Heart-Disease-Prediction/Heart_Disease_Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font>\n",
    "    <h1 style=\"text-align: center\">\n",
    "        AmirReza Azari <br>\n",
    "        99101087\n",
    "    </h1>\n",
    "</font>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8QU5zmW1lBAi"
   },
   "source": [
    "In this notebook, we are looking for the diagnosis of heart disease among patients using the [heart disease dataset](https://www.kaggle.com/johnsmith88/heart-disease-dataset). First, read the information about the different features, then answer the questions in each section using the attached data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "utNwgYmlKvfm"
   },
   "source": [
    "# Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bjntS4nxH4Hs"
   },
   "source": [
    "## Importing Libraries"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ycgOZilIJ-IO"
   },
   "source": [
    "In this notebook, you cannot use scikit learn and other libraries, except for those imported below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "RA04oziCHqpP"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TNo6F7XEH7p3"
   },
   "source": [
    "## Load Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Q1KtMuwCmomz"
   },
   "source": [
    "Load dataset csv file into a data frame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "m7Z3egeyH-g9"
   },
   "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>52</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>125</td>\n",
       "      <td>212</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>168</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>53</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>140</td>\n",
       "      <td>203</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>155</td>\n",
       "      <td>1</td>\n",
       "      <td>3.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>70</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>145</td>\n",
       "      <td>174</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>125</td>\n",
       "      <td>1</td>\n",
       "      <td>2.6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>61</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>148</td>\n",
       "      <td>203</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>161</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>62</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>138</td>\n",
       "      <td>294</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>106</td>\n",
       "      <td>0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  slope  \\\n",
       "0   52    1   0       125   212    0        1      168      0      1.0      2   \n",
       "1   53    1   0       140   203    1        0      155      1      3.1      0   \n",
       "2   70    1   0       145   174    0        1      125      1      2.6      0   \n",
       "3   61    1   0       148   203    0        1      161      0      0.0      2   \n",
       "4   62    0   0       138   294    1        1      106      0      1.9      1   \n",
       "\n",
       "   ca  thal  target  \n",
       "0   2     3       0  \n",
       "1   0     3       0  \n",
       "2   0     3       0  \n",
       "3   1     3       0  \n",
       "4   3     2       0  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"heart.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>1020</th>\n",
       "      <td>59</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>140</td>\n",
       "      <td>221</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>164</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1021</th>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>125</td>\n",
       "      <td>258</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>2.8</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1022</th>\n",
       "      <td>47</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>110</td>\n",
       "      <td>275</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>118</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1023</th>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>110</td>\n",
       "      <td>254</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>159</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1024</th>\n",
       "      <td>54</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>120</td>\n",
       "      <td>188</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>113</td>\n",
       "      <td>0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  \\\n",
       "1020   59    1   1       140   221    0        1      164      1      0.0   \n",
       "1021   60    1   0       125   258    0        0      141      1      2.8   \n",
       "1022   47    1   0       110   275    0        0      118      1      1.0   \n",
       "1023   50    0   0       110   254    0        0      159      0      0.0   \n",
       "1024   54    1   0       120   188    0        1      113      0      1.4   \n",
       "\n",
       "      slope  ca  thal  target  \n",
       "1020      2   0     2       1  \n",
       "1021      1   1     3       0  \n",
       "1022      1   1     2       0  \n",
       "1023      2   0     2       1  \n",
       "1024      1   1     3       0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "phw5_4uEIgyY"
   },
   "source": [
    "## EDA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NkxL1k5Onuvj"
   },
   "source": [
    "In statistics, exploratory data analysis (EDA) is an approach to analyzing data sets to summarize their main characteristics, often using statistical graphics and other data visualization methods. In this section, you should analyze the data. You can use the functions provided by the Pandas library. You can use analysis such as null checking of features, number of features and samples, and type of features. But don't stop there and search for it.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "-2qsSen8X1qv"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of features are:  14\n",
      "Number of samples  are:  1025\n"
     ]
    }
   ],
   "source": [
    "features = df.shape[1]\n",
    "samples = df.shape[0]\n",
    "print(\"Number of features are: \", df.shape[1])\n",
    "print(\"Number of samples  are: \", df.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1025 entries, 0 to 1024\n",
      "Data columns (total 14 columns):\n",
      " #   Column    Non-Null Count  Dtype  \n",
      "---  ------    --------------  -----  \n",
      " 0   age       1025 non-null   int64  \n",
      " 1   sex       1025 non-null   int64  \n",
      " 2   cp        1025 non-null   int64  \n",
      " 3   trestbps  1025 non-null   int64  \n",
      " 4   chol      1025 non-null   int64  \n",
      " 5   fbs       1025 non-null   int64  \n",
      " 6   restecg   1025 non-null   int64  \n",
      " 7   thalach   1025 non-null   int64  \n",
      " 8   exang     1025 non-null   int64  \n",
      " 9   oldpeak   1025 non-null   float64\n",
      " 10  slope     1025 non-null   int64  \n",
      " 11  ca        1025 non-null   int64  \n",
      " 12  thal      1025 non-null   int64  \n",
      " 13  target    1025 non-null   int64  \n",
      "dtypes: float64(1), int64(13)\n",
      "memory usage: 112.2 KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>age</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>54.434146</td>\n",
       "      <td>9.072290</td>\n",
       "      <td>29.0</td>\n",
       "      <td>48.0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>77.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sex</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>0.695610</td>\n",
       "      <td>0.460373</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cp</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>0.942439</td>\n",
       "      <td>1.029641</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>trestbps</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>131.611707</td>\n",
       "      <td>17.516718</td>\n",
       "      <td>94.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>140.0</td>\n",
       "      <td>200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chol</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>246.000000</td>\n",
       "      <td>51.592510</td>\n",
       "      <td>126.0</td>\n",
       "      <td>211.0</td>\n",
       "      <td>240.0</td>\n",
       "      <td>275.0</td>\n",
       "      <td>564.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fbs</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>0.149268</td>\n",
       "      <td>0.356527</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>restecg</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>0.529756</td>\n",
       "      <td>0.527878</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>thalach</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>149.114146</td>\n",
       "      <td>23.005724</td>\n",
       "      <td>71.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>152.0</td>\n",
       "      <td>166.0</td>\n",
       "      <td>202.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>exang</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>0.336585</td>\n",
       "      <td>0.472772</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>oldpeak</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>1.071512</td>\n",
       "      <td>1.175053</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>1.8</td>\n",
       "      <td>6.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>slope</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>1.385366</td>\n",
       "      <td>0.617755</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ca</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>0.754146</td>\n",
       "      <td>1.030798</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>thal</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>2.323902</td>\n",
       "      <td>0.620660</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>1025.0</td>\n",
       "      <td>0.513171</td>\n",
       "      <td>0.500070</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           count        mean        std    min    25%    50%    75%    max\n",
       "age       1025.0   54.434146   9.072290   29.0   48.0   56.0   61.0   77.0\n",
       "sex       1025.0    0.695610   0.460373    0.0    0.0    1.0    1.0    1.0\n",
       "cp        1025.0    0.942439   1.029641    0.0    0.0    1.0    2.0    3.0\n",
       "trestbps  1025.0  131.611707  17.516718   94.0  120.0  130.0  140.0  200.0\n",
       "chol      1025.0  246.000000  51.592510  126.0  211.0  240.0  275.0  564.0\n",
       "fbs       1025.0    0.149268   0.356527    0.0    0.0    0.0    0.0    1.0\n",
       "restecg   1025.0    0.529756   0.527878    0.0    0.0    1.0    1.0    2.0\n",
       "thalach   1025.0  149.114146  23.005724   71.0  132.0  152.0  166.0  202.0\n",
       "exang     1025.0    0.336585   0.472772    0.0    0.0    0.0    1.0    1.0\n",
       "oldpeak   1025.0    1.071512   1.175053    0.0    0.0    0.8    1.8    6.2\n",
       "slope     1025.0    1.385366   0.617755    0.0    1.0    1.0    2.0    2.0\n",
       "ca        1025.0    0.754146   1.030798    0.0    0.0    0.0    1.0    4.0\n",
       "thal      1025.0    2.323902   0.620660    0.0    2.0    2.0    3.0    3.0\n",
       "target    1025.0    0.513171   0.500070    0.0    0.0    1.0    1.0    1.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "age          41\n",
       "sex           2\n",
       "cp            4\n",
       "trestbps     49\n",
       "chol        152\n",
       "fbs           2\n",
       "restecg       3\n",
       "thalach      91\n",
       "exang         2\n",
       "oldpeak      40\n",
       "slope         3\n",
       "ca            5\n",
       "thal          4\n",
       "target        2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "58    68\n",
       "57    57\n",
       "54    53\n",
       "59    46\n",
       "52    43\n",
       "51    39\n",
       "56    39\n",
       "62    37\n",
       "60    37\n",
       "44    36\n",
       "64    34\n",
       "41    32\n",
       "63    32\n",
       "67    31\n",
       "61    31\n",
       "55    30\n",
       "65    27\n",
       "43    26\n",
       "42    26\n",
       "53    26\n",
       "66    25\n",
       "45    25\n",
       "48    23\n",
       "46    23\n",
       "50    21\n",
       "47    18\n",
       "49    17\n",
       "35    15\n",
       "39    14\n",
       "70    14\n",
       "68    12\n",
       "38    12\n",
       "71    11\n",
       "40    11\n",
       "69     9\n",
       "37     6\n",
       "34     6\n",
       "29     4\n",
       "76     3\n",
       "77     3\n",
       "74     3\n",
       "Name: age, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.age.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    713\n",
       "0    312\n",
       "Name: sex, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sex.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    497\n",
       "2    284\n",
       "1    167\n",
       "3     77\n",
       "Name: cp, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.cp.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    578\n",
       "1    226\n",
       "2    134\n",
       "3     69\n",
       "4     18\n",
       "Name: ca, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.ca.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    526\n",
       "0    499\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.target.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1020</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1021</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1022</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1023</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1024</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1025 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        age    sex     cp  trestbps   chol    fbs  restecg  thalach  exang  \\\n",
       "0     False  False  False     False  False  False    False    False  False   \n",
       "1     False  False  False     False  False  False    False    False  False   \n",
       "2     False  False  False     False  False  False    False    False  False   \n",
       "3     False  False  False     False  False  False    False    False  False   \n",
       "4     False  False  False     False  False  False    False    False  False   \n",
       "...     ...    ...    ...       ...    ...    ...      ...      ...    ...   \n",
       "1020  False  False  False     False  False  False    False    False  False   \n",
       "1021  False  False  False     False  False  False    False    False  False   \n",
       "1022  False  False  False     False  False  False    False    False  False   \n",
       "1023  False  False  False     False  False  False    False    False  False   \n",
       "1024  False  False  False     False  False  False    False    False  False   \n",
       "\n",
       "      oldpeak  slope     ca   thal  target  \n",
       "0       False  False  False  False   False  \n",
       "1       False  False  False  False   False  \n",
       "2       False  False  False  False   False  \n",
       "3       False  False  False  False   False  \n",
       "4       False  False  False  False   False  \n",
       "...       ...    ...    ...    ...     ...  \n",
       "1020    False  False  False  False   False  \n",
       "1021    False  False  False  False   False  \n",
       "1022    False  False  False  False   False  \n",
       "1023    False  False  False  False   False  \n",
       "1024    False  False  False  False   False  \n",
       "\n",
       "[1025 rows x 14 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "age         0\n",
       "sex         0\n",
       "cp          0\n",
       "trestbps    0\n",
       "chol        0\n",
       "fbs         0\n",
       "restecg     0\n",
       "thalach     0\n",
       "exang       0\n",
       "oldpeak     0\n",
       "slope       0\n",
       "ca          0\n",
       "thal        0\n",
