code.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This predictive classification analysis will try to predict whether a person has a heart disease or not based on 14 different features, where 8 of them are categorical features and the rest is numerical.\n",
    "\n",
    "Table of contents is shared below for understanding the flow of the project;"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of Contents\n",
    "    1- Importing Dataset\n",
    "    2- Data Exploration & Visualization\n",
    "    3- Data Wrangling and Feature Scaling\n",
    "    4- Predictive Machine Learning Models Implications\n",
    "    5- Hyperparameter Tunings\n",
    "    6- Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Importing Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Importing Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>cp</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>chol</th>\n",
       "      <th>fbs</th>\n",
       "      <th>restecg</th>\n",
       "      <th>thalach</th>\n",
       "      <th>exang</th>\n",
       "      <th>oldpeak</th>\n",
       "      <th>slope</th>\n",
       "      <th>ca</th>\n",
       "      <th>thal</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>63</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>145</td>\n",
       "      <td>233</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>37</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>130</td>\n",
       "      <td>250</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>187</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>130</td>\n",
       "      <td>204</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>172</td>\n",
       "      <td>0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>56</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>120</td>\n",
       "      <td>236</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>178</td>\n",
       "      <td>0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>120</td>\n",
       "      <td>354</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>163</td>\n",
       "      <td>1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  slope  \\\n",
       "0   63    1   3       145   233    1        0      150      0      2.3      0   \n",
       "1   37    1   2       130   250    0        1      187      0      3.5      0   \n",
       "2   41    0   1       130   204    0        0      172      0      1.4      2   \n",
       "3   56    1   1       120   236    0        1      178      0      0.8      2   \n",
       "4   57    0   0       120   354    0        1      163      1      0.6      2   \n",
       "\n",
       "   ca  thal  target  \n",
       "0   0     1       1  \n",
       "1   0     2       1  \n",
       "2   0     2       1  \n",
       "3   0     2       1  \n",
       "4   0     2       1  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"data/heart.csv\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Data Exploration & Visualization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "    Data contains;\n",
    "\n",
    "    age - age in years\n",
    "    sex - (1 = male; 0 = female)\n",
    "    cp - chest pain type\n",
    "    trestbps - resting blood pressure (in mm Hg on admission to the hospital)\n",
    "    chol - serum cholestoral in mg/dl\n",
    "    fbs - (fasting blood sugar > 120 mg/dl) (1 = true; 0 = false)\n",
    "    restecg - resting electrocardiographic results\n",
    "    thalach - maximum heart rate achieved\n",
    "    exang - exercise induced angina (1 = yes; 0 = no)\n",
    "    oldpeak - ST depression induced by exercise relative to rest\n",
    "    slope - the slope of the peak exercise ST segment\n",
    "    ca - number of major vessels (0-3) colored by flourosopy\n",
    "    thal - 3 = normal; 6 = fixed defect; 7 = reversable defect\n",
    "    target - have disease or not (1=yes, 0=no)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "categorical_columns = ['sex', 'cp', 'exang', 'ca', 'thal','fbs', 'restecg', 'slope']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(303, 14)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 303 entries, 0 to 302\n",
      "Data columns (total 14 columns):\n",
      " #   Column    Non-Null Count  Dtype  \n",
      "---  ------    --------------  -----  \n",
      " 0   age       303 non-null    int64  \n",
      " 1   sex       303 non-null    int64  \n",
      " 2   cp        303 non-null    int64  \n",
      " 3   trestbps  303 non-null    int64  \n",
      " 4   chol      303 non-null    int64  \n",
      " 5   fbs       303 non-null    int64  \n",
      " 6   restecg   303 non-null    int64  \n",
      " 7   thalach   303 non-null    int64  \n",
      " 8   exang     303 non-null    int64  \n",
      " 9   oldpeak   303 non-null    float64\n",
      " 10  slope     303 non-null    int64  \n",
      " 11  ca        303 non-null    int64  \n",
      " 12  thal      303 non-null    int64  \n",
      " 13  target    303 non-null    int64  \n",
      "dtypes: float64(1), int64(13)\n",
      "memory usage: 33.2 KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 0]\n",
      "[3 2 1 0]\n",
      "[0 1]\n",
      "[0 2 1 3 4]\n",
      "[1 2 3 0]\n",
      "[1 0]\n",
      "[1 0]\n",
      "[0 1 2]\n",
      "[0 2 1]\n"
     ]
    }
   ],
   "source": [
    "# Checking categorical columns for further analysis\n",
    "\n",
    "categorical_columns = [\"sex\",\"cp\",\"exang\",\"ca\",\"thal\",\"target\",\"fbs\",\"restecg\",\"slope\"]\n",
    "\n",
    "for col in categorical_columns:\n",
    "    print(data[col].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>age</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>chol</th>\n",
       "      <th>thalach</th>\n",
       "      <th>oldpeak</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>303.000000</td>\n",
       "      <td>303.000000</td>\n",
       "      <td>303.000000</td>\n",
       "      <td>303.000000</td>\n",
       "      <td>303.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>54.366337</td>\n",
       "      <td>131.623762</td>\n",
       "      <td>246.264026</td>\n",
       "      <td>149.646865</td>\n",
       "      <td>1.039604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.082101</td>\n",
       "      <td>17.538143</td>\n",
       "      <td>51.830751</td>\n",
       "      <td>22.905161</td>\n",
       "      <td>1.161075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>29.000000</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>126.000000</td>\n",
       "      <td>71.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>47.500000</td>\n",
       "      <td>120.000000</td>\n",
       "      <td>211.000000</td>\n",
       "      <td>133.500000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>55.000000</td>\n",
       "      <td>130.000000</td>\n",
       "      <td>240.000000</td>\n",
       "      <td>153.000000</td>\n",
       "      <td>0.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>61.000000</td>\n",
       "      <td>140.000000</td>\n",
       "      <td>274.500000</td>\n",
       "      <td>166.000000</td>\n",
       "      <td>1.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>77.000000</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>564.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>6.200000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              age    trestbps        chol     thalach     oldpeak\n",
       "count  303.000000  303.000000  303.000000  303.000000  303.000000\n",
       "mean    54.366337  131.623762  246.264026  149.646865    1.039604\n",
       "std      9.082101   17.538143   51.830751   22.905161    1.161075\n",
       "min     29.000000   94.000000  126.000000   71.000000    0.000000\n",
       "25%     47.500000  120.000000  211.000000  133.500000    0.000000\n",
       "50%     55.000000  130.000000  240.000000  153.000000    0.800000\n",
       "75%     61.000000  140.000000  274.500000  166.000000    1.600000\n",
       "max     77.000000  200.000000  564.000000  202.000000    6.200000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Descriptive statistics of the dataset\n",
    "\n",
    "data.drop(columns=categorical_columns).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Database is pretty clean and ready for further analysis.\n",
    "\n",
    "# No data-type conversion or removing missing value is needed at this point until further\n",
    "# classification process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Gender')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# checking target and gender distribution\n",
    "\n",
    "plt.rcParams[\"font.family\"] = \"agency fb\"\n",
    "\n",
    "fig, ax = plt.subplots(1,2,figsize=(12,5))\n",
    "\n",
    "sns.countplot(data=data,x=\"target\",palette=['lightgreen','firebrick'],edgecolor=\"k\",\n",
    "              linewidth=1.5,ax=ax[0])\n",
    "for p in ax[0].patches:\n",
    "    ax[0].annotate(f'\\n{p.get_height()}', \n",
    "                (p.get_x()+0.4, p.get_height()-65), ha='center', va='center', color='k', size=18)\n",
    "ax[0].set_title(\"Disease Distribution\",size=20,fontweight=\"bold\")\n",
    "ax[0].set_xticklabels([\"no disease\",\"disease\"],size=15)\n",
    "ax[0].set_xlabel(\"Target\",size=20)\n",
    "\n",
    "sns.countplot(data=data,x=\"sex\",palette=['pink',\"lightblue\"],edgecolor=\"k\",linewidth=1.5,ax=ax[1])\n",
    "for p in ax[1].patches:\n",
    "    ax[1].annotate(f'\\n{p.get_height()}', \n",
    "                (p.get_x()+0.4, p.get_height()-35), ha='center', va='center', color='k', size=18)\n",
    "ax[1].set_title(\"Gender Distribution\",size=20,fontweight=\"bold\")\n",
    "ax[1].set_xticklabels([\"female\",\"male\"],size=15)\n",
    "ax[1].set_xlabel(\"Gender\",size=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Disease percentage : 54.4\n",
      "Disease percentage : 45.5 \n",
      "\n",
      "Male percentage : 68.3\n",
      "Female percentage : 31.6\n"
     ]
    }
   ],
   "source": [
    "# There are 165 people with disease, whereas 138 people without disease in the dataset.\n",
    "\n",
    "disease_percentage = str((len(data[data[\"target\"]==1])/len(data))*100)[0:4]\n",
    "print(\"Disease percentage :\", disease_percentage)\n",
    "no_disease_percentage = str((len(data[data[\"target\"]==0])/len(data))*100)[0:4]\n",
    "print(\"Disease percentage :\", no_disease_percentage,\"\\n\")\n",
    "\n",
    "# There are total 207 male and 96 female in the dataset.\n",
    "male_percentage = str((len(data[data[\"sex\"]==1])/len(data))*100)[0:4]\n",
    "print(\"Male percentage :\", male_percentage)\n",
    "female_percentage = str(100-((len(data[data[\"sex\"]==1])/len(data))*100))[0:4]\n",
    "print(\"Female percentage :\", str(100-((len(data[data[\"sex\"]==1])/len(data))*100))[0:4])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2531b760>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting Heart Disease Frequency by Age\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"]=(16,9)\n",
    "plt.style.use(\"bmh\")\n",
    "\n",
    "pd.crosstab(index=data[\"age\"],columns=data[\"target\"]).plot(kind=\"bar\",color=['tan','darkred'],\n",
    "                                                          linewidth=1.3,edgecolor=\"k\")\n",
    "plt.title(\"Heart Disease Distribution by Age\",size=25,fontweight=\"bold\",pad=15)\n",
    "plt.xlabel('Age',size=23,labelpad=10)\n",
    "plt.ylabel('Count',size=23,labelpad=10)\n",
    "plt.xticks(size=13,rotation=45,fontweight=\"bold\")\n",
    "plt.yticks(size=18)\n",
    "plt.legend([\"No Disease\",\"Disease\"],fontsize=\"x-large\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x250addc0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting Heart Disease Frequency by Sex\n",
    "\n",
    "plt.rcParams[\"figure.figsize\"]=(14,5)\n",
    "plt.style.use(\"bmh\")\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "pd.crosstab(index=data[\"sex\"],columns=data[\"target\"]).plot(kind=\"bar\",color=['tan','darkred'],\n",
    "                                                          linewidth=1.3,edgecolor=\"k\",ax=ax)\n",
    "plt.title(\"Heart Disease Distribution by Sex\",size=22,fontweight=\"bold\",pad=15)\n",
    "plt.xlabel('Sex',size=17,labelpad=10)\n",
    "plt.ylabel('Count',size=17,labelpad=10)\n",
    "ax.set_xticklabels([\"female\",\"male\"])\n",
    "plt.xticks(size=13,rotation=0,fontweight=\"bold\")\n",
    "plt.yticks(size=18)\n",
    "plt.legend([\"No Disease\",\"Disease\"],fontsize=\"x-large\")\n",
    "\n",
