MNIST_digit_classification.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UsRVlAumzfJG"
      },
      "source": [
        "### Importing the necesary libraries"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "Zrw8IX8nbBgq"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "import random\n",
        "from tqdm import tqdm"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SE7xTBEtl21f"
      },
      "source": [
        "### Getting the data from keras.datasets -> MNIST\n",
        "### Reference LINK : http://yann.lecun.com/exdb/mnist/"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZIntyOChz4Fm",
        "outputId": "c98f86e5-7c41-435c-d57e-83d1bbf6b077"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n",
            "11501568/11490434 [==============================] - 0s 0us/step\n"
          ]
        }
      ],
      "source": [
        "mnist = tf.keras.datasets.mnist\n",
        "(train_images, train_labels) , (test_images,test_labels) = mnist.load_data()\n",
        "train_images = (np.expand_dims(train_images, axis=-1)/255.).astype(np.float32)\n",
        "train_labels = (train_labels).astype(np.int64)\n",
        "test_images = (np.expand_dims(test_images, axis=-1)/255.).astype(np.float32)\n",
        "test_labels = (test_labels).astype(np.int64) "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r6ew0ol6l21h"
      },
      "source": [
        "### Plotting the handwritten pictures"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 589
        },
        "id": "vEUr-QRl0RgI",
        "outputId": "bfafde4e-bca7-47cf-b133-53d9b40109a8"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj0AAAI8CAYAAAAazRqkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy92W9l55X29+wzz/NMHhZnsliqgSWWXLIky7bcstPd+JDuvuikETQQoO/zIQiCJH9EkLvcdBAEnUYCdCcNG3G3LX2w2tYslWogq1jFmYfk4Znnedy5KK9V+1BkDSzO5/0BhOwqitzaZ+/3Xe9az3qWJMsyBAKBQCAQCC46qtO+AIFAIBAIBIKTQAQ9AoFAIBAI+gIR9AgEAoFAIOgLRNAjEAgEAoGgLxBBj0AgEAgEgr5ABD0CgUAgEAj6As2rfLPH45GHh4eP6VIuJpubm0in09Lr/hxx7w/Hd999l5Zl2fu6P0fc/8Mh7v/pIdae00U8+6fLQff/lYKe4eFh3Llz5+iuqg+Ym5s7kp8j7v3hkCQpchQ/R9z/wyHu/+kh1p7TRTz7p8tB91+UtwQCgUAgEPQFIugRCAQCgUDQF7xSeUsgEAgEAoHgsHS7XXQ6Hciy3PNFSJIEjUYDlUoFSZIgSa8tS+tBBD0CgUAgEAiODVmW0el00O12kUwm8fDhQ+RyOVQqFRQKBXQ6HQBPAx6z2Yxr165hdHQUOp0ONpsNOp3uyK5FBD0CgUAgEAiODVmW0W630W63sbW1hV//+tfY2NhAMplEJBJBs9kE8DTo8fv9+Nu//VvodDpYrVbo9XoR9AjOPp1OB+12G91uF/V6Hc1mk9OWkiRBq9XCaDRCrVaf9qUKBAKB4BjpdruoVqtoNBooFArI5XLIZDLIZDLI5/O8P6jVaphMJjSbzZ4y2FEigh7BkSPLMpLJJGKxGEqlEh48eIDl5WXodDp4vV5YLBaEw2Hcvn0bXq8XkiRBpRKaeoFAILiIlMtl3Lt3Dzs7O1hZWcHy8jJisRjK5TK63S4kSYLBYIDBYIDNZoPZbIbJZDqWg7EIegRHTrfbRS6Xw9raGtLpNH7961/j97//PYxGI0ZHR+H1ejE7O4vp6Wm43W4R8AgEAsEFplqtYmlpCfPz84hGo9ja2kImk0G324Usy5AkCXq9HlarFRaLBQaDgctaR70/iKBHcGS0Wi3UajU0m00kk0lEo1Gk02kUi0U0m02oVCpUKhXodDpUKhUufx21Ov8iIssyarUaGo0G2u02yuUyGo0GCwQBQKfTwWg0QqPRwGg0wmQyccpYBJZni263i3a7DVmWUS6Xkcvl0O12+bSrVqthNpthMBhO+1IFgkMhyzKazSZarRaKxSLy+Tzy+TzK5TJarRa63S40Gg3UajW0Wi28Xi98Ph98Ph8cDgf0ej20Wq0IegRnl0wmg9XVVZRKJXzxxRf49NNPUSgUEIvFADwNiuLxOLLZLEKhEGq1GjqdDiRJ4mhfsD/tdhuRSASRSAT5fB7z8/PY2tpCp9Ph+xgMBjExMQGLxYKJiQlcvnwZer0eZrMZRqPxtP8TBAqazSby+TwajQbu3r2LTz75BJVKBSMjIxgeHobVasW1a9cwMjJy2pcqEByKdruNeDyOVCqFnZ0dLC4u4tGjR3xgAwCz2Qyv1wuTyYTZ2VncvHkTDocD169fh9fr5YDoKBFBj+DIqFariMViyGazWFlZwfz8PGq1GtrtNoCn4uZisQgAKBaLaDQanN4UPB8qGUYiESSTSXz55ZdYXFxEs9lEpVJBq9XC6OgoSqUSnE4njEYjwuEwut0u9Hr9aV++YA+dTgfVahWVSgXr6+v4wx/+gEKhgGvXrqFer8Pj8WB4eFgcBgTnlm63i2KxiGQyiUQiwV+tVgutVgsAoNfrYbfbYbFYMDo6ips3b8JisSAUCsFsNh/LdYmgR/DKUEml0+mgUqkgkUigUqlga2sLjx8/Rj6fRzKZ5PIVodVqYTKZoNfr2XtBLOjPhxaISqWCWCyGtbW1no6HdrvN5a1arYZ4PI5KpQKXywWr1QqbzYbJyUlYLBZxr88Q9Xod8Xgc+Xwe6XSaS5f0WQoE55VarYZarcYB/dLSEhKJBLLZLFqtFlQqFdxuN7RaLYaHhzEzMwObzYaxsTE4HA4YjcYjz+4oEUGP4JVRth9GIhF89NFH2NraQjKZxMbGBmq1GvL5POr1eo/plMFgQDgchsvlwqVLl2AymaBSqYTe5DnUajVks1kUCgXcv38f/+k//SdUq1Ukk0lUKhXIssyBZTabxf3796HX65FIJLC2tgaPx4O/+Iu/QDgcFkHPGSKfz2NhYQG7u7t48uQJstksms1mj1+JQHDe6Ha7yGaz2NnZQS6Xw+9+9zt89dVXfGirVqtwu92Ynp6Gw+HA3NwcPvzwQzgcDhYxH0dJS4kIegSvjCzLaLVa7LmwtbWF5eVlZLNZRKNR1Ot1zkDIssxCWrVaDYvFAqfTCavVCrVazYu7WOR7IWv2ZrOJWq2GUqmETCaDeDyOer2OarXKKWKi0Wig0WhApVJBp9NBrVajXq+jVCqh2+2K4PIM0Ww2kcvlkEqlUCgU0Gg00Gq1WOMGiHdCcL6gA1itVkOhUEA2m0U8HkckEkGj0eB9QaPRwOFwwOPxYGBgAOPj43A4HCd2nSLoEbw0tCiXSiU8fPgQOzs72N7exvr6OpdVqLtIp9PBYrFAr9djYGAAPp8PVqsVExMT8Hg8CIfDcDgcorNoH7LZLLa2tlAulxGLxRCJRFAsFrG0tIRarcadDwchyzIqlQrS6TRkWUY8HkcsFmMPDKHxeT7NZhPVapXLsxQwms3mI9MZVKtVbG1tYXV1lUvBZMHvdrvhcrmg1+tF4CM4F2SzWWxvb6NarWJ1dRVLS0t8IKayrSRJ0Ol0cDgcGB4exuDgIILBIDSakw1DRNAjeCmUWYdEIoHf//73+Oabb5DP57G+vo5CodDTPk2RvN1ux49+9CO8+eabsFqtGBkZgcvlYn3PST/w54F4PI5PPvkEiUQCy8vLePToEarVKkqlEiqVygtdSmVZRqlUQrVaRbVaxcbGBtbW1mCz2TA8PCyCnhdQr9eRSCQ4+9JqtaBWqzE4OMg2AK9LqVTCkydPsLCwwL+HgtJgMAiXywWTyXQE/zUCwfGzu7uLP/zhD0in01hcXMTCwgJqtRqKxSIqlUpP9tnj8eD69esYHR1FIBA40hETL8OZ2HGUBkWvcurfK5Tdi1LvoJzketCGQRNdtVrtsdYUzxN00qXW6HK5jGKxiEwmg1QqhXK5zPoetVrNmRuDwcBCWq/XC7/fD6vVCo/HA6fTedr/WWcOelZJL5XJZJBMJpFKpZBIJFgfRZ1wL4LKi41GA5VKBcViEWq1+qX//X6E1oVms8nPNT37Go3me+XEw0BrHR0gyuUyAHD5l7yWyKvnokP3nO7LflO390KDK18FpV+V8ktweGieVrfbRblc5j0hnU4jk8lwME97OxkOWq1W3hsMBsOJZzNPPejpdDrI5XIol8vQ6/VwOp0vZchVq9WwubmJdDq979/LssxmSO12mzfmbrfLH4YSrVYLvV4Pg8GAubk5zM3N9X3g0+12kUqlsLu7i1qthq2tLUSjUWQyGTx8+BDxeLxHfEltthaLBSMjIxgbG4PVasXMzAzC4TD0er3wizmAWq2GaDSKUqmEhYUFzM/PIxaLIZ1Oc3r4VTp7lBv42toarFYrQqEQm38JeqEAsd1uY21tDR9//DGy2SxcLheCwSDMZjNcLtdrtZA3m01kMhlUKhXE43HUajV0u10YjUYWcQ4MDGBoaIj//0VDOU+JnulGo4F0Oo1qtcoWDAcF5/V6HZFIhNf9l/0szGYzRkdH4XK54HA4EA6HYTabodFoeB6g4NXI5/NYXl5GLpfDw4cP8c033/BhjbzDNBoN7+u3bt1COBzmvcHv95/K/MVTD3ra7Tay2SySySQsFgtMJtNLBT31ep1neBDKB7fT6WB7exuRSAT1eh25XA6lUgmNRgPlchn1eh3As83BZDLBYrHAbrdDkiTcuHGjr4MeyjykUik8fPgQhUIBDx48wOPHj1GpVBCNRtlFlu6h2+3G7OwsPB4PZmZm8MYbb8BsNrNwWczYOhgK4mOxGBYWFrCwsIBYLPZK2Z29yLKMRqOBjY0NNBoNjI2N4datW0d85RcDZUfi+vo6Pv74Y2xtbeHy5cu4efMmXC4XRkZGXstTqtVqcfYuFouxh5XBYIDX64XD4cDg4CCGhoZgMpku5PrT6XRYk0aWC4VCAWtra8hmsyiVShzo70c+n8fnn3+OlZUVzsy/DB6PBz/+8Y8xPj7OAQ9l1kSJ/XAUCgXcu3cPkUgET548wbfffotCodDTwEIO44FAAD/60Y9w69Yt2Gw2hMNhLhVf6EyPMoVfq9VQrVZRr9cRi8U46NHr9ahWq/xAqlQqNJtNttwn8vk8EokEUqnUvr+r2+0ik8kgl8uh0WigWCyiWCyyhT8FPcrvp99J9vD9CqWXaSNIp9PI5XLIZrMoFos8aoIieUrFOxwOOBwOOJ1O2O12WK1WGAwGruUKDoYWf/Lgocn0+0ElWJVK1aONooCe3jF6juv1OiqVCpds6vU6/7vihPuUdruNYrGIcrmMfD6PUqnU451zmJLKXmjdK5fLHPBQ5khZGqZ/nsfPhg5B7XabM1kEBeGUcad1pFgsIh6Pc8Y/k8kcWErM5/MoFAqoVCoAwFmavVKGvRgMBmSzWaRSKbZ06HQ6sNvtHPicx/t90tBnS+aa+XyeLTWogiJJEo+XsNlscDqd8Hg8cDgcPExUo9Gc2gH4RIMe0hfUajXcvXsX3333HarVKhKJBHK5HJ94LBYL3G43RkZGYDAYsLW1hZWVlZ5NoF6vY2dnB5lMBsD305yyLKNaraJYLPaUtA46OdMHIB78pwsXCThXV1fx0UcfIZvNIpvNcrmQgka73Y7Lly/D5XJhamoKb731FpxOJwKBAJxOp9BHvSQ0jX5+fh7xeBzVavXAwFun0yEQCMBmsyEQCGBubg5+vx9LS0v48ssveWMoFApot9s87kCj0WB9fZ3fsUAgIGY7/ZF8Po+vv/4am5ubWFpaQiaTYVuAozoANRoNbG9vY3V1lbtagGdzuCiwUupbzhO0EbZaLSQSCTx8+BD5fJ7/vtvtolQq8Sw+OkQ1m00UCgXO/Cj9vfZC/x7wVJJAG2i73eb1XRnwE+VyGQsLC4hEInA6nVhaWoLVasXc3Bw++OAD2O12qNVqkfV5ATRKqFAoYHl5Gffv38fKygpyuRwHqgaDgUu2P/zhD3Ht2jV4PB5cvXoVgUDg1PeEE/2EW60WyuUySqUS5ufn8a//+q8olUrI5XIoFovs1EsmdnNzc7BarXjw4AE+//xz1Go1/lndbhf1er3nRLBf4PMyUIpNBDxPoXp7s9lELBbDvXv3kEql9r2fZrMZY2NjGBwcxPj4OCYnJ9lW3GKxiHLWS1KtVrG2tsajO/ZmIpVoNBo4nU74/X5MTk7iF7/4BcbGxvDpp59ie3sbKpUK7XYbpVKJM5vVahUWiwWxWAyxWAxutxtut1sEPX+kUqlwN1UsFkOhUOD15aiCj1arhXQ6je3tbaTTaV67KEtxFNmk04QOl+QMfvfuXZ67R3+fTqeRzWY5w08BzH4/6yDo86DyiV6v5wBSrVbva/DYaDSwubkJALBYLIhEIjCbzdDpdPjBD35wZF15Fx2So8TjcbYrWVtbQ7vd5qHSer2e/dhmZmbw4x//GBaLBYODgyfqx3MQxx70KBXemUwGm5ubKBQKSCQSnOZtNpv8fVTGKpVKiMfj3Cm0V3ysPBEdBvGAH4xyo6QUtfI+k98CmQ16PB74/X6e+STSxa8OtXTq9XpOEe+FUsJGoxFerxfhcBg+n4/LwlTy2lsa2dvBeN431+OA/KeoxEKZBqPRCI/HA4/Hc+gWcrrvNE6EymiUcVar1TCZTDCZTFzSP6+HMMqk1+t1HrFBdLvdHi8vWsNJ7EplPVo7qARLz/ReM1ODwcCNLxRs0WFtryao0Wggl8uhXq9DkiQOjPL5PHZ3d9FqtfggcB7v+3Gj9GiLxWLY3NzE7u4uqtVqT5OFSqWC3W7H0NAQXC4Xr09nyZ7k2K+i2WyygPjevXv4zW9+g3Q6zYZ2VEahl4BMjCqVClKpFDQaDfuT7K0Pizk1x0O5XObgNJlMfq++To6aJpMJIyMjuHnzJmd4PB6PCHoOgVqthtVq5Xb+vSdgSZJgNBphNBrh9/vxgx/8ALOzs3C73RgYGIDVaoXJZGJtyN6ghw4V9PUir59+g4Tkjx494k1TkiQEAgG8+eabbLtwmMylck7dzs4OVlZWWAMBPM08BINBOJ1OOJ1Ofn/OW5aUxMm1Wo39WtbW1nq+R6vVsm6SNIEmkwmBQIA1gBaLhUtXdrsdGo2GN07Si6hUKphMJvh8PhiNRtTrdQ4kyf1X+Xyn02l8+umn2NjY4M1blmUsLi7i448/hsfjwdzcHOx2uyjH70FZlozH4/jd736HO3fuoFAoIBaLodls8hqjVqsxMzODn//853C5XHjjjTcQDoc5sD0LHHvQQ2UoSnk+evQIqVSK67nKB5MWZAB8UjgIsWAfH81mkwVq5COihHx4bDYbHA4HAoEAt6TTxit4NSjTYzQaUa1Wv7fhkdjVYDDAbDZjYGAAExMTMJvNsFqtnOmhgGe/Uq+ykeA8akaOExKSKxsjKE0fCoXg9XoPVQJRlq5ItJvNZnvasmlTp9L+eXUpp8C61WqhWq1ylxqhUqlgtVpht9v5+aNsjt1u50yx0+mEXq9ng1OtVsuzmWguk1qthtlsht/vh8lk4nErZE+yN+ihGWexWAz1ep2zTalUis1VX7c776JCvlLlchm5XA7r6+t49OgRWq1WjyAfeLpOeTwe1nmGQiHYbLYzdQA+9qCnUCjg0aNHyGQyWFlZ4c6IozD6EhwtdPovFovY2tri4JSyb3q9nnVXFMGPjIzA7XbzpnuWHu7zhNlsxuXLl6HVahGNRiHLMgqFAsrlMsrlMlQqFS5duoSRkRH4/X5eTPrFxO44IH1CqVRiC33g6WdBmUyPx9NTOnxVqKGiWq1y11G5XOaSPgAWlXu9XrZ2OI9Q6bXb7SIUCuEHP/gBgsFgz9+T/5AsyyxvMJvNuHTpEhwOB4+voUCQAh2LxQKj0chlLrVaDb1eD7PZDL1ezyXBTqcDk8n0vfJWt9vF6OgoGo0GMpkMZ4Y6nQ4fyvu9a3cvNCmdbC+i0Sh32VH3LlVf6PBlNpvZfsFms0Gn05255/nYg55YLIZ/+7d/w+rqKqLRKHZ2dlCv119LjyM4eig13el0WIS4s7ODSCSCVqvFtVqHw4FgMIg///M/x1tvvQWz2YxAIACz2Sx8eF4Dt9uNDz74ALdv38aTJ09gNpuRTCYRiUSwtrYGrVaLmzdv4mc/+xkcDgcuX76MQCDAGgjBq9NoNLC0tIS1tTVsbGygUCgAAJxOJ9544w04nU6Mjo6yZuow97nT6SCfzyOZTGJ3dxepVAr5fJ47jCRJgsvlwszMDPx+/6FLaGcBjUYDm80Go9GIq1evwmw2o1Qq8d9LksTaJWpZbzQaXN6ixgfKVu7V9+zVOin/njJqyoymEpfLhUKhgFAohOXlZdaLUueYSqX6Xnaon5FlGblcjmcA3rt3j/3aNjc3UavVeiQmTqcTU1NTcDgcmJqawsDAAAeqZ41jXy0bjQaSySSb2T3Pf+RV2Bs9Kl8EejGUD/CraBjOWmR6ElCbOqWHc7kc0uk0+76QN4zJZOJW6UuXLkGr1Z6Kq+ZFQ6fTwefzsV+Mz+dDt9tFPp+HXq+HTqeD2+3G4OAgbDabGBz6mlAphnRr2WyWBeRUTnG5XJx12KuTepXfQ1YdVHZRZnkoY2Gz2VjTcl7XH6Xehgzo9mZcTCYTjEYjZFnmJhaj0Qifz3ess8aazSZcLhfK5TJsNhsHsBR8vmikUT9BgaPS3y6VSnGgSFkx4Nl+S+VIl8vFgS+J01/m9wEnt+8ee9BjsVgwMTEBrVaL9fV15PP5njr3y0BlFQBsbqXVavnmAs9umMvlwsDAAHQ6Hf+eZrPJbrfUWrfXq4d+nt1u75vpxkqjqVwuh8XFRaRSKTx+/BiRSIQ77KjDIhgMYnp6GsFgkNudD6s/UPpxaDSani6N83rSfR2UnSqBQABvvfUW8vk8xsfHMTU1BbVajdnZWRZuioDncNAE+nK5jGw2i8XFRXz99dfs2C5JEg9mDQaD8Pv9/LkcZk3odDo8ODYajfaY6pE+hdzgrVbriQ9fPEpId0adVQ6HY991ltZmEjRrtdpjz1buNYbsdrusTfR4PHC5XJyt7mfq9TqKxSIajQYePnzIvl8bGxvY3t7mUiDwtLPR5XLBYDDgypUreO+99+DxeDAxMQGDwfBSzSzkpSfLMjdqKA06j4NjD3rsdjuuXLkCr9cLSZKwurrKmZ5XCXqov5+6HtRqNbxeL7xeL4BnXjuTk5N4++23YbPZuGuiWCzi448/Zpt5Mm3b73c4HI5zfdp6FUig1mg0EIvF8Mknn+Dx48e8SFPA0+l0YDAYMDg4iLm5OXg8HgSDwe8FnK9Cq9ViQzLq2KBNoB+DHuWg23A4DL/fz8EoeST5fD643W52Dhe8OqSVisfjiMVi+O677/DJJ5+w0JjKuNPT07h06RJCoRBrRg5Dq9XCzs4OHjx4gEwmg2KxCKDXY4a6lC7CgYv0Ngd16yj/25Qn/ON+56ldvlAocJs18DTz5Pf74fF4zrWe6qioVqvY3d1FqVTCnTt38Ktf/Qr5fB6VSoXLf0o7h6GhIZ6r9Wd/9mfw+XzQ6XQvreUpl8uIRqNot9tsDUHZz3Mb9Gi1WlitVjSbTU7hkniMULbZ7u08oZPXXlMjg8EAl8sFt9vd871utxsejwc2mw21Wo3rxi9Ks2m1WhgMhjPlJ3ASUDaMjCPz+TwLLSndT4sYCTzpRHrQQ62sqyt9YZR/RiU0Sm93Oh32KzEajT1TkfsFup9Kx1LlPTObzftmHZQutPuNTKB3ijQQr1OuOe+QXw6Ji0ulErtf0/2hDjnS8hy2rEWZVBLNViqVnoMe+TKRUJo+l4sAvb+nDXUrNptNVKtVzvTQ50Cf98tmJi4qJP+o1WrI5XIoFAo8XoK0T9R8RM+q0WiEw+GA2+2G0+nk/f15v4M+D5q/ls/nkclkWN9GJVLyeyOzQ/psjuLzOZHy1uTkJHdGtFot5PN5FjVrtVoMDAzA4/HwxrrXJ4HaGQHwB2IwGDA+Pt4zMZpEgYODg9BoNHj48CEWFhaQy+Wws7PDs7f2ZpiUYkKPx8MC0X6AMjnUfki6A7pHNBndbrdjeHgYMzMz3F6738+i2nipVOJMTiqV4gAnl8vx76CXiVrddTodxsbGMDY2xn40Z8HB8zQhAzYA+778StO13d1dxONxJJNJlEoldDodqFQqFhT6/X6e4m2xWPoyW9TtdpHNZrG2tsb3qdvtQqfTwePx8Ol1cHAQAwMDMJlMh1oLqPMlm80iGo0iEomgWCxyectoNCIUCsFqtbJHD82B6teN96ihTGmxWMT29jbu3LmDe/fu9czuMpvNbElAw6b7jUajgWw2i1qthsXFRfz2t79FIpHA5uYmr9EUFOn1eoRCIbhcLoTDYfzsZz9DOBzGwMAAzGbzc39HMpnk2WobGxtsQLy5uYlWqwWj0chjRWjN8nq9eOutt1iyQoNiX4cTCXpIra9WqyHLMvL5PB4+fMgThm/cuIHx8XHo9frvWeNTpsflcgF4FvQYjUaMjY1x2Uz5/TSk9O7du3j8+DFyuRx2d3d7nFb34nA4MDExwSWzfgh6lL4tlH2jEiClnklk6XQ62RuGTkV7UYqhc7kcYrEYKpUKVlZWEIlEUK1WEYlEOACiNLPytPX222+j2+3CbrfDZDL1fdBDImZg/zJis9lEOp1GqVRCIpFAJpNBNptlsSxlz+hEFggEEAqFTn3+zWlBp8utrS0W6suyDK1WC7fbzd4iwWAQgUAAwKuXb2nIK3ldJRIJ7OzssB4RePq5+nw+dq2lAb2Co4M+a3IQXlhYwJ07d3qM9Ki85ff7YbFY+jLoofEotC//5je/QSQS2dcAWKfTYWBgAIODg5iamsJPfvITjI+Psw7neb8jmUxyR+pnn32GZDKJeDyOjY0NNJtNPsyRb5PZbMbk5CTrrcxm85E0zZxIHYfS8WQ8RaLYYrHIwxPJ64U0NUqUbpxUcqEMxH6nVdqwW61WjwfDXnW+clq10WjsERP2y8NPWZ5Wq8WLstIen06ibrcbNpuN75eyLEZGa+TGSkPpdnd32ZSSBjjm83nOuNXrdRYykqkZzXWp1+t8AqOyQz8EovvxvGdR6YJLo1r2moVR6ZbKKEoTw35C2ZVC/keUZlepVOw1QiXu17k/jUYDpVKJy2f0jik3EdLTUWMFZeb67XM5LsgYl0bqKBtY6B3Q6XT8bvRTeYvKvNQtmkgkkE6nkUwmUalUenz0lCNyyHWf9myj0Xjg4YneNdJvJhIJ7O7uIpFIcOmsVCqh2Wz2zEujgzAA7q7c3d3lSe2ve1g7MfEKOTVev34dzWYTU1NT7I/gdDpZyEoznZSQ4h94mo5stVrsyLkXZammWCwinU5z+n9vy7oys0RGe3Rj++Hhp44Gqt8mk0kkEgkODvV6Pd544w386Z/+KdxuN65evcoGhLSAJxIJ3L17F8lkkoOaZrOJeDyOaDSKVqvVs/BTJgkAb87Krq07d+4gGo3C4/FwcEsdFmfR8+G0aTabyGQyvJDQkEzlLBybzQa/38/mdwaDoe82V8pmNhoNpFIprK6ucgAOPBNlDg0NcSr9dX5XKpXC4uIi0uk0dnd3USgUej6XWq2G3d1d5PN5TE9PI5/Ps/2DGAJ7NHQ6HSSTSaysrGB3d5dLWnRgJl2oz+fru0xPu91GPB5HNpvFzs4OPvroI6yurnLHLkGHJnovvF4vfvrTn2JmZgZ2u71HU7uXVquFra0t7O7uIplM4pNPPsHq6irK5TLi8ThrbikQpXyFyNYAACAASURBVHvf6XT4z7e2tvDRRx9hfn4e169fh8vleu194EQVu1ar9bVTuM+rGwLPSiwU+JRKJRbm7oXcP61WK1wuF7xeLzweT9+0LlIkXq1WuY23UCiw2FWr1WJwcBBvv/02B4MajYa1O61WC7lcDo8ePcLGxkbPcFjK9OztknsekiShXq8jGo3C7/fjypUrGB8fh8Vigd1uF0HPPtA0dRKg0+mNoAyr0gemH8talKqnk20qleJJ6sDTg5XH48Hg4CDcbvdrNTOQq3k0GuVDl1I8CzzTYtFBgXRu1PXUD+vPcdPtdnleVCqV4nlqlNmnrtF+9L3qdDrcxbi5uYk7d+5gcXGRZ5ftlYy4XC4MDw8jEAjg2rVruHHjBttsPO93pNNpbG5uYmdnB/fu3cOjR496Gi4O8uih7I8sy5ifn8f6+joMBgO3y78OF6ZNieq0hUIB0WiUp8HuFeYCz24wiQkp2ldOCO8HaPCeUmhMUCmJFmGa/k3fl0gkWCC4s7PDmZ5iscgbLwVP5CtD2p29mTzlgEBKQdOpmP4pnFJ7oXtCM452d3eRy+V6Tk2SJEGn07Eey+/3920WoVKpIJPJoFQqsQZKmcYnIbhKpeKmCyp1mc3mV5p83m63sbm5ie3tbR67ozQjlCQJDocDw8PDcDgcCIfDLOTvx7LjUaPUKOZyOc5okG7EaDQiEAjAbrfD6/VyiaYf7j0lBMrlMra3t7G6uort7W3uLNw7WoIExUNDQxgdHeXWfqVDNvDsnpOOKpvNolqtYmVlBevr6yxipi4t2huAZ3IUOkwroeHWbrebh8++Lhdid1e2wK2uruLjjz9GMpnEw4cPkUgk0Gg0eoIeerj9fj/eeecdDA0NcScYDcy86A8/8Cz9OD8/z4Ei8EwMTkEKtSPG43EWvN25cwcbGxvIZDIsFqcHn0SCpI+6dOkSDwYk0SYhyzLS6TSi0ShPuiabc8pAabXal/Z06gdIx9NqtZBKpfDgwQP2V6LMBS0qZrMZU1NTeOedd+B0OrkhoJ+QZZk9ebLZLBYWFrC9vc16A2qu+Oyzz1hQT3ODhoaGMDExwe6yLyOibLfbWFhYwL1791CpVJDP59mSgQ5Vw8PD+Ku/+isMDQ3h0qVLCAQCbMwmeD0ajQZqtRqy2SxWVlZw584dVCoVFItFaDQa+Hw+3L59G8FgENeuXYPb7eYZXxcZmneWz+eRSqXw+9//Hp9//jnK5TIL7Un3JkkSvF4vj5Z4//33cfv2bV7DSeZA+6RSunDnzh188cUXKJVKWFtb4yGv8XgcnU6nZ74a8CxhQTYSSu0trV9DQ0MYGxs7kkPbhQh6gGdRYqFQwPr6OmKxGNcNlaIspWcJZXqGhoZYM9JPbbxk3Egp+L1lEarnkgCW7AaSySQ2Njbw+PFjFItFNrNS2tBT+pjE68FgEBaLBUNDQz0bryzLsFgsnHWKx+M9Xg5kES8yPb1QFkw5zbpYLPKCQUErNQdQe3Q/Z3qozEFTzpWZzXq93jMRHHgaOGazWXS7Xc4GvExQ0u12EYlEEI1Ge7LM5F2jVqvhcDgwNjaGiYkJOJ3OvvMHO05Iu0WZnkQiwYcEcmEOBAIIh8Os6eyHki9l0iuVCkqlEqLRKFZWVrgRQplRJ2G/1+vl8TfDw8MsaN77HtBBjIKbJ0+eoFAocBWAPhMaXq3T6aDVajnIoo7HvVCmx+/3w263H0lgeiHeMhIFVioVbG5uskK8UCh8r2NrrweA3+9HMBiEw+G48JG+EtI4kEkg+TEA4Fk45BQLgDuynjx5wt4jmUyGLd1pIQ8EAjAYDAgGgxgYGIDRaEQwGITH44HBYIDb7e6ZsdPtdpHJZJDL5VjbIHg+FKyShooMxOizIAE4mXqSiSeVUPoRnU7HJqnUDarRaDjdv3dWnyRJrAnZ3t7m8sfLBj2pVIqzSPRzTSYTpqen4fP5cPnyZV7IqTNV8Pp0u13s7u5iaWmJReTkM0MlGZvNhsHBQYyMjMDr9V74YFPZoUst4jRvjuQE9IySAa3BYMD09DRmZ2f50EpZSqVHGDVPZDIZbG1toVKp4OHDh2xXQmUzjUbD2X7ag9VqNarVKk9IoM+BDAm1Wi2cTidCoRDC4TC8Xu+RBKcX4tMulUp4/Pgx4vE47t+/jydPnrDL496yiFarhdfrhcPhwKVLlzA2Nobx8fEL5Yb6IpRuseSnQ0aCAGCz2TA+Ps4RNgmM19bW8Pnnn6NQKGB5eRnJZJJTk2RDcOvWLXg8Hty4cQNzc3N8iqJ0KNWB6Ro6nQ7PRUskEjybbe80ZcEzqKwVj8exs7PD/jzKyd10SiNDwnA43FNH7zeMRiNb3JM+oF6vo1Qq8aJPz6UyACHB86u6wVLpTHnocjgceO+99zA7Owu/34/x8XE4nc4XepwIXp5Op4OlpSX86le/QiaTwdLSEqrVKmd4yITyypUruHLlCjdsXGRIw0Nr+DfffMNdW1ROoueURrA4HA688847+MUvfvG9Ybj0HpTLZSwvLyOdTmN9fR3ffvstCoUCdnd3sbOzw+9Vt9uFzWbDG2+8gXA4zPsAlZ03NjZQrVZ7gh6LxQKz2YxgMIjJyUlcvXqVHdJflwuxAlJHBvX+ky/DflDphcyOaOxBP6HsZFF6vFCWgCJxmkZMIuNischZIZqQTDoFelCpAy4YDGJoaKjnIVWOpKDyFb0UyvZF2gRok+639uoXQf4jdJJS+lwAz8ooJECnr37dWOl+ULmV3n9JktiXSplmVy7s1OH4POiUrMzqKIMden51Oh3cbjeCwSBcLhdMJtOF33BPmm63i3K5zJkM2tTp/tOoIRLp9gPtdpsdwsmeJJ/Ps08aAF4blNPS3W436532K+22223WSmUyGcTjceRyOWQyGVQqFT4M0ywuMhmmfYD0PfuNntLpdDCZTNxhR9WZozgAX4igp1QqYXFxEU+ePEE0Gt23PZ0wGo2YnJzkevqLWuAvGiTapFr3zs4Op4CBpw+90+nE2NgYPB4Pkskk/uEf/gH1eh0PHjzA1tYWms0mNBoN3G43vF4vrl+/Dq/Xi1AohKmpKVitVoTD4e9tspRZIsPC+fl5ZDIZFpyXSiUuQ/h8PoTDYQwPDwvvkj20Wi1O4W9vbx9JG+dFx263Y2RkBNVqFe12Gw6HA5VKBTs7O2yzkM/n0Wg02C+HTNleZFZKgQ4ZqZHJJ838o0U7GAwiFAqxZb8IeI4OMlcl3zEql9frdc40XL9+HYODg3jjjTf2HaNzEZFlGevr6/jss8+QzWaxurqK5eVlvk/A0zWfutguX76M999/H36/HyMjI8816iQNbSaTQSaTQSqVQj6fR61W44BnZGQEwWAQdrsdoVAIdruddaCks0smkzwNAHhajQkGg7h06RKGh4fhdrthsViObBj1hQh6isUi5ufn8e2337KA7SBMJhNmZmZw69YteL3evgt6aPbQ5uYm4vE4tra2EI1Ge05CHo8Hk5OTCAQC+Pjjj/Ev//IvyOfzXBemEqHT6cTly5fx13/915iamoLJZILdbmf9w95SCtmdZ7NZ3L9/H//4j/+IjY0N9gqikSOBQADBYBDDw8MYHR3tu8GjL6LRaGBnZwePHj1CKpUSOqiXgMY8tNtteL1eXLlyBeVyGYuLi9ja2kK5XEYkEkGhUIDJZILL5eLTpsViee7zRxke6oQsl8ts3aBWq+F0OuHz+Xju2dDQ0L7vh+DwtFotnm2WzWaRSqVQKpV4L7DZbJibm8Obb74Jv9/fV0HPysoK/vmf/xmxWAz5fB6FQqFnxITBYGD3+5s3b+LnP/85BgYGuPR3UMBPFiY0XoI6ewHwPLvx8XHMzs7CYDCw1xqZE6bTaaTTaSQSiZ5SMPnDXb16FaFQCD6fjz+vczFw9LigkghZXFcqFT7Fkc6EUHYV0QRlm812JHM8ziPKmVv0RYu6ckI6lb/o3lJLOgkC6URMgllKHe/dIKik1Wg0uHOADPVIw0POzNTtRR0tz3vpLjL0/CpNvEijVigUesq4+xlA7i1hNhqNnlIh1dX7BXpmycmdspUulwvVahUGgwGVSoUDHZfLBa1W+8KgR/kuZbPZfU/FFODQdOq97b6Cw0H3XpblHh8masqgUjkd5miGoLJduh8gTQ/pevaOQ1Gr1VzypdKfstkE6N0XaN/NZrMoFosolUr8Z51Op8f2wWw2c2mdureoLE+yClrjqAxP+wlZpRz1WKhz+8mXSiV8+eWXWF1dxfr6Opdo9gY8ALgU43A4MDIywl8kbOs3KOqu1Wqw2WywWCw9QQ55HDmdTmQyGbhcLhgMBu4Y0mg0sNls8Hg8vIgclAZVTjrO5/PsCbS+vs4DYMkW3mg04vr163jnnXfgcrkwODjYlxsD6Z3o3qXTaTQaDSQSCW65vnv3LtbW1lCpVL6X2aRNIJ1OAwAikQiWl5dhMBh6XJmPYmLxeUOSJA7MyYV3dHSUMwVU3qJRHUoR/n6QXqLVaqHZbHK2WbkGUbBFGct+CjaPEzrwNhoNzM/P46OPPkI6ncbq6ipbDdAIFmXTCs3a6kcocFGi1Wrh9/sxNDS0bzcb6dqazSby+Ty++uorrK2tIZVKYWFhgTM89XodarUab7zxRk8TS7lcRq1Ww9bWFvL5PJe1arUa64o0Gg0mJiYwNTUFl8uFd999F1euXGE3/qPk3AY9lUoF9+7dw6effsoptoO0PCqVCg6HA8FgkOvqoVCoLzdUKmORkI8E3UodQi6Xw9raGiwWCwqFAux2OwwGA8/RooyMw+HgjNleZT9Bbb800G51dRUbGxuIx+M97YyUfZuYmMAPf/hD2Gw2eL3eU7pLpwt5XrTbbWSzWWxsbHCnxPr6OkqlEpaXlxGPx9m9WgktUoVCAWq1GrFYDFtbWzCbzfD5fLBarfyZ9VvQA4BPogDg8Xh6hMjEy64NdJ8bjQYikQgPTNzrAE8ZNiHKPzooUC2Xy3j06BF++ctfIhqNcpZTr9fDZDLxmj84OMjdQ/3IQV5n5NoeCAR4IPhems0me4J98cUXbGq4u7uLarXK2WidTofR0VF8+OGH0Gg0WFlZwfb2NjKZDObn57G5uQngmSEhXROVtG7fvg232425uTlMTk4ey/ty7oIeMqwj1TjNHNobvSohrQgNXaQNuh8hw0EaDUF+CGQaBTx9wMk7gbqDlO3/yin3arWaW38BfK9MRvb+29vbLJ7O5XIolUo9RnqU6TGZTKza76cNWZlpoDJgq9XCzs4O+18op9XTCYnKuXuhhb/RaCCdTiMSicBoNKJUKnGwS+8CGY6RNw0tYMo0OD0blC2irN55f49e97+BuoUoZa+cck9ldavVCq/XyzohwdFAc+eo1Ev3nrr1lMZ25BN23p/Xo4LWcK1Wy11VJBimdZfKtu12m4XKsVgM2WyWszf7jQkql8uIxWLQaDSs9cnlct+bjACgR3ZitVrhdDpZKnFc6/+5Cnq63S5yuRxyuRy2t7extraGtbU1doI8CDJaeu+99+DxePo2gwA8fdgtFgsvvl6vF16vlyeh1+t19uEhx0wAPMuGRkyQaZTJZEI0GuUJxhT0KGe8fPHFF5ifn0etVuOZXdRpATw1xPL7/bxAUSdBP20QuVwOX375JaLRKMrlMtLpNLubbm1tcSBEAX6lUuEyyn5BD5V62+02Pv30Uzx+/JjLktShNzIywp1yQ0ND0Gq1PZt3Nptlp20azXLp0iXcunWLh8/202e0H+VymT3ClpaWegw21Wo1TCYTLl++jNu3b8Pj8cDj8ZzyFZ9/KEtQKpWwsrLCmUyarUVuv3a7HVeuXOGByf1+75UBvk6nQzAYhNvtxujoKGZnZzEzM8NZYOBZdqdSqeDLL7/E119/jVwuhwcPHvBICeWsP9IdPnr0iN8BkjaQ6Hnv9SiHvo6NjeHatWsc/BwX5y7oqdVqPNAsnU4jmUzy3x8UxZMb5OTkJIur+hlqw61Wq6zHIVda4Kklf6PRYE0D1b+pjEKLud1uh06n41ZfJZSxKBaLePjwIb755hue4Lt3LIhWq4XVauUOm6MyoTpP1Go1bGxssH071bwTiQSi0ej3SlgvgkpkzWYTGxsb2NjYgEqlYu2V1+tFNpuF3W7H8PAw+/oUCgX+POlUpwx6ut0uZmZmWNy718yv36jX60gkEtja2kIikUC1WmWRtFqthlarhc/nw/j4OOx2e9+vPUcBBfr1eh2pVArRaBTZbLbHc4Z0WaFQiNf9fvHl2Q/lO0oaM5vNBp/Ph0AgwKU/pVlju91GvV5HuVzG+vo67t69i1KphN3dXRSLxZ6frTT3jMfj3MVVr9e/5yGmhDzhzGYze1iRJ89xcS6CHuo+oYWYRk08r1WXSloOhwMej4cHXZrN5r5S7j8P2vxGRkbYa4E8FpSdXbSYUNan1WohFov1qO33nvhJ2Fmr1ZDJZPhUQD/DYDBwup8EhjQQs1820W63y5skpY53d3dRLpfZIp6ya89jv/t10L9DnyXV50lX1W63odPpOOPXbDb5lAaAy1lutxs7OztotVpwuVzwer19VYbcS71eRzQaZWEn3V+TyQSr1coGby6Xi71GBIeHdG6VSgXb29uIRCKIRCI830mtVsPj8WBwcJCbIZSdoP1Cs9nklv1cLtfTtUXrOo2nUGaRaS3odDrIZDKcmScRcrVafeEBTNkODzwrYdHMLtJakVknzQUcGBiAwWDg7z8uzsXuXyqVkEqlUC6Xce/ePTx48IDFywehVqsxPDzMU3TJApv0LIKnplRTU1PQ6XSIRCJsEEgdKXt1HVRGqVQquH//PpaWlvjUsHfjVZbFyMFZ+TK4XC7cvHkTbrcbExMTmJ2dZVPDfglK2+02C7xXVlbw4MEDzM/Pc0aMFqD9WtIJ0o0o77/S+VqJUiDdbrdRLpd5ESK9Ay2Ge3+3MgvodrsRCATYrr6fg558Po87d+7gzp07qNfrqNVqUKlUcDqdGB4ehs/nw8TEBMbHx3uypoLDUa/Xsby8jJ2dHayvr+MPf/gDNjc3Ua/XuXtoenoaP/vZz+ByuXDjxg2MjIz0XSm2VCpxuXVjY4NtXShgoWYHcmlOp9OIx+M8RqJarfJQ6Uqlgkgkgt3dXQ6Ungd57iid4bVaLTweD0wmE5fWbTYbwuEwpqenYTabEQ6Hedp9Xwc9VCekSJSiz0Kh8FwdD2lXSMTmcDg4td/Pi7QS5ZDQSqXC0beyZLE3ageeOSvvrdE+D/p59E+9Xs+bJ3VXWK1WWK3Wvsn0dDoddkalki212r7sZHlaWA5ib2eSMnv3vPfnICgjSPqfl7nGi0yz2WQLfoKCG7vdDpvNxqUVse68PiReTqVSSKVSbIxH0H0Ph8M94tx+WVMIaowg0TGNW1EehGgdaLVaqNVqKJfLyOfzSCaTKJVK2NrawtraGqrVKjdQPK9hSAll3eiL/K/IK4ncmclKgP7uJHzZznzQAzy1d9/c3EQ+n8fW1ha2trZQrVb3td/XarUsZAsEApicnGSzu/0yEv0MCZKBpxqcS5cusWYqHo+jXq/3iNVeFdLr+P1+OJ1OaLVaFlEPDAzg1q1bPbOI+sE/Q5ZlNgorl8t48uQJNjY2sLOzg2Kx2NM9dRCkzSGxZjgcZr0Cdc2RnofKAcVikcXotPjRl7J1VPBi2u02p/oTiUSPnk3Z5eZwOGC327kzTnA4qAu00WgglUpxVjSZTLLEgXzDzGYzhoeHeVOlGWv9htLMdK88AXga8FAHbbfbxWeffYaVlRWeW1av15FMJpHL5dBsNr83PBdAT5Zfqfsjk1qlWNpsNmNgYAAOhwMWi4WbYLxeL2tDj9qE8CDOfNAjyzLS6TQWFxf5nysrK9yeuBfaWE0mE0ZGRnDz5k0eoHbQDJF+hQISj8cDnU6HK1euQKvV9uilqMzyqpuicsjj2NgYZmZmeh58r9eL6elpHltBZnAXoQ36eZBz7/b2NvL5PL799lssLCwgn88jk8m8lGBZq9XC4XBwOfDHP/4xBgYGeoKYarXKX48fP+ZxHzSEsdPpcIeXUrcleDHNZhPRaLSnsw54FvDsHb4ryumvBx0U8vk8dnZ2cO/ePXz22WfcaQo8nWB/5coVOJ1OzMzMYHx8HGazuW9n9inNZpVTCmgdp2w9zYvb2NiARqPhEjh9j3Io9F6UOh0aDq2cyWixWHDz5k2exzgyMgK3280NMhQ0UXbnpEw7z3zQQ+MLyA2Y7KsPWqRpoyW/F6vVCrPZfGJR5HmDHlTqxnI6nSiXyzCbzVzWepGuRImyJZL0UyQmV7amu91uOJ1OLmf1k0sttYKWy2Uejqistx8EBYQajQZGo5HdSj0eD/x+f0+WiFpNKRuhnKqs1Wo5pa3U77wMNEqhX+eh0QmaxqiQBk5ZvlV+RkajsW80ascF6U9ofA2NYaENmdZ85egCyhr3a0mRDp3kV0T/m3R7wLOmBlqPgP0bIJTBvDKjQ2NaaA/R6/XsjURO/bQ2WSwW+Hw+uFyuE7oDB3Nm30Y6idJYhNXVVWQyGdY8HITdbsfU1BQcDgeGhoa4Rb2flPuHwWazYXZ2FkNDQ4jFYgiHw+zXs7CwwEZUz9uYacYNichHR0dhtVpx8+ZNXL58mevtFJRSdqefIDM7Mhrc3d3F7u4udyceBI1NoIndV69exeDgIEKhEKanpxEIBPgkRwJ0SkuHw2Fu6a3Vamg2m3wNJL6l7M+LuHTpEnfaOZ3Ovvn8aOOlEsv9+/fx8OFDFn8CzzYassi4cuUKXC5XX3UkHgfNZhMLCwv46quvkEqlsLGxwYJxEr5OTEzgvffeg9/vx/DwMK9D/fJ87sVqtWJiYgKhUAjVahU7OzvIZrOIxWKIxWIv9a4TlNEhXQ51xhmNRvh8Pp7b5XA4eubV6XQ6DA4Owuv1spntWeBMBz2kyE8mk1hbW0M6nUa5XH5uqcXhcGBychJerxfhcJijfrHoPB+r1YobN26g2+0ilUphdHQUxWIRv/vd7xCJRPj7nhf00ANvNBoxMzODH/3oR3A6nbhy5QpGR0d7ylf9lt0hyFSN5mhFo1Hs7u5ymekgKOjx+Xxwu924ceMGpqen4Xa7MT09/b0TlFLArExt05+n02ksLy9zBjWXy71Utsfj8WBkZARms5l1cv0A6UpKpRKSySQePHiATz/9FNVqlYMeStdT0DMzMwOHwwGHwyHWn9eg2Wzi0aNH+NWvfoVKpYJUKoVqtcqbq9lsxvj4ON59910MDAxwlrmf77nFYsHExAQ6nQ6q1Sqi0SiSySS63S4SicQrBT3k3kwWMNevX8fo6Ch7fJE4ORgMckVFucYrR7CcBc5k0ENCTPIMIfX5QToemqRLE5TpFEpjEvr54X9ZKCUPgCenq1Qq+Hw+DA4OolAosFHdQa3U9DMo+CFLcVFefAZtnlRPp66Kg6B6uU6nY8dqyh7Y7XY+Ub1q8EGlMfr5KpXqpRZC+jyNRmPfbSzUrttqtVCv11GtVnuGi5KlP5XV+73E8rqQJoU6d6lc2+l0WBfidDpht9vhcDh4tI5Y859lHVUqFcxmMzweDyRJQiaTeeEBh55nZcmW1hq73c46UKvVysJkkpGch4rKmQp6qD1almUkEgncu3cP2WwWT5484Rlb+y3MJJC1WCyYnZ3FD3/4Q/h8Pvh8PrHgHAKz2YzBwUG0Wi2YTCZMTU2hWCzi66+/xt27dzldSq6bBNXWjUYjQqEQZmZm+HTQ74sQ0el0sLm5iS+++AL5fB6JRGLf76P7RffPYrHg/fffx9tvvw2LxYKhoSG43e5Dp43NZjMuXbrEZTASNb8IvV4Pi8UCjUbTV+VJWZa5szGXyyGfz6NQKPTooQYHBzE3NweXy4Vr165xCeA8bARnDWpg2d7eRjabRSQSQSaTQbvd5o18aGgIH374Ibc92+12EfDsQZIkhMNhfPDBB6jX63j33XdfKBHZ72dQ9kyn08Hr9XJ7udVq5T8/L3vtmQp6ALAYM5/P8yRpGrh4kOaBrN5pjsjly5fh9/tZvCV4NchpGQACgQBmZ2dRKpUgyzKfFGjhV0IvB7ktk9mU4BmdTgfJZBKLi4ssgj0ICiIpc3n16lX85Cc/6WkJPSx6vR4+n+/Q/36/QZoeshqg7IOy1E7BTiAQwOjoaM+MO8GrUygUEIlEkE6nkUgk+H6bzWbo9Xp4vV6eGUU+bP0ShL8K1EVIHMaeYm8geZ4DyzMV9HS7XS5hlctlpFIptsHem+FR1gxNJhMb3TmdTuh0Oq4lCl4PerjJUXNsbAylUgkajQbBYLDne6lN12w2iyzbAZD40u/3w2g08sR5JXq9nkuC4XAY4+Pj3Oav0+mO3aZdsD/UXkufgUaj6clOK92thT3G4SF9G2XWlLP91Go1XC4Xm6pSo4rwQno+ynvT7/fpTAU9nU6H54VsbW3hu+++QyQS2fdETOIqnU6HQCCAt956C+Pj4xgcHOT5Wv3+4R4lOp0Ob775JoaHh9FqtfbNvCk3hEAgcOGNBg+DWq3G6OgofvrTnyKTyeDOnTtYWlrq+R6n08lDEm/duoX3338fNpuNWz9JMCs4Och0jVzDzWYzTCYTjwxpt9uclXO73TCZTGL9OQSkm2q325wRzWazyGQykGWZvV+mp6cRDocxOjoKn88nnPYFL82ZCnrIGIms+aPRKLa3t/f9XhKyabVamM1mhEIhDA8P8xBLseAcLWq1GsFg8HvZHcGroZzLZDab8fjxY55OTBgMBi7XTkxM4ObNm6JMeMqQSJ/EyRTc08gWpRlnP4q8jwrKnFG2XzkIGQA77Y+Pj