Salinee_Kingbaisomboon-M03-DataModelFinal.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Milestone (Process and Tools) - Data Model Final\n",
    "## Author - Salinee Kingbaisomboon\n",
    "### UW NetID: 1950831"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Instructions\n",
    "Specifically, you’ll need to pick two (or more) classifiers and perform each one of the following tasks, in order:\n",
    "\n",
    "1. Split your dataset into training and testing sets\n",
    "2. Train your classifiers, using the training set partition\n",
    "3. Apply your (trained) classifiers on the test set\n",
    "4. Measure each classifier’s performance using at least 3 of the metrics we covered in this course (one of them has to be the ROC-based one). At one point, you’ll need to create a confusion matrix.\n",
    "5. Document your results and your conclusions, along with any relevant comments about your work\n",
    "\n",
    "Make sure that you state any assumptions you make (e.g. the parameters you use for the classifiers) in your report. Also, for this assignment you’ll need to provide your Python code, along with comments to show what each part of the script does. For additional traction with this material, you can examine how the performance of the winning classifier changes if you bin one of the variables you’ve used (you’ll need to binarize the binned variable after you create it).\n",
    "\n",
    "Note: if you have a multi-class problem, you can focus on a particular class, when evaluating your classifier. This way, you can simplify the problem and turn it into a binary one (class 1 = class you are focusing on, class 0 = any other class). This will enable you to use a confusion matrix for your evaluation of the classifier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import sklearn.preprocessing as preprocessing\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import *\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\") # To suppress warning for violin plots when python version is < 3.7\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Adult Data Set\n",
    "url = \"https://archive.ics.uci.edu/ml/machine-learning-databases/adult/adult.data\"\n",
    "Adults = pd.read_csv(url, header=None)\n",
    "\n",
    "# Find proper column names here:\n",
    "# https://archive.ics.uci.edu/ml/machine-learning-databases/adult/adult.names\n",
    "Adults.columns = ['age', 'workclass', 'fnlwgt', 'education', 'education-num', \n",
    "               'marital-status', 'occupation', 'relationship', 'race', 'sex', 'capital-gain',\n",
    "               'capital-loss', 'hours-per-week', 'native-country', 'income']\n",
    "\n",
    "# Turn off warning generated from 'Chained assignment' (https://www.dataquest.io/blog/settingwithcopywarning/)\n",
    "pd.set_option('mode.chained_assignment', None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Declare Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to replace outliers\n",
    "def replace_outlier(array):\n",
    "    # The high limit for acceptable values is the mean plus 2 standard deviations \n",
    "    LimitHi = np.mean(array) + 2*np.std(array)\n",
    "    # The low limit for acceptable values is the mean plus 2 standard deviations\n",
    "    LimitLo = np.mean(array) - 2*np.std(array)\n",
    "    #  Create Flag for values outside limits \n",
    "    FlagBad = (array < LimitLo) | (array > LimitHi)\n",
    "    \n",
    "    # FlagGood is the complement of FlagBad\n",
    "    FlagGood = ~FlagBad\n",
    "    # Replace outleiers with the mean of non-outliers\n",
    "    array[FlagBad] = np.mean(array[FlagGood])\n",
    "    # Return array result\n",
    "    return array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to replace missing numeric values\n",
    "def replace_missing_value(x, col):\n",
    "    # We determine the locations of the question marks\n",
    "    QuestionMark = x.loc[:, col].astype(str) == \"?\"\n",
    "    # If there are question marks\n",
    "    if sum(QuestionMark) > 0:\n",
    "        # Convert the current column to numeric data including nans\n",
    "        x.loc[:, col] = pd.to_numeric(x, errors='coerce')\n",
    "        # Get the Nan array from the current column\n",
    "        HasNan = np.isnan(x.loc[:, col]) \n",
    "        # Calculate the Median for current column without Nan\n",
    "        Median = np.nanmedian(x.loc[:, col])\n",
    "        # Replace the missing value with Median\n",
    "        x.loc[HasNan, col] = Median\n",
    "    return x.loc[:, col]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Function to bin the data\n",
    "def bin(x, b): # x = data array, b = boundaries array\n",
    "    nb = len(b)\n",
    "    N = len(x)\n",
    "    y = np.empty(N, float) # empty float array to store the bin numbers (output)\n",
    "    \n",
    "    for i in range(1, nb): # repeat for each pair of bin boundaries\n",
    "        y[(x >= b[i-1])&(x < b[i])] = i\n",
    "    \n",
    "    y[x == b[-1]] = nb - 1 # ensure that the borderline cases are also binned appropriately \n",
    "    # (Negative numbers mean that you count from the right instead of the left. So, list[-1] refers to the last element)\n",
    "    return y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Determine Model Accuracy\n",
      "\n",
      "sklearn accurate split:\n",
      "Test size should have been  10854.0 ; and is:  10854\n",
      "\n",
      "Use logistic regression to predict if the person make >50K per year from other variables in Adults with non-binarized columns\n",
      "\n",
      "Confusion Matrix and Metrics of Class 1: income>50K = true and Class 2 : income>50K = false with non-binarized columns\n",
      "Probability Threshold is chosen to be: 0.5\n",
      "Confusion Matrix:\n",
      "[[8032  218]\n",
      " [1746  858]]\n",
      "TP, TN, FP, FN: 858 , 8032 , 218 , 1746\n",
      "Accuracy rate: 0.82\n",
      "Precision: 0.8\n",
      "Recall: 0.33\n",
      "F1 score: 0.47\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJEAAAIGCAYAAADk/R5zAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdd5wV1fnH8c9D70VpKij23hANdsReYu8VE6PRxMSYn0nUqCRqYhJNMSaWGANYo6KiERsqlqhRsIsVASlK7yywsOf3x3Pu7uzdW3fv7qV836/Xfe3eqefOnDlz5pkzZyyEgIiIiIiIiIiISC7Nyp0AERERERERERFZ/SmIJCIiIiIiIiIieSmIJCIiIiIiIiIieSmIJCIiIiIiIiIieSmIJCIiIiIiIiIieSmIJCIiIiIiIiIieRUURDKzSWYW0j5LzexzM7vTzLYpYBktzew8M3vazL4xsxVmNsvMXjazS82sXYFpOdjMhpnZF2a22MyWmdlkM3vEzM40s9aFLKeplruuMbOdzOxJM5trZlUxrwwsYv52ZvZjM3vOzL42s+VmtsDM3jWzv5rZHmnT943rmFTq39JUzKy9mZ1hZn82s//GYyuY2cONtL6hcfmDG2P5Upgs5Wry84sc8+5kZg+a2YxYVn1qZtdmK0fNbEyufW5mO8ZyOZjZvWbWokQ/U1ZDDS0DzGwPM7vNzD6O5fNyM5sey/4LzKxjKddXbrH+cnAso98ys5nxN0+MdaCtmzAtA+O2HNNU65S6zGxInvL7kxzztovl9aex/J4Ry/Mds0w/ONc+N7NWZvZonGaKmW1Vop/ZYGa2SfxtM81sVUPLgcS5bGDpUrl6qG85uTbUg0utvtskcawNbZyUldfqei5eXfPw6pqu1UWxFwrPAN/E/3sCewDfBc4ws8NCCC9lmsnMtgAeB7YFVgKvA1OBbsDewL7ApWZ2TAhhXJZldAHuAw6Pgz4FngOWAxsD3waOA64zs+1CCEsL+UGNtdx1kZl1AP4D9AHeAD4HqqjJM/nm3x94AOgFrADexPNJO2B74IfAD83sxhDCZSX/AeWzJXBPuRMhZTMCWJxh+AeZJjazg/HjrBVelk7By9FfAkeY2f4hhEzLy8jMdgOeBdYD7gQuCCFUFfULGomZBYAQgpU7LQLmN1NuAwbHQZOBMcASYENgEHAEfr7sH0KYXIZkNob98WME/Jz0GrAK6EdNHejkEMITZUqflM97wLsZhn+daeIYYH0F2BnPSyPxOtNJwNFmdngI4cVCV25mbYFHgMOAL4EDQwiTivkBjcXMmuHnt92A9/FjaCXwRTnTtaaJAY1zgHNDCEPLmxoREVdsEOmGEMKY1BczWw8PDu0N3A7UaZFkZr3wE2YvPAj13RDCtMT4TsCfgXOBMWa2Rwjh47RltAGexyts7wLfCyGMTZumK3AJ8HP84ipvsKexlrsO2x2vDL0SQtivmBnNbB88eNcS+CtwdQhhfto0/YDf4EGXtcki4C5gLDAO2BW/UGsslwM3kKWSK03u/wqt9McLkHvxsmhwCGFYHN4GeAw4FN+3PyxweXsDo4BOwF+An4QQQrE/QNZ+Zmb4BeGRwCT8fDk6bZr2wPnA1UBXPMi0NqgCHgL+GEJ4IzXQzJoD1+P1g+FmtnkIYW4jp+VN/Iac6iKrh8dCCEOKmP4PeADpaeC4EMIy8BYQwL+A+2M+WpJvQfHG3RPAQOBj4KAQwvSiUt+4NsEDSJOAXVeXmxOrMdXNyu9R/Cb4gnInpJEoj0nJNKhPpFhZ+ln8urWZbZZhstvwANJrwNHJAFJcxsIQwnfwFigdgLszLONaPNDzGbBfeqAnLmdeCOEaPKC1vMCf0FjLXVf1iX+/LGameHf7fjyA9NsQwo/SA0gAIYS3QwiHATc2OKWrkRDChBDCd0MIt4YQ3qSR81kI4esQwichhLX1JNmo4mMt5epP7jtAd+CZVAAJIF6InIe3jjg/Br9zMrNBeGC/E/CbEMIlCiBJDhfhAaTZwD7pASSAEMKSEMKfgP7AjCZOX6MJIbwQQjg5GUCKw1cBV+AtmLvg26ex07I0lt9fNfa61kbxUcz1yrTu9fGWayvxIOyy1LjYwuRZvJX/uQUsqwt+420gfhN0/9UsgAQ1dcLJCiDlp7pZ+YUQFsR9sFYGWZTHpJRKcSH0fuL/nskRsZ+Ao+PXH4YQVuRYzo/xR5h2M7ODEsvoDFwYv14aQliUKzEhhHEhhIp8iW7ocq2mP5O+WZYfUo9iZBpu7kIzG2feB9N8835JQlx2xsc3zKxPfK58jpm1Shu3qZn93bxfp2VxmS+a2fG5t0ZmZraZeX8Pk837sJpt3ufFwWnTDYy/NXVRe07id44pYFVnAr3xyPg1+SYOIbxaYPpPiM//jjfvt6PCzD4xsxvNrFuWeTYys5viPAvNbJGZTTDvb+CEtGlT/TeNNe/fa5l5fwQvmtnlhaSxHCzLM9HJ4Wa2jZmNiPu8wszeMbPTciyztZldbGavxny3zMy+NLMHYrAiffqC8lZi+up+EMxsfzMbHffpXDN7zMy2jNM1M7OfmtlHMd3TzOx36cdK2rK/bWaj4j5cEffhXZY5KA5e0Z8a89Eu2ZbbSFLl6f3pI0IIU4FX8WDsEbkWYmZHAE8C7YHLQwhXljidDRLzYEh8r9XfSGJ4ql+SIWa2uZndY96X2iozuyROk7MPAMvRx0bMT2eb2Qsxry2P+fovZtYzw+Jy/aYeZnaJmT1rXsYvM7N55v0Cnp1lnur+b+Ix9ivz8n25mU2N6eiYZd5WZnaF1fS7Ms3Mbjez7sWkO7UdqLlhNCT9ZlC6GBjPWwk37y/lCjN7Kf6e5bEseMbMjsox3/GxDEjN8415OXxj+u8zs93M7H7zcnxZ3I/j4zHer5Dfn+e3VlFTD+rd0OXlY1n6RGpIXonz72ved820xDZ9JZanLdKmbW3el2Wq/rLE/Bzxc/PHq9KXXd3PiJmtb15PmWpeRr9niXOLme0T9/+8uOxRlqPfTSu+3nMR8LX5Of14a9r+Lo/AnwB4NZbX6VLl+jG5FmJef3kBGIC3mjgghDCrlAltCIv9iACpLi72t5rye1JiuoPivns/HpepesNtZrZJkevsamZXx2XNM+9bcpKZPWVm38syz75m9rD5OWNFzPMPWhHndTMbEH/X0xnGPW5Z+seKx2OwRNlvGc5VcTueE7/+y2qfCweTxvycdYl5HSjV39ZQK/58lTy3bmhm/4rbZ5l5X3g/yjFvp1j+fBT3w0Ize93Mvm/eerNk6yrwt7Q0s1+a2WdxmVPN7GbLcLPNsvSJlFaGdY77b4rV1Al+ZRn6krR61qvN7Fvm1yLjrKYPvinmdZwdssyTXNdu5nXjmeb90x6bPk1ivtT2z/UZk2F9O5r3JfxVTN8c83r8wBz7Ylcze8K8nF5sZm+Y2UnZpi+E+fXELXH/Vpgf/++a2Q1mtkGG6Y8zrz/Mi+meYN7n7oZFrDPjeTgxvpB8VJJzodXz3G9mLczsXPO+eL+xmr4tXzOz68yfcMgthJD3gzdFDcDADOM2jOMCsHnauEvj8A8KXM8Tcfo/J4YdF4fNAZoVspwC19Wg5Sa2Sd8s44Nv3szDgVuBSrwicD/w3zj+7WzbOo6/PI7/W9rwQ/DHogLerHkE3ldFRRz2myJ/3154c87U8u7HKwSr4rBfJKbdBhiKX8AG/Hn3ofHziwLWNTJ9vxeRzr5x3kkZxq0E5gP/Bf6NP7YzM04/EeieIS/PTPyGR4AH8VZ0S4CnE9M2A15O5KEn8L61XsTvwC9rQN4cHJf7cJ7pBmbLZ3nmGxrnG5xl+M14Hz1fxHz0v0S+PSvD8rrhj+KFON/TeMvC1/BHLh6rb95KzDMmjvtT3K+vxX0zIQ7/Oqbj4ZiG5/EgyeI4/q4s2+L2OH5ZzL8PAR/FYfOBPTLMMwxYmNgmH+CPs/SuZxnya+Bv8XMZsF2OeebEeXbIMv6mOP7GLNtvMHACHrCvAi6ubz5tzA+wTyI/BmrKk6HA0MR0Q+L4++L++go/1v8DnJ8rv2fYNgPThrei5py0CD+2RyTy3FRgsyJ+05lxvi/xFgQP4I96ryRDmZ52jP83pnMBHsR8MpEHRwOWNl9z4Clqjskn4rExO64/VeZm3CYZ0rFrnL4KWK8e+zPjPsD78UqVA0/FffdmYr9flmFZ18dxy/Hj/D68zPk8Dh+QmPawxPZ9Ky5/JPAOXt7kPT8V+PtS5+1zi5xvUjH7IS1PjClFXonz/iqxzcfhZfIzMY8HoEti2rYx34a4jsficTEvDnsT6Ji2/MFx3Ei85ffUuC9einkqAGcAx+P1ondjfk0dazOAbhnSXXS9Bzg9piH1e+fh54F9M22bHPthSJz/CeD3wB344/ZHAc2zzPNHMpTPifE7xvGzs2y/MXjL/g/j9xeBDqXIw6X84OfiofhxGfB+MYfGz42J6b7A6wj/i/v7cbwMT9Wrts6w7DGkldf4zZBP4/BpMU+mytd5wCcZlnNlnH5VXP+D1NRjlgPfLvC3No/HwRKgVdrw+Yl8tlHafB/E4X0Sw4aSVh7EYV/E4a9S+1y4T5ymbxw/CX/cfWncTo8Bs+K4j4DW9cjfd+H1q6lxeWOoqatdlWG+HsD4OH4mXqd6Im6f1PHSvBTrypP+1DaZHJdVgZeFD1FTz/8I6JrlWBuaZfhj8ffNxOsZz+H1xwD8M0M6Uvu02Hr1aLyeNi6u8xFqyq2l+NMz2db1zzjvp3hZ/hxwZI48dmxavkp+UsfV82nrOgcvqwN+Pn0IP/esxMv072dI36DEtvogpu31+P3PZLmWy7Ofj6Kmnj8xpmMkNXnw2LTpb4zDK/H6w/14nSh1ntkxSz6alDZ8IBnOw0Xko5KdC6l/PfFeauq3T+N1qdHUlMG98m7/AnfSJLIENvBmt6kTeHoC7yHLgZVlPVfH6V9ODLs2tQGKyVgFrKtBy6XhQaS5+DPi6eMvybXNEgfGHolhvWOmWQGcnDb9Nom0Dirwt7XBO+sNeOFuiXEH4oVAFbBXIQdNAetLrevMeuyHvmQpeIATgTZpw9ri/Q4E4La0cdfE4bdkWFYHYM/E9/2puTBpmzZt80K3dZbflNqO5QoiBeCqtP3+f3H4xAzLGxXHPUfdwq0L3sy+oXlrDDUVvmMTw1vjgdiAV6zHkzgm8Up5KmDSN22ZF1Nz0bRZ2rjvx3ETgBYZfnNb4BS80rsikbYX8DKxUwH7YVJieyc/VXFfpOerjolpOmdZ5k8y5Z3E9nsSP8mvAr5T3zyaIb8U+qlznOZZR878TU3lMwD/AFoWmt8zbJuBacNTAblngJ6J4c2oCWS8XMRv2Rbon2H4VtSctAekjRuY+H2vkzipA5tTc5Gyf9p8qfPIBGpfpHSmJtifdZtkSON34/RfNDCvpJc5u5P5QvFb+DmtMi39bfCLgYVkCODhfc30SHx/Ma73xAzTbkiOgG0Rv21QXMcyYIMi551UzH5IyxNjSpRXTqKmTnJA2jjD+1lrnRiWqoSPI1He4x3zpy6M0s+tgxNpu5/aF9vfi8On4oGDUxPjWif24TVpy2xQvSfmvT/jF62ptE0ErgO2KWA/DEnMl/75FNg5wzwj4vgfZ1lm18QyOiSGp7bfeGqCpaNIO0cUmfcG5kh/rs/AhubVxPhjSDtX4i21rovzPZ1hnjHp6cAvaAN+YZYeoGhN2gU33qI3tb93SRv3bbzcmU+BAXM8mBCAfRPD9ojD3o9/z06M646f5z9PW85QctfNMpYT1NSDA35humViXA9qLpLPKWLfJfP37dQ+Zk+MwxcD7dPmeziRP9snhm9MzYXwL0qxrjzpT26TaWnbpGMiH92RNt9gMlzHULsMG5nMt/j5KhU82TTLvgsUV68+lLo3uo2aummma+7kuq5JH19IXkqbdgB+vq2g9s2Zfni5O4+69aY98WNnBYlzO/6CpOlx3VekzXMSNcHCSUXs401jvqjCn2ZqljZ+RxKNW/CAU4jp7pcY3gJv2BHwwGKzxLi+mdJFw4NIgdKdCwcmllnQuR/vry7g58lMN2j2Atrl3QcF7qhJ1C20ewBnxx+6ENg7w3ypO6G/LXA9qYNjfGJYasfeX2jGKnBdDVouDQ8iZbwDGrdrJV45Sr+I7B/n/SRteOpi5+osyzwhjn+kwN92dmo/kKGVFh5RD8C/CzloClhf6q7hofXYDxkP8DzztIvbeFba8L/FZR1bwDJSFe+iW08VsOzUdswXRNoD+CQ9PxSw/KHkrqi8lmGelvhFRq08j3eaGfAWDhkDGyXKW2Pi8HsyzHNM4rg6MMP4R0mrQOGBvhn4iT9jaxJqWqEcnec3rQdcgLdMS91JWIrfXTiKDEGoON9fYz7aFA9KbYm33ky10ro/bfpkq89sy0ydhJ7Jsv1Sn6KO0SzrOo/sd6+yfTLegc+xjozlaGL8kET+y3hHPlt+z7BtBiaGdcODAnNIu1MZxzfD38oUgJ1KsC1T++0PacMHxuGrgG2z5KGAP2KWHJ66aDg5wzw7JfJpxm2SYZ6fx+lfr+fvy7kPssyTCtT9IDGsexz2boHLSLUq7FJMeotIYzdq6gLX1WP+5/Ey/Lgi5knliTElyiupFhF5b+Lg5VTqru+ADOP7xXHLqN16aXAcvgBYP22e5tS0lrg7wzJT5fuLacNLUu+J6z845tFkC9O38AuTHlnmOyseFzviF6U98JZv4+L8s0hrnYrfHQ7AeVmW2TKx/g0Swwcnhgf8oqBBeZqaFuTFfvIG2PLl1QLnnRbzcnqrtjHULa8vi8MuKXDZqX1UpzVH2rHyowKXl3rq4prEsF/EYcfh5e2wxLhT4rjb05YzlIYHkQ7JMP6nFHnep+bcOokMLZioKTeS+2GT+FtXABtnmOf4OM831A6mFL2uAtKf3CYXZBi/Q0xrtrJqaNr0qeELyXzRnaovZtt3BderC/ht/43zbJ9lXRnr14XkpcR0m+J15CrgpLRxqWD42VnmTeW3PyaGper+H5I5uJVa5qQitkPqOL2zwOlTN5wzPe3QNubLAByeIR9NSpt+IA0LIpXyXJhKS8HnfvwmSiDtSZFiP8W+ne1Fq9tVz3TgWyGETK/sLOVrmUMJl9UUy83n0UwDQwgzzewp/G7IsdTu+yT17PSwtNkOi39HZFlX6rn0AQWmLfVmtftC5s4Qh+KtOPYvcHmFKvm+MLNt8Yj+5niT51Q/YCuAbmbWNYQwLw5Ldaz+O/M+QJ4N2V+Vnnoc4rtm9jleUW3SjviCd8Kdta+IBqjzbH8IodLMvsSDRhvgJ3vwxwnAf38hHfU1NG89m2FYquypxCuY2cYnn3fuh1f6x4YQsnUE/xIeBBqAtzjKKPgLBm4HbjezjYHT8OaoJ8fPbDO7KITwUNp8F6ct6nPgj2b2Av5IyKlm9scQwltxfCnK01fwxzbONrPXQgh31HdBIYQ7gTtLkKZSeC7HsVofB+B3fp5IlA/VQghVZvYKHpAZQO2+AbMys5bAQXGennEdhh9T4K2SMpkc0t5aGn0a/1Y/929mffAK4HL8rnB62t83s/fxVjvFaowyui1wOH6TpBv+GCHUvIWzepuEEGaZ2WRgZzP7I/CPLNslZSywHf7Gq+uBN0IIK0uU7tb4OXcT/Lj6VbHLCCEcWIq0pCkmr2yAX0ylgt757IafRyeEtE7GAUIIb5vZB3hgZQB1zyXjQghz0uZZZd5PTje8NWu6TOU3lKjeE7xz9OeA58zs+3jd6ww8T/4ZuNHM/h1CODNtvvSXwCwCnjaz5/Hz0F549wM/SEyTKsOzHUf5jq8v8FbRvYCHzeyokOicuxghhE/wC5qyMu/76Ej8OO+IX0iBtw5oBmyB17dySdXdfmFmc4H/hCxvSTTvM60ffhPx5SzLewl/w+kA/MZWPi/Ev4OoKQcG4cfVk/gNh2TfkKn/Xyxg2cVIdZGRrs6xX4QXQgiZXvbyKV52JJe5L57HXw6ZO/5/FL947glsjQfQ67uuYmTqQ/JDM3sP2AVvSfRMgcsaG0KYnSWNR+VIYzH1asD7UcTLo+3wVsSpa/Ze8e9W+I2SdCOz1K8LYt5p/5N4HfkXybprvDY6GL/+eSzLIjKVvak6/f0hRjHS3I0HGYuRuv64K9+E5v1V7RW/3pM+PoRQYWYPUnP98VSRaSlWKc+FKQWf+/FjbzFwlJldiV+TTSw08SnFBpGewSN1zfC3LuyL/5gHzGzfULdD69SB1qPA5aemSx6gqf+L7gw0j8ZabqFyvVllOF5wnEUs/OIBcCoeFU4/APrGvx9mCPIlFfpbN4p/s2Wo1EV3TzNrUYJK+Wy8aXrJ9kXcXrfjb7PKpRPetBE8OLc/3jR6BLAqnmRexFvAvJuaKYTwhZn9GG/afwtwi5lNwFujjABGZSko1wSZOvwEL3DAL3xTNo5/Pytw2Q3NW5nSlnoV8jfxgiDb+GS6+8a//S1DB/hpCs6XIYSvzOz3+CND1+F3CLoB2+PPaheyjHfN7An8hHoEfkcc/CIlpT2ZX0HbPsO0SXfhJ/6bgNvMrDKE8K9C0rWaK/WbqvrGvyeWKn/EDhFHkj1QBF4eZVLMMZk6xqbkqEhOorggUqOcL81sb7wvklwdWqZvk7Pw/k5+AvzEzGbhx9sovCK0NDHtL/BA+2Hxs9TM/odX0IaFer7NKp5fHsSD4u/g/adU1mdZjaA+5ffEAtOfr/wGL8N3TExbSNqW5BifqfyG0td7iAGZh8zsNfzFKz/H68n7FLGMSjO7Ab/xkP6Cg1S53CHL7MnhmcrwaTFdL+KPfz9mZsdkufBe7ZnZdfgxWqez5YRsZWK1EMKLZvZbvPP/YUAws4/xYN6/04JFfePf7iU897+Hl5EDzKwd3sJ5b7wD9RXxxtClZrZVCOEz/CYFlD6I9E2W+nimY79Q9Tn3ZCwfQgipjtV3jtOmB5EKXpeZpcr2pNkhhP9LGzY/hLAwy3In4UGkYl6IUMz2qPd8ZnYRXk/L1blxtmOj3vWheKPrYfzx+ztDCL9Lm2R9PNgLsKCIsjeVNyZlmTbb8FyKuf5YH9/GK/ByNJPU9Uemc1eplfJcmG+ZdfJYCGGReefqd+LXKteZ2TT8GnYkMKKQa/tig0g3hBDGpL7EVh4v4lHU3+CVuqR38Ds6uxe4/P6J+VLejn93NbNmDYmupmms5aYitTllCLglPYEHNg4xs14hhG/wu2Ld8c7NpqRNn1rfffidiIbK1+IhZPm/vt7GC/H+eEdfpXAJHkCahufL14GZIb4h0Mym41HZ6t8a88DgGAQ4Cg8A7IXftfqpmV0bQrg6Mf3fzGxEnPZAPKh6bvw8b2aHr0YXFsWoz7FQaD5oaN7KlbZi0p06Zr4if0Xuf4Us0Mx2xDttPZWaiuok/LgsNlCTunNQfXEdQlhoZvPwfjP6kDmIlKoMTc624BDCH+NF8O+AO2Mgqc6dmXzM7DyKuLiKMlXySiHvGzlzyFRep4aNpyaIl02mO4GZPIwHkB7Dt/2nwMJ49+kQ/CZNtuOjMY/JQqTOl5ubWZcQwvyGLtDM2uOdhfbA+7O6Fe8zY3Fs6XU+fiOg1jYJIbxi/jbGQ+NnH/yRkeOAq8xsvxDC5Djt12a2Z5zmcDzosw9+EXeVmZ0UQniyyHQ3x2/kHI3nj0MKbIXZVMpZfudbZr601acML0m9x/xNTSfiZfh+cflL8eO2weV3lCqX+2SZL1V+z83WsjKE8LGZHYiftw4FRpjZ8SH324/riEHtXxQzT3RDbMXUIGZ2It659UK8vvYiHgRZFse/hvevUlCeCyFcYWZ34DdgD8DrYxcBF5nZ8BDCOXHSVL6Zi9e1cynod8bgyEv4I5T74K1A2+GPq4K3DroUONDMluKtLD8KIcwoZPlFKNl1TD2X2djlQ9Jh1G2xPhnvZyjfelLytQzMpL7buOD5zGx3/Ob0SjzfPIkHCCpiXrsPb/GebXs3pD50G349M5qaN5gnJZ/mqNPCK02mFlulrJeEtL+5NDRvFitfHKCU58J6zRNCGBFbzh6Jt5DfF89XpwEfxMZBOes2xQaR0hPwsfmrMx8HfmBmt4QQJiQmeRL4A7Cjme2SbMmRLjYxTTVN+09i1Iv4iXx9/IRZqiZmDV1u6oSd6Y5StgpCQUIIy83sAfwAPh1/o0e2R9nAOyreEu8bYEKG8cVKRTM3zTI+9erzGVlafhTrP3iF/CQzu6xEjxucGP9+P4SQzE/EO0W96s7iQgjj8YuD38eL7RPxx6x+aWb3J5sLxgBf9WM9ZvYtvGA9EA9i3V6C37I6S1WKty5w+qbOW9mkArFfhRAG13chZtYXL3BPx5tbgweA78AvNF+tZ4u09eLf9AuJd/EKcj/82fJ0/RPTZRVC+H2843QdMNTMVoYQHigyjftQ8+rhQmWq5DW2XGU1ZC6vU/nj7Ybkj5R4wbY93sfAiRny9hYNXUdC6i7bxjlukPQtcpnv4dukD57fb61/8qrtiweQxoUQzs8wPus2ia2NHo0f4mOkt+GBohtiGlPTVuF3116O03bCHzP6BV5uF/yIhPlt17vwPk2+AA7K8mjDmiJVfm9mZi0LuOmRr/xOjqtXK68iNLjeEx+lPBovvw/DH6WswuuHd+N3Y+vzmGy28jt1g7RflvkKLb8/MrOD8ODEkcC/Y0C0mLpTL4ovv8HrQg0OIlFTR7syS2vYosvEEMIkvP+Pv8Zj9WC8Pna2md0XQniGmrJ9aSnK9oQX8CDSIDyIBDVBpJfxgMAgaloTlLoV0uogZ/kQ90nf+LVB5UMIYWCBk3Y1s05ZWiNtEv9ma5lSLifgQY+bQwh/yjC+lPWFamZ2OX7d8hFeT8lUnszBg1St8L6mCm0FmdrGfbOMzzY8l6/wa4+t8Tc25zIbPy5b4y2NMrXaKebcVZ965Won3hC8N34ws+3wOEN/vI50ea7587aYKSABT+CFYUv8rkJy3CfURPr/Gi9asvkTvnPfCSFUPwsYf+Bt8etNZpZthwFgZrvFikG+dDd0ualMluni+ZAMw4o1PP49Kz6f+m28QvJIhmlTz9qemGFcfaSa/p6epVVVquLxUoZx9XE3vj03wt8okFN8BPxjIycAACAASURBVCKfVCUuvdUW5I7g1xJCSF1cvxzn2SHP9P+jpp+YnQpZxxou1UfR8fHiLJ+mzlvZvIXfidzDvA+ZgplZdzO7yMxexZu//ga/mHkEfwStVwjhghDCK/UJIJlZG7x1G9T09ZCS6pvp1AzzbYQ3oa/EH+3JKYRwPd5/Q3PgbjM7oZh0hhAGhxCsyE/fYtYRf0vq8aH6ylpWx+DOxunD8QuCSuCwfOeGAqXKo6+zBEfr7M/6ii1VJ+Ln0zp9DJjZDhRZNsVAzB/i11+Z96WTlZltlm8acpTRZtaKIvpHCN7/xnXxa87fFi8mrsArlL3iDaxC3Yrf0JmMd+LfpP3glVq8CfIB3mripAJmGYdfBG9uZnX6GjKzXfDtvwKo02dSidWr3mNmLczscDO7G39N9wN4IOljvIPmPiGEg0IIw+oZQIKabZlefo/Cgwn7xvI63Snxb7a+RqqFEN7HgyTziP1nxlZyBQkhjKlH+W3JpxEaKNfxfyANfHQ2uGep6TNrpzh8Op7ne8ebfqWS7BdpEL5f3onrXITnhQPwG4zJ6QuRumBt0I3/JvAK3opjvxjYT3cM3rfPDGpa6zWFU9IHxIvlnfHzQEGtzZtQrmNjG2DXUq/QzE7GX2YxAzgyWwuUGFgajdcbjy1iFak6/amW+Rm4M4pYVkrq+mNwvgljulOBpjPTx8d698nxayHXH6l65WZZYhuliAM0udiIIhW4zFtPbHAQKboq/j0r3plPuhA/Ue+DP79dq2JpZp3M7E48Ay3B+ztI90v8Tui2wEtmVucujpl1NrNr8L4RCn3utyHLTd1F+L/YLD81/W7AtQWuP6vgnVZ+ij+ve21c94gQwpIMk9+EPz8/xMy+m35xbmbNzGyQmR1a4OofwqPG2wJXJw94MzsAfxNVAP5S5M/KKDZfPh2/aPulmf3JzDqnT2dmO5t3On5ZAYtN3Sm7MC39uwC/zTSDmZ1tZnUKZzPrje8HiM8ax+15ePrFbbz4OSQ5bWMxsz3M7BMzK8VdwXoJIbyNtzhcH3jQzNZPjjezLmaWbHLcpHkrR7or8YvOVsBIM6vTP0xM+3lm1jNt1HT8TX574eXCBXjg6IQQwqOFPFZgZseb2cAMwzfEK70b4RWI9A7478LvqBxuZmcl5muNPxLUAn+OfQ4FCCEMwSsOLfCLkKMLma8Jpe5ebduAZaTK6rPMbPPUwLhf/0mG82C8uL4V78/qUTPbLH0aM+tlZj8uMMD1Od7CYQcz2zexDIt3//bNOmf9pDqD/W0sv1Lr6wT8nfp10v43/MK9O/BqPF5rMbO2ZnYxHmxIP27SpcqtQbFinFpGS7xD483TZzCzTeI5rmP6OPxGCyTKXTP7afL3JxyCn1MX4m+6ysvMbsSP9Wn4a+MbXL6b2fOxDD+uoctqgF/Hv381s/2SI2L+PCSWL6nH8FOta28xs26JabtS00LtX6EEjzzmUd96z614MOdM/EL/d8AOIYRdQgg3hgL6yYr58IL0ALOZNTfvK/HHcVCt81hstfYvvLy9I7Vd47xn462hZlLg43MhhHfwvLwAD6bdXUwgqcxSx//3Yr0JqG7dW1RLRzM7zsz2Sb84jfXIVNmaPF5TNyvvT6ufpOZrZ2anmXfZUZB403w63srsW/hbm5KtQF/A60kn4+eCYm6UleI82OiCP0b8KN6o4Na0a6M+1NyI+HM9W2jX1zVp5/4O+ONihr8Fq7HLqmKljo2zk2VMrF+nyo+SMb8hMAx/U93RcT/mci0eDP+7mdUJJJlZSzP7tvmj5CkPA1/j/eX9LG364ym+U23wp3SWAOeZ2Q8yHP87pNXdUsGRnyWv92OZeRPeOvNjCuhkPbZ6nIR3L/GjxLLMzK6gphPv1ZKZ7WpmJ5sHz5LDDW/dCgVcw5YkI4YQ/mtmT+MnwCuA8xPjpsdK8xN4J4OTzZ91no4XqPvgd8Gm46+5rdO/RPBe0wfhzVIPAcbFC+fxeIS+D/6685b4HdiCmtc1cLl/wyuUA4BPzTvr7BWn/wN5moAVaDh+gffD+D3To2yEECbHg/AhvBXMEDP7CK9Y9MZbSXTHK0uFHBwVZnYKXtG6BjjFzN7Bn+/fF7/oujyEkK/5YMFCCC/Fyt4D+PPxF8ZtOg1/9eJ21HRIm97RWyY34PnxAmCgmb2LPzqxH94h6l7UNGVNOR4YZmZT8ODigjjPvnjndg/GlkbgEdo/AfPNbBweve+AP8PfHe/oreBH2czsUWoeq0jdgTvAzJJ3cy+KQZuUdhT+GFljGoznq0Px4/sV/MJsY/yOybPEClM58lY2IYQ/mdmm+NsY3jbvRH0Cvq/74BW2VvFvsu+Cz4jNPws42WazH/DjRF5bRM32ahfXd2xIe/NO8H6RzsTL0+HmbxOaiufn3nFZPy8mISGEX8YL95/hncoeF0LI25KpiTyK92n2vHnnpIsBQgjnFbGMl/GOlA8G3jGzl/Ey/Vv49nodP27TXYZv0+OBT2I+nYR3Krkx3qlnc/w4z/kYSfC3it2G99HxopmNwV/l2g8PltxIaR/z+yt+PB4W0/48fk47AM9rj+OtLwoWvJ+i4/Bg5ZnAC+adpL6HPxreCz8ftsUvhDO+HSmxvLfNbBReL3g37t9F+L5YP/6G9DcYdsXPcX9L7I8WeHm8FZ4/kq1ZrwL+YGbj8Yr5CrzJeqoFwuWhgH7rzOwY/LXF4HWBqy1zh6KvBn9rYaE2x89DdW6aNJUQwsNmdi2+rV6K57PP8LvhO+AB7a7U1H9+iddz9gEmxP1Wheetrngrz0Ju9DQ03fWt9yzBg/H34Bf69bmY7Yq3ZL/JzMbi9dcu+AVSb/xGyFVZytHL8OPkCOCLWB/ujZfhK4AzimkBFUIYG+tOz+KtrCvN7NxQwn4+G8nNeMvjI4HPY32vI94X5Zt4+Vjohdj+eOBuppm9jT9y0xXPo53wV6JXt+IPITxqZj/HbyiOieVDqnPePnjZ3h5/PDbX2x/TvYjfEG9OzaNsKS/g10dt8CcucpaPaUYCVwOXmLcknYrnsbuaoq5UpAvxOtMRwJfmfUW1wVtntcdvOt7YhOn5Cn889MN4HlyK57Hu+L79WfZZy+ZfeL2nH17Gvoqf5w7Ay5rHKK4VUD7X4/voE7wPsYsyTPNJCOEGgBDCW+adMv8Tv8k2Ad+Wy/GybBv8nHYhXr8ihLDU/Mbnk8ANZnYG3iVDX/ycfzOJYEwhQgiTzOxU/M2it+B9176F3yDaEr9uPI7YYXYI4Qkzuwk/l/8vUQ/7Ft6VxizglCLKzl/i55Ebzewk/LjcGb+Wy1R/WZ1sgm+3JfGcPw3PA/3xMnAG8Pu8Swkh5P3gFbUADMwxzW5xmhXAxhnGt8SDS8/GxK3A76i/glee2xeYlkPxx58m4JWBZXgh8SheeLcqZDmlWC5eAXwIryxX4E1Xz43jgm/eOvNkHJ4lTX3w1ygGvPm85Zl+Q7zC9D5emV6KHzzP4ifYDYvcJpvjlbMpcX/NwS/+D8ky/eCY1qH12QdxGR3wINLoRD5ZFH/TLUD/tOn7xnVOypInn8IvaJbgFzuX4IGKVJ7um5h+P/zO4Vtx3cvjbx+NP27SPDHtFvijQC/GaZbF9bwV83OnIn93Kj25PgPT5hlYTH5KzDc0zje4kOGJ8WMypSOOa4Of9P6H392viHnv3izTF5u3cq07ax6I44fE8UOyjD8ADyxOjft8Lv5M+FD8BNSyvvk5xz4YiF/IvIefuCrxi5+3Ynq75Zl/Z/zOzqyY9z7HW1a1y7P9Mu7bOM1NcZpl2fZDU3/wgMRNMS+tSM/v+fZtYrr2eNA3tY8nUlNxypq34rzH4UG7b2IaZsX9ditwaBG/pRkeRHoPL4/m4P3B7UnNsTwmQz6pMzwxfjBZylw8AHolfnG0HL8L+E+8hdDQfPkhz28ZgPf79QlePq+Iy38K+D7QIW36jOvDK3tX4TduKvBy99/4RUid34ZfZP4Ev6iagJ/nFuDH659IlOdx+jPxmy8f4a1OluJ9GT0A7FXE702lJd+nzn7Is9xJxe6HxsgrcfwBeH0nlc+/xoP/lwAtMuy3n+Idri+J2/U9vA+FtvVY9xjqX76XtN5T4D5YH69gjyF2eIuXmxPwm38D8szfHi9/Pqem7vAwsFOefZdx38Zp9sKPxYCfW3PWF5viU0Ce3IKa1skVeHkyJOavjHki03C8pfjv8MdVplNT3r2KX3e0zrL+fnjZNDHuhwUxDQ/i9f6CrksSyzuXmrJg27RxbeI6AnBjlvmHkqU8wFuavZHYx9XTFXCM5NwPWeYZQu56U660dsJbOKbK9UUx7ReSqEeXYl050l+9TfBz4RC87F8e89tfgfVyHGtDCxme7zfkS3uOfN4LP19PivlmIn6O65JtmYVsp0zTJNKQ61Mn7+CBmr/hT80sxcvfz/GbVN/Lsn374fWeBfi54008+F29v4o55uIyt8Bv5n1JTR3+HbyriV4Zpj8eD/LOj9N/iV9fbpQrH2VZ98l4y+tleB3jMTx4Vd98lDE/5EoL9Tj3x/x1Od66fBJ+nM7FA66/BnoUsu0tLkxERERERERERCSrUvWJJCIiIiIiIiIiazEFkUREREREREREJC8FkUREREREREREJC8FkUREREREREREJC8FkUREREREREREJK8W5U6AiIhk161bt9C3b99yJ0NERKTJjBs3bnYIoXu50yEiInUpiCQishrr27cvY8eOLXcyREREmoyZTS53GkREJDM9ziYiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQiIiIiIiIiInkpiCQikmBml5jZQ2Y20cxC4jO4HsvaxMzuMLPJZrbczGaa2Ugz27sRki4iIiIiItKoWpQ7ASIiq5khQOeGLsTM+gGjga6Jwd2Bo4GjzOw7IYRhDV2PiIiIiIhIU1FLJBGR2j4A7gIuAmbWZwFm1gK4j5oA0ig8eHRT/N4MuM3MNmtYUkVERERERJqOWiKJiCSEEPZN/W9mP6/nYg4Hto7/LwRODCFUAE+Y2c7AQUAb4ELgsgYkV0REREREpMkoiCQiUnqDEv+/HQNIKf/Fg0jp04mIyBqgclUVyypXNXg5AZg6t4LKVVXVwxYvX8m0eRW0auEPC4ydPJeu7VphDV5baUycs5QJMxcT4vfFM5bQoWd7mjeD/1y8b855RURk7aAgkohI6SUfU/smbVzy++ZNkBYRkTVaVVVg+oIKKleFWsOnzaugonIVCyoq+WruUt75ah69u7bLuIxXPp9Fr05taN6sJhzzv4lzAejesXXBaZm1aHk9fsHaacFr05n/6nS6H7M5Hbddr9zJERGRJqIgkohI6bVP/L8ibVzye4dMM5vZ+cD5ABtvvHFpUyYikkMIgQUVlVTVjtfw/tT5fL1gGc2tbpuYCbMXM2XuUr6ctYQ+62UO4hTqufEz6N21La1bNKNixSqmL1jWoOUlTZ1XkXG4AkPFCSGw4NXpLHjtawBmPf4lLdq3LHOqRESkqSiIJCJSeksS/6ff4k5+X5xp5hDCHcAdAP379w+ZphERKVQIgZVVgUmzl/DR9IU8+s40mjczurZrxbLKVcxevJwFFZXMWbKiwQGVT75Z1OD0Zgv2rG7atmxeq2VTfSxevhKAnXrXvBR02rwKOrRpwS59ugAwYdZi9t68G+1alb/aHkLgybveY3QMIAEM2Ks3Nw85sIypEhGRplT+s5GIyNrny8T/vdLGbZD4f0ITpEVE1nILKipZumIlr30xh2fHf8Ok2Uvp0ak1zcx46bNZ5U5eyXRt15LObWtavFQF+GruUo7eeUPWa98KgIXLKtltk64Z569YsYotenSgdYvmtYZv2q09xcSCmjUzunUo/BG4tUUIgUsvfYbRD4yvHnbYYVvwyCMn07atWiKJiKwrFEQSESm9F4BL4v/9zKxdCGFp/L5f2nQiIgWbuXAZMxct5/OZi/jbixNYUFGZsfXQpzMa1iKodYtmtGtVE2xp0bwZVVWBzbt3YJP16z6ytrRyFeu3b0Xvrm3ZZP32dcYXa7Nu7Uk9OdezUxs6tG6BZXiUTppGVVXg4otH8fe/j60edvTRW/PggyfSurUuJ0RE1iUq9UVEEszsECB1hZS8UupnZvPj/6+GEGab2VDgnDjsVyGEIfH/p4DPgS2BjsDDZnYr/ja2/eM0y4HbGuVHiMgaacrcpYz+eAafzVjMjIXLqFixilmLl7N42UoWL19Z/ehTfXWMj0i1admcz2cs4thdN2LTbu1Zv31r2rZqTofWLejZyf9Pb60j666qqsAFFzzBnXe+Uz3shBO25b77TqBVK+UTEZF1jYJIIiK13QFskmH4xfEDcAAwJtsCQggrzex0YDTQGTg8fqonAS4KIehxNpF1xMyFy3j7q3nMX1pZ3Y/OZzMWMXH2EkKAj79e2KBOpLt3bE1zM+ZXrGC3Tbqya5+ubLdhJ9q2ak5zM/bcfH1aNm9Wqp8j65Bf//qlWgGkU0/dgbvvPo4WLZSfRETWRQoiiYg0ghDCWDPbFbgSOBToCSwEXgP+EEJ4pZzpE5HSCSEwZW4FK1atAqBiRRXvT5vPO1/NZ/KcJcxbWskXMzP2o19v22/Yic9nLOaMARvz88O2oU1LtQiRxnHRRbvz4IMf8fHHszn77J25666jaa6ApIjIOstC0It/RERWV/379w9jx47NP6GINKrKVVXMW7KClVWByXOWMm/pCp768BsWL6vkxU9L03l1i2bGLn26sM0GHdlhw8706NSaDq1bskHnNrRo7q2X1mvfSo+aSZP7+utF3HbbWK65ZiDNGvhGukKY2bgQQv9GX5GIiBRNLZFEREREoqqqwBtfzuGDaQv49JtFfDBtAZ+XoBVRq+bN2LpXRybOXsK2G3Rk4/W88+kOrZvTZ712MXjUiQ7qpFjKLIRQpxPzDTboyK9+dUCZUiQiIqsT1VRERERkrTR3yQqWrvDOqFdVBSbNWYoBVSHw0fSFAMxatJyqEHho7FRaNDMWNaDz6g07t6Ftq+Y0M2OLHh3YdeMubL9hZ9q2as7WPTvSXgEiWc0tW7aSU055mHPO2Znjj9+23MkREZHVkGozIiIissb7MLYcqlxVxXPjZ/D8JzMbZT2d27akclUVS1es4pDtelJRuYr9t+rOgM3WZ4eNOjfKOkWawtKllRxzzAOMHv0lo0Z9ziOPnMy3v711uZMlIiKrGQWRREREZI3z1ZylXPfkeJ4dP6PR1jFw6+7stFFnNuzSlp16d2HbDTrWecxHZG2wePEKvv3t+xkzZhIAK1dW8fbbXyuIJCIidSiIJCIiImuEEAJPfvA1f3txAh9/vbDg+Tbq0haAZZWrmLNkBftu2Y2qEPhg6gLO3XtT1mvfqnr5g7bpSY9OrfW2M1lnLFy4nCOOuJf//ndK9bBrrz2AX/5yvzKmSkREVlcKIomIiMhqZfKcJbw2YQ6fz1jMeu1b8ti70/lmwTIWF9Bf0Ta9OrJNr44sWraSc/bqy75bdlPrIZEs5s2r4LDD7uXNN6dVD/v97w/issv2LmOqRERkdaYgkoiIiJTdK5/PYthrkxn9cXGPp+24UWd+f+JObNa9Pa1bqPWQSKHmzFnKwQffzTvvfFM97M9/PpQf/3hAGVMlIiKrOwWRREREpMlVVQVe+GQmr34xm6GvTSp6/ltO35VDt+9Fy+bNSp84kbXczJlLOOig4XzwQU0H9LfeeiTf/37/MqZKRETWBAoiiYiISJNZuKySnX/1LCHkn7Zbh9b06tya/bfqzoZd2jJomx5069BagSORBli+fCWDBg3jo49mAWAGd955NN/5zq5lTpmIiKwJFEQSERGRRnXhPeN46sNv8k8IbL9hJ64+ajv22HQ99WUk0ghat27BD36wOxddNIpmzYxhw47lzDN3KneyRERkDaEgkoiIiJRMCIGqACurqvjsm8V8+5ZX885z7C4bcny/3uoEW6SJXHjh7lRWVtGzZ3tOOWWHcidHRETWIAoiiYiISIN9vaCCQ//0MguX5X+DGkDHNi145MK92LJnx0ZOmYhk8qMffavcSRARkTWQgkgiIiJSL09/+DU/+fd7VFSuKmj6Zy7Zjy16dKB5M7U2Emkqn302hyuvfIG77jqajh1blzs5IiKyhlMQSURERIr2z1cncu1/xuecpnvH1sxbsoK2rZrz6s8H0bltyyZKnYgAfPzxLAYNGs433yxm5swljBp1Ou3btyp3skREZA2mIJKIiIjkFULg/jencOOznzJ3yYqc074/5BA6tVHASKScPvhgBgceOJxZs5YCMHbsdD76aBZ77LFRmVMmIiJrMgWRREREJKe3Js3lpNtezzp+q54duOqo7di973q0adm8CVMmIpm8887XHHzw3cyZUwFAhw6tGDXqdAWQRESkwRREEhERkYxCCJx/9zieGz8j6zQn9OvNTSfv3ISpEpFc3nxzGoceeg/z5y8DoFOn1jz99BnsuWefMqdMRETWBgoiiYiISB3/fusrfj7ig4zjfnLQVvTv25V+G3elbSu1PBJZXbz22hQOO+weFi3yR067dGnDs8+eye67qwWSiIiUhoJIIiIiUsvId6dlDCAd328j/njyLmVIkYjk8/LLkzniiHtZsqQSgPXXb8vo0Wezyy69ypwyERFZmyiIJCIiItWeGz+DHz/wbp3hw76zB/tv1b0MKRKRfFItkCoqVgLQo0d7nn/+bHbYoUeZUyYiImsbBZFERESEylVVbHnlU3WG79ynC/84ezd6dGxThlSJSCG22aYbW221Pu+9N4MNNujA88+fzbbbKugrIiKl16zcCRAREZHyWbiski9nLc4YQOrZqTUjf7C3Akgiq7n11mvLc8+dxWGHbcFLLw1WAElERBqNWiKJiIisgxYuq2SnIc9mHT94r74MOXr7JkyRiDRE9+7teeqpM8qdDBERWcspiCQiIrIOmb14Of2vG51zmk+vO4zWLfTWNZHV1b///SGrVgVOP33HcidFRETWMQoiiYiIrAP+8fKXXD/q45zTbNWzA89csh9m1kSpEpFi3X33ewwePBIzaNWqOSeeuF25kyQiIusQBZFERETWUp/NWMT/PfQe709dkHO68b8+lHatVCUQWd3dddc7nHfe44Tg36+77mWOPXYbWrRQN6ciItI0VGMUERFZy4QQ+NPoz7n5+c9zTvfBkEPo2KZlE6VKRBri1lvf4qKLRlV/32mnnjz33FkKIImISJNSEElERGQtEkJg08tHZR1/Qr/e/Ob4HdTnkcga5C9/eYNLLnmm+nu/fhvw7LNnsv767cqYKhERWRcpiCQiIrIWGD99IZc++C6ffLMo4/g7ztqNQ7bv1cSpEpGG+v3v/8vPf17TGf4ee2zEM8+cSZcubcqYKhERWVcpiCQiIrIG+3LWYgbd9FLW8b8/YSdO6t9bnWWLrIGuvfYlrr56TPX3vffuw6hRZ9CpU+vyJUpERNZpCiKJiIisgcZNnsuZd75JReWqrNOM/MHe7NynSxOmSkRK5ZprXuTXv365+vvAgX154onT6NChVRlTJSIi6zoFkURERNYgIQR2v/55Zi9ennH8Ln26cO0xO7Bj785NnDIRKaVNN+1a/f9BB23GyJGn0q6dOsIXEZHyUhBJRERkDbF85SpOvPX1jAGkLXt0YNh39mDDLm3LkDIRKbXBg3dhxYpVjBz5KSNGnEybNqq2i4hI+VkIodxpEBGRLPr37x/Gjh1b7mTIamDqvKXs87sX6wz/yUFbcfGgLWjWTH0eiayNqqrCOnd8m9m4EEL/cqdDRETq0i0NERGR1dj46Qs54uZXMo57/Id7s1Nv9Xkksqarqgr87nevcuGFu9d569q6FkASEZHVW7NyJ0BEREQyW1BRmTWANP7XhyqAJLIWWLWqinPPHckVV7zAYYfdw8KFmfs7ExERWR0oiCQiIrKaqaoKDP7Xm+z8q2frjDtzwMZ8MOQQ2rVSY2KRNV1l5SrOPPNRhg9/D4D//W8at976VplTJSIikp1qoCIiIquRd6fM57Q73qCiclWdcRN/ewRmerRFZG2wYsUqTjttBI888nH1sO99rx+XXbZ3GVMlIiKSm4JIIiIiq5Fj//bfjMMVQBJZeyxfvpKTTnqIJ574rHrYD36wOzfffLj6QBIRkdWagkgiIiKrgZkLl7HXDS/UGf7WlQfRvWPrMqRIRBpDRUUlxx//IE8//UX1sEsvHcCNNx6iQLGIiKz2FEQSEREpsze+nMOpd7xRZ/h7Vx9C53Yty5AiEWkMS5as4JhjHuD55ydWD7v88n24/vpBCiCJiMgaQUEkERGRMlq8fGXGANI7Vx2sAJLIWmTRouUcddT9vPzy5OphQ4bsz9VX768AkoiIrDEURBIRESmThcsqOfzPr9QZPumGI8uQGhFpTFVVgYqKyurvv/nNIC6/fN8ypkhERKR4CiKJiIiUwdMffs3373m71rBDt+/J7Wf1L1OKRKQxde7chmeeOZMDDxzOGWfsyE9/ule5kyQiIlI0BZFERESaUAiBTS8fVWf4flt1VwBJZC3XtWtbXn/9u7RurSq4iIismZqVOwEiIiLrkqtGfphx+LBzd2/ilIhIY5oxYzEjRoyvM1wBJBERWZPpLCYiItJEfjvqY+5546taw0ZcuBe7bdK1TCkSkcYwffoiDjxwOJ9+Opvhw4/jzDN3KneSRERESkItkURERJrA2Elzuf3lL2sNe/PKAxVAElnLTJmygP33H8onn8wmBPjudx9nypQF5U6WiIhISaglkoiISCObMGsxJ972eq1hr18+iB4d77XCwgAAIABJREFU25QpRSLSGCZNms+gQcOYOHE+AC1aNOOee46jT5/OZU6ZiIhIaSiIJCIi0siueqx2P0h/OXUXNujctkypEZHGMGHCXAYNGs5XX3mro5Ytm/Hggydx7LHblDllIiIipaMgkoiISCP6z/vTeW3CnOrv391nU47ZZaMypkhESu3TT2czaNBwpk9fBEDr1s0ZMeJkjjxyqzKnTEREpLQURBIREWkkB/3xJb6Yubj6+wFbd+eqo7YrY4pEpNTGj5/FoEHDmDFjCQBt2rRg5MhTOeSQzcucMhERkdJTEElERKQR3P36pFoBpI26tOUvp+1avgSJSMm9//4MDjpoOLNmLQWgXbuW/Oc/p3HAAZuWOWUiIiKNQ0EkERGREps+v4LfPf1prWH/OLs/ndq0LFOKRKQxfPbZHObMqQCgQ4dWjBp1Ovvuu0mZUyUiItJ4FEQSEREpkQUVlez8q2frDL/iiG3YbsNOZUiRiDSmE0/cjqFDj+HHP36aUaPOYMCA3uVOkoiISKNSEElERKREDv/zy3WG7dynC+fts1kZUiMiTeGss3bmqKO2omtXvXFRRETWfs3KnQAREZG1wZDHP2L6gmW1hu22SVduP3M3mjWzMqVKRErp9denMGfO0jrDFUASEZF1hVoiiYiINNBVj33I3W9MrjXsi+sPp0Vz3asRWVuMHv0lRx99P9ts043nnz9bgSMREVknqXYrIiLSAKuqQp0A0v3fG6AAksha5KmnPueoo+6jomIl77zzDeeeO7LcSRIRESkL1XBFREQa4OL73671/d7zvsWem69fptSISKk9/vinHHvsv1m+fBUAvXt34g9/OLjMqRIRESkPPc4mIiJSDyEENr18VK1h/Tbuwt5bdCtTikSk1B5+eDynnTaClSurAOjbtwsvvHA2m27atcwpExERKQ+1RBIREamH34z6uM6wK4/ctgwpEZHGcN99H3DqqQ9XB5A237wrL700WAEkERFZp6klkoiISJFGjJvKP16ZWGvYeftsym6brFemFIlIKQ0b9i7nnjuSEPz71luvzwsvnMOGG3Ysb8JERETKTEEkERGRIjzy9lR++tB7tYY995P92LKnLi5F1gb/+Mc4LrjgP9UBpO23787zz59Nz54dypswERGR1YCCSCIiIgV67YvZXPpg7QDSmP8bSN9u7cuUIhEppVWrqhg+/P3qANLOO/fkuefOont3HeMiIiKgPpFEROows6PN7Dkzm2tmy8zsczO7ycwKfuWWmbU0s4vN7NW4nJVmttjMPjSzG82sZ2P+Bim9D6ct4PQ7/1dr2NVHbacAkshapHnzZjz55OnsscdG7LbbBrzwwjkKIImIiCRYSN1qERERzOxXwNVZRk8C9gshTClgOSOA43NMMgXYNYQwJ9dy+vfvH8aOHZtvddLIPv56IYf/5ZVaw87ecxN+fcwOZUqRiDSmefMqMDO6dGlT7qSsk8xsXAihf7nTISIidaklkohIZGb7AlfFr1XAFcBxwBtxWF/gzgKWswW1A0i3AQcDPwVWxWF9gJMbnGhpdMtXruKH971da9htZ+6mAJLIWiCEwNix0+sM79q1rQJIIiIiGSiIJCJS4xLA4v93hRB+G0J4DA/2pJptHmJm2+dZTpe075eFEEaHEP4IfJIY3qrBKZZGd9odbzBh1pLq79v06shhO/QqY4pEpBRCCFx11Yvsvvs/+Mc/xpU7OSIiImsEBZFERGoMTPz/auqf+PjaV4lxg/Is50Pg68T3P5jZgWZ2KbBNHLYYeKz+SZWmcMfLE3j7q/nV3686ajuevmS/MqZIREohhMDPfvYc11/vj6lecMF/ePrpL8qcKhERkdWfgkgiIoCZdQXWSwz6Jm2S5PfNcy0rhLAMOAJIPQP1fWA0cBPQPP6/ZwhhckPSLI3rhqc+4TejahqOdWnXku/s3bd8CRKRkgghcMklT3Pjja9XDzviiC0ZOLBv+RIlIiKyhlAQSUTEpb9+Z0WO7x0KWN4C4DO8b6V0ewEnmZllGIeZnW9mY81s7KxZswpYlZTasNcmcdtLE2oN++tpu5Jll4nIGqKqKnDRRU9y881vVg879thteOSRU2jTpkUZUyYiIrJm0NlSRMQtSfveOsf3xbkWZGZdgNeBnnHQd4F/4x1zjwC2xt8ANw/4c/r8IYQ7gDvA385WUOqlZPr+4sk6w/52ej/23bJ7GVIjIqWyalUV55//BHfd9W71sJNP3p577jmOli2blzFlIiIiaw61RBIRAUII8/CgTkp6z8kbJP6fQG4nUBNAei+EcFcIYUkI4SPg1sR0p9QrsdJoPpi6oM6woefuzpE7bZBhahFZU6xcWcXgwSNrBZDOOGNH7r33eAWQREREiqAgkohIjRcT/++b+sfMNgX6ZJkuk2STlY5pj611zvK/rAZ+8cj7tb6/eeWBDNy6R5lSIyKlUFm5ijPOeIR77qk5vgcP3oVhw46lRQtVhUVERIqhM6eISI2bE/8PNrMrzOxY/FG0lNEhhA8BzGyomYX4GZKY5r3E/5sBt5vZIWZ2PvCTxLi3Spx+aYCz73qTj6YvrP7++xN2okfHNmVMkYiUwtix03nkkY+rv19wwW78859H07y5qsEiIiLF0tlTRCQKIbwE/CZ+bQZcDzwK7B6HfQWcV8CingaeSnz/HvAMcDvQJQ6bBfy6gUmWEjn8L6/w8mc1nZjvtklXTurfu4wpEpFS2XPPPtx33/E0a2ZcfPEe3HrrkTRrpk7yRURE6kMda4uIJIQQrjSzt4CLgX5AO2AK8Djw2xBC3telhRCCmR2DB49OAXbAH11bAUwEngX+EEKY3ji/Qorxz1cn8vHXC2sN+/sZ/fQmNpG1yEknbc9mm3WlX78NdGyLiIg0gIWgF/+IiKyu+vfvH8aOHVvuZKy1ps5byj6/q93F1S2n78pRO21YphSJSEMtWbKCJUsq6dGjfbmTIvVkZuNCCP3LnQ4REalLj7OJiMg6a8jj42t9H7xXXwWQRNZgixYt5/DD72XQoGHMmrWk3MkRERFZ6+hxNhERWSf9ZtTHjP54RvX3+783gD03X7+MKRKRhpg/fxmHH34vb7wxFYCDD76bN944jzZtVN0VEREpFZ1VRURknTNx9hLuePnL6u+9u7blW5uuV8YUiUhDzJ1bwaGH3sPYsTVdzZ1zzs4KIImIiJSYzqwiIrLOeXDslFrf/35GP72tSWQNNXv2Ug4++G7effeb6mG33HI4P/jBHmVMlYiIyNpJQSQREVmnLF6+kpHvTKv+/u2dN2Sn3l3KmCIRqa8ZMxZz4IHD+egjf3GmGdx++1F873u7lTllIiIiaycFkUREZJ2ywzXP1Pp+9VHblSklItIQ06cv4sADh/PJJ7MBDyDdddcxDB68S5lTJiIisvZSEElERNYZMxYuq/X9woGb071j6zKlRkTqa8qUBQwaNJwvvpgLQPPmxvDhx3H66TuWOWUiIv/P3n3HV1kdfhz/nuwBIWEjW2SoqCi4BSEgoIAIKCBFQKtY29raqm2drUV/Wqvi1joZsipbHAiEoYgDtSob2SAjEFbIzj2/PxKe5IYVIMm5N/fzfr3uq+ecPPfmG6Dm5pvzPA9QuVEiAQBCxsLVqX7z33Ro5igJgNPx6KMLvAIpIiJMEyb00403sqsQAIDyFuY6AAAAFeX1Reu88UPXna1qcZEO0wA4VS+/fK3at2+kyMgwTZ58EwUSAAAVhJ1IAICQ8Ltx32l96iFv3vOCeg7TADgd8fFR+vDDQfr++x3q0KGx6zgAAIQMdiIBACq91TsO6sOftnvz6vFRqlct1mEiACcjLS3ziLWqVaMpkAAAqGCUSACASu+RGcv85m8O4fbfQLD44YcdatXqZb300leuowAAEPIokQAAldrmPRn6btNeb35T2wZq27i6w0QASuvbb39Rp06jlZqaoT/84RO99dZ3riMBABDSuCYSAKDS+npDmvr/Z4k3v7BRop6+8XyHiQCU1pdfblX37u9p//5sSVK1atE677zajlMBABDaKJEAAJXWfe//4Df/S7dWMsY4SgOgtD7/fLOuvXac0tNzJEnVq8fq008Hq23bMxwnAwAgtFEiAQAqpS1pGdqcluHNz6wZr8vO5DQ2INAtWLBRPXqMV0ZGriSpZs04zZ17iy64oK7jZAAAgBIJAFDpZOTkqf3T8715rarRSrmvo7tAAEplzpx16t17ojIz8yRJderEa968ITr3XE5jAwAgEFAiAQAqnSnfbvWbD7uiiZsgAErto4/Wqm/fScrOzpcknXFGVaWkDFHLljUdJwMAAIdRIgEAKhWfz+qxD1b4rf2u01mO0gAojT17MjRgwGSvQGrYMEEpKUN11lmcggoAQCAJcx0AAICydOaDHynPZ735a7+6yGEaAKVRo0acxo/vq4iIMDVpkqhFi26lQAIAIACxEwkAUCnk5PnUdsScI9a7t+ZivEAw6NWrpaZPH6Dzz6+jhg2ruY4DAACOghIJAFAptHj44yPWVvyzm4wxDtIAOJHs7DxFR/u/Fe3Ro4WjNAAAoDQ4nQ0AEPQOZecdsfb9I9coLorflQCB6I03vlWbNv/Rjh3prqMAAICTQIkEAAh673y+wW++8p/dlRQf5SgNgON5+eWvdeeds7Rq1W517jxGu3Ydch0JAACUEiUSACCoWWs1+but3rzbuXUUGxXuMBGAY3n22S90991Fp57Gx0cqMpK3owAABAu+awMAgtpLKT9r054Mbz6id2uHaQAcy5NPfqb77iu6+P3llzfQnDm3KCkp1mEqAABwMiiRAABB7bk5a7zx2fUSVDshxmEaACVZa/XYYwv04IMp3lqHDo01e/ZgVavG/18BAAgmXHEUABC0vli3229+V8dmjpIAOBprrR56KEVPPvm5t5ac3FQzZw5UPNctAwAg6FAiAQCC1qwft/vNr7/gDEdJAJRkrdV9932q55770lvr1q2Zpk0boNjYSIfJAADAqeJ0NgBAULLW6tPlO7x5u8ZJDtMAKOmNN771K5B69myh6dMHUiABABDEKJEAAEHprve+0+70HG/+t2tbOUwDoKQhQy5Qly5nSpL69j1bU6b0V0wMm+ABAAhmlEgAgKCTdihHnxTbhdSyTlW1a1LdYSIAJcXGRmrGjIF6/PFOmjixn6Kiwl1HAgAAp4lfBwEAgs68lTv95m8OaecoCYDDfD6rsDDjtxYXF6mHHurgKBEAAChr7EQCAASdmT/84o27n1tXjWrEOUwDIDc3XwMHTta//vX5iQ8GAABBi51IAICgsnlPhj5bu1uSFGak+7u3dJwICG3Z2XkaMGCyZsxYrfffX6GoqHD96U+Xu44FAADKASUSACCo/GXKD974sjNrqFmtKg7TAKEtKytP/fr9Vx99tNZb27Rpv8NEAACgPHE6GwAgaKzecVBfrk/z5klxUQ7TAKEtIyNX118/wa9A+stfrtDIkd0cpgIAAOWJEgkAEDS6Pb/Ibz7ihtaOkgChLT09Rz16jNecOeu9tYcfbq+nnuoiY8xxngkAAIIZp7MBAILCyDlr/OYjep+r6vHsRAIq2oED2bruunFavHiLt/bPf3bUI49c7S4UAACoEJRIAICAl5Wbr1cX/Oy3dsvlTdyEAULYvn1Z6t79PX311TZv7V//6qK//OVKh6kAAEBFoUQCAAS8rzakKTffevNF93dymAYITdZa9eo1wa9AGjmym+655zKHqQAAQEXimkgAgIA35ouN3rhTy1pqVCPOXRggRBlj9PDD7RUVFS5JevXV6yiQAAAIMexEAgAEtH0ZOZq3apc373x2HYdpgNDWrdtZmjKlv3buTNevf32R6zgAAKCCUSIBAALaBz9u98bGSNe3OcNhGgA9e7ZwHQEAADjC6WwAgID2wty13rhTy9pKiIl0mAYIHZs371fXrmO1Zct+11EAAECAoEQCAASsfJ/V7vRsb37L5Y0dpgFCx4YNe3X11aM0Z856de48Rr/8ctB1JAAAEAAokQAAAev7zXv95h1b1HKUBAgda9fuUYcOo7Rx4z5J0qZN+7Vs2a4TPAsAAIQCSiQAQMB6ePoybxwTGSZjjMM0QOW3cmWqrr56lLZuPSBJio4O14wZA9W1azPHyQAAQCDgwtoAgIC062CWVu0oOoXmdx3PcpgGqPyWLdulzp3HaNeuQ5Kk2NgIzZx5s7p0OdNxMgAAECgokQAAAWnErJV+c66HBJSf//1vh7p0GaM9ezIlSfHxkfrww0G6+uomboMBAICAQokEAAhIa3cW7UJqWD1WiXFRDtMAldfSpb+oa9ex2rs3S5JUtWqUPv74V7ryykaOkwEAgEBDiQQACDglT2X7z+B2DtMAldfy5QWnsB04UHAXxMTEGM2ePViXXFLfcTIAABCIuLA2ACDgPDt7jTe+oGGizjkjwWEaoPJq3ryG2rcv2HFUvXqs5s0bQoEEAACOiRIJABBQ9mXkaNLSLd78ymY1HKYBKreoqHBNntxfgwadp/nzh+qii+q5jgQAAAIYp7MBAALKX6f86De/swO3FgfKU0xMhMaN6+s6BgAACALsRAIABAyfz2r28p3evFXdqqoWF+kwEVC5zJq1Ro8/vsh1DAAAEKTYiQQACBjfb9nrN39tcFtHSYDKZ9q0lRowYLJyc30KDzd64IH2riMBAIAgw04kAEDAGDlnrd+8ac14R0mAyuW//12um256X7m5PknSW299r4MHsx2nAgAAwYYSCQAQML7dVLQT6eZLGjlMAlQe7733o26+eYry860kqXnz6lq4cJiqVo12nAwAAAQbSiQAQEDYn5mrzNx8b/6nLs0dpgEqh3fe+V5DhkyTz1dQIJ19dk0tXDhMDRokOE4GAACCESUSACAgjFq80W9eOyHGTRCgknj99aX69a9nyhb0RzrvvNpasGCY6tWr6jYYAAAIWpRIAADn8vJ9Gjl3jTdPblXbYRog+L344le6664PvfmFF9bV/PlDVbs21xkDAACnjhIJAODcP2et8Js/1ONsR0mA4Pf881/qj3/8xJtfckl9zZs3RDVqxDlMBQAAKgNKJACAU9ZajVmyyZu3aZioZrWqOEwEBLc2beoqNjZCknTFFQ01Z84tSkqKdZwKAABUBpRIAACnRs5Z4zefOPwyR0mAyqFjxyaaMWOgunVrptmzByshgbuwAQCAskGJBAAlGGOuN8bMMcakGWOyjDFrjTHPGmNqnMJrdTTGTDbG/GKMyTbGpBpjlhpjXjDGRJZH/mDz/ZZ9fvOYyHBHSYDK45prmunjj3+lKlWiXEcBAACVCCUSABRjjHlM0gxJXSQlSYqWdJakP0taaoxpeBKv9bSk+ZL6SaonKUpSTUltJf2h8LVDWurBbH22drc3f/amCxymAYKPtVYjRizUhg17j/iYMcZBIgAAUJlRIgFAIWNMe0mPFE59kh6U1EfSl4VrTSS9VcrXul3S/YXTg5KelNRbUndJd0oaJym/LHIHs36vfeGNoyLC1Pls7soGlJbPZ/X733+kRx9doE6dRmvz5v2uIwEAgEouwnUAAAgg90g6/Kv7d6y1T0qSMeZbSZsKP9bVGHOutXb5sV6k8DS1x4otXW+tXVDisDfKLHWQyvdZbU7L8ObXX3CGEuM49QYoDZ/P6s47P9Bbb30vSdq0ab9GjlyikSO7O04GAAAqM3YiAUCRjsXGnx8eWGu3SNpc7GPJJ3idyySdUTjeKinZGLOq8PpKmwqvh5RUFoGD2fJf/HdNPNLjHEdJgOCSn+/TbbfN8AokSbr55tb697+7OkwFAABCASUSAEgqLHWqF1vaUeKQ4vNmJ3i584uNG6jgFLmWKrgGUiMVXA/pC2NM4qmlDX7WWl3/8mJvHh5mVC2O64wDJ5KX59Mtt0zT6NE/eGtDhlygsWP7KCKCt3UAAKB88W4DAArEl5jnHGde5QSvVbIcWiHpBkl3SDpQuNZK0t+O9mRjzPDCO7gtTU1NPcGnCk5frk/zm9/XtaWjJEDwyM3N18CBkzVhwjJv7fbbL9S77/ZWeDhv6QAAQPnjHQcAFDhUYl7yzmnF5+kneK2sEvN7rbUzrLVvSXqt2Pp1R3uytfYNa207a227WrVqneBTBaf73v/Bb35H+6aOkgDBITs7Tzfe+L6mTFnprf32t+30n//0UlgYd2EDAAAVgxIJACRZa/dKKn6P7LolDqlXbLzuBC+3qcR8wzHG1UqXrnLJys3Xtn2Z3nzsry9RBLsogGPKyspTnz6TNHPmam/tnnsu1csvX0eBBAAAKhTv2gGgyPxi4/aHB8aYppIaHuO4o/lcUn6xeZNjjEuWTSFhybo9fvMrm9V0lAQIDuHhRnHFrhn2179eqeee6yZjKJAAAEDFokQCgCIvFhsPM8Y8aIy5QdKkYutzrbXLJMkYM8oYYwsf/zh8gLV2h6TJxZ7zjDGmtzHmNkl3FVsfX/ZfQuD7cWvRXdkSYiLYSQGcQGRkuMaP76devVro0Uc76MknO1MgAQAAJyJcBwCAQGGtXWiM+T9JD6qgZH+ixCGbJd1eypf7g6Q2KrgrW2tJ00t8fIakN089bXDaujdDI+eu8eZ/vbaVwzRA8IiKCtfUqQO4AxsAAHCKdyIAUIy19iFJfSSlSNqngruyrZM0UlI7a22pTkGz1u6SdKmk/5O0WlK2Ci7e/ZUKdiP1tdbmH/sVKqd+r33hN2/XuLqjJEDg2rcvS6+++o2stX7rFEgAAMA1diIBQAnW2uk6cufQ0Y4bJmnYcT6+X9JDhY+Qty41XTsPZHvzHufXU8u6VR0mAgLPnj0Z6tr1PX333Xbt2JGuf/6zk+tIAAAAHn6lBQCoEHNX7PSbvzTwQkdJgMC0a9chJSeP0XffbZckjRixyBsDAAAEAnYiAQAqxJMfr/LGgy9rxAW1gWK2bz+oLl3GasWKVEmSMdJbb12viy6q5zgZAABAEUokAEC5W7vzoN+8z4X1HSUBAs+2bQeUnDxGa9bskSSFhRmNHn2DBg8+33EyAAAAf5RIAIBy978t+/zmbbmgNiBJ2rRpn5KTx2j9+r2SpPBwo3Hj+mrAgNaOkwEAAByJEgkAUO6W/3LAG3c5u47DJEDgWL9+r5KTR2vTpv2SpMjIME2ceKP69j3bcTIAAICjo0QCAJS7NcVOZzv3jASHSYDAsHbtHiUnj9HWrQUFa1RUuKZM6a+ePVs4TgYAAHBslEgAgHKVk+fTF+v2ePOBlzR0mAYIDLt3Z2jv3kxJUkxMhKZPH6Bu3c5ynAoAAOD4wlwHAABUbuO/2uSNG1aPVb1qsQ7TAIHh8ssb6qOPfqWaNeM0a9bNFEgAACAosBMJAFCu1uxK98bxUXzbAQ7r0KGxNm78o+Ljo1xHAQAAKBV2IgEAytXWwlN2JGnAxZzKhtD0zTfbtHr17iPWKZAAAEAwoUQCAJQba60WrUn15pc0re4wDeDGF19sUZcuY5WcPEbr1qW5jgMAAHDKKJEAAOVmc1qGN46OCFPLOlUdpgEq3qJFm9S161gdOJCtX345qP79J8ta6zoWAADAKaFEAgCUm1U7Dnrj7DyfIsL5toPQMW/eenXv/p4OHcqVJNWuHa/Ro2+QMcZxMgAAgFPDu3kAQLlZvm2/Nx50aSOHSYCK9cknP6tnzwnKzMyTJNWrV0ULFgxV69a1HScDAAA4ddwmBwBQbqb9b5s3vqhRksMkQMX54IPVuvHG95WTky9JatAgQSkpQ9S8eQ3HyQAAAE4PJRIAoFzsz8zVlrSiO7M1rRnnMA1QMaZOXakBAyYrL88nSWrcuJrmzx+qpk0pUQEAQPDjdDYAlZIx5kpjzDzXOULZS/PW+s3Pq5/oKAlQMSZNWqb+/d/3CqRmzZK0aNGtFEgAAKDSYCcSgKBjjKkhqZmkNGvtzyU+dpmkf0rqLMnnIB4KvfX5Bm98V8dmiorg9xao3D766Gfl5xfcea1lyxqaN2+I6tdPcJwKAACg7FAiAQgaxphwSa9Iul2SKVz7WlJvSVmSXpc0QAXl0XhJT7hJih37s/zm/S5q4CgJUHHefvt6ZWbmavnyVM2bN0R161ZxHQkAAKBMUSIBCCZ3SxouaaukLyWdJelSFRRLDSRdImmspBHW2nWuQkL6bG2q37xZrXhHSYCKExERpnHj+urAgWzVqME1wAAAQOVDiQQgmNwi6SdJl1trMyTJGPOKpLsk7ZF0lbV2icN8KDR7+U5vPLzDmTLGOEwDlI8FCzbq6qsb+/37jowMp0ACAACVFheoABBMWkgac7hAKvRa4f/+iwIpcMxdWVQiXX4mtzVH5fOvf32uTp1G6/7758ha6zoOAABAhaBEAhBM4iXtKLF2eP5TBWfBMexOz/abX9SIO1OhchkxYqH+9reCmz8+++wSvfnmd44TAQAAVAxOZwMQbEr+yv/wPLeig+Dopn631RvXSYhWtbhIh2mAsmOt1aOPztfjj3/mrXXq1ESDBp3nLhQAAEAFokQCEGyuM8bULTaPU0GRdJMxpk2JY621dmTFRYMk/W/LPm/csUVth0mAsmOt1V//Olf//vcX3to115yp6dMHKo6iFAAAhAhKJADBZlDho6Q7j7JmJVEiVbCPfio643DAJQ0dJgHKhrVWf/rTbL3wwlfe2nXXNdeUKf0VE8NbKQAAEDp45wMgmHRyHQDH99X6PX7zxtW5SxWCm89n9fvff6TXXlvqrfXu3VKTJt2o6GjeRgEAgNDCux8AQcNau9B1BhzfgDe+9MY1q0SpenyUwzTA6cnP9+nOO2fp7be/99ZuuukcjRvXV5GR4Q6TAQAAuEGJBCDoGGPaSjpL0m5Ji6y1XFQ7ABzI8v9raNMwScYYR2mA07dx4z5NnrzCmw8adJ5Gj75BERHc3BYAAIQm3gUBCBrGmBhjzMeSvpY0XtKnklYaY5q5TQZJOpDpXyK9NbSdoyRA2WjWrLrhe468AAAgAElEQVRmzx6sqlWjNGxYG40ZQ4EEAABCGzuRAASTv0jqJul7SSmSWkjqJelNSckOc0FSdp7PG59ZM95hEqDsXHppA33zzR1q3ryGwsLYWQcAAEIbJRKAYHKTpC8ltbfW5kuSMeZxSQ8YY2pZa1OdpgtxP+9K98ZR7NZAEMrOztPWrQfUrFl1v/WWLWs6SgQAABBYeJcPIJicKWnS4QKp0HuSjAqukQSH/jzpf9549c6DDpMAJy8zM1c33DBJV1zxjlat2u06DgAAQECiRAIQTGIlldxtdHgeU8FZUEJ0sbtV9W/b0GES4OQcOpSjXr0m6JNPftauXYeUnDxau3dnuI4FAAAQcDidDUBlwcVKHNqXkaO0Qzne/NFe5zhMA5TewYPZ6tlzghYt2uStDR/eVjVqxDpMBQAAEJgokQAEmyHGmMuKzWMkWUm/N8bcUOJYa639Y8VFC12frS06/adV3aqKj+bbCwLf/v1ZuvbacVqyZKu39sQTyXrwwfYOUwEAAAQu3uUDCDZdCx8llSyQpIJyiRKpAizdmOaNGyTFOUwClM7evZnq1u09ffPNL97aM89co3vvvcJhKgAAgMBGiQQgmDR1HQBHN3pJ0alAN1/C9ZAQ2HbvzlDXrmP1/fc7vLUXX+yuu+++1GEqAACAwEeJBCCYWEmp1tpM10FQZPt+/7+Oi5tWP8aRgHu7dh1Sly5j9NNPu7y111/voTvvbOcwFQAAQHDg7mwAgskGSX1ch4C/H7bs95snxEQ6SgKc2CuvfO0VSMZI77xzPQUSAABAKVEiAQgm3IEtAL352XpvPLzDmQ6TACf26KNXa/Dg8xUWZjRmTB/deuuFriMBAAAEDU5nAwCclu37ik5nu6BBosMkwImFh4fp3Xd76ze/aasrr2zkOg4AAEBQYScSAOCUbdh9SL/sz/Lm7VvUdJgGONKOHeny+azfWkREGAUSAADAKWAnEoBg094YU+r/dllrx5RnmFD3xbrd3viceglcDwkBZfXq3UpOHqM+fVrppZeulTGcEQsAAHA6KJEABJvhhY8TMSq4mxslUjmas2KnN64eH+UwCeBvxYpUJSeP1s6dh/TKK9+oSpUoPfVUF9exAAAAgholEoBg84akL12HQIFl24ruzHbZmdUdJgGK/PjjTnXpMkapqRmSpLi4SHXr1sxxKgAAgOBHiQQg2HxmrR3vOgSkz9amand6jje/qV1Dh2mAAt99t13XXDNWaWkFF3yvUiVKH300SO3bN3acDAAAIPhRIgEATsmbn23wxgkxEaqTEOMwDSB9/fU2dev2nvbtK7jYe7Vq0frkk8G67LIGjpMBAABUDpRIAIBTsmhNqjfucX49h0kAafHizbr22nE6eLBgd1xSUow+/fQWtWt3huNkAAAAlUeY6wAAgOCTnZev2Mhwb37zJdwuHe4sWLBR3bq95xVINWvGKSVlKAUSAABAGWMnEoCgYa2l+A4Qb3++QZm5+d78vPrVHKZBKMvOztOQIdN06FCuJKlOnXjNnTtErVvXdpwMAACg8uEHMgDASXv6k9Xe+L6uLWSMcZgGoSw6OkIzZgxUYmKM6tWrogULhlEgAQAAlBN2IgEATsquA1l+835tuWgx3LrwwnqaM+cWJSbG6KyzqruOAwAAUGlRIgEATsqug9l+83rVYh0lQahKT89RlSpRfmtc/wgAAKD8cTobAOCkfLl+jze+ukUth0kQisaP/0nNmr2oH3/c6ToKAABAyKFEAgCclE+XF/3wnhAb6TAJQs3o0f/T4MFTtWvXIXXuPEYrV6a6jgQAABBSKJEAACfl641p3rhNw0SHSRBK3nzzW9166wxZWzCvUyde1atzKiUAAEBFokQCAJTatn2ZfvNOLTmdDeXvlVe+1vDhs7wC6YIL6mj+/KGqU6eK22AAAAAhhhIJAFBqz3662m9+Zi1+iEf5GjlyiX7/+4+9edu29ZSSMlS1asU7TAUAABCaKJEAAKWWdijHG/c4r57DJAgFTz31uf7850+9+WWXNdDcuUM4jQ0AAMARSiQAQKn9Uux0tg4tajpMgsrMWqt//nOhHnhgnrd21VWN9Omng5WYGOMwGQAAQGijRAIAlMrOA1laszPdm3dvzU4klI8PP1yrv/99gTfv1KmJPvnkV6paNdpZJgAAAFAiAQBKafbyHd64RZ0qqhYb6TANKrPrrmuuO+64SJLUtWszzZo1SPHxUY5TAQAAIMJ1AABAcJj49RZvHB/Ntw+Un7Awo9df76nzzqutO+5oq5gY/r0BAAAEAt6VAQBKZcX2A974xrYNHCZBZePzWVlrFR5etEE6LMzo7rsvdZgKAAAAJXE6GwCUYIy53hgzxxiTZozJMsasNcY8a4ypcYqvV8UY87MxxhZ7dCzj2OUq32f95r0uOMNRElQ2+fk+3XHHTA0f/oF8Jf6dAQAAILCwEwkAijHGPCbp0RLLZ0n6s6S+xpgO1totRz7zuEZKalYW+VxZl5ruN0+I4XpIOH15eT7deusMvffej5KkqKhwvfpqDxljHCcDAADA0bATCQAKGWPaS3qkcOqT9KCkPpK+LFxrIumtk3zNnpJul5RVNind+Gr9HtcRUMnk5uZr8OCpXoEkSVlZ+exGAgAACGCUSABQ5B5Jh7dAvGOtfdJaO11Sf0mHf7Ltaow5tzQvZoyppaLS6a9lmrSCbdyT4Y0vaFDNYRJUBjk5+RowYLImTVrurd15Z1u9/fb1ftdFAgAAQGDhnRoAFOlYbPz54UHh6Wubi30suZSv94akOpLmSnrpdMO5tDmtqETqx0W1cRqysvLUr99/NW3aKm/t7rsv0Wuv9VBYGKexAQAABDJKJACQZIxJklS92NKOEocUn5/w+kbGmNsk3SBpr6Rh1tqgPkdnzoqd3viKZjUdJkEwy8zMVe/eEzVr1hpv7d57L9cLL3TnOkgAAABBgBIJAArEl5jnHGde5XgvZIxpIun5wuld1tptJxPEGDPcGLPUGLM0NTX1ZJ5aLkpeVLtZrZJ/VMCJHTqUo549J+jTT9d5aw8+eJX+/e9rKJAAAACCBCUSABQ4VGIefZx5uo7vZUlVJY231k462SDW2jeste2ste1q1ap1sk8vcwtX+xdZ/MCPUzFs2AylpGzw5o891lGPP57MvycAAIAgQokEAJKstXtVcOrZYXVLHFKv2Hidju/wRYMGGWPs4UeJY+YXrieeQtwKNfGbostBtaxT1WESBLO///1q1awZJ0n6v/9L1qOPXk2BBAAAEGQokQCgyPxi4/aHB8aYppIaHuO4Si89K88bP9jjbIdJEMxat66tuXNv0UsvXasHHmh/4icAAAAg4ES4DgAAAeRFSX0Lx8OMMeskrZD0YLFj5lprl0mSMWaUpKGF649Za/9ROP63pKOdhzay2PgVST9LyiyT5OUkL9+nnQezvfmlTasf52igiLX2iJ1GF1xQVxdcUHKTHwAAAIIFJRIAFLLWLjTG/J8KSqMwSU+UOGSzpNtL8TrjjrZujCleIk221i44xagVZtu+TOX7Cs7Eq1klSjGR4Y4TIRjs3JmuG298X88/301t257hOg4AAADKCKezAUAx1tqHJPWRlCJpnwruyrZOBbuI2llrNzmMV+E2p2V44zNrHfemdIAk6ZdfDqpjx9H6/PPNuuaasfrhhx2uIwEAAKCMsBMJAEqw1k6XNL0Uxw2TNOwkXjforiKcnevzxvFR7ELC8W3Zsl/JyWP0889pkqQDB7K1evUeTmEDAACoJCiRAADHtONAljeOjqBEwrFt2LBXycljtHHjPklSRESYJkzopxtvPMdxMgAAAJQVSiQAwDF9uX6PN45lJxKO4eef05ScPFpbthyQJEVGhun9929S796tHCcDAABAWaJEAgAc08fLiq5nkxgX6TAJAtWqVbvVufMY/fLLQUlSdHS4pk0boGuvbe44GQAAAMoaJRIA4Jh81nrjnudzly34W7Zsl7p0GaOdOw9JkmJjIzRz5s3q0uVMx8kAAABQHiiRAADHVKxDUtOa8e6CIOBs23ZAnTqN1u7dBXfwi4+P1KxZg9SxYxO3wQAAAFBuwlwHAAAEppw8n9+8SjS/d0CRM86oqkGDWkuSqlaN0uzZgymQAAAAKjl+IgAAHNVP2/b5zaMi+L0Dihhj9Pzz3RUVFa4bbzxHl17awHUkAAAAlDNKJADAUa3Zme46AgKcMUb//ndX1zEAAABQQfi1MgDgqFJW7fLGfS+s7zAJAsH8+Rt0xx0zlZ/vO/HBAAAAqJTYiQQAOKoa8VHeeOfBLIdJ4Nqnn65T794TlZWVp9xcn955p7fCwozrWAAAAKhg7EQCABxVRk6+N+57Ide7CVUffrhGvXpNUFZWniRpzpz12r79oONUAAAAcIESCQBwVOnZed64agwbV0PR9Omr1KfPJOUUFoqNGlXTokXDVL9+guNkAAAAcIESCQBwVBt2H/LGiXFRxzkSldH77y/XTTe9r9zcgmsgNW2aqEWLhqlZs+qOkwEAAMAVSiQAwFEVL5HiosIdJkFFGzfuRw0cOEV5eQUFUvPm1bVo0a1q3DjRcTIAAAC4RIkEADiCz2f95nWrxThKgor27rvf65Zbpnn/Blq1qqmFC4epQQNOYQMAAAh1XOQCAHCEn1PT/ebVOZ0tJIwZ84Nuu22mN2/durbmzr1FdepUcZgKAAAAgYKdSACAI8xdudMb16wSze3cQ8Sll9ZX3boFhVGbNnU1f/5QCiQAAAB4KJEAAEeY/v02b3z9BWc4TIKK1LJlTc2bN