Hitchhiker's Guide to Machine Learning Algorithms in Python.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hitchhiker's Guide to Machine Learning Algorithms in Python"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook/post/project will aim to implemement the most common machine learning algorithms in python. I believe above all else, this will be a good learning exercise for me. I just finished Andrew Ng's Machine Learning course on Coursera and while it is an excellent course, it's quite light on practical implementation. This aims to bridge that gap. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Supervised Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The majority of the algorithms covered here will be of the supervised variety. What this means is that we know the correct answers going into implementation. So we can correct the algorithm and identify error points and other things of that nature, serving as something of a 'teacher' role for the learning algorithm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Linear Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Perhaps the most popular machine learning algorithm out there and definitely the most unappreciated. Many data scientists get caught up and forget that simpler is almost always perferred over complex when performance is similar. \n",
    "\n",
    "Anyways, linear regression is a supervised learning algorithm that predicts an outcome based on continuous features. Linear regression is a versitile algorithm that has the ability to be run on a single variable (simple linear regression) or on many features (multiple linear regression). The way it works is by assigning optimal weights to the variables in order to create a line (ax + b) that will be used to predict output."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>108</td>\n",
       "      <td>392.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19</td>\n",
       "      <td>46.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13</td>\n",
       "      <td>15.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>124</td>\n",
       "      <td>422.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>40</td>\n",
       "      <td>119.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     X      Y\n",
       "0  108  392.5\n",
       "1   19   46.2\n",
       "2   13   15.7\n",
       "3  124  422.2\n",
       "4   40  119.4"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import linear_model\n",
    "df = pd.read_csv('linear_regression_df.csv')\n",
    "df.columns = ['X', 'Y']\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x1176d8a58>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVgAAAFaCAYAAABbpEVLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1176d8b70>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_context(\"notebook\", font_scale=1.1)\n",
    "sns.set_style(\"ticks\")\n",
    "\n",
    "\n",
    "sns.lmplot('X','Y', data=df)\n",
    "plt.ylabel('Response')\n",
    "plt.xlabel('Explanatory')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficient: \n",
      " [[ 3.43948437]]\n",
      "Intercept: \n",
      " [ 24.00493255]\n",
      "R^2 Value: \n",
      " 0.703654905355\n"
     ]
    }
   ],
   "source": [
    "linear = linear_model.LinearRegression()\n",
    "\n",
    "trainX = np.asarray(df.X[20:len(df.X)]).reshape(-1, 1)\n",
    "trainY = np.asarray(df.Y[20:len(df.Y)]).reshape(-1, 1)\n",
    "\n",
    "testX = np.asarray(df.X[:20]).reshape(-1, 1)\n",
    "testY = np.asarray(df.Y[:20]).reshape(-1, 1)\n",
    "\n",
    "linear.fit(trainX, trainY)\n",
    "linear.score(trainX, trainY)\n",
    "\n",
    "print('Coefficient: \\n', linear.coef_)\n",
    "print('Intercept: \\n', linear.intercept_)\n",
    "print('R^2 Value: \\n', linear.score(trainX, trainY))\n",
    "\n",
    "predicted = linear.predict(testX)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First and foremost, logistic regression is a supervised classification algorithm and therefore is useful for estimating discrete values. It is typically used for predicting the probability of an event using the logistic function in order to get an output between 0 and 1. \n",
    "\n",
    "When first learning this alogrithm, I was under the impression that it was a sort of a niche thing and therefore didn't give it my full attention. I couldn't have been more wrong. Some of the underlying aspects of logistic regression come up in many other important machine learning algorithms like neural networks."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>26.6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>23.3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>28.1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>43.1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25.6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      X  Y\n",
       "0  26.6  0\n",
       "1  23.3  1\n",
       "2  28.1  0\n",
       "3  43.1  1\n",
       "4  25.6  0"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "df = pd.read_csv('logistic_regression_df.csv')\n",
    "df.columns = ['X', 'Y']\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x1199e3a20>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgQAAAFsCAYAAACzTaE8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xm0XGWd7/93zXXmeT5JIAkEMjEoQ0BEiQwaBhlEFE23\nrQKtpIF7V6v3eruVy7VZXu+6DSTe1tD82gmhQTGCIkQCIqOAhCQkJISETCdnnqrq1LDH3x+VU8nh\nDKkzT5/XWqxQVbv2fupM+7Of/Tzfx+O6rouIiIjMat7JboCIiIhMPgUCERERUSAQERERBQIRERFB\ngUBERESYJYFg7dq1k92EMTETPsdM+AwwMz7HTPgMoM8xlcyEzwAz43OM5DN4JnPaYSwW43Of+xw/\n/vGPqa2t7fPaM888w7p16wCor6/n7rvvpqCgYETHWbRoEbt27Rp1eyfbTPgcM+EzwMz4HDPhM4A+\nx1QyEz4DzIzPMZLPMGk9BFu3buXzn/88+/bt6/daLBbjzjvv5P7772fDhg2cfPLJ3HfffRPfSBER\nkVli0gLBo48+yne+8x0qKyv7vWZZFt/97nepqKgA0kmnqalpopsoIiIya/gn68B33XUXAAPdsSgu\nLmblypUAJJNJ1q9fz+rVqye0fSIiIrPJpAWCbESjUb7+9a+zePFirrrqqhHtI5lMArB//358Pt9Y\nNm9SHDp0aLKbMGoz4TPAzPgcM+EzgD7HVDITPgNM789h2zaQPv+Fw+Gs3zepgwoBLrroIn7xi1/0\nG1TY2trKl7/8Zc477zy+9a1vZbWvtWvXZgYiioiISF+33nora9asGfC1KRkIHMfhuuuu45JLLuGW\nW24Z1f7379/PJZdcwoMPPkh1dfVomysiIjKlNTU1ceONN7Jx40bmzZuX9fsm/ZaBx+PJ/P9NN93E\nbbfdRmNjIzt37sR1XZ566ik8Hg9Lly7NjDsYjt7bBNXV1dTX149Zu0VERKay4d4mn/RAsGnTpsz/\nr1+/HoAlS5awY8eOyWqSiIjIrDMrKhWKiIjI0BQIRERERIFAREREFAhEREQEBQIRERFBgUBERERQ\nIBAREREUCERERAQFAhEREUGBQERERFAgEBERERQIREREBAUCERERQYFAREREUCAQERERFAhEREQE\nBQIRERFBgUBERERQIBAREREUCERERAQFAhEREUGBQERERFAgEBERERQIREREBAUCERERQYFARERE\nUCAQERERFAhEREQEBQIRERFBgUBERERQIBAREREUCERERAQFAhEREUGBQERERFAgEBERESY5EMRi\nMa644goOHz7c77V33nmH6667jssuu4x/+qd/wnGcSWihiIjI7DBpgWDr1q18/vOfZ9++fQO+/o1v\nfIN//ud/5qmnnsJxHB555JGJbaCIiMgsMmmB4NFHH+U73/kOlZWV/V47fPgwqVSK5cuXA3D11Vfz\nhz/8YaKbKCIiMmv4J+vAd911FwCu6/Z7raWlhYqKiszjyspKmpubJ6xtU41lOax79C3eb+zmxJoi\nbv3M6fj9Gv4xnTiOy6bXD/B+Y4RYT4rdh7pJGRblxTnMryskmXLIzQlwQnUhruvy4pbDuLiUFYZp\n60pwsCWGz+dh2fwy8HjYtqcN23YBF8e2icRtjv1N8njA6/GQF/YTi5s4QDjgpaw4TFNbHLv/r924\n8nhggF/1jHDAS8p04APbeQCvFzweD14PWJaLc+T5cNB3ZHuXkoIwp59czvuHo7R2xkkYNrkhP8sW\nlnPqvBI2/HkPLZ0JbMdNXwV50vu3naPtK8oL4LiQMm18Xi8L64q48Mw5fOLsuTiOy33/+SYvb2vC\ntGxCQR9lhWHKinO44LRaPB4v+5sjnFBdyMqz5gIc+X53E09Yme8tuOxrihBPWOSE/Oxp6CKRSv//\ngvpi5tcWsfKsuXi9ngG/Tr0/R/uaIsyrKsBx4aUth8HjcsFpdXzi7Hn93nvse3rbN9j+h2O89jtT\n2zUdTFogGMpAIcHrnb0nwHWPvsWLW9LjLBpaegC4/XNnTmaTZJg2vX6AJ19+n0iPQVtXAufIj3hr\nV5L3D0dwXSjKD/KXtxtJpCwM08F2HBzHxXXJnOz/tLkBGPrk2vu67bpE4mbmuaTp0NAaH4dPd3zH\na2/SPHJm/sB2Lr0nbbff8wnDzjxubI/T/OqBzNcVIJ60eH5zAy+81YB1TAKyBzqOC10x85hnHLbt\nbaepI47HA9v3tvOnzQ2Zz5FI2Rxq7aGpI87ehm5yQn4K84Ls2Nue2UPv97s7ZmS+t726YwYeTzp8\neI4051BLjJ37OgC4+Jx5A36den+OAF7ddvRnBeBwaw8ej6ffe499T2/7Btv/cIzXfkdrqrZrOpiS\ngaCqqorW1tbM45aWlgFvLXzQ2rVrWbdu3Xg2bVK839g95GOZ+vY1RYD0CeCDJ0fDsgn4fKRMO7ON\nB086CLh9z13HO7HOZs4AXxvXdbFGOB7ZPdJbsK8pwvuN3QN+7Xu3OfYKtPd7DfT5nn6QcaRhrgt4\njm5z7Ps/6IP77v1Z6X080Hs/+NxQ+x+O8drvaE3Vdk2GlStX9nvu1ltvZc2aNQNuPyUDQW1tLaFQ\niM2bN3PGGWfw29/+lo9+9KPHfd+aNWv6fdBDhw4N+EWZTk6sKcr0DPQ+lunlhOpCduxtJxTw9es+\nD/p9uC6EAj4g3eVpmA4eT7orm2NCgefIeUfBoD+vp38o8Hg8+Lz06SHIlseT/p6cUF1IT9zM9OQM\ntE3v9w44cmuAzPc7mbL7vA6QTNkE/d70Cd2T/v72btP7/oH0/hxxZPven5XexwO999j3HG//wzFe\n+x2tqdqu0XBdF8dxcBwH27ZxHBfbtrGd9G1D1wXbcfDgwXFdOrrSIWjTpk3U19dnfZxJDwQez9Fk\nfdNNN3HbbbexZMkSfvCDH/BP//RPxONxTj31VL74xS9OYisn162fOR2gzxgCmV567ytrDMHApvIY\ngpVnzeXjH5qD4zjDGkMAjHgMwWB6XxtsDMFA7z32PR9s32iM135Hayq1q/dEbts2lmVjWla618p2\nMj/ntuPiuumTuuO6OEdSbe9j95irAQ8ePF4vXq83/Tvh9ZL5YT5mjoDpjOzU7nEHumE/g/T2EAw3\nKYmIiByr9+SeMkwsy86M83FcsI+c5HtP6o7r4joueL14PF58Pl/mRD4WTMuhO5aiO5aiK5Yi0mOk\n/4sZNDY28NR/fHP69RCIiIhMJtd1sSyLlGFiGGb6BO+42LaTOcHbtosLeL0+vD4fPp8POOZWkPfo\nP6MdAm9aNp3RFB3dSToiSTqjKbqi6X87I+kA0JMwB39/vHNEx1UgEBGRGa33hJ9MpkiZFq6bvsK2\nbQfbSZ/wPV4f/kAAr7fvSd4D+HzgC4xdexzHpSOSpK0rQWtXgvbuBG1dSdq7E7R3J4n0GKPafzg4\nslO7AoGIiMwIlmWRSCQxLBvbdjEtG8t2sB3weLwEgkE8niOnPW96fMp4TWh3XJeuaIrmjjjNHXFa\njvzb1pWgrStxZEDg8OSG/BQXhijOD1FckP63MP/Iv3lBivKDFOYFaW9t5vNPDb/NCgQiIjKt2LZN\nPJ4gZVhYtoNpOemBeh4vgUDw6FW+z4/fN74nOsdxae1K0NjWQ2NbD03tR/7t6MnMAMmG1+OhtDBE\nWVEOpUVhSguP/FcUpqQgHQBGeuWfLQUCERGZsizLItaTwDAtTCt98ndcD4FgEK/XDx7wBiA4hl36\ng4knTQ42xzjUEqWhJUZDa4zDbT2YWRa7CAd9VJTkUlGSQ0VxDpUluZQXhykryqGkMIRvkgvwKRCI\niMiU4LouiUSSeCJFyrTTJ388BIMhPJ4A+OADJR3GTSxhsr8xwv7GCAeaoxxsjtLenTzu+zxAWXEO\nNWV5VJflUlWWS1VJ+t+C3OCYzTIYDwoEIiIyKVzXpSeeIJ5IYZjpq3+vP4Df7weff8JO/qblcLA5\nyt6Gbt4/3M3+xghtWZz8SwvD1FXkUVuRT11FPjXleVSV5hKcqIaPMQUCERGZMMlkklhPkpRpY5gO\nvkAQny+Axw/jfIs8IxY3eO9QN+8d7GJPQxcHm6NDVrP0eT3UVuQzpzKfOVUFzKkqoK4in5zwzDqF\nzqxPIyIiU4rjOESiPSRTJknDBo+PQDAIPj/BCbqQjvSk2LW/k3cPdLH7YCdN7YMv8uX1eKgpz+OE\nmkLm1RQwr6aQ2vJ8ArNghVkFAhERGVOpVIpoLEHCsDAtl2AojMcTJBCamOMnUha79neyc18H7x7o\n5HBbz6DbFuQGOLG2iPl1RZxYW8QJNYWEJiqpTDEKBCIiMmqJRJJoT4JEysLBSzAYwuv3E5qAs4zj\nuhxsirJ9bzs73u9g7+HuzJoAH1RenMPC+mIWziliYX0xVaW5U3qg30RSIBARkRE5NgTg8eMPBPBP\nxPw/IJmyeGdfB9v2tPH2nvZBq/uVFoZZNK+ERXNLOHleCaWF4Qlp32SwbRvLNLDM4w+IHIgCgYiI\nZM0wDLqjcRIpCxffhIaA7liKLbtbeevdVnbt7xyw2l8o6GPR3BIWzy9j8YmlVBTnzMgeANd1MVIp\nfF6XgN9H0O8lnBskN6eAoCc1on0qEIiIyJAcx6ErEqMnYWDZHoKhEL7AxJw+OiJJ3tzZwuZ3W9h7\nqJuBbgTUVuSxbEE5S+eXMb+uCJ9v5g0AtCwLxzIJ+L0EA15CIT/5ZSVHFlkaGwoEIiIyoJ544sjg\nQIdAMITXH56QqYHdsRRv7mzh9Xea2dvQ3e91n9fDonklLF9YztIF5ZQX54x/oyaYkUrhwSHg9xIK\n+CgpCpGTUzSuvR0KBCIikuE4Dh1dEeLJ9LgAnz9IcAJmByQNi7d2tfLq243sOtCJ+4GugGDAy5L5\nZZx+ciXLFpSRG56Y2xQTwXVdDCOFz+MSCvgIBf1UFBUQDAYntB0KBCIiQjKZojPSQ/JIb8BE3BJw\nHJed+zv4y9tNbH63pd9iQAG/l2ULyvnQqZUsW1A+bSsADsQwUnhxMgGgurQ4XaFxEikQiIjMYrFY\nD13RBJabXilwInoDOrqTvLztMC9vbaQj0ndEvNfrYcmJZZy1uIrlC8sJT8S8xQmQHgNgEAr6CQV9\nk9IDcDwz4ystIiLDEonG6IomwBPA5w8z3h3wtuOwdXcbL245zI697f0GB86rLuDcpTV8eHEVBblT\n60Q5EulZAEkCfs+RMQBhcnOLJ7tZQ1IgEBGZRboj6SDg8QXxBcZ/MF6kJ8WLWw7zwlsNdEb6Tocr\nyA1w7rIazltWS0153ri3ZbzZloVtmYRDPnJCfgrLy/BO8pLGw6FAICIyC/T0xGnv7sHjDeIPjn8Q\n2NcYYdPrB3hzZ0ufegEeYPH8Ms5fXsvyk8rxT/MpgqlUEr8XwsHp0QswFAUCEZEZLJlM0dYZxXZ9\n+Me5R8BxXLa+18Yzr+3nvUN9pwvm5QT4yGm1XHB63bSeJth7KyDo9xIO+aisKiIQmBkzHhQIRERm\nINu2aWnvJmVCIBge1z/2hmnz8rZGNr12gNauRJ/X5tUU8vEz6/nQqZUE/NNzloDruphGiqDfQ144\nQE1Z6ZgWBJoqFAhERGYQ13Vp7+wmFrcIhMIExnF8XiJl8efNh3jmtQNE42bmeY8HTj+5govPnsf8\nuqLxa8A4OjYE5OcEKCwvnVbjAUZCgUBEZIaIxnpo7+rBH8whMI7T9WIJk2dfP8Bzfz2UXtjoiFDA\nx/mn1fLxD8+hYhreFui9HRAKeGdNCDiWAoGIyDRnmiYt7REsx0sglDtux+lJmDzz+gGefeMgKcPO\nPJ8X9nPRh+fwsQ/NIS9n+t1PT48JSN8OqJ1mMwPGkgKBiMg0lbk9kLAJBEOM1y36eNJk0+sH2fTG\nAZKpo0GgKD/ExWfP5SOn1xKeiEUOxpBhpPB7XHLCfqprxnaRoOlqen0HRUQEgHgiSWtHFF8gTGCc\nTsaGafPcXw/x9Cv7iB9za6AoP8QnV8zj/NPqCPinz9W0ZVngmOSE/JSX5xMKTUBZxmlEgUBEZBpx\nXZfW9i7iKZfAONUTsB2HV7c18cSLe+mKHi0mVJgX5LIVJ3DB6bXTZsZA77iA3JCPooIQ+fnTt07A\neFMgEBGZJvr2Coz9Mriu67JtTzuPPbebpvZ45vm8sJ/LVpzAhWfWT5sFhkzTwOdxyM8JzupxAcOh\nQCAiMsW5rktbexc949gr0NAa41ebdvPOvo7McwG/l5VnzeWSc+ZOi+WGHcfBMlPkhvyUleYRDuuW\nwHAoEIiITGGGYdDY2oXXHyYQHPur3Gjc4IkX9vLCWw24RyoMezzwkdNqWXX+fIoLpv5J1TQNfDgU\n5AUpqizH4xn73pPZQIFARGSK6uyK0N1jEgiO/VRC23H48+YGHv/z3j61BE49oZTrVp5EXUX+mB9z\nLLmui5lKEg75qFZvwJhQIBARmWIcx6GxpQPb9RMIjv2Jbs+hLh7euIuDLbHMc1WluVx30UksXVA2\npa+wbcvCdUzyczU2YKwpEIiITCHJZIqmtm4CoVzGevheLG7w2HPv8fK2xsxz4aCPyz8yn49/qB7f\nFF550EilCPhcSgtyNFNgnExaIHjiiSf40Y9+hGVZrF69mhtvvLHP6zt27OCf//mfsSyL6upq/s//\n+T/k50/tLiwRkdHI3CIY42qDruvyl7ebePTZ3fQkjq45cNbiKq79+ElTepyAkUqSE/RSV1lAMDiO\nCzPI5ASC5uZm7rnnHjZs2IDf7+eGG27g3HPPZcGCBZltvve973H77bfzkY98hO9///s88MAD3Hbb\nbZPRXBGRceW6Lk0tHZiOb8xvEbR2xnnw6V3sPGb2QE15HjdcvIhF80rG9FhjxXEcbDNFfk6AmtrZ\ntZ7AZJqUQPDKK6+wYsUKCgoKALj00kt5+umn+drXvpbZxnEcotEoAIlEguJidRGJyMxjmiaHWzrx\nBXLw+cfu3r3tOGx6/SBPvLAX03IA8Pu8rDr/RC45Z+6UvD1g2zaubVCUH9JsgUkwKYGgpaWFioqK\nzOOKigq2bdvWZ5tvfetbfOlLX+Jf/uVfyM3N5ZFHHpnoZoqIjKt4IklLe2zMbxE0tvXw09/vYF9j\nJPPcorklfP6yU6gqHb/Fj0bKMk282BQXhCksqDj+G2RcTEogcHsnux7j2C6hVCrFt7/9bX72s5+x\ndOlSfvKTn/CNb3yDH//4xxPZTBGRcdMdidIZNQiEwmO2T8dx+eNrB3jihb1YdrpXIDfk59qLTuK8\n5TVT7orbNA38Hofy4jzycqffcskzzaQEgqqqKt54443M49bWViorKzOP3333XcLhMEuXLgXgs5/9\nLPfee+9x97t27VrWrVs39g0WERlDLW2dJEzGdLxAc0ecn/xuO+8fPtorsHxhOTdedgpF+VNr0KBp\nGAR8LlUleeTkjF0gkr5WrlzZ77lbb72VNWvWDLj9pASCFStWsG7dOjo7OwmHw2zcuJG77ror8/q8\nefNoampi3759nHDCCTzzzDMsW7bsuPtds2ZNvw966NChAb8oIiITzXVdGls6sFw//jFaHMh1XV54\nq4FHN+3OjBXIDfm5/hMnc87S6inVK2AaBiG/S3VZvgoJTYBNmzZRX1+f9faT1kNwxx13sHr1akzT\n5Prrr2fZsmXcdNNN3HbbbSxZsoS77747M6ugrKyMu+++ezKaKiIyJhzHoaGpHY8/jM87NifpSE+K\nnz/5Dtv2tGeeW7qgjC9cduqUmkpoGgZBBYEpb9LqEKxatYpVq1b1eW79+vWZ/7/gggu44IILJrpZ\nIiJjLj2ToAv/GC5MtPW9Nn7+5A6i8XRdgWDAy3UXncQFp9dNmV4B0zQIeBUEpgtVKhQRGUfpyoNR\nAqGxCQOm5fDYc7t57q+HMs/Nqynk765YMmVmENi2DY5BRXG+BgtOIwoEIiLj5Oi0wrEZONfcEeff\nf/s2B5vTNVo8HvjkihNYdf6JU6KugOu6WEaSksIwRYWaPjjdKBCIiIyDWE+ctq7EmIWBV99u5KGn\nd5EybQBKCkL83ZVLOGnO5FcbdF0X00hSkBugrEIFhaYrBQIRkTEWifbQEUmOybRCw7R5+I+7eHnr\n0QWJli8sZ/WqxeTnBEa9/9EykgnycvzU1mrlwelOgUBEZAx1dUfpipljEgZaO+Os/822zDLFfp+H\naz5+Eh//UP2kX4UbqSS5IS81dQoCM4UCgYjIGOnsitAdtwiMwap8b73byk9/v4NEygKgvCjMV69e\nxrzqwlHvezRM0yDkc6mvKiIQmPweChk7CgQiImMgEwYCowsDtuPw2+f3svEv+zPPLV9Yzt9cvpi8\n8OSdgG3bxuOYVJbkk6vqgjOSAoGIyCh1dEWIJuxRh4FYwuTff/t2Zqlijweu+ugCLjl3Ht5JvEVg\nJBOUFoUpKiyftDbI+FMgEBEZhd4w4PeP7ur9UHOUHz22lbbuJAAFuQG+ctVSFs0rHYtmjohpGIQD\nME/jBGYFBQIRkREaqzDwxjvN/OzJHRhmei2CedUF3HzNckoLJ6dr3nEcXDtFZWmBbg/MIgoEIiIj\n0NUdJTLKMQOO6/L4n/fw1CtHxwucu7SGz1+6iGBgbBY/Gi4jlaQoP0hpsQoLzTYKBCIiw9QdidHd\nM7owkDQs/uOJHWzZ3QqA1+PhupWTN6XQtiy8WJo9MIspEIiIDEM01kNXzMA/ijDQ0Z3k//16C4eO\n1BfIC/u56erlLJo3OVUH04MGcygqLJ6U48vUoEAgIpKlnp447d2jq0C4t6Gbf/v1lswqhdVluXz9\nutOoKJn4hYlMwyAniIoLCaBAICKSlUQyRWtXYlRh4I13mvnJ73Zg2enBg0vml/GVK5eSE57YP8Ua\nNCgDUSAQETkOwzBoaosQHOESxq7r8vSr+9nw/J7Mcxd9eA7XXrQQ3wRfmWvQoAxGgUBEZAiWZdHY\n2j3iMGDbDr98ehcvbT0MpIsN3XDxIi48s34sm3n8dmjQoByHAoGIyCAcx6GhqQN/aGT39xNJi/Ub\ntvHOkcqDoaCPr1y1lGULJrbin5lKUFqUS2GBBg3K4BQIREQG4LouDU3tIw4DXdEUax95i4bW9EyC\novwQt37mNOZUFYxlM4dkmgZhP9RoaWLJggKBiMgAGls68PhHNuCusa2HtY+8RUckXYa4riKfWz9z\nGiUTVHnQdV0sI0FFST55eRM/e0GmJwUCEZEPaGnrxCYwogWF3jvUxf/71RbiyfSyxafMK+Hma5aT\nE5qYP7e9Uwnr6iompcCRTF8KBCIix+jsipA0Pfj8w+9if+vdVh54/G1MKz2t8KzFVfzNqsX4fePf\nXe+6LraZ1FRCGTEFAhGRIyLRnnRJ4uDwqxC+8FYDv3x6J66bfnzxOXO5+mMLJ2TZ4kyvQG25egVk\nxBQIRESAeCJJe3eCYGh4V9cfrDHgAa5beRIrz5o7Dq3sf2zLSFBZVqheARk1BQIRmfUMw6C5PTrs\nWgOO6/LrZ3ez6fWDAHi9Hr50+WLOWlw9Hs3swzQNcgIaKyBjR4FARGY1x3E43NJFcJjTC23b4ed/\neIdX324CIBjwcvPVy1kyv2w8mtmHkUxQUZpHvmYQyBhSIBCRWau31kBgmGHAtGzu3/A2W99rAyA3\n7OfWz5zO/Lqi8Whmhm1Z+DwWc2tL8fl843osmX0UCERk1mpsbh92rYGkYfGjX29l5/5OIF1w6LbP\nnk5tRf54NDHDSCUpLQxpiWIZNwoEIjIrtbR1YhHAN4z77/GkybpHt7C3oRuA8uIcbr/hDMqLR7bO\nQTYcx8G1UtRXFWsNAhlXCgQiMut0dUdJGOAPZN/tHukxuO8/N3OoJV2KuKY8j9s+ewbFBSNfDvl4\nTMMgP+ylvEorE8r4UyAQkVklnkjSFTMIBLM/kXdGktzz8GaaO+IAzK0u4B+uP5383OHXK8iGphPK\nZFAgEJFZwzRNWtpjBIZRa6CtK8E9D71JW3d6XYKF9cV8/TOnjVspYssyCflcTSeUCadAICKzguu6\nHG7pHNaMguaOOPc8/CadkRQAi08s5ZZrlhMcxq2G4TBSCcqLcynIzxuX/YsMRYFARGaFxpYOfIHs\nB/8dbotx78Ob6Y4ZACxfWM5XP72UgH/sw4Bj23hck7k1mk4ok0eBQERmvLaOLmzXj9ebXRf8oZYo\n9z68mWjcBOCMRRV8+cql47JIkZlKUlwQpLiofMz3LTIcWf10NzQ0jPmBn3jiCVatWsWll17Kgw8+\n2O/1vXv38sUvfpGrrrqKr3zlK0Sj0TFvg4jMfNFYDz1JF2+WV94HmqL86y/fzISBs5dU85Wrxj4M\n2LaNYyaoqyqiuKhwTPctMhJZ/YR/9rOf5ZZbbuH5558fk4M2Nzdzzz338PDDD7NhwwYeeeQR9uzZ\n02ebr33ta9xyyy389re/ZfHixaxfv35Mji0is0cymaK9O4k/y/n7+5si3PPwm/QkLQDOW17D365a\njM87tmHANFIUhD3Mqa1QbQGZMrL6KX/uuee44oor+Pd//3cuvvhiHnjgAbq6ukZ80FdeeYUVK1ZQ\nUFBATk4Ol156KU8//XTm9e3bt5Obm8v5558PwE033cSNN9444uOJyOxjWRZNbZGspxfua4xw70Ob\niR8JAxecXssXPnlq1rcZsuG6LraRoKa8gJJi9QrI1JJVIAgEAqxatYqf//znfO973+OXv/wlF154\nIf/tv/03mpubh33QlpYWKiqOFtqoqKigqakp83j//v2Ul5fzrW99iyuvvJI777yT3Fwt4iEi2XFd\nl8PNnQSyXL3w/cPd3PvwZuKpdBj46Bl1fO7SU/CO4bQ/yzTxYzCntpxQaHzqF4iMRlaBwHEcNm3a\nxM0338wdd9zBpZdeymOPPcZJJ53E1772tWEf1HXd/g05pkvOtm1ef/11Vq9ezeOPP059fT133333\ncfe7du1aFi1a1Oe/lStXDrt9IjK9NbZ04A1kV2tgb0M39/7nZhJHwsDHP1TP5y5ZNKZhwEglKc4P\nUF1ZptoCMmFWrlzZ75y4du3aQbfPapbBhRdeSFVVFZ///Oe57777CIXSXXALFizgl7/85bAbWVVV\nxRtvvJHxK3HHAAAgAElEQVR53NraSmVlZeZxeXk58+bNY/HixQBcfvnl3Hbbbcfd75o1a1izZk2f\n5w4dOqRQIDKLDGdGwd6Gbu57ZDPJlA3ARR+ew2dWnjRmJ23XdXHMJHWVRQSD6hWQibVp0ybq6+uz\n3j6rHoIf/vCH/OpXv+Kaa67JhIF33nkHgGeeeWbYjVyxYgWvvvoqnZ2dJBIJNm7cyAUXXJB5/cwz\nz6Sjo4Ndu3YB6Q+1ZMmSYR9HRGaXnp44sYST1YyC9w/3DQOfOGvumIYBx7bxuulbBAoDMh0MGQg6\nOjpob2/n29/+dub/29vbaW5uzuqKfTBVVVXccccdrF69mquvvporr7ySZcuWcdNNN7F9+3ZCoRDr\n1q3jf/yP/8EVV1zBa6+9xje/+c0RH09EZj7TNGnp7CGQxcn3/cPp2wS9YWDlWXO49qKFYxYGLNMg\nL+yhtkq3CGT6GPKWwT/+4z/y0ksvAXDeeecdfZPfz8UXXzyqA69atYpVq1b1ee7YqYXLly/n0Ucf\nHdUxRGR26C1LHMyiLPFAYeC6i8auZ8BMJakszSM3d/yWRBYZD0MGggceeACAb37zm3z/+9+fkAaJ\niAxXU5Zlifc1RvqEgYs+PHZhwHEcsFPUV5fg96sIrEw/Q/7U7t69m5NOOokbb7yRrVu39nt9+fLl\n49YwEZFsdHZFMB0fPv/QJ/UDTRHue3h8BhBapkluECqqKo6/scgUNWQg+MEPfsD69eu5/fbb+73m\n8XjYtGnTuDVMROR4enridPeYxy0+dLA5yj3H1Bn42Jn1YxYGjFSCipI88vNUK0WmtyEDQe89/Wef\nfXZCGiMikq3eQYTB4xQfamiJpcNA8mjRoc9efPKow4DruthmgvqqEpUflhlhyEBw//33D/nmr371\nq2PaGBGRbLiuS0Pz8QcRHm6LpdcmSKQXKvrIabXccMmiUYcBx7bxeSzqais0i0BmjCEDwd69eyeq\nHSIiWTvc3E7gOGGgqb2Hex46uoTxectq+Pxloy9HbJkmeWEP5aVlo9qPyFQzZCDIplywiMhEam3r\nxPEEhyyi0tIZ518f2kykxwDgnCXV6YWKRhkGzFSS8pJcjReQGWnIQLBmzRrWrl3LJz/5yQG7xZ58\n8slxa5iIyAd1R6LETfD7B48DbV0J/vWhN+mOpQD48KlV/M2qxaNetdAyEtRUFGlhIpmxhgwEN998\nMwB33nnnhDRGRGQw8USSzqgx5IyCjkiSf33oTToj6TBwxqIKvnT56MKAbdv4MZlTU9ZnETaRmWbI\nn+6lS5cCcPbZZ1NTU8P+/fs5fPgwc+bM4eyzz56QBoqImKZJS3tsyDDQFU3xrw+9SXt3EoDlC8v5\n8pVL8flGfhI3DYPcoEttdbnCgMx4Wf2E//73v+e6667j+eefZ+PGjVxzzTU8//zz4902EREcx6Gx\npYtAaPDljCM9Bvc8/CatnQkAlswv46ufXoZ/VGEgSXlxmPLS4hHvQ2Q6yaq+5o9//GMee+wx6urq\ngPTsgzvuuIMLL7xwXBsnItLY0oEvOHitgVjC5N6HN9PUHgfglBNKueWaZQSGGGcwlN76AnWVqi8g\ns0tWgcDv92fCAMD8+fPVfSYi467lODMK4sl0GGhojQFw0pxivnbtcgL+4y9/PBDbsgh4bdUXkFlp\nyLN673LHS5cuZf369cRiMeLxOD/72c9YsWLFRLVRRGah7kiUhMmgFx+JlMXaR97iYHMUgPl1RXz9\nutMIBkYWBkwjRWGujxotWSyz1JA9BOeffz4ejwfXdQH4v//3/2Ze83g8fOMb3xjf1onIrJRIpuiI\npAgOMm4gZdj88NG3eP9wBIB51QWs+czphEPDX2XQdV0sI0FVeRE54aHXRBCZyYb87dm5c+dEtUNE\nBEhP82tpjwy6RoFh2vy/X2/hvUPdANRX5vMPnz2DnPDww4BtWfi9NvPqdItAJKvfINu2efHFF4nH\n47iui+M47Nu3j1tvvXW82ycis8zh5g78gwwiNC2HH/9mG7v2dwJQU57HbTecQV7O8Af/mUaKkoIg\nRYWaRSACWQaC//pf/ys7duygu7ub+fPn884773DOOeeMd9tEZJZpbu0A38Dd9rbtcP+GbWzf2w5A\nZUkOt99wBgW5w6scmJ5FkKSmXFUHRY6V1VSBrVu38rvf/Y6VK1fyL//yL/ziF78glUqNd9tEZBbp\njkRJWp4BBxHajsMDT2xn63ttAJQXhbnjc2dSlD+8e/62beN1UsytLVcYEPmArAJBZWUlwWCQE044\ngd27d7N06VJisdh4t01EZolItIfOqInf37/r33Fcfvr7d3hzZwsAJQUhbv/cmZQUDl6oaCDHVh3U\neAGR/rIKBIFAgFdeeYWFCxfy/PPP09nZSWdn53i3TURmgXg8QUckSSDY/4rdcV0efGonr21vAqAo\nP8jtnzuT8uLBCxUNxEyp6qDI8WQVCP7xH/+Rxx57jI9+9KPs2bOH8847j6uuumq82yYiM1wimaKl\no2fANQpc1+Xhjbt4aethAApyA9x+w5lUlQ5v6WHLSFBbWagli0WOI6tBhcuXL+cHP/gBAA8//DCR\nSITCwsJxbZiIzGyGYdDcFh1wjQLXdfnVs7v58+YGAPLCfm674QxqyvOy3n/vKoVza3WLQCQbWQWC\nrq4uvv/97/PSSy8RCAT42Mc+xn/5L/+FvLzsfzlFRHrZts3hli4Cof5X7a7rsuH5PWx6/SAAOSE/\n/3DDGdRXFmS9f8s0yQt5KC8rH7M2i8x0Wd0y+M53vkNeXh7r169n7dq1GIbBnXfeOd5tE5EZyHVd\nGpo6BgwDAL978X2efnU/AKGgjzXXn8686ux7JE0jRWlhkPIyjRcQGY6segj27t3Lvffem3n8P//n\n/+TKK68ct0aJyMzV2NyONzDwDIEnX36f37/0PgDBgJc1nzmd+XVFWe/bTCWoLi8krBLEIsOWVQ9B\ncXExHR0dmceRSISCguy770REIL16oUVgwHv6G/+yn8f/vBeAgN/L1687jYVzsrvKT69HEKe+ukRh\nQGSEhuwh+O53vwukVxu79tprueSSS/B6vWzatIlFixZNRPtEZIbo7IqQMMA/wGqEz75xkMeeew8A\nv8/D31+7nEXzSrPar+M4+FxDSxaLjNKQgaC8vDzz71lnnZV5XrcLRGQ4enriROIW/kD/WgN/evMQ\njzzzLgA+r4ebr17O4hPLstqvZZrkBKGyXIMHRUZryEBw7OJFtm2zd+9eLMvipJNOwu8f/spiIjL7\nJJMpWjrjAy5l/MJbDTy8cRcAXq+Hr356GcsWZndyNw2Dojw/JcWaAi0yFrI6q+/Zs4e///u/Jx6P\np+f2+v3cf//9nHLKKePdPhGZxkzTpKlt4KWMX9pymAefSi+x7vV4+MqVSzn95Irs9mukKC/OUbEh\nkTGU1aDC733ve9x22228+OKLvPLKK3zzm9/kf/2v/zXebRORacxxHBqaOwkMEAZe3dbIL/7wDgAe\nD/zdlUs485TKrPZrphJUl+UrDIiMsawCQUdHB6tWrco8vvzyy+nu7h63RonI9JauNdA+YK2Bv2xv\n4qdP7sAFPMDfXr6ED59aldU+bSOhmQQi4ySrQGDbdp/FjI6dgigi8kFNLR14/P3HDLy2vYmf/G47\nrpsOA6tXLeacJdXH3Z9lmgQ8JnNqyzV+SWScZPWb9cUvfpHrr7+eyy67DIA//OEPfOUrXxnXhonI\n9NTS1onp+vF5+04BfG17E/9xTBj4widPZcWymuPuzzRSlBQEKSpU7ROR8ZRVD8EVV1zBXXfdheM4\nWJbFXXfdxQ033DCqAz/xxBOsWrWKSy+9lAcffHDQ7f70pz+xcuXKUR1LRCZGR1eEpOnB5+tba+D1\nHUfDAKTDwPmn1R53f2YqTnVZvsKAyATIqofg+uuv54knnuDcc88dk4M2Nzdzzz33sGHDBvx+Pzfc\ncAPnnnsuCxYs6LNde3s7//t//+8xOaaIjK9ItIdo3MYfCPR5/o13mvn/njgaBr6YRRhwXRfXSjK3\nthyvN6vrFhEZpax+0/Lz8zl06NCYHfSVV15hxYoVFBQUkJOTw6WXXsrTTz/db7tvf/vbfWohiMjU\nFI8n6Igk+4WB13c08cDjb2fCwI2XnXLcMGDbNl7XoL5GYUBkImXVQxCLxbjkkkuorKwkN/foqOEn\nn3xyRAdtaWmhouLofOOKigq2bdvWZ5uf//znLF26lNNOO21ExxCRiZFMpmjp6CHwgcJDf3m7kZ/8\nfkefMHDB6XVD7suyTHIDUFGeXaVCERk7xw0Ezz//PFdffTUej4clS5aMyUHd3r8Qxzj2SuDdd99l\n48aN/PSnP+Xw4cNZ73ft2rWsW7duTNooIsdnGAZNbdF+YeDVbY389PdHpxbe+MlT+MhpQ4cBDR4U\nGVsDjb+79dZbWbNmzYDbDxkIHnjgAR566CGWLl3KG2+8wX//7/+dT33qU6NuZFVVFW+88UbmcWtr\nK5WVR4uSPP3007S2tnLttddiGAbNzc184Qtf4Be/+MWQ+12zZk2/D3ro0CENShQZB6Zp0tja3a/w\n0MtbD/PzJ9/JhIEvfupUzls+9G0CI5mgqryA3JyBl0UWkeHbtGkT9fX1WW8/5A26xx9/nA0bNnDP\nPffw05/+lJ///OejbiDAihUrePXVV+ns7CSRSLBx40YuuOCCzOtr1qzhqaee4je/+Q3r16+nqqrq\nuGFARCaObdscbunCH+wbBl7c0tAnDKxetXjIMOC6Lo6ZYE5NicKAyCQbMhB4vV7y8/MBWLBgAbFY\nbEwOWlVVxR133MHq1au5+uqrufLKK1m2bBk33XQT27dvH5NjiMj4cByHQ03t/cLAn/56kF/8YWc6\nDHjgby9fPGSdgd5iQ/U1KjYkMhUM67fwg3OLR2PVqlV9yiEDrF+/vt92dXV1bNq0acyOKyIj11uS\n2B/sW5L4mdcO8KtndwPphYq+dMVizlo8eAVCyzQozg9ovIDIFDKsQODxeI6/kYjMSI7j0NDU3q8k\n8R9e3sdv/7wHAJ/Xw1euWsoZiwZfqMhMJakoySVPixOJTClDBoKdO3f2mfZnGAannXYaruvi8XjY\nsmXLuDdQRCaf4zgcamzHd8xtAtd1eeKFvTz58j4A/D4PN129nOULywfdj5lKUF1eqMWJRKagIQPB\nM888M1HtEJEpynEcDja29blN4Louj27azbNvHAQg4PdyyzXLWTJ/8PoBVipOfXWpxguITFFD/mbW\n1Q09b1hEZraBwoDjuDz49E5e2pKuERIK+PjadctZNK90wH30ziSYW1eh244iU5iiuogMyDTNI1ML\nj4YB23b4ye938PqOZgByQ35uvf505tcVDbgPx3HwuQZ1tQoDIlOdAoGI9JNKGf2KDpmWzf0b3mbr\ne20AFOQG+IfPnsGcqoFnCji2TcBrUV01+JgCEZk6FAhEpI94IklLe6xPGEikLP7t11t490AXAMUF\nIW6/4Qyqy/IG3IdtWeQEXK1JIDKNKBCISEY01kNbV4LgMWsTROMGax95iwNNUQDKi3O4/YYzKC/O\nGXAfpmlQmOuntLhwQtosImNDgUBEAGjr6CKWcPqEgY7uJPf+52aaO+IA1FXk8w+fPZ2i/IGnDZqG\nQUmBCg6JTEcKBCKznOM4HG7uwPUGCQSP/kloau/h3oc30xlNAbCgvoivXXcaeeHAgPsxjRSlhWEK\nCwa+jSAiU5sCgcgslkimaGmP4A/mcOwcgL0N3fzwV1voSZgALJlfxs1XLyMYGLh8uWmkqCjOUfVB\nkWlMgUBkluqOROmMGgQ+sEjR1vfauH/DNkzLAeCsxVX8zarF+H0Dr4VmppJUluVrtUKRaU6BQGQW\nam7tIGV5CQT7jgV4acthHnxqJ47rAnDRh+dw3cqT8A5SQ8BMJakuL1ApYpEZQIFAZBbpXaAIXwif\n/+gVv+u6/OHlfTz+wt7Mc9d8fCEXnz130IJCZipBTUURoVBw3NstIuNPgUBklkgmUzS1dRMI9b3P\nbzsODz29ixePlCL2ej38zarFnLNk8OWLzVScuqoSAoGBBxiKyPSjQCAyC3RHYnRGU/3CQDJlcf9v\n32b73nYAQkEfN1+9jMUnDrFIkZHQIkUiM5B+o0VmMNd1aWnrJGl5+o0X6Iwm+eGjWzjUEgOgKD/I\n1687nbnVA9cQcF0X10oyp6YMr3fgAYYiMn0pEIjMUKZp0tjShTcQxu/vOw7gUEuUHz66JVNjoLYi\nj1uvO53SooFnCtiWRdBnU1VTrkWKRGYoBQKRGainJ05LZw/BUP/ywlvfa+OBx98mZdgAnHJCKTd/\nehk54YH/HFimSV7YQ3mp1iUQmckUCERmkN5bBAmTfmHAdV02vX6QXz+7G/fIcyuW1fCFy07BN0iN\nASOVpKwoR9UHRWYBBQKRGSKZTNHcHsEXCBMI9O3Wt2yHhzbu4qUjMwkAPn3hAi49d94Q0wqTVKng\nkMisoUAgMgO0d3YTjVv9qg4CxBIm63+zNbN0cTDg5UuXL+GMRZWD7s8yEtRVFWtaocgsokAgMo1Z\nlkVjS+eRhYn6Vws81BLl3369lfbuJADFBSG+du1pg84kcBwHj2NoJoHILKRAIDJNRWM9tHXFBxw4\nCPDXnc389Pc7MMz0mgTzagr5+2uWU1wwcJlhx7YJeC2qa8rHrc0iMnUpEIhMM0MNHARwHJfHX9jL\nU6/syzx37tJqbrzsFAL+gVcrtCyTvJBmEojMZgoEItNIPJGktSM64MBBgJ6EyX/8bjtv70lXHvR6\nPFx70UIu+vCcQQcPWqZBcX6AosKBbyOIyOygQCAyDTiOQ2t7FwmDAQcOAhxoivDj32zLjBfIC/v5\n6qeXccoJpYPu1zRSlBfnkJ+XO+g2IjI7KBCITHHRWA/tXXECoRwCgyws+NKWwzy0cReWnR4vMKeq\ngJuvXkZ58cDhAY5MKywvIEdLF4sICgQiU5ZhGLR2RLEcL4FBBg4aps1//vFdXtp6tL7AectruOHi\nRQQDA48XcF0X20hotUIR6UOBQGSKcV2X1vYuepI2wVCYQcYB0tTew/0b3qahNb04kd/n5XOXLOL8\n02oH3bfjOHhdg7l1FVqTQET6UCAQmUK6IzG6ogl8gTDB0OBX769sa+ShjTszUwrLisLcdPUy5lUX\nDvoeyzTJCUJluaYVikh/CgQiU0AymaKtM4qDH/8ggwYBkobFwxt38erbTZnnzji5gi9+6lRyw4MH\nCNNIUVIQ1EwCERmUAoHIJLJtm9b2bpImBIJhBrk7AMD+xgj/3xPbae6IA+D3ebhu5clceEbdkN3/\nppGksiSP3NzBg4aIiAKByCRwXZf2jm5iCYtAKDzo7AFIFxra+Jf9PP7CXhwnvU5hZUkOX7lq2aAl\niHtZRoK6Sq1JICLHN2mB4IknnuBHP/oRlmWxevVqbrzxxj6vP/PMM6xbtw6A+vp67r77bgoK1N0p\n05vrunR0RYjEjHQQCA39K9jRneQ/fred3Qe7Ms+tWFbDZz9xMuEh3us4DtgprUkgIlmblEDQ3NzM\nPffcw4YNG/D7/dxwww2ce+65LFiwAIBYLMadd97JY489RkVFBffddx/33Xcf3/72tyejuSKj5rou\n3ZEo3bFUesBgeOjue9d1+cv2Jv7zj++SSFkA5Ib93HjZKXzolKoh32tbFiG/Q1VVxZi1X0Rmvkm5\ndHjllVdYsWIFBQUF5OTkcOmll/L0009nXrcsi+9+97tUVKT/oC1atIimpqbBdicyZbmuS3tnNwcO\ntxFNgj+Yc9zpft2xFP/266385Hc7MmFg0bwS/unvzjluGDCNFIW5PqoqBq9OKCIykEnpIWhpacmc\n7AEqKirYtm1b5nFxcTErV64EIJlMsn79elavXj3h7RQZKcdx6OiMEEtY+IOhIWcO9HJdlzfeaebh\njbvoSaaDgN/n5aqPzmfl2XPxHidImKkklWX55OaEx+QziMjsMimBwHXdfs8NdJ8zGo3y9a9/ncWL\nF3PVVVdNRNNERsV1XTq7o0RiBv5g6LhjBHp1x1I8vHEXm99tzTw3r6aQv121mJryvOMe0zZVeVBE\nRmdSAkFVVRVvvPFG5nFrayuVlZV9tmltbeXLX/4y5513Ht/61rey2u/atWszAxFFJlp3JEZnJJ5e\niTCU3VW667q8tPUwv372vcztAZ/Xw+Ufmc8l587Fd5wBgaZpkBOAulpVHhSRvnp72o916623smbN\nmgG3n5RAsGLFCtatW0dnZyfhcJiNGzdy1113ZV53HIebb76ZT33qU9xyyy1Z73fNmjX9PuihQ4cG\n/KKIjJVINF1dEG+QQCj7VQObO+I8+NQ7vHvg6AyCudUFrP7UqdRXHn9GjZFKUF6cS0H+0D0IIjI7\nbdq0ifr6+qy3n7QegjvuuIPVq1djmibXX389y5Yt46abbuK2226jsbGRnTt34rouTz31FB6Ph6VL\nl/YJDSKTrTsSozuawOML4gtkX/THtBw2/mU/T72yD9NKlx4O+L1c9dEFfPzD9cftFchMKawuwe9X\nKRERGRuT9tdk1apVrFq1qs9z69evB2DJkiXs2LFjMpolcly9QQBfEF8WgwWPteP9dh7euIuWzkTm\nuVNOKOXGy06hYoilinuZhkF+jpdyTSkUkTGmywuRLLiuS1d3lEhPCq8/NOwg0BlJ8uizu3lzZ0vm\nuYLcANd+/CTOWVqd1f1/I5WgsiSPvLzsb0uIiGRLgUBkCEenD5r4g+Gspg8eyzBt/vjaAZ5+dV9m\nZUIPcMEZdVx14QLyhliQqJfrujhmkvoqlSAWkfGjQCAyAMMw6OiKkTAcglmUGP4g13X5684WHnvu\nPToiyczz86oL+Nylp3BCzeDLFB+rdxZBZW25ZhGIyLhSIBA5Rk9PnK5oAsOCYChEMDT8fbx/uJtf\nP/se7x06OnugIDfAlR9dwPnLa/F6szuxm6kk5SW55OsWgYhMAAUCmfVc16WzK0I0buDxBvD5QwSH\nWod4EC2dcTY8v6fPOAGf18PHPzyHVeedSE44u18327bxuiZzakrw+UbQEBGREVAgkFnLNE06uqLE\nUzaBEYwP6BXpMfj9S+/zwlsNmeWJAZYvLOfai06iqjT7K3wjlaQoP0hpcfmI2iIiMlIKBDLrxGI9\ndMeSo7otABBLmPzxL/t57q8HMwMGAU6sLeSajy3kpLklWe/LcRxcK0VtRRGhUHBkDRIRGQUFApkV\nbNumsytKT9JMFxLyjey2AEA8afLMawd49o2DJA0783xlSQ6fvnAhZywaXhlhI5WkMC9AmWoLiMgk\nUiCQGS0SjRHtSWFYLsFQGH9w5D/ysYTJs68f4Lm/HsqsOwBQUhjiU+edyHnLavD5sl9R3LZtsA31\nCojIlKBAIDOOYRh0RXqIJy28/tH1BkB6JcJnXjvAnzc3kDKP9ggU5Qf55IoTOP+0OgL+7IMApIsM\nlRSEKC5Sr4CITA0KBDIjOI5DdyRGLGFgOR6CwRCB0OiK+DR3xNn0+gFe3tqIZR8dI1CUH+Tis+fx\n0TPqCAaGlzRMwyAcgHm1ZQMu+S0iMlkUCGTachyHSDRGT8IkZaYLCHn9YUbT+e66LnsOdfPH1w6w\ndXcr7jGvlRaGuWzFPFYsqyHgH14QcGwbHJPK0nxyc7JbGllEZCIpEMi04rou3ZEY8aRJ0rAJhsJ4\nfCFCo5yub9kOb+5s4bm/HuT9w5E+r1WX5XLpuSdw9uKqYY0R6G2vbSYpLsihqLBodI0UERlHCgQy\n5bmuSyTaQ0/COBoCvEFCY3Ch3RVN8cJbDbzwVgORHqPPa4vmlvCJs+eyZEEZ3hGUDTZTSfJz/ZRV\nqOywiEx9CgQyJbmuSzR2JASkbPzBEN4xCgGO67JzXwcvbjnMW++29ikm5PV6+NAplXzi7LnMq85u\nvYEPMo0UOUEP1ao0KCLTiAKBTBnpMQE9JFMmCcPGH0iHgOAY3XLvjqV4eWsjL21poK072ee1wrwg\nF5xexwWn11FcMLJKRaZpEPS51FYUEAxqGqGITC8KBDKp4vEEPYkUScPCPFIrwOMJjrh64AeZls3W\n99p4dVsj2/d24Lhun9cX1BfxsTPrOWNRJf5hjg/IHMM08HscKks0YFBEpi8FAplQqVSKaE+ClGFj\nWA5eXwC/34/X72eYKwwPynFd9jZ085e3m/jrO83EjykiBJAX9nPushrOP62W2vL8ER9HQUBEZhIF\nAhlXlmURjcVJGhYpwwaPj0AwCD7/qIoFfZDruhxsjvHGO8288U4zHZG+twQ8wCknlLJiWQ1nLKoY\n9rTBY5mGgd+rICAiM4sCgYwp27aJxnpIpixSpo3tposEeTxBAmN0G6BXOgRE2fxuK2/ubKG5I95v\nm+qyXM5dWsM5S6opKRzdydtIJckJeqmoyNcYARGZcRQIZFQsyyLWEydp2BimjWWnVxD0eIL4g2P/\nA2Y7Dnsbutnybhub322h/QODAwGKC0J8+JQqzlpcxdzqglFN+bNtG9c2yAsHqKktVXVBEZmxFAhk\nWFKpFD3xJIZpkzJtHNdLIBjE4wngCwTwja5a8IASSYvt77ez7b023t7TRk/S6rdNQW6AMxZVctap\nVSyYUzyiugHHMlJJQgEvpfkh8vNVUEhEZj4FAhmU67okEkniiRQp08YwHTzeI2MAPD7849Rr7rgu\nh5pj7Hi/ne1729nT0N2nVkCvkoIQp59cwZmLKllQX4zXO7oQYFsWrmOSFw6ohoCIzDoKBJJh2zax\nnjgpw8IwHUzLwetPzwIY60GAH9TenWDX/k527u9k576OflUDe82pzGfZwnKWn1TBvFHeDoB07QPL\nTJET8lNcFCYvt3hU+xMRma4UCGapzNV/MoVpORimje14jtz/D+DxQ3Acfzo6Ikl2H+xi94FOdu3v\npLUrMeB2oYCPRfNKWDK/jGULyykd5cBASH92I5UkHPRREA5QVKnSwiIiCgSzgOu66Xv/ifTJ3zRt\nTBZRAh0AABmoSURBVNvF13v172FcBgD2chyXxrYe9h7u5r2DXew+2NVvWuCx6iryWTK/jCXzS5lf\nV0zAP/qBfI7jYBopwkEfeWE/heVaflhE5FgKBDOM4zjEE0mSKQPLSnf7m7aL1+cnEAgAPryBAKFx\nGPzXK9JjsL8xwr7GCHsbunn/cDdJwx50+4qSHE6ZV8opJ5Rw8twSCnLHZnCCbdvYlkE46CM/J0ih\negJERAalQDBNua6LYRgkkikM08a0HCzbwXbAHwji86Wv/L0Bxv3kf7A5ysHmKPubIuxvjA559Q9Q\nW5HHwvpiTppTzMI5xZQUjF1xn1Qqid+bvtVQlB8kL69QIUBEJAsKBFPc0RO/gWkdPfFbtovX68Mf\nCODxBMAHft/4fUNtx6G5I05DS4yG1h4aWqIcaI7RHUsN+b5QwMcJNYXMry9ifm0RJ9YVkZ8zdgnF\nMk0c2yIU9BEK+qgqKU7fBhERkWHRX84pwrIskikDw7QwTfvIST99xX/0xO+H/7+9e4+Oor7/P/6c\nveYeAiQRjIDa04M/a8RWkCA3uSUCCUk8tUEJIl7qkYtordBSqtYrSqgUqwhq8Vh7VBTh5y2iICIC\nP9CfX6S/cjxQieGWkJAAue/u7Pz+2GQlJCKXkM2G1+OcObOTmZ39vHc383nvZ2Y+HxvYbZyT+/0B\nTNNP2ZE6DpbXBKbDgXnJ4Vp8pv+kz7XZDC5MjKFPj1h694ijzwVx9EiMxt6G5+q9Hg+W38TltOF2\n2ekeF4Xb3cZdIIqInIeUELQTv9+Px+OhvsGLzzQxTStY6ftNC8tmw+l0NV7oZjunv/j9lsWRqgbK\nKms5VFnHoYpaSg7XUlpRQ/mR+hYjArbG6bBxYWIMFyXH0uuCWC5KjqVn92hczra7NzFwN0ADdpuF\ny2nH5bTTPS5aCYCIyDmghKANWJaFaZqNF/L58Pr8mH4L0/xhbmFgsztwOByBX/oQrPTPRXlqG3xU\nHK3n8NF6yo/UUX6kjsNH6yg7Ukf5kfqf/LXfxDAgKSGKC7pFk5IUw4WJMVyYFENil8iz7gjoxDJ7\nvR7w+3E6bIEWALeDmG7qIEhEpD0oIfgJpmni8/nweLzHNeMHKnm/38L0W/gtMAwbdoejsfJqbCK3\ng93e9s379Q0+jlQ3cKSqgcqqBiqP1QfmVfVUHK2n4lj9Sa/qb43LaSM5IYrkblEkd42mR/doenSL\nJqlrVJvc9nc8n8+H6fNit4HDbgskAC4HMV3jdf5fRCREzsujr9/vD1TyXl+gcvJb+Bsn84R54Je9\nvfGXfWNzPoAdbPbg0lkzTT9VtV6O1XioqvVQVePhaI2HYzUNHKvxcLS6gaPVHo5UNwSGET4Dbped\n7vGRJCYEpqSEKJISIklMiKJLrPus+/8/nmVZgQv+/Gazit9htxEZE0FERLyu/hcR6UDOm4TgwKFK\nfLgJdIlvYNhs2O32ls3RNjBsZ/fGeH0mtfU+auq81NR7qak7/rGX6lov1XUeqhsfV9V6qG1lwJ7T\nFeG20y0ukq5xbrrGR9A1LpJu8RF07xJJ9/gIoiOdbVoJB1pPvBiWha2x0nc47DhsBk6nnciE2MZE\nShW/iEhHF7KE4N1332XJkiX4fD4mT57MzTff3Gz9zp07mTdvHtXV1fTv35+HH374rHqWszvcON2R\nJ93Gb1mBUfw8JvUek3qPj/qGxrnHpK7BR32Dj7oTptr6H+a19V68vlM7P3+qbDaD2CgXXWLddIlx\nER/jpkuMm/gYNwlxbhJiI0iIdRPhbruP0+/3Y/p8+P0mhhGo7O02IzA1PnZGOYlwR6uZX0SkEwjJ\nkby0tJRnnnmGVatW4XA4yMvLY+DAgVx66aXBbR544AEee+wxUlNTmTt3Lm+++SZ5eXln/JqrPtuN\nK7qSBo8ZHLq3qfJvOG7eXgK95zmJiXIRE+UkNspFXLSL2CgXsVFOYqNdxEe7iYtxER3pPOvmfMuy\n8Pv9+E0Tv9+PgYXNZmAYYLcZ2Joqe5sNu93A4XDgdkXqF76IyHkiJAnB5s2bSUtLIzY2FoD09HQ+\n+ugj7r77bgAOHDhAQ0MDqampAOTk5LB48eKzSgi2/acUZ5T37At/HJthEBnhIMrtaDaPjnASFeEk\nKsJBVISD6EgnMZFOohunmEgnzjO4vcDv9wcny/KDZWFYFobNhs0WKE+gkm+cY2EzAr/obTYDp8OF\n0xm48FH9+IuIyPFCkhAcOnSIxMTE4HJiYiI7duz40fVJSUmUlpae9esagMtlx+Ww4XY5cDsDvdu5\nG+9xj3DZiXDbcTsdjY8b5y4HEe7APNIdmALb2YO/ni3Lajb5/YEK28ICC7ACpxEMm4Xl82BagefZ\nDKNxHxY2mw2bQbNlw4Cjx+qZVrAeAwMMg5fnjiGxW3zjev16bysej8mfXviCooPHiHA7mDjq59hs\nBp99vY/d+wLjMVhW4FbR+Gg3LqcdpwPqGkzKjzRgATYDrvlfSWz+f4dCHU6HF/iWt8F+GnfUtC+X\nw6D3BXEcKK/BsBn0vyyJy/p044vtBzh8rJ6aOi+19V4cdhs9u0dSfrSBeo9Jnx5x/OWOQWz4n31s\n+J/9VBytp2tcBAmxLv7Pf0rx+PwkdYnkmVnD+Xz7PlZt+I4jVQ04HTai3Ha6J0QytF8Kowb0BmDt\ntmKKSo7R54I4Rvbv1Sa36fr91jnZb3u/hnRMIUkIrFY6vjn+F+tPrT8Tf771l6SkXIhhGI0HIqvx\ncaDZ3LIal43ActPByta43FSuYK97BtgMC/BjGIGK22bYMGyBK+mbKuumuRGs+E/frY+ux+H8ob//\nO+ev538XTDibt0Na8acXvmBnUSUQqOSXrv43kW4H1bXeZhWXz4TDx1rvstlvoWTgFLVFMgBw4uHC\n47PYte9ocHn9/93P5h0lwS6/mzR4/ezaVxVc3llUycyFn2L6LSqO1WP6LfaX1+D3//Ccg4drufPJ\nT/BbFlW1P7Q4VlZBaWUdB8trg//nH2zaA8B/vjsMwOhrep91rGu3FZ+T/bb3a0jHFJKEIDk5mS+/\n/DK4XFZWRlJSUrP1ZWVlweVDhw41W/9jFi9ezLPPPtvqul49u5HSs/tZlDp0TjxwttWBVJrbX1bT\nbNlnWjR4Tb3fYc6ywOMzAy1sP+Hw0XoiIxzBJKO1HydVtR4crfTNYVnQ4DUpKjnWYl1rfzsTJ+6n\nrfbb3q8h7WPkyJEt/jZ9+nRmzJjR6vYhOZGclpbGli1bqKyspK6ujjVr1jBkyJDg+p49e+J2u/n6\n668BWL16NUOHDv3J/c6YMYNvv/222bR27dpzFkd7OfEwpsa7c+PCxOhmyw67ETgtFKLySNswDHA5\n7JxKA123+IjGU4FNz235pNgoF+5Wuug2jMbBvC4IjONxvBOXz9S52m97v4a0j7Vr17aoE38sGYAQ\nthDce++9TJ48Ga/Xy4033sgVV1zBnXfeyT333MPll1/O008/zbx586itreWyyy4jPz8/FEXtEJbP\nG8OUR9ZgEUgGls8bE+oidUqP/vZaXUPQjjrrNQQj+/cKlu348/BtoWk/bb3f9n4N6ZgMq7U2sU5k\n3759jBw5krVr15KSkhLq4oiIiJxTZ1rv6d4zERERUUIgIiIiSghEREQEJQQiIiKCEgIRERFBCYGI\niIighEBERERQQiAiIiIoIRARERGUEIiIiAhKCERERAQlBCIiIoISAhEREUEJgYiIiKCEQERERFBC\nICIiIighEBEREZQQiIiICEoIREREBCUEIiIighICERERQQmBiIiIoIRAREREUEIgIiIiKCEQERER\nlBCIiIgISghEREQEJQQiIiKCEgIRERFBCYGIiIighEBERERQQiAiIiIoIRARERGUEIiIiAhKCERE\nRIQQJQQHDx5k0qRJjB07lmnTplFXV9dim7KyMm6//Xays7PJzc1ly5YtISipiIjI+SEkCcHDDz/M\nzTffzAcffMDll1/O3//+9xbbPPXUU1x33XWsWrWKgoIC7r//fizLCkFpRUREOr92Twh8Ph/btm0j\nPT0dgNzcXAoLC1tsl56eTlZWFgC9e/fG4/FQU1PTrmUVERE5Xzja+wUrKiqIjY3FZgvkIomJiZSW\nlrbYbtSoUcHHL774IpdddhkxMTHtVk4REZHzyTlNCAoLC3niiScwDAPLsjAMg969e7fYrik5aM3y\n5ctZsWIF//znP8+oDKZpAlBSUnJGzxcREQknTfVdU/13qgyrnU/M+3w+Bg4cyLZt2zAMg5KSEvLz\n8/n4449bbPvUU0/x+eef89JLL5GUlPST+168eDHPPvvsuSi2iIhI2Js+fTozZsxodV27JwQAd911\nF5mZmYwbN44lS5ZQVlbGvHnzmm2zfPly3n//ff7xj3+c1amC+vp6rrzyStasWYPdbj/boofUyJEj\nWbt2baiLcVY6QwzQOeLoDDGA4uhIOkMMEP5xmKbJmDFj2L59OxEREaf8vHa/hgDgz3/+M7Nnz+b5\n55+nR48eLFy4EIDXX3+dsrIyZsyYwXPPPUdMTAz5+fnB0w1Lly4lMTHxtF6r6c1o7VRFOEpJSQl1\nEc5aZ4gBOkccnSEGUBwdSWeIATpHHKeTDECIEoKePXvy6quvtvh7Xl5e8PHWrVvbs0giIiLnNfVU\nKCIiIkoIREREBOwPPfTQQ6EuRHu45pprQl2ENtEZ4ugMMUDniKMzxACKoyPpDDFA54jjdGMIyV0G\nIiIi0rHolIGIiIgoIRARERElBCIiIoISAhEREUEJgYiIiHAeJASLFi1qNuBRVVUVv/3tbxk7diz5\n+fkcPnw4hKU7Ne+++y7jxo0jPT2d1157LdTFOS3V1dVkZmZy4MABADZt2kRWVhbp6eksWrQoxKU7\nNc8++yzjx48nMzOTBQsWAOEZx6JFixg7diyZmZksX74cCM84AObPn88f/vAHIDxjmDx5MpmZmeTk\n5JCTk8M333wTdnGsW7eO3Nxcxo4dy2OPPQaE32exYsUKsrOzycnJITs7m6uvvppHH32UzZs3h1Uc\nAKtXrw4ep5566ingDD4Pq5Oqqqqy/vjHP1r9+vWzFi9eHPz7X/7yF2vp0qWWZVnWqlWrrFmzZoWq\niKekpKTEGjFihHXs2DGrtrbWysrKsnbv3h3qYp2S7du3W5mZmdYvfvELa//+/VZ9fb01fPhwa//+\n/ZZpmtbUqVOtDRs2hLqYJ7Vp0yZr4sSJls/ns7xerzVlyhTrvffeC7s4tm7dak2cONHy+/1WfX29\nNWLECGvnzp1hF4dlBT6TgQMHWnPmzAnL75RlWdbgwYMt0zSDy+EWR3FxsTVkyBCrtLTU8vl81s03\n32x99tlnYRXDiXbt2mWNGTPGKikpCbs46urqrAEDBlhHjhyxTNO0fv3rX1vr1q077Tg6bQvBJ598\nQp8+fbj11lub/X39+vVkZWUBMH78eDZs2HDaY0a3p82bN5OWlkZsbCyRkZGkp6fz0UcfhbpYp2TF\nihU8+OCDwaGrv/nmG/r06UPPnj2x2WxkZWVRWFgY4lKeXFJSEnPmzMFut+NwOLjkkkvYs2dP2MXR\nv39/Xn31VQzDoLy8HL/fT1VVVdjFceTIEZ555hnuuusuIDy/U9999x2GYXD77bczYcIEXnvttbCL\n45NPPmHcuHEkJSVht9v561//SmRkZFjFcKKHHnqI3/3udxQXF4ddHKZpYlkWtbW1eDwefD4fsbGx\npx1Hp00IsrOzueOOO7DZmod46NCh4IiJdrud2NhYKioqQlHEU3J8eQESExMpKSkJYYlO3SOPPMKv\nfvUrrMa+r06MJSkpqcPHcumll5KamgpAUVERH3zwAXa7PezigMD3ffHixYwfP56BAweG5efx4IMP\nct999xEXFweE53fq2LFjDBo0iOeff57ly5fz+uuvc+DAgbCK4/vvv8fn83HbbbeRnZ3Na6+9Fpaf\nRZNNmzbR0NDAmDFjwjKO6OhoZs6cSUZGBsOGDSMlJYXS0tLTjiPsE4LCwkKGDRvG8OHDg9PUqVNP\n+fmWZbVIGjoSq5WOJDtyeU8mnGPZtWsXU6dOZc6cOVx00UUt1odLHDNmzGDLli0cPHiQoqKiFus7\nchwrVqygZ8+ezbpjDcfvVL9+/XjyySdxu90kJCRwww038Le//a3Fdh05DtM02bJlCwUFBbz55pvs\n2LGDffv2tdiuI8dwvDfeeCPYmhyO36lvv/2Wt99+m/Xr17Nx40ZsNht79uxpsd1PxRGS4Y/bUkZG\nBhkZGae8fXJyMmVlZSQnJ2OaJjU1NXTp0uUclvDsJCcn8+WXXwaXy8rKgk3w4abpvW9y6NChsIjl\nq6++4p577mHu3Llcf/31bNu2Lezi+O677/B4PPTt2xe3283o0aP56KOPcDh+OAR09Dg+/PBDysrK\n2Lx5M0ePHqW2tpYDBw5gt9uD23T0GCDwffJ4PKSlpQX/lpKSQnl5eXC5o8fRvXt30tLSgsfOkSNH\nUlhYGFbfpyZer5cvv/yS+fPnA+F5nNq4cSNpaWkkJCQAkJOTw0svvXTan0fHTnvOgWHDhrF69WoA\n3n//fa6++upmB5SOJi0tjS1btlBZWUldXR1r1qxhyJAhoS7WGbnyyispKipi7969mKbJe++9x9Ch\nQ0NdrJM6ePAg06dPp6CggOuvvx4Izzj27t3Ln/70JzweDx6Ph3Xr1pGXl8eePXvCJo6XX36Zd999\nl1WrVjFz5kxGjBjBsmXLwioGCNzp9PTTT+PxeKiuruadd97hvvvuC6s4rrvuOjZu3Eh1dTWmabJh\nwwYyMjLCKoYm3377LX369CEiIgIIz//vvn37snnzZurq6rAsi3Xr1tGvX7/T/jzCvoXgdM2cOZM5\nc+Ywfvx44uLigreRdVTJycnce++9TJ48Ga/Xy4033sgVV1wR6mKdFsMwAHC5XDzxxBNMnz4dj8fD\n8OHDSU9PD3HpTu7ll1/G4/Hw5JNPYlkWhmGQl5cXdnEMGzaMHTt2kJOTg91uJz09nbFjx9K1a9ew\niuNELpeLxx9/PKxiGD58ON988w3Z2dn4/X4mTZrElVdeGVbfqdTUVG6//XYmTpyIz+fj2muv5aab\nbuKSSy4Jmxia7N27lwsuuCC4HI7HqWuvvZadO3eSm5uL0+kkNTWV6dOnM2jQoNOKQ6MdioiIyPl3\nykBERERaUkIgIiIiSghERERECYGIiIighEBERERQQiAiIiIoIRDp9Pr27UtmZmazYV5zcnLYtWvX\nGe1v69atwU6azoXbbrutQ48vItJZnXcdE4mcbwzD4JVXXqFr165tus9z5Ysvvmi1P3kRObfUQiDS\nyVmW9aMV7NatW7nmmmuCo6DdcccdwT7d+/bty4IFC8jNzSU9PZ133nmnxfMrKiqYMWMGeXl5jBkz\nhry8PPbv3w9Afn4+BQUF5OfnM2LECO677z78fj8A77zzDr/5zW/Izc1lxIgRLF26FIDZs2cDMGXK\nFPbu3cvu3bu55ZZbyMrKYsKECbz11lsA7N+/n6FDh3LbbbeRnp7OsmXLmDVrVrBcO3fuZPTo0W3x\n9omcN9RCIHIemDJlSnCgE8uy6N69Oy+++CIDBgwgPz+fOXPmMHjwYOrq6njggQeCz4uIiGDlypUU\nFxdzww03cNVVVzXb7/vvv8/ll1/OXXfdBcADDzzAv/71L37/+98DgWFyX331Verq6hg3bhwbN26k\nf//+vPHGGzz//PN07dqV4uJiMjIymDRpEvPnz2f16tW88sorxMfHk5GRwf333096ejrl5eXceOON\nXHTRRaSkpHDo0CEWLFjAgAEDqKys5IUXXqCyspKEhARef/11brrppnZ6d0U6ByUEIueBk50yuPvu\nu8nPz2fZsmW89957zU4H3HLLLQD06tWLAQMGsGnTJn72s58F1+fn5/P111/zyiuvUFxczPbt2xk0\naFBw/XXXXQdAZGQkF198MRUVFURGRrJ06VI+/fRTiouL2b17N5ZlUV9fT1RUFBBIWoqKiqirqwv2\nv969e3fGjh3L+vXrmTRpEna7nV/+8pcAJCQkMHr0aFauXEleXh4ff/wxH374YRu+gyKdn04ZiJwH\nTnZOvqamhtLSUlwuF7t372627vjx0y3LajEyaEFBAQsXLiQ+Pp7c3FxGjRrV7LXcbnfwcVOiUVpa\nyoQJE9i3bx/9+vVj9uzZrZ7WaDq9cOLffD4fAA6Ho9nwrhMnTuStt97iww8/ZNiwYcTHx/9ozCLS\nkhICkfPc3LlzycjIYOHChcyePbvZFf4rV64EoKioiK+++orBgwc3e+7nn39Ofn4+2dnZ9OjRg88+\n+6zVivx4O3bsIC4ujmnTpjFkyBA+/fRTDMPANE0gUNH7fD4uvvhiIiMjKSwsBKC8vJzCwsJgGU5M\nIFJTU4mOjmbJkiVMnDjx7N4UkfOQThmIdHKGYbS4hsAwDO68804qKyv5/vvvKSgowOl0Mm7cOGbP\nns2yZcsA+Pe//01ubi5er5fHH3+cCy+8MHjRIMC0adNYsGABS5YswWazcdVVV1FUVBR83RPLATB4\n8GBWrlxJeno60dHR/PznP6dXr158//33JCUlMXr0aPLz81m0aBHPPfccjzzyCM899xymaTJ16lSG\nDRvG/v37W73TITc3l7fffpvU1NRz8VaKdGoa/lhEWtW3b182bdrUprcrnks+n49Zs2YxatQosrOz\nQ10ckbCjUwYi0irDMMKmP4D//ve/pKWl4XK5yMrKCnVxRMKSWghERERELQQiIiKihEBERERQQiAi\nIiIoIRARERGUEIiIiAhKCERERAT4/5BEkoBgelPtAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x117735128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_context(\"notebook\", font_scale=1.1)\n",
    "sns.set_style(\"ticks\")\n",
    "\n",
    "sns.regplot('X','Y', data=df, logistic=True)\n",
    "plt.ylabel('Probability')\n",
    "plt.xlabel('Explanatory')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficient: \n",
      " [[ 0.07861888]]\n",
      "Intercept: \n",
      " [-3.17496229]\n",
      "R^2 Value: \n",
      " 0.670143415906\n"
     ]
    }
   ],
   "source": [
    "logistic = LogisticRegression()\n",
    "X = (np.asarray(df.X)).reshape(-1, 1)\n",
    "Y = (np.asarray(df.Y)).ravel()\n",
    "\n",
    "logistic.fit(X, Y)\n",
    "logistic.score(X, Y)\n",
    "\n",
    "print('Coefficient: \\n', logistic.coef_)\n",
    "print('Intercept: \\n', logistic.intercept_)\n",
    "print('R^2 Value: \\n', logistic.score(X, Y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decision Trees"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Decision trees are a form of supervised learning that can be used for both classification and regression purposes. With this being said, in my experience they are typically utilized for classification purposes. The model takes in an instance and goes down the tree, testing significant features against a determined conditional statement. Depending on the result, it will go down to the left or right child branch and onward after that. Typically the most significant feature in the process will fall closer to the root of the tree. Decision trees are becoming increasingly popular and can serve as a strong learning algorithm for any data scientist to have in their repatoire - especially when coupled with techniques like random forests, boosting, and bagging."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.4</td>\n",
       "      <td>3.9</td>\n",
       "      <td>1.7</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    X1   X2   X3   X4            Y\n",
       "0  4.9  3.0  1.4  0.2  Iris-setosa\n",
       "1  4.7  3.2  1.3  0.2  Iris-setosa\n",
       "2  4.6  3.1  1.5  0.2  Iris-setosa\n",
       "3  5.0  3.6  1.4  0.2  Iris-setosa\n",
       "4  5.4  3.9  1.7  0.4  Iris-setosa"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import tree\n",
    "df = pd.read_csv('iris_df.csv')\n",
    "df.columns = ['X1', 'X2', 'X3', 'X4', 'Y']\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: \n",
      " 0.933333333333\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "decision = tree.DecisionTreeClassifier(criterion='gini')\n",
    "\n",
    "X = df.values[:, 0:4]\n",
    "Y = df.values[:, 4]\n",
    "trainX, testX, trainY, testY = train_test_split( X, Y, test_size = 0.3)\n",
    "\n",
    "decision.fit(trainX, trainY)\n",
    "print('Accuracy: \\n', decision.score(testX, testY))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApsAAAJjCAYAAABDU9oQAAAAAXNSR0IArs4c6QAAQABJREFUeAHs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\ni22VoS26FVQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.externals.six import StringIO \n",
    "from IPython.display import Image\n",
    "import pydotplus as pydot\n",
    "\n",
    "dot_data = StringIO()\n",
    "tree.export_graphviz(decision, out_file=dot_data)\n",
    "graph = pydot.graph_from_dot_data(dot_data.getvalue())\n",
    "\n",
    "Image(graph.create_png())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Support Vector Machines"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Support vector machines are a popular supervised classification algorithm that towards creating a dividing line between the different categories of data. The way this vector is calculated, in simple terms, is by optimizing the line so that the closest point in each of the groups will be farthest away from each other. This vector is by default and often visualized as being linear, this doesn't have to always be the case. The vector can take a nonlinear form as well is the kernel type is changed from the default type of 'gaussian' or linear. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Geting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.7</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    X1   X2            Y\n",
       "0  1.4  0.2  Iris-setosa\n",
       "1  1.3  0.2  Iris-setosa\n",
       "2  1.5  0.2  Iris-setosa\n",
       "3  1.4  0.2  Iris-setosa\n",
       "4  1.7  0.4  Iris-setosa"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import svm\n",
    "df = pd.read_csv('iris_df.csv')\n",
    "df.columns = ['X4', 'X3', 'X1', 'X2', 'Y']\n",
    "df = df.drop(['X4', 'X3'], 1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: \n",
      " 0.977777777778\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "support = svm.SVC()\n",
    "\n",
    "X = df.values[:, 0:2]\n",
    "Y = df.values[:, 2]\n",
    "trainX, testX, trainY, testY = train_test_split( X, Y, test_size = 0.3)\n",
    "\n",
    "support.fit(trainX, trainY)\n",
    "print('Accuracy: \\n', support.score(testX, testY))\n",
    "\n",
