cyberbullying_v2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Detecting Cyberbullying in Tweets\n",
    "***"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/externals/joblib/__init__.py:15: DeprecationWarning: sklearn.externals.joblib is deprecated in 0.21 and will be removed in 0.23. Please import this functionality directly from joblib, which can be installed with: pip install joblib. If this warning is raised when loading pickled models, you may need to re-serialize those models with scikit-learn 0.21+.\n",
      "  warnings.warn(msg, category=DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np \n",
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.metrics import f1_score, accuracy_score, precision_score, recall_score, make_scorer\n",
    "from time import time\n",
    "\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.ensemble import AdaBoostClassifier, RandomForestClassifier, BaggingClassifier\n",
    "from sklearn.linear_model import LogisticRegression, SGDClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.externals import joblib"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading and Analyzing the data\n",
    "***"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scraped = pd.read_csv('../input/labeled-tweets/labeled_tweets.csv')\n",
    "df_public = pd.read_csv('../input/public-dataset/public_data_labeled.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scraped.drop_duplicates(inplace = True)\n",
    "df_scraped.drop('id', axis = 'columns', inplace = True)\n",
    "\n",
    "df_public.drop_duplicates(inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>full_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Offensive</td>\n",
       "      <td>So Drasko just said he was impressed the girls...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Offensive</td>\n",
       "      <td>Drasko they didn't cook half a bird you idiot ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       label                                          full_text\n",
       "0  Offensive  So Drasko just said he was impressed the girls...\n",
       "2  Offensive  Drasko they didn't cook half a bird you idiot ..."
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_scraped.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>full_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Non-offensive</td>\n",
       "      <td>!!! RT @mayasolovely: As a woman you shouldn't...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Offensive</td>\n",
       "      <td>!!!!! RT @mleew17: boy dats cold...tyga dwn ba...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           label                                          full_text\n",
       "0  Non-offensive  !!! RT @mayasolovely: As a woman you shouldn't...\n",
       "1      Offensive  !!!!! RT @mleew17: boy dats cold...tyga dwn ba..."
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_public.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([df_scraped, df_public])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(35787, 2)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeUAAAGgCAYAAACUpX+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3Xd4U+X7x/F30qSb7gWUUcAyC2W2CgKyQUVBhoqigIqiiCgiiuLWryIgIKAioqDwY4MDBJUhsqfsXaB7t2naZuf3RyRaWW2BJmnv13X1gp6cnHOftM0n5znPeR6F1Wq1IoQQQgiHUzq6ACGEEELYSCgLIYQQTkJCWQghhHASEspCCCGEk5BQFkIIIZyEhLIQQgjhJCSUXdzKlStp2LCh/Ss2NpYuXbrw7LPPsnbtWiwWS4n1k5KSaNiwIStXriz1Pnbt2sXMmTMv21Zp6kpKSrIv69KlC+PGjSv1NspbV3mOsSJZLBbef/99OnToQKNGjRg1atRl68ycObPEz/VqX85yjIcPH2bmzJlotdqbsr2xY8fSq1evm7ItgHnz5vH7779ftvyTTz6hSZMmN20/V3L27NkSP7OYmBg6dOjAsGHDWLhwIUVFReXabk5ODjNnzuTEiRM3ueKyu9rrK8pO5egCxM0xffp0IiIiMBgMpKSksGXLFl566SWWLl3K559/jqenJwBhYWEsWbKE2rVrl3rbu3fv5rPPPuOZZ55BqSzd57jOnTuzZMkSwsLCynU8N1JXeY6xIv3yyy8sWLCACRMmEBsbS0BAwGXrDBw4kDvvvNP+/ebNm5kzZ47953yJsxzjkSNH+Oyzzxg4cCC+vr43vL0XXniB4uLim1CZzddff03nzp3p2rXrTdtmWY0ePZoOHTpgMpnIzs5m9+7dTJ06lYULFzJv3jxq1apVpu3l5uby2WefUbt2bRo1anSLqi4dZ3h9KwsJ5UqicePG1KlTx/79/fffT69evRgzZgyTJ0/mjTfeAMDd3Z3Y2NhbVofRaESlUhEUFERQUNAt28+13OpjvFHnzp0D4LHHHrvqh5yIiIgS4XvpOf/9OVdWlfEYa9WqVeL3smfPnjz00EM89NBDjB07luXLlzuwOuEspPm6EuvZsyddu3Zl6dKl9rOOKzXtHjp0iGHDhhEXF0eLFi3o2rUrb731FmBrRv3ss88AaNq0qb0J7t/b+v777/n444/p0KEDMTExaDSaKzZfX7J06VK6d+9OTEwM/fr1Y+fOnSUef/TRR3n00Ucve16XLl2YMGFCqev6b9PumjVr6Nu3LzExMcTFxfHyyy+TkZFx2T7GjRvHzz//TO/evYmNjaV///7s3bu3VK/5H3/8weDBg2nevDmtW7dm1KhR9kC9tP2ZM2cCtoC9GU3Q3bt355133rF/n52dTaNGjejevXuJ9fr378/48ePt3xuNRmbNmkXPnj1p1qwZHTt2ZPLkyRgMhhLP02q1/O9//+Ouu+6iWbNmdOvWjblz53JpMMDFixfbf186depk/1lkZmYCtqbNXr160bx5c9q1a8eAAQPYtGnTNY/pv83Xl5qAV6xYwZQpU2jfvj1t27Zl1KhRl/0M/6t9+/ZkZWWxfPlye22TJk0qsc758+cZMWKE/fLPF198wX8HO8zKyuL111+nQ4cONGvWjD59+rBixYpr7vt6GjRowJNPPsnhw4fZt2+fffnq1at55JFHiI+Pp2XLlvTv358ff/yxxOvRp08fAMaPH28/rp9//hmALVu2MGLECNq3b09sbCz33nsvCxYsuOxSz8qVK+nbty+xsbG0bt2avn37XvbhYPv27Tz66KO0bNmSli1b8tRTT3H27Fn746V5fUXpyZlyJdepUyd+++03jhw5Qtu2bS97vLCwkCeeeIKYmBg+/PBDfHx8SE5O5sCBA4CtGTUtLY3ly5ezaNEi3NzcLtvG559/TkxMDO+++y5msxkPD4+r1rNnzx6OHj3K2LFjcXd3Z+7cuTz55JOsWbOGevXqlfq4SlPXvy1ZsoRJkybRp08fXnrpJTIyMpg6dSqHDh1i5cqV+Pj42Nfdt28fCQkJjBkzBg8PD6ZPn87TTz/Nxo0b8fPzu+o+/vjjD0aOHEl8fDzTpk2jqKiIGTNm8PDDD7NmzRrCw8P57LPPWLhwIStXrmTJkiXAjTdBx8fHl/hgs2vXLjw8PLh48SIpKSnUqFGD/Px8jh8/zpAhQ+zrvfDCC2zfvp2RI0fSvHlzTp06xYwZM0hLS2PKlCkAGAwGhg0bRlJSEqNGjaJ+/frs37+f6dOno9VqGTt2LD169CApKYmvvvqKOXPm2FtIAgICWLZsGdOmTeO5554jNjYWnU7HiRMnyM/PL9exzpo1i7Zt2/Lhhx+Snp7Oxx9/zKuvvsq8efOu+pwvv/yS4cOH06pVK0aOHAlAcHCw/XGr1cro0aN54IEHGDFiBBs2bGDq1KnUrFmTe+65B4D8/HwGDx4MwJgxY6hRowabN29m4sSJmM1mBg0aVK7jAdvf6JQpU9i/fz+tW7cGbB8s+/TpY28x2LVrF+PHj8doNNK/f38iIyOZNm0aY8eOtTeLwz8tDImJidx555089thjqNVqDh8+zJQpU8jPz2f06NEA7Nixg9dee41hw4bx6quvYjKZOHv2LAUFBfbaNmzYwJgxY+jevTuffPIJFouFL774giFDhvDDDz8QFhZ23ddXlI2EciVXvXp1APtZy3+dO3eO/Px8Xn755RLXpfr37w+UbEZt0aIFKtXlvzIhISHMmjULhUJx3Xqys7NZvHgxNWrUAGyB0qVLF+bMmcPkyZNLfVylqesSs9nM9OnTadeuHdOmTbMvj4qKYsiQIaxYsYKhQ4fal2u1WlavXo2/v7/9+AYMGMCWLVu49957r7qfTz/9lFq1ajF37lx7PbGxsfTq1Yuvv/6aV199lSZNmtivs9+sJva4uDiWLl1KZmYmoaGh7Ny5k86dO/PXX3+xa9cu+vXrx549e7BYLMTFxQGwbds2fvvtN6ZNm2Y/47rjjjvw8fHh9ddftwfw6tWrOXz4MEuWLKFFixb29UwmE19//TUjRowgODiYyMhIAJo0aVKi2f3gwYM0a9aMp59+2r6sc+fO5T7WqKgoPvroI/v3mZmZTJ8+nby8vCtemwdbS8qlSypXes0tFgsjR460B/Dtt9/Ojh07+Omnn+zLvv76a7Kzs/npp5/sx9q+fXvy8vKYMWMGAwYMKHV/i/+69Lfw77/R5557rkR9cXFxpKamsnjxYvr374+Hh4e9Zei/zeIAjzzyiP3/VquVtm3bUlRUxOLFi+2hfODAAUJCQnjllVfs6/67H8O/OyTOmDHDvrxt27Z069aNBQsWMG7cuOu+vqJspPm6krvefCN169bFz8+PN998kzVr1pCamlrmfXTt2rVUgQy2AL30JgTg6+tLp06dOHjwYJn3W1oJCQlkZ2fTt2/fEsvbtGlDzZo12bNnT4nlsbGx9kAG7G9+13ptioqKOHbsGL179y7xAaFWrVq0atXqsn3cTJeCdteuXfZ/4+PjiYuLs59B79y5k8jISHugbN26FW9vb7p164bJZLJ/XTrjutRcv3XrVurWrUvTpk0vW0+v13Po0KFr1hYTE8Nff/3FBx98wI4dO9DpdDd0rJ06dSrxfXR0NHDtn01p/PuDgkKhoEGDBiW2uXXrVlq3bk1ERMRlr0NmZiYXLlwo974v/Y3++2/o7NmzjBkzhg4dOtCkSROaNm3KDz/8QEJCQqm2mZaWxmuvvUbnzp1p2rQpTZs2Zc6cOWRnZ6PRaABo3rw5mZmZTJgwgS1btlzWc/706dOkpaXRt2/fEsfs6+tLTExMqS/piLKRM+VKLi0tDeCqvaCrVavGggULmD17Nm+//TaFhYXcdtttjB49mp49e5ZqH2XpYX2lZq2QkBDS09NLvY2yysvLAyA0NPSK+770+CX/DmSwdRwD0Ov1V92HRqPBarVe8bUICQkhOTm5zHWXVmhoKPXr12fXrl20bduW8+fPExcXh4eHh/0MZ9euXfbwBluLRVFRETExMVfc5qXXJDs7m4SEBJo2bXrN9a5m0KBBmM1mVqxYwYIFC3B3d6dz58689tprJc6oS6s8P5vrcXNzu6zHuLu7e4ltZmdnc/To0XK/DtdyKfwv/X5qNBqGDRuGv78/48ePJzIyErVazbfffsv69euvuz2TycRTTz1FQUEBzz77LFFRUXh4eLBu3TrmzZtnP64OHTowdepUvv/+e5555hkUCgXx8fFMmDCB2267jZycHADGjRt3xVsZ69atW+5jFlcnoVzJbd68GQ8Pj6u+mYCtw9HMmTMxmUwcOXKEL774ghdeeIE1a9bYz0SupbRnyWB7c/uvrKwswsPD7d+7u7tTWFh42XrlfeO71Kx5pSb8rKwsmjVrVq7t/pufnx8KheKq+7ha0+rNEhcXx59//knr1q0JCwujXr16eHp68uqrr3LgwAFOnz7NE088YV8/ICAAHx8fvv322ytu79LPIyAggKioqKteWrjebTxKpZIhQ4YwZMgQcnNz2bp1Kx999BHjxo3ju+++K+fRVryAgABq165doqPcv5WlP8R/bdmyBcB+PXnv3r2kp6cza9asEh+aTCZTqbZ39uxZTp48yfTp00t0lvvll18uW/fuu+/m7rvvRqvVsnPnTiZPnszIkSPZuHGj/Xd2woQJtGnT5rLnXqvviCg/CeVKbMOGDWzcuJGhQ4fi5eV13fVVKhWxsbGMGTOGjRs3cvbsWaKjo+1nIzqd7obvQf3rr79ITU21X+vWarVs2bKlRLNkzZo1Wb9+PQaDwb7vPXv2XBbUpa0rKiqKkJAQ1q5dy8CBA+3L9+/fT3JyMsOGDbuhYwLw9vamadOm/PLLL4wePdre8exSp7l/X+O7FeLi4li0aBGrV68mPj4esF2rrF27NtOnT8dqtdqXg+3a4YIFCzAYDPYwuJI777yTrVu34u/vf80Oaf/+WVxNYGAgffv2Zf/+/axbt66sh3hD3N3db6jp/M4772TVqlXUqlXrpn7AOnPmDHPnziU2NpaWLVsC/7yGarXavl52drY9vC+5WivBpef/+zKKwWCw98y+El9fX7p160ZCQgKffPIJWq2W6OhowsLCOHv27HX/Rm709RX/kFCuJI4fP05ubi5Go5GUlBQ2b97ML7/8Qvv27XnppZeu+rxNmzaxZMkSunXrRmRkJMXFxSxcuBAfHx/7m0T9+vUBmD9/Ph07dkSpVF612fN6goODGT58OKNHj7b3vi4qKioxqlWfPn1YsmQJr732Gv379ycpKYn58+dTrVq1EtsqbV1ubm48//zzTJo0iXHjxtG3b1/S09P59NNPqVu3rr1T240aM2YMI0eOZOTIkTz88MMUFRUxc+ZMfH19b0rwX0tcXBwKhYIdO3bw/vvvl1i+bNky6tatW6I1omPHjvTo0YNRo0YxbNgwe2tBUlISmzdv5o033qBmzZo88MADrF69mqFDhzJs2DCio6MxGAxcuHCBjRs3Mm/ePNzc3GjQoAEA3333Hffccw9ubm40btyY119/ncDAQGJjYwkKCuLcuXOsXbvWfu26olxq3t+yZQtBQUEEBweX6NtwPU888QTr16/n4Ycf5rHHHqNu3boUFhZy9uxZDh06ZL/N7VoSExM5ePAgZrPZPnjIihUrCAsLY+rUqfb1Wrdujbe3N5MmTeLZZ59Fq9Uya9YsQkJC7JejwNbZ0dfXlx9//JGoqCg8PT2pXbs2DRs2JCwsjMmTJ9uvV3/99dclQh6wh2+7du0IDQ0lJSWFxYsXExsba/+QO2nSJMaMGUNxcTE9e/YkICCArKws9u/fT926de0fNm/09RX/kFCuJMaMGQPYmpSCg4Np0qQJU6dOpVevXtdsXq5Tpw6enp7Mnj2bzMxMfHx8iImJYf78+fZrfnfddRcPP/wwixYtYtasWVitVk6ePFmuOtu2bUtcXBxTp04lLS2NBg0aMHfuXKKiouzrxMfH8/bbb/P111+zYcMGmjRpwuTJk3n++edLbKssdQ0ePBhPT0/mzZvHqFGj8PHxoWPHjrz88sslboe6ER07duSLL75g1qxZvPDCC6jVatq1a8fLL79cIhBvhcDAQKKjozl58mSJM+L4+HiWLVtW4nryJZ9++inffvstq1atYtasWXh4eBAZGUmHDh3sZ4Pu7u588803fP755