0Q9ejRXSsoQ1awZ5zoSAAAAAginswEA/Ph8Vmt2Fp3Ods4ZXAsnlJxzTi3NmjXIdQwAAAAEIHYiAQD87DqY7TfvcV49R0lQnqy1evLJz7RkyRbXUQAAABAkKJEAAH5+3LrPGyfGRSo2KtxhGpQHa60eeihFDz6You7dx+mbb7ad+EkAAAAIeZRIAAA/mbn53vhQdp7DJCgP1lrdd9+nevLJzyVJBw5k67nnvnScCgAAAMGAayIBAPz8tHW/Nx58WWOHSVDWrLX6wx8+1ssvf+Ot9ezZQu++29thKgAAAAQLSiQAgJ+3Pt/gja86q6bDJChLPp/VXXfN0htvfOet9e17tiZM6KcoTlkEAABAKXA6GwDAT0JM0e8XGteId5gEZSU/36df/3qmX4E0YMC5mjiRAgkAAAClx04kAIAnJ8+nA1lF10FqkBTrMA3KQl6eT0OHTtf48T95a7fccr7eeae3IiL4XRIAAABKjxIJAOBZuCbVG8dGhismkl0qwSw/36dBg6bo/fdXeGu33dZGb7zRS+HhFEgAAAA4ObyDBAB4nviwqGyIjuRbRLALCzNq2jTRm//mN2315pvXUyABAADglLATCQDg2bgnwxsPvbyJuyAoE8YYPfVUF+Xk5Mvns3r++e4yxriOBQAAgCBFiQQAkCRl5uT7zQdd2shREpQlY4yee66bNwYAAABOFfvZAQCSpCXrd/vN6yTEOEqCU5WenqMRIxYqN9e/EDTGUCABAADgtLETCQAgSVq5/aA3juauXUHnwIFsXXfdOC1evEXLl6fqvff6cvc1AAAAlCneXQIAJElb0oquh9S7zRkOk+Bk7duXpa5dx2rx4i2SpEmTluuTT352nAoAAACVDTuRAACSpJjIcG8cHsapT8EiLS1TXbuO1bffbvfWRo7spp49WzhMBQAAgMqIEgkAIEnal5Hjjds0TDzOkQgUqamH1KXLWP34405v7ZVXrtNvf3uxw1QAAACorCiRAACSpJ0Hsr1xvWqxDpOgNHbsSFfnzmO0YkWqJMkY6Y03eun22y9ynAwAAACVFSUSAECStGT9Hm9cLTbSYRKcyLZtB5ScPEZr1hT8nYWFGb37bm8NGXKB42QAAACozCiRAADKKnFL+LrVYhwlwYls3rxfycmjtW7dXklSeLjRe+/11cCBrR0nAwAAQGVHiQQA0IrtB/zmdRIokQJVfr5P2dkFpV9ERJgmTuynfv3OcZwKAAAAoSDMdQAAgHtL1hWdyta2cZLDJDiRpk2TlJIyRE2aJGrq1P4USAAAAKgw7EQCAOjjZUW3h+dUtsDXvHkNrVr1O0VH820cAAAAFYedSAAA5fuKxufXr+YuCI6wbNkuLVy48Yh1CiQAAABUNEokAIBWFrsmUqt6CQ6ToLjvv9+ujh1HqUeP8fr8882u4wAAACDEUSIBAFSrarQ3rp/I6WyB4Jtvtik5eYz27MnUoUO5uvnmKcrOznMdCwAAACGMEgkAQlxevk+707O9eeMa8Q7TQJKWLNmiLl3Gat++LElSYmKMpk7tzylsAAAAcIp3owAQ4uat2iVrC8Z1E2IUGc7vF1xatGiTevQYr/T0HElS9eqxmjv3Fl14YT3HyQAAABDqKJEAIMR9tjbVG7euz/WQXEpJ2aBevSYoIyNXklSrVpzmzRui886r4zgZAAAAQIkEACHvvS+LLthcs0r0cY5EeZo9+2fdcMMkZWUVXPeobt0qmjdviM45p5bjZAAAAEABSiQAgKdf2wauI4SkWbPWqF+//yonJ1+SVL9+VaWkDFWLFjUcJwMAAACKUCIBQAjbujfDb35e/WqOkoS2//1vh1cgNWpUTSkpQ9SsWXXHqQAAAAB/lEgAEMLmr071m8dEhjtKEtoeeqi9MjNzNXHicqWkDFHjxomuIwEAAABH4BY8ABDCFhYrkc6qXcVhktBmjNHjjyfr22+HUyABAAAgYFEiAUAIW7vroDe+rnVdh0lCy+zZPys7O89vzRijxMQYR4kAAACAE6NEAoAQlpPn88ZXNecuYBXhP/9Zqu7dx2nAgMnKzc13HQcAAAAoNUokAAhRWbn52r4/y5uf34CLape3l176Sr/5zYeSpBkzVuuxxxY6TgQAAACUHiUSAISoOSt2+s25qHb5euaZL/SHP3zizS+5pL7uvfdyh4kAAACAk0OJBAAhand6tusIIeOJJxbp/vvnePMrrmioOXNuUVJSrMNUAAAAwMmhRAKAELVk3R5vfP0FZzhMUnlZa/X3v8/Xww/P99Y6dGis2bMHKyEh2mEyAAAA4ORFuA4AAHCjakykN64WG3mcI3EqrLV68MF5euqpxd5a585NNWPGQMXHRzlMBgAAAJwaSiQACFE+a71x8zpVHCapfKy1uvfeTzVy5JfeWvfuZ2nq1P6KpbADAABAkOJ0NgAIUTl5Pm+cFMfOmLKUlpap6dNXefNevVpo+vQBFEgAAAAIapRIABCicvKLSqSoCL4dlKUaNeKUkjJUDRsmqF+/szV5cn9FR7P5FwAAAMGNd7QAEKL2ZeR44yoUHGWuSZNELVnya9WpU0URlHQAAACoBHhXCwAh6puNe71xA241f1ry8nz67rvtR6zXr59AgQQAAIBKg3e2ABCC9mfkemNjpAZJcQ7TBLfc3HwNHDhZV175jubNW+86DgAAAFBuKJEAoARjzPXGmDnGmDRjTJYxZq0x5lljTI1SPr+aMeZuY8wUY8waY8w+Y0yOMWaLMWacMaZNeX8NJ/Jzaro3tlYKDzMO0wSv7Ow83Xjj+5oyZaWysvJ0/fUTtWbNHtexAAAAgHJBiQQAxRhjHpM0Q1IXSUmSoiWdJenPkpYaYxqW4mXOlvSipL6SmkuqJilSUgNJgyR9bYy5tuzTl97Puw564xZ1qjhMErwyM3PVp88kzZy52lsbPvwiNW9e3WEqAAAAoPxQIgFAIWNMe0mPFE59kh6U1EfSl4VrTSS9VcqX80maJek2SddI+pukQ4Ufi5T08uknPnXfbdrnjTNz8x0mCU4ZGbm6/vqJ+vjjn721v/71Sj33XDcZw64uAAAAVE7cjgcAitwj6XAD8I619klJMsZ8K2lT4ce6GmPOtdYuP87rbJV0obX2x2Jrc40xuZKeLZyfaYypba3dVbZfQulMWrrFG1/UKMlFhKCVnp6jXr0maMGCjd7ao4920D/+0ZECCQAAAJUaO5EAoEjHYuPPDw+stVskbS72seTjvYi1dmuJAumw1SXmh45yTIVr14TTr0rrwIFsde/+nl+BNGJEJz32WCcKJAAAAFR6lEgAIMkYkySpeJuyo8QhxefNTvHTDCg2nmetdVIi7T2U4zfvd1F9FzGCzt69mbrmmrFavLhoF9fTT3fRww93cJgKAAAAqDiUSABQIL7EPOc485O+ErUx5j5JtxRO90u6+zjHDjfGLDXGLE1NTT3ZT3VCaRn+X1pcFGc2l8bMmav19dfbvPnzz3fT/fdf6TARAAAAULEokQCgQMldQdHHmaeX9kVNgWcl/btwaZ+k7tbalcd6jrX2DWttO2ttu1q1apX2U5Xa9n1Z3rh5be7MVlpDh7bRiBGdJEmvvdZDf/zjZY4TAQAAABWLXz8DgCRr7V5jzF5Jh68yXbfEIfWKjdeV5jWNMdGSRqvoNLatkq6z1v50OllP1+a0DG+8dlep+zBIevjhDrruuua66KJ6Jz4YAAAAqGTYiQQAReYXG7c/PDDGNJXU8BjHHZUxJlHSpyoqkH6UdJnrAkmS9mfmeuMOLcp+p1NlsX37QWUW+7M6jAIJAAAAoYoSCQCKvFhsPMwY86Ax5gZJk4qtz7XWLpMkY8woY4wtfPzj8AHGmNqSFks6fMXlLZIekNTUGHNVsUe1cv1qjmHpxjRvXDWGDalHs2nTPl111bvq2/e/ys7Ocx0HAAAACAj89AAAhay1C40x/yfpQRWU7E+UOGSzpNtL8VLnFD4Oayjpw6Mc10nSgpNPenrqVovxxtXjoir60we89ev3qlOn0dq8eb/Wr9+rwYOn6f33b3IdCwAAAHCOnUgAUIy19iFJfSSlqOAi2DkquAbSSEntrLWbHMYrE5m5+d74vPpONkMFrDVr9qhDh3e1efN+SVJUVLiGDr3AcSoAAAAgMLATCQBKsNZOlzS9FMcNkzTsKOsLJJmyzlVWvil2Olt8NN8GDluxIlWdO4/Rjh0FFxuPiYnQ9OkD1K3bWY6TAQAAAIGBnx4AIMRsScv0xs3rVHGYJHD89NNOde48RqmpBXeui4uL1Acf3Kzk5KaOkwEAAACBgxIJAELInvRsv/mZNeMdJQkc33+/XddcM1Z79hSUa1WqROnDDwepQ4fGjpMBAAAAgYUSCQBCyMI1qd74zFrxiggP7Uvjff31NnXr9p727cuSJCUkROuTT36lyy9v6DgZAAAAEHgokQAghCxZt8cbR4aFdoFkrdXdd3/sFUiJiTH69NPBuvji+o6TAQAAAIEptH+CAIAQEx1Z9J/9i5smOUzinjFGU6f2V7NmSapRI1bz5w+lQAIAAACOg51IABBCVu846I0bJMU5TBIY6tdPUErKUB04kK3WrWu7jgMAAAAENEokAAghGTn53rhGfJTDJG6kpWWqevVYv7VGjao5SgMAAAAEF05nA4AQsvyXA964WmykwyQV74MPVqtJk+f18cdrXUcBAAAAghIlEgCEqLrVYlxHqDBTp65U377/1cGDOerTZ5Lmz9/gOhIAAAAQdCiRACBEZOXm+81b1U1wlKRiTZy4TP37v6+8PJ+kgusgnXlmaF9UHAAAADgVlEgAECJ2Hsjym0dFVP5vAWPG/KBf/Wqq8vOtJKlFixpatGiYGjdOdJwMAAAACD6V/ycIAIAkKadwJ06oePvt7zRs2HT5fAUF0jnn1NLChcNUv35o7MACAAAAyholEgCEiNzC3TiS1KpuVYdJyt9rr32j22//QLbwSz7//DpasGCo6tat4jYYAAAAEMQokQAgROT7ikqkiHDjMEn5euGFL/Xb337kzS+6qJ5SUoaoVq14h6kAAACA4BfhOgAAoGLk+opOZwsPq5y/Q/juu+26557Z3vySS+pr9uzBSkwMnTvRAQAAAOWlcv4UAQA4wp70HG8cGVY5dyJddFE9PfPMNZKkK69sqDlzbqFAAgAAAMoIO5EAIEQUr41+2LrPWY7ydu+9V6hu3Srq3buVqlSJch0HAAAAqDQokQAgROTbomsiXdgwyWGSsmOtVU5OvqKj/b+d/epX5ztKBAAAAFRenM4GACHCV+zC2knxkQ6TlA1rrf70p9m69tpxysjIdR0HAAAAqPQokQAgRBTfiRQe5NdE8vmsfve7j/TCC19p/vyN6tNnkrKy8lzHAgAAACo1TmcDgBCRX2wnUpgJ3hIpP9+nO++cpbff/t5bq1YtWuHhwfs1AQAAAMGAEgkAQkTxEikiSHci5ef7dOutMzR27I/e2qBB52n06BsUEcHmWgAAAKA8USIBQIjw24kUhCVSbm6+hgyZrokTl3lrw4a10Vtv9VJ4OAUSAAAAUN4okQAgRGxOy/DG4UF2OltOTr5uvnmKpk5d6a3dccdFev31nkFZiAEAAADBiBIJAEJEbFS4N966N9NhkpOTnZ2nm256Xx98sMZb+93vLtaLL15LgQQAAABUIPb/A0CIKH4x7XqJMQ6TnJz775/jVyD9+c+X6aWXKJAAAACAikaJBAAhovg1keokBE+J9NBD7dWqVU1J0gMPXKVnnukqE2Sn4wEAAACVAaezAUCIKF4iBdM1kerUqaJ584Zo8uQVuvvuSyiQAAAAAEfYiQQAISJY7s7mK5bzsDPOqKo//OFSCiQAAADAIUokAAgRPltUzkQEaIm0d2+mrrrqHU2btvLEBwMAAACoUJRIABAidh3I9sbhAVgi7d6doeTkMVqyZKsGDJisWbPWnPhJAAAAACoM10QCgBCx/UCWN7b2yFPGXNq5M11duozVsmW7JEm5uT798stBx6kAAAAAFEeJBAAhol6xO7Jl5OQ7TOJv+/aDSk4eo1WrdkuSjJHefvt63XrrhY6TAQAAACiOEgkAQkRuvs8bN6tVxWGSIlu3HlBy8mitXZsmqeCC36NH36DBg893nAwAAABASZRIABAidh/K8caREe4vibdx4z4lJzDB62MAACAASURBVI/Whg37JEnh4Ubjx/dT//7nOk4GAAAA4GgokQAgRHy2NtUbRzq+sPa6dWlKTh6jzZv3F+SJDNN//3uTbrihldNcAAAAAI7N/a+iAQAV4vz61byxy7uzpafnqGPH0V6BFB0drmnTBlAgAQAAAAGOEgkAQlDtYhfZrmhVqkTpgQeukiTFxERo5syb1aNHC2d5AAAAAJQOp7MBQIiwrgMU89vfXiyfz+rcc2upU6emruMAAAAAKAVKJAAIQW6viFTg97+/xHUEAAAAACeB09kAAOXq66+3qWfP8UpPzznxwQAAAAACFiUSAIQI6+B8tsWLN6tLlzH68MO16tlzvDIycis+BAAAAIAyQYkEACHIVMD5bAsXblS3bu/p4MGCHUjLl6d6d2QDAAAAEHwokQAAZW7u3PW69tpxOnSoYOdRnTrxmj9/qFq1quk4GQAAAIBTxYW1ASBE2Aq6P9vHH69Vnz6TlJ2dL0mqV6+KUlIokAAAAIBgx04kAAhBppzuzzZz5mrdcENRgdSgQYIWLhxGgQQAAABUAuxEAgCUiSlTVmjgwCnKy/NJkpo0SVRKyhA1bZrkOBkAAACAskCJBAAhojzvzvbhh2s0YMBk5ecXfJJmzZKUkjJUjRpVK79PCgAAAKBCcTobAISgsr4722WXNdA559SSJLVsWUMLFw6jQAIAAAAqGXYiAQBOW40acZo7d4h++9sP9fLL16lu3SquI1Va+/fv1+7du5WTk+M6CgCUSnh4uKpWrarq1asrOjradRwAwGmgRAKAEFGep7NJUu3a8Zo8uX/5fpIQl5WVpZ07d6pBgwaKjY2VKestZQBQxqy1ys3N1YEDB7R582Y1atSIIgkAghinswEATtrzz3+pCRN+ch0j5KSmpqpWrVqKi4ujQAIQFIwxioqKUs2aNZWUlKS0tDTXkQAAp4GdSAAQIspqI9K//vW5/va3eQoPN4qKCle/fueU0SvjRLKyslS3bl3XMQDglCQkJGjjxo2qV6+e6ygAgFPETiQACEGnuollxIiF+tvf5kmS8vOtXn11qWx5nycHT15eniIi+P0PgOAUGRmp/Px81zEAAKeBd6IAgBOy1uqRR+briSc+89Y6dWqimTMHclpVBePPG0Cw4r9fABD8KJEAIESc6o4ha63+8pc5euaZJd5a167NNG3aAMXFRZZVPAAAAAABjhIJAEKQUel+G2yt1T33fKIXX/zaW+vRo7kmT+6vmBi+hQAAAAChhGsiAQCOyuezuuuuD/0KpD59Wmnq1AEUSACOa9WqVYqMjJQxRo8++qjrOChjb7/9towxMsYoJSXFdRwAQAWiRAIAHNWdd36g//znW2/ev/+5mjTpRkVFhTtMhVAwatQo7wfU4o/Y2FidddZZGjZsmJYvX37M5+/YsUOPPPKILr74YiUlJSkqKkq1a9dWly5d9MorrygzM/OYz/X5fJoxY4b69++vJk2aKC4uTgkJCTr77LM1ePBgffDBB1xMvhTuv/9+5eXlKT4+Xn/84x+PekxmZqYSEhL8/o5feeWVox67YMECv+M2btzo9/GS/2aOJi0tTU899ZSuvvpq1apVS1FRUapbt64uvvhiPfzww1q9evVpfc2na9OmTRo+fLgaN26s6Oho1a5dW71799bixYtP+rUWL16sm266SY0aNVJ0dLSioqLUoEED9e3bVwsXLjzqc8aOHauOHTsqMTFRUVFRaty4sYYPH65NmzYdceyQIUPUsGFDSdK9994rn8930hkBAEHKWsuDBw8ePAL00bZtW1tWuo1caBv/dZZt/NdZdsUv+094/Ntvf2elf1jpH3bw4Kk2Nze/zLLg1KxYscJ1hArx7rvvWknHfcTHx9sffvjhiOdOnz7dVq1a9bjPbdGihV25cuURz92xY4ft0KHDCT/33r17K+KPIWh9++233p/V7bfffszjxo4de8Sfbbt27Y567Pz58/2O27Bhg9/HS/6bKWnWrFm2evXqx/177d2792l93afj22+/tUlJSUfNFRYWZkeNGlXq15o2bZoNCws75tdpjLETJ070jvf5fPbmm28+5vFJSUl26dKlR3yeESNGeMdMmzat1PlK898xSUttAHwP5sGDBw8eRz7YiQQAOKrbbrtQr7/eQ7fd1kajRvVWRATfMuDGZ599ppSUFD399NMKDy/YCXfo0CG9/PLLfsctWbJE/fv318GDByVJ559/vsaMGaN58+bp2WefVY0aNSRJa9asUffu3bV3717vuRkZGerWrZsWLVokSQoLC9Ntt92mqVOnat68eRo1apQGDBigyMjAvph8enq66wh69dVXvfHNN998zONGjRp1xNrSpUuPu8vsVCxYsEB9+vRRWlqaJKlevXp66qmnNGfOHH388cd6+umn1aZNmzL9nCcjLy9PgwYN8v49XnfddZo5c6buvfdeSQW7437zm99o/fr1pXq9Z555xtsZ1Lp1a02fPl3Tpk1Tq1atJBX8Avm5557zjp84caImTJggSYqOjtYLL7yg2bNna/DgwZKkvXv3auDAgcrLy/P7PMX/bl9//fVT+dIBAMHIdYvFgwcPHjyO/SjLnUhdnzu5nUiH+Xy+/2/vzsOkqs59j/9emgYEGhSZRGhA0XbASKBVOAQFxAnUBuNA4CjK4HUIijnmSMhV8OZRcx3jcGOCIhEUNRpDY9QLthIjFzEyKODY4TAIFxQMg4D2gO/5o6o3u6qHKrqq09P38zz7Ye211l61alV1V9fLWmunrQ9ITWOdiRQ2YsSIIP+8886LKRswYEBQlp2d7Xv27Ikpf//9993Mgjq/+MUvgrK777475jGfe+65Cvv22WefeVFRUVLPY9WqVT5u3Djv2bOnN2/e3LOysvyUU07xn//850Gd6dOnB485bty4mOvPOuusoGz27NlB/rhx44L86dOn+9y5c71Pnz7evHlzz8vL8/79+1f6PDZu3BiMwWGHHea7du0Kyt577z0fPXq0d+3a1TMzM/3www/3s88+2/Pz85N6vu7upaWl3qZNm2C2WGlpaYX1Nm3aFMyWadGihY8ePTroc3h8ylR3JtKBAwf8hBNOCPK7d+/u27Ztq7BPa9asSeo5vvPOO0kf4fGtzIIFC4L+tWnTxvfv3x+UDRs2LCi79dZbk+rfySefHFzz6KOPBvn33XdfkN+nT58g/4orrgjyr7rqqiC/uLjYW7VqFZS9/PLL5R4rJycnmC319ddfJ9U/ZiJxcHBw1O+D/1YGgEYofsuQ774r1X/+5xvavfu7Cuomdyc34F+ta9euQfqLL77Qu+++G5xPmTJFWVlZMfVzc3M1fPjw4PyPf/xjkH766aeD9NChQzV69OgKH/P4449Xs2bNEvbtiSee0Gmnnaann35a69evV1FRkb755hutWbNGM2fOTPzkkjRv3jxdeeWV+uCDD1RUVCRJmjBhQlD+7LPPlqvvHtnT6dJLL1Xbtm0lRWYPDRgwQM8//7w2b96skpIS7dq1S2+++aby8vI0bdq0pPrzwQcfaM+ePZIi4102cyzenDlzgtkyF154oW688cag7JlnntGBAweSerxE/v73v+vTTz8Nzn/1q1+pU6dOFdbt3bt3Um0OGjQo6WPVqlUJ2wtvTN23b18ddthhwfnAgQMrrFeVc889N0j//ve/V35+vubPn69Zs2YF+VdeeWWQDs/Ia9WqVZDOzMyMea8vXry43GOdccYZkiKzpSrbawkA0LBwex0AaOT27y/RqFEvaNGidXrnnU1atOjflZXVvLa7hUPQY+qrtd2FSm349YiU21iyZIlKS0u1YsUKLVy4UJLUrFkzXX/99UGdDz/8MOaa3NzcCts67bTT9OqrkfFat26d9u3bJ0kxmyqHv4RXx8cff6zrr78+CIT06dNHt9xyizp16qRPP/1UL7zwQkrthxUWFmrgwIG66aab1Lp1a+3Zs0cjRozQlClTtG/fPi1cuFA7duxQ+/btJUWCSGXKgk0fffSRJk+erO+//15NmjTR1KlTNXjwYK1fv15Tp07Vzp07dc8992jYsGEaOnRolf1Zs2ZNkD7uuOMqrTdnzpwgPXbsWA0cOFA9evTQhg0btHXrVi1cuDAm4FddK1asiDlP9bWtCeFlap07d44pC5+vW7cuqfbuuusu7d27V08//bTWrl2rkSNHBmVHHXWU7rrrLl1zzTVB3oknnqhFixZJkl5++WVNmjRJOTk5euqpp2ICTPGbmUuRoGqZ1atXa9SoUUn1EQBQfzETCQAaCVf5O0rt21esCy+cp0WLIl9Oli3brBdeSO9+JECqBg0apCFDhujWW29VaWmpcnNz9eabb6pfv35Bnd27d8dc07Fjxwrbip+Fsnv3bu3atSsmr2zvpOqaPXt2EEDq2rWr3nnnHV111VU677zzdPPNN2vp0qUptR929NFHq6CgQJdffrmGDx+u0aNHKysrS5dffrkkqaSkRC+++KIkae3atUGQp1evXjrzzDOD/pbNCho6dKguuOACHXbYYTrppJNiggJPPPFEwv5s3749SFc2jkuXLtXnn38uSTriiCM0fPhwmZnGjBkT1AnPDEtFul9b6dC2ghg8eHDC9soCmZLKzXILnye731WzZs3Uo0cPHXHEEeXKtm7dqmeeeUZbtmwJ8m6++eZgRtqXX36pvn37qlWrVpo8eXLMtd99V36mart27YL0V199lVT/AAD1G0EkAIhjZheb2Rtm9k8z+87MCs3sATM7pG8fZtbdzGaa2UYzKzKzr8ws38wGJr66ZplM33xTpAsueFaLF28I8u+8c7AmTPhh7XUMSMLHH3+szZs3x+SVfQkuU9kX2i+//LLcdYcffnhM3tdff51y/8qcd955at26dUrtVWX48OFq0aJFufzx48cH6bIlbeGlbePHjw+Wqob7W1BQELMc66mnngrKktnwumypXHw6LLyh9mWXXRYESsaOHRvk5+fnx8yCadIk9k/W+LbD5+EluOl+baXIzLhkj/jgZkXCS8jKliRWdJ7s++iGG27QL3/5S3355ZfKy8vTli1btHPnTt1www2SIsviLrrooiBw2LNnT7399tsaMGBATDu9evVS//79g/OKglKVjTsAoOFiORsAhJjZnZLuiMvuJelnki4xszPd/Ysk2ukrqUBS+K/uDpIulnShmY139/T8V3uSwt+59uz+TleP+qOWLTv4Rfyee87W1Kk/+ld2CWmSjiVjdZm7a/v27br11ls1Z84c7d+/X+PGjdMPfvADnXTSSZIid2ILW758uQYNGlSureXLlwfpY445JvgCn5OTEyxpKygo0G233ZZSf8sk+mIdLo+/+1V4Vk9ljjrqqArzf/SjHwXPaenSpVq/fr2ef/55SVJGRoauvvrqhG3HK7vrXVXCM8DK7oYW9u2338bsRTVz5swK94gqKirS888/HyxZjN/faseOHerZs2dwHh6rcEAxPFtNiry24WBVdVT0vqrM4sWLE85GOuaYY4L0tm3bYsq2bt0apI899tiEj1dUVBSz99GMGTPUpUsXSZFlbmV3zlu1apUKCwuVk5MjSTr11FO1dOlSbdu2TZs2bVLHjh2VnZ2tc845J2jr1FNPLfd44de4Q4cOCfsHAKj/mIkEAFFmNkjS7dHT7yVNkzRK0rJoXg9JTybRTlNJ83QwgPSaIsGjB6LnTST9zsyOqeDyGnfg21KNv+JPMQGkBx88lwAS6rQOHTpo5syZQeCguLhYU6dODcqzs7NjZk08/PDDMcuEpMgX57L9kCTpiiuuCNLhoEpBQUGwBCxeYWGhiouLq+zrySefHKQXLlxYrh/hIFN4dkd4dlVhYWHMPk2VqSpIVTYbyd3105/+NNjTZvjw4THBpxNPPDFI/+QnP6lwWdb333+vtWvXJuzPKaecEqQr6v/8+fOTmp0jxS5py8nJUWZmZnD+2muvxdQNv65lgUVJOv3002Oe3+23317pLLVkZlrVhPA+UytXrtT+/fuD87/97W8V1qvMzp07YzYlL9vkPD4tlV8CKkX2YDr99NPVo0cPLV26NNhM28wq3O8o/BqHX3sAQANW27eH4+Dg4Kgrh6Q/6eAtop8I5XdTJKhUVnZygnYuCtXdLemwUNkbobL7EvWpX79+ni7DHvird538Z8/s+L9dmhEcjz32XtoeAzUrmVtjNwSV3a7d3f3JJ5+MKVu5cmVQtmTJEs/MzIy5jfncuXP9rbfe8gcffNDbt28flHXr1i3mluT79+/3U089NSjPyMjwiRMn+vz58/2tt97yuXPn+tixY71Zs2a+c+fOKvu/du1az8jICNrq16+fz5kzxxcuXOiPPvqoDxw4MKj7+uuvB/XMzG+55RZ/6KGHvGfPnjHPc/bs2cE148aNC/KnT59eaT+2bdvmTZs2jWlHks+fPz+m3urVq71JkyZB+bXXXusLFizwRYsW+axZs/ymm27yrl27xvShMqWlpd62bVuX5K1atfLS0tKY8nPPPTd4nIsvvtgff/zxmOM3v/lNTF8/+eST4NpLL700yG/SpImPHTvWb7/9dh88eHDMNb/97W9jHnPx4sUx74suXbr4vffe62+88YYvXLjQH3zwQe/Xr5/n5eUlfH41oaSkxI877rigfxdccIEvWLDAp0yZEuQ1b97c//GPf8Q8p7Ky7t27B/kHDhzwzp07B2W9e/f2F1980V955RUfNGhQkN+yZUvfu3dvcN3QoUN9+vTp/vLLL/urr77qd9xxh7dq1Sqof80111TY9+OPPz54727fvj2p55vM7zFJy70O/F3AwcHBwVH+qPUOcHBwcNSVQ9LXoS8i4+LKNoTKJido56FQ3cVxZTNCZSsS9SmdQaRBMxZ5ZvuDASSzGT5z5vK0tY+aRxDJvbi42LOzs4OykSNHxpS/9NJL3rp163KBk/DRq1cv/+ijj8o97tatW/3MM8+s8lpJCYNI7u6PP/54TCApfLRt2zaoV1pa6ieccEKFdbp165ZSEMndPS8vL6bdTp06eUlJSbl6jz32WEwgqaIjmSCSu/vEiRODawoKCoL8zZs3xzzGu+++W+H1ubm5QZ3bbrstyN+yZUu54Fr8MWLEiHKBK3f3v/zlL96uXbsqr62tIJK7+/vvvx8E3+IPM/NZs2bF1K8siOTu/uyzzyZ8LR955JGYa7p3715p3fPPP9/37dtXrs+FhYVBnXPOOSfp50oQiYODg6N+HyxnAwBJZnaEpHahrG1xVcLniTamCC9TS6WdtLKmJmsaWfrSpIlp9uw8TZrUL8FVQN2SmZkZs19Rfn6+Vq9eHZz/+Mc/1meffaZp06apb9++atOmjZo2bar27dtryJAheuSRR/Thhx/GLHkq07lzZy1evFh//vOfdemllyo7O1stWrRQ69atlZOTozFjxig/P7/cJt4Vue666/Tee+/pyiuvVI8ePdSsWTO1bt1avXv31qRJk4J6GRkZys/P1/nnn6+WLVsqKytLeXl5WrZsWcxeOdU1YcKEmPNx48apadPyW2LeeOONWrZsmcaOHavs7Gw1a9ZMbdq0UU5Oji677DLNmTNHl1xySVKPWbaPkSTNmzcvSM+dOzfYzPnoo4/WGWecUeH14ceZO3dusDyrS5cuWrlypWbMmKG+ffsqKytLGRkZOvLIIzVkyBA9+eSTWrBggTIyMsq1OWLECBUWFuqee+7RoEGD1L59e2VmZqpjx47q16+fpk2bpl//+tdJPb+akJubq1WrVmnChAnq2rWrMjMzdeSRR+qiiy7S22+/HbNReiJjxozRkiVLNHr06OC1zMzMVJcuXTRy5EgtWrSo3J3XJkyYoNNPPz0YlyOPPFLDhg3T3Llz9eqrr6ply5blHue5554L0uHXHADQsJm713YfAKDWmVlXSeENs4e6++JQ+d8kle2mOsvdJ1bRVoGks6Onc9x9XKhsvKSyXU8PuHu5b3Nmdq2kayUpOzu738aNG6vxjMob9uDb+mzjLn31x8/16+mDdct1p6WlXfzrfPLJJzH7uwB11cUXX6xXXnlFrVq10oYNG9S+ffva7hLSqLi4WMcee6w2b96sPn36aMWKFeXuoFeZZH6PmdkKd89NR18BAOnFTCQAiNgXd968ivO9h9DWIbfj7jPdPdfdc9N5t5vRp3XTjefnaPoTI/TvY3+Q+AIAqKZ7771XTZs21b59+/Twww/XdneQZnPnzg02gr///vuTDiABAOq/8vOZAaARcvedZrZTB++o1jmuSvg+2usSNPdfoXQq7aTVxEG1cjM4AI3QCSecoJKSktruBmrIhAkTyi2VBAA0Dvy3AQActDiULlu6JjPrqcgd2iqqV5G3Qum+ZhbeTOLMSuoBAAAAQJ1GEAkADnoklL7azKaZ2UhJL4TyC9x9rSSZ2R/MzKPHjFCd1yUVRtNZkl4ys4vM7CFJZ0XziyT9rkaeBQAAAADUAJazAUCUu79tZndLmqZIkP2uuCqbJFW6oXaonVIzGyOpQFJbSRdEj6CKpBvc/V+6nA0AAAAAUsFMJAAIcfdfShqlyFKzXZKKFdm76CFJue6e1K3S3H25pB8qcie2zZJKJH0t6RVJZ7n7U+nvPQAAAADUHGYiAUAcd58vaX4S9a6WdHUV5euVxMwl4FC4u8ystrsBAIfM3Wu7CwCAFDETCQCAeqJp06YqLS2t7W4AQLWUlJQoIyOjtrsBAEgBQSQAAOqJFi1aaO/evbXdDQColj179igrK6u2uwEASAFBJAAA6okOHTpo+/bt2r9/P8tCANQL7q7i4mLt2LFDO3fuVLt27Wq7SwCAFLAnEgAA9USLFi3UqVMnbdu2TUVFRbXdHQBISkZGhrKyspSdna3mzZvXdncAACkgiAQAQD3Stm1btW3btra7AQAAgEaI5WwAAAAAAABIiCASAAAAAAAAEiKIBAAAAAAAgIQIIgEAAAAAACAhgkgAAAAAAABIiCASAAAAAAAAEiKIBAAAAAAAgITM3Wu7DwCASpjZdkkb09xse0k70txmY8MYpo4xTB1jmDrGMHU1MYbd3b1DmtsEAKQBQSQAaGTMbLm759Z2P+ozxjB1jGHqGMPUMYapYwwBoHFhORsAAAAAAAASIogEAAAAAACAhAgiAUDjM7O2O9AAMIapYwxTxximjjFMHWMIAI0IeyIBAAAAAAAgIWYiAQAAAAAAICGCSAAAAAAAAEiIIBIA1HNmdrGZvWFm/zSz78ys0MweMLMjD7Gd7mY208w2mlmRmX1lZvlmNrCm+l5XpDqGZtbWzCab2Z/M7HMz22VmxWb2hZk9a2Z9avo51LZ0vQ9D7bU2s3+YmYeOwWnudp2SzjE0s8Fm9pKZ/f/oz/N2M1tuZg+bWWZN9L8uSMcYmllm9Od5SbSdUjPba2Zrzex+M+tUk8+hNpnZFDN70czWx/3sXV2NthrtZwoANGTsiQQA9ZiZ3SnpjkqKN0g6092/SKKdvpIKJB1RQfH3ksa7+9PV7Wddlo4xNLP+kt6tokqJpDx3f71anazj0vU+jGvzCUkT47KHuPtfD7mD9UA6x9DM7pX08yqqZLn73kPrYd2Xxt+Hf5J0SRVVvpD0Q3f/+pA7WceZ2S5JbSsousbd/3AI7TTazxQAaOiYiQQA9ZSZDZJ0e/T0e0nTJI2StCya10PSk0m001TSPB38Y/81SRdLeiB63kTS78zsmLR0vA5J1xiGrv+LpPGSzpE0VdK+aFmmpMdS73Hdk+YxLGvzQkUCSN+lp5d1WzrH0Mwm6mAA6RtJ90jKk3S+pP8h6VlJB9LR77okjb8Peyk2gPQ7RX6e/0MHx62bpMtT7nTdtEbSU5JukPRVdRpozJ8pANAYMBMJAOqpuP8tf9LdJ0Xzu0naKMmiZb3d/aMq2rlI0oLo6R5Jnd3922jZG5KGRcvud/eqZjfUO2kcw66S2rn76rj8n+ngFydJ6uTu1fpiVlelawxD7XVQ5ItsJ0k3S3o4VNwgZyKl8X2YqciMmy7RrAY5XhVJ4xjmSno/lBXM2jKztZJOjuZPcfeH469vSMxsg6Tu0dOkZyI15s8UAGgMmIkEAPXX4FB6SVkiulxjU6hsaIJ2wuUry/7Yj/p/h9BOfTQ4lK72GLr75vgAUtRncef7KqhT3w0OpVN5H5aZqUgAqUDSo6l2rp4YHEqnMob9dTCAtFnSUDP7NLo30MbofkgVLS9qCAaH0qmM4VpJW0Pn95nZ2dGA8AnRvL2S5le/qw1eY/5MAYAGjyASANRD0S+C7UJZ2+KqhM+PTdBceElBKu3UK2kew8pcEUq/6e4NKoiU7jE0s/GSRkraKelqbwTTpdM8hj8IpbsqsrwrR1JzSdmSbpK01MwOr15v66Z0jqG7fydpuKSV0azrFAloPiApI5oe4O4bU+lzA9coP1MAoLEgiAQA9VOruPPiKs5bH0JbqbRT36RzDMsxs1slXRk93S1p8qG2UQ+kbQzNrIek30RPr3f3LSn1rP5I5/swPjj0sSJBuUmKLCuSIrNpph5KB+uBdP8s75b0uSJ7K8X7N0mXmZlVUIaIxvqZAgCNAkEkAKif4me0NK/iPNFdmMJtpdJOfZPOMQxYxAOS7otm7ZJ0vrt/cuhdrPPSOYaPScqSNM/dX0i1Y/VIOscwfiPy/3D3fHd/UtLjofzhh9C/+iBtYxidpfWupNGK/J08QZFgR29Flqe2VOQOcDen0N+GrrF+pgBAo0AQCQDqIXffqciSnzKd46ocFUqvS9Dcf6WpnXolzWMoSTKz5pKek/SzaNZmRW4rvqzyq+qvNI9h1+i/Y8zMy464Oouj+Q1mOVaaxzB+idX6StIV3cK93krzGP5YkT25JOlDd3/K3fdFN+MOB+KuKH8pohrlZwoANBYEkQCg/locSg8qS5hZT0VuQV1RvYq8FUr3NbOWofMzK6nXUKRrDMtmMCzSwS+XqyX1d/c1aehnXZa2MWzEJvuPCgAABLdJREFU0jWGS3TwNvRS5Lb2FaUb4n4+6RrDDqF0VtyytbaVpBGrMX+mAECDRxAJAOqvR0Lpq81smpmNlBReClTg7mslycz+EJrhMSNU53VJhdF0lqSXzOwiM3tI0lnR/CJJv6uRZ1G70jKGZtZRkbsOlX1B+kLSLyT1NLMfhY6G+MUzXe/D+yTdUsER9n+ied+qYUnLGLr7Nkkvha6538zyohuWXx/Kn5f+p1Dr0vU+/DCUPkbS783sXDO7VrHvx/fT3P86IfpcR0bHLhz86VuWb2bto3X5TAGARqhpbXcAAFA97v62md0taZoi/ylwV1yVTZImJtFOqZmNUeSuQ20lXRA9giqSbnD3Brf0IF1jKOmk6FGmm6RXK6g3RNJfD72ndVca34fPVpQf/eJZ5iV3/2s1u1pnpfF9KEXuwNZHkbuy9Vb5W9HnS3qi+r2tm9I4hv9XkSBI2e/ASdEjbLuk/1X93tZpMyV1ryB/sg7eHKDK32ON+TMFABoDZiIBQD3m7r+UNEqRZQG7FLnzzTpJD0nKTfY21O6+XNIPJc1SZB+fEklfS3pF0lnu/lT6e183pGsMGzPGMHVp/Fn+StIZku5WZCPoIkU2On5PkdlIl7j7gcpbqL/SMYbu7pLyJN0o6W+S/qnIEsFvFbnb3W8k9SEAUrXG/JkCAA2dRT4rAQAAAAAAgMoxEwkAAAAAAAAJEUQCAAAAAABAQgSRAAAAAAAAkBBBJAAAAAAAACREEAkAAAAAAAAJEUQCAAAAAABAQgSRAAAAAAAAkBBBJAAAAAAAACREEAkAADRYZjbYzLyKo3+0Xnz+d2ZWaGYPmlm7UHs9Kqj7rZl9ZGZ3mlnL2nu2AAAANatpbXcAAADgX+A5Sa9VkP+PUPoDSQ9E0+0kDZd0i6RzzKyfuxeH6r4haU403UHSjyXdIWmApHPT2G8AAIA6gyASAABoDFa6+zMJ6myJq/OImb0i6UJJeZJeDJV9Hq5rZo9IWqaDAacV6eo4AABAXcFyNgAAgMotjP7bq6pK7n5A0l+jp8fVZIcAAABqCzORAABAY9DSzNrH5RW5+zcJrisLCO1I4jGOjf77z0PqGQAAQD1BEAkAADQGd0aPsBckjQ6dZ4YCTUcosoztekm7JeXHXdsiVLe9pMsljZK0WdLbaew3AABAnUEQCQAANAYzFbunkSRtizs/V9L2uLwPJV3r7l/F5U+IHmGLJV3n7kWpdBQAAKCuIogEAAAag0J3L0hQ5z1J/zOaLpK00d03VVI3X9JjkjIUWfL2n5K6Ra8DAABokAgiAQAAROxIItBUZnOo7kIze13SaknPm9m/ubvXTBcBAABqD3dnAwAASJG7r5N0v6T+kn5Sy90BAACoEQSRAAAA0uMhRTbhnm5mGbXdGQAAgHQjiAQAAJAG7r5LkX2Sjpc0ppa7AwAAkHYEkQAAANLnIUl7Jd3BbCQAANDQGPs+AgAAAAAAIBFmIgEAAAAAACAhgkgAAAAAAABIiCASAAAAAAAAEiKIBAAAAAAAgIQIIgEAAAAAACAhgkgAAAAAAABIiCASAAAAAAAAEiKIBAAAAAAAgIQIIgEAAAAAACAhgkgAAAAAAABI6L8BctOmWWZ0DH8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Use logistic regression to predict if the person make >50K per year from other variables in Adults with binarized columns\n",
      "\n",
      "Confusion Matrix and Metrics of Class 1: income>50K = true and Class 2 : income>50K = false with binarized columns\n",
      "Probability Threshold is chosen to be: 0.5\n",
      "Confusion Matrix:\n",
      "[[8088  162]\n",
      " [1809  795]]\n",
      "TP, TN, FP, FN: 795 , 8088 , 162 , 1809\n",
      "Accuracy rate: 0.82\n",
      "Precision: 0.83\n",
      "Recall: 0.31\n",
      "F1 score: 0.45\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Use logistic regression to predict if the person is female from other variables in Adults with non-binarized columns\n",
      "\n",
      "Confusion Matrix and Metrics of Class 1: Female = true and Class 2 : Female = false with non-binarized columns\n",
      "Probability Threshold is chosen to be: 0.5\n",
      "Confusion Matrix:\n",
      "[[6135 1114]\n",
      " [1972 1633]]\n",
      "TP, TN, FP, FN: 1633 , 6135 , 1114 , 1972\n",
      "Accuracy rate: 0.72\n",
      "Precision: 0.59\n",
      "Recall: 0.45\n",
      "F1 score: 0.51\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Bin numeric values using Equal-width Binning\n",
    "NB = 5 # number of bins\n",
    "\n",
    "for col in ['age', 'hours-per-week']:\n",
    "    # Replace missing values with Median for current column without Nan\n",
    "    Adults.loc[:, col] = replace_missing_value(Adults, col)\n",
    "    # To obtaining the boundaries of the bins\n",
    "    bounds = np.linspace(np.min(Adults.loc[:, col]), np.max(Adults.loc[:, col]), NB + 1) \n",
    "    # Get the bin data\n",
    "    bx = bin(Adults.loc[:, col].values, bounds)\n",
    "    # Assign the binned variable into the new binarized column\n",
    "    Adults.loc[:, col + '-bin'] = bx\n",
    "\n",
    "# Replace missing values & outliers. Then Normalize numeric values using Z-Normalize\n",
    "for col in ['age', 'fnlwgt', 'capital-gain', 'capital-loss', 'hours-per-week']:\n",
    "    # Replace missing values with Median for current column without Nan\n",
    "    Adults.loc[:, col] = replace_missing_value(Adults, col)\n",
    "    # Replace the outliers (if exists) by calling \"replace_outlier\" funtion\n",
    "    Adults.loc[:, col] = replace_outlier(Adults.loc[:, col])\n",
    "    # Coerce to numeric\n",
    "    Adults.loc[:, col] = pd.to_numeric(Adults.loc[:, col], errors='coerce')\n",
    "    X = pd.DataFrame(Adults.loc[:,col])\n",
    "    # Normalization numeric data using Z-Normalize by scikit library\n",
    "    standardization_scale = preprocessing.StandardScaler().fit(X)\n",
    "    z = standardization_scale.transform(X)\n",
    "    Adults.loc[:,col] = z\n",
    "    \n",
    "# Impute missing categories\n",
    "# Trim string (some daata has ? with spaces)\n",
    "Adults.replace('(^\\s+|\\s+$)', '', regex=True, inplace=True)\n",
    "    \n",
    "MissingValue = Adults.loc[:, \"workclass\"] == \"?\"\n",