    "# Apparently, the older the people become, there are higher chances of hearth disease."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x25229e20>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "pd.crosstab(index=data[\"cp\"],columns=data[\"target\"]).plot(kind=\"bar\",color=['tan','darkred'],\n",
    "                                                          linewidth=1.3,edgecolor=\"k\",ax=ax)\n",
    "plt.title(\"Heart Disease Distribution by Type of Chest Pain\",size=22,fontweight=\"bold\",pad=15)\n",
    "plt.xlabel('Type of Chest Pain',size=17,labelpad=10)\n",
    "plt.ylabel('Count',size=17,labelpad=10)\n",
    "#ax.set_xticklabels([\"female\",\"male\"])\n",
    "plt.xticks(size=13,rotation=0,fontweight=\"bold\")\n",
    "plt.yticks(size=13)\n",
    "plt.legend([\"No Disease\",\"Disease\"],fontsize=\"x-large\")\n",
    "\n",
    "# Apparently, chest pain type 1, 2 and 3 have some relationship with hearth disease occurence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Checking multi-collinearity between features for further analysis by using heatmap\n",
    "\n",
    "sns.heatmap(data.drop(columns=categorical_columns).corr(),annot=True)\n",
    "\n",
    "# None of the features are highly correlated with each other, which is nice for further analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col0,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col1,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col2,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col3,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col4{\n",
       "            color:  cyan;\n",
       "        }#T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col1,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col2,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col3,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col4,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col0,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col2,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col3,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col4,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col0,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col1,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col3,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col4,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col0,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col1,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col2,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col4,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col0,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col1,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col2,#T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col3{\n",
       "            color:  darkred;\n",
       "        }</style><table id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >age</th>        <th class=\"col_heading level0 col1\" >trestbps</th>        <th class=\"col_heading level0 col2\" >chol</th>        <th class=\"col_heading level0 col3\" >thalach</th>        <th class=\"col_heading level0 col4\" >oldpeak</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59level0_row0\" class=\"row_heading level0 row0\" >age</th>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col0\" class=\"data row0 col0\" >1.000000</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col1\" class=\"data row0 col1\" >0.279351</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col2\" class=\"data row0 col2\" >0.213678</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col3\" class=\"data row0 col3\" >-0.398522</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row0_col4\" class=\"data row0 col4\" >0.210013</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59level0_row1\" class=\"row_heading level0 row1\" >trestbps</th>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col0\" class=\"data row1 col0\" >0.279351</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col1\" class=\"data row1 col1\" >1.000000</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col2\" class=\"data row1 col2\" >0.123174</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col3\" class=\"data row1 col3\" >-0.046698</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row1_col4\" class=\"data row1 col4\" >0.193216</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59level0_row2\" class=\"row_heading level0 row2\" >chol</th>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col0\" class=\"data row2 col0\" >0.213678</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col1\" class=\"data row2 col1\" >0.123174</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col2\" class=\"data row2 col2\" >1.000000</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col3\" class=\"data row2 col3\" >-0.009940</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row2_col4\" class=\"data row2 col4\" >0.053952</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59level0_row3\" class=\"row_heading level0 row3\" >thalach</th>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col0\" class=\"data row3 col0\" >-0.398522</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col1\" class=\"data row3 col1\" >-0.046698</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col2\" class=\"data row3 col2\" >-0.009940</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col3\" class=\"data row3 col3\" >1.000000</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row3_col4\" class=\"data row3 col4\" >-0.344187</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59level0_row4\" class=\"row_heading level0 row4\" >oldpeak</th>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col0\" class=\"data row4 col0\" >0.210013</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col1\" class=\"data row4 col1\" >0.193216</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col2\" class=\"data row4 col2\" >0.053952</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col3\" class=\"data row4 col3\" >-0.344187</td>\n",
       "                        <td id=\"T_1a6b3379_30d4_11eb_9ea7_207918becd59row4_col4\" class=\"data row4 col4\" >1.000000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x252daaf0>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking multi-collinearity wih color map as well just to see if I miss any multi-correlation\n",
    "def color_negative_red(val):\n",
    "    color = 'darkred' if 0.5 > val > -0.5 else 'cyan'  # If there is moderate/strong relat. then cyan\n",
    "    return 'color: %s' % color\n",
    "\n",
    "data.drop(columns=categorical_columns).corr().style.applymap(color_negative_red)\n",
    "\n",
    "#No correlation between features, we are ready for predictive analysis with those numerical features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Data Wrangling and Feature Scaling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Creating Dummy Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_backup = data.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data_backup.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding dummy variables and dropping original features\n",
    "categorical_columns = ['sex', 'cp', 'exang', 'ca', 'thal','fbs', 'restecg', 'slope']\n",
    "\n",
    "for col in categorical_columns:\n",
    "    data = pd.get_dummies(data,columns=[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>age</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>chol</th>\n",
       "      <th>thalach</th>\n",
       "      <th>oldpeak</th>\n",
       "      <th>target</th>\n",
       "      <th>sex_0</th>\n",
       "      <th>sex_1</th>\n",
       "      <th>cp_0</th>\n",
       "      <th>cp_1</th>\n",
       "      <th>...</th>\n",
       "      <th>thal_2</th>\n",
       "      <th>thal_3</th>\n",
       "      <th>fbs_0</th>\n",
       "      <th>fbs_1</th>\n",
       "      <th>restecg_0</th>\n",
       "      <th>restecg_1</th>\n",
       "      <th>restecg_2</th>\n",
       "      <th>slope_0</th>\n",
       "      <th>slope_1</th>\n",
       "      <th>slope_2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>63</td>\n",
       "      <td>145</td>\n",
       "      <td>233</td>\n",
       "      <td>150</td>\n",
       "      <td>2.3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>37</td>\n",
       "      <td>130</td>\n",
       "      <td>250</td>\n",
       "      <td>187</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>41</td>\n",
       "      <td>130</td>\n",
       "      <td>204</td>\n",
       "      <td>172</td>\n",
       "      <td>1.4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>56</td>\n",
       "      <td>120</td>\n",
       "      <td>236</td>\n",
       "      <td>178</td>\n",
       "      <td>0.8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>57</td>\n",
       "      <td>120</td>\n",
       "      <td>354</td>\n",
       "      <td>163</td>\n",
       "      <td>0.6</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  trestbps  chol  thalach  oldpeak  target  sex_0  sex_1  cp_0  cp_1  \\\n",
       "0   63       145   233      150      2.3       1      0      1     0     0   \n",
       "1   37       130   250      187      3.5       1      0      1     0     0   \n",
       "2   41       130   204      172      1.4       1      1      0     0     1   \n",
       "3   56       120   236      178      0.8       1      0      1     0     1   \n",
       "4   57       120   354      163      0.6       1      1      0     1     0   \n",
       "\n",
       "   ...  thal_2  thal_3  fbs_0  fbs_1  restecg_0  restecg_1  restecg_2  \\\n",
       "0  ...       0       0      0      1          1          0          0   \n",
       "1  ...       1       0      1      0          0          1          0   \n",
       "2  ...       1       0      1      0          1          0          0   \n",
       "3  ...       1       0      1      0          0          1          0   \n",
       "4  ...       1       0      1      0          0          1          0   \n",
       "\n",
       "   slope_0  slope_1  slope_2  \n",
       "0        1        0        0  \n",
       "1        1        0        0  \n",
       "2        0        0        1  \n",
       "3        0        0        1  \n",
       "4        0        0        1  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Splitting dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X = data.drop(columns=\"target\",axis=1)\n",
    "y = data[\"target\"]\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Normalizing Numerical Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "# fitting scaler on training data\n",
    "norm = MinMaxScaler().fit(X_train)\n",
    "\n",
    "# transforming training data\n",
    "X_train_norm = norm.transform(X_train)\n",
    "\n",
    "# transforming testing data\n",
    "X_test_norm = norm.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Predictive Machine Learning Models Implications"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.metrics import confusion_matrix, accuracy_score, recall_score, f1_score\n",
    "from sklearn.metrics import classification_report, roc_auc_score\n",
    "\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### a) Applying Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Actual</th>\n",
       "      <th>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>155</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Actual  Predicted\n",
       "178       0          0\n",
       "113       1          1\n",
       "242       0          0\n",
       "10        1          1\n",
       "262       0          0\n",
       "200       0          1\n",
       "146       1          1\n",
       "155       1          1\n",
       "177       0          1\n",
       "183       0          0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creating model\n",
    "lr_model = LogisticRegression(max_iter=10000)\n",
    "\n",
    "# fitting model\n",
    "lr_model.fit(X_train,y_train)\n",
    "\n",
    "# predicting outcome\n",
    "y_pred_lr = lr_model.predict(X_test)\n",
    "\n",
    "pd.DataFrame({\"Actual\":y_test,\"Predicted\":y_pred_lr}).sample(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### a-1) Evaluating Logistic Regression Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.87.\n"
     ]
    }
   ],
   "source": [
    "# Confusion Matrix \n",
    "\n",
    "confusion_matrix(y_test,y_pred_lr)\n",
    "\n",
    "TN = confusion_matrix(y_test,y_pred_lr)[0][0]\n",
    "TP = confusion_matrix(y_test,y_pred_lr)[1][1]\n",
    "\n",