7PTu5hcL3gVzlTQ02q1kE6nkcvlkM1mn9tNYjQaMTk5iWAwiNHRUYRCIe4sERG/4KyiUqngdrsxNjYGj8eDer2OoaGhnqDH5/NhfHwcVqsVg4ODwtzuDEB6NSrhzs3NQa/Xc6ZHlmVcRpDqSAAAIABJREFUvXoVAwMDcLlcItNzSOr1OlKpFCqVCnZ3d9kHSaVSIRAIwO/3Y3BwEOFwGA6HQ2STBa/MmVpNabLr9vY2IpHIc83abDYbfvrTn+Ltt9+Gy+XC5OQkT3wWQY/grELGjX6/H+12G+++++73Bn/SWA6yYBAL++kjSRK36ttsNvzN3/wND1ylThgyaCPbBqG7enUKhQKePHmCXC6Hhw8fYnFxEdVqFeFwGJcuXUIoFMLs7Cxu3rzJ74lA8CqcqaCn0+nwqPsXTXTVarXwer0YHh6GxWJhE0KB4CwjSRKPJxCcL+hART5UgqOn1WrxiI9CoYBSqYRGo8FBp9VqZRsHgeAwnLmgp1wu8yyi/dp5qXOFRhp4PB6e+SEQCASC80u5XMb29jaPaaHOuEKhgFgsBp1Ox/oegeAwnKlIodVqoVAoIJ1Oo1gsfi/TQ2JBGmTpcrkQCAT4zwUCgUBwfqlUKlhbW8P29jYSiQTa7TY6nQ5yuRyazSY0Gg0qlcppX6bgHHOmgh6l5b4yiKF5NpIkwWKxwGQyse24yPAIBALBxYBa1mnsEAD2YyNvJHHAFbwOZypioOnRbreb23mLxSKcTif8fj/MZjOuX7+OmZkZuN1uTE5OnvYlCwQCgeAYoSnq09PTCIVC8Hq9p31JgnPMmQp6zGYzpqamMDAwgHK5zGJPmubqdrvxi1/8An/yJ3/CPhkCgUAguLhotVoMDw/j9u3bcLvdcLvdp31JgnPMmYoaSKDc7XbhcrkwPDwMrVaLcDiMwcFBOBwOOBwOtnkXCAQCwcXBaDTC6/Wi3W7zlHCtVotQKAS32w2HwyHmmQleizMVOWi1WjgcDlitVty+fRsulwuVSoVH1+v1egwODoqarkAgEFxAhoaG8Jd/+ZeoVquoVquoVCpQqVQIBoPw+/3Q6XRwOp2nfZmCc8yZCnpoeCgAWK1WjI+Pn/IVCQQCgeCkEOUrwXEjLEMFAoFAIBD0BZJy5s8Lv1mSUgAix3c5F5JLsiy/druBuPeHRtz/00Xc/9ND3PvTRdz/02Xf+/9KQY9AIBAIBALBeUWUtwQCgUAgEPQFIugRCAQCgUDQF5zroEeSpClJku4rvoqSJP3H076ufkKSJIckSf8sSdITSZIeS5L09mlfUz8gSVJYkqRPJElalCTpkSRJ/81pX1M/IkmSWpKke5Ik/X+nfS39hFj7T5fzvO5fGE2PJElqAFEAP5BlWYi+TghJkv4PAJ/Ksvz3kiTpAJhkWc6f9nVddCRJCgIIyrJ8V5IkK4DvAPznsiwvnvKl9RWSJP23AOYA2GRZ/vPTvp5+RKz9J895XvfPdaZnDx8AWBMP/ckhSZIdwI8A/G8AIMty87w8+OcdWZZjsizf/eP/LgF4DGDgdK+qv5AkaRDAnwH4+9O+lj5HrP0nyHlf9y9S0PNfAPi/Tvsi+owRACkA//sfU/x/L0mS+bQvqt+QJGkYwCyAr0/3SvqO/wXAfw+ge9oX0ueItf9kOdfr/oUIev6YXvsPAP7ptK+lz9AAuAngf5VleRZABcD/cLqX1F9IkmQB8P8A+I+yLBdP+3r6BUmS/hxAUpbl7077WvoZsfafCud63b8QQQ+A/wzAXVmWE6d9IX3GDoAdWZYpw/DPePoyCE4ASZK0eBrw/KMsy//vaV9Pn/EOgP8gSdImgP8bwE8lSfo/T/eS+hKx9p8853rdvyhBz38Jkd48cWRZjgPYliRp6o9/9AEAIaQ9ASRJkvC0pv5YluX/+bSvp9+QZfl/lGV5UJblYTwtr/xOluX/6pQvqx8Ra/8Jc97X/XPfvfXHWuIWgFFZlgunfT39hiRJN/BUyKkDsA7gv5ZlOXe6V3XxkSTpXQCfAljAM03J/yTL8r+e3lX1J5Ik/RjAfye6t04WsfafHud53T/3QY9AIBAIBALBy3BRylsCgUAgEAgEz0UEPQKBQCAQCPoCzat8s8fjkYeHh4/pUi4mm5ubSKfT0uv+HHHvD8d3332XlmXZ+7o/R9z/wyHu/+kh1p7TRTz7p8tB9/+Vgp7h4WHcuXPn6K6qD5ibmzuSnyPu/eGQJOlIXFrF/T8c4v6fHmLtOV3Es3+6HHT/RXlLIBAIBAJBXyCCHoFAIBAIBH3BK5W3TpJut4t2uw1ZlqFSqaBSPY3PVCoVnvqyCQQCwenQ7XbR7XYhyzL/U4lGo4FGc2aXV4GgbzlTbyUtIN1uF9vb27h79y5KpRI8Hg9CoRD0ej38fj88Ho8IfAQCwalQr9extbWFbDaLQqGASCSCYrEISZIgSRK0Wi1mZ2dx69Yt6PX6075cgUCg4MwFPe12G51OB2tra/inf/onbG9v48qVK3jzzTfhcDgAAG63WwQ9AoHgVKjX61heXsby8jK2t7fx7//+74hGo5AkCSqVCkajEX/3d3+Ha9euiaBHIDhjnKmgB3iaNu50Omg0GigWi8jn88hms8hms5BlGbVaDe12G2q1WpS6BALBidPpdFAul5HL5Xq+qAzfaDRQr9fR7XZf/MMEAsGJcqaCnm63i0ajgWaziWq1ilqthlqthq2tLbTbbdjtdlitVi51Wa1WcZISCAQnSqPRwObmJu7fv49MJoNSqYROp9NzAGu322i1Wmi1WnxAEwgEp8+ZC3qazSbq9Tp/1Wo1xONxpNNp2O12XL58GblcDmazGXq9XgQ9AoHgRGk2m9jd3cXS0hJqtRqq1SpkWe4RM1Opvt1uc9lLIBCcPmcq6Gm1WshkMigUCshms2g0Gmi1WrxoULqY0siitHU+oA2h0WggnU6jUqnAaDTC5XJBp9NBrVaLTpcjoNFooFarodvtQqfTQafT8bsiNt3Xh55j6ihVq9VQq9W8DlEXF5Xh8/k8AMBms4nn+xSQZRmtVgvtdhvNZhPZbBbVarXne6xWK5xOJzQaDbRaLbRa7Sld7cWj2+2iXC6jWq2i3W6jWq2i2Wz2rP0ajQY6ne5E9/Iz9Sbm83l89dVXWF9fx+rqKhKJBJ+i6GHU6XQwm80wGo1iITkHyLKMZrOJdruNaDSKX/7yl1hYWMD4+Dg+/PBDhEIhWCwW2O12qNXq077ccwlttslkEsvLy6jVaggGgwiFQvy+GAyG077Mcw9lbtrtNgwGAxwOB9RqNbLZLIBnmWpJkrC9vY3vvvsOHo8H09PTMJlM4pB2wnQ6HaTTaWSzWcTjcfz2t7/FkydP+O8lScKbb76JDz74AC6XCx6PBx6PRxwQjohGo4H5+XksLS2hUCjg8ePHSCQSGB8fx89//nOEQiE4nU4EAoET3cvPVNRQr9cRiUTw6NEjxGIxVCoVNBoN6HQ6Th2r1WqOyMXDeT7odDpo/f/svdlzG2eW9vkktsS+bwQBrqI2ypZlyyqXrSqHq11bd3R/UdF9Md03fTG3M/ffX/Ddz8XE3MzSNxM9UVFREVXdruruclXZ7bYWy7Y2ipIo7gCIHcgEkEgkkADmQj6vQRKURFkSAPL9RShsUSCRzEy8ed5znvOcdhuSJOHrr7/GJ598gkqlgtdffx1OpxMmkwlut3vYhzmWUGaBdlXJZBL1eh1GoxEejwedToeXgF8AdI51XUe324XZbIbNZoOmabvWoV6vh06ng2q1ip2dHWiahng8PsQjP77QZ6JUKiGZTOK//uu/cO3aNfbvgiBA13UsLi4CABwOx7AO9Uii6zoymQzu3buHQqGAzz//HJubm7h48SJOnz4Ni8UCs9mMTqdzfIOedruNSqWCQqEAWZaZOFAURXi9Xni9XjgcDnayeNAz+rTbbaRSKezs7CCZTKLRaLDSi8PhgNPphCiKfBd8COjB2ul0oCgKstksFEXBxsYG7t+/j0ajgW63C4PBAIfDgdnZWTgcDn6OvwPdbhf5fB75fB6ZTAY7OzuoVCpQFAW6rg/78Dh9kB5UURSsrKzg4cOHSKfTrNzIeXn0ej1UKhUmU1lfX0cymYQkSWg2mwAAVVWRSqWYrGFqauqVHuNIBT2qqmJrawtLS0usDmswGODxeJBIJBAMBhEKheB0OnnQMyaoqopr167h008/haIoyOfzsNvt8Hg8CAaDiEQivLvlkHS7XTQaDWiahmQyiT/+8Y9IpVJIJpNYXl6Gpmk4ffo0tra2EAgEYDabMTExwYOe70Cr1cL9+/dx/fp1FItF3Lp1C2tra9B1HZqmDfvwON/Q6/UgSRKy2SxKpRJ+97vf4U9/+hNUVUWhUBj24R15Op0OHj16hOvXr6NcLuPatWtYWlpCq9VCrVYDAJRKJXzxxRdYXV1Fp9PBuXPnYLfbX9kxjlTQ0+l0UKvVIEkSW6CpnOVyueB0OmG1WmEymbj+Y0zQdR3FYhEbGxtotVpotVos02O1WpnWhD+QDwdtCmq1GtLpNNbX15FOp5FKpaBpGhNokoBw75gEzuHodruQZRk7OzsolUool8tsER8E3c/8vn61kO1JtVpFpVJBKpXCo0ePuGfSK6LX67E1qVgsIpvNolAosEkLgiBA0zSUSiW0Wi1W0XmVjETQ02630el00Gw22eLc6/WYrXswGMTZs2cRCAT4CIoxgG78arWKYrGITCaDfD4Ps9mMSCQCl8uFeDzOA57npNVqscVkfX0dW1tbLIVMmpNarYZsNoter4dqtYpWq8XmQfHz/XxomgZFUVj5cBCCIDA9VTweh9/vh9PpfMVHevxoNpuo1WpMPHvjxg0Ui0Ukk0ke8L9Cer0eZFlGMplEpVJBtVrdd/4tFgtcLhe8Xi9sNtsrz/IPPeihjgdN06Bp2q6ojwKfWCyGd955B6FQCLFYjJdCRpxer4dSqYSNjQ0UCgWsra1he3sbgUAAb775Jk6dOoWZmRne0fKcaJrGOhw3Nzdx//59bG1todvtot1uAwCbCaWqKsrlMprNJmsN5V2Ph6fX66HZbKJaraJer6PVau17jcFgYC24oVAIJ0+eZFpEfp+/XBRFQSqVQrVaxSeffIJf//rXqFarPMv5iqG1/+HDh8ypfG8mx2KxwOv1IhwOw+12H6+ghzoiSHSmqio7QZTlMRgMEEURbrcbbrebd6KMAeTJU61WUa1WoaoqC2htNhtcLhfsdjsvUT4nlMKv1Wqo1WpQVZWJBIHHnx0a5dJsNlkmlaf4nw9apzqdDivRDnqQ9nsiUfnWbrfzIPMlQnYNzWYTsiyzsUWFQgGKoux6LT1T+o0ku90uM5hst9s8QPoO0Oek1WqxNWlQ6cpgMMBqtcJms71yjx5giEEPWbTX63Umatre3mZiM1EUYbfbYbPZEAgEEAgE4Pf7YbVa+a5pxOn1eshkMvj6669ZXZcM3VwuFwKBADds+w5QK26lUkG9XmfZHYKCTgCsrZq6vfiifjja7TZarRbq9TokSUKxWGRllL2YzWZ4PB7Y7XaW4XG5XLBYLEM48qOPpmksi7m0tIQ//vGPKBaLWF5e3peJM5vN8Pl8sNvtUFUVlUoFrVYLa2tr+O1vf4tAIICf//znSCQSfDP2HNAkBRofRYPDB220HA4HpqenkUgkEIlEXvlzYChPHXLKpNT7F198gU8++QS1Wg35fB7Atykwh8MBv9+PQCAAn8/HA54xoNvtsqCnVCqhUCig0+nwoOcFQa3qkiSxuU97oRl2drud7bhoV8x5dtrtNhRF2RX0UOZyL6IowuPxwO12w+v1wu12w+l08jXrJaFpGvL5PGRZxs2bN/Gb3/wG2WyWGUj2YzKZEAqF4Pf72Wah1Wqxlmq3242pqSl8+OGHvJrwHLTbbVSrVaZ5o6TGoKDHbrdjamoKJ06cOD5BDwBm9NVqtaAoCqu/0gJOhnUulwsOh2OX3TtnNKH0ZrvdZqJPivoBMF8Gsh/n1/NwUFqeRP+Kojxxmjddj/7v5Twb/WUTEuXTYt5f3iLhMnkihUIhNhiZr1kvHpJDkKdbLpdjvjCNRmNfmdfpdMJms8Fut7ONc7vdZg9aKhU3m03ut/QdaDabKJVKTNJAWeX+NYdGR1FHNo0gOhblrf4aLJkRZjIZJmgGgGAwiIsXLyIUCmF+fp7PRBkDNE1js1by+Tyy2SxkWWYiWtoJBwIBOJ1OnkY+JPTApe6Ihw8fMk0P58Wh6zoL1snwsVKpYH19nXXC0QOSOlHMZjNef/11/OxnP0M4HGaOs5wXi6qqWF5eRjqdRiaTwfXr15HNZpHJZFCtVne91m63491338X58+fZdRJFEQ8ePEAymYQsy0P6LY4em5ub+Oijj5DNZnHv3j1Uq9VdjUkGg4FNUrDZbMdT00MZHvLlKZfLuzQHbrcbCwsLiMfjiMVivBQyBui6zq4pBbS0EAmCALPZDLvdDpfLBavVyoOeQ0BTuzVNQ6PRQC6XQzKZZNlSzouDROCapiGXy+Hhw4colUrIZrNseCJBGR6bzYa5uTlcvnwZk5OTXMD8ktA0DZubm1heXsbW1hb+/Oc/Y3t7e+BrLRYLzpw5gw8//HDXtaChl5wXRz6fx9WrV5FMJlnWrb/sTuJ+s9kMURQhiiLL9Lxqhpbp0XUdzWYTmqYxbxFBEFjk53A44PV6mfiMp4lHH9I/UNml1+vBaDQyQTpdS7r5+TU9HP1ztqibiIuTXzyNRgPpdJq1QWezWUiShHq9vq+UKIoiQqEQ3G43AoEA273ygOfloOs6yuUydnZ2UCwWB2qrnE4nnE4nfD4fAoHAvmHGNpuNb7heAGR82m63IcsyGo3Gvk0BYTAYYLfb4XA4mNbNbrcPZQTR0IIeRVFQKBSY+p48LjweD6xWKxKJBE6fPo2ZmRk2lJIz2tDDQpIkSJKEXq8Hk8mE2dlZTE1NIR6PIx6Pw+v1wmAw8IXnkPS3TmuaBlVV2d85L450Oo2PP/4Y2WwWq6urWF5eRqPRQLVa3Rf0eL1eXLx4ETMzMzh16hSCwSCcTifTL3BeLI1GA3fv3sUf/vAHqKq6b56WwWDAzMwMFhcXEQwG8cYbb2B+fn7XWrO1tcVLjy8ARVGwuroKSZLw8OFD5HI5lMvlfX57AGC1WjE1NYVwOIyFhQUkEgnEYjGYzeZX/hwYass6RYZ0goxGI2w2G/NyoVlbJpOJGxKOATQWgXQ8FPR4vV7EYjFMTEywujrn8FCmh8TMz1LW4g/ew1Ov15mh5tbWFra2tg6cr2W1WjE5OYnZ2VlEo1HY7XauP3yJtNtt5HI5bGxsDPx3mtU4MzODYDCIaDQKn8+36/nB9YQvhlarhXK5jFwuh1KpxPScB3lY0bxFv9/Psj3DYGhBj8VigdPpRL1eZ1kcekDSH6r5vciAR9d11Go1NolXlmUWgNVqNfR6PSa0EkURExMTCAQCTIjFg6/dkBlVp9NhIrZisYhyuQybzQaz2YypqSmcOXOG7YI5Lx+qm5N+iubV8SBoMGSjQWZ19Xod9XodmqbtW8TNZjO8Xi9rvZ2cnEQsFoPP5+MP05eArutIpVLIZDLY2tpCqVTa9xqLxQK73c6qBKdOnYLP5xvohm2xWODz+eD3+9nX3G43l1EcEnqWUqv6oIwzrTl2ux0TExM4ceIEYrHYUDe+Qwl6SLMTDAbRbrchiiK63S5EUWQLSDweZ7umF3kjapqGra0tFItFFAoF3L9/nw0SXFtbQ6fTQSQSQTgcRiAQwI9//GO8+eab7AHC06K7oXpus9nEgwcP8Pvf/x6ZTAYWiwVutxt+vx9vvvkmfvzjH8Nms8Hr9Q77kI889PmiWXV033Id1cF0Oh00Gg20Wi3WUZrL5VCr1faVtKxWK+bn5xGPx7GwsIDz589jfn6eDdLlvFg0TcP169fx8ccfo1QqYX19fd9rHA4HEokE3G433nzzTXzwwQfM7mTvPW+z2TA5OYl6vb7r+7kP3OFotVooFotIpVLM7HGveJna030+H86dO4dLly7B5/PB4XAM7biHFvRQ5qR/YjqJnchb4UV6udBujVygaWFLpVIol8vY2trCw4cPoes6ZFlGvV5nBnCapkEQBG7jPwDaIVNpK5fLIZvNIhQKwel0ssUkHA6zlkXOy8dsNrMp9lQe5jqTg6GMJWWA+71b9n7uDQYDnE4n/H4/fD4fPB4PPB7PM71P/0DlvV8j+q/RcZ7W3j/WoFAoYHNzE7Is7xsvAXxrHeDxeOD3+xEKhQ58sFJzhdPpZOXiF/28OcrQ/UrWDuSPNOj5aDQaWeBD18blcg1VozuUd+52u6hUKtja2kI2m2URt8lkYqp7EgO+CDqdDmufLhQKuHbtGtbW1iDLMjY3N9FoNFCpVJghGU2o1nUda2tr8Pv98Hq9EEWRTQbnPIbMvVRVRavVYjd+NBrF+fPnWV2dBGt8UXn5CILAxrdQSZE2GPz8D6Zer+Pu3bvIZDJYWlpCPp9HvV5n2rR+jEYj/H4/JicnEQqFnjm7QxoIRVFY5wtlmKgzjDYKRqMRDoeDlSl9Pt+xWnt6vR42Njbw4MEDlMtlfP3110ilUvvmzBGxWAx/8Rd/gUgkgjNnzjxxcyWKIgKBAGq1GstUiKLI5A2cg2m320y/s7q6inv37mF1dRWFQmHfOByDwcA6sGOxGEKhEAKBAERRPJ5BT6lUwsrKCgqFAmRZ3tW9RSn5F3VidF1HNpvF9vY2UqkUPv74Y9y+fRudToeZkJFDNADUajVW019eXobRaMTk5CQmJyd31YE5YOdwb5t6PB7H5cuX4ff7EY/HuR7qFULeMeFwGKFQiInHud7kYCRJwvXr17G8vIxUKoV0Oo1qtTrQydpisSASiWB2dhbhcPiZgxFN05BMJpHL5aCqKorFIprNJjNn7XQ6iEajiEaj7D28Xi/raD1OQU+328WDBw/wy1/+EsViEaurq9jc3ESn0xnYEj07O4tf/OIXmJ2dhSiKTwx6bDYbYrEY2+BmMhkIggBJkng2/ym0Wi2mr3rw4AFu3LiBR48eDbwuRqMRwWAQiUQC09PTmJiYQDgcZpWeYTFUnx6aD0S+IzQF90WIl3u9HnN4bjabKJfLKJVKKJfLu2YW0fv374B1XWffX6vVUC6X4XK59kWynG+1EGTTT4uG1WplfhnDcN087lA9nUpb/PwPpj9VT/O1KBMzqKxFZXmr1cpMCWmt6nQ6bFJ3v5UArW+1Wo2tQYqisDleNJ+u2+3CZDLBZDKx3TDNTGs0GnA4HOy6HtXrSWNsWq0WqtUqisUiisUic/gFvr0ORqORmZzSiIlnLTMCjzOi5HVF7815MjQGhPzYqLxF9N+XZEjocDiYP9sobLxGomWdIsQXkVqkxabdbuPWrVv48ssvUa1WmclYtVpFOp1mi1N/jZ0uGH1N0zQ8evQIlUoFkiTh8uXL3/n4jgp07vL5PD755BOsra1hfX2dlSpp5ITH4+Et6pyRhMbhkBifgg9ZlvftWgVBgNvthtvtRiwWw+zsLE6cOMGcl1utFrLZLNbW1pgWkMSd1WqVDbgslUpQFIWtf+RiTuWtnZ0d1sDhcDhgtVoRi8XQaDQwOzsLj8cz9O6Xl0mxWMTKygpqtRpu3ryJ9fV1ZgwJPNaq+f1+OBwOxGIxvPPOO5iYmMDJkyefuUmi2Wwil8the3sblUoF7Xabaw2fEV3XUalUnmgOSRiNRoTDYZw4cQLRaHSo4uV+hu7ITIrvF1VL7c8iLS0t4Ve/+hUTLVP6ctAwNPrefprNJjY3N7G1tQXgcdmL8xg6f+VyGVevXsXXX38NVVWhqirbqbpcLqYnOao7U874Qtlc0tTQFPW9FvrAt0FPJBLBxMQEEokEZmZmWHa61Wohl8vh5s2bTCeyubmJZrOJTCaDYrG4y2eJ3h/4drgpvQ9ljugzMzc3h2AwCF3XMTk5iWAweGSDnkqlgqWlJeRyOSwvLyOZTO7qsjIajawxYnFxEX//93+PxcVFJo94FprNJvL5PHZ2dpiJKufZoEYfMiJ80pBWg8GAQCDAPJNGZfTH0DI9/R/8/izLd6XRaKBQKKBeryOfz7NdFnVkEIPeb2+mB3h8kalzi384HkMdWxS4NptNqKqKXq/HrMXJM4NrSTijBq05NLxVlmWUy2VmqU/l7X5MJhPcbjei0SjrROwvw2iahu3tbTayolKpoFar7SrhGwwGiKK4q3RPRpOUedZ1fd+DhBoFSDd31MowtJ50Oh0mLi4UCqhWq/vOBek+A4EAy/gcNgDcq0M8aufzZaBpGut8plmZtVptYNBDJXUqAZN1wKhMVRiJo3iRWYBHjx7ho48+YjuFTCZzYMAjCMKuHVb/v+/9Os9UfIuu6yiVSkwESBqpcDiMmZkZNiw2EonA6XQe2V0pZzxptVpMr/f1119jZWUF6XQam5ubqFQq0HWdZXoEQYDJZILdbsebb76JDz/8EA6HAw6HA+l0Gtvb2/jTn/6EVCoFSZKQy+WYlpACGbPZjHA4DJvNxsawUIBDnaX5fB7tdhvVapUJqIl2u418Ps/mFR01bSENdlUUBUtLS/jss8+QTCZZebAfm82GxcVFXLhwAZOTk4fS8BCqqiKVSmFjY4PZEnAOptPpMCuSTCaDL774Ajdu3ICqqvsm1VMThdfrRSAQwNzcHE6fPs0sAkaBkQh6iBcRWGSzWXz++efY2tpiOy4qZw0KbA56z0FeGZzHdDod1Ot1lMtl5pvRbDZhsVgwOTmJQCCAaDQKj8czMilNDgf4NrOiaRrq9To2Nzdx69YtpufpL6UA3/qMWK1WzMzM4NKlSzAYDGxt2dzcxJ///Gfcu3ePiY6Bbx2xyX6fjDlnZ2cxMTHBjqHb7bLOLTJH3Ns6TSUF2mgctVlr7Xabnc9UKoUHDx4cODldFEUkEgksLi4yV+zD0m9AyXk63W4XkiQhnU4jnU5jfX0dKysrAAbrcK1WK/PkCYfDmJiYGCm7l5EKevpnCh1G59Nut1kqmTwEVFVlFvIU3JDin0owVquVjbug3Vm322WdXmRCZrPZ4Pf7udgNj6+RpmlMs7C1tQVVVQGS7JhGAAAgAElEQVQ83oVFo1GEQiF4PB7eos4ZSUjXR+WtYrEIWZYHzjKjLkSv1wuj0QhVVdHpdJDJZJigk0q7/dlhr9eLaDTKprCTC+3s7CwCgQAz3ev1ehBFEe12G4qiMLHzoJLLUS2vq6qK9fV1pFKpgXPODAYDwuEwgsEgJiYmEIvF4PV6DzVDS9M0ZDIZ1Go1luHhPBn6nKiqylrUSTICYNc9D3ybHOjv2CL7gFHqOByJoKdf0NdsNpnD47N+yGu1Gu7du4dyuYylpSWk02nk8/ld7YhUZ3Q6nWzhicfjuHTpEgKBAMrlMrLZLBRFwdWrV3H16lUYjUbMzc1hamoKCwsLz5VKPUpQ660kSfjP//xPfPbZZ7t2TMFgEBcvXmTT1LmWhzOKtNttqKrKHoC3b99mguZ+BEGA3+/H7Ows3G43zGYzK8PcuHED6+vrKBQK+wSdJpMJCwsLuHz5MrxeL06dOoWpqSmYzeZdA0lJJ7ixsYF4PM4aLba3t49VySWfz+Nf/uVf8MUXX6BaraJSqez6d1EU8f3vfx8//OEP4ff7ceHCBUxPT7MM3LNQLpfxb//2b1haWkIymUSxWHwZv8qRotVqQVEUyLKMa9eu4de//jWzXTioEYiMUckjjLJxozQ0fCSCHuBby3ESBw4SEx4EOZ1ms1nmgUHZBwCsw4Lq8x6PB6FQCIlEAufOncPExARyuRzcbjdkWcbDhw93mSVGIhEEAoFjP1eHbnRN07Czs4OVlRUmsgQAu93OulvcbvfI3OQcDtHvoUMi5EKhcOBaI4oivF4vu58bjQZkWWbDL6vV6r6sgcFggM/nw8zMDAKBAM6ePYuFhYWBO106HkmSYLPZ4HK5RmZH/KpQVRWbm5u4d+/ewH83mUyIRqNYXFyEx+NBNBqF2+0+1Hs0m01sb29jaWkJlUqFZ3qeAfqMNBoNZLNZtt4/DRr+2p/pGaVnwdCCHk3TUK1WUavV0Gq1IAgCms0mtra20Gw2IYoiSqUSG5QoiuKuxaDT6UBRFJa23NzcRDqdRi6X2yX06w92LBYL6ygSRRGyLOP69euw2+0ol8vI5XJoNBrIZDJMvBiNRjE/P4+JiQmuTxkAGVAZDAbYbDbY7XY4HA5uSMgZSXq9HvPqymQy+7I7wLfiZaPRCJfLhWg0Crvdjkqlgps3b6JWq2F7exulUgmaprFOlUAggMnJSTgcDpw/fx5zc3PM2+dJVKtVbG5uolgsolwuH4tuom63y7pqyStnL/2lxXA4jHA4DKfT+VzaEF3XmdC83xuOM5her8eMMyuVyq4kwkGQhCQYDOL06dMIBoPw+Xyv4GgPx9DGUJAjqSRJrJZO829WVlbQ6/Vw8eJFlm3Z+xDVdR35fB6VSgUbGxu4desWtra2kM/nWTRKrzcajazOSH/sdjsKhQJu3LiBWq0GVVVRr9dZScxiscDr9eLkyZO4ePEiPB7PoXcXRxkKJmlIbK/Xg8vlgs/ng9fr5d48nJGk2+2iUCjg/v37yOfzkCRp32tovTCbzQiFQpifn4fZbMbNmzextLQETdOQz+dRq9XYa10uF06ePIkPP/wQoVAIp0+fxuLiIkRRfOIGoNfroVAo4NatW8jlctjZ2TkWQY+u66w1PZ1ODww+yYgxGAxidnb2mUZMHES73UY2m8Xm5ibTU3GeTL1ex87ODkqlEtPx9DOoy9lgMGB6ehrvv/8+fD4fotHoSGV5gCFmemjWVX97qK7rqNVqLIVMHjsmk4lNwQUen+B+74p6vQ5ZlpmF/KCuLHpA93vxkDNnpVKBpmnsg2ez2WCz2ZgrqsfjgdPpHBmfgWFB14y6VHq93q4bmhakUUtncjj95XMam0JZ5r3QVGjKCFMXVrPZZNmder0OTdN2vYbEyyS6dbvdB64Z1LTRP8ZFluVdvjE0bsFkMjEx6CgJQr8LNFpDlmW22dyLxWKB2+1mQ1hpTT4MVJLXdR2tVmtgWYuyenzN2g1p32g+5d7nav/faXwUPTNp8zsqHVv9DOUpTmInSn2Rj0u/W/LGxgZ++9vfsknRExMTMJvNsNlssFqtrG262WwilUohn8+jUCgwN9X+C9LfHtpqtbCyssKU/JS6o1k7NFvH4XDA6XTC5XLB5XLBZrMdW2EuXZdcLodSqYTt7W3W3muz2ZgfTzgcZqUuvoBwRolyuYxMJgNFUXD37l3cuXOHtUnv1fP4/X6cO3cOLpcLfr8fmUyGZZapZZym1kciEbz99tsIh8OYn5/HmTNnWLvuoM8AbRgqlQrW19dRrVZx9+5dtvmiTjAaOOrz+RCJRLC4uIipqSlMTU0dCd8rVVXx5Zdf4vr168jlcsjn87v+XRAERCIRXLhwgVlgHHZN6Xa7LKjK5/MDSzSRSASJRAI+nw+Tk5PHdo3fC5WBt7e32eyzJ2lsybjT7XYjHo+zYHUUO56HFvTY7XY2sVwURRaR0y5nfX0d6XQaRqMR0WgUiUQCVqsVfr8fPp+PaUjMZjO2traYnqffK4MCn16vh0ajAUEQUK/XUalU2NfJQKyf/rlRVJO3WCzHMtNDHVvtdhuZTAaPHj3Czs4Ocrkc6vU6rFYrJicnmR8DDQDkcEaFXq+HUqnERKxfffUVvvzyS6iqinK5vO/1kUgE7777LiKRCHZ2drC1tYV6vY5sNotarcbWL6vViqmpKfz0pz/F6dOn4fP52FysQZmDfsflbDaLK1eusM8UbcIo6DGbzYjH40xPeOHCBSQSiZHdPR8WRVFw7do1/PM//zMzi+xHEAQ2WysYDCIejx86w9XpdJhWk6wF9kJBaygUwtTUFN+sfUOv14Msy9jY2ECxWBy4OejH6/Wyz8D09DTr2hrF8zm0oIfaNxuNBkvZ9p9U+iCQSLDf2bfb7cJoNMJms0EURTbA70kdX/1BFY2WGPTa/iyUz+djZbWjklY+LBTwNJtNNvW4XC7vujZutxs+n+9QvhkczquCStmSJDF9AtliHGSjT5lek8kETdPYazudDoxGI0RRhM1mg9PphMfjgdfrhcvlGqg5oY0DDRfVNI1Z+dPxUCaaSvnkaktiULLy3zvGYtyg7tx6vY56vb7PiJEaI0RRZCMMXC7XoRojKDPdbDYhyzLTfvZ3HtGabrfbEQwGEQgEYLfbj+UaP4j++Ziapg0sP/ZLRsh8k84jBf2jeD6HEvSQwpsm496/fx8A9gUidMLq9TpSqRRMJhNyuRysViurIZpMJtTrdSa0Ouxoib3vZzKZcP78efz0pz+F3+9nFtrHteZLs3Co0+3TTz9Fo9GAqqqIRqOYm5vDD3/4Q8zPzyMajfION87I0K+bSafTuHbtGvL5PJLJJDMjPCjocbvd8Hq9MBgMu/Q2ZCY4OTmJWCyG+fl5xGIxVtrdG/R3u10WZFWrVWbwtr29jStXrrAZU6RriUajrKz1ox/9CK+99hocDgempqbYw39cNxa6rmNjYwMbGxvIZDLIZDL7Np6hUAinTp2Cx+PBW2+9hfn5eaapfJYHKGUoyC3+D3/4A65fv84chYHHms1QKASbzYbXX38dly9fRjAYRDQaHdtz+6Kg4Jy0b7VajQXle6HWdIvFgvn5ebz77rsIh8OYm5tjjSw86PkG8rGYnp6GKIrMA6PT6ewbD0GlKRIZDwo89nY7PMtoiYO+ZrFYcPLkSfzkJz9hU8KPQjr5eej1elAUhWl57t69i88++wwGg4F5F01NTeHChQs4d+4cswXgcEaB/mGehUKBaWdqtdq+DEM/pOkjzxxqlqASDGl5ZmdnMT09zRyXB0GC3XK5zLq01tbWsLOzg7t376JUKrFjtVgs8Pl8OHXqFEKhEN566y28/fbbrENs3B/INHLj5s2byOfzA0uLXq8Xi4uLCIfDOHXqFLMAeNaHZ6/XQ71eRyaTQaFQwLVr1/C73/1ul4xBFEUEAgF4vV7Mz8/jjTfeQDAYHNmH9KukP+hpNptQFIUN4t37eSFLB5op99prryEWizEjz1E9l0MTqZAomXxdnE4ndF1ng/r66Xd/PMgFcu/XgP0BDXUb0QWhhzQ5e9JCF4lEYLVamSj3OEMlQUoZU4BJFgCkqzrOJcBRhDKho5piflXQ2kHBD6Xqn6RP6M8EU+dUf0dit9uFqqosA0R/Br13q9VinkClUoltIOr1OnuQUKnMarUiEomwNm2Xy3WkOovI2LRarTJJwl6otOh2u1nzyLPcv2Skp+s6c4kvFApQFGWf7EEURVbS8ng8vNu0D+om1DQNiqLsC/j7EUURExMTrDWdyq+jrn0dWqbH5XIhFovBbDZjZmYGCwsLqNVqSKfTbDHoz/YQg75+mNdS6tpms8HtdiMWi8HhcGBmZgbnzp1jE8K9Xi9bcI4zg6wFzGYzgsEgEokEIpEIHA7HSEf2xxHKDpCo9jjSr+PTNA21Wg31ev1QxnR2ux2hUAhmsxmSJDG7DJrIXq1WkUgk2L/Rw7PT6TANz/Xr17G0tARFUbCzs8PmBJKwNhQK4cSJE3C73bh8+TLee+89OJ1O1rF6VDIQ1LW2ubkJSZJQq9X2vcbhcCCRSCCRSCAYDD5zMEJC80ajgdu3b+Orr76CJElIJpP7KgF+vx/f+973MD09jZMnTx6JbrgXhaIoSKVSbETL+vo6JEmCqqr7zuPExAT++q//GqdOnUIsFsPk5CQbOTHK9+vQQjISwLZaLfj9fgSDQRiNRuRyOfaaJ4mSB33tIJv3fkh8S14C8XgcHo8Hb7zxBn7wgx/A7XbDbrdzbco39Gd66FwaDAYmtHS5XHynNIKQZ8aoL0Avm36fFhIkPyv9DRftdpuVbtvtNht3Y7VakU6nmY8OfQ7a7TYbhHznzh188cUXTCfRv2sWBAFOpxOTk5Pw+/1YWFhgpob0M48KJFWoVCqQJOlAnQgJYh0OxzOvKyQOr9VqSKVSWF1dPTADRxqphYUFRCKRkc9MvEpo4j2J/qldfRAulwuLi4t46623YLfb4fF4xuJ+HVr3Vr9L8vz8POuucLlcKBaLqNfryOVyrKth72gJYHdAM8iMkNyXadCo1WqF3W5nU3q9Xi8SiQQcDgdisRgTZfEH+GPI14g0EIIgsG6KRCKBhYUFTE5O8gBxxKDrNDk5ycokxxHKvJhMJgQCAZw8eRIejweyLKNWq7EM5t7MjyzLWFlZgSRJ2NraQqlUQq1W2xUwtdttCIIAWZaxurqKer2+KyNDa5aqqigWi6z00l8epuGjZOTm8XhYSecolopptAE9VCnosdlsTKh99uxZJuR+WsuzrutQVRXtdhvb29u4desWKpUKVldXUSwWWacc8HiTnUgk4PF4sLi4yMoyhwmsjipU0mq320in03j48CFKpRIymcwT3cEFQWCjncYp0z9UTY/JZIIoivjRj36ES5cuscWmXC7j/v37+MMf/oB8Po9Wq8X0JIM6sgZ9zWg0IhgMYmZmBjabDTMzM0xkderUKYTDYZjNZlY3p/bTo1I//y70lwWo5ChJEkwmEyYmJhAKhfDuu+/i+9//Pux2O7xe77APmdOHwWBALBbDxYsX4ff7EQ6Hx2ZBepFQpsZoNGJubg4/+9nPUCqVcP/+fSwvLzMbhr1BTzKZxL/+678ySw3Sn0iStGvobrvdZp1ge0sktCHrdDqo1WpQFGVXttRut2NqagoOhwMnTpzAzMwMMzUkT7Cjds263S7K5TI2NjZQrVbZRjYSieAv//IvcebMGSaIJTHsk9biZrOJdDqNer2OK1eu4Je//CXzD6PzTVk1n8+Hn/70p3j99dcxMTGB8+fPw+fzHbls2vNA51GWZdy5cwcfffQRMpkMisXiEweMmkwmZtkwToa0Qwt66AQZjUZEIhFEIhFUq1X2wahWq0wYRYLAvSf1oK4t2uGR347T6UQ8HsfU1BSbp0UOn3ww5mCopNVqtaCqKjN3pDRmKBRCLBbjHVsjCF2nQCDAvKaO4z3er7GhzJfdbkc+n4fdbgfwWMOwl0ajgZ2dHZhMJhaoUOaGgpa93juHOSYKxlwuFzM/dTqdzIvsqIrPyQy2Xq/vMgoURRGJRAInT55kXVV0ffZ+/96fpSgK8+JZW1tDNpsd+N5kM3Dy5EnmweZ0Ol/8LzmG0HQDWZZRKpWQTCaRyWSgqupAfx5g/4iUcWKkipkkkLVarWi1WtA0DZVKhfkuUPcF1ctlWYaqqqw0pmkaPB4PIpEI82C4cOECnE4notEoM07iGZ399Gt3VFWFLMvQNA1ra2tYXl6GoigQRRGLi4usXX2vjoHDGUUEQWDtyZR1oXu80+ns0yx0u100m81dZXTSBT0PJpOJaQmj0SgTKb/++ussE5dIJGC325nb/FFCVVXUajVWVty7WTWbzfD7/ezc7NXYdDodZLNZFItFppMi48HV1VVUq1Xcv39/n+MylXntdjui0Sj7Q40Xxxla51utFrLZLG7evInt7W2sra1BkiQ0m82B3XVkq0CNSONo5zJSQQ+NNOh2u5iensalS5eYZXsqldoVBCmKgo2NDRQKBVQqFTx8+BCapiEYDOKNN96A3+/H5cuX8f777zNFOT2gKW16FHdTzwvtnDqdDiRJwsbGBmq1Gm7fvo2rV6+i3W7jRz/6Ed599134fD42cgI42BeJwxk2lFkJh8Pwer3QNI0NEy0UCigWi0gmk/vc4AcFOE9qc3/S+1M21O/348KFC5iensb09DTee+89hEIhposgs9WjFvTUajUkk0mm5dl7Hi0WC6LRKGZmZpieqZ92u421tTXcvHkTiqJge3sb5XIZkiRhbW2NmefV6/Vd32cymVgwNTs7i5mZGczMzLAMxXGGTDNlWcba2ho+/vhj3LlzB41GA6VSCa1Wa9dYKEIURUxPTyORSODEiRNwOBxD+g2en5G68pT2BR6fXJfLxTQ71H1hNpthsVhgs9kgSRLLCNFNbDKZ4HK5mDV8IBAYy2j0VUM7Weowoc4H8tSgeUAkuBRFkQc7nLHBZDKxVDxpZ3Rdh8vlgtVq3ecF1q/JGSTm7A9O9k6a7v+7IAgQRREWi4W9bygUQjAYZH+O+kOYBMeNRuOJ2TJBEJifEgBWWlRVFZVKZVeDS6lUgizLKBQK+1rf6XzSlHZqWrHb7bw9/Rv6J89T1pNGddDmdy9U0nK5XKxzdxz1UCP/SaOWzlgsBl3XEQwGmVOk1+tFqVTC+vo6NjY2UCqVYLVa4fV64ff7+SyVQ9BsNrG9vY1qtYrV1VVcuXIFpVIJlUqFlRzn5uawsLDAzCQ5nHGDBhgDgCRJaDQaCAQC0HWdefjQg0DXdeak3P+wpkGjfr+fZXIMBgMbWtmvoer/d6fTidnZWaZboYfGUV6jyNV9Z2eHBSp7Mz2KouDevXuwWCxQFAWSJLEyFpVatre3kU6nWWs6GegNGlRK+k2Xy4U33ngDJ0+ehN/vRywWe5W/+khDOlmadkB/+jsM+7FYLLscw1977TVMTEyM5XNg5IMeAHA6nSyNRruwZrOJUCiESqUCq9WKTz/9FADYjuo4Czifh2aziVQqhUwmg9u3b+P3v/89CoUCYrEYpqenEQgEMDs7ixMnTnBfHs7YYjAYEI1GEQqFoCgKDAYD4vE4ms0misUiGo0G6vU6e9hubGygXC7vC3pmZ2cxNzfHskdGoxFnzpzB5cuX4fF4dr0nZXxoUCm19x71gIeo1+vY2dlhni97H6r1eh1LS0vsGmxubqLRaCCTyWB7e5tlHij70O+oPWg0QiwWw9tvv41gMIgf/vCHePPNN5lvFecxFPSoqsr+HCTINxgMzDWcGoHefvttJr4fN8Yi6BnkSEolrlarBa/Xi1gsBlmWMTExwXZRNJiU83TIq4Fca5vNJtvtUlqT6u3jmNLkcAjS9lksFrhcLvj9ftaaq6oq6xolM8N+vxcA8Hg8mJiYQCQSYQ9To9EIv9/PfKz20t+1ddw2DDTLTFXVgd2yuq6zUhVpdWhA69MctOl8Wq1WOJ1OWCwWTExMIBwOw+/3w+12w2q1Hrtz/jT6gx5N0waeY9K90owtEt2TgS/p0MaNsQh6BkEiNZr/ZLfbIUkSAoEAEokEi0rH8aIMA1VVsb29jQcPHjDROIdzVCGtDZWpOp0Oms0mMyukGV1UYunPTlgsFoRCIbjdbtYWTx1i5C+z970G/f9xQBAExGIxvPfee5AkCZubm/sCEFmW2SBWGhei6zoajcaBLdMA2Kwym82GM2fO4P3332fu/pFIhAVAPODZj67rKBQKWF9fZ2VDo9G4S7xsMplgtVphs9nw/e9/H2+//Tb8fj/Onz/PnKzHMXs2tkEP1cidTicCgQDm5+d3GRUet8Xlu9JqtZDP55FMJlEsFp+7PZfDGRdMJhPC4TDC4TCAw429GbS+8DVnMH6/H36/H9VqlWXH+mk2m1hfXz/0zxVFkblZX7hwAX/3d3+HiYkJ/gx4BjqdDmRZZqJwXdf3nS8yD3a5XDhz5gw++OADuFwuRKPRfSXccWJsg569HJWhfK8a6pCgAYlUO6e0psPhQCgUYh5H/ByPLjRg1G63w+FwwGq1HuuBo4floHub3/MvBqPRiFAohIWFBVa6ohI6jRvqhzp1SVNCtiMWi4V1ks7NzcHj8SAWizFjR87TMRqN8Pl8iMVi6HQ6rCmo0+mwLH8gEGAjmyKRCJxOJ2w229h3Go730XO+E+TNo+s6ms0m+9PpdJjmIR6P43vf+x6CwSBisRhfVEYYp9OJRCIBl8uFqakphMNhuFwu7pjNGQnMZjPeeustAI87527evImNjQ0oioJMJrPPZ4eMZi0WCyKRCOsinZiYYJYkFPSQfofzbFitVpw7dw4zMzNYXV1FJpNhnXblchndbhcXLlzAT37yE/j9fjZJnbI/4wwPeo45lOGh4IdaFsnrwu12s8GVpGHgjCaiKDJPEo/HA7vdzjsYOSMDdVZ1Oh2Uy2WUSiVIkgSDwYBCobDv9aIowu12s4HQk5OTcDgcmJubQygUYkEPD3YOT39pV9d1hEIh+Hw+GI1GpqWKRqN47bXXEAgE2AbqKKwlPOg55phMJvR6Pfh8Prz++utwu90s2u90OnjzzTcRi8XgdDoHzsPhvDpol+Xz+XDp0iUmPCTIi8ThcGBqaoq1Rh+FhYoz/vTPhBNFEa+99hpcLhcURcGpU6fQaDR2vZ66hURRRCgUgt/vh9VqZe7aDodj7Esto4DT6cTi4iJsNhszgux0Ojh//jwCgQDLFh+VdYTfMccYaqE1mUyIx+P4m7/5G9a1ous6er0enE4n3G43q6UflRt/3Oh39hVFEf/4j/+Iv/3bv931GnKhNRqNbHglv16cUcFgMLBBn51OB9PT02g2m2ywcX+nFpk6Uks6rVN7/5+Xbr87wWAQP//5z5m0gXyQaLLBUbMp4UHPMYcyATQMkTO6kJ7KZrMhkUgM+Wg4nMNjNptZm/MgPyPOq8disbAOxuMAV6VyOBwOh8M5FvCgh8PhcDgczrGABz0cDofD4XCOBTzo4XA4HA6HcyzgQQ+Hw+FwOJxjAQ96OBwOh8PhHAt40MPhcDgcDudYwIMeDofD4XA4xwIe9HA4HA6HwzkWCP2ze576YkEoANh6eYdzJJnu9Xqh7/pD+Ll/bvj5Hy78/A8Pfu6HCz//w2Xg+T9U0MPhcDgcDoczrvDyFofD4XA4nGMBD3o4HA6Hw+EcC8Y+6BEEwSsIwq8EQXggCMJ9QRC+P+xjOi4IgvB/C4KQFwRhadjHchzh53+48LVnuAiC8DNBEB4KgrAqCMJ/H/bxHCfGee0Z+6AHwP8G4N96vd5pAOcB3B/y8Rwn/gnAz4Z9EMeYfwI//8OErz1DQhAEI4D/HcDPAZwF8PeCIJwd7lEdK/4JY7r2jHXQIwiCB8APAfxfANDr9Vq9Xk8a7lEdH3q93n8CKA/7OI4r/PwPD772DJ1LAFZ7vd56r9drAfj/APy3IR/TsWGc156xDnoAzAIoAPh/BEG4KQjC/ykIgmPYB8XhcI48fO0ZLpMAkn1/T33zNQ7niYx70GMC8CaA/6PX610AoADgtV0Oh/Oy4WsPhzOGjHvQkwKQ6vV617/5+6/weCHicDiclwlfe4ZLGkCi7+/xb77G4TyRsQ56er1eFkBSEIRT33zpLwAsD/GQOBzOMYCvPUPnBoAFQRBmBUGwAPifAPx2yMfEGQPGOuj5hv8VwP8rCMIdAG8A+B9DPp5jgyAI/wzgKoBTgiCkBEH4n4d9TMcJfv6HDl97hkSv19MB/C8A/h2Pu+Z+2ev17g33qI4P47z28DEUHA6Hw+FwjgVHIdPD4XA4HA6H81R40MPhcDgcDudYwIMeDofD4XA4xwLTYV4cDAZ7MzMzL+lQjiabm5soFovCd/05/Nw/H1999VWx1+uFvuvP4ef/+eDnf3jwtWe48Ht/uBx0/g8V9MzMzODLL798cUd1DLh48eIL+Tn83D8fgiBsvYifw8//88HP//Dga89w4ff+cDno/PPyFofD4XA4nGPBoTI9HA6Hw+FwOP30ej10Oh30ej30ej10u10AgMFggMHwOLdiNBohCN+52vqd4UEPh8PhcDic56ZQKGB9fR2KoiCXyyGbzcJgMGBychKhUAhOpxNzc3Pw+/3DPlQe9HA4HA6Hw3l+stksrly5gnw+jzt37uDmzZswGo24dOkSzp07h2g0CrfbzYMeDofD4XA440en00Gz2YSu65AkCcViEaVSCZIkoV6vw2w2Q9d1dLtdVu4aBXjQw+FwOBwO51AUi0V89dVXyOVyWFlZwZUrVyDLMnRdRywWg8PhwIkTJ3D69Gn4/X44nc5hHzIAHvRwOBwOh8M5JLIs46uvvsLKygq2trZw584dNBoNxGIxxONxeL1eJBIJTE9Pw+VywWazDfuQAYx50NPtdplqnNJsmqah0Wig2+2i0+lA1/Vd32MwGCCKIkwmE0RRhMvlgtlshomCh6wAACAASURBVNFohNFoHNJvwuFwxh1af54lla9pGqrVKtrtNlurAMDpdMLhcMBkMsFut0MUxZd92GOLrutoNBpot9tot9vs3LfbbbRarQO/z2AwsE4is9kMi8UCg8EAq9UKq9UKg8EAk8nEnwcD6HQ6UBQFmqahWCyiUqlAlmW0Wi2IoghBEBAIBDA1NQWv14tgMAin0wmbzTYy53Nsg55er4dWq4V2u41qtYqNjQ1Uq1Wk02ncv38fiqKgVquhXq+jf5K8zWbD5OQk3G43EokE3nnnHQSDQdhsNrhcLtZex+FwOIehWq0imUxC0zTWunsQqVQK169fR6lUQrPZhKIoMBqNOHfuHM6dOwe3242zZ89iamrqFf4G44Usy7h//z7K5TJKpRK2t7ehqipKpRIKhQILJPdiNpvhcrlgMpkQDAYRi8Vgs9kwOzuLmZkZiKI4UuWYUaLRaODOnTtIpVLY3t7G0tISkskkTCYTYrEYrFYrLl++jA8++AButxvRaBSBQABGo5Fner4rvV4Puq6j1WqhXq8jk8mgWCzi4cOHuHr1KiqVCiqVCkql0q7Fx+VyYWFhAeFwGLVaDQsLC7Db7TAajU9cpDgcDudJNJtNlEolKIqyy6tkEKurq/j888+RSqVQr9chyzLMZjMURYHVakUwGEQ8Hn+FRz9e9Ho9NJtN7OzsIJPJIJVK4f79+6jVakgmk0gmk/uy/ITVaoXP54MoiojH41hYWIDL5YLRaITX64XNZuMBzwG0221ks1msrKywc18sFuH3+xEKheByuXDixAlcvHgRTqcTFosFJtNohRmjdTTPgK7r0HUdiqLg0aNHyGazkCQJa2trkGUZqVQKsiyztOde2u02arUaDAYD8vk8MpkMjEYjIpEIPB7PyKTgOJxnQdd1qKrK/lur1dBut1mm8zBdE4IgwOVywel0wmQysTo8lVpGbfEaFt1uF81mk5VRJEmCpmlIpVJYXl5mQc+Tsj3b29vs+3RdZ6+VJAmbm5tQVZX9nFEwdBs2nU4HnU4H7XYblUoFiqJgZ2cHDx8+ZA/eQqEAVVWhquoTf1av12PZOEmSsLOzA7vdDpfLBVEU4fF4YLPZ4Ha7IQgCDAbDsb4GdL40TUO5XEY6ncbm5iZKpRIrKXq9Xpw9exZ+vx+Tk5OwWCwjY0a4l7FaxXq9HhqNBrvhf/nLX+Lq1avQNA2yLLO6rqIoTM+zd9FptVpIpVLI5XLodDqIx+MoFAo4e/YsYrEYLBbLkH47DufwaJqGTCaDer2OnZ0drKysoFarYW1tDSsrK7u0DYMewIIgsK+bzWacOHEC8/PzcDqdOHXqFCYnJ+F0OhGPx+FyuV7Z7zXK6LqOUqkEWZZRLBZx9+5dlEolbG1t4fbt2yxYOai8AmDXmkWBqa7r2NzcRLVaRTwef2