       "target      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "age\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3sAAAEGCAYAAADc7ozVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAAsTAAALEwEAmpwYAAATvElEQVR4nO3df7CldX0f8PdHVptUlF+7oQTYrmlpLJ2JSHco1kxKQhvBpq6xKaNTIyXb2XQGE51pp5L+E5OOHfNHmhjTOEOyImSqBk0R2nGIDLFJ06kimxDDD63EQGEHWH4YlDgxA/n0j/ssXsnu3bO799xz7ndfr5kz53m+5znnfPY7994vb77f53mquwMAAMBYXrToAgAAAFh/wh4AAMCAhD0AAIABCXsAAAADEvYAAAAGtGXRBRyPrVu39o4dOxZdBgAbYN++fU9097ZF17FZGCMBTgxrjY+bOuzt2LEjd95556LLAGADVNWDi65hMzFGApwY1hofLeMEAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGNCWRRcAbF7P7D1j0SU87+TdTy66BACApWJmDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAtiy6AAAANqfdV70tTzz19KLLOOFsPf2U7L3uhkWXwSYg7AEAcEyeeOrp/NxlDyy6jLl71607lurf+a5bdyy6BDYJYQ82mWf2nrHoEgAA2AScswcAADCguYW9qjq3qj5dVfdW1T1V9Y6p/fSquq2qvjQ9nza1V1X9UlXdX1Wfr6oL51UbAADA6OY5s/dskn/b3ecnuTjJ1VV1fpJrktze3ecluX3aT5LLk5w3PfYk+cAcawMAABja3MJedz/S3b8/bX8tyX1Jzk6yK8n102HXJ3njtL0ryQ294jNJTq2qs+ZVHwAAwMg25Jy9qtqR5NVJPpvkzO5+ZHrp0SRnTttnJ3lo1dsentpe+Fl7qurOqrrz8ccfn1/RAAAAm9jcw15VnZzkN5O8s7u/uvq17u4kfTSf193XdvfO7t65bdu2dawUAABgHHMNe1X14qwEvf/a3f9tan7s4PLM6fnA1L4/ybmr3n7O1AYAAMBRmufVOCvJ3iT3dfd/XvXSLUmunLavTHLzqva3TVflvDjJ06uWewIAAHAU5nlT9dcm+dEkf1RVd01t/yHJe5PcWFW7kzyY5IrptU8meX2S+5N8PclVc6wNAABgaHMLe939e0nqMC9feojjO8nV86oHAADgRLIhV+MEAABgYwl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgBwSLt27Vp0CQBDm/ffWWEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABiTsAQAADGhuYa+qPlhVB6rq7lVt766q/VV11/R4/arXfqqq7q+qL1bV6+ZVFwAAwIlgnjN7H0py2SHaf6G7L5gen0ySqjo/yZuT/L3pPb9SVSfNsTYAAIChzS3sdffvJnlqxsN3Jflod3+ju/8kyf1JLppXbQAAAKNbxDl7b6+qz0/LPE+b2s5O8tCqYx6e2v6KqtpTVXdW1Z2PP/74vGsFAADYlLZs8Pd9IMl/TNLT888n+bGj+YDuvjbJtUmyc+fOXu8Cgc3pmb1nLLqEJMnJu59cdAkAAEk2eGavux/r7ue6+y+T/Gq+uVRzf5JzVx16ztQGAADAMdjQsFdVZ63a/eEkB6/UeUuSN1fVX6uqVyQ5L8kdG1kbAADASOa2jLOqPpLkkiRbq+rhJD+d5JKquiAryzgfSPLjSdLd91TVjUnuTfJskqu7+7l51QYAADC6uYW97n7LIZr3rnH8e5K8Z171AAAAnEgWcTVOAAAA5kzYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGNFPYq6rbZ2kDAABgOWxZ68Wq+rYkfz3J1qo6LUlNL708ydlzrg0AAIBjtGbYS/LjSd6Z5DuT7Ms3w95Xk/zy/MqCFc/sPWPRJTzv5N1PLroEAACY2Zphr7vfl+R9VfUT3f3+DaoJAACA43Skmb0kSXe/v6r+YZIdq9/T3TfMqS4AAACOw0xhr6p+PcnfSnJXkuem5k4i7AEAACyhmcJekp1Jzu/unmcxsMyW6fxBAAA4klnvs3d3kr8xz0IAAABYP7PO7G1Ncm9V3ZHkGwcbu/sNc6kKAACA4zJr2Hv3PIsAAABgfc16Nc7fmXchAAAArJ9Zr8b5taxcfTNJXpLkxUn+rLtfPq/CAICjU1V7kuxJku3bt6/LZ+7atWtdPgdYX343mcWsM3svO7hdVZVkV5KL51UUAHD0uvvaJNcmyc6dO9flCto333zzenwMgxI4Fsfv5hjm/Ts069U4n9crPpHkdetfDgAAAOth1mWcb1q1+6Ks3Hfvz+dSEQAAAMdt1qtx/rNV288meSArSzkBAABYQrOes3fVvAsBAABg/cy6jPOcJO9P8tqp6X8leUd3PzyvwlisZ/aesegSAACA4zDrBVquS3JLku+cHv99agMAAGAJzRr2tnX3dd397PT4UJJtc6wLAACA4zBr2Huyqt5aVSdNj7cmeXKehQEAAHDsZg17P5bkiiSPJnkkyY8k+VdzqgkAAIDjNGvY+9kkV3b3tu7+jqyEv59Z6w1V9cGqOlBVd69qO72qbquqL03Pp03tVVW/VFX3V9Xnq+rCY/0HAQAAMHvY+57u/srBne5+Ksmrj/CeDyW57AVt1yS5vbvPS3L7tJ8klyc5b3rsSfKBGesCAADgEGYNey86OAuXrMzQ5Qi3beju303y1AuadyW5ftq+PskbV7Xf0Cs+k+TUqjprxtoAAAB4gZnus5fk55P8n6r62LT/L5K85xi+78zufmTafjTJmdP22UkeWnXcw1PbI3mBqtqTldm/bN++/RhKAAAAGN9MM3vdfUOSNyV5bHq8qbt//Xi+uLs7SR/D+67t7p3dvXPbNnd/AAAAOJRZZ/bS3fcmufc4v++xqjqrux+ZlmkemNr3Jzl31XHnTG0AAAAcg1nP2VsvtyS5ctq+MsnNq9rfNl2V8+IkT69a7gkAAMBRmnlm72hV1UeSXJJka1U9nOSnk7w3yY1VtTvJg1m5d1+SfDLJ65Pcn+TrSa6aV10AAAAngrmFve5+y2FeuvQQx3aSq+dVCwAAwIlmo5dxAgAAsAGEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQDgkG6++eZFlwAwtHn/nRX2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADCgLYv40qp6IMnXkjyX5Nnu3llVpyf5jSQ7kjyQ5Iru/soi6gMAANjsFjmz9/3dfUF375z2r0lye3efl+T2aR8AAIBjsEzLOHcluX7avj7JGxdXCgAAwOa2qLDXST5VVfuqas/UdmZ3PzJtP5rkzMWUBgAAsPkt5Jy9JN/b3fur6juS3FZVX1j9Ynd3VfWh3jiFwz1Jsn379vlXCgAAsAktZGavu/dPzweS3JTkoiSPVdVZSTI9HzjMe6/t7p3dvXPbtm0bVTIAAMCmsuFhr6peWlUvO7id5AeT3J3kliRXToddmeTmja4NAABgFItYxnlmkpuq6uD3f7i7b62qzyW5sap2J3kwyRULqA3guDyz94xFl/C8k3c/uegSAIAF2vCw191fTvKqQ7Q/meTSja4HAABgRMt06wUAAADWibAHAAAwIGEPAABgQMIeAADAgIQ9AACAAS3i1gscxjJdsh0AANjczOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAFtWXQBAABsTltPPyXvunXHosvYEMv079x6+imLLoFNQtgDAOCY7L3uhkWXAKzBMk4AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABrRl0QUAMB/P7D1j0SUkSU7e/eSiSwCAE5KZPQAAgAEJewAAAAMS9gAAAAYk7AEAAAzohL9Ay7JcwAAAAGA9Ld3MXlVdVlVfrKr7q+qaRdcDAACwGS1V2Kuqk5L8lySXJzk/yVuq6vzFVgUAALD5LFXYS3JRkvu7+8vd/RdJPppk14JrAgAA2HSW7Zy9s5M8tGr/4ST/YPUBVbUnyZ5p95mq+uKca9qa5Ik5f8dmpn/Wpn/Wpn/WNkb//Otar0/6m+v1QSeCffv2PVFVD87xK8b4+Zwf/bM2/bM2/bM2/fOtDjs+LlvYO6LuvjbJtRv1fVV1Z3fv3Kjv22z0z9r0z9r0z9r0D8eju7fN8/P9fK5N/6xN/6xN/6xN/8xu2ZZx7k9y7qr9c6Y2AAAAjsKyhb3PJTmvql5RVS9J8uYktyy4JgAAgE1nqZZxdvezVfX2JL+V5KQkH+zuexZc1oYtGd2k9M/a9M/a9M/a9A/LzM/n2vTP2vTP2vTP2vTPjKq7F10DAAAA62zZlnECAACwDoQ9AACAAQl7k6r6tqq6o6r+sKruqaqfmdpfUVWfrar7q+o3pgvHnLCq6qSq+oOq+h/Tvv6ZVNUDVfVHVXVXVd05tZ1eVbdV1Zem59MWXeeiVNWpVfXxqvpCVd1XVa/RPyuq6runn5uDj69W1Tv1D8vCGHlkxse1GSPXZow8PGPk8RH2vukbSX6gu1+V5IIkl1XVxUl+LskvdPffTvKVJLsXV+JSeEeS+1bt659v9f3dfcGqe79ck+T27j4vye3T/onqfUlu7e5XJnlVVn6O9E+S7v7i9HNzQZK/n+TrSW6K/mF5GCOPzPh4ZMbIwzNGHoYx8vgIe5Ne8cy0++Lp0Ul+IMnHp/brk7xx46tbDlV1TpJ/muTXpv2K/jmSXVnpl+QE7p+qOiXJ9yXZmyTd/Rfd/afRP4dyaZI/7u4Ho39YEsbItRkfj5m/cTFGHiVj5FES9laZlmDcleRAktuS/HGSP+3uZ6dDHk5y9oLKWwa/mOTfJ/nLaf+M6J/VOsmnqmpfVe2Z2s7s7kem7UeTnLmY0hbuFUkeT3LdtMzp16rqpdE/h/LmJB+ZtvUPS8MYuaZfjPHxSIyRh2eMnJ0x8igJe6t093PTFPE5SS5K8srFVrQ8quqHkhzo7n2LrmWJfW93X5jk8iRXV9X3rX6xV+5zcqLe62RLkguTfKC7X53kz/KC5RYneP8kSaZzet6Q5GMvfE3/sGjGyEMzPs7MGHl4xsgZGCOPjbB3CNPU+aeTvCbJqVV18Obz5yTZv6i6Fuy1Sd5QVQ8k+WhWlqe8L/rned29f3o+kJW15BcleayqzkqS6fnA4ipcqIeTPNzdn532P56VgU3/fKvLk/x+dz827esflo4x8q8wPs7AGLkmY+RsjJHHQNibVNW2qjp12v72JP8kKyfHfjrJj0yHXZnk5oUUuGDd/VPdfU5378jKFPpvd/e/jP5JklTVS6vqZQe3k/xgkruT3JKVfklO4P7p7keTPFRV3z01XZrk3uifF3pLvrk8JdE/LAlj5OEZH4/MGLk2Y+TMjJHHoFZmPamq78nKyZ0nZSUE39jdP1tV35WV/1N3epI/SPLW7v7G4ipdvKq6JMm/6+4f0j8rpn64adrdkuTD3f2eqjojyY1Jtid5MMkV3f3UgspcqKq6ICsXL3hJki8nuSrT71r0z8H/APp/Sb6ru5+e2vz8sBSMkbMxPh6aMfLIjJFrM0YeO2EPAABgQJZxAgAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAA2BSq6hNVta+q7qmqPVPb7qr6v1V1R1X9alX98tS+rap+s6o+Nz1eu9jqYeO5qToAAJtCVZ3e3U9V1bcn+VyS1yX530kuTPK1JL+d5A+7++1V9eEkv9Ldv1dV25P8Vnf/3YUVDwuwZdEFAADAjH6yqn542j43yY8m+Z3ufipJqupjSf7O9Po/TnJ+VR1878ur6uTufmYjC4ZFEvYAAFh6VXVJVgLca7r761X1P5N8IcnhZutelOTi7v7zDSkQlpBz9gAA2AxOSfKVKei9MsnFSV6a5B9V1WlVtSXJP191/KeS/MTBnaq6YCOLhWUg7AEAsBncmmRLVd2X5L1JPpNkf5L/lOSOrJy790CSp6fjfzLJzqr6fFXdm+TfbHjFsGAu0AIAwKZ18Dy8aWbvpiQf7O6bFl0XLAMzewAAbGbvrqq7ktyd5E+SfGKh1cASMbMHAAAwIDN7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMKD/D8WNpdKL6ccWAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sex\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cp\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trestbps\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "chol\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fbs\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3sAAAEGCAYAAADc7ozVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAAsTAAALEwEAmpwYAAAUiUlEQVR4nO3df7CldX0f8PdHVqQYFWQ3VhfIkoS0ZWgzMjuE1KlNQ9pRNKy1xsGpFS2T7SRqEklTaDsZ05jMlGkTo2lGS0QFxyYakspWTJ0MYpl2hLoESvhR65aoLEFZDRITY5T00z/ug7nhXODs3j333P3e12vmzH2e7/M9z/nsd+/d777v86u6OwAAAIzlKcsuAAAAgKNP2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAti27gPXYvn1779q1a9llALABbr311i92945l13GsMEcCbA1PND8e02Fv165d2b9//7LLAGADVNVnl13DscQcCbA1PNH86DROAACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAFtW3YBy/bTHz5/2SV801teesOySwAAAAbhyB4AAMCAhD0AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AMCaLrvssmWXAMA6CHsAwJruvPPOZZcAwDoIewAAAANaaNirqjdV1V1VdWdV/VpVnVBVZ1TVLVV1oKo+UFXHT32fNq0fmLbvWmRtAAAAI1tY2KuqnUl+LMnu7j47yXFJLkpyRZK3dvd3JnkoySXTWy5J8tDU/tapHwAAAEdg0adxbkvyV6pqW5ITkzyQ5PuTXDttvzrJy6blPdN6pu3nV1UtuD4AAIAhLSzsdff9Sf59ks9lJeQ9nOTWJF/u7kembgeT7JyWdya5b3rvI1P/Ux6736raW1X7q2r/oUOHFlU+AADAMW2Rp3GenJWjdWckeV6Spyd50Xr3291Xdvfu7t69Y8eO9e4OAABgSIs8jfMHkvx+dx/q7m8k+a0kL0hy0nRaZ5KcmuT+afn+JKclybT9WUm+tMD6AAAAhrXIsPe5JOdV1YnTtXfnJ7k7yY1JXjH1uTjJddPyvmk90/aPdXcvsD4AAIBhLfKavVuycqOV303ye9NnXZnksiSXVtWBrFyTd9X0lquSnDK1X5rk8kXVBgAAMLptT97lyHX3m5O8+THN9yY5d42+X0vyQ4usBwAAYKtY9KMXAAAAWAJhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGtNCwV1UnVdW1VfW/q+qeqvreqnp2Vf1OVX16+nry1Leq6u1VdaCq7qiqcxZZGwAAwMgWfWTvbUn+a3f/9STfneSeJJcnuaG7z0xyw7SeJC9Ocub02pvkHQuuDQAAYFgLC3tV9awkL0xyVZJ099e7+8tJ9iS5eup2dZKXTct7klzTK25OclJVPXdR9QEAAIxskUf2zkhyKMl7quq2qnpXVT09yXO6+4Gpz+eTPGda3pnkvlXvPzi1/SVVtbeq9lfV/kOHDi2wfAAAgGPXIsPetiTnJHlHdz8/yZ/kL07ZTJJ0dyfpw9lpd1/Z3bu7e/eOHTuOWrEAAAAjWWTYO5jkYHffMq1fm5Xw94VHT8+cvj44bb8/yWmr3n/q1AYAAMBhWljY6+7PJ7mvqv7a1HR+kruT7Ety8dR2cZLrpuV9SV4z3ZXzvCQPrzrdEwAAgMOwbcH7f2OS91fV8UnuTfK6rATMD1bVJUk+m+SVU9+PJLkgyYEkX536AgAAcAQWGva6+/Yku9fYdP4afTvJ6xdZDwAAwFax6OfsAQAAsATCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQHOFvaq6YZ42AAAANodtT7Sxqk5IcmKS7VV1cpKaNj0zyc4F1wYAAMAResKwl+SfJfmJJM9Lcmv+Iuz9UZL/sLiyAAAAWI8nDHvd/bYkb6uqN3b3L29QTQAAAKzTkx3ZS5J09y9X1d9Osmv1e7r7mgXVBQAAwDrMFfaq6n1JviPJ7Un+fGruJMIeAADAJjRX2EuyO8lZ3d2LLAYAAICjY97n7N2Z5K8ushAAAACOnnmP7G1PcndV/c8kf/ZoY3dfuJCqAAAABvWSl7xkpu36668/6p8zb9j7maP+yQAAAFvMWkHv0fajHfjmvRvnfzuqnwoAALCFrQ52jxcA12veu3F+JSt330yS45M8NcmfdPczF1IVAHDYqmpvkr1Jcvrppy+5GgCWbd4je894dLmqKsmeJOctqigA4PB195VJrkyS3bt3u4M2wBY37zV73zQ9fuFDVfXmJJcf/ZIAAADGtqhTN1eb9zTOl69afUpWnrv3tYVUBAAAMKjrr79+092N8wdXLT+S5DNZOZUTAACAw7CIYLeWea/Ze92iCwEAAODoeco8narq1Kr6z1X14PT6zao6ddHFAQAAcGTmCntJ3pNkX5LnTa//MrUBAACwCc0b9nZ093u6+5Hp9d4kOxZYFwAAAOswb9j7UlW9uqqOm16vTvKlRRYGAADAkZs37P3TJK9M8vkkDyR5RZLXLqgmAAAA1mnesPezSS7u7h3d/a1ZCX//Zp43TkcCb6uqD0/rZ1TVLVV1oKo+UFXHT+1Pm9YPTNt3HcGfBwAAgMwf9v5Wdz/06Ep3/2GS58/53h9Pcs+q9SuSvLW7vzPJQ0kumdovSfLQ1P7WqR8AAABHYN6w95SqOvnRlap6duZ4Rt/0eIaXJHnXtF5Jvj/JtVOXq5O8bFreM61n2n7+1B8AAIDDNNdD1ZP8QpJPVNVvTOs/lOTn53jfLyX5F0meMa2fkuTL3f3ItH4wyc5peWeS+5Kkux+pqoen/l9cvcOq2ptkb5Kcfvrpc5YPAACwtcx1ZK+7r0ny8iRfmF4v7+73PdF7quqlSR7s7lvXXeVfruXK7t7d3bt37PD0BwAAgLXMe2Qv3X13krsPY98vSHJhVV2Q5IQkz0zytiQnVdW26ejeqUnun/rfn+S0JAeraluSZ8XjHQAAAI7IvNfsHbbu/pfdfWp370pyUZKPdfc/TnJjVh7dkCQXJ7luWt43rWfa/rHu7kXVBwAAMLKFhb0ncFmSS6vqQFauybtqar8qySlT+6VJLl9CbQAAAEOY+zTO9ejujyf5+LR8b5Jz1+jztazc+AUAAIB1WsaRPQAAABZM2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAABgQMIeAADAgIQ9AACAAS0s7FXVaVV1Y1XdXVV3VdWPT+3PrqrfqapPT19Pntqrqt5eVQeq6o6qOmdRtQEAAIxukUf2Hknyk919VpLzkry+qs5KcnmSG7r7zCQ3TOtJ8uIkZ06vvUnescDaAAAAhrawsNfdD3T3707LX0lyT5KdSfYkuXrqdnWSl03Le5Jc0ytuTnJSVT13UfUBAACMbEOu2auqXUmen+SWJM/p7gemTZ9P8pxpeWeS+1a97eDU9th97a2q/VW1/9ChQ4srGgAA4Bi28LBXVd+S5DeT/ER3/9Hqbd3dSfpw9tfdV3b37u7evWPHjqNYKQAAwDgWGvaq6qlZCXrv7+7fmpq/8OjpmdPXB6f2+5Octurtp05tAAAAHKZF3o2zklyV5J7u/sVVm/YluXhavjjJdavaXzPdlfO8JA+vOt0TAACAw7Btgft+QZJ/kuT3qur2qe1fJfm3ST5YVZck+WySV07bPpLkgiQHknw1yesWWBsAAMDQFhb2uvu/J6nH2Xz+Gv07yesXVQ8AAMBWsiF34wQAAGBjCXsAAAADEvYAAAAGJOwBAAAMSNgDAAAY0CIfvQDAEv30h2dufLwUb3npDcsuAQC2JEf2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AYE1nn332sksAYB2EPQBgTVdcccWySwBgHYQ9AACAAQl7AAAAAxL2AAAABiTsAQAADEjYAwAAGJCwBwAAMCBhDwAAYEDCHgAAwICEPQAAgAEJewAAAAMS9gAAAAYk7AEAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABbaqwV1UvqqpPVdWBqrp82fUAAAAcqzZN2Kuq45L8SpIXJzkryauq6qzlVgUAAHBs2jRhL8m5SQ50973d/fUkv55kz5JrAgAAOCZtW3YBq+xMct+q9YNJvuexnapqb5K90+ofV9Wn1vm525N8cZ37OCp+LrXsElbbNOOyiRiTWcZkljF5jJ9LHa0x+bajsI8t49Zbb/1iVX12nbvx/TzLmMwyJrOMydqMy6yjMSaPOz9uprA3l+6+MsmVR2t/VbW/u3cfrf2NwrjMMiazjMksYzLLmCxHd+9Y7z783c0yJrOMySxjsjbjMmvRY7KZTuO8P8lpq9ZPndoAAAA4TJsp7H0yyZlVdUZVHZ/koiT7llwTAADAMWnTnMbZ3Y9U1RuSfDTJcUne3d13bcBHH7VTQgdjXGYZk1nGZJYxmWVMjl3+7mYZk1nGZJYxWZtxmbXQManuXuT+AQAAWILNdBonAAAAR4mwBwAAMKAtE/aq6kVV9amqOlBVl6+x/WlV9YFp+y1VtWsJZW6oOcbk0qq6u6ruqKobqmr4Z1w92Zis6vePqqqrakvcPniecamqV07fL3dV1X/a6Bo32hw/P6dX1Y1Vddv0M3TBMurcKFX17qp6sKrufJztVVVvn8brjqo6Z6Nr5PGZI2eZI2eZI2eZH2eZH2ctdY7s7uFfWbnhy/9N8u1Jjk/yv5Kc9Zg+P5rkndPyRUk+sOy6N8GY/L0kJ07LP2JMvtnvGUluSnJzkt3LrnszjEuSM5PcluTkaf1bl133JhiTK5P8yLR8VpLPLLvuBY/JC5Ock+TOx9l+QZLfTlJJzktyy7Jr9vrm34058sjGxBy5xedI8+MRj8mWmh+nP+fS5sitcmTv3CQHuvve7v56kl9PsucxffYkuXpavjbJ+VVVG1jjRnvSMenuG7v7q9PqzVl59uHI5vk+SZK3JLkiydc2srglmmdcfjjJr3T3Q0nS3Q9ucI0bbZ4x6STPnJafleQPNrC+DdfdNyX5wyfosifJNb3i5iQnVdVzN6Y6noQ5cpY5cpY5cpb5cZb5cQ3LnCO3StjbmeS+VesHp7Y1+3T3I0keTnLKhlS3HPOMyWqXZOU3DiN70jGZDquf1t3Xb2RhSzbP98p3JfmuqvofVXVzVb1ow6pbjnnG5GeSvLqqDib5SJI3bkxpm9bh/pvDxjFHzjJHzjJHzjI/zjI/HpmFzZGb5jl7bF5V9eoku5P83WXXskxV9ZQkv5jktUsuZTPalpVTVb4vK7/dvqmq/mZ3f3mZRS3Zq5K8t7t/oaq+N8n7qurs7v5/yy4MOHrMkSvMkY/L/DjL/LiBtsqRvfuTnLZq/dSpbc0+VbUtK4eVv7Qh1S3HPGOSqvqBJP86yYXd/WcbVNuyPNmYPCPJ2Uk+XlWfyco51fu2wAXo83yvHEyyr7u/0d2/n+T/ZGVyG9U8Y3JJkg8mSXd/IskJSbZvSHWb01z/5rAU5shZ5shZ5shZ5sdZ5scjs7A5cquEvU8mObOqzqiq47Nycfm+x/TZl+TiafkVST7W0xWTg3rSMamq5yf5j1mZxEY/xzx5kjHp7oe7e3t37+ruXVm5RuPC7t6/nHI3zDw/Px/Kym8tU1Xbs3Layr0bWONGm2dMPpfk/CSpqr+Rlcns0IZWubnsS/Ka6Y5j5yV5uLsfWHZRJDFHrsUcOcscOcv8OMv8eGQWNkduidM4u/uRqnpDko9m5S5B7+7uu6rqZ5Ps7+59Sa7KymHkA1m5gPKi5VW8eHOOyb9L8i1JfmO6Dv9z3X3h0opesDnHZMuZc1w+muQfVNXdSf48yU9197C/9Z9zTH4yya9W1ZuycjH6a0f+z3FV/VpW/kOzfboO481Jnpok3f3OrFyXcUGSA0m+muR1y6mUxzJHzjJHzjJHzjI/zjI/rm2Zc2QNPrYAAABb0lY5jRMAAGBLEfYAAAAGJOwBAAAMSNgDAAAYkLAHAAAwIGEPAIBjSlX9WFXdU1Xvr6p/vux6YLMS9gAAONb8aJK/n+TTyy4ENjNhDwCAY0ZVvTPJtyf57SRvSvLdVfWJqvp0Vf3w1Oe5VXVTVd1eVXdW1d9ZZs2wLB6qDgDAMaWqPpNkd5I3JPmHSc5L8vQktyX5niSvSnJCd/98VR2X5MTu/sqSyoWl2bbsAgAAYB2u6+4/TfKnVXVjknOTfDLJu6vqqUk+1N23L7NAWBancQIAcCx77Glq3d03JXlhkvuTvLeqXrPxZcHyCXsAABzL9lTVCVV1SpLvS/LJqvq2JF/o7l9N8q4k5yyzQFgWp3ECAHAsuyPJjUm2J3lLd/9BVV2c5Keq6htJ/jiJI3tsSW7QAgAAMCCncQIAAAxI2AMAABiQsAcAADAgYQ8AAGBAwh4AAMCAhD0AAIABCXsAAAAD+v/CHy1tOwQvcQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "restecg\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3sAAAEGCAYAAADc7ozVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAAsTAAALEwEAmpwYAAAY9UlEQVR4nO3de9Rld1kf8O9jJlyCIRcyxpgEBiSVxjZKOtKI1ALxEuIlqUWEJRowrvGCVhZdLVhsqy67hGVbFLW2WYAm1nIRBSLFS0xiXRUTmGCYhGBgDFAyBjKEEEAKGnz6x9lDT8a5nDcz+z3n7Pl81jrr3fu3f3uf593vnvc333dfTnV3AAAAmJYvWnYBAAAAHH3CHgAAwAQJewAAABMk7AEAAEyQsAcAADBBW5ZdwJE47bTTetu2bcsuA4BNcNNNN32su7cuu451YYwEODYcanxc67C3bdu27Ny5c9llALAJqupDy65hnRgjAY4NhxofXcYJAAAwQcIeAADABAl7AAAAEyTsAQAATJCwBwAAMEHCHgAAwAQJewAAABMk7AEAAEyQsAcAADBBW5ZdALAxLz/p0mWX8AUvvu/Nyy4BAICDcGYPAABggkYNe1X1waq6papurqqdQ9upVXVNVb1/+HrK0F5V9cqq2l1Vu6rq/DFrAwAAmLLNOLP3tO7+6u7ePsy/JMm13X1OkmuH+SR5RpJzhteOJL+yCbUBAABM0jIu47wkyZXD9JVJLp1rv6pnbkhyclWdsYT6AAAA1t7YYa+T/EFV3VRVO4a207v7rmH6I0lOH6bPTPLhuXXvHNoeoKp2VNXOqtq5d+/eseoGAABYa2M/jfMp3b2nqr4kyTVV9efzC7u7q6o3ssHuviLJFUmyffv2Da17IJ5sCAAATNGoZ/a6e8/w9e4kb0rypCQf3Xd55vD17qH7niRnz61+1tAGAADABo0W9qrqEVV14r7pJN+U5NYkVye5bOh2WZK3DNNXJ/ne4amcFyS5b+5yTwAAADZgzMs4T0/ypqra9z7/o7t/r6remeQNVXV5kg8ledbQ/21JLk6yO8lnkjx/xNoAAAAmbbSw1913JPmqA7Tfk+TCA7R3kheMVQ8AAMCxZOwHtAAAa+rCpz09e+/52LLLgIWd8PCH54Ybb1x2GbAyhD0A4ID23vOxfOsnHrvsMmBhb80Hll0CrJRlfKg6AAAAIxP2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkS9gAAACZI2AMAAJggYQ8AAGCChD0AAIAJEvYAAAAmSNgDAACYIGEPAABggoQ9AACACRL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkS9gAAACZI2AMAAJggYQ8AAGCChD0AAIAJEvYAAAAmSNgDAACYIGEPAABggoQ9AACACRL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkaPexV1XFV9WdV9dZh/rFVdWNV7a6q11fVQ4b2hw7zu4fl28auDQAAYKo248zejyV579z8y5O8orsfn+TeJJcP7ZcnuXdof8XQDwAAgAdh1LBXVWcl+ZYkrxrmK8nTk7xx6HJlkkuH6UuG+QzLLxz6AwAAsEFjn9n7+ST/OsnfDvOPSvKJ7r5/mL8zyZnD9JlJPpwkw/L7hv4PUFU7qmpnVe3cu3fviKUDAACsr9HCXlV9a5K7u/umo7nd7r6iu7d39/atW7cezU0DAABMxpYRt/11Sb69qi5O8rAkj0zyC0lOrqotw9m7s5LsGfrvSXJ2kjurakuSk5LcM2J9AAAAkzXamb3u/vHuPqu7tyV5dpLruvu7k1yf5JlDt8uSvGWYvnqYz7D8uu7useoDAACYsmV8zt6Lk7yoqnZndk/eq4f2Vyd51ND+oiQvWUJtAAAAkzDmZZxf0N1/lOSPhuk7kjzpAH0+m+Q7N6MeAACAqVvGmT0AAABGJuwBAABMkLAHAAAwQcIeAADABAl7AAAAS3DeeeeNun1hDwAAYIKEPQAAgAkS9gAAACZI2AMAAJggYQ8AAGCChD0AAIAJEvYAAAAmSNgDAACYIGEPAABggoQ9AACACRL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkS9gAAACZI2AMAAJggYQ8AAGCChD0AAIAJEvYAAAAmSNgDAACYIGEPAABggoQ9AACACRL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkS9gAAACZotLBXVQ+rqndU1bur6j1V9VND+2Or6saq2l1Vr6+qhwztDx3mdw/Lt41VGwAAwNSNeWbvc0me3t1fleSrk1xUVRckeXmSV3T345Pcm+Tyof/lSe4d2l8x9AMAAOBBGC3s9cynh9njh1cneXqSNw7tVya5dJi+ZJjPsPzCqqqx6gMAAJiyUe/Zq6rjqurmJHcnuSbJXyT5RHffP3S5M8mZw/SZST6cJMPy+5I86gDb3FFVO6tq5969e8csHwAAYG0tFPaq6tpF2vbX3Z/v7q9OclaSJyV5wkYLPMA2r+ju7d29fevWrUe6OQAAgEnacqiFVfWwJCckOa2qTkmy77LKR+b/n5E7rO7+RFVdn+Rrk5xcVVuGs3dnJdkzdNuT5Owkd1bVliQnJblnI98MAAAAM4c7s/cDSW7K7IzcTXOvtyT5pUOtWFVbq+rkYfrhSb4xyXuTXJ/kmUO3y4ZtJcnVw3yG5dd1d2/gewEAAGBwyDN73f0LSX6hqn60u39xg9s+I8mVVXVcZqHyDd391qq6LcnrqupnkvxZklcP/V+d5NeraneSjyd59gbfDwAAgMEhw94+3f2LVfXkJNvm1+nuqw6xzq4kTzxA+x2Z3b+3f/tnk3znIvUAAABwaAuFvar69SRfnuTmJJ8fmjvJQcMeAAAAy7NQ2EuyPcm57qEDAABYD4t+zt6tSb50zEIAAAA4ehY9s3daktuq6h1JPrevsbu/fZSqAAAAOCKLhr2fHLMIAAAAjq5Fn8b5v8YuBAAAgKNn0adxfiqzp28myUOSHJ/kr7r7kWMVBgBsTFXtSLIjSR796EcvuRpYjvPOO2/ZJcDKWPTM3on7pquqklyS5IKxigIANq67r0hyRZJs377dE7Q5Ju3atWvZJcDCxv7jxKJP4/yCnnlzkm8++uUAAABwNCx6Ged3zM1+UWafu/fZUSoCAADgiC36NM5vm5u+P8kHM7uUEwAAgBW06D17zx+7EAAAAI6ehe7Zq6qzqupNVXX38Pqtqjpr7OIAAAB4cBZ9QMuvJrk6yZcNr98Z2gAAAFhBi4a9rd39q919//D6tSRbR6wLAACAI7Bo2Lunqp5bVccNr+cmuWfMwgAAAHjwFg1735fkWUk+kuSuJM9M8ryRagIAAOAILfrRCz+d5LLuvjdJqurUJP8xsxAIAADAiln0zN55+4JeknT3x5M8cZySAAAAOFKLhr0vqqpT9s0MZ/YWPSsIAADAJls0sP2nJH9aVb85zH9nkv8wTkkAAAAcqYXCXndfVVU7kzx9aPqO7r5tvLIAAAA4EgtfijmEOwEPAABgDSx6zx4AAABrRNgDAACYIGEPAABggoQ9AACACRL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkaLexV1dlVdX1V3VZV76mqHxvaT62qa6rq/cPXU4b2qqpXVtXuqtpVVeePVRsAAMDUjXlm7/4k/7K7z01yQZIXVNW5SV6S5NruPifJtcN8kjwjyTnDa0eSXxmxNgAAgEkbLex1913d/a5h+lNJ3pvkzCSXJLly6HZlkkuH6UuSXNUzNyQ5uarOGKs+AACAKduUe/aqaluSJya5Mcnp3X3XsOgjSU4fps9M8uG51e4c2vbf1o6q2llVO/fu3Tte0QAAAGts9LBXVV+c5LeSvLC7Pzm/rLs7SW9ke919RXdv7+7tW7duPYqVAgAATMeoYa+qjs8s6P1Gd//20PzRfZdnDl/vHtr3JDl7bvWzhjYAAAA2aMyncVaSVyd5b3f/57lFVye5bJi+LMlb5tq/d3gq5wVJ7pu73BMAAIAN2DLitr8uyfckuaWqbh7a/k2SlyV5Q1VdnuRDSZ41LHtbkouT7E7ymSTPH7E2AACASRst7HX3/05SB1l84QH6d5IXjFUPAADAsWRTnsYJAADA5hL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkS9gAAACZI2AMAAJggYQ8AAGCChD0AAIAJEvYAAAAmSNgDAACYIGEPAABgCXbt2jXq9oU9AACACRL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkS9gAAACZI2AMAAJggYQ8AAGCChD0AAIAJEvYAAAAmSNgDAACYIGEPAABggoQ9AACACRL2AAAAJkjYAwAAmCBhDwAAYIKEPQAAgAkS9gAAACZI2AMAAJig0cJeVb2mqu6uqlvn2k6tqmuq6v3D11OG9qqqV1bV7qraVVXnj1UXAADAsWDMM3u/luSi/dpekuTa7j4nybXDfJI8I8k5w2tHkl8ZsS4AAIDJGy3sdfcfJ/n4fs2XJLlymL4yyaVz7Vf1zA1JTq6qM8aqDQAAYOo2+56907v7rmH6I0lOH6bPTPLhuX53Dm1/R1XtqKqdVbVz796941UKAACwxpb2gJbu7iT9INa7oru3d/f2rVu3jlAZAADA+tvssPfRfZdnDl/vHtr3JDl7rt9ZQxsAAAAPwmaHvauTXDZMX5bkLXPt3zs8lfOCJPfNXe4JAADABm0Za8NV9dokT01yWlXdmeTfJ3lZkjdU1eVJPpTkWUP3tyW5OMnuJJ9J8vyx6gIAADgWjBb2uvs5B1l04QH6dpIXjFULAADAsWZpD2gBAABgPMIeAADABAl7AAAAEyTsAQAATJCwBwAAMEHCHgAAwAQJewAAABMk7AEAAEyQsAcAADBBwh4AAMAECXsAAAATJOwBAABMkLAHAAAwQVuWXQAAsJq2Puq0vDUfWHYZsLATHv7wZZcAK0XYAwAO6Nrrr1t2CQAcAZdxAgAATJCwBwAAMEHCHgAAwAQJewAAABMk7AEAAEyQsAcAADBBwh4AAMAECXsAAAATJOwBAABMkLAHAAAwQcIeAADABAl7AAAAEyTsAQAATJCwBwAAMEFbll0AAON4+UmXLruEJMmL73vzsksAgGOSM3sAAAATJOwBAABM0EqFvaq6qKpur6rdVfWSZdcDAACwrlYm7FXVcUl+Ockzkpyb5DlVde5yqwIAAFhPKxP2kjwpye7uvqO7/zrJ65JcsuSaAAAA1lJ197JrSJJU1TOTXNTd3z/Mf0+Sf9zdP7Jfvx1JdgyzX5Hk9iN869OSfOwIt7GZ1qletY5jnWpN1qtetY7jaNX6mO7eehS2c0yoqr1JPnSEmzkWj7PNsE61JutVr1rHsU61JutV79Go9aDj49p99EJ3X5HkiqO1vara2d3bj9b2xrZO9ap1HOtUa7Je9ap1HOtU65QcjWC8Tj87tY5nnepV6zjWqdZkveodu9ZVuoxzT5Kz5+bPGtoAAADYoFUKe+9Mck5VPbaqHpLk2UmuXnJNAAAAa2llLuPs7vur6keS/H6S45K8prvfswlvfdQuCd0k61SvWsexTrUm61WvWsexTrXyQOv0s1PreNapXrWOY51qTdar3lFrXZkHtAAAAHD0rNJlnAAAABwlwh4AAMAETTrsVdVFVXV7Ve2uqpccYPlDq+r1w/Ibq2rb3LIfH9pvr6pvXoFaX1RVt1XVrqq6tqoeM7fs81V18/Aa/aE2C9T6vKraO1fT988tu6yq3j+8Lhu71gXrfcVcre+rqk/MLdu0fVtVr6mqu6vq1oMsr6p65fB97Kqq8+eWbep+XaDW7x5qvKWq3l5VXzW37IND+81VtXPsWhes96lVdd/cz/rfzS075PGzhFr/1Vydtw7H6KnDsk3dt1V1dlVdP/xuek9V/dgB+qzMccsDGSOXVuvKjJHrMj4O72eMXE6txscHV+vqjI/dPclXZg95+Yskj0vykCTvTnLufn1+OMl/HaafneT1w/S5Q/+HJnnssJ3jllzr05KcMEz/0L5ah/lPr9h+fV6SXzrAuqcmuWP4esowfcqy692v/49m9nCgZezbr09yfpJbD7L84iS/m6SSXJDkxiXu18PV+uR9NSR5xr5ah/kPJjlts/brgvU+Nclbj/T42Yxa9+v7bUmuW9a+TXJGkvOH6ROTvO8Avw9W5rj1esDPxRi5vFqflxUYIzf6+y1LHB+H9zNGLqfWp8b4+GBqXZnxccpn9p6UZHd339Hdf53kdUku2a/PJUmuHKbfmOTCqqqh/XXd/bnu/kCS3cP2llZrd1/f3Z8ZZm/I7HMIl2GR/Xow35zkmu7+eHffm+SaJBeNVOc+G633OUleO3JNB9Tdf5zk44fockmSq3rmhiQnV9UZWcJ+PVyt3f32oZZkucfrvnoOt28P5kiO9wdlg7Uu7XhNku6+q7vfNUx/Ksl7k5y5X7eVOW55AGPkONZpjFyb8TExRo7F+DiOVRofpxz2zkzy4bn5O/N3d/IX+nT3/UnuS/KoBdc9mjb6fpdn9peAfR5WVTur6oaqunSE+uYtWus/H05Jv7Gqzt7gukfTwu85XPbz2CTXzTVv5r49nIN9L8vYrxux//HaSf6gqm6qqh1LqulAvraq3l1Vv1tVXzm0rey+raoTMvvl/1tzzUvbtzW7xO+JSW7cb9G6HrdTZ4wcxzqNkVMaH5P1/V2zDmOk8fHI6tmWJY6PK/M5eyymqp6bZHuSfzrX/Jju3lNVj0tyXVXd0t1/sZwKkyS/k+S13f25qvqBzP4y/PQl1rOoZyd5Y3d/fq5t1fbtWqmqp2U2kD1lrvkpwz79kiTXVNWfD3+tW6Z3Zfaz/nRVXZzkzUnOWW5Jh/VtSf6ku+f/yrmUfVtVX5zZoPrC7v7k2O8HB2OMHI3xcQRrMkYaH4/AKoyPUz6ztyfJ2XPzZw1tB+xTVVuSnJTkngXXPZoWer+q+oYkL03y7d39uX3t3b1n+HpHkj/K7K8HS6u1u++Zq+9VSf7RouuOYCPv+ezsd8p/k/ft4Rzse1nGfj2sqjovs5//Jd19z772uX16d5I3ZdzLvxbS3Z/s7k8P029LcnxVnZYV3beDQx2vm7Zvq+r4zAay3+ju3z5Al7U6bo8hxsgl1bpCY+SUxsdkzX7XrMsYaXx88FZmfOxNulFxs1+ZnbW8I7PLDvbdOPqV+/V5QR548/kbhumvzANvPr8j4958vkitT8zsRthz9ms/JclDh+nTkrw/I94gu2CtZ8xN/7MkNwzTpyb5wFDzKcP0qcs+DoZ+T8js5t1a1r4d3mdbDn6T9LfkgTfyvmNZ+3WBWh+d2X08T96v/RFJTpybfnuSi8audYF6v3Tfzz6zAeD/DPt5oeNnM2sdlp+U2X0Lj1jmvh320VVJfv4QfVbquPX6ws/FGLm8WldijFz091tWZHwc3utQv8dX6nfNYWpdqTHyMLUaHx9cnSszPo76A1n2K7On3LwvswHgpUPbT2f2V78keViS3xz+wb0jyePm1n3psN7tSZ6xArX+YZKPJrl5eF09tD85yS3DP7Jbkly+ArX+bJL3DDVdn+QJc+t+37C/dyd5/iocB8P8TyZ52X7rbeq+zeyvUHcl+ZvMrs++PMkPJvnBYXkl+eXh+7glyfZl7dcFan1VknvnjtedQ/vjhv357uEYeekmHQOHq/dH5o7ZGzI3AB/o+FlmrUOf52X2gIz59TZ932Z26VEn2TX3s754VY9br7/z8zNGLqfWlRkjD1frMP+TWfL4OLynMXI5tRofH1ytKzM+7kvqAAAATMiU79kDAAA4Zgl7AAAAEyTsAQAATJCwBwAAMEHCHgAAwAQJewAArK2qemFVnbDsOmAV+egFAABWRlVVZv9H/dsF+38ws88o+9iohcEacmYPAIClqqptVXV7VV2V5NYk/7aq3llVu6rqp4Y+j6iq/1lV766qW6vqu6rqXyT5siTXV9X1Q79vqqo/rap3VdVvVtUXD+1fU1VvH9Z/R1WdWFUnVNUbquq2qnpTVd1YVduXtR/gaNuy7AIAACDJOUkuS/LIJM9M8qQkleTqqvr6JFuT/GV3f0uSVNVJ3X1fVb0oydO6+2NVdVqSn0jyDd39V1X14iQvqqqXJXl9ku/q7ndW1SOT/N8kL0xyb3efW1X/IMnNm/kNw9ic2QMAYBV8qLtvSPJNw+vPkrwryRMyC4K3JPnGqnp5Vf2T7r7vANu4IMm5Sf6kqm7OLDw+JslXJLmru9+ZJN39ye6+P8lTkrxuaLs1ya4Rvz/YdM7sAQCwCv5q+FpJfra7/9v+Harq/CQXJ/mZqrq2u396/y5Jrunu5+y33j8co2BYdc7sAQCwSn4/yffN3Wt3ZlV9SVV9WZLPdPd/T/JzSc4f+n8qyYnD9A1Jvq6qHj+s+4iq+ntJbk9yRlV9zdB+YlVtSfInSZ41tJ2bRChkUpzZAwBgZXT3H1TV30/yp7MHc+bTSZ6b5PFJfq6q/jbJ3yT5oWGVK5L8XlX9ZXc/raqel+S1VfXQYflPdPf7quq7kvxiVT08s/v1viHJf0lyZVXdluTPk7wnyYEuD4W15KMXAAA4JlXVcUmO7+7PVtWXJ/nDJF/R3X+95NLgqHBmDwCAY9UJmX1sw/GZ3e/3w4IeU+LMHgAAwAR5QAsAAMAECXsAAAATJOwBAABMkLAHAAAwQcIeAADABP0/VEmmrlkS9UAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "thalach\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "exang\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "oldpeak\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "slope\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ca\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "thal\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "target\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cols = df.select_dtypes(include=np.number).columns.tolist()\n",