yxatIjk5GR8fHyoXbs2nTp1svc4btGiBU8//TQrVqzg+++/x2Kx2JvTV69ezapVqygsLCQsLIwHHnigRO/iivDyyy/z5ptv8vzzz6PT6Rg8eHCJe7uvJyAggKVLlzJr1iw+//xzMjMz8fPzo169evTu3btU25g5cyYzZ85ErVbj7+9PdHQ0L774Ig888ADe3t729cLDw5kxYwaTJ09m9OjRRERE8Pjjj5OSklLicoNarea9995j+vTpPP7445hMJqZOncrdd9/N7Nmzee+99xg3bhwBAQEMGjSIgICAEsccGxvL999/z6+//kp+fj4hISHceeed9vcSsN0Dv2DBAj7//HMmTpyITqcjNDSUli1blug4eaOvr/iHwnq97rlCCCGEqBByS5QQQgjhJCSUhRBCCCchoSyEEEI4CQllIYQQwklIKAshhBBOQkJZCCGEcBISykIIIYSTkFAWQgghnISEshBCCOEkJJSFEEIIJyGhLIQQQjgJCWUhhBDCSUgoCyGEEE5CQlkIIYRwEhLKQgghhJOQUBZCCCGchISyEEII4SQklIUQQggnIaEshBBCOAkJZSGEEMJJSCgLIYQQTkJCWQghhHASEspCCCGEk5BQFkIIIZyEhLIQQgjhJCSUhRBCCCchoSyEEEI4CQllIYQQwklIKAshhBBOQkJZCCGEcBISykIIIYSTkFAWQgghnISEshBCCOEkJJSFEEIIJyGhLIQQQjgJCWUhhBDCSUgoCyGEEE5CQlkIIYRwEhLKQgghhJOQUBZCCCGchISyEEII4SQklIUQQggnoXJ0AUKIimO1WrFarFitANa/F9r+p7QaUVoMoFDYlisUgBLc3MFN3iqEqAgKq9VqdXQRQojyM5stWC1WFAoFSqUtUC1mC2aT7ctkMGMymDAazBh1JsxGM2aTxR7MVit4eKuJiPJH9eNjoHADpZvtX5UnePqDdyj4hNn+9QoCrwDw8AN3X1B5gcUIJj2gALW3hLgQ5SR/OUI4OavFisViC9FLoWvUm9AXGikq0KHTGtEXGdAXGjDqTVjMZf+c7RPgSUTdanBkSfmK9AmFgCgIjILA+hDaBIKjIaCOLcRNeltwu3mC2rN8+xCiCpBQFsKJWK1WLCaLvQm5uEBPQXYRxQU69IVGdIUGTAbzrdp5+Z9bmGn7St59+WNKN/CrBYH1IDwG6nSEGm3AtzoYi2zN42qv8u9biEpEmq+FcJB/AhgUCgXFWgMF2YUU5hZTmKdDX2SssFp8Ajy5rXU46o99KmyfqL0hvDlUbwV17oSa7WzhbSyyNYmr3CuuFiGchISyEBXIbLIAtmu++RlaNFlFFOYVo9MaHFqXT4Ant7UJR/1RBYbylag8IKIl1OsGjfpBWFMw6WzXrpVujq1NiAogoSzELWS1WDGbLSiVCrR5xeSmFKDJLERX6NgQ/i+nCeX/UntB7Q7QoDc0vNd2Jm3Sg0e1f3qJC1GJSCgLcZOZTWYUCgUGnYnc1ALyM7Roc4uxWpz3T81pQ/m/vIOh7l22gG54H6AAdx85ixaVhnT0EuImuHRtuLhAT+aFPHLTCm5dh6yqrCgbji23fSmUtmvRzR+FJgOQgBaVgZwpC1FOFrPt+rCh2EjGhTxyUjQYdSYHV1U+LnOmfDUKpa1Xd/NHoekA22goEtDCBUkoC1EGtnuGrZiNZjIT88hO0qB3suvD5eHyofxvlwK6xVBoOtB2q5dHNUdXJUSpSCgLUQpmkxlQkJWYR9bFPIo0ekeXdFNVqlD+N5WnLZjveNl2n7Sbh4w2JpyahLIQV2G1WrGYLZj0ZlLOZJOTnF+u0bJcQaUN5X8Lbw7xL0CzwWC12G6zEsLJSCgL8R+XrhVrsotIO5NNQXaRgyu69apEKF/i7gsxQ6D9y+AbbhuoRK49CychoSzE32xN1JBxPpeMhFwMLtppqzyqVCj/W2QcdHwdorqCUi1N28Lh5DdQVHlmkxmz0ULyyUyykzVOfT+xuMmSdsGieyGkIdz1LkTfA0oVuKkdXZmooiSURZVlNtmmMEw6lkF2ssbR5QhHyjoJywbZZrnq/LbtvmeFm4y/LSqcNF+LKsdsMmMxWUk8nkF2cr7tntYqrso2X1+NX6StWbvFUNtwniqZblJUDAllUWWYTWYsZitJJzLJTsy7oZkKKxsJ5avwDYcOr0LrJ0GhkjNncctJKItKz2yyYLXYwjgrMU+uGV+BhPJ1+NWEHp9Aw77g5glKpaMrEpWUXFMWlZbFYsFqgZTTWaSfy5EwFuWnSYblD0FEC7h7DoTHyH3O4paQUBaVjm3QDyt56QUkHs3AqK86tzaJWyztL5h3B9zWB+79EjwDbGNsC3GTSBuMqFTMJgvFBXpO7rjAuf0pEsji1ji9FqZHwR/vgaEQTK4//rlwDhLKolKwmC2YjWYuHkvj6JYECvN0ji5JVHZmI/z5P5hRH06usYWzEDdImq+FyzObLOSmaUg8miFzGIuKp0233ePcoBf0W2hrzlZ7Oboq4aLkTFm4LIvZgqHYyOndiSQcSJVAFo515heYUQ+OLpGzZlFuEsrCJZlNZrKS8ji86WyVmDBCuAh9AaweBov7gjYNjPK7KcpGQlm4FIvZglFv4uy+FC4cSq+0UykKF5ewEWY0gL8WSDCLMpFQFi7DZDSTl1HAkc3nyM/QOrocIa7NUAg/PQMLe4AmScJZlIqEsnB6FosFk9HMxSNpnN2bIteOhWu5uA1mNoQTP8i1ZnFdEsrCqZmMZrS5RRzdco7sJJnJSbgoYxGseAjWjrYFs8Xi6IqEk5JQFk7LZDSTeiaLk9sTMRTLICCiEjg4H+a2A00iGIsdXY1wQhLKwulYzBaMBhPn9ieTdibH0eUIcXNlHoPZTeHUT9KcLS4joSycikFvpFir4/jW8+RnyBuWqKQMhbYBR355wfZ/maxP/E1CWTgNvc5AfqaWE9suoi8yOrocIW69/V/BV/G2WahMekdXI5yAhLJwOKvVitFgJO1sDucPpMm9x6JqyTgCc5pB6gFpzhYSysKxzCaz7frxvlQyzuU6uhwhHEOXD990hJNy21RVJ6EsHEavM6Ar1HN86wU0WfJGJKo4sxFWPAzbPpZgrsIklIVDFBUWodPqObkjEUOxXD8Wwm7LO/DDExLMVZSEsqhQVqsVTX4BxfkGzuxOxmyUQRSEuMyR/7MNz6nLB4uMYFeVSCiLCmO1WtFqCinOM3Buf6p06BLiWhK3w5etoSBFemZXIRLKokKYzWaKtEVo0ou4eCgDJI+FuL6cs/B5rO1fCeYqQUJZ3HJGoxGD3kjWBQ0pJ7MdXY4QrqU4B76+A7JPgVHn6GrELSahLG4pg96A2WQh9WQ2GQl5ji5HCNeky4ev20PWcRkzu5KTUBa3THFRMVarlcTDGTLDkxA3Sl8A8++0DTYiwVxpSSiLW0KjKUChUJB4OJO8dK2jyxGicjAUwjedIe2gBHMlJaEsbrqc7BxUbiqSj2WRm1bg6HKEqFyMRfBtF0jZA4YiR1cjbjIJZXFT5eTk4u7uQfrpXHJSJJCFuCVMOtt9zMk7bSEtKg0JZXHT5OXmoXJTkXVeQ1ZivqPLEaJyM+nhu96Qdkh6ZVciEsriptDkawAFecmFZJ6XXtZCVAizAb7rAZpE29jZwuVJKIsbVlCgxWQ0o83Qk35WZnoSokLpC2B+RyjOBosMW+vqJJTFDSnUFtoml8gzkXpKBgYRwiG0abZgNkg/DlcnoSzKrbi4mLTUNNzMapKPZzm6HCGqtuzTts5fMruUS5NQFuWi1+k5fvQEocHhXPgr3dHlCCEAknfDsoHSI9uFSSiLMjOZzOzZvZfGjRtzbp/M9iSEUzm9Dn5+Ts6YXZSEsigTq9XKnt17adO2DQn7UjHqTI4uSQjxXwfnw45pEswuSEJZlMmRQ0dp1qwpSUczKcyXeyOFcFqbJ0HSDttAI8JlSCiLUrtw/iKhoSHkpRSSmyq9PIVwalYrLB0ARXKrlCuRUBalkpWVja5Ih5vZg7TTOY4uRwhRGrp8+K4nmGTyClchoSyuq6iwiBNHT1A9ogYXDklPayFcSsZR+OEJub7sIiSUxTWZTCb+2PQHbdu14fyBNKwW6WkthMs58n9wYL4EswuQUBZXZbVa2fz7FuLviCfpWCb6IhlbVwiXtf4FyDxmGy9bOC0JZXFVh/86Qlh4GIYCi0zDKISrs5hh0T22sbKF05JQFleUlZnF+YTz1K/XgMSjGY4uRwhxMxRmwJL+YJARv5yVhLK4jMFgYOOvm+jRqwfnD8iIXUJUKhf+gAPzJJidlISyuMwfm/+kY+eOpJ/NoUijd3Q5Qoib7deXoSjDdi+zcCqlCuW33noLhUJBz549L3tswIABdO7c+WbXdUMsFgvPPvss4eHhKBQK3nrrLQDWrFlD48aNcXd3p27duhVa0+bNm1EoFBw5cqRC91tWZ06dwcfbG3eFJxkJeY4uRwhxK5j08H/95P5lJ1SmM+UNGzawZ8+eW1XLTbNy5Upmz57Nhx9+yI4dO3jiiScwm80MHTqUFi1asHHjRlatWlWhNbVq1YodO3ZQv379Ct1vWWjyNezds5+WrVrK/chCVHZpB+HPj8CgdXQl4l9UpV0xKCiIyMhI3n//fVavXn0ra7phJ06cIDAwkOHDh9uXJSUlodFoePjhh+nQoUOF1+Tn50d8fHyF77e0zGYzP/+wjt539yLxaCYmg9nRJQkhbrWt70OTARDaBJRujq5GUIYzZYVCwWuvvcYPP/zA4cOHr7rewYMH6dq1K97e3gQGBjJkyBDS0/856zp//jwKhYKlS5cycuRI/P39iYyM5M0338RSivFZi4qKeP7554mIiMDT05O2bduyYcMG++OdO3fmjTfeIDcBwIsMAAAgAElEQVQ3F4VCgUKh4JtvvqFWrVoA3HfffSWatC0WC//73/9o0KABHh4eREdH8+2335bYZ+fOnRkwYACLFi2iQYMG+Pn50bt3b5KSkkqs9+GHH9KgQQM8PT0JDw+nV69epKWlAZc3X3fq1IlBgwZddnzjxo2jdu3aWP++1qPT6Rg/fjy1atXCw8ODFi1asHbt2uu+TmW1a8duImvVxKyzkpcmt0wIUSVYzLD0AZm0womUqfl64MCBREdH8/7771/x8czMTDp37kxRURGLFi1i5syZbNmyhe7du2MwlLxhffz48fj6+rJ8+XIeeeQR3nnnHZYvX37dGp588knmz5/PxIkTWbVqFbVq1eLuu+/mzz//BGD27NmMGDECf39/duzYwY4dO+jduzcrV64E4JNPPrE3aQOMHj2a9957j6eeeoqff/6Zfv36MXz4cH766acS+921axefffYZU6ZM4csvv2T//v089dRT9scXLFjABx98wIsvvsj69euZM2cODRo0oLDwyiPoPPjgg/z0008lHrdarSxbtoxBgwahUCgA2zX7b775htdee40ff/yRtm3b0rdvXw4ePHjd16q0UpJTOfTXYeLi25F4NPOmbVcI4QKyT8Nvr0oztpModfM1gFKpZMKECYwYMYJ33nmH6OjoEo9PmTIFgPXr1+Pn5wdAdHQ0cXFxrFixgoceesi+bseOHe3rd+/enV9++YWVK1de8ezxkuPHj7N48WLmz5/PY489BkDPnj1p3rw57777LuvXr6dJkyZERkaiUqlKNBe3bNkSgIYNG9qXnzlzhjlz5pTYXrdu3UhNTeXtt9/mnnvusT9fo9Hw888/ExgYCEBaWhpjx46luLgYLy8vdu/eTY8ePRg1apT9Of3797/qsQwYMIDRo0fz448/8uCDDwKwc+dOLl68aP/+999/5+eff2bz5s106tQJgB49enDq1Cnef/99li1bdtXtl5bJZOLnH9Zy3/19ST2dLfMjC1EV7Z4JsY9BREtQyk05jlTmV/+RRx6hdu3afPjhh5c9dimYLgUyQLt27ahbt679TPaSHj16lPi+SZMmJZqDTSaT/ctstl3f3LNnD1arlYEDB/5zAEolAwcOvGz7pfH777+jVCrp169fif117dqVgwcP2vcL0LZtW3sgX6oXIDk5GYDY2FjWrl3Lm2++ye7du0s890pCQ0Pp0qULS5YssS9bsmQJ9evXp02bNgD89ttvRERE0L59+8vq27t3b5mP90r27dlPjZrV8fHyJeO89LYWospaNRTM0oztaGUOZZVKxfjx4/nuu++4cOFCicdSU1MJDw+/7Dnh4eHk5JSc7i8gIKDE9+7u7uh0tl+I8+fPo1ar7V+Xeiynpqbi6+uLt7f3ZdsvKipCry/bPbVZWVmYzWb8/f1L7O/xxx/HZDKRmpp6zXoBe83Dhw/ngw8+YOnSpcTFxREeHs4bb7xxzXB+8MEHWbduHRqNBovFwrJlyxg8eHCJ+tLS0krUplareeutt0hMTCzTsV5Jbk4ue3buoWv3rlw8JKN2CVGlZR6DvV/IoCIOVqbm60uGDx/Oe++9x0cffVRiefXq1cnIuPzNPT09ndatW5d6+zVq1Chx65WHh4d9+1qtlqKiohLBnJ6ejre3t3290goKCkKlUrFt2zaUV2iyCQsLK/W2lEolY8eOZezYsSQmJvL9998zceJEatasydNPP33F5/Tr149nnnmGNWvWUKdOHVJSUkqEclBQEDVr1rwlvd2tVivr126ge6/u5CRpKC6QQUKEqPI2vQHNHwF37+uvK26JcoWyh4cH48aN49VXX6V169ao1WoA4uLimDNnDgUFBVSrVg2wNTmfP3++TLchubu725tw/61t27YoFAqWL1/O0KFDAVu4LF++vFy3OXXp0gWz2Ux+fj7du3cv8/OvplatWkyYMIH58+dz7Nixq64XGBhIjx49WLJkCXXq1KFx48Y0b97c/njXrl2ZMmUKvr6+NGrU6KbVB3D86HGsViu1atXm+B/nb+q2hRAuylAIP42EfgvA3dfR1VRJ5QplgJEjR/LBBx+wfft2eyekF198kTlz5tCzZ09eeeUVtFotEyZMICYmhgceeOCGi23cuDEPPfQQzz33HBqNhgYNGjB37lxOnDjBnDlzyry9hg0b8vTTT/Pggw8yfvx42rRpg06n4+jRo5w6dYqvvvqq1NsaOXIkQUFBxMfH4+/vz6ZNmzh9+vRlrQn/NXjwYIYPH46/vz/PPfdcice6d+9Oz5496d69O6+88gpNmzZFo9Fw8OBBdDrdFa/rl0ZRYREb1v3G8KeGkXQ0Q8a2FkL84/gqaH8EarSTTl8OUO5Q9vb2ZuzYsUycONG+LDQ0lE2bNvHSSy/x0EMP4e7uTp8+fZg2bZr9GuyNmjt3Lq+88grvvvsueXl5xMTE8NNPP5V7QJBZs2YRHR3N3LlzmTRpEn5+fjRp0oQRI0aUaTu33347c+fO5YsvvkCn09k/MNx///3XfN59992HSqUiKyvL3uv6EoVCwcqVK/nggw/49NNPuXjxIkFBQcTGxjJ69OgyH+slmzduoWlME0w6C/kZMum5EOI/fngCntwNSmnGrmgKq1VGJK9KLl64yJLFyxg95jnO7EqWa8kCAJ8AT25rE476Ix9HlyKcxb1fQPOhoPZ0dCVVirRNVCEmk5l1P62nS5e7yE8vlEAWQlzdbxPAIuMWVDQJ5Srk6OGjGAwGYlo0I/VUtqPLEUI4s+Jc2PaxrfOXqDASylWEXqdn42+b6N6zG5nn8zDq5ROwEOI6dk4Fq0xOU5EklKuIPbv34e3lRVS9KNLP5Tq6HCGEKzAUwh/vy9lyBZJQrgIKNAVs27qNHr17kH4uF7Pp+rNxCSEEYBsX22y4/nrippBQrgJ2/LmTsLAwqlevTuZ5OUsWQpSBsRg2vy2zSFUQCeVKLic7h31799O1exfSzuTIQCFCiLLb9zmY5G6NiiChXMlt27qd0NBgwiPCyU7Kd3Q5QghXZNLDxjdAL2fLt5qEciWWkZ7Job8Oc2fnjmReyJOzZCFE+R34Cowyg9StJqFciW3buo2AgACi6keRKXMlCyFuhNkImybJ2fItJqFcSeXm5HL86Anu6HAHOckaTAa511AIcYMOLQCkxe1WklCupPbt2Y+HhwdNmjYiI0F6XAshbgJjMeybK52+bqFyzxIlnFehtpB9e/bToWN7CnKK0RcaHV2ScCIKpQIPbzVqDxUKpQKlUoFnNQ8UCiU07me7J9VsBEMB5J0HbRrIvDXikp3ToO0oR1dRaUkoV0KH/jqMxWqlRcsWJP6V6ehyhAMo3RR4VfPAw8fdFsBeKjy8VXj6eKD2UFGkLUaTn0+xVoO3twdWq5787EKMrSehVCpRKt1w9/DAyy8YN3cvzPnJWHPPo8w9jSr3NOSeg/RDtn9F1aJJgoTfoUFvmW/5FpBQrmQMBgM7/txJqzYtMestaHOLHV2SqCBefh74h/rgG+JFtUBvNJoCNPkFZOdno0kvoEBj+yoqKsJqtZKekkTSqQO0vz3mmttVqd3xqRaAr18APn518Atrhn+DagSGRaIw61Cc+w3VuV8gYSMUZlTQ0QqH2voB1OkEHr6OrqTSkVCuZE4eP4lOp6N161ZkJMh9yZWZysPNHsL+ob4YDAaSEpM4sf8wqalpmIz/TDpitVoxGAyYTCasFitWq5XiYh0Wi5mCvBwUSgUKhRI3NzfUHp64uf3z1mAyGsjPySA/5/LA9QsMIaJWPSJbTiDknrlY85NRnvsFt3Pr4eyvMvVfZZW4HQpSwCPa0ZVUOgqrVS4WVRZms5kvZs3F39+fBwb15/Dv525pR8nft61n0ZpvuZCcgE5XTERYDXp37svQ/sNRq90vW3/q3A/5vx8XMuT+xxkzfPw1j+O71fP5c89mEhLPAtC4flOeeXQMTW7756zOZDLyyZcf8OvWdYQGhzHxuXeJadSixOMPjb6fUUPHctft3W7ikTtWQLgvIXX88Q30IjUljcTERJKTUtDka9Dr9OgNBgx6PVarFaVCgRUrViv4+ftRrVo13NzccFMpsZotFGcnEBzkjdlowmIxUVykRZOdiclkABQoFAosFgsqtRq1hyde3r6orvCzBVAolASF1aB6rShq16mDr58/yv1zUe77HPITK/ZFErdei0ehzyzwqOboSioVOVOuRM6fO09uTi53dupATpLmlt+5kF+QR+uYdjzSbzjVfKpx9PRhvlo8i5zcLF5++vUS6567eIYff1uJj/f1m7v0Bh0Lln/FPd368fiAJ1EoFCz7eRFPvvIIX328iMYNmgKw5tcVbN/3B2+9+