    "pred = support.predict(testX)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x119edb4e0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbgAAAFaCAYAAABsRkg6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XtclHX6P/7XzHAQEQ+1QIuKB36tB0wrZZFUVLB0RZTx\nrKWltWYFqBWgedbtoHgq0I/m6q9do1VIwSxPBZrrglJmiop4ROUgYAqIIMMw9/cPc5rhOMDM3DM3\nr+fjweMx97zvwzVYXHO/7/u6L5kgCAKIiIgkRi52AERERKbABEdERJLEBEdERJLEBEdERJLEBEdE\nRJLEBEdERJIkWoL79NNPMXLkSAQGBuKLL76oNp6eno7x48djxIgRWLx4MTQajfmDJCIiqyVKgvvp\np59w8uRJfPfdd/j666+xY8cOZGZm6q0THh6OJUuW4ODBg9BoNIiNjRUjVCIislKiJDgvLy/s2LED\nMpkMd+7cgUajgYODg3Y8JycH5eXl6N27NwBAqVTiwIEDYoRKRERWSrQpSoVCgaioKIwaNQr9+/eH\nq6urdiw/Px/Ozs7aZRcXF+Tl5YkRJhERWSlRbzIJCQnBiRMnkJubi7i4OO37NT09TC5vXKhqtRpZ\nWVlQq9WNjpOIiKyPjRgHvXbtGlQqFbp37w57e3u8+OKLyMjI0I67urqioKBAu5yfnw8XF5d69xsV\nFYXo6OgaxxITE9GhQ4emB09ERFZBlDO4W7duYdGiRVCpVFCpVEhKSsLzzz+vHXdzc4O9vT1Onz4N\nANi7dy98fX3r3W9ISAgyMjL0fhITE032OYiIyHKJkuAGDx6MIUOGQKlUYvz48Xj++ecxcuRIzJo1\nC+fPnwcAREZG4qOPPkJAQADKysowbdo0MUIlIiIrJZN6u5ysrCz4+/tzipKIqJnhk0yIiEiSmOCI\niEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiS\nmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCI\niEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiSmOCIiEiS\nmOCIiEiSmOCIiEiSmOCIiEiSbMQ6cHR0NA4ePAiZTIbBgwfj/fff1xvfuHEjdu/ejTZt2gAAJkyY\ngKlTp4oRKhERWSFRElxKSgqSk5Oxd+9eCIKAN954Az/88AOGDRumXSctLQ3r169Hnz59xAiRiIis\nnCgJzsXFBfPnz4dCoQAAeHh4IDc3V2+dc+fOYevWrbh58ya8vLwQEREBOzs7McIlslqCRoP8pCN4\nkHkTjp3d4eI3FDJ59SsTVddzHjIYBUd/rHc7IksmSoLz8PDQvs7MzMSBAwewc+dO7XulpaXw9PRE\nREQE2rdvj4iICGzatAlz584VI1wiq5WfdAS5+w8BAIovpAMAXIf517te8YV0PMi8We92RJZM1K9k\nly9fxsyZMxEREQF3d3ft+y1btsSWLVvQsWNHyOVyzJw5E8eOHat3f1FRUejWrZvej78//6ek5utx\nkqptudb1rmUatB2RJRMtwZ06dQozZsxAWFgYxowZozeWm5uL3bt3a5cFQYCNTf0nmyEhIcjIyND7\nSUxMNHrsRNbCsbN7ncu1rte1s0HbEVkyUaYoc3NzERwcjA0bNsDb27vauL29PdasWYP+/fvDzc0N\nMTExejegEJFhXPyGAoDetTRD1qvpGhyRtRElwW3fvh0qlQqffPIJBEGATCbDpEmTcOTIEYSGhsLT\n0xMrVqzA7NmzUVFRgb59+2LmzJlihEpk1WRyuUHXzmpaj9fcyNrJBEEQxA7ClLKysuDv74/ExER0\n6NBB7HCIiMhMeN8vERFJkmhPMiGi6gytW6uLRq3G1U2b8eBaJhy7dobH27MhN+AmLSKp4X/1RBbE\n0Lq1ulzdtBl3jicDAMpycgAAT4cGGzFKIuvAKUoiC2Jo3Vqd+6haw1Zlmai5YIIjsiCG1q3VuY+q\nNWxVlomaC05RElkQQ+vW6uLx9uxH+9C5BkfUHDHBEVkQQ+vW6iK3seE1NyJwipKIiCSKCY6IiCSJ\nU5REFkS3Dq5lp46AAJTevFXt+ZBVxwzt86a7XmPHTPmZ2XuOjIkJjsiC6NbB/XbiJGQywMapdbUe\nbVXHAMP6vOmu19gxYzPnsah54dckIguiW/emUamgKVf9MaZTz1ZtzNA+bzrLjR0zNnMei5oXJjgi\nC6Jb9ya3s4Pc3u6PMZ16tmpjhvZ501lu7JixmfNY1LxwipLIgujWwTX0Glx9+6u6XmPHjM2cx6Lm\nhe1yiIhIkjhFSUREksQER0REksQER0REksSbTIgkxlIKuA2NichUmOCIJMZSCrgNjYnIVPgVikhi\nLKWA29CYiEyFCY5IYiylgNvQmIhMhVOURBJjKQXchsZEZCpMcEQSU1fTVGM0VG0MsY5LzRunKImI\nSJKY4IiISJI4RUlkhfQao7p3BGRA6Y0aHspcZcycNXGsfSOxMcERWSHdurK7J09CEADb1tUbo1Yd\nA8xXE8faNxIbExyRFdJrjFqugm5LkAfXMoHfz5SqjZmxJo61byQ2zhcQWSG9xqj2dpDb1dIYteqY\nGWviWPtGYuMZHJEV0muM2sBrcDXtwxS1aax9I7Gx4SkREUmSaFOU0dHRGDVqFAIDA7FmzZpq4+np\n6Rg/fjxGjBiBxYsXQ6PRiBAlERFZK1ESXEpKCpKTk7F3717Ex8fj3Llz+OGHH/TWCQ8Px5IlS3Dw\n4EFoNBrExsaKESoREVkpUa7Bubi4YP78+VAoFAAADw8P5ObmasdzcnJQXl6O3r17AwCUSiWioqIw\nefJkMcIlMiqDa9g6dQQEoPRm9Ro2S2ForRtr4kgMoiQ4Dw8P7evMzEwcOHAAO3fu1L6Xn58PZ2dn\n7bKLiwvy8vLMGiORqRhaw/bbiZOQyQAbp+o1bJbC0Fo31sSRGES9i/Ly5ct48803ERERAXf3P24h\nrum+F7kB3/aioqIQHR1t1BiJjM3gGjaVCjIAcKq+naUwtNaNNXEkBtHmCE6dOoUZM2YgLCwMY8aM\n0RtzdXVFQUGBdjk/Px8uLi717jMkJAQZGRl6P4mJiUaPnagpDK5hs7OD3L7mGjZLYWitG2viSAyi\nnMHl5uYiODgYGzZsgLe3d7VxNzc32Nvb4/Tp03juueewd+9e+Pr6ihApkfEZXMNWwzU4S2NorRtr\n4kgMotTBffjhh9izZw/c3d0hCAJkMhkmTZqEI0eOIDQ0FJ6enrh48SIWL16M0tJS9OjRAx9//DFs\nbW0bfCzWwRERNU8s9CYiIknifbpERCRJfBYlkYiq1of9yXcQrm3+HA+uZcKxa2d4vD0bchubGtc1\nZS0Z69ZICpjgiERUtT4s74ckPLh2HQBQlpMDAHg6NLjGdQHT1ZKxbo2kgF/JiERUtR6sLDtHf/xa\nZq3rmrKWjHVrJAVMcEQiqloP5tDeTX9cpy7OnLVkrFsjKeAUJZGIqtaH1XQNrrZ1TVlLxro1kgKW\nCRARkSRxipKIiCSJCY6IiCSJCY6IiCSJN5kQmYG6QoX/rf4H1DezYdPBDa6tnPHwxk2jFHNr1Gpc\n3bRZe2NK19mzcOfYf2vcR237b8hxWQRO1oIJjsgM/rf6H8Av6bABIM8rxB3IoLCzM0ox99VNm3Hn\neDKAR8XhZTm50KgqatxHbftvyHFZBE7Wgl+7iMxAfTNb+1omANC5ebmpxdy62wM1FIvr7KO2/Tfk\nuCwCJ2vBBEdkBjbu7bWvBRkAmUy73NRibt3tgRqKxXX2Udv+G3JcFoGTteAUJZEZDAhfVOc1uMca\nU2D9ePu6rsHVt/+GHJdF4GQtWOhNRESSxClKIiKSJCY4IiKSJF6DI2qCptaEaQQNjl4/gZtF2XBv\n0x5DuvSHXMbvnUTGwARH1ARNrQk7ev0EDl/5EQBwseAKAMCv6wtGjpKoeeJXRaImaGpN2M2i7DqX\niajxmOCImqCpNWHubdrXuUxEjccpSqImaGpN2JAu/QFA7xocERkHExxRE8jk8iY9h1Euk/OaG5GJ\ncIqSiIgkiQmOiIgkiVOURE1QWx1bY+vbTF0Xx15u1JwwwRE1QW11bI2tbzN1XRx7uVFzwq9uRE1Q\nWx1bY+vbTF0Xx15u1JwwwRE1QW11bI2tbzN1XRx7uVFzwilKoiaorY6tsfVtpq6LYy83ak7YD46I\niCRJ1CnKkpISBAYGIicnp9rYxo0b4efnB6VSCaVSia+++kqECImIyFqJNkV59uxZLFq0CJmZmTWO\np6WlYf369ejTp495AyMiIkkQ7QwuLi4OS5cuhYuLS43j586dw9atWzF69GisXLkSKpXKzBESEZE1\nE+0MbuXKlQCAmi4BlpaWwtPTExEREWjfvj0iIiKwadMmzJ0719xhUjPUkGLr0opShH67FCUVpWhl\n2xITegYgt7QAHVu7ARBwqzi32j7UGjU+/+krZBZmoXOb9ggqdsPDm1lw7OwO5yGDUXD0RxZiExmB\nRd5F2bJlS2zZskW7PHPmTCxcuJAJjsyiIcXWod8uRbGqBABQrCrB//9rLDq364jUrF8hA+Bk36ra\nPj7/6Ssk3zoFAGh9+hrOX9fA2fFPKL6QjuIL6draNBZiEzWNRSa43NxcJCcnY9y4cQAeneXZ2NQf\nalRUFKKjo00dHklcQ4qtSypK9ZY1eDQjoaqsAAA41bCPzMIs7esn7lVAVanRLj+4lgnonLGxEJuo\n8Sxy7sPe3h5r1qxBdnY2BEFATEwMhg0bVu92ISEhyMjI0PtJTEw0Q8QkJQ0ptm5l21JvWQ4ZAMBO\nYQt7hW2N++jc9o9ylbvtbGGns55j1856+2MhNlHjiX4GJ5PJtK9nzZqFOXPmwNPTEytWrMDs2bNR\nUVGBvn37YubMmSJGSc1JQ4qtPxu13OBrcI/N8poK4NGZ3BND2sPz+bqvwRFR47DQm4iIJMkipyiJ\niIiaigmOiIgkSfRrcETmYoxmono1bG07YJbXVNjIbartv0Prp5BRcA03irLRuW0HvNFvMo7f+Nlk\njUx1sakp0SNMcNRsGKOZqG4NW879PADA297Tq+3/WOZJlFU8hEKuQM79POTez0eFpqJJxzYUm5oS\nPcKvddRsGKOZqG4NW9Vl3f2pKlUQ8Mf9W7kleU0+tqHY1JToESY4ajaM0UxUt4at6rLu/uwUdpDh\njxKYP7dybfKxDcWmpkSPcIqSmg1jNBPVrWF7fA2upv0P8xhY7zU4U2FTU6rJkiVLYGdnh0WLFmnf\ne+2116BUKjFmzBgRIzMd1sERETUDZWVlGDt2LFasWAEvLy/8+9//xvnz57Fq1SqxQzMZTlESETUD\nDg4OWLVqFZYuXYorV65gz549WLp0qdhhmRSnKImImonevXtj5MiRmDZtGrZt24aWLVvWv5EVY4Ij\nq9fY+jZVpQorj3yG3JI8uDo6489OzrhZlPv7DSACbhbloHPbDpjZdyK2n4pFZmEWOrVtj25PdkXW\n/dtwb9Mevp3/imOZqbhZlF1nDzgiS6FUKnH06FH07NlT7FBMrs4EV1JSgrt378LdXf8urIsXL6J7\n9+4mDYzIUI2tb1t55DNc+u0aAKC4vARX7mbCVmH7+y38Amzktsi5n4eLBVdw92ERACCrOAenctLg\n4vgkLhZcwcWCK9pb/uvqAUdkSSR+64VWrV8vjx8/jiFDhmDs2LEYO3Ys8vL+qONZsGCBWYIjMkRj\n69uq1qY9rlsTBAEanT8Adx8Wal9rBAGqSpV2WbcOTlVZgfLf+8A1JA4ic9Pt4iJltSa4devWISYm\nBidPnsTAgQMxffp0FBcXA2g+2Z+sQ2Pr26rWpj2uW5PJZJDr/AF4okVb7Wu5TAY7hZ12WbcOrq4e\ncESWon379vj666/FDsMsap2iFAQB3bp1AwC8++67KC0txZw5c7Bt2zazBUdkiMbWty0eGmrya3BE\nJJ5a6+DGjRuH1atXw8PDAwBQWVmJGTNmoFOnTjhz5gy++eYbswbaWKyDIyJqnmqdopwzZw6mTp2K\n/fv3AwAUCgWioqJw7tw5XL582WwBEhERNUatU5R3797F999/D7VarX2vTZs22LlzJ0aOHGmW4IiI\niBqr1gS3bds2ZGRkIDw8XPteVlYWQkND4erqWttmRBarar2c7vUz3bq1uurqDK25M0bvOSJqmloT\n3H/+8x+EhYXhjTfewPr163H69GmEh4cjKCgIYWFh5oyRyCiq1svp1rDp1q3VVVdnaM2dMXrPEVHT\n1PqVslWrVti0aROeffZZBAQEIDw8HCtXrsSCBQtgY8MHoJD1qVqXVrW32+PxuurqDK25M0bvOaLa\n1PagjZ07d2LXrl0mOWZWVhYWLlxokn2bSp1zJkVFRThz5gxat24NhUKB0tJSc8VFZHRV69Kq9nZ7\nPF5XXZ2hNXfG6D1HVJvaCrUnT56MSZMmmeSY2dnZuHXrlkn2bSq1noqdPXsWc+bMgY+PDzZu3Ijr\n168jODgYv/76KxYuXAhbW9vaNiWySFXr5Wq6BlfTerr1bIbW3Bmj9xxZt+yCEhw7nQ17WzmG/bUT\nWjva1b9RA6WmpiIyMhIajQZ/+ctf0L79oy9Ss2fPxgcffIArVx5Nj0+ZMgUTJkyotv2qVauQkpIC\nuVwOPz8/BAcHo7S0FCtWrMDly5eh0Wjw97//HSNHjsSHH36IrKwsrFy5EosXL8bmzZuxb98+KBQK\nDBgwAOHh4Xjw4AHee+893LlzBwAQHByMoUOHIjU1FRs2bMDDhw9RXFyMsLAwDB8+3Oi/j2qEWvTt\n21eIi4vTe+/evXvCa6+9JowfP762zSzOrVu3hL/85S/CrVu3xA6FiJqJe8UPhfkb/yvMXXdEmLvu\niPDxv1KFykqN0fbfvXt3QRAE4eTJk4KXl5dQUlIiCIIgREVFCVFRUUJqaqowa9asR7HcuyfMnz+/\n2j6ys7OFgIAAQRAEoby8XAgLCxPKy8uFNWvWCDt27BAEQRDu378vjBo1Srh165Zw8uRJYdq0aYIg\nCMLRo0eFSZMmCeXl5UJlZaXw1ltvCV9++aUQHx8vrFixQhAEQbhy5YqwevVqQRAEITQ0VLh27Zog\nCIKQkpIijBo1ymi/i7rUega3Y8cO9OjRQ++9tm3bYtu2bYiMjDR54iUislZZ+ffxsPyPEqu83x7g\nfqkKbVrZG/1YXbp0gaOjo957Tz/9NDIzM/H6669j8ODBNd4Y6OrqihYtWmDKlCkYOnQo5s6dCzs7\nOyQnJ6O8vFz7OK+ysjJcuXJFr7XOiRMnEBAQADu7R2el48aNw969e/H+++9j3bp1uH37NoYMGYK3\n334bABAZGYkjR47gwIEDOHPmDMrKyoz+e6hJrdfgqiY37QZyOSIiIkwWEBGRtXN5oiUUij/+vLZp\nZY9WDqa5rGNvXz1ptm3bFvv27cP06dNx/fp1BAUF4f79+wgKCkJQUBCUSiUUCgViY2Mxd+5cFBYW\nYuLEicjMzIRGo0FkZCQSEhKQkJCAXbt2YdCgQXr7F6o8AEsQBKjVari7u+PgwYMYPXo0fv75Z4wf\nPx7AoynStLQ09OrVC7Nnzzbb84xZmENEZGQu7VpiZqAnPDq0Rc8uT2L22N56Ca+p6ksQSUlJCAsL\nw+DBg7Fw4UI4Ojri9u3b2qQVHx+P9PR0vPLKK/Dy8kJ4eLj2rK9///746quvAAD5+fkYPXo0cnJy\noFAoUFlZCQDo378/vvvuO5SXl0OtVmPPnj3w9vZGTEwMPvvsMwwfPhxLlizB3bt3UVRUhJs3byI0\nNBS+vr44fvw4NBqN0X4XdeH9/mR25iyC1j1Wx9Z/BiDDreIcFl+TyfXs8iR6dnnSJPuur93N4MGD\ncejQIQQEBMDe3h4vvfQSnn76ab11evTogeeeew4BAQFwcHBAz5494evri379+mH58uUIDAyERqNB\neHg4OnbsCCcnJxQXFyMiIgKrVq1Ceno6xo0bh8rKSgwcOBDTpk1DaWkp3nvvPQQGBsLW1hahoaFo\n06YNxo8fj4CAADg5OeHZZ59FWVkZHj58iBYtWpjk9/NYrQ9blgo+bNnyJF1L1hZBA8BL/99gkxVB\n6x7rfnkJBACt7VuZ/LhEJD6ewZHZmbMIWnffus1ITX1cIhIf52fI7MxZBK27b3uFLezYkJSo2eAZ\nHJmdOYugdY9V0zU4IpIuURNcSUkJpkyZgi1btsDNzU1vLD09HYsXL0ZJSQm8vLywfPlyyOU84ZQC\nuUxutmtf5jwWEVkW0TLG2bNnMXXqVGRmZtY4Hh4ejiVLluDgwYPQaDSIjY01b4BERGTVREtwcXFx\nWLp0KVxcXKqN5eTkoLy8HL179wYAKJVKHDhwwNwhEhGRFRNtinLlypUAai5YzM/Ph7Ozs3bZxcUF\neXl5ZouNLINao8bnP32FzMIsdG7bAbO8psJGblPnWENq7PRr5NwACLhVnMsaOSKJsMibTGpKerz+\n1vx8/tNXSL51CgCQc//RF5y3vafXOdaQRqO666Zm/QoZACf7VmxQShave/fuuHjxYrX3d+7cCZlM\nZrKWOfXJz8/H4sWLsWXLlgZvW9tnagqLTHCurq4oKCjQLufn59c4lVlVVFQUoqOjTRkamVHVhqS6\ny7WNNaTGTndM9XuNnJMB2xEZqvBhMWzlNnC0a1n/yg1QVz84Mbm4uDQquQH1P52lMSwywbm5ucHe\n3h6nT5/Gc889h71798LX17fe7UJCQhASEqL33uMnmZD16dy2g/bs7PFyfWPubdprz8AeL9dGd107\nhS1kVcaIGksQBMSd/w6/5KRBJpNhVDd/DHD3MvpxmtIPLikpCbGxsdi8eTMAICYmBpmZmViwYAFW\nr16N1NRUaDQaKJVKvPrqq9WONXr0aERGRkIul6NNmzZYu3YtHjx4gGnTpiEpKQk5OTlYsGAB7t69\nCwcHB6xcuRLdunXD7t278cUXX0Amk8HT0xNLliyBg4ODNq6HDx9i0aJFyMjIgFwux4wZMxAUFIT4\n+HjEx8ejsLAQQ4cOxbx58+r9/Yie4HSz9qxZszBnzhx4enoiMjISixcvRmlpKXr06IFp06aJGCWJ\nYZbXVADQu85W31hDauz0a+SqX4Mjaqzr927il5w0AI+S3bcZiejr1hstbIzfLufGjRs4cuQIHB0d\ntTNYp0+fRlFREfbs2YPCwkKsWrWqWoLz9fXFsmXLcP/+fTg5OeHbb7/FBx98gNjYWMhkMuzZswcq\nlQqvv/46evXqVe1Y06dPx4oVK9CrVy98+eWXuHDhAjp16qT9m758+XKMGDECU6ZMwbFjx7B582a8\n9dZb2LJlC77++mu0bt0aK1asQHR0tF47n88++wzt2rXDvn37cO/ePUyYMAE9e/YEAOTl5eHgwYMG\nn+2JnuASExO1rz///HPt6+7duyMuLk6MkMhC2MhttNfcDB1rSN0ba+TIVNSaSr1lQRCgEUzzBP3G\n9oOzsbHBSy+9hEOHDuGFF15AUVERnnnmGWzduhUZGRlISUkB8Kgf3KVLl+Dh4aF3LH9/f7zzzjsY\nNmwY/P398cILLyA7+4+p/dTUVKxbtw7Ao2Tq6+uLmJgY+Pn5oXXr1gCAiRMn4oMPPtCL6+TJk/jo\no48AAO3atcOwYcOQmpoKR0dHeHp6NmgqU/QER0QkNR5PdELXJzrh2t0bAICBnbzQ0tahnq0ap65+\ncCkpKTh69CiCgoLw3XffaWfCZDIZ4uPjERgYiE8//RRFRUUYNWoUAECj0SAsLAzDhg0DANy7dw+O\njo749ddf9Y716quvws/PD0eOHEFkZCRGjBih3QcA2Nrq97+7evVqjW1yHrfgeazqTYYajQZqtbrW\nz1oX3ppIRGRkCrkCbzw/GbP6vYxg79cwqtswo+7fGP3gAKBPnz7Iz8/HN998g9GjRwN41Ott165d\nUKvVePDgAaZOnYozZ85UO8bEiRNRUlKC6dOn49VXX8X58+f1xr28vLB//34AwP/+9z8sWbIE3t7e\nSEpKQnFxMQAgNjYW/fv31/tM3t7e2m7id+/eRWJiIry9vRv1e+IZHBGRCcjlcnR9wt0k+zZGP7jH\n/va3v+H48ePadmKTJ0/GjRs3oFQqUVlZifHjx8PLywupqal627377ruYP38+FAoFHBwcsHz5cr3x\nxYsXY+HChYiJiYGDgwM+/PBDdO3aFbNmzcLLL7+MyspKeHp6ard7/JneeecdbT86QRDw1ltvoUeP\nHo0qIWA/OCIikiROURIRkSQxwRERkSQxwRERkSQxwRERkSQxwRERkSQxwRERkSQxwRERWZnu3bvX\n+P7OnTuxa9euJu07KSkJUVFRDdpGqVQafZ/GwDo4IiIr06NHD6Snp4sdhsXjk0yIiEygLDsHBf89\nDrmdHVyH+cO2tVP9GzWQsdvlXL9+HZ6enkhNTcXHH38MPz8/9OnTBxcvXkRMTAy+/fZbxMTEoHXr\n1ujSpQvc3d0RHBysbVYaHR2NvLw8ZGZmIjc3FxMmTMCbb76J+Ph47T6Tk5OxatUqCIIANzc3rF27\nFoIgYOHChcjLy0N+fj68vLywatWqJv9+OEVJRGRkqsIiXNm0GXdTf8ad48m4unkLhBoeNGwMN27c\nwL///W98/PHH2vd02+Vs374dv/zyS7XtfH19ceHCBdy/fx8A8O2332LMmDHV1hs8eDAOHDiAgoIC\n/Oc//0F8fDxiYmJw48YN7Tq6jw67dOkSvvjiC8TGxmLLli0oKSnRjqlUKoSFhWH16tX45ptv0K1b\nNyQkJODHH39Ez549sXPnThw6dAinT5/GhQsXmvy74RkcEZGRlWVnobLsoXb54e18qEtKYPt7mxhj\nMna7nMdnfY/17t0bAJCSkoIhQ4agZctH3ckDAgK0D03W5e3tDYVCgSeeeAJt27bVJlDgUfJzdXVF\nt27dAECvaenZs2fxr3/9C1evXkVRURFKS0sb+RvR+YxN3gMREelp4eICuY0CGvWjVjB2bVvDpkoS\nMhZjt8v5rtUBAAASKklEQVSp9llatADw6OHRhtyyYWdnp30tk8n0trGx0U85JSUlePDgAQ4fPozD\nhw9j8uTJGDBgAC5fvmzQserDBEdEZGT2zs7oPONV5B85CrmdHdxGjYRMoTDa/g1pl7N37158+umn\nGDBgAE6cOKFtl6NLt13Oxo0b69ynj48PQkNDERISAltbWxw+fBg+Pj51xlP1/a5du+LevXu4evUq\nPDw8sHXrVshkMly6dAmTJ09GQEAALl++jIsXL1brE9cYTHBERCbQukd3tO5R8+38TWXKdjm1Hefp\np5/GK6+8gsmTJ6Nly5Zo166d9uyutniqvm9nZ4fVq1cjPDwcarUa7u7uWL16Nc6cOYNly5Zh27Zt\ncHR0xPPPP4+srKw6P6MhWCZARET1yszMxNGjR/Haa68BAN5++21MnDgRQ4YMETWuuvAMjoiI6uXm\n5oa0tDQEBgZCJpNh4MCBFp3cACY4IiIygJ2dHdauXSt2GA3COjgiIpIkJjgiIpIkJjgiIpIkJjgi\nIpIkJjgiIpIkJjgiIpIklglYIY1GQOJPN5F5uxidn2oNfy93yOV1P9mAiKi5YYKzQok/3cT+5OsA\ngAvXfgMAvOjdScyQiIgsDqcorVDm7eI6l4mIiAnOKnV+qnWdy0RExClKq+Tv5Q4AetfgiIhIHxOc\nFZLLZbzmRkRUD9GmKPft24eAgAAMHz4cMTEx1cY3btwIPz8/KJVKKJVKfPXVVyJESURE1kqUM7i8\nvDxs2LABCQkJsLGxweTJk9G/f394eHho10lLS8P69evRp08fMUIkIiIrJ8oZXEpKCnx8fODk5AQH\nBwcMHz4chw4d0lvn3Llz2Lp1K0aPHo2VK1dCpVKJEapV0GgEfH/yBrbuTcP3J29Ao5F0D1siIoOI\nkuDy8/Ph7OysXXZ2dsbt27e1y6WlpfD09ERERAQSEhJQXFyMTZs2iRGqVXhcF3fh2m/Yn3wdiT/d\nFDskIiLRiZLgBKH6GYZc/kcoLVu2xJYtW9CxY0fI5XLMnDkTx44dq3e/UVFR6Natm96Pv7+/UWO3\nRKyLIyKqTpQE5+rqioKCAu1yQUEBXFxctMu5ubnYvXu3dlkQBNjY1H+5MCQkBBkZGXo/iYmJxg3e\nArEujoioOlESnI+PD06cOIF79+6hrKwMhw8fxqBBg7Tj9vb2WLNmDbKzsyEIAmJiYjBs2DAxQrUK\n/l7uGPlCF/Ts+iRGvtCFdXFERBDpLkpXV1fMmzcP06dPR0VFBSZOnIhnnnkGs2bNwpw5c+Dp6YkV\nK1Zg9uzZqKioQN++fTFz5kwxQrUKrIsjIqpOJtR0QUxCsrKy4O/vj8TERHTo0EHscIiIyEz4LEoi\nIpIkPqrLgtTV502lqsSiLf9DdsEDuP3JEX79OiKroASdn2qNoX074sipW9W2Y984ImrOmOAsSF19\n3hZt+R8u3rgHALh4Q4XM3GJ0cGmFC9d+w/lrv+HG76UButuxbxwRNWecorQgddWzZRc80BtTqSu1\nr6/nFtW4HevjiKg5Y4KzIHXVs7V3dtQbs7NRaF93+XObGrdjfRwRNWecorQgdfV5+8ebAxp0Da6+\n/RERSR3LBIiISJI4RUlERJLEBEdERJLEBEdERJLEm0wsiFqtQXTcr7ieWwR3VyfcvvMAOb+Vor2z\nI1b8/QX890x2jUXbLOgmIqqOCc6CRMf9iuNncgAA13OKIQiATPaosDt03RE4OtgCqF60zYJuIqLq\nOEVpQXQLtqve2/pb0UO9Zd2ibRZ0ExFVxwRnQXQLtmVVZhifbNNCb1m3aJsF3URE1XGK0oIET3gW\nAAy+BvcYC7qJiKpjoTcREUkSpyiJiEiSmOCIiEiSeA3ODOqqU9OrfXNphYs3C3G36CHatrLF3eJy\nqDWArUKG6PChmLv2R5SpKmFvI4OdrRwPHlbCqaUdNof7I+V8LhueEhHpYIIzg7rq1HRr365l/3F7\nf35hufZ1RaWANz9O0i6XqwWU/94PruiBCjM/PIz2Lq2q7Z/1cUTUnHGK0gzqqlOr2qy0McpUlXrL\nbHhKRMQEZxZ11alVbVbaGA52Cr1lNjwlIuIUpVnUVaemV/tmxGtw9R2XiEjqWAdHRESSxClKIiKS\nJCY4IiKSJF6DMzNDa9OqrjfgGTfM3XAUvxU9xJNtWuCzd4eiRQv+8xER1YZ/Ic3M0Nq0quv967sL\nKHqgAgDk/laK0HVH8PkHL5opaiIi68MpSjMztDat6vv3S1V6y1X7wxERkT4mODMztDat6vtOLe30\nlqv2hyMiIn2cojQzQ2vTqq5X0zU4IiKqHevgiIhIkkSboty3bx8CAgIwfPhwxMTEVBtPT0/H+PHj\nMWLECCxevBgajUaEKImIyFqJkuDy8vKwYcMG7Ny5EwkJCYiNjcXVq1f11gkPD8eSJUtw8OBBaDQa\nxMbGihEqERFZKVESXEpKCnx8fODk5AQHBwcMHz4chw4d0o7n5OSgvLwcvXv3BgAolUocOHBAjFCJ\niMhKiZLg8vPz4ezsrF12dnbG7du3ax13cXFBXl6eWWMkIiLrJspdlDXd1yKXyw0er01UVBSio6Ob\nFhwREUmCKGdwrq6uKCgo0C4XFBTAxcWl1vH8/Hy98dqEhIQgIyND7ycxMdG4wRMRkVUQJcH5+Pjg\nxIkTuHfvHsrKynD48GEMGjRIO+7m5gZ7e3ucPn0aALB37174+vqKESoREVkp0c7g5s2bh+nTp0Op\nVGL06NF45plnMGvWLJw/fx4AEBkZiY8++ggBAQEoKyvDtGnTxAiViIisFAu9iYhIkvgsSiIikiQm\nOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIi\nkiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQm\nOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIi\nkiQmOCIikiQmOCIikiQbMQ6am5uLsLAw3L17F126dMGaNWvg4OBQbZ2AgAB06tQJAPDkk0/in//8\npxjhEhGRFRLlDG758uV4+eWXsX//fnh6emLjxo3V1klLS0NgYCDi4+MRHx/P5EZERA1i9gSnVqvx\n008/Yfjw4QCAsWPH4uDBg9XWS0tLw6VLlzBu3Di89tpruHTpkrlDJSIiK2b2BHf37l04OTlBLn90\naGdnZ+Tl5VVbz97eHkqlErt378aMGTPwzjvvQK1WmztcIiKyUia9Bnfw4EF8/PHHkMlkEAQBMplM\ne01N1+Nkpys4OFj7evDgwVi/fj2uXr2Kbt26NSiGyspKAMDt27cbGD0REfDUU0/BxkaU2xWoiUz6\nrzZixAiMGDFC7z21Wo3+/ftrE15BQQFcXFyqbfvll18iMDAQbdq0AQBoNJp6/yOLiopCdHR0jWMv\nv/xyIz8FETVniYmJ6NChg9hhUCOY/WuJjY0N+vXrh/379yMgIAAJCQnw9fWttl5qairKy8vx+uuv\nIzU1FYIgwMPDo859h4SEICQkRO+9hw8fok+fPjh8+DAUCoVRP0tD+fv7IzExUdQYGIdlxmEJMTCO\nmuN46qmnxA6DGkmU8+4lS5YgIiIC//d//4c///nPWLduHQBg586dKCgoQEhICBYuXIgFCxYgISEB\nLVq0wNq1axt1rBYtWgBAjVOjYrCUb4KMQ58lxGEJMQCMoypOT1ovUf7l3NzcsGPHjmrvT548Wfva\n1dUV27dvN2dYREQkIXySCRERSRITHBERSZJi2bJly8QOwhy8vb3FDgEA46iKcVhWDADjqMpS4qCG\nkwmCIIgdBBERkbFxipKIiCSJCY6IiCSJCY6IiCSJCY6IiCSJCY6IiCRJ0glu3759CAgIwPDhwxET\nEyNqLCUlJQgMDEROTo5oMURHR2PUqFEIDAzEmjVrRIvj008/xciRIxEYGIgvvvhCtDgeW7VqFRYs\nWCDa8adPn47AwEAolUoolUqcPXtWlDiSkpIwduxYjBw5Eh9++KEoMcTFxSEoKAhKpRJBQUHo168f\n/vGPf4gSy969e7X/v6xevVqUGKiJBIm6ffu24OfnJxQXFwulpaXC6NGjhStXrogSy5kzZ4TAwECh\nV69eQnZ2tigxJCcnC1OmTBHUarVQUVEhvPrqq8L3339v9jhSU1OFKVOmCBqNRnj48KHg5+cnXL9+\n3exxPJacnCz0799fmD9/vmgxDBw4UKisrBTt+IIgCDdv3hQGDRok5OXlCWq1Wnj55ZeFY8eOiRrT\n5cuXhZdeekkoLCw0+7HLysqEv/71r0JhYaFQWVkpTJgwQUhOTjZ7HNQ0kj2DS0lJgY+PD5ycnODg\n4IDhw4fj0KFDosQSFxeHpUuX1tgWyFxcXFwwf/58KBQK2NjYwMPDA7m5uWaPw8vLCzt27IBMJsOd\nO3eg0Wjg4OBg9jgAoLCwEBs2bMDs2bNFOT4AXLt2DTKZDG+88QbGjBkj2kzDDz/8gICAALi4uECh\nUGD9+vXo06ePKLE8tmzZMrz33nvallnmVFlZCUEQUFpaCpVKBbVarX1wO1kPyT4mOz8/H87Oztpl\nZ2dnpKWliRLLypUrAQCCiDX1uq2GMjMzceDAAezcuVOUWBQKBaKiorB9+3aMGDECrq6uosSxdOlS\nvPvuu6JOGxcXF+OFF17A8uXLUVpaiunTp6Nr167w8fExaxw3btyAra0tXn/9dfz2228YMmQI5s6d\na9YYdCUnJ6O8vBwvvfSSKMd3dHREaGgoRowYgRYtWsDb2xvPPfecKLFQ40n2DK6mZFJT5/Dm5vLl\ny5g5cyYiIiLg7u4uWhwhISE4ceIEcnNzERcXZ/bjx8XFwc3NTfTHMD377LP45JNPYG9vj3bt2mHc\nuHH48ccfzR5HZWUlTpw4gbVr1yI2NhZpaWlISEgwexyP7dq1CzNmzBDt+BkZGdi9ezeOHj2K48eP\nQyaTsbuJFZLsX3xXV1cUFBRol2vrHN6cnDp1CjNmzEBYWBjGjBkjSgzXrl3DxYsXAQD29vZ48cUX\nkZGRYfY4Dhw4gOPHjyMoKAifffYZkpKS8Mknn5g9jlOnTiElJUXvPTH6j/3pT3+Cj48P2rZtCzs7\nO/j7+4t2s0tFRQV+/vln+Pn5iXJ8ADh+/Dh8fHzQrl072NraYuzYsUhNTRUtHmocySY4Hx8fnDhx\nAvfu3UNZWRkOHz6MQYMGiR2WaHJzcxEcHIy1a9fib3/7m2hx3Lp1C4sWLYJKpYJKpUJSUhKef/55\ns8exfft27Nu3DwkJCQgNDYWfnx/mz59v9jju37+PyMhIqFQqlJSUID4+Hi+++KLZ4xg6dCiOHz+O\nkpISVFZW4tixY/D09DR7HMCjs6fOnTuLes2re/fuSElJQVlZGQRBQFJSEnr16iVaPNQ4kr0G5+rq\ninnz5mH69OmoqKjAxIkT8cwzz4gak0wmE+3Y27dvh0qlwieffAJBECCTyTB58mRMmjTJrHEMHjwY\naWlpUCqVUCgUGD58OEaOHGnWGCzJkCFDcPbsWQQFBUGj0eCVV14R5eaO3r1744033sCUKVOgVqsx\nYMAAjBs3zuxxAI++BD311FOiHPuxAQMGID09HWPHjoWtrS169+6NWbNmiRoTNRy7CRARkSRJdoqS\niIiaNyY4IiKSJCY4IiKSJCY4IiKSJCY4IiKSJCY4IiKSJCY4krz09HT069dP78kcj9sX6T6O6uuv\nv8bf//53MUIkIhNggiPJ69GjB+bPn485c+agsLAQgiAgLCwMAwcORFBQEAoLC7FkyRJ89NFHYodK\nREbEQm9qNj744APk5+ejT58++PXXX/HPf/4TMpkMMTExuH//Pp588kkcPnwYW7duFTtUIjICJjhq\nNsrLyzF27FgUFxfju+++Q+vWrfXG4+PjsX//fiY4IongFCU1G9evX8e9e/egUqlw+vRpscMhIhOT\n7MOWiXQVFhYiODgY4eHhaNmyJcLDw7F792506NBB7NCIyESY4EjyNBoN5s2bp72pBAB++eUXhISE\nYNeuXbCzsxM5QiIyBU5RkuRFRkaipKQECxcu1L73/vvvw8HBAcuWLRMvMCIyKd5kQkREksQzOCIi\nkiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikiQmOCIikqT/B6A5jbQ/tFNsAAAA\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11aee4a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_context(\"notebook\", font_scale=1.1)\n",