    "colors = [\"#F39C12\", \"#1ABC9C\", \"#BB8FCE\", \"#E74C3C\", \"#707B7C\", \"#7CB342\", \"#880E4F\", \"#8D6E63\", \"#99FF66\", \"#FF9966\", \"#2471A3\", \"#C0392B\", \"#FF33CC\", \"#27AE60\"]\n",
    "color_index = 0\n",
    "for col in cols:\n",
    "    print(col)\n",
    "    plt.figure(figsize = (15, 4))\n",
    "    plt.subplot(1, 2, 1)\n",
    "    df[col].hist(grid=False, color=colors[color_index])\n",
    "    plt.ylabel('count')\n",
    "    plt.subplot(1, 2, 2)\n",
    "    sns.boxplot(x=df[col], color=colors[color_index])\n",
    "    color_index += 1\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x24f70d615e0>]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(df.thalach, color=\"SeaGreen\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 15))\n",
    "sns.heatmap(df.corr(), annot=True, square=True, fmt='0.2f')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NOo1N_ZaKHPg"
   },
   "source": [
    "## Analysis of Features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1fnTddu9sTll"
   },
   "source": [
    "As you have probably noticed, in the dataset that you checked, some features are discrete and the rest are continuous. The following code specifies continuous and discrete features for you. **Note**: In the cells below, `df` is the same dataframe that you loaded in the previous section. You can change its name according to your code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "RIcPajwSJj11",
    "outputId": "d9551a1e-e668-4d00-8dce-87c48bb62673"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 14 features found in the data\n"
     ]
    }
   ],
   "source": [
    "feature_list = [feature for feature in df.columns]\n",
    "print(\"There are\",len(feature_list),\"features found in the data\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "o3n2gCAoJmdr",
    "outputId": "65132134-22aa-49bb-8d40-384f016f09ab"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Discrete Variables Count: 9\n",
      "Discrete features are  ['sex', 'cp', 'fbs', 'restecg', 'exang', 'slope', 'ca', 'thal', 'target']\n",
      "Continuous Variables Count: 5\n",
      "Continuous features are  ['age', 'trestbps', 'chol', 'thalach', 'oldpeak']\n"
     ]
    }
   ],
   "source": [
    "#Print Discrete Feature Data\n",
    "discrete_feature=[feature for feature in feature_list if len(df[feature].unique())<25]\n",
    "print(\"Discrete Variables Count: {}\".format(len(discrete_feature)))\n",
    "print(\"Discrete features are \",discrete_feature)\n",
    "\n",
    "#Print Continuous Feature Data\n",
    "cont_feature=[feature for feature in feature_list if len(df[feature].unique())>25]\n",
    "print(\"Continuous Variables Count: {}\".format(len(cont_feature)))\n",
    "print(\"Continuous features are \",cont_feature)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zygi7E6ekk4q"
   },
   "source": [
    "# Perceptron"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "js3fx5A2lnR6"
   },
   "source": [
    "### Stochastic Gradient Descent\n",
    "\n",
    "We will implement the perceptron algorithm in python 3 and numpy. The perceptron will learn using the stochastic gradient descent algorithm (SGD). Gradient Descent minimizes a function by following the gradients of the cost function.\n",
    "\n",
    "### Calculating the Error\n",
    "\n",
    "To calculate the error of a prediction we first need to define the objective function of the perceptron.\n",
    "\n",
    "#### Hinge Loss Function\n",
    "\n",
    "To do this, we need to define the loss function, to calculate the prediction error. We will use hinge loss for our perceptron:\n",
    "\n",
    "$$c(x, y, f(x)) = (1 - y * f(x))_+$$\n",
    "\n",
    "$c$ is the loss function, $x$ the sample, $y$ is the true label, $f(x)$ the predicted label."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XQRK4oiglp2z"
   },
   "source": [
    "This means the following:\n",
    "$$\n",
    "c(x, y, f(x))=\n",
    "\\begin{cases}\n",
    "    0,& \\text{if } y * f(x)\\geq 1\\\\\n",
    "    1-y*f(x),              & \\text{else}\n",
    "\\end{cases}\n",
    "$$\n",
    "\n",
    "So consider, if y and f(x) are signed values $(+1,-1)$:\n",
    "\n",
    "<ul>\n",
    "    <li>the loss is 0, if $y*f(x)$ are positive, respective both values have the same sign.</li>\n",
    "    <li>loss is $1-y*f(x)$ if $y*f(x)$ is negative</li>\n",
    "</ul>\n",
    "\n",
    "#### Objective Function\n",
    "\n",
    "As we defined the loss function, we can now define the objective function for the perceptron:\n",
    "\n",
    "$$l_i(w) = \\big(-y_i \\langle x_i,w \\rangle\\big)_+$$\n",
    "\n",
    "We can write this without the dot product with a sum sign:\n",
    "\n",
    "$$l_i(w) = (-y_i \\sum_{i=1}^n x_iw)_+$$\n",
    "\n",
    "So the sample $x_i$ is misclassified, if $y_i \\langle x_i,w \\rangle \\leq 0$. The general goal is, to find the global minima of this function, respectively find a parameter $w$, where the error is zero.\n",
    "\n",
    "#### Derive the Objective Function\n",
    "\n",
    "To do this we need the gradients of the objective function. The gradient of a function $f$ is the vector of its partial derivatives. The gradient can be calculated by the partially derivative of the objective function.\n",
    "\n",
    "$$ \\nabla l_i(w) = -y_i x_i $$\n",
    "\n",
    "\n",
    "This means, if we have a misclassified sample $x_i$, respectively $ y_i \\langle x_i,w \\rangle \\leq 0 $, update the weight vector\n",
    "$w$ by moving it in the direction of the misclassified sample.\n",
    "\n",
    "\n",
    "$$w = w + y_i x_i$$\n",
    "\n",
    "With this update rule in mind, we can start writing our perceptron algorithm in python."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_eiYih0TmZhr"
   },
   "source": [
    "## Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "i_91Dsdlmwvr"
   },
   "source": [
    "\n",
    "Finally we can code our SGD algorithm using our update rule. To keep it simple, we will linearly loop over the sample set. For larger data sets it makes sence, to randomly pick a sample during each iteration in the for-loop."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "dpCynLlEmxQY"
   },
   "outputs": [],
   "source": [
    "def perceptron_sgd(X, Y):\n",
    "    '''\n",
    "    # TODO: implement the below\n",
    "\n",
    "    Initialize the weight vector for the perceptron with zeros\n",
    "    Set the learning rate\n",
    "    Set the number of epochs\n",
    "    Iterate n times over the whole dataset.\n",
    "    Iterate over each sample in the dataset\n",
    "    Misclassification condition y_i * (x_i . w) <= 0\n",
    "    Update rule for the weights w = w + y_i * x_i including the learning rate\n",
    "\n",
    "    '''\n",
    "    w = np.zeros(X.shape[1])\n",
    "    learning_rate = 0.1\n",
    "    epochs = 130\n",
    "\n",
    "    for epoch in range(epochs):\n",
    "        for idx, x in enumerate(X):\n",
    "            if (Y[idx] * np.dot(X[idx], w)) <= 0:\n",
    "                w = w + learning_rate * X[idx] * Y[idx]\n",
    "\n",
    "    return w"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1tIY7wrXm5FZ"
   },
   "source": [
    "### Let the Perceptron learn!\n",
    "\n",
    "Next we can execute our code to train a classifier model. To see the learning progress of the perceptron, we add a plotting feature to our algorithm, counting the total error in each epoch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "MJFiM-IBm9xV"
   },
   "outputs": [],
   "source": [
    "def perceptron_sgd_plot(X, Y):\n",
    "    '''\n",
    "    TODO: train perceptron and plot the total loss in each epoch. You may not use the above perceptron_sgd(X, Y) function.\n",
    "\n",
    "    :param X: data samples\n",
    "    :param Y: data labels\n",
    "    :return: weight vector as a numpy array\n",
    "    '''\n",
    "\n",
    "    w = np.zeros(X.shape[1])\n",
    "    learning_rate = 0.1\n",
    "    epochs = 130\n",
    "    errors = []\n",
    "\n",
    "    for epoch in range(epochs):\n",
    "        count = 0\n",
    "        for idx, x in enumerate(X):\n",
    "            if (Y[idx] * np.dot(X[idx], w)) <= 0:\n",
    "                count += (np.dot(X[idx], w) * Y[idx])\n",
    "                w = w + learning_rate * X[idx] * Y[idx]\n",
    "        errors.append(count * -1)\n",
    "    \n",
    "    plt.figure(figsize = (15, 4))\n",
    "    plt.plot(errors)\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylabel('Total Loss')\n",
    "\n",
    "    return w"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "hG3b6rbvp_T0",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 0, 1, 0], dtype=int64)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"heart.csv\")\n",
    "X = df[['sex', 'cp', 'fbs', 'restecg', 'exang', 'slope', 'ca', 'thal', 'age', 'trestbps', 'chol', 'thalach', 'oldpeak']].to_numpy()\n",
    "y = df.loc[:, \"target\"].to_numpy()\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1, -1, -1, ..., -1,  1, -1], dtype=int64)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for i in range(len(y)):\n",
    "    if y[i] == 0:\n",
    "        y[i] = -1\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "z0kQs1yip0Qm"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  1. ,   0. ,   0. , ..., 168. ,   1. ,  -1. ],\n",
       "       [  1. ,   0. ,   1. , ..., 155. ,   3.1,  -1. ],\n",
       "       [  1. ,   0. ,   0. , ..., 125. ,   2.6,  -1. ],\n",
       "       ...,\n",
       "       [  1. ,   0. ,   0. , ..., 118. ,   1. ,  -1. ],\n",
       "       [  0. ,   0. ,   0. , ..., 159. ,   0. ,  -1. ],\n",
       "       [  1. ,   0. ,   0. , ..., 113. ,   1.4,  -1. ]])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TODO: Add a bias term -1 into the dataset. This is needed for the SGD to work.\n",
    "bias_column = np.full((X.shape[0], 1), -1)\n",
    "X = np.concatenate((X, bias_column), axis=1)\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "yz313F2ep3St"
   },
   "outputs": [],
   "source": [
    "# TODO: Split the data to train and test with a 0.2 ratio\n",
    "train_count = int(X.shape[0] * 0.8)\n",
    "X_train, X_val, y_train, y_val = X[:train_count], X[train_count:], y[:train_count], y[train_count:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "vza_Q_Gim_cf"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "w = perceptron_sgd_plot(X_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GoPThTyLnd0N"
   },
   "source": [
    "### Accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ARKTa6f2n4v0"
   },
   "source": [
    "\n",
    "Implement the `accuracy_score` function by considering the predictions and true labels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "lkHOtSEYnePG"
   },
   "outputs": [],
   "source": [
    "def accuracy_score(y_true, y_pred):\n",
    "    corrects = np.sum(y_true == y_pred)\n",
    "    return corrects / len(y_true)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9sE-fMyLn9b9"
   },
   "source": [
    "Get the accuracy of your model using the function you implemented."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "DEvjAMiUn9t_"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "74.14634146341463"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = np.sign(np.dot(X_val, w))\n",
    "accuracy_score(y_val, predictions) * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "54V2QMyQoAQr"
   },
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "f5cheWvWnjVR"
   },
   "source": [
    "Implement the `confusion_matrix` function to generate the confusion matrix by receiving the predicted labels, the real labels and the labels themselves."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "Kv6PgYmgoGtJ"
   },
   "outputs": [],
   "source": [
    "def confusion_matrix(real, predicted):\n",
    "    matrix = np.array([[0, 0], [0, 0]])\n",
    "    count1 = 0\n",
    "    count2 = 0\n",
    "    count3 = 0\n",
    "    count4 = 0\n",
    "    for i in range(len(real)):\n",
    "        if real[i] == 1 and predicted[i] == 1:\n",
    "            matrix[1][1] += 1\n",
    "            count1 += 1\n",
    "        elif real[i] == -1 and predicted[i] == -1:\n",
    "            matrix[0][0] += 1\n",
    "            count4 += 1\n",
    "        elif real[i] == 1 and predicted[i] == -1:\n",
    "            matrix[1][0] += 1\n",
    "            count2 += 1\n",
    "        elif real[i] == -1 and predicted[i] == 1:\n",
    "            matrix[0][1] += 1\n",
    "            count3 += 1\n",
    "\n",
    "    print(\"TP: \", count1)\n",
    "    print(\"FN: \", count2)\n",
    "    print(\"FP: \", count3)\n",
    "    print(\"TN: \", count4)\n",
    "    return matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JhtEA4xRoKMx"
   },
   "source": [
    "Implement the following function to plot the confusion matrix obtained from the previous part."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "id": "1zBnY2ALoNY5"
   },
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(matrix, labels):\n",
    "    plt.figure(figsize=(12, 10))\n",
    "    sns.heatmap(matrix, annot=True, fmt='d', cmap='Greens', xticklabels=labels, yticklabels=labels)\n",
    "    plt.xlabel('Predicted', color=\"lightBlue\")\n",
    "    plt.ylabel('Real', color=\"lightBlue\")\n",
    "    plt.title('Confusion Matrix', color=\"lightBlue\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CQhj0HS8odjN"
   },
   "source": [
    "Now plot the confusion matrix for your model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "id": "KNswewkkodPS"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TP:  70\n",
      "FN:  29\n",
      "FP:  24\n",
      "TN:  82\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = [-1, 1]\n",
    "cm = confusion_matrix(y_val, predictions)\n",
    "plot_confusion_matrix(cm, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0u9hzezSo48b"
   },
   "source": [
    "### F1 Score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LV2lhF-xo7tG"
   },
   "source": [
    "Implement the following functions to calculate precision, recall and [F1 score](https://en.wikipedia.org/wiki/F-score) using confusion matrix. Then calculate all three metrics for your model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "id": "5H5x1seOo7-m"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7070707070707071"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def recall_score(cm):\n",
    "    return cm[1][1] / (cm[1][1] + cm[1][0])\n",
    "\n",
    "recall = recall_score(cm)\n",
    "recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "1-jnEGfXo-CO"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7446808510638298"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def precision_score(cm):\n",
    "    return cm[1][1] / (cm[1][1] + cm[0][1])\n",
    "\n",
    "precision = precision_score(cm)\n",
    "precision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "id": "elp_Y21uo_11"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7253886010362693"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def f1_score(cm):\n",
    "    recall = cm[1][1] / (cm[1][1] + cm[1][0])\n",
    "    precision = cm[1][1] / (cm[1][1] + cm[0][1])\n",
    "    return (2 * recall * precision)/(precision + recall)\n",
    "\n",
    "f1_score(cm)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GSMb1HLNL5BB"
   },
   "source": [
    "# Naive Bayes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "BRS6sygsLV-G"
   },
   "source": [
    "## Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xdyNTg7Is-Z1"
   },
   "source": [
    "To predict the target column we want to use a Naive Bayes classifier. In this section, you must implement the functions of the `MixedDataNB` class. Assume continuous features follow a normal distribution. **Hint**: You need to train a Naive Bayes model that implements the likelihood function for categorical and continuous values in two different ways. Search for Gaussian NB and Multinomial NB. **Note**: Please feel free to modify the following class and its functions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "id": "7EnbOik9KW9f"
   },
   "outputs": [],
   "source": [
    "class MixedDataNB:\n",
    "    # Help a little from gpt and https://github.com/The-Swapster/NaiveBayes\n",
    "    def __init__(self):\n",
    "        self.prior = None\n",
    "        self.countsss = None\n",
    "        self.category = {}\n",
    "        self.samples = None\n",
    "        self.features = None\n",
    "        self.mean_data = {}\n",
    "        self.var_data = {}\n",