Gzs8Ydeug2m03Isoxbt25ha2sL6XQan332GXZ2dtBut9FsNllg9KRz3263Ua/XIQgCFEVBNpuFyWRCNpvF+vo6wuEw7HY7IpEIjEYj06ocV7rdLiqVCorFItLpNG7cuIErV66g0+mwAHNmZga/+MUvMDk5CZ/PB6fTucuNeZQYu6CHxIL1eh1bW1tYXl7eJWJ7Gt1ul+0GZFlGuVyG1WpFvV4fKS+BcYIemk8rDwqCcKwXj5cBZT2r1SqKxSKSySQkScKDBw9w584daJr2zD/LbDaj3W7DYDDA6/XC7/ezQOegUsFxhBb7Wq2GSqWCdDqNbDaLra0tbGxssLXkecrljUYDvV4PVqsVzWbzJRz9eELBTLPZRKFQQDKZxM7ODlKpFHZ2dva9/mlrDd3PJI8wmUyw2WywWCzodDosYwo8fmaM4sP7VUFBZ61WQ7VaRalUQj6fZ4Jwk8kEp9PJOrXoa6PK6B7ZALrdLkqlEpLJJFKpFKrVKlqt1lNFgwfRbrchSRIsFsuhSwHHmf70vqIoyGQyUBQF9XodpVJpV1nRYDDA6XTCarXC7Xbj5MmTCIfDPAA6BP3zbKjTR1EUtFotZLNZfPHFF8jlcsjlcswKvlgsHvp+7nQ6qFQqWFtbg9PpRKvVQjKZRDgcZpbyVquV6R+OMhS0kMiY/t7r9SDLMr788kusr69DkiSsrq4yczbK3DxtPbJYLLDb7TCbzeyzYDabMTk5iUQigUgkgomJiVf024423W6X3dOFQgFra2tYXl5GpVJBo9EAAJhMJlZS8fv9CAaDB5alOp0O2yTTdaP3yOVy0HUd9+/fh8vlgsfjwfz8PHw+36v+tYcO3e+apmFzcxO3b99GLpdDPp+Hruvw+Xw4efIkfD4fTp8+DYfDAaPROPIB4lgFPbquY2dnB3fu3EE2m0WhUGA72ecJejRNQz6fZ8HPk1KinG/pdDqoVquspHL16lVkMhkkk0ksLy+jXq+z15pMJjZ/ZWpqCv/wD/+AQCDAUp888Hky/YP82u02dF1Hs9lkWrbV1VX87ne/Y8GOJEnQdZ299rDvlc1mUSwWYTKZsLKyArvdjtnZWTgcDrRaLfj9foiiODKdGC8DOuf0UMxms9B1nX0tm83io48+whdffMFKJaQ1pE3Y0xBFEeFwGDabDQaDARaLBaIo4p133sGlS5fg8XgwNzfHPx94HPTIsoydnR2k02ncunUL165dY8EL8DiIJE3OmTNn8Nprr8FisQzcXFGZrNVqYXV1FbVaDY1GA5IkoVqtolwu49q1a5BlGfF4HD6f71gGPbTBUlUVS0tL+P3vf8+uQ7vdhsfjwQ9+8APMz89jbm4Obrd7pDM8xOgf4R6ovNVqtXbtZOnGJuEZgKfaX9PuuX86LGcw/ZkGKi/KsgxJkpDP55konLpRCLPZDOBxoGSz2VCr1aCqKkwmE6xWK1/Un8Le866qKprNJiRJYh2KxWIRxWKRXZf+4P2w57fVaqHVarFuRlVV4fV6Ua1WoSgK7Hb7kc+I9mfW+h+QpBGsVCooFAooFAos2HnShmmQ54vdbofb7WadoxaLBVarFYFAAOFwGC6XC1ar9WX/qiNP/4O3VquxEkutVmOvoSyZ3W6H3W5n/jCkxRkU9JjNZmiahkKhAKvVyspn9BmTZRmFQoFlPOnZcJzWq/7nY6PRQLlcRq1Wg6ZpLFD3eDwIBoNjlf0dq6CHtAaTk5MwGo1wu93MTIrSxJR+FwSBPRAOWqStViui0SjC4TB8Pt9YRKmvGloMdF1nWYBqtYp79+4hlUrtSu+TMHPv98uyzHbJn3/+OSqVCsLhMF577TV4PB6e8XkCJMxvNpu4e/cuHjx4wNLwkiShVCphZ2eHdSu+qICk2+2yjQWVvABgfn4eU1NTcDgcL+R9RhEyO221WlheXsZ//Md/ML+vbreLer2O7e1ttNvtp7ami6KI6elpTExMsHXKYDAgGAzi5MmT8Hg8LMtjMpkwPT2NRCIBURRht9tf4W89eqiqinq9DkVRWHanXC4jl8sBANOOGI1GzM3N4Qc/+AH8fj+mp6cxNzcHk8k0cF2hbGm73UY0GoXFYoEkSdje3sbW1hba7TbS6TTrrMvn84jFYjCbzex5cxygTRQ9R/P5PDRNgyiKEEWRlWBjsdhYPT/H4yi/wWAwwOPxMP8Kt9sNURTZzUg+C+FwGEajERsbG6hUKk8MeiKRCOLxOPx+/9hEqq8Syqw1m01sbm7iwYMHKJVK+PTTT1l3EKX3KbDph0oEsixDVVVcvXoV29vbOHv2LBKJBBPK8nM/GHLulWUZ169fx7//+79DURQUCgXIsszE/c8rnD2I/jZVSZKwsbGBZrMJi8VyKHH0OEJBT7PZxPLyMn7zm9+gWCyyf6fy17OUD81mM+bm5nD+/Hnm/ms2mzExMYELFy6wcqHdbmfCUPosHJeH60GQ0aAkSbh9+zb++Mc/MtE+8DjrQj4w8/Pz+Ku/+ivWPeT3+594/ihYDQaDMBqNKBaL6PV6SCaTuzp8AaBQKKBarTKh83G5Lq1Wi2Xzi8UiOx+RSARutxuhUIgFPaPkuPw0xirooZucUsPRaBRTU1NstES/lTgw+EEqCALbAdhsNng8Hni9Xtjtdp5t6KNfxFatVll6kybpyrIMRVGeST9CP4s+RHa7HZVKBeVymYmcqcWR85j+1DKVnMi7RVEUNBqNZ+7uoft6b0atP0vxpICp3/dHVdUjWd7qH2WjKArK5TK75ynTdhDkR9JfYjebzTCbzXC73QgEAggGgzCbzRBFEUajET6fDy6XCw6HAxaLBTabja8/2K2nqlaryOfzqFQqkCSJXYdOpwNBECCKInw+H2w2GwKBANxuN/OWIj3P097L4XDA6/Wi0+nsKrf3j7NoNptQVRVGo/FI3vv99Jd26/U6W+9ptJPZbIbH40E4HEYgEIDNZmNjnMbl/h2roMdgMMDv98NqtSIUCsHhcOAnP/kJezh0u13s7Ozg0aNHqNVqAy+E2WxmH5S5uTm89dZbmJ+fh9/v5x4939Bf0qLsDnWs0FC5bDbLHoDPKgCnbFE+n0e9XofT6cTExAQWFhbwxhtvHPt0PkECTRKMl8tlJrJUFAWqqh5KdG82m2EymVh5xWQy7dKiUGB1UODTarWQTqchyzJisdgT5xqNG6TnkyQJ9+/fZ2l8ErguLy8/0eyOtCRUZqHzG4/HkUgk4PV68b3vfQ+Li4swmUzsOtjtdgQCgZE2cRsGpFdrNpu4ceMG/vznP6NSqeDBgweo1Wq7Ap7JyUl88MEHmJycxOnTpxGPx1m58FnOpyAICAaDeP311yFJElZWViCKIsuakuFnJpPB5uYmwuEw+/lHFfqdW60Wbt++jT/96U/s/He7XbhcLrz33nt46623EA6HkUgk4HQ6x+oeHqugRxAEOJ1OZhE+MzPDbs56vY52u42bN2+yB/KgzIHBYIDD4YDH42EP3FOnTrHUMudbPUe73UY+n2fzbW7duoWbN28ONH18FqjFGgBreUyn0zAajVhcXHzRv8bYQrvM/gwLdZjQdXnWoIeE/fQwJodlEocOci3fS6vVYin+SqVypDx7KLNQr9exurqKra0tbG9v4+bNm5Akia0rB0HZZYvFwrQ5ZrMZs7OzeOONN+Dz+XDhwgWcPn2adSzyjObBUDZYURQ8fPgQn3zyCXNgpvlklEXz+/24ePEizp49i2AwyMTLh8Hj8cBsNqNarSIQCDCBM20GNE1DpVJBPp+HKIpHvsOX1gXaoF65cgWlUgmlUgndbhdWqxVnz57F+++/zzJs4ya4H6ugZy/U1UIOzfV6HalUCoVCAeVyGaqqsodz/w5rYmIC4XAYkUiECdPGJUp9WfSnlWVZRiaTQaPRwObmJjKZDCRJYjutg5x9RVHcFfVTRqFer+/q6AIel0yoXPMiBbhHATJgazabSKVSWF9fhyzLzAPpMEGn0WiEy+Vion/qDCKtTqvVYu26tMsj7xOCPjd2u/1Iddx1Oh0W2GQyGaTTaezs7KBYLLLy4aCAx2w2w+l0QhRFOBwOhEIh2Gw2dv9bLBbMzc0hGo3C5XIxvQ4PeJ5Os9lEPp9nxrEkKO/1ejCZTBBFEdFoFG63GzMzM8xA87sIjKk0SXqq/jIWdfBR9uOor1NU3qXJB6qqQtM0mEym/5+99wiOK02zxc5N773PhAcIgqZIVrHIYtk2Uz1vumNmdhPvRWijxUgLSSFpp6322itCMTJvoXmKp5l+M/OmjXp6uru6q7q76AmS8C4TSKT3edNnXi1Y31c3QQAEDQAi854IRrFIIHnx33v//zPnOwdms5kd0ynQP4vP85kNemiMrtlsIhaL4Ze//CUbzT19+pRZ/51Oh6e6DAYDgsEgPvnkE7adcDqdStADcLbbarWwtLSEX/7yl0ilUtje3sba2hqP7u730lOVLBAIYG5uDmazmRn+zWYT9+7dw/Lyct9BTRpJlEkN+mbyMshms7h9+zZyuRwWFxfx4MED1Go1Ji/3er0jV1s0Gg3Gx8cxNTUFj8eDDz74AOPj4yiVStje3oYoitjZ2cH6+jpEUcTW1hai0Wjf59N743K54Pf7B6a832g0sLi4iPX1dezs7OBXv/oVtra20Gq1UC6XORin7J6CebPZjMuXL2NkZIQF2hwOB6xWK7esLBYLO3nTOO+w7zFHQSaTYd2vp0+fcquLXLpdLhe+//3v4+LFi/D7/bh8+TJ8Ph9XMl8WVAmlCpLBYOirpHY6HRSLRaRSKeb+DDJqtRq3suPxONLpNERRhN1uh9vtRjgchsfjYZ/Ks9gdObNBD4m1NRoNFAoFrK6uYmVlBel0Gjs7O33EQ5VKxVmCxWLB6Ogozp8/z8Zyymb0bUuLqgw0kp5OpxGPxw8s8csN+cxmM4LBIOx2O8xmM8xmM6rVKtbX1/s8ngCwqrAgCGg0GkrQI4MoiojH40gkElheXsbjx4+fq74cFRqNBg6HA8FgEOFwGNeuXcPc3Bzy+TxcLhfK5TJMJhProGSz2efeB5VKBavVyp46Z3Gj2w+dTgfZbBZbW1tsori1tXXo91BF0+/3Y2JiAh6PB5cuXeKDYJCCwtOAKIrY3t5GLBZDOp3mSg9NB1ksFm4d0nq/jiccVXn2Bj/0DhDBXRTFI1sdnWXIJ7ZocKHRaMDtdsNut7O2FA0OKZWeY4ScUV6r1VCr1bC5uck+LNvb28jlciwZLwdt/H6/nyNVp9PJwmDDil6vx9UyInLmcjl2Li4Wi+wFtB+IYxUMBmEymTA7O4vLly/DZrNxFlAsFvclKOt0OrhcLjidTv76YQaJD3Y6HW6ttFqtF2aWtEmT/DsFoDS2a7FYcPHiRczMzMDj8bC+jk6ng9PphF6vR6VSgd/v53bN3qCH3judTsdmvoMAst3Y3t5GNpt9IUFbbhfh9XoxPj4Om80Gs9nMlQYlgXp5UAJLnle5XA6ZTIatgUgIj6bdSPBULh3wqgcwJc2lUomF98iBHQC3dex2O7cpBw1yAdRisYitrS1kMhmWadDr9YhEIrhw4QICgUBfd+QsPu9nIuihm9Jut9lkLpvN4osvvsDi4iIqlQri8XjfCLUcarWaqzvhcBiTk5OIRCJM8BxWkPBWNptFNBrFj3/8Yxa/I07JQa0U2mTcbjc+/PBDhMNhnDt3Drdu3YLVamW+TiaTwe9+9zuoVKq+YNRsNmN0dJTFIc+KsNVxod1uo1AosDZJpVJBtVp9YXap0Wh4ooSqmTqdDpcuXcLVq1dhtVoxOTnJvlkulwsAYDQaMTIygm63C5VKhVqtxiKEezcyUq6t1WrsLzUIaLVaiMVimJ+fR7Va7VP53Q+0yZtMJszMzODGjRu85uT9dBYPgdMG8f6o5b25uYmNjQ2WECBemsfj4eldWmcaKydV65dd/0qlglgsxsryVN2gwF6j0cDtdiMYDJ4pAb6XAcmJkADtnTt3EIvFWJfHbDbjypUr+NGPfgS73Y6RkZEz3a59q+8gRdsUzbdaLYiiyAqROzs72NzcRLPZRKlUOlA0Ta1Wc2uLiFgkUT7MoEoPOedub29jY2Pjha71RFKmyRWPx4NQKIRgMMjl5nK5zIepvFxMoCk6IiEO872gjJXK6DSlRZWew0jLpF1F2iT0X6/Xy+KPgUCAdWKo9UIcCUmSYDKZYDQa2R5kv+sj6XlSIR4EkKHo3oPuRSCNL4vF0qfrQhM/8pbJMD/XRwVVesgCgsyLaXScWk8k7EhrTd+j0WhY6XpvBeKg9adEmsRVy+Uyk5WJB0pVPYPB0NfSGTTQBHS73e7TY6vVarwG1CKn8/MsP9dvXdBDG2q9Xufpkkwmg3g8jlqtho2NDUSjUZTLZUSjUZ4oOmzD6nQ6SCQS0Gg0qFaruHTpEhMOiWw4TKAprWq1isXFxT4DV5pQ2O9go4zWaDRidHQUHo8HkUgE165dQyQSgc/ng06ngyRJyOVy2NjYQCaTQbFYfO7zyLeFWjCDuJm8CGTRQQ7S9+7dQzKZRDKZ5ApmJpM5tNJjt9tx8eJF1hChicTR0VFMTExAr9fD5XIxJ2K/NmImk8HDhw/5PduvPUzCb4PE6SFH7pGREW5vHMadonUpFov4zW9+g3Q63ff3pBdDUy7UtqUDU8H+qNfriEajnHiREB7t6bRXpVIp1Go1fPnll9jY2ODKJVkPud1uFiykIN9ut7PECe1rVLGs1+tYXFzE7du3USgUsLGxgV6vB41Gw/y18fFxTE5OYmpqCg6HYyD5Ws1mE8lkEuVyGbFYDJlMBoVCgW2faJ+mfeSsr8Fbd9qTIiSVHavVKh4/fozbt2/zxMnOzg4LqlHmeVj2SWVsGmW/du0aPB4Pj/EOY9BDhLX79+/j5z//OURRRDKZ7Otny0EETpPJBKfTiZs3b+L8+fMIBAK4ceMG/H4/80u63S7S6TSePHmCdDrNGg9y0GFMI9TDGPR0Oh1kMhmkUilsbm7iH//xH7G0tIRGo4FyucwB6GHtJIfDgffffx8zMzMIhUKYnZ2F2Wzm7JdGcWl992ZokiQhkUjgq6++QjKZRC6Xe66dSS00l8s1UPwrrVYLv9+P6elppFIp5rEdBEqucrkcfvrTn+KXv/xl399fvnwZP/jBD+DxeBAOhxGJRLiKqQQ9B6NWq2FtbQ1bW1s8RSjnV5G0AFWPt7a2uJ1Fh7DP58Pk5CTMZjNmZ2cxNzcHk8nEf0afQ7yVlZUV5PN53L59Gz/72c9Y8bnT6UCv17O9wvT0NC5cuICLFy/yQMygodFoIBaLIZFIYGNjA4lEAul0Gh6PB16vFx6PBz6fD16v98yOqcvxVt1BuY9QvV5HoVBAuVxmUhWJo5Hn0Mt8Lmk91Go17t+TSNuwgUq7RJotlUqsSbJ3XeX6FSaTiW07SFqfKjVGo7Hv82nEvVKp7CvhT5971iTMXwdyiXfS/yDF5Xw+j0KhwEGHvN1CayO3O6BfcisVp9PJBP2jQG5zQbocB5F5B7FdIzcoFkXxyAdap9Nh/yc58vk8crlcX/uL/tvr9QZu/d4UaJKTFMf325Pl1XyiMVDVmYJ6q9WKRqOBXC6HfD6PVquFSqUCh8PBLVoaQaevyefzKJVKKJVKAMBWIRaLhascg16pI20eWn9q8ZHcgtls5uLAQQkP7SVycrl8r5OD2pUAuOUunwI+brxVQQ8RJkVRxOLiIn7xi18glUohk8lgZ2cH7XYb5XL5pTkFFEyR3PzCwgK63S4mJiYQCAS4/DkMkAc8RAKUV8z2wmKx8Gj/1atXceHCBW6phEIhHk0HwJ/bbDaxu7uLR48e8UG+32fLe/CDjlarxZsubbKlUglfffUVFhYWkMvlWGpBvlFQUKjRaJiTRpUXk8mE6elpnD9/HpOTk7DZbEcm5lP/njgNh/GH6FDSaDTcehgE6HQ6jI2NodfrYXNzE48fP8bu7i5v3MDhnmR7kUwm8eWXX8JsNsPv98Pv98PhcOCTTz7hQ4MI50oA9AySJHF7a2FhAclk8sg2J6SYLAgCstksut0udDodyxBYLBZcuHAB4+PjaLfbSCQSnDhvbm6iUqkgkUigWq1CkiQ4nU4OdG7cuIELFy7A5/PB6XQe8yqcDuRitIuLi1haWkI0GkW9XocgCAgEAnjvvffg9XoRCoUOrfBUKhUkk0k0Gg2WHCCurSiKfV9L1i16vR5TU1O4cuUKt4T3mx5903irgh7i7+TzeTx58gS/+MUvsLOzw5vQi9pYh4G8hiqVCtbX13nzvnHjxhv+Kd5+UCROAQoFPftF5CaTCR6PBw6HAzdu3MD3vvc9mM1m+Hw+2Gy2vqhdHvQQL4X0Hg4Les56ufQooGyKrDh2d3eRyWTw1Vdf4euvv2aF6r2tLDLT1Wq1XF0zGo3sJj06OoqpqSmMjIywt9NRr6darbICM2XS+wU0pHquUqkGynBUq9UiHA7zKDK17uQDFHv1pQ4DtXLJUJSsEahNItc2oUrnMEPO34zH41hdXeUA/KjfT1/barVQLBYhCALi8ThWVlZgNpuRy+WQTCYhiiKWlpb49+l0mhX7O50Om8OOjIzA4/Hg6tWruH79Ole3Bw10ppJMwNraGh4+fMhVfwDw+Xy4cuUKt7cOC0aq1SpisRjK5TLu3buHr7/+GqIocqtMDjpTjEYjPvvsM35PSPl9qIIe4FtrCfnE1n6Q+wjR2CiNRVNVh1pZ8k2LXpTDqhuDDhLhIv0ij8eDRqMBQRDYs4ymIWgKiL6OSvZklLgXFJhSFkY+NgTKcHU6HQwGA7v0DmrWS+sh32h3d3cRj8eRz+dRLpe5ArRfMKHRaFjy3e12IxKJwGg0IhAIsEoqVRFepmomtweRl7T3I7HTyLDD4YDZbB6Yw5qeQzrYwuEwCoUCms0myuUyG7LuXRO5N50cdIDKJ/H0ej0SiQQ2NzdhMBjgcDhYxt9qtfKzP6jP/36Qa66RREO1WuUWO70H8sBwr7Dpfp9Jf06itYIgoFgsstwC8XZIm6zdbvfJPNAgAAnxDfLEFk3u1ut1rsbU63UOADUaDUwmE6xWK6xW677TzkSPII/GRCLBwoaku6RSqXhKFHh2n9RqNZ/rhUIB8XicVbflSs/H9U68dUEPHRIHBSN0YJvNZszNzSEcDsNqtfJ4Lj3cjUYDDx8+xP379/tG2SkoIvG3Qclajwra6NVqNXw+H27evAmDwcCGoqlUCmazGR6PBwaDAVeuXMGHH34Im82GcDgMv9/PG8VeyLOHZrPJLxURY+Vjpw6HA5FIhAXeBpEgKB9jXl5eZsfi3d1d7O7uolarYXd3l8nj+1XaqFVis9nwve99Dx999BH0ej0HnuTW/bLCeOVyGUtLS8hkMlhbW2MPqv0OFKvViitXrmB8fJwJooMAtVoNm83GQchf//Vfo1AoYGdnBw8fPuR9hARP6R5RCT+Xy+37udSy6fV6qFQq+Kd/+ifcvn0bFosFU1NTHLzevHkTHo/nORXgQQa9E9VqFXfv3sXq6ipb3eTzAfANnwAAIABJREFUefaWA8CJEdAvoEcB+kGgVi3ZjMRiMXS7XZTLZTQaDU6ogWcaNDSpdePGDXz88cew2WyYnp7m92oQtdyazSZWV1cRjUaxubmJzc1NpFIpTq70ej1GRkYwPT3NAwz0fNL5nM/nMT8/j2w2i7W1NXz99dfMdSPhWiI/7xWgpEm8+fl5ZDIZWK1W/Omf/il7/BGP6DjwVp40h1VfqDxMY7mzs7MsBe9yuVAqlZBOp1GtVlEoFDA/P98X9Mg1CV7VLfysg6pjVqsV09PT0Gq12N3dRSqVQrPZhNPpxMjICMxmM65evYrPPvuMJ6wO25jlRF3KtvbagZAbtclkgsvlYmfkQcymKAhst9tIpVK4d+8eTwklEokj+Wfp9Xqu6MzNzeGTTz55IyOj8ooTVaAO0rkiiYK5uTlEIpEzP7JKIMKx0WiEzWZDJBIBACwsLECSJKRSKd5H5O110rU6KOgBwFVOAKxsa7fbceXKFQSDQVQqFbbCATCQB+t+kJP4NzY2cPfuXZ5g3CsOSVVO+j6qRL7ovZG37vfySfaCRtztdjsmJydx9epVJjEPMteThAhXVlbYbqhUKsFut8Pr9XLA4vf74XQ6+6q79B5Uq1VsbGwgFothYWEBt2/fRrlcRigUwsjICAwGA4vP0j2nf5eqe9FoFGtrazCZTJiYmMCtW7cA4FiJ429V0EOlMJvNxuZm8qkRqvAYjUY2+hsbG+PpFYvFgk6nA4PBwOxzBQdDpVLBYrHA4/EAAObm5tg40efzMXfnRVmoXGYgHo/zobC3cqHValnZlLgU1JYZRMifXdrAadrksGCbCK/0vJNGzpsQ1JQLvpEEAbUY5cJvcrRaLeTzeSQSCZjN5iObnZ5VmM1mfj5rtRrbIVBAX6lU0Ol0YDKZeEKIWuYHST4A3+oyEedna2sLnU4HbrcbgUBgKAIfqgLTdG4ymeR2CPBtdYcq0cFgEFqtlpNU4guWSiX+rKM8j/QOyqunBoMBXq8X4XAYFouF77lerx+YFu5hoHdd3r7V6/Ww2Wx8nu5tM9Ez3Gg0kEwmsbOzg+3tbbTbbYRCIfh8PoyMjGBychJ6vZ71w4hW0u12WVuJzouNjQ3mPObzefR6PR6OOQ68VVGBXq9HIBCAw+FAvV7Hd77zHWSzWa5MaDQajI2NIRQKsYy+y+ViNji1XdrtNlcThqFk/KrQarWIRCLwer1oNps4d+4cms0mtFot86VIMfmgwISqGb1eD9FoFL/+9a+RTqexsLDwHOfBbrfj1q1bmJubw9jYGPx+P6t7DuJ9kpsYUhuKTC4P+3nlomsjIyMYHx/nqZLXWSd5iVkQBLakINkBavvuDXqq1SoWFhaQyWTQ7Xbx3nvvvfI1nAX4fD589NFHvFZ0qNLhUKlU8OTJE7bDefz4MXK5HLtxH8RDbDQa2NzcRDweR7FYhEajgc/nw9WrV+FwOAY+6CEpi3K5jGw2i5WVFTx48IBb4cAzkuvIyAhMJhOuXbuGW7duwWg0otFooNlsolAo4Pe//z3W1tYgiiJSqdSRgh5SINdqtZiZmcEPf/hDRCIRuFwuBAIB6PV6uN1uuN1u5jsOMqjjQYMsFPRYrVZMTU3B6/UiEAhw8kX7DlXodnd3sb6+jt/+9rfY3NzE+Pg4Pv30UzgcDszOzuLChQvMlaK1pPdHPjn6z//8z/ibv/kbVCoVpNNprK6uwuVywWw28/70pvFW3Vmq9Gi1WrhcLoRCIZhMJp7l12q1OHfuHCYmJvhg2BsR1ut1GI1GtFqtoemTvyrICsJsNkOSJHi9Xq4CvMxLTxkDCUomEgmO2OXQ6XQIBoOYmprigGcYMiqqUlKl50X8G9qgSePFarXCZrO9treTXK4AAFd6DAYDkxf3ux+dTgf5fJ4NOge90kNrvxfyimaz2WQbld3dXa7yHHZ/5Po+er0e0WgU1WoV4XB44NcU6D9oSXE/nU73VcZIJd9msyEUCmFubg5ms5mnDLPZLNbX13ki6Kj7FCUgdLbMzs5iamoKTqcTfr+/z7R3GEB7wd6JTZLDcDqdPLSwX6UnnU7zFOrOzg5CoRDC4TACgQAHPS9qg3e7XTx58gRqtRrdbpf9/9Rq9YGt9jeBtyrokXs6UftKFEWOvDUaDYLBIE89DHo0fpKgTYGCnqOCWh+NRgPxeByxWIxL1vQy0T01Go08wjsMmW21WkUikYAoilhdXcXGxgZXBPYGhERmpTHqa9euwel0IhQKYXx8nMnlrxP0kJN7s9lEKpXCw4cPkUgksLa2xi2a/YjMer0e4XCYVWoHhdPzKiAvIjLJtVgskCQJhUIBa2trrCXWaDRQq9UObHURiZ0c7h8/fgyXywWv1wuv1ztQh6/cKysajWJpaQmpVIq5TnLVcI/Hg8uXL8Pr9WJmZgZOpxMGgwEGgwEWiwUGgwHXr1+Hz+dDuVzG1tYWa16RjU6z2USj0egbrKB/B/jWOd1qtfaJGw5TgkzTc/l8HqIowmAwwO12IxQK4dy5cwiFQggEApwEERdLFEVsbm7i4cOHqNVqOHfuHCKRCC5duoRz587B7XbD5XIdmszKBQzl+41cP+44ubZvVdRARFdJkhAKheByufrGF2nDoUx5GKoEJ4lX2Wjr9Tp2dnZQKBSwtLSE+fl5pNNp7sFTIKvVamG1Wtntnoz8BhnFYhEPHjxAKpXC/fv3ce/ePbaX2BtckL+VwWDA7Ows/vzP/xyhUAhutxter5fX63UOQxpTrVarWF9fx89+9jOsrq7yAb2feioAnpScnZ3F+Pj4vlWQYQAdigaDAWNjYwiHw2i327h69Sra7TZ+97vfoVgsIpFIIJPJHMo3qVQqWF5ehlqtRq/XYx2mGzduwOVyDVTQI5/mfPToEX7605+iWCxia2uLR5ipjTI6Ooo/+ZM/wdjYGDweD0+LUpWt0+lgbGwMzWYTxWIR6+vrKJVKPD1ULBZZEJX+XXrXaAhGp9OxjhJpYA3Seh8FVLUlPo5877l16xZGR0e5ywJ8Sw4vFou4f/8+/uVf/gUjIyP43ve+h0gkgpGREVy8eBEWi+VAORMCJV8kjkt7jrzdRoMDxxGIvlVBD9Cv4/IqGaWcYzJs4+inAbmEeaVSYYdkAgWq1DKgccRhECUkLZxyuczrs3dCRQ4icRKRPBgMcqn5TQT41E8n24lsNotUKvXC76PM2GazsZDfMIMEI/V6PSRJgtVqRa/Xg8/ng8PhgCiKrGAtCMK+5HB6NtRqNYrFIlc95OPxg1J5oAy+1WqhXC4jlUpxNQxAX2JkNps50Cfyvvx5o5Ys8CwYbzQasFgsKJfLbDfRarVY6JNsVug66N8jrt2wVXgItE6kAE/td1LgdzgcvD50plIVrVKpoFgssuJ4OBxmwvJREiKa7qWAlKo6e61uhkan53UhiiILv+3XRlDwZiAf311cXEQ0GsXGxkZfL5ZGgt955x0ugwaDwT4PqUEGuTw3m01YLJYXkpeDwSDcbjdGRkbgdrvhcDheu7ojR7vdRi6XYy+7YeCRHDfoOVapVAiFQvj00085g47FYqjVatja2kIymezTbSLQNFg8HketVuPWMGkHDUIrsVKpYHd3F6VSCTs7O0ilUiyGB4BNQl0uFy5cuMBm0AdNK5JKNo1E2+12njqq1Wo8Al+tVvHo0SOsr69zlUgURZRKJaRSKR4O8Hg8Q0eVkLe31Go1D04Qd1DOb+p2u0gmk4jH45wkeb1eTsxo+OioiVm5XMbKygpX+6gj4HQ6MT4+DrfbDavVemw/+8Dd6Uqlgmg0imw2yyVOBW8eVEkrFot4/Pgxnj59inQ63Ze9UdDz3nvv4fPPP2dBwmFpS2q1WjidTvR6PVgslkODF4PBgHA4jNHRUS7tO53ONxocttttZDIZRKNRpNPpI8v9KzgcdF9HR0fxgx/8AI1Ggx3DC4UCvvjiCxSLRS7fy/ck0juJxWIoFotIJpPI5/P8vJz1oIcSo83NTeTzeUSjURbkpHWwWq24fPkyJicnMTMzA7/fz5OKBz37ZBYbDAYhSRJGR0dx7do1dLtdpFIpJJNJZLNZtn0h8nSr1UKhUGD5hV6vB6fTOXRBj1wk0G63IxKJcKKl1+v71qPb7SIej2N+fh6lUgmCIDC/LxwOs3bXUdewWCxifn4eu7u72NjYQKvVglqthsfjwblz52C324/V+uNU7jSV1mikmUqMcvuDl4V8HI7aLfsxwOnfIWL0oFcbjgMkNEVrTa71RB4EwFLiVLJ2uVxMQB9k9Ho9VqGWt/zkayN3TacNRu6STuv0uhUemtAg8nKhUEChUGA5/sMSAvJdIy6WfPxUeWf2h1arZU6Dw+GA2+3mdSTuzn4gIU+dTsfv1UG2JGcRnU4H9Xodoigyz0leZVSr1bBYLC/97Mt5nSRXQmPxJIRnNpuh0+n4vaQzotFocBA0jAK1wLeJa6/X6yOT732/5SKrkiRxC5Iq0TT59iItN3qmyTUhn8+z7REF+AaD4dh1kk406JGz+NfX17GxsQFJkpiVb7VaMT4+/tJRHr1UnU4H8Xgcjx494nLc3o1drVbD4XBwP3LQD+HjQKPRwMrKCit6km6DXOPFaDTCbDazQ+/o6CiPbA8ySqUSbt++jWg0ilwuh9XVVZRKJcRiMa6CyVWpr1y5gomJCbjdbrzzzjvs/fO6Vg/tdpv1MO7fv4/l5WWUSiUsLy8jnU4jm80eyi+yWCy4fv06Zmdn4fV6cfXq1aGZuntVUNDT7XYxPj4Op9OJbDaLR48esfDjfod5vV5nVex8Po9qtQqVSjUwRpdkPJnJZFAul58LMoxGIyYnJ/HOO+/A5XK9lhov2R/QIer3++Fyudhviya7UqkUc1iUbsDhIF4mTdGNjo6yrhFpHL2IG9VsNlm49unTp/jjH/+IaDSKdrvN5qPE4yLNvePCiQY9ckJUNBrF119/zfowJH9NhKiXAUXxzWYT6XQaa2tr2NnZgSiKz/EWaEqG/ESGpdXyJtFsNhGLxbC0tIStra2+Xi8AVvul6gUJXQ1DCblareLBgwe4d+8e8vk8Njc3udLTbrf71JkdDgeuXLmCDz74AHa7nX1uDvI2exnQ5Eq5XMajR4/wq1/9CtVqFTs7O9xqkVuE7IXBYMDly5fxne98BxaLBaOjozzhoQQ9+0Pucm+xWBAMBpFOp+F2u6HT6Q48XOkwbrfbqFQqqNVq0Ol0A8O5IlPRbDa77xi/wWBAJBLBzMwMVxRfB+TfRAKc1L7K5/MAniUEhUIBRqMRoVBoYCpqLwu5CeiLQHuS0WjEhQsXEA6H+0RsX4RWq8UWPCsrK5ifn8fW1hZ8Ph8ikQirQJNC/MAEPc1mE6VSCbVajQmVNK9frVaZXCU/HA4DkWkbjQb7bWWz2T7mPoGE18jCQqn0vDxIV4HuXyqVQj6ff66NSP12u91+KCFxEEG+cGazGaIo9smvUwtW7sZN01SdTofbrq8biEuSxAEPaZjQe0fB115Rsv1+DqPRCKvVCrPZzKO9ZzVJIONQSZJYD0luvvumQc/7y0wHyQ+hQWq50J5B7STgW+NjcvMmKZI3OU1F75Ner0e9XufPJTNSUgUepLV+GdB6EMFbr9dzoUA+PUh7AWkmmUym58jO+4FaWY1GA8ViETs7O+zzRdQWi8WCkZEROBwOTviO02EdOMGgR5IkpNNpLC4uolAo4A9/+AO+/PJL9rAxGAyYmZlh1229Xg+r1XrghiQ/OBKJBH7zm99ge3sbKysrSKfTqFQqPA5HHlNEFr1+/Tref/992O32gTaVe5OgKROaNLl79y5+//vfM39KDiKlXbhwAV6vFy6Xa2iCHr1ej7GxMXQ6HaytrWFxcZEdt2lzpSBIp9Mhk8lgZ2eHp3ooy31d5WVyPs5kMtyCpBFV2tQO2+zpHo6Pj/NUB5Wxz9rIuiRJ2N3dxZMnT9BsNuH1euHxeNh7aVDaSG8jJElCrVZDOp3miSpJkqDRaODxeGC32xEKhfj5epOcMeJT0Sg7JdGtVoslAsjDa5hBxPt4PA6v14tarcZJGHEzR0ZGWDqDrIkOI5pT4rW6uoqtrS2k02l88cUXWF9fZ66jTqfD3Nwc/uqv/gp+vx8jIyOwWCzH7sd4opUeKq/ncjm2s2+1Wn0EyVwux864JIN9EKhKVKlUsL6+jqWlJSSTSZaJJ1B/l9xzR0ZG2F38ON1cBwn0EFerVRSLRQ4wSXNBDkEQYLVaEQgE4PF4jtU87m0DTWwFAgHeWPdOSRGRs9FooFqtolQqsdbLm3jZaSKIVG8zmcxLTzKS2anL5eozgTyroAkiaoOrVCqYTCbYbLbTvrSBR6vV4mEHMmSl54sczg0Gwxuv9Mg1wur1Ogc9VNnQarWHqmYPAwRBQKfTQS6XgyRJyOfzLBgor/Q4HA44HI6X+uxOp4NMJoP19XXs7u7i8ePHWF5e7hsm8vv9uHbtGoLBIGtfHTdOvL1FapnEpKdf8lJ/q9ViJv5eyCdSEokEq3ImEgnkcrl9iXJ004LBIPx+PywWS5+ys4IXo9frIZvNMmmZnKflFQwKIg0GA/x+PyYnJ5k7NSzr3G63kc1mnyvjykGbO/F6fD4fBxdHQaPR4OQBeL4n3+l0sL6+juXlZRSLRRSLxUM/z+PxIBgMskCcSqWC3+9nHtZZq+zsB0q4yuUyT48YjUbk83l4PB6YTCYEAgEWznyVAI/2MHlVlKQz9ho7ymE2m2GxWGCz2bj6MQyTjpS00lSPvL33JvYL8hZ0Op3odDpDwSk8Kqj74XQ6mXZCHnsLCwsQRREulws+n4/bwYc9j3I5hlKphGKxiGq1isXFRayuriKfz6NWq0GlUsFms7Gz/djYGHN4TmqfOdH2VqlUwsbGBk+PEKeAODytVguiKKJSqUCj0ezLOaCXpFAo4M6dO1heXuaJrd3dXVZ6lEOr1WJsbAyXLl1COByG1+tld+9B2NBPAu12GysrK/jiiy+QyWQQj8e5gkGbldFoZGEpIsHabDZYrdahCXpqtRqWlpZw584dpNNprlrKodPpmOg9Pj6Oy5cvw2azHbkiViqV8POf/xx37tzZV+2XWsnpdBrNZhPZbPbAKo8gCJiensYPf/hDJhLSL3JKPostLTkkSUIymcTt27eRTqd5ipA8xVwuFyKRCD7//HNMTk6+suVHq9VCqVRCs9nE2toalpeXkc/nsbq6imq1+pzXEMHj8WB2dhZOpxOzs7OIRCKsXj6okCsnU8tVPjr9JqBWq+F2uzExMQGdTjdUFecXgaosMzMzKBaLqNVqKJfLWF9fxz/8wz/A7XbzkIXVamUhx4NQr9eRyWR4/3v8+DFKpRLm5+fZk45kGUZGRvCjH/0IkUgE586dY67QSZ0RJxr6ttttLnPKM2A5uZMiRrlOhTwDoGyq0Wggk8kgFoshlUqhUCg8xy0hkA6E1+uF0+kcGnfvNwW6N6VSCbu7u6yvsDdrVavVfGBSi2fYOFOkz5PL5ZhXJoe8tEsaRhTwqNXqF2a51GaMx+NYXl5mif+994IqPCT/vx/ogKHgy+PxwGKxsNYM9fAHQT270Wgw+Z6qWXq9Ho1GA/l8nqszlIDJy/uH8RbkoJYljZ7TpFy5XOYK0N5Ejkj/TqcTbrebrT6o3TMIkK8f/Z4q9hQIykn+bwLy6jNNBO2355/lYP51QORku93Oz3y320W1WsX29jbbTJBVislkQrfbPfBdIGsbURSRTqfZCHZnZwfxeJwr2xqNBlarFZFIBJOTk/D5fNDpdCd6H96qt4rG2ra3t3nSwmg0cqBEFR4qn927dw+bm5solUosaU4QBAE2m41H4c6dO4dLly7B4XAcq8T1IEGSJKRSKezs7KBUKmFxcRGxWIy9hQhUCSDyMrn1DmNgqdfrEYlEMDs7i93dXSZvEkgcTRAElmOn1sbExAScTicsFgtvBlS5pGe/WCwiHo9ze4vI+vLDQpIk5q7I/444DiqViv29TCYTzp07h2AwCKfTyUErTb0MQsCzFxQEdjodpNNp1Go1AMCdO3eQyWRgtVrh8XiYy0QHpsVigclkQrvdZv8oORKJBJaXl7mVFovFWOp/76FO06mkx0OKuE6n80hib2cJJpMJfr+fn3mykaA9noJRUkl2u90v3dprNps8eUTnRaVSwYMHD7C4uIh8Po9SqQSgf8JSp9MNzDq/DHQ6HSYmJtBoNBCPx9mWRq1WI5vNolwu48GDB+h2uxwcUet3P9Bz3mw2sbm5ifX1dd6DzGYzzGYzZmZm4PF4MD09jcnJSdb9Oulz4q0Keur1OqLRKOsrVKtV6PV6JBIJRKNR9rGJx+M8pk7Z9N5slj5jYmICXq8X169fx0cffcQbmYIXQ5IkbG1t4de//jWy2Szu3LmDxcVFVowFvj1IBUFAKBTChx9+iEAggKmpqYHJVF8GFETo9XosLi5icXGRuTcEGlGXJAl3795FIpGAy+XC5cuX4fP5MDo6CqvVygq9FNQvLi7yRBa1dQ/S2pC3veQq2eTnFAqFcP36dXg8Hly8eBETExM8lSFvaQ1aJiw3XQXAAWkmk2GhNI/Hg8nJSeY8eL1envyksd7l5WVsbm72fe7CwgJ+85vfoFAosPZOr9fj+y2HXI3b6/ViZmYGXq8XXq/3ORuAswwaahgfH4fZbGZeYKvVYrVyh8OBWCzGa/0qfCbyLavX69jZ2UE0GkW5XMbdu3exsLCAZrPJOj1UYZVPIQ0bjEYjLl26hJGREayvr2N7e5tVs3d3d9FqtbCzs4N79+5xVfQwzqHcvLRer3MRggQgg8Egvvvd7+LChQvw+Xy4ePEi7HY71Gr1iT/rJ/5mHZY59no9NoSjbFSv17OmjyiKSCaTSCQS3D/fK7BGJXuaDiCBPNKMGaQM6rggLz2LosjTP+VymZWvCRqNhjUbqK3lcrlYHGzYQEEgSanvtwbyNi1pS1HGSyOhVFYml/Z6vY5CocD3gjgi9FkHXYu8lUbVCjJCdbvd8Hg8cDgczHEhtehBA1nc6PV6lgegti0A1p6iP7darWwdQhwf4qeJoohCocDTecCzvSudTiOZTKJQKBx6X4D+AJQOYJLVGLT3hlpMZrOZEyQAfB/IubtQKHBQCRzuuk2BPFXQaAqyVqshn89ztYKMpylRo3eCFONPurXytoBI3jTkY7VaYbVa+94LMoWV64cdtFadTmdfbTySniFjV1LINplMpzY5faJBz96Hfy9EUcTS0hISiQSMRiMcDgc0Gg1KpRIKhQLa7Tby+TzK5fJzbsUEq9UKt9sNs9mMGzdu4OOPP2bOwpschxxkUKAjiiIWFxfx6NEjFAoFpFKp53xaqHXodDpx7do1zM7O8kE6jJtJvV7H+vo6nj59yi7bB6Hb7aJQKKDVajGZ0GKxwOPx4OnTpzCbzSyqR5lXIpGAKIrI5/OHHqo6nY7ft3fffRfnz5/nTUij0cDtdmN6ehpWqxVer5d5PINSYdiLYDCIW7duIZVKYW1tDSsrK33BT7PZRDKZRKlU4nYLOXeTNYLf74fH44Eoinj8+DHi8Xjfv5FKpVCr1Y4kDeB2u/Huu+/C4/HgnXfeYaNFl8s1cO+N1WrF2NgYrFYrlpeXodVq+9a+WCziq6++wtraGpxOJxO5KTChSgOdGfJ7ViwWefook8nwhHA2m0Wj0UAikeCqAw0QkFbb+Pg4RkdHBzLIfxFINV+tVsPv9+P999+Hz+djPatSqYRyucwBY7fbPbQNRYERVZHD4TCMRiPGxsbYhf3y5cv856e55ie2w1EPm0hl+/VSRVHE6uoqByd7SW/Atw88/flemM1mjI2NweFw4ObNm/izP/szWCwWLtkreDFEUWQy28LCAjPx905YqFQquFwuXLlyBaOjozh//jxmZmY44BnGALPRaGBzc5MDxb1cMzloRLRYLEKlUmF7e5tVw+12O7RaLRORya26VqsdSF6WQ6PRwGg0wuVy4bPPPsOPfvSjvmyNkgrKdAf5fgmCgEAggA8++ICrOeSFRhWZZrOJTCbTV10gcTubzQaNRsOGmK1WC1tbW32VHgBHui8Et9uNDz74AOPj45iYmMDs7CyMRuPA3QfiVo6NjXFQR88hrT0FPXLtHr1eD6fTCY/Hw55mJpOJ25O9Xg/lchmxWIytO0qlEh/QZI4pF/0kZXFq7c7NzZ2YNszbBiLQS5IEv9+P69evY2pqCqurqyy7QWfAfrIbe0EcKb1ej8nJSbz//vsc6ExNTUGn03Hn5rSnpk+80mO1WlGv1/d90OTl5qOANgf5iCmV0MixmiYhlIDncNAGQdLh5MhNVQY5Z4qyBBJ8dLlcPKp+XLL+ZwUkeme1WtFsNqHX65lQvN+zLV93Ug8Hvm2TyadbiCfyItD3UtuECP1EnCWxzkHijrwIJBEgSRJcLhdcLhdzD+ig3Ov+DTxrQ9VqNeYeUIWBDCyPAtrkKfHTaDR994aI44P63pDXHL0XFNATH5Oeb1ojrVaLRqPB66bT6dBsNjmBkGshlUolVKtVtpWgxEzOOaQ2s9zFm8wzh+X5Pwi0PiaTCZ1Oh3lVGo2GubIHTX/KodfruWUVCATg9Xphs9nY9UBOgzhtnOgd9/v9eO+995DP55FOpzE/P3+kKHI/0ENMgc7c3BzcbjcmJyfx3nvvwW63Y2RkhAMeJeg5HGTCV6/XsbKygj/84Q9Ip9NYWVl5bnOnEqbX68X09DRu3LiByclJVlYdZtjtdty6dQuRSASrq6tsk0L6Uwe1Pij4oUOVKmryPz+qorJarUYgEMDc3Bw8Hg9GR0f7xs/pa96GDegkIAgCPB4PLl++jHq9zmP6VNHMZDKoVqtIJBLPTWSROq1arUaj0UChUGAvqaP+28STMhqNCAaDsNvtmJ2dxcTEBEZGRuByuQb6XpDwo81mwzvvvINsNotCocCmk0C/6CxxOkVRRC5hLBaGAAAgAElEQVSXYz4WBSiUIJCuG/F19soC0NqTKjNdg9frZb+vQausvQqodUuFglAoxNNw5XL5SIasdI9IG8nr9UKn0zF/5206g0+0veVwODA1NQW32w2fz/daD5s8C3A4HJibm+NR4Vu3bjEz/G1Z6LcdpC9D3lqLi4tIJBJIp9PPHbY6nQ5erxejo6OYmJjAuXPncO7cuVO68rcLJpMJc3NzGBkZgclkwpMnT1jqXhTFQwMXeTn+KNnVQRAEAU6nkycXfT4fLBbLUL8LVO2iqo7BYEC5XIbZbIbBYECxWOTpHjn2q/68LIhAbTabEQ6H4fP5MD4+3mfTMshBD7WQzGYzJiYmcPnyZRY4jUajfXSFdrv9yonwfqC1p3YuHezDVOV8ETQaDVtM+Hw+TE1N8d+9jG7S3vP8bQ0mT/SuU1m32+3C7/djenqaNRXkjtS9Xo8fVjm/h7gIFL3T1EkgEMDo6Ch8Ph87pysR/MuBTPhoUo4CIGrNyEHaIj6fD06ncyiJgAeBsktyJR4fH4darUY0GkU+n38jGzqNO8uHAQRB4BFrvV6P2dlZdi8e1km6/UDtR7L9iEQiEASBq5x2ux31ep35Ift5yx0Fcr0jl8vF0zETExPw+/0IBoM8PTTIAY8c9Iz6fD5oNBqMjIywcCONO9PoMyUAhwWcOp2OHb/lIBI08dqotUammaFQaCh5PEfFfmKSg4QTDXqMRiN8Ph/sdjs+/fRTWK1WFItF3L9/H2tra2g0Gshms6jX6zCbzQgEAhzkUKAzMjKCUCgEk8mE6elp+P1+GAwGuN1ufriJ8T+IN+xNgwTTSqUS7t27h7W1NcRiMWxsbKBYLLIjuBwGgwEzMzO4desWPB6PIvYog0ajYYVlsnEolUr413/9V2xvbz8nsfCq/0YoFGL3eqponj9/Hp988gmcTiccDgcL7LndbuVd+AYqlQqBQABWqxWdTgeTk5Msk7GysoJisYiNjQ3cuXOH/YOq1epLm7WGw2FMTU3BYrFgcnISkUgEFosFU1NT7PVFgdewVKQ1Gg1GR0dhs9lQqVTgcrlw7do1iKKIeDyOcrmMQqHA70m9XkelUtm32kAE6fPnz8PpdPKfq9VqjI+PY3Jyknk7xOORK467XK6T/NEVvEU4cSIzVXBGR0fR6/WQy+WQSCSQTCZ57I2+lrJU0hAxGo2YmprCxMQEM8MpUxuGTeO4IEkSGo0GkskkNjc3kUgkUCgUUKlU9v16rVYLj8eDSCTC5GUFz0Akb+Dbikyj0cDq6uobM5DUaDS8cRPBVqVSYXp6Grdu3eozCVTejX4IgsAKsURqbrfbKJfLMBqNKBQKUKlUWF1dZZPQlw0YqaIRDodhtVoxOzvLYocTExNDe+AKgsDEViKQW61WlEolaLVa5HI5aDQaFAoFAM+qzyqV6kArF71eD6/XC7/fz3+n1WoxNzeHK1eu9HmsEQ1i2DmHCk5JkZk2BVI6vXbtGux2OxMFqcwcCAS4jE/VnlAohEAgwOXjQZTJPymQI261WsXu7i52d3eRTCYPHFOktabMdNBHnV8XRO6jsdArV64gHA7z38sFwFqtFsrlMprNJgwGA+x2O7fJTCZTXwvEYrHg/PnzCAaDfXYGMzMzMJvNPAmkvBsvBiVaZJar1+tZLZi0kw6qNhwEQRC42mA2mzE6Ogq3233q+iRvAyhgoaS22+3CZrMxcblYLCISifBkHdEe9n4GADgcDszOzjIfBXiWaBBxnxJsmphTgn8FwCkFPcTwttls6Ha7mJ6e5p4ujRyq1WqezpLrZlAQRFm08iC/OlqtFmKxGHZ2drC+vs62CXKZfjlo4ocOVfqlHKz7gzR3dDod5ubm8Bd/8Rd9nmWNRgMbGxscaFJ1weFw4Pz587DZbAgGgxgdHe2rElksFly6dKmvyikIAoxGY5/quHJfDgeNR5NitcFgQK/Xw+TkJK5fv86k8pdpbRGorSKX8KfDfphBz6rBYEAkEoHf70ev18OlS5eYw0O8TrmVyn6gYFVOSKYzgpSt5b+GhTul4HCcGn1dLndvs9lO6zKGGmRMSaRl4i8clFnR4SDPnJSD9WDINUJsNhvC4XCfWGG9Xke1WuUAkwIkmjRxOBzw+/2IRCJ9FQKqHoTDYaV99ZqQD0nQ4Umq2AqOB7TmFBgqUHCSUGb2hhiU6RIB3OVywe/3o9FosLeT0WhkReuZmRmMj4/D5/PhwoULbI447NnrQZAHJHa7HePj431tw3a7DafTiUKhgFqthnfffReVSgUOh4Ol+B0OB3w+X1+Wqtfrh34EXYECBQpeBUrQM8SQq5RarVb4/X5Uq1UUi0U0Gg10u10eMXU4HPj888/x/e9/HyaTiUXWlLLx4SDek9vtZkVgglyXhxSbqbWr0Wg4aNrbQiQej1JlU6BAgYKXgxL0DDHo8KTJOHLC1Wg0bHlAFhM0Au3xeJhcq4h7HR3DpICsQIECBW8rlFNriKHRaODz+WAymeD3++FwOHiCiJzs97a3yDBQaWkpUKBAgYKzBiXoGWLQFJ3L5YIkSZidnd3XwV7u16SQlxUoUKBAwVmFEvQMOeSjzUr7RYECBQoUDDKU8Q8FChQoUKBAwVBACXoUKFCgQIECBUMBJehRoECBAgUKFAwFhJf0lMkAiB7f5QwkxiRJ8r7uhyhr/8pQ1v90oaz/6UFZ+9OFsv6ni33X/6WCHgUKFChQoECBgrMKpb2lQIECBQoUKBgKKEGPAgUKFChQoGAocKaDHkEQDIIg3BYE4ZEgCE8FQfifT/uahgmCIMwKgvBQ9qssCML/cNrXNSwQBOHfCIKwLAjCmiAI/9NpX8+wQRCELUEQHn/z7N897esZNgiC8D9+s+8/EQThPwiCoFi2nxAEQXAIgvB3giAsCYKwKAjCrdO+pqPiTHN6hGeqemZJkqqCIGgBfAngv5ck6Y+nfGlDB0EQ1ADiAG5KkqSQ7o4Z36z3CoDPAewAuAPg30mStHCqFzZEEARhC8B1SZKyp30twwZBEMJ4tt9fkCSpLgjCfwTwU0mS/q/TvbLhgCAI/x7A7yRJ+htBEHQATJIkFU/7uo6CM13pkZ6h+s3/ar/5dXajuLON7wNYVwKeE8MNAGuSJG1IktQC8P8A+MtTviYFCk4SGgBGQRA0AEwAdk/5eoYCgiDYAXwK4H8HAEmSWmcl4AHOeNADPMt4BUF4CCAN4F8kSfr6tK9pSPFvAfyH076IIUIYwLbs/3e++TMFJwcJwC8EQbgnCMJ/ddoXM0yQJCkO4H8BEAOQAFCSJOkXp3tVQ4MJABkA/6cgCA8EQfgbQRDMp31RR8WZD3okSepKknQVQATADUEQLp32NQ0bvilv/gWA//e0r0WBghPEx5IkvQvgzwD8N4IgfHraFzQsEATBiWeVzQkAIQBmQRD+i9O9qqGBBsC7AP5XSZKuARABnBlO4ZkPegjflNd+DeDfnPa1DCH+DMB9SZJSp30hQ4Q4gBHZ/0e++TMFJ4Rvqg2QJCkN4D/hWctRwcngTwBsSpKUkSSpDeDHAD485WsaFuwA2JF1Vf4Oz4KgM4EzHfQIguAVBMHxze+NeEbqXDrdqxpK/Dsora2Txh0AM4IgTHxTafu3AP7plK9paCAIglkQBCv9HsAPADw53asaKsQAfCAIgumbgZbvA1g85WsaCkiSlASwLQjC7Dd/9H0AZ2aAQnPaF/CaCAL4999MsqgA/EdJkv75lK9pqPDNhv85gP/6tK9lmCBJUkcQhP8WwP8HQA3g/5Ak6ekpX9YwwQ/gPz07b6EB8LeSJP38dC9peCBJ0teCIPwdgPsAOgAeAPjfTveqhgr/HYD/+5uEawPAf3nK13NknOmRdQUKFChQoECBgqPiTLe3FChQoECBAgUKjgol6FGgQIECBQoUDAWUoEeBAgUKFChQMBR4KSKzx+ORxsfHj+lSBhNbW1vIZrPC636Osvavhnv37mUlSfK+7uco6/9qUNb/9KDsPacL5dk/XRy0/i8V9IyPj+PuXcVX72Vw/fr1N/I5ytq/GgRBeCO2GMr6vxqU9T89KHvP6UJ59k8XB62/0t5SoECBAgUKFAwFlKBHgQIFChQoUDAUUIIeBQoUKFCgQMFQQAl6FChQoECBAgVDgbNuQ6HgBCFJEiRJQq/XQ7vdRq/XQ7fbRafTQa/XAwB8I8sPtVoNjUYDQRCg1Wr59/T3ChScBdAzLkkSut0uP+cEQRCgVquhUh2cP77o7xUoUHByUIIeBUeCJEmo1+toNBqo1+uIxWLI5XIol8uIRqOoVCoc6KhUKvh8Pvj9fphMJoyNjSEYDEKj0UCv10OtVp/2j6NAwZFQLpeRz+fRarWQz+dRLBb7Ah+TyQSPxwOj0bjv96vVarjdbtjt9pO6ZAUKFBwCJehRcCRIkoRms4lKpYJSqYSlpSVEo1Ekk0ncu3cP6XQaarUaer0eOp0OU1NTOH/+PBwOB7RaLRwOB3Q6HTQajRL0KDgTkCQJtVoNqVQKtVoNsVgM8Xi8L+ix2+2YnJyEzWbb9zN0Oh2MRiNsNptS5VSg4C2AEvQoeCGotC+KIme7mUwGyWQSmUwG1WoV9XodarUavV4PrVYLxWIRyWQS9Xod29vbMBqNMJlMCIfDsNvtUKlUSsn/GCFvPeZyORSLxb6/12q1MJlMUKvVMJlMsNlsyv34Bs1mE6Ioot1uY2dnB+vr66jX60gkEkgmk31fK4oiVCoVLBYL/5kgCLDb7bDb7dwSVqBAwdsBJehRcCjo8Gw0Gtja2sLS0hKy2Sy+/PJLrKysoNlsolQqodlsAviWv1Cr1bC9vQ29Xo/l5WUEg0GEQiH88Ic/xNzcHLRaLQwGg1L1OSY0Gg2IoohisYj//J//M373u9+h2+3y37vdbpw7dw52ux1zc3O4ceMGzGbzKV7x24NMJoOHDx+iWCziyZMnuH37Nmq1Grd35UGMRqOB2WyGRvPtVqrVavHRRx/hs88+g8FgOI0fQYECBQdACXoUHAqq8nQ6HRQKBWxvbyOdTmN9fR2rq6sHfl+lUgHw7ADI5/OIxWKYmJjA9evXMTk5CQDQ6/Un8jMMGyRJQrvdRr1eR7FYxIMHD/DTn/4U7XabvyYSiaBcLsPn88FqteLatWuneMVvF6rVKra2tpBKpTA/P48//vGPqNfrR/5+nU6HSCSCTqdzjFepQIGCV8FABD20yfd6PdTrdZRKJXQ6HdRqNdRqNXQ6HbTb7b5Nfy90Oh30ej00Gg3cbjfcbjc0Gg10Oh20Wu0J/jRvD3q9HtLpNHZ2dlCtVrG0tISNjQ0Ui0XUarXnvp4mWQRB4IkX4gJVq1UUi0Vsb2/D7XbD5XJhbGysL0NW8GYgSRJKpRJisRjS6TQqlcpzLZZ2u41CoQBJklCpVJ6bSho2UGWyUqlwQJ/JZJDJZF56bXq9HpLJJB48eACHw8H7kF6vh9vthslkOqafQoGCNwua1s3lcigUCmi327z/63Q6WCwWaLVaqNVq6HS6Pt6aJEloNBpot9vodDr8e+J+qlQqaLXavnPX5XJBpVIdK/9tIE6cTqeDarWKdruN7e1tLCwsoFwuIx6PIxqNotFooFAo9PEa9i6qw+GAz+eDyWTCrVu38MEHH8BkMsHr9Q5d0EPBSqfTweLiIn72s58hl8thdXUV6+vraLfbqFarz30fBYmCIKDT6fAoe7VaRa1WQ6/Xwx//+EdkMhlMT0/D5XIpLZVjQK/XQywWw5dffolMJoNEIgGg/5kXRRGbm5swGAy4cOHC0Fcl0uk0fvKTn2BpaQnJZBKLi4uoVCqo1+totVov9VmdTgcPHjzA1tYWbDYbPvzwQ1y+fBkejwc3b97E6OjoMf0UChS8OdA50Gg08OTJEzx48ACFQgGPHz9GLBaD0+nE9PQ0HA4H8wLlSWyn00Emk0GhUEC9Xsfu7i7K5TKMRiNcLhcMBgMcDge8Xi9MJhNu3ryJmzdvQqfTHWvgc6aDHspeu90uWq0Wms0myuUykskkisUiNjY2sLKygkajgVQqhUKhcOBneTwehEIhWK1WjI+PQxRFPryHEcTloXXM5XLY3t5GMpl8rmpADydF7vLPAJ5VFSRJQrVaRTabhdFohNPpPLTypuDVIUkSRFFEOp1GLpfjqpycqNzpdCCKIjqdDur1+tBWemhjp0rPysoKstks4vH4S7W09qJYLKJYLMJsNiMUCsHlcqHb7XLwr2hWHQ/ke9NRCeR0H5T70Q85tSGfz2N7exvZbBZPnjzBxsYGXC4XJEmCy+WC1WqFy+Xq2//b7TZ2d3eRTqdRq9WwtbWFQqEAs9kMv98Po9EIj8cDURRhtVoxOzuLTqdz7JpuZyrooYOzVqtxma1er6NSqXBpOplMYm1tjTf9QqGAZrP5wgO22WyiUCig1WohmUwiHo/D4XDA4XDA5XKd0E/4doDWURRFrK+vY3d3F8VikXk6APihtFgsCAQCMJlMcDqdiEQi0Gg0SKfT2N3dRavVQiqVQjab5SBKr9cjGAyiXq+j3W5DpVIphOY3gFarhVqthkajwetfKBS4KqdSqVg2wOfzYW5uDi6XCxcvXhxKfpUoirh//z7W1taQSCSwsLCAZDKJarXaR/p+HXQ6HcTjcWg0Gv7v5uYmrFYrQqEQTCYTjEYjrFarMj33kqDWS6/X48nSVquFSqWCcrnMFIdGo3HgZ7hcLgQCAW49ejyeoQ5+5CKcqVQK8XgclUoF8/PzWFpaQqlU4v2k3W7z+2IwGJBMJvue4V6vh0KhgEqlwvQStVrNLfVGo8F7lslkQiAQgMPhgNVqxejoKLxe77H8jGcq6KFFTKVSqFarWF9fRzKZRDabxfz8PLLZLKrVKkqlEkeonU6HWzWHoV6vo9lswmAwIBaLYW1tDW63G8Fg8IR+urcHhUIB8/PzyOVymJ+fx9raGiqVCvN0gG+ntJxOJ65du4ZAIIDJyUlcv34dJpMJ8/PzuH//PorFIu7evYtcLodWq4V0Oo1ms4lAIABRFNFsNqHVao+9jzsMaDQaLCEQjUaxvr6OcrmMcrkMSZKgVqthsVhgMBgwMzODv/zLv2ThyGHkmZRKJfz4xz/G3//936PVavGYOu0dbwLNZhMrKyvY3NyExWLB1tYWvF4vxsbG8PHHH8Pv98Pn88FoNEKn072Rf3NYQFzObreLVCqFpaUlVCoVbG1tYXNzE/V6HalUCvl8ft/vFQQBc3Nz+OCDD+B0OnH58mU4nc6h5hl2u10uEqytreHLL79EPp/HvXv38PjxY7TbbW73iqKIra0t5nHuF7R3u110u11W5tfr9cwRonug0Wj4DGg0GvB4PPjud797bAHoW3135dySTqeDVquFcrnM2Ws2m0U2m0U6nUY6nUY2m0Wz2USz2eTDmUpltKFQdkCfTa0XujkqlQqtVgutVovJ0XRzhgWktdNsNtFqtdDpdDjzpfUkZWWLxQKn0wmPxwOfz8dVn0QiwRUyuQpzu91Go9FAs9nkz1WqPG8GrVYL1WoV5XIZoihCFEXU6/W+qoVarWaNHrfbDZ/PB5vNNpT3gCYSd3Z2+B2nzVur1T73ztOmTl9z2J5Aewr9vtlsQqVSIZvNotPpwGg0Ip/PQ6fTwWQyodPpcCIxTHvNq4CSr3a7zYMqpJxdLpf5TBBFEclkErlcru/76WwQBAEul4sP4Gq1ygf6sOqI0d5PFbNsNsvabPIKqCAI/FwfBgqI5MK1zWaTSc3As8RAp9OhVCohn89Dq9W+NI/uZfDWBj3yzGtpaQmPHj2CKIrI5XJcxsxms6hUKqjValxy83q9cLvd0Ov1cLlccLvdfRt6u93mCkMmk8HTp09RKpX47wVBgMFggM1mg9VqfY6RPgywWq2YmpqC1+tFJpPBysoKSqUSbxYGgwETExNsNfHee+/B5/PB7XZzphSJRNBoNJDL5bC+vo719XUuQ1MrkQ5nSZKg1+uHbp3fNLa3t/GTn/wE6XQai4uLKBQKXEIGngWsJpMJVqsVDocDHo8HXq8XRqNxKDd4OXQ6HZMriWgp5ycIggCHwwG73Q6tVgun03kgCZ+m50qlErfhq9UqD1Rsb2+jWq2iUqnAbrfj6tWrAJ4NU9hsNtjtduVdOADNZpPb7cViEWtrayyWGo1GUavVUCgUkMvl0Ol0mIx+EL9nd3cXjx49YkVtSZJgsVgQiUSOrb3yNqNQKGB5eRnlchn37t3Do0ePUCqVXmmKkSay7HY7bDYbpqen4ff7EY1G8fvf/x7pdJq/lnh1uVwOGo3m0Jbk6+KtDnqKxSJEUcTXX3+Nv/3bv+VRuUajgV6vx60rjUYDo9HIi3zlyhWWh5+amurL2prNJlKpFCqVChYXF7G9vd0X9ADPDnW6UcM2uQV8G/TQBjM2NoZSqcTkZqfTiY8//hjnzp2Dy+XC+fPn4XK5OKKXJAnhcBhGoxGZTAa3b9+G0WhEs9nkNiL1eqvVKis5D/vB+zqQJAmxWAw///nPEY1GudIjVwRWqVQwmUyw2+19QY9Cqn0W9Hi9XjgcDrjdboyPj/f5aWm1WoyMjCAUCsFsNmN8fBwej2ffz+r1eojH49je3ka9Xkc0GkUikUA2m8UXX3zBdhaLi4tQqVQolUoYHR2F3+9HJBJRLCsOQavVwvb2NqLRKHZ2dvDb3/6Wp4IoGQa+HaIgzs9BSCQSaDQafffa5XLx5O6wgQQ50+k0Hjx4gEePHjH5/mWVxdVqNbxeLyKRCHw+H77zne9genoad+7cwcLCwr5BTyaTgVarHc6gh4IaElmjsj21XIBvnbyNRiPcbjeMRiP3yG02G1wuFxwOBzQaDbev6OCmvqWco0IlOMr2DAbDUPZ3ifAKPPMWIuNQ2kDsdjs8Hg+cTidsNhuMRuNzwaFGo4HBYOA11Gg0zJOgsmg+n0c6neYsmkqhCl4N/z97b/IcV3ql/T0353kekUjME2eAZI2sKpY+Kbpld7fcrd7IHfbKqw47PIQ3/iMc3nnjCDu8cajD4f4UbbVaXVapmyWVVMUqsIoTCBAzkJkAcp7nm5leUOfwJgiQ4IQEkO8vAlFFEkPi3pvve95znvMc6g4ql8toNBoHLvbKEk6/pvAJjUYDl8uFoaEhGI1GhEIhXjcoA7b/c6k912q1HprpabfbsNlscDgc0Ov1XBbodDqcQaYyL/Bko4nH4+h0OnA4HGJsxQGQ3KBYLCKdTiORSCCZTCKXy6FQKKBSqRypYQXo7uoit3kAXBrrdDr8Pakc3C/vE7rGlJ08zN+ONDqUbACeBpqNRgO1Wo31OrSfWiwWWK1WGI3GA9d5CqxeJcB6GU7sjt5sNlmESYpxijhbrRb0ej1CoRAcDgcCgQDeffdd+P1+uFwuhEIhGAwGWK1W2Gw2tNttpFIp5HI5ZLNZ3Lt3D+vr62zcBoBPwJTaHB0dhc1m65qp0y+o1WoYDAbodDpcvnwZDocDjUaDswZ6vR5+vx92u52DRCWSJMFoNEKtVkOWZS4TkqANeHLC+vWvf435+Xm89957PKtIDCR9dcgArJ9b0F8Gu92On/70p5idnYVGo+HnVK/Xw2q1dh14qFPRZDLx6InDUKlU/CzLsoxAIMCi2na7jcHBQfYCyufzWF5exj/8wz/A5XLhpz/9KcbGxsR7QAF1wMViMaRSKfz2t7/FwsICSqUSYrEYKpUKGo3GK3XckQ6uVqthcXERmUyGD9A0U42sTPqBXC6HxcVFbG5uYnd39xlBPwUyKpUKPp8Ps7OzcLvd/O+tVgvLy8t49OgRVCoVbDYbnE4nG/56PB4uEfeKExv0KE+tNPNGKZqiUtbAwADGx8fx6aefYmxsjCcaK7MGjUaDXZpLpRI2NjZw//59/r7AE7Et1exJkGs2m7tOe/2CMtMzNjaG0dHRZz7nRRkZrVYLrVaLSqXCAZQsy3xiymQymJ+fh9lsht1uxyeffAKz2Sza11+DdruNZrOJer3e5Z0kOBiz2YwbN27gww8/BPDiZ/plspBms7krMOp0OtxwYTAYYDabsbm5iXw+zxu63W7H3NzcG2uXPyvQoXV9fR27u7u4e/cuvvvuO25yUV6vl80Uk1swAPZzc7lcrD+hDbtfgp5KpYLt7W1sbGxwx+5+VCoVNBoNHA4HLly4gNHRUc7MUAPM2toaH34pu0kHW7PZ3NM1/sQGPXSaonKKw+EAAA5+lCMnSqUSl71os97/8NNmQB+1Wo0fdrVazaaE1IlEfiZi03g9067DvpbSypIk8f0UE6lfHtJHVatVJJPJrpKtJEldGR+VSsUZUIPBIJ5t4Nj0TNRB6vF4UKlUUCwW4XA4uIRQr9fRarVQKBSwu7vLG0U/HroI6oCr1+tIpVLY3NxEMpnksSnKUgjdQ/p/kit4vV44nU50Oh3OBpFkguQTym5fMrolvze/349QKASLxcJyirNWglfKPaiDjcrjtH5Q5xU9wzabDYODgxgYGOjSPjUaDdhsNuj1er4/1ExxUoL5Exv0mM1mDA8Po1arYXp6GhcvXkQqlUI0GsXOzg6azSZ2d3eRy+Wg1Wqxt7cHp9MJ4EnaWkm73UalUkEmk0Emk+GRFMq29unpafzt3/4tfD4fRkdHRanlLaMUqpN5FWmuBEeHxLHRaBR3795FoVAAcLAbrVarhc/nw/DwMPx+f18aEvYSi8WCubk5TE9Pw+PxYGNjAzqdDplMBvF4HLIsY21tDf/+7/8Ol8uF2dlZjI+P9/pl9wzSPeVyOdy5cwe/+MUvUC6XObgHnupIlIEPrdt2ux0//vGPcePGDR6JQDMAt7a2UCqVkEwmEY1GOfAhreGXX36JBw8eYHJyEk6nEzqdjp3kz5rOs1arIRqNIp/PY2tri/U8ykYhSga43W5uYnE4HJiYmOjab+v1OjY3N+FyuTgRkUqlYDAY+Br3+mB7Yu+eTqeDzWZjPxHqMslkMlCpVOTW/psAACAASURBVF3tiNTVValUYLFYntk4KcqvVqv8UavVOHLXaDTweDw4f/48AoEAbDYbDAbDmYvoTxLKEwB59oiA5+WpVCpYX1/H6uoqYrFYl+hwv7+USqWC2WyGw+Hgk6vg+NBqtfD7/QCAdDoNr9fLjvGUlctms9jY2ECpVMLExETfeYQpoaxLrVbDzs4OlpaWuiQO+zdPpdcSCWjHx8fxwQcfcNdXLpdDIpFgKwEyRyUvOPrY2toC8GSdomBJpVKdyTVKlmXk83mk02nk83meN0e/K5WpaE7WuXPncO3aNe5wU2o66/U6XC4XjEYjZ+rK5TJqtVrXEOpecmKDHgCs73C73ZiamoLX62VHyHq9zor9fD6PhYUFFItFjI6OQqfTwWq18hugVCphc3MTi4uLSCaTqNVqrOEZGRmB3W7HzMxMVydSvy40gpOPUstAi8lBWbL9zzC1ZQ8NDcHn8/WlHcNJQavVwm63w+VyoVgsntkN9WUhXVqr1cLOzg62traQTqe5u00JGeQp/6xWq+F0OhEIBOD1euHz+WA2m2EwGOD3+7mDSKvVolwuw+12Q6vVolgsIpFIIB6Pd5Vhms0mMpkM9vb20Gq14HK5zpxzdr1eRzQa5SqKstxHXVrDw8OYm5uDx+Phch9NSj+MdrvNiQmNRoPFxUVIkoS1tTWeB0iQ4S1ZxbzNa3xigx56gCVJwujoKDQaDYrFIte5c7kcVlZWUCwWEYvF8C//8i8wmUx49913eXGnum4ul8M333yDW7duoV6vo1gswmQyYXJyEn/913+NoaEhDA0NYWBgQBi1CU48VK6t1WpdM2xkWe4anrg/6KFn/vr167BarX05euKkQC3ydC81Gs1bdaE9LVDLdL1ex/379/H5558jm81ieXn5UAsGZTZMpVJhZGQE7777LjweDyYnJ+F2u9l9maxQrl27hlarhaWlJQwODiKbzeKbb75BOp3uCnpqtRq2t7eh1+sxNjaGwcHBM/e+KRQKuHfvHh48eIDd3V0ubdH+aTQa8e677+JnP/sZzGYz3G43LBbLC5tOZFnmgceJRIIDn+3tbeRyua7PlSQJdrsdwWAQPp/vrV7jExv0EGSoRr4ZHo+HRc1UWyXDQbVajXA4zEMt6aaR/wK1jKrVauh0OlgsFoRCITYa61dfnl4iAsyXR1muVY4KodTx/pIW8LTVlMaGiGe9t6hUKuj1epFZ/hOUWSDbBXKv3t3dRSaTQbFYPLTUp/SdIlmE1+tlwa1Op+Prvf9nFgoFBAIBaLVa7rZT/oxWq8U6ILJMOWtQeSuZTHKXM6HMng0ODsJoNEKv1x9p7SCNVLFYhCRJ2NvbA/BEh6gM8Gn8il6vh8lkYluIt8WpWPXoQdbpdBgfH0en00EqleIUGel7ms0mIpEIvvjiC1itVr6YpHuo1Wrc6k5TXcnMkCJXwfFACxQNoROdckenVCrh22+/xerqKvb29vDo0SPE43Fks9lnfDXIa8RiscDn87EXjdhoe0uhUMCjR4+wurqKaDTa11meer3elRFYXl5GoVDAwsICNjc3udlhPyRR0Ov18Hq9CIfDsFgsOH/+PC5dugSr1cpZnoOQJAkmkwnBYBA6nQ4OhwM6nY67lmgOVTweh0qlgsvlOpL54WmDyuX7m0na7TZrzWi8R6PRYMPfw6Ass/K603Uko2G6jk6nk9el2dlZ9mx7m9PuT0XQYzAYoNfrIcsyLly4gGAwiGQyydOjqdRVrVaxtraGRCLBpTFJkiDLMsrlMgudaTjmyMgIQqEQBgYG+t6d9rghJ229Xs9ZOTFs8WgUCgV88cUX+M1vfoNqtYpEIsGGhPsXZZVKxc87OQ4bDAYR9PSYbDaL+fl53Lt3jwW7Z00rclRqtRq2traQSqWwvLyML774AslkEqlUCnt7e6xf259lMRgMCIfDcDgcOH/+PG7evAmXywW/38/ZmxfNTrRarQiHwxwg6fV6bmunZotIJIJisQiv1/vCAZunERreun9KAf1bq9Xi4aP1eh1Go/FI5Sdl4EO6IZVKxT9PkiR4PB5cvnwZLpcLN27cwKeffsr7wtviVAQ9SkU+mR3V63U4nU6u02o0Gk775/N5rjVSzZeM8ais5XQ6YbVaj5yqE7xZaEI7pTOVQarg+dDg1lwuxw7MtEgrBYhUc9fr9bBYLDCbzZzqF9f6+KHFnoxXqZOUUFr7GwyGM9ldRyVYpRi/WCzykNB0Os2TvYvFIpdu6VlWTrgnY1PaBzweD1wuV5cB3ouecdpTGo0GHwSU7w0KfMgLrtedR28D8pAyGAwsOm40GpwEoH2V5iTuzybvR6VSQavVdlm+0LOv/Jnk/E8+fDQB4W0Pnz5Vu71yvIFWq8XHH3+MqakpLC0tIRqNIpVK8agD5amARFJGoxHBYBA3b97EhQsX4PV6n/H0ERwPfr8fP/jBDxAIBDA3N8eGViLb9mYgPZtOp8Pg4CAuXryIQCDAk8RFVu34KZfLWFlZQTKZxMOHD58p2ahUKgwPD+PGjRs88+4s3SMaKkktzFtbW0gkEkilUvjuu++wt7eHTCaDaDTKejUaREzu+Hq9Hi6XCxaLBX6/H1evXuUsZjgc5s876vNtMBjgcrm6xpBQoAU8KctQAHZQ+fgsYLFYMD09DbVaje3tbdTrdeRyua7sz/r6Ov74xz/C6/XCZDLB4XBwdUR5nSkY9Xq90Gg0KBQKz3TZabVaDurD4TDm5ubg9XoRCoXY/FEEPX9CkiQudVG2ptVqwW6341e/+hWAZyNK4IkmyGw2c0nrvffew3vvvcebguD48Xq9uHHjBqampuD3+7mNVPBmoJOUXq9HIBDA9PQ03G436xYEx0+1WsXy8jJWVlawurr6TNuuRqNBKBTCtWvXzqTGkIKebDaLfD6Pe/fuYXl5GYlEArdv38bu7i67+AJPjQepg8jpdMJisWB0dBRutxuDg4P44IMPEAwGYbVa4fF4XrpsS07DNAGAfNvoe5CfD4BnRL5nBerqpCzLxsYGms0myuUyBz2RSARqtRqhUAjnz5/H+Pg4SxKUkE7K4/FApVJhZ2fnmZ+nUqk4gA0Gg7hw4QJ8Ph/cbvexDJ0+VUEPoSx3USrtKGlM0o+QnbbQ8fQO2pBNJhPfC8Gbg8oBVM6l9LEILI8fKmmVy2VkMhkkEgnkcjlujTYajTyXyOFwQKvVnqnSlrKclcvlEIlEUCgUEI/HeRA02S4QypKLTqfjphMSHtPwSrIwoSzxy26YpKdSGvKRiR6ArrLaWRxBATwdw1Sr1eByueD1erlTrlKpoNPpsD0GDQDP5XK8tigPUWQJQF56+0dPUOKC/JM8Hg9n545LZ3gqg56XhaJPMj4i4exZfIBPC2TO5nQ6edCo4M2h1Wo5DT02NobZ2VneXAXHhyzLvLFvbW3hq6++wnfffYdSqcTlrenpafzgBz+Ax+PB3NzcmQtMKbCoVCr4+uuv8etf/xqlUgl7e3sc8BxU6gsEAhgbG4PNZsPc3BwuXLjAHVu0UXo8HhiNxlceGZTP57G3t4d8Po9EIsH+V5RlosGaBoPhhV1LpxWTyYTx8XGEQiFeH9LpNO7evcv3J5lMolwuI5/PY35+HrIsw+VyYWZmhsc/AeAxHtvb28hms2w1AICDxuHhYfzkJz9BOBzG2NgYwuEwH35F0POGoFODyWRifxKxyfYWSlmbzea3LlzrN5QiQaPRyPO2zuKCfdJpt9s842l3dxcrKytYWFjgf5ckCcFgEB988AECgQDC4fCZyvIA3dmUlZUV3Lp1i930KbtzkNsyBewulwvXr1/H+++/D51OB71e/8YCQxrUSyU3pecV8GSjNplMbOZ5FvcNnU7H41HUajVKpRIymQySySQ0Gg0Pwi0UCpBlGRsbGzCbzRgYGMDg4CAsFgt/LxpamslkOGACnlZnqPX/2rVrmJmZYVfy41ybTu0qSKp/mr111Fqr2FyPB0oTNxoN7ixSTu2lrohqtQq1Wn0muyKOEzGl/mTSbrf55Lu7u8ubAM2GokyCw+GA3W7vmmN02qH3O4mTae4VzbeiZ5ZmIKrVapjNZhbbz8zM8EBLl8vFZb/XDTxIr9JsNhGNRrG6usrdY1TaotdGQ3pp2OZZy8LtR6/Xw+1283NpNBrRbDa7PpLJJDY3N1Gr1eDxeLqyOTShnjJmVN7SarVwuVwwm82s4TzO7I6SUxv01Ot1bG1tIR6PHzjLQ9Bb6vU6qtVq15gEpQcETVlPpVIAALvdfuZOuG8TZZCj/AAOnrAu6A31eh2Li4v44osvkE6nkUwmATwdQ0GlhfHxcXi93jOjb6MDjyzLWFxcxL/+678ilUpx15pywLCyhDQxMYEPPvgAHo8HIyMjmJqa4rLsy3RlPY9isYjV1VUWU3/xxRfI5XJsEqnUoVBpbWJiAhMTE2duBMV+aA5lvV7H6uoq/H4/d2E1m01UKhXcvXsXy8vL8Hq9iEQi8Hq9AJ42Ed27dw+JRKIr6DGZTLh48SIGBwdx/vx5hMPhrlFRx8mpDXoo00PW2WfRKfM0Q4seLXz7zcXI7bRWqz1jiCV4PuJanR5arRay2Sy2t7dRKBTYl4d8qmw2G2sNrVZrj1/tm4NM7WRZRiaTwerqKhKJBJLJJK8HBIlbyTh2cnKSS33j4+NvPLvSaDSQzWaRyWSws7ODtbU15HI5lMvlZ4S3NMcxHA5zd9hZhpp8ZFlm7RTZDEiSxJke4IkFg9lsRjab5a9vt9usjVLuyTqdjjvuaAJCr7KapyroabfbyOfzKBaLyGQyuHfvHlZWVrCzs8NCOPIAkCQJ9Xr9TDponnSoNTWVSiGVSvFirzxFUTcEpaxF2fH5tNtt5HI5FAoFRCKRro4XpSFhp9NhF+bx8XH4fD44nU5xfY+Ber3OFvv07OdyOSwuLiKRSKBcLrOGxePx4J133sHAwABmZmbeqgNtL6hUKohGoyiVSuzHQ6MmgKdTtanscfXqVQwMDGB4eBjDw8NwOp2w2WxvLOtFhyxZlpFIJLC4uIjd3V1sbGygVCrx4Qt4knkig8PBwUGEQiEEg0E4nc6+0cVJkoRQKIQPPvgAqVQK33//PQqFQldQSDMvlVUW2qOpbEmBUzAYxMzMDM6dO4dAIACj0diLXwvAKQt6Wq0WYrEYNjc3sbe3h88//xx3795FrVZDPp8H8NRsSqVSsfBKcLx0Oh3k83lsbm4ilUohkUjwbDTlBk028f2ykLwO9Oyvrq5id3eX6+akQVAGNRqNBuFwGDdu3IDb7UYwGBRBzzFQKpUQiURQKpVw9+5d3LlzB/l8HouLi9ja2uKNFwBCoRD+5m/+BhcvXoTJZOJhl2eFfD6Phw8fIpFI4P79+1hbW0M+n+eMLw2/NZvNmJiYwF/91V/xvCyfz8dO+W+q9EGVAZrD+Lvf/Y5fE5naUiaaDD2DwSDGx8cxMzOD6elptjvpB1QqFc6dOweLxYJ0Oo16vY6lpaWuoKdSqWBzc/OZe9RsNiHLMvR6PQYGBjA0NIRwOIyPP/4YV65c4aaiXnEqdptWq8XlElKGZzIZ3lAJpQ/Pi8be9zNKA7CDNCEvAynylbRaLVSrVZTLZbZvJzEz1eRpQRMzt44GDf3L5/PcZULvC8ru0D2lcgEJZCnzKXg+Sn+Wo0JlnE6ng3K5zF0uqVQKsVgMxWIRuVyOy1rUOUoDYIPB4Nv4VXoGrSP1eh35fB6ZTAaFQgG1Wo2z7uSPRiUtm83G18JoNMJisbyRgxCtc+12G/V6ndejQqHA+qpqtfpMuU0pqLbb7RycvQkR9WmBbF78fj/bi+h0OvZboud+fznwoO9B6xCNC6F/6xUnNuhRXtTNzU0sLy+jWCzi8ePHWF9f5weXaoWXLl2Cy+VCPp9HMplErVZDpVIRi/2foAWg1Wohl8vxxFxalGRZ7pp+e1TMZjMCgQBMJhN/fxIvPnz4EPl8np1WaRqy2WzG8PAwu3AKn57DofdAtVrF7u4uHj9+zGUTOp0e1O5LQY/NZuub0+nrUCqVsLm5ye67R6VYLGJzcxPFYpEPZPV6Hdvb29ja2kK9Xuf0v91ux5UrVzAwMIDLly+fOc8kWk8qlQrW1tawuLiIWCyGnZ0dnp9lMBg42Hn33XcxNTWFYDCIoaEhWCwWaLXa114LSEvYaDQQiUQQj8dRLpexubnJnXQ7OzuoVCpd5WFqh3c6nZidncU777wDt9sNn8/XNZerX9DpdKwzO3fuHD755BPk83msrq4iHo8f6euHh4dx/fp1eL1eOBwOAL3voD6xQQ9NjG42m1haWsI//dM/IZvNYm1tDdvb21wz1Ol0CIfD+MlPfoKpqSksLCzgt7/9LW8Mvb7AJwVyymw2m4jH41hfX0exWMTKygrPutlfnz0KPp8Pc3Nz8Hg87MTZaDTw6NEj3L9/H+VyGdlsloOeUCgEv9+PsbExBINBeL1e4Yz9HFqtFp9So9EoFhYW+AT9vNKtwWDgk9VZaoN+W5RKJTx48ABbW1sv9XXRaBS3bt1CLBbranVWtmVTBs7hcODmzZt499134fV6+dR7Vmg0Gtjd3UUymcTjx4/x4MEDblVvNBpdJ3+/34+bN2/yVG06DAGvvyk2Gg0Ui0UUi0V8//33uH//PnK5HB48eIDd3V3WXikdoJWZJ6/Xi3feeQd/+Zd/yQNgz7qA+SCU0wsuX76MTqeDvb09lEqlIwc9Y2Nj+PDDD2GxWOByuU7Efnxig55Go4FCoYB6vY5sNstqeyqXKMVmLpcLLpcLbreb/65SqfS9VmR/epeG+NEk42KxiHQ6jUwmg1qthlwuh0qlwmWn5z2glMYmUTkADnpoXg0JBGlzJh2Pct6NEDE/H2UXHE17Ji3PfrRaLeukzGYzz6kTZd5uarUau+5StoxGIlBnylGh989RMkQ0MNlisXD79VmCvLfK5TJ/VCoVyLLMWRKTycTrtdPphN1u5/XgResArTkHZTcpw0xC2mw2y2VGuj/UCEABaafT4TWIBo7SQcFms3FJq1/3EbofKpWKKy8HXfvnQbPMTpIh8Im6m8r6eCQSwe3bt5FOp3H//n0sLS2hWq2i0+nA5/PBbrfjgw8+wMTEBPx+P6anp+Hz+TA4OIiZmRl4PB5UKhVsbGz0+tfqGRQw1ut1RKNRrK2tcXZnZWUFtVoN2WyWWzWptk019sMyBJR9oI14fn4ewNMgS5ZlDlRp0waeBj00K6efauSvAulE9vb2uO2ZTMEOysh5vV5MTk7Cbrfj0qVLCIfDrEcQgeUTms0m7t69i/n5edRqNaTTaeTzeVQqFWxvbz8zDuFFKMdJ9DuNRoNbwLe3t5HJZLiTR6PRQKfTYW5uDjdv3oTL5cKFCxfYn+tFgTmJY2VZRj6fZ40UkcvlsLe3h2q1is3NTayvr6NarSISiSCVSvG9pnKWMvsWDAZhMpkwPT2N6elpOJ1OTExM9GVJSwldp3q9jsePH+Pzzz/nmWlH/fpyuYxcLod2uw232/2WX/HROJFBT6vVQjwex1dffYVYLIatrS1sbm6i3W6zDmRgYAA3btzAe++9x1NdDQYDgsEgRkZGYLVasb6+DrVa3eUP00+QH0W5XMbKygr++Mc/IpPJYGVlBcvLy6zfoUXJYDDw8DmHw9FlL65ElmU2HaQaL3XPEQcJo2kcCLWqiizPi6lUKkgmk0in04jH49jb2ztUd+V0OnHhwgV4PB4+DBgMhr5MzR9Gs9nE48eP8dlnnyGXy2FzcxPxeJzXnpdFeCY9hVr1I5EIdnd3kcvlUCwW+T1vMpkwMzODP//zP4fVamWR8Iuge0NDLClQVV77WCyGpaUlFAoFPHjwAHfv3uUWdco0K5s3gCfrkcViwdDQEOx2O65fv44PPviAXYP7PUNKQU+j0cDW1hbm5+fZeuQo0OG4WCxCrVafmE7qExX0NBoNJJNJVCoV7OzsIJvNdnU+UN3VZrNx+tFkMnUNEKULTWWVfg14gKfBCaV7M5lMV2YHeOI4SmUQssH3er0YGho6tK2Q5gkpFyAyriqXy4de806nwwsXTZ5WLkCCZ6lWq0ilUkin0yiXy8/dZEkb4ff7YbPZOKUsri247EKZGcockG7wOKBp1aVSiYX/Z42DOkHVajWXsMjDizK8+zdQ5ZTzZrPJk8/Jc428YfYHPYlEAvF4nLU8dF+VmirCYDDwWkft1DabDV6vl0uP/XpQoMGwrVaLO92y2SwSiQS/Z8gigzS1KpWK75XynpD8IZFI8D08CZyooCeTyeDf/u3fsL6+js3NTZ7y2mg02MguFAphenoagUAAoVAIbre7q+5aqVSwu7vb9cbo10W/UCjg8ePH2Nvbw6NHj/Dw4UMUCgUOesxmM65cuYLJyUlYrVZMTk7C6XRyO+1hXT/0AMuyjGg0Cr/fj1gshlgshsePH6NcLh/4da1WC/l8HgaDAdlslg3BVCpV39bNn0e73UYsFsNXX32FVCqFaDT63I0yEAjgxo0bGBoa4lZ1IRJ/svjGYjE8ePAA+Xwe6+vrr5TVeV1ow97Y2ECr1cLo6Oixv4ZeYDQa4Xa7YbFYUKlUsLy8DJ1Ox9le4GkWZr/B497eHusEaWMlbafyvUBBLQ28rNfrBwY8ABAMBnHz5k34/X4MDQ1henoaZrMZHo+HRyOcNbPIo5LJZLC4uIh8Po+lpSV8//33KBaLWFtbQ7VaRbvdZssRamk3Go1cXlQGPlQWa7VaCIfDGB4eRigU6nnJ8ETtNOVymVX/e3t7iMViKJVKfEqgbEQgEIDX64Xdbn8mG9FsNpHP57syRP1KvV5HMpnEzs4Ot46WSiX+d41Gg4GBAZw/fx5utxtXrlxhLY/Vaj00EKFTWKvVgsfjwe7uLrRaLZrNJtbW1g59PRQslUolVKtV/h70Pfs1OD0MpcljMplELpc7NNMjSRKsVivGxsb6ZjN9GSjYoYxnL7Is7XabrTYcDkdPAq9eoDQibDQaSCQSz6wtdD8qlQq3/ZMOURn0yLLMIyMOei8cZQ2x2WyYmZnB6OgowuEwJicnObvTrxkegrpEk8kkvvvuO/zmN7/hQzKt0ZRB1uv1cDgcsFqtaLfbSKVS3KVHpTHy0aPqwEnI7Pc86Ol0OlwnJPErdRaR6t/hcCAQCMBqtfJwPrK33g+VWCjaF5vp4ShHQVBrIultnpcdoBp7s9lEtVpFJpPhGWitVguSJMFut8PpdHZpqqgEWSgUEIvFMD8/j93dXQSDQYyNjUGv14vMBJ6cWsnQLR6Pc/dJrVbrEmEq6fdrdhCFQoFHITx69AgrKyvswEsdi8eZcqeWbspujI6OdmVFyV+JtHVGo/FUeSxRqz4dZpTZG+rEjUajXBo56OtJA9JoNLr2AfqghouDhuvSOk9rmVqt5kGl5Dmj0+kwMTGBkZERBAIBNt3r56YKyp5Vq1Wsr69jdXUVyWQSyWSSgx26X0ajESMjI3C73bDZbNwsEYlEoFKpUCqVWE7RbrdRrVY5CZHJZNhb700ZUL4KPQ96ZFnmRWhjYwOrq6tYXV3l6F6lUmF4eJgt9a9du4aLFy9Cr9cfKIKrVqtIJBLY3d1FoVBgF2DBs9D8G7LBt1qtsNlsB7osK6HFiQKe9fV1LC4uIpfLodlsQq1WY3R0FHNzc9Dr9ZxmLhQKePToEd+bSCQCk8mEH/3oR/i7v/s7uN1uXrD6GZrXlMvlsLCwgNXVVRQKhWcmQCs5LJXfz0SjUfzjP/4jtre3EYlE8PjxYz5g0cZ8nIOKaTzF0tISd5UqHZlVKhUCgQCXDAYGBk5Mx8tRINFrpVJBtVrtcqqORqNQqVTY3d3FvXv3njmIUvCi1PQotX/Kz1Nmh/d/DxInO51OmEwmnDt3DsPDw7DZbJiamoLX64XVamX3Zzro9brk0ksKhQK+//577OzsYHV1Fbdu3eLOO7J2oHKk3+/HX/zFX+C9995jwbfRaMTS0hK+/PJLpNNpPHz4EHfv3kW9XmfrEo1Gg9XVVbYqGB0d7d+gh7wdSIBWLBZRKBT4RKvRaGA2mzEwMACPx4NgMIhgMHjgpkwnDRqBICavvxjS05Au6igPovJER/eO2n4pyLTZbBgYGOgKeuhERYMyS6US1Go1zp07h3q9/kpjAM4ijUaDT0a0aCjLkvuhIFUE991QG/ry8jL29vaws7Nz5M6To3DUTZKeaSrNAIDJZGKncoLefwaDAbIs83ti/4Z8kjdnpV8O/W5KsfhBz/GrlqmojEL/TxkkrVYLs9nM2sTBwUG4XC7MzMwgFAqxbUa/d2cRjUYD6XQau7u7PNcvnU53BZx0bU0mE4aGhjAzMwOTyQSv1wuDwYBarYb19XVoNBpsbm7y9ybxOTUQFItF6PX6npZ2exb0UGt6qVTC8vIylpaWeHp0p9Nhd0yDwYCRkRFMTk5ylLj/DUElLer+oq6vSqXCbwra3EU3y6tDixcN7UskElhZWUEqleIAxu/3Q6fTYXx8HJcuXYLBYOBujlwuB5VKhdHRUe5iIRsCKqmJe/P0xExi8edhsVjYZ2RoaKjvs2RKyJslHo9z1vdVUalU8Pv9/KxarVYYDAYumxymBZFlGXt7e8+YHtrtdlQqlWc8T1KpFBYXF6HRaLp0iwMDA/wzSW9nMplgsVhOTLAryzJrlvL5/JGvtzJ4oT8fhlarhc1mY3NHOkiR4y+NugkGg7x3DAwMwGQywel08uf3+zqjHMAaiUSwvLzMw4wrlUrXrESNRoPx8XEMDw8jGAxicHAQdrsdWq2W9Z1Ukqcy5n6odGsymbjBolf0LOhpNptsnnfnzh3cunUL+XweiUQCsizDZDJheHgYdrsdFy9exNWrV9m9c/8DS+K4UqnEkSp9n1arxS3tJ80Z8jRB7bY0VuL+/ftYWVlhgXQul4PX6+VU8tzcHD766COYTCY+hZVKJYyOjrLvzOrqKsrlMsLhIxlX/gAAIABJREFUMOsY+n0xAp4sSGRAqDR2PAiHw4ErV64gGAxiZmZGjJxQUKvVEIvFsLGxwaaZr4pOp8Pk5CRmZ2dhNpsxNDTEuoahoSGeUbSfSqWCr776Cvfv3+/a2EulEhKJRFcwRP5k1OJLa5bP58Mnn3yCcDgMl8uFoaEhLi2YTKYTs57Jsox0Oo1oNNr17B4FCnxe9P7X6XTcyKLX62Gz2Xi8DU0FDwaDnNGhwFSlUnV1i/X7OiPLMrf5r6+vY35+HgsLC+y9RtISClYuXbqETz/9FC6XC+fOnYPH4+GMf6PRQLlc5kHgB4nMKUiniQm9zLL1JOihkz/V1Ol0oCxJ0SRim83Gyv/9zrJ0kqBSST6ff2aqN0GeAuSy2Q/QG13ZHkqW4srUM7kr04cSpf04DVVU+v4Ui0VeoMlDyeVysaqf7hmJpt1uN5diMpkMp6Ip4OmXe/M8yGuKJkDTFHUldJ1oHIvL5RKDW/dBGbOXLWnRs0plQ41GA6PRCKfTCY/HA4vFAr/fD4/HA7vdjsHBwUODnnK5zBki5Xqk0WjYjoOgU3O5XOb7Tvc+nU7DYrFArVaz+Hm/L8pJQKVS8XuZgrb9WRx6RqkERn+3/71P65Xy72njJEsGh8PB2R2/3w+r1crXW4y5eRa6F61Wi92S94/ooOeUmlyMRiMcDgcPDaUsG5UyqRRLBzXl4YLuIZUdtVotv7d6RU8zPXSRcrkckskkCwyBJ+nf6elp+P1+jtqVmyLVx6vVKmKxGP7whz8gFothZWWFN2LlzQsGgzh//jx8Pt+hC9RZg7IAQ0NDAICNjQ0Yjcau7jZyZvZ6vWg0GhgYGOCv73Q6yGQySCQSLEorFosol8tYX19HMpmEWq3G2NgYJicnMTo6iuvXr8PhcGBmZobn6dA902q18Pl8sFgscLvdcLvdaDQaCAaDMJvNb2TC8lkglUrh22+/xebmJmKx2DObNpl06nQ6+Hw+jI6O8gBXUd56fdxuN8bGxmA2mxEKhTAyMgKTyYRwOIxQKMStumazme00DkOr1WJ8fBwmk6lr869Wq88EPbIsIxKJYHt7G81mE9lsFqVSCTqdjieGVyoVaLVaPggODQ2dmDZrs9mMy5cvQ61WIx6P86BPygYAT9Ykp9PJjQ2VSgVGoxF+v7+rG1etVsPj8SAQCHB2nspZSvNNk8kEjUYDm80Gv98PvV7P5T+xlnRDndI0fuj27du4e/cuN/7UajXuGlSpVBgZGcHMzAzsdjvef/99TE5OslwhnU5z01ClUsHCwgKWlpaQTCaRSqXQarWg0Wjg8XjgcDi6BkzT3MVe0bOfLMsyBz2U6VEKWe12O6ampnih2Z+hoflO6XQay8vLuHXrFlZXV3lTVgql1Go1Bz12u71vgh673Y7Lly9zy+j8/DyfvMrlMptHxWIx+P1+AMDg4CB/fafTwdraGhYXF9k/gxYx0puEQiF8+umnGBwcxPT0ND7++GM+fe0fIqjVauHxePh7T05O8klWnMiekk6n8e233+Lx48cHdmWpVCqYTCYYjUZ4PB6Mjo5iamoKTqfzxGyApxmXy4XZ2Vn4fD5cvXqVy7TKzIMyA/G8zZUmTY+MjHT9/UHWAxT0UHCztbWFeDyOcrmMSCSCZDKJWq3GOqL92aNeQ0GPz+fD2toa8vk8dDodqtUqCoUCJEniMUGtVgvRaBSZTAZOpxOXLl2Cx+Pha6lSqTA1NYVLly6x6JgyNyaTiUtW9KG8JyJjfDgU8Ozu7uKbb77Bb37zG7YJqNfrXVrac+fO4cc//jFcLhcmJycxNjbGJUzSba2srKBQKODhw4dYXFxEOp3m6oBer4fP58Pw8DDrgZSZ/l7R0+6t/VNz91tYU4cQ6UiUmQMSAZKfADkNk2sk8HTqNJ2KjEYja0f6ASrpSZIEm82GQCAAtVrN5RPaUCuVCkqlEjKZTJcTaafTYWGaLMtsSEXXlFpsfT4fR/S0GR8WxCgf9n65D0eFgn4SMZMg8CALAb1ez880fRxlUnU/QbP4qBOF2m8Pgsqz5Pru8/k4nU+jCV6VowzUBJ7cf5p7p9fr2cyNzDy1Wi2cTiesVissFsuB+sZeQgEJlVsDgQAkSUKtVoPdbockSRgYGIDf7+dnnUpUtIYAT593l8vFOk4KeCjLSYdgkc05OsqB0JR0oFIqJQloz6BWfuX4EOq6TafTyGazyOfzvPcq9wkqD1Nw7na7uRx5Etb8ngY9FMAoI3QKfAqFAjY2NlAqlRCJRLCwsNBVFywWi7h37x62t7fZbZUEWNTm6fV62etibGyMBbP9IvZUBikXLlzAz372Mx7I9+DBA5TLZWxvbyMej6PZbOL27dtdi7skSbyI0SBS2gDC4TBbyw8MDMBisfAG0eua7Wmk2WzyaetFnUZ6vR6BQIBNHUOhEAKBAHemCJ4wMTGBv//7v0c6ncb29jYWFxcPdWmnYa3UMUU+Lm63+9iyZyqVCg6HAzqdDrIsIxgMsnM5GfZR+Uar1R7razsKJCh2u90IBAIYGhpiXVqz2WQPHbPZzDO46vU6dDodXC5X17pMBqcul4sPULRP9LOR4OtCWXrS4NBBgIIenU7H67rZbEar1eqaVl8oFPDNN9/wFHvKPlK5CwAfwmw2G65cuYL3338fLpcLHo/nROwLPQt6XpSGrNVq2Nvb6zqdqVQqGI1G6PV65HI5fP3111hdXWUB4H7PC3KMpBZGWiRO0kLxNqGIGwCGh4fh8/lQrVZhMpkgyzLrdarVKqrVKrLZbNfXq1QqTExM8CC+4eFhDA4OwuFwYHZ2FsPDw13pZXHyenVIWEgfzwt61Go1nE4ngsEgAoEA3G43nE7nMb7a08HAwACCwSDa7Tbu3r0Lk8mEYrF46Of+6Ec/wujoaM88cSgoUJquPk+ofBI2ECXUqAAAoVAIMzMzr/X9Ttrvdxag5iHSz9LUAkKn0/FAb/JYozlo9N8//OEPePDgARtRKvdeSZKg1Wr5sDw6OorZ2Vn+80mgZ0EPDXUjg6NwOIx6vd5lD5/JZFjYTCp/SreRdodKY7TZGo1GTv2OjIxgbGyM06T9fEJQphydTieGhoZgt9tRKpVgsVgO3GTVajUGBwcxMTHBDrEkRCYBIQDhr/MGaDabyGQy7DH1vPZqjUbDQY/L5RLi5edAgbjZbEYgEIDNZjvw87xeLws4TxKn+X11ml97vyLLMorFIjqdDhKJBDY2NqDT6biCQl1eZJxJ+69Op+Pyld/v58OY1+tlb56TkoXuSdBDZRMSo3344YcwGAzY29vDV199hWg0imKxiEePHj1zoSizIMsy8vk8a02opDI0NIS5uTk4HA5cvnwZc3NzMJvNrBrv1w2auhm0Wi0uXryIcDiMZrOJP/uzPzt0KjoAHlFBTrFULrNarcLz4g1C5dqNjQ0sLS2hUqkc+rlGoxHnzp3DzZs3YbfbD93IBU+QJAmhUAgmk+lQJ1i9Xg+Xy3XMr0wgOFmUSiVsbm5Co9EgEong7t27XTYnNBONtLP0flLO4bpy5QquXbsGm82G6elptg84KRWWnmV6yCjQYrFgcHAQlUoFBoMBDx48gEqlYpHVUaFMhs1mw8jICLxeL2ZmZnhOV79vzFQLV6vV8Hq98Hq9vX5JAgWNRgPxeBxbW1tIJBJdpoT7sw9arZaNICkQFTwfq9XaN12bAsGromygSKVSR/46g8EAt9vNJa3Lly/DYrGwNOIk0fPZW9TL32g0oNFocOHCBRiNRhSLRcTjcTQaDfZ5kCSJO1aUFuRmsxk+nw9msxljY2OYnp6Gw+Hg9jiB4KRjMpkwMTEBg8GAwcFB+P1+FgbuD9j9fj97xyhdZgUCgeB5GAwG2O121Ot1jI2N4fLly+y3Q9kb5RR7Qmk2qTRCJW+1cDiMS5cuwW63Y3x8nKfbn5TsjpKeBz16vR5jY2MYHBzEyMgINBoNYrEY1tfXcfv2bba1brVaUKlU3Bmg1+vhdDphNBoRCARw9epV+Hw+uFwuNhBTtjYKBCcZp9OJTz75hNs+ldYL+6HOHbINEIG9QCB4ESqViue3mUwm3LhxAw6HA/F4nGUlsiyjUqk8Uwam4IbGQoXDYZY5kBfV9evXYbPZeNgryU5OGj1/RaTvoXp7MBgE8ETjYDabUavV2FeGBsuRj4XL5YLVakUgEMDIyAgCgQD7ApzEiy0QHAYFMgKBQPC2oMywLMvweDzswG+z2WAymXiEkzJ7TB1Z5GXldDp5sLRy9tng4CBsNtuJ7+I9UZGBwWBAOByGzWaDw+GAy+XicQnVapUNqxwOB89sonRdMBiEzWbr+QRXgUAgEAhOKpIkQafTIRQK8X/tdjuPj2g0Gs9kmUkPajQaWTNLfyaJCmlnT3pl5UQFPSaTCWNjY6wK//jjj7tcm4Fud1Ol/Til3056lCkQCAQCQS+goMRgMGBsbAzDw8NotVr46KOP2CX7IG8oCmSUWR9lgEPGwSc94AFOWNBDESihHEAnEAgEAoHg9aHg5SQKjd82IiUiEAgEAoGgLxBBj0AgEAgEgr5ABD0CgUAgEAj6AhH0