    "fp = confusion_matrix(y_test,y_pred_lr)[0][1]\n",
    "fn = confusion_matrix(y_test,y_pred_lr)[1][0]\n",
    "\n",
    "# Checking Accuracy\n",
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_test,y_pred_lr)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sensivity score is 0.925.\n"
     ]
    }
   ],
   "source": [
    "# Recall / Sensivity\n",
    "\n",
    "'''\n",
    "-Something we want to maximize\n",
    "-How \"sensitive\" is the classifier to detecting positive instances?\n",
    "-Also known as \"True Positive Rate\" or \"Recall\"\n",
    "-TP / all positive\n",
    "    all positive = TP + FN\n",
    "\n",
    "Applying the same understanding, we know that Recall shall be the model metric we use to select \n",
    "our best model when there is a high cost associated with False Negative.\n",
    "\n",
    "Mostly used in fraud detection or sick patient detection.\n",
    "'''\n",
    "\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_test,y_pred_lr)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision score is 0.880.\n"
     ]
    }
   ],
   "source": [
    "# Precision \n",
    "\n",
    "'''\n",
    "Precision is a good measure to determine, when the costs of False Positive is high. \n",
    "\n",
    "For instance, email spam detection. In email spam detection, a false positive means that an email \n",
    "that is non-spam (actual negative) has been identified as spam (predicted spam). \n",
    "\n",
    "The email user might lose important emails if the precision is not high for the spam detection model.\n",
    "\n",
    "Precision = TP / (TP+FP)\n",
    "'''\n",
    "\n",
    "\n",
    "precision = TP/(TP+fp)\n",
    "print(f\"Precision score is {str(precision)[:5]}.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F-1 score is 0.9024390243902439.\n"
     ]
    }
   ],
   "source": [
    "# F1 Score\n",
    "'''\n",
    "F1 Score is needed when you want to seek a balance between Precision and Recall. \n",
    "Right…so what is the difference between F1 Score and Accuracy then? We have previously \n",
    "seen that accuracy can be largely contributed by a large number of True Negatives which in\n",
    "most business circumstances, we do not focus on much whereas False Negative and False Positive \n",
    "usually has business costs (tangible & intangible) thus\n",
    "\n",
    "F1 Score might be a better measure to use if we need to seek a balance between Precision \n",
    "and Recall AND there is an uneven class distribution (large number of Actual Negatives).\n",
    "'''\n",
    "print(f\"F-1 score is {f1_score(y_test,y_pred_lr)}.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### a-2) Summary of Logistic Regression Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.87.\n",
      "Sensivity score is 0.925.\n",
      "F-1 score is 0.9024390243902439.\n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.76      0.80        21\n",
      "           1       0.88      0.93      0.90        40\n",
      "\n",
      "    accuracy                           0.87        61\n",
      "   macro avg       0.86      0.84      0.85        61\n",
      "weighted avg       0.87      0.87      0.87        61\n",
      "\n",
      "\n",
      "\n",
      "ROC AUC Score: 0.8434523809523808\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_test,y_pred_lr)))\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_test,y_pred_lr)))\n",
    "#print(f\"Precision score is {str(precision)[:5]}.\")\n",
    "print(f\"F-1 score is {f1_score(y_test,y_pred_lr)}.\\n\\n\")\n",
    "\n",
    "print(classification_report(y_test,y_pred_lr))\n",
    "print(f\"\\n\\nROC AUC Score: {roc_auc_score(y_test,y_pred_lr)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### b) Applying Random Forest Model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>Actual</th>\n",
       "      <th>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>253</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Actual  Predicted\n",
       "178       0          0\n",
       "253       0          0\n",
       "262       0          0\n",
       "31        1          1\n",
       "298       0          0\n",
       "121       1          1\n",
       "132       1          1\n",
       "192       0          0\n",
       "41        1          1\n",
       "213       0          0"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creating model\n",
    "rf_model = RandomForestClassifier()\n",
    "\n",
    "# fitting model\n",
    "rf_model.fit(X_train,y_train)\n",
    "\n",
    "\n",
    "# predicting outcome \n",
    "y_pred_rf = rf_model.predict(X_test)\n",
    "\n",
    "y_pred_rf_training = rf_model.predict(X_train)\n",
    "\n",
    "pd.DataFrame({\"Actual\":y_test,\"Predicted\":y_pred_rf}).sample(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### b-1) Evaluating Random Forest Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.82.\n",
      "Sensivity score is 0.800.\n",
      "F-1 score is 0.85.\n",
      "\n",
      "\n",
      "ROC AUC Score: 0.8285714285714286\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_test,y_pred_rf)))\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_test,y_pred_rf)))\n",
    "#print(f\"Precision score is {str(precision)[:5]}.\")\n",
    "print(f\"F-1 score is {str(f1_score(y_test,y_pred_rf))[:4]}.\")\n",
    "\n",
    "print(f\"\\n\\nROC AUC Score: {roc_auc_score(y_test,y_pred_rf)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### b-2) Hyperparameter Tuning for Random Forest Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC Curve for N_estimators\n",
    "\n",
    "n_estimators = [1, 2, 4, 8, 16, 32, 64, 100, 200, 400]\n",
    "train_results = []\n",
    "test_results = []\n",
    "for estimator in n_estimators:\n",
    "    rf = RandomForestClassifier(n_estimators=estimator, n_jobs=-1,random_state=0,min_samples_split=2,\n",
    "                                 min_samples_leaf=1,max_features=\"sqrt\",\n",
    "                                 max_depth=3)\n",
    "    rf.fit(X_train, y_train)\n",
    "    train_pred = rf.predict(X_train)\n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_train, train_pred)\n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    train_results.append(roc_auc)\n",
    "    y_pred_rf = rf.predict(X_test)\n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_test, y_pred_rf)\n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    test_results.append(roc_auc)\n",
    "    \n",
    "from matplotlib.legend_handler import HandlerLine2D\n",
    "line1, = plt.plot(n_estimators, train_results, \"b\", label=\"Train AUC\")\n",
    "line2, = plt.plot(n_estimators, test_results, 'r', label=\"Test AUC\")\n",
    "plt.legend(handler_map={line1: HandlerLine2D(numpoints=2)})\n",
    "plt.ylabel(\"AUC score\",size=25)\n",
    "plt.xlabel(\"n_estimators\",size=25)\n",
    "plt.xticks(size=20)\n",
    "plt.yticks(size=20)\n",
    "#plt.savefig(\"nestimator.png\",dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC Curve for Max_Depth\n",
    "\n",
    "max_depths = np.linspace(1, 32, 32, endpoint=True)\n",
    "train_results = []\n",
    "test_results = []\n",
    "\n",
    "for max_depth in max_depths:\n",
    "    \n",
    "    rf = RandomForestClassifier(max_depth=max_depth)\n",
    "    rf.fit(X_train, y_train)\n",
    "    \n",
    "    train_pred = rf.predict(X_train)\n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_train, train_pred)\n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    train_results.append(roc_auc)\n",
    "    \n",
    "    y_pred_rf = rf.predict(X_test)\n",
    "    \n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_test, y_pred_rf)\n",
    "    \n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    test_results.append(roc_auc)\n",
    "    \n",
    "from matplotlib.legend_handler import HandlerLine2D\n",
    "\n",
    "line1, = plt.plot(max_depths, train_results, \"b\", label=\"Train AUC\")\n",
    "line2, = plt.plot(max_depths, test_results, \"r\", label=\"Test AUC\")\n",
    "plt.legend(handler_map={line1: HandlerLine2D(numpoints=2)})\n",
    "plt.ylabel(\"AUC score\",size=25)\n",
    "plt.xlabel(\"Tree depth\",size=25)\n",
    "plt.xticks(size=20)\n",
    "plt.yticks(size=20)\n",
    "#plt.savefig(\"treedepth.png\",dpi=300)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAAFRCAYAAABpDwxJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAB7JElEQVR4nO3dd3wUdf7H8dd3+6ZSEikieCqKFAugYDkVsZ2n3okF8E5PxY4IeOqPu7PdnefZRYpYUFHvFHvvDVAPC2ABRASVKCUhgZi6fb+/P3YTNiGE1Jmd2c/z8dgHZmZ29jPk7ZBPZub7VVprhBBCCCGEEEK0n8PsAoQQQgghhBDCLqTBEkIIIYQQQogOIg2WEEIIIYQQQnQQabCEEEIIIYQQooNIgyWEEEIIIYQQHUQaLCGEEEIIIYToIC6zC0g3CxYs0F6v1+wyRDtFo1FcLom3MJbkTphBcifMILkTRku3zNXW1paNHj26sKl16VNlmvB6vQwYMMDsMkQ7hUIhpFEWRpPcCTNI7oQZJHfCaOmWuWXLlhXtaJ3cIihsqbi42OwSRAaS3AkzSO6EGSR3wmhWypw0WMKW3G632SWIDCS5E2aQ3AkzSO6E0ayUOWmwhC3l5+ebXYLIQJI7YQbJnTCD5E4YzUqZS8tnsJRS1wM30LABrASO1lovbbTt28CxjXbxidb6kOT6Y4CXAX/K+ihwhdZ6TkfXLtJDWVkZ2dnZZpchMozkTphBcifMILmzP6011dXVaK3NLgWAmpoaYrGY4Z+rlCInJwelVIvfk5YNFjAKeAK4J2XZy8BIYGmjbY8FpgIfJb8eBtynlMrWWtcAhwHrgHNS3nMDcDQgDZZNWem3HMI+JHfCDJI7YQbJnf1VV1fj9XrxeDxmlwJAVlaWKaMIhsNhqquryc3NbfF70rXBUkCJ1npJ/QKlwsnlTVlbt61SKidlH3V/1jba1xYgB2Fb4XDY7BJEBpLcCTNI7oQZJHf2p7VOm+YKMO1KmsfjIRgMtuo98gyWsKVAIGB2CSIDSe6EGSR3wgySO2G0eDxudgktJg2WsKWePXuaXYLIQJI7YQbJnTCD5E50tptuuomhQ4dy4YUXUlRU1K5RBB966CEefPBBAJ544gneffddACZMmADAMcccw8UXX8wxxxzTIc95SYOVpr791sH69Yo0ea7Qcqw0V4KwD8mdMIPkTphBcic627XXXsshhxzCzJkz6devH5FIpMntYrEY1157bbP72rBhAxs2bNjh+n79+nH//fezzz777PBzWiNdn8HKeNdck8VHH7np2jXO4MGx+teQITH23jtGGt0Sm5bS6Z5hkTkkd8IMkjthBsmdMNqORvFzOp3cdNNNO31/S24xPPPMMztkvi1psBrZvHkzEyZMwOVyEYvFGDNmDBMnTqS4uJjs7GycTieVlZUUFhaydetWtNYUFhZSUlJCTk5i3Izq6mp69OhBaWkpSim6detGaWkpeXl5xGIxampq6NmzJ8XFxbjdbvLz8ykrKyM/P59wOEwgECAvbwD5+Yrychcffujgww+3fbNdrjh77x1jjz2qGDQowv77Q69em9lzzy5UVVURDofr9+/3+/F4PFRUVFBQUEBFRQWRSKR+vZHHVLfe4/GQm5vLli1b6Nq1K4FAgGAwWL/e5/Ph9/spLy+ne/fubTomrTVFRUW2OiY7fp/sdkxOp5Py8nJbHZMdv092O6bq6mrKyspsdUx2/D7Z7Zjy8vIoKiqy1THZ8fvUnmNyu91kZ2cTjUZxOp1orYnH4/Tq1b7bQ8vKthCJRHA4HDgcDqLRaP3P3Vpr3G53/fp4PE4oFGLTpk1cfPHFdOvWjbPPPptnn32WXXbZhaKiIh599FEuvfRSzj77bGbPnk1+fj7Dhw/nwgsvBKi/5S8ajRIOh4nH40QiEaLRaP3+tdaEQiEOOeQQAEKhUINjjsViFBUVNfg+NUely9j2qZRSC4AlWuurUpatA+7QWs9qtO0aYK8mdpOrta5WSt0InKS1Hp7ynnlAjtb69MZvWrx4sR4wYEBHHEa7aQ0bNihWrnSxYoWT5cudrFzp5IcfHGi9fRffu3ecwYOjDBkSY9CgxNWuX/0qjiMDbwQtKiqiX79+ZpchMozkTphBcifMILmzv8rKSvLy8rZb3q1b13btd+vW8hZvO3HiRO688042b97M3LlzGTt2LJWVlQSDQUaNGsWECRN46KGHmDBhAueddx5r1qzhsMMO4/nnn2fatGkAlJSU8OKLLxIKhRg7dizvvfceu+yyC8ccc0yD9z/00EM7rKOpv4tly5YtHT169PCmtrfDFaxhwK/YNiz7cOBB88rpOEpBnz6aPn0iHH/8tvtBq6vhm2+crFjhZMUKF8uXO1m1ysnGjQ42bvTw9tvb9pGdrRk4sO72wiiDBsUYODCG3ecG7Nq1ff/zC9EWkjthBsmdMIPkLnO1pkHqSEopBg0aBMANN9zAokWLmn2uqs6qVat466230Fqz//77N7vtk08+yRlnnNHu+bYs32BprSuBr+q+Vkp1Ma8aY+TkwMEHxzj44BiQmIciFoMffnAkm65E47VihZNNmxx8/rmLzz93AV4AlNLsuWe8/irX4MFRBg+O0auXphWTVKe1xG2W2//WRYjOJLkTZpDcCTNI7oTRUu+6W79+ff2Vp51ZtWoVt912G5FIhEWLFjU7YfBHH33EqaeeKg2WUuoK4AOt9XKzazGT0wn9+8fp3z/Oqaduu9q1ZYtqcHvh8uVOvvvOydq1iddLL23bR7du8Qa3Fw4enBhQowOe9TNcayeEE6IjSO6EGSR3wgySO2GmqqoqrrvuOjZt2rTTbVevXs15551HNBrl22+/5YQTTmDOnDm88MILOJ1OIHHL68SJE1m9enWHDHJhh2ewyoCrtNbzkl8fBXyADZ7B6iyhEKxe7dyu8aqo2P5hLbdbM2BArMFIhoMHx+jaNf1ykyoUCuH1es0uQ2QYyZ0wg+ROmEFyZ387egbLLPF4HIdJAwtk4jNYopW8Xthvvxj77bdtIrW6ATXqnumqu9Xwxx+dLF/uYvnyhlHZddd4/TNddVe7dt89fQbUKC4ulodvheEkd8IMkjthBsmdMFokErFMU5+uDZYGeiilUrtCD6CVUr2B3lrrJSnbDlRKHZP8ev+U5XV/ZjXaV3dArm2nSB1Q44QTtt1iWFmZGFBj5cptjdeqVU42bHCwYYOHN9/cto+cnMSAGnWN1+DBiQE1srKMPx6fz2f8h4qMJ7kTZpDcCTNI7oTRzLp61Rbp2mAtAK4H/piyrAr4FLgIuIFtowZ+AFydfNX5TGtdk/zvxcA04POU9VFgcodXbUN5eTByZIyRI7dd7YrF4PvvHSm3F7pYudJJcbGDzz5z8dln22LlcCQG1Nh2e2FiQI2ePTt3QA2/3995OxdiByR3wgySO2EGyZ0wmjRY7aS1/jvw9x2sXgLcmLLtmTvZ19uAnAU6kNMJe+8dZ++945x22rarXaWlKmUUw8RIht9952DNGidr1jh54YVt+ygoiDe4vXDw4Cj9+8c7bECN8vLytLpvWGQGyZ0wg+ROmEFyJ4xWN+GxFaRlgyWsqbBQM2pUlFGjovXLgsHEgBqpz3WtWOGkrMzBwoUOFi7c1lF5PA0H1KhrvvLzWz+gRvfu3TvkmIRoDcmdMIPkTphBcieM1t6h041knUqFJfl8sP/+Mfbfv+GAGj//7GgwiuGKFU7WrXPy9dcuvv66YSx3263hCIZDhsTo27f5ATWqqqrIycnprMMSokmSO2EGyZ0wg+ROGC0Wi8kVLCF2RCno2zdO375xTjxx+wE1li931V/pWrXKyc8/J15vvLFtHzk5uv55rrpbDffdN0bdLeHhcNjgoxJCcifMIbkTZpDcic5200038fzzzzNs2DCuvfZaevbs2ar3f/vtt/z73//m0Ucf5aeffuKRRx7hhhtuYOLEidx5551ceeWVxGIx1q9fz0033cSBBx7YYbVLgyXSRlMDakSjsHato34wjRUrEle8SkocfPKJm08+2XaLocOh2WuvxIAaw4btwfjxcbp0Se/5uoS9tPbkL0RHkNwJM0juRGe79tpr2bRpE3feeSc+n494PL7DbT/66CP23XffBreurlq1isrKyh2+JxQK8dBDD/HEE0+wZcuWDq1dGiyR1lwuGDAgzoABDQfU2Ly54YAay5e7WLvWwXffOfnuOyfPP+/hX//SnHFGmAsuCDFoUKyZTxGiY8i8MMIMkjthBsmdMFpz82B99NFH7LLLLg0arDVr1jBw4MCdXm0dPnw4WR08p5A0WMKSdtlFc/TRUY4+etuAGoEAfPtt4rmup5+G//0vm0cf9fLoo14OOSTCBReEOOmkSIeNVChEYzJssTCD5E6YQXKXWbp269Yp+y3furVF223dupU//OEP7LXXXgwYMIC99tqLefPmEQgEuOmmm3j77bf54YcfeOCBB+rfEwwGGTx4MGvXrm32ecG999673cfRmHUGlBdiJ/x+OPDAGOecE+Y//9nM4sUVXHhhkJwczeLFbiZMyOGAA/K57TYfJSWdOAmXyFgej8fsEkQGktwJM0juhJGqq6s5+OCDufrqqykrK0MpxWGHHca0adPw+/0cd9xxXHXVVQ3eo5Rin332YdWqVYbXK1ewhC1VVFSwzz5duPXWAH/7W4Cnn/by4INe1qxxcsstfu6808cpp0SYMCHIiBGxTp30WGSOiooKunTpYnYZIsNI7oQZJHeZpaVXmjrT//73PyKRCIMGDeK4445j+fLlLFiwgGnTpm23bSgU4r333uPnn3+mX79+HHTQQTvc72effUZeXh4DBgzosFrlCpawpYKCgvr/zsuDCy4I8cknlbzwQhUnnRQmFoPnnvNw4ol5HHVULo8/7qG21sSChS2k5k4Io0juhBkkd8Johx56KNOmTaNv37588MEHjB49Gq01X3755Xbbfvfdd5x77rk88MADlJaWNrvftWvXsn79+g6tVRosYUsVFRXbLVMKjjwyymOP1fDllxVMnRqge/c4y5e7mDw5m8GD87n+ej/r1sn/FqJtmsqdEJ1NcifMILkTRsrJyeGTTz5h2rRpfPXVV4TDYW699VY++eQTXC4XBQUF3HHHHfXbf/vtt+y2224A1NTUkJOTw4cffsjEiRNZvXo1brcbr9fLxIkTefLJJ+nWwc+YKa1lGOtUixcv1h15iVCYo6ioqEWjGwWD8OKLHubO9bJsWeKOWaU0xx2XGBRj1KhosxMaC5GqpbkToiNJ7oQZJHf2V1lZSV5entll1AuFQjscRbCzNfV3sWzZsqWjR48e3tT28qOjsKWWzs/h88G4cWHefbeKd96pZNy4EG43vPWWhzPOyOXgg/OYM8dLRYU8pCV2TuaFEWaQ3AkzSO6E0dwWGgZaGixhS8XFxa1+z7BhMe69t5YVKyq47roAu+4a54cfnPztb1kMGpTPlVdm8c038r+M2LG25E6I9pLcCTNI7oTRIpHIzjdKE/LTorCl7OzsNr+3oEAzdWqQL76o4PHHqznyyAi1tYp587wcfng+J5+cw0svubHQ/+fCIO3JnRBtJbkTZpDcCaM5LPTMRlpWqpS6XikVU0rplFeFUmpYE9vOb7SdVkqtVEp5kut/38R6rZQ6xvgjE0ZxOp3t3ofLBb/9bYQXXqjmf/+r4IILEnNqffyxm/POS8ypdfvtMqeW2KYjcidEa0nuhBkkd8JoykJz6qRlgwWMAp4ADkp51QAjm9h2NPD3lO3GAQOBXsn1HiDcaF8HAZ92XvnCbJWVlR26vwED4tx2W4AVK37h1ltr6d8/xqZNDv79bz/77ZfPRRdl8dlnTmTMmMzW0bkToiUkd8IMkjv7U0oRDofNLqNeLBYz5XPD4XCrm7t0nWhYASVa6yX1C5QKJ5c3te26um2VUsGU5ZBosGpT9yXsr7CwsFP2m5cHF14Y4oILQixa5GLuXC9vvOHm2We9PPusl/33j3LBBSHGjAnj93dKCSKNdVbuhGiO5E6YQXJnfzk5OVRXVxMMBne+sQGi0SihUMjwz1VKkZOT06r3pGuD1ZE8QMDsIoSxtm7dSlZWVqftv25OrSOPjPLzzw4eecTDY495+eorF5Mmubj+ej9//GOY888P0a9fvNPqEOmls3MnRFMkd8IMkjv7U0qRm5trdhn11q9fT58+fcwuo0XS9RbBjiQNVgYycn633XaLc/31QVasqGD27BoOPDBKebmDmTN9DB2ax1lnZfP++y7i0mfZnswrKMwguRNmkNwJo1kpc3ZosGJA6sD4df8dTf4pDVYGMuPWBZ8Pxo8P8957iTm1xo5NzKn15pseTj89l5Ej87j/fi9y27o9aQ0FBXLLjDCe3KolzCC5E0azUubs0GA9BtyllLpAKXUU8CLwPrAhud4DDGo0guBWpdRAU6oVhigpKTH184cNizFnTi3Ll1dw7bUBeveOs3atk7/8JYtBg7rw5z/LnFpWpjWsX694+WU3N9zg5+STc+jbtwv9++/C6NG5XHppFnff7eO119ysWeMgGt35PoVoK7PPdyIzSe6E0ayUOcs/g6W1vlop9RPwYHLRw8CFett1xP8ACxq97V0SIwl+03h/mzdvZsKECbhcLmKxGGPGjGHixIkUFxeTnZ2N0+mksrKSwsJCtm7ditaawsJCSkpK6h+Aq66upkePHpSWlqKUolu3bpSWlpKXl0csFqOmpoaePXtSXFyM2+0mPz+fsrIy8vPzCYfDBAKB+vUej4fc3Fy2bNlC165dCQQCBIPB+vU+nw+/3095eTndu3enqqqKcDhcv97v9+PxeKioqKCgoICKigoikUj9erseE0BRUVFaHNNpp61j7FgfH3/cjYce8vP557k88oiXRx7xcuihIU49dSO/+U2YrKzM+z5Z5ZiysnrxwQdVfPttPitXZvHFFx7KypqaUd7JF1/AF180PLW63Zq+fYP07x9jzz3D7LprJcOHZ5OfX4zfj3yf5JjadUyBQICysjJbHZMdv092Oyav10tRUZGtjsmO3yc7HVM4HKaoqChtjqk5Kh3vZ1RKLQCWaK2vSlm2DrhDaz2r0banAXOBt4AfgCnAtVrru5rZfxlwldZ6XuN1ixcv1gMGDGj/QQhTbdmyhe7du5tdRpNWrXLw8MNe5s/3UlOTGOyyd+84550X4pxzQhQWpt//k5kkHIZvvnGydKmLZcucLFniYs2a7ed7yc+PM3RojGHDogwbFuPAA6NUVm6lrKyA1audfPedM/mng/Xrm54vRilN375x9t47zt57x9h77xj77BNjn33i5OdLDkTLpPP5TtiX5E4YLd0yt2zZsqWjR48e3tQ6y1/BAh4FbtBa3wn1zdlbSqnntNZFplYmTFNdXZ1W/xOm2nffOLffHuC66wLMn+9l7lwva9c6+de//Nx+u4/f/S7MBReEGD48hoXm1LMkreHHHx0sW5ZoqJYudbF8uZNQqOFfvMejGTIk0UzVNVV77BHf7vsTCFQycmRXRo5sOFdHdTWsXZtour77zlHffP34o4OiIidFRU7eeafhFbEePbY1XakNWM+eWnIhGkjn852wL8mdMJqVMmeHBisbWJ3y9XfJP3MBlFI9gTyt9XeN3yjsq0ePHmaXsFN5eXDRRYk5tRYudPHQQ17efNPNM894eeYZLwcckJhT69RTZU6tjrJli2rQTC1b5qS8fPtn4fr3jzF06LZmatCgGF7vzve/o9zl5MABB8Q44ICGjVc4DD/84Eg2XtuarzVrnJSUOCgpcfDhhw0br7y8OP37xxtc7dp77xh9+8ZxNn2hTNicFc53wn4kd8JoVsqcHRqsnbkEOAlo8hKesKfS0lJ22203s8toEYcDRo2KMmpUlJ9+cvDII14ee8zDl1+6uPzyxJxaZ58d5rzzQvTtK2O9t1QgAF9/XXerX6KZWrdu+w6ksDBef5vf0KFRDjwwRpcubbs9r7W583hgwIA4AwbEgUj98ngc1q93sHp1avPlZPVqB7/84mDpUgdLlzY8fXu9mr32im13u+Gee8Zb1BwK67LS+U7Yh+ROGM1KmcuEBktkIGXRe6j69o1zww0BrrkmwAsveHjwwcTkxffc42PmTC/HHx/hggtCHHVUVG4TSxGPw3ffORo0UytXOolGG/4lZWVp9t9/WzM1fHiUXXftuFvuOip3DkciC337xjn22G1DEGoNpaVqu1sNv/vOyaZNDlaudLFyZeN9aX71q/h2txr27x8jL69DyhUms+r5Tlib5E4YzUqZkwZL2FK3bt3MLqFd/H4466ww48eHWbLEyUMPeXnhBQ9vvJF49e8fY8KEEOPGhTLyh+RNm1T9LX5Ll7r44gsX1dUNT7wOh2bgwNRmKnFFx9WJZ73Ozp1SsMsuml12iXL44Q3XVVbCmjUNB9f47jsn69Y5+P57J99/7+SNNxq+p1evbbca1jVg++wTo6BAnvOyEquf74Q1Se6E0ayUuXRtsDTQQymVelufB9BKqd5Ab631kpR1e6Vs2ytlH3V/5iqljknZ3p2yXthQaWkp/fr1M7uMdlMKDjooxkEH1fKPfwR4/HEvDz/sZc0aJ9OmZXHTTX7Gjg0xYUIoeZuZ/VRVwZdfuho8O7Vp0/bPTfXpE2Po0G3N1H77RUmO0GoYM3OXl5eYf23YsIbPeQWDiee86q501V39Wrs2cdVr0yYHCxc2fM6ra9eGIxsmmrA4ffrEccj0bWnHLuc7YS2SO2E0K2UuXYdpvwG4noYTIVcBR5N4nuoGrbVKbvtecnmqNcCBWusapdRxwEuAL2V9BfBrrfXyxp8tw7TbQ3l5OV27djW7jE4RjcLrr7uZO9fLRx9t+8H4iCMStw+ecEKkU6/SdKZIBFatcrJ06bZnp1avdpD8371eXl6cAw+MMXx4YiCKAw+M0rOn+ecyK+UuFoOffnLUX+1KbcCqqpq+fJWVVfec17bBNfbeO8Yee8RxNzUtmDCElXIn7ENyJ4yWbplrbpj2tGywzCQNlj2UlZVRUFBgdhmd7ptvEnNqPfXUtjm1dt01MafW2Wen95xaWid+wF+yxJl8bsrF1187CQQa/nDvdmsGD942RPrQoVH22is9r6TYIXdaQ3Gx2m5kw9WrnWze3PRfusuVeM4r9VbDuue8kvN+i05kh9wJ65HcCaOlW+akwWoFabDsoaioyDKXkTtCZSU8+WRiTq3vv0+MlOfxaE49NTGnVuPbxsxQXq5YutRZPwjFsmUuysq2/4F9jz0azjc1eHAMn6+JHaYhu+ful1/UdiMbfvedg59+2v4qY50+fRpe7Uo0YXG6dZN/ezqK3XMn0pPkThgt3TInDVYrSINlD6FQCG8Gjk0dj8OCBS7mzvXy1lvu+h96hw5NzKn1+9+HDWlWgkFYvjzRRNU1VT/8sP0Q6d27xxs0U0OHxuja1brnpEzNXW0tfP99otn69tttzdcPPziIRJpuvAoKGo5seNRREds+R9jZMjV3wlySO2G0dMucNFitIA2WPaTbbznMUFSUmFPr8cc99ZPpdu8e5+yzQ5x3XpjdduuYH2bjcVi71tGgmVqxwrndD9Y+n2b//esaqcTofn37xm01Wp3krqFIBNat234i5e++c9bf0prq2GMjTJoU5LDDZBqC1pDcCTNI7oTR0i1z0mC1gjRY9rBx40Z69+5tdhlpIRCA559PzKn19deJ0S8cDs1vfpMYFOOII1r3w2xJSeMh0p1UVja81U8pzT771E3gm2imBgyI2X4gBMldy2gNGzao+oE1li938tJLnvrn7w48MMrllwc5+WTrDthiJMmdMIPkThgt3TInDVYrSINlDzU1NWTL0/UNaA2ff+5k7lwvL73kqb/C1L9/jAsuCDF27PZzalVXw1dfuRqM6rdhw/bPTfXuHU9elUo0U/vvHyU314ijSi+Su7bbskUxd27iOcItWxIZ69cvxmWXhTjrrJAMltEMyZ0wg+ROGC3dMicNVitIg2UP6XYZOd1s3qx47DEvjzzirZ9TKidHM25ciIEDY/UDUXz7rZN4vOHlrZwcXd9M1Y3q16uXnEdActcRamth/nwPs2f7+PHHxHN7XbvGmTAhxIUXpvfImGaR3AkzSO6E0dItc9JgtYI0WPbwyy+/0KVLF7PLSHuRyLY5tT7+ePv791wuzaBBDYdI33vv9BwiPR1I7jpOLAavveZm5kwfS5cm7hP0+TTjxoW57LIge+0lA2LUkdwJM0juhNHSLXPNNVhyd7uwpXA4bHYJluB2w+9+F+F3v4vwzTcOHn3US3m54sADE03VkCEx/H6zq7QOyV3HcTrhlFMinHxyhE8/dTJzpo833vAwb56XRx/1cOKJES6/PMiIEeZPQWA2yZ0wg+ROGM1KmZMGS9hSIBAwuwTLGTgwzq23yt9bm4RCuBYtQtXWwimnIEPgdRylYOTIGCNH1vDddwFmz/bx1FMeXnst8Tr44CiTJgX5zW8iGXtlVc53wgySO2E0K2UuQ/85EnbXs2dPs0sQdqc1zqVL8V99Nfn77kvu2LHsfd555P7mN7jefTcxqojoUHvvHeeee2r56qsKrrwyQH5+nM8+c3H22TmMHJnHvHkegkGzqzSenO+EGSR3wmhWypw0WMKWiouLzS5B2JTauBHv9OnkjRxJ3rHH4nvoIRy//EJ00CCi+fm4PvuM3DPPJPeYY3C/8YY0Wp2gRw/NtdcGWb68gptvrmW33WKsXevkyiuz2X//fO64w0d5eeZcRZTznTCD5E4YzUqZkwZL2JLH4zG7BGEntbW4n32WnNNOI3+//cj6xz9wrllDvLCQ4KWXUrloEVUffsiad96h9sYbiRcW4vriC3L+8AdyjzwS90svJWZkFh0qJwcuuSTE0qWVPPhgNfvtF6W01MHNN/sZMiSfadP8FBXZ/585Od8JM0juhNGslDkZRbARGUXQHqqrq8nJyTG7DGFlWuP89FO8Tz6J58UXUVVVicUeD5ETTiA8fjyRo48mdfbk+tzV1uJ97DF8M2fi2LQJgNg++xD485+JnHpqYgQH0eG0hkWLXMyc6eP99xPfF4dD87vfRZg0KcgBB9hzQAw53wkzSO6E0dItc82NImj/X+2JjLRlyxazSxAW5fjpJ3y33Ube8OHknXgi3scfR1VVER06lNrbb6di1Spq5s0jcvzxDZorSMldVhahSy6hYulSau64g/iuu+JcvZqciy4ib+RIPE88kRgjX3QopeDII6M8+2w1H35YydixIRwOeOEFD0cfncfvfpfDO++4bHfXppzvhBkkd8JoVspc2jVYSqnrlVIxpZROeVUopYY1se38RttppdRKpZQnuX6wUqq80fqYUupm449MGKlr165mlyCspLoazxNPkHPKKeQfcAD+W27B+eOPxHv1Ijh5MhWLF1P17ruEJkxAN5Ot7XLn8xE+//xEozV9OrHdd8f5/fdkX345eQcfjGfePAiFOvfYMtSgQTHmzKll2bIKJk4MkpOj+fBDN2PH5nL44Xk8+aQHC4342yw53wkzSO6E0ayUubRrsIBRwBPAQSmvGmBkE9uOBv6est04YCDQK7l+GKAa7esh4NjOK1+kAysN5SlMEo/jWrSIrMsuo8u++5J9+eW4P/oI7fMROv10qp59loqvvyZwww3E99mnRbvcYe48HsLnnEPlZ59Rc++9xPbaC2dREdlXXkn+sGF4H3yQjBz+zgB9+mj++c8AK1b8wo031tKrV5xVq5xMnJjNgQfmM2OGl8pKs6tsHznfCTNI7oTRrJS5dGywFFCitV5S9wLCyeVNbbsuZbuVKcvr/ow22tfGHexL2EhQflgVO+D4/nt8//oXeQccQO7vf493/nxUTQ2RkSOpmT6dX779ltoHHiB69NGtflZqp7lzuQiPG0fl4sVUP/ggsQEDcGzcSNb//R/5Bx6Id/ZsqKlpx9GJHcnLgyuuCPHFFxXMnl3DvvvG2LTJwY03ZjF4cBeuu87Phg3W/KdBznfCDJI7YTQrZU4mGhbWFo/jWLcO54oVOJcvx7lyJa7lyxkSjRI7/HAiRx5J5Kij0H36mF2pMJGqqMD9wgt458/H9dln9ctju+1GeOxYwuPGEd9jj3Z/Tovn6HA6iZx2GpFTT8X92mv47rwT19dfk3XddfjuuYfgZZcRmjABcnPbXZNoyOOB8ePDjBsX5t13Xcya5ePDD93Mnu3j/vu9nHZamMsvDzFokHUGxLDS3DDCPiR3wmhWypw0WMI6amtxrlqVaKZWrMC1YgXOlStR1dVNbu587jk8zz0HQGzPPYkecQSRI48k+utfN/scjbCJWAzXBx/gnT8f9+uvo5K/+dLZ2YRPOYXwuHFEDzsMHB13Ib+4uJh+/fq1/A0OB5GTTyZy0km4334b3+2341q2jKx//APfjBmELrmE0MUXo/PzO6xGkaAUHHtslGOPreaLL5zMmuXjpZfcPPWUl6ee8nL00YmRB484IopK8wtbrc6dEB1AcieMZqXMWb3BigGpw3jV/XfUhFpEB1IlJdsaqeXLca5YgWPtWlQTcwnFe/YkNngw0SFDiA0aRGzIELaUltJz5UpcCxfi/vBDnN9/j/P77/E+8ghaKWIHHJBoto48kuiIEeDzmXCUojM4vv0W7/z5eJ5+GkfKpISRX/+a8PjxhE86KTGBUifwtTVHShE5/ngixx2H64MP8N1xB+5PPsF/yy34Zs8meNFFhC69FN2tW8cWLAA48MAYDz1Uw/XXO5gzx8t//uPl/ffdvP++m/32izJpUpDf/S6CK03/xWxz7oRoB8mdMJqVMpd282AppRYAS7TWV6UsWwfcobWe1Wjb24FLgKnAWuDR5J/HaK21Uurc5PsKUt5zI3CS1rrJcetlHiyDRaM41q6tv7WvrqlybN683aba6STev3+DRio2eDC6sHC7bSsrK8nLy6v/DOcXX+BeuBDXwoW4PvsMlTJEtvb5iI4YUd9wxfbbT+Ypshi1dSue557DM38+ri++qF8e22MPwuPGER47lvhuu3V6HQ1y1x5a4/r440SjtWhRYlF2NqHzzyc4cSJ6l13a/xlih7ZuVTz8sJcHH/RSWpq4wrnbbjEuvTTEH/8Y6qz+vM06LHdCtILkThgt3TLX3DxYlm6wkusmATOSXz4MXKi1jifXnUsrG6yXXnpJ/+Uvf8HlchGLxRgzZgwTJ06kuLiY7OxsnE4nlZWVFBYWsnXrVrTWFBYWUlJSUj/5WXV1NT169KC0tBSlFN26daO0tJS8vDxisRg1NTX07NmT4uJi3G43+fn5lJWVkZ+fTzgcJhAI1K/3eDzk5uayZcsWunbtSiAQIBgM1q/3+Xz4/X7Ky8vp3r07VVVVhMPh+vV+vx+Px0NFRQUFBQVUVFQQiUTq1xt5TL1ycqj86CNy1q4l9/vvUcuX41+7FkcTw1THc3Op3Wsv9P77U7XHHlTtuSf5hx5K8S+/tOiYysrK8Pv9TR5TrsOBf+lSXAsW0HXpUtwrVzb47Fh+PrUjR/LLsGH4TjqJDT4fbo8nY75Pljkmvx/fBx+Q9cwz5H/4YX3THM/NZeuxxxIeN46qIUMIpNTc2cdUW1tLz549O/T7lLdiBb0fegjfggWJ4/P52DJmDPrqq9kI6f99snD2tmypYdGifsye7aeoyAtAXl6UP/yhij/+sZysrMq0OKbVq1fTu3fvjP0+yTGZc0yxWAyn02mrY7Lj98lOx/Tjjz+Sm5ubNse0ZcsWezZYSqnTgLnAW8APwBTgWq31Xcn15yJXsIynNWrDhsQzUsuX1w8+4fzxxyY3j+22W+JqVN1VqSFDiPftS3sefGjNbN+qrAzXokX1V7icP/20XX3RI49MXOE64ogmr5gJ4ziXL8fz5JN4nn0WR1kZANrhIHrUUYTGjydy4ong95tSW2fOMu9ctgzfnXfieeMNALTHQ+jsswlOniyDuHSyeBzeeMPNzJk+PvsscZ+gx6MZOzbMxIlB9t57+1uXjdSZuRNiRyR3wmjpljnbXsFSSlUDN2it70x+fRyJZmt3rXWRNFgGCIdxrl7dYBQ/5/LlOH75ZbtNtcdDbMAAYoMHJ17Jpkp36dLhZW3atIlevXrtfMMmONatw7VgQaLhWrQIR3l5g/XRQYO2NVyHHNJpz/OIbdTmzXieeSZxC2DKFcfYPvsQGj+e8BlnoNv4/e5I7cldSzlXrEjcOvjKKyit0W434XHjCE6dSnz33Tv1swV8+mliQIzXX3ejdeKXQCecEGbSpBAjR5ozIIYRuROiMcmdMFq6Za65BitNH9ltsWxgdcrX3yX/lLGNO4EqL9/WSCWflXKuXt3geaY68W7dGlyVig4ZQrx/f3C7m9hzxwuHw21+b3z33Qmfey7hc8+FeBzn8uWJwTIWLMD1ySe4Vq7EtXIlvnvvRbvdRA86KNFwHXEEsaFDDTtG2wuFcL/5Jp7583G/+y4qlhg2O961K+HTTiM8bhyxAw9s15XOjtae3LVUbPBgaubNw/Htt/juugvP88/jffxxPE88QfiMMxKNVv/+nV5HphoxIsaIETWsXevg3nt9PPmkhzffTLyGDUsMiPHb30YMfYzTiNwJ0ZjkThjNSpmz+hUsDZystX41+fXuwI/AEK31CrmC1UaN55ZKDonu2LBhu021UsT32KP+qlS0buCJXr1M/cE3FArh9Xo7fsfBIK7PP69vuJxfftlgZEOdk0Pk8MPrr3DF99knrRqAtKc1zmXL8Myfj+e55+qvhGqnk8ixxxIeP57IccdBZ3xvO0Cn5a4ZjrVr8d19N56nn0bFYmiliPz+9wT+/GfiAwcaWksmKi1VPPigl4ce8lJenhgQY489Ylx2WZDx48OG3K1qRu6EkNwJo6Vb5ux8i6A0WO3VirmltN9PbODABo1UbODAtLxFrqioyJC5EtQvv+D66KNEw7VwIc61axusj/fsWT86YeSII9C9e3d6TVakNm7E8/TTeJ98EueaNfXLo4MHJ4ZWP/10Szz7ZlTumuIoKsI3fTqeJ56ov6ocPukkglddlRgZU3Sqmhp44gkv997rpagocfmqe/c4F1wQ4oILQnTv3nn/1pqZO5G5JHfCaOmWOTvfIihaoU1zS9U9LzV4MPE997TM8OV+gwY50F26EDnpJCInnUQAUOvX4160qL7hchQX433qKbxPPQVArH9/IkcdlWi4Dj8c0mi4UcPV1uJ+/XW8Tz6Ja+HC+hzGCwsJn3464fHjiQ0ebHKRrWNU7poS79eP2rvvJvDnP+ObORPvY4/hefVVPK++Svj44xON1rBhptVnd9nZcOGFIc47L8Qrr7iZNcvHF1+4uPVWPzNm+DjrrBCXXRbiV7/q+AExzMydyFySO2E0K2UuHa9gfQCsB+5JWfwy8C/gBaC31npJcltNYg6sj5Lb9UpuO1hrvVIp9afkfo5J2ddFwIFa64Oa+nxbXMFqy9xSqQNP7GBuKSv55Zdf6NIJg2e0itY4Vq2qH53Q/fHHqJqabasdDmJDhyaucB11FNHhw9P21rcOozXOTz/F++STeF58EVVVlVjs8RA5/vjELYCjR1v2Oba0yF2SKi7GN2sW3nnzULW1AERGjSJw9dXERo40uTr70xo+/tjFzJk+3nknkWeHQ3PSSREmTQoybFiswz4rnXInMofkThgt3TJntVsEbwCuBxwpi6uAo4GTSIwaqJLbvpdcnmoNiQaqRik1BPgQyE9Zr4HbtNbTmvp8yzVYVVWJRip1FL9vvkEFg9ttqnNytt3aV9dMDRgAFpoZu6XS7TIyAJEIzqVLt014vGQJKhqtX639fqKHHFLfcMUGDQKHo5kdWofjp58Sz1U99VSD4fqjQ4cmbgEcMwbdtauJFXaMdMydKivDe++9+ObOrb/1N3L44QSvuoror38tzwgaYNUqB7Nm+Xj2WQ+RSOLv+9BDI0yaFOLYYyPt/t88HXMn7E9yJ4yWbpmzVINltrRtsNoyt1Sjq1Lxvn1t8wP7ztTU1JCdnW12Gc2rqsK1eDHu5JDwzlWrGqyOd+9O9Igj6huueN++JhXaRtXVeF5+OTEK4Ecf1S+O9+pF+MwzCY0blxgExEbSOXeqvBzvfffhvf9+HJWVAEQPPpjAVVcRHT1aGi0DbNyoeOABH4884qWqKvH3vffeMS6/PMgZZ4TbfAE7nXMn7EtyJ4yWbpmTBqsV0qXBcq5cifPrr3c+t5TbTWzffbdN0ptsqjpjbikr2bhxI70tNqCEKinB9eGHiYZrwQIcGzc2WB/bffeGEx5362ZSpc2Ix3F99FHiatUrr9TfEql9PiK//S2hceOIHnWUZZ7lay1L5K6yEt8DD+CdM6d+jrfogQcSvOoqIiecII2WASor4bHHvNx3n4+NGxO/9OrZM85FFwU599wwXbq07t9lS+RO2I7kThgt3TInDVYrpEuDlXfwwduPSNet23ZXpWL9+4PHY1KV6SvdLiO3mtY4vv9+2+2EixbVX3WAxPD4sf3229ZwjRgBWVmmlev4/vtttwCuX1+/PDpiRGIi4N//PiMG9LBU7qqq8D7yCL7Zs3GUlgKJURuDf/4zkZNPzpir3WaKROD55z3MmuVl5crEmFM5OZo//jHEZZcF6dOnZf8+Wyp3wjYkd8Jo6ZY5abBaIV0aLP/f/oZjw4b6SXpjgwYlhviW3y63SLrNldBusRjOL7+sH6HQ9cknqJQJ97THQ3TEiPqGK3bAAZ1+lUhVVOB+4QW88+fj+uyzbaX26UN43DjC48YR32OPTq0h3Vgyd7W1eB99FN/MmTiKiwGI7bMPgT//mcipp9r2amM60Ro++CAxIMbChYkBMZxOzamnhpk0KcSQIc0PiGHJ3AnLk9wJo6Vb5qTBaoV0abBE+6Tbbzk6XG0trk8/rb/C5fz6a1TK/8vx/Hyiv/71tgmP99yzY5rzWAzXBx/gnT8f9+uv1w+morOzCZ9yCuFx44gedljGXv2wdO6CQTxPPIH/7rvrJxWP7bknwalTCZ9xhmVHdrSar792MmuWlxde8BCLJf6fPfLIxMiDo0ZFm/zf2NK5E5YluRNGS7fMSYPVCtJg2UNZWRkFBQU739Am1JYtiee3kle4Gg9+Eu/du36wjMgRR6B79GjV/h3ffot3/nw8Tz9df5UDIPLrXyeuVp18clpOOG00W+QuHMYzfz6+6dNxrlsHQKxvX4JTphAeP97+UwmkiZ9/djBnjpfHH/dSU5PoqgYPjnL55SFOPTXcoN+1Re6E5UjuhNHSLXPSYLWCNFj2UF5eTlcbDPvdVo6iovrJjl2LFuHYsqXB+ti++xI58sjEpMeHHAK5udvtQ23diue55/DMn4/riy+2vfdXv0o0VWPHWm9kw05mq9xFo3iefRbfXXfVPw8a792b4OTJhM4+25bTO6SjX35RPPKIlwce8FJSkrgyvOuucS65JMg554TIzbVZ7oRlSO6E0dItc9JgtYI0WPaQbpeRTRWPJ+ZKq2u4/vc/VCBQv1q7XMSGDatvuBzl5XiefBL3W2+hIpHENrm5hE89ldC4ccRGjJBnAXfAlrmLxXC/+CL+O+/E+e23AMR79CB4+eWEzj0X0mjIXDsLheDppz3MmuVjzZrEc3F5eXHOOy/MqaeuYb/9+phcocg0tjzfibSWbpmTBqsVpMGyh9raWrJMHFUvrYVCuJYswVU3/9ayZah4fLvNtMNB9KijCI0fT+TEE8HvN6FYa7F17uJx3K+9hu+OO3AtX55YVFBA8LLLCE2Y0ORVUNHx4nF4+203M2d6Wbw4cZ/giBEhXn+9Vn7vIQxl6/OdSEvpljlpsFpBGix7WL9+PX36yG90W6SyEvdHH+FatAj3woVot5vwaacRPuOMxMiVosUyInda4377bXy3345r2TIA4l26ELrkEkIXX4zOzze5wMyxZImTsWNzKC938OqrVRx6aNTskkQGyYjznUgr6Za55hqszBzqS9ie/OKgFfLyiJx4IoFbbqFy8WKqFi0iNHmyNFdtkBG5U4rI8cdT9c47VD37LJGRI3H88gv+W24hf7/98P3rX6itW82uMiMMHx7jwgtDANx9tzwTJ4yVEec7kVaslDlpsIQtFRYWml2CyEAZlTuliB59NNWvvUbVyy8TOeIIVFUV/jvvJH///fHfcANq82azq7S9iy4KkZUV57333Hz9tcxZJoyTUec7kRaslDlpsIQtlZSUmF2CyEAZmTuliB5+ONUvvkjlG28QGT0aVVODb+ZM8g88EP9f/oLatMnsKm2rWzfNqaeWATB9ulzFEsbJyPOdMJWVMicNlrClHJmTSZgg03MXGzGC6meeofLddwn/5jeoQADf/fcnGq2rr0atX292ibZ08cU1uN2al1928/338s+6MEamn++E8ayUuTadiZVSv1dKPaiUejjldY9SKje5/ozksgeVUqe28TOuV0rFlFI65VWhlBrWaLtzG21T93o4uX76Dta72lKXEEKI5sWGDqXmv/+lcuFCwqecApEIvoceIn/YMLImT8aRnMBYdIxevWKMHRsmHlfMnClXsYQQwmxt/VXX48AE4NyU1+XAqOT6g5PLJgDz2vgZo4AngINSXjXAyCa2rWi0XRnwbXKdB1jcaP1BWmsZbsnGqqurzS5BZCDJXUOxIUOomTePyo8/JnT66RCL4X38cfIOOoisyy7DsWaN2SXaQnV1NZMmBVFKM3++h02bZLx20fnkfCeMZqXMtbXBygamaK0dWmsHsAdQf0bXWl+dXH4+0NbreQoo0VovqXsB4dTPSRFN2WY9UAB8mlznabyf5HbCxnr06GF2CSIDSe6aFh8wgNoHHqDyk08IjR8PgHf+fPJGjiR7wgQc33xjcoXW1qNHD/r3j3PyyRHCYcW998pVLNH55HwnjGalzHXUzdo7GjfRiPEUVwFzU74eAcSAuibKAwQMqEOkkdLSUrNLEBlIcte8+F57UTt7NpVLlhD605/A5cLzwgvkH3442WefjfOzz8wu0ZLqcjd1ahCAefO8lJfLVSzRueR8J4xmpcxZ/mlYrfWnWutpKYtGACu01jXJr6XBykBKyQ8XwniSu5aJ9+tH7d13U7F0KcELLkB7vXhee428E04g58QTcb/5JsTjZpdpGXW523//GKNGRaipUcyd6zW5KmF3cr4TRrNS5traYMWBs+oGuABuJ3G16orUgS9IPINl9L+SI9h2eyBIg5WRunXrZnYJIgNJ7lpH9+lD4LbbqPjySwJXXkk8Px/3J5+Qc9ZZ5B16KJ7//hdCIbPLTHupuZsyJXEV6/77vdTU7OgdQrSfnO+E0ayUubY2WM8Dw9k2wMXpJJ6NOpqGA18cmtzWEEopR7Kuxg3WxEYjCH6tlPIYVZcwnpUuIwv7kNy1je7Rg+C111Lx9dfU3nQT8d69cX73HdmTJpE/dCjeGTOgstLsMtNWau4OPzzKsGFRtm518PjjchVLdB453wmjWSlzbWqwtNZnaq1ddYNcNPNyaa3HdnTRzdgXyKNhgzWVhiMIjgOGAL0MrEsYLC8vz+wSRAaS3LVTbi6hyy6j4osvqJkzh9i+++LYtImsG2+ky+DB+G+4AbVxo9lVpp3U3Cm17VmsWbN8hMNmVSXsTs53wmhWypzd5oIaAVSRGPgCAK11g3GAlVLBuv9sagebN29mwoQJuFwuYrEYY8aMYeLEiRQXF5OdnY3T6aSyspLCwkK2bt2K1prCwkJKSkrqJ0Crrq6mR48elJaWopSiW7dulJaWkpeXRywWo6amhp49e1JcXIzb7SY/P5+ysjLy8/MJh8MEAoH69R6Ph9zcXLZs2ULXrl0JBAIEg8H69T6fD7/fT3l5Od27d6eqqopwOFy/3u/34/F4qKiooKCggIqKCiKRSP16ux5TdXW17Y7Jjt8nux1TLBYDsNUxmfZ9+u1v2TJiBL2++oqcOXPI/uwzfDNn4pkzh8pTTqH20kvZXFBgrWPq5OzVHdMRRxSy554uvv/ex5NPKo46ap0lj8mO3yc7HZPH46GoqMhWx2TH75Odjqm0tJTKysq0OabmKK1bP9CfUuou4BQSV8B+1lof2eqd7PwzFgBLtNZXpSxbB9yhtZ61g/c8AOyptR7dzH4HA8uBX2mt1zVev3jxYj1gwID2FS9MV1RURL9+/cwuQ2QYyV3ncS5bhm/mTNyvvIJKDoARPuEEgldcQWzEiMSlmwzVVO6eesrDpZdm079/jMWLK3FYfkgrkW7kfCeMlm6ZW7Zs2dLRo0cPb2pdW0+5E0kMalEEpNP9Go0HuEAptZdSqrdJ9QiT9OzZ0+wSRAaS3HWe2NCh1DzyCJWffUbw/PPRPh+eN98k78QTyT3hBNyvvZaxIw82lbsxY8LstluMNWucvPaa24SqhN3J+U4YzUqZa2uDVQPM1FqP0lqP78iC2koplQ0MolGDRWKOrCuNr0iYqbi42OwSRAaS3HW++B57ELjjDiq++orAVVcR79IF1+efk3P22eSNHInnsccgGNz5jmykqdy53XD55YkRGKdP99GGm1WEaJac74TRrJS5tjZYGzq0io4xHHCyfYMlMpDbLb+xFcaT3BlHFxYS/OtfEyMP/vvfxHbbDefatWRPmUL+AQfgu/tu1C+/mF2mIXaUuz/8IURBQZwvvnCxcKHdHrkWZpPznTCalTLX1gYrAOyllDqiJa+OLLgZI4CftNbWaW9Fp8nPzze7BJGBJHcmyMkhdPHFVC5dSvWDDxIdMgTH5s34//lP8vfbD/+116LWrze7yk61o9xlZcEllySuYt1zj8/IkkQGkPOdMJqVMtfWBms4ieewPtjJa0Hyz7bQQA+l1PC6F4k5rbRSqnfy61TbPX+Vsp++SqljlFLHACNTlgubKisrM7sEkYEkdyZyuYicdhpVCxZQ9dxzRI48ElVdje/ee8kfOpSsyy7D8c03ZlfZKZrL3YQJIXJyNAsXulm2zGlgVcLu5HwnjGalzLX1noHzOrSKpi0Argf+mLKsikQTdRFwAw2HWh8B3N3Mfs5IWbYK2NRxpYp0Y6Xfcgj7kNylAaWIjhpF9ahROL/6KjHy4Isv4p0/H+/8+USOOYbgFVcQPeww24w82Fzu8vM1558fYsYMH9On+3jssRoDKxN2Juc7YTQrZa5Nw7TbmQzTbg+bN29ml112MbsMkWEkd+nJUVSEd84cvI8/jgoEAIgOHUpw0iQiJ50ETmtf2dlZ7kpKFAcckE8opFi8uIJ99snM0RZFx5LznTBaumWuM4ZpFyKtBZI/RAlhJMldeor360fglluo+PprAtOmEe/eHdeyZeScdx55Bx+M55FHwMLfu53lrkcPzVlnhQGYMUOexRIdQ853wmhWylxbJxq+BTiNxKh9zYkBz2qt/9KG2kwhV7DsIRQK4fV6zS5DZBjJnUXU1uJ98km8s2fjXLcOgHhBAaGLLiI0YQK6a1dz62ulluTuxx8dHHRQHg4HLFtWQZ8+cveKaB853wmjpVvmOuMK1tTkn0U7eYHMQSVMYKW5EoR9SO4sIiuL0IQJVH72GdUPPUT0gANwlJXhv/lm8ocMwT9tGo6ffjK7yhZrSe5+9as4p54aIRpVzJolV7FE+8n5ThjNSplra4PlZttEwzt8AbNp+0AaQrSZx+MxuwSRgSR3FuNyETn1VKree4+ql14iMno0qrYW3wMPkDdsGFkXXYRz+XKzq9ypluZuypTEBMyPP+6lrMweA3wI88j5ThjNSpnr7Gew5B4EYYrc3FyzSxAZSHJnUUoR/fWvqX7mGSoXLSJ05pkAeJ99lrwjjyTntNNwLVwIaTooVEtzN2hQjOOOCxMIKO6/P31usxHWJOc7YTQrZU4GuRC2tGXLFrNLEBlIcmd9scGDqb3vPiqXLSN4ySXo7GzcH3xA7qmnknv00bifew6iUbPLbKA1uau7ijV3rpeqqs6qSGQCOd8Jo1kpc21tsOLAWUqph5t7AeOT2wphqK4We0hd2IPkzj7iu+1G4OabEyMP/u1vxAsLcX31FTkXXkje8OF4H3wQatJjTqnW5G7kyBgjR0aoqHAwb55cxRJtJ+c7YTQrZa6tDdZzwHDg3J28hgPPtr08IdrGSkN5CvuQ3NmP7tqV4J//TMVXX1Fz113E9twT508/kfV//0f+/vvju+UWlMm/VW1t7qZOTVzFmjPHRyjUGRWJTCDnO2E0K2WuTQ2W1nqs1tqltXbs5OXSWo/v6KKF2JlgMGh2CSIDSe5szOcjfO65VH7yCdWPPkp06FAcW7fiv+028vfbD/811+BIDvlutNbm7phjogweHKW42MGTT1rnoXGRXuR8J4xmpczJM1jClnr27Gl2CSIDSe4ygNNJ5OSTqXrnHapefZXwccehAgF8c+eSN3w42RMm4PzyS0NLam3ulILJkxM/qMyc6SMW64yqhN3J+U4YzUqZa1ODpZTaXSn1B6XUOUqp3zRal51cfk5ym907pFIhWsFKcyUI+5DcZRCliB56KDXz51Px0UeExo8HhwPPCy+Qd/TR5Jx6Kq733jNk5MG25O53v4uw++4xfvzRyUsvuTuhKmF3cr4TRrNS5tp6BWsR8BgwD7iu0bqC5PJHktssbONnCNFmPp9MpCmMJ7nLTPGBA6mdPZuKL74gOHEiOicH98KF5J5xBrlHHonnmWcgEum0z29L7lwuuOKKxFWs6dN96ToCvUhjcr4TRrNS5traYPUhMYnwKGBio3XFyeVHA3OS2wphKL/fb3YJIgNJ7jKb3nVXAv/8JxXLl1N7ww3Ee/TAtWIF2RdfTN6wYXjvuw+qqzv8c9uau3HjwvToEWfFChfvvuvq4KqE3cn5ThjNSplrzzNYa7XWC7XWX6Qu1FqHkssXAmvaV54QbVNeXm52CSIDSe4EgM7PJzR5MhVffknNPfcQ698f5/r1ZP31r+Tvtx++f/0LVVraYZ/X1tz5fHDppduuYgnRGnK+E0azUubSdpALpdT1SqmYUkqnvCqUUsMabXduo23qXg+nrI80WhdQSp1kzpEJI3Tv3t3sEkQGktyJBrxewmefTeXixVT/5z9EDz4Yxy+/4L/zTvL324+sK6/E8f337f6Y9uTu3HND5OfHWbzYzSefONtdi8gccr4TRrNS5trTYB2cMphFky/g4HbsfxTwBHBQyqsGGNnEthWNtisDvk2uO5LEM2Op679JLhc2VVVVZXYJIgNJ7kSTHA4iJ55I1ZtvUvn664RPPBEVCuGdN4+8gw8m+09/wrl0aZt3357c5eXBBRckJsO65x65iiVaTs53wmhWylxbb7peDZyVfO3MqjZ+hgJKtNZL6hcoFU4ubyxat51SqieJgTY+TdlPeaP9VO1gP8ImwuGw2SWIDCS5EzsTGzmSmpEjCaxejW/2bDxPP43nlVfwvPIKkcMOI3jFFUSPOSYxlnoLtTd3F18c4t57fbz1lodvvgkwcGC8XfsTmUHOd8JoVspcW69gjQCOIHGVqbnXEcAh7S+zWauAuY1qiwFLmt5cZAIrzZUg7ENyJ1oqvs8+1M6YkRh5cPJkdG4u7o8/JnfsWPIOPxzP/PnQwh8m2pu7ggLN2WcnrmLJs1iipeR8J4xmpcy1qcHSWldqrT+qG8yimddHWuvKji66US2faq2npSwaAazQWtd05ueK9GaluRKEfUjuRGvpXr0I3HADvyxfTu3f/068Vy+cq1aRfdll5B94IN5Zs6Cy+X9GOyJ3EyeGcLk0zz/vYd26tH08W6QROd8Jo1kpc22daFgppXZVSvVVSvVoYn3v5LpdlWrFfQ4dYwTbbg8UGcpKQ3kK+5DciTbLyyM0aRIVX3xBzezZxPbZB8emTWRdf31i5MF//AO1gx8uOiJ3u+0W5/TTw8TjilmzvO3en7A/Od8Jo1kpc239NdUC4CfgR+CVuoVKqV2UUsuBn5PrfkpuawillAMYjjRYGc/j8ZhdgshAkjvRbh4P4fHjqfz4Y6rmzydy6KE4KivxT59O/gEHkDV5Mo41axq9pWNyVzfx8H//66WkRB5TFs2T850wmpUy19YG63DgbeAfwIMASik38DywF/Dv5Lo3gcPaX2aL7QvkIQ1WxquoqDC7BJGBJHeiwzgcRI87jupXX6XyrbcIn3QSRCJ4H3+cvJEjyT77bJyffQZ0XO4GDIjz29+GCYUU990nz2KJ5sn5ThjNSplr6yiCCnhDaz0jZdl9JAa0OEdr/V8ApdRE4IT2ldgqI4Aq2j5yIZs3b2bChAm4XC5isRhjxoxh4sSJFBcXk52djdPppLKyksLCQrZu3YrWmsLCQkpKSsjJyQGgurqaHj16UFpailKKbt26UVpaSl5eHrFYjJqaGnr27ElxcTFut5v8/HzKysrIz88nHA4TCATq13s8HnJzc9myZQtdu3YlEAgQDAbr1/t8Pvx+P+Xl5XTv3p2qqirC4XD9er/fj8fjoaKigoKCAioqKohEIvXr7XpMHo+HoqIiWx2THb9Pdjsml8tFeXm5rY7Jjt8nyx3TLrvgv+MOsqZMwT9nDt1ffRXPa6/hee01wgcdhHviRMo66JjGji3ltdf25qGH3Iwb9yN9+uTK90mOqcljys/Pp6ioyFbHZMfvk52OSWtNUVFR2hxTc5TWutVNiFIqDkypa7CUUlOAu4DbUgecSDZYM7TWrZ69UCm1AFiitb4qZdk64A6t9awdvOcBYE+t9eiUZfOAHK316c3tu87ixYv1gAEDWluuSDMbN26kd+/eZpchMozkThhBbd6M98EH8c6di6Oigkj37tR89hm6a9cO2f/vfpfDhx+6ue66AFOnBjtkn8J+5HwnjJZumVu2bNnS0aNHD29qXXuGCtpTKXWEUupc4HbgVa31NKWUN7n8CKB/O/bfFjLAhQAgEomYXYLIQJI7YQS9yy4E//Y3Kr7+msghh+DesgX/tdd22P6nTEk0Vffd5yUQ6LDdCpuR850wmpUy19YGaz0wCfgAeJjELXl1kw73JDGwxQfAFcltO51SKhsYhDRYAmvNlSDsQ3InDJWbS+0996B9PrxPPonrvfc6ZLdHHRXlgAOilJY6+O9/ZURB0TQ53wmjWSlzbW2wjgTOAc4HzgQO01pXJ9eVAecl150DHNXOGltqOOBEGiyBteZKEPYhuRNGi++1F5suvhiArKlToaqq3ftUattVrJkzvVjol8bCQHK+E0azUubaOtHwj1rr/2itH9VaP6u1rkpZV5Nc/mhymx87rtxmjQB+0lpb529fdJrs7GyzSxAZSHInzFBz8cVE998f5/r1+G+6qUP2edJJEfr3j/Hzz05eeME6QyML48j5ThjNSplL5+naNdBDKTW87gV4AJ2cyLjxQ2U7ev5KA10b7Sc3uVzYlNPZ6nFVhGg3yZ0wg9PrpXbmTLTLhXfuXJyffNLufTocMGlS4irW9Ok+4vF271LYjJzvhNGslLl0brAWkHiu6/OUVw6JJuqi5NepdtRgfQgc0Wg/g5LLhU1VVlaaXYLIQJI7YYbKykpigwcTnDwZpTXZkydDsP2j/515ZpjeveN8+62Tt95yd0Clwk7kfCeMZqXMpW2DpbX+u9baqbVWKa88rfUSrfWNWmvVaPs+Wus7m9jPw1prd6P9+LTWLxt3NMJohYWFZpcgMpDkTpihLnfBq64i1r8/zjVr8N1+e7v36/HAxImJRu3uu320YVYXYWNyvhNGs1Lm0rbBEqI9tm7danYJIgNJ7oQZ6nPn9VIzYwZaKXwzZuD8+ut27/ucc0J07RpnyRIX//ufq937E/Yh5zthNCtlThosYUttmUBbiPaS3AkzpOYuNmIEoQsvRMViZE2aRHuHAMzOhosuCgGJq1hC1JHznTCalTInDZawJStdRhb2IbkTZmicu8C11xLr2xfX8uV4Z89u9/4vuihEdrbm/ffdfPWVdR4yF51LznfCaFbK3A4bLKXUg0qpiFIqlvJ6qZnthzXaNqaUCiulHuic0oXYsZKSErNLEBlIcifMsF3ucnKovftuAPy33opjzZp27b9rV82f/pS4ijV9ulzFEglyvhNGs1LmmruCdS6wDliU8lrZzPZVJEbmS91+I4lJh4UwVE5OjtkliAwkuRNmaCp30VGjCJ11FioUIvuKK2jvOOuXXRbE7da8/LKbtWvl5hch5zthPCtlrrmzpBOYqbUelfL664421lp/p7U+KnV74PGdfIYQQgghOkHgppuI9+iB69NP8T78cLv21bu3Zty4MForZs6Uq1hCCNEcaX6ELVVXV5tdgshAkjthhh3lTnfpQm1yuHb/P/6B4+ef2/U5kyYFUUozf76HjRvVzt8gbE3Od8JoVsqcNFjClnr06GF2CSIDSe6EGZrLXeSkkwifcgqqupqsqVNpz2RWe+0V55RTIkQiinvvlatYmU7Od8JoVsrczhos64yHKESK0tJSs0sQGUhyJ8yws9zV3nor8S5dcL//Pp6nnmrXZ02dmph4+NFHvWzdKlexMpmc74TRrJS55hqsamB6EyMDtvgF/JXE4BdCGEop+YdfGE9yJ8yws9zpHj0I3HwzAP6//hW1eXObP2u//WIcfXSEmhrF3LneNu9HWJ+c74TRrJS55qZlHweMITHYRVvFgOfb8X4h2qRbt25mlyAykOROmKEluQuPHYvn2Wdxv/8+WddcQ828eW3+vClTgrz/vpsHHvAycWKQ7Ow270pYmJzvhNGslLkdNlha69eB1w2sRYgOU1paSr9+/cwuQ2QYyZ0wQ4typxS1d99N3qGH4nn5ZcKvvELk5JPb9HmHHRZl+PAoS5a4eOwxL5deGmrTfoS1yflOGM1KmZNBLoQt5eXlmV2CyECSO2GGluYuvttuBK6/HoCsa65B/fJLmz5PqW3PYs2e7SMcbtNuhMXJ+U4YzUqZ2+EVLKXUOR2w/yiwUGu9oQP2JUSLxWIxs0sQGUhyJ8zQmtyFJkzA8/zzuD79FP9111E7c2abPvP44yMMGBDj22+dPP20hz/+UbqsTCPnO2E0K2WuuWew5pEYRbA9T5Rp4Aegf2vepJS6HriBhlfYKoGjtdZLG207F5jQaBergAO11iGl1O+BF5r4mGO11u+2pi5hHTU1NRQUFJhdhsgwkjthhlblzuGgZsYM8o44Au9//0t4zBiio0a1+jMdDpg8Ocill2Yzc6aP8ePDONvzxLawHDnfCaNZKXPNNVitP+Nu7wzg0ja8bxTwBHBPyrKXgZHA0kbbjgZuB55Ofr0nMB/oCRQBHiAMHNbofavbUJewiJ49e5pdgshAkjthhtbmLt6/P8FrrsH/z3+SNXUqlR99BDk5rf7cMWPC3HyzjzVrnLz2mptTTom0eh/CuuR8J4xmpcw1N8jFwvbuXCl1QFvfCpRorZek7CtM01fTFPBT3bZKqWDKckg0WLWp+xL2V1xcbJkHIYV9SO6EGdqSu+Dll+N+6SVcX3+N/1//IvDvf7f6c91umDQpxDXXZDF9uo+TT45goVGURTvJ+U4YzUqZy4RBLjxAwOwihLHcbrfZJYgMJLkTZmhT7txuamfMQDudeB94AOenn7bps//whxCFhXG+/NLFggXN3RQj7EbOd8JoVsqcHRusMuBOoCL5tTRYGSg/P9/sEkQGktwJM7Q1d7H99iN4xRUorcm+4goIBnf+pkb8frjkksQw7ffc42tTHcKa5HwnjGalzO2wwVJK9W3vCzB8RjCtdbHW+iqtdXlykTRYGaisrMzsEkQGktwJM7Qnd8GrrybWvz/ONWvw3Xlnm/Zx/vkhcnI0ixa5WbpURrrIFHK+E0azUuaau4K1Dvixna/rgNLOKb3FPMAgpZROeW1VSg00uS7Riaz0Ww5hH5I7YYZ25c7no+aee9BK4bvnHpwrVrTh8zUTJiSuYk2fLlexMoWc74TRrJS55m6Y/geJYdbbIwa83s59tMRMpVTqZB5rgQO01jXAf4AFjbZ/FzgI+KbxjjZv3syECRNwuVzEYjHGjBnDxIkTKS4uJjs7G6fTSWVlJYWFhWzduhWtNYWFhZSUlJCTHIWpurqaHj16UFpailKKbt26UVpaSl5eHrFYjJqaGnr27ElxcTFut5v8/HzKysrIz88nHA4TCATq13s8HnJzc9myZQtdu3YlEAgQDAbr1/t8Pvx+P+Xl5XTv3p2qqirC4XD9er/fj8fjoaKigoKCAioqKohEIvXr7XpMVVVVVFRU2OqY7Ph9stsxRaNRtNa2OiY7fp/sdkwbN24kGo22/Zh69eJXf/wj+Y8/juvii6l9+22Ky8padUx/+EN35szZldde8/D11xXk52+Q75PNj8ntdlNRUWGrY7Lj98lOx1RSUkJFRUXaHFNzlNbt7aE6nlJqAbBEa31VyrJ1wB1a61mNtl0HPAk8lVzUG3gNGKy1XrmD/ZcBV2mt5zVet3jxYj1gwID2H4QwVVFRkWVGmhH2IbkTZuiQ3FVVkXfYYTjXr6f2xhsJXXFFq3dx1VV+Hn7Yx/jxIWbPrm1fPSLtyflOGC3dMrds2bKlo0ePHt7Uuk4f5EIptWdnfwawQWv9pdb6S7ZdlZLBYjOYleZKEPYhuRNm6JDc5eZSe/fdAPhvuQXH2rWt3sXll4dwODTPPONh/Xr5J9ju5HwnjGalzHVYg6WU6qKUOkIpdblS6n6l1GKlVBXwXUd9Rhvr6qmU2tvMGoTxiouLzS5BZCDJnTBDR+UuOno0ofHjUcEgWVOmQDzeqvfvvnucMWPCRKOKWbPkWSy7k/OdMJqVMtfqBksp5VRKDVZKnaWU+rdS6jWl1M/AFuAD4B7gHCAHeBu4tUMrbr1LgCdMrkEYzOPxmF2CyECSO2GGjsxd4KabiBcW4v7f//DMm9fq90+Zkhjq/fHHvZSVyVUsO5PznTCalTLXbIOllOqtlDpBKXW1UupxpdRXQA3wFYnBIy4EcoFXgSkkbsubDmRprYdorU/TWv+1Mw9AiKbk5uaaXYLIQJI7YYaOzJ3u2pXa224DIOvGG1Hr17fq/QMHxjn++DCBgOL++70dVpdIP3K+E0azUuaamwerFPiZxIAR1wC7Af8DrgKOBXpprQu01kdorS/VWteN4lej03HkDJFRdja6ixCdQXInzNDRuYv87neETzoJVV1N9pVXQiv/SZ88OXEVa+5cL5WVHVqaSCNyvhNGs1LmdnaLoEq+AkAtUJ18VSX/7Cwa6KGUGl73IjGflU5eVRveaNu+KdsNSVle92euUuqYuhfgpv1D0Is01rVrV7NLEBlIcifM0Bm5q73tNuL5+bjffRfPM8+06r0jR8Y45JAIFRUO5s2Tq1h2Jec7YTQrZW6HDZbWuhDYFfgNMJPEM1bHAfcBnwCVSqkflVKvKqVuVUr9KfnW5ubWaqkFwFnA5ymvHOBT4KLk13XeB65O2e5lEvNgrUuuXwz0Bd5JeWlgWQfUKdJUIBAwuwSRgSR3wgydkTvdsyeBm24CwP+Xv6BKS1v1/rpnsebM8REMdnh5Ig3I+U4YzUqZa/YKltZ6k9b6La317Vrrs7XW+5NodA4gMZDF04AT+CPwSPJt/6eUKlNKfayUekQpNa21RWmt/661dmqtVcorT2u9RGt9o9ZapWw7odF2SmvdPznJMFrrt7XW/kbru2itl7e2LmEdQfkXXZhAcifM0Fm5C591FpGjjsJRXk7WtNb9U37MMVGGDIlSUuJg/nzrPJguWk7Od8JoVspcq0cR1FpHtdbLtdb/1Vr/n9b6N1rrXYECYDRwJfASiVv6zgT+1aEVC9ECVporQdiH5E6YodNypxS106ejs7PxvPAC7tdfb81b65/FmjHDRzTaOSUK88j5ThjNSpnrsHmwtNZbtdYfaK3vSV5VOojE1a59O+ozhGgpK82VIOxDcifM0Jm5i/ftS+DaawHIuuoqVEVFi997yikRfvWrGOvWOXnpJXdnlShMIuc7YTQrZa7DGqym6ARTJxoWmcnnk0kuhfEkd8IMnZ270AUXED3oIBzFxfivv77F73O5YNKkxFWse+7xtXYwQpHm5HwnjGalzHVqgyWEWfx+v9kliAwkuRNm6PTcOZ3UzJiB9njwPv44roULW/zW8ePD9OwZZ8UKF+++2xFjYIl0Iec7YTQrZU4aLGFL5eXlZpcgMpDkTpjBiNzF99mH4NVXA5A1dSrU1LTofV4vXHpp4irW3Xdb57fPYufkfCeMZqXMSYMlbKl79+5mlyAykOROmMGo3AWvuILooEE4163Df/PNLX7fueeGyM+P88knbj75xNmJFQojyflOGM1KmZMGS9hSVVWV2SWIDCS5E2YwLHduN7UzZ6KdTrz33Yfz8893/h4gNxcuuCAEwPTpchXLLuR8J4xmpcxJgyVsKRwOm12CyECSO2EGI3MXO+AAQpdfjtKa7CuugFCoRe+7+OIQfr/m7bc9rFwpV7HsQM53wmhWypw0WMKWrDRXgrAPyZ0wg9G5C1xzDbE998S5ejW+u+5q0XsKCjRnny1XsexEznfCaFbKnDRYwpasNFeCsA/JnTCD4bnz+6m95x4AfHffjXPlyha97fLLg7hcmhdecLNunfz4YXVyvhNGs1Lm5AwnbMlKQ3kK+5DcCTOYkbvooYcSPP98VDRK1hVXQDS60/f06aM544ww8bhi5ky5imV1cr4TRrNS5qTBErbk8XjMLkFkIMmdMINZuQtcfz3xXXfF9cUXeO+7r0XvueKKIEppnnjCQ0mJ6uQKRWeS850wmpUyJw2WsKWKigqzSxAZSHInzGBa7vLyqEk+g+W/+WYcP/yw07fss0+c3/42QiikmDNHrmJZmZzvhNGslDlpsIQtFRQUmF2CyECSO2EGM3MXPfZYQmeeiQoGyZo8GeLxnb5n8uTExMMPP+ylokKuYlmVnO+E0ayUubRrsJRS1yulYkopnfKqUEoNa2LbuY2200qpb5RS3uT6wUqp8kbrY0qpls+QKCzJSr/lEPYhuRNmMDt3gX/9i3hBAe6PP8bz2GM73X7YsBhHHBGhulrx0ENeAyoUncHs3InMY6XMpV2DBYwCngAOSnnVACOb2HY0cHvKduOAfYG6cRyHAarRvh4Cju288kU6iEQiZpcgMpDkTpjB7Nzp7t2pvfVWALKuvx61YcNO3zNlSuIq1n33eamt7dTyRCcxO3ci81gpc+nYYCmgRGu9pO4FhJPLm9r2p5TtVqYsr/sz2mhfG3ewL2EjVporQdiH5E6YIR1yF/n97wmfeCKqupqsP/8ZtG52+yOPjHLggVHKyhz8979yFcuK0iF3IrNYKXPp2GAJ0W5WmitB2IfkTpghLXKnFLW33048Lw/P22/jfv75nW1efxVr5kwvFvrFtEhKi9yJjGKlzNmtwSoD7gSsc5Om6BTZ2dlmlyAykOROmCFdcqd79SLwz38CkDVtGqqsrNntf/vbCP37x1i/3snzz1tn+GWRkC65E5nDSpmzVYOltS7WWl+ltS43uxZhLqfTaXYJIgNJ7oQZ0il34T/+kcgRR+DYsoWsv/yl2W0djsS8WADTp/taMgChSCPplDuRGayUOVs1WELUqaysNLsEkYEkd8IMaZU7paidPh3t9+N57jncb77Z7OZnnBGmd+84q1c7efNNt0FFio6QVrkTGcFKmXOZXUAHmKmUmpny9VrgAK11TVt2tnnzZiZMmIDL5SIWizFmzBgmTpxIcXEx2dnZOJ1OKisrKSwsZOvWrWitKSwspKSkhJycHACqq6vp0aMHpaWlKKXo1q0bpaWl5OXlEYvFqKmpoWfPnhQXF+N2u8nPz6esrIz8/HzC4TCBQKB+vcfjITc3ly1bttC1a1cCgQDBYLB+vc/nw+/3U15eTvfu3amqqiIcDtev9/v9eDweKioqKCgooKKigkgkUr/ersfk9XopKiqy1THZ8ftkt2NyuVyUl5fb6pjs+H2y2zFFIhHKysrS55h69aL6ssvY9c478V15JT/06UPX3Xff4TH94Q813H57H267zcHAgUX06mXP75Pdjik/P5+ioiJbHZMdv092OiatNUVFRWlzTM1Reicj/RhNKbUAWKK1vipl2TrgDq31rEbbrgOeBJ5KLuoNvAYM1lqvVEqdm3xfQcp7bgRO0loPb+rzFy9erAcMGNBRhyNMsn79evr06WN2GSLDSO6EGdIyd7EYuSecgGvpUkLnnkvtXXftcNOaGth//3y2bnXw8stVHH541MBCRVulZe6EraVb5pYtW7Z09OjRTfYTdrhFcIPW+kut9ZfAN8llMgx7hku3XxyIzCC5E2ZIy9w5ndTMmIF2u/HOm4fro492uGl2Nlx0UQiAu+/2GVWhaKe0zJ2wNStlzg4NlhDbKSwsNLsEkYEkd8IM6Zq7+L77EvzznwHImjyZ5mYUvvDCEDk5mg8+cPPll9Z5kD2TpWvuhH1ZKXPSYAlbKikpMbsEkYEkd8IM6Zy74JQpRAcOxPnjj/hvuWWH23XtqvnTnxJXsaZPl6tYVpDOuRP2ZKXMSYMlbKnuwUYhjCS5E2ZI69x5PNTOmIF2OPDeey/OZct2uOlllwXxeDSvvOJmzRr58STdpXXuhC1ZKXNyBhNCCCFEp4kNHUrosstQ8TjZkyZBONzkdr16acaODaO1YuZMuYolhLCudGywNNBDKTW87gV4AK2U6p38OnXbvinbDUlZXvenq9G+eqesFzZVXV1tdgkiA0nuhBmskLvAtGnE9tgD56pV+KZP3+F2V1wRxOHQPPWUhw0bZLyqdGaF3Al7sVLm0rHBWgCcBXye8soBPgUuSn5d533g6pTtXiYxD9a65Pq6exFS93UB8F4n1i/SQI8ePcwuQWQgyZ0wgyVyl5VFbbKx8t15J45Vq5rcbM8945xySoRIRHHvvXIVK51ZInfCVqyUubRrsLTWf9daO7XWKuWVp7VeorW+UWutUrad0Gg7pbXuXzfJsNZ6uda6S6P1Dq31NPOOUBihtLTU7BJEBpLcCTNYJXfRww8ndO65qEiE7CuugFisye2mTAkC8NhjXrZulatY6coquRP2YaXMpV2DJURHUEr+URbGk9wJM1gpd7U33ki8Vy9cS5fivf/+JrfZb78Yo0dHqKlRPPig1+AKRUtZKXfCHqyUOWmwhC1169bN7BJEBpLcCTNYKnd5edTedRcA/n/9C8e6dU1uVncV64EHvFjosYuMYqncCVuwUuakwRK2ZKXLyMI+JHfCDFbLXeT44wmdfjoqECBryhTQ2487deihUQ46KEp5uYPHHpOrWOnIarkT1melzEmDJWwpLy/P7BJEBpLcCTNYMXeBm28m3r077kWL8Dz++HbrlYKpUxNXsWbP9hEKGV2h2Bkr5k5Ym5UyJw2WsKXYDh6eFqIzSe6EGayYO11QQO0ttwCQdd11qI0bt9vmuOMi7LtvjE2bHDz9tMfoEsVOWDF3wtqslDlpsIQt1dTUmF2CyECSO2EGq+YuMmYM4eOPR1VVkXX11dvdKuhwwOTJiatYM2f6djTooDCJVXMnrMtKmZMGS9hSz549zS5BZCDJnTCDZXOnFLV33IHOzcXzxhu4X3hhu03GjAnTt2+MtWudvPqq24QixY5YNnfCsqyUOWmwhC0VFxebXYLIQJI7YQYr507vuiu1f/87AFnTpqG2bGmw3uWCSZMSD2BNn+5rajwMYRIr505Yk5UyJw2WsCW3W37TKYwnuRNmsHruwuecQ+Tww3GUleH/29+2W3/WWSEKC+N89ZWLDz5wmVChaIrVcyesx0qZkwZL2FJ+fr7ZJYgMJLkTZrB87hwOaqdPR/v9eJ9+Gtc77zRY7ffDJZckrmLdc4/PjApFEyyfO2E5VsqcNFjClsrKyswuQWQgyZ0wgx1yF99jDwJ/+QsA2VOnQmVlg/UTJgTJzdV8+KGbJUucZpQoGrFD7oS1WClz0mAJW7LSbzmEfUjuhBnskrvQJZcQHToUx8aN+P/xjwbr8vISTRYknsUS5rNL7oR1WClz0mAJWwqHw2aXIDKQ5E6YwTa5c7momTED7Xbje/hhXP/7X4PVl1wSwufTvP66h1Wr5McXs9kmd8IyrJQ5OUMJWwoEAmaXIDKQ5E6YwU65iw8cSHDqVACyJk+GlGPbZRfNH/6QeBZr5ky5imU2O+VOWIOVMpeWDZZS6nqlVEwppVNeFUqpYU1s+2Wj7bRS6qOU9S82sX69sUckjGaluRKEfUjuhBnslrvg1KnEBgzA+f33+G+9tcG6yy8P4XRqnn3Ww88/p+WPMBnDbrkT6c9KmUvXs9Mo4AngoJRXDTCyiW09wKxG205otP75RutP7KzCRXqw0lwJwj4kd8IMtsud15u4VVApvLNm4fzii/pV/frFGTMmTDSqmDXLa2KRwna5E2nPSplL1wZLASVa6yV1LyCcXN6YB1ibuq3WenWj9T81Wv+1AccgTOTxeMwuQWQgyZ0wgx1zFxs+nNAll6DicbKuuAIikfp1kycnBrt4/HEvpaVN/VggjGDH3In0ZqXMpWuD1RoeoLmbMne2XthQbm6u2SWIDCS5E2awa+4Cf/0rsd13x7VyJb577qlfPnBgnBNOCBMMKu6/X65imcWuuRPpy0qZkwZL2NKWLVvMLkFkIMmdMINtc5edTe306QD47rgDx7ff1q+qu4o1d6638ZRZwiC2zZ1IW1bKnDRYwpa6du1qdgkiA0nuhBnsnLvoEUcQOvtsVDhM9uTJEIsBMGJEjEMPjVBZ6WDePLmKZQY7506kJytlzi4N1jMpIwTGlVIPNFp/e6NRBN8yqVZhECsN5SnsQ3InzGD33AX+8Q/ivXrh+vxzvHPn1i+fMiVxFWvOHB/BoFnVZS67506kHytlzmV2AR3gCBo2imcCZ6R8PRbwp3x9OHD3jna2efNmJkyYgMvlIhaLMWbMGCZOnEhxcTHZ2dk4nU4qKyspLCxk69ataK0pLCykpKSEnJwcAKqrq+nRowelpaUopejWrRulpaXk5eURi8WoqamhZ8+eFBcX43a7yc/Pp6ysjPz8fMLhMIFAoH69x+MhNzeXLVu20LVrVwKBAMFgsH69z+fD7/dTXl5O9+7dqaqqIhwO16/3+/14PB4qKiooKCigoqKCSCRSv96ux1ReXk4wGLTVMdnx+2S3Y6qtrcXj8djqmOz4fbLbMZWUlOB0Om11TI2/T7v96190P/98fH//O9WjRrHR6+WAA3IYONDNN9/4eOwxB7/5TZGljsnq36dYLEZRUZGtjsmO3yc7HVNpaWmD95t9TM1RWutmNzCDUmoBsERrfVXKsnXAHVrrWTt57+XAVVrr3Xew/iTgFa11k0MPLV68WA8YMKCNlYt0EQqF8HrlthFhLMmdMEOm5C57wgQ8L7xA5MgjqX7+eVCKF15wM2FCDrvvHuOzzypx2eHXxhaRKbkT6SPdMrds2bKlo0ePHt7UOjvcIijEdqw0V4KwD8mdMEOm5K72lluId+uGe+FCPE88AcApp0TYY48Y69Y5eeklt8kVZpZMyZ1IH1bKnOUbLKXUAUqpvGbWD1RKFRhZkzCfz+czuwSRgSR3wgyZkjtdWEjg3/8GwH/ttajiYpxOmDQp8QDW9Ok+0vCmHNvKlNyJ9GGlzFm+wQK+IPEc1o68DowzqBaRJvx+/843EqKDSe6EGTIpd+HTTydy7LE4KirIuuYa0Jpx48L06hVn5UoX77wj9wgaJZNyJ9KDlTJnhwZLiO2Ul5ebXYLIQJI7YYaMyp1S1Nx5JzonB8+rr+J++WW8Xrj00sRVrLvvts4PYFaXUbkTacFKmZMGS9hS9+7dzS5BZCDJnTBDpuVO9+lD7Y03ApB1zTWo8nL+9KcQXbrE+fRTF5984jS3wAyRabkT5rNS5tK1wdJAD6XU8LoXifmstFKqd/LrVPsrpY5RSh0D7JN8f+q+9klZv78hRyBMVVVVZXYJIgNJ7oQZMjF34XPPJXLooThKS/H/7W/k5sIFF4QAuYpllEzMnTCXlTKXrg3WAuAs4POUVw7wKXBR8us67wM3Ae8kX5cDH6Ss/yC5rG79Tcn3CBsLh8NmlyAykOROmCEjc+dwUDt9Otrnwzt/Pq533+Xii0NkZWneecfNihVyFauzZWTuhKmslLm0bLC01n/XWju11irllae1XqK1vjF1Diut9ehG2ymt9fkp689vYv1oc45MGKVnz55mlyAykOROmCFTcxffay8C06YBkHXllXT3VHL22YmrWNOnW2e0MavK1NwJ81gpc2nZYAnRXlaaK0HYh+ROmCGTcxe67DKiBxyAc/16/DfdxMSJQVwuzYsvuvnxR/kRpzNlcu6EOayUOTn7CFuy0lCewj4kd8IMGZ07l4vaGTPQLhe+Bx+k3/rFnHFGmHhcMXOmXMXqTBmdO2EKK2VOGixhSx6Px+wSRAaS3AkzZHruYoMHE5w8GYDsK65gyiW/oJTmiSc8FBernbxbtFWm504Yz0qZkwZL2FJFRYXZJYgMJLkTZpDcQfCqq4jtvTfOtWsZ8uKt/Pa3EcJhxZw5chWrs0juhNGslDlpsIQtFRQUmF2CyECSO2EGyR3g9VIzYwZaKXwzZnDtbz8F4JFHvPzyi1zF6gySO2E0K2VOGixhS1b6LYewD8mdMIPkLiF28MGELroIFYsx7N6JHP3rWqqrFQ895DW7NFuS3AmjWSlz0mAJW4pEImaXIDKQ5E6YQXK3TeBvfyPWty+u5cuZ0e82AO67z0ttrcmF2ZDkThjNSpmTBkvYkpXmShD2IbkTZpDcpcjJofbuuwEY+MwtjBm4gi1bHPznP3IVq6NJ7oTRrJQ5abCELVlprgRhH5I7YQbJXUPRUaMI/eEPqFCIOZELUcSZNcuLhX75bQmSO2E0K2VOGixhS9nZ2WaXIDKQ5E6YQXK3vcA//0m8Rw92WfMJNxbOYv16J889Z50hnq1AcieMZqXMSYMlbMnpdJpdgshAkjthBsnd9nSXLtTefjsAf6n4K30pYvp0H/G4yYXZiOROGM1KmZMGS9hSZWWl2SWIDCS5E2aQ3DUtctJJhH/3O9zhGh71XsR33zl44w232WXZhuROGM1KmZMGS9hSYWGh2SWIDCS5E2aQ3O1Y7a23Eu/ShaNCb3MOj3H33T60Nrsqe5DcCaNZKXPSYAlb2rp1q9kliAwkuRNmkNztmN5lFwI33wzAdDWVDcvK+Ogjl8lV2YPkThjNSplLywZLKXW9UiqmlNIprwql1LAmtv2y0XZaKfVRW/Yl7EPLryiFCSR3wgySu+aFx44lMno0XXU5M5nE3Xf7zC7JFiR3wmhWylxaNljAKOAJ4KCUVw0wsoltPcCsRttOaOO+hE1Y6TKysA/JnTCD5G4nlKL2rruIZ+dwBs/SdcErfPmldR6WT1eSO2E0K2UuXRssBZRorZfUvYBwcnljHmBt6rZa69Vt3JewiZKSErNLEBlIcifMILnbufhuuxG84XoAZjORB24NmFyR9UnuhNGslLl0bbBawwPImVI0kJOTY3YJIgNJ7oQZJHctEzr/fGoPHEEvijnurWl8950dfgQyj+ROGM1KmbPD2UUaLCGEEEI0z+Egct8Mwg4v5zGPD/760c7fI4QQbWCHoXSkwRLbqa6upnv37maXITKM5E6YQXLXcvH+/dl82TT6zPo7496fyLpP36F7H4/ZZVlS6YbNeIIyr5gwTumGzeR6c/HkpP//s3ZpsJ5Rqv6RKg3M1VpfZF5Jwmw9evQwuwSRgSR3wgySu9bJvu4yfnj8Jfao+JLdf7O32eVYVl+zCxAZpy/w7thZDJtzltml7JQdGqwjaHir45nAGW3d2ebNm5kwYQIul4tYLMaYMWOYOHEixcXFZGdn43Q6qayspLCwkK1bt6K1prCwkJKSkvp7Q6urq+nRowelpaUopejWrRulpaXk5eURi8WoqamhZ8+eFBcX43a7yc/Pp6ysjPz8fMLhMIFAoH69x+MhNzeXLVu20LVrVwKBAMFgsH69z+fD7/dTXl5O9+7dqaqqIhwO16/3+/14PB4qKiooKCigoqKCSCRSv96ux7R161a8Xq+tjsmO3ye7HVMgEKBHjx62OiY7fp/sdkxr1qyhV69etjqmzv4+/XjDDXiunkxerLytPy4IIUxQHQ4SCoXS4lzeHJWOY8orpRYAS7TWV6UsWwfcobWetZP3Xg5cpbXevS37Wrx4sR4wYED7D0KYav369fTp08fsMkSGkdwJM0juhBkkd8Jo6Za5ZcuWLR09evTwptbZYZALIbbTrVs3s0sQGUhyJ8wguRNmkNwJo1kpc5ZvsJRSByil8syuQ6SX0tJSs0sQGUhyJ8wguRNmkNwJo1kpc5ZvsIAvSDyHJUS9vDzpuYXxJHfCDJI7YQbJnTCalTJnhwZLiO3EYjGzSxAZSHInzCC5E2aQ3AmjWSlz0mAJW6qpqTG7BJGBJHfCDJI7YQbJnTCalTKXrsO0a6CHUip1ZA4PoJVSvYHeWuslKev2V0oFk/+9T/L9O91XJ9Qt0kTPnj3NLkFkIMmdMIPkTphBcieMZqXMpesVrAXAWcDnKa8c4FPgouTXdd4HbgLeSb4uBz5o4b6ETRUXF5tdgshAkjthBsmdMIPkThjNSplLywZLa/13rbVTa61SXnla6yVa6xu11ipl29GNtlNa6/Nbsi9zjk4Y4cUXXzS7BJGBJHfCDJI7YQbJnTCalTKXlg2WEO31/PPPm12CyECSO2EGyZ0wg+ROGM1KmZMGS9hSNBo1uwSRgSR3wgySO2EGyZ0wmpUyp7SWsR5Svffee6VAkdl1iPbZunVrQbdu3crMrkNkFsmdMIPkTphBcieMloaZ6zd69OjCplZIgyWEEEIIIYQQHURuERRCCCGEEEKIDiINlhBCCCGEEEJ0EGmwhBBCCCGEEKKDSIMlLEkpdb1SKqaU0imvCqXUsCa2fafRdjGl1PVm1C2sqzWZS3nPwUqpuFKqj5G1Cvto5blubqPttFLqG6WU14zahXW1Mnfzm8jdSqWUx4zahfUppf5PKfVsE8uPUUrVNspaRCl1qRl1NkcaLGFVo4AngINSXjXAyCa2PQaYmrLdE8n3C9EarclcnTHAZ1rr9Z1fnrCp1uRuNHB7ynbjgH2BnoZUKuyktbn7Ow1zNxDoZUilwo4G72D5YcA6GubyTeBoY8pqOZfZBQjRRgoo0VovqV+gVDi5vClr67ZVSpUAu3V+icJmWps5SDRYD3Z2YcLWWpM7BfyUcq4LpiwXojVam7t1kjvRgQYD3zexXAG1jXK5BcgxqrCWkitYQgjRCZRSQ4D+gHWmnhdCCCFMpJRykrjybmnSYIlMsBqobrTMOtOBC6s6DfhKa93Ub+GEMEIZcCdQYXYhQgjRQv0Ay0/SKw2WsD2t9QCt9YKURcOBL82pRmSQ04DnlFKXK6XWmV2MyDxa62Kt9VVa63KzaxG2FgPcKV/X/bf8IlO0mtb6B+Aps+toL3kGS2QUpdRBwK+Bi8yuRdiXUqo/iXvIx5KGD98KIUQHegy4SymlgbXAo8D7wAZTqxLCRHIFS2Sa/wNe0Vp/Z3YhwtZOA1Zrrb8xuxCRcWY2GsJ4jVIq2+yihH1pra8G/kpiQJ8PgHeBY7XWlr/NS4i2kitYImMopfYETgWOMLsWYXunAc+ZXYTISLew7faa3sBrwO7ASrMKEvamlDoN+AeJ3P0ATCGRt7tMLEsIU0mDJTLJVSTmJPrY7EKEfSml+pJ4zu9is2sRGWmD1vpLAKXUL8llMly26EyPAjdore8EUEotAN5SSj2ntS4ytTIhTCINlsgISqlC4E/A2WbXImxvDFAEfK2UcpG8FVsp5dJay0PfQgi7ySYxWm+dulvwc02oRYi0IA2WyBRXABuBF8wuRNje8SSGmY00Wh5BriQIIYQQticNlrC95APelwHXaa3jZtcjbG8KDX9zeybwR+AUU6oRQgghhKGkwRKZ4AISk9bNM7kOkQG01qm3yqCUGgmEtdZLTCpJCCGEEAaSBktYlQZ6KKWGpyzzAFop1RvorbVeknwGZirwFjBQqfo7tKoa/yAsxE60KHPmlCZsrDW500DflG17pSwXojVae77bS3InOppSygPsB/yotd5CIlNZjXLZHQiaUV9zpMESVrUAuJ7ErVd1qoBPSUwifAOJ511GkHgeph9wVsq2C4GjDKhT2McCWpY5ITrSAlqeu/eBq5OvOmuBdZ1dpLCdBbQud3c3ev8aJHei/XoDnwPnkbgLaTEwLbmsThSYbHhlO6FkHjghhBBCCCGE6BgOswsQQgghhBBCCLuQBksIIYQQQgghOog0WEIIIYQQQgjRQaTBEkIIIYQQQogOIg2WEEIIIYQQQnQQabCEEEIIIYQQooNIgyWEEEIIIYQQHUQaLCGEEKKdlFJaKTXP7DpaSym1Tim1wOw6hBDCTqTBEkIIIYQQQogOIg2WEEKIDqOU2j15NedGs2sRraeUujH5/dvd7FqEEMKqpMESQgghxA4ppX4vDbMQQrScNFhCCCE60k9AIXCb2YWINrmNxPfvp5RlvwduMKUaIYSwIJfZBQghhLAPrXUcKDO7DtE2WutaoNbsOoQQwsrkCpYQQgghhBBCdBBpsIQQQmxHKTUvOdjBu0qplUqpLUqpaPK/X1VKfamUKlFKRZRS3ymluibft90gF0qpBUqpWqXUB0qpVUqp8uS+/qeUcjZTwyFKqbLk/upeMaXUkynb7KWU2tBoG538nPyU7bRS6mel1MdKqR+VUjVKqZBS6h2l1PtKqdVKqQqlVFgpNT3lfXWDPnyolPoqecxhpdR6pdRezdR+mlKquonan0rZZg+l1E+NtoknP6u9fy/nJpd/ppRappTamDzeLUqpw5rZd4NBLpRSGvhTyt+hDIAhhBA7IbcICiGEaM6rKf99AnA88BzwbnLZwcB4YDDwYTP7CQIvNbGv3sDPO3jPMUB34J/A1uSys4Azk58JcEhyH7OBtY1q2q9RTZuAZ1K+vh74FTArZdnlwMnAlEa1fJBSw17AxORnr6VpJwLZwNSUZWcBp6d8fRiwW6PaO+rvpc5nKfvuBlwHHA18vIN9NzY1ue+DUo5l6443F0IIIQ2WEEKIHdJaT6/7b6VUFxI//D+stV6XXHYuiR/qd3jFJalyB/tq7n1161I/7wASP+w33uZZrfWCndT0TaMapgDrGy37PbB7E7Wk1nAUiQZrp7U32ndLau9Cx/y91End9+4kGqydfa/qaa2n1+079ViEEELsmNwiKIQQIl3VJP/cI2XZj8DSlK8/Bi4EVhtVVAdq6vg2kji+UCvf1/jvRQghhEmkwRJCCJGungQ2AK8qpcYCaK3/rrUeXreB1noNiVsWvcnnv3YHCswotil1NSXrymm0+i1gBfCAUmoqgNb6Aa31cK31pmZ2u9O/FyGEEOaRWwSFEEKkJa31eqXUQcCLwHyl1BDgOq21rttGKfUb4BVacdubwX5s9HXd1Se01pXJASeeAO5KHt8lWutwcztsyd+LEEII80iDJYQQIm1prTcppY4E5gJ/AwYrpf6ota5ObnIwiebqYmBzctnRwCTDi23aqY2+/j71i2STdQpwC3A1sI9SaozWuqS5nbbg70UIIYRJpMESQgiRlpIDPnQhMRDFH5VSy4GbSYwE+Ju6zZJ/vp0y4EMXQwtthtb6xR2tU0rlkLidsVhrfU3y+B4E3ldKDd7RFakW/r0IIYQwiTyDJYQQIl1NIXGLXR8ArfWtwHzgWBNrajOl1Fil1B0pi04ncXwjAbTWjwN3AgOBXZvZ1RRs9PcihBB2Iw2WEEIIK4lg/vNWdRMYB1v5vt8Af97JNpHkn629w6Qlfy9trVsIIUQryC2CQggh0lUs+ec5Sqm6yW0HAPEmtjk/ZZuDO6GWuv17gPNJNDSfNrN9DOrn5KqzF03XfqpSanDyvw9utG6H+6b5v5c6pyfnsXICZySXfdTMvnf4ecl5wyAx/1ZlK/chhBAZQxosIYQQ6epdErfD/T1lWZzEs0Z1FgAVwDVAGYl5pLydUMt1Kf+9Djhba914hMBUrwNjgUdSlsVJDClf52NgPXBFyjKdXL6xmX235O+lzsSU/94ETNFat7bBeg04E7g7+fWLgDRYQgixA0pGdRVCCCGappS6EbgB+FXdIBpWkLxy9ggwSmu9wNxqhBAis8gzWEIIIYQQQgjRQaTBEkIIIYQQQogOIg2WEEIIIYQQQnQQabCEEEIIIYQQooPIIBdCCCGEEEII0UHkCpYQQgghhBBCdBBpsIQQQgghhBCig0iDJYQQQgghhBAdRBosIYQQQgghhOgg0mAJIYQQQgghRAf5fzQWbwfdqxPMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC Curve for min_samples_split\n",
    "\n",