    "Adults.loc[MissingValue, \"workclass\"] = \"Private\"\n",
    "\n",
    "MissingValue = Adults.loc[:, \"occupation\"] == \"?\"\n",
    "Adults.loc[MissingValue, \"occupation\"] = \"Other-service\"\n",
    "\n",
    "MissingValue = Adults.loc[:, \"native-country\"] == \"?\"\n",
    "Adults.loc[MissingValue, \"native-country\"] = \"Unknown\"\n",
    "\n",
    "# Categorical columns are consolidated\n",
    "Adults.loc[ Adults.loc[:, \"workclass\"] == \"Self-emp-not-inc\", \"workclass\"] = \"Private\"\n",
    "Adults.loc[ Adults.loc[:, \"workclass\"] == \"Self-emp-inc\", \"workclass\"] = \"Private\"\n",
    "Adults.loc[ Adults.loc[:, \"workclass\"] == \"Federal-gov\", \"workclass\"] = \"Government\"\n",
    "Adults.loc[ Adults.loc[:, \"workclass\"] == \"Local-gov\", \"workclass\"] = \"Government\"\n",
    "Adults.loc[ Adults.loc[:, \"workclass\"] == \"State-gov\", \"workclass\"] = \"Government\"\n",
    "Adults.loc[ Adults.loc[:, \"workclass\"] == \"Without-pay\", \"workclass\"] = \"NoIncome\"\n",
    "Adults.loc[ Adults.loc[:, \"workclass\"] == \"Never-worked\", \"workclass\"] = \"NoIncome\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"marital-status\"] == \"Married-civ-spouse\", \"marital-status\"] = \"Couple\"\n",
    "Adults.loc[ Adults.loc[:, \"marital-status\"] == \"Married-AF-spouse\", \"marital-status\"] = \"Couple\"\n",
    "Adults.loc[ Adults.loc[:, \"marital-status\"] == \"Divorced\", \"marital-status\"] = \"Single\"\n",
    "Adults.loc[ Adults.loc[:, \"marital-status\"] == \"Never-married\", \"marital-status\"] = \"Single\"\n",
    "Adults.loc[ Adults.loc[:, \"marital-status\"] == \"Separated\", \"marital-status\"] = \"Single\"\n",
    "Adults.loc[ Adults.loc[:, \"marital-status\"] == \"Widowed\", \"marital-status\"] = \"Single\"\n",
    "Adults.loc[ Adults.loc[:, \"marital-status\"] == \"Married-spouse-absent\", \"marital-status\"] = \"Single\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Tech-support\", \"occupation\"] = \"WhiteCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Sales\", \"occupation\"] = \"WhiteCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Exec-managerial\", \"occupation\"] = \"WhiteCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Prof-specialty\", \"occupation\"] = \"WhiteCollar\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Craft-repair\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Other-service\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Handlers-cleaners\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Machine-op-inspct\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Adm-clerical\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Farming-fishing\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Transport-moving\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Priv-house-serv\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Protective-serv\", \"occupation\"] = \"BlueCollar\"\n",
    "Adults.loc[ Adults.loc[:, \"occupation\"] == \"Armed-Forces\", \"occupation\"] = \"BlueCollar\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"relationship\"] == \"Wife\", \"relationship\"] = \"Family\"\n",
    "Adults.loc[ Adults.loc[:, \"relationship\"] == \"Own-child\", \"relationship\"] = \"Family\"\n",
    "Adults.loc[ Adults.loc[:, \"relationship\"] == \"Husband\", \"relationship\"] = \"Family\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"relationship\"] == \"Other-relative\", \"relationship\"] = \"Not-in-family\"\n",
    "Adults.loc[ Adults.loc[:, \"relationship\"] == \"Unmarried\", \"relationship\"] = \"Not-in-family\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"race\"] == \"Asian-Pac-Islander\", \"race\"] = \"NotWhite\"\n",
    "Adults.loc[ Adults.loc[:, \"race\"] == \"Amer-Indian-Eskimo\", \"race\"] = \"NotWhite\"\n",
    "Adults.loc[ Adults.loc[:, \"race\"] == \"Other\", \"race\"] = \"NotWhite\"\n",
    "Adults.loc[ Adults.loc[:, \"race\"] == \"Black\", \"race\"] = \"NotWhite\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Cambodia\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"India\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Japan\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"South\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"China\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Iran\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Philippines\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Vietnam\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Iran\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Laos\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Taiwan\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Thailand\", \"native-country\"] = \"Asia\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Hong\", \"native-country\"] = \"Asia\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"United-States\", \"native-country\"] = \"NorthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Canada\", \"native-country\"] = \"NorthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Outlying-US(Guam-USVI-etc)\", \"native-country\"] = \"NorthAmerica\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Puerto-Rico\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Cuba\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Honduras\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Jamaica\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Mexico\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Portugal\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Jamaica\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Dominican-Republic\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Ecuador\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Haiti\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Columbia\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Dominican-Republic\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Guatemala\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Nicaragua\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Guatemala\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"El-Salvador\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Trinadad&Tobago\", \"native-country\"] = \"SouthAmerica\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Peru\", \"native-country\"] = \"SouthAmerica\"\n",
    "\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"England\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Germany\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Greece\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Italy\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Poland\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Greece\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Ireland\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"France\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Hungary\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Italy\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Poland\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Scotland\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Yugoslavia\", \"native-country\"] = \"Europe\"\n",
    "Adults.loc[ Adults.loc[:, \"native-country\"] == \"Holand-Netherlands\", \"native-country\"] = \"Europe\"\n",
    "\n",
    "# One-hot encode categorical data\n",
    "Adults.loc[:, \"Female\"] = (Adults.loc[:, \"sex\"] == \"Female\").astype(int)\n",
    "Adults.loc[:, \"Male\"] = (Adults.loc[:, \"sex\"] == \"Male\").astype(int)\n",
    "Adults.loc[:, \"income<=50K\"] = (Adults.loc[:, \"income\"] == \"<=50K\").astype(int)\n",
    "Adults.loc[:, \"income>50K\"] = (Adults.loc[:, \"income\"] == \">50K\").astype(int)\n",
    "Adults.loc[:, \"Private\"] = (Adults.loc[:, \"workclass\"] == \"Private\").astype(int)\n",
    "Adults.loc[:, \"Government\"] = (Adults.loc[:, \"workclass\"] == \"Government\").astype(int)\n",
    "Adults.loc[:, \"NoIncome\"] = (Adults.loc[:, \"workclass\"] == \"NoIncome\").astype(int)\n",
    "Adults.loc[:, \"Couple\"] = (Adults.loc[:, \"marital-status\"] == \"Couple\").astype(int)\n",
    "Adults.loc[:, \"Single\"] = (Adults.loc[:, \"marital-status\"] == \"Single\").astype(int)\n",
    "Adults.loc[:, \"WhiteCollar\"] = (Adults.loc[:, \"occupation\"] == \"WhiteCollar\").astype(int)\n",
    "Adults.loc[:, \"BlueCollar\"] = (Adults.loc[:, \"occupation\"] == \"BlueCollar\").astype(int)\n",
    "Adults.loc[:, \"Family\"] = (Adults.loc[:, \"relationship\"] == \"Family\").astype(int)\n",
    "Adults.loc[:, \"Not-in-family\"] = (Adults.loc[:, \"relationship\"] == \"Not-in-family\").astype(int)\n",
    "Adults.loc[:, \"White\"] = (Adults.loc[:, \"race\"] == \"White\").astype(int)\n",
    "Adults.loc[:, \"NotWhite\"] = (Adults.loc[:, \"race\"] == \"NotWhite\").astype(int)\n",
    "Adults.loc[:, \"Asia\"] = (Adults.loc[:, \"native-country\"] == \"Asia\").astype(int)\n",
    "Adults.loc[:, \"NorthAmerica\"] = (Adults.loc[:, \"native-country\"] == \"NorthAmerica\").astype(int)\n",
    "Adults.loc[:, \"SouthAmerica\"] = (Adults.loc[:, \"native-country\"] == \"SouthAmerica\").astype(int)\n",
    "Adults.loc[:, \"Europe\"] = (Adults.loc[:, \"native-country\"] == \"Europe\").astype(int)\n",
    "\n",
    "# Remove obsolete columns\n",
    "Adults = Adults.drop(\"education\", axis=1)\n",
    "Adults = Adults.drop(\"sex\", axis=1)\n",
    "Adults = Adults.drop(\"income\", axis=1)\n",
    "Adults = Adults.drop(\"workclass\", axis=1)\n",
    "Adults = Adults.drop(\"marital-status\", axis=1)\n",
    "Adults = Adults.drop(\"occupation\", axis=1)\n",
    "Adults = Adults.drop(\"relationship\", axis=1)\n",
    "Adults = Adults.drop(\"race\", axis=1)\n",
    "Adults = Adults.drop(\"native-country\", axis=1)\n",
    "Adults = Adults.drop(\"income<=50K\", axis=1)\n",
    "Adults = Adults.drop(\"Male\", axis=1) # Redundant since Female = 0 means Male\n",
    "Adults = Adults.drop(\"Couple\", axis=1) # Redundant since Couple = 0 means Single\n",
    "Adults = Adults.drop(\"BlueCollar\", axis=1) # Redundant since BlueCollar = 0 means WhiteCollar\n",
    "Adults = Adults.drop(\"Family\", axis=1) # Redundant since Family = 0 means Not-in-family\n",
    "Adults = Adults.drop(\"NotWhite\", axis=1) # Redundant since NotWhite = 0 means White\n",
    "\n",
    "# Initialize parameters used in data analysis\n",
    "print ('\\nDetermine Model Accuracy')\n",
    "TestFraction = 1/3 # ratio of test data over all data (ratio = 1/3 based on the specification from adult.name file)\n",
    "C_parameter = 50. / len(Adults) # parameter for regularization of the model\n",
    "class_parameter = 'ovr' # parameter for dealing with multiple classes\n",
    "penalty_parameter = 'l1' # parameter for the optimizer (solver) in the function\n",
    "solver_parameter = 'saga' # optimization system used\n",
    "tolerance_parameter = 0.1 # termination parameter\n",