CAQCgUAg6AtE0CMQCAQCgaAvEEGPQCAQCASCvkA6qCf/0E+WpCSArbf3cs4kw51O57Wne4pr/8qI699bxPXvHeLa9xZx/XvLgdf/pYIegUAgEAgEgtOKKG8JBAKBQCDoC0TQIxAIBAKBoC849UGPJEn/gyRJC5IkPZQk6eeSJBl6/Zr6BUmSwpIk/bskSY/+dA/+u16/pn5CkqT/Q5KkhCRJD3v9WvoN8ez3HkmSfixJ0mNJklYlSfqfev16+onTvO+e6qBHkqQQgP8WwPVOp3MRgBrAz3r7qvoKGcD/2Ol0zgN4H8B/LUnS+R6/pn7i/wTw416/iD5FPPs9RJIkNYD/FcB/AuA8gP9cXP/j4bTvu6c66PkTGgBGSZI0AEwAdnr8evqGTqez2+l0vvvT/xcBLAII9fZV9Q+dTud3ADK9fh39iHj2e867AFY7nc56p9NpAPgHAP9Zj19TP3Fq991THfR0Op0YgP8ZwDaAXQD5Tqfz//X2VfUnkiSNAJgDcLu3r0QgOF7Es98TQgAiij9HIYLOY+G077unOuiRJMmJJ9H9KIABAGZJkv6L3r6q/kOSJAuAfwTw33c6nUKvX49AcFyIZ1/Qb5z2ffdUBz0AfgRgo9PpJDudThPAfwTwYY9fU18hSZIWTxb9/6vT6fzHXr8egeC4EM9+T4kBCCv+PPinvxO8fU71vnvag55tAO9LkmSSJEkC8EM8qa0LjoE/XfP/HcBip9P5X3r9egSC40I8+z3nWwCTkiSNSpKkwxMh7f/b49fUL5zqffdUBz2dTuc2gP8HwHcAHuDJ7/O/9fRF9Rc3APyXAP6DJEl3//Txn/b6RfULkiT9HMBXAKYlSYpKkvRf9fo19RHi2e8hnU5HBvDfAPgMTzbc/7vT6Sz09lX1B6d93xVjKAQCgUAgEPQFpzrTIxAIBAKBQHBURNAjEAgEAoGgLxBBj0AgEAgEgr5A8zKf7PF4OiMjI2/ppZxNNjc3kUqlpNf9PuLavxp37txJdTod7+t+H3H9Xw1x/XuHWHt6i3j2e8th1/+lgp6RkRHMz8+/uVfVB1y/fv2NfB9x7V8NSZK23sT3Edf/1RDXv3eItae3iGe/txx2/UV5SyAQCAQCQV8ggh6BQCAQCAR9gQh6BAKBQCAQ9AUi6BEIBAKBQNAXiKBHIBAIBAJBX/BS3VsCgeBk0el00G630el00Gg0UKvV0G63odVqodVqIUkStFotNBrxVhcIBAKxEgoEp5hGo4FsNot6vY61tTXcuXMH5XIZ4XAYw8PDMJvNGB4eRjAYxJOByAKBQNC/iKBHIDjFNJtN5HI5FItFPHz4EL/4xS+QzWYxOzuLd955B06nE1arFcFgsNcvVSAQCHrOmQt6lOn+Wq2GarWKVquFcrmMarV6pO+hVquhVquh0WjgcDhgsVigUqmgVqvFaVlwImi1Wmi32yiXy9jd3UUmk8He3h4/57IsQ5IkqFRCttdLqOwoyzLa7TYajQbfO/o7+v9Op8Pri1qtht1uh8lkgkqlgkajEfdSIHgDnLmgR5ZlXvTX19exurqKYrGIhYUFbGxsoN1uP/frJUmC2WyGyWSCw+HAj370I8zOzkKn08FqtUKn0x3TbyIQHAwF8fV6Hevr6/j1r3+N1dVV7O7uIhqNotVqQZZlGI1GGI1G8cz2gE6ng06ng2aziXg8jkwmg1qthkQigVKphFqthmw2i0ajgWKxiEKhgFarBUmSoFarYbVa8eGHH+LChQswGAxwOp0wGAy9/rUEglPPmQt6Wq0WGo0G6vU6EokEVlZWkE6n8fvf/x537959YdADAA6HAw6HA4FAAJOTk5icnES73YbJZDqG30AgeD6UPahUKkin03j06BEWFhZQLpdRLBYhSRKLmXU6ncgQ9IhOp4NWq4VCoYBkMolyuYytrS1ks1kUi0XE43FUKhXkcjkkk0nIsgy1Wg2VSgWPx4OBgQGEw2G0Wi3YbLZe/zoCwZngTAQ9nU6HU8WJRAIbGxsoFotYWlrC2toaCoUCisUiOp3Okb5fq9VCrVZDPp/H2toaXC4XnE4nZmZm4Ha7oVarodPpRKnriNCpt16vIx6Po1QqoV6vo1AooNlswmq1wul0QqfTweFwwOl0imt7CFS23drawu7uLlZWVpDL5VCtVmEwGDAyMsL/DQaDsNls0Ov1nEVQqVTi2r5haF2p1+uoVCqQZRnZbBbpdBq1Wg2RSATxeByyLPOz32634XA4YLPZYDKZoNPpurI+lUoFxWIR2WwW7XYbHo+nx7/l60EBYDKZRDKZRLPZ5GwlAJYOmM1meDweaLVaPry2220u71E3olqtPvDnSJIEg8EAo9EIlUolOhcFz3AmngZZllEul9FoNPDo0SP86le/QiKRQCQSQTQaRaPRQKlUOvL3q1araDQaqFar+Pzzz/Hw4UOMjIzgpz/9KWZmZmA0GuFwOMSb6Yi02220Wi2k02n87ne/w8rKCpLJJJaWllAoFDA1NYWrV6/C4XBgbm4Oc3Nz4toeAOk/crkcvvzyS8zPzyOZTGJ9fR25XA6Tk5N4//334XK58M477+DKlSswGo3QaDSo1+tQqVTQ6XSHbhiCl4c0hHRftre3US6Xce/ePe6kSyaTyGaz0Ol0CIVCsNvtcDgcGBkZgdlsRqlUQj6fR7VaxYMHDxCNRtFut7Gzs4PNzU14PJ5TLUSn93+9XsedO3fw5ZdfolAoYHNzE3t7e1Cr1TAYDFCr1RgbG8P7778Ph8OBbDaLZDKJVqsFg8EAnU4HrVYLm80Go9F44M9SqVTw+/0IhULQ6XSw2+2wWCzH/BsLTjJnYmchgWC9Xkc2m8Xm5iZ2dnaQTCaxt7cHWZb5c49yym21Wmi1Wuh0OiwOVavVyOfzqNVq0Gg0RyqTCZ5uCpQ9i8fj2NzcRCwWw71795DL5dBut+Hz+VCr1VAsFsW1PQQ6LTcaDb6O+XyeA36tVgu/3w+/3w+fzweHwwGtVotms4lms8kCfSp3iYzP60P3pN1uo1aroVAooFAoIBaLYWVlBaVSCel0Gvl8HjabDVarlbU5FosFDocDer0eWq2Ws3WyLHP5slQqwWKxoNVq9fg3fT0oG5/NZrGxsYFsNoulpSVEIhFoNBoOzmVZxvDwMBqNBpLJJHZ2djjoMRgM0Ov1qFQqh0oNNBoN9Ho9HA4HSxJoPVE+7/347FNGkv57Uq4HVQKUSJL01l7TqQ16SCQoyzJSqRQePnzI/93Z2UEmk0G5XH6txaLdbnPHVy6XQyqV4jS1w+F4U7/KmYSC0Eaj0RWAPnr0CGtra8jlcmg2m5AkiUtdlJEQPEur1UI8HkcqlUIsFkM0GkU8Hkez2eQNIxAIYHR0FD6fD4VCAb/73e/QarVQrVZRr9eh1Wrhcrk4UxkOh2E2m6FSqYTu5yVot9uoVCpoNBpcAs/n80in04hGo1waJz2ORqOBWq2GyWTC6OgoXC4XbDYbBgcHYTQa2VSyUqnw+8RoNGJoaAiDg4NwOp3Q6/W9/rVfGQoMZVlGLpdDJBJBoVBAuVwG8OR6NptNlifcvXsXFosFxWKRD0WkT1Or1TCbzXw9lJsllW/9fj9WV1eh0+lgs9lgNps5sNJqtTCbzfD5fBxsku6NDD3PGvl8HqlUCvV6HXt7e9jZ2YFarYbX6+Xy6tDQUE90Y5FIBI8fP0atVoMsy5BlGQaDAVNTUwiFQiwleZPr06kNemjhIX3DZ599hpWVFcTjcayvr6NarXLr+qtCXTKVSgV2ux2xWAx2ux2dTgehUOjQFGu/0+l0UK1W2T/mm2++wcOHD5FOp/Htt98iGo3yaRZ4ooVIp9PodDool8uvdc/OKrIsY3t7Gw8fPsTe3h6Wl5exvb0Ng8EAl8vFOp5Lly7B7XbjD3/4A27duoVyuYx8Po9SqQSTyYTh4WG4XC6Mj4/jhz/8IWseyL1Z8GJo8y4UClhfX8cvf/lLbG5usgYHAGZmZnDhwgVYLBaMjo4iGAzCaDTC5/PBYrFArVZDq9VCpVJxNpQCn1wuB4PBgOnpaUxMTMBkMp3qtYYyvY1Gg5tLSqUSH0hJmA882QQTiQRUKhV/HfD05K/8oK8l6O9tNhvLD4xGI/R6PQwGAzweDywWCwYGBnD16lW4XC6YzWbYbDZoNBpYLBZoNJoz9T7odDpIpVK4f/8+stksvvrqK3z99dfQaDSYnZ3F6OgoQqEQLBbLsQc97XYbS0tL+PnPf450Oo1KpYJqtQq3242f/exnsFgs0Ov1XJZ/U5zKoIcWiWaziVqthnK5jEwmg1QqhVwuh1qtxm8igk4B9LFf16AMkOh7A+A3XaPR4OxFs9kUG/MBUJqSFnASZabTacTjcQ6C6IQHoGvxet0g9SxDnjzZbBbZbBaVSgX1eh06nQ4GgwFWqxUWiwUmkwlms5k1JsViEalUCoVCAWazmcsndrsdhUIBVqsVer2+6/1xlhb9NwmVsUgjmM/nkcvleO2hjA356tjtdthsNni9XgSDQRgMBrjd7gNLM51Oh7MYtNgrdSynPRNHawNlfEhCsD94qdfrR872HrRWSJIEWZa5nKvX66HT6VjMXy6XodVqkclkOLCibM9ZtQQggXw+n0cymUQ0GoVer0coFILD4YDVauX97jggXyrqbEwkEkgkEqhUKiiXy2g2mygUCqjX69yJ+iY5dUFPo9FAs9lEpVLB+vo6EokEVldXsb29jd3dXVQqla7TAQU4VquVF5SJiQm43e6u71utVjllvbGxgUgk0vXmFLyYarXKuqcHDx7g+++/R6FQwMrKCiKRCKrVapegXOlJEgqF4Ha7YbfbxaargBbwUqmEjY0NzM/PI5vNIpPJAADsdjtmZ2cRCoVw8eJFOJ1OGI1G2O12+Hw+aLVaPgzQgkKlsXw+D6fTidHRUVy6dAlmsxl2u13cAwUUjMuyzOXFXC6H77//HltbW2g0GtBoNBgeHobD4cDAwAAMBgOGh4e5k46ea8qo7YeynrVajUuQarUatVoNe3t7HNDSpnza7o0kSdBoNNDpdPD7/ZienkY+n+eD0JuGglL6uWQ0Wy6XodPpsLOzg52dHS43nj9/HlarFWNjY7BarWdO6F8ul7Gzs4NEIoFsNssBTqFQQCaTgcPh6NK9vm0ymQwWFxeRy+Vw584dbG9vI5fLcVJBq9UiGo1idXUVDoeD9VxvilMV9Cj9SfL5PNbX17G5uYmtrS1sb29jb2+vK1tAAY9Wq4XT6YTb7UYgEMAPf/hDTE1NdZ0ycrkcG4dpNBokEglIksSnE8Hz6XQ6qFQqiMfjXNL65S9/iUKhwGJb5bWkgEelUsFqtWJgYABer1dsuPuQZRm1Wg2lUgnr6+uYn5/n5x8ArFYrZmdnMTMzwyc36lrxeDx8LQuFAtrtNtLpNCRJwtraGhYWFmAwGHDjxg3enAHAZrOJe/AnKOCp1+uIxWJ4/PgxEokEPvvsMywsLMDj8WB2dhZDQ0OYmZnBRx99BIfDAbPZzHoppTP2QdeVDFWpScLpdAIAC/+bzSaCwWCXY/NpgjJfOp0OPp8Pk5OTyGazqNVqbyXoIS2h8ucD4DVdkiTcuXMHGo0Gc3NzkGUZHo8HNpsNw8PDZz7oqdfraLVaKBaLyGQycLvdx5rpSafT+OabbxCJRLC4uIhIJML623a7DZVKhWg0irW1Ne7Eo/fEm+BUBT3tdhulUolPW6lUCslkkkWxSkdT8msgxb/P54PP54Pf74fb7X5GiCxJElqtFvR6PSwWC3Q6HS94wFN/FKXPRj+jTFfXajW0Wi0ehUAlrWKxyNmzg95UtBjq9XouBRgMhlO5sL9pKHgvFotIJpPIZDLsxyPLMnQ6HTQaDWw2G2w2G48soGffaDTC5XIBALxeL/x+PwdQ9EyTNQOZ47XbbfZMok2q360DqNWaxMmpVArZbJYFyiQK93g8POeMsjJH1UnRe4g0PeVymdeXZrMJjUbTVVI/je8PCvzoenU6Hc4qHgZtgvT1BI3soGaWg8pcyr877NCqlCuQz9tZQyk5oI5k5b/t/7vjgJoryKeJSp6UkaNyJL2H3nRp91StaI1GA/Pz8/j666+Rz+exurqKvb29rpOvVqvlixUOhzE6OgqbzYbZ2VlMTk7CbDZjYGCg681GGSQKamKxGBYXFzn9R6Zja2trqFarUKvVmJub69Vl6DlkNEjDLskgb319HXfu3EEul0MsFkMmk+G5Q/vRarUwmUzQ6/UYHh7GtWvX4Pf7ObXfz1Bmodls4vbt2/jss8+QTqexuLiIYrEIjUaDkZERuFwuTE9PY2ZmBpOTk9ydolKpMDo6CqPRiEqlgqmpKUSjUeTzeSwtLSEej6NQKPB7Z3FxkV1/ySfJYrFwVNiTAAAgAElEQVRgeHgYPp+v15ejZ3Q6Hb5OlL38/e9/z8aCH330EQYGBnDz5k3OsrlcLuj1+pea01csFhGJRFAsFnH//n3cvn0brVaLg6eRkRGMjY3B7/dz5ug0BT60ael0OgwODuLq1asoFAqcWTls0y2VSmxhQYJ7WZa5hE6Zolqtdpy/juA1kWWZNUblcpnnBJIA3efz4eLFi7h+/ToLzd8kpyroaTabWFtbw61bt1AsFhGNRpHL5breNCqVirM7pHNwuVz46KOPMDs7+0KhZrlc5gUMAGtQ6vU6dnZ2UK/XMTAw8IxQup8goTf5Iq2srGB3dxcLCwu4desWMpnMC08PdJ+oo2VychKBQOCYfoOTDYlla7UalpeX8c///M9Ip9N8MqWFYWxsDOPj4wiHwxgYGOj6HoFAgLM7Q0NDSKfTSCaTbAS3t7eH3d1dHo2Qy+XYxdZiscDpdMLlcvV10AMAlUoFqVQKmUwGS0tLuHPnDqxWK27cuIGZmRkMDQ3h6tWrGBwcfKVAhDodycBwY2MDjx49QqvVgslkgsFg4FIEZSJO46GAsroejwfj4+PsfXaY9Uen00E2m0UikUC73YZer4der2ePKtowX8Z0VnAyoEzP/qqJ2WyG1+tFIBDAyMgIpqamunzF3hSnKuihdCZNkj4oNUeiWKvVypuBclL6yy5Myo6uRqPBKbl+7DJSdrDRwhOPx9lm//9n701640qz9L/nxjzPI8kIkuJMihJTc2ZW/luZ3Y1qV3W1F90LG/BHMGx4Y9hL773wxhvDMGDAgA2ju9GoMirRnYnKoTKlTGVJoiRKnIfgEPM8R9yIuF4oz8lLah7IICPuDxA0cQjeuPd9z3vOc56TzWY57W8ymThCp5tb3rlBIyeoHHAWF/LjolwuIxKJcJah0WjwachgMLBV/+DgILxe7ws9XKikoNfrYTab0Ww2EQqFAABGoxH5fB4Wi4Xfk2azyd0Uco+qfoPS/u12m12W6XBF3ibBYBADAwPweDzc/fYmUDOGKIqIxWJsNFmv11lT5fP5YLfbEQ6HYTabD2mDziKCIECv18Nms7GR5ot+Hiq1ms1mSJIEnU4HnU6HRqMBg8GAQqGAQqHAppzUyfs6+ksSOKtUKn6WvF4vTCbTmcqgvQp5Jy1dr3q93rUyHpXa5MJ9ueyBuu0oU3pcWc0zF/TQRkDpzaOEw2H85je/gdfrxdjYGCYmJtih82UXT173pGBKHtiIoohsNss+Giepdj8N0BiJTCaDXC6HW7duYWNjA/l8HhsbG7wASZIEk8mEubk5XLhwAZIkYXV1Ffv7+2g0GmwaabPZMDMzA7/fj+Hh4Z40BXtbdnZ28M///M84ODjA2toaT+C2WCywWCwIBoO4du0arl+/DqvV+lKjTJVKxQZkLpcLLpcLlUoFu7u7GBoa4u7H+/fvcwlXkiQEAgFMT0+fWfHsu0ABSbPZxOPHj/GHP/yBnZE//PBDuFwufPrppzwB/WW6lBdBXY2FQgF3797FF198gXK5DIfDgcnJSVgsFly9ehUTExOw2WwIh8OsETqr74dKpeLuwna7jVAo9NL29Fqthlqtxu38arWaBeWUBVpcXEQqlUIymcTW1tZrBT1HS+tXrlxhrWevHL4ouCDz3rW1NcTjcRQKha4EPZSwIH89st4g/Rpl/h0OB+x2OwwGw7GZpp65oIe6t+SCTDlWqxXj4+MYGhpCKBTCyMjIa4sxKdB5ni02nXwpcu5F0dvLkBsOJpNJrK2t4eHDh6yBKpfL0Ol0MBqN0Ol08Hq9PJ2e/ExolAcAFpcHg0E4HI6eWWzeB4VCAU+ePMHW1hY7qZKuwWQywW63Y2BgAOfOnWPR34ug0zVle5xOJyRJgtVqRTabhc1m4ywGlSuj0SiAp6WdfkTuAUb3erVaxYULFzA0NMSlxdHR0bf+Ho1GA+l0Gul0Gjs7O1hZWUGlUsH58+cxNjYGj8eD6elpXLx4EVqtlkuPZ503aT+WG5jSqb9er8NsNiObzcJisSCVSkGlUnHn2+sgCAI/N2Qz4Pf73ypjd1qRz4Sr1WrIZrNIpVJdO6zLrR/o2aKDBe21pMc1GAzHahJ5poKe4ySbzSIajaJQKGB3dxe5XA6lUqmvtTsAOA1JJa29vT2kUimk02kUCgWIosimdx6Ph4co0smUPr9UKrHZFHXI+Xw+Lj/2e9DTbre53JHL5XiGE10znU6HcDiMyclJnq2l1+s5Tf+60EJiMpkQCoVgsViQyWQQCoVQLBZhMBgOvefNZpNP2b2yIbyKTqfDJnlU0m2323A4HDxGwmw2v/bXo/Jho9FgW4xYLIZHjx5xowRl1WZnZzE7O8sdYTR6oV+u/fOgA5coiiiVSlhbW8Pe3h7S6TS2t7c5+/yqDZ3Kg1arFeFwGFarFYFAgDfZXggqjyIvKckdrskfzefzwel0HnumnYIv2gvoF8keNBoNzGYz3G43O8wfF0rQg6c3RiQSwVdffYV0Oo07d+5gd3eX35B+plwuI5VKoVqtYmVlBU+ePEEmk8HW1hb29/dhNpsRCARgNptx8eJF/PrXv4bT6UQ+n0cul2OjsEQiwbV5m83GJmVjY2Pwer19X96ioJI6eWKxGBKJBACwuPjy5cv4zW9+A7vdjvHxcVit1rfSeQiCAJfLhUuXLqHZbEIQBMTjcWQyGbaBMBqNrJ2jE1i/BKatVouHfVYqFdTrdUiShFAoxCXFo+amL4KsLur1OtLpNL7//ntsb28jkUhgaWkJhUIB09PT+Nu//VvYbDZMTU1xSd5qtcJoNJ55Lc+70ul0UCwWkc/nkUgk8Kc//QmLi4uoVCqIx+NsvfCyUpkgCFwedLvduHz5Mpdw5a7kvYTcVoQaTwiNRgOfz4fx8XEMDg6+cIDr+4JGtxQKBSSTSaTTaWQyGbYfoNInGQfT2nYc9H3QQzdFuVxGPB7nLgo6WciR2/T32gNyFPKJaTQa3CWRz+eRyWSQz+dRrVb5ZEWiZZ/Px63U1BEkLxVQ3VY+OsFut/PC3s9QpodOQHQqks/GcjqdGBoaei/ib61Wy3PkXC4XnE4nRFFEoVBgXRtpAmg+VL8g10PI9X0mk4nHSLzuLKCj43JSqRQODg5Yh1IulyFJEvx+PzweD4LBIAKBADsJ9/NzcdQLjNqck8kkawTJbI84er3k41Xkg1/dbjd8Ph/P3eq1oFI+Uf15BrvyjLvRaDz2Aw3tJbSuHTWQlDdpmM3mYz0E93XQQ4FOtVrF2toaNjc3kcvleLOWYzabMTMzA6/Xi8nJyZ6c00IPSD6fRzQaRbVaxfb2NusN9vf3EYvF0Ol0EAqFEAqF4HK5MD8/D5fLhZGREbjdbu6uWFlZQTab5RZ2all1OBwIBoNs6NYreoW3gYLLQqGABw8eYGdnB8vLy6hWq9zBEwqFeFyE2+1m3dT7wm63Y3R0FFarlQ0QS6USyuUyqtUqtwz3m1EhbQzUBEG+OVRyehlUFqtWq3jw4AGLlsnRWavV4tq1a9BqtZibm8Pk5CTsdjvcbndPbsJvA3USVqtVPHz4EMvLy8jlctjY2ECpVHqhoSAZ1Go0GgwNDcHj8fB0dZ1Oh+HhYSwsLMDv98Pv9/dllpmCDKvVyqamxwlZDRwcHCAWiz2TlSMrDYfDAZvN9l7Xt6P01yp2hGKxiOXlZaRSKTx48ADLy8vs/3D0YSLjtomJCZw7d+6NavpnAbk7ZyqVwt27d5FOp/HgwQP88MMPfCIlY7aPP/6YvXWuXr2KQCAArVYLvV7PnW6Li4vIZrPcAq3VahEIBBAMBjE8PMyn235O31OdPZPJ4Pbt27h37x7ryQRBwMDAAG7cuAG3242pqSn4/f73uikKggCn04mpqSl4vV5sbGzwjCLK8HU6HVgslvfy/c4aJpMJXq+XzdMoy/OqTtBGo4FSqYRMJoNvv/0WX3zxBXevdDodjI2N4ebNmxgYGMDg4CDGx8e5hNgvZcSXQYevzc1NZLNZ/Pu//zu+/vprdsemgOd5QQ9dQ7PZjOnpaVy4cAF6vR5Op5PL8RcvXmTn8X4Meuj62O12ni5/nDQaDezt7WF1dZWlIwRl4sxmM1wuFxwOx0ubM96Vngt6SIRI5Sl5eo8eEhpZkc1muQ1b7vL5PGtzSovSwtcrmzRlGsgwimqvpO+ga9RoNLgLiMZGeDweFp45HA6+viTapInqnU6Hx01YLBb2TXqdE3MvI+9ooEnINBiU7j9arEk8Syn69wn5Y9AIECptkaBZo9H0XbciQaVsCsxfVtqm97LT6aBSqSCfz7O2jWZMGY1GGI1GHvdBw0gpg6TwC+Qd1Wg0UC6XDw3LBA6PmiDvHRrBQtfY7XZz9tnpdMJkMnHwYzQau/FjnRoo2DiJMirtL1S6p3KbSqU6VMLX6/XHvi/0XNCTTqdx9+5dRCIRAMDQ0BA0Gg2b5xWLRU4x53I57OzsoFQqIZlMIp/PPyNeJlt/ahV2uVw9Y6bXbrdRLBbZN2FpaQnJZBLRaBQPHz7k9HKtVoNGo8H8/DxP8r506RJ3atlsNqhUKr6O5XIZe3t7XDqUZ3iuXr2K2dlZBAKBvs0eEJ1Oh8tI1BGXSqUgiiJEUYRKpYLX68X8/Dy8Xi98Pt+JBdutVgvRaBQrKyvweDx8Iuw3aNSBIAgol8ssan7eTKBsNsvDE7e3t7G+vo5SqYSNjQ0AgNPpxIcffshZTurSMpvNPbGedBOyYzAYDDh//jwuXboEi8WCkZERBINBnpNGXUK9KE84zTQaDezu7uLJkydIp9PssUeBKXlRDQ0NwWKxHKuwuueCnkwmg0ePHsFms2FwcJDdbKPRKCKRCOLxOL744gusrq6i0Whw2/XzBrJRbfio8LZXMj206dJi/dVXX2F9fR3ZbBY7OzvsS0RzmSYnJ/HrX/+au4f8fv+h0lStVkM0GmWvFxqdEAgE4HK5MDg4iIsXL+Lq1avsG9PP0ABd0j2RYRfdg2q1Gm63G9PT0/D5fCfqGNtqtZBIJLC5uYl6vY7JyckT+b6nCfIFKxaLAMC+InQ6ProGFAoFrK2tIZvN4v79+/jpp594cRcEAXa7HdevX8f169c580DZnX4WLL8PDAYDZ81+9atf4R//8R9htVo5gybP0PVDI8ppo9ls4uDgAOvr66jValzeMhqNCAaDcDqdGBgYQCAQOPbGlp4LetrtNs91SafT2N/fh0qlwsHBAaLRKE9lr1arPNTxqBcPbeQajYanf3s8Hi5tkVvkWYUCmVqthkwmw+KyXC53aAhcp9OB0Wjk2i9F5NTiKT+dkpdGMpnkciFlzMi11mazwWg08udSmVFuBkmpzl5flKhDqFQqIZ1Oc5aR2jflk4Zp2vlx33NHJzDLxzH0U/cWAA5sgF+eFyqx0CnUaDSyd0yz2UQ0GkUsFuP1RaPRwGAwcIcMeVLJn4Fev89PCiqtt1otLi22223Y7XZeT45jjtNphWQezWbzmS5k+n8S2pvN5pdas8iNDoFfgkb5ui0v/9LH0ufJfZZoz6XP0+l0MJlMMJvN0Ov1bzUq6k3puaCnWq0iHo8jl8vhm2++QTweR6fTwcHBAZLJJOr1OttxkzX2UbRaLbRaLaxWK27cuIGpqSkEg0HMz88jHA5Dp9OdWfGbJEmoVCqc4fnyyy/xl7/8BcViEVtbW4fq5nq9HnNzc7hx4wacTieuXLmCc+fOHcrSyB+InZ0dfP755zg4OMDu7i5EUYROp8PAwADm5+e5Y8tgMPD7QFPYyS/G5/PB6/W+Vrr/JB6Q44ACiVKphEePHmFxcZEzZMDTlG8wGITFYoHf73/tjqF34UVBTb8FO8AvGidq16fFemtrC7du3YLdbuc1IRaL4fvvv0csFsPBwQFWVlZQq9XgdDoxMTEBs9mM8+fP49y5c7BarRgbG4PNZlMEy++ZRqOBVCqFUqmEO3fuoFarweFwYGFhAVNTU+wPRrq1Xg9+6vU6EokEZ5KPavLa7TYSiQS2t7fRaDTYPV+eBZNrYcmzi/yONBoN+/9QdydVQGq1GqrVKh+sG40GIpEIcrkci9ApyLJarRgeHuZO3pN4X3ou6Gk2m8jn89BoNJxSkySJNTuvs4iTuMpisWB8fBxXr16Fy+VCKBSCx+M5gZ/i+DjaWfL48WP8+c9/RqPR4OyXRqPhwI4M2cht2efzPbNg00mYur12dnY4ojcajXC73QiFQvB6vbBardBqtTzDrFKpoNlsolarsY+Py+V64euXp6jP6lwoOpHWajXs7u7i0aNH3LEFPM2Meb1e2O12OByOY3fl7cfA5mXQ809lEVrc4/E4VldX4fF4eJFOJpO4d+8eVldXkUqlsL29DVEUcenSJczOzsLj8eDq1au4dOmS4r1zjNAg6nq9zh2ITqcTDocDfr8fRqMRBoOBtVi9XuIiz61CocAdmHLo0JVMJmEwGHjUjfwgKW9MqdfrKBaLvEZrtVr+91arxd46arWaPZUoC1qtVpFOp9nbTZ7pMRqNJz7w9UwEPSTspA3yZYu03HZbFEUeWPeqFL1Go2EFudvthtfrhcvl4knKx+0dcBLQZpvL5VhoTJomiuiPUqvVWGdCgaTcpFHe9RWJRDjCp2tNwdD29jZyuRy0Wi3P6iIPDnmmh8aBvOgUTGlqtVoNr9eLYDDID9tZOTnLu1JKpRJyuRzK5TKffkwmE4aGhuD1euH1erm0dVwLQi8v/m8DTabvdDpwu90Ih8Ns2XBwcMDjOjKZDI9DqNVqnNXRaDSYnZ3FuXPnuKyrBDtvDnnrNJtN+Hw+DA0NodFocPlQjtyAj2YEdjodbGxsQKfTwWKxHPK5kuupehF5ifx5Pyd16R4cHAAAe4PJ13bKyLRaLTalVavV7BNGBzdRFLkbUaPRoFAo8FgQmpNJ1RdyYKb9odVqcdf0Sc0FO9VBD5VO5GJPWnxeFMDQ5k3dRCTGfdUFpQfBaDTi4sWLuHjxIhwOBy5fvoyxsTEe9nhWoZJKvV7H+vo6vvvuO2QyGWxvb6NUKh2aywL8EuWn02ksLy/DZDJhZWXlGbv2er2ObDaLer2Ora0tpFIp1Ot1/lqNRgNLS0vY3t6GTqfD559/Dr1ezxs/BUhy/c/LLOFpqKler8fNmzfxu9/9jsXlZ+H9kWfaKPhcXV3lhx8AAoEAPv30UwwPDyMUCrF52HFumi/62v24Uet0OtjtdjYk/eu//mvkcjksLy/jm2++gSAI+Oabb2AwGPh9kyQJCwsL+PWvf83Bqs/n47JKP+jU3jcWiwUDAwPskaZWq5HNZnHv3j3uziVoMxUEgbMcGo0GBwcH+OKLL+D1evGrX/0K586dQzAYxKVLl3o66NFqtRxsP6/xptFoYH19Hfv7+3A4HNjb24PL5Tp0eKQMZ6vV4tERgiCwppMSEaIowm638wGNhpvSntJqtXjALq1xBO0fgiBwguK4ObVBj1wE1Ww2UalU+AK/6sLIBZivMzCUPB6MRiPrKMbGxmC32xEIBF57zs5ph27AbDaL/f19Fi6/6BrR3CAKNp9HrVZDIpFgUTRlfYhWq8XTpN8HNJPIYDBgZGQElUqFg6CzwtFMz9Gyq9lsxtDQEEZGRuBwOLq2Ycq/Zz9t2HRK1mg0XNY2Go1YWlpCNBo9pAMkx2ay0J+bm+O2W4vF0vPakeOChuxaLBbW+g0NDcFgMGB1dfWQ7oR+b7VanKGgzTWTyQAA/H4/BgcHWR/X64Ok6R5ut9s8c0wOucAXCgWUSiVIksTWI3TPyoOeRCKBZDLJA1upJEYdjVRG1Gg0SKfTSCaTrxXAkOD6Rf54x8GpCHoojUbGgTRgLh6Pc2psb28PxWIR29vbz3T8vCm0iVBbu8lk4oeKTnfhcJhV5b0AReXlchnpdBp7e3soFArPpImBXwJOURSRz+cRiUReKNyWT1CnjNGbQsJxcuV8mSWAxWJho75wOMzW8meltEUBOc12et49TGl9ufDyOKFMqPz1kOaByr79FPQQZHRHOj7y55JnROVlx2KxiLW1NRQKBc70UIaY3sde15K8T+g50Gq1CIfDnEUol8twOByoVqt80KIN+EUZfRqDQIaEw8PDqNVqsFqtcDgcZ2b9eF2oDKVWq3m4s9Pp5CaWTqeDer3OjSR0+JV3uMkns1MQSUkI+n+63mT/QhIIKg+/KpCR6+dOqjGl60EPRXrkp0MCqK2tLdy+fRuZTAY7Ozvsq0PtoW8b8JAmRKvVYmhoCJ999hkGBgYQCoUwPT3NTp60wJ3VLq2j1Ot1zuxEIhE8efKEM2dHket74vE4px+fB22Y8kDpTaETnV6vZ1v+Fy1CNOPLYrFgamqKB5aepfeJHKupvHcUnU7Hbr3yk9dxQEHY814PaQKed1LsF8g0jczvNBrNIQG9/Nokk0n8+OOP3KU1Pj4Ok8mEYDDI72U/X8s3hbIykiTB6XRienoaxWIRXq8XkUgEiUQCi4uLSKVSrAV8UdBTq9VYV5jP5zmLT+9trwU9VN5qt9sYGxvDRx99xBq0vb091Go1pNNpzjhnMpnnrjO0z9K6Ttl/steg/ydNDnV3yTPvL8uq0cyt9z1P8GV0Leg5qgwnQVS5XEaj0UA2m0UikUAqlUIsFkM0Gn1mSNnrIvdooEFr1J3l8/kQDAYxMDCAgYEBmEwmzjz0EvIMg9wv4VWZGRqcCODQQ0GlxxchPzHIP+/ogk9u1xT0OBwOri0/DzpBWywW1kqcNZHo8zIrcmhzPKl7UD4KgzYN+eGg3wZgym0YqI2XtA40o4iyX/LAsNPpsD9VPp/nJgH5UEcayUJfE+iv0uGbIA/45Vkfv98PURQhCAIcDgcH7Eddlo/6TNVqNajVah4NotPp4HA4uBX7OEa8dAv6eSh77na72VU8n89zB+2r9lS6N183IKGgx2AwHMoKPQ9a52i00Uld+xMPeuhGpECG2naTySS7n5LoiVoPs9nsS82TXgR19ej1em6Z1uv1cLlcMBqNCIVCWFhYgNfrZRFzry7wNMNJp9NhcnIS165dQ6lUQiwWYx8HyvDQNZMvOhSckHgzm83yBG65iJCCHRo7QdebTMKoHHX0tdHG4XQ6X+rXQA+wTqeDy+XiIY1n5T3rdDrseB2Px7lNvZuQizmdmMlMkrKfDofjTIjE3xUKQKmrpV6vIxKJYH19HeVyGaVSCaOjo9Dr9RgbG+NBpMDTxZ4y0aVSCU+ePMH29ja0Wi1cLheXKr1eL8+foywDLfxK8PNqtFotgsEgSxIsFgtrUxKJBGcgqIKwurqKnZ0dADjkBHzr1i3YbDYsLy9jeXkZFosFs7OzmJmZ4VL7WVlTXoYgCPD7/bhy5Qqq1SpmZmZ4liJ1VL1v6CBQqVTw008/YWlp6dD/0zR1g8GA6elptkQJhUInEvicaNAjbyePRCK4e/cu8vk87t27h/X1dYiiyEEPZSXkQxnfFEqdWa1WnD9/HhcuXIDJZMLg4CDsdjucTidGR0dZcHjWMgZvAol9rVYrZmZmeOTBvXv3+HrT73q9HjabjVPxdG3cbjecTifXd+kURQETfRz5+1y+fBk2mw0TExMIh8MwGAy8UMmRt0m+qvVc/rHyP58VKOjZ3NxEKpXiEQfdgjys7t27h3g8jnw+z0HPyMgIzp8/3zcjQ2h9ymazePz4Mf9+9+5dNBoNeL1eTE1NwePx4NNPP8Xs7Cx3prRaLTx58gTffvst8vk8EokEYrEYOp0OZ5ZplAt5XlHW4nUmtys8hQ6wg4ODaDabmJqaYmfhYrF4qM06n8/j97//PetVaKh0Pp/H3t4eVCoVC3BtNhv+6Z/+CcPDw5zt75WgJxgMwuv18j5KjRTHFfTQHkBt7k+ePDlUUTAYDAgGg7Db7Zibm8Mnn3wCj8dzYtrMEw16yDeH9CWZTAa5XA7ZbBb5fP5QeetdoM3TbDbzRG+atms0GvnkRWMRerl1kaBrotFo+Hqo1Wp4PB6+9vV6nUdP2O32Z4Ieh8MBu92OSqXCJ1O5MFOj0XA7o9PpZH8jCpbIFuBo0NNvc3EokO/meAd5ubNWq6FcLqNarUKSJB5/Qb8oY9rLUJmdTNtofapUKpxlNpvN8Hg8cLvdcLlccDgchwzaHA4HbDYbaxONRiPa7Tb7WdG0cL1ej2KxiEKhwAJnpdPr9aGDEY08oCwxicxpY1epVDw2p1qtolQqPXNIIyNW8ht7Vdn+LCI/SMq1l/V6/VjWWiqtiaL43K5amnZAHlYmkwlGo/HE7v8TC3okSUIkEsH9+/eRz+exuLiIR48eoVKpIJVKoVwuH1KDvy2UKtZqtbh06RKuXbvGtvHDw8Os5ZHX4vsJtVqN4eFhGAwG1Go1jI+PI5lMspFUq9XiDilS1NPNSJoPmgieSCS4rCUIAmd3nE4nZmdnsbCwAJPJBKfTCavVyh0FR2/ufmqNJk2Z3W5Ho9HoWqt9rVZDKpVCtVrF7u4uDg4OkM1muUWbhsSaTKaeLfkCv2R3qPOKsnA//PAD8vk8TCYT5ufnYTQaMTs7y6MlQqEQbDYbOp0OzGYzOp0Ozp8/D7vdzm7j2WwWjUYDsVgM6XSa9XTpdJoDXqvVitnZWVy/fp0Dn15/Bt4XVIbX6XTs3yWf+2S1WjE3N8ddXvfu3WPrDbr+FNQ2m03uauq1oEcOHWJJSvC+y9byA129XueMmTzTY7fbMT8/j8HBQZw7d45nWZ7UfX+imZ5YLIbbt28jkUhgeXkZKysr3Bb3vm40ssk2mUyYmpriqeDkt9NPG+zzUKvVCAaDCAQCaLfbmJqa4mCHurmsVivrnyjoabfbbFC1v7+PO3fu8MmKrqPP58O1a9cQDAYxOTmJubk5zhb147V+EXq9HhaLBdVqtWtBN80qKhQKiMfjPPTU5/MhEAjA7/dzZ1wvv3e09lQqFWxubmJ3dxdbW1tYXFxEoVDABx98gEuXLsHlcuHy5VMkTn8AACAASURBVMuYn59nIfLR6+J0OjE2NoZOp4NqtYpKpYJqtYoHDx5gY2ODAytyFCZBqU6nw8WLF9mGv5ev9/uEhOQAnnt4MJlMOHfuHCqVCqLRKNbX1/n/5CJnynCTyeS72KGcBSjrb7VaYbVa3+vXJvPVRqPBGtmjWCwWTExMYGxsDKFQ6MRLu8ce9IiiyAZGmUyGS1o0ruBtBMovg1J5pCA3m80snu3V0+qbIl9YdTrdIc0UzcuST/am60blRxoHQh4MlM6k7iun08ltoMo1PwxdK7PZ/MJF4bihtlMyjSwWi5AkiUvCXq+XB8P2MrRA07y+VCqFeDyOarXKY2doLpDD4YDZbH5p1oueK7kxXKfTYbdwGrUiN32jv/fDZnuU563/VHJ6H9DG7vV6IYoiSxxIAySvKvTTdT9OjkoVXvZx3cpqHvuKS0KmXC6HH3/8EYuLizxc8X0HPMAv9ULq8iGR2lly7D0pVCoVBzgkMKcZW0fNotrtNqLRKB4/fox4PM5zswBwCtPv92N2drZnvS/eByqVCk6nE+FwmIOMk4TS/8lkEt999x0ikQh2d3dZ/zA1NYVPPvkETqcTgUCgp7MOoigikUjwOJZbt25heXkZgUAAV65cgd1ux8zMDC5dugSz2Qy73f7aQTw9ExqNBl6v91CqnzRxlEmlbFM/bbw0XojWELrPqPHkfawdWq0W4+Pj8Pl8ODg4QDqdhslkQjqdxtraWtebCBS6w7EHPbVaDQcHB4jH44hEIuwEfFxQypNKXDab7RnhrMIvkOPuq6BFigIecmGmhVyn08FqtXKrusLzIZdfh8OBSqVyoiJ6uf9MqVTC1tYWVldXUalUIAgC9Ho9vF4vxsbGWADay9A9nU6nEY/Hsb29jfX1ddbsDA4OYnJyktvU3wS5eNRiscBut7OepNVqQaPRcCao13Ukz4Nayp/nCP++DkzUcep2u6HX6xEOh3nvoTZ2hf7j2IMe+QytVz3YZNlODr00sI8QRREHBweIxWJd7XzpJ0ibUC6Xsb+/j0gkwrOirFYrbDYbRkZG2IVWyai9GrVazeJLCs5J5EoGdxsbG+xE6/F43slOgZ69er2OWCyGQqGAra0t5PN5NBoNGAwGHrvi9/s5U9qrIn8qLZVKJUQiEaytrSGVSsFut2NsbAyjo6MYHBxEMBjkeUSvQ7PZRK1WQ7vd5knT9Xodu7u7bAfQaDRY3B8Oh2G1WhEMBk/Uhr+b0LWv1WrY3NzE/v4+/x95yqjValgsFmg0mne+B+n7VSoVNiV8kRO9wvuBbBzIAuW0cWJBj3y0wYswGAwYHh6G3W7H+Pg4/tN/+k+Hhn2Wy2X8y7/8Cz7//PNDE7oVjgdJkpDJZHiDvHPnDr777ju0Wi0uZ42OjuK3v/0tRkZGMDg4CJvN1u2Xfeqh4J4yAOQqS9mzWCyGL7/8EktLS1hYWMDHH3/M88jeZlOk5y+bzeLbb7/F2toaotEoZ13HxsZw8eJFOJ1O7qrQ6/Vd0RsdN5IkcftyKpXCrVu38Oc//xkmkwmhUAhzc3OYnp7G1atX4fF43ug6VCoVHBwcoFqt4tGjR1hcXES5XEY0GkUqleKOGRp/8PHHH8Pv98Pv9/Mm38tBD7kil0olpNNpfPnll/jxxx95X1CpVLhw4QL7uJDr+tvqAimTV6lUkEgksL+/j83NTR51pPD+IRdmMul8Hdf/k+ZEVjV5tudlkMbEbrfD7/djcnISfr+fv0ahUOAZNi9aHOTeMb28gBw39J6RmVcmk+HpuYIgwO1284l1ZGQEk5OTbGio8GJIwEdlRbJOoHEHwNOScCwWQ7PZRDgc5oXjbe5peZdKrVZDIpHAzs4OO2qTlwaJl8lPqVffR9Ku1et13gz39vbYeHBgYAB+v5+9pV73a9JiT87NsVgMGxsbKJfLSCQSSKfTnBmlNS4YDHI2qV9mcsmvfSwWw+bm5qF73+PxoFQqweFwsNbwTaHPoVb0Wq2GWq3GGWvKxikcD3TdXzRb8OgefdL3/ak7ytFDUa/XubRSqVTYzCubzXKb+/Ow2WwYGxuDy+WC1+tVuofegkKhgFgshmq1ipWVFdy/fx+FQgEHBweQJInLIYODgxgdHYXP5+P2ZuV6vxoqb1Hr5vXr15HJZPDgwQPU63WUy2Vsb28jlUrBYrHwZun3++HxeKDRaNg0kIIo4Jdnp9PpoFarcVCTyWTYpv/JkyfY2dmBRqNhV+C5uTnMzc3B6XT2/DPT6XRwcHCA5eVlpFIp5HI5CILA7c1TU1Pw+/0vLdO2222Uy2WeGUii/ng8zqNz9vf3Wa8yOjqKmZkZWK1WTE1NwefzwefzcZNFr9sCEKQli8ViSCaTz21oiUQiuHXrFnw+H8bHx3H+/PlX+rhQIEvO2NlsFqlUCs1mE9lsFsViEclkEru7u+z4T/sHaa/IDVg5ML8bkiShVCohGo0imUyiUqnw/5GBJHXUBQKBN2oOeF+cqqCHPBMo2CkWi9DpdIjFYohEIigWi0gkEi8V/3k8HiwsLMDv9yMcDvdkiv64SafTuHPnDlKpFB48eIDbt2+zUzY5Nk9PT+PixYsIBAI816xX5tUcNzTE0+FwYGFhAQ6HA1tbW9jb20MikUA+n8fS0hK0Wi0KhQLS6TTsdjv7xBgMBng8nkNmdjT7KZ/Po9lsclauUqlgZWWFtVhLS0uIx+MYHR3FzZs3EQwGMTMzgxs3bsBisfS8+3Kr1cLGxga+/PJLFAoFRKNRCILAhmnXrl1j75yXfY10Os1Dke/du4dkMomDgwOsrKygVqvx5GiLxYLz58/j/PnzcDgcmJ6eht/v58CVnpl+eG5oBMvW1haSySTi8Tiy2SxnlQFgeXkZ5XIZFosFn376KdxuN2w22yEXZkK+/pNmp16vY2lpCQ8ePEC1WkUul0OxWORANJ/Pc7kX+MXgUH6IUAKet4fGuGxubiKdTqNQKECSJPbPo1l04XCYGwROeo8+ke/2utEzRexkkU2W7SQKLJVKhwykngd5oNA4BOUGfjPIu4Rs+NPpNDKZDGq1Gk/+ptlcTqeTBykqweWbQX49NpsNHo8HuVwOVqsVBoOBy4qNRgPFYhGpVAr1ep1n2dAcNXmpTKVS8QRlURR5vEulUuEAqFQqoVarQRRFqFQq2O12LuPQJPBeh4wDacQEjTEwGAycVXse8s63er2OUqmEfD7PzsvZbBaFQoEzQFS2pBIwef14PB64XK4T/qlPD/Lp9fLuNdJnVqtVpNNpVCoVvt/JRuN5wTjtA+VymYXj5BZPQQ9p5ejeJ2giOPkoKfPP3h0qpzcaDR7PAvziv0QNHPTMdWPe5bHvVBRJkx8MBT/PC1pEUeSFmuZ0mUwmFItFvqFpiN+LoDETVqu1L2ZqvS86nQ7XYBOJBB4+fIj9/X0cHByg2WwCAAYGBljzMD8/j9nZWTbZU3hzyMOFxqF8+umnGB4eRiwWw9LSEgcxGxsbPK9paWmJB7fSCZhsA/L5PGuBSL/QbDaRTCaRz+chCAI8Hg9ndxYWFhAOh+HxeHpWw0OQAWe9Xkc2m8XBwQE6nQ5nKUdGRuBwOJ75PNJDiaKIeDzOLtb379/H7u4ul1UAYHBwkCdF+/1+DAwMcNlsYGCAO/X6FXKDFwQB8Xgc9+7dg9ls5vdFFEWIoohCoYBarYaffvoJtVrtkDP8UWgfIe1Oq9VCPB7HwcEBd4k1Gg2IosjrGM0gVKvVGB8fx6VLl+B2uzE7O8t6tn7IvJ0kWq2WR9vQc9Gt8TbHHvSoVKpD6UNy6X1eeYrS8iqVikdVUIAkF6e9TIRGmR6LxcLD/BReDZ1gm80mEokEHj9+jM3NTT4h6fV6BINBLh3Ozc1hZmaGLc0V3hwKemgAriiKmJmZweLiIvb29lCtVlEoFNhbZGVlhZ8n0oNQ0KNWq5HJZLC/v8/ZUIL+7HA4cPHiRQwODmJubo6Dnn54D0lcKQ96dDodPvzwQ1y7dg0ej+eFQQ+dWiORCFZWVpBKpfD1119jZWUFDocD4+PjcDgcGBwcxPT0NKxWK6fv6WSrlE6e7gWBQAA+n4+Db6vVyq7YAA79OZ/P4+HDhwCe7+4rv8flB2HKNtCf6f/o48lXTKPRYGJiAr/97W/h9XoRCoVgNBp7/lnoBnq9HoFAAENDQ4eCnm5wIpkek8nEcz4sFsuhPv6jgY+82+RNvgelyQwGAw+h60bq7CwhSRJEUWQtVTqd5vQyjQ5ptVqscbBYLDxpvVsjFHoN0uRQyZDs8v1+P3vr0PRvuVCZ3GTJjFOlUqFUKqFaraLRaBwqe9GBg8pZfr8fTqez78qSdHiSl1aoKaLZbKJYLEKr1fIaRNefsgXZbJa1bTqdDjabDXa7HS6XC06nk38n4bnJZDp00FMAZ2x0Oh3sdju8Xi+XBOma09pPHUAv4nnVgpet9/I9wuVywWAwwO12w263w263K4fkY0Y+DLybz8Oxr3g2mw0zMzOceiSNyM7ODiKRCM9vehtoMTEYDLyIB4NB+Hw+eL1emM1m5SZ+CaIoIhqNIpPJIJlM4s6dO4hGo9jd3UUikeAp4EajEVarFRMTE7hy5QrreRTeHZVKBUmSYDabuQzidDrhcDiQz+fZW4RafBOJBICnws1arcZZGhIyi6IIQRBgs9ngcDhgMBh4arrb7cb169cxPDzMwWu/IJ9tRYFPo9FAIpHA+vo69Ho99vf3YTAYUK/XUSwWIYoiSqUScrkc2u02L9gajQYffvghPvvsM9hsNgwPD7Oho9Pp5BJ7vxgOvg0mkwlXrlyByWRCNBrFN998g0gkwiWp45jJSKWr4eFhfPLJJ/D7/ZiamsL4+DgsFovSfXqMqFQqnpBAnmPd4tiDHhKJdTod5HI5bsWtVCo88+dtbNhJxEnpfqfTCYvFwoJBJXJ/Ne12G5lMBru7u4hEIvj666+xurqKRqOBcrnMw0fNZjNsNhsGBgYwPj7OmTuF9wNlegKBAICn07rdbjeq1SqWlpag0WhQKBQgiiLy+TxvDC8yWKPTLGUdaGH3er24fv06RkZG+vK5OBr4tNtt5HI5RKNRqFQqRCIRAECpVEIymUStVkM+n0cikWD9x8jICNxuN6anpzE5OQmLxYJgMMgHrH68rm+DwWDA9PQ0XC4XNjc3uWxIWbX3HfTIZRYDAwO4fv06222QI7bC8aFWq3kAOHUtdosTy20LggCz2YyBgQEYjUaeuF6v11m4RuWW1wmAtFotPB7PoVEINpsNQ0NDbKGv1GZfTqvVQiqVwtbWFqLRKJ9uKeUvCAKsVisGBwfhdDo5m0bCQoXjQaPR8GnI6/VieHgY5XIZgiDAYrEcmtL9PFQqFVwuFzweD0wmE4aHhxEIBDjz048bM2XEtFot3G43RkZGIIoitFotSqUSl1woe0wnUoPBwGXA0dFRFjy73W7ODnRLkHmWocDcarWywWmz2WQ/KdJRVavVNz4Qy+UO1OlI4n+z2YxwOAyXy8Xdksp79+5Q9lQURVQqFeRyObYQOG2caNBD07drtRpCoRBGRkaQz+fx6NEjbG9vs301dUO8DLPZjOvXr2NmZgYejwdzc3NwuVyw2+0IBAKKpuc1qFaruHPnDv74xz+iUqkgGo2iWq2y5oEM7P7mb/4GLpcLc3NzcLvd72UmjsKLMRgM8Hq9aLfbPJKF5gfRJvCy7CjpfEisSZsz6Sj6ETppCoKADz74gOdjLS0tYX19na85NUGMjIzw9aNrOTY2huHhYej1ejgcDh6MqWQJ3hyNRsMePHa7HVqtFqlUCnt7e7h37x6y2Syi0SjvC28Cdcnp9XqMjIwgGAzCbrdjZmaGDT5nZmb4+yqH43eHss/1eh17e3t49OgRMpkM+zCdJk5UxWiz2VisSfqDTCaDWCyGeDwOQRB4EvGr0Ov1CIVCmJ2dhc/nw8LCAjwezzH/BL0FDXB98uQJZw6OppUdDgdGR0fh9Xrh9XqVuvcJQOJ/ADy5XuHdoFK4JEkIBoOYm5tDKpXC6uoqcrkcjEYjbDYb63bI/JHE+3q9HufOnUM4HFY2yfcAZWEoGJckCaFQCFarFclkEnq9nrObb/p1ySbFZDLB4/EgHA7D7XZjbm4OQ0NDLIMwGAzH9NP1H