    "sns.set_style(\"ticks\")\n",
    "\n",
    "sns.lmplot('X1','X2', scatter=True, fit_reg=False, data=df, hue='Y')\n",
    "plt.ylabel('X2')\n",
    "plt.xlabel('X1')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### K-Nearest Neighbors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "K-Nearest Neighbors or KNN for short, is a supervised learning algorithm specializing in classification. The algorithm  looks at different centroids and compares distance using some sort of function (usually Euclidean), then analyzes those results and assigns each point to the group so that it is optimized to be with all of its 'nearest neighbors'. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.4</td>\n",
       "      <td>3.9</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    X1   X2            Y\n",
       "0  4.9  3.0  Iris-setosa\n",
       "1  4.7  3.2  Iris-setosa\n",
       "2  4.6  3.1  Iris-setosa\n",
       "3  5.0  3.6  Iris-setosa\n",
       "4  5.4  3.9  Iris-setosa"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "df = pd.read_csv('iris_df.csv')\n",
    "df.columns = ['X1', 'X2', 'X3', 'X4', 'Y']\n",
    "df = df.drop(['X4', 'X3'], 1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x11aea39b0>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAboAAAFaCAYAAABos5gHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIlljoiMiIlljoiMiIlljoiMiIln7//u5zeNd2DeKAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11aee1518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_context(\"notebook\", font_scale=1.1)\n",
    "sns.set_style(\"ticks\")\n",
    "\n",
    "sns.lmplot('X1','X2', scatter=True, fit_reg=False, data=df, hue='Y')\n",
    "plt.ylabel('X2')\n",
    "plt.xlabel('X1')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: \n",
      " 0.688888888889\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "neighbors = KNeighborsClassifier(n_neighbors=5)\n",
    "\n",
    "X = df.values[:, 0:2]\n",
    "Y = df.values[:, 2]\n",
    "trainX, testX, trainY, testY = train_test_split( X, Y, test_size = 0.3)\n",
    "\n",
    "neighbors.fit(trainX, trainY)\n",
    "print('Accuracy: \\n', neighbors.score(testX, testY))\n",
    "\n",
    "pred = neighbors.predict(testX)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forests"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Random forests are a popular supervised ensemble learning algorithm. 'Ensemble' meaning that it takes a bunch of 'weak learners' and has them work together to form one strong predictor. In this case, the weak learners are all randomly implemented decision trees that are brought together to form the strong predictor - a random forest. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.4</td>\n",
       "      <td>3.9</td>\n",
       "      <td>1.7</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    X1   X2   X3   X4            Y\n",
       "0  4.9  3.0  1.4  0.2  Iris-setosa\n",
       "1  4.7  3.2  1.3  0.2  Iris-setosa\n",
       "2  4.6  3.1  1.5  0.2  Iris-setosa\n",
       "3  5.0  3.6  1.4  0.2  Iris-setosa\n",
       "4  5.4  3.9  1.7  0.4  Iris-setosa"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "df = pd.read_csv('iris_df.csv')\n",
    "df.columns = ['X1', 'X2', 'X3', 'X4', 'Y']\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: \n",
      " 0.955555555556\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "forest = RandomForestClassifier()\n",
    "\n",
    "X = df.values[:, 0:4]\n",
    "Y = df.values[:, 4]\n",
    "trainX, testX, trainY, testY = train_test_split( X, Y, test_size = 0.3)\n",
    "\n",
    "forest.fit(trainX, trainY)\n",
    "print('Accuracy: \\n', forest.score(testX, testY))\n",
    "\n",
    "pred = forest.predict(testX)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Unsupervised Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For unsupervised learning, we don't necessarily have a specific feature that we are trying to predict. Instead the algorithm will attempt to find something on its own without the help of us serving as the 'teacher'. This is in large part used for insights regarding things like clustering. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### K-Means Clustering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "K-Means is a popular unsupervised learning classification algorithm typically used to address the clustering problem. The 'K' refers to the user inputted number of clusters. The algorithm begins with randomly selected points and then optimizes the clusters using a distance formula to find the best grouping of data points. It is ultimately up to the data scientist to select the correct 'K' value. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.4</td>\n",
       "      <td>3.9</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    X1   X2            Y\n",
       "0  4.9  3.0  Iris-setosa\n",
       "1  4.7  3.2  Iris-setosa\n",
       "2  4.6  3.1  Iris-setosa\n",
       "3  5.0  3.6  Iris-setosa\n",
       "4  5.4  3.9  Iris-setosa"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "df = pd.read_csv('iris_df.csv')\n",
    "df.columns = ['X1', 'X2', 'X3', 'X4', 'Y']\n",
    "df = df.drop(['X4', 'X3'], 1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>Y</th>\n",
       "      <th>Pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Iris-setosa</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>Iris-setosa</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>Iris-setosa</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.4</td>\n",
       "      <td>3.9</td>\n",
       "      <td>Iris-setosa</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    X1   X2            Y  Pred\n",
       "0  4.9  3.0  Iris-setosa     2\n",
       "1  4.7  3.2  Iris-setosa     2\n",
       "2  4.6  3.1  Iris-setosa     2\n",
       "3  5.0  3.6  Iris-setosa     2\n",
       "4  5.4  3.9  Iris-setosa     2"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "kmeans = KMeans(n_clusters=3)\n",
    "X = df.values[:, 0:2]\n",
    "kmeans.fit(X)\n",
    "df['Pred'] = kmeans.predict(X)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x11c187748>"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAFaCAYAAAAEtcoCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11c172208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_context(\"notebook\", font_scale=1.1)\n",
    "sns.set_style(\"ticks\")\n",
    "sns.lmplot('X1','X2', scatter=True, fit_reg=False, data=df, hue = 'Pred')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Principal Components Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PCA is a dimensionality reduction algorithm that can do a couple of things for data scientists. Most importantly, it can dramatically reduce the computational footprint of a model or analysis when dealing with hundreds or thousands of different features. It is unsupervised but the user should analyze the results and make sure they are keeping 95% or so of the original dataset's behavior."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting Started"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.4</td>\n",
       "      <td>3.9</td>\n",
       "      <td>1.7</td>\n",
       "      <td>0.4</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    X1   X2   X3   X4            Y\n",
       "0  4.9  3.0  1.4  0.2  Iris-setosa\n",
       "1  4.7  3.2  1.3  0.2  Iris-setosa\n",
       "2  4.6  3.1  1.5  0.2  Iris-setosa\n",
       "3  5.0  3.6  1.4  0.2  Iris-setosa\n",
       "4  5.4  3.9  1.7  0.4  Iris-setosa"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import decomposition\n",
    "df = pd.read_csv('iris_df.csv')\n",
    "df.columns = ['X1', 'X2', 'X3', 'X4', 'Y']\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import decomposition\n",
    "\n",
    "pca = decomposition.PCA()\n",
    "fa = decomposition.FactorAnalysis()\n",
    "\n",
    "X = df.values[:, 0:4]\n",
    "Y = df.values[:, 4]\n",
    "train, test = train_test_split(X,test_size = 0.3)\n",
    "\n",
    "train_reduced = pca.fit_transform(train)\n",
    "test_reduced = pca.transform(test)\n",
    "\n",
    "pca.n_components_"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}