    "\n",
    "    def fit(self, X, y):\n",
    "        self.samples = X.shape[0]\n",
    "        self.features = X.shape[1]\n",
    "        number_of_samps = X.shape[0]\n",
    "        number_of_feats = X.shape[1]\n",
    "\n",
    "        # self.countsss = np.zeros(n_classes)\n",
    "        # for i, c in enumerate(self.classes):\n",
    "        #     self.countsss[i] = np.sum(y == c) / n_samples\n",
    "\n",
    "        self.types_of_cla = np.unique(y)\n",
    "        nums_cla = len(self.types_of_cla)\n",
    "        self.prior = np.zeros(nums_cla)\n",
    "        self.countsss = np.zeros(nums_cla)\n",
    "        alll = np.all(np.array(X, dtype=float) == X, axis=0)\n",
    "        cont_features = np.argmax(alll)\n",
    "\n",
    "        for i, classessse in enumerate(self.types_of_cla):\n",
    "            self.samples = X.shape[0]\n",
    "            self.features = X.shape[1]\n",
    "            X_help = X[y == classessse]\n",
    "            self.prior[i] = len(X_help) / number_of_samps\n",
    "            self.countsss[i] = len(X_help)\n",
    "\n",
    "            continuous_features = X_help[:, :cont_features]\n",
    "\n",
    "            self.category[classessse] = {}\n",
    "            categorical_features = X_help[:, cont_features:]\n",
    "            for feat in range(cont_features, number_of_feats):\n",
    "                feature_values, counts = np.unique(categorical_features[:, feat - cont_features],\n",
    "                                                   return_counts=True)\n",
    "                self.category[classessse][feat] = dict(zip(feature_values, counts / len(X_help)))\n",
    "\n",
    "            # for i in range(len(X)):\n",
    "            #     if y[i] == 1:\n",
    "            #         for j in range(self.numberOfFeature):\n",
    "            #             self.var_1[j] += ((X[i][j + 8] - self.miu_1[j]) ** 2) / self.count_1\n",
    "            self.mean_data[classessse] = np.mean(continuous_features, axis=0)\n",
    "            self.var_data[classessse] = np.var(continuous_features, axis=0)\n",
    "\n",
    "    def predict(self, X):\n",
    "        self.samples = X.shape[0]\n",
    "        self.features = X.shape[1]\n",
    "        n_samples, n_features = X.shape\n",
    "        utilss = np.zeros((n_samples, len(self.types_of_cla)))\n",
    "\n",
    "        for i, cla in enumerate(self.types_of_cla):\n",
    "            cont = len(self.mean_data[cla])\n",
    "            likelihood1 = np.zeros(X.shape[0])\n",
    "            #     likelihood1 = np.sum(np.log(self._pdf(i, j, x[j])) for j in range(len(x)))\n",
    "            for i in range(cont):\n",
    "                likelihood1 += -0.5 * np.log(2 * np.pi * self.var_data[cla][i]) - 0.5 * (\n",
    "                        (X[:, i] - self.mean_data[cla][i]) ** 2 / self.var_data[cla][i])\n",
    "\n",
    "            cont = len(self.mean_data[cla])\n",
    "            n_features = X.shape[1]\n",
    "            likelihood2 = np.zeros(X.shape[0])\n",
    "            for j in range(cont, n_features):\n",
    "                likelihood2 += np.log([self.category[cla][j].get(x, 1e-6) for x in X[:, j]])\n",
    "            # for i, c in enumerate(self.classes):\n",
    "            #     prior = np.log(self.class_probabilities[i])\n",
    "            #     likelihood = np.sum(np.log(self._pdf(i, j, x[j])) for j in range(len(x)))\n",
    "            #     posterior = prior + likelihood\n",
    "            #     posteriors.append(posterior)\n",
    "            utilss[:, i] = np.log(self.prior[i]) + likelihood1 + likelihood2\n",
    "\n",
    "        return self.types_of_cla[np.argmax(utilss, axis=1)]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RCcNBC_AOddW"
   },
   "source": [
    "## Train and Validation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "6KjeHMceOJTE"
   },
   "source": [
    "\n",
    "### Data splitting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0ljWeBvFv65G"
   },
   "source": [
    "Split the training and validation data. Separate 20% of the data for validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "do-NG_s2bvql"
   },
   "outputs": [],
   "source": [
    "#np.random.shuffle(X)\n",
    "X = df[['sex', 'cp', 'fbs', 'restecg', 'exang', 'slope', 'ca', 'thal', 'age', 'trestbps', 'chol', 'thalach', 'oldpeak']].to_numpy()\n",
    "y = df.loc[:, \"target\"].to_numpy()\n",
    "y = np.array([1 if i == 1 else -1 for i in y])\n",
    "train_count = int(X.shape[0] * 0.8)\n",
    "X_train, X_val, y_train, y_val = X[:train_count], X[train_count:], y[:train_count], y[train_count:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7HbVGE0BwFoZ"
   },
   "source": [
    "### Model training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "id": "3uHjoYgQb5ed"
   },
   "outputs": [],
   "source": [
    "model = MixedDataNB()\n",
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vfHEirlXwyeb"
   },
   "source": [
    "### Make predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "id": "PuN_GWEhwwUN"
   },
   "outputs": [],
   "source": [
    "predictions = model.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AiVuMqFNw2rS"
   },
   "source": [
    "### Accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zOsIA_F9xf3u"
   },
   "source": [
    "Get the accuracy of your model using the function you implemented."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "id": "Q5auaGaCiZ2I"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8878048780487805\n"
     ]
    }
   ],
   "source": [
    "print(accuracy_score(y_val, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0BmtJ-iqQhN_"
   },
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "lvj1QkeP3XPk"
   },
   "source": [
    "Now, using the functions you wrote, plot the confusion matrix for the model you trained."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "id": "-nN1RNVopf-_"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TP:  88\n",
      "FN:  11\n",
      "FP:  12\n",
      "TN:  94\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApUAAAJcCAYAAACotl/bAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAAsTAAALEwEAmpwYAAAmHElEQVR4nO3de7RkVX0u7Hc2qFyUawiWoJFELKIYiBIOiBrkYsQQwYoxRo+n9aB9NAQ06hBUjpfzmQwZn9EYPzQ2IBJvgWAhJFGEcFQ0GhQJQRQLFIQABSgBVEAFnN8fVZims6t3Q3XPvXft5+mxxq5aa9VcczVjbH79zjlXlVprAABgGisWugMAACx9ikoAAKamqAQAYGqKSgAApqaoBABgaopKAACmpqgENor+YLh5fzD8+/5geHt/MPy7Kdp5cX8wPHdD9m0h9AfDz/QHw5UL3Q+AjaV4TiUsb/3B8EVJXptktyQ/SnJJkj/rdTtfmrLdlyQ5KslTe93OPdP2c0PrD4b7J/lckk/1up3nrbF/j4z+Dr7Q63b2X4923pbkcb1u579vlI4CLBGSSljG+oPha5P8ZZI/T7JjksckeX+SwzZA87+S5IrFWFCu4ftJ9u0PhtuvsW9lkis21AX6g2HpD4Z+1wIzT1IJy1R/MNw6yfVJXtbrduYcnu4Phg9LcnySF4x3nZ7kmF6389Nx0vfRJO9JckySe5O8qdftnNIfDN+e5I1JSpKfJnl1kkdnjUSvPxg+NsnVSR7S63bu6Q+GL03yliQ7JPlBkuN63c7Hxvtf3ut2njb+3FOTvDfJ4zMq/l7d63a+PD72+SRfTHJAkt9I8pUkL+p1Oz+Y497u6/8/JPlGr9s5oT8YbpLk2iQfTHLAfUllfzB8b5Jekq2TXJnkNb1u54v9wfDZSc5e4z6/2+t29hj345+T7J/kyUmelOSkJB/tdTsn9QfDDyT55V638/vj9o9PsleSg3rdjl/KwJLkX8+wfO2bZLMkZ67jnDcn2SfJnkn2SLJ3kuPWOP7IjAqtnZIckeSE/mC4ba/beWtG6edpvW7n4b1u5+R1daQ/GG6Z5K+SHNLrdh6R5KkZDUGvfd52Sf5xfO72Sd6d5B/XShpflORlSX45yUOTvH5d107yN0n+x/j17yS5LMkNa53ztYz+DrZL8vEkf9cfDDfrdTvnrHWfe6zxmZckWZXkEUmuWau91yV5Un8wfGl/MHx6Rn93KxWUwFK26UJ3AFgw2yf5wTzD0y9OclSv27k5ScYJ5AeT/O/x8buT/J9xG5/uD4Y/TtJN8i8Poj8/T7J7fzC8ttftDJMM5zjnd5Nc2et2PjJ+/4n+YHh0kt9L8uHxvlN63c4V4/6enuS567por9v5cn8w3K4/GHYzKi7/Jsnma53z0TXe/kV/MDwuo/v8t3U0/eFet/PN+970B/95O71u587xnNPPZDSP9ahet3PduvoJsNhJKmH5uiXJL/UHw3X94/JRuX/Kds143y/aWKsovTPJwx9oR3rdzh1J/jDJK5MM+4PhP/YHw93Woz/39WmnNd7f+CD685Ekf5LkmZkjue0Phq/vD4aXj1ey35ZROvtL87T57+s62Ot2LkxyVUZD56evRx8BFjVFJSxfX8loHuDh6zjnhowW3NznMfmvQ8Pr644kW6zx/pFrHux1O5/tdTsHJ+kk+XaSE9ejP/f16foH2af7fCTJHyf5dK/buXPNA+Ph6TdkNK902163s02S2zMqBpNk0pD1Ooey+4PhkUkeltE9veFB9xxgkTD8DctUr9u5vT8YviWjeZD3JDk3o+Hsg5I8s9ftvCHJJ5Ic1x8Mv5ZRkfSWjBa3PBiXJDmmPxg+JqOi7I33HegPhjtmNHfzn5LcleTHGQ2Hr+3TSd43fgzS6Ul+P8kTMlps86D1up2r+4Phb2eUHK7tEUnuyWil+Kb9wfDYJFutcfymJAf3B8MVvW5nrj7/F/3B8PFJ3pHRQp47k3y1Pxh+ptftXPLg7wJgYUkqYRnrdTt/kdEzKo/LqGj694yGgT81PuUdSS5KcmmSbyS5eLzvwVzrvCSnjdv6eu5fCK4Y9+OGJP+R5LeTvGqONm5JcmhGC11uySjhO3Su1d0Pon9f6nU7c6Wwn01yTkYrza9J8pPcf2j7vpXzt/QHw4vnu854usFHkxzf63b+rdftXJnkTUk+Ml5tD7AkeaQQAABTk1QCADA1RSUAAFNTVAIAMDVFJQAAU1u0jxQqB+9sBRGwXu74zLcXugvAErHFpg8v85+1cbWucep51zW5Z0klAABTW7RJJQDATCoLHpZuFJJKAACmJqkEAGhpRiO9Gb0tAABaUlQCADA1w98AAC1ZqAMAAHOTVAIAtDSbQaWkEgCA6UkqAQBaMqcSAADmJqkEAGhpRiO9Gb0tAABaklQCALRkTiUAAMxNUgkA0NJsBpWSSgAApiepBABoacVsRpWSSgAApiapBABoaTaDSkklAADTU1QCACxjpZRXl1IuK6V8s5TymvG+7Uop55VSrhz/3Ha+dhSVAAAtldJ2W2dXyu5JXpFk7yR7JDm0lPK4JMcmOb/WumuS88fv10lRCQCwfP16kgtrrXfWWu9J8oUkvSSHJTl1fM6pSQ6fryFFJQBAS6XtVkpZVUq5aI1t1Rq9uSzJ00sp25dStkjynCSPTrJjrXU4PufGJDvOd1tWfwMAzLBa6+okqyccu7yUcnySc5PckeSSJPeudU4tpdT5riOpBABoaUVpu82j1npyrfUptdZnJLk1yRVJbiqldJJk/PPmeW9ryr8WAACWsFLKL49/Piaj+ZQfT3J2kpXjU1YmOWu+dgx/AwC0tPgefv7JUsr2Se5OcmSt9bZSyjuTnF5KOSLJNUleMF8jikoAgGWs1vr0OfbdkuTAB9KOohIAoKV5nh25VJlTCQDA1CSVAAAtrceK7KVIUgkAwNQklQAALc1mUCmpBABgepJKAICWrP4GAIC5KSoBAJia4W8AgJZmc/RbUgkAwPQklQAALXn4OQAAzE1SCQDQ0mwGlZJKAACmJ6kEAGjJw88BAGBukkoAgJZmNNKb0dsCAKAlSSUAQEvmVAIAwNwklQAALc1mUCmpBABgepJKAICWzKkEAIC5KSoBAJia4W8AgJZmNNKb0dsCAKAlSSUAQEsW6gAAwNwklQAALc1mUCmpBABgepJKAICWVsxmVCmpBABgapJKAICWrP4GAIC5SSoBAFqazaBSUgkAwPQklQAADRVzKgEAYG6SSgCAhiSVAAAwgaISAICpGf4GAGhoRke/JZUAAExPUgkA0NCKGY0qJZUAAExNUgkA0JBHCgEAwASSSgCAhiSVAAAwgaQSAKAhSSUAAEwgqQQAaGhGg0pJJQAA05NUAgA0ZE4lAABMoKgEAGiolNJ0W4/+/Gkp5ZullMtKKZ8opWxWStmllHJhKeU7pZTTSikPna8dRSUAwDJVStkpydFJ9qq17p5kkyQvTHJ8kvfUWh+X5NYkR8zXlqISAGB52zTJ5qWUTZNskWSY5IAkZ4yPn5rk8PkaUVQCADRUWv8pZVUp5aI1tlX39aXWen2SdyW5NqNi8vYkX09yW631nvFp1yXZab77svobAGCG1VpXJ1k917FSyrZJDkuyS5Lbkvxdkmc/mOsoKgEAGlpkjxQ6KMnVtdbvJ0kppZ9kvyTblFI2HaeVOye5fr6GDH8DACxf1ybZp5SyRRlVuwcm+VaSzyV5/viclUnOmq8hRSUAQEOltN3WpdZ6YUYLci5O8o2MasPVSY5J8tpSyneSbJ/k5Pnuy/A3AMAyVmt9a5K3rrX7qiR7P5B2FJUAAA2tWFxzKjcYw98AAExNUgkA0NAiW/29wUgqAQCYmqQSAKAhSSUAAEwgqQQAaGhGg0pJJQAA05NUAgA0ZE4lAABMIKkEAGhIUgkAABMoKgEAmJrhbwCAhgx/AwDABJJKAICGJJUAADCBpBIAoKEZDSollQAATE9SCQDQkDmVAAAwgaQSAKAhSSUAAEwgqQQAaGiFpBIAAOYmqQQAaGhGg0pJJQAA05NUAgA0ZPU3AABMoKgEAGBqhr8BABoqMfwNAABzklQCADRkoQ4AAEygqGTROvp5R+Qbq/8pl514fl79vCPud+y1z1+Vet512X6rbReod8Bi8rbj3p4Dnn5Qnn/YC36x7z3v+ss879BeXvC8P8xrj35dfvTDHy1gD+E/lVKabq0oKlmUnvjYbl5xyB9l76MOzR7/61k5dJ+D8muPemySZOcdOnnWU56Ra266bmE7CSwav3f47+WED77vfvv22fe/5e8+dXpOP/O0/Mqv/Eo+dOIpC9Q7WB6aF5WllIe3viZLz68/5nG58NuX5K6f/iT3/vzefOHSf0nvaYckSd7zyrflDSf+WWqtC9xLYLF4yl5PztZbb32/ffvut2823XS0dOBJe+yem266aSG6Bv9FKW23VhYiqfzWAlyTJeay7w3y9Cftne0esU02f9hmec7eB+TROzwqz933Wbn+lhtz6VWXL3QXgSXkrP7Z2e/p+y10N2CmbZTV36WU1046lGRiUllKWZVkVZJkt22Snbfc4H1jafj2td/J8ae9P+e+8+O54yd35pLvfjMPe8hD86Y/OirPOvZFC909YAk56YMnZ5NNN8lzDj1kobsCSaz+fqD+PMm2SR6x1vbwdV2z1rq61rpXrXUvBSUfOudvs9eRz8lvv+75ufXHt+eb11yRXR756PzbB8/N1R/5SnbeoZOLP3BOdtx2h4XuKrBInX3m2bngC1/Mnx3/jpn9HzksFhvrOZUXJ/lUrfXrax8opbx8I12TGbPDNtvn+7fdkkfv8Kj09jsk+xz93PzVmSf/4vjVH/lK9jryObnlh7cuYC+Bxeqfv/jlfPhDf5OTTj0xm2+++UJ3B35hVv+Bs7GKypcluWXNHaWUR9Zab0yy10a6JjPmk29Zne232jZ333NPjvz/3pzb7/jhQncJWKSOff2b8vWvXZTbbrstv3PAIXnlkf8rp5x4Sn5299151cv/OEnypD2elOPe+qYF7inMrtJqBW0p5eJa65PX+/yDd7a0F1gvd3zm2wvdBWCJ2GLThy94TPj4dz+7aY1zxWvPaXLPLVd/L/h/RAAANo6W3/19YsNrAQAsSjM6pbJdUllrfX+rawEA0JavaQQAYGoth78BAJa9WX2kkKQSAICpSSoBABqSVAIAwASSSgCAhiSVAAAwgaQSAKChGQ0qJZUAAExPUgkA0JA5lQAAzJRSSreUcska2w9LKa8ppWxXSjmvlHLl+Oe287WlqAQAaKiU0nRbl1rroNa6Z611zyRPSXJnkjOTHJvk/FrrrknOH79fJ0UlAABJcmCS79Zar0lyWJJTx/tPTXL4fB82pxIAoKHWcypLKauSrFpj1+pa6+o5Tn1hkk+MX+9Yax2OX9+YZMf5rqOoBACYYeMCcq4i8hdKKQ9N8twkb5zj87WUUue7jqISAKChRbr4+5AkF9dabxq/v6mU0qm1DkspnSQ3z9eAOZUAAPxR/nPoO0nOTrJy/HplkrPma0BRCQCwjJVStkxycJL+GrvfmeTgUsqVSQ4av18nw98AAA0ttoef11rvSLL9WvtuyWg1+HqTVAIAMDVJJQBAS4ssqdxQJJUAAExNUgkA0NBim1O5oUgqAQCYmqQSAKChGQ0qJZUAAExPUgkA0JA5lQAAMIGkEgCgIUklAABMIKkEAGhIUgkAABNIKgEAGprRoFJSCQDA9BSVAABMzfA3AEBDFuoAAMAEkkoAgIYklQAAMIGkEgCgIUklAABMIKkEAGhIUgkAABNIKgEAGprRoFJSCQDA9CSVAAANmVMJAAATSCoBABqSVAIAwASSSgCAhiSVAAAwgaISAICpGf4GAGhoRke/JZUAAExPUgkA0JCFOgAAMIGkEgCgJUklAADMTVIJANCQOZUAADCBpBIAoKEVsxlUSioBAJiepBIAoCFzKgEAYAJJJQBAQysklQAAMDdJJQBAQ+ZUAgDABIpKAACmZvgbAKChWU30ZvW+AABoSFIJANCQRwoBAMAEkkoAgIY8UggAACaQVAIANGROJQAAM6eUsk0p5YxSyrdLKZeXUvYtpWxXSjmvlHLl+Oe287WjqAQAaKiU0nRbD+9Nck6tdbckeyS5PMmxSc6vte6a5Pzx+3VSVAIALFOllK2TPCPJyUlSa/1ZrfW2JIclOXV82qlJDp+vLUUlAEBDKxpvpZRVpZSL1thWrdGdXZJ8P8kppZR/LaWcVErZMsmOtdbh+Jwbk+w4331ZqAMAMMNqrauTrJ5weNMkT05yVK31wlLKe7PWUHettZZS6nzXUVQCADS0yFZ/X5fkulrrheP3Z2RUVN5USunUWoellE6Sm+dryPA3AMAyVWu9Mcm/l1K6410HJvlWkrOTrBzvW5nkrPnaklQCADS0CL9R56gkHyulPDTJVUlellHweHop5Ygk1yR5wXyNKCoBAJaxWuslSfaa49CBD6Qdw98AAExNUgkA0NAiW6izwUgqAQCYmqQSAKCh2cwpJZUAAGwAkkoAgIbMqQQAgAkklQAADUkqAQBgAkklAEBDi/BrGjcISSUAAFOTVAIANGROJQAATCCpBABoaDZzSkklAAAbgKQSAKAhcyoBAGACRSUAAFMz/A0A0JDhbwAAmEBSCQDQkK9pBACACSSVAAANmVMJAAATSCoBABqazZxSUgkAwAYgqQQAaMicSgAAmEBSCQDQkKQSAAAmkFQCADTkG3UAAGACSSUAQEOzmujN6n0BANCQohIAgKkZ/gYAaGhWF+qss6jsD4bvS1InHe91O0dv8B4BALDkzJdUXtSkFwAAy8SsPvx8nUVlr9s5tVVHAABYutZrTmV/MNwhyTFJnpBks/v297qdAzZSvwAAZtKsJpXru/r7Y0kuT7JLkrcn+V6Sr22kPgEAsMSsb1G5fa/bOTnJ3b1u5wu9bud/JpFSAgA8QKWUplsr6/tIobvHP4f9wfB3k9yQZLuN0yUAAJaa9S0q39EfDLdO8rok70uyVZI/3Wi9SnLXOVdszOaBGbL5yj0XugvAElE/uvD1xYrM5pzK9Soqe93OP4xf3p7kmRuvOwAALEXru/r78Uk+kGTHXreze38w/I0kz+11O+/YqL0DAJgxs/qNOuu7UOfEJG/MeG5lr9u5NMkLN1anAABYWta3qNyi1+18da1992zozgAAzLoVpTTdmt3Xep73g/5g+GsZfw94fzB8fpLhRusVAABLyvqu/j4yyeoku/UHw+uTXJ3kxRutVwAAM6os89XfVyU5qD8YbplRunlnRnMqr9mIfQMAYIlYZ1HZHwy3yiil3CnJWUn+afz+dUkuzejrGwEAWE+zuvp7vqTyI0luTfKVJK9I8uYkJcnzet3OJRu3awAALBXzFZW/2ut2npQk/cHwpIwW5zym1+38ZKP3DACAJWO+1d/3fed3et3OvUmuU1ACADx4s/pIofmSyj36g+EPx69Lks3H70uS2ut2ttqovQMAYElYZ1HZ63Y2adURAIDloKz3Y8LbKKV8L8mPktyb5J5a616llO2SnJbksUm+l+QFtdZb19XO4rorAAAWwjNrrXvWWvcavz82yfm11l2TnD9+v07r+/BzAAA2gJbzHKdwWJL9x69PTfL5JMes6wOSSgCAGVZKWVVKuWiNbdVap9Qk55ZSvr7GsR1rrfd9JfeNSXac7zqSSgCAhlo//LzWujqjr9ue5Gm11utLKb+c5LxSyrfX+nwtpdT5riOpBABYxmqt149/3pzkzCR7J7mplNJJkvHPm+drR1EJANBQafxnnX0pZctSyiPue53kWUkuS3J2kpXj01Zm9HXd62T4GwBg+doxyZnjIflNk3y81npOKeVrSU4vpRyR5JokL5ivIUUlAEBDi2n1d631qiR7zLH/liQHPpC2DH8DADA1SSUAQEOtV3+3IqkEAGBqkkoAgIZWzGimN5t3BQBAU4pKAACmZvgbAKAhC3UAAGACSSUAQEOSSgAAmEBSCQDQ0IpIKgEAYE6SSgCAhsypBACACSSVAAANrZBUAgDA3CSVAAANFau/AQBgbpJKAICGVpTZzPRm864AAGhKUgkA0JDnVAIAwASKSgAApmb4GwCgIY8UAgCACSSVAAAN+ZpGAACYQFIJANCQOZUAADCBpBIAoCFzKgEAYAJJJQBAQ6XMZqY3m3cFAEBTkkoAgIas/gYAgAkklQAADVn9DQAAE0gqAQAaKpJKAACYm6ISAICpGf4GAGhohUcKAQDA3CSVAAANWagDAAATSCoBABoqZTYzvdm8KwAAmpJUAgA0ZPU3AABMIKkEAGjI6m8AAJhAUgkA0FAxpxIAAOYmqQQAaMicSgAAmEBSCQDQkOdUAgDABIpKAACmZvgbAKChUhZfpldK2STJRUmur7UeWkrZJcnfJtk+ydeTvKTW+rN1tbH47goAgNZeneTyNd4fn+Q9tdbHJbk1yRHzNaCoBABoqDT+M29/Stk5ye8mOWn8viQ5IMkZ41NOTXL4fO0oKgEAZlgpZVUp5aI1tlVrnfKXSd6Q5Ofj99snua3Wes/4/XVJdprvOuZUAgA01Prh57XW1UlWT+jLoUlurrV+vZSy/zTXUVQCACxf+yV5binlOUk2S7JVkvcm2aaUsuk4rdw5yfXzNWT4GwCgocU0p7LW+sZa68611scmeWGS/1trfXGSzyV5/vi0lUnOmu++FJUAAKztmCSvLaV8J6M5lifP9wHD3wAADbWeU7m+aq2fT/L58eurkuz9QD4vqQQAYGqSSgCAhlasx7MjlyJJJQAAU5NUAgA0tFjnVE5LUgkAwNQklQAADZUZzfRm864AAGhKUQkAwNQMfwMANGShDgAATCCpBABoqHj4OQAAzE1SCQDQ0ApzKgEAYG6SSgCAhsypBACACSSVAAANeU4lAABMIKkEAGiozGimN5t3BQBAU5JKAICGzKkEAIAJJJUAAA2t8JxKAACYm6ISAICpGf4GAGjIQh0AAJhAUgkA0FCxUAcAAOYmqQQAaMicSgAAmEBSCQDQUJnRTG827woAgKYklQAADa0wpxIAAOYmqQQAaMhzKgEAYAJJJQBAQ55TCQAAE0gqAQAaMqcSAAAmUFQCADA1w98AAA1ZqAMAABNIKlm03vLmt+WCL1yQ7bbbLv2zz0iSnHvOefnACX+dq6+6Oh877SN54u5PXOBeAovBa5790rx8/z9IrTXfuO6KvGz1sdnv8U/J//tHb8iKsiI//smdeenqY/Ldm65d6K5CVsxopjebd8VMOOx5v5cPrD7hfvset+uv5T1/9Rd5yl5PXqBeAYvNo7bdMUc/6yXZ63/38qQ3HppNVqzIC/f53XzgpW/Li9//+vzmmw/Lx7/y9znusD9e6K7CTJNUsmg9Za+n5Prrb7jfvl/9tV9doN4Ai9mmm2yazR+6We6+955s8dDNc8OtN6emZqvNt0ySbL35I3LDbTcvcC9hZFbnVCoqAVjSbrj1przr0yfn2vd+Pnf97Kc59xtfynmX/XNeftJx+fTrT8xdd/80P7zrx9nnbX+w0F2FmdZ8+LuU8rJ1HFtVSrmolHLRySd+qGW3AFiittliqxz25AOzy58ekEcd9bRs+bAt8uL9nps/ffZL85x3vSKPPvoZOeWCT+bdL37TQncVkoweft7yTysLkVS+Pckpcx2ota5OsjpJfnLvnbVlpwBYmg7a/am5+vvX5Qc/ujVJ0r/o3Oy365Ozx2N2y1e/e2mS5LR/+XTOecPJC9lNmHkbpagspVw66VCSHTfGNQFYnq695Ybs87g9s/lDN8tdP/tJDnzivrnoqsvyB//tkOz6yMfmyhu/l4N33y+XX//dhe4qJDGn8oHaMcnvJLl1rf0lyZc30jWZMce8/thc9NWv57bbbsvBz/ydvOpPXpmtt9467/yz43Prf9yaP3nV0enu1s1fn/j+he4qsIC++t1Lc8ZXP5uL3/Gp3HPvPfnXay7P6s/9ba679cZ88tXvy89/XnPrnbfnf642/A0bU6l1w48yl1JOTnJKrfVLcxz7eK31RfO1YfgbWF+br9xzobsALBH1o1cseEz4te9/qWmN81s7PK3JPW+UpLLWesQ6js1bUAIAsLR4pBAAQEMtV2S35Bt1AACYmqISAKClUtpu6+xK2ayU8tVSyr+VUr5ZSnn7eP8upZQLSynfKaWcVkp56Hy3pagEAFi+fprkgFrrHkn2TPLsUso+SY5P8p5a6+MyeprPxPUy91FUAgAsU3Xkx+O3DxlvNckBSc4Y7z81yeHztaWoBABoqPXXNK75NdjjbdX9+lPKJqWUS5LcnOS8JN9Nclut9Z7xKdcl2Wm++7L6GwBghq35NdgTjt+bZM9SyjZJzkyy24O5jqISAKChxfo1jbXW20opn0uyb5JtSimbjtPKnZNcP9/nDX8DACxTpZQdxgllSimbJzk4yeVJPpfk+ePTViY5a762JJUAAA0tsoefd5KcWkrZJKOw8fRa6z+UUr6V5G9LKe9I8q9JTp6vIUUlAMAyVWu9NMlvzrH/qiR7P5C2FJUAAA0tsqRygzGnEgCAqUkqAQAaWqyrv6clqQQAYGqSSgCAhsypBACACSSVAAANSSoBAGACSSUAQENWfwMAwASKSgAApmb4GwCgIQt1AABgAkklAEBDFuoAAMAEkkoAgIbMqQQAgAkklQAADUkqAQBgAkklAEBDVn8DAMAEkkoAgIbMqQQAgAkklQAADUkqAQBgAkklAEBDVn8DAMAEkkoAgKYklQAAMCdFJQAAUzP8DQDQkIU6AAAwgaQSAKAhDz8HAIAJJJUAAA1JKgEAYAJJJQBAQ1Z/AwDABJJKAICGzKkEAIAJJJUAAA1JKgEAYAJJJQBAQ1Z/AwDABJJKAICGzKkEAIAJFJUAAEzN8DcAQEMW6gAAwASSSgCAhizUAQCACSSVAABNSSoBAGBOkkoAgIZmM6eUVAIAsAFIKgEAGvKcSgAAmEBRCQDQVGm8raMnpTy6lPK5Usq3SinfLKW8erx/u1LKeaWUK8c/t53vrhSVAADL1z1JXldrfUKSfZIcWUp5QpJjk5xfa901yfnj9+ukqAQAaGjx5JRJrXVYa714/PpHSS5PslOSw5KcOj7t1CSHz3dfikoAgBlWSllVSrlojW3VhPMem+Q3k1yYZMda63B86MYkO853Hau/AQCaarv6u9a6OsnqdZ1TSnl4kk8meU2t9YdrrlCvtdZSSp3vOpJKAIBlrJTykIwKyo/VWvvj3TeVUjrj450kN8/XjqISAGCZKqNI8uQkl9da373GobOTrBy/XpnkrPnaMvwNANDQInv4+X5JXpLkG6WUS8b73pTknUlOL6UckeSaJC+YryFFJQDAMlVr/VImT/I88IG0ZfgbAICpKSoBAJia4W8AgIZK40cKtSKpBABgapJKAICGJJUAADCBohIAgKkpKgEAmJo5lQAADS2yb9TZYCSVAABMTVEJAMDUFJUAAEzNnEoAgIY8pxIAACZQVAIAMDXD3wAATRn+BgCAOUkqAQAams2cUlIJAMAGIKkEAGjI1zQCAMAEkkoAgKYklQAAMCdJJQBAQ7OZU0oqAQDYACSVAABNzWZWKakEAGBqkkoAgIY8pxIAACZQVAIAMDVFJQAAU1NUAgAwNQt1AAAaKh4pBAAAc5NUAgA0JakEAIA5SSoBABqazZxSUgkAwAYgqQQAaMjXNAIAwASSSgCApiSVAAAwJ0klAEBDs5lTSioBANgAJJUAAE3NZlYpqQQAYGqSSgCAhjynEgAAJlBUAgAwNUUlAABTU1QCADA1C3UAABoqHikEAABzK7XWhe4DrLdSyqpa6+qF7gew+Pl9AW1JKllqVi10B4Alw+8LaEhRCQDA1BSVAABMTVHJUmN+FLC+/L6AhizUAQBgapJKAACmpqgEAGBqikqWjFLKbqWUr5RSflpKef1C9wdYnEopHyql3FxKuWyh+wLLiaKSpeQ/khyd5F0L3RFgUftwkmcvdCdguVFUsmTUWm+utX4tyd0L3Rdg8aq1XpDRP0KBhhSVAABMTVEJAMDUFJUsaqWUI0spl4y3Ry10fwCAuW260B2Adam1npDkhIXuBwCwbr5RhyWjlPLIJBcl2SrJz5P8OMkTaq0/XNCOAYtKKeUTSfZP8ktJbkry1lrryQvaKVgGFJUAAEzNnEoAAKamqAQAYGqKSgAApqaoBABgaopKAACm5jmVQDP9wfDeJN/I6HfP5UlW9rqdOx9kWx9O8g+9bueM/mB4UpJ397qdb004d/8kP+t1O19+gNf4XpK9et3ODx5MHwGWE0kl0NJdvW5nz163s3uSnyV55ZoH+4Phg/qHbq/befmkgnJs/yRPfTBtA7B+JJXAQvlikt8Yp4j/T5Jbk+zWHwx/Pck7MyoEH5bkhF6388H+YFiSvC/JwUn+PaOiNEnSHww/n+T1vW7nov5g+Owkf55kkyQ/SHJERsXrvf3B8L8nOSrJt5P8dZLHjJt4Ta/b+ef+YLh9kk8k2SnJV5KUjXb3ADNGUgk0N04kD8loKDxJnpzk1b1u5/EZFYG397qd30ryW0le0R8Md0nyvCTdJE9I8j8yR/LYHwx3SHJikt/vdTt7JPmDXrfzvYwKyPeMU9IvJnnv+P1vJfn9JCeNm3hrki/1up0nJjkz/1l0AjAPSSXQ0ub9wfCS8esvJjk5o+Lwq71u5+rx/mdllGA+f/x+6yS7JnlGkk/0up17k9zQHwz/7xzt75Pkgvva6nU7/zGhHwcleUJ/MLzv/Vb9wfDh42v0xp/9x/5geOuDu02A5UdRCbR0V6/b2XPNHePC7o41dpUkR/W6nc+udd5zNmA/ViTZp9ft/GSOvgDwIBj+BhabzyZ5VX8wfEiS9AfDx/cHwy2TXJDkD/uD4Sb9wbCT5JlzfPZfkjxjPFye/mC43Xj/j5I8Yo3zzs1obmXG5+05fnlBkheN9x2SZNsNdVMAs05SCSw2JyV5bJKLx4tzvp/k8IzmOB6Q5FtJrs1oIc399Lqd7/cHw1VJ+v3BcEWSmzNa2PP3Sc7oD4aHZVRMHp3khP5geGlGvwcvyGgxz9uTfKI/GH4zyZfH1wFgPZRa60L3AQCAJc7wNwAAU1NUAgAwNUUlAABTU1QCADA1RSUAAFNTVAIAMDVFJQAAU/v/AaFocX3EtmTLAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_val, predictions)\n",