    "binarizedColumns = ['age-bin', 'hours-per-week-bin'] # use to filter the columns from analysis\n",
    "noneBinarizedColumns = ['age', 'hours-per-week'] # use to filter the columns from analysis\n",
    "\n",
    "print ('\\nsklearn accurate split:')\n",
    "TrainSet, TestSet = train_test_split(Adults, test_size=TestFraction)\n",
    "print ('Test size should have been ', \n",
    "       np.round(TestFraction*len(Adults), 0), \"; and is: \", len(TestSet))\n",
    "\n",
    "##########################################################################\n",
    "# 1. Pick Class 1: income>50K = true (1), Class 0: income>50K = false (0)\n",
    "# Target columns: income>50K\n",
    "# Input columns: age, fnlwgt, education-num, capital-gain, capital-loss,\n",
    "#                hours-per-week, Female, Private, Government, NoIncome\n",
    "#                Single, WhiteCollar, Not-in-family, White, Asia, NorthAmerica\n",
    "#                SouthAmerica, Europe\n",
    "# Note: We use non-binarized columns (age, hours-per-week) to analyzied data\n",
    "##########################################################################\n",
    "print ('\\nUse logistic regression to predict if the person make >50K per year from other variables in Adults with non-binarized columns')\n",
    "Target = \"income>50K\"\n",
    "Inputs = list(Adults.columns)\n",
    "Inputs.remove(Target)\n",
    "Inputs = [x for x in Inputs if x not in binarizedColumns]\n",
    "\n",
    "clf = LogisticRegression(C=C_parameter, multi_class=class_parameter, penalty=penalty_parameter, solver=solver_parameter, tol=tolerance_parameter)\n",
    "clf.fit(TrainSet.loc[:,Inputs], TrainSet.loc[:,Target])\n",
    "BothProbabilities = clf.predict_proba(TestSet.loc[:,Inputs])\n",
    "probabilities = BothProbabilities[:,1]\n",
    "\n",
    "print ('\\nConfusion Matrix and Metrics of Class 1: income>50K = true and Class 2 : income>50K = false with non-binarized columns')\n",
    "Threshold = 0.5 # Some number between 0 and 1\n",
    "print (\"Probability Threshold is chosen to be:\", Threshold)\n",
    "predictions = (probabilities > Threshold).astype(int)\n",
    "CM = confusion_matrix(TestSet.loc[:,Target], predictions)\n",
    "print(\"Confusion Matrix:\")\n",
    "print(\"{}\".format(CM))\n",
    "tn, fp, fn, tp = CM.ravel()\n",
    "print (\"TP, TN, FP, FN:\", tp, \",\", tn, \",\", fp, \",\", fn)\n",
    "AR = accuracy_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Accuracy rate:\", np.round(AR, 2))\n",
    "P = precision_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Precision:\", np.round(P, 2))\n",
    "R = recall_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Recall:\", np.round(R, 2))\n",
    "F1 = f1_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"F1 score:\", np.round(F1, 2))\n",
    "\n",
    " # False Positive Rate, True Posisive Rate, probability thresholds\n",
    "fpr, tpr, th = roc_curve(TestSet.loc[:,Target], probabilities)\n",
    "AUC = auc(fpr, tpr)\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8, 8] # Square\n",
    "font = {'family' : 'Sans', 'weight' : 'bold', 'size' : 18}\n",
    "matplotlib.rc('font', **font)\n",
    "plt.figure()\n",
    "plt.title('ROC Curve of Class 1: income>50K = true and Class 2 : income>50K = false with non-binnarized columns')\n",
    "plt.xlabel('FPR')\n",
    "plt.ylabel('TPR')\n",
    "plt.plot(fpr, tpr, LW=3, label='ROC curve (AUC = %0.2f)' % AUC)\n",
    "plt.plot([0, 1], [0, 1], color='navy', LW=3, linestyle='--') # reference line for random classifier\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "##########################################################################\n",
    "# 2. Pick Class 1: income>50K = true (1), Class 0: income>50K = false (0)\n",
    "# Target columns: income>50K\n",
    "# Input columns: age-bin, fnlwgt, education-num, capital-gain, capital-loss,\n",
    "#                hours-per-week-bin, Female, Private, Government, NoIncome\n",
    "#                Single, WhiteCollar, Not-in-family, White, Asia, NorthAmerica\n",
    "#                SouthAmerica, Europe\n",
    "# Assumption: using binarized columns should make the accuray decrease since we lost of data's frequencies\n",
    "# Note: We use binarized columns (age-bin, hours-per-week-bin) to analyzied data\n",
    "##########################################################################\n",
    "print ('\\nUse logistic regression to predict if the person make >50K per year from other variables in Adults with binarized columns')\n",
    "Target = \"income>50K\"\n",
    "Inputs = list(Adults.columns)\n",
    "Inputs.remove(Target)\n",
    "Inputs = [x for x in Inputs if x not in noneBinarizedColumns]\n",
    "\n",
    "clf = LogisticRegression(C=C_parameter, multi_class=class_parameter, penalty=penalty_parameter, solver=solver_parameter, tol=tolerance_parameter)\n",
    "clf.fit(TrainSet.loc[:,Inputs], TrainSet.loc[:,Target])\n",
    "BothProbabilities = clf.predict_proba(TestSet.loc[:,Inputs])\n",
    "probabilities = BothProbabilities[:,1]\n",
    "\n",
    "print ('\\nConfusion Matrix and Metrics of Class 1: income>50K = true and Class 2 : income>50K = false with binarized columns')\n",
    "Threshold = 0.5 # Some number between 0 and 1\n",
    "print (\"Probability Threshold is chosen to be:\", Threshold)\n",
    "predictions = (probabilities > Threshold).astype(int)\n",
    "CM = confusion_matrix(TestSet.loc[:,Target], predictions)\n",
    "print(\"Confusion Matrix:\")\n",
    "print(\"{}\".format(CM))\n",
    "tn, fp, fn, tp = CM.ravel()\n",
    "print (\"TP, TN, FP, FN:\", tp, \",\", tn, \",\", fp, \",\", fn)\n",
    "AR = accuracy_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Accuracy rate:\", np.round(AR, 2))\n",
    "P = precision_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Precision:\", np.round(P, 2))\n",
    "R = recall_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Recall:\", np.round(R, 2))\n",
    "F1 = f1_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"F1 score:\", np.round(F1, 2))\n",
    "\n",
    " # False Positive Rate, True Posisive Rate, probability thresholds\n",
    "fpr, tpr, th = roc_curve(TestSet.loc[:,Target], probabilities)\n",
    "AUC = auc(fpr, tpr)\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8, 8] # Square\n",
    "font = {'family' : 'Sans', 'weight' : 'bold', 'size' : 18}\n",
    "matplotlib.rc('font', **font)\n",
    "plt.figure()\n",
    "plt.title('ROC Curve of Class 1: income>50K = true and Class 2 : income>50K = false with binarized columns')\n",
    "plt.xlabel('FPR')\n",
    "plt.ylabel('TPR')\n",
    "plt.plot(fpr, tpr, LW=3, label='ROC curve (AUC = %0.2f)' % AUC)\n",
    "plt.plot([0, 1], [0, 1], color='navy', LW=3, linestyle='--') # reference line for random classifier\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "##########################################################################\n",
    "# 3. Pick Class 1: Female = true (1), Class 0: Female = false (0)\n",
    "# Target columns: Female\n",
    "# Input columns: age, fnlwgt, education-num, capital-gain, capital-loss,\n",
    "#                hours-per-week, Female, Private, Government, NoIncome\n",
    "#                Single, WhiteCollar, Not-in-family, White, Asia, NorthAmerica\n",
    "#                SouthAmerica, Europe\n",
    "# Note: We use non-binarized columns (age, hours-per-week) to analyzied data\n",
    "##########################################################################\n",
    "print ('\\nUse logistic regression to predict if the person is female from other variables in Adults with non-binarized columns')\n",
    "Target = \"Female\"\n",
    "Inputs = list(Adults.columns)\n",
    "Inputs.remove(Target)\n",
    "Inputs = [x for x in Inputs if x not in binarizedColumns]\n",
    "\n",
    "clf = LogisticRegression(C=C_parameter, multi_class=class_parameter, penalty=penalty_parameter, solver=solver_parameter, tol=tolerance_parameter)\n",
    "clf.fit(TrainSet.loc[:,Inputs], TrainSet.loc[:,Target])\n",
    "BothProbabilities = clf.predict_proba(TestSet.loc[:,Inputs])\n",
    "probabilities = BothProbabilities[:,1]\n",
    "\n",
    "print ('\\nConfusion Matrix and Metrics of Class 1: Female = true and Class 2 : Female = false with non-binarized columns')\n",
    "Threshold = 0.5 # Some number between 0 and 1\n",
    "print (\"Probability Threshold is chosen to be:\", Threshold)\n",
    "predictions = (probabilities > Threshold).astype(int)\n",
    "CM = confusion_matrix(TestSet.loc[:,Target], predictions)\n",
    "print(\"Confusion Matrix:\")\n",
    "print(\"{}\".format(CM))\n",
    "tn, fp, fn, tp = CM.ravel()\n",
    "print (\"TP, TN, FP, FN:\", tp, \",\", tn, \",\", fp, \",\", fn)\n",
    "AR = accuracy_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Accuracy rate:\", np.round(AR, 2))\n",
    "P = precision_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Precision:\", np.round(P, 2))\n",
    "R = recall_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"Recall:\", np.round(R, 2))\n",
    "F1 = f1_score(TestSet.loc[:,Target], predictions)\n",
    "print (\"F1 score:\", np.round(F1, 2))\n",
    "\n",
    " # False Positive Rate, True Posisive Rate, probability thresholds\n",
    "fpr, tpr, th = roc_curve(TestSet.loc[:,Target], probabilities)\n",
    "AUC = auc(fpr, tpr)\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"] = [8, 8] # Square\n",
    "font = {'family' : 'Sans', 'weight' : 'bold', 'size' : 18}\n",
    "matplotlib.rc('font', **font)\n",
    "plt.figure()\n",
    "plt.title('ROC Curve of Class 1: Female = true and Class 2 : Female = false with non-binarized columns')\n",
    "plt.xlabel('FPR')\n",
    "plt.ylabel('TPR')\n",
    "plt.plot(fpr, tpr, LW=3, label='ROC curve (AUC = %0.2f)' % AUC)\n",
    "plt.plot([0, 1], [0, 1], color='navy', LW=3, linestyle='--') # reference line for random classifier\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***\n",
    "**Summary:** \n",
    "The classification that we interested are:\n",
    "1. if the person make >50K per year (income>50K)\n",
    "2. if the person is female (Female)\n",
    "\n",
    "    - Based on the Probability Threshold 0.5\n",
    "    - From the Sample of 32,561\n",
    "    - We split the data to Train (21,707) and Test (10,854) using scikit learn package\n",
    "    - From model, the Classifier: income>50K yeild better AUC (0.89) than Classifier: Female (0.77)\n",
    "    - This model does quite a good job of distinguishing the positive and the negative values since the AUC = 0.89 which is large\n",
    "    - Therefore, the logistic regression seems to fit for prediction if the person make >50K per year\n",
    "    - The performance of the Classifier: income>50K changes a bit (from AUC = 0.89 to AUC = 0.88) when using binarized columns as the initial assumption\n",
    "***"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}