D/2HNzBxMkvsvKLX1Cp1AAs/XkRIUFhlSKQ3dRKQmr5E1IngOLiIg4dPsi5swkUFxdToCnAarGAQkFISDA1ImsQGhZCUFAg1fz8qOZXDV9fX1Qqt1Lty2q1oisqRKvJpVCThzY/h5yMFLIzUki7eBaDXgdY8fDyxcfP335WbbVayE5PIjs9iSN7t+IfFEp0005EPf0c1sQdqPdMhzPrwSoTolQKR5ZA7xmOrqLSkTPlSmTB19+Rk5PDyFFPcWZXMjptxc/sMnvhpyxfu5jfF+1EoVDYlz/7xnBiGsaybvMPdLmjx3XPlAuLtfj5+tuXGY0GBjzTh9Yx7Zg05gMAJvzvBZpGx/Bo/xGYzCa6PRzPvMmLqV/7NvI0uQx85m5mvzef26Ia3roDvsW8qnkQWsefoJp+JF5M4q+Dh0g4m4BOp0OhAH9/fxo2aUS9enWpUbMGHp4et7Qei8VCTkYKqRfOcObofpLPHcdstt0D713NH29fvxI/dwA3lZo60c1o3DQGH08PlPtmo9z7ORTn3NJaRQXo9hHEvwCqK7eeiLKTM+VKIjsrh6TEJFq3bY2h2OSQQAbwrxaA0VjyFqzft63nfOI5Ppn4Ges2/3Ddbbi5uZUIZAC12p16tRuQm//PG7nRZMTD3RMAlZsKlUqN6e99z/luOl3b93TZQPb286BGoxA8q7lz/Ohx1qzbR1ZmJmp3d+o3qE90w9uIrB2Jv//lIXgrKZVKQiIiCYmIJCauMyajkczUiySfO8mpQ7vITLmI0s0N/+BwVGpbi4XZZOTcsQOcO3aAoLAaNGp2H5GjX0a5fTLKXdNs974K13RwPrR7FpBQvlkklCuJ48eOo1AoadioITlJBRW6b7PZjNFk4MTZYyz96Tse6D3YHhQ6vY7pX3/Ms4+9iJend7n3YTAaOH7mKH3u6mtf1qh+E9Zt/oGu7Xuy88A2zGYztWvW5XTCSTZu28CyOT/f8LFVNLWHihoNg/EL9WbPnr3s2r4bg8FAzcgaDHhwAPXqR6FSOc+frUqtpnrt+lSvXZ/WnXqTlZbE0T1/cGT3FkwmIz5+ASXOnnMyUti+MYVqAbtoFTeA8HbPo9z8OsoDX0uztivKOmG7RSrENT/8OiPn+esW5WY2m9m3ez/h4WHUjKxu6+BVgToNao3BaDsz73PXfTw/7GX7Y98un0tIYCi9O997Q/uYv/QLCrQa7us+wL5s8L2PsHXPZno/1hGVSsVrz76Dl6c3U7/6H8MGjSTAL/CG9lmRFAoIqxdERP1Ajh45xsJFmzCajDSPjaFVm5aER4RX6BlxeSgUCkKr16Jz3yHc3r0fZ48dYN8f68hMuYBK7Y5/cDhubrbr2gV52WxZv4bg8Jq0bf8K1do8h/rnpyB5t4OPQpTZvi/grvfAvfwfusU/JJQrgcSLSRQVFtIuvi156VrMpoo94/jq40Xo9MUcPXWYeUvmMPmL93jlmUkkpyXx3ar5zH5v/g0Fyp97tjB/2ReMGTaeOpFR9uV+vv58O2UpSakXCfQPwtenGpu2/0pWTgaD7n6YsxdP8/7MSVxMTqBti9uZOPpdfEvR0ayi+QR4UjsmnLz8XOZ+Pg+zycSdd7WnabOm+Pj6OLq8cvHw8qZJ6/Y0bnUH6UkJHNq5ieP7/sRNrSYgJALl34NOZKcn88vK76jbMIY2D61DeeoHVBvGgk4mUHEZhxdD1w8cXUWlIaFcCfx18BAqtZqGjRqSeaZim67B1owMENukNQF+gbz96asMuf9xZi/8lDta30ndyCgKtBrA1lHIYDRQoNXg61PtumF97PRhJk5+kX49B/HQfUMve1yhUFCrRh3A1sQ945tPGPfUa6hUat6aOoEeHfswoM9DvDFlPPP+b/Y1O5hVNIVSQc3GIQRE+LJh3XrOJ1yka48uNGrSqNQ9pZ2dQqEgolY9ImrVo3XH3uz4dSWnD+/B09sXv8AQ+8///MnDJCecomV8J+o+fRjV0n6QstfB1YtS0aZB6n6odYejK6kUZIw0F1dUVMzxI8eJqh+Ft7c3msxCh9ZzKaBT0pO4kJzAph2/0vXhePtXelYay35eRNeH48nITr/mti4kn2fsO8/Qtnk8456aeN19L1r9DbVr1KF9m05oi7ScPHecB3o/iJenN/f1GMC+w87TNKr2VBF9eyRaYx5fzp5LcGgoTz/3FM2aN600gfxfweE1uOeR53jw2UkEBoeTkZSArvif31ejQc/uPzawffs2jEN+xRL3vAOrFWWyZw7oNY6uolKQM2UXd/b0GSxWKw1uq09OsuP/KP46vh+AGuGRTHzuXYp1JSdFnzj5JVo1a8sDvR8k0D/oqtvJysnk+TefpGZELd57+RP7tcirrp+byXervuarjxfZFvx9p59Or8PbywedrhhnufvPN8iLqJbV2b5tO3m5eTz+xFD8A/yv/8RKonrt+gwa9TonD+5ky4/fU5CXTXBYTZR//4yTzh5nbWYanXu9iE+dLqjWDJU3fGd3YjXc+6Wjq6gUJJRdmNVqZc+uffj4+FC3bl1Sj1XsFI3Pv/kU7WLjqVe7AUqlG4eO7+f71d/Q/c7eRFavfcXneLh7EB4SQeuYdvZlP29cw3szXmfll79QPawmOr2OMW8/RUGhhpdHvs7phFP2dd3Vahr+fTb+b7MXTKPPXfdRN7IeAL4+1bgtqiGfzvuIPl3u45vlX9KmefxNfgXKLrRuAOH1A/ll7S80jWnKXd06O30HrltBqVTSuNUd1G3YnB2/ruTgtt8IDI3Aw8vWWahQk8u65Qto06ErdUceQrWsH6QecHDV4qoMWji7ARr2tfVaFOUmoezCsrOySUtNo36Denh7e6HNTanQ/Te5rRk//b6a1Ixk3NxU1AyPZNTQsTzQa3CZtmO1WjBbzPbZAXPysjmdcBKAF999psS61cNqsOar30osO3b6CNv2/nHZLVBvvfA/3p35OhP+N4Z2Le7gicElt1WRFEoFtWPCUHpa2LFjJ73u6Ymnp6fD6nEWXj6+3HXfo0TWb8yGpXPRFWnxDw4DwGIxs/uPDaQ1aELcI7/jtmEsyr++dXDF4qr2fwV1O4Nn1Wn1uRVkRC8XtuPPHWzeuJXOXTpSv3Y0F/669jVa4RhqDxX12lQnJTUZnV5Pi5YxVfLs+HpyM9NYu2g2makXCYmoZW/OBtvEF936Poj7ljdsY2kL56P2gldyQCUfNm+EdPRyYYcPHaVaNV/q1K2DJqPo+k8QFU7l7kaDuJocPXYEXz9fYls1l0C+isDQCAY9M5Hm8V3ISL6Avvif32lNbhYbVn2PoePbWNqNdmCV4qqMxbZe2OKGSCi7qLy8fLIzs/Hzr0ZoeCj5GY7tdS0u56ZW0qBdTc6eO0PTmCZE1qrp6JKcntrdg7vue5S7H3kWrSYXTW6W/TGtJpf1qxahv2MilttfcmCV4qqOLAGDnCDcCAllF5V44SIorNSsFUlBTiEWswxR6EyUKiX129YgKyeTJs0a4+fv5+iSXIZCoaBhizgeHv0WSqUbmtxM+2OFBXmsX/U9unbjMLef4MAqxRWdXsstn56ukpNQdlFHjxzD08uLOnVqU5AhA/o7E6Wbgvqtq6M3FlO3Xh3Uf0/MIMomOLwmA0e+ispNTX5Ohn15kVZjC+bWozF3nOTACsVlcs7I7Ws3SELZBel1es6fu4Cfnx+1akeSl17xo3iJK1MoFUS1qo4RA2ERYXL9+AYFhkYw4OnXcHf3JC/rn46MxYUFrF+1CF3sSCytnnBgheIyJ38Ai7TclZeEsgtKSkzCarUSHhGOQWfCUGxydEnib3VbhmNxMxESGuzoUiqNgOAwBox8FU9v3xLBrCvSsmntSsxdP4aabR1YoSjhxCowyIlCeUkou6CTJ06hUqmoXiOCgkzpVOEswusHovZSEhwSJGfIN5l/UCgDnpqAt68fuRmp9uWa3Cx2bF6PadBq8A5xYIXC7vwWUHk4ugqXJaHsYsxmMyeOncTPz4/q1atTmKdzdEkC20xPEfWD8Avwk0C+RfwCg3ngyVfwruZXold20rkTnD51CtPAFaCsnOOGuxSTTiYTuQESyi4mPS0dvV6P2l1NaFgI2hzp5OVobiol9VrVQKVWSSDfYtUCguj72AsYDfoS9zEf3LWFXGUo5q4fObA6YXfqJzDpHV2FS5JQdjFJickABAYGYNSbMBnMDq5I1I4JR+2hss8RLG6t4PCa9Bo8ktysVMxmW38Kq9XKlg0/YGz2KNYmAxxcoSBxB5jkhKE85F3ExZw7ew4vLy/CIsLR5sovvaMFR/oTEOaLWyWdbtFZ3RbThrgu95GVmmif/cugK2bTulWY7/kS/Gs5uMIqLmUvqL0dXYVLklB2IWazmcQLSXj7+FC9egSFudI85Eievu7UbhaOyl3mdXGE+G73U+e2ZuRk/DMRS25mKsf/2oepzxcOrExgLIL8REdX4ZIklF1ITnYOZrMZlcqN8PAwtDnS89qRolpWx00lf0KO4qZS0XPwU3h5+6LN/2fa0qP7d6APaQ6N+zuwOsGFLY6uwCXJO4oLyUjPxGqx4OXlhdrdHZ3W4OiSqqzgSD88fTykY5eD+VTz555Hn6dIq8FkNAK2KR+3b96AqfdscPd1cIVV2PnNoJf7lctKQtmFJJxLQO3uTlhEGAW5MgGFoyjdFNRqGo5KLdeRnUFErShu796PnIxk+7LMlAskJydi7jDRgZVVcUk7QT60lpmEsouwWq0knDuPj68P4eFhFOXKWbKjVG8QIj2tnUyrO3vhHxSGVpNnX7Z/xx9Y2z4D/rUdWFkVln3a0RW4JHlncRHaAi3aAi3u7u6EhoRSlC+DhjiCyt2NiPpBci3Zyajd3ek+YASF+blY/h53ubhQw4m/9mLqPtXB1VVhaQcdXYHLkXcWF5GRbpslR6FQ4Bfgh04rPa8doUa0DOXorCLrNaRxqzvIzfxnGM6jB3ZiqXsXhDZxYGVVWMImsMhYCmUhoewikpKSUSqVqNVq1GqVTELhAGoPFaG1A1C6yZ+Ns7qjxwNYLVZMRtvlHbPJyMkjBzDFjXVwZVVU5jEwSP+XspB3FxeReCERLy8vAgIDKNJK07UjVJezZKfnFxRCuy73lpi04vSR/dB0EHgFOrCyKir7FGB1dBUuRULZBVitVjLSM/H08iQgwB9dgXTyqmgKpYKQSH85S3YBLTv0wN3TC4Pe9uFVV1xISsJJLK2edHBlVVD2KRnZq4zkHcYFFBcVo9frUalUBAQGoC+UpuuKFhhRDazyid8VeHh60erOXuRnZ9iXHT20H0u7MTKLVEUzaG1fotQklF1AXl4+SqXtfr+gwCD0MmhIhQuLCsRN7kt2GU1adwDAYrZ1MsrNTKVAWwiN7ndkWVVT7jlHV+BSJJRdQF5eHhaL7SzNP8CfYul5XaHUnip8/D0dXYYog2oBQTSMjSc/55+z5SOHDmCMH+fAqqqo9MOOrsClSCi7gKyMLBRKJW5ubnj5eKIvlDPlihQS6e/oEkQ5xN7RDaNBb59FKunsCSyBDSCkkYMrq2LSD4JJOqeWloSyC0hNTcPTwwM/fz90hXq5tFnBwuoGSgcvFxQeGUVYzSgK/x7ly2q1kJRwGkv0PQ6urIrJOimhXAbyTuMCMtIz8PTyxNfXB52cJVco30Av3NTyZ+KKFAoFbTvfTVFBvn1Z0oUETNFyXblCZZ8CpUxvWlrybuPk9HoD2gItarUaL28vTAYZHaciBUf6o5Bxrl1WVKMWePr42m+PSktKwK16S/Co5uDKqpD8C6DycnQVLkPebZxcfl4+CqUChUKBl5cXJr3F0SVVKX6hPvae78L1qNRqopu3Q5ufA9hG+MpNOw9RXR1bWFViMYOp2NFVuAwJZSen1Wrt98f6ePvImXIFUropcPdSO7oMcYOiGsViNhnt31+4cAFTdF8HVlQFFeddfx0BSCg7veKiYnvHLm9vb4w6GTikovgEemExS8uEq6tRpwGgsM8elXLhDDTo49iiqpqiTEdX4DIklJ1cYWGh/ZYOL28vjHoJ5YpSLchbel1XAh5e3tSoc5u9w1dBXjZGixXCYxxcWRWiTXN0BS5D3nGcXH5+Pmq1rQnVy8sTk16aryuKv1xPrjSiW7SjuLDA/n1mSiLUaOvAiqoYTZKjK3AZEspOTpOnQaW23U7g4eUpZ8oVyFtG8ao0IuuVHDAkJy8fS4jMsVxh8hPBIpeCSkNC2clpCgpQq9S4ubmhVCgwm+QXuyJ4VfOwXzYQri8orAZePtXst0YV5GVjCmnq4KqqkMJ06YFdShLKTk5boEWlVuHl7YVBb7z+E8RN4RMgZ8mViUKhoPZtTSnWagDQ5GajCIl2cFVVSGEGWKSVrzQklJ2Y1WqlUFuISqXCw8MDo0F+qSuK2kOFUgYNqVTCatTBoLedrRXk5+DmX1NGmqoohRmAtDyVhvxGOjG93jaYvlKpRKlUYrXIL3VFcfdSo7hGJ6/c3Fwef/xxALKyslAqlQQFBQGwbNky3N3dS6yfl5fHunXreOihh665X5PJRHx8PHv37rUvy87OZvjw4Vfc18qVK3Fzu7lTSi5btowuXboQHBxc5udeuHCBF4lH9mEAACAASURBVF54AYvFgtls5vHHH2fAgAEADBo0CI1Gg4eHBwALFy7Ez8+Pt99+m/379wNQVFREUVER27Ztw2q18sEHH7B9+3YsFgudO3fmlVdeKfHaA6SmpjJo0CDGjbv2DFABIeGA7WdqMZswaHPxDIyC7NNlPk5RRnoNl157cW0Syk6suKgYhcL2i+zmprRP3yhuvesNGhIYGMiaNWsAmDlzJt7e3owYMeKq6+fn5/N///d/1w3lKwkODrbva9q0aQQGBpYIpZtt2bJltGjRolyhXL16dZYsWYK7uzsFBQXcfffddO3alcDAQABmzJhBdHTJZuM333zT/v+vvvqKpCRbT92dO3dy8uRJfvzxRywWCwMHDuTgwYPExsbaXw+Au+++mx49ely3tmoBwfa/J4CC3Ew8gxtKKFcEkw4U0vJUGvIqOTG9/p95k5VKNzlTrkDunuX/vDp37lzuuece7rnnHhYuXAjAlClTSEhI4L777uOTTz5Bq9UydOhQ+vXrx7333sumTZvKta/PPvuMJUuWALZwe/LJJwHYvHkzEydOBGDTpk0MGjSI+++/nxdffBGdztbZ6eDBgwwZMoT+/9/efcdHVeb9/3+dqWkz6b0XSgidIL0XC6iAiGVvda1rd73dW73dddet6M9yr1931V1cC+ruIgpSF5UmSpUuNdISQhJSSZkkU87M74+BwUiRQMiZST7PxyMPyJmZM58MYd5zlXNd06Zx7733UlVVxcKFCykoKODRRx/l+uuvR1VVZs6cyTXXXMO1117LK6+8ct56TCaTr5fA4XDg8XhaNWFuyZIlTJ7s3cVJURTsdjsulwu73Y6qqmd8UNi3bx92u53evXv/6LmtkTG43W5fPTUn6iBaxpXbhYTyBZOWsh9TVZVTH+z1eum+bk8G88V1Ce/cuZNFixYxd+5cVFXlxhtvZODAgTzxxBMUFhb6WnhOp5PXX3+dsLAwqqqquOWWWxgzZkyrn2/gwIF8/PHH3HTTTezfvx+Xy4Xb7WbLli3k5+dTUVHB22+/zezZswkKCuIvf/kL77//PrfffjvPP/88b775JhEREcyfP5/XXnuN5557jg8++IA//OEPdO3alePHj7Nu3TqWLl0KQF2dd6LUsmXLOHz4MA888MAZNRUVFfHQQw9RVFTEr371K19XO8ATTzyBXq/nmmuu4b777mvxuMOHD1NTU8OAAQMAGDRoEHl5eQwbNgxVVbn77rtJTU1t8ZilS5cyadKkC3qtzEHBBIeG4XI6MZpMNDXb8QRHS6dqe3A1SShfIAllP+Zd4tH7liFjyu3LYLy4UN68eTMTJ04kONi7K8748ePZsmULw4cPb3E/j8fDSy+9xJYtW9DpdJSWllJdXY3Vam3V8/Xp04dnn32WmpoaLBYLiYmJ7Nu3j82bNzNjxgy2bNnCgQMHuOmmmwDvh4ErrriCgoICDhw4wB133AGA2+0mOTn5jPNHRkbidrv51a9+xejRoxk1ahQAV1111TlrSktLY9GiRZSWlvLwww8zYcIEIiIieO2114iPj6e+vp4HH3yQlJQUrrnm9HKXS5Ys4eqrr/Z1MR86dIjS0lLWrFmDy+XijjvuYMSIES1axUuWLOHNN9+84NcrOj6Z6ooyjCYTLpcDt8lC247Ii7NyNcukugskr5IfU7/X1abX62VMuZ2c2pXrYlxoV+2CBQuor69n/vz5GAwGRo4cicPR+r2yg4KCiIqKYtGiRQwYMIDExETWrl1LRUUFqamp7Nq1i9GjRzNz5swWj9u5cye5ubm+7vVzMZlMzJs3j7Vr17JkyRLmzJnDrFmzLqi2xMRE0tPT2bZtG2PGjCE+Ph4Ai8XCNddcw7ffftsilJcuXcqLL77o+37ZsmUMGDDA9wFn+PDhbN++3RfKO3bsICQkhC5dulxQPeCdgV129BBYI3A5nXhCZQvHduF2wUX+n+pspD/Bj7m/twKOTqfz7RYlLi+D6eI/AA0cOJDly5fT3NyMzWZjxYoV5OfnExoais1m892vvr6e6OhoDAYDa9eu5fjx4xddb35+Pm+//Tb5+fnk5+fzwQcf+IJrwIABrF+/3jd5ymazUVhYSNeuXTl27Bi7du0CvOO/Bw4cAGhRa0NDAzabjbFjx/L000+zZ8+e89ZSWlrqmwtRXV3Njh07yMjIwOFwUFNT43uuL7/8skWY7t27F1VVycs7vaBHUlISmzZtQlVVHA4H33zzDdnZ2b7bFy9e7Bt/vlBh4VG4Tu4YpTodeEwSyu3CrSKzry+MtJT9mFtV8Zy8tk+n1+GRxbz8Xu/evZk0aZLvMqBbbrmFbt26AdCzZ0+uvfZaRo0axZ133sn999/PtGnTyMvLIyMj46KfMz8/n9mzZ9O7d29MJhOKovjGZePi4vj973/Po48+itPpRFEUfvGLX5Cens6rr77KH/7wBxobG3G73dx9993k5ORwww038PTTTxMUFMTrr7/Oo48+6mvFP/3008C5x5T379/Pyy+/7OtpeOihh8jMzKSuro677rrLN+Y9cuRIpkyZ4nvc9yd4nXLttdeyadMmrr32WgDGjh3LsGHDAO8H1s8++4wPP/ywVa+VyRwEJz9wuT1uPIpszdkuPG4ZU75AikfWEvRb+/cVMO