5TloYYAMlklV/LTRFdaN8gYLRQKweFwoFarIRAIoNFooFKpvFZk73A4MD8/j6GhIdjtduW09RZoNBoEAgFMTU3xYMRSqcTiZb1eD7fbzYZq/WKXr9C7qFQqWCwW+Hw+GAwGXL16FXa7HXq9nsuBlC0mV2UqlyuNEccDiVxVKhXcbjfC4TAsFgtEUWQ35WKxyB2LhMViQSAQgNlsBvBL5xaVdo1GI8bHxzE0NASr1Qqfz8clLSVwfb+0Wq1DkxQKhQKKxeIbZ+lOgq4EPWq1GgMDA3C73Wi32/jggw/YW+R1J6eTmy35KihR+5tjMBgwNzfHpo/ffvstSqUSjEYjLzznzp3DuXPn4HQ6lUVf4cyjVqvh9Xpht9vRbrcxMTGBer3O7rxy51j5iAi5NYDC+0Wr1cLhcPC4gna7zUNHtVotisUiVldXUalUDmWi3W43PvnkE4RCIf432ltCoRDreFwu1zPz6vrJquEkIP8ryvLQAfo0TrPv2jtPAkEAfW3L3k3UajVsNhsCgQBEUWQrdrmvErXikueIgsJZhzQ6APpW43SaoIATAI/t0Ol0nGUmw9mj+huTyQSfz4eBgQE+jKnVagwNDWF4eJj9eGw2W1d+rn5CXt4iEbpcm0udjfJZgd1CCXf7GJ1Oh3A4zL87nU6k02mYzWZ4PB4YDAbMzMz0xSBKBQWF7kMldavVyt2KtVoN09PTPGya8Hg8uHDhAtxuN/+bSqXiTlPK2CkcP5RJazabnFHTarVotVqQJAlarRZ+vx8jIyPcaNS119q176zQdXQ6HcbHxzE6Oop2u41PP/0U7XYbKpWKO99oOFy3o3MFBYXeh7SEJDg/f/48D6I+atFAguWjhzH5+qWsWScD2Wx0Oh22eqDSliiK0Ol0GBwcxOTkJDcGdO21du07K3QdCmoUFBQUTgPyOXCK7ubsQDopmlrvcrmg1Wp5lhqZB/eVOaGCgoKCgoJC70HeVSaTCTdv3oTP52Nn7Xa7DYfDgQ8++ACBQIBtIbqFEvQoKCgoKCgovDU0kw4Abty4gatXr/L/UVeefPCuImRWUFBQUFBQOLNQIKPRaE51aVIxnVBQUFBQUFDoC5SgR0FBQUFBQaEvUIIeBQUFBQUFhb5AeJMJqIIgpABEju/l9CTDkiR53/WLKNf+rVGuf3dRrn/3UK59d1Guf3d57vV/o6BHQUFBQUFBQeGsopS3FBQUFBQUFPoCJehRUFBQUFBQ6AvOfNAjCMJ/JwjCY0EQlgRB+L8FQTB0+zX1E4IgqAVBuC8Iwv/X7dfSbwiC8HeCIKwKgrAhCML/0O3X028oa0/3EATBIAjCHUEQHvz8HvxP3X5N/cZZXfvPdNAjCMIggP8GwBVJks4DUAP4L7r7qvqO/xbAcrdfRL8hCIIawP8K4D8DMAvgvxQEYba7r6p/UNaertMA8JkkSRcBLAD4O0EQbnT5NfUbZ3LtP9NBz89oABgFQdAAMAGIdvn19A2CIAwB+C2A/73br6UPuQZgQ5KkLUmSmgD+HwD/eZdfU7+hrD1dQnpK+ee/an/+pXTlnBBnee0/00GPJEkHAP5nALsAYgAKkiT9R3dfVV/xvwD47wF0uv1C+pBBAHuyv+///G8KJ4Cy9nSfn8sriwCSAL6QJOnHbr+mPuLMrv1nOugRBMGJp6fbUQADAMyCIPxX3X1V/YEgCH8PIClJ0t1uvxYFhZNGWXu6jyRJbUmSFgAMAbgmCML5br+mfuCsr/1nOugB8DcAtiVJSkmSJAL4VwAfdfk19QsfA/gHQRB28LS08pkgCP9Xd19SX3EAICT7+9DP/6ZwMihrzylBkqQ8gK8A/F23X0ufcKbX/rMe9OwCuCEIgkl4OuL1r3EGhVVnEUmS/kdJkoYkSRrBUwHnnyRJUk66J8dPACYEQRgVBEGHp+/B77v8mvoJZe3pIoIgeAVBcPz8ZyOAvwWw0t1X1R+c9bX/9M5/fw0kSfpREIR/BnAPQAvAfQD/W3dflYLC8SNJUksQhP8awL/jaefQ/yFJ0uMuv6y+QVl7uk4QwP/5cxejCsD/K0nSmWqdVugOyhgKBQUFBQUFhb7grJe3FBQUFBQUFBReCyXoUVBQUFBQUOgL3kjT4/F4pJGRkWN6Kb3Jzs4O0um08K5fR7n2b8fdu3fTkiR53/XrKNf/7VCuf/dQ1p7uotz73eVF1/+Ngp6RkRH85S9/eX+vqg+4cuXKe/k6yrV/OwRBiLyPr6Nc/7dDuf7dQ1l7uoty73eXF11/pbyloKCgoKCg0BcoQY+CgoKCgoJCX6AEPQoKCgoKCgp9gRL0KCgoKCgoKPQFStCjoKCgoKCg0Bec6TEUCsdDp9NBu91Gp9OBKIpoNBrodDqQJAlHHbwFQeBfOp0ORqMRarWa/01BQeEp7XYbtVoNoihCkiS0Wi1IkgStVgu9Xg+VSgWtVguNRlmWFU4v8r2A9gn6JUkSdDoddDrdqd0DlKdL4RnK5TISiQRqtRo2NjawtLSEWq2GRqOBer1+6GO1Wi1MJhM0Gg2mpqbw0UcfweFwwGAwwGg0nsqbXkHhJKHNIJPJ4IcffsDOzg4qlQpSqRSazSZGR0cxNzcHi8WCUCiEUCgElUpJwiucLiRJQqfTQbVaRbPZRLVaRTKZRK1WQ7lcRi6XgyRJmJiYwNTUFPR6PTQaDdRqdbdf+iGUoEfhGWq1GmKxGPL5PH744Qf88Y9/RLFYRKlUQqVSgSRJHMzo9XoOcm7evImxsTHO9BgMBiXoUeh7aLMoFAr46aefcOfOHWSzWWxsbKBareL69esQRRFerxcGgwGDg4NK0KNwqqAMf7vdRrVaRbVaRT6fx+bmJgqFAjKZDHZ3d/njwuEwZ3qUoOcEoNRbrVZDvV5Hq9VCpVJBtVo99HEqlQoajQYqlQoWiwVOpxNqtRoqlapvFh26SVutFqfek8kk9vf3kc/nkUqlUKlUUK/X0Wg0IIoigF/KWq1WC/V6He12G5lMBjs7O6jVavD5fGi329BoNDAYDKc63amgcFx0Oh1UKhXUajXkcjnk83kUCgWUy2WIosglLnmZ+KxBP0ez2UShUECtVoNarYZWq31mHaW1lcrhlAnQ6/W8OZ7Fa9Dr0J7aaDQQjUaRSCRQLBaxu7uLUqmEfD6PTCaDTqeDVCqFRCLBeyq93/R7t+m5oKfdbqPVaqHVamFrawubm5soFotYXFzExsbGIU2KwWCA0+mEXq/HpUuX8Nlnn8Fut8NgMMBgMHTxpzg52u022u02CoUC1tbWkMvlsLm5ie+//x6ZTAbxeByZTAaiKKLT6UCr1fINrNFoIEkSKpUKAGBxcRGVSgU2mw3z8/NYWFiA1WrFyMgIgsEg1Go1B5UKCv2AKIrY2dnB7u4u9vb28OTJE2xubqLVaqHdbkOr1UKn08FsNsNisUCn03X7Jb8R7XYb2WwW+XweyWQS33zzDTY2NmAymeDz+aDX6w8FdHq9Hnq9HjqdDj6fDw6HA1arFUNDQ7BarVCpVJwpVjg9dDodtFotJJNJ/Nu//Ru+++47iKKIcrmMdruNZrOJWq3G71ur1YLD4cDFixcxPj7Oh9/ToFfr/it4z5BAUBRFZDIZbG5uIpPJ4Pvvv8dPP/10KOixWCzw+XywWCwwGo24evUqvzHyEk6vQml3EljG43FEo1Gsra3h0aNHSCaTXLuVJImzYiS4VKvVaLfbaDQaaLVaiEajyOVyMBgMaLfbcDgccDqdcLlc8Pl8AHDqUp0KCscJBQW7u7vY399HMplEOp2GIAj8DFHgo9PpztyGTxqPXC6HWCyGn376Cffu3YPdbkc4HIbRaOQTviAIMJlMMBqNMBqNqFarCAQCqNfrcLvdMJlMAJQ14jRC+0S5XMajR4/wpz/96bkfp9Fo4HK54HQ64Xa7MTg4iFAoxALn08CZDXoouKH0cTQaRbVaRaPRQKVSgSiKWF9fx9raGorFIgqFwjNfg0o6kiQhEongxx9/hNvtZjEhdSPp9fou/ITHT6vVQiqVQqFQQCKRwNraGvb397G/v8/XxWq1wuv1QqvV8mlUpVKxSK1WqyGVSqFarUIURVQqFTSbTaRSKaysrMDhcMBsNvO19Hq9MJvN3f7RzySiKEIURbTbbVQqFTQajUOnaKqzUwaBfnk8HjgcDjQaDcTjcZRKJT65SZIEj8eDwcFB6HS6virtHheSJKFer6NWq3EJYGNjA8lkEuVyGQBgMpng9XphNBoRCoUQCATg8XhgsVjOVNCjUqlgNBrhcDhQr9cxNzcHtVoNnU4Hg8EAlUqFer2OarWKTqfD2QCNRoNGo8FlkGQyCZvNBp1OB5PJxOuN1WqFRqOBzWaD1Wrt9o+r8Ao6nQ5KpRLvIalUCvl8nvfR0xD4nNmghx6gRqOB7e1t/PGPf8TOzg5KpRJyuRxEUUShUEAul0Or1UKpVHqm3brZbCKbzUIQBNy6dQtbW1uwWq24efMmbt68CavVilAo1LNBT6PRwNraGlZXVxGNRvHVV19hZ2cHoihy0BMMBjE/Pw+z2Yzh4WEMDg5yx5ZOp0MqlcKjR4+QSqUQiUTw8OFDlMtlPH78GJFIBBaLBcViEfl8Hm63G5cvX1aCnreg0+mgXC6jVCqhWq3SBO1DurTd3V0sLS2hUqnAbrfD4XDAZrPho48+wsWLF5FKpfDFF19gdXUVzWYT5XIZkiTh448/xj/8wz/A6XRCp9P17P1+ElDrbjqdRjQaRSaTwa1bt/Dtt9+iVqshnU4DAFwuFy5dugS/34/Lly9jYWEBdrudW9fPCiqVCi6XCzabDR6PBzabUtwiJwAAIABJREFUjbVL29vbKJfLiMViSKfTqFaryGQyyGazAMCaHo1Gw1YX9HWMRiNGR0cxNTUFq9WKmZkZmM3mM3Vt+pFOp4P9/X3kcjk4nU4MDQ0hEAjAbrfDbDZzNq+bnNmgR5Ik9pApFArY2dnB6uoqSqUS0uk0Go3GMy3W9MCQKKvT6aDRaAAAkskkisUiTCYTJiYmUCgUIAgCC3d7EdLyJBIJxONxzvJoNBrOElgsFr5pR0dHMTIywqcw0kTlcjlotVoUCgWo1WpIksTBp81mQzQaxdDQEARB4Out8PrQvUqlxnK5jHQ6jUQiwaVGlUqF/f19rKysoFKpwOVywePxwOl0YnZ2ljOgBwcH2NjY4OdGkiSMjIygWq3CYrGcipr7WUW+rtTrdRQKBWSzWSSTScRiMYiiiGazCeBp16PX60UgEIDX64XT6TyTmQwSJOt0Omi1WoTDYfh8PsTjcc6u63Q61lpSKazVagHAIRG3SqWC1WrFwMAA211QhrJSqaDdbvP3lH9/hdNFrVZDrVZDq9VCoVBApVLhe+A0cGZXuHq9juXlZezv7yMSiWB3d5c1KI1G41BpRqfTYWBgAG63G7VaDdvb28hms2g2m6hUKlwmazab0Gg0LNolf41egwLCbDaLvb09rK+vI5PJoNVqscBwYmICdrsdU1NTWFhYgMVigd/vh9vthkaj4fKWx+PB7OwsBgcHYbfboVarUSgUsLu7i52dHUiSxIFUKBTC7OwshoaGOEOhLFrPIjf9yuVySKVSaDQa2NzcRCQSOWQpQKUCQRCQSCQ44DeZTGwuGYlEoNVqkUwmsbW1hWg0ClEUUa/XIUkSNjc3cfv2bbjdbkxOTmJiYkLRVbwFFOg0Gg0sLS3hzp07yOVyiEQiaDQaUKvVcLvd0Ov1OHfuHObm5jA0NIShoSFotdpuv/x3RqVSwWAwQK1Ww+PxYGJiApVKBYFAAKFQCI1GA7lcjjPx+XwelUqF71N50CgIApLJJO7evQuDwYBEIoHHjx/DaDTC7/fDbrfDaDTC5/Oxbki5ZxVehzMb9JTLZfzlL3/BDz/8gEwmg+XlZeRyOd5ENRoNnE4nQqEQbDYbPvnkE1y4cAHJZBK///3v8eTJExSLRRbhUjcFBT30914LeqjbKp/PI5FIYHl5GXfv3kW9Xke9Xoder8fo6Ch+97vfYWBgAMPDwxgfHz9kNCXXkVitVm5Pn5ycxMjICPL5PL766iuk02nUajWsra1hfX0d09PT+PjjjzE5OcnZCWWheha5OHx3dxcPHjxAPp/H7du3cf/+fbRarWfsAwBw2zAA2Gw2bjF98uQJtra2kMvl8PDhw0N+GgDw8OFDiKIIp9OJv//7v8fo6KjyvrwF5XIZ+/v7KBaL+POf/4x//dd/5XJko9HgLIbX68X8/Dw++ugjjIyMsKXDWUetVsNkMkGSJG4SoYCm2WyyjkwURVSrVWxsbHAAXiqVWAu4sbGBcrmM3d1d3Lt3D51OB3a7HVarFQ6HA9euXcPo6Cj8fj8uXboEr9fLJV7lEKXwKs5c0CNP9RcKBRZKkb6HNAkajQYmk4k7iHw+HwYGBjhlajKZOLMDgE2UaCPu1dZqEoCT0JLKJa1WC2q1GkajERaLBV6vlzM7drv9hSdREi0CQLVahc/ng06n48+hzBu9X+TmCaBvbAHeFNooqCU0m80im80ilUohmUxyQN7pdA59HgUyarWaP6bRaHDXBaWaj5YYK5UKe2xQp54kSWg2m5ySpmBXaSl+PlRuL5fLKBaLyOVyyGQyKJfLfM00Gg2sViucTiecTifsdjtsNltPicfp56CuNOCXsh+t3aTHpJJVs9mE0WjkQN9isbCGje5j8v+hwMhqtXJJnSxG5Pdmr1zP0wDtGbSOH113zhpnKuihdH+hUMDBwQH29/cRjUY54AGeCgQnJydhtVoxOzuLCxcuwGKxYHR0FG63G61WCxMTE2i329jb20MqlUK5XIbdbufuiXA4jIGBAVit1lMhvHqfkMhybW0NyWQSyWQSpVIJVqsVExMTcDqdOH/+PMbHx+H3+2Gz2V771E/Xzu12Y3x8HJOTkygUCojFYkgmk6jX6+xV4nK5MDY2BovFcsw/8dmBMov5fB57e3sol8u4d+8evv/+exQKBezt7XEp9mULDz0nW1tbhwzf6vU6dw/JIfNOrVbLZp71eh137tzB2toaDAYD/H4/LBYLPB4PxsbGeu65eFsoQCUB5w8//IBkMont7W20Wi0uATscDvh8Pnz22WeYmJhAIBBgM9ReDyDp55Nrd8h52mw2s7ay3W6jWCyyBo1Ku/V6Hdvb29jb2wMAbG5uYm9vD3a7nTtEw+Ewzp8/z0Gl2+1WAp/3BHVCb2xs4ODgAIlEotsv6Z04k0EPCW739/cRi8W4PAUAXq8XCwsL8Hq9uHr1Km7cuMF1ZlpgyCxJrVZjaWkJAGC1WhEOh+FwODA8PIyBgQGYTKaeW9wp6FlfX2cxbKlUgsPhwPT0NCYnJ3Hu3DmMj48fctN8HSwWC2fQxsfHMTExgUwmg0qlcijosdvtCIVCGBwcVIKen5HreAqFAjY3N5HNZrG4uIjbt2+jVCqxkeTrfK1cLodyucyu2fLvcRQSSOt0Og56isUivv32W3z++eew2+2Yn5+H1+vF1NQUPxsK4KyzKIo4ODjAjz/+iIODA8TjcQ56vF4vwuEwwuEwbt68iQ8++ABqtZqzE/2APPABnmaCBgYGEAwGOUMp9w0jKxIaf/Mf//Ef+Oabb1Aqldh7TafTwWKxQK/X4/LlywAAn8+H0dFRXrsU3p1Wq4WNjQ18/fXXSKfTSCaT3X5J78SJBD1y/cC7nGpo0abUv1yLQydfWkxMJhOnPeUtuCSgpe4kOnkYjUY2VSJfGaoT9xq0sJAmRKPRQKfTweFwcPspGae9CVQiJPdNq9WKZrPJ5S+5nsjlcp0aNf9pQD6Bu1AoIJ1OI5PJoFAooF6vc0nwTZBng+gZpPudfqnVapjNZthsNi4rZLNZFAoF5PN5lEolAEAul4NarWbtRbvdVjQUeLohlMtl1Ot1FItFFItFHjFBJR673Q6fz8eZZAp2+v3avWhtPdpAolKp4Ha7EQgEYDQaUSwW+d6moL5arSKVSgF4mu2v1+s9vYafJDTSqVAooFQqnfmO5mMPeuTKfPnclbd96KnjiuZBNZtNPiEAYD2Jw+F45ZRv6nzR6XQYGhrCr371K+5gsVgsb7XxnyU0Gg3PRwkGg5idncXly5dhtVphNBrf+uuqVCp4PB5MT09zx5AgCKjValhfX+esxfXr19/jT3O2IWFnJpPBysoKvvzyS55h8zYBDxlIyru7SHfSbrdhMBjYF2ZkZASzs7MwmUyoVqv4wx/+gGKxiMePH3P5l7ohNRoNO5fr9fpXPmO9Cm3K+Xwe9+/fRzKZxOLiIvb29pDJZAA8zR7bbDbcuHEDf/VXfwWHw4HBwcFDzQAKz6JSqSBJEnsWGQwGfPjhhxgZGUGlUsHOzv/f3pnFxnWm6fk9tW+n9pVVxX0TKdmSLWlsy8uM2z09M30xA8xFAsxcJJjcBciC3AS5S66DALnKTYIgF0Ewg3RuGggybvd0N+xut2XLkiVKJMW9yCoWa9/3qpML+ft0ilosSiKLZP0PQEiWJeronFP///3f8r7bSKVSKBQKWF9fR6FQQKVSwS9+8Qs2PiYhVTrECV6ebreLdDrNQr90EDqrHHvQo1ZOpizAy5rrHW6oooyPGpogoLTniwY9fr8fCwsLrJsxDKlnckKXZRlutxvj4+OYnZ195UZASZJgt9sRiURgNBpZxbnZbGJ/fx/VahV+v19o9qhoNpuIx+OIx+NYXl7GN99880q1c3qv1ZYhpBTcarVgsVjg9XphsVgwPj6OS5cuwWAwYHV1FXfv3kW5XMbOzg4qlUrfkMDIyAiKxSLcbjdvSMO6eVPmcmNjA1tbW1hfX2eFc7vdDqvVCqfTiQsXLuD999/nIYvzvq68Dsimg5qhZVnG/Pw8Go0Gtre3+6QANBoN0uk0VldXoSgKAoEA3njjDX7PZVke2nf0dUAl93g8zua5Z5ljD3poHLHdbrMVgdpv5mX4oTHy573g6gklj8eDsbExNsejD1ilUkGv1+Npi1fJepxGKBBptVo86UMCYMVikaXgX2VxNplMsNvtaLVa3OtD34/KkfQlTr3gPhoqaR2WS6D3lt5F9cGBHOzpM0XlWovF0ufxREEPTchQuYB6qzQaDeunUBmr1+tBo9H0GflSb9FZn+J4WdQ2INQXR5kHKtmazWZ4vV62XVHLPQiOjnpy0Gq1otPpwOv1IhqNcptCIpFAs9lEpVJBOp1Gt9uFx+MZ9KWfaaiKop7AexHUZfTT9s4fe9BTrVaxubnJzbIkFkh9Iyd9Q/R6PQKBAAwGA6vW1mo1jIyMwGq1otvtYnNzE6lUCmazGZcvX8bU1NSJXuNxotFo4HK5MD4+zh431WoV+Xwe29vbsNvtcLvdbBb4sn+H2+3mPqHR0VGEQiF0u11WuSbtDpINGHYl4Fqtho2NDSwtLSEejz+RBdPpdIhEIhgbG+uTVCBVbCpXBQIBdrKmLAxtuNR02+12eVM2GAy8mJF8wfLyMqrVKmq1GgCwwjAFxvRzk8l07nSsXgSyvkkmk1hbW8Pvfvc7bGxsoFarodFoQKvVIhKJsIbMyMgIjEbjuZXBOAnUAX4gEOC12+l0olwu44svvsDDhw9Rr9eRSCRw8+ZNeL1eHpo4bRvvWYB0leig86KoJRrU/YOnhWPfaVqtFgqFAvut0IjioKY/tFotbDYbFEXh03C73YbZbIZer+fJl+3tbVitVkxPT58rx3VJkvoMArVaLavzFgoFZDIZ6PX6V24yNpvNMBgM6PV6LLFPQc7hbAGdJM7LPX4Z2u028vk8Dg4OUCgUeOJKjcPhQCgU4iCRTr2BQICzlWNjY3zv1T091CehDnqcTicMBgMKhQJSqRSPDpNPEqEekSdxOSpZDyOdToddxROJBPb29hCLxfj/6/V6yLKMcDgMt9sNWZZFwPOaoPYFAGxiSUEoBfDFYpFFD6vV6oCv+OxyeJruKJ93tTzBy7azHBfHHvSQxgdNiFCGZ1Cy65T6Bx6bllLnf61WQ6fTQTKZRDKZhNPp7Fv8zwMU9FDa980334SiKAgGg5icnEQwGITT6XzlzAu98DqdDg6HA8FgkM0HSSjv4OCABdo8Hs+5bhr/IahfjYISWmBsNht/zc7O4s033+RyFfVkOZ1OGI1GbuBXGzmqT1qKonDGp91us+5PIpHA+vo6e9iR+zotUkajER6PB2azGSMjIzyBNIhM7aCg0iDZt2xubmJ5eRnxeJzXCLPZDLvdDrPZjNHRUUxOTsLtdsPhcAzNfTpJDk8i0vtOpWKj0ciDLoKjoSgKEokEWzbt7u6yeOTzgh9JkljYlioGpPF1WlTHjz3okWUZc3NzbPGgXogHsRDodDp2n1aPB29sbOC7777ra9IKBAL48MMPT/wajxMqb8myjFAoBIfDgQ8//BAWiwWRSAR2u52bYF8FGl83Go0IhUKYn59HOp1GIpFALpdDIpHAysoKms0mxsbGjiSCeB4h9WRSnaWF2uPx8Ob5R3/0R/j444/7gg31Yk+9curTFdDf40alrHg8jm+//RaZTAb37t3Dl19+iUKhgGKx2DctptFoYLfbcenSJQSDQSwuLvLo9Q8NCpwner0eCoUCstksEokEvvjiC9y8eRPVapWz2A6HA7Ozs3A4HLh69So+/PBD7mcblvt00lAZhb70ej1qtRoODg64ZCs4Or1eD/fv38fPfvYzZDIZrKyssEL284JInU6HyclJvPvuu/B6vbh27Rrm5uZeqYf3dXPsQQ+NRb9Ofihd9rymq8Pj8tSnkMvluDudSgNU+jpvUKaNpnHsdjsMBgP3hrxOqO9ElmVONbdaLTQaDZTLZbamGNZSiRp1yY/eXXK6l2UZXq8XoVDolbKk9L3b7TZyuRwODg6QSCSwubmJUqnEv099iqbeLJ/PB6fTCZPJNHT6J1QaJJuJbDaLg4MDlgCgrJvD4WBFYI/HA6vVOuhLP7ccVnqmL3XWVKwrL4eiKCiVStjd3eUmfZKeeR6SJMFqtbIuldPp5CGJ08KZ6h5V96NUKhWeZqEGTRJRSqVS0Gg0GBkZeeKlr1QqPFq6u7uL3//+90in00gmkygUCmi1Wqxxct6hhVrd7Pq60Wg07HavFipstVrI5/OwWCwIBoNDvzjR4cBut7PwH/DofaXNtVAocIPzUXpEKNCpVqvY2tpCNptFLBbDrVu3kE6nsbOz05fdofeBshSRSASXL1/G7Ows/H4/6/sMU3mr0+lge3ubNXmSySQajQY0Gg1PZ01MTODdd9+Fz+fD5OTkuXBOP82QKCH5eDUaDbTbbVYO93q98Hq9Q/OOCl6MMxX0aDQaWCwWuN1u1Go1rp/T6bjdbqNer7ME/OTk5BORablcxu9//3ssLS0hkUjg22+/RS6X4+ZMSZJgNBqH4oRG95N6nI4jGtdqtdyA22q1OJPUbDY5OB0dHR16dWZqsPd4PGi329xTRc7dlI1sNBoAcCQdKWpGLBQK+Oqrr7CysoL9/X3cuXOHvyc1TqttAlwuF7xeL6anp/Hee++xfYK6hDYstNttPHz4EL/85S9RLBYRi8VQrVZhtVo5o7O4uIgf//jH3Pckgp7jpdPpoFwuo1KpoFwus6J5KBTCe++9B7fbzSbTAgFxpoIe4NGJmEZyqUxDWi8AuGPfaDTyCCkAHr3L5/PIZDLIZDLI5XJPKEyqXYJJSPE8cxJd9dRv8qyNUjQaPi4DUoMwvYekoURmoYVCoW965XlCkjR9Ua/XUa/Xkc/nkcvlkMlkkM/nUS6XuXmfPkP0nAwGQ58juCzLQ3EQOAytG7VaDeVyuU/DiFTmrVYrZFmGLMvcmE/aSYLXD2UuW61WX4mcMpRmsxkOhwMOh+PU9JGcd9Rrh3q69DRNbRFnKuihpkqdTod2u82LsU6n4+Amn8/j/v37sNls7Jyu0WiwtbWFeDyOfD6Pe/fuIZFIsNrs4b+DapI+n+/cCROeNNQ/Qqq+lFEwGo3w+XwIhUJwOp1D3cQMPMrcRCIRfo/X19cBPPbkAoAvvvgClUoFHo8H7777Lqampth0kbIK1MtG2c9KpYIvv/wSS0tLyOfzuHv3LhKJBGvyqC1iJEniz43T6cSHH36IS5cuwe/3w+/3D+bGDBBq+F5fX0c+n8fS0hL29vY4M2Y0GrlZMxqNYmFhgTfaYX+fjwuazKrVaojFYvjss8+474TKsRMTE5ifn+f+qtO26Z5HqJ+NprUDgQCcTuepNCY+U0EPLcpWqxX1eh0Oh4ObpGmRKRaL3IxMwmC9Xg+ff/457t69i2aziUKhgHq9/lTtAfo7fD4f3G73a2/sHTYo6Gk0Gmg0GpzVMRgM8Hg8CAQCsNvtp6rRbRAYjUbubSqVSlzeorJru93GN998g9XVVUQiEV7QyViXNKboFEyGsqVSCV9//TV+/vOfo1arIZVKoVqtPuG4Tk3LdrsdgUAAfr+fPaOMRiNkWR7UrRkYiqIgmUzi1q1byGazWFtbQzKZZJVqnU4Ht9uNy5cvY35+HsFgELIsi+zCMdLtdtkMd2NjA7/61a+wvLwMWZa5zDg6OoqpqSnYbDYR8JwQRqMRLpeL904ShjyN++eZCnoISqWRtw0A3iRIRVKj0aBSqSCTyfDpoFKpsGfXYfE3vV7PJTOXy8Xd56fxoZ0lFEXpU/IFHgU8JpOJp5KGwefsh9BqtVxOcjqdcDgcqNVq/L5Ssz5laGh8utlsctmp3W6j2WxydoiUtrPZbF+jJ30+6DNDxqF6vR6hUAiRSIQntejXh+n50PQPOahTOZzsaSRJ4l44elbUX6i+T5R1IzFOdQngVf3thgW6hyRsWq/XcXBwgEwmg2w2i16vxxOOoVAIdruddcbE/T06am9Lks84bInzNPR6PcxmM6xWKywWC38+TuOww5kMeoBH/Qxzc3NotVrY29tDJpNBsVgE8Djw2d7e5v+OxWIoFov8AToMCSnZbDZ88MEHuHHjBmRZFhLmrwiJ3z148AD5fB7Ao3sdCoUwOzvLuibDbkNhtVoxMzPD79v29jYSiQQSiQRPV1F5MJ1O49tvv0W1WoXH48Gbb74Jj8fD+kf1eh3xeJwboJeXl5HP5/sMesnLS6/XIxwOY3Z2FjabDRcvXsTi4iIbkVqtVt6sh4VGo4FisYh6vY7l5WV8/vnnyOfzSKfTaLfbsNlsmJubQzQaxdTUFObn5zExMcEZN+Cx/ECn02E/Nb1ez4GkTqcTwf4PQA347XYbsVgMiUQC+XweX3/9NTY3N9HpdGC1WrGwsID5+XncuHEDbrcbY2NjItv2khQKBWxtbaFUKmF9fR25XA7FYhHNZvO5gY/VakU0GmWLo4mJCVgsllPZB3hmdxo6lVIAc7j3hhYbyi5Qd//TIG2BcDgMl8uF+fl5XLlyhWXOBS8PTQ2plWspozEyMoJIJHJs4/JnCYPBgGAwCOBRX9rExAQMBgMajQb29vYAgN/1UqnEysmhUAherxfdbheJRAKrq6sol8tYW1vD6uoqN0DTu08lLWqcpr93YWEBbrcbV69exdWrV4d68oimgqrVKhKJBB4+fMgChMCjrHIwGMTMzAzGx8cRCoWe6HlSlxiLxSJSqRQrZdOEKJnDCp6O+h5mMhn2RPz666+xvLwMp9OJubk5uFwuzM3N4b333oPX6xVefq9ArVZDPB5HNptFKpXqm4p7Hkajkf3QqLx1Wvthz+ybQZ36BoPhmZMSJFB1uH+BFh7q9DcYDIhGo5iYmOAHp5b6Fxwd6pdqt9uoVCrI5XJceqH+EbVxprjPj7FYLBgZGYFOp0OxWMTe3h4qlQrbIHS7XZTLZWQyGQDA6uoq0uk0MpkMYrEYarUastnsE+lp9UmNGg5tNhui0SgikQicTidkWR76jbhSqSAWi3F2h9YOalAm3algMAi73Y5Go4F8Po9ms8nTcNVqlZ3Y9/f3kcvloNfrkUwmWY8pGo3CarVyeXHY7zsAXjPosJTJZFCv17G5uYnd3V2Uy2U4HA5MT0/D5XJhZmaGM8dk6irWkpeHgnSaUiRj6KNKipzmZ3Bmgx7SmJFlGRaL5akLBm0QALiHh4QHKdU8OjoKWZbx1ltv4Uc/+hErz1qtVvEBegW63S5arRaq1Sr29vawvLwM4JGXFJVV6Evc4378fj9u3LiBSqUCq9XKo9KJRAIHBwdoNpvY3d3FwcEBDAYD7t27B71ez3V46ump1+vc3EwN+/Sjx+PB9evXEQwGMT8/j2vXrkGW5VOnnjoIEokEfvGLX2Bvbw8PHz5Es9mERqPhYCUcDmNxcRHXrl0DAORyOVa33tzc5AxRPB5nw+VKpQKdTgdZlmE0GjE1NYWPP/4YoVAIwWBQlGS+hwZNms0m7t27h6+++grlchn7+/s4ODiAxWLBhQsX8P7778PtdnOGkqYOh0Fm5DipVCpcWo/H41zmPU86amc26KFMjdFofGZ5hJo/1VADJzVeeTweOJ1OhMNhTExMsHnjMKf3Xwfqeny5XEY+n2eHZLVJ4LBvsE+DMj2NRgObm5twu91QFAWZTAaSJHGmh9jf3wfwdL2jp20AZLobCAQ4yxMOh0/leOkgqFariMVi2Nra4kwPaReR+avb7YbP50OtVmNX+mQyyf0Q29vb2NzcRLPZ5BKBTqeDxWKBTqdDq9XCwsICf89hVyQnut0uGo0GqtUqUqkUG+Fms1nk83lerycnJ+HxeDA9PQ232z3oyz430MQnBeo0SHGeOFNBj6IoXGunvoW9vT3s7Oywr9PTkCSJy1WyLGN2dpbrjtPT05wutVqt/PsEL4+iKMjn80gmkzg4OOAGcjJ79fl8cLlcIrB8BpSNlCQJfr+fF/ZSqcQBzrMmKtS/RhNC1GBrs9kwPj7OTfuLi4usRzXs73yv12P9HUrvFwoFllkgCxyy4Dg4OMCDBw84wCE/rt3dXdTrdaTTaVSrVdZLor+Dyu3pdBp37tzB/v4+Go0GSzdQuX6YoOGSXq+HeDyO27dvI5/PIxaLod1uw2g0YnZ2lqd1L1y4gEgkwqa3AsFROFNBT7fbRTqdRjweRy6Xw82bN7G2toZsNttnlngYmlQxm80IhUL45JNPcPHiRbhcLkxMTMBms8FkMsFsNouS1mug1+uxxUc6nUY6nUav14PBYEAoFML4+DhGRkaGbnF/UXQ6HaxWK0wmEyYmJtBqtZDNZpFMJrlMeLhPjX7tMOR9RtnMP//zP8fFixe5TGOxWE6VA/KgIG8z8u5LJpNIJpNcItfr9bDb7fD7/TCZTNjY2EAymUQ6nWY/LirpUl8KKV3Tc+n1ejwOHIvFUC6XYTKZUKlUMD09jU6nA4fDMXQl3263y/IMKysr+NnPfoa9vT2YzWbWZXvvvfdw7do1mM1m+Hw+7j0b9vdWcHTORNCjVpmt1WooFAosqU9WEs+qOZIMNrmmOxyOPiVgv99/KsfqTpLDKr7A0x2M1Tyt+VitS1Kr1ZDL5djElTybrFYr7HY7LBbL0GcXngbdQ9ostVotB+RHsTZQm8iazWbY7Xa4XC6EQiGMjY3BbDbD5XKJTeN7KCAhA0u1PhJleuieAuCJOAqQUqlU3/dTf2bUP6fvR30SBoOBM0rUwDtsqKe0SEAzmUyyOCzpInm9Xm76psMp+SUCT7qtD1Pg+KpQ5pg0pShgf1HoGdLw0OFJ6cMaVfRrg+BUBz20AVQqFZRKJVSrVXz99df45ptvUKlUsLa2hlQqhWazye7TatQihjdu3MBbb70Ft9uNN954AyMjI9xQO6z0ej3UajU0m00Ui0Wsrq4im80CAC8qTqcTXq+3L0DLdfcKAAAgAElEQVTR6/Xwer2w2Wx9BpTlchm5XA7VahVLS0u4desWSqUSWq0WwuEwgsEg3nrrLVy8eBE+n4/VtAWP7SY6nQ4KhQL29vZQr9exvb2NlZUVlMtlbGxscCbheQsSlcUooL98+TKmp6fh8XgwMzPD2QQRdD6G9IzUY/8U9ACPs8xA/wJeKpV4gadRaUmS2KtMq9VyoNPpdPjzQZsEBVcvs9GcF6i0SF6JlCWrVqvIZDKoVqu4efMmyuVyXx8nZUSpV1CWZR5QcblcPN07zGv8i0DNy4VCAXfv3sXq6iqSySQymcwLB+GFQgErKyuwWq0oFotYW1vru+8Oh4P3XZqaps8KfZHf5XFzaoMe9dRJsVjE7u4ucrkcfvvb3+LTTz9Fo9HgExktKIehkWhZlvGHf/iH+Ku/+ituUqYbPsyNtOTNVCwWsbOzg5///OdYW1vjcXKdToeJiQnMzc311c6tVivm5uYQCoWg1+u5ObNYLGJjYwPFYhF37tzB7373O7Tbbfh8PkSjUYyPj+MP/uAPcP36dWi1WlGPV0HNyY1GA1tbW7h58yZyuRzW1tawtLSEer2OSqXy3I2RTk46nQ6BQACLi4vwer344z/+Y7zzzjvcvK9ebASPKJfLuHfvHh48eIDd3V0O1olOp4ODg4O+wAd4LIQKgLNyOp2OLXDUQQ89Xwp6yPeMAp/DmaVhQd28TIG/2mNLq9WiVCrh/v37fe8tlbrMZjM7qpP3FvXFUSAkeDb07m9vb2N1dRX379/nBv4XbWKmiotWq8Xt27efcDIYGxvDX//1X+Pq1auwWCxwu90wGo19Ay0UzB43pyrooeCFFgRSoM1ms0in08jn86wfQP/vacHOYWhaRSj/PoKCyVarxQ2YNB2RzWY54NFqtbDb7chkMn0BSq1Wg8vl4pq61WqFTqfj/p1iscibRqfT4UWJvFl+yB18mKB0MpVtK5UKP49sNsvvO2n0qAMeytTQr6l7egwGA+x2O3+RMa/QRHo63W4X1WqVN9qnnXDJgoLeXbqflOk0GAwcVJKNjU6n48+bwWBAIpFgnzR1AEvP7ofk/s8j6lM+lV2p/Ef3Q6PR8DOhHyVJYlkGvV7PZUJyvDcajWg2m+zBRc9LnZ0Wn4XHVkH1eh2NRoMtP47yLqrf5VarxaLAhCzLyGQySKVSsFgs6HQ6LNZJz4X2Z3oux/VsTlUE0Gg0cHBwgEqlwuOK5XIZiUQCsVgM1WoVGxsbaDQaLPMuOBqKovD0W6FQwK9//WvcvXuX3elpLJoW9lwuh+3t7b5SCBktkrAaOUurbRDS6TSb0F29ehXXrl2Dx+NBKBQSwc73tNttJBIJZLNZZDIZ3Lp1C/v7+8hkMtja2kK1WmUDXQqOgMeyC1qtlvuw6Isc06mU5fF44Ha7RcDzAzQaDSSTSWxubqJUKj1xwqWSFU0MjY6OwuVywW63Y2xsjINK+rLZbHA6ndBoNPxsEokEPy8qKb/Ioe28o9fr4Xa7Icsy3n77bRiNRpTLZZRKJeTz+T4/qF6vxxl+CjQ1Gg1PzvV6PfZ/IlNju93OYpxWqxUejwdTU1P8zIY9E6SuqqjXkaMG4PT7nzZZmk6n8emnn+Lu3bsc3FJZknzrrly5grfeegtGo5EHOY6DUxX0tFotHBwcIJvNYn19Hb/97W+RTqexv7/PsvsvYn4meDaKonCT8cHBAW7evIl/+Id/QLPZRKlUQqPR6NsYE4nEM7VeqEGcUpWVSgX5fB7dbpcnhpxOJxYWFvDRRx+xQaPYeB/R6XSQSqWws7ODvb09/PrXv8b6+jpnfZ7mEQc8blKmwIdOZXQaliQJDocDY2NjcLvd7GIv7vuz6XQ6yGQySCQSnKFUQw3hdD8vXLiAsbExBAIBXL16FX6/n9P0kiTBYDCw7AAFOpubm7h79y729/dRKpX6tJaGGZ1OB4fDAeBR78fExAQ6nQ6SySR2d3fRaDQ4E02KwbVajbPV3W4XpVIJa2trqFarXCXQarXw+/0cnM7MzMDn82F0dJSbounvH/bPBgU+T1NvP8r3UH8vNYVCAV999RUfeOl+0zARBaDj4+Ow2WxsSn0cDDzoURSFU2rZbBbxeBwHBwdIJBLI5XIs7EWZHaPRyDdDPXVUr9eHsgnwqFAqs1QqcSqfmgeBR2URi8XCTsW0gFDvAaUxqQ9Bo9Gg0Wj06ZxQRoIWE5rmAh6lOYcdknYnB29Smy2VSn0n2sOaO6ThYjAY4PV6IcsyZwyazWZfvwj9OgnhCZ4PBZJ6vZ4DRwB8v/V6PUZGRjirE41GEQgE4PV6+VSqbsqkvkH63r1ej5tw6XtSMzT9HvWPw4q6XEjTtrTe63Q6LpfTmkOfFb1ez2tQvV5HrVaDJEk8KWowGFCpVFgvbGNjA4VCgZWcqTeRSu/D/hyOg6cNYLTbbdTrdej1etRqNT4I2O32Y7uOgQc9pFlxcHCAWCyGv//7v8f6+jqq1SrS6XSfu7ROp0M0GkU4HIYkSVzmKpVKiMfj4uT0AvR6PaRSKaysrCCdTiOZTKJcLrP4ml6vx9zcHD7++GO43W5kMhmW00+n0+zpFI/HUSgUWMGTFnYKePR6PTempVIpNgg0mUywWCxDu7D0ej3uUcvlcvj8889x+/ZtlMtl7O7uolAo8CSPGp1OB6/XC6/XC7fbjQ8++ADT09PI5XJYWlri5/TgwQO0Wi0kk0ncuXMHPp8PgUAAk5OTQ3m/XxS9Xg+Xy4VAIMDS+51OB06nkz22PvnkE7z//vswGo1wOp2wWCxc3qWsDqEu4ap7gGw2G1wuF2v/tNttDn6ERtjjoIcmRK1Wa98hixrHKXtGB99ms8mN/vl8nrNCpVIJlUoF5XIZKysr3FNy8+ZNmM1mRKNRzM7OQpZlzM/PY25ujjOow/4sTgKSKVAUBfF4HOvr69z7eVxK2wMPerrdLvL5PPft3Lt3D8vLy31TEQQ1O1HQQ6lMrVaLg4ODAf0LzhYkAZBKpdhFt9lscjreZDJhZGQEV69eRSQSQSwWg9PpRK1Wg8lkgkaj4WZb4PG4qRr1CKJGo0G1WsXBwQG63S5njej3DRtUXsxkMkin09ja2sLy8nJfxuZZdhJWqxVerxehUAiXL1/G1atXkUgk0Ol0sLe3h1arBZ1Ox5m8eDyOdrvN2R/Bs6HMgt1uR7PZ5HfTZDLB4/HA4/FgcXER77///pH0koD+zwPpzNDkCgVEh3V9hhW6T8CjKdGjaqiR2nUqlUKj0WCPuv39fRSLRWxubkKr1WJjYwNarRazs7PodDpwu93c66MuUwqOF9oTJElCuVxmIdvDe8rrZCBBT6/X40mfSqWCpaUl7OzscLbmcBqMtGLIbG5hYQG9Xo+nXfR6Pba2tgbxTzk3kBkiSb07HA52k6aNk17KZ9XASdmXTrnVahW9Xg+xWAzdbpe1M5LJJKf6aVSRAioKvtTGsOcBEr5rNpvY2NjA0tIS91VReZcCE7WIF4kKms1mzMzMYGJiAl6vl5WBSbjQarUeeTMWPIb0vOx2O2q1GpemqAxpNBqfKDs+716reyRoAm9vb49NY7vdLiKRCAAgHA7D4XDwMxS8GuRnZjAY4PP5WKqBdNrICJmqBxQQUSZalmWEw2F4vd5B/1POPTqdDiaTCTabDV6vF9FoFE6n81gFgwcS9LTbbdy7dw+ff/45ixrFYjE0Gg1kMpm+DI9Go8H4+DjeffddOJ1OXL16FZcuXUK73cbu7i6y2SxrCySTyUH8c84FZBHh9/vZJmJkZARerxeTk5OoVCqo1WrY3d3ljAKhPhW7XC5otVrU63UcHBxAkiR2A7darbh16xY8Hg8sFgs8Hg//Gb/fD6PRCI/HA5fLBZPJBK/Xyw2OZ30zJ62pUqmEzz77DJ9++imq1Sry+Tz3GlBpUKvVcs/H9PQ03n77bTgcDly+fBmLi4vcPG61Wlk+oNFosFik4OhQn1QkEuHsMSknp1IpdDodlMtl7i38ofKHWmH4/v37WF9fx/7+Pm7fvo2VlRVEIhFcv34dfr8fV65cQTQahd1uF2WVV0SSJLYV6vV6CAQCaLfbaDQauHDhAkqlEpLJJJaWlpDP59nrq9vtIh6PY2VlBV6vF3/2Z38Gj8cjnsUxQxN2DocDc3NzeOedd14qw3cUBpbpyefz2N7eRj6f5+kVUigF+oXW7HY7b8DRaBSjo6OsJSBJEpLJ5DMXeyFJ/iRPuxc0nULjnuRVZjQaeREhPZ7DqV91z4LaRZomK8rlMhRFgcViQa/XQyaTgc1mQ6lUgtlsZrNYdban2+3CbrdzU+lZTjdTEziN4O7v72Nzc5M1Rg4391FTrcFggNPpxMjICNxuN8bHxzE5Odn3DCi7phbcFBwdMmZVl56AR1Nd1BxLysm9Xu8J2QV1+ZA0eGgIgJ55MplEPp9HsVhEKBRi0U6fz8f9QYJXhyYb1XQ6HVgsFtbxyWaz3G9IPVy0/lBD7bCJRAJ44r1+3RzeN6ipn6xyvF4vzGbzsV7DiQY99Xod1WoVlUoF+/v72NvbYz0G8lAhIzmPx8Pja4uLi7h06RJkWYbD4UC73Ua5XMbW1hY2NzexsbHxVK8Pj8eDQCCAQCAgRqVVULqXAhQA7C6t1+uRSqUQj8dZGJK0k27duoXNzU3O+lCP1cTEBHuajY6Owmg0si9aq9XC/v4+8vk8b/DZbJa9i/R6PZLJJOLxOJs60sTG6OgofD4frFYrIpEIZFnm6bKzkNHodDq8Wa6treE3v/kNyzFQOUutwEtGkyMjI7h06RIcDgdmZmZw8eJFyLLcZwdCRpjFYpGnHal582VHTocZaqA1mUx9JVoKesrlMh48eIBf/vKXkGUZoVAILpeLew97vR5n7lqtFju11+t1rK+v86FucnISY2NjGB8fxxtvvIFgMIhgMDi0oqmHvebUoqWv87BK5XKNRgOfz4eFhQWUy2V4PB54vV62vEgmk+h0Otjf30cqleorlZ1XDAYDAoEAi6T6fD4Wfnxd/YDUK0oChPRFrStk5nsS6/qJftIqlQri8Tg3lK2vr/MG2mq1YDKZ4HA4YLFYcPHiRfzpn/4p/H4/wuEwRkdHeZy00Wjw1MqtW7eQzWZRLBb7/i6tVotQKIQrV67A7/cjEAiIoAePo2tZlvt6F9rtNrLZLJrNJlwuF9bW1pDL5fDgwQP20Nrc3MTOzg6fdkl59qOPPsLk5CSi0SguXboEi8XCQU+5XMatW7fw4MEDlEolPHjw4AlzRnUPC1mHmEwmTE1NIRwOIxQK4YMPPsD4+DjsdjuXfk47rVYLuVwO9Xodt2/fxt/93d8hnU4/0cMD9Je0ZmZm8Jd/+ZeIRCLw+XyIRCKc0aGxXVJPzWQyiMVi2NraQjKZ7POLErw4tClSxoXWCpoaajab+PLLL3FwcACPx4N33nkHU1NTLMdAQpOrq6uoVCrY29tDLBbj6dN2u41gMIif/vSnuHDhAgKBAC5evAin0zm0LvcU6JArPWXQ1BZBr+tzTqr8dNjz+/18aPjwww9RKpXwt3/7t/jmm2+Qz+extbWFnZ0dlic4z8/HbDZjbGwMDoeDne4NBgOy2SzLkLwqsizjzTffxNTUFB9uqTIgyzJMJhPGx8dPRLj2RIMeGj0kyet6vc4nVrXcOGUiXC4X93WQsVy5XEatVkOtVuNyAdXa1dBLTk25InX8GCqdmEwm1g3RaDRc+65UKigUCtBoNCwOSR5dZOxK4+h2ux0+n48Dy0AgwBsHlcZ8Ph+8Xi9PyFDwSo2h6hMdNX8ajUb+YGi1WhQKBb4mUhg+7VYWavdoygLQ1NthqKRFG6/L5YLP5+MmZuoxoU24Wq1y1rRSqaBarbJNhTqAPMtlwZOEgh5ys6d7p25ILpVK7EmUzWbhdrv5WZC44cHBAcrlMpLJJE/Wqb+X3W5HMBhkpeBhN92l+0uWQhqNps/LjD7nr+NdprWCSir0d5DIKjWr06GEjE/P+yGC9spOp8OtDLS+v8z9psCVfk6Tp5TVUXuikYQJDbOcxFp1okEPjaep/T2oTn6YRqPB5n7Ao01WURQsLy9jfX0dmUwGd+/e5QZo6gWitJnJZEIkEsHly5fhdrvh9XrF4g+wSuni4iKy2Sx2dnb4mdCm/PDhQ3S7XZjNZtZPovJjIBCAzWbDwsICotEogsEgrl27hkAgwCJfkiRxg7LFYsFbb72FcDiMcrmMubk5ZDIZPk3RZAzVcWmqD3jU/EvZjEajAZfLhenpady4cQNutxtOpxNut/tMZH1+CDJKdLvdGBsbg9FoRLfbRS6X49IgBTe1Wg1bW1usm3Tv3j1ks1lUq1XeHJxOJ0KhEE89inf/+VgsFszOzsLv9wMAlpaWIEkS93eQtAZNnjabTXz33XdsKaEoCgqFAmuLUcler9djcnISIyMjCIfDWFxcxOTk5ND38JC/ImUtk8kkarUaqtUqcrkc2u0291gZDAZEIhGEQiH2zXpdDd90+KLDN/DoAEICrVar9dxMkD4L0qiyWq2Ix+NwuVycVDjqPaY9IhwOw2g0wufzsTL/lStXEA6H+0yqSRBUq9XC7Xafv0yP+gWjr2epxTYaDaRSKZ6UIJOyb7/9Fp9//jlKpRLW19d5rp+icSqNmM1mhMNhXLp0iTdHsfA/irz9fj/cbjfy+TxPY6XTaXz11VcsmrexsQHgsYqmVqtlBdpgMIif/OQnePvtt2G1WtndmPQtgEcpU5PJBEVR4PP5OODd399HuVzGzs4OvvjiC6TTaciyzP0Rq6urXMIkSQKtVov19XVotVpcu3YNTqcTkUgE0WgUDofjXAQ9JpOJS4TRaJSzYYVCgTNs5PJdLBZx+/ZtxGIxNJtNlMvlPi8iqpUHg0Ge8hI8H6vVitnZWe5rCIfDXPKl8hU9C41Gg62trSfWE7V3EX0WdDodpqam2KpiYWGBhSJPc5byuKEsKEkCxONxZLNZpFIpPHz4ENVqFbIss6Ly9evXuaePMj+v2misvgYauqCMq9VqhcvlOjHn70FC3meKorBtB6nFH/UdpX3i0qVLsNvtmJubQzQahSzL7FdHelWH1chPKit9ok+TFgFS3NTr9X1+Wur6br1eRy6XY9lwg8HAXkXFYhGVSoWVmvv+Qd+/sPRFm+95f3GPgjqlSx3z3W4XDofjCVVrKi/pdDoEAgH4/X4Omsgojjbaw6iNSylVKssyJEniMXX6NYfDwadlsscAHjc6kuIzBbdnYSKP3mV1NvNZ16x2OqYySr1eZ2PYVquFVCqFTCbDzf+kQEv9EFQWNplM7K5OZrCn/V4NGmok12q1rBlCwxFkcEkHtacJpwKPs8yU6aR+BZ/P1yfFcB6C9FdFvdFRP5tadwoAZ1gURUG5XEYqlYJer2ejYwryKbupFntUCz5KktTXLE2BDpUpk8kkisUiWq0WbDYbl3jUz/O8Q/eJ1MZrtRp7Kar7BKnMe1g8UD1FSqV5WZbhdrvhdru5l4qeG1UEBsGJRgJms5n1WLxeLzweD6fRqGEqn8+jVCpxNz31OMiyjF6vh729Pe6wp41RjcvlwoULF+ByuTA5OQm3233uu++PCi0MZrMZc3NzcDgcSKfTMJvNbDBKC7PdbofT6eQUczAYhMViwfj4OHw+35FcislY0GKxsK9OvV7vq7FfvnyZG6Dv3LmD9fV1NJtNFAoFNJtNhEIh3kAou3