    "min_samples_splits = np.linspace(0.1, 1.0, 10, endpoint=True)\n",
    "train_results = []\n",
    "test_results = []\n",
    "\n",
    "for min_samples_split in min_samples_splits:\n",
    "    rf = RandomForestClassifier(max_depth=5,n_estimators=10,min_samples_split=min_samples_split)\n",
    "    rf.fit(X_train, y_train)\n",
    "    \n",
    "    train_pred = rf.predict(X_train)\n",
    "    \n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_train, train_pred)\n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    train_results.append(roc_auc)\n",
    "    \n",
    "    y_pred_rf = rf.predict(X_test)\n",
    "    \n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_test, y_pred_rf)\n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    test_results.append(roc_auc)\n",
    "    \n",
    "from matplotlib.legend_handler import HandlerLine2D\n",
    "\n",
    "line1, = plt.plot(min_samples_splits, train_results, \"b\" ,label=\"Train AUC\")\n",
    "line2, = plt.plot(min_samples_splits, test_results, \"r\", label=\"Test AUC\")\n",
    "\n",
    "plt.legend(handler_map={line1: HandlerLine2D(numpoints=2)})\n",
    "\n",
    "plt.ylabel(\"AUC score\",size=25)\n",
    "plt.xlabel(\"min samples split\",size=25)\n",
    "plt.xticks(size=20)\n",
    "plt.yticks(size=20)\n",
    "#plt.savefig(\"minsamplesplit.png\",dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### b-3) Random Forest Hyperparameter Grid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bootstrap': [True, False],\n",
      " 'max_depth': [0, 1, 2, 3, 4, 5, 6, 7, 8, 10, None],\n",
      " 'max_features': ['auto', 'sqrt'],\n",
      " 'min_samples_leaf': [1, 5, 10],\n",
      " 'min_samples_split': [1, 5, 10],\n",
      " 'n_estimators': [1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  1,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  2,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  3,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  4,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  5,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  6,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  7,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  8,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  9,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  10,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  11,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  12,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  13,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  14,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  15,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  16,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  17,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  18,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  19,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  20,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  21,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  22,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  23,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  24,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  25,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  26,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  27,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  28,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  29,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  30,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  31,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  32,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  33,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  34,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  35,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  36,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  37,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  38,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  39,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  40,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  41,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  42,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  43,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  44,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  45,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  46,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  47,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  48,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  49,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  50,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  51,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  52,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  53,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  54,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  55,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  56,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  57,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  58,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  59,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  60,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  61,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  62,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  63,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  64,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  65,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  66,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  67,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  68,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  69,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  70,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  71,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  72,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  73,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  74,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  75,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  76,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  77,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  78,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  79,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  80,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  81,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  82,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  83,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  84,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  85,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  86,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  87,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  88,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  89,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  90,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  91,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  92,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  93,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  94,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  95,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  96,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  97,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  98,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  99,\n",
      "                  100]}\n"
     ]
    }
   ],
   "source": [
    "from pprint import pprint\n",
    "\n",
    "# Number of trees in random forest\n",
    "n_estimators = [int(x) for x in np.linspace(start = 1, stop = 100, num = 1000)]\n",
    "\n",
    "# Number of features to consider at every split\n",
    "max_features = ['auto', 'sqrt']\n",
    "\n",
    "# Maximum number of levels in tree\n",
    "max_depth = [int(x) for x in np.linspace(0, 10, num = 10)]\n",
    "max_depth.append(None)\n",
    "\n",
    "# Minimum number of samples required to split a node\n",
    "min_samples_split = [1, 5, 10]\n",
    "\n",
    "# Minimum number of samples required at each leaf node\n",
    "min_samples_leaf = [1, 5, 10]\n",
    "\n",
    "# Method of selecting samples for training each tree\n",
    "bootstrap = [True, False]\n",
    "\n",
    "\n",
    "# Create the random grid\n",
    "random_grid = {'n_estimators': n_estimators,\n",
    "               'max_features': max_features,\n",
    "               'max_depth': max_depth,\n",
    "               'min_samples_split': min_samples_split,\n",
    "               'min_samples_leaf': min_samples_leaf,\n",
    "               'bootstrap': bootstrap}\n",
    "\n",
    "pprint(random_grid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### b-4-a) Random Forest Hyperparameter Grid Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  17 tasks      | elapsed:    3.1s\n",
      "[Parallel(n_jobs=-1)]: Done 249 tasks      | elapsed:    5.3s\n",
      "[Parallel(n_jobs=-1)]: Done 277 out of 300 | elapsed:    5.6s remaining:    0.4s\n",
      "[Parallel(n_jobs=-1)]: Done 300 out of 300 | elapsed:    5.9s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=3, estimator=RandomForestClassifier(), n_iter=100,\n",
       "                   n_jobs=-1,\n",
       "                   param_distributions={'bootstrap': [True, False],\n",
       "                                        'max_depth': [0, 1, 2, 3, 4, 5, 6, 7, 8,\n",
       "                                                      10, None],\n",
       "                                        'max_features': ['auto', 'sqrt'],\n",
       "                                        'min_samples_leaf': [1, 5, 10],\n",
       "                                        'min_samples_split': [1, 5, 10],\n",
       "                                        'n_estimators': [1, 1, 1, 1, 1, 1, 1, 1,\n",
       "                                                         1, 1, 1, 2, 2, 2, 2, 2,\n",
       "                                                         2, 2, 2, 2, 2, 3, 3, 3,\n",
       "                                                         3, 3, 3, 3, 3, 3, ...]},\n",
       "                   random_state=42, verbose=2)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use the random grid to search for best hyperparameters\n",
    "\n",
    "# First create the base model to tune\n",
    "rf_model = RandomForestClassifier()\n",
    "\n",
    "# Random search of parameters, using 3 fold cross validation, \n",
    "# search across 100 different combinations, and use all available cores\n",
    "\n",
    "rf_random = RandomizedSearchCV(estimator = rf_model, param_distributions = random_grid, \n",
    "                               n_iter = 100, cv = 3, verbose=2, random_state=42, n_jobs = -1)\n",
    "\n",
    "# Fit the random search model\n",
    "rf_random.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 44,\n",
       " 'min_samples_split': 5,\n",
       " 'min_samples_leaf': 5,\n",
       " 'max_features': 'sqrt',\n",
       " 'max_depth': 3,\n",
       " 'bootstrap': True}"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_random.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.82.\n",
      "Sensivity score is 0.825.\n",
      "F-1 score is 0.85.\n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.71      0.81      0.76        21\n",
      "           1       0.89      0.82      0.86        40\n",
      "\n",
      "    accuracy                           0.82        61\n",
      "   macro avg       0.80      0.82      0.81        61\n",
      "weighted avg       0.83      0.82      0.82        61\n",
      "\n",
      "\n",
      "\n",
      "ROC AUC Score: 0.8172619047619047\n"
     ]
    }
   ],
   "source": [
    "# creating a new model with new parameters based on above information\n",
    "rf_model = RandomForestClassifier(random_state=0,\n",
    "                                  min_samples_split=rf_random.best_params_[\"min_samples_split\"],\n",
    "                                 min_samples_leaf=rf_random.best_params_[\"min_samples_leaf\"],\n",
    "                                  n_estimators=rf_random.best_params_[\"n_estimators\"],\n",
    "                                  max_features=rf_random.best_params_[\"max_features\"],\n",
    "                                 max_depth=rf_random.best_params_[\"max_depth\"])\n",
    "\n",
    "# fitting model\n",
    "rf_model.fit(X_train,y_train)\n",
    "\n",
    "# predicting outcome\n",
    "y_pred_rf = rf_model.predict(X_test)\n",
    "\n",
    "pd.DataFrame({\"Actual\":y_test,\"Predicted\":y_pred_rf}).sample(10)\n",