+j+cQnxJOb153MxByO7pYt0C43g0lPn/E5skNUB7Rv23o+++jvxCalk9YljytSjZjmTtW6rM7hOYmaCyHvOn7M7Xb7OnwURSe91+3E4/ZIT1sHZTCaZGxTK9LVd0EklP2YW3X7FqZzq+p5V5gSbcfj8aBIKndIiqL4VntUUGSeRnvRG+W1vkASyn7M7VZ9v8eqqsqQTDtxq9JS7gyCQkLR2Uq1LqNzMIbKhhQXSN7m/ZiqulFOfqxXVRWdXpKivbhV+VTf0YWGhmGoK9S6jM7BJKF8oSSU/Zher/O12FTpvm5Xsm91x/T9ea2WsBCoK9awmk7EGAoeVesqAoKEsh/TGwycSmWXS5XZwO1IdckbSEdkb270BXNIaJiEcnsxhcqY8gWSd3k/ptfrfRNFnU6nhHI7am5o/epawv/VVlX4troMDguH+mMaV9RJmMKQ/ZQvjLzL+zGDXu+7fMPpcGAwyD9Xe2mobpJlTTugE1XHMZjMgIIpLArqJJTbhTFUMvkCybu8H9Pr9b5fZIfTif4iN0kQrddU33xyQxDRkdRVV2A0mTEHh+BxNHo3ShCXnylUrg+/QBLKfsxgNJ7uvnY4L3rnItF6jXV2dDKxrsOpO1GF0WQmzBqJKq3k9mMMlWU2L5C8Sn7MaDL65kY4nU70Bgnl9uJsdsm8lA5GVVUaG+owGE3EJqaiP/q11iV1HmaLbN14gSSU/ZjJeHqxfI/H420tmySY20uzza51CaINNdnq0Z3cljMpMR79kZVal9R5WFLAEKR1FQFBQtmPeVvKp5trjbYmTMGyq017sdXIeGNH0lBbg/cSQ4WY5Cwo+krrkjqP6C4ypnyBJJT9mNFoavG9zWbDFCxdQO3FdqJJFhHpQCpKCnF73IRHxeJpOgH1ssRmu4lI17qCgCGh7MdMJiN6vR5V9S5k0VDfgClIWsrtpa6yUT7cdyBH9u8gKDiUuOR0lMIvtS6ncwlL1LqCgCGh7McURcEabsXpdALQYGvAGCQt5fbiaHLiaJb1ejsCt9vN0QN7CQkLJykxAUPhCq1L6lxCYrSuIGBIKPu5iIhwnA5vKNtsNgxB8k/WnqqOnpDrlTuAmooynE4HBqOR2KQMKFyjdUmdR1C41hUEFHmH93ORUZE4TraUbQ2NmKSl3K6qS+vl0qgOoPzYEfB4iI5PRnE2QM1hrUvqPCzJskhLK0go+7nIqEhcztMtZZl93b6aGxy4nLI5RaArLPgWY1AwXXJ7od/2d63L6VysKbIZRStIKPs5i9Xi+3ujrRFzsFnDajqnquJaWQc7gHk8HgoLvsUaGUNaTh66He9pXVLnYk2RhUNaQULZz4WGhqCcnAKsqioupywg0t5qSurwSCgHrIqSIpobbWTl9sVTvFE2oWhvkdlgDNG6ioAhoeznQkPDWnxfX9tAUJjpHPcWl0NjnR23XK8csPZuXYdOr6db9x4Yt72pdTmdT8pg0EnUXCh5pfxcaFhIi1W9qqqqCLHKcnXtrfRgpSwkEoCcDju7v/mSpMxuhMfEw/5FWpfU+cT10LqCgCKh7OfMZnOLBUSqqqoIsshkr/ZWUXhCJqsEoKLvduNwNNO91wCUnR+A6tC6pM7FGALB0VpXEVAklP2coihERUdht3s3R6iuriHIKt3X7c2teig9WC2t5QCzY91yQq2RdOnRB/22WVqX0/nE9gBno9ZVBBQJ5QCQnJJEU5P3Or+a6hpCLcEaV9Q5lR+u1roE0Qp11ZUUHdxL/2ETUI5tgOPfal1S5xPfCxSZmNoaEsoBICk5CafD2+3mdDqx2x2YQ6ULu72pLjflR6pRZYWvgFCwcxNGk5leAwZjXPW/WpfTOSX0B1Oo1lUEFAnlABAVHYXue7MXa6pqZLKXRsoOViN7VPg/p8PBtrWfM2jMZO/mE9JK1kbKFbJlYytJKAeAqKhI3G63bxZ2ZVUlQRYZV9aCy6FSKeth+739Ozagqi76DBourWQtRXfTuoKAI6EcAEJCQwgODsbl8u5YVF1VLTOwNVRSUCUTsf2Y02Fn3WefMHbyDHTffgiV+7UuqXMKiQGD9Oi1loRyAFAUhcSkBJqamgDvDGzpvtaO0+6ieG+5zMT2U3u2rCUoJJSsbj3Rr35W63I6r5RBshHFRZBQDhDJqSk0n5yBXV9Xj8Gol+U2NVR+pAZHo6PFwi5Ce022BtZ/MY9JN/4U3Ze/gaYarUvqvDLGginsx+8nWpBQDhBx8bG+APB4PJSXVWCJlvVktaLTKzQ1OHCfHFIQ/mHLl0vpP3QcVscxdN/8VetyOrecq0AnDYfWklAOEFFRkS2+P1ZyjLBo6cLWgiU6hF5jswkPrkO/421w2LQuSQA1FWWUFh2k/+DhGOffKiuwackYAtE5WlcRkCSUA0RkVFSL5TZLS8oIk5Zyu9LpdaT3iqfLFcmYNs5E/2oqLH0YThSCW/Zc1pKqqny9bC6TbroTw8I7oaFM65I6t5RB4GzSuoqAJKEcIAwGPWkZqdgavK2ymuoaTGaDjCu3E2tsKL3HZhEZUo3+tWxY9RvvDW4XfDQNXHZtC+zkdqxbTl6fAZi/mw8FS7QuR2SMke0aL5KEcgDp0jXHNwNbxpXbh96gI6NvAjkDkjB+/WuMr3eB2qKWd6rcDwvvkW5sjRwvPkLl8WJSE6LQf/641uUIgC7XgF4u27wYEsoBJCk5GeV7q+PIuPLlFREfRq9x2UQYS71d1V8/f+477/oXrHtJgrmd2Zub2LD8U8ZcPQ3jx9Olx8If6I0Q11PrKgKWhHIAiY2LRafTybjyZWYw6ckakERW3wSMy/8b45t5FzZGufo5KFgMDtkVpz14PB42r17CldP+C8OS+6B8t9YlCYDEAXJ98iWQUA4gMq58+UUmWug1Jotw9yH0/5cAm15r3Qnm3wbl30qLrR0c3r+TfoOGY1r1FMqej7UuR5ySNV5W8roEEsoB5ofjysfLKrDGSGv5UhnMenIGJpPROxbDsgcwvNUfmi5iq0bVCR9c5W1Zy4zsy+ZEZTmx0VGYN72EbtvbWpcjvq/XrWAwa11FwJJQDjA/HFc+WlSEJVZC+VJEJVvpNSYbi303hpfjYOtbl3bC5hPw3hhw1Mu1spdB3YlqFFcjIXveQ7/+Za3LEd9nSYSIDK2rCGgSygHmh+PKhUeKiEywoOhke7TWMgYZ6DoohfS8aAwLb8PwzlCw17XNyWsOw9vDoakKVFn1q63Y6mtxNlQRemQJ+lWyrrXf6XYdeGRN+Eth0LoA0TqnxpXLSo5jDbfS1NTEiZparDEh1JbLzN8LFZMWTlpePJ6irzDMmgTOyzA5q3w3vNkP7voKwhKlS+8SNTfasNdVYK3YgOGzR9r9+V1ueGkd/GMbFNVCbAjc2AP+76rT98n4MxTWtnxcfCiU/eL853ao8PzXMHsHHKuHZAv8pBc8MwLMJ9+l6+1w90L47CDkxsDsqdA1+vQ5apqg21/gPz+BAUlt8zO3Wu/bwRSq0ZN3DBLKAah7bncOHTiCNdwKwKFDh8hIzJZQvgCmYCNZ/RIJDlXQf3wj7Jt/eZ+wrhj+1h9uXwEx3cEYfHmfr4NyOuzgaiT8wBzvzk8aDAvcuQBWHILfjILuMXC0DvZUnHm/W3vBI1ec/v5C5mE+vRze3Ax/GAv9EmBrKfxqFZxohlev9t7nj19BQRV8NB3e3QE//RTW3X36HM+thsldNQxkswWSBmj05B2HhHIAyszKQME70UtRFAoPF9GvX1+KdpWDDGGeU2xGBKm5cXDwc/SvTwHV0T5P3FQD/xgKNy+AtGHSkmgl1eVCcTVjXvYwyp6PNKlh2QH49y7YcT/0iD3/fRPDYHBK687/z2/hgXz47yHe78dkelvMH357OpSXH4JfjoArc6BvAiS8DDYHhJpgbwW8vxP2PNT6n63N5FztvepAeoQuiYwpB6DwiHDiEuOw2bwt44aGBhoabFiiZMLX2ZhDTeQOTycl24L+X5PQ/+ua9gvkU1zN8M9rYPdccDS073MHMGezDU9jJYb3x2oWyABvb4OxmT8eyBfL6YbwH1xFFBHUskPAoULwyUWyQoynjwE8/hk8NQwStNwpsdetEGTVsICOQUI5QPXp29t3vTJ4u7DDE6QF9kMJWVHkjcwguGQZhpci4eBn2hXjVmHBnbDqOe8CI26ZEHM+DlsNVO7D8Pd+ULpV01o2HoOuUfDwUrDOhJA/wrQ5UFJ/5n3f3gam30P48zD9Iyg88ePnv6cf/G0LrC2CBgd8VQhvbIaHv9cNPiAJZm2FqkZ4dSNkRUJkMCwpgO+q4eeD2+7nbTW9EbInaFhAxyHd1wEqKzsTj8fj68I+criQnj17Urz7LINcnVBQmIns/kmYDA7074+FwjVal3Ta+pfhwH/gpnlgTZHu7LNwNdag+24JhsX3+MVCLGUN3nHcPvHw7+neSVdPLoepc2DD3XDqKsXru3m7rlOssLcSfvsljHgHvn3gzJbw9z0/HppcMPyd08cezIdfjzr9/W9GwfjZEPMihJngkxngVOGJz+GlCacnhGkiY7T3Gn1Z7vqSSSgHqMioSKJjomhqaiIkJIS62jrsDjuhEUHYTnTeJe4UBRJyoknMiUbZ829082/3z0s0KvbAG71g9O9g8KOyo85JHmczHmcjhiX3e7v6/YTH4/1acDOcWtk20QKj3oWVh2FclvfYqfFfgBHpMDQV+r4J72w/f0v2xXXwwU547WroHQ87yuDZVd7n+t0Y730yImDfw3Coxhv6IUZ4ZT0kW2Fqrrd1/dBSKG2A6bneWtptsb/8+8Fkaacn69ik+zpAKYpC7769qa87PT556OAhIhM773+MYKuZvFGZJCTr0b87DN28//LPQD5FdcKK/4V3x3hnaV+Oy7ICiMfegHv/YnR/6eZXgQzebuJe8acDGWB4mjf0zjYD+5SecdAtxjub+lwqG+FXK+GF8d7u6pHp8Mgg7/czv4bvX1Rh0Hkvgwoxeh/3p6/gz1eC3QUzPoZfjYTvHoGtZfD3LZf+c1+QoHDvJC+dxElbkFcxgOV0yeb7060PFBwkKsXa6RYSUXQKyd1iyB2Wjnn/+xhejoHijVqXdeGObYLXusLOD7wbw3e25TkdNjx1xShzb/ReptZYqXVFZ8iNOftxjwcu5L+bcp77HKrxTvTqm9DyeL9E77XR5xqT/uUK73XSveJhX6W3K3tGnneC2G29YdWRH6+rTfS8pfP9zl5GEsoBLCY2BqvVQnOTt7u6vr6e6qpqIhM6T2s5NCKInqMziY13o5/VH93i+7Qu6eI4m2DRz2DWQDiyunNsAemy43E04v7qeZRXs+HAMq0rOqfJXWHncW/r9JQ1hd4w7ZNw7sftKof9lTAg8dz3SQ/3/vnD1vSWEu+fGRFnPmbncfhkL/x+7OljDhXUkx1DNkc7Xso96FEwazntu2ORUA5giqLQu18f6upOLyG0d89eolI7figrOoWUHrF0G5JG0M43MP5fAhzfqXVZl658N8weDx9eDWXbwd7Q8dbPdti8reNvXkd5rQu6r/7Q/peotdJ9AyA6GK79Fyza772u+Lb5MD7L240N3lnQt3wCH+6EVYfhjW/gyg8gLRx+2vf0uWbvAMPvTreA48NgSnd4ajm8usH72P9bD0+fbAnHnmUe4GPLvF3Vp/ai6Rbj7dJ+8gtvHX/9BkZnXNaXxCuuJ4SntcMTdR4y0SvA5eZ156vVX/lmYRcVHmXo8KGYQ03Ybf79RnexwqKCyeqfhK7pOPo3hkPVfq1LanuFX3mX6MwYDeOf9775GYIDe9zOXg9uJ56v/z+UzW+gtNU64+3AaoaVd8Cj/4GbP/GOJV/fDf7vytP3SQ33jv/+/DPvSlzRwXBVDvxpnPfxp7g9oHparvPz3hT43Zfw/zZ5L7NKtsDPBsCzI8+sZd5eKK2HhwaePhZkgH/dAA8u9S4DOr0H3J/f5i/DmfJ/BnpTOzxR56F4PB3tY3jnM/udD6iurPYtuzlwUD7RIfEc2+d/Y3OXQqdXSOkRR0yKFf36F2HFM1qX1H5Sh8Dgx6HrZHA7wRxAizTY670bc6z6Dez6l3eCmwh8eiM8WRlYv4sBQFrKHcAVgwYyf+58Xyjv37ufa6d0paSgCo+7Y3zmssaEkNkvCaXhKPrX+kNtodYlta+j671fxhDvTjz590PKIO8EG3+7ztmtelct0xm8i7VsmeXX48XiInWdrHUFHZKEcgeQlZOJyWzC6XBiNBmpq6vnRM0JIuLDqCk9y5JDAURv0JGaF0dUogX9V8/Bmj9qXZK2nI2w69/er9BYyLvJG9CRWd4WqClMmy5ulwNUu/drzyewe463C94t21Z2WIN+Lq3ky0C6rzuIVctXs2nDJmLj4gDIzM6kb69+HNh4TOPKLl54XBiZfRNRThzA8ME4qC/RuiT/ZU2BlMGQPhIyxkBMV++Mbr2p7XemUp3eDwc6PeiMUHMI9s6DPR97J6eJji+6C9y/Q3Y9uwykpdxB9OrTk/VrN3xv56hChg4bjDnEiL0xsMbw9EYd6T3jiYgPRb/yKVj/itYl+b+6Ym8o7vnY+73eCPF9vEGdNR7iekFwpHd7Pbfqne3sUQHFG9yKzrvQiscNeE5ed+rx3q7owRAEDaXeGe7FG7x/lu+CE4c73uxw8eNG/Mo7PCHanLSUO5DZb79PdXUNVqu3Syn/igHEWRM5GkDrYUcmWMjokwCVe7ytYz9cSCLgBUVAaNzpr7B4bzCrTm9Yu53ey5aaa7zbTjbXQN0xv79sSbST0Dj4+RFpJV8m8lGnAxk4eCDzP/7UF8q7vt3N9BndKT1Qjcvu3yvuGEx60nsnYI0OwvDZw7Dlb1qX1HE1n/B+VRVoXYkIQJ4h/41C51o1sD1JKHcg2TlZmIwmnE4nRqOR5qZmDnx3kPjMaL++PCoq2UJ6r0Qo3YLhlSu9gSGE8D+mUJSBD4LxPFteiUsSwCsRiB8ymUzkDxpATXWN79i3O74lJi0CvdH//qmNQQa6DEohPS8Gw+K7MLw9SAJZCD/m6X+vd/6BuGzk1e1gBuT3BwVcLm93tc1mo+hIEXEZkRpX1lJ0aji9RmdhsW3D8FI07JitdUlCiPPR6VFG/K//XRffwUgodzAWq4X+/ftRU13lO7Z9+w7iMiLQ6bX/5zYFG+g2JJW07pHo59+M/r1RnX7LQiECQo8bvUu9istK+3dp0eYGDs7H7fbgdnu3jKmrraOkpJTYU9vRaCQ2PYKeo7MIqVmP4aXI05fvCCH835jfeS+pE5eVhHIHFBkVSc/eeVRXV/uObd+6nbisSE32WjaHGMkdlk5KFwv6Oddh+HAiuOztXocQ4iJ1nwJh59mjUrQZCeUOavDQQTidTl9rubq6hsqKSmJS27e1HJcVSd6oTIKPf4HhxUj4bmm7Pr8Q4hIpOrjqz9JKbicSyh1UbFwsubndqak5PZt529btJOREodNf/tZyUJiJHiMzSM4IQf/hRPRzpsg6yEIEIE/v2yA4SusyOg0J5Q5syPAhOB0OTi3aVlFeQVlZGXFZl/E/mAKJOVH0GJFBcOEC79jx4ZWX7/mEEJeP3gQTXpBWcjuSUO7AEhLjyczO5MT3WsvfbPyG+MwIjOa2Xzcm2Gqm56hM4lOM6N8dgW7eLSfXUhZCBCLPwIdQjG1zCdS7777LoEGDCA0NxWq1MmrUKBYuXNjiPm63m4ceeoj4+HgUReG5554DYMGCBeTm5mIymcjIyGiTei7U6tWrURSFXbt2tcvzSSh3YIqiMGLUcJqbm31jy/X1DezfV0Bit7ZrLSsKJHWLIXdYOuaCf2J8ORqK17fZ+YUQGggKhzG/BXPYJZ/qgQce4J577mHQoEF8+umnzJkzh4yMDK6//npeeOEF3/3mzZvH66+/zsyZM1m/fj333HMPqqpy++2306dPH1auXMn8+fMvuZ7W6N+/P+vXryc7O7tdnk+W2ezgklOS6N6jG4cOHCI6JgaA7Vt3cOPNOYRYzTTWXdos6JDwILL7J6F316F/K1+27hOig/CMeg6lDXaC+vTTT3nzzTd54403uP/++33Hr776ahISEnjmmWeYMGEC/fv3Z9++fURGRnLXXXf57ldcXExdXR233norw4cPv+R6WstqtTJ48OB2ez5pKXdwiqIwZtxoXC7Vt8qX0+lk6+ZtJOfGXPx5dQopubF0H5qGadffMb4SJ4EsREcRkY6Sf1+b7AT16quvkpOTw7333nvGbc888wwWi4W//OUvjB49mmeffZaamhoURUFRFN59911SU1MBuP7661t0abvdbp5//nlyUHCxDwAAHF9JREFUcnIwm8107dqV9957r8X5R48ezfTp0/nnP/9JTk4OVquVq6++muLi4hb3mzlzJjk5OQQFBREfH89VV11FWVkZcGb39ahRo5gxY8YZP8svfvEL0tLSfHN4mpubefLJJ0lNTcVsNtOnTx+WLv3xq0+kpdwJREVHMWjoFWxat4m4hHgA9u8rIK9nHhHxYZw43tCq84VFBpPVPwmdvQL93/pAxe7LUbYQQiOeq/4fis54yedxuVysX7+eBx98EL1ef8bt4eHhjBkzhjVr1rBw4UJeeeUVPv74Y5YtWwZAZmYm8+bNY9q0abz00ksMGzaMlJQUAB555BHee+89fv3rX9O/f3+++OIL7rrrLqKjo5k8ebLvOTZu3EhJSQkvv/wyTU1NPPbYY9x3332+gJw9ezZ/+tOfeOGFF8jLy6OqqoqVK1dis9nO+jPdfPPNPPHEE9hsNkJDvePtHo+HuXPnMmPGDBTFe3XL9OnT2bRpE7/97W/Jzs7mo48+4rrrrmPz5s307dv3nK+ZhHInMXjoYLZt3o692Y45yIzH42HD+o2MGDGc2vKGC9qnXqdXSM6NJTY1Av3GV+CLJy9/4UKI9pVzFUrWONBfeihXVlZit9tJT08/533S09NZtmwZPXr0ICUlBYPB0KK7uF+/fgB069bNd/zAgQO88cYbvPPOO9xxxx0AjB8/ntLSUn7729+2COW6ujqWLFlCZKR3/f+ysjIef/xxmpqaCA4OZtOmTUycOJEHH3zQ95hp06ads97p06fzyCOPsGjRIm6++WYANmzYQFFRke/7FStWsGTJElavXs2oUaMAmDhxIgUFBfzxj39k7ty55zy/dF93EiEhwYwZP6bFKl/Hio9xovYEsekRP/p4S3QIvcZkE22tR//XbhLIQnRExmA817/j95tOrFixAp1Ox9SpU3G5XL6vcePGsX37dlT19P7xAwcO9AUyQI8ePQA4duwYAH379mXp0qX85je/YdOmTS0eezaxsbGMHTuWOXPm+I7NmTOH7Oxs8vPzAVi+fDkJCQkMGzbsjPo2b9583vNLS7kT6d23JxvXb6ShvoEwi3dG5YZ1G5h03SRqShtw2s9c3ENn0JHaI5boJCv6r/8AX/62vcsWQrQT95g/oAuyttn5YmJiMJvNFBYWnvM+hYWFJCcnt+q8lZWVqKpKePjZVygsLS31dXNHRLRsdJhMJsA75gtw1113UV9fz9///nd+97vfER0dzQMPPMBzzz131i538HZhP/jgg9TV1REWFsbcuXP56U9/2qK+srIyjMYzexvOdc5TJJQ7EYPBwMSrJ/Cv9/9NaFgoiqJw4kQte/fsIyUvjcNbS1vc3xobSla/RJTaw+hf6wN1xec4sxAi4MX3Qhl4PxhD2uyUBoOBIUOGsGTJEl566SV0upads3V1daxevZqpU6e26rxRUVEYDAbWrl17xjkB4uLiLvhcOp2Oxx9/nMcff5yjR4/y4Ycf8stf/pLk5OQWs8W/b+rUqTzwwAMsWLCA9PR0SkpKuOmmm1rUl5yczKefftqqnwsklDudrOxMsrKzKC0pJTLK26WzY9sOsqZnEplgoaasHr1RR1rPeCLjQ9Gv/iWsfVHjqoUQl5Wiwz31A3T6oDY/9WOPPcbUqVN56623uO+++1rc9vzzz1NXV8fDDz/cqnOOHTsWVVWpra1lwoQJbVZramoqTz/9NO+88w579uw55/0iIyOZOHEic+bMIT09ndzcXHr37u27fdy4cbz88suEhYXRvXv3VtUgodzJKIrC+CvH8Y8338blcmEwGFBVla++/Ipx48eh6BXSe8ZD1T70fx4HtnKtSxZCXGbu/AfQRWXBWVqdl2rKlCncf//9PPTQQ+zZs4fJkyfjcrmYM2cO7777LjNnzqR///6tOme3bt24//77ufnmm3nyySfJz8+nubmZ3bt3U1BQwFtvvXXB5/rZz35GVFQUgwcPJjw8nFWrVvHdd9+1WNTkbG666SbuuusuwsPDz/hQMWHCBK688komTJjAU089RV5eHnV1dWzfvp3m5mZmzpx5zvNKKHdCcfGxDBs5hK/XrCMh0bsd2/GycsrKysjolYz+88fgm9c1rlII0S4siSjjnwfTpa/cdS6vv/46gwYN4o033mDWrFnodDr69+/PggULuO666y7qnH/961/p2rUrs2bN4te//jVWq5UePXpw9913t+o8Q4YMYdasWfztb3+jubmZnJwcZs2axZQpU877uOuvvx6DwUBlZaVv1vUpiqIwb948/vSnP/HnP/+ZoqIioqKi6Nu3L4888sh5z6t4PBdyMYzoaBwOB2+98Q+cLhcWiwWH3c7h/bv55XNPY/3PnXDwc61LFEK0A/cti9DlTPRuPiE0J5dEdVImk4nJUybRUN/AsaIjbFzzBe7GMtav+g+u6971rnsrhOjQPH1uR8kcLYHsRySUO7G09DSyM5M5sncrvbrE0q9fD6rKjnLkyGFcV76mdXlCiMspKhuu+SvKZey2Fq0nodzJTbnxBgb1y8RqOb3G7db1q3FmToRuFzfWI4Twc3oj6ox5KG2wtrVoWxLKnVxQSChX3XQf9TWVuE+uZONyOljz+SJc170DEedeHk8IEZjUMX9EF5kFuvMvZCHan4SyIK1LHn2GjKO6/JjvWNXxYr7duhHnjfPaZA1cIYSfyByDMvABlDbYJ1m0PQllAcCwq24k1BJBQ90J37G92zdS2WxEnfCKhpUJIdpMSDTu6XPQSSD7LQllAYA5OISrb32QxroTuJwO3/GvVyzBmTsDclu3DJ4Qwv+4rp+NYm67ta1F25NQFj5J6TmMmHQzlWXFvo26nfZmVi/7FNfktyAyU+MKhRAXyz3wQXTpI1AMZq1LEechoSxa6Dd8Itl5/ak+fnp8ubq8hB2b1+O8cb5czyhEIEofAeNfQBdk0boS8SMklEULOp2OCTfcTXBIGA11Nb7j+3d+Q7kN1Ctf1bA6IUSrRaSjzvhUxpEDhISyOENImIVJtz1CY30tTsfp8eV1K5fi6DIFd5/bNaxOCHHBTKG4bvkPillayIFCQlmcVWJaNiMn30rV8aOnx5cddlYs+QR14quQPlLjCoUQ56UoOKf+C09YEjqDXNYYKCSUxTn1HTqeLr2uoKqs2HesrqaCNZ8vwDVjHkR31bA6IcT5qKN/D6nDMIbKOvaBREJZnJNOp2P8tDsJtUZQV1PlO368+DCb163B9ZPPISRGwwqFEGfj6TEDz8CHMIZFaV2KaCUJZXFewaFhXP/Tx1GddppsDb7jh/btoODAIZy3LAW5xEII/5HYH/e1s9AHSws5EEkoix8Vk5DC5Nsfo66mAqfD7ju+feMayhoNuKZ8CIqiYYVCCAAi0nHdsgRMoSjyfzIgSSiLC5LeJY8JN9xFVdkxVNV18qiHtSuWUBfVF3XsTE3rE6LTC0vAdfuXeExW9HrZaCJQSSiLC5Y3cCSDxl1HZUmRb0a2W3Wxaskn2Hvegbv/vRpXKEQnFRyJ87aVOI1WjEEhWlcjLoGEsrhgiqIweMJUuvUd0iKY7c2NrFj0Ec4xM3H3/i+NqxSikzGF4fzJFzQbIgm2RGpdjbhEEsqiVfR6PeNvuJOEtGxqykt9x+trq/liwb9xTXwVT94MDSsUohMxmHHevJhGYxyW6AStqxFtQEJZtJrJHMTk2x4hOMxCbXWF73hdTQXLF87Bec2beGRXKSEuL50B5w1zaQrLJDw+VetqRBuRUBYXJdQSzpQ7n0BRFOpPnL6G+UTVcVYsnINr8j+g27UaVihEB6YoOK97j+boPlhiJZA7EgllcdGi45O44d6ncKtu6murfcdrKstYsXguzutmQ5erNaxQiA5I0eGa9HfsySMJi02VS586GAllcUliE1OZft9TqE5ni12lqstLWLnkY5xTPoTsCRpWKEQHojPgnPpPGlMnEBKdJIHcAUkoi0sWm5TGDfc+hbO5GVvdCd/xquPFrP7PfFw3fARZ4zSsUIgOwBCE86ZF1Eb2IzQmBZ1O3r47IvlXFW0iPiWDG+57CntzI7b6Wt/xitIiVv1nHs4b5uLpeZOGFQoRwMwWnD/5ggp9EhEJmbI4SAcmoSzaTEJqFtPueZLmxgYaG+p8xytKivh8wb9wTPwL7sGPa1ihEAEoJBrn7V9Sag8hNq0bBqNsw9iRKZ5TK0AI0UaOHS7gk1kvEGIJJyTM6jseEhbOhOtmELTv3+iX/w/Ir54Q52dJwnnbSoor6knt2kcCuROQUBaXRfGh/Sx452UM5iAs4ae3jzOZgxk7eTrWik0YFt4BqlPDKoXwY5GZOG9bxdHSCtK795Mu605CQllcNhUlRcz7x4uoqouI6Hjfcb3ewMgrryeWCgxzrgdHw3nOIkQnlDoU543zKDpaTGZuP5nU1YlIKIvL6kRVOZ++/TINtTVExiX6LuFQFB2DRk0kLSYEwwcTwFaucaVC+Ad3v7twjXuRI4cP0qVnvlz21MlIKIvLrrGhjoXvvcrx4kPEJKa1eJPpPXAE3bt3w/DvyVC2Q8MqhdCYTo864RUac6ZRWlJC1z5XaF2R0ICEsmgX9uYmls35G4f2bCM2Kb1Fd1x6Th6DRk1At+xRdDvf17BKITQSFI5z+jwqPFE0NDvp2mug1hUJjUgoi3ajulysWvA+OzeuJDYpHb3e4LstPDqOMVdPw/zdfPSfPSYTwETnEd0V5y1LKThwiMiEDJIyumhdkdCQhLJoV263mw3LP2XDF58SGZeIOSjYd5vRFMSICZOJMTZi+GgK1Jee50xCdADZE3FO/Sdb1q8hd8BwwqNita5IaExCWbQ7j8fD/h0b+XzuW5iDgrFERH/vVoVe+cPI7dkbw9xpULRWszqFuGwUHerwZ3ANfIwt69cwYNTVLT6gis5LQllopqKkiEXvv0ZDbTXRCSktJoAlpucwfNwk9F/+Bt2m1zSsUog2Zk3BOW0OlWoYpSUl9Bk6Qa5BFj4SykJTjQ31fPHxWxzas43oxFQMhtMrFoVZIxlzzTSCy7/BsPgeaKo5z5mECAC5U3FNmsU3a1cSnZQhE7rEGSSUheZUVeWbVYtZ9/knRETHERQS5rtNrzfQb8hosrK7YFj4UziwTLtChbhYxhBcV72GPfMqvlgwh6ETpxGXnK51VcIPSSgLv3F43w6Wfvg6Or2e8Oi4FrfFJWcwfPwkDAULMHz+c3DYNKpSiFZK6INz+sccLizmu/17GH3tTwi1hGtdlfBTEsrCr9RUlLH4g9eoOl5CdEJKi7E2o8nMFSPGk5wQh2H+rXB0nYaVCvEjFAX3oJ+jjniWlYs/IiGtC70GjZYlM8V5SSgLv+OwN7Pus3ls/XoZlojoM1oVKVndGTxqIvrtb6Ff9SyoDo0qFeIcorvgvPYdanXRrPl8ESMm3UxsYqrWVYkAIKEs/FbRgT18NufvNNrqiY5PbtHCMAeHMnTMlcSEGTDOv1WW6BT+QWfAPfRJ3EP/h6+/WIgHHcOumo7JHKR1ZSJASCgLv9Zka+DLxf9kz+aviYiJJygktMXtmd37kD90DMquf2JY9Qw012pUqej0kvJxXvcuFfVOVi6ey9CrbiSzex/ZUEK0ioSy8Hsej4eDu7fyxdy3cKkuouKSWrzRmYNC6Dd4JGmZ2eiX/w/Kjtkgv9aivQRF4Br3Au7c6axYNIcmWz0Tpt+DJSLqxx8rxA9IKIuA0VBbw4pP3+Pgrq1nLNEJEBWXxOCR4wl1VWNceh+UbtOoUtFZeHr/F+rEVyjYvZN1KxYzZOI0cvsPk8lc4qJJKIuA4na72bt1LV8u+hCX00lUXBK6FqshKWT36Ev/wSPR7ZmLfuXTsuiIaHspg3FO/DM2YyxL584mPCqGkZNvldaxuGQSyiIg2eprWf/5PHZtWk1QqAVLRHSLLm2TOZi+g0aQkd0V3cr/RbftbfC4NaxYdAgx3XCOexE1eTBffb6QI/u/ZeyU28nq0U/GjkWbkFAWAa206CArP51NefFhImLP7NKOjElg0IhxWMwKxpVPwt75GlUqApolCdfo3+PJvYHNX3/B5jWf0W/oBAaOmYw5OETr6kQHIqEsAp6qquzdupY1i/+Fy2knKi75B13a3g0u8gePIMhZg3HFL+DQCo2qFQElKBx12P/iyb+fvds3sWbZJ8QmpDJmyu1y3bG4LCSURYfR2FDHhi8+ZceGFQSFhGGNjPlBl6JCWpce9L9iOMa6wxhXPwNHvtSsXuHHDEG4Bz6Ie/gvKTy4j1WLP0bRKYycdAvZef1lIpe4bCSURYdTdvQwqxd+QGnhd4SGRxJmjWxxu6LoSO/ak775QzDWH8G46hk4slqbYoV/CY7CPfBh3Fc8QnlpMSsXf0xDXQ3DrrqR7v2GYjSZtK5QdHASyqJDcrvdFBbs4qsl/6a6/BiWiBiCwywt7tMinBuKMK5/AfYvBLeqUdVCMxHpqEP+B0/v2zh6cC9rly+ipvI4g8ZeR58h42TcWLQbCWXRoamqysHdW/lq6b+pr6nCEhlNcOiZ4ZyanUte775YQkPQbXoV3dZZ0HxCo6pFu0noi2vYM5BzJQW7trJx5RLqairpM3QC+aOvIcwaoXWFopORUBadgsvp5Ltvv2HdZx9Tf6L6rOEM3gVIevTqR1JmN9g9B8OGV6BynwYVi8sqewLOYb/EHduTXVs38s2XS2lubKBbn8EMHj+FyNgErSsUnZSEsuhUfOH8+SfU11QRYgkn1BpxxjWmQSFhdO3Zn649+6GUbce44SU48B9ZvjOQWZJw9/0par/7sKsK2zZvYPvaL/C43fQePIY+Q8ZLGAvNSSiLTsnldHJo73Y2r15M+bFCjGYz4VFxZ1xKpdPpSe/ak7ze/Qg26tDveAfdt+9D9UGNKhetojdBt2tx9rsfJWUQRQd2s3vbJg7u2YrJHMyAkVeTN3CkdFMLvyGhLDo1j8dD2dFD7Fi7nP07NwAK4dFxZ91qLyouiayuPcjomgc1RzDu/AfKrjnQWNn+hYvzi++F2u9ePL3/i9rK4+zft5u92zZQX1NBmDWSQeOup2ufQWcsNiOE1iSUhTip/kQ1uzd/xbavP8Pe1EhoeCQhYdYzurYVRUdCahY5XbuTlNkdd9F6jDvfhv0LwNmkUfWCyCzc3aeg9vop7rBEDuzdyb5vt1ByeD+q6iI2MY0rxl5HVm5f9AaD1tUKcVYSykL8gNNh5+DurXyzajFV5SXoDXqskXFnvUbVYDSRktmNnK7diE5Mw7N/CYZ9H8Oh5WCv06D6TiZpAO7u01C734AnNI5jhws4fGA/3337DU22ekymIPIGjiK3/1Bik9JkfWrh9ySUhTgHt9vN8eLDFOzYxO7Na3DYmzGazVgjY9H/YOwZICg4lLScHqSnpxOVlIVatgNDwQJ0B5bC8W81+Ak6IL0RMkZ7Q7jbFBwulcJD33H08AFKjhRQW12Bx+MhLTuXPkPHk9YlD6PJrHXVQlwwCWUhLoDL6eTY4f3s3vIVB3Ztwa26CA61EGqNPOuSi3qDkfjkDFLSM0lOz8KoU+DAUgwFC+HwCrDXa/BTBCBFgdg8SB+JM2MCuuyx1FeXc+TwYYoP76fq+DHqaipRnU7CwiPpO3QCXXtfgTUqRuvKhbgoEspCtFJzo43C73axa9OXFB/ciwcPwWFWQi0R51wT2RIRTVJ6DulpaUQmZqKW70F/9Cv0xWuheAPUHWvnn8JP6QyQ2A9P+khcGRPQpQ7C0dTI8dKjlJaUUlL4HbVV5TTUVuPxuDEHh9K972Byeg4kMT3nrD0YQgQSCWUhLkH9iWoO799BwY5NHDuyHzyg0+uxRESfdQY3eFvR0fHJxCYkkxAfR1RCOorbgad4E4ajX6IUb4DSLZ1j0pg1BeJ74U7Mx5U+Dn1KPk11VZSVFFNWcozykkKabPU01tfS2FALHrBGxZLbfxgZ3XoTm5QmQSw6FAllIdqIvamRsqOHOLxvOwU7vRONPB4PIWFWQizh591ZKNQaQUx8CnHxicQlJGCJTsJVfQilfBf6yl3oqvZD5X6o/i4wwzo4EuJ6QlwvXPF9ccf3RR/bHVV1UV9ZQkVlFcdLiqkoPYq9uRGHvZnG+hM4HQ4UID4lk9wBw0jLySMiJl4mbIkOS0JZiMvA4/FQXV5C8eH9FOzYRMmRAlAAj3e1sOAwCwaD8ZyP1+kNRMYkYI2MITwykqjwcO/SoOFxqI1VeKq+Q1+5B33lHqg+ALbjYCsHWwWojvb7QU8JivC2ek9+eawpqNYM3BGZ6GK7o5hCqa8qoaaqiqrqak5UlVNbVY69uRGPx4O9yYatvha3qoLHQ1hEFJndepOak0dieo4s7iE6DQllIdqBvamRipIijh87QtGB3ZQcKUB1OvHgQa83EhxmJSgk9EdbgIqiEGqJwBoZgyUimsiIcKxWC0EhYZhCLBiDw3E7G3HbqvDYKsF2HJ2tFH1DKUpzDbid3tBWT/3pOPOYTg+GoJNfwd4/TWFgtuI2h+M2R+Ixh+OxJKFYktBbE3CrLuwNNTQ11NLQYKPe1kijrYHGhlrqqiux1Z/e3MPldNLcZKO5sR6P2wN4iIxJIKN7X1KzuxObmHbWpU+F6AwklIXQgNvtpraqnMqyYkqOFFB0YDfV5SUoioLb7UGn12MOCsYcFILRHNSKgFIwmYMICgnFHBxKUEgoQcHeL7PJgF6nQ6dT0Ov16HQ67/cn/67T6dHp9Xg8blSXC9XlwqWqqKqKw+nC7nDhdDpwOuy4HHaammw0NdTR2FCHy3lm69ytqtibGmluasDpcKDT6fB4PJiCgolNTCM5syvJGV2JSUwjJOzMzUGE6IwklIXwE/bmJqrLS6irqaS6vITyY0dOXvJT5Q00POD2YDCZMQUFYzAYMRhN6PR6TVqVbrcbp8PuC2mn0w4eD4qiAzzodHqi4lNISM0kITWLiOh4wqPjzrpKmhDCS0JZCD/ndDhoqK2m/kQVJ6rKqSw7SnV5CQ21NTQ21OGwN6EoupPB7R3P9qiqtwWs16MoOhRF8X7pTv1dByeP6RRv4LvdbjxuNx6PG7fbjVtVcasu3G4VUHznAPB43Cg6PZbwSCwRMURExxERHU9YRBQhYVbCo2KxRESfd3KbEOJMEspCBDiX04m9uZHmRhv2JhtNjQ3YG2001J+gubHB1xWtqk5UlxNVdeF2qbh8f3eh6HQYjCYMJhNGowmD0Yw5OITgkDDMwaGYzEEYTWaCQsK8s8nDrJiDQ6TFK0Qbk1AWQggh/IT0LQkhhBB+QkJZCCGE8BMSykIIIYSfkFAWQggh/ISEshBCCOEnJJSFEEIIPyGhLIQQQvgJCWUhhBDCT0goCyGEEH5CQlkIIYTwExLKQgghhJ+QUBZCCCH8hISyEEII4ScklIUQQgg/IaEshBBC+AkJZSGEEMJPSCgLIYQQfkJCWQghhPATEspCCCGEn5BQFkIIIfyEhLIQQgjhJySUhRBCCD8hoSyEEEL4CQllIYQQwk9IKAshhBB+QkJZCCGE8BMSykIIIYSfkFAWQggh/ISEshBCCOEnJJSFEEIIPyGhLIQQQvgJCWUhhBDCT0goCyGEEH5CQlkIIYTwExLKQgghhJ+QUBZCCCH8hISyEEII4ScklIUQQgg/IaEshBBC+AkJZSGEEMJPSCgLIYQQfkJCWQghhPATEspCCCGEn5BQFkIIIfzE/w9okaMo/izhyQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (7,7))\n",
    "sorted_counts = df['label'].value_counts()\n",
    "plt.pie(sorted_counts, labels = sorted_counts.index, startangle = 90, counterclock = False, wedgeprops = {'width' : 0.6},\n",
    "       autopct='%1.1f%%', pctdistance = 0.7, textprops = {'color': 'black', 'fontsize' : 15}, shadow = True,\n",
    "        colors = sns.color_palette(\"Paired\")[7:])\n",
    "plt.text(x = -0.35, y = 0, s = 'Total Tweets: {}'.format(df.shape[0]))\n",