RaoUZmCuSe1xtAIp3b29usZ0Sj/yTTTj5p9H1psoJOpx6PB5OTk5BlGVevXsV7773HU1/DsHC/Cuq+kdHRUfz0pz9FoVDA5uYmVlZWUKlUsL6+ju3tbW7kP4zdbkckEuHPx9TUFE+fTkxMcK+W4PFABR2qZmdn2UpiZmaGFcqz2Sw6nQ7u37/PhpUU9NhsNoTDYZbZoCEHOvBSUESHa+p7KxaLiMVi/Ezv3LnDe8n169fh8XjYY40GK4aFYDCIjz76iA9bdP9JnyqdTuOzzz7DvXv3+v6cy+Xiaet3330XH3zwASvIq73NyFdtkOvRiQY95FOj0WjgcrngcDggSRL3lJAAGPDIlXVvb++p3+d5I3Q2mw3RaBR+v589c47LrfUsQ1E96YRQn0IgEODMDI13BoNBmEwmRKNRhEKhl/YJogULeKSyHQ6Hn/g9lBkpFot8+qtUKizsR47JNpvtCb+j0wa9z6QD8rygp16vY2dnB8Dj7Bq5pqsb/Q9/D1qUtVotb7putxvT09OYn5+HxWI5vn/gOYNOu36/H9evX0ez2YTf74fBYEA+n0ehUMDu7i5nHw+vQ7QJOxwOXLlyBdevX4fNZkMkEhHTo0+BssM0Fg48OiiMj4+j3W5jZ2cHDx8+RLFYxN27d/Hdd98BAGdi3G43FhYW4HK5uJxrMplYnoTWMK1Wy0FPrVZDMpnE8vIycrkcvvvuO/zmN79Bu93GtWvXcO3aNR7MIPPRYcLpdOLSpUucQaYpU1qXY7EY7t+//0TQY7PZMDY2xs/k8uXLnNk5bQmHEy9vkc0ANVpaLBZON9J0Ci3szwtuKB1NqTJqOAuFQggEAvB4PLBarUP30h4VytRYLBaWXaf0sFar7Zt+o9T86yotPe17UMBFCrnRaJR9v6rVKkZHR3nEcZAp0heBDCZpKoKE76js9Sxook09OaSGRj4pBW+z2aDX6zE6OsqaSVQ7FxwdWqckSWJxSFmWkc/nATzu3Tm8PgWDQc60hUKhPofv0/yenibUkz9WqxVutxsGgwETExO8N5CWEjXDUh8WCQ6Wy2Xkcrm+dYx6sihrTFphGo0GwWAQiqIgEonwoWFYm/+fJhlA8iQkljo7O4tcLtf358LhMKampuBwOODxeLhUfBoz8Sca9FD91Wg0YmFhAbVaDYVCAbdu3YJWq+UJohcxG6PF3mKxwO12Y25uDi6XC9FoFG+88QZkWUYkEhEL/3MgFVpyRDebzX2K2BRYUh2WFEuPczGgD4rdbseVK1cwPT3Njc3dbpdH4KnB8DR+qAgqc9TrdWxubiIYDEKj0aBYLD6zRAKgrxlW/Xso4JRlGZcvX+Z0MjUK+nw+jI+Pw2w2w+12n/upk+OCgsperweDwYCRkRG0223cuHGDlXqfdiCjkgtJOZCtxDBPaR0Vul90351OJzqdDi5cuMD3ng4CJFBbKpVQKBSwvr7OYpJUOSgWi/znqLeOnpNer4fNZsMnn3zC3lxXrlzhz88wQto56h5b4LG0gN1ux9/8zd/gL/7iL/r+HCkq63Q6dkkXQQ8ep+KpbDIxMYFcLoft7W22kjjcSPssqDxDo4VjY2MIBAI8Gmqz2WC320/lTT9N0DOhWvigoWCLVEIDgcCgL+mlMRgM3MPkdDphs9lQqVRYC+lZ0Ib6tOZnymyGQiFMTU3B5XJhZmaGU/wjIyOnLp181qBMD/DYZV1wMqj7CWVZhizLAMClcGr4JzHI/f195HI5VCoVJJNJtv3IZrOsBn84K+F0OjE5OclmvpQdnZmZwdTU1FAPvVBZ8FlYrdYz/3kYyNMl92hKAVerVfj9fn5hXyTTQz0MJEA4OzvLo+nklivSyoJBol5AfD4f5ufn4fF4uE9BURSetqJTp9FoRKfTYd0QmoqgqSKbzQaHw4E33ngDExMT3CxIWU8R5AvOO1SCsdlsXO6uVqsIh8PsoF4ul3kw5vBB2mazwe/3w2KxIBAIYGxsjA8mYr84/wwk6NFqtQgGg3C5XOh2u7h48SKPgj4v7a9GPW2h0+lYtZY2iZdtthUIXhdq3ZypqSn8+Mc/5l6DWCwGRVFYQNBut3PmhiYX2+02ZyyNRiPGxsZYEykcDrMwo/p9F+VcwXmGSu46nY59+Ug1mX6kfYSMX9WDAAD6PjPUY0rle/H5Of8MLI9Hst8AxBin4NxCwbnVakUgEIBer+dMpKIorFHkcrkQDAbh9XpRLpeh0WjQarU4BU8GrGNjY1xyGXYLA8Fwoi6Bi1Ku4KgMb/FSIDhBLBYLSyh89NFHrIDqdrshyzL/f6vVyirLnU6HRe70ej28Xi8bVIomZYFAIDg6IugRCE4ACmwURcHY2Bh+8pOfQFGUPiNFmkZTq8eq7QqofHUW1KgFAoHgNCKCHoHgBFD7komUvEAgEAwGMeohEAgEAoFgKBBBj0AgEAgEgqFABD0CgUAgEAiGAhH0CAQCgUAgGApE0CMQCAQCgWAoEEGPQCAQCASCoUB6mlvwM3+zJKUB7Bzf5ZxLxhRF8b3qNxH3/qUR93+wiPs/OMS9Hyzi/g+Wp97/IwU9AoFAIBAIBGcVUd4SCAQCgUAwFIigRyAQCAQCwVBw5oMeSZL+tSRJ9yVJWpIk6X9JkmQa9DUNC5IkRSVJ+pUkSQ++fwb/ctDXNEyId3+wSJLklCTpf0uStCJJ0rIkSe8O+pqGBUmSTJIk3ZQk6bvvPwP/ftDXNEyc5Xf/TAc9kiSFAfwLAFcVRbkIQAvgHw/2qoaKDoB/oyjKAoB3APxzSZIWBnxNQ4F4908F/xnA/1MUZR7AmwCWB3w9w0QTwMeKorwJ4DKAP5Ek6Z0BX9MwcWbf/fNgOKoDYJYkqQ3AAiAx4OsZGhRF2Qew//3Py5IkLQMIA3gw0AsbHsS7PyAkSXIA+BDAPwEARVFaAFqDvKZhQnk0gVP5/j/133+JqZwT4Ky/+2c606MoShzAfwQQw6PNt6goyqeDvarhRJKkcQBXAHw12CsZDsS7P3AmAKQB/HdJkm5LkvRfJUmyDvqihglJkrSSJN0BkALwC0VRxNpzMpzpd/9MBz2SJLkA/DkePYQRAFZJkv56sFc1fEiSZAPwMwD/SlGU0qCvZxgQ7/7A0QF4C8B/URTlCoAqgH872EsaLhRF6SqKchlABMB1SZIuDvqahoQz/e6f6aAHwCcAthRFSSuK0gbwfwC8N+BrGiokSdLjUcDzPxVF+T+Dvp4hQrz7g2UPwJ4qu/C/8acFGEEAAAEsSURBVGgjEJwwiqIUAPwKwJ8M+lqGhDP97p/1oCcG4B1JkiySJEkAfoQz1FB11vn+nv83AMuKovynQV/PkCHe/QGiKEoSwK4kSXPf/9KPIHrZTgxJknySJDm//7kZwI8BrAz2qoaDs/7un3lF5u9HFf8RHk0S3QbwzxRFaQ72qoYDSZLeB/A5gHsAet//8r9TFOX/Du6qhgfx7g8WSZIuA/ivAAwANgH8U0VR8oO9quFAkqQ3APwPPJpa1AD4O0VR/sNgr2p4OMvv/pkPegQCgUAgEAhehLNe3hIIBAKBQCB4IUTQIxAIBAKBYCgQQY9AIBAIBIKhQAQ9AoFAIBAIhgIR9AgEAoFAIBgKRNAjEAgEAoFgKBBBj0AgEAgEgqFABD0CgUAgEAiGgv8PKGl6iG2A+kQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x720 with 36 Axes>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.figure(figsize = (10,10))\n",
        "random_inds = np.random.choice(60000,36)\n",
        "for i in range(36):\n",
        "  plt.subplot(6,6,1+i)\n",
        "  plt.xticks([])\n",
        "  plt.yticks([])\n",
        "  plt.grid(False)\n",
        "  image_ind = random_inds[i]\n",
        "  plt.imshow(np.squeeze(train_images[image_ind]), cmap=plt.cm.binary)\n",
        "  plt.xlabel(train_labels[image_ind])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a71z8Ygll21i"
      },
      "source": [
        "## Note: I have used two models here.\n",
        "### 1. Fully connected neural network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IdOVVXKpl21j"
      },
      "source": [
        "#### Approach: \n",
        "1. We first flatten the images data to a 1-D vector\n",
        "2. We use one hidden layer which has 128 nodes with non linearity function : ReLU\n",
        "3. We use a output layer which has 10 outputs[0-9] which will have the probability of each possibility for each input using function : softmax\n",
        "4. After neural network architecture we give optimizer Stochastic gradient descent(SGD) because this optimizer approaches the local minima much fasster than ADAM as its locally unstable\n",
        "5. We use loss = 'sparse_categorical_crossentropy' because we have want the digit only to belong one particulat class"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "bCptcoGF06dY"
      },
      "outputs": [],
      "source": [
        "def build_fc_model():\n",
        "  fc_model = tf.keras.Sequential([\n",
        "                                  tf.keras.layers.Flatten(),\n",
        "                                  tf.keras.layers.Dense(128,activation = 'relu'),\n",
        "                                  tf.keras.layers.Dense(10,activation = 'softmax')\n",
        "  ]) \n",
        "  return fc_model\n",
        "model = build_fc_model()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "-QvXHCHhv5Vt"
      },
      "outputs": [],
      "source": [
        "model.compile(optimizer = tf.keras.optimizers.SGD(learning_rate = 1e-1),\n",
        "              loss = 'sparse_categorical_crossentropy',\n",
        "              metrics = ['accuracy'])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NjFpPP-4l21l"
      },
      "source": [
        "#### We train the model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JCLYKQolxDMy",
        "outputId": "fe1de74e-1df9-48ad-e850-6d936d56b274"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3765 - accuracy: 0.8957\n",
            "Epoch 2/5\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.2017 - accuracy: 0.9430\n",
            "Epoch 3/5\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.1521 - accuracy: 0.9567\n",
            "Epoch 4/5\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.1230 - accuracy: 0.9651\n",
            "Epoch 5/5\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.1048 - accuracy: 0.9699\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fa1d2ae3090>"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ],
      "source": [
        "model.fit(train_images,train_labels,64,5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sye8zD_9l21m"
      },
      "source": [
        "#### We take the test accuracy using model.evaluate function"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZlBoevuqxecU",
        "outputId": "de67af29-f8c5-434c-a6f5-c1741dc670a4"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 1s 2ms/step - loss: 0.1111 - accuracy: 0.9662\n",
            "0.9661999940872192\n"
          ]
        }
      ],
      "source": [
        "test_loss, test_acc = model.evaluate(test_images, test_labels)\n",
        "print(test_acc)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9IA2MLtil21n"
      },
      "source": [
        "#### Note: However we can see that the training accuracy is greater than testing set and is quite efficient however I would like to see if we can increase its efficiency even more using Convolutional neural network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lq_oDRlJl21n"
      },
      "source": [
        "### 2. Convolutional Neural Network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Xo20JApFl21n"
      },
      "source": [
        "#### Approach: \n",
        "1. We use two convolution and pooling operations\n",
        "2. Convolution: To get the feature map using the filters, Pooling: To reduce the dimension of feature map\n",
        "3. We use keras.layers.Flatten() to convert the feature map to a vector input so we could use it for training network\n",
        "4. We use one hidden layer which has 128 nodes with non linearity function : ReLU\n",
        "5. We use a output layer which has 10 outputs[0-9] which will have the probability of each possibility for each input\n",
        "6. We used optimizer SGD above but why ADAM here? \n",
        "7. Reason : Basically the reason for CNN approach is that we want to find the feature for each part of a image. So if the pixels are closer to each we would want them to have a same feature set however SGD is not good at that as it gives different results even there is small noise.\n",
        "8. We use loss = 'sparse_categorical_crossentropy' because we have want the digit only to belong one particulat class"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "jNr9SqVax_yi"
      },
      "outputs": [],
      "source": [
        "def build_cnn_models():\n",
        "  cnn_model = tf.keras.Sequential([\n",
        "                                   tf.keras.layers.Conv2D(filters = 24,kernel_size = (3,3), activation = 'relu' ),\n",
        "                                   tf.keras.layers.MaxPool2D(pool_size = (2,2)),\n",
        "                                   tf.keras.layers.Conv2D(filters = 36, kernel_size = (3,3), activation = 'relu'),\n",
        "                                   tf.keras.layers.MaxPool2D(pool_size = (2,2)),\n",
        "                                   tf.keras.layers.Flatten(),\n",
        "                                   tf.keras.layers.Dense(128, activation = 'relu'),\n",
        "                                   tf.keras.layers.Dense(10,activation = 'softmax')\n",
        "\n",
        "  ])\n",
        "  return cnn_model\n",
        "model1 = build_cnn_models()\n",
        "        "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "s0b-a3gm2kOE",
        "outputId": "a5b51634-e093-49e3-b132-217b47173b16"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.09153599, 0.0974744 , 0.08563238, 0.09499917, 0.08801459,\n",
              "        0.09992997, 0.13172424, 0.10569081, 0.11593387, 0.08906455]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ],
      "source": [
        "model1.predict(train_images[[1]])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "r762FxOo2wNi",
        "outputId": "9a0f7600-2f5c-46c8-e278-8e6485a07d06"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 26, 26, 24)        240       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 13, 13, 24)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 11, 11, 36)        7812      \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 5, 5, 36)         0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 900)               0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 128)               115328    \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 10)                1290      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 124,670\n",
            "Trainable params: 124,670\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ]
        }
      ],
      "source": [
        "print(model1.summary())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "VZWXaQB_20ju"
      },
      "outputs": [],
      "source": [
        "model1.compile(optimizer=tf.keras.optimizers.Adam(learning_rate = 1e-3),\n",
        "              loss = 'sparse_categorical_crossentropy',\n",
        "              metrics = ['accuracy'])\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ker65bWVl21q"
      },
      "source": [
        "### Training the model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zWMGpw_T3H7y",
        "outputId": "75c0504f-e35f-4823-b3a3-ca8dc6ad34af"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n",
            "938/938 [==============================] - 35s 37ms/step - loss: 0.1834 - accuracy: 0.9453\n",
            "Epoch 2/5\n",
            "938/938 [==============================] - 35s 37ms/step - loss: 0.0567 - accuracy: 0.9823\n",
            "Epoch 3/5\n",
            "938/938 [==============================] - 34s 36ms/step - loss: 0.0401 - accuracy: 0.9872\n",
            "Epoch 4/5\n",
            "938/938 [==============================] - 34s 36ms/step - loss: 0.0312 - accuracy: 0.9905\n",
            "Epoch 5/5\n",
            "938/938 [==============================] - 34s 36ms/step - loss: 0.0233 - accuracy: 0.9929\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fa1d2c8d850>"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ],
      "source": [
        "model1.fit(train_images, train_labels, 64,5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4tbaagYm3RgX",
        "outputId": "b76e3a95-8f8d-4683-8fbf-6e5df68aee5e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 3s 8ms/step - loss: 0.0342 - accuracy: 0.9903\n",
            "0.9902999997138977\n"
          ]
        }
      ],
      "source": [
        "test_loss , test_acc = model1.evaluate(test_images, test_labels)\n",
        "print(test_acc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qSb7nMZe7v-R",
        "outputId": "ec93e55a-4d92-46c6-c2c2-de07caf448ae"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9.9129238e-09, 1.1098036e-07, 1.4663065e-07, 2.4885549e-05,\n",
              "       1.2328418e-07, 2.0192454e-08, 1.3266777e-12, 9.9997294e-01,\n",
              "       5.0571121e-08, 1.7608695e-06], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "pred = model1.predict(test_images)\n",
        "pred[0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "62Yt5ljV76KF",
        "outputId": "602c6933-2fb9-4c1d-b6fc-226b21cbad6f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "5\n"
          ]
        }
      ],
      "source": [
        "print(np.argmax(pred[15]))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "iCG8DGaN8LEx",
        "outputId": "9b798c31-e558-4233-eeea-e004664c7040"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Label of digit is: 5\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fa1d2f53ed0>"
            ]
          },
          "metadata": {},
          "execution_count": 19
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAANpklEQVR4nO3dXYxc9XnH8d+vJLkwyQWwK8siUKcRkm1VqhMPVqUYiyoqAm5gjYRimciVEGskkGLJF+VNCpeoamL1osQsNYoLKVGQX/AF0FArAnITeWy5YPxSKLIdLONdyxch3KTYTy/2gDaw8z/rmTMv9vP9SKOZOc+cPY9G/vnMnP+c83dECMCV7y+G3QCAwSDsQBKEHUiCsANJEHYgia8McmNjY2OxdOnSQW4SSOXEiRM6d+6c56v1FHbbt0v6F0lXSfq3iHiq9PqlS5eq3W73skkABa1Wq2Ot64/xtq+S9K+S7pC0QtJ62yu6/XsA+quX7+yrJb0fER9ExJ8k/VLSXc20BaBpvYT9ekm/n/P8w2rZn7E9abttuz0zM9PD5gD0ou9H4yNiKiJaEdEaHx/v9+YAdNBL2E9LumHO829WywCMoF7Cvl/STba/Zftrkn4gaW8zbQFoWtdDbxHxqe2HJf2nZofenouIdxvrDECjehpnj4hXJL3SUC8A+oifywJJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQxECnbMbgHThwoFjfvXt3sb5z585i/fjx48V6RHSs2fPOLPy5VatWFevLly8v1h999NGu170SsWcHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQYZx+AqampYv3YsWPF+ltvvdX1tuvG2evGukvj5AtZf9OmTR1rExMTxXVvu+22Yh2Xpqew2z4h6WNJFyR9GhGtJpoC0Lwm9ux/FxHnGvg7APqI7+xAEr2GPST92vYB25PzvcD2pO227fbMzEyPmwPQrV7DviYivivpDkkP2V77xRdExFREtCKiNT4+3uPmAHSrp7BHxOnqflrSbkmrm2gKQPO6Drvtq21/47PHkm6TdLipxgA0q5ej8Ysl7a7GWb8i6T8i4rVGurrClMaapfqx6kWLFhXrpXOzN2/eXFx32bJlxfrY2Fixvm7dumIdo6PrsEfEB5L+psFeAPQRQ29AEoQdSIKwA0kQdiAJwg4kwSmuA1A3PLVnz55ive6yx/v377/knpAPe3YgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIJx9gHYtm1bsX7w4MFi/eTJk8X6qVOnOtZuvPHG4rrIgz07kARhB5Ig7EAShB1IgrADSRB2IAnCDiTBOPsA1M2E88ADDxTrTzzxRLF+7lzneTUZZ8dn2LMDSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKMs4+AixcvFusRUawfOXKk63V7VXdN+7rppjE4tXt228/ZnrZ9eM6ya22/bvu96v6a/rYJoFcL+Rj/c0m3f2HZI5L2RcRNkvZVzwGMsNqwR8Sbks5/YfFdknZUj3dIurvhvgA0rNsDdIsj4kz1+CNJizu90Pak7bbt9szMTJebA9Crno/Gx+wRoI5HgSJiKiJaEdGqOyEEQP90G/aztpdIUnU/3VxLAPqh27DvlbSxerxR0svNtAOgX2rH2W2/KOlWSWO2P5T0Y0lPSfqV7fslnZR0bz+bvNzVHavYvn17sW67WN+4cWPHWt04e93frlt/YmKiWN+wYUPHWt289WhWbdgjYn2H0vcb7gVAH/FzWSAJwg4kQdiBJAg7kARhB5LgFNcG1A2trV27tlivm5J51apVxXrpNNM1a9YU163z7LPPFut1003v2rWrY61u2G///v3FOqfXXhr27EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBOPsDTh27Fixfvz48WL9nnvuKdZfeumlS+6pKZOTk8V6abpoSXrhhRc61vbs2VNc9+abby7WV6xYUayX3re6MforEXt2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUjC/Z7Sd65WqxXtdntg28PlbWpqqlivO9e+dJ2AV199tbhu3TUERlWr1VK73Z73QgHs2YEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcbZcdmqO5e+dL3+8+fPF9d9+umni/VRnW66p3F228/ZnrZ9eM6yJ22ftn2out3ZZMMAmreQj/E/l3T7PMu3RsTK6vZKs20BaFpt2CPiTUnlzzwARl4vB+getv129TH/mk4vsj1pu227XTcnGoD+6TbsP5P0bUkrJZ2R9JNOL4yIqYhoRURrfHy8y80B6FVXYY+IsxFxISIuSnpW0upm2wLQtK7CbnvJnKcTkg53ei2A0VB73XjbL0q6VdKY7Q8l/VjSrbZXSgpJJyRt6mOPwLzGxsaK9W3btnWsbdmypbjugw8+WKyfOnWqWN+8eXOxPgy1YY+I9fMs3t6HXgD0ET+XBZIg7EAShB1IgrADSRB2IAlOcUVKvZweK9VPw33hwoVL7qkJXEoaAGEHsiDsQBKEHUiCsANJEHYgCcIOJFF71htwJao7PfaWW24p1o8dO9ZkOwPBnh1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkmCcHSkdPXq0WN+zZ0+xvmLFiibbGQj27EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBOPsA7B169ZifXx8vFi/7777mmwnjZMnT3asPf7448V1P/nkk2L9jTfe6KqnYards9u+wfZvbB+x/a7tH1XLr7X9uu33qvtr+t8ugG4t5GP8p5K2RMQKSX8r6SHbKyQ9ImlfRNwkaV/1HMCIqg17RJyJiIPV448lHZV0vaS7JO2oXrZD0t39ahJA7y7pAJ3tpZK+I+l3khZHxJmq9JGkxR3WmbTdtt2emZnpoVUAvVhw2G1/XdJOSZsj4g9zazE7O+S8M0RGxFREtCKiVXcgCkD/LCjstr+q2aD/IiJ2VYvP2l5S1ZdImu5PiwCaUDv0ZtuStks6GhE/nVPaK2mjpKeq+5f70uFlYNeuXcX6li1bivVNmzYV65fz0Fvpq9vu3bt7+tt16x88eLBjre5T5vPPP1+sL1u2rFgfRQsZZ/+epB9Kesf2oWrZY5oN+a9s3y/ppKR7+9MigCbUhj0ifitp3sndJX2/2XYA9As/lwWSIOxAEoQdSIKwA0kQdiAJTnEdgNkfGHb2zDPPFOs7d+4s1tetW9f1tuumHr7uuuuK9bpLLpe2P/sTju7WlaTly5cX6xs2bOhYe+yxx4rr1k3pfDlizw4kQdiBJAg7kARhB5Ig7EAShB1IgrADSTDO3oDSOLckvfbaa8V63Vh1ndJ53dPT5WuK1E09XDcWXncufmm8emJiorhunbpzyhctWtTT37/SsGcHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSRcd85wk1qtVrTb7YFtD8im1Wqp3W7P++MI9uxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kERt2G3fYPs3to/Yftf2j6rlT9o+bftQdbuz/+0C6NZCLl7xqaQtEXHQ9jckHbD9elXbGhH/3L/2ADRlIfOzn5F0pnr8se2jkq7vd2MAmnVJ39ltL5X0HUm/qxY9bPtt28/ZvqbDOpO227bbMzMzPTULoHsLDrvtr0vaKWlzRPxB0s8kfVvSSs3u+X8y33oRMRURrYhojY+PN9AygG4sKOy2v6rZoP8iInZJUkScjYgLEXFR0rOSVvevTQC9WsjReEvaLuloRPx0zvIlc142Ielw8+0BaMpCjsZ/T9IPJb1j+1C17DFJ622vlBSSTkgqX1MYwFAt5Gj8byXNd37sK823A6Bf+AUdkARhB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJEHYgiYFO2Wx7RtLJOYvGJJ0bWAOXZlR7G9W+JHrrVpO9/WVEzHv9t4GG/Usbt9sR0RpaAwWj2tuo9iXRW7cG1Rsf44EkCDuQxLDDPjXk7ZeMam+j2pdEb90aSG9D/c4OYHCGvWcHMCCEHUhiKGG3fbvt47bft/3IMHroxPYJ2+9U01C3h9zLc7anbR+es+xa26/bfq+6n3eOvSH1NhLTeBemGR/qezfs6c8H/p3d9lWS/kfS30v6UNJ+Sesj4shAG+nA9glJrYgY+g8wbK+V9EdJ/x4Rf10t+ydJ5yPiqeo/ymsi4h9HpLcnJf1x2NN4V7MVLZk7zbikuyX9g4b43hX6ulcDeN+GsWdfLen9iPggIv4k6ZeS7hpCHyMvIt6UdP4Li++StKN6vEOz/1gGrkNvIyEizkTEwerxx5I+m2Z8qO9doa+BGEbYr5f0+znPP9Rozfcekn5t+4DtyWE3M4/FEXGmevyRpMXDbGYetdN4D9IXphkfmfeum+nPe8UBui9bExHflXSHpIeqj6sjKWa/g43S2OmCpvEelHmmGf/cMN+7bqc/79Uwwn5a0g1znn+zWjYSIuJ0dT8tabdGbyrqs5/NoFvdTw+5n8+N0jTe800zrhF474Y5/fkwwr5f0k22v2X7a5J+IGnvEPr4EttXVwdOZPtqSbdp9Kai3itpY/V4o6SXh9jLnxmVabw7TTOuIb93Q5/+PCIGfpN0p2aPyP+vpMeH0UOHvv5K0n9Xt3eH3ZukFzX7se7/NHts435J10naJ+k9Sf8l6doR6u15Se9IeluzwVoypN7WaPYj+tuSDlW3O4f93hX6Gsj7xs9lgSQ4QAckQdiBJAg7kARhB5Ig7EAShB1IgrADSfw/BQNZyUulZ4kAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "print(\"Label of digit is:\", test_labels[15])\n",
        "plt.imshow(test_images[15,:,:,0], cmap=plt.cm.binary)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "id": "0KD-FdQE8SpM",
        "outputId": "56193535-705b-4dd9-d988-59eef96b6bb0"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "image_index = 45 \n",
        "plt.subplot(1,2,1)\n",
        "mdl.lab2.plot_image_prediction(image_index, pred ,test_labels, test_images)\n",
        "plt.subplot(1,2,2)\n",
        "mdl.lab2.plot_value_prediction(image_index, pred,  test_labels)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 589
        },
        "id": "AHd349Ws8hPh",
        "outputId": "16e09a30-2a73-4915-ab1e-11c5c40aedcb"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x720 with 40 Axes>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "nr = 5\n",
        "nc = 4\n",
        "ni = nr*nc\n",
        "plt.figure(figsize=(2*2*nc, 2*nr))\n",
        "for i in range(ni):\n",
        "  plt.subplot(nr, 2*nc, 2*i+1)\n",
        "  mdl.lab2.plot_image_prediction(i, pred, test_labels, test_images)\n",
        "  plt.subplot(nr, 2*nc, 2*i+2)\n",
        "  mdl.lab2.plot_value_prediction(i, pred, test_labels)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gNboYCVul21s",
        "outputId": "dfad129f-14ab-484d-ad46-fb02d5570b3c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting mitdeeplearning\n",
            "  Downloading mitdeeplearning-0.2.0.tar.gz (2.1 MB)\n",
            "\u001b[K     |████████████████████████████████| 2.1 MB 4.4 MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from mitdeeplearning) (1.19.5)\n",
            "Requirement already satisfied: regex in /usr/local/lib/python3.7/dist-packages (from mitdeeplearning) (2019.12.20)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.7/dist-packages (from mitdeeplearning) (4.62.3)\n",
            "Requirement already satisfied: gym in /usr/local/lib/python3.7/dist-packages (from mitdeeplearning) (0.17.3)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from gym->mitdeeplearning) (1.4.1)\n",
            "Requirement already satisfied: cloudpickle<1.7.0,>=1.2.0 in /usr/local/lib/python3.7/dist-packages (from gym->mitdeeplearning) (1.3.0)\n",
            "Requirement already satisfied: pyglet<=1.5.0,>=1.4.0 in /usr/local/lib/python3.7/dist-packages (from gym->mitdeeplearning) (1.5.0)\n",
            "Requirement already satisfied: future in /usr/local/lib/python3.7/dist-packages (from pyglet<=1.5.0,>=1.4.0->gym->mitdeeplearning) (0.16.0)\n",
            "Building wheels for collected packages: mitdeeplearning\n",
            "  Building wheel for mitdeeplearning (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for mitdeeplearning: filename=mitdeeplearning-0.2.0-py3-none-any.whl size=2115442 sha256=b8cc49f0f23412808e3eb4ef0b2b5e91213a58c16e64f94ce878566b9c634184\n",
            "  Stored in directory: /root/.cache/pip/wheels/9a/b9/4f/99b7c8c5c75355550b83e1fcfc02956fb40c35eb01e2262877\n",
            "Successfully built mitdeeplearning\n",
            "Installing collected packages: mitdeeplearning\n",
            "Successfully installed mitdeeplearning-0.2.0\n"
          ]
        }
      ],
      "source": [
        "!pip install mitdeeplearning\n",
        "import mitdeeplearning as mdl"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 558
        },
        "id": "OgP-PBIx8w6x",
