzyxeg3qAfXlmWeeKfrRNikLUrpSeUoCq627bVI/SA9KpV4tNZxgFGB2XouWoH67/nGtWpRY7Zgk/x37OV3MsAnK2W2zVkXdwyEoM37n9kfd8yOM+DvK1e81iurzjq7r8n19DtQt0lu+J3PtqMNe7b1WdV8q90Dey4+HqWW4eejETTfdtGg/txfqIuXkVDvlSZ2S6keC1WjPpZ5rKoUnKhAIBAKBQGAgxEtUIBAIBAKBwACIl6hAIBAIBAKBATDWmKiUUuYanYvk51ZVYMYrkHv12AV+fuqpp4rvGNPE6zsvfswxx+Q2YyhYjVkq+Xlee+PGjUU/xiAwJurtb3979XrOgXOdOH+PHxhV+jSv3/P0nrrKeTJmwGXJlFTy1D4XxkvxN1JZeoC64fOnPtxwww257Xz+0UcfndtHHHFEbjtPT/79+eefz22PJ2qdTk7ZUuYeWzJIVeRB0OtTq4RHKw6R86E8PUaCa+KyXrFiRW4feuihue2xOBwH7dvLijDmkW0fO2NdGLPj+sF5uY4xRop66nPs79064X57wBganydlQXt0HWP8Sq0sglTGl/qezn3hkEMOyW23YdrtAw88kNu+vpwLYyE9TpKxMq0K14yvdfumvtFufY6ul6MAT4dolThoVSyvlQ7xmFvqqsd7cV+nrbs8GY9EW7r88suLfk888URuP/vss7ntcVqHHXbYomPy8XEcHnfI8VK3o2J5IBAIBAKBwA5CvEQFAoFAIBAIDICxVyzv3Y+essz0aFJddNtKpbuRNIO7L/mdU2J0MdKdSbeyVLqCmTLqpRDofqRL1Wk/uixZedvdl61qsHQnc8089XbUdB6r6LpLm2PmurmLle5Y0gTuYqab1teK83zhhRdy+5577in6XXPNNbl97bXX5vbq1auLfqQHSc94yjDvxbIGvha1A6Mlac2aNblNSsIpwUEOIB4EvQ21KkDTlryice00AZc79cMPOyY9xD3CKcZ169bl9iOPPJLbXmKC4127dm1uO+3He7HdovNcnqS9uDa+Tv36jkquXdflfaJFBVPOTtPRrrhWTg9yj/QD13l92pWXC6G+c61uu+22oh/18OSTT170PlKpQ7y2z7F1OC6pPuqr66Gn7Y8CMzMzeQ/0fZ26Sjtz2pG6Vjtc2r/zZyptofV84b7GUAeXE6+xefPm3HbbpO5QFv68ZiiAj51rQ9lyH5deXbfl0HrhiQoEAoFAIBAYAPESFQgEAoFAIDAAxk7n9e5Cd5fR1Uq3rbtZa4cgOnXGz569QRqpluUjla57Vk/2iuV0N9J9zuwDqXQxcl50P/v4nArhNbgWThmM42DMHi2arpUBxHkzo8NdrKR8vNrxvffem9ukV5iBJ0lf+9rXcpsZeH/5l39Z9GOGJtfUq1OT1qDr3CtmU7/Y9t9R/51+9sNRRwFmADmFTPtpZcS2aGiC9uM2Qnrhhz/8YW7T/iTprrvuym1S/t6Pa0e99AOlWdGeFJDbESkDpzQ4Ds7DdafPHhtVdl5KKeuQ2xzvSVn6XChLyp8hCVJJXzvVzrAHrpvTMLQz2pXvi7w+q117dhjnTIqxdUqG01q0Qe5hTuG2ThoYFnigdOugaO4nrRALXsPpaupBK2ORbVJx/Xh78Fnu+x9t8GMf+1hue1b74Ycfnts8fcRDfTgvfyZR1uxHulF6db+IA4gDgUAgEAgERox4iQoEAoFAIBAYAPESFQgEAoFAIDAAxhoTxdRbrzZKzpb8JcsdOPibVozVcccdV3zHNM6HH3540etJ0nnnnbfoODxmhZXNOQ6mekplKjbjJDxGgNf3lOJa1XNPvXWue9hIKWXe2eMJyDNTzh7bQ46c8Q+tE7TvvPPO4jNj38jhs/KxJJ1//vm5fcEFF+T2n//5nxf9GP+xYcOG3GZ8jlTGBJA/d1myNILHrXFtGM/g5T+85MMo8Morr+RYILdN2lLrpHSOk/ErHuNGG2F8mlTGWXHNPRaENn3uuefmtscwMSbj7rvvzu3HH3+86McxUqc8BoY67HEXnBfn7+s0arBiuceUUs+433ncqMc39XAb5nqvWrWq+I7xKxdeeGFuf+hDHyr6/eAHP8htlh/xchXc72h/Bx10UNGPsTbcS92uuOe0YmioQ60SJqNC13V57F66gPbIsbnesh9LCLhd+VoSfH5xz/P9gmPks9HX+P3vf39uU3e8VAZjYilbX4tWjB+/437m/Xo7aD2DHOGJCgQCgUAgEBgA8RIVCAQCgUAgMADGSuex8qqnhpL2oFvS3Y2kRFoHY9Lt5y5tuqBrB49KJdXHKrutSsV0MXoKON3HvJfTPEwR9TRozpPXdzrTXdfDxtzcXHazukubrmO6/+kWl0o5c/zuMmfZAJ8nSxLwd6TspHK9zzjjjNy+/fbbi37UFcrSKVyvYN7D5UXd8FRsUkhcQ6dWPNV7VOjX1ikxVten+7vldueYvQIxU5a9enWt0rmXH+Hh0Byvl5ioHUD86KOPFv0ot9ph6A7XZ8qTsvYwgX6Oo7JRUu1OeVB+rTFy3qTn/ZB2Xo9ylaTjjz8+t4866qjcZnkCqdQb2vA73vGOoh+/4728hARLVLQOpW2VkuHveF+373FULKc8W5XVCd+vGOrAvYt25N/5s5f0Kvcrp774mePzcgI8WJjz4gHSfj3SzH5f6rDLk89KPkP92dvvEVHiIBAIBAKBQGDE2OZLVErpSymlF1NKD+Df9k4pXZ9S2rjw/71a1whMDkKe04OQ5XQh5Dk9CFnuPFiKJ+pKSRfbv10maR9V300AACAASURBVF3XdaslrVv4HHht4EqFPKcFVypkOU24UiHPacGVClnuFNhmTFTXdTellFbZP39Q0rkL7ask3Sjp77d1rfn5+a3iXXqQw2Xci3PA5CrJhbeOUPAYD0/J7OFlAlhWnvEOHhPFMTHN29PdmdL52GOP5baXwyff7Nw2P7fm7GPEb4YqT2nreDTyzOTEXQ7ktMln+/V4ZITrA2XJeBjnxG+++eZFr+H3YuwG031Xr15d9GPMHefIsUplvIfHczH+oHYch/cjhinLlFKOG3A7YCwR9dv7MQ6K8Q9+xAqPDvEjlGpxYh5Pxmsw3sPjx2699dbcZkyHj4nypQx9v6L9eTwMda61h/WfPWZyWPJkSrzHdnAdGfPic+HYWFrAY6doI4ceemjxHWOiuFd7rA33QsY6XXTRRdUxMWaLJRKkcu15rJDv+604HMqSMX2+h9VKHAzTNilPX3+uCdu+xnwWUYdZgkAq4/z8XrR3rrHHDVLWPBKGJUak0va5h3qMG/WvNg+p3Jv8O9oBr8f9WVpeaYP8m2X/YgtWdF3Xr87zkla0OgcmHiHP6UHIcroQ8pwehCynENsdWN5t+RO66hJJKX0ypbQ+pbR+HAepBrYPLXlSlq0DZgOTgbDN6cJSbXMcGWOB7UPY5vRg0BIHL6SUVnZdtzmltFLSi7WOXdddIekKSTr44IO73g3o7ka60Zj665QV3W81Gk0q3Xl0F0ul+59VdV1Z6Sb2E6MJujZPOOGE3PaqyEzpJEXj1AJdy16xnSngdL22TiFfApYkT8py//337/p7OC1BcIysBC2VOsCUcy81QTe8V0U+6aSTcpv68J3vfKfoR5roiiuuyO2zzjqr6Ed3Pcfnc2Q/VkP3SumkVT3Vn7rMVH93MbfS7BfBQLZ50EEHdf18/f6suE17bFWvblXsrlGwUmkjvL7r99q1a3Ob1Mtdd91V9ON4eYKAX4/7B1PkW7bJflKpI61U8Z5yXGKJg4Fss5+PlzjgPkvddBqDNDdlQnpGKu3PqSHuXdy33JYov3e/+9257Xr/5JNP5jZt3fX1ueeeW/S+/oxg2IFT6DXb97EvJxVe22GbvR47rU25cR2dduRcWabCnxOkU53+JLXNNqvMS+VJA7SzQw45pOjH5zLX0csOEKQA/USQWjV+qdR1rpPrfS9314cWBvVEXS3p0oX2pZK+PuB1ApOBkOf0IGQ5XQh5Tg9CllOIpZQ4+LKk70s6OqW0KaX0CUmfk3RRSmmjpAsXPgdeAwh5Tg9CltOFkOf0IGS582Ap2XkfrXx1QeXfm6gdpEi3HzMePFKfbuZWxfJWZVNSTHRVuwuPbmK6iL1S82c/+9ncpoveK/jSpch+Pj66Yt2lTTcl18wzpWoxS8OW52KgC50y8rn4wbQ9nGrh3EjjSKWMzj777Ny+//77i37M5uKYPPuKmVn33HNPbrvb2/W3h+sG9dIzgDhPjskzfmpZmMOUJQ8Hd9Rc2y5PuuSZ7eZVkemu96rflCflTppcqh9y7RliBxxwQG6TivHsVdoq6X6nILgPtLJ0adNOyfcZRk7zDUuec3Nz+Z4tKrh1YCvnxmt4FXHao+sJDwsnle2HG1NGpOycTvrWt76V29Qbp9j4mTbmWa7Minab477Qikly2fYY9j7br62vMe/POXg/2hyfLy537smeacxrcD91+o32Q/rNTyfgZ+49vt+RfqPMfK/ld057EhyT36tfj1bmuyMqlgcCgUAgEAgMgHiJCgQCgUAgEBgA8RIVCAQCgUAgMAAGLXEwEFJKOQ7AuVjGEJCPdq6a/Zhu6/EFvL7HBzHFc/369bntKbBM1WSMzWmnnVb0I9/PE+/ZlsrYAlbS9TkydZr8so+RvK1z8z6XYYMVrj1Vm58ZT+CxBYwXYhyDc+fky7/97W8X3zHFmtf32Bim07JMgsetrVu3Lrc3btyY276erJzLmADXDcYmeHo446yoo64P46gTMzs7m2NQPEaPcUosV+BxF9Rv6kCryrOnp7PiOGNirrvuuqIfY5UYG+GxkRwj42M83oZxOdSJ22+/vehXi8+QSn3hODzuq9fn5cRdLBf9vN02aycZeIV/gnrqJR8oh2uuuab47qGHHsptxjr5vsjSJ7z+VVddVfTjnnnKKafktlep5/5OW/fSKdRXxnH6Z+5NrvNejX7U8HjRWlyw2wFjnbj+jPGTyueIx65xHZ566qlFxyCVewT3SS/xQrugbDx2jb+rnXgilWvjOkE7aJ0m0K/bOEocBAKBQCAQCOzUiJeoQCAQCAQCgQEwVjqP8LRUus/o5nYXIN2N/E3r4EC6d6XSpU13r1c7vvzyyxe9hlfmpeuQafF0OUtlyihdmU7zkNpxNzvHTlrE+7k7c9ggnedrT9dpq5p5bZ5OYTKd3dNpuR683pFHHln0I71HGsZpUNK7dOmTjpBKipHz94rqTLV1OoE0FKkFdyU71TAq9HbXopA5b9cxusZ5DafzuOZO05GG53qRKnOcc845ue0UG9PsaS9ucyeeeGJuU4+c7uUpBF6ig+tEG/DyGL3cl1ixfNmYmZnJsnCqpaa3rUPESc/wwGGp1E1f00ceeSS3a4e+S2WZmdtuuy23XTeIU089Nbedaufzo1UJm3uw05y1MjPebxxU+8zMTKa4PDSFOlQ70FyqV3h3ufDkCLdvypr27ev6tre9LbepOzzcXSpPBCF16CeM8LlJm/NnC+Xpz0N+5nidHu3XdzmV6MMTFQgEAoFAIDAA4iUqEAgEAoFAYACMlc6bmZnJrlF3i5IKoMuylcFCt5xTSnQ9Op1HVx0rontF6TPPPDO3SbF4dgldv8we8KrZzNzifN3dz0rN7m7k2Oma92rTo6IKOI5+bH6v2uHHTlN5lfUeXm2W2Yq+HqRMmWnirl5SBpSz0zWkDqlD7rK+++67c5sucT+UtpaBIpV6Tr3x9VvmYdIDgfSs34/0G+kyp4BIq5A6c1snJebX4JqwKrLbN7N+uHZ+wgEzv2644Ybc/tjHPlb0oz22qLjWob3Ub8rd6cxe/1pU9/aip/H83tRV7iWepUR9JzXu68HrexYis2yp+6effvqSruH0D9eXeuN0NylMZg96ZiH3Et9XOH/atNN348jO67ou651n7nLcfFa26Fnqt9Ne/N3DDz9cfHffffflNmXzgQ98oOjHPZl7+S233FL049o9/fTTue0nHHCO3GOccqOtu21y3Xg9f3b16xTZeYFAIBAIBAIjRrxEBQKBQCAQCAyAeIkKBAKBQCAQGABjjYnqui5zkM5Bk2snp+3xMeS/ye97ujFjJtj2azK18oEHHij6MWbnuOOOy+1LLrmk6MdToVml109QJ5/OGB0fO/nc1ne1sUpbc+fDxszMTL6HxwkwPqsWxyCVHD7jlDyGjXL2uAum5DLGxOPRarELrofnnXdebjPV1tedpQw43xNOOKHox98x3Vcq42uYduxxLOOIu5ibm8tj8DiB2mkCvia0K8rCT1tnHIbHMFGPGZPhMQrnnntubrM8gZesoB6wKrmnUfMzU6rdrhiH4fbN2Bl+5zrbr81y4i6Wi36cHstH+XGf9dhTjp/y8zgu7k+MSZFKu2XsoZetYWo+1+Sd73xn0Y/xMLRbr7rNeJ0DDzywOnbO2W2OKfytuFzf00aFmq7weci2l0Lg3FnuxfeWo48+Orc9vpNyYhkDL/nDcVDffI0ZG0n4Pslx0JZ8X+FzyOMwabeUmdtHL+uIiQoEAoFAIBAYMeIlKhAIBAKBQGAAjP0A4t6V76mydDfS3e9uSVIGpB3oqne4y46p03TtMc1SKl3QLGvgbkhSAaTz1q5dW/SrVeX2taBb0ikTrgdTdp2CGWX6tLTF7dmvnY+frlNWvfUxknIknecUCilXdz/TXc+1clc0KQRez2lPVmR+73vfm9vuticNRTq3dtis31cq16lWukIafbkKqSw/4mtMioxu8lZVYFJFTklTv31urHLNcXjaM+kE0vWu97RbXsNp3Fq5EO4VUkkh+Ng5f5bKcH3u7z2qSvQsP+Jp4LRV0lStKvm0Hb8eaWjfq2kjpOIYyiCVlCvtzCmZWvV/Vi+XSnskNeSH11Kv3eZITXKP8GeJ/24U4HPT50AZUh99XLSLWkiMVOrq+973vuI76gGfjV4V/5lnnsltnvRA25bKPZ/2TDpdKvd42pXTg/zsuki0qLqa3bQQnqhAIBAIBAKBARAvUYFAIBAIBAIDYIdl57m7jK5buh7dfUmXOTNIWgcau9ucruCbb745t31MvBfd/56Fwn7HH398bjMLQiqpANJ5LarIqRW6Tumy9Tk61TJszM/P5zm4O5eUQas6LmVGWsAPwGXVYc/W5L3pmvbq83Tx8zBbzwQh9cTrrV69uujH7zgvz+Shrv3whz8svuPaUF9r9M8oMTc3l2k21znaGWXjh/0SpFaZ8SOVWXdut6RYKBt38ZNOIN1CPfKx075ZTVsq58Xv3Nb5nWcW1u7rttlTQq1D07cHpH+ccuRYSKu6HdSq1HsmLn/HKv5SaSO8ntMwzMaljbiMSMdyfUn3OCgjz+YiVes2RjumnFonSIwS/X18r+VnPkdc7qTBGHLgVclJq/nBzpQ17cXDW7h/cD9ds2ZN0Y9rx/3VQyJ4DVKFHrLBfcBti3rL63u/fuzLsc3wRAUCgUAgEAgMgHiJCgQCgUAgEBgA8RIVCAQCgUAgMADGGhMlvcqDsvqpVPK5TAX3CsSMjWD8g1+PcSVMc5bKuADGVji3vmHDhtwm/+o8OO/9jne8I7edv2YphKeeeiq3nXsm3+wxI4wT4PU9jsVjQ4YNpsR7iQOuFcflabfksFkB3MF1c76cKcecs8+fJ5Bff/31ue2xIDyBnLFYDsY6Mf7A07Ipy1bMHeNHvAKzx5CMArOzs3ltPe6lVn7BdY6xLbTnY489tuh3zz335Pa9995bfEdbYvxJ67R5ytZj4RijyHIbHvNAeVJmntLOmB2XJ+2RsZwexzdOeIxeLSW+Ba6Hrxuv5ycNsGTMkUcemdseA8prsvyBV4Rn3CBT2L00DWOd+CzxPYHX81gn6iH3YLcNLxEwCszPz+f5ui5x3Nx3fY25L23evDm3fW/hM9D3Wt6b9/Jr8LlJuXs/jpG/cdDmqCu+/7SeNfyuVU5mkNJA4YkKBAKBQCAQGADbfIlKKR2cUvpuSumhlNKDKaW/Wfj3vVNK16eUNi78f69tXSuwYxGynC6EPKcHIcvpQshz58FS6LxXJH2267q7UkpvlnRnSul6Sf+fpHVd130upXSZpMsk/X3rQl3XVSu8khKiS83d80zVZAVVvy7pIT8Ulgcm0j3rKZOkBN2NS9DFSDezu0PpRiSF52712sG8UklrkOp0Ss1d4QsYmiwJpzW4blxTd0XTZc40aj+g9Kyzzsptp8vomqZb/6677ir68TPX29OoSevw0Gmnlbm+pAlcD/mdu9h5b9IOTl016KChyrNfF6ewOG5WB/d+dNezvIe7yE8++eTc9kO/WfGfdK8fGEyKifrhZSQoN47J7ZlzpOufFKBU0nSt6tHUI9fnXteNGhuaLFNKmVr08grcd2ibTnWRmqR9tw4qdkqe61MrZyKV+s7QC9q2JN144425TdrPZcl70a5aYRgeUsF1o027LjeeC0OT58zMTB6r06k1ytvHyc98Rvn4aVdeWZ5ryT2+tTdyvCxPIJWlCzh2p+m451P/fF/kM9D3Yeoz71XrN9QDiLuu29x13V0L7V9IeljSgZI+KOmqhW5XSfrQku8a2CEIWU4XQp7Tg5DldCHkufNgWTFRKaVVkk6WdJukFV3X9X8qPC9pReU3n0wprU8prfciioEdh+2Vpf8VHtixCNucHmyvLFvnhgXGj7DN6caSs/NSSm+S9J+SPtN13c/p7uq6rkspdYv9ruu6KyRdIUmrVq3qehe9U1h0ebeynfgdXfDuIqZL0N2I7HvOOefktmf4rV+/PreZAeTZIMwYZGVsz86ji7hVUdwPxSX4O17PXaBOTRLDkOXKlSu73lXrLmGnFnHf4jMzQbhWpHukMoPOr3HHHXfkNg8iJZ0iSaeddtqi43P3OKneRx99NLddJvxMyscpO352ipUvonzwuXt8Wxkjw5DngQce2PX3dTuo6ZzLvWa3XFOplKfTQxz7TTfdtNiwJZUVjlvVhUmp+7wIUpOkBL16PMfr9s1xcB5+3xZlMAxZrlixoqtRs9Q5UlguS2bakfLwLDZWAXd6paa3fi9WtCeFd/vttxf9SMkzg9VpJ9oZwyF87JSfv6hUwiG22mfHYZuHHHJI19/H9wbaJq/t46JsKE/f11h93J/RJ554Ym4/8cQTue2HgxN8Vvqasjo6sxzdXngoPGXoa8G9yfcVUrns53tHrwfLOfR9SZ6olNKu2qII/9513X8t/PMLKaWVC9+vlFQ/AyEwMQhZThdCntODkOV0IeS5c2Ap2XlJ0r9Jerjrui/gq6slXbrQvlTS14c/vMAwEbKcLoQ8pwchy+lCyHPnwVLovLMk/Zmk+1NKfZW8/y3pc5L+I6X0CUlPS/qT0QwxMESELKcLIc/pQchyuhDy3EmwzZeorutukVTL97tgOTebn5/PfLhz9YxdIGfpMVGMMSGn7am85Gy/+93vFt8xRoOxFV59lnzzk08+mdvO2ZJPJ9/vabPkWcnteiAo18Zjp5wHrsHvLQ1XloTLknw819RPvWe5CnLW5N6lduwb12flypW5/c53vrPox3gBpuIzbVcqY0YYE+VjZz9ez1NmWRLA9eall17K7VaVbB9jj2HLs48jcJ3jeKir3o/jpGx97c4444zcPvPMM4vvuEbve9/7cpv2J0nvec97cru1dtQd6gDTq6UydoP6y3lIZUyMx9gw5qMV3N3rCOM2hilLlpJpxexwP/L4LMaocs4e98TfeRwmbZ/7M2MXpXKvuuGGG3LbbZ2xT4zD8bg1xrnw2l7+gmP3WKeaLH3+tZI9w5Znr2u+r/P+fB56rBNthPFuXsKjVbqAJw1wX+O+K5U2x+v56RDUCY9rI7hPUk6+/3D+/t5AO3BZE/1zaKglDgKBQCAQCAQCWyNeogKBQCAQCAQGwFgPIJ6Zmcnudq/iSzc+XfKeDktXHF2F7tKlm5JpuJL0jW98I7fpUvzQh8q6Z0y3pXvXaySx8jTdrU4Z0N3IObqLmO5Qd8vy3uy31DIDw8L8/Hwei6eu0v3N77zsAOkQtq+99tqin1dtJ+iaJp3AVF2ppGt4L9cNunFZ1uKAAw6ojp2659RbSw+pA6QgPHXedXsU6Louu7JbZQdoL06hk9akfbNSuCTdcsstuU1aQCrpda75xRdfXPTj4dCk+tyNT3m0SlbQrihbnyP3KZcTP5Nmcfn5GIeNlNJWtGaPmt62DodtpYTT1v0a3KtZBsYPKmapE54S4JQo9zjOjynwUkldca2d2mxVyeb8+Tuvet4qrzEspJSqJXE4Hq6xH2LOfZj27Cc28D5PPfVU8R0pN8qsdWA57Yrjk0rb5HqTvpPqJ4I4fUwZ+jOJ1HXrEPV+HL4HthCeqEAgEAgEAoEBEC9RgUAgEAgEAgMgXqICgUAgEAgEBsBYY6IYR+P8NHl38p6eFslYBsbKeLxJf1K6tHV8EOMaePI4yyJIZUosOXiPkyCPTH7YeVmOl/2clyVn3To3iTENXqa+lhY/TPTjdL6ecyP/7P34HTlxP72d8U0ed8E15Zw9JoTxWLyXxw5Qv6hDPnby9Lyvx8vxen4N6jnjaTzuYjlHEAyKmZmZrK+tI5mo0x6TwPgmS98v+jFOzL+jPHg9j51iXCPtpRVPQb10+6C+sO1xT7yG70201ZYNjyNesU9Pdzvgvbn2njrP8gLs1zo6xePMqN9su81xn2VJCV97ptJTD11fWS6H5TV8jtwHWsdwteKeRh3fJpUlK1rHmVDn3F74vOUe5c8olqVolTph+QPXZ+rIhg0bcpvxjn596puvN/efpca4ObhOfjQX0evwcmLdwhMVCAQCgUAgMADiJSoQCAQCgUBgAIyVzkspZdefu8tqqbfuviSVRvfl4YcfXvTj79zdyOszjdpTkUmzPfzww7ntLlC66/mdz5FuZ37nLnLO0atcc850kfs6uWt92Jidnc3u8Va5Cq6Nz4XuV1J77p6n+9kpFLqBWye233///blN124tFdyv7eUZqBt0D7vMKRefF13TpC6csmyNcVjoui6Px93k1KVa+rgkPfvss7lNSsVT1bn+mzZtKr6jTjhdSHD9qX9uB6SCSVV4P96XVJ+fYsDPTtO5fvfwdertezlp1MsBSxy4LGkXnHOrwjNt0/vRNp2Gru2FbiNcb16/tT6Uv1cN5zU4x1Y4QYuGJ4Xr++q4Sxz4HlIrq+G6yD2Ke5mHpvAarjvcl0jJu07wecvnlVPovD7X0a/HOXIv8fFx7C4X7ke8vt+rH6/rcgvhiQoEAoFAIBAYAPESFQgEAoFAIDAAxkrndV23lfuwB91tdBt6hgbdmXQXtzLEnM6jG5CZYJ7ZQ1efU3gEsz44Pj/ckeOo/UYqXa9OS9UOFHX3ZStTYVjoXZ8uI8q4lZ1GOXOefj3+zqkhuvVbmZF0YfM7z4ykLOjib2WCtOhBUoItipXX9/nXDjkdJpid53PlmvA778dsxlrmpcP3A1azpqydMqYtcb1cnszo4ng9m5fX4Bjc/njf1kGmLXq29bthoZeZ7y0cP/Xbda5Wpdtlyev7NWgL7PfMM88U/WiP/I1X0/brL/YbqZRZjdqTtj4pguB+yt/5GMZB53Vdl9fP58D5kcLzfZJyb4WLkEJ1WfNe3Gtba8DredZ1rcK407iUJ3XRqfYWOHauRWu/XirCExUIBAKBQCAwAOIlKhAIBAKBQGAAxEtUIBAIBAKBwAAYa0zUzMxM5iM9zqPGezqnz1giphh7SiL5bueHH3vssdwmL9uKj+F9/Xq1sgteIZdjrJ1I7p9bacOtuK9xcPX9PfzejLWocfaOFifO079b1a+Xen3GJrHyrlSuG6/n8T/UPcrL4whqqbU+jlbcVy0WZJhgVWSfA9eE+ui6STugTngpAMYyuMwoT8ps48aN1Wu0quJTNrQX2qlU6iz3AbfNVso2x8TrUQekV9dzlDba24mvL/c06l+rnAD10edSu55/16rszbXiOFxG1Cnao8ec1eKgfI7U89b4OA6PmfGyOKNA13V5rG6bnF8r1ql2OoI/X1tlL7j+jBt0O6hVwvd9rfYs9+dwLR6Z9iyV+uclKwjOq6Zj/u8thCcqEAgEAoFAYADES1QgEAgEAoHAAEijqpq76M1SeknSryS9vK2+Y8BbtePHMa4xHNp13b7b7rZ0LMjyae1c67gtjGMcQ5elFLa5g8YwSlmGbZZ4rcszbHO8Y1iyLMf6EiVJKaX1XdetHetNJ3QckzCG7cUkzGESxjBJ4xgUkzL+SRjHJIxhezEJc5iEMUzSOAbFpIx/EsYxCWMggs4LBAKBQCAQGADxEhUIBAKBQCAwAHbES9QVO+Cei2ESxjEJY9heTMIcJmEM0uSMY1BMyvgnYRyTMIbtxSTMYRLGIE3OOAbFpIx/EsYxCWPIGHtMVCAQCAQCgcA0IOi8QCAQCAQCgQEQL1GBQCAQCAQCA2CsL1EppYtTShtSSo+llC4b432/lFJ6MaX0AP5t75TS9SmljQv/32vEYzg4pfTdlNJDKaUHU0p/syPGMSyELKdHllLIc5rkGbKcHllKIc9Jl+fYXqJSSrOS/q+k90paI+mjKaU1Y7r9lZIutn+7TNK6rutWS1q38HmUeEXSZ7uuWyPpdEl/vTD/cY9juxGynB5ZSiFPTZE8Q5bTI0sp5KnXgjy7rhvLf5LOkPRtfP4HSf8wxvuvkvQAPm+QtHKhvVLShnGNZeGeX5d00Y4eR8hy55ZlyHO65BmynB5ZhjxfG/IcJ513oKRn8XnTwr/tKKzoum7zQvt5SSvGdeOU0ipJJ0u6bUeOYzsQslzAFMhSCnlmTIE8Q5YLmAJZSiHPjEmVZwSWS+q2vM6OpdZDSulNkv5T0me6rvv5jhrHtCJkOV0IeU4PQpbThZDnFozzJeo5SQfj80EL/7aj8EJKaaUkLfz/xVHfMKW0q7Yowr93XfdfO2ocQ0DIcnpkKYU8p0meIcvpkaUU8px4eY7zJeoOSatTSoellF4n6X9JunqM93dcLenShfal2sK1jgwppSTp3yQ93HXdF3bUOIaEkOX0yFIKeU6TPEOW0yNLKeQ5+fIcc1DYJZIelfS4pP8zxvt+WdJmSX/QFk75E5L20Zao/o2SviNp7xGP4WxtcTneJ+mehf8uGfc4QpYhy5DndMszZDk9sgx5Tr4849iXQCAQCAQCgQGwXXTejioCFhgNQp7Tg5DldCHkOT0IWU4XBvZELRQBe1RbajZs0hbu9qNd1z00vOEFxoWQ5/QgZDldCHlOD0KW04ddtuO3p0l6rOu6JyQppfT/JH1QUlUZdt99926vvbZUZ5+fny++2xI/tgWvvPJKbs/MlM4yvvTxN7Ozs9WB+osi783rt/rx+nNzc9Wxs91CbQyt+zq4Tq1+mzdvfrnrun23MaRlyfP1r399t/vuu281XqmcT2suvo6DoCYXX1OCMnKZ8xq77LLLov/u9yWW80cJ+9bWzPHTn/506LKUpN12261785vfvNW4Fvu82Ji3Ne7W7wa5Rk2Grf2ihVo/t+fa/rOt7xbDL37xC/32t79dyoaxLHnuvvvu3Z577rnVmBb7XMNS9zR+19rT2R50/+TYa3bq/WrPFam99xO8fusZ8fzzz4/cNn19lvqs4Lj5G79ea01q6+rXqO3rS92TW3tA6xqt/b+mf7Xr/fznP9dvfvObJT3Mt+clarEiYO9o/WCvvfbSpz/9aUnS7373u+I7KsCPfvSj3N5tt92KfjQEKnevZD24SAiASAAAIABJREFUaL///e+L7/j5DW94Q27/4Q9/KPr95je/ye1+U5KkH//4x0W/173udYu2W5sKr/3GN76x6PfrX/960fv6NTmO/oWmB9fzH//xH5/WtrEsee6+++668MILJW0tS87nF7/4RW77XPgd0TJeV/q3vOUtuf2zn/0st11viNe//vW5/dvf/rb47pe//GVuv/Wtb83tn/70p9X7tnSN4/XNhbpMPfQxcf5f+9rXhi5LaYv9fOQjH9nqftLWc+rhekt5tjZ06qqvCefe2kwpQ9ot/92/az2g2Y/z9+tRZrR1qVynXXfdddHrEV/96ler4zEsS5577rmnPvWpT201Xv/s3xEcP23JZcI9uLWns0079evTXlxevP6vfvWr3Kadej/Oo7Vvt/6g23vvvXPbnxGU+T/90z+N3DZdH2kvb3rTm3Lb90mOk88X12HuQy5rzp3r6vfiPkAZ+n7BNee8+Gz0fnzO+/i4R/vzkPNkP66Z9KpNfOUrX9FSMfISBymlT6aU1qeU1lPxA689UJa+YQZee6A8feMKvLYQ++x0IWzztYPt8UQtqQhY13VXSLpCklauXNn9/Odbio36G/0+++yT2/RY+IbAv1paXiS+ofvbNvvy+u4W5r1aisw3bP61s99++xX9+NcAr+33ZT+/L+fFv8Baf1UuEduUJ2W5xx57dL3XhnKQ6n/F+l+C9OZQDi1vAv8qlMq/UvhXhY+J16h5gKRy7fkXVWtM/E3L3e9/OXEuL7/88qLzkErv2BKxbNvcd999u37sPk7qd+svdl/LHr52XC//K5YeC66Dr0HNy9EaE/8C9zFRZzlHvzav5zZX88q0vOlLxLJsc+XKlV0/19baUA5uVwTXzWVM1sDXlPsu5bzHHnsU/Whn/M49sjVPonu2qK+t+7b2nJqMWnv1ErFs29x///0zned2QF2lzvnzkGvCOfizsfXcrIWP+J7Xh+xIWz+/CV6f4/U9psZsuIy4X7Q8hi1vWz+X5YRlbI8natKKgAW2DyHP6UHIcroQ8pwehCynDAN7orqueyWl9GlJ35Y0K+lLXdc9OLSRBcaKkOf0IGQ5XQh5Tg9CltOH7aHz1HXdNyV9c0hjCexghDynByHL6ULIc3oQspwubNdL1HIxMzOTeUvGgEgl796KReJnthlfI5UxDs6t83fkTv0a/B25aOeKH3/88dwmx+oZXeR2OT6/HmPCWmUBeC+f46gxMzOT4wg8nsD79fDsjFomjvPUXAMPaPfMnB6tWIVW/Ae58FbsDuXHeD7n0n/yk59Ux0T9oMxb8QyjQkopz9F1iXNvlWKopQ63Yrrc5vi7WrabVK5dK1OI8R/77vtq9rnHXdRinVoxbkvNYvQ9rJfnUsuhDIJ+HTxuhHFB1CvXOe5J7OeypO17vBT3ca6V74v8bvPmzbndx8/2qMnlwAMPLPpRBwbJGPV78RquX0OIRd0muq7L93F7YczRUuPwaC8e40W7cJ3g9WsZ6VK55zE+yvWd+yZtx58njHmj/bl+UL4tuXAeLvfaWFsY5wHEgUAgEAgEAlODeIkKBAKBQCAQGABjpfMId/3SZfziiy/mtpcJoDuPbj53rZMe9BRYpkLSFUlXslS6bukCdBqj5sb34pJ0o9LF7xQV3ah+r/333z+3uU7uZvbio8NGSimvndMVdKXSPewyoiuZbmp3RdeKskn1lPhW5eoXXnght33tuaYsyeDrWaMTXF85PqfJqKOEpwW7y30UIGXg1Fmtqq9TYtRV2pW71unGb5UQoLu+tlZSnWaQSjqBtIzLk2vOdqs8itOsXDe2W0UoR4VeNk5XUL9bpzCQ8m4VtnTKh6jtz6R7pDK0g22nVGg/tLlWcV7K3+2I9/I9jOvG6zmF1CrqOyx0XZfXuUX1c290e6mV7fDnMG3d16RGu7p902b4m6OPPrrod9hhh+U291p/ntaKGfvYW3Rm7Zk6DDo2PFGBQCAQCAQCAyBeogKBQCAQCAQGwFjpvK7rskvWXat0M5Pacfcp3ZetM65IGXilWmaY0C187LHHFv2efPLJ3N60adOiY5Ckhx9+OLdrVJZU0gm1rDJJOuCAA3LbqZUadUG39WL3HgV6t2jLJVo7204qXc6kwXx96X51WdLVy+rJTsPQTU1X/bPPPlv0e+qpp3KbOupzpEuc89i4cWPRr0WrUratDMdxZV729uSUAcfWyqLkXOl2Z1acVLrhfR8ghUAZ+hrUMlh9vVesWKHF8OCDZWke6iazx5wW4Gen5bg2XEOvQL+casiDYGZmJtuQy5L7RE2HpZJeIbXHfdrh9Btl9Nxzrxblfv7554t+TzzxxKLj8HAI6sChhx6a229729uKftQ32jrHIJUy9/2SlCPl1zqjdBxoZWtz/X2c/I7z9j2ZFKc/e2tZ7a2zU4844ojc9mfekUcemdu0K7dN6gSf6/fee2/Rj+NtVbunPH3f7fe0yM4LBAKBQCAQGDHiJSoQCAQCgUBgAMRLVCAQCAQCgcAAGGtMFNPiPY2b8SEtnpn8KzlzcqVSyel7vNBBBx2U24ccckhuM2ZJKvlixgV4tXWme3JeDzzwQNGP8VcXXnhhbnvcBssYeMr8D3/4w9wm98x/l0YfdzE3N5fn6rwy+W3y1B7jQDCGxvls8vutCvaM0fH4B+LRRx/NbY/P8M89PDaNcWuUOdN2pTK2wuO0uE7U63HEsy2GPm7M44oYH8D19riBWuyarx11s1Wd/qWXXsptxidKZdwL27RnqbSlVvXqp59+Orc5fy/vwDm7njJOhDrrc/TPw8bc3Fzeu1pp4Jyb2ybtkTEprRR732cZW8bYNz8x4JFHHslt2pLvYYxLbJV34fV5X+9HGXl5G9og2x7fNo54xZRS1juPIaQ+cq/xfpQT9c/jmWgjXmqgdvKAy732jPbnK2NYGUvK57NU7pvUKR8PdbNVTofj8H79ei7n+RmeqEAgEAgEAoEBEC9RgUAgEAgEAgNgrHTe/Px8dg27u5HuddIcfuAi3Zd0VbvrkS5YT4E96qijFu3nNArdiKR5PB34nHPOye3vfe97uX322WcX/UgxMt3zpJNOKvqRwvPDNek6JYXHa0ujdzPvsssu2W3u6f81F7pTAaRaSPu1quh6iYNnnnkmt0n5uOuYa0VdcxqDYyKt43pIHeCYPAWcrm1P8SUFzes7tTCuNOrehe2UKd3/pOZcnpRbS+61Q1O9L2meDRs2FP24D5x88sm5vXLlyqIfx/vYY48tOgapdN/XDmuVSt12udTohFHTdw6GTfj4uce1TgngvsP1cOqee5VTIGeddVZuk17/4he/WPTjmjIl3qlxlh8hhev3pe5xf2iV0/AyHPwddc2fEeMqP1I7UJrPKMrQ5UmqjzpBSk0q19xthCEMpAc9TOHUU09d9Poup6uvvjq3v/zlL+e2lxo644wzcpu25M/hxx9/PLddnrW1qZ0mEHReIBAIBAKBwIgRL1GBQCAQCAQCA2CHHUDskfV0sdEt7plwdCPSPetua2bpuNu9VhHd3YN0Z77jHe/IbdIMUkkX0vV75plnFv1Y2XzVqlW5vXr16urYmY0glXReiwJzV+ywMTc3l8fm60sXKakud/vSnU4XvLvI6cb3aua1TAvPoiF1Rgpv7dq1Rb9TTjklt0kBsqqyVK4vs8icuuK8nArhPEkderXh2gHXwwQPIG5lIHH9ndqoHbbs2Vicj1dF5rq2Dkol7UO98uuRxqCc/BBcjpe26fsK9dlpT+5h/J3vKz0lMSqaL6WU165VeZl7ic+TNB3bThOx4r/Tb6S2aUsuy9NPPz23uZe2Dt4m7e5UL58t1AffO2hzzNiVSp1qHVjruj0KMAzG9wbOlfuaPzeoB9yHPBOOtu9ZuqTKjzvuuNz2jFjaz0033ZTb9913X9HvP/7jP3Kb6+qnPpAe5LOb9KJU7heeiUn9rp16Ir26NrVD1xdDeKICgUAgEAgEBkC8RAUCgUAgEAgMgHiJCgQCgUAgEBgAY42J6roux0N4CiLjJBhr4PEZ5D1rFcWlkjv2yuFMu+SJ0R7jwNIDvO/FF19c9CPH7Hw/cdddd+U243y8KjnH7twsuWPy/R4347ELw8bs7Gzmp51XrlVrdj6f2LRpU267HBj/wMrSUhk3QZ1i/IsknX/++bnNE+CPOeaY6r0YE+CpwFxfrr2vBeXM+0plDEMrJmwcJQ5mZmayHnu8Yk2/PT241s/LUjBGyEsSMI6GMvSq/ieeeGJus8SBxzqx9AfjrXxfede73rVoP79v7UR5qbRHxsb5ntDLczlxF8sBSxy4LBkHxX3Hx3LCCSfkdi2eRCrjYRgf5b+74447ctv3/sMPPzy3KVePiXrhhRdym3rjcV+0QcZL0bal0q58/+R68Dufv8e7jQKzs7N5r/CYP9oc9w23OdoF199jjlmGxfcdlqygHnk5GZb5YRzUN77xjaIf45v4bKDuSaUN0m79mc+YPB8T9wHahNtmv4ZR4iAQCAQCgUBgxIiXqEAgEAgEAoEBMFY6jxSQuxHpVqOLkpSdVLpW2fZ0Ybr67r333uK7G264YdHrM71dKl2AdM+zMqpUupaZ9u2lEG6++eZF+5Hykcp0YLq6pbJyOvv5/MdxAHHvWvYDKLlWdOm7251zocy9FAApTE/JJejuP+2004rvOEamUR955JFFv/Xr1+c29cspCK49qT1fd6bityom03XuVJOnK48CXddlubUodI7ZqQ32a1ElXEs/8JnXp7vfaT+mxTPV2cto0IY5L6f9SM1R35xaJtXXqgZOG3C595TBKCuZ9+vfKvVBOtbTxTl+UkNOE9EOuL9J5QkCpJrchrlXs7K5V+5fs2ZNbpNGdIqLvzvvvPNy2/X1zjvvzO3WnkM5+Xq6Ho0CLD/ioL5zLB46wecraVfXb1LoXjn8mmuuyW1So05p8tnOMZEOlMoSQA899FBuuyxo+xyvl1ihPrtt0VZJD3rYR/+7oPMCgUAgEAgERoxtvkSllL6UUnoxpfQA/m3vlNL1KaWNC//fq3WNwOQg5Dk9CFlOF0Ke04OQ5c6DpXiirpR0sf3bZZLWdV23WtK6hc+B1wauVMhzWnClQpbThCsV8pwWXKmQ5U6BbcZEdV13U0pplf3zByWdu9C+StKNkv5+CdfKPKbHLpAHZQqi8578zFgUnjoulcd0eEwGeVByzc6DMn6A3O7//M//FP1YuuD222+vXo/xV+RlmdYtlRy/x1WR1ycX7Zx5jdMdpjz7WA+PSWA8DGMXXA7k6XkkjsdYca08Roz3ZlyHl42o6Y3H5jFe46tf/Wq1H+MKGNfjsTvUIY+N4fypX75OteMvhilL6dWYNU+Lp86xLIMfLcQ4Ca43y1dIZayTl0mgTnNNPHaNv+MxER6vyFRn2qkfycR4PcZseVwO43fc5rintUqd9L/zsgLDkmfrmBDqEsfo8UL8zPgSP5KDsqUNS6UNUua+91OPGEflMTQEx+H6ymO0jj766Ny+9dZbi360dY/roT5wv3Bb9PidHsO2zV4GXrqmFgfkscScD2NH/TnB4808Xoj2UyuVIZV6xf2Q5Ssk6Z3vfGduc14skSCV5WVa5WSoR15+hLFUXq6G6GMBxxETtaLrun53eV7SilrHlNInU0rrU0rrx1FTIzAQliRPynIc57kFBsJAtum1qQITg2XbZuyzE4uBbNODvwOThe0OLO+2vLJVX9u6rrui67q1Xdet9b92ApOHljwpS/9rMjB5WI5ttjwngcnAUm0z9tnJx3Jsc9SFkwPbh0FLHLyQUlrZdd3mlNJKSS9u8xcq06jp3pVKlywrRXsKNF3gfEP3v6Tp2nPX3Hve857cpoJ66i3dwpdffnlu+2nU999/f24zpZquUamc8zPPPJPbLa+O0xh0d/MvTr+Gp+JuA8uWZ0opy4Lp0FJJh5CS8Rcv0pici4+dlb49Ffucc87J7drp7T6Oxx57LLedkiENVUutlUq9YZVfh7uVa/ciBVpLiV8iBrJNjsFpJsqtVYmbFAK/q6VnS1vLiZ95UrzT9aRR6J73KsukmEgLe2o615x65LKgDfu8SK1wLTwkYZlp1APLU9pad1gug5SMUxxMCac+OE3E9XVah7bKPcFpv6997Wu5Tbre937uraRZL7jggqLfqlWrtBhY2kSSbrnlltz26tdcN87fZeklH7aBgWQ5MzOT7cl1jjQdn5seIkLdpG063cs15rNGKstPcG/08AvaIPd1p+45doZLPProo0U/joO/8bFTx9wbS/lyT3b76G1yOadEDOqJulrSpQvtSyV9fcDrBCYDIc/pQchyuhDynB6ELKcQSylx8GVJ35d0dEppU0rpE5I+J+milNJGSRcufA68BhDynB6ELKcLIc/pQchy58FSsvM+Wvnqgsq/N9G7slsZQHSxeaVougRJMzDjQCrprQ984APFdzxYmNXMWX1XKik30jlOBdCNyqwfVquWysMwmeXkdBDpLHcr0rVMasGrWtcolGHJs+u6LEuvRM7PpAU824PUUCsYlm5bp3VYVZfyo+tZKjNzmMHl7nhSurWsIR87aRmnuCgHd2d7xevFru3XIIZpm6Rnnc4jxUbdbGXOkgajrkv1ivYOZhu5flAetE2nm+jGbx3yTJ3leGsHlC4G3pvzd3n23y1ycO7Q5NmvXUsfSd041UVZcs733HNP0W/dunW5/f73v7/47uyzz87tL3zhC7l91VVXFf14fdqtH/TO0AaOg9WupXI/5v7AjG2/r++fpCZpt24btUyvYcuyFvLBOfAZ6nFUfN6SznIak3Sq6zpP1qB++57MzGXqvtOzX/nKV3Kbz9pvfvObRT/aFZ/5Rx11VNGP9LFTfbU91OfY70dRsTwQCAQCgUBgxIiXqEAgEAgEAoEBEC9RgUAgEAgEAgNg0BIHA6HrumoKIfl5wjnoWoqxx2cwHsljD8jTkkf2FHzGzjA+g6eJS2WsE6uyetwXuV3Ol1W9/Xc+L3K4jM3yVGyf87AxMzOTeffWSfeMyfCTtRnnwt84d85YJ6apOygvXw+uPWMcPHaAOvDggw/mtlcsZ2wQddJLMPBeHt/HObPt/L3bwCgwPz+f5ei6w/iCVrkCrjn7uR3weotU7c5t6rrHhNB+GJfi8Vf8HcfkNse4RF7D9yVew+MpqN+1PUZ6dY6jslGeDOH7B+2A8WiMSZHK2BiugccV0fY99ovxprQXn/fatWtzm/unxyvSzhgPc8011xT9KFvGTrm8uP8wjkoq9aZ22sFiYxw1fF+rVfD20wRYJobz8YrrXAe/F2OpuJd5HCL1naUW3A6ofyx74TbHNacO+Ni5Fl5CibFfPMXA5d6vzXJsMzxRgUAgEAgEAgMgXqICgUAgEAgEBsBY6byZmZns0nM3M13oTIX3Krh0H/M37mZlyqrf6/rrr8/tp59+Ord5gKpUuilJQfghszzkki5KPzz4jjvuyG26vk855ZSiHykDd7OTHqPL0l3V7lofNlJK2eXpFA/pG9K2Tt3QnU5X7wMPPFD0Y5qyp7/S5Ux60ylG6gNTcl2/uG6keGqHAEulm9tT9jlnT9On+5n65fo6LvR0VCs9mPNzHSOdRbm7LLheXvaB68XDbVslMEjBttafOuBlRUgN8BpecZ7X8PR2jqNVCqGfs9Pbw8Ls7GyWoadqk6aoVaKXSmqcc3GaiHbrlaZptywx4vcipcI93VPYSUlRR1uVyFs0+THHHJPbroekH2n7XmF/HFR7SqlaQbt2IL1TZ6SwSHd6tXHSkywFJJUUHvXDqS+uOXXs7W9/e9GPa8wSRS4LljOh3boNU4+8an0tRMb32n5PXo5thicqEAgEAoFAYADES1QgEAgEAoHAABg7nde7zd21Srcz3XJOBdD9xsh/j8anu/fuu+8uviPts379+tx2yoBuT2aF+aHAHC/dnDzoVipdv8wW8PvSHerZXnTL1ipEe79RYG5uLrtI3fVZo1CcaiEFy0wed0XzgOdWBWa6fZ2SIm1Hisbd2aSQeC93Hfs4enjGCHXU51+jyVru8VFhZmYm67HrDuXRygCiDrLta0VKxN3pvGbrgG3aZsuWSBmQJnB74XfUZ79v7ZBlqaS9qH9O7fX2MarsvLm5ubzGfg/uVbRTp1C4BpSXnwTA/dNpOma0MuP2tNNOK/qReuGB4n49UvnMEvRQDl6PduVZX6R4fM8hqKOuXy2af1hg5qzrI+kyUqF++gZpOs7Hw0W4N7ay33j9G2+8sej37ne/O7cZOnHbbbcV/c4444zc5mHjblc333xzbpNCP+yww1SD6w7lxus7TdrrelQsDwQCgUAgEBgx4iUqEAgEAoFAYADES1QgEAgEAoHAABhrTBTjaDxmhXEY5GJbMRPku3nCtFTGSXjsBqvYss1SBT4OxiY5D07elzw7q+9KJX9N7tmrJzvHTzBGjLyt89cefzNszMzMZG7Z4wQ4LsYaeNwF4zW4hl4Bl7FOHl/CtWP8FWMrpJI/ZwkNxkD5NbyaLcEYD+qDj4/XcK6/Fvvn8RnjKHnQdV2Wj8+b8QWMo/Fx8jvGGiy1cr9Uxjxs3rw5t88666zq2Kl/noLONPnzzz9/0ftI5Zx5PY8p4mfqilTOuaUT/XfLibtYDrquyzFNrVhJxot5rE2tpITHaJ599tm57XGunPef/Mmf5Pbxxx9f9KNdsOSMr+8NN9yQ26yo7qnzjF+kTv7gBz8o+vFEApcz5895+N7sMbujQNd1eW09/pQ2SBn6XkNd4/p4WR/CYyP5O8rwwgsvLPpxjNy7WOJHkjZu3JjbjKPyshG8L2NYPQ6aqJWEkOolGKRX3xtav3eEJyoQCAQCgUBgAMRLVCAQCAQCgcAAGPsBxL3L0V2/tTRSpwzodifN4HQW3fXuNudnVq310gWkBFmuwCvksjIvq7o6LUJ6j3SQp9nTFevrxDmzVEPLfTsKMO3WqVlSiUxzdlnSZUo3ubusTz755NxurRXX1w8qpixZ8sLd8/9/e+fza0d53vHvwyVdZRO3yFixBRUyQpZABAUwFFClgEi9qcSiahcVi0jdZJFIWZSUf6Cr7LpBSlUWIVIkRwqIBUpQCuKHQgFV4B/YbhERrRw7XbXKCl+mi3vO5DPf3Blfj8+Zexh/PxJizj1zZt55n/d9Z/x8n+cZSo68rg8++KC37UMVyykTeD/xXNx2iWDopb+rZDlmvP/7Xqbr9mS7aXeXSigxeSoyJSamqrvbnRIvj+fV/48fP77rfj4/mOo9lPpPiaOvzIXUlYD6pPZ1lThgxfIhCZ3yj88Dlubok+eX51pCSV6SHnzwwV2/87nJ8iaUfN5///3OfnyTAeeZy/+U+HlfGZLCPBzE32SwZGgsr4utra32XuJrQ5/U7msNP7PvKJlL3b50GZr3yiNHjrTblGqlrnz/7LPPttu0s9Rdr1km4fDhw539KOsy5MblPM5Nf2E8YV+4bOfzZS/EExVCCCGEMII8RIUQQgghjGBSOa+qWvcn3fZSN2OKrntWWpW6rrih6slnz55ttz0DiC5ZSkX+0kK6PelGdUmJrmC6d70aLDl69Oiuv5G6fTGUKcV+8owGVq9dB1tbW20/eBYN3apDFct5nXSrsm+krsw65H7lmHLZgTaidOHHoxTAMeTuYV4LXf+Uc6WuhOCyg8+B3drnx5gCd+Pz/ENZhBybnCOe5UN5kllWUrcvuU35QJLOnTvXbtM2J06c6OzHrB++eJxyutS9ZsoRLvsNvXCW8hj382OsO9uyaZr2HC4/cT3l3PQ28jtei8uqvBZ/WTOz5vgmh5dffrmz35kzZ9ptrtWedUcZkNfhlbC5ZjJb0+UfzjOfm4QZbL5WrztsQtrpY19jl9A2nJtDL+JmtXG3J8MWXOKk9Pfkk0+2257hzIx3zjOfw8zOYzsoJUvdezslXR8frKZ/6NChznd9zwqeObzsp1QsDyGEEEJYM3mICiGEEEIYQR6iQgghhBBGMGlMFFM1GQ8jdWNJ+J3HIDBuianD1FelrrbuaZzUzBkL4fFXLCFAHdkrrxKm8no8E3VWbvt+jPNhKrfUjU+hjuxxM30a+qq4cuVKG4vi6bSMeeG1eVwRv2PshqeEM77GKyZzX8Y8eEwY2/jKK6/semypWxqDMRgeV8drZFyVxxgwTsFT4vvi27x69BRUVRtH4Wm+7EvGmHhMF2NiOIfdFoyT8Hgp7stKxT5HuB/HhMcyMK2asTMvvPBCb9t5bF9/hsYYxwRjZzx2bHnN64p149sEvP19sZ0eG8KYGo5Njwni/PNxw75niRhWJZe66e2cI4899lhnP7514L333mu3fQwxFnWofR9//HG77bF5bAfHvM9h3iPWxdbWVluixcc37ckYL9+PcUa0u99fGC/kdmL/sdSAl39hiaKhkj+8p3L99z7mdfEePRSn6vd8/q6vGr80rmTFVT1RVXWkqn5ZVWeq6nRVfWfx9wNV9fOqurD4/1eudqywv8SW8yL2nA+x5byIPW8c9iLnXZH0vaZpjkk6LunbVXVM0jOSXm2a5qikVxefw2YTW86L2HM+xJbzIva8Qbiq76ppmouSLi62/6+qzkr6qqS/lPTni92el/Svkv5+6FifffZZ6+Z3NzNd23SpuVRE1z3lEFYNl7oSi8tvTMFk+qpXVH3ggQfabbpt+eJKbzulGLqfpa5rk65Rl4o+/fTTdtvTzfk7Hs/TpndLn1+lLauqtaFX0aV8QVnRZTr2KaUzb3tfFXGp2990U7vrmDIMK+V6v9E17S8sJZSaKOu4REMXs6eb91Uid3e2p+MvWaU9pd/PGbcT3d+0jc9hyj4sT+Djg+nk/h0ry1NiYvqyt4P96JLNyZMn2+3nn3++3fbyI/fee++ux/ax2CfzSF3bsw9dWlnKE1YGYWW23N7ebiUuD1Hgmsm54+n/nra+2+95LdIfppWzdAHt4uNz/HDqAAAN8UlEQVSZazfX4yeeeKL33G+//Xa77fLMO++8026zZInPN5ZS8XWWcjTn8JAURFY9N5f4mONay/Hk6xDXQ95vvJwKw2J8baTkzXuPy9qcq5zP/rJjzgtKqz5mWRqBa8fQy7W93EZfJXJfw5Zr4LW8TeCaAsur6nZJX5P0K0kHFwNFkn4j6WDPz8IGElvOi9hzPsSW8yL2nDd7foiqqi9LOinpu03TdB6Hm51Hyl2rU1XV31XVu1X1rv+rM+wPq7DlfgQ/h93J3JwPq7DlXr0lYf1kbs6fPYWiV9WXtDMQftQ0zU8Xf75UVYeaprlYVYckXd7tt03TPCfpOUk6ePBgO2DcBcisgKEquLx5UwIaegmuu2r53W233dZu+wuImVEy9NJCuvzZXncf08XP7IHz58939uO5PIuRbln2014r6a7KlgcOHGiWrmCf5HQl043qrlNeC3/j1XY5Vlym42e6mN11zEwQuo5d6n399dd3PbaPV44V9r27gZlB4pl7HMvc9pugSy1kVfa85ZZbmqWtvP/7Mpx8HnC8U2ZwWYB959Ih5QVKZ94mShLsV0r1kvTGG2/segx/WS7XCLbB5Sva06UQztu+Ku/eXrIqW956661NnxzBccb2epXoO+64o91mxqrPP8plnuHGOUhp3P8BxvHw+OOPt9ueics5TDndbUQJifKPh1ewHT6+uH4OPZR6f5BV3jf75ChK3hxXPje59rLNbgvazOVZhrdQgvXwA5576O0EzAzkWPSM6b411LMyKQP6ddGGXJt8nox5KfhesvNK0g8lnW2a5gf46kVJTy+2n5b0s2s+e5iU2HJexJ7zIbacF7HnjcNePFF/JulvJX1YVct/5v2DpH+U9JOq+pakX0v6q/U0MayQ2HJexJ7zIbacF7HnDcJesvPekNTn4/rGapsT1klsOS9iz/kQW86L2PPGYfKK5cv4Dtd4qUH3VS+XujEU1E497umRRx5pt10f5e+os3vsAt9UzeqtH374Ye+5huJjWLqAab4ej0DNln0h9b8pnjFl0jRvF19q354GzjZSp2aMmdTtH8Yc8Y3vUjfGgTEukvTQQw+124yRY19L3bgcpkC/9NJLnf04Nhg38/DDD3f247kuXbrUbvt4JR53wfIPPK/Hjk1B0zRtH/m4HarqTxgLwTiJocrc7Ef/HcfRRx991NmPY46xEZ4+T9swLsJjsRjLyDZ4XCNtODTHuJ9f/3KurqtiedM0bbt9XeS1sf0eU0pbMq7I47lYpoT7SV27MIaG5Qn8M0tN+Fw6d+5cu8010/uRY5TX4fNqyH6MQ+Q1+hyegsLbBPxe1ldCxWOi+mBsmdQtK+GxjHfddVe7zTWZMXNS957K8ebxp/wdxwrPI3VjuHiP9orlHiNFeH9hDJ3f85f372uZm3l3XgghhBDCCPIQFUIIIYQwgknlPOn3bjKX8+hqpQvaU/fp1meKsbuZ6Yp0FyirFfO8dP1L/ZIE03D93Cxd4NVlKTHRFekSCV2P3k90M/MaXVJbt5z3+eeft+m1nuZLm1E2ufPOOzv70U3usgOhu9/7itVxvQwBeeutt9pt9un999/f2e/uu+9ut5li7cemnelGdkmDY8PlCR6T48GlJv+8DliBfqg0B6/Pxxx/x+2h8iOenk55j1WzXUbh3KQk4S59puBz7nif8hhDZTmIl57gGsG56fLoct0ak069F26++eZWbhlKz2ff+zrLkAKmpns6O6tJ+xzmmKaU7/OAsvapU6fabS9xwLHBNXgoNKTvRcpS9xr9rRGffPJJu81xvh9y3vb2drtW+njkNTCkw9d/9jnHpl8PbeZlYtgnlOJee+21zn4cL7SZj48+qdXtxDE8NGZ5LV5+hMfn9fu9a0gS7COeqBBCCCGEEeQhKoQQQghhBHmICiGEEEIYwaQxUVXVauquxVLPpX7pWjV1csYYuX7LVGePuzh8+HC7ffr06XbbYxzYJh7PywnwM1OlPVWfGi71YY8ZYdwJNW+H2rG3yY+5am666aY2VoCvHpC6ejTjPjyeie1nf3iaOmPLfDywr3he7zfGq3CsMPVa6sbG8Lo8FZYxBkM6Oq/LU4YZf0B7+Xj1GKV1sRyffW82l7r29HYxDoNxJB6f0Tc+pG7MI2MUPcaBMU1DZSU4xniModgWXr/HffXtJ/XHeLjdl9e8rpiopmnaMeQxH4xDpF28LZxz7CuPiWK8iscDMtWdr/hgnJokPfXUU+32o48+2m7fc889nf0uXLjQbnNe+dxkfzN2x8cQ560fg3ZnTK2n1U8Rr7i1tdXGuPnaQLuxLT42aUP2z9Dr13wN5Tx78803e/fre4XV0DrJeFkvL0C77fWdrR4LyH7j8fxcS7tfS5mZeKJCCCGEEEaQh6gQQgghhBFMKudtb2+3Ll93H9MlOOQ+psRCl667VbmfV2+lS5Bv9vbKq+fPn9/1Onw/npup70PnpczgqZp9FXelrtuZfej9NCZV81pomqZNLR5yCdMtOpSSSimVlcKl7nW6251uWvavu735mdtuS5cSl1AClrr9S1nL7UW7DFWPZp+5i9lTftdBVbX2cfmJbdvr9fgxCPvfJSCei1WRfXyzTyhROZSsuMa4dM9xynHpsh/389T6vpIofRLMusqQcG76uTkH+8af1O0frkcub1Iu9HHLMcC+8rWfx6dc5lXq+0rTuJzLOcg2+Tjhd75e8vhD66zLpetge3u7LQPh45ZjiPc8bxfnC+eBz2F+55IYZbD77rtv1zZI/dW+ve/6fuP3CdqTMrO/7YDyv89btp3j0iXeMXMynqgQQgghhBHkISqEEEIIYQSTVyxfutJcAqILj5LNUEYFM6mYCSJ1JTZ30dF1Tbenu7TZJroOXYLoO5e7+3ldPK9LQHRtejYa3a10M3vWwlAF8FXQNE3bzqF+Yzs846EvO8ld0exvSjx+DLpp3RXNNtLt7cej/MH2uhRAdzHb67IAbekZk2wHx7mP1ymy8ygBuTudUgCvx93z7H/2o2dl8hje/3wRM8/rc4T9z4xYl2eZicmx6BIp28Rt34/HGHpJKfvQ7bns53W+VWB5fg9zoF04d4ZkKa5jvqZxXvi4ofTC8e379WXm+hsfKPVxbvq6wjnIccPK6FK3/30sc91l33gmmh9zXSzb6mOOawPb6XI6+6TvHiINh1+wv3hel3i5XrP/fQ7zdxyXvnZzTvM+5zIu1w7PGO97g4Cfa3nN1zI344kKIYQQQhhBHqJCCCGEEEaQh6gQQgghhBHsW8Vyj+GhBs04iaHqpXzLu0ONe+it9MS19b7K6Z5mz/2oxbrOzvNSi/W4C2rH1Hml/oqtrgFfS8XVsSx15qHYEF6bt6kv/divkXq5x0lwrDDuyW3MeCSmCXtMA+MF2A6PMeB3Q7EVbFOf/i4Nx7d5mvq6WI47v1ZeE8eZ2539yjl3+fLlzn60jcfu9VUV9xiFvrgOjztjGxmD4SnQ7PO9xEz4dUhdO3EOexzRso3rmqOMb/M28jr7xrB/ZuyJr0fcz8cpx83Qetw3zzwOkdfCeeX927duexzuUJkEjmWey/vJ43zWjZ+f9yL28VD5DcZHDY1hPwb7hMfwGGHOTca4+X2ub575vOC8ZZs8jo/f+bk4xmgzH4vL609MVAghhBDCmslDVAghhBDCCGqdabZ/cLKq30r6naT/udq+E/An2v92TNWG25qmuWWVB1zY8te6sfrxakzRjpXbUsrc3Kc2rNOWmZtdvuj2zNyctg17tuWkD1GSVFXvNk3z9UlPuqHt2IQ2XC+bcA2b0IZNasdYNqX9m9COTWjD9bIJ17AJbdikdoxlU9q/Ce3YhDaQyHkhhBBCCCPIQ1QIIYQQwgj24yHquX04525sQjs2oQ3XyyZcwya0QdqcdoxlU9q/Ce3YhDZcL5twDZvQBmlz2jGWTWn/JrRjE9rQMnlMVAghhBDCHIicF0IIIYQwgkkfoqrqm1V1rqr+o6qemfC8/1xVl6vqFP52oKp+XlUXFv//ytAxVtCGI1X1y6o6U1Wnq+o7+9GOVRFbzseWUuw5J3vGlvOxpRR7bro9J3uIqqotSf8k6S8kHZP0N1V1bKLT/4ukb9rfnpH0atM0RyW9uvi8Tq5I+l7TNMckHZf07cX1T92O6ya2nI8tpdhTM7JnbDkfW0qxp74I9myaZpL/JD0k6RV8/r6k7094/tslncLnc5IOLbYPSTo3VVsW5/yZpCf2ux2x5Y1ty9hzXvaMLedjy9jzi2HPKeW8r0r6FJ//a/G3/eJg0zQXF9u/kXRwqhNX1e2SvibpV/vZjusgtlwwA1tKsWfLDOwZWy6YgS2l2LNlU+2ZwHJJzc7j7CRpilX1ZUknJX23aZr/3a92zJXYcl7EnvMhtpwXsecOUz5E/bekI/h8ePG3/eJSVR2SpMX/L6/7hFX1Je0MhB81TfPT/WrHCogt52NLKfackz1jy/nYUoo9N96eUz5E/Zuko1X1p1X1R5L+WtKLE57feVHS04vtp7Wjta6NqipJP5R0tmmaH+xXO1ZEbDkfW0qx55zsGVvOx5ZS7Ln59pw4KOyEpPOS/lPSsxOe98eSLkr6TDua8rck/bF2ovovSPqFpANrbsMj2nE5fiDp3xf/nZi6HbFlbBl7ztueseV8bBl7br49U7E8hBBCCGEECSwPIYQQQhhBHqJCCCGEEEaQh6gQQgghhBHkISqEEEIIYQR5iAohhBBCGEEeokIIIYQQRpCHqBBCCCGEEeQhKoQQQghhBP8PEFj31Af8g0YAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x720 with 25 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising layer 1\n",