    "plt.title('Distribution of Tweets in the Dataset', fontsize = 16);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['label'] = df.label.map({'Offensive': 1, 'Non-offensive': 0})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Implementing a training and predicting Pipeline\n",
    "***"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Separate training and testing data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of rows in the total set: 35787\n",
      "Number of rows in the training set: 26840\n",
      "Number of rows in the test set: 8947\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(df['full_text'], \n",
    "                                                    df['label'], \n",
    "                                                    random_state=42)\n",
    "\n",
    "print('Number of rows in the total set: {}'.format(df.shape[0]))\n",
    "print('Number of rows in the training set: {}'.format(X_train.shape[0]))\n",
    "print('Number of rows in the test set: {}'.format(X_test.shape[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vectorize the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiate the CountVectorizer method\n",
    "count_vector = CountVectorizer(stop_words = 'english', lowercase = True)\n",
    "\n",
    "# Fit the training data and then return the matrix\n",
    "training_data = count_vector.fit_transform(X_train)\n",
    "\n",
    "# Transform testing data and return the matrix. Note we are not fitting the testing data into the CountVectorizer()\n",
    "testing_data = count_vector.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Implement Pipeline:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pipeline(learner_list, X_train, y_train, X_test, y_test): \n",
    "    '''\n",
    "    inputs:\n",
    "       - learner: the learning algorithm to be trained and predicted on\n",
    "       - X_train: features training set\n",
    "       - y_train: income training set\n",
    "       - X_test: features testing set\n",
    "       - y_test: income testing set\n",
    "    '''\n",
    "    \n",
    "    # Get length of Training Data:\n",
    "    size = len(y_train)\n",
    "    \n",
    "    results = {}\n",
    "    final_results = []\n",
    "    \n",
    "    for learner in learner_list:\n",
    "        \n",
    "        # Store the learner name:\n",
    "        results['Algorithm'] = learner.__class__.__name__\n",
    "\n",
    "        # Fit the learner:\n",
    "        start = time() # Get start time\n",
    "        print(\"Training {}\".format(learner.__class__.__name__))\n",
    "        learner = learner.fit(X_train, y_train)\n",
    "        end = time() # Get end time\n",
    "\n",
    "        # Store the training time\n",
    "        results['Training Time'] = end - start\n",
    "\n",
    "        start = time() # Get start time\n",
    "        predictions_test = learner.predict(X_test)\n",
    "        predictions_train = learner.predict(X_train)\n",
    "        end = time() # Get end time\n",
    "\n",
    "        # Store the prediction time\n",
    "        results['Prediction Time'] = end - start\n",
    "\n",
    "        # Compute the Accuracy on Test Set\n",
    "        results['Accuracy: Test'] = accuracy_score(y_test, predictions_test)\n",
    "\n",
    "        # Compute the Accuracy on Training Set\n",
    "        results['Accuracy: Train'] = accuracy_score(y_train, predictions_train)\n",
    "\n",
    "        # Compute the F1 Score on Test Set\n",
    "        results['F1 Score: Test'] = f1_score(y_test, predictions_test)\n",
    "\n",
    "        # Compute the F1 Score on Training Set\n",
    "        results['F1 Score: Train'] = f1_score(y_train, predictions_train)\n",
    "\n",
    "        # Compute the Precision on Test Set\n",
    "        results['Precision: Test'] = precision_score(y_test, predictions_test)\n",
    "\n",
    "        # Compute the Precision on Training Set\n",
    "        results['Precision: Train'] = precision_score(y_train, predictions_train)\n",
    "\n",
    "        # Compute the Recall on Test Set\n",
    "        results['Recall: Test'] = recall_score(y_test, predictions_test)\n",
    "\n",
    "        # Compute the Recall on Training Set\n",
    "        results['Recall: Train'] = recall_score(y_train, predictions_train)\n",
    "\n",
    "        # Success\n",
    "        print(\"Training {} finished in {:.2f} sec\".format(learner.__class__.__name__, results['Training Time']))\n",
    "        print('----------------------------------------------------')\n",
    "        \n",
    "        final_results.append(results.copy())\n",
    "    # Return a dataframe of the results\n",
    "    return final_results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Put Algorithms in Pipeline:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# make a list of models\n",
    "models = [MultinomialNB(), DecisionTreeClassifier(), LinearSVC(), AdaBoostClassifier(), \n",
    "          RandomForestClassifier(), BaggingClassifier(),\n",
    "         LogisticRegression(), SGDClassifier(), KNeighborsClassifier()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training MultinomialNB\n",
      "Training MultinomialNB finished in 0.03 sec\n",
      "----------------------------------------------------\n",
      "Training DecisionTreeClassifier\n",
      "Training DecisionTreeClassifier finished in 5.97 sec\n",
      "----------------------------------------------------\n",
      "Training LinearSVC\n",
      "Training LinearSVC finished in 0.68 sec\n",
      "----------------------------------------------------\n",
      "Training AdaBoostClassifier\n",
      "Training AdaBoostClassifier finished in 1.96 sec\n",
      "----------------------------------------------------\n",
      "Training RandomForestClassifier\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/ensemble/forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RandomForestClassifier finished in 6.24 sec\n",
      "----------------------------------------------------\n",
      "Training BaggingClassifier\n",
      "Training BaggingClassifier finished in 41.75 sec\n",
      "----------------------------------------------------\n",
      "Training LogisticRegression\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training LogisticRegression finished in 0.47 sec\n",
      "----------------------------------------------------\n",
      "Training SGDClassifier\n",
      "Training SGDClassifier finished in 0.07 sec\n",
      "----------------------------------------------------\n",
      "Training KNeighborsClassifier\n",
      "Training KNeighborsClassifier finished in 0.00 sec\n",
      "----------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "re = pipeline(models, training_data, y_train, testing_data, y_test)\n",
    "results = pd.DataFrame(re)\n",
    "results = results.reindex(columns = ['Algorithm', 'Accuracy: Test', 'Precision: Test', 'Recall: Test', 'F1 Score: Test', 'Prediction Time',\n",
    "                          'Accuracy: Train', 'Precision: Train', 'Recall: Train', 'F1 Score: Train', 'Training Time'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = results.reindex(columns = ['Algorithm', 'Accuracy: Test', 'Precision: Test', 'Recall: Test', 'F1 Score: Test', 'Prediction Time',\n",
    "                          'Accuracy: Train', 'Precision: Train', 'Recall: Train', 'F1 Score: Train', 'Training Time'])\n",
    "\n",
    "results.sort_values(by = 'F1 Score: Test', inplace = True, ascending = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analyze the results of the Pipeline\n",
    "***"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Algorithm</th>\n",
       "      <th>Accuracy: Test</th>\n",
       "      <th>Precision: Test</th>\n",
       "      <th>Recall: Test</th>\n",
       "      <th>F1 Score: Test</th>\n",
       "      <th>Prediction Time</th>\n",
       "      <th>Accuracy: Train</th>\n",
       "      <th>Precision: Train</th>\n",
       "      <th>Recall: Train</th>\n",
       "      <th>F1 Score: Train</th>\n",
       "      <th>Training Time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BaggingClassifier</td>\n",
       "      <td>0.927797</td>\n",
       "      <td>0.965493</td>\n",
       "      <td>0.923129</td>\n",
       "      <td>0.943836</td>\n",
       "      <td>0.419815</td>\n",
       "      <td>0.988376</td>\n",
       "      <td>0.996507</td>\n",
       "      <td>0.985780</td>\n",
       "      <td>0.991114</td>\n",
       "      <td>41.754808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SGDClassifier</td>\n",
       "      <td>0.927462</td>\n",
       "      <td>0.961044</td>\n",
       "      <td>0.927211</td>\n",
       "      <td>0.943824</td>\n",
       "      <td>0.001418</td>\n",
       "      <td>0.982526</td>\n",
       "      <td>0.992377</td>\n",
       "      <td>0.980964</td>\n",
       "      <td>0.986638</td>\n",
       "      <td>0.068617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>0.926344</td>\n",
       "      <td>0.964089</td>\n",
       "      <td>0.922279</td>\n",
       "      <td>0.942721</td>\n",