    "\n",
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_test,y_pred_rf)))\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_test,y_pred_rf)))\n",
    "#print(f\"Precision score is {str(precision)[:5]}.\")\n",
    "print(f\"F-1 score is {str(f1_score(y_test,y_pred_rf))[:4]}.\\n\\n\")\n",
    "\n",
    "'''\n",
    "Sensivity score increased from 0.868 to 0.921 on the test data.\n",
    "Also, accuracy increased from 0.87 to 0.90\n",
    "F-1 increased from 0.89 to 0.92\n",
    "'''\n",
    "\n",
    "print(classification_report(y_test,y_pred_rf))\n",
    "print(f\"\\n\\nROC AUC Score: {roc_auc_score(y_test,y_pred_rf)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.18032786885245902"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_error = np.mean(np.square(np.array(y_pred_rf)-np.array(y_test)))\n",
    "training_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.88.\n",
      "Sensivity score is 0.904.\n",
      "F-1 score is 0.88.\n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.89      0.86      0.88       117\n",
      "           1       0.88      0.90      0.89       125\n",
      "\n",
      "    accuracy                           0.88       242\n",
      "   macro avg       0.88      0.88      0.88       242\n",
      "weighted avg       0.88      0.88      0.88       242\n",
      "\n",
      "\n",
      "\n",
      "ROC AUC Score: 0.8836239316239316\n"
     ]
    }
   ],
   "source": [
    "# testing training data with a new model by using new parameters based on above information\n",
    "\n",
    "rf_model = RandomForestClassifier(random_state=0,\n",
    "                                  min_samples_split=rf_random.best_params_[\"min_samples_split\"],\n",
    "                                 min_samples_leaf=rf_random.best_params_[\"min_samples_leaf\"],\n",
    "                                  n_estimators=rf_random.best_params_[\"n_estimators\"],\n",
    "                                  max_features=rf_random.best_params_[\"max_features\"],\n",
    "                                 max_depth=rf_random.best_params_[\"max_depth\"])\n",
    "\n",
    "rf_model.fit(X_train,y_train)\n",
    "\n",
    "\n",
    "y_pred_rf_tr = rf_model.predict(X_train)\n",
    "\n",
    "\n",
    "pd.DataFrame({\"Actual\":y_test,\"Predicted\":y_pred_rf}).sample(10)\n",
    "\n",
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_train,y_pred_rf_tr)))\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_train,y_pred_rf_tr)))\n",
    "#print(f\"Precision score is {str(precision)[:5]}.\")\n",
    "print(f\"F-1 score is {str(f1_score(y_train,y_pred_rf_tr))[:4]}.\\n\\n\")\n",
    "\n",
    "'''\n",
    "Sensivity score increased from 0.868 to 0.921 on the test data.\n",
    "Also, accuracy increased from 0.87 to 0.90\n",
    "F-1 increased from 0.89 to 0.92\n",
    "'''\n",
    "\n",
    "print(classification_report(y_train,y_pred_rf_tr))\n",
    "print(f\"\\n\\nROC AUC Score: {roc_auc_score(y_train,y_pred_rf_tr)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\nSummmary;\\n\\nModel's initial performance was;\\n\\nAccuracy: 0.823\\nRecall: 0.828\\nROC AUC Score: 0.824\\n\\nAfter hyperparameter tuning;\\n\\nAccuracy: 0.872\\nRecall: 0.862 \\nROC AUC Score: 0.868\\n\\nFlow;\\n\\n1- Get initial score\\n2- Draw graphs for parameters to see boundaries\\n3- Set parameter limits for GridSearch\\n4- Choose best parameters\\n5- Re-evaluate and compare\\n\""
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "Summmary;\n",
    "\n",
    "Model's initial performance was;\n",
    "\n",
    "Accuracy: 0.823\n",
    "Recall: 0.828\n",
    "ROC AUC Score: 0.824\n",
    "\n",
    "After hyperparameter tuning;\n",
    "\n",
    "Accuracy: 0.872\n",
    "Recall: 0.862 \n",
    "ROC AUC Score: 0.868\n",
    "\n",
    "Flow;\n",
    "\n",
    "1- Get initial score\n",
    "2- Draw graphs for parameters to see boundaries\n",
    "3- Set parameter limits for GridSearch\n",
    "4- Choose best parameters\n",
    "5- Re-evaluate and compare\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### c) Applying K- Nearest Neighbour Model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>Actual</th>\n",
       "      <th>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>137</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Actual  Predicted\n",
       "106       1          0\n",
       "131       1          1\n",
       "3         1          0\n",
       "137       1          0\n",
       "146       1          0\n",
       "252       0          0\n",
       "35        1          0\n",
       "298       0          0\n",
       "120       1          0\n",
       "200       0          1"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creating a new model with new parameters based on above information\n",
    "\n",
    "kn_model = KNeighborsClassifier(n_neighbors=2)\n",
    "\n",
    "# fitting model\n",
    "kn_model.fit(X_train,y_train)\n",
    "\n",
    "# predicting outcome\n",
    "y_pred_kn = kn_model.predict(X_test)\n",
    "\n",
    "pd.DataFrame({\"Actual\":y_test,\"Predicted\":y_pred_kn}).sample(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### c-1) Evaluating KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.51.\n",
      "Sensivity score is 0.375.\n",
      "F-1 score is 0.5.\n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.39      0.76      0.52        21\n",
      "           1       0.75      0.38      0.50        40\n",
      "\n",
      "    accuracy                           0.51        61\n",
      "   macro avg       0.57      0.57      0.51        61\n",
      "weighted avg       0.63      0.51      0.51        61\n",
      "\n",
      "\n",
      "\n",
      "ROC AUC Score: 0.5684523809523808\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_test,y_pred_kn)))\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_test,y_pred_kn)))\n",
    "#print(f\"Precision score is {str(precision)[:5]}.\")\n",
    "print(f\"F-1 score is {str(f1_score(y_test,y_pred_kn))[:4]}.\\n\\n\")\n",
    "\n",
    "print(classification_report(y_test,y_pred_kn))\n",
    "print(f\"\\n\\nROC AUC Score: {roc_auc_score(y_test,y_pred_kn)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### c-2) Hyperparameter Tuning for KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC Curve for Number of Neighbors\n",
    "\n",
    "neighbors = list(range(1,30))\n",
    "train_results = []\n",
    "test_results = []\n",
    "\n",
    "for n in neighbors:\n",
    "    model = KNeighborsClassifier(n_neighbors=n)\n",
    "    model.fit(X_train, y_train)\n",
    "    train_pred = model.predict(X_train)\n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_train, train_pred)\n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    train_results.append(roc_auc)\n",
    "    y_pred = model.predict(X_test)\n",
    "    false_positive_rate, true_positive_rate, thresholds = roc_curve(y_test, y_pred)\n",
    "    roc_auc = auc(false_positive_rate, true_positive_rate)\n",
    "    test_results.append(roc_auc)\n",
    "    \n",
    "from matplotlib.legend_handler import HandlerLine2D\n",
    "line1, = plt.plot(neighbors, train_results, \"b\", label=\"Train AUC\")\n",
    "line2, = plt.plot(neighbors, test_results, \"r\", label=\"Test AUC\")\n",
    "plt.legend(handler_map={line1: HandlerLine2D(numpoints=2)})\n",
    "plt.ylabel(\"AUC score\",size=25)\n",
    "plt.xlabel(\"n_neighbors\",size=25)\n",
    "plt.xticks(size=20)\n",
    "plt.yticks(size=20)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### c-3) Re-evaluating KNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.62.\n",
      "Sensivity score is 0.600.\n",
      "F-1 score is 0.67.\n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.47      0.67      0.55        21\n",
      "           1       0.77      0.60      0.68        40\n",
      "\n",
      "    accuracy                           0.62        61\n",
      "   macro avg       0.62      0.63      0.61        61\n",
      "weighted avg       0.67      0.62      0.63        61\n",
      "\n",
      "\n",
      "\n",
      "ROC AUC Score: 0.6333333333333334\n"
     ]
    }
   ],
   "source": [
    "# creating a new model with new parameters based on above information\n",
    "\n",
    "kn_model = KNeighborsClassifier(n_neighbors=10)\n",
    "\n",
    "# fitting model\n",
    "kn_model.fit(X_train,y_train)\n",
    "\n",
    "# predicting outcome\n",
    "y_pred_kn = kn_model.predict(X_test)\n",
    "\n",
    "pd.DataFrame({\"Actual\":y_test,\"Predicted\":y_pred_kn}).sample(10)\n",
    "\n",
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_test,y_pred_kn)))\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_test,y_pred_kn)))\n",
    "#print(f\"Precision score is {str(precision)[:5]}.\")\n",
    "print(f\"F-1 score is {str(f1_score(y_test,y_pred_kn))[:4]}.\\n\\n\")\n",
    "\n",
    "print(classification_report(y_test,y_pred_kn))\n",
    "print(f\"\\n\\nROC AUC Score: {roc_auc_score(y_test,y_pred_kn)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nSummary\\n\\nAfter hyperparameter tuning ROC AUC score increased from 0.46 to 0.69, yet still this \\nscore is not good enough.\\n\\n'"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "Summary\n",
    "\n",
    "After hyperparameter tuning ROC AUC score increased from 0.46 to 0.69, yet still this \n",
    "score is not good enough.\n",
    "\n",
    "''' "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### d) Support Vector Machine (SVM) Model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>Actual</th>\n",
       "      <th>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>279</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Actual  Predicted\n",
       "36        1          1\n",
       "41        1          1\n",
       "106       1          0\n",
       "2         1          1\n",
       "252       0          0\n",
       "279       0          0\n",
       "295       0          1\n",
       "289       0          0\n",
       "79        1          1\n",
       "107       1          1"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creating the model\n",
    "svm_model = SVC()\n",
    "\n",
    "# fitting model\n",
    "svm_model.fit(X_train,y_train)\n",
    "    \n",
    "# predicting outcome\n",
    "y_pred_svm = svm_model.predict(X_test)\n",
    "\n",
    "pd.DataFrame({\"Actual\":y_test,\"Predicted\":y_pred_svm}).sample(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### d-1) Evaluating SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score is 0.72.\n",
      "Sensivity score is 0.775.\n",
      "F-1 score is 0.78.\n",
      "\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.59      0.62      0.60        21\n",
      "           1       0.79      0.78      0.78        40\n",
      "\n",
      "    accuracy                           0.72        61\n",
      "   macro avg       0.69      0.70      0.69        61\n",
      "weighted avg       0.72      0.72      0.72        61\n",
      "\n",
      "\n",
      "\n",
      "ROC AUC Score: 0.6970238095238096\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy score is {:.2f}.\".format(accuracy_score(y_test,y_pred_svm)))\n",
    "print(\"Sensivity score is {:.3f}.\".format(recall_score(y_test,y_pred_svm)))\n",
    "#print(f\"Precision score is {str(precision)[:5]}.\")\n",
    "print(f\"F-1 score is {str(f1_score(y_test,y_pred_svm))[:4]}.\\n\\n\")\n",
    "\n",
    "print(classification_report(y_test,y_pred_svm))\n",
    "print(f\"\\n\\nROC AUC Score: {roc_auc_score(y_test,y_pred_svm)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nSummary\\n\\nInitial performance score is 0.62, no further hyperparameter tuning is needed since I obtained\\nbetter initial results in other models.\\n\\n'"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "Summary\n",
    "\n",
    "Initial performance score is 0.62, no further hyperparameter tuning is needed since I obtained\n",
    "better initial results in other models.\n",
    "\n",
    "''' "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Best result was achieved by using Random Forest Classification Model for this predictive analysis.\n",
    "\n",
    "Tuning parameters in random forest model help me to achieve even better results than logistic regression.\n",
    "\n",
    "I could have tried to increase performance more for random forest yet due to limited time this scores are good enough to end the project."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Gorkem Guneser - 27.11.2020"
   ]
  }
 ],
 "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}