    "learned_weights = dbn_mnist.rbm_layers[0].W.transpose(0,1).numpy()\n",
    "plt.show()\n",
    "fig = plt.figure(3, figsize=(10,10))\n",
    "for i in range(25): \n",
    "    sub = fig.add_subplot(5, 5, i+1)\n",
    "    sub.imshow(learned_weights[i,:].reshape((28,28)), cmap=plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlEAAAJCCAYAAADgPpLrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzsnXeYFFX2hr/rDDkNSURFMSsqa8CAEQMqZjERzCgmXNOKmF0VlTXrqogBMGLCHNBFDKyiYkBBFDABLqACAgoS6/cHw/zqvNXTM9PTM9O0530eH/noma7qOvdWFXW//k6IokiO4ziO4zhOxVijpnfAcRzHcRxndcRvohzHcRzHcTLAb6Icx3Ecx3EywG+iHMdxHMdxMsBvohzHcRzHcTLAb6Icx3Ecx3EywG+iHMdxHMdxMsBvohzHcRzHcTKgUjdRIYQDQgjfhBCmhBD6ZWunnJrB65k/eC3zC69n/uC1zC9CponlIYQCSZMkdZY0XdLHkrpHUfRVab9Tv379qEmTJiV62bJl5vXly5cbvXDhQqNr1apldMOGDRPbqF27ttFLly41ukGDBkb/8ccfRjdq1MjoP//80+gVK1YYPX36dKNbtmxpdJ06dYwuKioyevbs2SIFBQVG87jwdR7HJUuWGD1nzpxfoyiyOwYqWs+6detG8eO/xhr2fnzRokVG169f32geRx634v02mrXkezRv3tzoxYsXG81ahhCM5lzg+ONnoP75559FWCvWf/78+UZzfNarV8/o77//Puu1LN5u1KxZs/h7mNd57DmuuZ/8XFJynJb1nvFzhSRNmzYt7TY599daay2jOZ74+19//bXR6623nsjvv/9uNOtZWFhodLq5O3PmTM2bN88e6BRUtJ6NGjWKWrRoUaI5D1hbjjnOE/6+lJx7rBXPQawN5xrHBn+e+8DzS926ddNuv2nTpiIzZsxIu01qfmZeKyZPnlwlc7NevXrmusn94rzhvOK5mZ9bSo5jnrfmzp1rNM+1HDPUfH9euxcsWGA061fWNS/Ve/Jz83zO+sXPWfPmzdPChQvLnJuSVFj2j5TKjpKmRFH0nSSFEIZJOkxSqYOhSZMmOumkk0o0C8Mbii+++MLoNddc0+jdd989sY3111/f6JkzZxq93XbbGf3pp58avdtuuxk9ZcoUo3nTdeGFFxp97LHHpt2frl27Gj148GARnpC4zcaNGxv9yy+/GM2LzaOPPvpjYiNJKlTPhg0b6uCDDy7RvKGYMGGC0TzunDRnnHFGYhtPPvmk0Zz8nBTxsSVJ3377bdp94kWUJ5+PP/7YaH4G6nvuuUeEF6jDDjvM6JEjRxq94447Gt2+fXuju3fvnvVaSlKzZs10wQUXlGj+g+V///uf0RtttFHa/RwxYkRiG7zJnDVrltEbbrih0QceeKDR8f2TpK233troddZZx+i+ffsaPWzYMKO32moro3fZZRejL7nkEpH333/f6IMOOshonqN4Mxk/0Z911lmJ9y+FCtWzRYsWuuaaa0r05MmTzeu8CO+www5G82byxx+TQ47npAMOOMBonoPatm1rNG+CeIPLn+dcHjdunNFbbrml0T/88IPRPO9KUv/+/dNus02bNkbzM++zzz5Gd+7cuUrmZpMmTXTccceVup/UnGe8wbz22msT2zjiiCOM5g3Gs88+a/QJJ5xgNMfYN998Y/Qhhxxi9K677mr0u+++a/TUqVPT7g/HS6r35OfmmOnUqZPR8XNWqutyaVRmOW8dSfGZMr347wwhhN4hhLEhhLG86Dk5RZn1jNeS/9JwcooKz01eIJycokJzk/9AcXIKv27mGVVuLI+iaFAURR2iKOrApxXO6kW8lrzLd1Y/4vXkEzNn9SJeS/6r3Vn98Ovm6kNllvN+khR/3rlu8d+VvrHCQuN94b+YjjnmGKNvvPFGo/v06WN0Kg8KH9F/9tlnRn///fdGz5s3z2g+rueSI5/AcDnv888/N3rbbbc1+s477zSajy2l5GNJnhTLWv9P5S8qBxWqZ0FBgeIeGi7/8FE4l3+4z7fffntiGxtvvLHRPXr0MHrnnXc2+qqrrjKaSytczuMyRnx5Ukp6IDiW+Jkvv/xykZtuusnoUaNGGc3PyMfiv/32W+I9y0GF52YIwXwejvMtttjCaHpUNt10U6O5TCklfTBcjqengXORvpb999/faJ5PzjzzTKO55MTzy4ABA4zmspeUXDrhEzx6N7g0Fl9C5pJWGipUzxUrVpil7u233968/uuvvxq9wQYbGM1zZKtWrRLb4NI6/SedO3c2mkupXO456qijjOZyIOcBz7v/+te/jH7ooYeMvu+++0T4Dwees7iczzH+0ksvJd6zHFR4bi5atEhffvlliaZFhEud7733ntH77bef0V26dElsg8vOvI7xXErLCZfiea4eP3680bTRfPDBB0aXtcR8+umni3AJlzeftGvwOMWvZ/Q2pqMyT6I+lrRJCGGDEEJtSd0kvViJ93NqFq9n/uC1zC+8nvmD1zLPyPhJVBRFy0IIfSSNkFQg6aEoiiaU8WtOjuL1zB+8lvmF1zN/8FrmH5VZzlMURa9KejVL++LUMF7P/MFrmV94PfMHr2V+UambqIry22+/mXVkrrU/99xzRvNrjMx94bq5JI0ePdroN99802j6jR588EGjb775ZqP5NXt+rZ3eAPo8uO7K9Wx+zV5KenHo9eC3NeLr5VLqr/Nmm1q1aql169Ylmv405nhwzZ1fmd9pp50S26Cvgdugpj+F/iL60+gFueyyy4zmV4HpwWK0BD1TUtI7dOihhxodz/ORkmv/9NBVFbVq1TJeOh67iRMnGs0xyRwY+pUk6T//+Y/R9Ch89ZX9ljc9DfQ4vfiiXQWhR41ZNhwfjJtgRAK1JK277rpGv/baa0bTu8NMpriXhP6pbBFCMO/91ltvmde53VSZXnFSfduP45Zev8cff9xo+o3i/hMpeZ6k/43XCkbPEEam0CsrJTO86DW9+OKLjWZGWKr5XhXUqVPHeEq538yJoneLHrhU2ZD0zfE6xjHEucm5RT8Zr2m9evVK+/7t2rUzmtf6V19N3oPyPbhN+nK7detmdPxcW9aciONtXxzHcRzHcTLAb6Icx3Ecx3EywG+iHMdxHMdxMqBaPVFNmzY1fp2hQ4ea1+lZofeHa72pUrOZZ0NfBNex6XF65ZVXjGbm0ttvv210hw4djKbvg2vx9OEwTl9K5pcwkp/eMWZ2DBo0KPGe2SaEYLJhWBt6GngcmL9F75kk0+pAkt555x2jWZvNNtvMaNaCtX799deN5po5c4rYboftMLbZZhsR7hM9M/Sb0Et2yimnGH3LLbcktpENFi1aZDxJzMyiR4r7Sb8IM7lSQT8hPZCc32wlw6yg559/Pu32OF54fmHWEee6lPQCsXUNs67YSibu46rKlPi4F4ves7iXUUp6KpnXN2nSpLTvLyU9jqwljy09dPRI0UPFdly9e/c2+pxzzjH6k08+MZptRyTpu+++M5o+GH4Gao756oJ+sFNPPdVoHmt6qFJlz3EMsOXS2muvbTTrx3Pf0UcfbTTbobEe1Myp4rk5VUbkkUceafTw4cON5rk03TbGjh2beP/S8CdRjuM4juM4GeA3UY7jOI7jOBngN1GO4ziO4zgZUK2eqNq1a5ucpE022cS8zrVbwrXcv/3tb4mfYZ4NPSmzZ882mh4HZs1wLZgeCK4/l9VjiD4beomkZF4Js6+4lsv1Z/oy2KcoG9StW9dkedBLRu8ZjxM9FfSyScl6s9fYeuutZzTX9ZkrxnV0emLow+jUqZPRTzzxhNHs78SxISX9bewZyAwX9pC69NJLE+9ZFYQQTN4MfXcdO3Y0mvlI7BuYyhNFDxA9ZsyOeeCBB4xmLzt6Q1g/1pdzn9lGzBZKlV1GX8zhhx9uNL0hfM+494P5Ptli0aJF5vjzPHvPPfcYve+++xrNvnapjsO9995rNMc1z6M77rij0Tyv0nPH8wfHI31dgwcPTvv+zJyTpPbt2xvNfmnMQuJ4o8/nuuuuS2wjGyxevNj0hWOGE8+lnDf//ve/jeY5RpJ69uxp9MyZM42mZ+2ZZ54xmt5O+nj5fmeddZbRnIv01NJzx/1JtY0bbrjBaPbW45i98sorS/7M3LJ0+JMox3Ecx3GcDPCbKMdxHMdxnAzwmyjHcRzHcZwMqFZP1IIFCxL9beJwTZp977beemuj6XGQkvkV9CAxG6RRo0ZGMx+D/by4j9OnTzeauVEvv/yy0cy+6d+/vwjX7+kFYl+qO+64w2hmbtAvkA0WLFigkSNHlmj6VZgLRT8S17jpYZCSPZ7okXrssceM5jo5/Ub0frA3Gr0Fb7zxhtH0WNFPlyp/hX60E0880WiOYXroeJyqiqVLl5qxzByotm3bGn377bcbfd555xmdqvfUlltuaTS9N+yPVVYmE314e+21l9H0RNGDRa8Pe22l8o5wHHPc02dB32b8M/Jns0XLli2NF4u1Yg9CZumxTx37i0pSv379jGZu06233mr03//+d6PpHePcY+81enbY640+UJ4jOT4l6b777jOaY5Y9BumJ2nDDDRPvWRXUr1/f+Lf4WXn+Z24U/T3MDZOk7t27G/30008n9iEO/Yf0G3Gud+7c2Wj2BOW5nn5BeqTYY1RKXpv5uXkd/emnn4z+8MMPS/5ckQw3fxLlOI7jOI6TAX4T5TiO4ziOkwF+E+U4juM4jpMBfhPlOI7jOI6TAdVqLC8oKEgYv+OwySCNvwxkS/VeNOoyBGzjjTc2Oh4YKUnjxo0zmmZiGtG5D2xaettttxl90EEHGc3QOilpVqUhNt7EVEo2VqXZvSoIIZj9pNGQgaBsYnrAAQcYzWbBUjKINB7UKkl77LGH0QxWHTZsmNFxI7yUbHhNMzwbjNJ4TON6qrBNjuF3333X6Hnz5hlNc2SqkMCqgOGpNOmzPjRd8wsVNL9KZTclZiArAxtp9KW5mMeORlI2/+ZnpEGXIZVSstEtm2KPHz/eaIatxs8/DJDMFnPmzDFfulh33XXN64888ojRrB11KoM9zer8Mgs/2yWXXGI0v7DD4NzTTjvN6LvuustomoJp+Gd478CBA0V69Ohh9NSpU41mE9p69eoZnerLMFVBvXr1zLmJ5xmGhtLkzesqr3lS8rrIL7RwLvJLITTlf/vtt0azfjSe89hyewz7Xbx4sQjPB3wPfqmL54MXXnih5M+pviRUGv4kynEcx3EcJwP8JspxHMdxHCcD/CbKcRzHcRwnA6rVE7V48WITdknPCdd6uW7KtdztttsusY25c+cazVCvXXbZxWh6GhiOx5Av+iwYEsa1YIbG0dfD9Wwp6Scoy6PAEMmq8lrEWbJkSdrgQHoS+LlZ61RNTocPH250ly5djOY6/C233GI0A/a4jwwRpL+trDX1jz76yOjNN99cZOeddzaaoXVPPvmk0WzcyYDZL774IrGNbFC7dm3j32KDYfryeKwOPfRQo2+66abENugP23XXXdO+Bz2QDz/8sNH02e23335Gs15du3Y1mo1ZjznmGKNT+S54XBgQSm8JfXTx8wk9XtmiUaNG2nPPPUs0m/nS39atWzej6U8699xzE9tgKCw9cAw6pf+EIbRsHstanXzyyUbz2vDoo48a/d///tdonpdTbYPhjKwlPXsMFK0qFi1aZDxkv/zyi3mdflReo3ie5OtSsp70tNIvRh8uvcec2/SP0QPF6zLPzQw2Pfroo0U4n+jrYvNw3ivEr1H0JafDn0Q5juM4juNkgN9EOY7jOI7jZIDfRDmO4ziO42RAtXqiGjdubBoR0m900UUXGU1/yJAhQ4xmNoiU9ElxPZi/w2yg5cuXG03/AH02fJ05MvRk0deTqiEzPU4TJ040mk1s6RXZYYcdjKbvKxssX77cZGnQb8T1aa4/M+Np6NChiW0wW4Y+C2aC9enTx2g2d6aPgh6aUaNGGc11embb0Ofx6quvipx11llGX3fddUbTj0LvAL0IVcXvv/+uMWPGlGjWh96f3Xbbzeh4804pWTspmZnGJp/0LOyzzz5GM2ONvgrObeZGXXPNNUbTY8ljzQboUjJbjP6U/fff32j6tuKe0FSeq2wwbdo0XXjhhSWa/jQeB+4jPVCpfHjMN2NuEH06bN7Lpuj/+Mc/jD799NONpr+N2z/qqKOMZp3oX5SSDYjZ1Ja1ZM4ZGzdXFfXq1TPeWWZk0ePEZtD0n6W6brIBNM+NzE+k149zib5demR5feD7M0eQc4UN0qXkdZLXfs5nXg/ivryKzE1/EuU4juM4jpMBfhPlOI7jOI6TAX4T5TiO4ziOkwHV6omaNm2a/v73v5doen+4Ds61eObM0IcjJddFmS9Bzd54Tz31lNF777230fTFMP+I/ZYuuOACo5nhdOSRR4o8++yzRtP7wbV8Zmq8//77ifesapgLQr8S/W/cx1QZS6NHjzaaniWuaTMbhplf7HvIfaAfiTkjM2bMMJpjjTlVkjRo0CCj6dtib7WlS5caXbdu3cR7VgXNmjUz/ix69ehXoncw7vWRkn0PpWSvMnpM2Kvu2muvNZpeQHp5mGXFDDeOMWZ0xc9NkvTjjz+K0MdFHyf70tErxNywqmCNNdYweWTMw9prr72MPuGEE4xmRhB7DErJ+tOD9MwzzxhNTyTPsw888IDR22yzjdH00DCn7KWXXjKa3lSe56WkV6hFixZG81xOvy3PR1XFsmXLjI+JXkxmGvG6yblJH7CUrDFzn3ieoh+MfUSZh8drNTO36IniNY/+6E8++USEv8NzKXP/mFMZz+RL5YcsDX8S5TiO4ziOkwFl3kSFEB4KIfwcQhgf+7tmIYQ3QwiTi//fNN17OLmD1zN/8FrmF17P/MFr+dehPE+ihkg6AH/XT9LIKIo2kTSyWDurB0Pk9cwXhshrmU8MkdczXxgir+VfgjI9UVEUvRtCaIu/PkxSp+I/D5X0tqSLy3qv5s2bq2fPniWa6+T0IHBNk33oUvXO69Spk9HMH6Lvgtkw7JfETI63337b6O7duxvN3J/vv//eaK7ljxw5UqRjx45Gt27d2mh6QV555RWjN9tss8R7riJb9WzcuLHJEimrTxE/w5prrmk0+wNKySwZ5gT98MMPRjNbhLlS7777rtG777670fSiMdOL+Sv0Htx9990i9G3RR7X99tsbTS8CfV1xsjk358+fr9dff71E0/tDzTHI3JaHHnoosQ3WnJr5NfQvMouK2TM8XzAbiOOF3hKefw477DAR+jJvv/32tPtMHc/i4jHNVj2bNGli5gLnJjOB2CuPXjXmJ0nSY489ZjT9gvxs9NTQw8TebfTxsHb0ebHv3eOPP240vWlSsg/iHXfcYTTHE3089MPGyebcnDdvnjnHn3LKKeZ1+s/oNY6POSl5LKVkJhb9oyNGjDCanjjOC/qT6JHi79NjxTH6xhtvGJ3KE8VeiPT+0afLfopxj1Sq/oKlkaknqlUURatmzUxJrTJ8Hyc38HrmD17L/MLrmT94LfOQShvLo5VfgYlKez2E0DuEMDaEMLYid3dOzZCunvFa8psNTu5RkbnJJwdO7lHeuem1zH0qMjdTfQvdyR0yvYmaFUJoLUnF/0/myBcTRdGgKIo6RFHUgY/4nJyhXPWM17Jhw4bVuoNOuclobnK5xckZKjw3vZY5S0ZzkxEGTm6RaU7Ui5JOlHRj8f9fKNfGCgtNHsVOO+1kXr/xxhuN7tWrl9HMeeG6vJT0AzGXhdlAO+64o9HM5eG66osvvmg0PQ/03bDnU7p+PatgP664V0WS9txzz7TvkWrNuwwqXM9Fixbpyy+/LNFc8z7ppJOMnjx5stH0n9ALJElt27Y1mhld9Fm9/PLLRvNicuCBBxp9//33G82eYvTT1alTx+iWLVsazZ6FUrJv2T333GM0PXrsOcW+iOUgo7kZQjBjn7kxzIHhmOTcZL6alPQT0sNGzyN9GvRl0G9IP+Pxxx9vND0wzLLik/JUPSfpV+Q26dPiuI5nMNEXUgoVrufvv/9uMtOYf0QvGp9ctWplV5mYtyVJAwcONJp9I+OZO1LSl8Pjxl5t9ERxrp199tlGMxeKczmVt/DQQw81mseBczPeK1RK5kyVg4zmZuPGjc1coH+M1zj6FW+++WajWQspOZ832mgjozmXuA367vj0jOcTZrSdeuqpRj/66KNG0+dF/5OUzDdbd9110+4zPW5xjzV74qajPBEHT0j6QNJmIYTpIYReWjkIOocQJkvat1g7qwFez/zBa5lfeD3zB6/lX4fyfDuveykv7VPK3zs5jNczf/Ba5hdez/zBa/nXwRPLHcdxHMdxMqBae+cVFBSY9dy+ffua15kDxKyhFStWGE1PjJTM9uA69m677WY08yiYG8UsK669c52VOVD0YDHHKlVfKn5OfqZ99rH/mGGfM3qJqoKioiLjKWAOC/vIsd8f/TD0nklJzxv9Z8wWoaeFuS/U3Cb9bJ999pnRrO306dONpv9Nkm677Taj6fuhf421Z1YR+3llixUrVpicFPbXmj9/vtHMkWHfy+HDhye2wb6BzIWiJ4XHhj4N+gvpUSmrdxb9j/zGKfs9SkmvB7Om6PXgOI97qpgzli3WWWcdXX/99SWa/iB+KYQ+LubOxbP9VkF/G3029I7RU8OcJ3qg6Nt64QVrHzrmmGOM5nn64ott/NK9994rQr9i+/btjb7sssuMpicyXU5UNlmwYIE53jxHcN7wuslxzfw0SercubPR9Hcyi4rzYJdddjF6ypQpRjNrjP5U9qpjn02OB/obpeT9AMcMDfrcZx7X8uJPohzHcRzHcTLAb6Icx3Ecx3EywG+iHMdxHMdxMiCkygCpKoqKiqJ4vzKuu9KvxHVzZoNw7VdKrvfSt8K1VvZ8oq+Cniluk7lU7MfG/m9ch2W+iiSdccYZRrNPEH1Y9IpwH++9995PoihKNqerBE2aNIl23XXXEn3aaaeZ1+mh4edkLghzpKSkP4290+hRoKeO26QHht4QZpEwZ6p///5Gc4091Zr6wQcfbPTaa69tNHsz0kvA/KUrrrgi67WUpObNm0fx40cPwtFHH200/WPMNps2bVpiG5xL9JSU5QcbPXq00fTl8FjTo8bt0UPHWqT6DMxQYoAw+3HNmzfP6Pj5dsSIEZozZ449KFlg7bXXjuLzkXlUHNc8LgUFBUY///zziW3QE0d/GmtDfxl9WPTEsG8lM+KYAcTPxNwinuelZJbVrFmzjKZPi70xOR579uxZJXOzbt26UdznymxDegsvv/xyo9nHkplc0sosqjj0j/K6xZw/egN53mJvQ/q0OB44r3ju5vlGSo5JwnnAvrVx316/fv307bfflmtu+pMox3Ecx3GcDPCbKMdxHMdxnAzwmyjHcRzHcZwMqNacqDXXXNPkLDG7g+ve3bvb0Ff6C1L1nWMPN3oYmC9BfxHzcRYuXGg0PU9c+6WPh1lE7PeVKmuEfYyYgcLPTT/R008/bXSqjJTKssEGG5j+Rv/617/M6zwOrAu9avQCSUkfA9e86Q3gcWL/Nvoo6B2gB4d97ugtuPXWW41O5S9kNhF9FPRd0KdRWFg9U7R+/fomJ4d+IvYRJPQ3ch5JSd8L+6fRT/Tcc88ZzV5XAwYMSLuPHB881uydx+ybI488UmTChAlGM1uI3hB6geLHgJ6+bPHHH38Y/+B5551nXmc/RmYA0Rt41FFHJbbB8+i2225rNM/VzOvj+aFbt25Gs1cj8/Q4t+mR4bWEuVJS0jtKjxzPs1dccYXRzK2rKho1amT8wvSfPvvss0aX5T/iNU+Sli1bZjSzCOlnPOSQQ4zmNYteYfoLmUs1e/Zso+nLo07lJT755JONpleYPiueT+J9S3nM0uFPohzHcRzHcTLAb6Icx3Ecx3EywG+iHMdxHMdxMqBaPVHff/+9TjzxxBJNrw49AszlmDhxotHx3m2rYD+122+/3WhmUTEPh9k/XBenr2PBggVG03dD7w+9Ii1bthRhLzxugz4rridfeeWViffMNlOmTDH9j/72t7+Z19kzrKzcKHoqJJk+i1Ky79ygQYOM3nPPPY1mtsiQIUOMprfg888/N5qeLGbftGnTxuhU/iV6Q+jdoGeK/Z/4GaqKhQsXmoy0//73v+b1k046yWhmbNGHwT6DUrKe48aNM/qss84y+h//+IfR+++/v9HsS8i5Rw8dPRGPP/640YcffrjRTz31lMj48eONZm/MU045xWh6c+J5RvRkZIv58+frP//5T4nmOYfjlBlgrPVVV12V2Ab7Ei5evNhoeknpiaGXrF27dkbH/SlS0m/48ssvG03PDs+Z7C+XCuYIcfzRh8V+oFVF/fr1tf3225doZiHyWNPPw2saj7WUvOZQ33nnnUbTR0c/Es/N9BLTt8tem6wvPVSpcr/4O2uttZbRvPYz6yqexVURv6I/iXIcx3Ecx8kAv4lyHMdxHMfJAL+JchzHcRzHyYBq9UQ1adLEeGWYTTFs2DCjmcPBdXSuo0rJtdcnn3zSaPYIoheHXg9mydDr8fXXXxvNHCB6nl588UWjU2WN0F/A3JZHHnnEaGZ6MFOlKmjevLlOOOGEEh33YEhJ/xG9PezfdMsttyS2wewprv3Tz8ba0J/yww8/pN0n9s5iPg77O9EvN2bMGJF4PyYp2QeRr3OtnxkwVUWTJk1M7zn6vegpoZeLuTBxf8EqODcfeOABozmGOnXqZDTnEr0ePD8we6wsjwR9N/TZSNImm2xiNHu80UvCfYz7tKrKE9W6dWvj92MteVx53Pv27Ws0+9BJyXMMeymee+65RvM8Sa8o+/PxnMd50aGDbVHHutDDk2pu0mPHnCiOB3oomWtWVRQUFJhzG708O++8s9Ecg7zGrbPOOolt0Esc97tKyT62nO+PPfaY0V27djV64MCBRvNcygw+ZrYxh4rXZSmZ0cbPzfsNHsd4D0J6/tLhT6Icx3Ecx3EywG+iHMdxHMdxMsBvohzHcRzHcTKgWj1RK1asMD256JtZe+21jY57NKRk1lAq2HvsggsuMPqggw4ymr4qrqPS88Cf59opvSJbbrml0fQGsLdWqr/jejV7vL3zzjtGp/KMcoKCAAAgAElEQVSjZJulS5eazKPjjjvOvM71Z/qN6CdgHzQp6Zugn4x5WcwFYq7UHnvsYTRzgZivxfwt9vNinegDkKSffvqpQvtIH8axxx5rND132SKKIuNDoP+I/iLOM35O9tKTpH79+hkd76MpJX1XzFiib475R/GcKynpeRk8eLDR9ETRL0nfh5TMolpvvfXSboPnn3jvTPYDqyro9WGt6CVhn8t4tt8qevfubTR7iDKfr6w+Zsz8YyYbvWUPPvig0cyyokdqp512EmGGG7fB/oAcDytWrDCaGWLZYsmSJeZcyAyjsq5B9BftsssuiW3Qs0S/J89brCf7t9KXxWt73H8kJccPfcHMiKS/UUpmizGTj7/DjLdLLrmk5M/MwUqHP4lyHMdxHMfJAL+JchzHcRzHyQC/iXIcx3Ecx8kAv4lyHMdxHMfJgGo1lteuXdsYM9nwlSZNGr8Yrhg3qa+CZlQac/keLVq0MJqGNzZJZnAdDXY0ks+dO9domvpSmcBpaqNpr2PHjkYz+K48zTYry+LFi/XNN9+UaJr42AyY4XijRo0ymoZ7Kfk51l9/faPjTTmlZHBlPAxUStYibvKVksZSml/ZZJmGXQbCSUlDJc3trD+/JPDbb78l3rMqCCEYw+quu+5qXudc45c8OO4nTJiQ2AaDZePjR0p+VjZS3WabbRL7HIdjiOODxn9+cYBNsRkIKCUbMzOwj+cHhjzGxxi/fJEtCgoKjBGYwYQjR440mrWmgZpfbJGSpn3OLdafwaIMvmVTXYY38osNHCv8wgWNzTw/SclzEq8NNP6zQTYN3S+99FJiG9lg/vz5GjFiRIk+5phjzOsMiGYoMY9VqsbavNYywHnq1KlGs+E4g3Rpyud5jiZvzgWOD34xgONFSjbS5hcmGBjK+g8dOrTkzzw3pMOfRDmO4ziO42SA30Q5juM4juNkgN9EOY7jOI7jZEC1eqKWLVtm1qbLahDIdXaug6YKDWPgGX0U9MFwbZahXgzXa9CggdFcq2/VqpXR9H0xoC0VbCjJ96Tv64gjjjC6OkL8GjdubJpJMwiTPi56y+gvir/XKugPYkggfRODBg0ymsGqfJ0eKjYkXr58udH8TPTPMbxTsuvsUrJ2DHjs06eP0WwOW1X8+eefZt8YjsrwQoYh0h/CeSIlxyWDRadNm2Y0fRn0SM2ZM8doBr5++OGHRtMXQ38aw1tTNQimN4yBsAym5LyIe0voqcsWy5YtMz4YNhDu3r270S+//LLRDFK9/vrrE9vo1auX0ZdeeqnRTZo0MZrBlqeeeqrRHC/XXHON0Rw7fH9+xosuushoen6kZC3pReW1gZ6nivhmKkOTJk1M6DOPJb2j9ALSn8YwTinpceL5mOOcnrhU+xyH4Znbbbed0fRMsUky60vvoZT0K9IbRji/45+ZvsG071Pun3Qcx3Ecx3FKKPMmKoTQJoQwKoTwVQhhQgjh3OK/bxZCeDOEMLn4/8n+JU5O4bXML7ye+YPXMr/wev51KM+TqGWSLoyiqJ2knSWdHUJoJ6mfpJFRFG0iaWSxdnIbr2V+4fXMH7yW+YXX8y9CmZ6oKIpmSJpR/OcFIYSJktaRdJikTsU/NlTS25IuTvdeBQUFKioqKtHjxo0zr7MJIf0i9KSkyoliVgTXYseOHWv0mWeeaTQzUrgPzJViM0iumzM/g54Y5mFIUmGhLQtzQNj8k/lEpTWpzWYtly1bZjwphxxyiHmdxyFedynpN+B6tpT0JLGR5sknn2w0fTxspNq/f3+jeVw5NpiNwoa19OQ8++yzItwHHid66rjNBQsWJN5zFdmsZ+3atY0vgQ1hmbHC3BaO2VR+ItbzjTfeMJoNxzmu6RXksX3llVcS24xDD9zTTz9tNJuNp8r94Zhp3bq10fyM9DPGPZbxeZrNWi5dutT4B+nDpJenR48eRtOPdPHFyc3Rj8hxzfrTv0oPFMcbM9uY4UU/LWvJz7zxxhurLOifvfzyy41mc+kdd9yx1PfKZj0XLVqkL7/8skQzF5DzgF5QzptUOVHMAuPY33fffY2mR4oNiqnpeRo2bJjRe++9t9E87zGHjs3gpeS1mWOIc5Ee2Ey9xBXyRIUQ2kraVtKHkloVDxRJmimpVSm/5uQgXsv8wuuZP3gt8wuvZ35T7puoEEJDSc9KOi+KIhNXHK38mk1Uyu/1DiGMDSGMrapvozgVIxu1rI5UdKd8ZKOe6Z54OdVHNmqZ6gm9UzN4PfOfct1EhRBqaeVAeCyKouHFfz0rhNC6+PXWkn5O9btRFA2KoqhDFEUd2EbDqX6yVctUX2F3qp9s1ZOP353qJ1u15DKFUzN4Pf8alOmJCisbVD0oaWIURbfGXnpR0omSbiz+/wtlvdcaa6xh1m/ZS4tru/QbPPfcc0Yz90eSXnjB7sbdd99t9HfffWc0B+hWW21l9GeffWb0m2++aTR9OcwjoY+GeTrMqpGSvhh6wZiXxHwd+o9Wkc1a1qlTx/Qmog+LeUf0Kw0fPtxorvNL0tlnn230gw8+aDR7451//vlGM++GuWPMrmrXrp3RZY2/AQMGGP3pp5+K0AP3ySefGM11+oEDBxqdKgttFdms54IFC0w2CjNWOG7Z947zJFXuFz0HLVu2NJoZbqwXe+PR+8F605/EXC++HvedSKn7WjK/hrl0zDZjvk18H+O9/7JZyxUrVpgMKp5neRzZi43zgJ9BSvY5pXf0iy++MJq5UsyJYqYXM9eYbcW+ifQz0etKf5OUHH88B3GMM/uIYx77l7V6FhUVmSxA5inys9GvyJ/ndVZKjms+mWafOV7n6LMrKyeK44djkjlUzHij30mS7rvvPqNTzd849EDyfFNeyhO2uauk4yV9GUJYlRx5qVYOgqdCCL0k/SjpmFJ+38kdvJb5hdczf/Ba5hdez78I5fl23mhJoZSX9ynl750cxGuZX3g98wevZX7h9fzr4InljuM4juM4GVCtvfMWLlxoeslx3Zv5FTS7DhkyxGiu3UvJvKGrr7467Xsyq4rZH1wnjfuAJGnSpElGH3bYYUZfd911iX2Mw5wpKemJ4no015+ZbUUD/5gxY9LuQyb89ttvxn9G39YWW2xhNHOhmja1Qb30J0nJ7Bl6ah566CGjmW2y5pprGk1vQMeOHY1mTtkNN9xg9H777Wc083Xo0ZKSXg2u7dN7wKypjz/+2OhUxykbMMONPSTpo2jfvr3R9B+w156U9DFwHDPbh94Ozou33nrLaH77l9ujf4l+NI7JVD499i2j5uemx3L99dcv+XNVfTljyZIlxid51113mdfp5WPt6C1kJpOU9MzRE8XeaJz/PG8yZ4zvx958nNv04LDvYip/DPsasnZ9+/Y1+pxzzjE6VVZRVRGfGxyn7J06atQoo+kveu211xLvf9ZZZxnN/nucq/Q40W/IPDXm4dHTRg8t+xjSL5nq/ML8xPvvv9/oM844w2jmAsa9fxX5trI/iXIcx3Ecx8kAv4lyHMdxHMfJAL+JchzHcRzHyYCwMjS1mjYWwi9a+bXOFpKqb0E5M3J9Hyuyf+tHUdSy7B8rP17LrFPefcx6LSWvZ5bxWpaffNrHv3o9c33/pCqoZbXeRJVsNISxURR1KPsna45c38dc2b9c2Y90+D6ufvuRjlzfx1zZv1zZj3T4Pq5++1Eaub5/UtXsoy/nOY7jOI7jZIDfRDmO4ziO42RATd1EJZve5R65vo+5sn+5sh/p8H0sP7myH+nI9X3Mlf3Llf1Ih+9j+cmV/SiNXN8/qQr2sUY8UY7jOI7jOKs7vpznOI7jOI6TAdV6ExVCOCCE8E0IYUoIoV91brs0QggPhRB+DiGMj/1dsxDCmyGEycX/b5ruPaphH9uEEEaFEL4KIUwIIZybC/vp9cxo/7yW5STXa1m8P17PcpLr9fRalp9cr2Xx/lRLPavtJiqEUCDpbkldJLWT1D2EkGx+V/0MkXQA/q6fpJFRFG0iaWSxrkmWSbowiqJ2knaWdHbxsaux/fR6ZozXsvwMUW7XUvJ6VoQhyu16ei3LzxDldi2l6qpnFEXV8p+kjpJGxPQlki6pru2XsW9tJY2P6W8ktS7+c2tJ39T0PmJ/X5DUuSb30+vptfRaej3zuZ5ey/ypZVXWszqX89aRNC2mpxf/XS7SKoqiGcV/nimpVbofrk5CCG0lbSvpQ9Xsfno9K4nXMiNyspaS1zNDcrKeXsuMyMlaSlVbTzeWl0G08nY1J77CGEJoKOlZSedFUTQ//lou7WcukyvHyWtZeXLpOHk9K0+uHCevZeXJpeNU1fWszpuonyS1iel1i/8uF5kVQmgtScX//7mG90chhFpaORAei6JoePFf1+R+ej0zxGtZKXKqlsX74fXMnJyqp9eyUuRULYv3o8rrWZ03UR9L2iSEsEEIobakbpJerMbtV4QXJZ1Y/OcTtXIttcYIIQRJD0qaGEXRrbGXanI/vZ4Z4LWsNDlTS8nrmQVypp5ey0qTM7WUqrGe1WzsOlDSJEnfSrqspo1mxfv0hKQZkpZq5XpzL0nNtdK1P1nSfyQ1q+F93E0rHzl+Ienz4v8OrOn99Hp6Lf/KtfR65lc9vZb5U8vqrKcnljuO4ziO42SAG8sdx3Ecx3EyoFI3UbmYpOpkjtczf/Ba5hdez/zBa5lfZLycV5ykOkkrw6uma6UBrnsURV9lb/ec6sLrmT94LfMLr2f+4LXMPwor8bs7SpoSRdF3khRCGCbpMEmlDoYGDRpETZv+f5uaFStWmNdr165t9G+//Zb2df6+JDVu3Njo5cuXG71o0SKjeRNZWGgPyZ9//pn2/ZcuXZpWr/yCwP/TsmVLo5ctWybC9+A+NGzY0OhZs2al3cfp06f/GkWR3XCSCtWzfv36UVFRUYnm5+BxXLhwodGsXbNmzRLbWLJkidGsZUFBgdGsba1atdJus0mTJkb/+uuvaX+fPz9nzhyj48djFaw/azt37ty0+8jj8tNPP2W9lpJUt27dqFGjRiW6QYMG5nXqn3+23wpmvX///ffENtZee22jOb85hljf+vXrG/3TT/ab3jz+9erVM7pOnTpGc0xyvPEzScl6lnV+YT3jn2HBggVatGiRfcPUVKiejRs3juLnmfnzTTSOmjdvbvS0adOM5jmKx614H4xmvTl3ynqdx7Fu3bpGc2zwdR73Ndawiyycd6n2gfM/Ph9SvSf3efbs2VUyNxs0aGDOtQsWLDCvc15Qz549m++X2AavMZwLHBOcOzyWvEaxPpxbf/zxh9Gs7+LFi9O+X6r35D6uueaaRvMzx89HS5cu1fLly8szNyt1E5UqSXWndL/QtGlT/f3vfy/RHAwbbrih0S+8YL95uO666xrNgyRJe++9t9Eszrhx44zmROAJ5uuvvzZ6//33N5onct7Q8Mavd+/eRvNiJEm//PKL0d98843RO+1kD/Mdd9xh9D777GP0RRdd9GNiI0kqVM+ioiLzWXgC4nH85JNPjOYA7t69e2IbP/5od5vjhRfNzz//3Oi11lrLaE68Aw880OghQ4ak/f0uXboY/cQTTxh9+OGHi3Bi/+9//zP6ueeeM5rjlcelX79+Wa+ltPKCEd9/jrEddtjB6H//+99Gs97//e9/E9u49tprjX7++eeN5k0p/zGw3XbbGX3xxRcb3bVrV6O33npro9dff32jv/zyS6N/+OEHo1u0aCHCiwVvUHh+4cVo2223Lfnzk08+mXj/UqhQPVu2bKkbb7yxRI8cOdK83qNHD6MvvPBCo88880yj27Ztm9gGz2sffPCB0Zw777zzjtE8l/M4brbZZkbzRoCvs5a8CM+YMUOE+/Dggw8aveeeexrNmw/+I2Dw4MFVMjeLiopMTd59913zenxMSdI222xj9NChQ43ecccdE9uYPHmy0VOnTjX6jDPOMJrnc9Z7t912M3r8+PFGxx+mpHq/LbbYwuhvv/3WaJ7rpeRN0meffWZ0nz59jJ40aZLR8fsNngvSUeXG8hBC7xDC2BDCWF4gnNWLeC1T3cA6qxfxevKm1lm9iNeSNyTO6odfN1cfKvMkqlxJqlEUDZI0SJJatGgRffXV/z+15CM6Tv6dd97Z6C+++MJo3q1K0tixY43mY+B7773X6NNPP93ozTff3Gg+Gv3www+N5tMQ/ovpyCOPNJpP1/hkQkr+K2r33Xc3mv+q69Chg9Gp7tLLQZn1jNeySZMm0ZgxY0pe40X4oIMOMnqTTTYx+r333jM61Yniu+++M5r/ern++uuNfvrpp43meCGvv/660YcddpjRjzzyiNGdO3c2mmPrqquuSmyDY5RPSzp27Gh0u3a2QXuqJ5XlIKO5GX9qMmDAAPOzfGq3wQYbGM2nSKmeLHLsT5gwwWiOEZ4f+GTo8ssvN5rzgk8K+RTppJNOMjo+niVp4403FnnppZeM3nfffY3eaqutjOb5JH6cuD9pqNDc3HzzzaPWrVuXvManABy3PL+MHj3a6LfeeiuxQ506dTKaT4KmT59uNJfCuEzNpzyc+3xSwqf1fD8+xeZ4lZLHoVUr20KNT6qoeb4pJxWemxtuuGEUfxrI1RbOPc5dPlXh0pyUXEHhU9tBgwYZzVUBnr95bPmPbp6bOT74ZOvjjz82mtdhKTkmeW7l9eP99983Ov70nSsr6ajMk6jVKUnVKRuvZ/7gtcwvvJ75g9cyz8j4SVQURctCCH0kjZBUIOmhKIomlPFrTo7i9cwfvJb5hdczf/Ba5h+VWc5TFEWvSno1S/vi1DBez/zBa5lfeD3zB69lflGtbV8aNWoUxb85cOyxx5rX+U05rvW+8cYbRvMrzFLyGzz8GX7rgJ4V+om4lrvlllsaTc/T448/bjQ9T/x9fqNASq7303dF/xF/nl8Nv+GGGz6JosgapypJUVFRFPdScJv8NkxZfpdU/hD6SfhNtk033dToU0891egXX7RPyempYW05/rhOz2/FxL9pKiW/9p3qPblP9NDQv8Bvg3br1i3rtZSkVq1aRT179izRPFb8Jhn9BfzGF78RJkl/+9vfjKZvpayvUdN/9NFHHxlNXw19GXGfkCS9/PLLRtMb9P3334vQK8K516ZNG6Pp5Yh/xqFDh2rmzJnl+hp1RWjQoEEUP6/R+8Xa9Otn8x7pb+K3KKXkeZXjlN5UepZ4nuXc5Tcz77zzTqMZHUEvIf21qeJwPv30U6M5n+kD4vzmeKmK86y00q948MEHl2j6j+j9YZQIPW68RkrJbyZeffXVRr/22mtG85ttnMv0GnMe8Bvn/AYov6nP+5RUMQ38xij3kT4r+rrivrA33nhDc+bMKdfc9LYvjuM4juM4GeA3UY7jOI7jOBngN1GO4ziO4zgZUCljeUVp1aqVzj///BL9zDPPmNeZQsvYfeZAMNtCSnoS6Fnge9K7s8suuxh93XXXGU1fxcCBA43mWjzj79nmI1V8PZOWR40aZTTXfnkcUrU4yDYtWrTQKaecUqLp5Zk3b57RTG5m0juPiySts846RjOZd7311jN6xIgRRp977rlG33LLLUYzj4ueB/rV6D2gb6N///4iHI/0brD+TItmUnBVEUWRqRGzhfjZ6JthejMThqVk7hv9YvTJcC7Sw0KfCzO12KaHXhEmNw8fPtxo+r6kZAcDZvbQ+0NfVjzfpqpCFJs0aWJy2nhc6W8pK48rVXrzDTfcYPQ999xjNI8djwu9oczoYkI+z+N77LFHYp/ivPLKK0YzfVySbr31VqPpA9p1112NpkeT7XKqisLCQpNhxevko48+mvj5dDAPL9V7cq7R28vrIK9z9FjxXM65GPdjSklv8ZQpU4w+55xzROiroqeRrYd4bo7/fqqOC6XhT6Icx3Ecx3EywG+iHMdxHMdxMsBvohzHcRzHcTKgWj1Rf/zxh8kP4RozO8e/+eabRjNLKNU6N3OgmCVEfwC9OFyb79Kli9HMr6Cvg9kU7LfEnJlUvfO41l6nTh2jufbLtV16fdizLBssXLjQdMlmryz6h7iPzPyiz0tK9keiB4r9lzie6ONhLhn3mR4aZuGwtvQisHO5lPTIMWuqffv2RrO3Y/PmzRPvWRUUFRXp0EMPLdH0NDDni13X6R/r1q1bYhv04jBbiJ4F9k876qijjGY2DecN95HnBtabWUWpfF30wbG/Hs8n7L0ZH4OpcqiyAWvJHoL0G7HPIf0n9AZJyX56PM/Sl7nRRhsZzVrxfMHacawQ+hnvv/9+o5mVJSXnFnPJxo0bZzQ/U3U1YW/QoIG23377Es3zHj1OrB99wMxokpLXWo5rnqc4zpk9xfnPcznrRd8vPXX0XF1zzTUiF198sdH0bfL8weMY91im8iqXhj+JchzHcRzHyQC/iXIcx3Ecx8kAv4lyHMdxHMfJgGr3RMU9R8wKYTYFs2e4jsosECmZHUPvzdlnn200c1y4Ta4fc+336aefNpoeqmeffdZorvUyP0lK5rQwU4n5MuwhdtpppyXeM9vMmTPHZHkw32jmzJlGM4OHPQvpsZCkK6+80mhmd+y3335G00/E/k2s5QYbbGA0vQRcM2et2IuJnj4p6fX66quvjOZ4pc+C443ehWwxdepUMzfYp+722283mvOGn525QVLSg0KvHz2PPD+wHxs9a++9957R8RwzKdlDbPLkyUbT37jXXnuJMIvqqquuMprznzl07B9YFfz5559mrDMDiBk///nPf4zmPtKrJiV7hjJzib00OZePOeYYo9nHjPOAvdfOO+88o5kbxHnF842U9PlxPDJ37ogjjjC6qjxtpHbt2sYPyt6HHLfxjDApmXWYyrvJ+TphwgSj6SfcZ599jObc5XWRfUg5j+hnYq9EeovjPXhXwXMreyEyN/Dwww83evbs2SV/TtVrsTT8SZTjOI7jOE4G+E2U4ziO4zhOBvhNlOM4juM4Tgb4TZTjOI7jOE4GVKuxvEGDBsaA+uuvv5rX33//faNpiGTIJMP4pKRh7cADDzSaJmwGLtLAxnBNBvyxeSdD5BhSRsN1qkC/li1bGs2GxA0aNEj786nC1LJNw4YNTRPQuClPSgZfsoEojacMrpSkSy65xOgOHToYTdMujy0DIN9++22jWauyjMUM32NAHH8/1d+xqS6DDx955BGjaeisKoqKitS1a9cSzXBEGi0ZRscvUPD3pWR92OSYhmeGq9LIyzA9BipOnDjRaAYI0gzLLzdwTEvJ+comxvyyQrxxrGTnMo3x2WLOnDkaNmxYiea8YeAvjcaci6lM2Zx7PG8y8Jchx2yazoa0NCIzMPj11183msZkvj+b10vJsGaOn6OPPtrooqIio6vjPCutnDc333xzieZ1MB6sKiXPS5w3bJotJb+UwbkXD1aWkuclXhdp4ub5g3Pr7rvvNprN5hlKnar588EHH2z0bbfdZjSvxTwOV1xxRcmfKxKk6k+iHMdxHMdxMsBvohzHcRzHcTLAb6Icx3Ecx3EyoFo9UfXq1TPrku+88455nT4KrpMzZJLr5lKy8SED9hioRp/E119/bTS9O/RQsTknPVn0VPH32bhTkn744Ye0+8BGmPQb0DNVFdSpU8d8NjaXZbhm3759jWZd6AWSpBtuuMHoG2+80WiGMV577bVGM0SQ3o7nnnsu7T6yaSY9EWxgS++JlAzsY23j3hVpZVBinLPOOsvoVMcpG9SqVct4ku644w7zOudJ586djeZ+shaS1KdPH6Ppozj55JON/uc//2k0w1C32morozkGCwvt6Y0+Do4HNmqlhyvVPjGElL4uej3ijXXpE8wW9erVM8eGga/0+vG4sGk2g1WlpJePHjn6W9mAnL4eBunSc8UGxPRIcSwwePf8888XOe6444zefffdjeb1heHOnKtVRdOmTU24aVleTzY5Z0hoqmbvbPDL+U6vH8cEw1TpsaIfie9PjxODcFkbNkBOtQ9srM35xuPUo0ePkj/fd999ifcvDX8S5TiO4ziOkwF+E+U4juM4jpMBfhPlOI7jOI6TAdXqiVq4cKHJm2BzVXp7uG5Kjwu9QlJyLZ4eBfow6FFiJhPfj14Arg0ze4ZNUxs3bmw0c2akZCYHG4IyA4U5IPSOVQV16tQxvgb6g5inw4aVzAiiv0lKHvt11lnHaOYEcY27X79+ad+PGT5smklPDPN12Gw2VW5MPN9FSvow6KPieJw0aVLiPauCFStWmEazzLOiZ4HNl6+//nqjr7766sQ2hgwZYjSzaOhRYm4PPSrt2rUzmj4JZhWxfvRIsCluFEUibITLnLhmzZoZzfkd9yPRc5ct6Fe86KKLzOvMz2Gjbh5H1klKfk5m9rFpMY89G71zLvN8wJyhM844w2h6LHltYc6QJO28885Gs6E5vWL0DXH+M7sqWyxfvtwcL3qgOBd5jqE3kOdJKZmx9PDDDxvNuTZw4MC02+S5k/XjdZPHmnOR50Huj5T0H9PjyKwwNnOPj0F6/NLhT6Icx3Ecx3EywG+iHMdxHMdxMsBvohzHcRzHcTKgWj1RtWrVMuu3zOlhFsg999xjNHOjNtlkk8Q2PvjgA6NHjhxpNNdab731VqO5Ls6eUMw7oQ/miy++MJr9ebhOyxwZKekfoEeBfqJDDjnE6BCC0S+99FJiG5Xl559/Nvt+2GGHmdfpH+BxZe4HM4Ck5Odmfg17KzIr5quvvjKataOXgOv0zPCi/4V5Xalyf9544w2jOR6YI8V1e/akqyroiWImFvuMMa+K+Vf0SEnJ3Bb6DuhHZHZQz549jaavhh4VZg0xA46eKfp44plOq2CfSva/4+/QC/LAAw+U/Jm9Q7PF/Pnzjd+DPs4BAwaU+ftxOEalpP+E2UM8zx1wwAFGM+eJ70fPVf369Y2m/2ibbbYxmufQ7bffXmTw4MFGMweKPk7Wi57KqmLOnDl6/PHHSzS9oZybrB+PXapeiPQHsbcdj+/9999vNHOVeN3k3OZ1s3fv3kbzus15ljUk5SYAACAASURBVMqvyH6tL7/8stE8bpzL8XmSau6Xhj+JchzHcRzHyQC/iXIcx3Ecx8mAMm+iQggPhRB+DiGMj/1dsxDCmyGEycX/b1q1u+lkC69n/uC1zC+8nvmD1/KvQ0i1tmh+IIQ9JP0u6eEoirYq/rt/SZoTRdGNIYR+kppGUXRxWRsrKiqK4uu3zKJh3hH56aefjE7Vq4yZSswHoXeD3h36Wph/xPfjWn7Hjh2N5to/f59rvZK00047GU1/Af1DzOTg75922mmfRFHUQcpePdu0aRNdcMEFJZpZMqzlkiVLjF6+fLnR9I5ISe8XfQ7MP5kwYYLRzBXimjnX6Ql7Y7GW7Cd14IEHJt7jqaeeMpo5YkcffbTR9PlwH44//vis17L4s0TxvnDM/WG+2rbbbms0x/WcOXMS22DPNnpKRo0aZTTr9+mnnxp97LHHGk1PC31d9FTSI8fcoFQZXWXl43CfOUbjXpPrr79eP/74Y4mBMZtz88ILLyzRZfX75Dlx3rx5aT9TaX+HfTCaGW38fXpk+vfvbzQ9KqwNPZdjxowxmh4rSXrrrbeMZoYfry/MSuP5o1evXlUyNxs3bhzF94XXKHreOPfoieLclZI+KXp16V/kdYtjjB7Zpk3t/SLnIs+T9PUefvjhRqeam/S90bfFMcn6H3XUUSV/HjJkiGbMmGF3ohTKfBIVRdG7knhGPEzS0OI/D5V0uJzVAq9n/uC1zC+8nvmD1/KvQ6aeqFZRFK366stMSaV+TSGE0DuEMDaEMJZPI5ycoVz1jNeyOlLRnYzIaG4uWLCgevbOqSg+N/MHv27mIZU2lkcr1wNLXROMomhQFEUdoijqwK/BOrlHunrGa9mgQYNq3jOnolRkbnLJ2Mk9fG7mD37dzB8yzYmaFUJoHUXRjBBCa0k/l+eXGjdurM6dO5do+mCY28CcF65700MhJb029CQwG4br3lxPpreH2ULMUKHnhf396JlKlUfBSUOvB31c9OYwv6gcVLies2bNMn3h6BWhR4bHgf6leB++VXANm94N9uej16NHjx5G04Nz/vnnG12nTh2jmb/CdX7WbtiwYSLs/3jnnXcaTZ8FPXb0O5SDjOZmQUGB8f/xX7+XXnqp0ddee63REydONJreRCnZE/Kmm24ymn3pnnzySaMvueQSo/v27Wv0ySefbDTHC30Z9Ljx3JCqfxZvNtmHjvXiU6G4Zp5PKVS4noWFhcYPRn8L94n5OTynMT9Nko4//nijmcFFfwrPYYRznb5P+iPLyoTjuWDcuHGJbfIcRM8dMwR5Ttttt90S71kGGc3Npk2bmnzDjz76yLzO6+ARRxxhNK8PqXo28tzHnrPM5WrevLnR7HXIPrecmxyDnJu8F+A1kXNbSub4MWeSPSLXXHNNo+PzPZVHtzQyfRL1oqQTi/98oqQX0vysk/t4PfMHr2V+4fXMH7yWeUh5Ig6ekPSBpM1CCNNDCL0k3SipcwhhsqR9i7WzGuD1zB+8lvmF1zN/8Fr+dShzOS+Kou6lvLRPKX/v5DBez/zBa5lfeD3zB6/lX4dq7Z03Z84cPfHEEyWa2UILFy40mr1v2PeO2TVSsscPewB9+eWXRnO9n5lL9EAxx4dr8/QXcB/pqeD+Siv70sVhVlCvXr2MptcklZcj29SrV0/t27cv0czcYF9DHve4N05K+puk5LGmX43ZJvQfxfuUSTYHREp6y+gLYL+vZ555xujLLrvM6FReg4cfftjoeBaTlOyt165dO6OZl1JVzJ0713iQ2OOPfhD6eY488kijU3m56LNgLs/ZZ59tNOc7PVFXXHGF0eznRz/kHnvsYTR789GDxT6GUtK7ceKJJyZ+Jg4zeuLenqoyDC9cuND08LzyyivN6+zHyLlHr89DDz2U2AZ7ozFvkLU49NBDjaZfhX5D9kBlRiD9ssyQmzp1qtEcz1Ly+kPfH8/lHNP0TFUVy5cvN/ON58Vu3boZzUwszt3HHnsssQ2ed5h5Rz8y+0zyOsnz3tixY43mNer55583mp46HuuLLrpIhP5l5r7xPXhNiveYLadfUZK3fXEcx3Ecx8kIv4lyHMdxHMfJAL+JchzHcRzHyYAye+dlkzZt2kTxbJ5ff/017c9z7Ze5PPRMSElfCtfKW7RoYTTX1jfddFOjmX/C/mvMk6Cvgl4R5u+kCjnkGjb7TPG40MPAjI5BgwaV9HTKFuuvv34U9xBwH1mrfffd1+g77rjDaK6xS9J5551nND1OvXv3Npr+obhnS5JeffVVo7nP7JPIfl5du3Y1evLkyUan8hoMGDDAaGb2MA+FeTf777+/0eecc07WaymtrGe/fv1KNP2J1BtttJHRHHPMS5OSx5f+InqYOIbocWA/NmYDMYWdn4H5Oa+99prRqTxL9DDSR8f8G46x+Hy/6aabNHXq1Kyb3jbccMPouuuuK9H0ltJTyVwofqZU/T2Zx8ecNx775557zmieR9nvk16yu+66y2h6Kulh4Vjo3j3p86Zvc+7cuUafc845RtO3w7l5xRVXVMncbN68eRTfFq9ZHMf0htLnmyozi8eLWYOc3wcddJDRHA/0ZdETRQ/tjTfaLyryXHr77bcbTU+dlPRccozxMzA3MH59mT17tpYuXZqd3nmO4ziO4zhOEr+JchzHcRzHyQC/iXIcx3Ecx8mAas2J+vPPP43PhOuozAGin+noo482mn4DKemzYLYHfQ8HH3yw0al8LXHY44m+Ca6Ts68R8zRSedLYh2jWrFlGn3DCCUbT01AdLF++3NTrzTffNK+zdxrX4elN4xq6lPRRMJOL44UeJvqL6D+jj4K5UXydWTX8zNyelPSjMJeMPaiYZcR1/KoiiiIzH9mfi7k77DlJbxdzZqRkRhZzuLjNc8891+gHH3zQaPbjiue8SMl8NY4xejKZlcO8Ninpo2GeEd+TczmeC1aRLJqKUFhYaPqC/fLLL+Z1HkfOK2b4vPvuu4ltMPeJfkTWkn44+lnpedp9992NZo7UpEmTjKY3lZlu2267rcqCvkz2wuR4TeUtqgoWLFigd955p0TTX8pzCs+tnBddunRJbIPzl9cceqbiOWRSMjeMflT6k3h9GDlypNH05bF3IjPmpKRnjXOPY4TXj7jHlv7bdPiTKMdxHMdxnAzwmyjHcRzHcZwM8Jsox3Ecx3GcDKhWT1Tz5s1Nryn6CSZMmGA0+87RT8S1YCnZP4dr81yrpWeJvhiu5afq8RaH67Abbrhh2vfjOq1Udj8/9vz59NNPjY77IaqKpUuXmnVz5l9NnDjRaPpT2NOQa+5S0vNGzwqPJT0mXKfnmjl7Y3F79Baw5xj9L/TwSDJeBknabLPNjGYWEjPCevbsafQLL7yQ2EY2WLZsmfHOcNw2aNDAaOanMfeFOUGStMsuuxhNnwTn3vXXX5/29aeeespoZlcxs4keyx9++MHo9957z2iONynpi+OY2XrrrY1mfeOeKp6bssXy5ctNllw8m0+S7r///sTPx6FvlF5UKVnvtdde2+jFixen3cbgwYON5nFlZlNRUZHRr7zyitEcS/S7HHfccSIco+ylx/5v9OnQZ0UfVrZo1aqVObdwHuy5555G85yz4447Gs15ICX9XcxQYp9J+nJZn2nTphnNTD3uA6/9ZPTo0Ubz+iIlryEcE/TR0QMXv24ysy4d/iTKcRzHcRwnA/wmynEcx3EcJwP8JspxHMdxHCcDqtUTNW/ePJObQ69Pjx49jKafYMqUKUYfddRRiW0wz4Zru5999pnR7Mf1+OOPG33aaacZ/eijjxrNvByus7LPHf1KzJ2SpI8//thoejfo5enYsaPRqdaLsw19F8x9iefhSMlcGeZjpfK3cS2/b9++Rh9//PFG09PA/CzW/qSTTjL6u+++M5r9u9ibiev6qbKR6Lnj+KC3qCzfT1VRq1YtM67ocaNHhb3S2NuKHhcp6VFiDtiZZ55pNL1+9NWxvxfrx33kuYH9/diDbIstthBhNhjrd8oppxh96qmnGh33ftBjky1mzpxpepFddNFF5vXtt9/eaPo/brvtNqNTebfYn425QTwf0IdFPyM9ePSv7Lfffol9iFOWF5XnYSmZn8T6M6uKvq/qOM9KK88ZcU8Ssw3vvPNOo5n5RD9SqnxFeth4neK1mrlOzEvkPhLOPebl0c9ET2ZBQUHiPZnTx/egV4zX0biPi+e/dPiTKMdxHMdxnAzwmyjHcRzHcZwM8Jsox3Ecx3GcDKhWT1RhYaHpX0XfxNChQ40+8sgjjeaadKq+YvTacO2U69xjx441mv3V6Mvg61xL5ror11Y33XRTo5lrJSXzS/i5+Z70jpWnT1RliaLI+JiYI8JML+aG0LPw4osvJrbB3BV6Nfg7zACjhyaeUSYls27osWMfRXpqmHt0zTXXiNBXw/5+zDahP2X69OmJ96wKCgsLTd9JzkXOG3og6HdKlX/G3Df6WPr372/0EUccYTTrTf8Y94HzhJ5K+np4rqDnKtV78riwVyI/c7z+9Jlli+bNm5scH44hfi6eZ5nB9Pbbbye2Qf/QjBkzjGbvMfpy6JlhX0P276NPdKeddjKax5KezFSZX/Sj0qPG8cT5nSoXriqoX7++tttuuxLNno70Au61115G81yc6pzC6xK9mzx30n9Efyivw8wF41xnLz1+JnqieK6Xkhlt7OfIvLMPP/zQ6PgYYq5ZOvxJlOM4juM4Tgb4TZTjOI7jOE4G+E2U4ziO4zhOBvhNlOM4juM4TgZUq7F8zpw5evLJJ0s0G8TutttuRjOosHHjxkbPmzcvsQ2aAxlUSYMhTdz16tUzulu3bkbTXPzqq68azc/A7bN5MJvYSlLnzp2NpqmOhnqaMqsjBK6goMAYkfklARrw+TqNhKnM8DRZM9Bx3333NZqBcHyd4ynecFeSBg4caDQN/TvssIPR/FICw92kZOAjxyz3gbWk4bOqmDt3rgn2ZCAjP0e89lIypHaPPfZIbKNu3bpG07y6cOFCo2nSZkgg5xIDWxmmRxM/Tf78jJy7UjLglY1T+RkYKhk/bhVpcloRQgjmPMjjQNM2jeP8wkWnTp0S22AYJrdBszrN0DSC81pAaFy+8MILjR40aJDRDMaluVpKNkEfM2aM0ZzvbLrLn68q6tWrZ86Xr7/+unmdRnKa+nndZLNgKfllAxr/+SUd1ptzj/t48cUXG83zHveRgaGcZzzfSNJZZ51lNL94wms3zyfxsGZ+cSYd/iTKcRzHcRwnA/wmynEcx3EcJwP8JspxHMdxHCcDqtUTVbduXeMbYZgZPRD9+vUzetiwYUa3aNEisY3x48cbzYagDBqjD4LvSV8EQ+MY8MWQQWquR++9994iXGun/6Qsj0OqoMNss2LFClM/Nv9l01J6xe6++26j6ZeRkr4K+oXoFaN/iN6AffbZx2j6VRj+GG/iKiX9cvRAbbLJJiIDBgwwOh6aJyV9V59//rnRl156aeI9q4LCwkLTrJtBdfTZseHse++9Z3SqBsT0mDBA8bjjjjP6tddeM5p+Ro45Bj6yiSoDArk/c+bMMZpBulLy/MG5Sv8QQynjTbXHjRuXeP9sEa8fz4n0l/Ac2L17d6NTNWNl4/UHH3zQaPqNymq0zWbfHH8M6+W8YKgtgxQZBiwlPXI83zAoldcC+q6qil9++UX33ntviWajc16zevXqZTQDoBkgLUnHHnus0fys7777rtEc1zyvMciW3mE2g+e5m2OSpDrXMgyVXjF6anv27Gl0vLl4qibNpeFPohzHcRzHcTLAb6Icx3Ecx3EyoMybqBBCmxDCqBDCVyGECSGEc4v/vlkI4c0QwuTi/zct672cmsVrmV94PfMHr2V+4fX86xDYKDDxAyG0ltQ6iqJPQwiNJH0i6XBJJ0maE0XRjSGEfpKaRlF0cZq30kYbbRTdcMMNJZprs127djWa66hsWst19VTvSZ8D/UPMM2I+DrNo2LSS6+bMKurYsaPR9G3QFyYl19r5GZipwfVrrvXuvffen0RR1CGbtSwqKoriWUD0j4wYMcLok08+2WjmYzVv3jyxjf32289oepzoaWC2BzN7qOnDoseJx50eKvqX6POSpDvuuMNo5uPQW8D5WKtWLaN79uz5SRRFHaTszs2ioqJozz33LNEcY2y0y7wreqQ4r6Skh+m+++4zmjkv3AZ9F//73/8S24jDejPLhg2H2XT01ltvTbzn1VdfbTT9QtTMworP98cff1yzZs0KUnZrWbdu3Sieq8SMHjb2ZmZP3H8jpfbQ0PPIWnHc0vPE11kL+h95XuX5gmOLtaRnS5IOOOAAo5kZSC8qrzfPPfec0R9++GGVzc34uZbezHvuucdoXkfp9UzlV+R1KH6dlpJjgh4petiYdcicKfoBmcdIRo8ebTSbCUvSzTffbDQbxHMMxjMrJevD+vzzz7VgwQL7oUqhzCdRURTNiKLo0+I/L5A0UdI6kg6TtKrV+1CtHCBODuO1zC+8nvmD1zK/8Hr+daiQJyqE0FbStpI+lNQqiqJVj1VmSmpVyu/0DiGMDSGMnT9/fiV21ckmla1lqm/sODWH1zN/qGwt+RTGqVl8buY35b6JCiE0lPSspPOiKDJ3Q9HKNYiU64JRFA2KoqhDFEUduETg1AzZqGVFYvGdqsXrmT9ko5ZcGnNqDp+b+U+5cqJCCLW0ciA8FkXR8OK/nhVCaB1F0Yzi9d+fS3+Hlfz555+m5xLXtYcMGWI08y/ogWDeiZTsj0NP0mabbWb0t99+azTXxen1OOSQQ4xmngTzLbjPn3zyidHMOpKSmSj8lyU/A71i7GsVJ1u1LCgoMJlXPM70jtAjw1yXVE8pX3rpJaMPP9w++aYXjP256C9iVhVv6tkzjJ4a5g7RDxfPAFoFc4ToL+HaPo8TxyfJVj1r1apl+ovRIzFlypS0+8lxzCwbSfrtt9+MPvHEE9Nug8f/qKOOMnrWrFlGM0eKHgnuE+vH8XL22WeLsH701dE7xn2Me4F4s5OtWtapU0cbbbRRiaYf6LLLLjOavTjZRy3VTRn7Dp566qlGMzeMvevoL+Sxv+SSS4xmbzxmCtKTSe/qEUccIcLss/bt2xvNMc6sqdNOO81oZlNlq56LFy/W999/X6Lp7TnjjDOMpteLXs833ngjsQ36j4cPH270eeedZzR7SPbt29do9kalR/aqq64ymp4n5rHx/Tj+pGS+Gb1hffr0MZr9GuOfifcR6SjPt/OCpAclTYyiKO60fFHSqrPgiZJeKPdWnRrBa5lfeD3zB69lfuH1/OtQnidRu0o6XtKXIYRVX0e6VNKNkp4KIfSS9KOkY6pmF50s4rXML7ye+YPXMr/wev5FKPMmKoqi0ZJK+6rfPqX8vZODeC3zC69n/uC1zC+8nn8dqrV33uLFi/Xdd9+VaPaAY185rpvz9fh7rSKedSMlvR3MM6GngR4nrpty7ZgeCK7N0xNFXw7zM6RkhhI1/UTM7GDuS1XQqFEj89npm2BtunTpYjQ9EPRhSEk/Gnuf0RvAfkrMlmE2DdfV+fvMCGPGGPNX6MOQkj3DCgvtlKMXjH636vpmDv2KHGOEx6Jly5ZGP/XUU4nf4WenR4reP/YyY+9M/j7nWtwXJCX7ubGnGH049MhJyX57H3zwgdHMDqNvM55tVpH+XBWhWbNmOuaY/3/AwXMM95HnC3p76EeRpHgOlSSddNJJRtNDs9VWWxnNuczzB/u/cbzxM/H8Qb8krzVSMm8p3jtNWukti8NMt+roUSqt9BvG8wsPOugg8/rgwYONHjVqlNGvv/660QcffHBiG/Swcj7379/faHqDeR1kHh49RtxH5oLxvMffZ56jlPR60X/M7CqO4bgPqyK19bYvjuM4juM4GeA3UY7jOI7jOBngN1GO4ziO4zgZUGbvvKxuLIRftPIbCS0k/VrGj9c0ub6PFdm/9aMoaln2j5Ufr2XWKe8+Zr2Wktczy3gty08+7eNfvZ65vn9SFdSyWm+iSjYawthVjRpzlVzfx1zZv1zZj3T4Pq5++5GOXN/HXNm/XNmPdPg+rn77URq5vn9S1eyjL+c5juM4juNkgN9EOY7jOI7jZEBN3UQNKvtHapxc38dc2b9c2Y90+D6Wn1zZj3Tk+j7myv7lyn6kw/ex/OTKfpRGru+fVAX7WCOeKMdxHMdxnNUdX85zHMdxHMfJAL+JchzHcRzHyYBqvYkKIRwQQvgmhDAlhNCvOrddGiGEh0IIP4cQxsf+rlkI4c0QwuTi/zet4X1sE0IYFUL4KoQwIYRwbi7sp9czo/3zWpaTXK9l8f54PctJrtfTa1l+cr2WxftTLfWstpuoEEKBpLsldZHUTlL3EEK76tp+GoZIOgB/10/SyCiKNpE0sljXJMskXRhFUTtJO0s6u/jY1dh+ej0zxmtZfoYot2speT0rwhDldj29luVniHK7llJ11TOKomr5T1JHSSNi+hJJl1TX9svYt7aSxsf0N5JaF/+5taRvanofsb8vSOpck/vp9fRaei29nvlcT69l/tSyKutZnct560iaFtPTi/8uF2kVRdGM4j/PlNSqJncmTgihraRtJX2omt1Pr2cl8VpmRE7WUvJ6ZkhO1tNrmRE5WUupauvpxvIyiFberuZEDkQIoaGkZyWdF0XR/PhrubSfuUyuHCevZeXJpePk9aw8uXKcvJaVJ5eOU1XXszpvon6S1Cam1y3+u1xkVgihtSQV///nGt4fhRBqaeVAeCyKouHFf12T++n1zBCvZaXIqVoW74fXM3Nyqp5ey0qRU7Us3o8qr2d13kR9LGmTEMIGIYTakrpJerEat18RXpR0YvGfT9TKtdQaI4QQJD0oaWIURbfGXqrJ/fR6ZoDXstLkTC0lr2cWyJl6ei0rTc7UUqrGelazsetASZMkfSvpspo2mhXv0xOSZkhaqpXrzb0kNddK1/5kSf+R1KyG93E3rXzk+IWkz4v/O7Cm99Pr6bX8K9fS65lf9fRa5k8tq7Oe3vbFcRzHcRwnA9xY7jiO4ziOkwGVuonKxSRVJ3O8nvmD1zK/8HrmD17L/CLj5bziJNVJWhleNV0rDXDdoyj6qrTfqVOnTtSgQYMS3bhxY/P6okWL0m5z9uzZRhcVFSV+prCw0Gh+vsWLFxvNffj5Z2vULygoMLp27dpp35+6VSsbQfHbb7+l1ZK01lprGc3PvdIv9//Ej6kkLV++nL//axRFLRMbsu9ZoXoWFhZGtWrVKtEtW9q3nzdvntFNmjQxmp+7Tp06iW38+eefRtevXz/dR9D8+ebbq2rRooXRCxcuNLpevXpGL1iwwOhGjRoZHf+8kjRt2jSjmzVrltgnjh+OcW6DtZ0zZ47Rf/zxR9ZrKUkNGzaM4vu/ZMmStPvJcc6f5+co3obRHNesN39+xYoVRi9btsxo1odz+9dffzWa44nb47lCktZYw/67k2OOc5Hni/hn/PXXX7VgwYLkgQIVrWeDBg2i+Lnx999/N69z3NetW9dozoNU1wgeO85nvidrw+PC8fLLL78Y3bSp7czBfWLtOJZSfQZeK5YuXWo0xxd/nnO5Ks6zklSrVq0ofjzXXHNN8zo/K8+l/Bz8eSk5rlm/uXPnpt0Gz308N5Z1LqVmvXiu4PtJyc/Az8nzP88P8bm8ePFiLVu2rMy5KUmFZf9IqewoaUoURd9JUghhmKTDJJU6GBo0aKD999+/RO+7777m9fHjxxvNm4HHHnvM6IMPPjixDV7M+R7ffvut0Z07dzb6jjvuMJo3am3atDGaFw+e6M8991yjX375ZaOff/55kb59+xo9dOhQo3lh7tixo9G8gRk8ePCPiY0kqVA9a9WqpY022qhEn3766eb11157zegDDzzQ6BdesF+I2GCDDRLbmDhxotHbb7+90Zw0I0aMMPrUU081+uOPPza6ffv2Rr/zzjtG77nnnka3bt3a6PPPP9/oY445RoSTnWOc2+DJi2N+zJgxWa+ltPIk+I9//KNE//ST/RZ1p06djOYNxtSpU41OdVO86667Gv3www8b/dVXdvf22GMPo3kzwJsi/uODc3vIkCFGczzttNNORn///fcirM/IkSPTvsd6661ndPwz/vOf/0y8fylUqJ5FRUVmPo4ZM8a8vuWWWxq9+eabGz1q1CijeQ6VpG233dbo4cOHG833ZG14XHhjN3DgQKO7du1qNM+7u+++u9Fff/112p+XkteKGTNmGM0bOd68fPHFF0YPGTKkSuZm3bp1td1225Xoc845x7zOebPhhhsazRtc/ryUvPnfbLPNjGZ9119/faN79Ohh9AUXXGA0z3McD+uuu67RvAF69NFHjeb5SEre2PP68emnnxrN88Obb75Z8meOn3RUZjmvXEmqIYTeIYSxIYSxqf5l5+QMZdYzXstUJ1YnZ6jw3OQNipNTVGhu/vHHH9W6c06FqPDc5BMyJ7eocmN5FEWDoijqEEVRh1T/OnVWH+K15NMwZ/UjXk8uhzirF/Fa8qmCs/oRryeXupzcojLLeRVOUl2yZIl++OGHEj1s2DDzOv0fXbp0MZqPfPfbb7/ENvgIdvLkyUbHty9JTz75ZFp97LHHGt2unW2gzX8lcLnw5ptvNppLEHx/KbksxWWozz//3GguU3DJsZxUqJ4NGzY0yzNcQuTj2FdffdXoddax//jq2bNnYht8pM9jy3X48847z2gug2666aZG0/9G/xqXNWbOnGk0xwLHlpT83BwPfLTep08fo7kUQ10KFZ6bIQSzVMVlSHpYbrvttrSv89hIyUfkrB+X1+iTobeDS+1cgvjoo4+M5vINl5Q4XnbeeWeRL7/80ujRo0cbzaWU999/3+j4GEnlhyyFCtWzsLDQLD1xub9t27ZG87hybtLPmOrvDjjgAKO5XMPac3y89dZbRu+yyy5Gc6w89dRTRvNaQP9LKr/ifffdZzSXjzfeeGOjJ0yYYDRrYsInrgAAIABJREFUXU4qPDebNWtmrhP8bDxW9I/dddddRp9wwgmJbXC5jOOWy3G8RvXu3dvoiy66yGgujdPzxLnAMbj11lsbPWnSJBHeH3AMcszyfBM/HwwYMCDx/qVRmSdRq1OSqlM2Xs//Y++8w6Qos7f9vAw5SDIsJnAVFLMu5gQGVjGACigmVBSzmBdzzj/FHFARjCiYMAAKiq5ZWMwoplVxBUQBReIM9f3BOF+du4qemaZ7pm3PfV1e8tBDV3Wdet+q6fep5xQPXsviwutZPHgti4ysv4mKoqg0hHCypLGSSiQNiaLok0r+mVOgeD2LB69lceH1LB68lsXHiiznKYqiFyS9UOkPOn8KvJ7Fg9eyuPB6Fg9ey+JihW6iqsvixYvNo9NHHnmkeZ35JHyEdMcddzSafgRJ2myzzYyeNGmS0bvttpvR9Mmce+65RvNRTq7V8im1zTff3Gh6WOiBYiSClHyc97777kv8TByu1fPx43zQtGlTsxZPvxDX6en94eOoI0eOTGyDkQL0HHGdnOv4HTt2NJrnD7fJBx/okeK5wogEPnosSRdeeKHRfAque/fuRo8aZb/ZT3vMPh/UqVPH+Jp4Xo4ZM8Zo+ke+/dY+3b366qunbiMOHzvneOfP05dBjxx9WRyLX375pdHvvPOO0fQGMc5ASubZcKxxPmCWUHz+yJdhOIoiE0FBbxkjVG6//Xaj+RR1Wn4f52pmKPGReD7C/vnnnxvN84cPrnDs0Tv44osvGs1app2P++67b8b3YEwDPzMjEfJFWVmZmTfoq6X/lHMM/abjxo1LbCMeoSAlxyI/O8fGkCFDjObY5j7Sr0jf3rPPPms052KeL1IyF5D+aF7b+Rlfe+21ij/T45sJb/viOI7jOI6TBX4T5TiO4ziOkwV+E+U4juM4jpMFNeqJat68uWn7wpwNtoFgBhPXyemJ+GMbcbhOSs8C12bpUWG+CddRmTUxbdo0o5l/UVnrACnZKoYR+vG1WymZ0UGPVD5YsGCBaWHSvn178zqPC9fQ//3vfxvdunXrxDaYJVNZphKzSXguMF+F/jR6apjPxXVy9pNi6wMp6cN64403jGYbIH5GtrOg7ytXlJSUmNwl7kd83EqV9xnjuEv7Gfok6Hugr465TvSLTZ482WjmxvCcZD8u+j7uueceka222spoZk/RC3TUUUcZHfdy5MsTtXTpUnOu06/IdkXM56Mnit4/KelRYm4Ta1VZX8JOnToZTV/PW2+9ZTTPler2zZSSvhrm67G3Gs+ntOypfLBkyRLjJWb2IT2z8fYlUtILRK+XlPSP8vjxPGeGGuc11oNzMeduekF5PeGxT2vjw7ZS9N1xvHO+P+GEEyr+zJYxmfBvohzHcRzHcbLAb6Icx3Ecx3GywG+iHMdxHMdxsqBGPVGNGjXSxhtvXKGff/558/rAgQON5ron/UWHH354Yhtcx2ZWzAcffGA0+2/R08AsIvpomEMV9wlJyX5eEydONPr7778X2XvvvY1mRs/XX3+dUdcEZWVlxiNEPwozneg3OOmkk4xOy+T517/+ZTTrfdhhhxnNdX16qphVw9559JLR48TX2V8prV8cs2fogbvjjjuMprdg9OjRiffMB3PmzNHTTz9dodl3iv0a6f3jOIl7OP6Avgbm3dADRS9PZTlO9HHRt8FxwlwwZtXQeygl83TYw4t9yfgZ49t48MEHE++fC6IoMseqMu/V+PHjjd51110r3caECROMpnc0Ps9LyVrRA8M5jz/PscfsKuZU0b/IXm1S0hvE91h33XWNpk+rd+/eRl933XWJbeSK+P7T1/vII48YTf9Rly5djL7ooosS78/rZN++fY1mphq9nvRMcr6gb4t9CemPZH2ZbbX//vuL0GfLay3HJu8V4nNWmudqefg3UY7jOI7jOFngN1GO4ziO4zhZ4DdRjuM4juM4WVCjnqhffvlFI0aMqNBck+S6K9eoX3nlFaP79euX2AbzSdgziVkwXFu96aabjP7999+N5lopfTnU9Dz16tXLaK5vS0kvCHNcmNnBz0SfVz5YsmSJ/ve//1Vo9o1jBgtzYa6++mqjjz/++MQ26KNiLtPll19uNPeBfjR68LbZZhujmUXDvJUTTzzR6HiuiJRck5ekddZZx2hmm/F84bo9ezXmi0aNGpm+k/SL0ONCP0+zZs2MpmdKkmbMmGE0c3Z4ntNPRE8Vzw964ujLi5+vUjI3hjl0HKtSsp8j85G4D/TixD1U9IHkkvhn4Xbosdxzzz2N5nGnF1WSXnjB9s/lWKS3j+cP+84xl4d+RXqkmAHUrl07o9kv8KyzzhJhhhd9Pccee6zRzPhL63uaD+rWrWtym3gsOW7o++U1iHOvlMxMomeJuXDfffed0ZxbWT9e53itZ44Y52Zm9vEclZKeRnoB6ZPjPsWvWWn9IpeHfxPlOI7jOI6TBX4T5TiO4ziOkwV+E+U4juM4jpMFfhPlOI7jOI6TBTVqLG/YsKEJMGOTwcaNGxvNQDYa/2jAlpJmUgbqffnll0bTVHnhhRca/dRTTxn90EMPGf3YY48ZTfMxwxRpPqYZTkqa29u2bWt0t27djGYjVgYl5gOaHf/zn/+Y13/66SejX375ZaNpjmRYn5QMs2PjTR5LBj6OHDnSaB5HBmH26dPHaJqlDzjgAKP5EAMbfUrSrbfeavRee+1lNIPxGJx32mmnJd4zH8yfP9+YN2nC5Fh7++23jWaQaVrDVzYcp3l00KBBRnM+4IMmNLNeccUVRjPYlOdcz549jeZckNZEmaF/DDLkceM+x8/hfBrL42GbfLiB88ezzz5rNPc57UEV1vuGG24wmmOX5w/DHS+77DKjGQg5duxYoxm8S2N6kyZNjE4bR1tvvbXRNGAzcJJzWjycNp+UlpYacz9DY8855xyjKzNhs75SMjiYD47Q+M2HMjjeGZzLB1X4wAUfPOLcfskllxg9fPhwEc5Je+yxh9Ecu7xuxhuUpx2j5eHfRDmO4ziO42SB30Q5juM4juNkgd9EOY7jOI7jZEGNeqIWLlxoQtK4tkuPAP0jbHKbFnbGkEeGutE3Q98DPVQMO2STU/pu6KPhWjLD+OrUSd7H9u/f32g2NebaL4PP2Hg1HzRu3Ng0Y11ttdXM6/QocM2da94PP/xwYhv0KMXDIKVk81cGAPJYM4CPnrxff/3VaDabveCCC4xmWGRaI2j6FVq2bGk01+lZO37mfFG/fn0TWMj9YHge601PQ9qxYDgmz2t6+Xh8GQhI3w3rz6BdNkUeMGCA0fREpfki6OXYeeedjb7++uuN7tGjh9Fxbw+PWa4oLS01Phh6Yujj3GGHHYy+5ZZbjKbXTJI6d+5sNINK6cNiOCY9Swy25Pu1bt3aaHos2cCYted1QEpeP7hNNqPnZ+JYzhfNmjUzDX95DWKjXQZZ8jxnqHDa33H8sxkzvcX0Yb355ptGT5o0yeju3bsbzbmBHjqG2qZ5Ljnf04PJBub0bcU9rrwPyIR/E+U4juM4jpMFfhPlOI7jOI6TBX4T5TiO4ziOkwU16olq0qSJWYfkumqmZp1SsgEpm06mwaaR9FWxqTFzoZg1s+GGGxrNnJ958+YZzSySyvZHSvqq6POit+OYY44xmp8pX8Q9Rtddd515jcfl1VdfNfqjjz4y+qSTTkq8PxvG8t+wKenFF19sNGvBxqv0SXTt2jXjz9NncfvttxvN7Jy0fWA+zm677WY0/QxscpwvVlppJbMvbGLKnB56JthceIsttkhsgw0/6YuhJ4p+QW6DuU/0N9LjRk/E7rvvbjSPfVo+Ev1Djz76qNH83Jzj4tvkMc4VixYtMnl59Fgy44fj6LbbbjM6zX9Cbyl9OvSv0V/Gxu5sRs9zg8eVPlBmONGLet9994l06NDBaNafuYSVXZ/yxeLFi82czybmvF7wvKWfKG1ssn6c6+g3pBdw3333NZoZj3FPlyS9//77RvNaTs39S2tY/8QTTxhNbx89lhyb8fmjOs2l/Zsox3Ecx3GcLPCbKMdxHMdxnCzwmyjHcRzHcZwsqFFPVP369U3uEtfauY5O/xD7s/3222+JbdCzwLXUeH8cKdnTJ96jKA16Adh36uqrrzaa+Rv0adAjIyV7xDEviX3mfvnlF6OZ6ZMP6tSpY3wH/NxDhw41mj3hmOExYcKExDboa+D5stNOOxnNY8veVlznpxdk9uzZRjNPh733zjjjDKPpE5CSn4v5J8yioXdkl112MZq5Urli7ty5xgNGf9HAgQONps+CHgZ6AaWkD4aeBPa6omeI3g965tg3c7311jN6lVVWybh9el6GDRsmwvdgv8TRo0cbTR9nvC8mz9dc0ahRI2200UYVml6yTz75xOg2bdoYvWTJEqM5v0hJfyK9o5ybOU8eeOCBRjMjkLlE9HHR30J/4+DBg43mPC8lezPSv0iPHvO2mDPEvqq5IoRgcosOOugg8zo9UMzQ4rxJf5OUnJfYF5Sa5wTPKc7vPFac55gbyGPP7dGvKiV9dKwv529+pnjWFeeqTPg3UY7jOI7jOFlQ6U1UCGFICGFmCOHj2N+1CiG8FEL4ovz/NRPd6qwwXs/iwWtZXHg9iwev5V+HqnwTNVTSnvi7gZLGR1HUXtL4cu38ORgqr2exMFRey2JiqLyexcJQeS3/ElTqiYqi6LUQQjv8dXdJncv/PEzSBEn/qnRjdesaTwE9K1yXZfYEfRb0O0lJXwR9E1wH3W677Yy+4YYbjGbfIa61c3365ptvNpq99+hxSevRQ08U/QT777+/0ewJRy9R3K+Qq3qWlpaaeo0fP968zuNK6C1J84bxWPfq1cvouXPnGs1sEvouunTpYjSzQOiJoneA/Z3YH479oKTk+cAeYZ06dTKaPiz6euLkcmw2btzY9Omjl+fBBx80mr3zWIshQ4YktsEcrspyW+hvPProo41mZhuPFfsOcr6gb6devXpGc66QkhlsrCezqNhDMp6Hxn5yuapn3bp1Ta85+jzZs5CaPeR23XXXxDY23XRTo5kbRk/k2WefbTTnC/oJ77333oyvM1+P5yt7GPI6ICW9o8yZGz58uNHrr7++0Wm9Ff8gl2OzQYMGpm/f3XffbV7fb7/9jKbXjtcY1k5Knvv0WTHHi35SeuQ4r9HDxGs5539mdvEcZO6UJN14441GDxo0yOiLLrrI6EMPPdTouNc4raft8sjWE7VaFEV/OPmmS1ot0w87BY/Xs3jwWhYXXs/iwWtZhKywsTxa9khLtLzXQwj9QwgTQwgT056mcwqLTPWM17I6Ty84tUN1xma+nhRzckdVxyafMnMKj+qMTX7L6xQW2d5EzQghtJGk8v8v95n6KIoGR1HUKYqiTnxE2SkYqlTPeC35OKlTMGQ1Nvl1vVMwVHtscjnfKRiyGptsoeUUFtnmRI2S1FfSNeX/f6Yq/2jBggXGn7PPPvuY17k2z2+u2H8tbd1yq622MpprseyXRP/RlltuaTR9G/Q80NfAHCl6IujLiGdTLO89mPPBfbr//vuNjmfEVJFq13P+/Pkmx6djx47mdXrBOBFwzZt9zKTkujo9SMwR48/Tx8N/z9wXemzoyYv7TKTkuXTPPfeInHzyyUazfxy9B5deeqnR9CrQY5VCVmOzrKzMnGfMFlqwYIHRzHHh56L3T0qOFeadsf/i6aefbvRRRx1lNHPjuM8cF/RMMXumb9++RnPsSsnPzfOY3+jR5xk/B6v4DUO167lw4UJzbJm/17lzZ6PpDeQNddpNGefql156yWjOoxx79JLRa0YPFH09PDeYtzZt2jSjv//+exHWhrXkPMvjwp6mVSCrsTl//nxNnDixQnNO4TWFv+DyPKN/SZK22WYbo+kHfOYZu6u8NtOny/mfmXrs/8qxG/eAScnzh3OJlPSk8ksbeoffffddo+O5Ujn1RIUQHpX0lqT1QwjTQgj9tOwk2COE8IWk3cu18yfA61k8eC2LC69n8eC1/OtQlafz+iznpd2W8/dOAeP1LB68lsWF17N48Fr+dfDEcsdxHMdxnCyo0d55devWNT6Wnj17mte/+eYbo9mPi+uU7McmJdfvqSdNmmQ01+7peWD+CHs6cZ2Vvhyu9XJtl3lJadug5nGgl4frz48++mhiGyvK0qVLjYfmp59+Mq/fcsstRq+55ppG//Of/zSaPjAp2ZeMHigeO/oemOHF/luE/iT2j2M+C70Haebs7777zmjuM88/7iM9Ufli7ty5ev755ys089EqM57Ti5iWE8WxxPP2gQceMJpjl/lF7I1Fbw+zyti/i1lF9Gil+br45BvnE3px2rdvb3SaHyXXtG7dWkceeWSFZl8ynufMkWLvPPb/lJJ+QfalfPjhh43msedxZO14HLmPzANkrzZmvKXV8o033jCaPiCen9wn5srlC3rcKusjSK8W680sRCnZ45M+K86FzGBjRh89kqwPM7roTeY1j/cC7NUnJT1tld1P8HqRrYHfv4lyHMdxHMfJAr+JchzHcRzHyQK/iXIcx3Ecx8mCGvVENW3a1PRYoqeB65xcw2RO1E477ZTYBr0bzKsgXFunJ2rkyJFGMwemR48eRnO9mhlPzLKI+1CW9zN777230ewJx22mZU/lmmbNmpl+Rsxxefvtt42uzJuW1geRve7oiaGngfk3/Hn2vqOmf4k5UvRxcZ/T+tyxp9jUqVMzbvPNN980uqZS/ps3b65u3bpVaGao0C/AcXXVVVcZzZwfKekhOeWUU4y+9tprjabvZty4cUZX5l/ksWNfQma+0b+UFg5M7w6PC30zHJvx/n9jx45NvH8umD17tslN4pzIfC1+TvYxS/MrxjPipKSfiL3N6ImM5x5JUsuWLY1mthX9KwMH2t69nEfpofrqq69EeH3p3bu30fT5VNYvMl8sWrTIzBPsfUcPFK8XPLYnnnhiYhu77WYfGmRvPGbiLQtcX/7Pxz15knTfffcZTQ/W+eefbzSvD/Qet23bVoS5kzxH2T+R+xzPnlq4cGHi/ZeHfxPlOI7jOI6TBX4T5TiO4ziOkwV+E+U4juM4jpMFNeqJWrJkickTGjFihHmduS7M1OHr9DNJyZ489D2wJw97vtHL06tXL6Ppg7n99tuNZr7FEUccYTT9SvRxSMmeX+wr9+OPPxrNz/T+++8n3jPXzJ0713gGuEZ9wQUXGE2vSN269tQ79NBDE9vgGjbPB3ri6BU777zzjKYf6ZBDDjGaPg1mlxxzzDFGM7cqzdc1bNgwo0899VSj6fOhr+u5555LvGc+KC0tNdktrFebNm2MZg4MfRppuTzMdaPfkJ4Tnuc8r/fcc0+j6T9iPg7PH2ZbcXv03ElJLxh9FXGfoJT0q8QzfTgGckWDBg1MLg/9QPQC7rDDDkazj12a1+/LL780mtl1kydPNpp96ngc6R3lseG8+tRTTxk9evRoo5m/tnjxYpENNtjA6Lvuusto5kQxj6mmGgOXlZUZjxe9w8zoYv2oOZalpIeM9WGuF/MO2a+VczfHNuvD+YaeKx57Xuel5Nz6+OOPG835mnPz2WefXfHntLl8efg3UY7jOI7jOFngN1GO4ziO4zhZ4DdRjuM4juM4WVCjnqh58+bp9ddfr9D0nHDdlL1vmF1Dj4qUXIvfeuutjWb/M66Vc32ZeRZcG6aHZf/99zf6H//4h9HsnbX++uuLMNeFP8MsoQ8//NBoZtnkgxYtWuiAAw6o0PR6cc2cngdmLqXlZX388cdGcy2fnjh6N9gzjH4VrsPTc8dzgZ6Z+LksSV27dhW55pprjOZa/8Ybb2w0zw9m9HCbuSKKIjMWfv/9d/M6fTPxTBUp2Xduv/32S2yDHjZm/9C/SF/FjTfeaDQ9DxzL7N/GsUjNrLO0jDlmA9FnwfOc53A8/4bnU66YP3++2Q69P8yF4jjh/EHvmiRts802RjPXifMgfTyc43hcOW/zM/C4MvOJ+5PWa43zKH1b9FjyOC1YsCDxnvmgRYsWZjyxTynnEF5Heaz581JyXuJcyrmW/kP6w5jJRs8Uxxoz35hlxt6LHLuSdOyxxxrNazPrzSyzeH2rk8/n30Q5juM4juNkgd9EOY7jOI7jZIHfRDmO4ziO42RBjXqimjVrZtYpua7JnB6uUbOXFrNKpGS/NHpt6O1gvglzm+jb4tosvSPjx483mj6cc845x+i0z0DPAv0nzH1hrz3+fD4oLS016+L0hrzwwgtG9+zZ0+jWrVsbvfLKKye20adPH6OZ20Nf1bnnnms0zy/6Lti3jrkwzJ5hTyr2m+L7SdKjjz5qdN++fTPuE48jvQr5ok6dOsY7QR8FNevLzKWrr746sQ1msvF40xPFXnuDBw82mr0NWW/2QuOxpueSvTyZTSQl/UF77bWX0cwvom8z7uvkMcwVJSUlxqPC+YJzYDxTSpLuvPNOo9n/UUp64OhXfOaZZ4zmWKXnhjlz7CtIDxS9gfT5VOa5kpIZX/RdsT8fP9MVV1yReM98sGTJEnMt5LHmvHfmmWcazXHFcZMGPU2cv+nn47W7sswm+rZuuukmo5kz+NprrxnNXqxSMv/s1ltvNZr9PJk7F+8dSk9dJvybKMdxHMdxnCzwmyjHcRzHcZws8Jsox3Ecx3GcLKhRT9Svv/5q1rrpB2JWBPOR7r33XqO5ji4l1/uZ/cPeeD169KjW68y7oE+HWVZcu6Vnij2EpKQX5K233jK6X79+Rg8ZMsToww47zOhXX301sY0VpUmTJsYHQ/8Q/S+sHXvlxfvw/UGXLl2Mpt+NfjbmELG/G9ftmVXCdfCLL77YaHrN6Aug90RKZpUwp+z44483mjlRaWv/+aCkpMRsq3379uZ1ZqzQ68M8NfZzlJL1WbhwodE8T3/99Vej6aNgLzyeH8yZYv4R68XtpXmi2MOLcxj9RZwf4udcCCHx/rkiPjboBWGGW5MmTYymB4q+UCnp22R+Ho8Da8UcINaenibO4zNmzDCaPiG+Pz17UtLfGs/WkpK9NceMGWM0ezPmiwYNGphzkV5h9pDkOUcvMf1mUtKzxvmb5wBzvDj26FmjJ+qMM84wmv4z5lKxf+z1118vQk8k7wX4GQ4//HCj48eFeW+Z8G+iHMdxHMdxssBvohzHcRzHcbLAb6Icx3Ecx3GywG+iHMdxHMdxsqBGjeV169Y1IYtsGEozF0Mk2bB00aJFiW3QTEyjLg2yV155pdFs5kuD24MPPmh0vAmvlAxwo2GOQWcM65SSBtgBAwYYzc9NgyRNnflg9uzZGjFiRIVu2bKleX3zzTc3+oQTTjD6qKOOMppGVUnadNNNjWbjzQceeMBoGpcZvkYjL5tm0hzN4NVOnToZTZMxH4yQkoGOPC48X2hWz6f5OE79+vXNgxo0H9eta6cKGkc322wzo9NMt3z4gIbXtdde2+h33nnHaB4L1vuOO+4wmuZ4mt2ffvppo2kkZsCjJB199NFG02zMc4ZNa7fYYouKP3P+yxVLly41Ru+ZM2ea17ldHncGvtJELEkvvfSS0Qy7pDGZx4nbYCgym8k+++yzRvPBBhrT2SA5rXE3689znPMBG1Lz/MoX9erVM8eL8yCvB5xj+PM04UvSc889ZzSva3yAgg9QcS7kWD3iiCOMpumbY53meDaQ5sNVUvJBkvfee89ozvc0mnfo0KHiz3zoKBP+TZTjOI7jOE4W+E2U4ziO4zhOFvhNlOM4juM4ThbUqCdq/vz5mjx5coWurCEgA7ZOO+00o+nDkZLr/fQgsSEwvTp8T3qWuFZKj1XXrl2Npm+Ha8P0zKTtA4Pj2PiWjVy5NpwP6Ltg7ehNo/eHtWfwopT0pNC7cdBBBxnNNfApU6YYTQ8Naxf3q0hJPxyby/LcYZinlGxqyvOJ6/QLFiwwmr6ufLFgwQJzvNdcc03z+g8//GA0PVBXXXWV0aecckpiG/RNsDHqBx98YPSRRx5p9KBBg4xmc+f999/faAY2Tps2zWj6dthAmueoJN1zzz1Gn3zyyUazcepZZ51l9LBhwyr+nObpzAVRFGnJkiUVmmOP/kM2d33xxReN5pwmJccajxUb2DLINO4/kZKeKjYoZu3uuusuoxm0WK9ePaPTGnnTr8rzj74+zlH0L+aLhQsXGk8Sm7XTE8tQ2htvvNFonvdS8hpCHx3HKgNaObfynPr222+N5nWZvi2eT+3atTM6LQyTnlReq3kOMQC2f//+FX/mdTsT/k2U4ziO4zhOFlR6ExVCWCuE8EoI4dMQwichhAHlf98qhPBSCOGL8v8nvxZyCgqvZXHh9SwevJbFhdfzr0NVvokqlXRmFEUbStpW0kkhhA0lDZQ0Poqi9pLGl2unsPFaFhdez+LBa1lceD3/IlTqiYqi6EdJP5b/+bcQwhRJa0jqLqlz+Y8NkzRB0r8yvVdJSYlZZ6angU0Q2ZyXOR5pjTH5Hsx1YoYG1725FszGl8zYYJ7OnDlzjGaTZK5PpzXGpCeK68P0UXGflpcTlctaNmzY0Kyj05/EhqTMfbnllluMTsuJ4jo6s2YuvPBCo5nRc+mllxrNnBfy2WefGU1PDn1eXPc/+OCDE+9J70afPn0y7iN9GvT4xRs557KeTZo0MXlAce+OJPXq1cvoqVOnGt27d2+j6eOTkp4FNnxdffXVjeZYZV4aPSz0GP38888Zt8+xSd8MfX5S0l/E/BrWl96g+HiPNyPPZS0bN25s5s7KGr1zvqAfjo2+paSHhs25K/t5zrPMDWKTXV4L6G8hPDfo2ZGSHkYeJ9aaPh7qOLmsJ+E8xmNBPw/9S6yFJOOhk5K5XHvvvbfRvC4yc4uetOOOO85ozp3McKMfdZdddjF6yJAhIuedd57RQ4cONZrN3Ok/feuttyr+nJbfuDyq5YkKIbSTtIWkdyStVn6iSNJ0Sast5585BYjXsrjwehYPXsviwutZ3FT5JiqE0FTSE5JOi6LIPGoRLfsVLvkSTp8aAAAgAElEQVT4w7J/1z+EMDGEMDHtCSyn5slFLdOejnBqh1zUc+7cuTWwp05l5KKW/BbYqT18ri1+qnQTFUKop2UnwsNRFD1Z/tczQghtyl9vI2lm2r+NomhwFEWdoijqxK9HnZonV7Vs3Lhxzeywk5Fc1ZNRC07Nk6tactnCqR18rv1rUKknKixrgnOfpClRFMUNPaMk9ZV0Tfn/n6nsverXr29yb+j9oQeGWTTMFnnhhRcS22CfJ3o34jlVktS3b1+jmbHB3wKYwfTyyy8bTV8GMzmYf7LXXnuJcD2aN5+zZ882mhfASy65JPGeUm5rWVpaarI9rr76avM6vWn0M9ETwTV0SXr88ceNZq8qepjocWHuE/ttDR48OOM+3XfffUYzd4QXK/ZdlKQ99tjDaH5LwH2M95aUpJEjRybe8w9yWc958+aZPm+HHnqoef3cc881mnlrhx12mNG33XZbYhv0QN10001G0+fA489xQA8LfRqffvqp0czsYrYQM574GSVp/fXXN5reCfoVmX8T74EW93jmspa//fabXnnllQrNHKHffvvNaPato9eQvk8ps9dLSs5RHLsca/TEMOOLHhnu00UXXWQ0+96lrYLEe39KyX5xzA3baKONjKYXKX7Mcz3Xxv1Xo0aNMq9fdtllRrPvHHOl0uap22+/3Wj2iGTf0K+++spo+uboZ2SuGP1k9KxxHHF7Xbp0EeF8Tg91SUmJ0Zwf4p5XepszUZWwzR0kHS7poxDCH07L87TsJHg8hNBP0reSei/n3zuFg9eyuPB6Fg9ey+LC6/kXoSpP570uaXmt5Hdbzt87BYjXsrjwehYPXsviwuv518ETyx3HcRzHcbKgRnvn0Ufzyy+/mNeZ2UQPxCOPPGI0e5lJUr9+/Yxmbkt83VqSNtlkE6O5ls4sqk033dRo5lc8+eSTRjOTg54Y5lhJ0rbbbms01+afecYuo3P9mt4P9j3LBQ0aNDAeNXp5ZsyYYTQzeOhp4Jq3JB111FFG00fB3nj0UdDzMm7cOKOZPdOzZ0+jr7vuusQ+xaFPI813QR8FPTR8nd6xyvJwckXdunWNd4LHmjlf7KXFfmtpvcrY/4q9x5jrRE8SX6fvgvlH9Kwx54t+sx49ehjNYyAlc6Hokbz88suNZnZZPA+NHo1c0bRpU+Mxoe/u7bffNpo+zWuuucZoeg0lqXPnzkYz54ljgV5Uwswe+n74Geh34T4ztyrNQ3PggQcazVrSg8vPtOeeexodz3DLJSUlJea84fxOL1D37t2NjucfScnrriSdeuqpRrMnJD2R9AbS/M5jyd6I33//vdH0Qy9dutRoXus59qXkdY45b7xenHjiicvdZtr8tTz8myjHcRzHcZws8Jsox3Ecx3GcLPCbKMdxHMdxnCwI1Vn7W+GNhfCTlj3WubKk5TceKgwKfR+rs39toyhaJZcb91rmnKruY85rKXk9c4zXsuoU0z7+1etZ6Psn5aGWNXoTVbHRECZGUdSp8p+sPQp9Hwtl/wplPzLh+/jn249MFPo+Fsr+Fcp+ZML38c+3H8uj0PdPys8++nKe4ziO4zhOFvhNlOM4juM4ThbU1k3U4Mp/pNYp9H0slP0rlP3IhO9j1SmU/chEoe9joexfoexHJnwfq06h7MfyKPT9k/Kwj7XiiXIcx3Ecx/mz48t5juM4juM4WVCjN1EhhD1DCJ+HEL4MIQysyW0vjxDCkBDCzBDCx7G/axVCeCmE8EX5/1vW8j6uFUJ4JYTwaQjhkxDCgELYT69nVvvntawihV7L8v3xelaRQq+n17LqFHoty/enRupZYzdRIYQSSbdL2kvShpL6hBA2zPyvaoShkvbE3w2UND6KovaSxpfr2qRU0plRFG0oaVtJJ5Ufu1rbT69n1ngtq85QFXYtJa9ndRiqwq6n17LqDFVh11KqqXpGUVQj/0naTtLYmD5X0rk1tf1K9q2dpI9j+nNJbcr/3EbS57W9j9jfZyTtUZv76fX0WnotvZ7FXE+vZfHUMp/1rMnlvDUkxVs3Tyv/u0JktSiK/mgxP13SarW5M3FCCO0kbSHpHdXufno9VxCvZVYUZC0lr2eWFGQ9vZZZUZC1lPJbTzeWV0K07Ha1IB5hDCE0lfSEpNOiKPo1/loh7WchUyjHyWu54hTScfJ6rjiFcpy8litOIR2nfNezJm+ifpC0VkyvWf53hciMEEIbSSr//8xa3h+FEOpp2YnwcBRFT5b/dW3up9czS7yWK0RB1bJ8P7ye2VNQ9fRarhAFVcvy/ch7PWvyJuo9Se1DCOuEEOpLOljSqBrcfnUYJalv+Z/7atlaaq0RQgiS7pM0JYqiG2Mv1eZ+ej2zwGu5whRMLSWvZw4omHp6LVeYgqmlVIP1rGFjVzdJUyV9Jen82jaale/To5J+lLREy9ab+0lqrWWu/S8kjZPUqpb3cUct+8rxQ0nvl//Xrbb30+vptfwr19LrWVz19FoWTy1rsp6eWO44juM4jpMFbix3HMdxHMfJghW6iSrEJFUne7yexYPXsrjwehYPXsviIuvlvPIk1alaFl41TcsMcH2iKPo0d7vn1BRez+LBa1lceD2LB69l8VF3Bf7t1pK+jKLoa0kKIQyX1F3Sck+GRo0aRc2aNavQDRo0MK/zho568eLFRpeWlia2Ub9+faOXLFnCfcj4Htzm0qVLM77euHHjjNurV6+e0XXr2kPOY5C2T2ThwoVGL1iwwOhlDyX8f37++edZURStkvFNq1nP+vXrR/HPzn1q0aKF0ZXVIX5exPbb6IYNGxrNz/n7778b3bKlbYn0668mIiRRy9mzZxu95pprGv3LL78YvdZaaxnNYyAla/nTTz8Z3aRJE6NLSkoy7uOMGTNyXktJatiwYdS0adMKzfN00aJFRvPYk3nz5lWyi1KbNm0ybuO3334zum3btkbzWLZu3dro7777zujmzZtnfP/11lvP6I8++iixz9wGxzfPQZ7n8bE6f/58LV68OPOBXEa16tmgQQMzNrkPZWVlRtepYxckOJ+suuqqiW3wZziPcSzxPOdx4jjhcebPc5/5/pWNO0lad911M+4zxyLfc/78+fz3eRmbTZs2jVq1alWhOU44t3LeI6ydlJy7OP459/JazOskzzFel3ls+ZkI34/ntJT8DDxHeE7xM8SPy9y5czV//vyqjM0VuolKS1LdJtM/aNasmXr27FmhOWnxZKD+73//azQvspLUrl07o3/4wUZqbL755kZPnz7daF60eDFgsTt16mT0jz/+aPTqq69udHwwSNLf//53kVmzZhnNk2HKlClGf/jhh0bzhL///vu/TWwkSbXq2bhxY+20004V+osvvjCv77fffkbzOG+yySZG77LLLoltDB061OgNN7Qtozgw33jjDaPj55okvfLKK0Zz0D311FNGX3LJJUY/9thjRt90001Gf/bZZyIzZ9oIkrvuusvo7bbbzmjeTHLyuO6663JeS0lq2rSp9t133wq98sorm9e/+eYbozkJ8qaKtZCSY2vgQLuS8fXXXxs9YcIEo++++26jeSyPOOIIo0855RSj99prL6Nfe+01o59++mmjedMmJc+pv/3tb0ZPmjTJ6I022sjoTz75pOLPr776auL9l0O1x2bnzp0r9GabbWZe5y8TnC8+/dRez48//vjENvgzvCEZOXKk0dtsY3f37bffNppz+eGHH270e++9ZzQvottuu23G97v99ttFuI9PPvmk0SuttJLR/CVq4sSJRo8YMSIvY7NVq1Y6++yzKzTHCa9xPEd5/fjggw8S2/jyyy+N5vhv37690dOmTTOav5BQr7322kbzFxrOL2Tu3LlG81ogSVOnTjWaN/ZHHnmk0d9//73R8eMybNiwjPsTJ+/G8hBC/xDCxBDCxLQ7YOfPQ7yWvIt3/nzE65n2LZrz5yFey8p+q3cKn3g9q/KtrlN7rMg3UVVKUo2iaLCkwZK00korRfHf1vlNAn972H777Y3++OOPjY5/E/IH/MaDJyB/g956662N5le6/FaI/56/kX3++edGH3bYYUbzt2v+Vigl75D5W0B82UVKLp3xuN1///2JbaRQaT3jtVx33XWjvn37VrzG38BXWcV+q81vAa6//nqjn3kmmXfG35C5vMLfbvl1LY/jFltskfF1fjvG3zTXWMO2rHrhhReMHj16tAi/iTzkkEOM5rdf/PYj7T2rQLXHZqtWraL4LzlcqnzrrbeM5m+7/C2we/fuiZ3ieXzvvfcazW8j+a3cu+++a/QGG2xg9JgxY4zefffdje7Tp4/R/JbwoosuMrpfv34iO++8s9G8+bzqqquM5jhYZ511Kv7Mb2MyUK2x2apVqyg+duLffknJb9O32morozlW99xzz8QOcd7jNwlbbrml0fzmgMeey3GsJed6jgsuJ1H37t1bpFevXkZffvnlRnM5eP311zeay9EjRoxIbCOFao/Ntm3bRvFr5f/+9z/zs/y2c86cOUbzmyh+cyVJXbt2zfhvuGz9/vvvG73pppsazaVOriJRc158/PHHjeY3vmkrF6w55zDePxB+5qqyIt9E/ZmSVJ3K8XoWD17L4sLrWTx4LYuMrL+JiqKoNIRwsqSxkkokDYmi6JNK/plToHg9iwevZXHh9SwevJbFx4os5ymKohckvVDpDzp/CryexYPXsrjwehYPXsviYoVuoqpLy5YtjZeCfiN6YMaPH2804wTSHPpcO6dHip4lPspJ8zufIuA+8+m8+FMxUnJ9mk94pH0GPtnAJyHom6HJ+4knnki8Zz6IP5HFp2Xo2zrwwAONHjBggNGsiyTts88+Rt9yyy1G05/WsWNHo/lEDz1R9AHwM/CJng4dOhj93HPPGU1viZT0enBt/6STTjL6gQceMPr00083+sILL0xsIxc0btzYPLlKz0ncyyMlxwE9MvSCScka09vBJ1uPPfZYo9955x2jK3sSrlu3bkZfccUVRvOJJI6rHj16iPAJzH//+99G84lA+rbiXo98mflLS0tNfbp06WJe55NOPA6c84466qjENnbYYQejOUfxMfvXX389sY+Z9mnHHXc0mj4ujgM+TcX942eUpIMPPthoRprQY0PfZlo8TT5YsmSJ8THRyzVo0CCj+WTsueeeazS9oFJyPPPJePoZOa+NGzfOaNaf8x6PNZ9U5XWZT19yvpGS9eI5xyfheY3iv68q3vbFcRzHcRwnC/wmynEcx3EcJwv8JspxHMdxHCcLatQT1bBhQ+NbYHIqPSfMZGHCMNfhpeRaOz0JzK+gB4rrzXfeeafR9ESwnQAzm5jJcfLJJxud9hm4fsz34DaY0k6fFpOYc0GdOnWMJ4B+lttuu81o+lt43OhfkpI+Bya105PA1PT+/fsbHc+1kpJJzsyZ4vszXZz+N+bISNJ1111nNLOohg8fbjSzjP7zn/8k3jMfLFmyRDNmzKjQzIU65phjjGaaPD0LPJZScmzec889RtPDwEw1Jp5znDCrih4nZsDFE9ql5LEeNSr55DnPoZdfftlo+omYNXbmmWdW/JlzQa5o3ry59t577wrN48JsPPpXmOzPvB0p6WdjphL9ZPQPMuGc3SuYr8XPcMcddxjNvD1+prS2L/TQcb5hyjbHbk0FDpeWlhovLc85zvfM+eLYZN6dlDy+9EQxb5HvwVwwHm/OpXyd1w9etzlP0pMlVZ6KTk91fL6TrC+sstZrcfybKMdxHMdxnCzwmyjHcRzHcZws8Jsox3Ecx3GcLKhRT9T06dN1zTXXVOh//vOf5nVmR7CXFfMt6DeRkr4ZrtUzl4lrwfTRsPMzewbRx8U8jbFjxxrN/jxp+Uj0fnDtnj3kmLHBbvb5IIRgeh/SW8ZeSFxj55o4+/1J0r/+9S+jeRy4bk2vB7fJnCj6fg444ACjma/Dzubs9cj3S4O+G/qq2E+NPcbyxeLFi834Ou2008zr/Ozsp8bzOO0cpO/t6quvNppjkd4OZtnQo0KfDvtv8VjTU8XX085J7tM555xjNI8TvT/xPpb8PLmiXr16Zj94nrJvGb2B9JKl+TbpL4tnjElJfxEzeOiROvroo41mBhz7rJ544olG33fffUY/+eSTRqf5FemjYk5hPAdPkn7//Xej07xF+SJ+/Ojt4WejR5YZb2kZbquttprRL774otG8rtF/ykxHegd57abm3M73Z28+jjMp6VHjOcfjQA9m3KPI/LdM+DdRjuM4juM4WeA3UY7jOI7jOFngN1GO4ziO4zhZUKOeqLKyMrMOzWwI9sJizgPXQdMydOirOv74442++OKLjeb6P3syMaOJPgnmm7z55ptGf/vtt0az5w/7w0nSyJEjjf7kE9vkm74Zene4vp0P6KGh14vZJGeddZbR9JbUq1cvsQ2ugzPXg96xjz/+2GhmFdG3QU8M1+mZv0LvGeG5J0nXXnut0fTI8TPRY8NtPv/88xn3IVtKSkq00korVWgeO26X/hH6Zpg1lPaerC9zm+gn4tjiOGA20ZVXXmn0pZdeajTPWfo60npQ0sdJ7wS9I5zD4vMLvY25YtasWSbHi9ljnJPYk5Ck9YhjT1B6T9n7jj4c+nbogWKvO879N910k9Ecq8zSo19OSvbjo+Y5zXk1rX9bPmjevLmZL9nTk58tnkUmJf1pafXkZ/3HP/5hNM9zZjgyR4pjk/5Fzh/0k9K7zJ/n+Ze2j5xfmC3FbcRzKNm/NhP+TZTjOI7jOE4W+E2U4ziO4zhOFvhNlOM4juM4Thb4TZTjOI7jOE4W1KixfPXVVzfmTjZSZFgigyxpLG3UqFFiGwxxpNmY5uKtttrK6HijR0n65ptvjD744IONvuuuu4z+6quvjKYZkeGJNJFLSdPcfvvtZzQN1Qx9q05QWLbMnz/fPAhAcyOPO0NMaeJ+5plnEttYddVVjWYj1NGjRxvNppXvvvuu0az9xhtvbDQfdBgzZozRND4z7DPN7Pjwww9n3AeGlA4cONBohl7mi5KSEmOCZjAmgwcZjscgO5q0pWQoIB8+mDx5stEM8KQhlubXDz/80GiavGksp9Gdeu7cuSI0yPLBkscee8xoBiPGDbzTp09PvH8uaNKkiTFa00DPOY5zEuestHBGBhnyc9Jozgcq9t9/f6NvueUWo3ktoAmYYZt8gIhzAa8lUnJ8M2y3a9euRtOgXbduzVw+586da45f9+7dzevz5883evDgwUa3atXK6LQHCSqbOzt27Gg0r3Ns+Lxw4UKjOXYZ1vvGG28YzbHL84lBypJ0+OGHG83rQ/zBmTTiD1xUp7m0fxPlOI7jOI6TBX4T5TiO4ziOkwV+E+U4juM4jpMFNeqJWrhwoT777LMKXVZWZl4fMWKE0dtss43RO+20k9HNmjVLbIM+ihdeeMForv0yyJIBXVxHpfenadOmRnPtn02U2QSXa8dSskEoG4iyCTIbtzJ8Mx/UrVvXeJa4Zv7AAw8YvcsuuxjNoDoeNym57s0mpfQLMbyQHgY2xWSTS56P9BLQB8I1d54LUtLH9cEHHxh92GGHGU0vEtf580VZWZnxdDGElv4deqToy+O/l5JhhvTREPpgGLj44IMPZvx5+jS4T7vvvrvRHEdpDYI5f/Az0XsTn+8k62ekpy5X1KtXz/iY6F+hl5ShpE899ZTREydOTGyDx5Z+N4bn0ndFn9Zxxx1nNK8F9K5yvmGjeDYsTvMr0ofF+eP88883mh6oU045xegbbrghsY1cUKdOHRMU+fXXX5vXeV4fc8wxRvOaOHz48MQ26AfcddddjWazZs6NvCaxqTWhl5ghtnw/htbymiglAzK5D7yuMpw37l+l7zgT/k2U4ziO4zhOFvhNlOM4juM4Thb4TZTjOI7jOE4W1Kgn6ueff9b9999foZnzwtwYeqLYPJjr5FIy44SeJvpe6HHg+j/Xm9kwlmvtXEs+6KCDjGY+EtflpWRTYq4H08vBXKi05q+5JoRgPAL0/vTu3dtoZoLxMz399NOJbTz33HNGr7322kbHzyUp6Yvo1q2b0Vz3p6+HOUP0ipxxxhlGH3LIIUbzfJWkCRMmGM2m2awdfV/MwznvvPMS28gFLVu2VK9evSo0z8u9997baGZqsaktPStS0svBY0PfxYEHHmg0GxJvueWWRtOTxp9nhtytt95qNH0+9GlIyaw6esXo0+R4j+9jdbJoqgObg9PLw8/AZrI8TmkNa+nt5NihZ4a5YvzszDpihhvHCfWxxx5r9LPPPms0zz0p6bHl3M/rEY9LZY2bc8X8+fONV5LeQPooWW9eo5jRJSU9QFOmTDGaXmHmwtF/yvmC103mgPH8oWeOtYo32P4DXlPoaaT/lNuMj002xM6EfxPlOI7jOI6TBX4T5TiO4ziOkwV+E+U4juM4jpMFNeqJat68uekDt9tuu5nXX3rpJaPpSWGfqrTeRey5w3VPeqa+/PJLo5ln0qJFi4zvz3XyAQMGGM2MlkGDBhndsGFDEWacMOeDfhX266K/gLkvuWDevHnGW8FcD3rBmE3Trl07o5kblfYz9F2xhxTX/s855xyjH330UaOZ00NP1Q8//GD0DjvsYDS9JWlZNOy/xn1ijhl9GGk9BfPBjBkzdOONN1Zoelo4Tthbi2P5kUceSWyDvcnoT6TPZt68eUafeuqpRjO3i/MBs2Toc+BYZq3S6sl9juf3SElvCX2ccd/FWWedlXj/XNC4cWPjN6U3hP4k5p1xXNCfIiXnNfpROOfQ37bKKqsYzcw2Zohx7LMOzB3i6/369RNhD0H2COTnXnPNNY1+/fXXE++ZD+rWrWuuWxwXzEfj9YJ+JF4DpeRY4nWPx4I5Xszs4znHHrSsD3uKsqcsz1l6RSVp1KhRRjNjr7LMyPhxTLsuLw//JspxHMdxHCcL/CbKcRzHcRwnCyq9iQohDAkhzAwhfBz7u1YhhJdCCF+U/79lpvdwCgevZ/HgtSwuvJ7Fg9fyr0NVPFFDJd0mKd4MbaCk8VEUXRNCGFiu/1XZG5WVlWn27NkVmmuUL774otEDBw40mmvBt99+e2Ib8X5uUjLjhOvgf//7341mLz3mG3Xq1MloZl1tsskmRk+aNMloelzoqZKSWULMVLn++uuNZj8/+sDAUOWgns2bNzc5OjxO9B9xTZ0ZTs8//3ymzUlKZu7Qs9SmTRuj6QWg74qeK2bJMKuGOVD0haT1QuM+sLb0HhB69MBQ5WhsNm7c2PgS6FmgX4ieN3qm7rzzzsQ2Hn/8caOZqUZ/ET2RzGxjPg7H5t/+9jejmb/G/m70XL322msizCPieOfYZO+8eP9G9t1TjupZVlZm5jmOPW6Xfpg+ffoYzTlVkh566CGjd9xxR6O5TeYM0UNJXw8z4TjPsm8d95GeK54bknTCCScYzZwyei453pl1hFy7ocrhdTM+/nhe08vFPobMMmOPOSnpcWRGGuda5gCOGzfOaNb71VdfNZrZZBwn9E/SL5k2v3Tt2tVo9inlnMZemfHXq5PhVuk3UVEUvSbpF/x1d0l/pGsNk9RDzp8Cr2fx4LUsLryexYPX8q9Dtp6o1aIo+iOyeLqkpN2/nBBC/xDCxBDCxAULFmS5OSfPVKme8VryW0GnYMhqbPKJTqdg8LFZPGQ1NvOVbO/khhU2lkfLnumNMrw+OIqiTlEUdeJj7k7hkame8VqyxYZTeFRnbPKRcqfw8LFZPFRnbDLyxiksss2JmhFCaBNF0Y8hhDaSMho3/qBRo0Ym7+GJJ54wr9PjNHbsWKPpH0rzRA0fPtxoZsXwN+4lS5YYzZ5ACxcuNPrJJ580mlkT9HHRp8O+R/SBSclMFWZwcM2bPcIq89mkUO16rrrqqiYTi9kk/Nz0D/BzM1dESvZGZA4QPTHjx483mtki9NywvxN9P8w2Ya893nik5cbsscceRt99991Gx3PTpGReDv0qVSCrsdm6dWuTk8WMFdaP/hHmqrCvoZT0SbBXGT87/Yr02V144YVGs59ajx52tYTnB8c25wqOOyn5Odk/jX3J9txzT6PjXqC0/KUUql3P7777TieddFKF7tmzp3md5yBrxby+tB6lnCfpmaEnafLkyUZzVYJjlb006UdirhD9iJwr2CdRSs5R9N0wF44eG3omq0BWY7Nx48amTyT9RJyHeJ2srN+rlMynY+YSM/TogeXxveaaa4w++OCDjeaNPv1KnPeuvfZao+mhk5KeV2Y2MtuK39jGvcQpfsXlku03UaMk/eGQ7CupZhIBnXzh9SwevJbFhdezePBaFiFViTh4VNJbktYPIUwLIfSTdI2kPUIIX0javVw7fwK8nsWD17K48HoWD17Lvw6VLudFUdRnOS/ttpy/dwoYr2fx4LUsLryexYPX8q9DjfbOq1u3rlkrP+SQQ8zr9DQw44lr1Fxnl6T+/fsbTe8NM5VatrR5Z9OmTTOauVLMFuL7ca32v//9r9Hs8ZT25AW9OnF/Q9o22Qspradgrpk+fbpZ92YWSYcOHYyO92WTkj3G6B1Je096Zjp27Gg0vQHM+aHHhtlG9MfR/8JaffTRR0Yzd0aS3njjDaMvvfRSo9kP8r777jOaeSzDhg1TPpgzZ47pd8beVfSgPPDAA0bTZ8McMKny8c3jzRwnjgtmy3Bs3nbbbUbTA8PzhTlW9D+mQV8VvX3nn3++0WussUbFn3///fdK3z8b2rZta/xi7Pl2wQUXGL3ZZpsZTa8hfZxSMr+MY5MPEdGnw/FODwp7N3777bcZ9/nHH380Op5HKFUt94fzJjMDef6wn1++aNSokfHe8rPRW7zTTjsZzXGR5ldkBh8z09gLk2N3zJgxRtMjy/dnJh/PsdatWxvN8yOttx33iX1KmzVrllHHs6mYY5YJb/viOI7jOI6TBX4T5TiO4ziOkwV+E+U4juM4jpMFNeqJmjNnjsmfaN68ecaf32677Yymn+TNN99M/Bv+zPHHH280s2q4lltZTxc9a9oAACAASURBVCf27+JaP7NDmIfCHk705UjJXkj0H3B9mHk71cm4yJYGDRoYj9GECRPM61OmTDGava3od+HPS9Kzzz5rNHtZrbTSSkYzX4tZJsy2iWevSEkPFveJ/bvoLbnkkktEWH/6+OgloXeMPazyRaNGjUwfuPPOO8+8To/CHXfcYTT9JMwJkpIeEuYydenSxWj21uQ4OOqoo4ymx4k5MOw5xmwavh/9aVLyc7/11ltGM/eLnrbvvvuu4s/5ClGsW7eu8RSxv9+DDz5oNHODmDOX5t1iLekvYa04D/O8Z79P5j7xMzDDjTlmnBPTPDSci0ePHm00s6/ifjYpObbzRVlZmcllo7+Ln50+XHo9mdkkJXt6Mv9st92sH57nzNChQ43mOUS9yy67GM368zOy72Var1XOWcz9Yz3ph45nkdHrmAn/JspxHMdxHCcL/CbKcRzHcRwnC/wmynEcx3EcJwtq1BO1yiqrmCydW265xbzOtVv6k9jbiDkeUnLtlVkx9NXQy0NfBvNIuPZLHw57DtH7w/yM3377TYQ5H+wjRO/NPffcY3RJSUniPXPNzz//bLKCmMHELCuugdNTkdZ5nv2yuNbPfCyu0zOrhv3e6JvgGjk/A9fpX3nlFaNPPPFEEfqqmC1FL0A8q0SSLr/8cqPZmy1XzJw504xH+izYI5K5L+xJyZ+Xkj3ZeDzpgeLYYm8sjgNqfobBgwcbfdxxxxndq1evjO8nJT2V7CvHOYlZU88991zFn5nvkytKS0vNWGDPQWYAEY7FtIbG9IbyvGRGG31VnKP4fhx7nAOfecZ2TGGOFDOd2IdVsv40KTmH0TPF8/Whhx5KvGc+aNKkibbZZpsKHc9zk5LzGHur8hrGOUdKesyY48V56+OPPzaaWWT0n1a2j/RH8hpIDx3zIKXk/EGPK6+19DMee+yxFX9m5lQm/Jsox3Ecx3GcLPCbKMdxHMdxnCzwmyjHcRzHcZwsCOxhlk9WXnnlaN99963Qe+yxh3mdfeu4Dk7/El+XkhlL7MnDtVuuxdKXNWPGDKO5br7RRhsZzbV+9hyij4deAGnZGngc5rpcddVVRtO7wUyNs846a1IURXaBfwVp2bJlFM8OGTBggHm9styR7bff3mjmSEnJvof8XCeffLLR7K1ID028n5iU9O2wpxR9GfTTMSOM/bukpB+B+Ukcf/QrMNtk0KBBOa+lJK2++upR3CPEzCVmZLHPHPORmNcmJTPX6IugD4E5YOxDyH5+9FCy/vQzXnnllRn3J63f2quvvmo0zyF6c+izi3tbbrjhBn3//fchsZEVZN11142uvvrqCs28Nc5xnIOY+ZQ2z06dOtVonh98D2Ys8Tx/++23jaZnctdddzX69ddfN5p5auzTuPPOO4u8++67RnNsstbrrruu0Tz/DjjggLyMzWbNmkXxbXXu3Nm8zkwjjiPWL+2aTx8cc9+YPchziO/JuTN+3ZeSY5PXVebSMVMtba6tLBeKuZPMmYzP56+++qrmzJlTpbHp30Q5juM4juNkgd9EOY7jOI7jZIHfRDmO4ziO42RBjeZE1atXz+R30G/EdU76m9hrr1u3boltcF2b2RBc5+Z7cu2cuS/MH2H2DdeS6S354IMPjO7atavI8OHDjWbGEjNSmKlB31U+WLp0qcl+Yd8pZjoxT4eZThdffHFiG/SvrbfeekazHxf7ENLDxNwQrsvTs8NsG37G9957z2iu+0vSZ599lnEfWeuHH37Y6HiuWj4pKyszNWE2EDOx6LOg34x9CaWk55HZUu3btzea+0AfDn0U9DfSI0UPHX08rBXnJ0kaNGiQ0fRVsHcmvT3xfJ00r1EuWLBggZm3uE/sI0f/Up069ndrfgYpWV96P5mn9csvvxhNj0vHjh2Npq/n6aefNprHjplN7JPIzDkp6WnktYJ+V2a+pb1nPmjRooXJTJs9e7Z5ndcw+sFYq6+//jqxDXqc6LvidZWZe/Qejxs3zmhmMjFHivvEccXrZFqvVR4H+lHjPkEpOR/Es7CY75gJ/ybKcRzHcRwnC/wmynEcx3EcJwv8JspxHMdxHCcLatQT1bx5c9OThz2d6JPhuvbee+9tNPszSck8EXqYuJY+cOBAo5mpwn1gDyDmYbD3Htdu+/TpYzTzTqRkzy9mLtEbRL/B2muvnXjPXNOwYUPjY+BxJlxTJ1znl5JesLvvvttoZptcdNFFRnfo0MFo9hhkf6h47yRJOuKII4ymr4s5McwlkaQdd9zR6NNPP91o+hXY3409yZinkyuWLFli9p/b4djbZJNNjOZ5zVpIyUwl1m/SpElGs3deaWmp0ez3NXnyZKOZC8N603dx3nnnGU0fjiRdcMEFRsdzn6Skt4Rej3g989V7rUmTJuZcZd4ac+jYz5O5Q/RISckcH441zgfMy6J++eWXjeb5RX8S50361zjPs9+clPSrMquuXbt2Gd+TOWb5okGDBibLi+OGXiDOGfyc9KtJyWsxxxY9TPSD0evLvMTDDz/caPqT6H+NZxBKyRyq0047TYR5ivQrc77mPsfrT29zJvybKMdxHMdxnCzwmyjHcRzHcZws8Jsox3Ecx3GcLPCbKMdxHMdxnCyoUWP5zJkzdeutt1boTp1sr8Z4EJ2UNLcxDC3N2Mf3oHmsX79+RjM8j9vk+x1yyCFG8zPQfDhz5kyjaYhkUKaUNLTS5MYwzVatWhn93HPPJd4z14QQzH7xc5100klGjxkzxmiG6aUZBVu0aGF0PAxNSpq2aeL/7rvvMv48zY80GrOZ7M0332w0jc8HHXSQyG233WZ0CLanJRun8jPWRHCqtMwoHDfS8wELNv9lgCNN/wxClZIhsTyvOTaff/55oxmAx4bBZ511ltE0y7KBOQ3WNDdzf6VkjRkYykBXBrbeddddFX9m2GOumD9/vgka5XFmw2Geg8OGDat0G5zXeJ7edNNNRvfv39/oRo0aGc2gUz6wQ+MxDftsgM1QW4Z7SskwZz6ww/ow/JX/Pl/8/vvvJtiX8yKvOXzgig9YpDXW5rzEB4EY4MkG8gzfZKArH7o59dRTjeZ1luOKnyE+jv6ABntem1k/NlWOH8e0hymWh38T5TiO4ziOkwV+E+U4juM4jpMFfhPlOI7jOI6TBTXqiVq6dKlZj2WIJIPr2JCYHigGdknSYYcdZjTXud955x2jGeDH0Ed6oGbNmmU0g80YGHjmmWcaPXbsWKMZ5iklAzs//PBDo+mzqMzjcP311ye2saI0b97cbIdr3jwODPCjRyGt4SMb0jKokh4ohhcyfI2+jTvvvNNoHlc2B6a3hJ6ptFrSv8BQyrPPPttofgZ6Q/LFokWLTNNo+sXYgJieKDYITQurY3gqfQ/jx483+osvvjB6wIABRvP4X3PNNUbz/OE5eOihhxrNhtIM55SSn4teEAYG8zyPHzc2Q88VZWVlmjt3boWmN5DnKb1+9LPQSyhJ33//vdHx7UlJzws9TO+++67RCxYsMJrjhg1u+f707PHf0x8jJT009MAdfPDBRrPW1fHNrAhlZWXGD0xvD3149JfyHGRQqpScf+lRYwNhetT22WefjD/PoGOeUww25bWf1820Y89rEMcixz8DXeMNi+nZzYR/E+U4juM4jpMFfhPlOI7jOI6TBZXeRIUQ1gohvBJC+DSE8EkIYUD537cKIbwUQvii/P8tK3svp3bxWhYXXs/iwWtZXHg9/zpUxRNVKunMKIr+E0JoJmlSCOElSUdKGh9F0TUhhIGSBkr6V6Y3YmNM+pHoYeAaNJse0iMjJZvx0jfBdfCysjKjt9hiC6PZzJf+ATYk5vuzaS1h3ookTZ061Wg2TuSaNjM8tt122+VtLme1nDZtms4555wKTT/bXnvtZfSQIUOMpvcsrQExvV30anCN+4YbbjCa5ws9KGw4yvOxWbNmRh955JFGM0fm008/FWH+Cr0C559/fsbXK1mbz1k9Z82apXvvvbdC0wvInK/HH3/caI5dHnspmX/Gpsb0f40bN85oegWZd0NfBs/J0aNHG00P1lprrWU0zwcpOR/Q68M5i76aeIYbPHg5q2WjRo1Mw15mJtFTQz8R88/ouZSSuT7MFePczEweHjfmAMX9KVLSr8gG2fTb0UNDj42UzKpik1x6bPbYYw+j2bgZ5KyeURSZuY6NvEeOHGk05xAee/rLpOR4ZV4dPY8ce8zY43xATxo9TZwneR2nF7Eq/tPu3bsbzSbV3Mf777+/4s/0+GWi0m+ioij6MYqi/5T/+TdJUyStIam7pD9S2YZJ6lHlrTq1gteyuPB6Fg9ey+LC6/nXoVqeqBBCO0lbSHpH0mpRFP1xez9dUmq0cgihfwhhYghhIn+bcGqPFa1l2m/pTu3h9SweVrSW7Ozg1C4rWk9+4+4UFlW+iQohNJX0hKTToigy31lHURRJitL+XRRFg6Mo6hRFUSe2WXBqh1zUMu0Rdqd28HoWD7moJZehndojF/WkjcEpLKo0c4YQ6mnZifBwFEVPlv/1jBBCmyiKfgwhtJGUNPeAKIqMB4k5Dlx779ixo9H0ZdB/JCXXMulbYa86rnuz7xw9TfREsd8W/QfMlWJu1bx580SYnxPP70nbR3pF0rw5f5CrWrZo0cJ4J5jzwuySCy+80Gh+xjQPDY/tFVdcYTS9HLwRYCbYuuuua3Tv3r2Nvvjii41m7S+55BKje/bsaTQ/s5T0n3Cb9AHR6xH3tqSRq3o2a9bMnEfM8WFmE7OH6C/gv5eS45vev2uvvdbo4447zuh11lnHaHoDd9hhB6OZVUafBbOwJk+ebHSad4THgdtglhXPibgPj9k8uaol+1rSb8JfZk888USjuV/0kknJzD5m7owaNcpoep7oR2LuFOeDeC/AtH38+eefM74/510pWbunnnrKaPZypAeK8xPJZT3jn5eZS/RI8ZtIHjvOxVJyPPP40VfF6xrh2FljjTWM5rjgseV1e9n95v8nzafHuZU5gZxb6ZmK52nltHdeWDYC75M0JYqiG2MvjZLUt/zPfSUlu3U6BYXXsrjwehYPXsviwuv516Eq30TtIOlwSR+FEP64pT1P0jWSHg8h9JP0raTey/n3TuHgtSwuvJ7Fg9eyuPB6/kWo9CYqiqLXJYXlvLxbbnfHySdey+LC61k8eC2LC6/nX4cadZMuWrTIrHVzXXPChAlGc12UhklmsqT9HX0zm266qdH0D7GXGbNk6KPo0KGD0fRtMR+DeSr0kkjJvBv6rAYOHGj0ddddZ3QmT1SuWLx4sb755psKTW8Xn8Tceeedjb788suNTstDoueFPaGY6UV/EY8tz58bb7zRaNaKOUVt2rQxmj6CtHV0ZlMxi4jeEj4ll5Yjlg8aNGhgPEf0UdAjQY/DlltuafRLL72U2AazwOhxYu8yzg/MpqIP54477jCaPj164nj+sF8YfRhSsv8i84zo26MnLp4zRZ9Jrli6dKl57x497FP08XErJT8TjyPnGynpN+R78HPTp7N06VKjOQ46d+5s9Pbbb2/0oEGDjN51112NZlYSrwOSdPTRRxvNscoMQXrJ0ubufFBaWmr8fJWNRZ73++23n9G8nkhJjxKzwpi3yM/esqXNDGUmHHvtcex/8sknRvMz0f+c1v9vxIgRRjNfkR5s+rri5xjn/kx42xfHcRzHcZws8Jsox3Ecx3GcLPCbKMdxHMdxnCwIaev+edtYCD9p2RMJK0vKHDRR+xT6PlZn/9pGUbRKLjfutcw5Vd3HnNdS8nrmGK9l1Smmffyr17PQ90/KQy1r9CaqYqMhTIyiqFPlP1l7FPo+Fsr+Fcp+ZML38c+3H5ko9H0slP0rlP3IhO/jn28/lkeh75+Un3305TzHcRzHcZws8Jsox3Ecx3GcLKitm6jBtbTd6lDo+1go+1co+5EJ38eqUyj7kYlC38dC2b9C2Y9M+D5WnULZj+VR6Psn5WEfa8UT5TiO4ziO82fHl/Mcx3Ecx3GywG+iHMdxHMdxsqBGb6JCCHuGED4PIXwZQkg2ZKoFQghDQggzQwgfx/6uVQjhpRDCF+X/b5npPWpgH9cKIbwSQvg0hPBJCGFAIeyn1zOr/fNaVpFCr2X5/ng9q0ih19NrWXUKvZbl+1Mj9ayxm6gQQomk2yXtJWlDSX1CCBvW1PYzMFTSnvi7gZLGR1HUXtL4cl2blEo6M4qiDSVtK+mk8mNXa/vp9cwar2XVGarCrqXk9awOQ1XY9fRaVp2hKuxaSjVVzyiKauQ/SdtJGhvT50o6t6a2X8m+tZP0cUx/LqlN+Z/bSPq8tvcR+/uMpD1qcz+9nl5Lr6XXs5jr6bUsnlrms541uZy3hqTvY3pa+d8VIqtFUfRj+Z+nS1qtNncmTgihnaQtJL2j2t1Pr+cK4rXMioKspeT1zJKCrKfXMisKspZSfuvpxvJKiJbdrhZEDkQIoamkJySdFkXRr/HXCmk/C5lCOU5eyxWnkI6T13PFKZTj5LVccQrpOOW7njV5E/WDpLVies3yvytEZoQQ2khS+f9n1vL+KIRQT8tOhIejKHqy/K9rcz+9nlnitVwhCqqW5fvh9cyegqqn13KFKKhalu9H3utZkzdR70lqH0JYJ4RQX9LBkkbV4ParwyhJfcv/3FfL1lJrjRBCkHSfpClRFN0Ye6k299PrmQVeyxWmYGopeT1zQMHU02u5whRMLaUarGcNG7u6SZoq6StJ59e20ax8nx6V9KOkJVq23txPUmstc+1/IWmcpFa1vI87atlXjh9Ker/8v261vZ9eT6/lX7mWXs/iqqfXsnhqWZP19LYvjuM4juM4WeDGcsdxHMdxnCxYoZuoQkxSdbLH61k8eC2LC69n8eC1LC6yXs4rT1KdqmXhVdO0zADXJ4qiT5f3b5o0aRK1aNGiQv/+++/m9fhrkvTrr+ZpRDVu3NjoVq1aJbbx3XffZdzvVVZZxehZs2YZvdJKKxnNfVy8eLHRTZo0yfjzq6++utH/+9//jG7dunViH7lPJSUlRjds2NDolVde2egff/zR6F9++WVWFEX2g4Pq1rNBgwZR/LPXr1/fvM7j1KxZM6MXLFhg9OzZsxPb4LHhcSgrKzN66dKlabu63H3k+82dOzfj+y9atMhoHvdlPkYL32PevHlGb7DBBkZ/9NFHRvMcnz59es5rKUmNGzc2Y5OfpbS01Gieg82bNze6Tp3k72fffvut0fXq1TOacxHr06hRI6MXLlxoNM8p/vvK3o+fsWnTpiIc34T1Zf3in3HWrFn67bffkicNqG4969evH8U/W8uWtqsF5xfWksclDR57jh2OXZ4P/Pc8F6j5/quuuqrRPO6cx/mZ07bBc57/prLPlK+x2axZsyg+18yfP9+8znmPY5HnNa+jkrRkyRKjOVZ+/vlno3l8Z860D7jx2FZ2XeX5wGPNz8DPKCXrxc9Zt27djDp+js2ZM0fz58+vdGxKUt3Kf2S5bC3pyyiKvpakEMJwSd0lLfdkaNGihU444YQK/e6775rXu3fvbvS4ceOM3nzzzY0+5JBDEts4+eSTjebE3L9/f6OHDh1q9B577GH0O++8Y/R///tfo7fffnuj33zzTaMvu+wyoy+55BKjjzjiCJH777/faJ4wHTp0MLpfv35GX3755UY//PDD9uqVTrXq2aRJE+2+++4Veu211zav84IZ/1kpebMwcuTIxDYOO+wwo3kcOLFycuEkt+aaaxrNi8vo0aON5sTx9ddfG81zKe0mir8IvPHGG0a/9dZbRrdt29boPn36GH3ttdfmvJbSsrF5zDHHVGheWDlBtW/f3uh9993XaP57STr++OON5i80HKusd8eOHY3+8ssvjf7444+N5o079cYbb2z0nDlzjN52221FJk6cmHGfX3/9daMPOuggo+M31ZwLMlCtejZq1Ejbbbddhe7Vq5d5/d577zV6/fXXN3qTTTYxOu0X7c8//9zo559/3uijjz7aaJ4PU6dONbpNmzZG8yaJ73/aaacZzXHVpUsXoznPS9Jqq9mMRV74hwwZYjSvN7zZvPrqq/MyNldeeWVzrnzwwQfmdd5AduvWzWiO3X/84x+Jbfzwg01N4NgbNmyY0XvuaTu+3HzzzUaznryuTpo0yegpU6YYzesib9L22WcfEV43t9hiC6P5JQ3PsTFjxlT8+e677068//JYkeW8KiWphhD6hxAmhhAmVvZbnFOrVFrPeC35rYxTUFR7bPIG1CkoqjU2+S2wU1BUe2z+9ttvNbZzTvXJu7E8iqLBURR1iqKoE5e+nD8X8Vo2aNCgtnfHWUHi9Uz7it/58xCvJZetnT8f8XryG1SnsFiR5bxqJ6nWrVvX+Ej22msv8zqXT84991yjTzzxRKN/+eWXxDa22moro9dYw97k82tkrr3ST0RPy5Zbbmk0v2Y8//zzjX7//feN5leIEyZMEIl/FS9JG25om3a/9NJLRo8dO9ZoLlNUkWrVs6SkxCyH8StlLpV98803RnMdn1+9Ssklo/HjxxvN5Re+J5d/6Lvi0iyX+/h1f6dOnYzm8l3ab4ysBZeguPyz6aabGs1lkypS7bFZUlJivsL/4osvzOsbbbSR0Z9+alcfRowYYXTa1+1czuMSzYwZM4xed911jR41ymYMcrl2s802M3q99dYzumfPnkZz7F555ZVGv/LKKyKcX+jV4PzApbMdd9yx4s/0gWSgWvWMosh4EjmncQ7inPXwww8bneYN4+e+9tprjeZy/X/+8x+jv//+e6N33nlnozlPc/mIc//f/vY3o7m0y+UlKVkr2iRYH87LWf7iUe2xGUIwHqW///3v5nUeG9pg6MN94YUXEtvgZ+VcyXPi8ccfN5p2DV43OZ9wWZzzP1+nf2n69OkiXHbmOcS5dfjw4UZ/8sknFX/m9SwTK/JN1J8pSdWpHK9n8eC1LC68nsWD17LIyPqbqCiKSkMIJ0saK6lE0pAoij6p5J85BYrXs3jwWhYXXs/iwWtZfKzIcp6iKHpBUvK7QedPidezePBaFhdez+LBa1lcrNBNVHUpKSkx6+tcq73nnnuM5mPT9FnQ7yQl1+q5FssnBOnD2GGHHYzu2rWr0XzU8+WXXzaahutp06YZzbX5tLX6nXbayWh6geK+Ckl67bXXjF5rrbWUb0pKSkz2B71lzFCi/4SP/nPNW0p6jrbeemujGTlAjwz9aNwGH2Gnx+G5554zmh4d6osvvliEPirWlrWjD6uy7KtcsXDhQuO/4nnJsfrVV18Zzc/OYy8lvXzMpqFPhseO8Q8cu3vvvbfRAwfaHEN6LuMeCCk5V9BnI0kffvhhxm3Q48ZtxjV9I7miQYMGxt/DOY8xLPTD0Jie9uQmH+9/9NFHjaYnivE19KcwT4vzOJ84TPPDxuFnou9Lki699FKjOZfzWsAIhBdffDHjPuSKkpIS83g+PU30TdL7R69mmneTsRbMiaKXk+OgR48eRnO+oP/xwAMPNJpjn3EEPPZPPfWUyH777Wc0vcKcPxgvE4+s4fyWCW/74jiO4ziOkwV+E+U4juM4jpMFfhPlOI7jOI6TBTXqiSorKzPr6/TFMIvirrvuMprrtGl98uhj4Dr3scceazTzJuhhYn4JWyT07dvXaHpcmCtDGLcvJTMq2B7j1Vdfzfie9Ezlg19//dV4XBjrz2wa+rTouaEXSErmndADR08JfRDMaGJWFb0FzKVK26c4TzzxhNFsSyQlPze9H2x9xEyeyZMnZ9yHXNG4cWPTDoLHlm0Q+DkeeeQRo9Na4NDTeOGFFxo9YMAAozmW6JGk74bvx3w1ekGYE8QWO8yZkpJenIceeshoeqLuuOMOo+PBiZWN42xZvHixabtEPxL1Aw88YDTz09Ky7P7v//7PaM6755xzjtH0TLGXGv2NDJik35FzA72GzCBMG0f0yHA877bbbkZfd911RtNHlC/KyspMJh79QxyLlXmgmOkkJXO8mEXFc4BZY5x7eV1kph/nZuZOseUXr7vXX3+9SGW5XbfeeqvRvH+I+zh/+umnjO8Vx7+JchzHcRzHyQK/iXIcx3Ecx8kCv4lyHMdxHMfJghr1RC1cuND0NKL3h1kzzH1hfkla3gV7JtH3wGyYo48+2mh6WJhHQR8XPVf0ObAfG7OJ0nxdd955p9Fs3MyMlc6dOxv92GOPJd4z1zRq1Misa3PNOu7JkJJ+Jq6xs++ZlMzcWrRokdGsP9fd6Wmg14O+CfaLYm2YGUafUBrHHHOM0ZdddpnR7NfG43L66acbzeykXFGvXj3jAaMnapdddjGax5oZXvQzSUkfC2vObBmeM88++6zRPBY8B+nTa9iwodHsz/Xee+8ZneZXZLbQcccdZzTPmTfeeMPouFeMc0mu+PXXXzVmzJgKTa/YUUcdZTTnk8rmQEl68MEHjeY8x89Gz9KQIUOM3n777TNq7sP+++9vNP1M9OSkZV1xH+nr4flx6KGHGs3MwHwxb948k+3Vu3dv8/q///1vo5m3yHGU5t2szF/KnCiOvXjfTSlZL469wYMHZ9xHzv38jMx0k6RTTz3V6EmTJhnNevH6EJ9rzzjjjMT7Lw//JspxHMdxHCcL/CbKcRzHcRwnC/wmynEcx3EcJwtCFEU1trEOHTpEt912W4X+4IMPzOv0uDCLgn6mI444IrGNp59+2mj6ibiWSg/VdtttZzTXevn+9IpwLXno0KFGH3744UYzs0VKer/mzp2bcRtcP541a5bRZ5555qQoinJqpll11VWj+No8jyv9S/Q4sLbt2rVLbOOKK64wurL8K2aPvP3220bTg7f66qsbzTwu5rHQB8S+dmm+C2ZTsR8XM1uuvfbajPs0evTonNdSktq2bRvF+8DxszNf7YADDjD6vPPOM5r5a1LSd8GfYX14vE866SSj2VMy3i9Okjp27Gg0c2LY/40eCmYZScn5gV4O9qnjec3+XlEUJQO1VpDWrVtHLDKWtAAAIABJREFU//znPys08/W6dOlidNw/JSUz3tjHTkrWimOTx4W90w466CCj2QeRvdI4tvj+1PRkpuVEcZ498sgjjb733nuN5vnAf3/zzTfnZWyuscYa0QknnFChmaHF+Z+ZWrxe0BsqJf2J9NWyFyo9cMy3o9+MOWCcezk30A9dWaaclLzuTZ061Wj2TuU1KN4L9fzzz9fXX39dpbHp30Q5juM4juNkgd9EOY7jOI7jZIHfRDmO4ziO42SB30Q5juM4juNkQY2Gbc6cOVO33HJLhT7rrLPM62wIS9M3jWM0lkrJ8DsGItKASEMd35MBoGxqSkMdTXsnn3yy0euss47R/Expf8cAPxo/GSy30UYbJd4z15SWlmrmzJkVesmSJeZ1hpx+9tlnRrOR6yuvvJLYBpsUL1682Gg+NMDgU/78IYccYjTDG2lkpvmRZmp+pnfeeUeEDw6w0Scb1HKbaWGs+WDevHkm0I7n+bnnnmv0JZdcYjSP5QYbbJDYBh8codGbQaRsYsyxRgMtGxJze5WZoWnYpak/7T2nTJliNJtWMyRw+PDhFX+mGT9XlJSUqEWLFhWaoaE8B3kceM5xfpGS9abxmI2Yt9lmG6M5T9KAz3PhlFNOMZqN4UeNGmV0t27djOacKSXrzZ/hPn/11VdGpzXZzgezZ882YaL87HzAhfVm0CXHiZR8yOPggw82msGjDGTlQzz/r71zD7aqrN/486pchMLh+FMyNbHxkiSaioa30UlUJLGLeau8lGVjliia11JLSwMTLe3iICAkirfCW6PIyKhpqY2KASE/G0VMxfFXKijGZf3+4HhmfT9rs89hsfc+283zmXHgOfuw17vXd73vXu732c930003DZrXGI3k/GISv6DBevILXlIxRJRr2P333x80v3CRD4TlFxuq4U+ijDHGGGNK4JsoY4wxxpgS+CbKGGOMMaYEDfVE9e3bN+wzMwCNDQFvv/32oOkXqRQCd/nllwdNT8rs2bODzvt6JOkTn/hE0GyaSp9Wnz59gmYIGRs1MuCPzy8VAzzZxJSPc2++0p53rWEDYvqP2tragqZXhGPkeZeK+/CDBw8OOu/7kIr7+gxP5D48rxU2j6Zvg00z6QNhbaXi3jp9GPRZ0d/wpz/9KehnnnmmcIxasHLlyuBDGD58eHicPr0bbrghaL52BttV+h0G0dLrcc011wT9wAMPBM2GxQzvpaeJfkh6gziXK3ncRo4cGTRfZ//+/YNmaOnYsWM7/l4pkLQWcG5yjKwlQ2q/+MUvBs3GzFLx2qf3i15TNoe++eabg+Z6wTDF9957L2jOi+222y7oGTNmVB2PJL355ptBM3SUvhx6bugbuvrqqwvHqAX9+/fXkUce2aF5Xd95551B77HHHkFznWQzaKl4LdKz+rGPfSxoepLY7Jv/nmsvf5/vcWx4zHWTflSp2MydzcLpy+J6kQ8brtR8fE34kyhjjDHGmBL4JsoYY4wxpgS+iTLGGGOMKUFDPVHLly8PHiT6i9isley1115B5xsGfsCsWbOCnjp1atBshPvJT34yaI6JTQzZWJfZQ7vsskvQ+X1WqZjZUSlrhD6pfDNRqejVoZeDDUbrwYYbbhiaM7NBKP1FbJI5dOjQoCdOnFg4Bvf26RdiFgn9KPRJ0HfFjCZef/Qz7bnnnkEzl2r77bcXYS2ZG8YxjRs3LuhKeUv1IKUUXi/nAT0N9PrwGq3k92HDTzbWvuqqq4JmE9KbbropaDZF5tyn94e+PHqi6HGZPHmyyBNPPBE01wNmjdFDlb/OebxasWrVquAhYr7WgQceGDTXJGYuVcqyo5eUuUHM5KJHhr6rs846K+jTTz89aHqmWAf6m5hbRD+TVGwG3pnvhnOA56BeLF26NLxe+su4TtIzxXXusMMOKxyDmUp838rnO0qd++Y417le/P73vw+a7+2dNfI+6qijROjTevzxx4Pmez29xflriBlS1fAnUcYYY4wxJfBNlDHGGGNMCXwTZYwxxhhTgoZ6oj760Y+GLBj6LuiBOPfcc4NmNgmzaqSiz4DZU+zHNm3atKD79esXNPvQcT/6jTfeCJo9xrgX/5WvfCXo0aNHi3zta18Lmt4bjon5Jm+//XbhOWvNf/7zH02fPr1DZ1kWHmevI+Z+MCOMvbGkYtYLc8GYM3TppZcGzZwo7pGz9xo9UfQJ8Pj0w/H5pGK+zZw5c4Jm7ejZa5Tvglk0jz32WHicfcS22WaboPn7w4YNKxyD1wAzsnhdn3TSSUHn+85JxXycnj17Bk1fBbPJuBb84Ac/CJoeGan4uugvosdx0aJFQed7eXLtqBXvv/9+ONfHH398eJx+FM5V+peYpyMV/YnsY8lemp31/+R5ZS3y+VpS0QNFzw3fK+hdk4q1oeeJ8/ull16qOmb2gqwVPXv2DN5JjpP9HOmZZf2Z+SRJ99xzT9D0f9LzRD8Z1ylmU3G9YMYb/Yn//ve/g6Y/tVKuINdS+hGZsUf/cn6MHE81/EmUMcYYY0wJOr2JSilNSCktTin9PfeztpTSjJTSgvY/+1d7DtM8uJ6tg2vZWrierYNruf7QlU+iJkkajp+dJ2lmlmXbS5rZrs2Hg0lyPVuFSXItW4lJcj1bhUlyLdcLOvVEZVn2cEppIH78BUkHtv/9RkmzJJ2rTli5cmXID6FHhf4h7svefffdQVfKQ2K/K/aNYv+c3XbbLWju9eezkKRiZgr9Ag899FDQ9ExceOGFQbPXnlT0C3Fvd8KECUEze+g3v/lN4Tk/oFb1fO+99zR37twOzcwu7omzbxpzo+hvkYr1ZS1Ya+7b33vvvUHvvffeVZ+PvRl5PdLTQ+8B+z9J0gknnBA0PXXst0VvwhlnnFF4zg+o5dzceOONgweMOSnMaOK5PPPMM4OmH00q+iaYw0MfzRVXXBE0PWuca/RD0ht43HHHBX3ttdcGTQ8Ws4akYo9Ieid4TXDu5j2h7P9Wq3r26NEj9AlkD1L2E73vvvuCZl2+9a1vFY6R93ZJxUw2Xj98Tmb2MMuKa/8RRxwRNHPmmNFG/xxzjaRifTkm+qy4prG3Y55av2/mfWrMN2JOINdBzpNK+YrMdWIGG9favB9Wkr7xjW9UHQPf6+mhpc/rkksuCZr+aV4vUtFLzPwz9mKlznuk6K+qRllP1IAsyz5wBr4maUC1XzZNj+vZOriWrYXr2Tq4li3IOhvLs9W3lNmaHk8pnZJSeiql9BRTbE3zUa2e+VryEzvTfKzN3OS3YUzz0dW5yW8lmuZjbebmsmXLGjgys7aUvYl6PaW0hSS1/7nG72FnWXZ9lmVDsiwbUukjONMUdKme+VpW+jq/aQpKzU22zDFNw1rPTcYNmKah1NzktpRpLsrmRN0l6URJV7T/Ob36r69mww03DHki7H/EfXX6l3gx0e8kFXs4MY+Cva2YN8I8Ef5fAPMpuI/+wx/+MGju/R5zzDFBM7tCKvbr43lgvg771DGTowusdT3b2trCa/nud78bHj/55JODzntBJGnw4MFB8zVKxV5ozO055JBDgubef2degVdeeSVo+nyY40M/DHOGhgwZIsLMnYsvvjhoZoDRQ0d/XBcoNTfnzp0benBxHrG3Gc8Fr8FKCz/9hLvvvnvQ9O7QQ8WsGGaudZYNxAw35ufQA0dPllS8puj1oa/usssuCzrv8+M5XANrXc9Vq1YF7ym9ZI8++mjQ9Abuv//+QVfyK9KXw7lArx99WFxH6YdjThTnbt7zVenfMwNowYIFIsxLYhYa87S4ZnHtf+CBBwrHAKXmJnvO8n94OC84z3guK32YQQ8ZMxiZmUd/KDP9Ovtkm/4y6ilTpgRNj12l3nb0YfJ+gvXlen3bbbd1/J35ftXoSsTBzZIel7RjSmlRSulkrb4IDk4pLZA0rF2bDwGuZ+vgWrYWrmfr4FquP3Tl23nHreGhg9bwc9PEuJ6tg2vZWrierYNruf5gY4sxxhhjTAka2jtPiv3F6FFgzg73Sdm77NRTTy08P/fiuW/NvAp6df7whz8ETU8TvSLs18VvrTEficevZALl6/7tb38bNPebd9ppp6CZVVMP3n33Xc2ePbtDX3TRReHxww8/PGga0elHoqdGKnqguE/NnlHMGuF5GzduXND0Tey6665Bt7W1BU0vyeuvvx40PT6Vfof+hTvuuCNoZvbMmzev8Jz1YPPNNw85K7yu+TqY28LXXqn31NFHHx00zycfZ33poWTGGr07fJy9Ofka6bv55je/KcLzMGrUqKDZA5K9EPPZavX61tUGG2wQ1hW+TnoHuQZxXPTUSMWsOuafsXb0JDF3iF7VnXfeOWiuafTgMKeMuUeVvrFIPyyvWXpo7rrrrqAr9aCrB+xred1114XH+R7Hc0mP1OWXX144xiOPPBJ03h8kSePHjw+avrm99toraHqm2DOUHjpmPPHegP0d2cdQKs5X5oDl36+kolcwf42szbfP/UmUMcYYY0wJfBNljDHGGFMC30QZY4wxxpSgoZ6oZcuWhVwVek6Y63DzzTcHzdyX66+/vnAM9gWi/uc//xk0e+Swx9LChQuDpr+gM1/Hq6++GjT3r7m3LxUzVJhXxB5h7BvFHoMzZ84sHGNd6dOnT+g7SI/T8OGx9yb7lPE8MqNJKmbw0BPDPXDmn/zsZz8Lmt4A7uMzs4nXIzPH6JGgx0qq3Kcqz/333x80/SgDBw6s+u9rRc+ePYPfb7PNNguP0wPFjCV6B+n9k4o5TvTmcC6xXxo9cdOmTQuac4/+xh122CFo5jTRW/jTn/5UhD4r+qh4HbPHW95TVWnu14INNtgg+AmZh9OvX7+g6dvkvPrlL39ZOAaz6F544YWg2b+TGT7MYDr33NhCjrlFXOOYO8VMOV6/9CJKxfVi8uTJQbN/30EHxS/WVfL91YN33nknvD5ep4S5X5/61KeCPvTQQwv/hu8pfG9m/0R6NW+88cagef6ZK0Z/M/vY5T1glX6f9ZakAQNiF53Pfe5zVcdIb9mPf/zjjr/TX1sNfxJljDHGGFMC30QZY4wxxpTAN1HGGGOMMSVoqCdqo402Cnul9Mmw3xr7b9FfQM+DVOxVxnwQeha23XbboJmxwX1zamas0L/ELCv6C9j/Syr2LaK/iFlYJ554YtCNyC9ZunRpyHrJ9wSTpOOOi4G99HnRK1RpzOwryCwRZpHke79JRc8MoR+JvgvmRs2fPz9oZmHRFyIV+yDyPPAY9IrwPFbKeKkFK1asCLk59PpwLtIfRq8f/SNSMduF/a9GjhwZdD5TSZJOOOGEoJklxLlHTV8O5yY9MczCkYpeIGbRfP3rXw+a9eSY60FKSSmlDt1Z43f6RC+44IKgK/X4o8+K+Xn540vFWgwbNixo1prr+Kc//emgef3Ru0r/Iz0+UtGTxxwxXsPMTuK6XC+22morjR07tkP/8Y9/DI8z74jeIPb0q9RHkN5d+hHpeWR9mfs0ceLEoNlzkrmBXAdZ/4cffjjoSl5TZoPRY8n3KJ7H/Guo5KFbE/4kyhhjjDGmBL6JMsYYY4wpgW+ijDHGGGNK0PCcqLyvhPuoixcvDpr77sx0oodFkpYsWRI0917ZT+3ggw8Omn6k/fbbL2hmcAwaNCjo5cuXB01PC8fM/B1JOuWUU4LO5zFJxQwePudbb71VeM5as8kmmwRPAM/7UUcdFTT74HEPnrkyUtEDw33xjTaKly89SfTYMdeJteM+PK835kZxX56erErH4DXPWjIPZcyYMYXnrAfLly/Xyy+/3KGZV0WvIOtJTwv9R1KxP+KVV14ZNHuVPfvss0HTq8Mek+wpyXkxZcqUoPmapk6dGnSla5L9+ejtoQeK10w+K4vXb61YuXJlWAPYk5R5WdTM86LfSCrWinOHmV6vvfZa0KzNpEmTgh49enTQ9Exy/aCfjvPu3nvvFRk8eHDQfL/ha2K+Vq9evQrPWQ9efPHFkN3FucccKPoweW74PisV/ab0+uY9WVJx7WOvRHp/n3766aDp06MvjzmC++67b9XjS8XcLnqk6NvjNfTuu+92/N2984wxxhhj6oxvoowxxhhjSuCbKGOMMcaYEjTUE0WYNcEsGvYyY28y+pek4n4+92q/853vBM2+UfQ0MQeGz8/+fvQzMReIGR70ZVQ6BvOwmHdBnxb3guvBf//739Avj33L6JG56aabgmbezj777FM4Bj0q9Djlc6ok6ec//3nQ7H1G78DHP/7xoJlLtfvuuwf93HPPBU2fRyUfEP0HZ555ZtD0ZbBnE3PO6tEH8YPj5ucKe+PRw0K/CPPWmOMjSb/+9a+D5vljffv37x80s3+YHUQPE3OgmMnWWT+2Qw45RIT1mz59etBcP5hlls8m4+upJXm/Fa8p9v9j7ZjfteWWWxaeP98DUCr2PmP9OdfmzJkTNLPuuK5y3WQfVfZJGz9+fNDskykV5yv9qZ/5zGeCpoeNnpp60aNHj+AJY/04Dvrw6BWt1NeSWVNcl773ve8FTX8q68e5zfdu+heZZcX+gDz3la5J9tPj+s/1hz6rfFYVM8Sq4U+ijDHGGGNK4JsoY4wxxpgS+CbKGGOMMaYEDfVE9e3bN+xNcy9+0aJFQXMflPu29DtJxb1xZiwx64M+