        "outputId": "939c6be0-8fbe-42e2-e0fb-25c080f92ab7"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 5000/5000 [02:02<00:00, 40.78it/s]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEGCAYAAAB7DNKzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO2dd5wU5f3HP9/da1ylF2kHgjQFQUSJYBALCBL92U00iSWKRo1J1GCMmliJCSbGaJRYUGPsUVEQFKVYQHo76XD0crTrbXef3x8zz+4zszOzs3e7t3e73/frxYvd2SnP7M0+3+fbSQgBhmEYhrHDk+gBMAzDMM0bFhQMwzCMIywoGIZhGEdYUDAMwzCOsKBgGIZhHElL9ADiQfv27UVhYWGih8EwDNNiWLFixWEhRAerz5JKUBDRJACT+vTpg+XLlyd6OAzDMC0GItpp91lSmZ6EEB8LIW4uKChI9FAYhmGShqQSFAzDMEzsYUHBMAzDOMKCgmEYhnGEBQXDMAzjCAsKhmEYxhEWFAzDMIwjLCgUXvlmBz4rOpDoYTAMwzQrWFDoCCHwnyU7cfPrK/DR6r2JHg7DMEyzgQWFDhHh01+djQFd8nHve2sTPRyGYZhmAwsKhYw0D4Z0K0CtL4CjlXWJHg7DMEyzgAWFiXEndwYAbD1UkeCRMAzDNA9YUJjo0yEXALCthAUFwzAMkGSCgogmEdH00tLSBp+ja+tWaJXuZY2CYRhGJ6kERSyqx3o8hF7tc7CdNQqGYRgASSYoYkWHvEwcYWc2wzAMABYUlmw9VIG1e0rx2uLiRA+FYRgm4bCgsGDv8WoAwIMfFSV4JAzDMImHBYUFeZmhDrFCiASOhGEYJvGwoLBg+k+HB19X1PoSOBKGYZjEw4LCgpEntsM94/oBAGdoMwyT8rCgsGFAlzwAwPaSygSPhGEYJrGwoLChdXYGAOD6GcsSPBKGYZjEwoLCBtWhzTAMk8qwoLChe9vsRA+BYRimWcCCwoasdC+G9miNVuneRA+FYRgmobB9xYFVu44neggMwzAJhzUKB24a1QsAUFXHuRQMw6QuLCgcGNAlHwBwqKw2wSNhGIZJHCwoHOiYnwkAOFhWk+CRMAzDJA4WFA50ys8CABwqZ42CYZjUhQWFA53yNEHBGgXDMKkMCwoH8lulITPNwxoFwzApTbMPjyWiHADPAagDsEAI8UYTXhud8rNYo2AYJqVJiEZBRC8T0SEiWm/aPp6INhHRViKaom++FMB7QohfAPhRU4+1U34mCwqGYVKaRJmeZgAYr24gIi+AZwFcCGAggGuIaCCAbgB267v5m3CMAIB2OZlYsv1oU1+WYRim2ZAQQSGEWATAPPuOALBVCLFdCFEH4C0AFwPYA01YAAkY75yiAwCAon2lTX1phmGYZkFzcmZ3RUhzADQB0RXA/wBcRkT/AvCx3cFEdDMRLSei5SUlJTEb1MWnngAAOFLBDYwYhklNmpOgsEQIUSmEuF4IcauTI1sIMV0IMVwIMbxDhw4xu/4dY/sAAI5VsaBgGCY1aU6CYi+A7sr7bvq2hNJBz6XgMh4Mw6QqzUlQLAPQl4h6EVEGgKsBzIzmBEQ0iYiml5bGzp+Qn5WGjDQP9h6vjtk5GYZhWhKJCo99E8BiAP2IaA8R3SiE8AG4HcBcABsAvCOEKIrmvEKIj4UQNxcUFMRyrKjzBTDj22IIIWJ2XoZhmJZCQhLuhBDX2GyfDWB2Ew/HNXuOVXPnO4ZhUo7mZHpq9nApD4ZhUpGkEhTx8FEAwK/O7QsAKD5cGdPzMgzDtASSSlDEw0cBABMHdwEA/PbdNTE9L8MwTEsgqQRFvOiQm5noITAMwyQMFhQuaJOTkeghMAzDJAwWFAzDMIwjSSUo4uXMBoBrRnRHhzw2QTEMk3oklaCIlzMbAPKy0lFeUx/z8zIMwzR3kkpQxJO8zDTU1AdQ5wskeigMwzBNCgsKl2Rnakns1XVN3juJYRgmobCgcEl2hhcAUFXvS/BIGIZhmpakEhTxdGZneLWvat0e7nTHMExqkVSCIp7O7J1HtPIdN7++IubnZhiGac4klaCIJzmZCSm0yzAMk3BYULhkSPfWAMC5FAzDpBwsKFxyZu92AIASLjXOMEyKwYKCYRiGcSSpBEU8o54A4MZRvYJhsgzDMKlCUgmKeEY9AUBeVhqq6vzwB7h3NsMwqUNSCYp4k6tHPlXUctIdwzCpAwuKKMjLYkHBMEzqwYIiCnIz0wEAFTUsKBiGSR1YUERBblCj4HLjDMOkDiwookD6KMpZo2AYJoVgQREF0kdxoLQGt/93JY5W1gEAhBAQgiOhGIZJTpJKUMQ7j0LWe3rp6x34ZO1+/PPLrQCAqXM24vTH5sXlmgzDMIkmqQRFvPMopOlJ5lFU1/tQ7w/ghYXbcbiijrUKhmGSkqQSFPFGCorth7WS45W1fpRVhxzba7lXBcMwSQgLiijwesjwftXuYzhQVhN872eNgmGYJIQFRSMY3K01Zq3dH3zP+RUMwyQj3I2nEahCAtCc3Gef1CFBo2EYhokPrFFEyag+7W0/W7i5BDX1/iYcDcMwTPxhQREl903o7/j5pGe+bqKRMAzDNA0sKKKkbU6G4+c92mY30UgYhmGahqQSFPFOuAOA9rlaz2yzQChsp73v2S4nbtdmGIZJBEklKOKdcAcA6V4PiqdOxO1j+xi2v/iz09E6Ox3Hquridm2GYZhEkFSCoilpZzJBZWd4cbyqHh+s2os56/ejcMosXPzsNwkaHcMwTOxgQdFA2ukmKElGWuirnPyflQCANbuPWx5b7w/gWGXjNY9lxUfx2KzvG30ehmEYJ1hQNBCzRqEKikjc8+4aDH3k8wbXhjpUVgN/QODnLy/Fv7/agbIa7o/BMEz8YEHRQNrlmgSF14PO+Vmujv1ET9RbvvNY1Nc9UlGLEY9/gSfnbERlnZaz8cTsDVGfh2EYxi0sKBpIdkZasD8FoAmKSUO6hO0ntYbJr6/A3KIDAIBW6V4AwBXPL3a8Rp0vgMIpszB22oLgtipdOLywaDtO7d4aAPDm0t0NvxGGYZgIsKBoBF/+dkzwtcdDmHLhgLB9auoDAIA5RQdwy+srAABZGV5X5z+oFxzcXlJpuI6kZzvO2WAYJv6woGgEHfKMDm1zdVkAKK+tN/gi1u0pDWoUkkWbS1B8uNJ8KGp94eVAAoHQudTz+ANcuZZhmPjAgiLOXPC3RSgprw2+n/TPr5GVHvraD1fU4qcvL8WYvy4IO1aamQBg7R4tgsqnCISAIoA+Wr03lsNmGIYJwoIizhyvqseIx78wbMtMC2kCD31UZHusKijq/ZoJyx8IBLf5/CFBwa0wGIaJF1xmvJH8YeIALC8ORS+1yU7HsSrncNWK2lDfilnr9lvus35vKa6evkQ5RhMaqkZRr7zOb5Ue3cAZhmFcwhpFI7lpdG88f91pwff/u+0sAMDdF5xke8wOC38EAIMvY+mOo4bPvtpcguo6P95eFopw+njNvqBfpLLWF5aX8dRnm7C82HgehmGYaKGGJn01Z4YPHy6WL1+esOvX+vzI8HrQ677ZUR33/cPjkJ2hKXlzlSgpSW5mmkEbAbRqtkf1LO+8rDSs++O44GeFU2YBAIqnToz6HhiGSS2IaIUQYrjVZ0mlUTRF9Vg3ZKZ5QRQeARUJVQjkZIRbBc1CAgAKFJNTeZStWPeXVgd9HwzDMHYklaBoiuqx8aSyNuS8FnCn6Zl9E5/a+DzCr+XDyCe+xAMfrnc/QIZhUpKkEhTNjeKpE7HxkfFYcPcYV/tX1YU0gjveXOXqmL3Hqgzvb31DK0gYiJBXUalrJ/M2HHR1nVTn+YXbgqY8hkk1WFDEmax0Lwrb52BYj9YR962q82PQg3Pw70XbcTxC5JTkcIV1FVp/BN9TdQN7e3+4ai/eW7GnQce2ZKZ+uhGAdRIkwyQ7LCiaiNys8PDVp64cYni/dMdRVNb58ZhS5O/hiwehdXZ0oa9CCEOmtlXWtjRzqZ/V+wP4Zuthy3P+Z8lOPPPFFtz19mrc/e4afLnxIL7fVxa231dbSrBuT2J9RPFAZsFXROkHYpim4HhVHeZvPBS387OgaCL+evlg3DbmRMO2S4d1w1XDuwdDaf8yd1PYcTkZaejTIdf2vOf27xi2rbzWZ8jaVk1a5m3HqkIlRv72+Wb85MXvsGJneEjtHz5cj2mfbw6+v2HGckz4x1dh+1330lJM+ufXtuNtqWTr9bmOxKCPCMPEmtveWInrZyzDkYrayDs3ABYUTUTH/CzcO75/2PY/Xz4YPxrS1fa40Se1xyOXnGz7uZq9LXnooyKDpmAVDbVWWfXX+rTIp+IjWn7HvuM1+OPMImwvqbC9bqqRpWsUFz2TfEKQafns1n2VVpGRsYAFRTPghNbWfSzuGNsHHfOyMKBLPpbef67hs0cuHgTA2tfwwaq9UCp9oLzGh6fnbcH/PfdNUFt4+JNQZ7wa/RwZXu1xmLV2P2Z8W4yx0xY2/KZs2HSgHE99tqnBTZsShdQo6nzxDSfefbQKB0pr4noNJvnI0ssCyWrVscaVoCCiHCLy6K9PIqIfERHXjGgAax68IGxbmtf6z5ChbO+YFxIm700eiX6d8wEA1XV+XDasG0b3bY87xvYJ7lOn5EeU1dTjb/M2Y9Wu47jsX1oPjJtG9Qp+LrWSdP16c/S+GfHgrrdX4x9fbsXBsvioyPEiWykNH8+J/OrpSzDqz1/G7fyNwR8Q7MxvpkiN98oXnHvcNBS3GsUiAFlE1BXAZwCuAzAjLiNKcgqicEyb26sWT52I4qkTMbywbXDi2nyoHNOuHILXbzwDt43RBMWEUzrjqumhB6bUIoKqc0FI8Ow+qqmt6VG0c5WomoEbLUGG5T7z5RYUTpkVk97hTUG2kgA522WuSkPYe7zaUM+rOXHDjGXo94c5iR4GY4GsSF1aXR8Xbd3tzEBCiCoAlwJ4TghxBYBBMR9NCqNqA5JMh4m7lS4o1GeiVYYXPdpmI8PrMTQ7OlJpXL0LIfDorFBk1UMztQq2GRaazV1vrXLMH6hVTDGvL9lpu59E3tMb3+0CAHzeQvI41Na3NSm6ql64uSTRQ4gJk19fgb/P2xx5xxZEltKbpjYO5lHXgoKIRgL4CQA5a7hr08aE8c4tI/HOLSMN27ItSnZkpNl/xSd2yMUz1wzF6gfPN2zPzUwLc16b3y8w/eAz9YfMSjB9uHqf7RgAzfQledChZLrEXFn33vfWRjymOaD+EK00tOZAWU09/rNkZ9z9Py3Nv2RmTtEB/H3elkQPI6aorQuy0mM/NbstM34XgPsAfCCEKCKi3gDmx3w0KcKIXm3DtuVkhv9xzaYnM5OGnBC2zapwoDl57/pXlhnet9HNYW6T/FTKa3yYt+Eg7jFN+PX+QNDnoXI4TuF78UbNDdkfJx9FYyfgP84swv9W7kWfjrk4s3e7GI0qnDp/wDAxtSTimWuQSNRmaPHA1dmFEAuFED8SQvxZd2ofFkLcGdeRpRjm9qhAZEFheZ4MbzCKCdAeoC8i/DgWbCqBEAJvL9/tuJ8VU+dswLvLwzO1GyJ0mjObDpYHX3dv2you14iUJBkJ6f9x6/fZeqgcv3xjZdQCqqq2eZjeRj/5Jd5auiuqY66fscz2s6+3HMZ7K/YYtOSWQjy0CBW3UU//JaJ8IsoBsB7A90R0T1xHlmJYrb6dfBR2ZGd4UVXnx4Au+RjWozVyM9OwYX94BrWZugZWkV27pxSZFqsZsw24stbX4AKEB0pr8PjsDU3aF/yzogN45Zsdlp9ZmQljgerEbkiJFVlxuNLlRHfeU4swa91+fLo+uii3eMXqR4M/ILD7aDWm/G9dzM557Uvf4e531+C8p2IfFh5vmoVGAWCgEKIMwCUAPgXQC1rkExMjWmXETqOoqvPD6wHaZGcgJzMNHfIyAQBXDe9ue1xDV4kDu+QbNBiJdFZL3ly6y9LZ7UYY3v/BOkxftB3//mp7g8bYEG5+fQX+9PH3lp+VlNfG3LH77dbDeOnrkGCqasBkLJ+hiprotLk9psKSdsjn0SrJs6mJZ3n8vcer43bueJEVZ1Og25koXc+buATATCFEPeCyDjbjCivVMdMmv8KJVuleVNf74Q8ARIScjDTU6D/swd2N5ddldzxAq+XUEE7pWoCS8pDfoVf7HMv97CIxan2BiJrC8Wpt4pv66caEx/HnZaVhxrfF+NnLS2Oauf7jF78zlHBxqxWoSOG851h0E93jszdi+KPzIu4nkw0rLUrCuGH30SqURynE7GiuIcSJolmYngC8AKAYQA6ARUTUE0BkewbjmpNPyA/b1pCfgmZ60tqiekiPgtJXp7mZRpPJPeP6oVO+pm1M+3wzuraO3vZeVa8l/EnaK2GkKlZ1rO4d3w9A5EJ7coxA9M2Z3FA4ZRZue2NF5B1hbChlV7k3FlQ2wrzz4tfWJjMnDldoWpKbEOeGap+jn5yPS579pkHHmvFxwy0Dad7oG6VFg1tn9j+EEF2FEBOExk4A58R1ZClGu9zMsG0NWT23ykhDTX0AvoCA10OGaKpO+cZSIV4iTLvi1OD7smrn1d5lw7rhujN7GrZV1/kNq5l2OeH3YUX/znnYclBbkQ95+DPDZ9tLKvD2spDpSo3gkZNUICCCiYKxYPa6A/jCIqdDnZDa52YgW/k+rUxusSIR5p2fvbzUlR8pkkYhhLD1Y2wrse4XHy31/sZrFC09zLcpcevMLiCip4houf5vGjTtgokjhOhXCXt1s8PWQxXwECFb0SK8HsI/fzw0+N7jIeRmhT4vr/Vh4ildbM9d2C7bUKCwfW4GKk2ValVfixACT3y6Ade++F3YuWbfOdogmGSjpcMVtRg7bSF+9/66oB1aNU3JCei5BVsx+sn52BZD84/ZrwJoJqCderHETvlZaK10FHx2/lZc+byxZIIQAo988j3W721cqfWGmHeijcYa069DVPtLS6VVNWKVv8/bgpMfmoubXo1f3/q1e443+hyqObSlC414D9+t6ellAOUArtT/lQF4JV6DUiGi3kT0EhG91xTXSyRPXjYYAHDegE64Y2wfnH1SdD9kADhUHorxJwJylImbAOQrfTHSPBRmjsrO8KJtjmY+Sjeps/X6hH3DWb3w6CUnI79VOqrq/VDNxV4P4Xd6ldya+gBeWLgdX5t6XEw4pTM8HjK0cZUTo2orl9FaqqCQ+323QytuuPtoFd5fsScmtm8rDWHm6r1YvVublIr2lRnG/N2Oo1habCzJXlXnx0tf78AVz0dXc8f8XV//yjIcLIsuX6NLfkhQHK9yNotV1vosM/HdUBnB9PT0F1oym133xO+2H2nQdVVeXdwwn5qKat5r6S6PeA/f7ZNyohDiISHEdv3fnwD0jnQQEb1MRIeIaL1p+3gi2kREW4loitM59Ovd6HKcLZpOev2lPh1z8dsL+hmczW5RBYGHyBDKSWScnIk056xKmpfw0S/Pwj9/PBQPTtKqtHx8+yi8dsMI/OZ8rW/Gg5MG4toze2r+kFqfoYhgmoeCdajsVp5tsjVBdFKnvOC2Uguzl0zOUzWW6jo/DpTWBCe5tXtK8dt318Qkw9vK4f7AR0WGyKxsi+g0q7az0YYb51k0trrupXBNzAnD9+RgFgsEBAY9NBeffW89kUey/0fSKKxQV+xXTV8S9fEqxyrrIgpCO/p2DPV2qarzo6LWh/s/WGd4/roUWFdztqO0uh5bD5VH3jGexFmlcCsoqololHxDRGcBcBNaMQPAeHUDEXkBPAvgQgADAVxDRAOJ6BQi+sT0L7wrTxJzdt/2eO4nw/Dr8/s2+BzqxO8hY8a3h4B85fO3lu4O0yi8HkL3ttm4aPAJuPaMHvhmylic0q3AUrvJzkjD/E0lWLP7uOF4mTy4Yb/1jydNF4A//0FhcJvVvjfM0EwXaoTLq98W48wnvggmwMkJcbcS4jln/QGMfvLLiH3Dzdj5HNQcl1bp4TkU5crK1Gwuq6j14dn5W20juypqffAHBI5aJMmVVtdjW0kFAgGBv8zdaJvXIVEFhdOqP1KtKqtjy2vqg6vuSBqFFeb733ig4bEwQx/53NBPJRrUUfx36S68sHAb3vhuF15UQq+jzRO56oXFOO+pRQ0aT6xoLhrFZADPElExERUD+CeAWyIdJIRYBMDcLm0EgK26plAH4C0AFwsh1gkhLjL9S858exuICBNO6dKo8gindAuFwHqIkJNpr1EcLKsJWyGrfhEicoyEssoAJgr5KQ5EMJ2oSUL3vrfG1k6sTviyTpUMAZWjrdXr8C/cXILJ/1mB3Uersf1wuP+itLreNvNWCgqzyUd1nFppFK99Wxx8/ZIp4uipzzbjL3M34ZO14TWzhBA4+aG5uOvt1ZbjOVhWi3OnLcTUORvx7PxttnkdEnUurqrz4YsNB/HYrPBjIvUsKK8N1+7UybMhGoU5nHX838O7IzYF/oBAnv6b+NeCbXjmy60AgH167kRmmgdVdf6ofBYbD2iLlpbu53DCbdTTGiHEEACDAQwWQgwFMLaB1+wKQK0VsUffZgkRtSOi5wEMJaL7HPa7WTrbS0qSo8plQ/jxiB7B17uPVRnCOT0mU1O61wMio3nLTXikxKqPRr1PBCdTf8B6QvrxGVrklHrtY1X12HrI2jGtWkLMK1P5bot+7M9eXmp5nGTInz7D2GkLLK8jJ9AzHv/CsL26XpsYf3VuX0tBoSZoqTklQGhStVqFS/PUx2ucCy9OX+Qu0VAglBRXWevHja8ux7+/CtdCIkVrWY3Vq/ytosnx2HywHHW+gGXeQzTPWqzwBQJh5lYgVPwyv1W63nfDnenw9cXFwdfxqNrqlubizAYACCHK9AxtAPhNHMZjdc0jQojJQogThRBPOOw3XQgxXAgxvEOH6J3AyYI6+db5hWFi8xAZtJXf6L26Lz41vLigG/ItfnB1/gCKj2hmoN+9b11eoV/nkG9CNWnNtJkw33GoQeXUcc5u5WtX1M9cjl0iq8WO6NXWMoN+5Imh8F1z4pP0M/ktfsmN6UZWUl6LAQ/MMUT/CBFaLZvv/UBpDUY+8QW2lVRELA/ypVIbTAoVdZ7/73e7cOHTX2H30aqIZpoL/rYIJ/3hU0vHfENLupixy9I+UlEb5jsIBKz9QZICXeN2a356QKmYnMgaUSLOxqfGFAhpaIbHXgBqLYlu+jYmxtTW+02mJ+PnUtv46xVDGnR+qxIjdb5AVL6BGT8/PfhamgFUAgHhWFLBadIz/9gjtTGtqQ9Ymg8O6lqCh8iyu526kjRnpgcFhT6ZCSGC12hMlvm32w6jut5v0DYCQgRXy+qqv84XwB8+XIf9pTV4YeE2S3+IyrsrNMG8ZPsR9H9gDr7ddjhsItqwvwyjn5yPy//1ravxWhWOjBVqzokQmj+naF8pTnt0XpjvwE6jkIseKSisEh5//spS3PL6csxZv9/SmR5tSPORitqYCZdmpVGYaOjQlgHoS0S9iCgDwNUAZjZiHEGIaBIRTS8tbVwMe7JQ5wsYNArpfzipkxb50bNdNgCjs/aBiwa6Pr+VL6XWF8CEwVouxsldjdnms+4chY2PGGIb4IkQ2TVDsf+bBR3gvIr7rMgY1fPMl5F7EFg5nWVuioeMDmOJWpdJ9bsEAgKv6WGc0s3R677ZuFSfXGsboVFIzaWmPgCfP4Bz/roA6/eWBfNi1DE9/cVmzNugaQlF+8rw5JyNjue+Uq8JtkwPQf56y2Hb8FFpnw+9t3ZSv7a42PGajUHVVur8ATw7fxsmPfN1aJsiyP0BYSkoyvSM/5CgCH+uFmwqwdyig5j8n5W4+fXwTP5onfynPTovZpnqCXVmE1E5EZVZ/CsHENFeQURvAlgMoB8R7SGiG4UQPgC3A5gLYAOAd4QQkTveuEAI8bEQ4uaCgoLIOycxn9yhBaiV1fgMUU0e/a89966z8e2UsTi5a+h7kuGml0RhhspIC5+56/wBdNEzwI+YSlwMOqEg6po03yg5GNkWx6omFrMJwmwDt9IG1GN6tc+xtKWX6TkaduHK6updPf6Hfw21bPl6S8hvtmqXZi5qiE1baiPye6z1+bG/tAY7DmtJgfLvrY7p2fnb0E8PRS7aV4bzBnRyvMbUTzVBIjXGer+zlnj/ByETo/lvLrHLNI9F7a4L/hbSGl7VFxbqcKtNfx8n01N+UCNz1g6W7jgapn1GE7IrFyRq+fpYMP/uMTE9n8RRUAgh8oQQ+Rb/8oQQEWstCyGuEUJ0EUKkCyG6CSFe0rfPFkKcpPsdHovVzTAaITtrvSGPwqMvyYkIJ5iima7VS3NkRjGRz14Xyp84t78WyVzvC8DjIWSmeRrV4Ec2U6pXfvFWuQnqBBTJSasKKfkjf1iJJAoIYalRyBh7IsLphVrTqdZK73NVWKkT6u6jIZPZ/E3hARYNmSSlcJG5HTX1foPfRAqKRz753mAaVEuP9Gib7epa8vhIJjs1oz3aisdW+TMNobSqHmc/OR+Pzw7XllTzpJ1GIcmPwkdhflTMzbuciHX1WyG0xZ5dUc7GEt8i5kxCkA97TX3AkEfhZOS5f+IArH7w/LC8CifUAoCndm8NAMHSFY2JADmnXwec1ac9AGPyl1V9H6OgcL6maoZ7f6XmFvtodcg9tvNIFXwW15ClRjwEXH5aN3z3+3PxA8WBrZoc7PIlLj+tW9g28yq7e9tW2PTo+LD9zMes2Hk0mMltvmfVjJilTNpSiwFCZoo3bjoDd1r0apc8oWsWdX5haXKzwqoBl4p5ki6rNk7ItT4/CqfMcgxgSLPQ7Gp8fuyyqf2lCnJ/QDiOUX7mpvCh+W+962iV66ZRsUZANNxr7IKkEhTso9DIUyZ7c2a2HV4PoXW2deVXO2bfOTr4WgqnWIQIej2EP0zUfCVjdU3FaqIFgBU7jwVfR9Io1JX39EXbsO94ddA2LbHKvZD7eD0EIkKn/CyDNqVORHblr62+lydmbzC89/lFxByaN5fuwmX/Woy/ztUaQ9XU+w2OTI/yNzbfm0Tu3z43M1gNwMy87w8GNYm5RQeCq2c1SdIJtaaYSng/d6NGISsJm78bFfl3VBcqTr6qKpPpySqsWyI1NTvUEZsAACAASURBVLsIOBUr4SnNdpW1PkdznVvn8+GKWjy/cFvkHI34yonkEhTso9DwKJnP5szsWNJRqUbbJicD79wyEiseOC9sv99P6I/Pfn226/MSEdrkaILn+YXbAADvrTBGzailzSVWEVC3/3dl8LWqUWw+WIEfTP0ybP9/fBHu8C5Rop6sUP0BdivvFcVHgzWjAK3e0cpdxsJ2UtC9O3kkrj+r0PI8y/XaUov1ekk1Pr8hIonIOilQRe5PpH2PVte66bVQQb+jlXXB+xraozV+MbqX5XmPVNQGa421SvcGTaBOmE1PchI/ZtNKd+/x6qCwURc+TqYi+b1W1PpQ5wvgZYcM9wJ9seSmla+dmbKm3o9BD83FE5/aCzu3Gtq9763F1E83YtVu5yKIAtbBHrEiqQQFE6J46kT88UeDDGq23UTXGGQFUiEERvRqa+ko7Nspz1DXyYxZgHlIs7d6PWTb82HaleEhvVYO00/W7kfhlFmoqPVFNIsA1r4EiWrWkdpUr/Y52F5SGZw07EzP+0prDBEu5kKJANBd9x2cXtgWPzmjZ9jnADBc95FI05M5cspDFLFEuZyjCJrf5iG9pperY4jQyqIV7KvfFuO0R+cFy654POTYnlP6Sarq/Hh9yc5QMUuH+bNwyiycpQj3EzuE7PFO9yw/k85mJ5+LEAKZaR5X/UCscmOq6/3Bv8nby4zms/mbDuGSZ7+BP+DelCe11RoXYbQNqTbtFhYUSY666orHikOahjrn2xdSixQG+toNZ+CCgaFIHA9pJh41ysmuIZKKU6hhSXlt1BFXC0wRJGqF14En5KN46kTsOFyJDfvLcOLvZwNwv1K0Sty778IBwdd2WoE0YUl/TU293zC5RtIa070U3N3qebjDxmchTR8eAkotonsemmkMXPQS4WBZuPlGmq7O1//eWw5W4IEP1+MXr2nhpnbfn9Xq/flrTwu+dopSkoJCmvWcim3W+gLIzUxz58y2GNPCzSX448fad2EOvvjde2uxevdxHKmsDQulrfcH8Oz8rfh03X6DCTVDH3NtBOd3vMuHsKBIIeKhUVx3Zk/MvetsnKE0FwJgqBEVqZLqqL7tMf2nw4P2YTlOdTJVJ9FHlZ4YbqnzBRzvv7BdeCRQYfscg2/ETUTY6ggmAolaEmPaFUPw7uSRhgxvO0FRbZoQa3wBwyL8ULmzbd3rIWVSCf8+7CKC7n53DQDtb+OmxLfdqv2m0b3w4k+H49YxJwIAPtSDCWStJbvpzup8qk/tsVn2Zh458cqv/EGHXCGfXyAnMy1MQ3nZomug3XP9waq9lmPOTA+VV3ngo1BW+u6jVbhm+hL8Ze4m3PrGSjzyyfc4UFqDFTuPIUNfnFz/yjI89flm23ELwaYn17Az25l4PEhEZCjJIfnlOaGVaa3LTnBycpQ+FkP5EeVJlXkAN46ytpVbUV3vd0xK+rPeC8RuTAAitor1BwQ+tyjd/X9Dw0uZTVN+9KNPah8Mu5VYaRxAeJ2lOl/A4Bj9dptzr4ea+kDYxKmSm2ntV1ijV2slaIItEkdtcgq8HsJ5AzsFo+tk/ocsk2K3Mo602LCrEwaENAp5aiet6/pRhcjO8AaTLCUPfxJeXNEcsWXGrHDISgiVtT7D+Yv2leLEDqHy5298twvnTluAy/71bTBRErD2n0kE2JntGnZmO+MU9RRrVCd6jsuQWxmhJX/IrSxyQIBQ/sEVw60joS4b1g0bHjaGmR6pqHVUz7uZcguks9xuwrbCrpxIb4vYdnW1aaXp2DUV+q9FFz7Vmd0uJyOiM1vW4LJ6GnIdcgwAvaJwm8id9OzMfPJeM035FnX+AB76aL2l3R9oXI9saecPfk82v4OBXfKRn5WOjQfKsbT4aMQuhdM+C+8D74R8lhZvOxLsmghoVYLVPBcgusKLgNQo2EfBxIBYRz05odYTuvDkzq6OkT8kOZmYCxreP0EzP0mTg12GbducdLTK8OIcpdXnja8uD64o+1k41nNMk+vd47SCidkW/SfssGvPucVhtQsYzVCShv7o6/0B29BUN9c4UOrcZsZDkXMlAPu/uZr0aebVxTttJ2cZdtoQgjZ/xYk/0mQqBcITIO98axVGPDbPsG14zzbBAI5P1x9AJIr2lQZLzMvn+bHZGwyC4KGZRXjlm2I3t+IIaxRMTIiHj8LNtdxOenICkrubBcVNo3thxxMTgmGX5sldsq2kUj/eOMnLmkDHTGaREwqywvaV48/JdJ4Un7761ODrH/9b60h3Zu+QGenzX58dMQvX7u/yzDVDXYUVq4vwshofBnTJt99ZQb3q89cOwwiT+cvyGCJXWlaah4JO40sV05uqKZm1CiDUrEqlrKYe75rCoyXr/zTO8D4jzYP3b/2BYRxB05O+zUOEN246A9sen2A41py8uL2kEofKjZroOf074lqbiDQrJv7ja9z+31VYvO0I6n3xczg35+qxTAujCRUKS79FJOQE5LUwT3hIm6RUoWMXpy8jZ8wTmvQLmE1Ez117WljpCXmVSJOi1aSsTtx9O+Whk0NEGACQza9w0pATHMOKJW+ZwjC7FLTCF7/9YfD9pcOs272o8mn8yV3wzuSRaJ+b6XgtDwFZLhprERF66uY8mWUPGEuJuDXrWWXLS8yVBOp8AZzWsw0euXgQHr54EFpleIOCQj4XRJofzBz9ZFdSpaTCGCBgNhO54Zp/Lwnrrx5LRJydFEklKNiZbY38PThlpMaaM3u3w21jTsTnUSTaZZtMT6ojzwo7TUWG49ppHOYVviw/Yjy5cUx2WJlhzOdvEyHjPZKmZzk+BSsnpyqcnrryVDx99amY/MMTI173R0Oci0J6yDk/QuWCQZr5qUvr0FiMbWXdCopwjaxTvrNAu25kIX46shDlNb5gBWI1f0SiBhrYhXGPeMzYyCrHIo+kOcCmJ5ewM9saGT+e5m1KnQK4d3x/9HWxIpaEop7CP4umNIhcqVolhgHuGgYFQ3R1H4W5ZLrEylFvjtCJNLFa+ShUXr9xBF65/nTHfSRyAjVPwhef2hXDe7aJeLx5MWGO9CKCoZ2uE/eM64dvp4w1RPSoRBIUUuDO32RcMIzt3xH//PGw4Hu18ZaVOQvQku2s8kf+cvlgPPcT7VyReonLY90GZzQlQgh2ZjON4+9Xn4r+nfOa7UpIIidlqwc+Ulc2SZeCrGCehZ02oE76V9jUkDKbnuy0Aqtr1NQHMKBLflCTO9PCcWq4VoTfd15WOs7p19F5JwBDurfGwnvOAWCdVFZl+g7dzCvm1q5EhKx0LzY+Mh7bH5+Ar393ju2xXo9WpdguNyOS6Uk6oc1O45d/frohnFg1z9kJpTvfWh30NagZzGleDzrr9a5kEmObbHtBSAj30dw25kS8dfOZjvfSFHAeBdMoxg3qjDl3ne2YkdoceH+l5rA0F44DgFlr9zse++Tlg7HqgfOx+L5zgyXU7QTF8J6hSUYWHzRDQY1CNgiyFlRWK9hDZTX45I5R2PzohQC0CdypKmysggyy0jyO2eejFV8B4C7IwKwdyUcoK90Lj4fQpSCkcdjVgLLTHCJpFNV1fjzw4XosUMqqWGlW6j2n22jN6/eWwi7P0PycvDt5pO2YyCLq697x/XFm73Z4+GKtFIpVRFWssAuMSGjjIoZJBJ8VaStIdSU6oIuzCau23o82OcZVv9WKNSvdg46qfdtmrgzmcuiTgp1GYzXZltX44PWQwYzjVBXWrQB3yigGtCxtJ8zfh5ur3nVeX8N7s1BTx35/BKFrNx67RMbqer+h+VTrbGvNSjXt2eVhqOHa5tGYQ6D7dLR/1ghGwaJm9P90ZCGKp07EiF6Ro8ck0a7d7BYsgqvHMqlCV5Mm8MrPT0dWugfjBnXC7ef0dTrUUu+2MrWd0audIWrHLnJKmifkJBSr3sZmNj063rWgiBQ9NaLQ2Qdh1n7sFAq12F5bk/B1o/wMieB8l0hN4FfnWf9tzcL5mWus80PUifv6Hxi1GrXkvlrY0HC8RRST+b6D54Dxe7QSgtE0b3Lr0JfYLVgE2EfhGo56atk8qzsVZSLd8MK22PjIhXjhuuEocLAbA0aHpsRKo/AQgpnFd19wkv0JFRML4M4B7hZ1oonUf0IlUljmL87u7fi5eSKxqzaqdsAz2/ytjin60zgU6fkMRX8ah3dvsTfdqMhJ0s6ZbxbOdgK1dSttUv/hSR1wmcnntPj35wLQMu2D5dVNx1uZKIfaCLva+oCWSyJzfiz2iRQ5pmLOwI4UJVhTZ2N6Yo3CPRz11LKR/YojlclWkZOuVckLqwlg/qYSXDCwE56/9rSwcFF1JSznLhnh0s5F9VrJvN/80PFzOa6hPdytvCWRug+azUJf/vaHWHSP0dn85i9CTle7Baia8X6Wya+xdEd4LkBOZlrwe8rJTHO9opZ5C2U11r0fquv9hm52aVbhcEAw96OnRWHH3Mw09Gqfgzp/QNEojPtY5YXYOdrlij74ucV32L1tNtb+8QLL452478LIUYJOkVnszGZSAllnyFwh1Yn/3fYD3DrmREunsp0zm4gw/uTOYaGgbyuRK3LSbZuTgb9fdSpe/OlwV+PJyfCiT0fryBvZPlVqEdH2N45kpjALit4dctHDNHmeoOQ02M0r0uwycXCXsM+62HTEi8Rlw7phwinGsh5zi7QCiq8v2YniqRPRMc+YG2HuB24X3ntKtwK8esMI/H7CAMvPszO8qKz1GTKzVTwWmordsyN9BE4aBQDkZ6UbTHhucOOQtjKB7jlWFbGveWNhQcE0G+SK2RzG6cSgEwrwu/H9Le2zraKo0wQYo2fUs10ytKuhm58TTvke5sknUv6E0/juGdcv7HM3ZzOslG0OkPsMtMg6j1YLkky7cgie+8lphm3B29dnyPsm9Dd8Xl0XMAhHp6/rhyd1sI34yslM01qTKpnZkTCXdAmNSXs2pe9KlouJ5hx27NDP9ctzTrTdx8qZPerP8/USJ+yjYFIAOSn0tVmRR4tapylatTya/QedEJpQ7XpmA+EJgNGGxapa0y/P6YNVD5xv+NzN6fIVs5Kdj0Ka8WR5+PeUcNFoEigjMfXSUwAAw3UnvPn7qK73G4V3A20rORleVNb5XPepBiKbntz4lqJ1VMv2tic4lLO/avoS26ZK8TQ9Ne8MLCalICK8c8tI9I5SZbdD/aHKScJN4TsguraSH9x2FlbtOoarpi9x3E927FN7VkeDORLHPJnVO9REkmS5WKGPG9QZT3+xBeNP1kxPWlDBeIO/IBZcOqwbth6qwG1jtN4lZt9Udb0/6snWipzMNOw8UgUEv/fw+3j/1pGGpMpsm+tKQfH9/rKI142mRD2AYJfH9Aildt5YshO36P41tWAhO7OZlGFEr7YRC9O5Rf5Q1SgUs53cjmgm8Yw0j+MqUGKOWoqmLAkQchQP6aYFa5j9MpGc3WbsNBrZ5nWgoillpXtjXiss3evB/RMHBvNfzKa5mjo/th92LtHuBq+HsP1wpWWtJ8lpPduitxLhZTfJ2+UxWOFGyI0bpAmHiwZ3wd26OdGuF0loDKHnxqpFbDxgQcEkLXlZ6fj812fjyctD3evcmi+iXe27qX8kBZbsFx3NpCPZ/OiF+Oj2UQC0e5Ea0q/O7Rv1CrapJhm3mH0MVXV+V1pSJD5arfWDKK3Woqvc/G3N9ZzO6qMFIkSTT2PnEC+eOhGAlkA4Xu/bcduYPsH7lxqF7HvRs102rhrePXi8Gvmk/gnZ9OQSIpoEYFKfPtYN4pnUQ9rUO+dn4UBZjesfUzSmJ8Dd6nFojzYonjoRH6zag/8s2eW6fpUT70weiQWbDoWFsbohEI3RvgkY29+YdV0VRfSbG2TOgpu/rXmSl9pXNKHbWQ6Ce82DF8DrJeRmpmFsv06GPCFZvDPN48GsO0ehR9ts+AMCby/XysmrC4yAwfTEzmxXcB4FY8fovtpE6vanFK05PppsXJkxHqts7zH9Oka0a6tcfbq2Oo2U6d3UqPeQ4fVENSk78Ve9x7cMu3azWDALfllOJpo+K06Lh4Ls9KCp0JxMKutV+QIBDDqhAHlZ6QbnuSooVK2QNQqGaSTSLBPJ3NKnYy62HqqIazkEadZoiOkpFky9bDCmXjY48o4JpE1OuqHOU2OQiZwVtVKjiIw5tPXK4d1x46jewZDh9rkZOFxRh2tGdLc6HEDk8vJ2dMzTBHhhu1BQh3oudYHhbyJnNgsKJiUIFfdzdiC/dfOZWLenNK6VdkOCIr5JUi2ZWPpP5PddFoWPQvX3bHnswjCNTZqizNn9Kuv3GiOj/nbVEHRrE549bubkrgX4z41n4PReoQRDdeGimiwD7MxmmNghf/iRsr7b52binP6Rez/YYVVGwkxupnNF2lRGCvRYNgeS/oa/6a1wF24ucdrdcAxgHa46rIc2iXfIs4/Q23W0Kvh66f3n4v+GdjP00XBiVN/2YbkawYTUOn8wLFYVqPtKa1yduyGwRsGkBJHKhceCVQ+cj0wX5gZp1mBBEc7KB86HgMAVzy+O2Tnl912uJ6odqahz2l0/xjk4YdqVQ3DXsb6O2ddqJJw0JzWGrHQvKmp9+GrLYdz19mo8ffVQ27LqsYY1CiYlkD/8eE7ObXIyXJVtCJqe4lS6vCXTKsOL7Iw0FO2LnNDmFvm3H6VHht04yrrBknkcTuRkpqF/Z+v2uJLWLlvGukX1U8iQX18MwofdwIKCSQlkjHq1TZnmpiRHn4SiaXDDNBwpmGV3uEgl64Ho6zRZ4dRStSGY80wCAREzh38k2PTEpARyhZioSCOVNK8HX/z2hw2uxMpEh9QoKmV4rIv4oFiUDrl1TB98uHof3nHZnyMS5jE9/Mn3ETs/xgrWKJiUoCl8FNFwYofcmKxaU4E7x/bB9w+Pa/DxmWkeeCgUheQm6ikWUW/9OudF3RrVCbOgmPFtscH3McCi2m+sSCpBwR3uGDtkxEgco16ZGPKnHw0Kvq71BxolVInI0Ba3pT4CVoESQuliUevQ1KixJJWg4Mxsxo7TC9virvP64nG9tDXTvJGZ0ABQVdv4CTBSG9mWgKwRpqIGPdXGMS8nqQQFw9jh8RDuOu+kmIQpMvFHDU+NRSmPg2W1wdfxLHURT647s2ewFI3EIChYo2AYJpVQI3yq62NbHLClGp+IyDGzmzUKhmFSCtVxWxkD05NKS9UogPBEQDWLItr+JtHAgoJhmGZHfyWCJxZVdq89s0fwdQuWE+GCQrE91flZUDAMk0IUtErHzNvPAgBUxcD0ZIh6cqlSTBpyQlxDThuCOWO8qTqKcCA3wzDNEplRHYtyRmqRQbcaxTPXDG38hWNMLBIBGwILCoZhmiW92+fgzrF9cNlp3Rp9LoOgaMG2p3DTU9NclwUFwzDNEiLCby7oF5Nz5SgTbDPrABsVrUyJh/H0S6iwj4JhmKRH1Shq4phvEG+yTaanpqpAzIKCYZikJ1cRFLHqxZ0IcrOMGkVT1S5jQcEwTNKj2vZ7to3chbC5kmvq/KcKvV7tc8y7x4yk8lEQ0SQAk/r06ZPooTAM04yQpqeTOuWiXa59+9LmTp6NRvHGTWdgSPfWcbtuUmkUXBSQYRgr5Eo81lneTY25l7jsAd8hLzNM24glSSUoGIZhrJAluv2BFhzyhHDTk9Qo4h3xy4KCYZikp0DvX33l8MbnZCSSzDQPJpzSGWP6dQDQdK19k8pHwTAMY0V2Rho2PjIemWkte21MRHjuJ6dh44EyLNhUgvdX7tG3x/e6LCgYhkkJshJU/iIeZKUZ78Xria8AbNnilWEYJgUxFwf0xlmlYEHBMAzTwjBrR/441yVhQcEwDNPCyEo3Tt31ca75xIKCYRimhZHh9cCjWJta61Fd8YIFBcMwTAuDiIK9KTLTPOiYnxXX67GgYBiGaYFIP0WnOAsJgAUFwzBMi0QKiqZoxMSCgmEYpgUiQ2TTvfGfxllQMAzDtEBk5FNTZJuzoGAYhmmBSGd2GmsUDMMwjBXSR+FhHwXDMAxjRUhQxF9SsKBgGIZpgUjTU7zrPAEsKBiGYVokrTg8lmEYhnFCRj01hemp2fejIKJLAEwEkA/gJSHEZwkeEsMwTMLJ0vMo4tyKQrtGPE9ORC8T0SEiWm/aPp6INhHRViKa4nQOIcSHQohfAJgM4Kp4jpdhGKalEDQ9xb1jdvw1ihkA/gngNbmBiLwAngVwPoA9AJYR0UwAXgBPmI6/QQhxSH/9B/04hmGYlEdGPdX54t83O66CQgixiIgKTZtHANgqhNgOAET0FoCLhRBPALjIfA4iIgBTAXwqhFhpdy0iuhnAzQDQo0ePmIyfYRimuSI1ito496IAEuPM7gpgt/J+j77NjjsAnAfgciKabLeTEGK6EGK4EGJ4hw4dYjNShmGYZkpQUNT7436tZu/MFkL8A8A/Ej0OhmGY5kSmHvVUl6QaxV4A3ZX33fRtjYaIJhHR9NLS0licjmEYptnSqgl9FIkQFMsA9CWiXkSUAeBqADNjcWIhxMdCiJsLCgpicTqGYZhmiywzXtvSBQURvQlgMYB+RLSHiG4UQvgA3A5gLoANAN4RQhTFcxwMwzDJRjJFPV1js302gNnxvDbDMEwyE3Rm++LvzE6qEh7so2AYJlVoSo0iqQQF+ygYhkkVZK2ngIj/tZJKUDAMw6QK0vTUFLCgYBiGaYHIqKemgAUFwzBMCyQrjQVFg2BnNsMwqYKnKZply2s12ZWaAHZmMwyTSvxx0kB8cseouF+n2dd6YhiGYaz5+Vm9muQ6SaVRMAzDMLGHBQXDMAzjSFIJCnZmMwzDxJ6kEhTszGYYhok9SSUoGIZhmNjDgoJhGIZxhAUFwzAM4wgLCoZhGMYREqIJatQ2MURUAmBnAw9vD+BwDIfTEuB7Tg34nlODht5zTyFEB6sPklJQNAYiWi6EGJ7ocTQlfM+pAd9zahCPe2bTE8MwDOMICwqGYRjGERYU4UxP9AASAN9zasD3nBrE/J7ZR8EwDMM4whoFwzAM4wgLCoZhGMYRFhQ6RDSeiDYR0VYimpLo8TQGInqZiA4R0XplW1si+pyItuj/t9G3ExH9Q7/vtUQ0TDnmZ/r+W4joZ4m4F7cQUXcimk9E3xNRERH9St+etPdNRFlEtJSI1uj3/Cd9ey8i+k6/t7eJKEPfnqm/36p/Xqic6z59+yYiGpeYO3IPEXmJaBURfaK/T+p7JqJiIlpHRKuJaLm+remebSFEyv8D4AWwDUBvABkA1gAYmOhxNeJ+zgYwDMB6ZduTAKbor6cA+LP+egKATwEQgDMBfKdvbwtgu/5/G/11m0Tfm8M9dwEwTH+dB2AzgIHJfN/62HP11+kAvtPv5R0AV+vbnwdwq/76NgDP66+vBvC2/nqg/sxnAuil/xa8ib6/CPf+GwD/BfCJ/j6p7xlAMYD2pm1N9myzRqExAsBWIcR2IUQdgLcAXJzgMTUYIcQiAEdNmy8G8Kr++lUAlyjbXxMaSwC0JqIuAMYB+FwIcVQIcQzA5wDGx3/0DUMIsV8IsVJ/XQ5gA4CuSOL71sdeob9N1/8JAGMBvKdvN9+z/C7eA3AuEZG+/S0hRK0QYgeArdB+E80SIuoGYCKAF/X3hCS/Zxua7NlmQaHRFcBu5f0efVsy0UkIsV9/fQBAJ/213b232O9ENy8MhbbCTur71k0wqwEcgvbD3wbguBDCp++ijj94b/rnpQDaoYXdM4C/A7gXQEB/3w7Jf88CwGdEtIKIbta3NdmzndbQUTMtFyGEIKKkjIsmolwA7wO4SwhRpi0eNZLxvoUQfgCnElFrAB8A6J/gIcUVIroIwCEhxAoiGpPo8TQho4QQe4moI4DPiWij+mG8n23WKDT2AuiuvO+mb0smDurqJ/T/D+nb7e69xX0nRJQOTUi8IYT4n7456e8bAIQQxwHMBzASmqlBLgLV8QfvTf+8AMARtKx7PgvAj4ioGJqJeCyAp5Hc9wwhxF79/0PQFgQj0ITPNgsKjWUA+uqRExnQnF4zEzymWDMTgIxy+BmAj5TtP9UjJc4EUKqrs3MBXEBEbfRoigv0bc0S3e78EoANQoinlI+S9r6JqIOuSYCIWgE4H5pvZj6Ay/XdzPcsv4vLAXwpNC/nTABX6xFCvQD0BbC0ae4iOoQQ9wkhugkhCqH9Tr8UQvwESXzPRJRDRHnyNbRncj2a8tlOtDe/ufyDFimwGZqN9/5Ej6eR9/ImgP0A6qHZIW+EZpf9AsAWAPMAtNX3JQDP6ve9DsBw5Tw3QHPybQVwfaLvK8I9j4Jmx10LYLX+b0Iy3zeAwQBW6fe8HsCD+vbe0Ca9rQDeBZCpb8/S32/VP++tnOt+/bvYBODCRN+by/sfg1DUU9Les35va/R/RXJ+aspnm0t4MAzDMI6w6YlhGIZxhAUFwzAM4wgLCoZhGMYRFhQMwzCMIywoGIZhGEdYUDCMCSKq0P8vJKIfx/jcvze9/zaW52eYeMCCgmHsKQQQlaBQsoPtMAgKIcQPohwTwzQ5LCgYxp6pAEbrPQB+rRfg+wsRLdPr/N8CAEQ0hoi+IqKZAL7Xt32oF3ArkkXciGgqgFb6+d7Qt0nthfRzr9f7DlylnHsBEb1HRBuJ6A09Cx1ENJW0/htrieivTf7tMCkDFwVkGHumALhbCHERAOgTfqkQ4nQiygTwDRF9pu87DMDJQitZDQA3CCGO6qU1lhHR+0KIKUR0uxDiVItrXQrgVABDALTXj1mkfzYUwCAA+wB8A+AsItoA4P8A9BdCCFnKg2HiAWsUDOOeC6DV0FkNrYR5O2g1ggBgqSIkAOBOIloDYAm0Qmx94cwoAG8KIfxCiIMAFgI4XTn3HiFEAFppkkJo5bJrALxERJcCqGr03TGMDSwoGMY9BOAOIcSp+r9eQgipUVQGd9LKX58HYKQQYgi0ekxZjbhurfLaDyBNaL0VRkBrxnMRgDmNOD/D/1rdYAAAANdJREFUOMKCgmHsKYfWVlUyF8CtejlzENFJejVPMwUAjgkhqoioP7R2lJJ6ebyJrwBcpftBOkBrZ2tbzVTvu1EghJgN4NfQTFYMExfYR8Ew9qwF4NdNSDOg9T0oBLBSdyiXINR+UmUOgMm6H2ETNPOTZDqAtUS0UmjlsSUfQOslsQZaFdx7hRAHdEFjRR6Aj4goC5qm85uG3SLDRIarxzIMwzCOsOmJYRiGcYQFBcMwDOMICwqGYRjGERYUDMMwjCMsKBiGYRhHWFAwDMMwjrCgYBiGYRz5fxV3s3IGD9fSAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "cnn_model = build_cnn_models()\n",
        "\n",
        "batch_size = 12\n",
        "loss_history = mdl.util.LossHistory(smoothing_factor=0.95) \n",
        "plotter = mdl.util.PeriodicPlotter(sec=2, xlabel='Iterations', ylabel='Loss', scale='semilogy')\n",
        "optimizer = tf.keras.optimizers.SGD(learning_rate=1e-2) \n",
        "\n",
        "if hasattr(tqdm, '_instances'): tqdm._instances.clear() \n",
        "\n",
        "for idx in tqdm(range(0, train_images.shape[0], batch_size)):\n",
        "  \n",
        "  (images, labels) = (train_images[idx:idx+batch_size], train_labels[idx:idx+batch_size])\n",
        "  images = tf.convert_to_tensor(images, dtype=tf.float32)\n",
        "  with tf.GradientTape() as tape:\n",
        "    logits = cnn_model(images)\n",
        "\n",
        "    loss_value = tf.keras.backend.sparse_categorical_crossentropy(labels, logits) \n",
        "\n",
        "  loss_history.append(loss_value.numpy().mean()) \n",
        "  plotter.plot(loss_history.get())\n",
        "\n",
        "  grads = tape.gradient(loss_value , cnn_model.trainable_variables)\n",
        "  optimizer.apply_gradients(zip(grads, cnn_model.trainable_variables))"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "-PxaXAskn5un"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "name": "MNIST_digit_classification.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "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": "3.9.7"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}