    "plot_confusion_matrix(cm, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3Snbk3FeQyph"
   },
   "source": [
    "### F1 Score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "EGMYWaiT3rqc"
   },
   "source": [
    "Using your implemented functions in the previous section, calculate precision, recall and F1 score using confusion matrix. Then calculate all three metrics for your model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "id": "ElFj7NblojAf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8888888888888888\n"
     ]
    }
   ],
   "source": [
    "print(recall_score(cm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "id": "MySaWgxwojsw"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.88\n"
     ]
    }
   ],
   "source": [
    "print(precision_score(cm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "id": "p_Sx4kv8ojz2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8844221105527638\n"
     ]
    }
   ],
   "source": [
    "print(f1_score(cm))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dEbb0kriPDu2"
   },
   "source": [
    "### Heatmap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "46_IPabaxvVL"
   },
   "source": [
    "By ploting the heatmap obtained for chol and oldpeak features from your model, compare the usefulness of these two features.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "id": "F9x7emkB6Uzn",
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqcAAAJXCAYAAACud/0dAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOydeXxU5dn+vw9ZWMKakIQEQVYREnbEti4oKFJARK3W2ld9rWjR2revtbW1/lxarbXW2rp00brW9q1VBK0KZbFWFEQJyKICKgQCIQkkspgEsszcvz9mkk7CZGZyknlyz+Hc/czHZObc53ud55xJD+c857qMiOCVV1555ZVXXnnllVcaqlNHC/DKK6+88sorr7zyyquG8k5OvfLKK6+88sorr7xSU97JqVdeeeWVV1555ZVXaso7OfXKK6+88sorr7zySk15J6deeeWVV1555ZVXXqkp7+TUK6+88sorr7zyyis15Z2ceuWVV1555ZVXXnnV5jLGPGWM2WeM+TDkvXRjzHJjzKfB//aJup54+ZwaY54CZgP7RCQ/+N6vgPOBWmA7cLWIHAx+NgZ4DOgJ+IFTRORos3WmA38HBgE7gUtF5IATfSNGjJgBPAQkAU9s27btvjj1dAFWAp2BZGDBtm3b7owHqw0arfS4laVdn02Wpy9xWNr12WRp12eTpV2fTZZTfcdzGWPOBCqBP4ec+90PfC4i9xljfgz0EZEfRVxPHE9OwwmcDvxLROqNMb8EEJEfGWOSgfXAFSKy0RiTARwUEV+zdbZ6A8PViBEjkoBPgHOBPcBa4Bvbtm37uD17gn0GSNu2bVvliBEjUoB3gO9t27ZtTRxYVrZLuz6bLO36bLI8fYnD0q7PJku7Ppss7fpsspzq8wqMMYOA10LO/bYBZ4lIiTEmB/i3iIyItI643dYXkZXA583eWyYi9cFf1wAnBH+eDmwSkY3B5Sqan5gG6wLg2eDPzwJzHcqbDHy2bdu2Hdu2basFng+uu7172LZtm2zbtq0y+GtK8BXtXwSOWA77rI2FS1na9dlkefoSh6Vdn02Wdn02Wdr12WQ51efVsZUtIiXBn0uB7GgNyfHVE7G+ReAWPcBJgBhjlgKZwPMicn+YnlZvIMBRX3WTk8EHfnM/q95Z3fj+PffdzeZNH3LUV31vS+uIteetkjeO6fX7/Pzq+t+S2iWVMy74ChdcN3vN0j2vNn4+JWeaI5ZTjR3R41aWdn3eWCSOPm8svLHoaJZ2fRrHoktSNxNJr40y555gL4d+RfG3getC3nlcRB6PtV1ExBgTVW+HnJwaY24D6oG/hug4HTgFqAbeMMasE5Fjz/SCFesGtlRlpWXMmTkXv8/PyXknk56eHpcegK3rPqGupo4evbuz4a1NnHLuRHIH58SFZWu7tOuzyXLSU1dXx7//9W8u+fAj6ut9nDhoIFnZ0f+t5cax8PR1DEu7Ppss7fpssrTrs8lyqs/NFTwRjflkNFhlxpickNv6+6I1WH9a3xjz3wQelPqm/GfC6x5gpYiUi0g1sBiYEKa9LLhhRNtAY8x1xpgCY0zBk396qslnfTP7sm7den7/2KMsevUlPihYT3JSUkTdTnogcNX0xYcXMf8X87jtmVuoranlvWUFcWHZ2i7t+myynOrL7Z/L8BEn8eKiF3hh4fN8/OEW/L5wM1naztI+Fp4++yzt+myytOuzydKuzybLqb4OKWPsvZzVP4Crgj9fBbwSrcHqyakxZgZwCzAneBLaUEuB0caYbsGHo6YA4SYdx7yBIvK4iEwSkUnXXPut5p8FZn22YqCd9ABsKdhGenYf+uZm4PcLyakpVB6ojNjjlGVru7Trs8lyqi9/dD7Fe4rZs6eYo9VHOXjwIBMmhvv3WNtZ2sfC02efpV2fTZZ2fTZZ2vXZZDnVd7yXMeZvwLvACGPMHmPMNcB9wLnGmE+Bc4K/R6y43dYPCjwL6GuM2QPcCdxKwFJpuQns8DUiMl9EDhhjHiTwNJwAi0Xk9eB6ngD+KCIFBDboheDG7gIudaKtoryCwUMGc8HMuQjC4EGDqY9y1SrWnubzR3ceKKb4k73cPONWBGHgiQMZkjnsmOXaqi/e2xVapSWlJKckM3fWhYgIAwYOUKXPJivWnnBzkUdPHcWs6bMBSM/uQ9rE1CbLNT9G3DIWnr6OZ2nXZ5OlXZ9NlnZ9NllO9XVIKXKsF5FvtPBRyyc9YSqeT+t/Q0RyRCRFRE4QkSdFZJiIDBCRccHX/JDl/yIieSKSLyK3hLw/L3hi2vAU/zQRGS4i54jI5+HY0crv97O7qIhXXl/EewXvUln5BYcOHWr3HoB+/bIxBl5+bSFr1q6murIqbixb25WcnMzUaWdTsPF91qxbTV1dHeXl5Wr02WQ51ufzs3b5eu547lbuWXAnlYeq2LTqw8g9bh0LT591lnZ9Nlna9dlkaddnk+VUn1ftUx35tH6H1RdfVJKcksIJAwJOVoOHDObQgYPt3uNWVna/bMrLKwCor6+n5mgNGX36qNFnk+VU366tRWT2z6BvbgYAg0YNZNVraxhzWn67s7SPhafPPku7Ppss7fpssrTrs8lyqs+r9ilFF4PtVffu3amrrWPPnmLqauso3FFIr969273HLay3St5o8qroW8a27VuZPuM8zjztLI7UVjN0+qBjlnPjWDjtmZIzrcmr16F0huYMZ0rONE7tcxo1FXX07ZrZZBm3joWnr+NZ2vXZZGnXZ5OlXZ9NllN9HVL6H4hqdcVzzukx8aUhn90MPABkikh5MGf1KWAocBT4logcc4/TGDOYgBFuBrCOQKJUbWu1JSV1YuCJA5vMJenVu1e797iVlZSUxKkzJrPsLysQEZKSk4glacydY+FM3+HDX/DmG28yaezkxrnIuf1z48KyORZjx41h/rzrKS0ppVevXur0aR4/myzt+myytOuzydKuzybLqT6v2qfieeX0GWBG8zeNMQMIJEIVhbz9E2CDiIwBriSQZRuufgn8RkSGAQeAa5wI65vZl8LCwsa5JAcPHojJVqK1PW5l+X1+3l9awG1P38IDr99LckpSVHssm/psspzqGzN2ND6/r3Eu8uGDhxJ+LLKys8AYLr3sEs6ZPo3uPdLIzspUo0/7+Nlkaddnk6Vdn02Wdn02WU71dUgZiy9LZTW+NFi/IWAnFXqpbRTwr2DfVmCQMaaJI7kJPN4/FVgQfOtZHMaXutH2wibLiT2WTX02Wdr12WTl5edRuL2QFctWMOeC89m/bz9Tzj5LjT7t42eTpV2fTZZ2fTZZ2vXZZHlWUh1bVh+IMsZcABSLyEbTdIdvBC4C3jbGTAZOBE4AykKWyQAOikh98Pc9QH8nOirKK5gwcXyT24+JbnsRT1Z72GPFU19HsrTriyfr2W1PH9NXm1zDps2buWH+jXRJ68wq/0pWbVvZ+PlVI662pq89etzK0q7PJku7Ppss7fpsshLKSsqFJ9DWHogyxnQjcPv+jjAf3wf0NsZsAL4LfADE9Sjol9Ov8fZjRt/okWTabS9sspzYY9nUZ5OlXZ9N1vb3dlBZUcXVj17JRbdfgK/eR8XuCjX6tI+fTZZ2fTZZ2vXZZGnXZ5PlWUl1bNm8cjoUGAw0XDU9AVhvjJksIqXA1dB4+74Q2NGsv4LACWxy8OrpCUBxSzBjzHXAdQCP/uERQlOisrKz2LWriOI9e5n37Wu49+57o86N0257YZOlXZ9NlnZ9NlmfvPspvnofC+5YSH2dj/paH4sfXMoVv7lchT7t42eTpV2fTZZ2fTZZ2vXZZCWUlZQLfZesbZKIbBaRLBEZJCKDCNyWnyAipcaY3saY1OCi84CVInK4Wb8AbwJfC751FQ7jS/Py8/j4o4+5/IrL8fv8Mc2N0257YZOlXZ9NlnZ9NllDJw9h5JQRzHv8ambdPIP0AenkjozsQGBTn/bxs8nSrs8mS7s+myzt+myyEspKyoVlNb5URJ5sYfGRwLPGGAE+IuQpfGPMYmCeiOwFfgQ8b4y5h8Ct/5bW16Saz4379N3P8Bs/3//f7yN+oXO31Khz47TbXthkaddnk6VdXzxZzb8jy3cuZ+HqRbx00ytUV1eDwMl9RoadZ2pDX3v0uJWlXZ9NlnZ9Nlna9dlkJZSVlDfnNPYKF1/a7PNBIlIe/PldETlJREaIyEUiciBkuZnBE1NEZIeITA7GoF4iIjVOtJV8WkpKl2S690mjS48uHK2q4aWfvhyxR7vthU2Wdn02Wdr12WT1zezLunXr+f1jj7Jk2esgok6f5vGzydKuzyZLuz6bLO36bLISykrKhXVcxpeeeeXpnHnl6QAUrt/J6w/8k69840sRe7TbXthkaddnk6Vdn02Wpy9xWNr12WRp12eTpV2fTVZCWUklgMTWlvWEKGPMd4HvEHga/3URuSXks4HAx8BdIvJAmHUOxkFC1KCeA49576M1W3jyrmeor/PRrUc3vjxpcsR1aLe9sMnSrs8mS7s+myxPX+KwtOuzydKuzyZLuz6brNKSUpJTkpk760JEhAEDB+i1knJhWU2IMsacDVwAjBWRPAIRpqH1ILAkwjrbJSHK7/Oz4NGX+clTt3DPgjupr6tn06pj0lKb9ii3vbDJ0q7PJku7PpssT1/isLTrs8nSrs8mS7s+m6zk5GSmTjubgo3vs2bdaurq6igvL4+qr0Oqk7H3slRxu3IqIiuNMYOavX09cF/DXFER2dfwgTFmLgELqapw6wtJiGrwpXkWuAv4Q2u17dpaRGb/DPrmZgAwaNRAVr22hjGn5bfYo932wiZLuz6bLO36bLI8fYnD0q7PJku7Ppss7fpssrL7ZVNeHvBprq+vp+ZoDRl9+kTV51X7lG13rJOAM4wx7xlj3jLGnAJgjOlO4En8n0bobbeEqNKiMrr37gFAbU0dB8oOYqLMK9Fue2GTpV2fTZZ2fTZZnr7EYWnXZ5OlXZ9NlnZ9Nll5+Xns2rmLubMv4uzTp1Hvq+frl18WVV+HlLH4slS2H4hKBtKBLwGnAC8YY4YQuAL6GxGpjHaS6KSOid80xbyy+vUm8Zujh46JGL+p3fbCJku7Ppss7fpssjx9icPSrs8mS7u+trDGjhvTJKZbmz5NrLdK3jimb+y5o1n2lxWBh6OSYO3+dynuvrPJMuedcH5UzV61vmxfOd0DLJRAvQ/4gb7AqcD9xpidwP8CPzHG3NistzEhKvh71IQoY0yBMabgyT891eSzMWNH4/P7GuM3Dx88lPC2FzZZ2vXZZGnXZ5Pl6UsclnZ9Nlna9Tnty8rOAmMaY7q790iLmoTo1rFw0uP3+Xl/aQG3PX0LD7x+L8kpSby3rCCqPq/ap2xfOX0ZOBt40xhzEpAKlIvIGQ0LGGPuAipF5NHQRhERY8ybBBKinieGhCjgcYCjvmpp9pnrbC9ssrTrs8nSrs8my9OXOCzt+myytOtz2peXn0fh9kKqq6qYf8O3WfnvlVGTEN06Fk56thRsIz27D31zM6itqSM5NYXKA5Wt0mqtEsHuqpVlNSEKeAp4yhjzIVALXCUi0vJa2ichqnlVlFcwYeJ4rr/2Bvx+P+PGj4/JiqK1PW5laddnk6Vdn02Wpy9xWNr12WRp1xdrX/MkRIBOvQ3b9+zgBz/6Iam9oichumUsnPQcM/3vQDH7CvfzyPWP4ff7GT96PJmZmRGn/3nVfhXPp/W/0cJH/xWl765mv88M+XkHENmQNMYyxgReGExSbP/qcNLjVpZ2fTZZ2vXZZGnX1y+nH7977FG+cck32b27iIzMDFX6bLK067PJ0q7PSd+OtYV0TutMyuEj1B6ti/lhFjeOhZOe3Nwc8vJHUVy8F7/PT1Ky4nQo9104tT7nVEWFxiwuevUlPihY36poxlh73MrSrs8mS7s+myzt+rKysygtLeOvz/0fQ4YOpqamJuocPJv6bLK067PJ0q7Pad+eLcXs/GAXdUfr8Pv8HNh7kEX3/EONPu0sp/q8ap86LuNL3Tg/xiZLuz6bLO36bLK063MyB8+mPpss7fpssrTrc9o3bPJQ9heWc/Gdc9n94R5W/OFf9B+Zo0afdlZCxZdaNMe3VSriS40xKcATwISgpj+LyC/CrHMwDuJLm5cbo9ZssrTrs8nSrs8my2ncX7z0hZuDV5tcw6bNm7lh/o10Set8zBw8CD8Pz237Srs+myzt+mLta37cLt+5nKK0Pbx00ytUV1eTJMn09w88Zjlb+hKN5VSfV+1TWuJLLwE6i8hoYCLw7TDpUtBe8aUujFqzydKuzyZLuz6bLKdxf7b0bX9vB5UVVVz96JVcdPsF+Op9VOyuUKPPJku7Ppss7frawtq4YSO/f+xRlix7neSkJHX6NLOc6uuQMhZflkpLfKkAaUEP064EnuQ/HNrYnvGlboxas8nSrs8mS7s+myyncX+29H3y7qf46n0suGMh9XU+6mt9LH5wKVf85vKIfW7cV9r12WRp12eTpV2fTZZTfV61T6mILwUWAFVACVAEPCAinzfrbbf4UjdGrdlkaddnk6VdXzxZb5W80eRV0beMbdu3Mn3GeZx52lkcqa1m6PRBTZaxqe+qEVc3eV0z9xrmzDmft99aycO/fYhhw4cx9SvTjlnOlj6bY5Fo+myytOuzydKuzybLqb4OKWPsvSyVlvjSyQTmoOYCfYC3jTErgtZR7V5uiFrrSJZ2fTZZ2vXZZCUlJXHqjMmNcX9JyUlEsTG2qg+grLSMOTPnUl1dHZM22/rcety2NkbT5nZp31c2Wdr12WQ51edV+5SW+NLLgX+KSF3wVv8qYFKz3naLL3Vj1JpNlnZ9Nlna9dlkOY37s7mvGqxhlix7HURUjZ9NlvYYTZvbpX1f2WRp12eT5VRfh1QnY+9la5OskQL1MnA2QGh8KYFb+VOD76cRuLK6NbQxmCT1JoH4UoghvlREJonIpGuu/Vbzz1xne2GTpV2fTZZ2fTZZoXF/fr/EHPfn7Sv7rI6I0VyxbAVzLjif/fv2q7Lw0r6vbLK067PJSigrKReWivhSY8zvgKeNMR8ReB7saRHZFFxPu8eXOrG80W57YZOlXZ9NlnZ98WSFi/sr/mQvN8+4FUEYeOJAhmQOixr35+0r+6x46nNi4RVurq8bxiLRWNr12WQllJWUC8+f43blVES+ISI5IpIiIieIyJMiUisi/yUi+SIyQUT+FVy2UkQuEZE8ERklIr8KWc/M4IkpIrJDRCaLyLDg8jVOtDmxvNFue2GTpV2fTZZ2fTZZ/fplYwy8/NpC1qxdTXVllSp92sfPJsumPu0WXtr3lU2Wdn02WQllJeXCOi4TopxY3mi3vbDJ0q7PJku7PpssT1/isGzq027hpX1f2WRp12eTlVBWUi6cemB7zqmKysvPY9fOXcydfRFnnz6Nel89X7/8sog92m0vbLK067PJ0q7PJsvTlzgsm/qGTh7CyCkjmPf41cy6eQbpA9LJHZkbF5b2sdDO0q7PJiuhrKRcWFbjS40xfwdGBBfpTcC3dFxIz0DgY+AuEXmAZuU0vjScv+LYc0c3Wt6QBGv3v0tx952NnzefJ6fd9sImS7s+myzt+myyPH2Jw2qLvmi2UOFiNBeuXtQYo4nAyX1GRozRtLld2veVTZZ2fTZZnpVUx5bV+FIR+bqIjAuekL4ELGzW8yCwJMI62ye+1IHljXbbC5ss7fpssrTrs8ny9CUOy6k+J7ZQ2i28tO8rmyzt+myyEspKylh8Wap4PhC1Emie8gQ0RpFeCvwt5L25QCHwUYSeqQTSpCAQXzrXiTYnljfabS9ssrTrs8nSrs8my9OXOCyn+pzYQrl1LNzI0q7PJsuzkurY6qgHos4AykTkUwBjTHcCNlHnAj9oocdxfGk4y5tpXzqHk/1jkCThkQcfYck/l0S0vNFue2GT5cSKy6Y+myzt+myyPH2Jw4q1pz1sodwyFscDS7s+m6yEspKyaI5vqzrqgahvEHLVFLiLwO366I7d7VBDBg/mqquuIrt/Jv0GZNGvXz/y8/Mj9mi3vbDJcmLFZVOfTZZ2fTZZnr7EYTnV58QWyq1j4UaWdn02WZ6VVMeW9SunwfjRi4CJIW+fCnzNGHM/gQel/MaYoyLyaMgyjfGlwaunUeNLgesAHv3DI4SmRGVk9kWQxt999fUkJ0U+T9due2GT5cSKy6Y+myzt+myyPH2Jw3Kqz4ktlFvHwo0s7fpsshLLSqqjBbR/dcSV03OArSKyp+ENETlDRAaJyCDgt8C9zU5M2zW+9KQRw/nb3/5GqulCaqcu7Ny5iyHDh0YUrd32wibLiRWXTX02Wdr12WR5+hKH5VSfE1sot46FG1na9dlkeVZSHVtW40tF5EngMpre0o+2njbHlw68e3qT33t16cH3B15M3uh8/H4f539tLtXj01jw/LLGZYpuX9akR7vtRTxZTqy4oH3suGyOhd/vw3QyTebSatKnneXpSxxWrD3tYQvllrE4Hlja9dlkJZSVlAsf2rIaXxp8/79F5I8R+pp4nMYjvvSMQRN49g9PceCrvTjwtb68ufItBnfKitij3fbCJsuJFZdNfU77cvvnkpef1ziXtrqqGn+UCfDa95VNlqcvcVht0ddaWyg3j4XbWNr12WQllJWUC+u4TIjase0zTjzxRDr36QZJhi+d9RXeXbkqYo922wubLCdWXDb1Oe3LH51P8Z5i9uwp5mj1UQ4ePMiEiRPU6NPO8vQlDku7Ppss7fpssrTrs8lKKCupThZflqqjrKSsVvNb9MuXLufdmjX87Sd/BeCVl1/mg00b+L//91iL66gor2DCxPFcf+0N+P1+xo0fH7PtRWv7tLHCWXHtK9zPI9c/ht/vZ/zo8WRmZka04oqnPqd94aYrzJw/nTmzAk8hD8kfTNrE1CbLNd9GbfuqI1mevsRhaddnk6Vdn02Wdn02WU71edU+FbfzYGPMU8aYfcaYD0Pe+7sxZkPwtdMYsyH4/jdD3t9gjPEbY8aFWWe6MWa5MebT4H+jPyLeQvnxc6Suirlz5/L0M8d6+bWwTYEXBpMU+7+mnPQ56amrq2Prlq106dKZ1NRUfP766E0OWLm5OeTlj6K2tpbamlqSkmO/1WFrLJz2JSUlkZGdTkZ2Oof2H2JvYYkqfZqPC6c9Nlna9dlk9cvpx8uvLSQtLY3du4vU6bPJ0q7PJku7Ppssp/qslzH2XpZKRXypiPw15P0rgEIR2RBmnT8G3hCR4cAbwd9bXVnZWZSWlvHX5/6PIUMHU1NT06oIvkWvvsQHBetbPUcr1j6nrNz+uQwfcRIvLnqBFxY+z8cfbolpzqQtfdpZfp+fFx9exPxfzOO2Z26htqY2plhb7WPhHReJoc8my8nfQJv6bLK067PJ0q7PJsupPq/ap9TEl4bUN4DnW1jtBQRiS6EN8aVujOADe3MmE2EsbM2lTYSx8I6LxNBnk+Xkb6BNfTZZ2vXZZGnXZ5OVUHNOjcWXpVIRX9qsvk7gJDRcZYtIw33WUiA7FljzCL762noqDlawb/8+bph/I527pSZcBF+4+ZIAo6eOYtb02QCkZ/eJac6k22LnYu0LN5e2+JO93DzjVgRh4IkDGZI5LKFibb3jInH1xZPlJIYU2ufvoLaxSDR9Nlna9dlkJVR8qQtLS3wpAMaYU4FqEfnw2JamFTTlFyfwXRuKGHH6cP53wY1cdMcFiF/Yuy3y3ELtUWsQuC29dvl67njuVu5ZcCeVh6rYtCryULoxds5pX79+2RgDL7+2kDVrV1NdWaVKn3dcuFufTZaTGFKb+myytOuzydKuzyYroeJLvTmnba+Q+NK/h/k4mkF/mTEmJ7ieHGBfBM51xpgCY0zB2y+80+SzvdtK2PnBLp647mkW/2YpNUdqeff59yLqDo0yS0lNcRS1FmufU9aurUVk9s+gb24GPfv0YNCogax6bY0afW5kadcH3nGRKPpsskJjSJc8tKwxhlSLPpss7fpssrTrs8lyqs+r9ikV8aUAxphOBOahtjTfFOAfBGJLoRXxpVfMu5xBPQc2vq74zje452930qNnD+qP1JHRL4Obf/0/TZZpXtqi1qbkTDvm1etQOkNzhjMlZxqn9jmNmoo6+nbNbLKMLX02x6IjWdr0ecdF4uqLJ+uqEVc3eV0z9xrmzDmft99aycO/fYhhw4cx9SvTjlnOjWORaPpssrTrs8ny4ks7tjTFl54J7BaRHc3W8wTwRxEpAO4DXjDGXAPsInAy66g6JXXih3/4X+6+6j4O7DvA3sIScgfntLi89qg1gMOHv+DNN95k0tjJjXMmc/tHzr12Y+ycTZZ2feAdF4mizzarrLSMOTPnUl1dHXj4I4Zy41ho12eTpV2fTVZCxZe6ME5JTXypiPxbRL4U5v15wRNTRKRCRKaJyHAROUdEwroBxFr/Xvg2g0aeSEZOOlvWbou4rPaoNYAxY0fj8/sa50wePnhIVaybG1na9YF3XCSKPpssJzGkNvXZZGnXZ5OlXZ9Nlhdf2rHlwvPt6PXFwUpKdpby8XtbmHTORKoOVZE9ICtij3bbC5ss7fpssrTrs8ny9CUOS7s+myzt+myytOuzyUosKyn3PRAVz9v6TwGzgX0ikh98bxzwR6ALUA/cICLvG2O+CfyIgIvWF8D1IrIxzDoHE5iTmgGsA64QkdpoWnYebpp+Urq9jL/9+AVMJ8Nnm3aQ0jWF7nndmyw3pdkdfu22FzZZ2vXZZGnXZ5Pl6UsclnZ9Nlna9dlkaddnk+VZSXVsWU2IAu4HfhpMgroj+DtAITBFREYDdwOPt7DOXwK/EZFhwAHgGifCqg5UkTd1JP/74o1cePsccIuVlHKLDTeytOuzyfL0JQ5Luz6bLO36bLK067PJSiwrKYsvSxW3K6cistIYM6j520DP4M+9gL3BZVeHLLMGOKH5+oKpUlOBy4NvPQvcBfyhtdpCraTqausbraQuvnNuiz2hthKAI9uLWPu0s7Trs8nSrs8my9OXOCzt+myytOuzydKuzybLqT6v2qdszzn9X+BXxpjdwAPArWGWuQZYEub9DOCgiNQHf98D9Hci4owrTmPeY1fTOa0ztUdq6ZXVM+KJKei3vbDJ0q7PJku7PpssT1/isLTrs8nSrs8mS7s+m6yEspLqZOy9LJXt+NLrgZtE5CVjzKXAkwR8TwEwxpxN4OT09PaEhvMt/fj9rZh6Q1r3blQfOEJqRUrCW0lpt9hwI0u7voa+MWNHM3fWhSQlJXH6Gacfl2OhXZ9NlnZ9Nlna9dlkaddnk5VQVlIuLNtXTq8CFgZ/fhGY3PCBMWYM8ARwgYiEy9GrAHoHE6YgcOu/uCVQaELU4r/+s8lnfp+fFx9exPxfzOOu/7uNLmmdeW9ZQUTh2m0vbLK067PJ0q4PICs7i5KSUs6ZPo1TTj2FUXknk52VqUaft6/ss7Trs8nSrs8mS7s+m6yEspJy4dP6tk9O9wJTgj9PBT4FMMYMJHDSeoWIfBKuUQIu0W8CXwu+dRUxJkTN/GbT57K2FGwjPbsPfXMz8PuF5NQUKg9URhSu3fbCJku7Ppss7foA+vbtS2lJKWeceQbi9/PPJUuZcvZZavR5+8o+S7s+myzt+myytOuzyUooKykXltWEKOBa4KHg1c+jwHXBxe8gMKf094HnnqgXkUnB9SwG5onIXgJ2U88bY+4BPiAwLSBqNY9n3HmgmH2F+3nk+sfw+/2MHz2ezMzMsDGODVVRXsGEieO5/tob8Pv9jBs/Pmbbi9b2aWdp12eTpU3fs9uePqbv1ftf58v/dSo//+W91FTWcOrXTmGVfyWrtq1sXKZ5VKUbxiLR9Nlkaddnk6Vdn02Wdn02WU71dUi58Pw5nk/rf6OFjyaGWXYeMK+F9cwM+XkHIVMBnFZubg55+aMoLt6L3+cnKTm2S/XGmMALg0mK/Whw0qedpV2fTZZmfTvWFtKtVzf65PbGGAMGTIyT2t02Fk576urq2LplK5lZmdTX+/D566M3OWQ57XPjvrLJ0q7PJku7Ppssp/q8ansdlwlRodF9i159iQ8K1rcq7i/WHreytOuzydKur3jrXj57fzsv//xV6mvrEb+wdtE6KnaHm9ZtX59NllN9uf1zGT7iJF5c9AIvLHyejz/cgj/KFRS3joUbWdr12WRp12eT5VRfR1TjSbSFl62y/bS+inLj/BibLO36bLK06zvjitMYesoQ3v37e0y+eBIFL6+n/8gctr+/g4wBGR2uzybLqb780fkU7ylmz55ievXoycGDB5kwcUJcWNrHwo0s7fpssrTrs8ny5px2bGmJLz2LwMNNhcH2hSLyszDrHIyD+NLm5caoNZss7fpssrTpaz53FGD5zuUUpe1h5e9WcejgIYZlDic9PT3ssvHW15GsWHveKnnjmPdGTx3FrOmzAUjP7kPaxNQmyzWfr+6WsUg0VmlJKckpycyddSEiwoCBA1Tp087Srs8mK5HiS21e0bRVWuJLAd4WkXHB1zEnpsFql/hSN0at2WRp12eTpV1fQ9/GDRt56tkneGfNSjZv2HRcjoVjfT4/a5ev547nbuWeBXdSeaiKTas+jA9L+1goZyUnJzN12tkUbHyfNetWU1dXR3l5uRp92lna9dlkJVR8qQtLRXxpLNWe8aVujFqzydKuzyZLuz6bLLfq27W1iMz+GfTNDUyDGDRqIKteW8OY0/LbnaV9LLSzsvtlU14emE9dX19PzdEaMvr0UaNPO0u7PpssL760Y0tTfOmXjTEbjTFLjDF5YXrbLb7UjVFrNlna9dlkaddnk+UWfVNypjV59TqUztCc4UzJmcapfU6jpqKOvl0zmyzj1rHQznqr5I0mr4q+ZWzbvpXpM87jzNPO4khtNUOnD2qyjFvH4njQZ5OVSPGlxth72Sot8aXrgRNFpNIYMxN4GRgeLxFJSZ0YO24M8+ddT2lJKb169Ur4qDWbLO36bLK067PJcqu+w4e/4M033mTS2MkIwsATB5LbPzcuLO1joZ+VxKkzJrPsLysQEZKSkwIPtqjRp5ulXZ9Nlhdf2rGlIr5URA6LSGXw58VAijGmb7Nex/GlT/7pqSafZWVngTFcetklnDN9Gt17pMUlztFpn3aWdn02Wdr12WS5Vd+YsaPx+X28/NpC1qxdzeGDh47bsdDO8vv8vL+0gNuevoUHXr+X5JQkL5raRfpsshIpvrSTMdZetsr2ldOG+NJ/0zS+tB9QJiJijJlM4KS5iRFj8LM3CcSXPk8M8aXA4wBHfdVN/umcl59H4fZCqquqmH/Dt1n575VxiXN02qedpV2fTZZ2fTZZnr7EYWnX57QvNJq6tqbOi6Z2mT6bLM9KqmNLS3zp14DrjTH1wBHgMgneizHtEF8aLtKxNrmGTZs3c8P8G+mS1jmmOEfNthc2Wdr12WRp12eT5elLHJZ2fbH2hYumLv5kLzfPuJWGKRhDModFjaZ2w1h0tL4JE8c3mSp3vI5FR5RnJdWKEpFviEiOiKSIyAki8qSIvCMiE0VkrIicKiLrgss+KiJ5wfe/JCKrQ9YzM3hiiojsEJHJIjJMRC4RkRon2ra/t4PKiiqufvRKLrr9Anz1vqiJOdptL2yytOuzydKuzybL05c4LO36nPb165eNMTROwaiurFKlTzurLfZJ/XL6NU6Vy+ibHpdtsrldnpVUx9ZxmRD1ybuf4qv3seCOhdTX+aiv9bH4waVc8ZvLW+zRbnthk6Vdn02Wdn02WZ6+xGFp12eTpV2fTZZTfVnZWezaVUTxnr3M+/Y13Hv3vVGf43DrWHREeVdOE7SuGnF1k9c1c69hzpzzefutlTz824cYNnwYU78yrckyzUu77YVNlnZ9Nlna9dlkefoSh6Vdn02Wdn02WbH2PLvt6Sav9akFbNq8iZypWSzdtYTi0r0cHHygyTJuHQuv4lNxvXLaygjTXsBfgIFBXQ+IyDFHtDFmIoH0qa7AYuB7DfNTW1NlpWXMmTmX6urqqFYjoN/2wiZLuz6bLO36APx+H6aTaRLpeDyOhRMLOZv6bLK067PJ0q7PJsupvp3rd5ExIJ1X71+MiNAlrTN9B2bEhaV9LDqivCunra9niD3C9DvAxyIylsCDVL82xqSGWecfCDxYNTz4ar7+qNU3sy/r1q3n9489ypJlr4NIwtte2GRp12eTpV0fQG7/XPLy8xojHaurqvFHmdjvxrFwYiFnU59NlnZ9Nlna9dlkOdW3Z0sx+3bsp1uvrnTp0YUjlUdZdM8/4sLSPhZetU/F9eRURFYCnzd/m/ARpgL0CMaUdg/21Yc2GmNygJ4isiZ4tfTPwFwHulxne2GTpV2fTZZ2fQD5o/Mp3lPMnj3FHK0+ysGDB5kwcYIafbZYDRZyK5atYM4F57N/3/6oFnI29dlkaddnk6Vdn02WU33DJg9l4JgBXPfkNcz+wVfpk9Ob/iNz4sLSPhYdUe2ZABXtZas64oGo/wWWGmMeIHBy/JXg+48C/yBwstoD+LqI+Jv19icQW9pQjiJM3Wh7AVBaUkpySnKT27eaLDbcyNKmL1w8I8DoqaOYNX02AOnZfUibmNpk2eZWO24YCycWctA+NnLaxiLR9NlkaddnkxVrT/PvyPKdyylK28NLN71CdXU1SZJMf//AsM9vtEVfvLerPfR51T7VEQ9ENUSYDgBu4j9epecBG4BcYBzwqDGmZ7gVtEe5zfYCIDk5manTzm68fVtXV0d5ebkafW5kadcHgdSctcvXc8dzt3LPgjupPFTFplUf6tFnieXEQs6mPpss7fpssrTrs8lqi76NGzY2TpVLTko6bseiI8oYY+1lqzri5PQqwkSYAlcDCyVQnwGFwMnNeosJxJY2VIsRptHiS3ftKuLtt97hwosvpKamplW2FympKY5sL2Ltc8rK7pdNeXng/2zr6+upOVpDRp8+avS5kaVdH8CurUVk9s+gb24GPfv0YNCogax6bY0afbZYoRZySx5a1mghp0WfTZZ2fTZZ2vXZZGnXZ5PlVJ9X7VMdcXLaEGEKIRGmQBEwDcAYkw2MAHaENopICXDYGPOl4NzUK2khwlREHheRSSIy6Zprv9Xks7z8PD7+6GMuv+Jy/D5/THPPtNteNGzXrp27mDv7Is4+fRr1vnq+fvllavS5kaVdH0BpURnde/cAoLamjgNlB6P+C9iNYzF08hBGThnBvMevZtbNM0gfkE7uyFw1+myytOuzydKuzyZLuz6brESyknLjldN4W0m1JsL0buAZY8xmwAA/EpHy4Ho2BJ/uB7iB/1hJLQm+IlbzuWc71hbSq39Pvv+97zfaXkSLL9Vme9HS3MKx545m2V9WBCZzJ8Ha/e9S3H1n4+fN5xa60Q7FJkubvnAxjTtNMa+sfr1JpOPooWMiRjq6YSzCzYtbuHpR47w4BE7uMzLivLh46utIlnZ9Nlk29TmxdbO5Xdr3lU1WIllJubHi/bR+ayJM94rIdBEZLSL5IvKXkPWMC/m5IPj5UBG50YnHqRttLyAwt/D9pQXc9vQtPPD6vSSnJPHesgI1+tzI0q4PYMzY0fj8vsZIx8MHDx2XY+HEQs6mPpss7fpssmzqc2LrZnO7tO8rmyzPSqpj67iMLx02eSj7C8u5+M657P5wDyv+8K+Et70A2FKwjfTsPvTNzaC2po7k1BQqD1Sq0edGlnZ9NlmevsRhaddnk2VTX6itW68ePWOydXPK0j4W2lkJZSWFfo2trePi5DTc7b2aoXVcNeJq1h4qYGPPD2OyvZgwcTzXX3sDfr+fcePHx2x70dq+WHrC3r49UMy+wv08cv1j+P1+xo8eT2ZmZsTbt/HSd7ywtOuzyfL0JQ5Luz6brHjqCzf9aub86cyZFXCLGJI/OKqtm83t0r6vbLKc6vOqfSpuJ6ctRJeOJRBd2h3YCXxTRA4bYwYBW4BtwfY1IjI/zDrTgb8Dg4L9l4rIgdZqq6ur49//+jeXfPgR9fU+Thw0MNZtCrwwmKTY/6XipM9JT25uDnn5oygu3ovf5ycpObZbELb0uZWlXZ9NlqcvcVja9dlk2dSXlJRERnY6vnofh/YfYm9hCbmDI9+5c8rSPhbaWU712S6bDyrZqnjOOX2GY6NFnwB+LCKjgUXAD0M+2y4i44KvY05Mg/Vj4A0RGQ68Efy91ZXbP5fhI07ixUUv8MLC5/n4wy0xxTk2zFdb9OpLfFCwPub5Ma3t087Srs8mS7s+myxPX+KwtOuzybKpz+/z8+LDi5j/i3nc9swt1NbURn0uwOZ2ad9XNllO9XnVPhW3k9MWoktPAhoeiV8OXNzK1V4APBv8+VkcRJeCvThHp33aWdr12WRp12eT5elLHJZ2fTZZNvWFPhfg90tMzwU4ZWkfC+2shJpzauy9bJXtOacfETjBfBm4BBgQ8tlgY8wHwGHg/4nI22H6s4NepwClQHYs0Obzfupq66iVmsY4x95ZvWOKcxw3fiwXzr4IESE1NZUNH2yMyj6eI+6OB5Z2fTZZnr7EYWnXZ5MVT33