mnweh1T0ZeR9QFKxPw99NMybYP7FjjvuKMI8Hfoq6LOiT2ZtMi7K0qtXr1A/Zv7Q3zZixIigb7/99qCvuuqqwjGY08SMHmZ40TfBrJp8DohUzCHi47zeuA/Pf8++aVKxlm+//XbQ9PWwrxWvl3rRv3//4BniuX/kkUeCpr+ss7kqFX1Ul1xySdDM8qEn6Utf+lLQzCb7xS9+ETR9Fl/96leDZmbXAQccEHSlnnFnn3120PRV0PvBLKu8J5PjqSV5HxR9mjNmzAiaayavQfpXpKLHkeee6yz7VPL64RrHnCfm7f3ud78LmnmAXCMr+YDoy6SfbcKECUEzR4r9IutF7969w/sEr3t6Yvk66PXkuZSKc5Ov/aKLLgqanjeua8z94tpMzTHRQ3fLLbcEzfcXqegV43r+k5/8pOoY8z5evp5q+JMoY4wxxpgS+CbKGGOMMaYEvokyxhhjjCmBb6KMMcYYY0rQUGP5qlWrgimOxj4aPV988cWgaR5kAJhUNCjec889Qffp0ydoNrl94403gmYgG43nNHUzKJMmPzaPpPFYKpoi2WyRYXg0UOcbydaLxYsXhyaTNKcyRJRfAmDQKk3GUtE8eOCBBwbN2tLky8DPk046KWh+kYFj7CzgbbPNNguaBk6peB5YKxo4GcDIOVIvVqxYEcylhx12WNVxcW4y4HXPPfcsHIPBtuecc07QvCY4VyZPnhx0Z8by73//+0EzoJGGaQYIzp49W+TII48MurPgyn/84x9B33333R1/rxS0WwuWL18eAhh53rlG8TzwNcybN69wDNbmwQcfDJrH5PXEufv5z38+aK7D+fMmFRvYMiiTzYEZWCwVjf00irMpLkOOubbXi1WrVoV1gwG+DFNl8CnX5iOOOKJwjKuvvjro008/PWgGX3Nt5trHLydwDPzCBc3x/Pc0hdMIL0kHHXRQ0PzyQ48ePYJm2G5+Lq/NF3r8SZQxxhhjTAl8E2WMMcYYUwLfRBljjDHGlKChnij6LrIsC4//+c9/DpqBa9yH5d6wVPRJ0SfBvVc2wuSYuHdLHw0bxNJzxednyBx/Xyp6w9jMkUF2Bx98cNA/+tGPCs9Za3r16hU8BDwPo0ePDpqvk+eVr0kqnru33noraAabcp+cniWGJ7LJKcdAn8bzzz8fNBsSDxw4UIShbVtvvXXQU6ZMCZq+C77GSk23a8HGG28cgj/p1xk6dGjQvCY5r+hpkaSdd9456MceeyxoelK4HnBu0odFH8cee+wRND1UDBTl3GRDZKnY5Jg+LjYk/va3vx10PmSWYYC1IqUU1kpep21tbUHTA8X1p5KfiB6X4cOHB33fffdVfQ56mrh+0I/00ksvBb1s2bKg6W/h9Uh/nFRcy3meLrvssqD53nLaaacVnrMerFixIswnNn9m42w2W2Yj5UoN6uk3Zbgl/YRcW6+77rqg2cCcPl56jemBOvbYY4NmMOa4ceNEeA3SIztt2rSgjznmmKDzXmKuNdXwJ1HGGGOMMSXo9CYqpbR1SumhlNLclNKclNKo9p+3pZRmpJQWtP/Zv7PnMt2La9lauJ6tg2vZWrie6w9d+SRqhaSzsiwbJGmopNNSSoMknSdpZpZl20ua2a5Nc+NathauZ+vgWrYWrud6QqeeqCzLXpX0avvf30kpzZO0paQvSDqw/ddulDRL0rnVnqt3795h75pNBumBYX7JPvvsEzQ9DVIxr4K5UYceemjQ9MnQp7HTTjsFzdwX5ufQK8C9YO7lr1ixQmTTTTcNml4OvsbOGrt+4JmqZS179OgRcjXocbr11luDZsYSmwfzPErSCy+8EDTzbebPnx80PTdvvvlm0MwBopeMTVL5GtgQd7vttqv6fFLRS8C8FWbV5PN9OqOW9Vy6dKn+8pe/dGj6LH71q18FTS8QPS70l0jFpqNsDs7XzkwmZmw9+eSTQdMDwSa1+QbLUtEzNXjw4KCZfSQVM9t4jbGRNv2Jea9Hfh7XspZsJs2cJ84b1vLSSy8N+qyzziocg2svG8jSo8RcKa4HzORis2l68jiXmadG/2SlhtjMS5o1a1bQ11xzTdD0VdG3laeW9VyyZElooHzxxReHx7n+Mx+ps3VQKs5fNrGnP5m+OnrQ6Jnq27dv0FwLmCvGdZPXKD1YlX6H3rFRo0YFzabU+dyvunmiUkoDJe0m6a+SBrRfKJL0mqQBa/hnpglxLVsL17N1cC1bC9eztenyTVRK6SOS7pB0RpZl4StH2erbtoq3bimlU1JKT6WUnuI3lUz3UIta8v8UTPdRi3ry21Gme6hFLeuVhG7WnlrUk4nkprno0k1USqmHVl8IN2VZdmf7j19PKW3R/vgWkop5A5KyLLs+y7IhWZYNqfQ1dtNYalVLtlUw3UOt6smP203jqVUtubVuuoda1ZOWENNcpM72/tLqjcYbJf1flmVn5H4+VtKbWZZdkVI6T1JblmXnrOl5JGmTTTbJ8r4F5qTQT8R91yeeeCLoAw44oNIxgma+CPMp6KN59tlngz711FODZo4PvQC84JnBwn5s9OFI0mc/+9mg6fVgfg57Ai1cuDDoW2655W9Zlg2pZS232Wab7MILL+zQrM3+++8fNL1g7GtXKS+LfjH6bHju+Zz/+te/gh4yZEjQPG98DfTI8NM35kbxepWKvh/u/U+dOrXq7zM/ZcqUKX/LsmyIVNu5OWDAgCzv16Gn5Mtf/nLQ9JPwdTBrSJLOPvvsoOkXymcoSUVfBr0dzMOhD4O5csy+GTNmTNC8fsaOHStCrw89lrwZve2224LOZ9s999xzWrJkSZJqW8t+/fpl+TWE/UHZI5Iey876IkqrfVd5+GnJ3nvvHTTnLv1tXPvZK4+wtxm9rfRUMfNJKnpyOb933XXXoJkrxWPeeuutdZmbm2++eXb00Ud3aPalZP9VXoOcBzz3UrH34LBhw4Lm/L/22muDZr9Xeqw49/iByogRI4Jm5tOVV14ZNN8jpeJayd6GfG/mNZw/b2PGjNHChQujyWoNdCVsc19Jx0t6LqX0wWy7QNIVkm5NKZ0s6SVJR6/h35vmwbVsLVzP1sG1bC1cz/WErnw771FJa7ojO2gNPzdNiGvZWrierYNr2Vq4nusPTiw3xhhjjClBQ3vn9enTJ+wzs78W912ZZ0F/Ej0sUjG/hPvc3E/m3iw9KcyFYq8zPk4fD/fNmdHCTCipuKfNPBJmrjBPix4q7kfXgvfffz/sQW+wQbwf5+ukJ4qeBmbZSMV9dWYXsfcdH2cvu6effjpoGnA5Jp5XZtnQY1XpixMcI69h5q/w+nzqqacKz1kPsiwLHgJ+k5b+Mvayoi/j8MMPLxyDfQPpYaAPg5lavO75jUI+zvwcziuuFbxe6H+q9Bz0OBKuJ3nfFedIrejdu3fIMOM1xHwzZjbRK8LzKhU9c/SrUecz5aSiF5WZbLxW2Fdx5MiRQZ9//vlBjx8/PugddthBhDlCnO+s9S677BI0zxNfQ73g3KP/kOsS/WrMTpSK54f99ugP5Hs13we5XjDjke/LzAk877yYQcrXVClfkd495goSvvdOnDix4+/MZqyGP4kyxhhjjCmBb6KMMcYYY0rgmyhjjDHGmBJ0mhNV04Ol9IZWf63zfyQVm6U1F80+xrUZ3zZZlm3W+a91Hdey5nR1jDWvpeR61hjXsuu00hjX93o2+/ikOtSyoTdRHQdN6akPQsmalWYfY7OMr1nGUQ2P8cM3jmo0+xibZXzNMo5qeIwfvnGsiWYfn1SfMXo7zxhjjDGmBL6JMsYYY4wpQXfdRF3fTcddG5p9jM0yvmYZRzU8xq7TLOOoRrOPsVnG1yzjqIbH2HWaZRxrotnHJ9VhjN3iiTLGGGOM+bDj7TxjjDHGmBI09CYqpTQ8pTQ/pfS/KaXzOv8X9SelNCGltDil9Pfcz9pSSjNSSgva/+xf7TkaMMatU0oPpZTmppTmpJRGNcM4Xc9S43Mtu0iz17J9PK5nF2n2erqWXafZa9k+nobUs2E3USmlDSVdJ+kwSYMkHZdSGlT9XzWESZKG42fnSZqZZdn2kma26+5khaSzsiwbJGmopNPaz123jdP1LI1r2XUmqblrKbmea8MkNXc9XcuuM0nNXUupUfXMsqwh/0naW9L9OX2+pPMbdfxOxjZQ0t9zer6kLdr/voWk+d09Rox3uqSDu3Ocrqdr6Vq6nq1cT9eydWpZz3o2cjtvS0kv5/Si9p81IwOyLHu1/e+vSRrQnYPJk1IaKGk3SX9V947T9VxHXMtSNGUtJdezJE1ZT9eyFE1ZS6m+9bSxvBOy1berTfEVxpTSRyTdIemMLMtrjQhHAAABV0lEQVTezj/WTONsZprlPLmW604znSfXc91plvPkWq47zXSe6l3PRt5EvSJp65zeqv1nzcjrKaUtJKn9z8XdPB6llHpo9YVwU5Zld7b/uDvH6XqWxLVcJ5qqlu3jcD3L01T1dC3XiaaqZfs46l7PRt5EPSlp+5TStimlnpKOlXRXA4+/Ntwl6cT2v5+o1Xup3UZKKUm6QdK8LMuuyj3UneN0PUvgWq4zTVNLyfWsAU1TT9dynWmaWkoNrGeDjV0jJD0v6QVJF3a30ax9TDdLelXScq3ebz5Z0qZa7dpfIOlBSW3dPMb9tPojx9mSnmn/b0R3j9P1dC3X51q6nq1VT9eydWrZyHo6sdwYY4wxpgQ2lhtjjDHGlMA3UcYYY4wxJfBNlDHGGGNMCXwTZYwxxhhTAt9EGWOMMcaUwDdRxhhjjDEl8E2UMcYYY0wJfBNljDHGGFOC/wermsT+boUW0gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x720 with 25 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising layer 2\n",
    "learned_weights = dbn_mnist.rbm_layers[1].W.transpose(0,1).numpy()\n",
    "plt.show()\n",
    "fig = plt.figure(3, figsize=(10,10))\n",
    "for i in range(25): \n",
    "    sub = fig.add_subplot(5, 5, i+1)\n",
    "    sub.imshow(learned_weights[i,:].reshape((23,23)), cmap=plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5421"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "number = 5 #A number between 0 and 10.\n",
    "\n",
    "particular_mnist = []\n",
    "\n",
    "limit = mnist_data.train_data.shape[0]\n",
    "# limit = 60000\n",
    "for i in range(limit):\n",
    "    if(mnist_data.train_labels[i] == number):\n",
    "        particular_mnist.append(mnist_data.train_data[i].numpy())\n",
    "# particular_mnist = np.array(particular_mnist)\n",
    "len(particular_mnist)\n",
    "# mnist_data.train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = torch.stack([torch.Tensor(i) for i in particular_mnist])\n",
    "train_label = torch.stack([torch.Tensor(number) for i in range(len(particular_mnist))])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------\n",
      "Training the 1 st rbm layer\n",
      "RBM fitting: 542it [00:04, 108.43it/s]\n",
      "Epoch:1 ,avg_cost = 0.4802815914154053 ,std_cost = 0.0413631871342659 ,avg_grad = 10701.537109375 ,std_grad = 539.1286010742188\n",
      "RBM fitting: 542it [00:04, 109.53it/s]\n",
      "Epoch:2 ,avg_cost = 0.4664861857891083 ,std_cost = 0.041570115834474564 ,avg_grad = 10595.7509765625 ,std_grad = 523.930419921875\n",
      "RBM fitting: 542it [00:04, 110.73it/s]\n",
      "Epoch:3 ,avg_cost = 0.46115565299987793 ,std_cost = 0.040436387062072754 ,avg_grad = 10577.158203125 ,std_grad = 532.3255004882812\n",
      "RBM fitting: 542it [00:04, 110.07it/s]\n",
      "Epoch:4 ,avg_cost = 0.4562234878540039 ,std_cost = 0.03937240317463875 ,avg_grad = 10561.625 ,std_grad = 523.265869140625\n",
      "RBM fitting: 542it [00:04, 110.76it/s]\n",
      "Epoch:5 ,avg_cost = 0.45389828085899353 ,std_cost = 0.03856254741549492 ,avg_grad = 10551.380859375 ,std_grad = 516.8515625\n",
      "RBM fitting: 542it [00:04, 110.14it/s]\n",
      "Epoch:6 ,avg_cost = 0.4521692395210266 ,std_cost = 0.0385347343981266 ,avg_grad = 10561.982421875 ,std_grad = 526.7361450195312\n",
      "RBM fitting: 542it [00:04, 109.88it/s]\n",
      "Epoch:7 ,avg_cost = 0.4512341618537903 ,std_cost = 0.040520261973142624 ,avg_grad = 10576.3837890625 ,std_grad = 537.3392333984375\n",
      "RBM fitting: 542it [00:05, 107.39it/s]\n",
      "Epoch:8 ,avg_cost = 0.4483449459075928 ,std_cost = 0.038684044033288956 ,avg_grad = 10562.4140625 ,std_grad = 529.1563720703125\n",
      "RBM fitting: 542it [00:05, 100.49it/s]\n",
      "Epoch:9 ,avg_cost = 0.4483908414840698 ,std_cost = 0.03887860104441643 ,avg_grad = 10570.890625 ,std_grad = 517.0592041015625\n",
      "RBM fitting: 542it [00:05, 107.83it/s]\n",
      "Epoch:10 ,avg_cost = 0.44661352038383484 ,std_cost = 0.038788244128227234 ,avg_grad = 10568.0517578125 ,std_grad = 537.1287231445312\n",
      "RBM fitting: 542it [00:05, 103.37it/s]\n",
      "Epoch:11 ,avg_cost = 0.446227103471756 ,std_cost = 0.03991188481450081 ,avg_grad = 10570.048828125 ,std_grad = 520.0927124023438\n",
      "RBM fitting: 542it [00:05, 105.15it/s]\n",
      "Epoch:12 ,avg_cost = 0.4455302953720093 ,std_cost = 0.038111329078674316 ,avg_grad = 10561.7841796875 ,std_grad = 505.67425537109375\n",
      "RBM fitting: 542it [00:05, 107.28it/s]\n",
      "Epoch:13 ,avg_cost = 0.44449758529663086 ,std_cost = 0.0386669784784317 ,avg_grad = 10574.5341796875 ,std_grad = 528.5682373046875\n",
      "RBM fitting: 542it [00:05, 103.68it/s]\n",
      "Epoch:14 ,avg_cost = 0.4438527822494507 ,std_cost = 0.03935395926237106 ,avg_grad = 10570.8857421875 ,std_grad = 520.3591918945312\n",
      "RBM fitting: 542it [00:05, 101.95it/s]\n",
      "Epoch:15 ,avg_cost = 0.4432593584060669 ,std_cost = 0.03835182264447212 ,avg_grad = 10566.8583984375 ,std_grad = 531.2382202148438\n",
      "RBM fitting: 542it [00:05, 105.89it/s]\n",
      "Epoch:16 ,avg_cost = 0.442916601896286 ,std_cost = 0.03817553073167801 ,avg_grad = 10568.498046875 ,std_grad = 510.5917053222656\n",
      "RBM fitting: 542it [00:06, 89.50it/s] \n",
      "Epoch:17 ,avg_cost = 0.4430980682373047 ,std_cost = 0.03871399909257889 ,avg_grad = 10569.3212890625 ,std_grad = 518.69580078125\n",
      "RBM fitting: 542it [00:05, 90.67it/s]\n",
      "Epoch:18 ,avg_cost = 0.4411771297454834 ,std_cost = 0.03849467262625694 ,avg_grad = 10569.5478515625 ,std_grad = 532.6403198242188\n",
      "RBM fitting: 542it [00:04, 120.08it/s]\n",
      "Epoch:19 ,avg_cost = 0.4421692192554474 ,std_cost = 0.03850514069199562 ,avg_grad = 10569.7490234375 ,std_grad = 524.4596557617188\n",
      "RBM fitting: 542it [00:07, 73.98it/s]\n",
      "Epoch:20 ,avg_cost = 0.4407064616680145 ,std_cost = 0.03797527030110359 ,avg_grad = 10547.3173828125 ,std_grad = 524.4125366210938\n",
      "--------------------\n",
      "Training the 2 st rbm layer\n",
      "RBM fitting: 542it [00:04, 114.23it/s]\n",
      "Epoch:1 ,avg_cost = 1.5668307542800903 ,std_cost = 0.11685563623905182 ,avg_grad = 12726.79296875 ,std_grad = 366.77044677734375\n",
      "RBM fitting: 542it [00:04, 118.72it/s]\n",
      "Epoch:2 ,avg_cost = 1.5317744016647339 ,std_cost = 0.11219735443592072 ,avg_grad = 12569.76171875 ,std_grad = 337.7249450683594\n",
      "RBM fitting: 542it [00:03, 154.26it/s]\n",
      "Epoch:3 ,avg_cost = 1.518539547920227 ,std_cost = 0.10964541137218475 ,avg_grad = 12604.826171875 ,std_grad = 355.7041320800781\n",
      "RBM fitting: 542it [00:02, 208.87it/s]\n",
      "Epoch:4 ,avg_cost = 1.5130383968353271 ,std_cost = 0.10972438007593155 ,avg_grad = 12630.9111328125 ,std_grad = 349.7060546875\n",
      "RBM fitting: 542it [00:02, 205.58it/s]\n",
      "Epoch:5 ,avg_cost = 1.5077815055847168 ,std_cost = 0.10994377732276917 ,avg_grad = 12639.5830078125 ,std_grad = 319.8765869140625\n",
      "RBM fitting: 542it [00:02, 214.74it/s]\n",
      "Epoch:6 ,avg_cost = 1.505225419998169 ,std_cost = 0.1097317636013031 ,avg_grad = 12649.5703125 ,std_grad = 337.6900329589844\n",
      "RBM fitting: 542it [00:02, 204.31it/s]\n",
      "Epoch:7 ,avg_cost = 1.501486897468567 ,std_cost = 0.1077805757522583 ,avg_grad = 12670.5439453125 ,std_grad = 347.1653747558594\n",
      "RBM fitting: 542it [00:02, 213.08it/s]\n",
      "Epoch:8 ,avg_cost = 1.5023223161697388 ,std_cost = 0.1103823110461235 ,avg_grad = 12675.0146484375 ,std_grad = 338.91192626953125\n",
      "RBM fitting: 542it [00:02, 212.54it/s]\n",
      "Epoch:9 ,avg_cost = 1.5007609128952026 ,std_cost = 0.1093381941318512 ,avg_grad = 12670.7109375 ,std_grad = 332.9410400390625\n",
      "RBM fitting: 542it [00:02, 214.17it/s]\n",
      "Epoch:10 ,avg_cost = 1.4969537258148193 ,std_cost = 0.11052418500185013 ,avg_grad = 12669.576171875 ,std_grad = 336.015625\n",
      "RBM fitting: 542it [00:02, 205.17it/s]\n",
      "Epoch:11 ,avg_cost = 1.5004069805145264 ,std_cost = 0.10908368974924088 ,avg_grad = 12675.5947265625 ,std_grad = 335.63433837890625\n",
      "RBM fitting: 542it [00:02, 214.33it/s]\n",
      "Epoch:12 ,avg_cost = 1.4975590705871582 ,std_cost = 0.10822778940200806 ,avg_grad = 12671.4599609375 ,std_grad = 342.0650634765625\n",
      "RBM fitting: 542it [00:02, 204.07it/s]\n",
      "Epoch:13 ,avg_cost = 1.4951459169387817 ,std_cost = 0.10798324644565582 ,avg_grad = 12665.1689453125 ,std_grad = 330.3404541015625\n",
      "RBM fitting: 542it [00:02, 208.01it/s]\n",
      "Epoch:14 ,avg_cost = 1.4953104257583618 ,std_cost = 0.10966561734676361 ,avg_grad = 12680.44921875 ,std_grad = 328.8835754394531\n",
      "RBM fitting: 542it [00:02, 212.67it/s]\n",
      "Epoch:15 ,avg_cost = 1.4985558986663818 ,std_cost = 0.10990628600120544 ,avg_grad = 12676.5625 ,std_grad = 357.8343811035156\n",
      "RBM fitting: 542it [00:02, 200.57it/s]\n",
      "Epoch:16 ,avg_cost = 1.4960503578186035 ,std_cost = 0.10869167745113373 ,avg_grad = 12664.625 ,std_grad = 343.3152770996094\n",
      "RBM fitting: 542it [00:02, 217.28it/s]\n",
      "Epoch:17 ,avg_cost = 1.4949637651443481 ,std_cost = 0.10916182398796082 ,avg_grad = 12672.474609375 ,std_grad = 328.18701171875\n",
      "RBM fitting: 542it [00:02, 210.07it/s]\n",
      "Epoch:18 ,avg_cost = 1.4928606748580933 ,std_cost = 0.11190258711576462 ,avg_grad = 12687.2626953125 ,std_grad = 333.9811096191406\n",
      "RBM fitting: 542it [00:02, 215.98it/s]\n",
      "Epoch:19 ,avg_cost = 1.4931063652038574 ,std_cost = 0.10892508178949356 ,avg_grad = 12677.8271484375 ,std_grad = 344.97174072265625\n",
      "RBM fitting: 542it [00:02, 194.22it/s]\n",
      "Epoch:20 ,avg_cost = 1.4950965642929077 ,std_cost = 0.10956544429063797 ,avg_grad = 12682.3818359375 ,std_grad = 336.6941223144531\n"
     ]
    }
   ],
   "source": [
    "dbn_mnist.train_static(train_data,train_label,20 , batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The original number: 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The reconstructed image\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "idx = 3\n",
    "img = mnist_data.train_data[idx]\n",
    "reconstructed_img = img.view(1,-1).type(torch.FloatTensor)\n",
    "\n",
    "_,reconstructed_img= dbn_mnist.reconstruct(reconstructed_img)\n",
    "\n",
    "reconstructed_img = reconstructed_img.view((28,28))\n",
    "print(\"The original number: {}\".format(mnist_data.train_labels[idx]))\n",
    "plt.imshow(img , cmap = 'gray')\n",
    "plt.show()\n",
    "print(\"The reconstructed image\")\n",
    "plt.imshow(reconstructed_img , cmap = 'gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}