       "      <td>0.002266</td>\n",
       "      <td>0.978763</td>\n",
       "      <td>0.990242</td>\n",
       "      <td>0.977338</td>\n",
       "      <td>0.983748</td>\n",
       "      <td>0.467848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DecisionTreeClassifier</td>\n",
       "      <td>0.923438</td>\n",
       "      <td>0.952132</td>\n",
       "      <td>0.930272</td>\n",
       "      <td>0.941075</td>\n",
       "      <td>0.027815</td>\n",
       "      <td>0.998845</td>\n",
       "      <td>0.999943</td>\n",
       "      <td>0.998300</td>\n",
       "      <td>0.999121</td>\n",
       "      <td>5.974688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LinearSVC</td>\n",
       "      <td>0.916732</td>\n",
       "      <td>0.946599</td>\n",
       "      <td>0.925510</td>\n",
       "      <td>0.935936</td>\n",
       "      <td>0.001932</td>\n",
       "      <td>0.997019</td>\n",
       "      <td>0.998298</td>\n",
       "      <td>0.997167</td>\n",
       "      <td>0.997733</td>\n",
       "      <td>0.681417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RandomForestClassifier</td>\n",
       "      <td>0.910137</td>\n",
       "      <td>0.951120</td>\n",
       "      <td>0.910034</td>\n",
       "      <td>0.930123</td>\n",
       "      <td>0.347698</td>\n",
       "      <td>0.991990</td>\n",
       "      <td>0.997773</td>\n",
       "      <td>0.990029</td>\n",
       "      <td>0.993886</td>\n",
       "      <td>6.237272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>AdaBoostClassifier</td>\n",
       "      <td>0.907567</td>\n",
       "      <td>0.972508</td>\n",
       "      <td>0.884354</td>\n",
       "      <td>0.926338</td>\n",
       "      <td>0.341812</td>\n",
       "      <td>0.909650</td>\n",
       "      <td>0.971744</td>\n",
       "      <td>0.888448</td>\n",
       "      <td>0.928231</td>\n",
       "      <td>1.964057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>MultinomialNB</td>\n",
       "      <td>0.893372</td>\n",
       "      <td>0.901663</td>\n",
       "      <td>0.940306</td>\n",
       "      <td>0.920579</td>\n",
       "      <td>0.004461</td>\n",
       "      <td>0.944560</td>\n",
       "      <td>0.956763</td>\n",
       "      <td>0.959039</td>\n",
       "      <td>0.957900</td>\n",
       "      <td>0.025667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>KNeighborsClassifier</td>\n",
       "      <td>0.857606</td>\n",
       "      <td>0.895161</td>\n",
       "      <td>0.887245</td>\n",
       "      <td>0.891186</td>\n",
       "      <td>32.587579</td>\n",
       "      <td>0.897727</td>\n",
       "      <td>0.927596</td>\n",
       "      <td>0.915982</td>\n",
       "      <td>0.921753</td>\n",
       "      <td>0.002175</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Algorithm  Accuracy: Test  Precision: Test  Recall: Test  \\\n",
       "0       BaggingClassifier        0.927797         0.965493      0.923129   \n",
       "1           SGDClassifier        0.927462         0.961044      0.927211   \n",
       "2      LogisticRegression        0.926344         0.964089      0.922279   \n",
       "3  DecisionTreeClassifier        0.923438         0.952132      0.930272   \n",
       "4               LinearSVC        0.916732         0.946599      0.925510   \n",
       "5  RandomForestClassifier        0.910137         0.951120      0.910034   \n",
       "6      AdaBoostClassifier        0.907567         0.972508      0.884354   \n",
       "7           MultinomialNB        0.893372         0.901663      0.940306   \n",
       "8    KNeighborsClassifier        0.857606         0.895161      0.887245   \n",
       "\n",
       "   F1 Score: Test  Prediction Time  Accuracy: Train  Precision: Train  \\\n",
       "0        0.943836         0.419815         0.988376          0.996507   \n",
       "1        0.943824         0.001418         0.982526          0.992377   \n",
       "2        0.942721         0.002266         0.978763          0.990242   \n",
       "3        0.941075         0.027815         0.998845          0.999943   \n",
       "4        0.935936         0.001932         0.997019          0.998298   \n",
       "5        0.930123         0.347698         0.991990          0.997773   \n",
       "6        0.926338         0.341812         0.909650          0.971744   \n",
       "7        0.920579         0.004461         0.944560          0.956763   \n",
       "8        0.891186        32.587579         0.897727          0.927596   \n",
       "\n",
       "   Recall: Train  F1 Score: Train  Training Time  \n",
       "0       0.985780         0.991114      41.754808  \n",
       "1       0.980964         0.986638       0.068617  \n",
       "2       0.977338         0.983748       0.467848  \n",
       "3       0.998300         0.999121       5.974688  \n",
       "4       0.997167         0.997733       0.681417  \n",
       "5       0.990029         0.993886       6.237272  \n",
       "6       0.888448         0.928231       1.964057  \n",
       "7       0.959039         0.957900       0.025667  \n",
       "8       0.915982         0.921753       0.002175  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.reset_index(drop = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Accuracy: Test</th>\n",
       "      <th>Precision: Test</th>\n",
       "      <th>Recall: Test</th>\n",
       "      <th>F1 Score: Test</th>\n",
       "      <th>Prediction Time</th>\n",
       "      <th>Accuracy: Train</th>\n",
       "      <th>Precision: Train</th>\n",
       "      <th>Recall: Train</th>\n",
       "      <th>F1 Score: Train</th>\n",
       "      <th>Training Time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.857606</td>\n",
       "      <td>0.895161</td>\n",
       "      <td>0.884354</td>\n",
       "      <td>0.891186</td>\n",
       "      <td>0.001418</td>\n",
       "      <td>0.897727</td>\n",
       "      <td>0.927596</td>\n",
       "      <td>0.888448</td>\n",
       "      <td>0.921753</td>\n",
       "      <td>0.002175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.927797</td>\n",
       "      <td>0.972508</td>\n",
       "      <td>0.940306</td>\n",
       "      <td>0.943836</td>\n",
       "      <td>32.587579</td>\n",
       "      <td>0.998845</td>\n",
       "      <td>0.999943</td>\n",
       "      <td>0.998300</td>\n",
       "      <td>0.999121</td>\n",
       "      <td>41.754808</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Accuracy: Test  Precision: Test  Recall: Test  F1 Score: Test  \\\n",
       "min        0.857606         0.895161      0.884354        0.891186   \n",
       "max        0.927797         0.972508      0.940306        0.943836   \n",
       "\n",
       "     Prediction Time  Accuracy: Train  Precision: Train  Recall: Train  \\\n",
       "min         0.001418         0.897727          0.927596       0.888448   \n",
       "max        32.587579         0.998845          0.999943       0.998300   \n",
       "\n",
       "     F1 Score: Train  Training Time  \n",
       "min         0.921753       0.002175  \n",
       "max         0.999121      41.754808  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.describe().loc[['min', 'max'], :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# store the best metric scores\n",
    "best_acc = results[results['Accuracy: Test'] == results['Accuracy: Test'].max()]\n",
    "best_f1 = results[results['F1 Score: Test'] == results['F1 Score: Test'].max()]\n",
    "best_precision = results[results['Precision: Test'] == results['Precision: Test'].max()]\n",
    "best_recall = results[results['Recall: Test'] == results['Recall: Test'].max()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('darkgrid')\n",
    "plt.figure(figsize = (17, 7))\n",
    "\n",
    "barWidth = 0.17\n",
    " \n",
    "# set height of bar\n",
    "bars1 = results['Accuracy: Test']\n",
    "bars2 = results['F1 Score: Test']\n",
    "bars3 = results['Precision: Test']\n",
    "bars4 = results['Recall: Test']\n",
    "\n",
    " \n",
    "# Set position of bar on X axis\n",
    "r1 = np.arange(len(bars1))\n",
    "r2 = [x + barWidth for x in r1]\n",
    "r3 = [x + barWidth for x in r2]\n",
    "r4 = [x + barWidth for x in r3]\n",
    "\n",
    " \n",
    "# Make the plot\n",
    "pal = sns.color_palette()\n",
    "plt.bar(r1, bars1, color= pal[0], width=barWidth, edgecolor='white', label='Test Accuracy')\n",
    "plt.bar(r2, bars2, color= pal[1], width=barWidth, edgecolor='white', label='F1 Score')\n",
    "plt.bar(r3, bars3, color= pal[2], width=barWidth, edgecolor='white', label='Precision')\n",
    "plt.bar(r4, bars4, color= pal[4], width=barWidth, edgecolor='white', label='Recall')\n",
    "\n",
    " \n",
    "# Add xticks on the middle of the group bars\n",
    "plt.xlabel('Algorithm', fontweight='bold', fontsize = 13)\n",
    "plt.ylabel('Score', fontweight = 'bold', fontsize = 13)\n",
    "plt.xticks([r + barWidth for r in range(len(bars1))], results['Algorithm'], rotation = 15, fontsize = 11)\n",
    " \n",
    "# Create legend & Show graphic\n",
    "plt.legend(fontsize = 13)\n",
    "\n",