N/39k54Fiij/Z28TWbUjmsIjPBdjcLu37yibLiy/t2LJtwv8t4AZjzDqgB1AbfL8EGCgi44HvA/9nokSXBi2kxImI5JRkbnnsJh5+4wF+/tJdHKk8wtoV66L25eTm8NaqNynY+D63/OSHlJSUsGnjpog9XsSdu1na9dlkefoSh6Vdn02WTX39+mVjDI22btWVVcftWGhnJVJ8aSdjrL1sldUrpyKyFZgOYIw5CZgVfL8GqAn+vM4Ys53AFIDmk3HKjDE5IlJijMkB9rXEMsZcR9Dg/3/uu4GZ35wR+hmdu3YGoHOXVDp3SWXnliJOOWdii9qzsrMoK9tHt7RuAJTuLSGpUxJEsXAIjUADWh2bFmuPTZZ2fTZZ2vXZZHn6EoelXZ9NlnZ9Nlna9dlkOdXnVfuU1Sunxpis4H87Af+PwJP7GGMyjTFJwZ+HAMNpFl0arH8AVwV/vooWokuBJvGloSemAF8crKTyUBW/vO5BfnLxXZhOnRg5aUSTZd4qeaPJq6JvGZ/s2MaCguf56uyZ/PH3jzN04iAOZO1rslzz8mLd3M3Srs8my9OXOCzt+myytOuzydKuzybLiy8N/7JV8bSSChdd2t0Y853gIguBp4M/nwn8zBhTB/iB+SLyeXA9TwB/FJEC4D7gBWPMNcAu4FIn2g5XHKZ6ax1P/v4pAF559WV69UuL2JOUlMTk807h3m89gIgwbOwQyks+j2oD4tZYt7HjxjB/3vWUlpTSq1cvVfpsspzGETrp0z4Wnr7EYWnXZ5OlXZ9NlnZ9NllefGnHVjyf1g8XXfqQiJwUfP24IXpURF4SkbygjdQEEXk1ZD3zgiemiEiFiEwTkeEick7DCWxra8SIEVz1rSv5LGULnyZ/xFdnz2Tfp+URe5pHg1YdrqbfoGy2rN0Wsc+NsW5Z2VlgDJdedgnnTJ9G9x5pZGdlqtFnk+U0jtBJn/ax8PQlDku7Ppss7fpssrTrs8lKpPhSN145tf1AlIrqVJtMRVU5fvFRU1PLe2ve48uTvhyxZ9fWIvpk9aZbz24kpyQz9owxfLRmC9kDsiL2udFiIy8/j8LthaxYtoI5F5zP/n37mXL2WWr02WQ5sSVz2qd9LDx9icPSrs8mS7s+myzt+myyEslKyo2lJSEqhYBB/4Sgpj+LyC/CrHMw8DyQAawDrhCR2ubLNa8ZP72lye+j007kxB0pfLr9M0SECy++iEPZnfifR37euMyRPzZ9er9203LSTBq3X/IzEEhNTWXwkEF856LvRmS7wWLj2W1PH9NXm1zDps2buWH+jXRJ68wq/0pWbVvZZJnmcbBuGItw84pHTx3VaEuWnt3nGFuylipaXzg7M01jEa5nwsTxTaZ7aNOnefxssrTrs8nSrs8mS7s+m6xEspJy4/mzloSoS4DOwfcnAt82gUjT5vVL4DciMgw4AFzjRNjmyp1knJ/HXxcv4A8LnmH1+2so3r47Yo/f7+ezTz/j5VcX8u7aVfTo0Z0TBg6I2NPQ5zaLje3v7aCyooqrH72Si24PZERX7K5Qo88my+/zs3b5eu547lbuWXAnlYeq2LTqw+j6HPRpHwuAfjn9Gqd7ZPRNj7q8TX2JMH7eWNhnaddnk6Vdn01WIllJaStjzE3GmI+MMR8aY/5mjOnS2nVoSYgSIM0Ykwx0JeB/eji00QQmO0wFFgTfehaHCVEYw7Pvv8Kke77GOQ/8N3V1dez+vCRiS6itREpqiiPbi1j7tLM+efdTfPU+FtyxkCUPLaO+1sfiB5eq0WeTtWtrEZn9M+ibm0HPPj0YNGogq15bE1Wfkz7tY5GVncWuXUW8/dY7XHjxhdTU1MQ0F9mWPu3jZ5OlXZ9NlnZ9Nlna9dlkOdXXEaVpzqkxpj/wP8Ck4F3zJOCy1m6TloSoBcH3S4BuwE1hHnbKAA6KSH3w9z1A/1ig/7zz/mPeS5Zkfv7t+ykvLuf8y7/Kqf81nivl3BbXEWorkZ2VReGOQsZNGB+V7aRPG6v57fncuSewqu9q7rr7Tta+X8D99/2K8RPGH7OcLX02Wc1vtX++ehFDc4YzJWcaR48e5dcVjzBwUE7UxBcnfdrGovl0D3+qn02bNzHr+zNYumsJxaV7OTj4QJPlwh0jto4LbePXkSzt+myytOuzydKuzybLqT6vgMC5ZdegA1M3YK+TFdisbwEPG2NuJ+BZ2jBfdDLgA3KBPsDbxpgVIhLO67TNVVdbxx9/+CTdktLo1y+F1xe/Rs5X+ra7JZTTPpssp1ZIZaVlzJk5l+rqaoKmC3HRp33cDx/+gjffeJNJYyfTEEeY2z83qj4nfU7HwpZt1c71u8gYkM6r9y9GROiS1pm+AzOi6tNuDeNGlnZ9Nlna9dlkaddnk5VIVlI2n6KPViJSbIx5ACgCjgDLRGRZa9dj9Wl9EdkqItNFZCLwN2B78KPLgX+KSJ2I7ANWAZOatVcAvYO3/gFOAIpbYhljrjPGFBhjChb/9Z9NPktOSebq+/6L7/7h2/zvYzdQV18X9ZaqdtsLp31OLY3WrVvP7x97lCXLXgcRV4yFk54xY0fj8/sa4wgPHzwUkz4nfU7HwpZt1Z4txezbsZ9uvbrSpUcXjlQeZdE9/4iqT7s1jBtZ2vXZZGnXZ5OlXZ9NViJZSdms0HOr4Ou6Zp/3IXAnfDCBC45pxpj/ai1HRUIUgTPsqcHP0oAvAVtDe4OeqG8CXwu+dRUOE6IqD1Xh8/kBOFJ1lCOVR8nI7hNRu3bbC6d9brQ0ssnSrg/s7eNhk4cycMwArnvyGmb/4Kv0yelN/5Et341oC8ut+8obC/ss7fpssrTrs8nyrKTCV+i5VfD1eLNFzgEKRWS/iNQRCFz6Sms5WhKifgc8bYz5iEBY/dMisim4nsXAPBHZC/wIeN4Ycw/wAfCkE22HKw7zpzuf5vOSAwD0ye7N1EvPitij3fYi1j4nVkiJZmkEUFpSSnJKcpNb2ZrsRmxaXUF89nHz+aPLdy6nKG0PL930CtXV1SRJMv39A6PORdZuDeNGlnZ9Nlna9dlkaddnk5VYVlKqTqCLgC8ZY7oRuK0/DSho7Uq0JERVisglEkiJGiUivwpZz8zgiSkiskNEJovIsODyNU605Qzqh6FTo41P1aFqV9j42LJCSoSxSE5OZuq0sxtvZdfV1VFeHiUFTPu+csqytI/9fj8bN2xsnO6RnJSkaiwSYl95Y2GdpV2fTZZ2fTZZnpWUsxKR9wg85L4e2EzgPLP51dWoZfuBKBUVauMDNNr4jDktv8WeUFsJwJHtRax9NlluHYvsftmUlwe8V+vr66k5WkNGn8hTN7TvK6csW/tY+1ho12eTpV2fTZZ2fTZZ2vXZZDnV1xGl7MopInIngbvljuu4jC8tLSqje+8eANTW1HGg7GDUnRtqK1FXW0fhjkJ69e4dleWkzybLrWORl5/Hrp27mDv7Is4+fRr1vnq+fnlkqzXt+8opy9Y+1j4W2vXZZGnXZ5OlXZ9NlnZ9NllO9XnVPhXPOacDgD8D2QSmFT8uIg8ZY9KBvwODCESYXioiB4wxJxOYgzoBuE1EHmhhvYNpZYRp8/l0n/l28cKbC9m0cjOC0L9/f0YPHRPRY1K77UWsfc23cacp5pXVr3PzjFtpsDTSNBax2CC1NM9y7LmjWfaXFYGJ7Umwdv+7FHff2fh5823Utq+c9ITbb072sS37Kad9bthXHcnSrs8mS7s+myzt+myyEstKqqMVtH/F88ppPXCziIwi8PT9d4wxo4AfA2+IyHDgjeDvEEiT+h8g7ElpSLU5wnTipAlMPGUiaze8x7trV3Gk+khcrHWc9tlk2bQ0smV1BYF5lu8vLeC2p2/hgdfvJTklifeWRZ6TrX1fOWU52ce27Kec9rl1X3ljYZ+lXZ9NlnZ9NlmelVTHVjwfiCoRkfXBn78AthBIdLqAQPQohESQisg+EVkL1LW0zvaKMPXskxJHn5N9BbClYBvp2X3om5uB3y8kp6ZQeaCy3fVp31dO+9z4HdGuzyZLuz6bLO36bLK067PJSiQrKU3xpe1VVh6IMsYMAsYD7wHZItIQZF9K4LZ/rOUowjTcbd+Z86czZ9YF+Op9DMkfHJO1zoSJ47n+2hvw+/2MGz8+ZtuL1vZpZ8VTX3vsK4CdB4rZV7ifR65/DL/fz/jR48nMzIx4K1vbWNhkHQ/fEe36bLK067PJ0q7PJku7Ppssp/q8ap+K+wNRxpjuwEvA/4rI4dDPglZSEm8N4Wr2uefz/gfvsXbj+/z4ph9TtnNf1J7Gfz1gMEmx/wvCSZ92lk19SUlJZGSnk5GdzqH9h9hbWBK1Jzc3h7z8UdTW1lJbU0tScmy3Y7SPhfZx1z4W2vXZZGnXZ5PVL6cfL7+2kLS0NHbvLlKnzyZLuz6bLKf6bJcbr5zG9eTUGJNC4MT0ryKyMPh2mTEmJ/h5DhD9rPA/FXOEqYkQX9q5UxcGdh/Mu2UreXf/Snqm96RzZVpEcGhk56JXX+KDgvUxz49pbZ92lk19fp+fFx9exPxfzOO2Z26htqY26txRm9ulfV857XMy7trHQrs+myzt+myysrKzKC0t46/P/R9Dhg6mpqaG7KxMNfpssrTrs8lyqs+r9qm4nZyawCn2k8AWEXkw5KN/EIgehSgRpM0reKX1TWKIMJVI8aWHq0AgySRRV1NPbVUdSd0i/4tA+/wYmyyb+pzMHXXK0j4W2sdd+1ho12eTpV2fTVZefh6F2wtZsWwFcy44n/379jPl7LPU6LPJ0q7PJiuR5px2Msbay1bFc87pacAVwGZjzIbgez8B7gNeMMZcA+wCLgUwxvQjEHHVE/AbY/4XGCUih00bI0yveOKXTX4/KSWXSdd+mSk50xG/HzlReHXFGh5/4sXGZYpuP3Y+3bjxY7lw9kWICKmpqWz4YGPUQfBi3VrXd4zV1YFiij/Z28QGaUjmsIhzR9uyXRMmjmf+vOspLSmlV69eCb+vILYo1/YYd+1joV2fTZZ2ffFkPbvt6WP6apNr2LR5MzfMv5EuaZ1Z5V/Jqm0rmyzTPIbXDWORaPpsshIpvtSNFc+n9d8RESMiY0RkXPC1WEQqRGSaiAwXkXNE5PPg8qUSiDntKSK9gz8fDn7WrhGmZUlf8GnVHvIfmMOgX5yLH4lp5mtObg5vrXqTgo3vc8tPfkhJSQmbNm6K2OPFurWtr1+/bIyh0QapurIqbiwIzD279LJLOGf6NDL6pseNoz3K1cm4ax8L7fpssrTrs8na/t4OKiuquPrRK7no9sADgBW7K9Tos8nSrs8mK5HiS42x97JVx2V86ZlDJlF0oISK6oOcNfQU9hwqY2jfgRF7srKzKCvbR7e0bgCU7i0hqVMSEHlvebFuicPKys5i164iivfsZd63r+Heu++NOvcsEcbCi3JNDH02Wdr12WR98u6n+Op9LLhjIfV1PuprfSx+cClX/OZyFfpssrTrs8lKpPhSN9ZxEV+a2y+jyasm+ShfGjSGISfkcPkps6iur6asZn+TZd4qeaPJq6JvGZ/s2MaCguf56uyZ/PH3jzN04iAOZO1rslzz8mLd9LKe3fZ0k9f61AI2bd5EztQslu5aQnHpXg4OPtBkGe1j0fy4bTh2t23fyvQZ53HmaWdxpLaaodMHecetMn02Wdr1xZN11Yirm7yumXsNc+acz9tvreTh3z7EsOHDmPqVaccs58axSDR9NlmJFF/qxqf1NcWXnkXg4abC4CoWisjPwqx3MK2ML21em/du5epv/je9O3Xm8YUPMvzUUazOKYzYk5SUxOTzTuHebz2AiDBs7BDKSz5nb2EJuYNzIvR5sW6Jwtq5fhcZA9J59f7FiAhd0jrTd2CGGn3OWUmcOmNyY5RrUnJSYLK/Gn32jtux48Y0mVOsSZ9NlnZ9tlllpWXMmTmX6urqqN+NjtDnHRf2WYkUX+rG0hRfCvB2yPzUY05Mg9Xm+FKSDIe+msbJ3z2dW3/3U5b+azk1xV9EbGkeh1l1uJp+g7LZsnZbxD4v1i1xWHu2FLNvx3669epKlx5dOFJ5lEX3/EONPqcsL8o1UFnZWWBM45zi7j3SYrYMcttYaNdnkxVqGbRk2esgokqfTZZ2fTZZXnxpx5aa+NJYKmhP1eb4UmMMJrUTM4adyT8/WQn+6P9S3rW1iD5ZvenWsxvJKcmMPWMMH63ZQvaArIh9nsVG4rCGTR7KwDEDuO7Ja5j9g6/SJ6c3/Ue2fFXctj6nLC/KNVCeZVDi6LPJ0q7PJku7PpusRLKSMhb/Z6u0xZd+2RizEdgL/EBEPmq2Kkfxpb+cPe9YTX7DI9+5n08//ZT0fn2489rvNvm8uWVO7ablpJk0br/kZyCQmprK4CGD+M5FTfual2exoZfVfB7Z8p3LKUrbw0s3vUJ1dTVJkkx//8Cw881s6HPS01KUqw1bKG1jcTxZBrnRAs37e2afpV2fTZZnJdWxpSm+dD1wooiMBR4BXo6nLukkjJo6gjFn5HOk8mjUaEa/389nn37Gy68u5N21q+jRozsnDBwQleNZbCQWa+OGjY2395KTktTpc2SPZckWSvtYuNUyCNxpgeb9PbPP0q7PJiuxrKTc90CUmvhSETksIpXBnxcDKcaYvs1W2S7xpQAH9h/k4/e2cMacr5DWq1vUuaOhthIpqSmObC9i7dPO0q7PJku7Ppss7fpCLYOWPLSs0TJIiz6nfQ0WaG+/9Q4XXnxhTPGb2veVTZZ2fTZZ2vXZZDnV51X7lJr4UmNMv2APxpjJQW1NLmsEr7S+SRvjS784WMmLDy1kznWzqa/3U3WoKurcUe22FzZZ2vXZZGnXZ5OlXd/QyUMYOWUE8x6/mlk3zyB9QDq5I3PV6HPal5efx8cffczlV1yO3+ePaS6t9n1lk6Vdn02Wdn02WZ6VlEutpGhlfCmBE87rjTH1wBHgsuDJKG2NL915uKjJ7x8s2ci2DZ+yZe1W/H6hc1pnuud1b7LclGbPwWi3vbDJ0q7PJku7Ppss2/qi2UKFm1O8cPWixjnFCJzcZ2TEOcVt0RevsWg+l3bH2kJ69e/J97/3/UYLtOZzaZtvo/ZjySZLuz6bLO36bLI8K6mOLU3xpY+KSJ6IjBWRL4nI6pB1tWt86RcVX1BfU0/XngHLoKMusQzSbrHhRpZ2fTZZNvU5sYVyq2WQWy3Q3Hjcamdp12eTlUhWUuFiRuP1slXHZXzpsMlD2V9YzsV3zmX3h3tY8Yd/ucIySLvFhhtZ2vXZZNnU12ALVV1Vxfwbvs3Kf6+Meitb+/g57fP+nrlbn02Wdn02WYlkJeXG0pQQ1Qv4CzAwqOsBETnGC8YYMxF4BugKLAa+13D7v6Vqfktr8aeLWbprOa/9eAmVX1RiajolnGVQR7K067PJ0q7PJiue+pzYQrWHJZTTPs8CrWNYpSWlJKckM3fWhYgIAwYOUKVPO0u7PpusRLKSsjkX1FZpSoj6DvBx0ErqLODXxpjUMOv9A3AtMDz4mhFmmYiVnJzM6WeezouLXuAfS16hW7dulJeXR+zRbnthk6Vdn02Wdn02WTb1ObGF0j5+bWG50QLNSV9ycjJTp51Nwcb3WbNuNXV1dd7fdhfps8lKJCspN1bcrpwGjfZLgj9/YYwJTYg6K7jYs8C/CTzkJECP4BP73YHPCZzgNlbQeqqniKwJ/v5nAglRS1qjLbtfNuXlgf8jq6+vp+ZoDRl9+kTsCbWVABzZXsTap52lXZ9NlnZ9Nlk29YXaQtXX+Rptoa74zeUq9GlnadfntM/72+5ufTZZTvV1RHlXTh1WjAlRjwIjCaRDbSZwu97fbFX9CaRCNVRMCVHNKy8/j107dzF39kWcffo06n31fP3yyyL2aLe9sMnSrs8mS7s+myyb+pzYQmkfP5ss7fqc9nl/292tzyYrkayk3FhxfyCqeUJU6Bm+iIgxpmG+6HnABmAqMBRYbox5u3mqlJN6q+SNJr93S07j7wuf51D5YUSE/rn9+bB0Y5Plmkc7are9sMnSrs8mS7s+m6x46msPWyjt42eTpV1frH3N/7YDjD13NMv+siLwQEsSrN3/LsXddzZ+7v1tT1x9NlmJZCXlXTltZbUmIQq4GlgogfoMKARObrbKYgKpUA3lKCGqur6KDyrfZ0eXrRR23YZPfLzxr39F3Bbtthc2Wdr12WRp12eTZVtfa22htI+fTZZ2fU77/D4/7y8t4Lanb+GB1+8lOSWJ95YVqNGnnaVdn01WIllJubHUJEQBRcC0YG82MALYEbrO4HSAw8aYLwXXfyUOE6KqK48A0KtTH3bv2U3n3ikRt0e77YVNlnZ9Nlna9dlkefoSh6Vdn9O+LQXbSM/uQ9/cDPx+ITk1hcoDlWr0aWdp12eTlUhWUp7PaeuqtQlRdwPPGGM2Awb4kYiUAxhjNojIuOByN/AfK6klxPAw1I9ee6LJ7/79R6ldWgICP7/1p2w98D4vVq2G1xp9/1lzbdNbPxXlFUyYOJ7rr70Bv9/PuPHjY7a9aG2fdpZ2fTZZ2vXZZHn6EoelXV+sfc1v0e88UMy+wv08cv1j+P1+xo8eT2Zm5jHL2dKXaCzt+myynOrzqn0qnk/rv0PgJDNcHfOXIpgANb2FdY0L+bkAyG+Ltk6ZXUg5PZO6f+/j9799lMNDDYxPi9rXmC+LwSTF/k8IJ33aWU566urq2LplK5lZmdTX+/D566M3OWQ57XPjvrLJ8vQlDku7Pid9ubk55OWPorh4L36fn6Tk2G7DunEs3KrPJsupPq/aXlae1tdW4hfq/lXGmTfO4q7H7+OLD/fhr4icgho6x23Rqy/xQcH6mOfHtLZPO8upvtz+uQwfcRIvLnqBFxY+z8cfbsEf5V+ibh0LN7I8fYnD0q7PJku7Ppss7fpsspzq64hqPIm28LJVx+XJqb/0KKZ3Khd9aRYrdr1L0oie+LZ785LirS9/dD7Fe4rZs6eYo9VHOXjwIBMmTogLS/tYuJHl6UsclnZ9Nlna9dlkaddnk5VIc07dWB0RX3oJcBcBT9PJwdv0GGMygAXAKcAzInJjC+sNG38aScuaa//W5PfV77zLyfNHkpOTw/Shp/H/Tr+eFW+s4Oxzp7S4Du1RazZZsfaEs3kZPXUUs6bPBiA9uw9pE1MjWni5ZSwSjWUzBtIWy637yo36bLK067PJ0q7PJsuLL+3Y6oj40g+Bi4CVzZY/CtwO/CDKeluKP425qqoqefjRh6jxH6HWfxSfr56Pt34csUd71JpNlmN9Pj9rl6/njudu5Z4Fd1J5qIpNqz6MD0v7WChn2YyBtMVy675yoz6bLO36bLK067PJ8uJLO7asx5eKyHI49kxfRKqAd4wxw6Ks+gLCx5/GXFnZWZSWlgHQiU5UHPicLp1TI/Zoj1qzyXKqb9fWIjL7Z9A3NwOAQaMGsuq1NYw5reXn29w6FtpZNmMgbbHcuq/cqM8mS7s+myzt+myyvPjSjq2OiC9ta7UUf9pifelP32jyuu69X7BmcwGTH7iYxZ+8wxN/e4bHKl5vskzz0h61ZpMVa8+UnGlNXr0OpTM0ZzhTcqZxap/TqKmoo2/XzCbLuHUstLPeKnmjyauibxnbtm9l+ozzOPO0szhSW83Q6YOaLNMenHiybI5forG067PJ0q7PJku7PpssL760Y8t6fGl7rrtZ/GnsmjoZkvJ6UfNsIY8s+wUHux2h06zMiD1JSZ0YO24M8+ddT2lJKb169VIVtWaT5VTf4cNf8OYbbzJp7GQEYeCJA8ntHzkP3fa4t5alfV85ZyVx6ozJjTGQSclJgQcEIpTf78N0Mk3mjsamr/Usp+M3ekw+F86+CBEhNTWV3UW7Y9CnfV+5T59NlnZ9Nlna9dlkJVZ8aUcraP/qiPjStlZL8afN2Y3xpRVv72zymfgF38eHmPKTudz1p/uo+rwyqpVUVnYWGMOll13COdOn0b1HGtlZkU9owYt1C60xY0fj8/t4+bWFrFm7msMHD0XtsznuTlja95XTPicxkLn9c8nLz2ucO1pdVR3VKswpy+n+NZ068daqNynY+D7//a0rKd9fzqaNm9qdpf240K7PJku7Ppss7fpssrz40o6tjogvbWu1FH/apELjSzPOGNTkMydWUnn5eRRuL2TFshXMueB89u/bz5Szz4oq1rPYaFufzXF3wtK+r5z2OYmBdGIV5pTldP/uLtrN5wcOUFdbx9J/LqNbWjdazgpxztJ+XGjXZ5OlXZ9NlnZ9NlmJZCXlRp/Tjogv7Qw8AmQCr5tANOl5AMaYnUBPINUYMxeYLiIfG2OeAP4YtJ1qKf60xWpuJbV86XIW7lnEb75zDz6fj7PzxpGens5Prv3Pg//Pbnv6mPXUJtewafNmbph/I13SOrPKv5JV25qaDlw14uomv9u05NFusRFLn5Nxbz7mNlna9xXEdgyGi4Es/mQvN8+4tXEKxpDMYU2Waw+rsFhZTsYinL4x5+Qz56tz8Pn8mE6GqZdM4UDWvna3M9N+XGjXZ5OlXZ9NlnZ9NlmJZCXlxorblVMReUdEjIiMEZFxwddiEVkkIieISGcRyW44MQ32DBKRdBHpHlzm4+D78xr8UEWkQkSmichwETlHRD5vrTa/38/GDRv53WOPsOjVl9i8YVNUi4jt7+2gsqKKqx+9kotuvwBfvY+K3RVRWTYtebRbbDjpczrutlja9xU4Owb79cvGGBqnYFRXVkXX58AqzDHLyfg1TB945kc8+M/7yB6YxacbtrO3sCRyn/LviBv12WRp12eTpV2fTVYiWUl5V05dUk4sIj5591N89T4W3LGQ+jof9bU+Fj+4lCt+c3nEPpuWPNotNmyOuy2W9n0F9o5BJ1ZhTlntoW/StAl88sGnbFm7jdzBOR2uzyZLuz6bLO36bLK067PJSiQrKTfWcRlf6sQiYujkIYycMoJ5j1/NrJtnkD4gndyRkZ80h8A8t107dzF39kWcffo06n31fP3yy9pdn9M+7RYgTsfdFkv7vgJ7x2BpURnde/cAoLamjgNlB2P6l7atsThYfoi0XmlUVx4BoEfv7pTuLCN7QJYKfTZZ2vXZZGnXZ5OlXZ9NViJZSXlXTltRrY0vDekbCHwM3CUiD4RZ72DgeSADWAdcISK1rdEWi0VE87mFy3cuZ+HqRbx00ytUV1eDwMl9Rh6zXLh5bmPPHd1ok0MSrN3/LsXddzZ+3nyOm1stNuI57h3F0ravwh1/EJ9jsHnPZ75dvPDmQjat3Iwg9O/fn9FDx0ScOxrPsWjOre3jZ4d/F3d9/ef4/X58Ph/pGel856Lvdoi+jmRp12eTpV2fTZZ2fTZZiWQl5cbSFF/aUA8CSyKs95fAb0RkGHAAuKa1wpzaSqxbt57fP/YoS5a9DiIx2UrYsslpy3ZptwBxMu62WNr3Fdg7BidOmsDEUyaydsN7vLt2FUeqj8RkJWVrLLKys6ipqeXNt99g7Yb3mH/DdXQynTwrKYX6bLK067PJ0q7PJiuRrKSMsfeyVfF8IKpERNYHf/4CaIgv3SIi28L1BJ/QLwQ+auFzA0wFFgTfehaY60CbNVsJWzY5TvvcaAFik6VdH9g7Bp1aSdkaC89KKnH02WRp12eTpV2fTVYiWUm5saw8EBVLfGkwSepHwLnAD1pYLAM4KCL1wd/3AP1bqyee9k5OLHmcstqjz40WIDZZ2vR1tFWTEyupeI2FZyWVuPpssrTrs8nSrs8my7OS6tiK+wNRrYgvvYvA7frIl3TaoWzaO9myyXHa50YLEJss7frArlWTEyspa/vKs5JKGH02Wdr12WRp12eT5VlJufSBKGh1fOmpwNeMMfcDvQG/MeaoiDwaskwF0NsYkxy8enoCUNwC+zrgOoBH//AI11z7rcbP3GjvZJOlXZ9NlnZ9NlmelVTbemyytOuzydKuzyZLuz6bLM9KqmNLTXypiJwRNOEfBPwWuLfZiSkiIsCbwNeCb11FDPGloSem4E57J5ss7fpssrTrs8nyrKTa1mOTpV2fTZZ2fTZZ2vXZZCWSlZQbn4hSFV/aUhljFgPzRGQvgXmpzxtj7gE+IHACHLGazz2rq62jztRSVlSGiCCdJOHtnWyytOuzydKuL56sY+ZXm2JeWf16k7mt2qyk1lds5PZLfgYCqampDB4yKC5WUn6/D9PJNJnXrum40HYsdSRLuz6bLO36bLI8K6mOLVXxpSG9TTxORWRm8MQUEdkhIpNFZJiIXCIiNa3VlpySzC1/vInfLP0lv178Czp3TWXVa2si9mi3vbDJ0q7PJku7PpusMWNH4/P7Gue2Hj54SJW+vpl92bp1Gy+/upB3166ia9cujB03Ni76cvvnkpef1zivvbqqWpWtlvZjySZLuz6bLO36bLISy0rKfXNOj8uEqMpDVfh8fgCOVB3lSOVRMrIjzznVbnthk6Vdn02Wdn02WZ6+/5R2Wy3t+8omS7s+myzt+myyPCupji0rVlLa6nDFYZ775d/Yt3s/vjofg/JOZOqlZ0XsqSivYMLE8Vx/7Q34/X7GjR8fs+1Fa/u0s7Trs8nSrs8m63jWF862aub86cyZdQG+eh9D8gfHbKuV6GORaCzt+myytOuzyXKqryPKjefP8XwgaoAx5k1jzMfGmI+MMd8Lvn9J8He/MWZSyPLfNMZsCHn5jTHjwqw33Riz3BjzafC/kS95hqn+Q3OZM28W6Vl9SM/uw+Hyw1HtZILswAuDSYr9aHDSp52lXZ9NlnZ9Nlmevv9UUlISGdnpZGSnc2j/oZj+xjhlaR8L7Szt+myytOtz2tcvpx8vv7aQtLQ0du8uiqs+r9peauJLReSvDXNTCTxIVSgiG8Ks98fAGyIyHHgj+Huryu/z8+LDi5j/i3nc9swt1NbUxhTn2BBtuejVl/igYH3M82Na26edpV2fTZZ2fTZZnr7/lJO/MTa3S/u+ssnSrs8mS7s+p31Z2VmUlpbx1+f+jyFDB1NTU0N2VmZc9HVEeXNOW1HiIL40pL4BPN/CZxcQiC0Fh/GlbowUtcnSrs8mS7s+myxP33/Kyd8YpyztY6GdpV2fTZZ2fU778vLzKNxeyIplK5hzwfns37efKWefFRd9XrVPqYkvbVZfJ3ASGq6yRaTh/lgpkB1tZcdDpKhNlnZ9Nlna9dlkJYK+CRPHM3/e9ZSWlNKrV6920xfub8y+wv08PP+PFO0qokfPHlH/xrRlu9y2r2yytOuzydKuL9a+Z7c9fUxfbXINmzZv5ob5N9IlrTOr/CtZte0/N3CvGnF1u+jriLJ5RdNWaYovbVj+VKBaRKLmHgZN+aW1mtwYKWqTpV2fTZZ2fTZZ2vVBYN7ZpZddwjnTp5HRNz3q8k5Zubk5TD3nbC64cA7Tzp1K374ZcWO5dV95Y2GfpV2f077t7+2gsqKKqx+9kotuDzykWLG7Ii76vGqfiuvJqWldfGlDXQb8LcLnZcaYnOD6c4B9LbCvM8YUGGMKnvzTU00+C40lS0lNaXWUWaw9bmVp12eTpV2fTZZ2fVnZWezaVcTbb73DhRdfGNO8s0RguXFf2WRp12eTpV2f075P3v0UX72PBXcsZMlDy6iv9bH4waVx0dcR5c05bUWZwFbEHF8a7OkEXErL800B/kEgthQcxpe6MWrNJku7Ppss7fpssrTpe3bb001e61ML2LR5EzlTs1i6awnFpXs5OPjAMcvZYtkci/bocStLuz6bLO36Yu27asTVTV7XzL2GOXPO5+23VvLwbx9i2PBhTP3KtCbLtJc+r9qntMWXngnsFpEdoSsyxjwB/FFECoD7gBeMMdcAuwiczLaqnEQLao9as8nSrs8mS7s+myztkZ071+8iY0A6r96/GBGhS1pn+g6MfrtdO8uNx5JNlnZ9Nlna9bWlr6y0jDkz51JdXR142ClOnI4oF0451RVfKiL/FpEvhVnXvOCJKSJSISLTRGS4iJwjIp+3VpuTaEHtUWs2Wdr12WRp12eTpT2yc8+WYvbt2E+3Xl3p0qMLRyqPsuief8RFn02WG48lmyzt+myytOtrC6vBFmrJstdBxFXxpW6s4zIhKjRasFePnjFFC7rVYsON+myytOuzyXLyvbKpb9jkoewvLOfiO+ey+8M9rPjDv+g/MifhWW48lmyytOuzydKuzybLs5Lq2IrbyakxZgDwZwJWTwI8LiIPGWN+BZwP1ALbgatF5GCw51bgGsAH/I+IHDNj2RgzmMCc1AxgHXCFiNRG0hIuWvD2X98GdYYD5Yd4+PcPUzeoKmK0YKJYbIyfMI65sy4kKSmJ08843bMb8axX4sYK970aPXUUs6bPBiA9u0/MkZ3x0Nd8HtnyncspStvDSze9QnV1NUmSTH//wLDzzTSzOqrHrSzt+myytOuzyfKspDq2OiIhajmQLyJjgE+AWwGCn10G5AEzgN8bY8JdQ/8l8BsRGQYcIHAy26rq3KkLp+Z/mcIu2/jIv4GkpCS6VKVF7EkEiw2Aysoqzpk+jVNOPYWzpk1Rpc+NLO36bLL8Pj9rl6/njudu5Z4Fd1J5qIpNq6I6wlndVxs3bGy8tZeclBTX8bPJctuxZJOlXZ9NlnZ9NlmelVTHVkckRC0TkfrgYmuAE4I/XwA8LyI1IlIIfAZMDl1n0AFgKrAg+NazOEiIAjAYkkwSad27kd63D1s/jBxalQgWG8nJyRQVFXHhxRcCsK+0LKp9jXYLEO0s7fpssnZtLSKzfwZ9czPo2acHg0YNZNVra9To0z5+Nlna9dlkaddnk6Vdn02WZyXlUiup0IqQEPUtYEnw5/7A7pDP9gTfC60M4GDIyW24ZaJW+eflfFqxlTNzzmFKzrnsK9nHYf/BiD2JYLHx6iuv0blzZyoqKhC/n38uWRo1ok2bBUiisbTrs8kqLSqje+8eANTW1HGg7GBMf8y8fWWfpV2fTZZ2fTZZ2vXZZHlWUh1bcX8gqqWEKGPMbQRu/f813hqueOKXTX6f2H04C26Yw47CQnw+H2ecdiblH9ZxxRN3Ny5TdHvTuXHaLDaa+yTuWFtIefJ+hk0byk9+dBviF04cOzBqRJt2CxDtLO364sk6JrLTFPPK6tebxAKPHjomamSnt6/ss7Trs8nSrs8mS7s+m6zEspLy5py2qlpKiDLG/DcwG/im/MdwrBgYENJ+QvC90KoAehtjkiMs08BoTIg6sq60yWe5A0/g5a1vcvpzV3P2366juu4oQmTfM+0WG8Vb9/LZ+9tZ8/f36JzWmaSUJEo/K4sa0ZYIFiCaWdr12WSNGTsan9/XGAt8+OAhVfq0j59NlnZ9Nlna9dlkaddnk+VZSXVsWU+IMsbMAG4B5ohIdUjLP4DLjDGdg0/kDwfeD11n8ET2TeBrwbeuIoaEqK4T+zX5rPhQGRNPyKNrSmfOHDKR2vo6PijeEnF7tNtenHHFaZz/w1kMHDOA82+ZycAxAzjlwglsf39HxD43WoDYZGnXZ5Pl6UsclnZ9Nlna9dlkaddnk5VIVlJunHPaEQlRDxNIiVoe3NA1IjJfRD4yxrwAfEzgdv93RMQHYIxZDMwTkb3Aj4DnjTH3AB8QOAGOWLn9miazlPrKeKvoPf59w7NkdOtN0aG9rNjzzjHLhVZFeQUTJo7n+mtvwO/3M278+JhtL1rbF0tPOEua5TuXUzO0jvMGfJXS7vs5L386mzd9GNG+Jl76jheWdn02WZ6+xGFp12eTpV2fTZZ2fTZZTvV51T4Vt5NTEXkHCHeavThCz8+Bn4d5f2bIzzto9hR/q7XV+/nN93/Bb/1wYo8cynNrqT+lR9S+xn89YDBJsf8LwkmfU1ZZaRl333UPfp+ft/69kvT0dFX63Mhyqq9fTj9+99ijfOOSb7J7dxEZmdHjLd06Fp4++yzt+myytOuzydKuzybLqT7blQAXd1tdVp7WV1dJhs5fG8h5/+9Sbvv93VR+VoG/5EjEltD4s0WvvsQHBetjnh/T2j7tLO36bLKc6svKzqK0tIy/Pvd/DBk6mJqamqi2X24dC0+ffZZ2fTZZ2vXZZGnXZ5PlVJ9X7VPHZXypMQZSDTOGnck/P1kJ/sgPQ4H++TE2Wdr12WQ51ZeXn0fh9kKqq6qYf8O3WfnvlVFtv9w6Fp4++yzt+myytOuzydKuzyYr0eacuq3UxJcaYyYDjze0A3eJyKIw6x1MK+NLfzl7XpPf62rreOh/f88tv/8OtbW1DBo1kJuu/W7E7bEZtVZaUkpySjJzZ12IiDBg4ABXxLq5kRVrT3PrL4Da5Bo2bd7MDfNvpEta56i2X24ZC09fx7O067PJ0q7PJku7PpusRIovdWOpiS8FPgQmicg4AvGlj4VYRoVWm+NLk1OS+e6v5/OrxT/nngV3srewlLUr1kXssRm1lpyczNRpZ1Ow8X3WrFtNXV0d5eXlcWG5MXbOJsupvu3v7aCyooqrH72Si26/AF+9L6rtl1vHwtNnn6Vdn02Wdn02Wdr12WQlUnyp97R+K0pESoCS4M9fGGMa40tDFltD0Baqma1UFwJXW5uUCYzMVODy4FvPAncBf2iNNmMMnbt2BqBzl1Q6d0ll55YiTjlnYos9oVFmgKOotVj7svtlU14eOFGpr6+n5mgNGX36xIVlq8etLKf6Pnn3U3z1PhbcsZD6Oh/1tT4WP7iUK35zeYs9bh0LT599lnZ9Nlna9dlkaddnk+VUn1ftU5riSzHGnGqM+QjYDMwPiSltqHaJL/3iYCWVh6r45XUP8pOL78J06sTISSMi9sQzau2tkjeavCr6lrFt+1amzziPM087iyO11QydPqjJMu2l0Y2xczZZsfZcNeLqJq9r5l7DnDnn8/ZbK3n4tw8xbPgwpn5lWpNl3DoWnr6OZ2nXZ5OlXZ9NlnZ9NlmJFF/qXTl1UKYV8aUi8h6QZ4wZCTxrjFkiIkfbW9PhisM8ceczHCw/BCLUHq0lvV/kK5N2o9aSOHXGZJb9ZQUiQlJyEv8J0mpvlvti52yy2hJxV1ZaxpyZc6muro7b/rW5XZ6+xGFp12eT1RZ9Y8eNYf686yktKaVXr17H9Vi4jZVI8aVuLE3xpY0lIluASiC/2UeO4ksX//WfTT7LGdQPMNz21C386rV7SU5J4r1lBRG3xWbUmt/n5/2lBdz29C088Hps+pyy3Bg7Z5PVFn0NNiVLlr0OIsf1WHj67LK067PJcqovKzsLjOHSyy7hnOnT6N4jLSY7ODeOhRtZXnxpx5aa+FJjzOCGk05jzInAycDO0HU6jS+d+c0ZTT7bUrCN9Ow+9M3NwO8XklNTqDxQGXF7bNpeONHnlOVGCxCbLO36bLI8fYnD0q7PJqutdnArlq1gzgXns3/ffs8OzkWsxLKSsveyVWriS4HTgR8bY+oAP3CDiJQDmDbGl07Jmdbk950Hiin+ZC83z7gVQRh44kCGZA47ZrnQiqftRXvoc6rRjRYgbWFNmDi+yW06bfo0szx9icPS/r2yuV2eHVzi6LPJ8qykOrbiduVURN4RESMiY0RkXPC1WESGiciAkPfmB5d/TkTygu9NEJGXQ9Y1M3hiiojsEJHJwfVcIiI1rdXWr182xsDLry1kzdrVVFdWqbK9cKLPKcuNFiBt6euX06/xNl1G3+jRr24dC0+fu1nav1c2t8uzg0scfTZZnpVUxz4QdVzGl4ZaRKSkprTaViLWHreytOtz2peVncWuXUW8/dY7XHjxhTFFirp1LDx97mZp/17Z3C6nYxFqB7fkoWWNdnDxYGkfCzeynOrzqn3quDw5daPthU2Wdn2x9j277ekmr/WpBWzavImcqVks3bWE4tK9HBx8oMkybh0LT9/xxdL+vdI4Fp4d3PHFSiQrKTdOOlUTXxrsGQM8BvQkMO/0lOZWUsaYdODvwCACD0xdKiIHoulJNqkkmSQEabQAueX7P+K2n9zGiy8u4IsvvojYr932wiZLuz6nfTvX7yJjQDqv3r8YEaFLWmf6DsxQo087y9OXOCyn+vx+H6aTaRKtHI/vlc3t8uzgEkefTZZnJdWxpSa+NPik/l8ImO/nAWcBdWHW+2PgDREZDrwR/D1q+aSeWn/gPLfBAuSFF//OmPGjefGlv7NvfxnJJqXFfu22FzZZ2vU57duzpZh9O/bTrVdXuvTowpHKoyy65x9q9GlnefoSh+VUX27/XPLy8xqjlaurqvFHeUjEyffK5nZ5dnCJo88mK5GspLw5p60oESkRkfXBn78AGuNLQxKe1hDwKgWYDmwSkY3BngoRCfdX7wICsaUE/zs3Jj34G3/Oy8+jaFcRfp9QU3OUfy5ZSp+MPnQyLR942m0vbLK063PaN2zyUAaOGcB1T17D7B98lT45vek/MkeNPu0sT1/isJzqyx+dT/GeYvbsKeZo9VEOHjzIhIkTIvY4+V451ejGfWWTpV2fTVYiWUm5seKeEAUQJb7078GfTwLEGLMUyASeF5H7w6wuW0RKgj+XEpg2ELG6zjgJgBOzT+C1u59h9HXnkFaTRFVlFZO/dCr1ycLIG87lwKKP6HNhHgBH/vlJk3Vot72wydJuDRNrX/M5Yct3LqcobQ8v3fQK1dXVJEky/f0Dw84ds6Ev0VievsRhxdoTLip59NRRzJo+G4D07D6kTUxtslx7fK/ivV1t7XErS7s+m6xEspLq5MLz57g/ENWK+NJkAl6n3wz+90JjTERjz6Apf/SJPmGqqquPv/39b1R281Gb7G9YX4vLa7e9sM3SbA3TFtbGDRsbb9MlJyWp06eZ5elLHJZjfT4/a5ev547nbuWeBXdSeaiKTas+jMpq7ffK5nZp31c2Wdr12WQlkpWUGyuuV05jiC+dJv85I9wDrGxmvD+BwLzS0CozxuSISIkxJgfY1wL7OuA6AE7uDSekNfk8ud7w5r//TVJOd6S4kn7pWew72LJHXaitBODI9iLWPu2sBmuY4j17mffta7j37ntbbQ0TT302Wdr12WR5+hKH5VTfrq1FZPbPoG9u4IGmQaMGsuq1NYw5rXnSdNtZ2sfCjSzt+myynOrriLI5F9RWqYkvBZYCo40x3YIPR00BPg6z6n8QiC2FGONLm5+YAmQeSGXIqXmcN2lKYEXnXsIrq5e1uD3abS9ssvLy8/j4o4+5/IrL8fv8McX22dRnk6Vdn02Wpy9xWE71lRaV0b13DwBqa+o4UHYw6v8xunUs3MjSrs8mK6GspFxYauJLReSAMeZBYC2BW/WLReR1AGPME8AfRaQAuA94wRhzDbALuDSakD8+dCvT+p9LTrf+dEnuwuu3PM5Tf3mSla8uwdR2Iq/fUG687L9YsWcpf3zo1rDr0G57EU9Wcx/CHWsL6dW/J9//3vcbrWGax/bBsXPPEmEsxo4b02QurTZ9mlmevvZhjRk7mrmzLiQpKYnTzzi9Q8fimGhlU8wrq19vEq08euiYiNHK2sdduz6bLO36bLISyUqqk3flNPZqbXxpsOcvwQjTfBG5JeT9ecET04an+KeJyHAROUdEPo9FzxvFy/nLp8/wxJY/8tc3nuPD9z+i9kgt1bXV7Nq1i3nXzaPG33ISqnbbC5ss7dYwTvsaLMYa5tJ275EWdbqC9n1lk+XpazsrKzuLkpJSzpk+jVNOPYVReSfH5Rh0qm/M2NH4/L7GaOXDBw8l/Lhr12eTpV2fTVYiWUm5sY7LhCjPMqhtPdqtYZz25eXnUbi9kBXLVjDngvNjmq6gfV/ZZHn62s7q27cvpSWlnHHmGYjfzz+XLI3LMZgIY+EdF/ZZ2vXZZCWSlZQbfU6tWEl1dIWzNqkZWsdVI65m7aECNvb8MCbLoAkTx3P9tTfg9/sZN358zLYXre3TxmqP8YunPqd94WITO/U2bN+zgx/86Iek9ko9ZrpC823Utq86kuXpa11PuOPv1ftf58v/dSo//+W91FTWcOrXTonLMahtLDqSpV2fTZZ2fTZZTvV51T4VzweiBhhj3jTGfGyM+cgY873g+78yxmw1xmwyxiwyxvQOvp9qjHnaGLPZGLPRGHNWC+tNN8YsN8Z8GvxvHyf6GmLnfnr7zzhx0ImxblPghcEkxf4vCCd9TlnZ/bLp2rUrg4cM5qxpU+LGcjJ+Tlm2xn3H2kIyB/Vl3h//m5TOKRytanmaR0foSwSWp895z461hXTr1Y0+ub0DVygMmBgNDG2ORb+cfrz82kLS0tLYvbsorizvuLDP0q7PJsupPq/aXmriS4FrAURkNHAu8GtjTDh9juJLQys0dm7Rqy/xQcH6mOaftLbHNisrO4sP1m9gyNDBAOwrLYtpvpobx8JJX/HWvWxfu4M/XPUnvqio5Ojhoyz+zVI1+rSzPH1t6yneupfP3t/Oyz9/lfraesQvrF20jordLVvc2dQHgb8xpaVl/PW5/2PI0MHU1NTE5W+Mze3Srs8mS7s+myyn+jqiOll82SpN8aWjgH8Fl98HHAQmhVn1BTiIL22mzXXzY8Cbr9bWvjOuOI1v3HcpWUMyOf2/vkyXnl2YedN5avRpZ3n62tZzxhWncf4PZzFwzADOv2UmA8cM4JQLJ7D9/R0q9IE752Vr12eTpV2fTVYizTnVVsaY3saYBcG75FuMMV9u7To0xZduBOYYY/4GDAAmBv/7frOeVseXNi8n8ZvaotZamq920lnDue3W/4f4hRPHDoxpvprbYufa0vf6r//JFxWVvLeggPra+qjLJ8pYjJ8wrok9kaa4P+1xhPFihZuj3RD1ufJ3qzh08BDDMoeTnp7e7hG6FeUVjBs/lgtnX4SIkJqayoYPNh6zXLi/M7XJNWzavJkb5t8Y1kauPf7GtGW7Ev246EiWdn02WYkVX6ruBPoh4J8i8jVjTCrQrbUr0BRf+hSBlKgC4LfAaiDikdCW+NLWxm9qj1rbsbaQrj27UrShiBnfm86g8SdSdag66i1BN8bOOe3b/t4OyndVcMnPLmLW92fQqZOJy/g57WtLnF5lZVWjPVEsc5G17ys36mvo27hhI089+wTvrFnJ5g2b4sbKyc3hrVVvUrDxfW75yQ8pKSlh08ZNEXu2v7eDyooqrn70Si66/QJ89b6E/45o12eTpV2fTZYXX+qsjDG9gDMJhDAhIrUicrC164nryamJHl/6zYb4UhGpF5Gbgt6nFwC9CcxJbV5lJhBbiokSX2qMKTDGFDz5p6eafNYQv/n2W+9w4cUXxjRvKjTKLCU1xVHUWqx9TnqKt+7l03c/o/LzKt56+m32fFSMgai3BG3pSwTWJ+9+iq/ex4I7FrLkoWXU1/pY/GDkOaeJMBbJyckUFRVx4cUXArHNRda+r9yozyYrKzuLsrJ9dEsLXNAo3VtCUqckIPIVGDd+R7Trs8nSrs8my6m+jqj2somK5RVDDQb2A08bYz4wxjxhjDk2pjNKxe22vglsRaT40ikSEl9qjOkGGBGpMsacC9SLSKT40vuIEl8KPA7w2Me/k9DbU/5UP5s2b2LW92ewdNcSikv3cnDwgSa3sJrfmgqNMsvOyqJwRyHjJoyPOg5O+mLpaa7vqnuuZvkZy1n1zmpmnT+LZ5/+M+d9dTqbN30Y8ZZgvPQlAqv5uOTOPYFVfVdz1913svb9Au6/71eMnzC+3cfPaV8sPeFuw/7jr6/jT/Lx8uaF7P6iiC0vb2HmTee1+/GubSwSTV88WW+VvNHkd19fH5/s2MaCgud58q4/U1JYyqnnTeJA1r4my7rxO5Jo+myytOuzyXKqz+1ljLkOuC7krceD51sNlQxMAL4rIu8ZYx4i8OD67a3hqIkvBbKApcYYP1Ac7AXAtDG+tHntXL+LjAHpvHr/4sb4zb4DMyL2aI9aa6iy0jJu+/H/49DBQ3Tp2oX09MhTFtwYO9eWvgaLrOrqaoIX9dXoc9KzY20hab27Mfq80Sx5aFnjXOR4HO/ax0K7PpuspKQkJp93Cvd+6wFEhGFjh1Be8jl7C0vIHRw5UMNt3xHt+pz2+f0+TCfD3FkXIiIMGDggLhHOTvVpH3cvvjR8hV74a6H2AHtEpOEZowU4cFVSE18qIjtFZISIjJRALOmukHW1Ob40tJzEb2qPWmvoW7dufeN8tVjtMtwWO9cWVoN1yJJlr4OIOn2t7WmwJ1rz9/fonNaZpJQkSj8ri8meSPu+cps+myy/z8/7Swu47elbeOD1e6k6XE2/QdlsWbstKstt3xHt+pz25fbPJS8/j4KN77Nm3Wqqq6rxR3mgx0mEs1N92sfdiy91ViJSCuw2xowIvjUNCHcXPGJ58aVefKnr9NlkadeXCPZE3r6yz9q1tYg+Wb3p1rMbySnJjD1jDB+t2UL2gCwV+myytOtz2pc/Op/iPcXs2VPM0eqjHDx4kAkTJ0TscWIV5lSf9nFPJCspZXNOAb4L/NUYswkYB9zb2m2K55zTAcCfCVg9CYF5CQ8ZY+4m4FXqJ/Aw03+LyF5jzDeBHxGYkf8FcL2IbAyz3sHA80AGsA64QkRqI2lpKX7zv4ZdyQXfu5DaQ3UxxZdqtr2wydKuzyZLm76OtCcqLSklOSW5yW1Eb1/pYE3Jmdbk99pNy0kzadx+yc9AIDU1lcFDBvGdi77bIfo6kqVdX6x9zecVA4yeOopZ02cDkJ7dh7SJqU2W23n42ISvaFZhcHxYEiaSlZS2EpENhPepj7k6IiHqVw23+oHXgDuCyxcSeEhqNHA3Lc9p+CXwGxEZBhwArnEq8K/P/R/5o/M4YUD/qMtqt72wydKuzyZLu76GPhv2RMnJyUyddnbjbcS6ujrKy8tj0uftK7ssv9/PZ59+xsuvLuTdtavo0aM7JwwcoEafTZZ2fY5ZPj9rl6/njudu5Z4Fd1J5qIpNqz6M2OPEKsyxPuXjnkhWUl5CVCtKWk6IOhyyWBqBq6qIyGoRORB8PzQ5qrGCDgBTCUywBYcJUW60krLJ0q7PJku7Ppus7H7ZlJcH/o+svr6emqM1ZPTpo0af9vGzydKuzyZLuz6nfbu2FpHZP4O+uRn07NODQaMGsuq1NRF7nFiFOdWnfdwTyUrKjWXlRLh5QpQx5ufGmN3AN/nPldPQugZYEub9DOCg/Cf+dA8Q/bJns8rLz+Pjjz7m8isux+/zxzSvJtRWoq62jsIdhfTq3Tsqy0mfdpZ2fTZZ2vXZZOXl57Fr5y7mzr6Is0+fRr2vnq9ffpkafdrHzyZLuz6bLO36nPaVFpXRvXcPAGpr6jhQdjDqnMGhk4cwcsoI5j1+NbNunkH6gHRyR+bGRZ/2cXeqryOqkzHWXrYq7vGl4RKiROQ24DZjzK3AjcCdIcufTeDk9PT20tDc+3HH2kJ69e/J97/3/UYrqWgRfNptL8C5dYjbLEBssrTriycr3By3seeOZtlfVgQeJkiCtfvfpbj7zsbPm899jKe+9uixzfK+w/ZZ2vXF2tf8u7XTFPPK6te5ecatCMLAEwcyeuiYJstNafYc8PKdy1m4ehEv3fQK1dXVIHByn5ER56drHIuO1OdV+1SHJESF1F+Bi0OWHwM8AVwgIuEmulQAvY0xDSfVJxDwRA3HbkyIevuFd5p85lYrKSfWIW60ALHJ0q7PJqu5PVFyShLvLStQo0/7+IH3He4IlnZ9TvvGjB2Nz+/j5dcWsmbtag4fPBQXOz2n+rSPeyJZSSl8Wr/NFbeT0wgJUcNDFrsA2Bp8fyCwkMDT9+FiSwlGnb4JfC341lVESIgSkUkiMumMS5tehHWrlZQT6xA3WoDYZGnXZ5O1pWAb6dl96Jubgd8vJKemUHmgUo0+7eMH3ne4I1ja9dlkaddnk5VIVlJurI5IiLomaM7qJ5DwND/42R0E5pT+Pnh2Xi8ikwCMMYuBeSKyl4Dd1PPGmHuADwicAEes5rckFn+6mKW7lvPaj5dQ+UUlpqZTTFZSM2d9FVPfibLi/fz97y/QN7MvSaYTvsYpsOH74mFhEe6WKkS3Dml+68eJ/U9FeQUTJo5vkiKiyQLEJku7vniyjrmNeKCY4k/2NrmNOCRzWNhb+Tb0tUdPPFnt9R22PRat/e5rG/dE02eTpV2fTZZnJdWx1REJUReLSH7w/fNFpDi4/DwR6ROy7KSQdc0MnpgiIjtEZLIEkqYuEZGa1mpLTk7m9DNP58VFL/CPJa/QrVu3mCxvMNB/UA4nDM7l3bXvUldXh08iH6xWLTYcWIc4tf/pl9OvMUUko2/kiNS2bJdb7UbcyOrXLxtjaLyNWF1ZpUqf9vEDZ99h2zY5rf3uax937fpssrTrs8lKKCspFz4QdVwmRDmxvMnKzqK0tKzx987JKVR+8QWB6/4tl00LCyfWIU7HorVWXE63y612I25kefraznLyHbapz402fNr12WRp12eT5VlJdWwdlyenTixv8vLzKNpV1Ggr0bVrGhLDPyJsWlg4sQ5xOhatteJyul1utRtxI8vT13aWk++wTX1utOHTrs8mS7s+m6xEspIyFl+2SlN86VkEHm4qDK5ioYj8LMx6B9PK+NJwc7u+9r9zGdVrDIMHDQbgwwMbKC7Z2fj5j1574pieugE1zJoxGzDccMP1/KL4OT4/0vQy/5pr/9bk93hZWISbxxeLdYgT+5/mEXexWHFB+9hxudVuxI0sT1/repx+h23pg+PDhk+7Ppss7fpssjwrqY4tTfGlAG+HzDk95sQ0WO0SXzp32kVkjk5nW9JmHvnbb3l/1dqIy4tf8H18iM5XDeG8X13Oq4tfo3zPvqgcmxYWTqxDnNj/OLHicrpdbrUbcSPL09d2llP7H1v63GjDp12fTZZ2fTZZiWQl5c05bUW1Nr40lmqv+NIjh2vondKH4uoiamvq+HjtVnpl9YzY4y89iumdSqfeqcw8+Wz6jMnFt12PTY7TPif2P06suJzqc6vdiBtZnr7EYTnV50YbPu36bLK067PJ8qykOrbinhAF4eNLgSuBQ8DZIYt+2RizEdgL/EBEPmq2KkfxpY+se7nJ7yfU9mWoOZmjO4WTho7i+pu/x7Of/4M31/3HXrX57fnlS5ezqn41d117J507deWLL5WwedOH/OTaH0dkN1ivXH/tDfj9fsaNHx+zVVNremLtC2f/s69wP49c/xh+v5/xo8eTmZkZNUWkZmgdV424mrWHCtjY88OoVlwax+J40GeT5elLHFasPc2/006++24Zi+OBpV2fTZZTfR1RNq9o2ipN8aXrgRNFpNIYMxN4GRgefq1tqy5ZaVTsK+env7ib2vpazjn/PC65aCZ/2fpyxL6y0jLmzJyL3+fn5LyTSU+PbqNSV1fH1i1byczKpL7eh8/fsidqaDUmMmAwSbEfeK3ty83NIS9/FMXFe/H7/CQlx3bbovlYxEuf0/FzwrLZ41aWpy9xWE71Ofnuu3Us3MjSrs8my6k+r9peauJLReSwiFQGf14MpBhj+jZb3lF8adHSrU0+21/1OT/92U/J/O+TOPHmCaxasZIuByIPRWis26JXX+KDgvUxxxEOH3ESLy56gRcWPs/HH26JKY7QCctJn60ep31Oxs/mdmnfVzZZnr7EYWnXZ5OlXZ9NlnZ9NllO9XVEefGlrSgH8aX9gj0YYyYHtVWErlPEWXzpwPOa/uu+9LNisnKzGXTiiZjkTkw6+1TeeONfEbdHexyh0z7tc4WcjJ9Tlvax0M7y9CUOS7s+myzt+myytOuzyfLmnHZsaYov/RpwvTGmHjgCXBY8GaWt8aULZj/S5PflS5ezcMPL7Lx/HX6/n9RRI+mVnsavZt/d4jpijTILZ9VkI47QaZ+22Ln2GL+2bJfbohltsjx9icPSrs8mS7s+myzt+myyEim+1I1zTjXFlz4qInkiMlZEviQiq0PW1a7xpX6/n40bNvC7xx5h4T8WsGHDhoSPI3Tapz52zsH4tUWj26IZbbI8fYnD0q7PJku7Ppss7fpsshIpvtSNZeVpfW0VGksGtDrKLNYeaBpHCDTGEY45Lb/dWba2y6Y+J+PnlNUQzVi8Zy/zvn0N9959b6uiGWPlOO3TzvL0JQ5Luz6bLO36bLK067PJcqqvI8p9102P0/jSeEaZTcmZ1uTV61A6Q3OGMyVnGqf2OY2aijr6ds1sskx76Iv3dtnS1x7jFyvr2W1PN3mtTy1g0+ZN5EzNYumuJRSX7uXg4ANNlrE5FonG8vQlDku7Ppss7fpssrTrs8lKpPhSN5am+NI+wFPAUOAo8C0ROeb+rZP40uZlM8rs8OEvePONN5k0djINcYS5/XPjwtIe6+akz8n4OWXtXL+LjAHpvHr/4sZoxr4DM9qd09D35S9/ib1FJaSazvz97y/w6uuvxo2l+bjw+32YToa5sy5ERBgwcIAqfW5k2dY3dtyYJnO5j+ex0MzSrs8mK5HiS705p62r1saX/gTYICJjCBj0P9TCetscX+rGOEKb22VTn5Pxc8qyGc2YlZ3F5d/8Jl8+7VSy+vdl0SsvEfj3WvuztB8Xuf1zycvPo2Dj+6xZt5rqquqY7NY0H7faWTb1ZWVngTGNc7m790iLOl3G5nZp31c2Wdr12WQlUnypG0tTfOko4F/B5bcCg4wx2aHrbK/4UjfaXthkadfntM9mNGNefh5Dhwxl1+4i6mrrWPz6Er582lfiwtJ+XNiyW9M+fjZZNvXl5edRuL2QFctWMOeC89m/bz9Tzj4rLiztY6GdpV2fTZZnJdWxpSm+dCNwEfB20Of0RAIm+2Uhq3IUX9q8PMugtvWUlpSSnJLc5DZsvPS1hTV+wjjmzrqQpKQkTj/j9GP6wkUzFqXt4aWbXqG6upokSY4pmjGWbfraa99t8vvgngP40qGT+f0Dj5CcnMz40yZx52e/p2bbf2aoNLdA035cOLUKAzt2a9rHzyYrnvrCzc2uTa5h0+bN3DD/RrqkdWaVfyWrtq1s/Dzcd8wNY5FoLO36bLI8K6mOrbg/ENVSfKmIDCCQEHVjcNH7CKQ/bQC+S8DDNG5HgmcZ5LwnOTmZqdPObrwNW1dXR3l5eVz0OWUBVFZWcc70aZxy6imcNW1KTPo2btjI7x97lCXLXic5KSlu+6oT8PSjTzD6ptPp/t0hbFj3AVM6R75aqP240G63pn38bLJs6tv+3g4qK6q4+tEruej2C/DV+6jYXRGxx+Z2ad9XNlna9dlkeVZSHVva4kuvDs5FvRLIBHY0W95RfOmTf3qqyWcNlkFvv/UOF158ITU1Na2yDEpJTXFkexFrn3ZWdr9syssD/+dSX19PzdEaMvr0iYs+p6zk5GSKioq48OILAdhXWhaXfex0X+3aupN+J+SwJ7kck9yJPuP7sX/j3riwbG2X87H4j11Yzz49Gu3CtOhzI8umvk/e/RRfvY8FdyxkyUPLqK/1sfjBpXFhaR8L7Szt+myynOrriPLiS1tRDuJLextjUoPvzwNWNpuf6ji+9Jprv9Xks7z8PD7+6GMuv+Jy/D5/THOgtNte2GTl5eexa+cu5s6+iLNPn0a9r56vX35ZXPQ5Zb36ymt07tyZiooKxO/nn0uWxmUfO91XB8oP0LNvL/p3D0yrPmngMIr27o4LS7v1SmlRGd179wCgtqaOA2UHo/4RdOP3yibLpr6hk4cwcsoI5j1+NbNunkH6gHRyR0Z33HDjWGhnaddnk+VZSXVsaYovHQk8a4wR4CNCnsJva3xp8zlQO9YW0qt/T77/ve83WgZFmwOl3SbHaV8sPeHmCY49dzTL/rIiMGk8Cdbuf5fi7jubLNN8nmC8WDsPFzVZfsfaQsqT9zNs2lB+8qPbEL9w4tiBcdnHsfYcE6GbspyFC15m7/2f4vf7OToqjRPTc3ls9o/bzGqPvnj1hPOl3WmKeWX169w841Ya7MJGDx3TooetzW1yKyue+sLN5V64elHjXG4ETu4zMuJcbpvbpX1f2WRp12eT5VlJdWxpii99V0ROEpERInKRiBwIWVe7xpfatAyyZZPjtM9Jj9/n5/2lBdz29C088Pq9JKck8d6ygrjoc8Iq3rqXz97fzpq/v0fntM4kpSRR+llZ1Hlutu1Q1q1b1xihW1BQcNxar9iyW9M+fjZZtvWtW7e+cS43Isf1WGhmaddnk+VZSXVsHZfxpcMmD2V/YTkX3zmX3R/uYcUf/hU3y6BQm5xePXrGzSbHaZ+Tni0F20jP7kPf3Axqa+pITk2h8kClGtYZV5zG0FOG8O7f32PyxZMoeHk9/UfmsP39HWQMaNlU3412KDZZnr7EYWnXZ5OlXZ9NlnZ9NlmJZCWlX2Hry3pCVMjnNwMPAJkiUm6M+SaBW/YG+AK4XkQ2hlnvYFqZEGXTMijcbenW2uTYtGqKpae5vp0Hiin+ZG+T27BDModFvA0bT9aUMP+uaNjHK3+3ikMHDzEsczjp6entso/b2uNWlqcvcVja9dlkaddnk6Vdn01WIllJubE6IiGq4cR1OhA6WbAQmCIio4G7gcdbWG+bE6JsWgY5scmxadXkpKdfv2yMofE2bHVlVdwsQNrC2rhhI089+wTvrFnJ5g2bVNmNuJHl6UsclnZ9Nlna9dlkaddnk5VIVlKdjLH2slVxu3IqIiVASfDnL4wxWwgY5n8M/Aa4hZAn7UVkdUj7GgI2UU0qJCHq8uBbzwJ3AX9ojbZQiwig1bYSsfZAU5scoNEmZ8xp+S32tIdVUzy3y+lYuJGlXZ9NlqcvcVja9dlkaddnk6Vdn02WU31etU/F3YQfmiZEGWMuAIrD3bIPqWuAJWHeb5eEKO02OTatmtxoAWKTpV2fTZanL3FY2vXZZGnXZ5OlXZ9NViJZSXlXTh1UaEIUgVv9PyFwS7+l5c8mcHJ6erw0xdPe6Zg5kzHY5DixTwo3v1O7xYZ2lhtsvzqS5elLHNbxbI2XaPpssrTrs8lKJCspN5bthKihwGBgozFmJ4Fb9+uNMf2Cy48BngAuEJFwvj/tkhBl097JiU2OTasmN1qAOO1zo+2XTZanL3FYnjVe4uizydKuzyYrkayk3JgQFc+n9Y9JiBKRzUBWyDI7gUnBp/UHAgsJPH3/Sbh1iogYY94kkBD1PFESogg+VHXUVy2hn2m3d7Jp1eRGCxCnfdqPC+0sT1/isDxrvMTRZ5OlXZ9NViJZSbmxrCdEicjiFpa/g8Cc0t8Hz87rRWQStD0hKtxt85nzpzNn1gX46n0MyR8c1d6poryCCRPHc/21N+D3+xk3fnzMthfR+sLZJ+0r3M8j1z+G3+9n/OjxZGZmxmTV1FqNtno0srQfF+3RY5Pl6UscVqw9x8N3RLs+myzt+myynOrriLLy8JDliufT+u8QxRtWRAaF/DwPmNfCcjNDft4BTG6rvqSkJDKy0/HV+zi0/xB7C0vIHRzZiL/x0jYGkxT7v6Za25ebm0Ne/iiKi/fi9/lJSo79VoITjbZ6nPTV1dWxdctWMrMyqa/34fPXR+1pi8bWHhe29TnpcapR83HhVn02WU71af7babPHrSzt+myynOrzqu3lxhPuqOX3+Xnx4UXM/8U8bnvmFmpraqPO6QyN4Fv06kt8ULC+FTGVrevTzrKpL7d/LsNHnMSLi17ghYXP8/GHW6LOcXPKcnJc2NTndNydaNR+XLhRn02WU31u/NupXZ9NlnZ9NllO9XnVPhXxyqkxZjOBWRfHfERgCuiYuKiKczmZ06l9foxNlvZ5oE5ZTo4Lm/q0zxP0jtvEYTnV58a/ndr12WRp12eTlUhzTm0+qGSrot3Wn+10xQ7iS88i8HBTYXCRhSLyszDrHUwr40ubz4H6bP8udmwu5Adf/QmC0L9//6jxm06jzJxEkbo11i2WvvaIf42V5SQq1Ym+cKx4jXs4fbFotKWvvfrcqM9pXzz/xrRHdLH2cdeuzyZLuz6bLC++tGMr4m19EdnV8AKOAqODryPB9yJVa+NLAd4WkXHB1zEnpsFqc3zpgAEnMHP2V1m74T3eXbuK+vr6qPGgTqPMnESRujXWzRHLQfyrU5aTqFSb+mxG6Ko/Llyoz2mfzb8xjr4jysdduz6bLO36bLK8+NKONeGPac6pMeZS4H3gEuBSAklPX4vUIyIlIrI++PMXQEN8KfwnvjTclIFIOgyB+NIFwbeeBea2Zh3gLB40NMosJTUl5igzmywnfbZ6nPaFxr/27NOjMf41Hiw36nOqUftx4UZ9Tvvc+DfGJku7Ppss7fpsspzq86p9KtYHom4DThGRq0TkSgJPy98eK6QV8aVfNsZsNMYsMcbkhfncUXzpWyVvNHlV9C1j2/atTJ9xHmeedhZHaqsZOn1Qk2WaV6xRZjZZ7dGnLXZuSs60Jq9eh9IZmjOcKTnTOLXPadRU1NG3a2aTZWxulxN94TTa0ud0DLUdF8eDvlj7joe/MTZZ2vXZZGnXZ5PlxZcmRnxpJxHZF/J7BbFfdY01vnQ9cKKIVBpjZgIvA8Nj1NeqSkpK4r9u/gajeo1h8KDBAHx4YEOUHqdRa0mcOmNyYxRpUnJSYKJ1XFi6Y92c9B0+/AVvvvEmk8ZObpzjlts/Ny4sN+pzqlH7cZEI+saOG8P8eddTWlJKr1694shy398Ymyzt+myytOuzyfLiSzu2Yr1y+k9jzFJjzH8bY/4beB1oyUy/sUwr4ktF5LCIVAIEjfpTjDF9m63SUXzp4r/+85jP5067iMzR6WxL2swjf/st769aG3FbnEaZOYkidWusm5M+J/GvNrdLuz6nGrUfF9r1ZWVngTFcetklnDN9Gt17pJGdlRkXlhv/xthkaddnk6Vdn02WF18a/mWrYjo5FZEfEogCHRN8PS4iP4rUE5wfekx8qYhkicigoAH/HmCCiJQaY/oFezDGTA5qq2imQ4A3CcSXQpT4UhGZJCKTZn5zRpPPjhyuoXdKH4qri6itqePjtVvpldUz2hi02XrF7xfPesVlLO36bLI8ff+pvPw8CrcXsmLZCuZccD779+1nytlnxYXlxr8xNlna9dlkaddnk5VIVlJurJgTokTkJQJXQWOt1saXfg243hhTDxwBLguejGLaGF/6yLqXm/x+Qm1fDn9whMf+8BjJyUmMP20Sq7vu5M11nzQuM2X2sdY648aP5cLZFyEipKamsuGDY6fNetYrxxdLuz6nfZ4FWuv6nt329DF9tck1bNq8mRvm30iXtM6s8q9k1baVjZ9fNeJqR6zj4W+MTZZ2fTZZ2vXZZCWSlVSnyGGcCVmxzhu9yBjzqTHmkDHmsDHmC2PM4Ug9IvKOiBgRGRNiD7W42TKDRKQ8+POjIpInImNF5EsisjpkuZnBE1NEZIeITBaRYSJyiYjUtHajO/ftyp+ffpbRN51G9+8OYcO6D5jSObpxek5uDm+tepOCje9zy09+SElJCZs2borY41mvuJulXZ/TPs8CrW1929/bQWVFFVc/eiUX3R7Ioa/YXRGxxynLjX9jbLK067PJ0q7PJiuRrKTcWLFeOb0fOF9EtsRTjK3atXUn/U7IYU9yOYZO9Bnfj/0b90Z8/CorO4uysn10S+sGQOneEpI6JUGUf7GE2lEArbawiLXHJku7Ppss7fqc9rXVnije+rSP3yfvfoqv3seCOxZSX+ejvtbH4geXcsVvLlexXdr3lU2Wdn02Wdr12WQ51dcR5caEqFgfiCpzy4kpwIHyA/Ts24v+3bMBOGngMIr27o7Yk5efR9GuInbtKuKSuZfyx98/zmlnfIUxY0dH7PMsNtzN0q7PaV9efh67du5i7uyLOPv0adT76vn65Zep0ad9/IZOHsLIKSOY9/jVzLp5BukD0skdGd3FwY1joZ2lXZ9NlnZ9NlmJZCXlxop45dQYc1HwxwJjzN8J2Ds13kYPPoHfUm9r40t7AX8BBgZ1PSAix0zkMsZMBJ4BuhJwDPhew9zUlmrB7Eea/L48ZTkLF7zM3vs/xe/3c3RUGiem5/LY7B83LhPOG3DMOflceH5gzumwsUP4cPuH/G31X8gdnNO4TPN5Xm602PD7fZhOpsl8RE36bLK064u1L9zxPvbc0Y32RCTB2v3vUtx9Z+Pn7XGsO+3TNn7N548u37mchasX8dJNr1BdXQ0CJ/cZGXaeaUdsl/Z9ZZOlXZ9NlnZ9NlmJZCVl03/UVkW7cnp+8NUDqCbgT9rw3uwova2NL/0O8LGIjAXOAn5tjEkNs94/ANcSuAk/HJgRZpmI1TezL+vWreN3jz3Cwn8soKCgoNV2LVWHq+k3KJsta7dFZbnNYiO3fy55+XmN8xGrq6rxxzBR3I1joV2f0z432hPZ1rdu3Xp+/9ijLFn2Oogc12OhmaVdn02Wdn02WYlkJeXGinhyKiJXi8jVQBJwU8jv34+2YgfxpQL0CNpJdQc+J3CC21jGmBygp4isCV4t/TMO4kudWETs2lpEn6zedOvZjeSUZMaeMYaP1mwhe0BWu7O0W2zkj86neE8xe/YUc7T6KAcPHmTCxOgPlLlxLLTrc9rnRnsi7fpssrTrs8nSrs8mS7s+m6xEspIyFv9nq2J9IGqMiBxs+EVEDhhjxscKaSm+tNkk3keBfwB7CVyp/bqI+Jutqj8Bb9SGiim+tHlVlFcwYeJ4rr/2Bvx+P+PGj49q11K7aTkDMwfy9I/+gt/v59DBQ/TPzuU7F323zaz26IknK9wt35nzpzNnVuAp5CH5g0mbmHrMcs3H0A1jkWj6Yu0LZ0+0r3A/j1z/GH6/n/Gjx5OZmRnVnkjzWGjXZ5OlXZ9NlnZ9Nlna9dlkOdXnVftUrA9EdTLGND6qa4xJJ8YT2xbiS+8Is+h5wAYgFxgHPGqMieyM77Dq6urYumUrXbp0JjU1FZ+/PnoTcPTIUY4ePcrRI0cZO2EseaPzY+prTFfAYJJi+5eHkx6brKSkJDKy08nITufQ/kPsLSyJG6tfTj9efm0haWlp7N5dFL2hDSyb+8rJdtk6LnJzc8jLH0VtbS21NbUkJcd2O0v7catdn02Wdn02Wdr12WRp12eT5VSf7YpHElRLL1sV68npr4F3jTF3G2PuBlYTsJeKWKYV8aXA1cBCCdRnQCFwcrNVFgd7Giqm+NIn//RUk89y++cyfMRJvLjoBV5Y+Dwff7gl6pzJ0Dlki159iQ8K1sc8P6a1fdpZfp+fFx9exPxfzOO2Z26htqY26nxEp6ys7CxKS8v463P/x5Chg6mpqYk5BtLGWDjdV062S/tx4elLHJZ2fTZZ2vXZZGnXZ5PlVJ9X7VMxXf0UkT8bYwqAqcG3LhKRjyP1BOeOHhNfCmSFLLMTmBR8Wr8ImAa8bYzJBkYAO5rpKDGBEIAvAe8BVwJNH8X/z7KPE4hc5aivusnT/KFzJnv16BnTnEnt82NsskLnI9bW1MU0H9EpqyEGsrqqivk3fJuV/14ZtxhIm/vKyXZpPy48fYnD0q7PJku7Ppss7fpsshJpzqkbqzXxpR8DEU9Im1Vr40vvBp4xxmwm4Gz/IwmmRxljNojIuOByN/AfK6klwVfEaj4Xsq62jlqpYdb0gOFA76zex8yZDDdfUnPUWjxZ7RGXGCvLZgykzZ722C6nx4WtKFJtx22i6bPJ0q7PJku7PvC+wx3BSqj4UheeQMd6W7/VJa2PL90rItNFZLSI5IvIX0KWGxfyc0Hw86EicmPwqf1WVXJKMrc8dhMPv/EAP3/pLo5UHmHtinURe7RHrdlkOYlLdMqyGQNpc1852S6nLFtRpNqPW+36bLK067PJ0q4PvO9wR7C8+NKOrZivnLqpjDF07toZgM5dUuncJZWdW4o45ZyJLfZoj1qzyfJiINvW43S7nLJsRZF6x23isLTrs8nSrg+873BHsBIqvjR+1xk7rNy3RTHUFwcrqTxUxS+ve5CfXHwXplMnRk4aEbFHe9SaTVY89V014uomr2vmXsOcOefz9lsrefi3DzFs+DCmfmVak2USYSzaY7tiYb1V8sYxr4q+ZWzbvpXpM87jzNPO4khtNUOnD2qyTEeNn02Wdn02Wdr12WRp0+d9h3WwvPjSjq24XTk1LcSXGmPuIpDwtD+46E9EZLEx5lzgPiAVqAV+KCL/CrPedODvwCBgJ3CpiBxojbbPSz/n4e//ARFB/ELNkRrS+0X+V6j2qDWbLJv6AMpKy5gzcy7V1dXEOotD+1g42S7n457EqTMmN0aRJiUnReVpPy7cqM8mS7s+myzt+gJ93nfYNsuLL+3YiueV0xbjS4HfhJmHWg6cLyKjgauA51pY74+BN0RkOPBG8PdW1cARA7j3pbt4cMl9PPD6vXTumsqq19ZE7NEetWaT5cVAtq3H6XY5ZdmKIvWO28Rhaddnk6VdH3jf4Y5gefGlHVtxu3IqIiVASfDnL4wxofGl4Zb/IOTXj4CuxpjOIlLTbNELgLOCPz8L/Bv4UWu0VR6qajQVP1J1lCOVR8nIjnzlVLvthU2Wdn02Wdr1gTPrLzeOhXZ9Nlna9dlkadcH3ne4I1iJZCVl0xzfVll5IMqExJcSsJi60RhzJVBA4Opq89vyFwPrw5yYAmQHT3wBSglMG2hVHa44zHO//BtlRfvw1fno2qMrUy89K2KPdtsLmyzt+myytOkLZ+f12f5d7NhcyA+++hMEoX///lGtv9wwFommzyZLuz6bLG36wn0vndj3uWEsOpKVSFZSbqy4PxBlQuJLReQw8AcCSVHjCFxZ/XWz5fOAXwLfjrbuoI1Uq62k+g/N5ceP38z518xk7JmjMZio8ZvabS9ssrTrs8nSrg9gwIATmDn7q6zd8B7vrl1FfX39cWlDo12fTZZ2fTZZ2vWBM/s+t46F9uOiI8pY/J+tiuuV0zDxpYhIWcjnfwJeC/n9BGARcKWIbG9htWXGmBwJpEXlAPtaYF8HXAfwP/fdwMxvzmjy+YH9B/n4vS1M/+Y09u4oYcvabeQOzmlxW7TbXthkaddnk6VdH3g2NImizyZLuz6bLO36bLK067PJSiQrKTdW3K6chosvDb4fegZ4IfBh8P3ewOvAj0VkVYRV/4PAA1ME//tKuIVE5HERmSQik0adP4qdh4saXx/v3sqzv/4rp3xzEsWHSjl88AtMhmmyTPPSbnthk6Vdn02WNn2eDU3i6rPJ0q7PJku7Ppss7fpsshLJSqqTMdZetiqeV07DxpcC3zDGjCNwO34n/7l9fyMwDLjDGHNH8L3pIrLPGPME8EcRKSBgN/WCMeYaYBdwaWuFffzvbRSu28muDUWIX0jpmsKQUwZH7HFqK+H3+zCdTJPYuePVYsONLO36An1JXHzjXJ7+6Z8xphOpXTsflzY0SUmdGD0mnwtnX4SIkJqayu6i3Wr02WTZ1jd23Bjmz7ue0pJSevXqdVyPhWaWdn02WYlkJeXGsh5fKiJXBCNKx4jInIaHm0TkHhFJC1l2nIjsC342L3hiiohUiMg0ERkuIueIyOet1VZ9qIpuvbuS1rsbnbt3pqaqhpd++nLEHqe2Ern9c8nLz2uMnauuqsYfZVK1Wy023MjSrq+h9u8pZ+wZYzhp/DBOPW8SW9ZuU6PPFisrOwvTqRNvrXqTgo3v89/fupLy/eVs2rhJhT6bLJv6srKzwBguvewSzpk+je490sjOyowLS/tYaGdp12eTlUhWUsYYay9bdVwmRJ155el8+8l5zHv8amb8z7mkdknlK9/4UsQep7YS+aPzKd5TzJ49xRytPsrBgweZMHFCXFjaLTbcyNKuD2DP9mI2r/6IL8+cjN/vZ9u6T8kekKVGny1WXn4eu4t28/mBA9TV1rH0n8voltYNokzyd+NY2NSXl59H4fZCVixbwZwLzmf/vv1MOfusuLC0j4V2lnZ9NlmJZCXlxtKUEDUI2AI0XNJZIyLzw6y31QlRg3oOPOa9Tv5OpBZ159zTvsr8M75Dadc9HKpteTWx2kqEm783euooZk2fDUB6dh/SJqY2Wa65HYhbLTbcyNKmL9x86QWPLqJiz+f8/kd/wu/3M3D0ALrndW+y7JRmzwK6YSzCfRfHnJPPnK/OwefzYzoZpl4yhQNZ+9r9+6htLGzqe3bb08f01SbXsGnzZm6YfyNd0jqzyr+SVdtWNn4eLq7XDWORaCzt+myyEslKqpMLrzNqSogC2B7y/jEnpsFqc0IUwKiMMWSM6s3GmgK+9+P/Yftnn0Vc3rHthc/P2uXrueO5W7lnwZ1UHqpi06oP48NSbrHhRpZ2fTvWFtI7pxfJKUmc9z/nMmjciVQfPkLF7goV+myyGlN2nvkRD/7zPrIHZvHphu1xsZFTPxYW9W1/bweVFVVc/eiVXHT7BfjqfVGPP5vbpX1f2WRp12eTlUhWUm6seM45LRGR9cGfvyBwVbTFhKhW1AUEkqEI/ndua1eQbJLp0zmD4uoiunXvyuDRJ7JxzeaIPaG2EimpKTHbSuzaWkRm/wz65mbQs08PBo0aGDUq1SnLSZ+tHreytOsr3rqXT9/9jMrPq3jr6bfZ81ExBtj+/g4V+myyQr+LySnJTJo2gS5pnaPOv3XjWNjU98m7n+Kr97HgjoUseWgZ9bU+Fj+4NC4s7WOhnaVdn02WU31etU9ZuRbcLCEKAglRm4wxTxljQg0XBxtjPjDGvGWMOaOF1bU5Icp/BI7WHSG/zzi+lHkGMybPov+gyOfNTm0lSovK6N67BwC1NXUcKDsYdVKxWy023MjSru+MK05j2rfPZuSUEcy6eQYDRp/AxAvG80VFlQp9NlkHyw+R1iuN6sojAPTo3Z3SnWVR59+6cSxs6hs6eQgjp4xg3uNXM+vmGaQPSCd3ZG5cWNrHQjtLuz6brESyknLjA1Fxjy9tnhBljPkDcDeBeah3E0iI+haBtKiBIlJhjJkIvGyMyQumSoUtERFjTNSEqEfWvdzk9xPrszl78Ll854bv8cE7BQwfOYJv/nxek+WmzG467yxWW4nm89V2mmJeWf16k9i50UPHRIyds2mx4dTqSrM1jE2WNn3h5u4t37mchasXsfJ3qzh08BDDModzcvrIsMvGW59NVvPvWG0fP+srNnL7JT8DgdTUVAYPGcR3Lvpuh+jrSFY89TU/rhqOv5dueoXq6moQOLlP5OPP5nZp31c2Wdr12WR5VlIdW3G9ctpSQpSI+ETED/wJmBx8v0ZEKoI/rwO2AyeFWW1Zg5F/tIQoY0yBMaagaOnWJp8dSjtCRc1BirqXk5bfl0OmkqG9Toy4LU5tJcaMHY3P72uMnTt88JAqiw0nVlfarWFssrTra+hbt249Tz37BO+sWckHBeuPy7Hom9mXrVu38fKrC3l37Sq6du3C2HFj1eizybKtb9269fz+