    "textstr = '\\n'.join(['Best Accuracy: {:.3f} - {}'.format(best_acc['Accuracy: Test'].values[0], best_acc['Algorithm'].values[0]), \n",
    "                     'Best F1 Score: {:.3f} - {}'.format(best_f1['F1 Score: Test'].values[0], best_f1['Algorithm'].values[0]),\n",
    "                   'Best Precision: {:.3f} - {}'.format(best_precision['Precision: Test'].values[0], best_precision['Algorithm'].values[0]), \n",
    "                    'Best Recall: {:.3f} - {}'.format(best_recall['Recall: Test'].values[0], best_recall['Algorithm'].values[0])])\n",
    "props = dict(boxstyle='round', facecolor='lightgrey', alpha=0.5)\n",
    "\n",
    "#place a text box\n",
    "plt.text(9.2, 1, textstr, fontsize=14,\n",
    "        verticalalignment='top', bbox=props)\n",
    "\n",
    "plt.title('Classification Summary of Algorithms', fontweight = 'bold', fontsize = 17);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# store the best/worst performance time\n",
    "best_train_time = results[results['Training Time'] == results['Training Time'].min()]\n",
    "worst_train_time = results[results['Training Time'] == results['Training Time'].max()]\n",
    "best_prediction_time = results[results['Prediction Time'] == results['Prediction Time'].min()]\n",
    "worst_prediction_time = results[results['Prediction Time'] == results['Prediction Time'].max()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (12, 7))\n",
    "\n",
    "barWidth = 0.17\n",
    " \n",
    "# set height of bar\n",
    "bars1 = results['Training Time']\n",
    "bars2 = results['Prediction Time']\n",
    " \n",
    "# Set position of bar on X axis\n",
    "r1 = np.arange(len(bars1))\n",
    "r2 = [x + barWidth for x in r1]\n",
    " \n",
    "# Make the plot\n",
    "plt.bar(r1, bars1, color= pal[0], width=barWidth, edgecolor='white', label='Training Time')\n",
    "plt.bar(r2, bars2, color= pal[1], width=barWidth, edgecolor='white', label='Prediction Time')\n",
    " \n",
    "# Add xticks on the middle of the group bars\n",
    "plt.xlabel('Algorithm', fontweight='bold', fontsize = 13)\n",
    "plt.ylabel('Time (seconds)', fontweight = 'bold', fontsize = 13)\n",
    "plt.xticks([r + barWidth for r in range(len(bars1))], results['Algorithm'], rotation = 15, fontsize = 11)\n",
    " \n",
    "# Create legend & Show graphic\n",
    "plt.legend(fontsize = 13)\n",
    "\n",
    "textstr = '\\n'.join(('Best Training Time: {:.3f} - {}'.format(best_train_time['Training Time'].values[0], best_train_time['Algorithm'].values[0]), \n",
    "                     'Worst Training Time: {:.3f} - {}'.format(worst_train_time['Training Time'].values[0], worst_train_time['Algorithm'].values[0]),\n",
    "                   'Best Prediction Time: {:.3f} - {}'.format(best_prediction_time['Training Time'].values[0], best_prediction_time['Algorithm'].values[0]), \n",
    "                    'Worst Prediction Time: {:.3f} - {}'.format(worst_prediction_time['Training Time'].values[0], worst_prediction_time['Algorithm'].values[0])))\n",
    "\n",
    "props = dict(boxstyle='round', facecolor='lightgrey', alpha=0.5)\n",
    "\n",
    "#place a text box\n",
    "plt.text(9, 36, textstr, fontsize=14,  bbox=props)\n",
    "\n",
    "plt.title('Time Complexity of Algorithms', fontweight = 'bold', fontsize = 17);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> The Algorithms Bagging, SGD, Logistic Regression and Decision Tree and Random Forest and LinearSVC have more or less similar performance. We will tune the hyperparameters of these algorithms. However, the training time for Bagging is very high as compared to the others. Hence we drop it. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuning Hyperparameters\n",
    "***"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def param_tuning(clf, param_dict, X_train, y_train, X_test, y_test):\n",
    "    \n",
    "    # make scorer object\n",
    "    scorer = make_scorer(f1_score)\n",
    "\n",
    "    # perform Grid Search for Parameters\n",
    "    grid_obj = GridSearchCV(estimator = clf,\n",
    "                           param_grid = param_dict,\n",
    "                           scoring = scorer,\n",
    "                           cv = 5)\n",
    "\n",
    "    grid_fit = grid_obj.fit(X_train, y_train)\n",
    "\n",
    "    # Get the estimator\n",
    "    best_clf = grid_fit.best_estimator_\n",
    "\n",
    "    # Make predictions using the unoptimized and model\n",
    "    predictions = (clf.fit(X_train, y_train)).predict(X_test)\n",
    "    best_predictions = best_clf.predict(X_test)\n",
    "    \n",
    "    # Report the before-and-afterscores\n",
    "    print(clf.__class__.__name__)\n",
    "    print(\"\\nOptimized Model\\n------\")\n",
    "    print(\"Best Parameters: {}\".format(grid_fit.best_params_))\n",
    "    print(\"Accuracy: {:.4f}\".format(accuracy_score(y_test, best_predictions)))\n",
    "    print(\"F1-score: {:.4f}\".format(f1_score(y_test, best_predictions)))\n",
    "    print(\"Precision: {:.4f}\".format(precision_score(y_test, best_predictions)))\n",
    "    print(\"Recall: {:.4f}\".format(recall_score(y_test, best_predictions)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Stochastic Gradient Classifier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SGDClassifier\n",
      "\n",
      "Optimized Model\n",
      "------\n",
      "Best Parameters: {'alpha': 0.0002, 'max_iter': 2500}\n",
      "Accuracy: 0.9281\n",
      "F1-score: 0.9439\n",
      "Precision: 0.9697\n",
      "Recall: 0.9194\n"
     ]
    }
   ],
   "source": [
    "# Dict for parameters\n",
    "param_grid = {\n",
    "    'alpha' : [0.095, 0.0002, 0.0003],\n",
    "    'max_iter' : [2500, 3000, 4000]\n",
    "}\n",
    "\n",
    "clf_sgd = SGDClassifier()\n",
    "\n",
    "param_tuning(clf_sgd, param_grid, training_data, y_train, testing_data, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic Regression Classifier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression\n",
      "\n",
      "Optimized Model\n",
      "------\n",
      "Best Parameters: {'C': 1.2}\n",
      "Accuracy: 0.9270\n",
      "F1-score: 0.9433\n",
      "Precision: 0.9638\n",
      "Recall: 0.9236\n"
     ]
    }
   ],
   "source": [
    "# Dict for parameters\n",
    "param_grid = {\n",
    "    'C': [1, 1.2, 1.3, 1.4]\n",
    "}\n",
    "\n",
    "clf_lr = LogisticRegression()\n",
    "\n",
    "param_tuning(clf_lr, param_grid, training_data, y_train, testing_data, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Decision Tree Classifier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTreeClassifier\n",
      "\n",
      "Optimized Model\n",
      "------\n",
      "Best Parameters: {'min_samples_leaf': 1, 'min_samples_split': 5}\n",
      "Accuracy: 0.9244\n",
      "F1-score: 0.9417\n",
      "Precision: 0.9554\n",
      "Recall: 0.9284\n"
     ]
    }
   ],
   "source": [
    "param_grid = {\n",
    "    'min_samples_split': [2, 5, 8],\n",
    "    'min_samples_leaf': [1, 2, 5, 8]\n",
    "}\n",
    "\n",
    "clf_dt = DecisionTreeClassifier()\n",
    "\n",
    "param_tuning(clf_dt, param_grid, training_data, y_train, testing_data, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Random Forest Classifier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/ensemble/forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestClassifier\n",
      "\n",
      "Optimized Model\n",
      "------\n",
      "Best Parameters: {'min_samples_leaf': 1, 'min_samples_split': 2, 'n_estimators': 150}\n",
      "Accuracy: 0.9157\n",
      "F1-score: 0.9351\n",
      "Precision: 0.9462\n",
      "Recall: 0.9243\n"
     ]
    }
   ],
   "source": [
    "param_grid = {\n",
    "    'n_estimators': [50,150],\n",
    "    'min_samples_leaf': [1, 5],\n",
    "    'min_samples_split': [2, 5]\n",
    "}\n",
    "\n",
    "clf_rf = RandomForestClassifier()\n",
    "\n",
    "param_tuning(clf_rf, param_grid, training_data, y_train, testing_data, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Linear SVC:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/svm/base.py:929: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearSVC\n",
      "\n",
      "Optimized Model\n",
      "------\n",
      "Best Parameters: {'C': 0.25}\n",
      "Accuracy: 0.9257\n",
      "F1-score: 0.9425\n",
      "Precision: 0.9591\n",
      "Recall: 0.9264\n"
     ]
    }
   ],
   "source": [
    "param_grid = {\n",
    "    'C': [0.25, 0.5, 0.75, 1, 1.2]\n",
    "}\n",
    "\n",
    "clf_linsvc = LinearSVC()\n",
    "\n",
    "param_tuning(clf_linsvc, param_grid, training_data, y_train, testing_data, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusion:\n",
    "***\n",
    "We found Stochastic Gradient to be the best suited model for our data. We achieved the following performance parameters:\n",
    "-\tAccuracy: 92.81 %\n",
    "-\tPrecision: 96.97 %\n",
    "-\tRecall: 91.94 %\n",
    "-\tF1-Score: 94.39 %\n",
    "\n",
    "Save the model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['cb_sgd_final.sav']"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filename = 'cb_sgd_final.sav'\n",
    "joblib.dump(clf_sgd, filename)"
   ]
  }
 ],
 "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}