sUdZsux1EDmux0IzS7s+myzPSsqlVlIOEqIyjTFJwZ+HAMOBcBPjWp0QNfC8k5t8drDmMFt3fUL9p5X0mpxNz9TuFH2xN+L2aLe9cNrnxOpKuzWMTZZ2fTZZnr7EYWnXZ5OlXZ9NlnZ9NlmelVTHlqaEqDOBnxlj6gA/ML/BYL+tCVHnDz/2ZOuhX/2KB+78BUerj/Dnj5/jsG9P2OUaqqK8ggkTx3P9tTfg9/sZN358zLYXre2LJyucvc7M+dOZMyvwFO2Q/MHHWF2Fsyfq1Nuwfc8OfvCjH5LaK/UYaxg49vaetrHoqB63sjx9icPSrs8mS7s+myzt+myynOrriOoUxas5EStuJ6ci8g7h3a0Xh3kPEXmJwBSAcJ/NC/m5Amh5wmYM9em7n7Hjw0J+/JMfU1tVQ6fkTpzuj2zCDyGXzjGYpNgPBid9NllJSUlkZKfjq/dxaP8h9haWkDs4p8Xld6wtpHNaZ1IOH6H2aF00D/M269M+7tr12WR5+hKHpV2fTZZ2fTZZ2vXZZDnV51Xby33OrTFUyaeldEoy1FbVUFdbz8HSQzHFlzbMm1r06ksxzdtz2meT5ff5efHhRcz/xTxue+YWamtqeW9ZQcSePVuK2fnBLuqO1uH3+Tmw9yCL7vlHXPRpH3ft+myyPH2Jw9KuzyZLuz6bLO36bLKc6uuIcuOc07g/ra+xzrzydM688nQACtfv5PUH/hm3+FLtc3G2FGwjPbsPfXMzqK2pIzk1hcoDlRF7hk0eyv7Cci6+cy67P9zDij/8i/4jW77S2hZ92sdduz6bLE9f4rC067PJ0q7PJku7Ppssb85px5am+NIU4AlgQlDXn0XkF2HWOxh4HsgA1gFXiEhtJC3h4ks/WrOFJ+96hvo6H916dOPLkyZH3B7tUWsNfeMnjGPurAtJSkri9DNOP6bvGKurA8UUf7K3idXVkMxhTZZrHm25fOdyitL2NFrDJEky/f0Do1rDeBF37mZ5+hKH5VRfaUkpySnJTaznovU56bG5Xdr3lU2Wdn02WQkVX+rCE2hN8aWX8P/bO/P4qMqz/X8fMiyykwAhQSKrVsIii2iLioJFBEXQuva1vha0au3i21Zr/Vl931pbra1VUau1UtvaqkVwA2Sx7soSkE0BlSVISFgiiyGSZeb+/XFO4mSYTCaHzJ1nDuficz5M5pz7fK/zzJKTmedcN7QWkcHACOB7xgnvj9U9bn1/YC8wrbHGIuEIs2a8wC+evJm7Zt1BdVV12rcUrVFZ2UHOHj+Ok085mTPHjWlw+x49sjGG2qir8rKDSflbvWp1bTRMKCPDurGwvcWdH1mBv/RhefUXCoUYO+6s2ui5qqoq9uzZ0+Q1msdl+2OlybLdnyYraF/avErlBVHFOMH6iMgXxpiG2pcK0M4YEwKOASqBOgH8bjzVWOAK966ngDuBRxvjLbqNIVDbUnTI6EH11kS3MgM8tVpLts4rKxQKsW3bNn540w/4+1P/YFfJzgbzRzX9+ZFluz9NVuAvfVhe/WX3yGbPnlIAqqurqThUQVaXLk1e49WjHx8rTZbt/jRZXv01h4wPr9a3qX3pLOAgzgntNuC+miipKGUB+0Sk2v15O4lPeAHnq+zopdP+TPrlDGBMzjhO6TKaitIquh7Trc42sbKt1dpTG2cetsx4+mEiGWFeWDubz77Yxj9feIZ9ffbW2UbLn+ZYNCfLdn+arMBf+rCSrXmz+LU6S2nXnWzctIHxE87hjNFn8mVlOf3G9z5su8bWpMNYHA0s2/1pstKpfakfZVP70lFAGMgFugBvG2MWi0i8IP4j0oEDX/D6a68zcuio2nmWuT0T93q2vdXa5uVbaNe5LYPPGcz8BxYiEeG4oXl0zcuywp9fWbb702QF/tKH5d1fBqdMGMXCfyxGRMgIZTgXjjRxjVePXtsx2/xYabJs96fJSqf2pS2M/4KXrGlfivNV/asiUiUiu4B3gZExuywFOrtf/QMcCxTVw65tX/qXPz9ZZ53tLUW91BRt2MGnyzax5NmltG7XmoyWGZR8upPSz0qt8OdXlu3+NFmBv/RhefUXCUdYtqCA22bezH1z7ybUMqPB6DkvNV49emnHbPtjpcmy3Z8mK53al/pRqbxa31BP+1J3PipEtS/F+Sp/LPB3Y0w7nIuo/hi9TxERY8zrwLdwrti/igTtS4HHAQ6FyyVmne9iL06/cjT9Tu7L+88uZdRFIyl4YSU9T8xh07LNZPWq/9NTP46FJst2f5qswF/6sLz68xI956XGq8fodsydOnRMqh2z7Y+VJst2f5qsdIqS0swf1ZJN7UsfBmYaYz7E6Tk0U0TWABhj5gHTRWQHcAvwjDHmLuADnBPgRslLtIltsRfxYptqIp7eevhd9u/bT/9uA8jMzEwY8eTHCBBNlu3+NFmBv/RhJVvjJXouVl5qkvUYb77q4LEDmTT+PAAys7sc1o45lmv7Y3UkrOEjhnHd9OspKS6hU6dO1vmzmZVOUVJ+VMq+1heRd0TEiMiQ6NgoEblSRAa790+u+RRVRMpE5GIRyReRgSLyu6h9TXRPTBGRzSIySkT6u9tXNNabl2gT22MvaupWr1rNk089wTtL3mLtqjVWRWz4kWW7P01W4C99WF79eYme81Lj+bjCEZYvWskv/34rd826g7L9B30RE+j58crpwSWXXczZ48eR1TUzZRzbxyKIkko/HZUdorxEm9gee6HJst2fJst2f5qswF/6sGz357XOrzGBXuq6Z3ensHAbRdt3MP1707j7V3enJFrQa53trCBKqnnlv0u84qgpok1sj73QZNnuT5Nluz9NVuAvfVi2+0u2zo8xgV7rYqMFV7YqYM3aNeSM7c6CwvkUlexISbSgjWPRnP4CNY3U25e6634AfB8nOmquiNxsjMnCyTo9GfiriNxYz34zgWeB3jhzVi8Rkb2N8RYJR8gIZbBnRykiQouMFknEoejGXgw9aUiduUJ+iNjwI8t2f5qswF/6sGz357XOjzGBXuu2riwkq1cmL987DxGhTbvWKYkW9FpnOyu9oqSCT04bo7jtS40xZwEXAENFJB+4z93+EHA78NMG9vtz4DURGQC85v7cKIVahrj5Tzdx/4J7+P2839D6mFa8+8qShDWasRfds7uDMbVzhdp3aNfg1zFeWX6MANFk2e5PkxX4Sx+W7f681vkxJtBr3fb1RezavJu2nY6hTYc2fFl2iDl3vWSNP9tZQZRU8yqVF0QVi8hK9/YXQE370uuB39ZcyORmmiIiB0XkHZyT1ES6AKdtKe7/UxrrrWz/QcLhCABfHjzEl2WHyMpOPOdUM/Yif1A+WzZtYfHCxUy+4Hx279rNmLPOTAnLjxEgmizb/WmyAn/pw7LdnybLdn9e6/qP6kfekF5c+5dpnPfTc+mS05meJ+ZY4892VlpFSSn+05LKBVEx7Ut/B5xujPk1zonoT0VkeSN2lx2Vk1qCM22gUfq85HMe/J9HqaqsAoHO3Tox9pIzE9akMvYi3tyfylAFa9au5YbrbqRNu9a8G3mLdze+Vbs+XjyU7REbfmTZ7k+TFfhLH5bt/jRZtvtLti72d0JNtODzN71IeXk5GRKiZySvyaMFvdbZzgqipI5MxpgMoAAoEpHzGluf8guiYtuX4pwQZ+J81f8z4DnjMUFWnImiiSeLxlHeCb24+/k7eXDxffz6+Tv58uAhli9ekbBGM/Zi09LNlJUe5OoZ3+HC2y8gXB1usNOTV5YfI0A0Wbb702QF/tKHZbs/TZbt/o6EtXrVah55bAbzF84llJFhnT+bWekUJdXCGLWlEfoRzjfm3o7Ja2Eyite+FNgOzBZHy4AI0LURu91pjMlx958D7KqHXdu+dN7Tr8auo/UxrQFo3aYVrdu0Yuv6bQmh0bESLVu19BR7kWzdx+9/Qrg6zKxfzmb+Awuprgwz7w8LUsLSqvEry3Z/mqzAX/qwbPenybLdnybLdn+aLK/+AoEx5lhgEvCE132k7OS0vvalwAvAWe42xwOtgMQJ+HX1Ek7bUmigfamIjBSRkRO/PaHOui/2lVG2/yD3XPsHfnHRnZgWLThx5AkJoZqxF/1G9eXEMScw/fGrmfSTCWT2yiT3xMRXnHpl+TECRJNluz9NVuAvfVi2+9Nk2e5Pk2W7P01WOkVJGdNCbUlSfwRuxvnw0ZOao33pk8CTxph1QCVwlfv1PMaYrUBHoJUxZgowXkQ+MsY8AfxJRAqA3+JMBZgGFAKXNNZYzZxTEUEiQsWXFWT2SHxBVCpjL+LNFZr93pzauUIIfK3LiQnnCnn16McIEE3WkfhrbFyYn8ci8KfLst2fJst2f5os2/1pstIpSkpTxphrgWuj7npcRB6PWn8esEtEVhhjzvTKaY72pZUi8l8iMkhEhovIf6JqeotIpoi0F5FjReQj9/7p7okpIlIqIuNEZICInC0inzfWW82c0z/M/y33zb3buiiprt26smLFytq5Qoj4ImLDjyyv/rzEhfl1LAJ/+izb/WmybPenybLdnyYrnaKkNK/Wj/5W2l0ej7EzGpjsftj4DDDWGPOPxh7TUdEhKla2R0nZzrLdnybLqz8vcWF+HYvAnz7Ldn+aLNv9abJs96fJSqcoKZskIre6Hy72Bi4D/iMi/9XY/ahESdmmA6UH+Ps9/2LXZ7sJV4XpnX9cUlFSw0cM4/prbiASiXDSsGFJx140ts52lu3+NFnJ1sSLC2vR2bBp+2Z+esvPaNWpVYNxYX4Zi8Bf87Ns96fJst2fJst2f5osr/6aQ0GHqEbIGNPLGPO6MeYjY8yHxpgfRa37gTFmg3v/vTF1ecaYMmNM3E5Rxpg+xpilxphPjTHPGmNaNdZbz365TJ4+iczuXcjM7sKBPQfYsaW4wTpjjLNgMBnJPxm81HmpqaqqYsP6DbRp05pWrVoRjlRb5c+vLC81n7z/KXuL99GmfRsyQhkNts89EpbXuuCx8jfLdn+aLNv9abJs96fJ8uovkCMRecNLxinY1b60Rn8A5ifY7z3A/SLSH9gLTGussUg4wr8fnMN1v5nObX+9mcqKSpYuLEhYEz0PdM7Lz/NBwcqk58c0ts4rK7dnLgNOOJ5/z3mO52Y/w0fr1hNp4C89TX9+ZHn1V/xJCS0yDJUHK6iqrGZfyX6e/98XUsKyfSwCf/os2/1psmz3p8my3Z8my6u/5lDtSbTCoqWUfa3vdnEqdm9/YYypaV96DXHalwIY5wr9LcDBePs0zsiMBa5w73oKuBN4tDHe1hdsJDO7C11zs6isqCLUqiVle8saOh6r58cADBo8iKLtRWzfXkSnDh3Zt28fw0cMt8afH1le/Z3xndM44zunAbBl5Vbm3vcq37j81JSwbB+LwJ8+y3Z/mizb/WmybPenyQrmnDavrGlfapxOUrcA3wTifqUPZAH7RKTm++rtOCe8CTUmZ1ydn7fuLWLXlt08eN2f2Fa4jQ4dO9C3W//DtouWba3W3ix+LW7t4LEDmTTe+RQ9M7sL7Ua0qrNt7DF68VdSXEKoZYgpk6YiIvTK62VV2zlNVrI1vTvmHXbfh0vW85c7/0p1VZi2Hdry9ZGjmtxfqo8r8JeeLNv9abJs96fJst2fJitoX9q8sql96Z04X9cn/gizCZSbm8PYs8/igqmTGffNsXTtmtVgje2t1sCZrrB80Up++fdbuWvWHZTtP8iad9c1OSsUCjF23FkUrF7GkhXvUVVVxZ49DfdRCFrcRdWFI8ya8QK/ePJm7pp1B9VV1Sl5rDSPK/CXPizb/WmybPenybLdnyYrrdqXYtQWvWNKoUzj2peeAtxrnGysHwO/MMbcGLPLUqCzMabmE99jgaJ62LXtS//y5yfrrOue3Z3Cwm28/eY7TL1oKhUVFQ1mTNreag2gcMM2uvXMomtuFh27dKD3wLwG81u9sLJ7ZLNnTykA1dXVVByqIKtLlwb9BS3uvpLWY6V5XIG/9GHZ7k+TZbs/TZbt/jRZQfvS5pU17UtF5HQ3hL83Tuuru0VkRvQ+xbmk+XXgW+5dV5FE+9LQGYanNs6sXVa2KmDN2jXkjO3OgsL5FJXsYF+fvXW2iZVtrdbG5Iw7bOm0P5N+OQMYkzOOU7qMpqK0iq7HdKuzjRfWm8Wv1VlKu+5k46YNjJ9wDmeMPpMvK8vpN773YdtpjUVT1aWyRuuxSoexCPw1P8t2f5os2/1psmz3p8lKr/alwQVRjVGj25fWJ2PMPGC6iOzAmZf6jDHmLuADnBPgRmnrykKyemXy8r3zEBHatGtN17zEX+3b3moN4MCBL3j9tdcZOXQUgpB3XB65PXObnBUJR8gIZbBnRykiQouMFklFIQUt7r6S1mOleVyBv68UiYQxLUydedlH61jYzrLdnybLdn+arKB9afPKqvalUbV3ish9UT9PdE9MEZHNIjJKRPqLyMU1V/03RtvXF7Fr827adjqGNh3a8GXZIebc9VLCGttbrQEMGTqYcCTMC6/MZsny9ziwb39KWKGWIW7+003cv+Aefj/vN0m1f/XKsn3cbX+sNI8r8PeVcnvmkj8ov3ZedvnB8gZj3TSPy/bHSpNluz9Nlu3+NFlp1b7UtFBbtHRUdojqP6ofu7fs4aI7pvDZuu0sfvQ/9DwxJ2GN7bEXmqyy/QfJCDkv0mTbv2r602TZ7k+TFfj7Sl5i3byybB8L21m2+9Nk2e5PkxVESTWvUnZyaozpBfwNyMZ5iB8XkQfcdT8Avg+EgbkicrMx5ts4V+/XaAgwXERWxew3E3gW6A1sBS4Rkb2JvMS2gZz3yTwWFC7ilZ/Pp+yLMkxFC3pG8g7bLlq2x15osg6UHuDPd8zk82Jn2Ltkd26w/aumP02W7f40WV79eYkms2384s2xbmysm+Zx2f5c0mTZ7k+TZbs/TVY6RUlpXkWvJWs6RInI0zVf/+PMVd0Se2Lq6ufAayIyAHjN/blRCoVCnHbGafx7znO8NP9F2rZt22AUku2xF5qsnN49MLSojaw6uL+8wRgkTX+aLNv9abK8+vMSTWb9+HmIddM8LtufS5os2/1psmz3p8lKpygpP8qqDlFRuhx4pp5dXwCc6d5+CngD5yKppOUlCik6VgLwFHuRbJ3trOgYJKA2BmnI6EFW+NNk2e5Pk+XVn9brUXP8bH+N2P5c0mTZ7k+TZbs/TZZXf80hzavotaQyuzWmQ9TxOB2ilhpj3jTGnByn5FLgX/XsLts98QUowZk20CjlD8qncGshU867kLNOG0d1uJpLr7gsYY3tsRearJJtO2nfuQMAlRVV7N25L6kXhx/HwnZ/miyv/rRej5rjZ/trxPbnkibLdn+aLNv9abLSKUrKj0r5BVGxHaLcAP2aDlEn43SI6lsTJ2WMOQUoF5EGvwMTETHGNJhhFDsfrKqyiipTyc5tOxERpIWwfPf7FLXfWrtN7Hww22MvUsk6rP2rKeLF9+bykwm3UhODNLjfkITtX1PprzlZtvvTZCVTU1/b3aHfHMzCfyx2LkLIICWvx1SOX7q9Rmx/LmmybPd3JKyhJw3huunXU1JcQqdOnazzZzMrnaKkTDDntHFqZIeoGl1G/Z+aAuw0xuS4+88B4k0LqNMhat7Tr9ZZ5yUKyfbYC02WlxgkTX+aLNv9abK8+ouEIyxbUMBtM2/mvrl3E2qZwdKFBdb48+NrxPbnkibLdn9e67pndwdjuOSyizl7/Djad2jXYCdEv46F36Ok/ChrOkS5P7cALqH++aYAL+F0hoIkO0RN/PaEOuvK9h8kHI4AyUch2R57ocmy3Z8my3Z/miyv/tYXbCQzuwtdc7OIRIRQq5aU7S2zxp8fWbb702TZ7s9rXf6gfLZs2sLihYuZfMH57N61mzFnnWmNP9tZ6RQlFXSIapy8dIg6A/hMRDZH78gY8wTwJxEpAH6LMxVgGlCIczLbKH1e8jkP/s+jVFVWgUDnbp0ajEKyPSbHa53tESBexk/zuGx/rDRZydTE+1p7694idm7exc8m3kZGRganfuNUunXrlvArcD+OnybLdn/gj4gxTVa8ttuVoQrWrF3LDdfdSJt2rXk38hbvbnyrdn1sfKJfxqI5/QVqGlnVIUpE3hCRU+Psa7p7YoqIlIrIOBEZICJni8jnjfWWd0Iv7n7+Th5cfB+/fv5Ovjx4iOWLVySssT0mx2ud7REgXsZP87hsf6w0WV795ebmkHdcHmPPPosRJ49g7NlnWXNMfmXZ7g98GjGmyNq0dDNlpQe5esZ3uPD2CwhXhyn9rNQaf7az0ilKqgVGbdE7pqNQxhhaH9MagNZtWtG6TSu2rt+WsCY6VqJlq5ZqMTmNYXmp06rxWudl/DSPy/bHSpPl1V8oFGLbtm1MvWgqALtKdjY4N86P46fJst0f6L13psNYeKn7+P1PCFeHmfXL2cx/YCHVlWHm/WGBNf5sZ3n1F6hpdFSenH6xr4yy/Qe559o/8IuL7sS0aMGJI09IWGN7TI7XOtsjQLyMn+Zx2f5YabK8+nv5xVdo3bo1paWlSCTCq/MXNDg3zo/jp8my3R/4M2JMk9VvVF9OHHMC0x+/mkk/mUBmr0xyT8y1xp/trCBKqnml3r7UGPMsUHMm2BnY53aFwhhzKzANp63pD0XksD/zjDF9cC6YygJWAFeKSGVjvB0oPcATd/yVfXv2gwiVhyrJ7JH4L/JkYyXiReVoxOR4rbMtAqQpYr80j8v2x0qTlUxN7Ly4zcu3sCe0m/7j+vGLW25DIsJxQ/ManBvnx/HTZNnmz68RY5qs2NfIoq2LmP3eHJ6/6UXKy8tB4GtdTkzYptsvY9Gc/ppDzrXk/pJ6+1IRuTSqTenzwGwAY8xAnBipfGAC8IgxJl5uwz3A/SLSH9iLczLbKOX07gEYbnvyZn73Suqia0AvJsdrne0RIF5ivzSPy/bHSpPlpaZoww4+XbaJJc8upXW71mS0zKDk050Nzo3z4/hpsmz3B/6MGNNmrVixkkcem8H8hXNBxDp/NrOCKKnmVSoviCoWkZXu7S+AmvalQG3U1CV8lWl6AfCMiFSIyBbgU2BU9D7dmrHALPeup4ApjfWmFV2jzbI9YsNLnZfYL68s28fCdpaXmtOvHM35P5tE3pBenH/zRPKG9OLkqcPZtGxzwjo/jp8my3Z/EESMNQfLdn+arLSKklL8p6WUd4iCw9qX1uh0YKeIfOL+3BOI/khsO1Ens66ycKYBVCfY5jAd1r1lbxG7tuzmoesfIxKJMGzwsKSia4aPGMb119xAJBLhpGHD4sZKaLKaok6rxmvdgdID/P2ef7Hrs92Eq8L0zj+uwdgvzeOy/bHSZCVTE+8rxUVbF1HRr4pzep1LSfvdnDNoPGvXrEv49aMfx0+TZZu/+iLGNN47bRuL5mTZ7k+T5dVfoKaRevvSqFWXk7gTVMqUm5vDxPPO5fbbb8fQgrKyMp597tkG62qDaDGYjOT+gsjNzSF/0ECKinYQCUfICCX3tYAXltc6rRovdT375TJ5+iRmP/wi4eowB/YcYMeWYnL75KTEo81jkQ4sr/52luzkV3feRSQc4c033iIzM9Mqf35k2e5P873T9rHQZNnuT5Pl1Z+2NMPxtdQc7UsxxoSAC4HoM8IioFfUz8e690WrFOjs1te3TQ2jtn3pX/78ZJ113bO7c/a4bxKRMJWRQzz9z7/jdFGtX9Hzd+a8/DwfFKxMen5MY+tsZ2n6i4Qj/PvBOVz3m+nc9tebqayobHDemeZx2f5YabICf+nDst2fJst2f5os2/1psrz6C9Q0ao72pQBnAxtEZHvUfS8BlxljWrtX5A8AlkUXuZ2kXge+5d51FUm0L512zXfrrMsflE+/vv0o/GwbVZVVzJs7n6+P/kbC47F9fowmS9Ofl3lnXlm2j4XtrMBf+rBs96fJst2fJst2f5qsYM5p/H9aUm9fKiLzcK7Kr/OVvoh8aIx5DvgI50r/74tIGMAYMw+YLiI7gFuAZ4wxdwEf4JwAJ9S3XvlBnZ/7dOzFqfu/xiP3PUQoFGLY6JHc8ekjVGz8KpFq1nkP1amxvdWa1zrb2qvGm7Nb9PEOfjLhVgQh77g8+nbrn3DemVePtj9WtrMCf+nDst2fJst2f5os2/1psoL2pc0r9fal7rr/FpE/xan5tYj0E5ETRGR+1P0T3RNTRGSziIwSkf4icrGIVDTWWwtg5ownGHzTabT/QV9WrfiAMa2HJ6yxvdWa1zrb26v26JGNMfDCK7NZsvw9yssOWjXufn1eBP78zbLdnybLdn+aLNv9abLSqX1p7dxYhUVLKlfr26bCDVvpcWwO20N7MLSgy7Ae7F69w5lIUI+iW5kBnlqtJVunyTrSFoGp9mc7y3Z/mqzAX/qwbPenybLdnybLdn+aLK/+AjWN/NdWII5mnfdQnWXacVPJ69mL58+bwazzHuLKUReRcVDqbBMr21utJVv3ZvFrdZbSrjvZuGkD4yecwxmjz+TLynL6je9dZxu/jsXR4E+TFfhLH5bt/jRZtvvTZNnuT5OVTu1LW2DUFi1Z1b7UrcvDmXd6p4jcF2e/fTjC9qUAxcXFTJ44hUgkwsCBJ9Ips3PC7W1vtea1riaiZc+OUkSEFhktnInglvizneXVXyQSxrQwdeb6pvtYeDkmTX+2j58my3Z/mizb/WmybPenyUqn9qV+lDXtS6P0B2A+9euI25c6ERErePixh5j90iwKCgrSvtWa1zov7UH9Ohaa/nJ75pI/KL92rm/5wXIiDUy2t30svByTpj/bx0+TZbs/TZbt/jRZtvvTZKVT+1I/zjm1qX0pxpgpwBbgw3j7dGuOuH2pH2MvvNZ5aQ/q17HQ9Ddo8CCKthexfXsRh8oPsW/fPoaPSHxRnu1j4eWYNP3ZPn6aLNv9abJs96fJst2fJiudoqT8KGvalxqnk9QtwDeBn9azK0/tS2OlGSthW1RTrA6UHuDPd8zk8+K9AHTJ7txge1DbI0COhDV8xDCum349JcUldOrUqcn8xZu7O3jsQCaNPw+AzOwutBvRqs52sXFZto17UxxTKv01RY1fWbb702TZ7k+TZbs/TVY6RUkZH14+lPIjMsm3L70T5+v6hhPWj1CasRK2RzXl9O6BoQW//Put3DXrDg7uL2fNu+us8afJAuiR04NLLruYs8ePI6trwy00PfsLR1i+aGXtuJftP5j24+7lmFT9WT5+mizb/WmybPenybLdnyYrnaKk/Cib2peeAtxrjNkK/Bj4hTHmxphdNkn70uiIiJatWjY6ViLZGjjyqKbGsLzUFW7YRreeWXTNzaJjlw70HpjX4JxTTX+arO7Z3Sks3Mbbb77D1IumUlFRQXb3binx58dx93JMmv5sHz9Nlu3+NFm2+9Nk2e5Pk+XVX6CmkTXtS0XkdBHpLSK9gT8Cd4vIjOgikaZpX6oZK5E/KJ/CrYVMOe9CzjptHNXhai694rKUsLzUlWzbSfvOHQCorKhi7859DU56tj0CxGtd/qB8PvrwI6648goi4Qi7d+1mzFlnpsSfH8fdyzFp+rN9/DRZtvvTZNnuT5Nluz9NVjpFSfnxgihr2pcm0pG2L41VKmMl4s3DG/rNwSz8x2JngnUGLN/9PkXtt9auj52Hl8qIjcPag5oiXnxvbp32oIP7DUnYHjQjowVDTxpSZ26mTREgydY9tXFmnZ83L99Cp54d+Z8f/Q8iQpt2rXk38hbvbnyrdpurTrjak7+mGnebolea4phS6a8pavzKst2fJst2f5os2/1psoIoqeaVVe1Lo2rrZJw2dftSzViJSDjCsgUF3DbzZu6bezehlhksXViQEpaXuiFDBxOOhGvbgx7Yt7/Bmu7Z3cGY2rmZ7Tu0a/Drb6/+NMdi+/oidm3eTdtOx9CmQxu+LDvEnLteSok/L+Nue/SKl2PS9Gf7+GmybPenybLdnybLdn+arLSKklL8pyX/XeKVhDRjJdYXbCQzuwtdc7OIRIRQq5aU7U18zZftERv5g/LZsmkLixcuZvIF5yf19bemP691/Uf1I29IL679yzTO++m5dMnpTM8Tc6zxZzsr8Jc+LNv9abJs96fJst2fJiuIkmpeqURJ2aaayKDrr7mBSCTCScOGJR0z1FDNYV917i1i15bdPHT9Y0QiEYYNHka3bt0SftXpxV8qjyv262+AFp0Nm7Zv5qe3/IxWnVod9vU3HP4VeCrH3UtdrL9FWxdR0a+Kq064muX7C1jdcR09I3mHbaflL91Ygb/0YdnuT5Nluz9Nlu3+NFle/TWHWvjwBNqa9qXGmG8DP4vaxRBguIisitlvJs5V/r2BrcAlIrK3Md6qqqrYsH4D3bp3o7o6TDhS3XARUZOOMZiM5J4Mubk55A8aSFHRjtpWoaliea1rbM3m5Vto3a41LQ98SeWhKhrzSb+GP691VVVVvPGfN7h43YdUV4c5rneeVf7SgRX4Sx+W7f40Wbb702TZ7k+T5dVfoCOXNe1LReTpqPuvBLbEnpi6+jnwmogMAF5zf26UcnvmMuCE4/n3nOd4bvYzfLRufVKtI1esWMkjj81gzsvP80HByqTnxzS2znbW9vVFbP2gkKpDVUTCEfbu2Nfg3ExNf17rgudF4O9oYdnuT5Nluz9Nlu3+NFle/TWHTDDnNHmJh/alUboceKaeXV+A07YUPLYv9WPrSE2Wl7mZmv681gXPi8Df0cKy3Z8my3Z/mizb/WmygjmnzStr2pfG6FKck9B4yhaRYvd2Cc60gYSKF+808brxnD9xMpHqCG07HJN2rSM1WfHmZm5rt53nb3qR8vJyMiTU4NzMVPrzWufHlqLNyQr8pQ/Ldn+aLNv9abJs96fJSqv2pT48gbapfWnN9qcA5SLSYN9DN5RfvPga/PV8Jk+fxNAzBgOGHVuKE25ve6s1TVYkEmH1qtU88tgM5i+cSygjwyp/nlk+bCmqyQr8pQ/Ldn+aLNv9abJs96fJCtqXNq9S+slpEu1LR8Qpayigf6cxJkdEio0xOcCuetjXAtcC/PC3NzDx2xPqrN+7ex8fLV3P+G+PY8fmYtYv30hun/q/mo5uZQZ4arWWbJ3tLNv9ea2Lbr8J1LbfHDJ6kBX+bGcF/tKHZbs/TZbt/jRZtvvTZHn11xwyPkwFtaZ9qVvTAmcean3zTQFewmlbCkm2Lx14/kC2HthWu3z02Qae+v3TnPztkRTtL+HAvi8wWabONrGyvdWaJst2f8nWjckZV2fptD+TfjkDGJMzjlO6jKaitIqux3Srs41fxyLwd3SxbPenybLdnybLdn+arHRqX+pH2da+9AzgMxHZHH2nMeYJ4E8iUgD8FnjOGDMNKMQ5mW2UNr7zCZ+t207hB4VEIkLrdq3pe3KfhDW2t1rTZNnuz2vdgQNf8PprrzNy6Chq2m/m9sy1xt+RsIYMHcyUSVPJyMjgtNNPOyqfF7b702TZ7k+TZbs/TZbt/jRZ6dS+NJhz2gh5aV8qIm+IyKlx7p/unpgiIqUiMk5EBojI2SLyeWO9fVH6BdUV1RzT0WlTeSiFbSqDtm7pw/JjS1Fw2s0WF5dw9vhxnHzKyQzM/1qD7WZtf6z86E+TZbs/TZbt/jRZtvvTZKVT+1I/yn8TFZJQ0KbS3/40Wbb7A+jatSslxSWcfsbpSCTCq/MXNNhu1o9jYbs/TZbt/jRZtvvTZNnuT5OVTlFSLTBqi5aao0PUScCfgDY4Qf03iMgyt0PULTj9hr4ArheR1XH22wdnTmoWsAK4UkQqE3mJjTia98k8FhQu4pWfz6fsizJMRYuk2lTaHHuhybLdnybLNn/xWs2+fO9cTr1iFP/vF7cjEeEbl596WLvZeK1m030s4tUMHzGM66ZfT0lxCZ06dbLKnyZL019JcQmhliGmTJqKiNArr9dROxa2s2z3p8lKpygpP0q9QxRwL/C/bieoX7o/A2wBxojIYOBXwOP17Pce4H4R6Q/sBaY11lgoFOK0M07j33Oe46X5L9K2bVv27NmTsMb22AtNlu3+NFm2+9u8fAttO7UFEXoOzKVNxzaccvHJ1vjTZvXI6cEll13M2ePHkdU1s8Httf358XkbCoUYO+4sClYvY8mK96iqqmrw/VbzuGx/rDRZtvvTZAVRUs2rlH1y6gblF7u3vzDG1HSIEqCju1knYIe7zXtR5UuAY2P36SYAjAWucO96CrgTeLQx3rJ7ZLNnTykA1dXVVByqIKtLl4Q1tsdeaLJs96fJst1f0YYdfLLkUyoOVhJqFaKyvJJ59y9g4k3nWOFPk9U9uzuFhdso2r6D6d+bxt2/urvBubea/jRZmv68vN96Zdk+FrazbPenyUqrKKk0mHrQWKnMOY3pEPVj4HfGmM+A+4Bb45RMA+bHuT8L2Cci1e7P24lqiVqf3ix+rc5S2nUnGzdtYPyEczhj9Jl8WVlOv/G962wTK9tjLzRZtvvTZNnu7/QrR9PzxFwu+83FnHrpKNp0bNPgiammv1Synto4s86yslUBa9auIWdsdxYUzqeoZAf7+uw9bDs/joWmv6Z4v/XLWKQby3Z/mqwgSqp5lfL2pbEdoowxdwE3icjzxphLcLJQz47a/iyck9PTUuUpEo6QEcpgz45SRIQWGS2cyc8JZHvshSbLdn9e6yKRMKaFqTM3ziZ/XmpqvtYv31/OslkFHCo7xLLnCxh10ciU+Bt60pA6czptGoutKwvJ6pXJy/fOQ0Ro0641XfOyrPGnydL1l8EpE0ax8B+LEREyQhkNvt9qHpftj5Umy3Z/mqy0ipJSvFBJSyn95LSeDlFXATW3/w2Mitp+CPAEcIGIlMbZZSnQ2e0wBc5X/0X1sK81xhQYYwrmPf1qnXWhliFu/tNN3L/gHn4/7ze0PqYV776yJOGx2B57ocmy3Z/XutyeueQPyq+dG1d+sJxIAxPgbR+Log07+HTZJl749cu1f4gtn7OC0s/ivbyOjNU9uzsYUzuns32Hdkl9ba41FtvXF7Fr827adnIi5L5MIkJO058mS9NfJBxh2YICbpt5M/fNvZtQywyWLixICcv2sbCdZbs/TVYQJdW8SuXV+vV1iNoBjAHewJk/+om7fR7OSeuVIvJxvH2KiBhjXge+hXPF/lUk6BCFe1HVgu0v1/kzvWz/QTJCzpPsy4OH+LLsEFnZiedA2R57ocmy3Z/XukGDB1G0vYjt24vo1KEj+/btY/iI4db481Jz+pWj6XdyX95/dimjLhpJwQsr6XliDpuWbSarV/2fGnph5Q/KZ8umLZQfPMh1N3yPt954q8HIKq8sLzX9R/Vj95Y9XHTHFD5bt53Fj/6nwQg5TX+aLE1/6ws2kpndha65WVRWVBFq1ZKyvWUpYdk+FrazbPenyUqnKCk/zjlV7xAFXAM84H76eQi41l33S5w5pY+4A10tIiMBjDHzgOkisgMnbuoZd3rABzgnwI3SgdID/PmOmXxevBeALtmdGXvJmQlrbI+90GTZ7i/Zunhz3QaPHcik8ecBkJndhXYjWtXZLraFqW1jES8ObdHWRWxrt523Hn6X/fv207/bADIzM484Oi3e/MzKUAVr1q7lhutupE271g1GVtWwhg0/qU4Hq1SMRc04PH/Ti5SXl5MhoQYj5JJlNUWNJiuV/mJfI1v3FlH08Q5+MuFWajqv9e3WP2474OY4LtsfK02W7f40WUGUVPNKvUOUe/8IERkqIqeIyAp3++ki0iVq25FR+5ronpgiIptFZJSI9BeRi0WkorHecnr3wNCCX/79Vu6adQcH95ez5t11CWtsj73QZNnuzzMrHGH5opW1z4uy/Qd987xYvWo1Tz71BO8seYu1q9akhLVp6WbKSg9y9YzvcOHtFxCuDjc4faBGZWUHaztYnTluTFLH5GX8Vq9azSOPzWD+wrmEMjKsfKz89hru0SMbY6jtvFZedvCoHQvbWbb702SlU5SUXgS/Xt+mo7JDVOGGbXTrmUXX3Cw6dulA74F5Dc45jY6VaNmqpafYi2TrbGfZ7s9rXfC8OLKaj9//hHB1mFm/nM38BxZSXRlm3h8WNOgvFAqxbds2pl40FYBdJTsbnKvqx/HTZNnuT5Nluz9Nlu3+NFle/QVqGh2VJ6cl23bSvnMHACorqti7c1+DczZsj73QZNnuz2td8Lw4spp+o/py4pgTmP741Uz6yQQye2WSe2Jug/5efvEVWrduTWlpadLtVf04fpos2/1psmz3p8my3Z8mK52ipFoYo7ZoyZr2pW7NmcAfgZbAHhE57Ps9L+1LD5sDZYp48b25deZADe43JOEcKNtjLzRZtvtLti54XiSuaSgWKt6cztnvzamd04nA17qcWGe72Hmqm5dvYU9oN/3H9eMXt9yGRITjhuY1OFc1HcbPZpbt/jRZR+KvsdFpfh4Lv7HSKUrKj7KmfakxpjPwCDBZRPKBi+vZ7xG3Lx0ydDDhSLh2DtSBffvTPvZCk2W7P691wfPiK3mJherarSsrVqysndOJSNJRV0ueXUrrdq3JaJlByac7UxJ15dfHyo/+NFle/Xl9jfhxLPzISqcoKaP4T0upvCCqWERWure/ABK2L8VpSTpbRLa5Nbti9xnVvnSWe9dTwBQP3nwXe6HJst2fJst2f17ramKhFi9czOQLzmf3rt0NftXuNerq/J9NIm9IL86/eSJ5Q3px8tThbFq2uclZfn2s/OhPk+XVn9ZrxGudHx8rTVY6RUn5USnvEAVx25cuMMbch3Ny/A13s+OBlsaYN4AOwAMi8reYXXlqXxqr0j2lDB8xjOuvuYFIJMJJw4YlFSvR2Bq/smz3p8my3V+ydfFioVp0Nmzavpmf3vIzWnVq1eBX7clw6ou6quhXxTm9zqWk/W7OGTSetWvWNRh1ZdP4pRvLdn+arGRrtF4j2seVTv40WV79NYf8mHOa8guiYtuXAtfjtC/tBdzEVzmlIWAEMAk4B7jdGHN8KjxVVVWxYf0G2rRpTatWrQhHqhsuwnkCGON+uJ2R/JOhR04PXnhlNu3ateOzz7allOWlTqvGryzb/Xmp27x8C63btaZlq5DzyUGSKK/+dpbs5Fd33sXWzVt58423Gi7wyPLjY+VXf5osLzXarxGbxwKC33OBmlY2tS/dDiwQkYMisgd4Cxgas0tP7Uv/8ucn66zL7ZnLgBOO599znuO52c/w0br1SbWprJlPN+fl5/mgYGXS85JKSnby9N//Sd9+faioqGjU3L3GsLzUadX4lWW7P69129cXsfWDQqoOVREJR9i7Y1+DrT5tHwvb/WmybPenyfLqL3iNfKXg91zzygRzTpNXEu1LIap9KU4b0tOMMSFjTFvgFJx5qrUSEQFex2lfCg20LxWRkSIycto1362zLrpN5aHyQyltUxnMS/I3y3Z/Xuv6j+pH3pBeXPuXaZz303PpktO5wVafto+F7f40Wbb702R59Re8Rr5S8HsuUFPLmvalIrLeGPMqsAaIAE+IyDqAI21fqtmmsilaOvq1rZtX1vARw+rEtdjEsv2xSrauKVp92j4WtvvTZNnuT5OVbE3wGvlKwe+5QKmWNe1L3ZrfichAERkkIn+Mur9J25dqtqn00tLRr23dvNb1yOlRG9eS1TWzwe01WbY/VkfCamyrT9vHwnZ/mizb/WmyjrQtcPAaCX7PNbdq58YqLFo6KjtEabap9NLS0a9t3bzUdc/uTmHhNt5+8x2mXjQ1qblMmizbHytNVuAvfVi2+9Nk2e5PkxX8njtyf4GaRkfFyemYnHF1lk77M+mXM4AxOeM4pctoKkqr6HpMtzrbxCrZVmZXnXB1nWXalGlMnnw+b7/5Fg/+8QH6D+jP2G+Mq7ONV1ZT1NnWdu6pjTPrLCtbFbBm7RpyxnZnQeF8ikp2sK/P3jrbNCfL9sdKkxX4Sx+W7f40Wbb702QFv+eO3F9zqIXiPy01R/vSoTjtS9sDW4FvuxFTGGNuxen4FAZ+KCKH/enlpX1prD7/fC+vznuVRQsWIQg9e/Ykt2fiHuBH0spsZ8lOJk+cQnl5uTPJugH5ta2bl7qtKwvJ6pXJy/fOQ0Ro0641XfOyrGHZ/lhpsgJ/6cOy3Z8my3Z/mqzg99yR+wvUNGqO9qVPAD8XkcHAHOBnAO66y4B8YALwiDEmXm7DEbcvHTFyOCNOHsHyVUt5f/m7fFn+ZVJRUl5brTW2paNf27p5qdu+vohdm3fTttMxtOnQhi/LDjUY16LJsv2x0mQF/tKHZbs/TZbt/jRZwe+5I/fXHArmnDZCCdqXHo+TYQqwCLjIvX0B8IyIVIjIFuBTvspABZqufalmlFQQvXJkdV7iWjRZtj9WmqzAX/qwbPenybLdnybLdn+arCBKqnnVHO1LP8Q5EX0BuBjo5W7WE4i+Kilea1JP7Us1o6RidTRHr3ipa4q4Fk3WkYyF32KrAn/pw7LdnybLdn+aLNv9abLSKUpKMxxfS83RvvS7wA3GmBVAB6BR80WbQppRUkH0ypGzGhvXosk6krgRv8VWBf7Sh2W7P02W7f40Wbb702SlU5SUH5XST07jtS8VkQ3AeHf98cAkd/MivvoUFeK3Jq1tX+p+epqwfSluwP8Pf3sDE789oXZddJQUUBslNWT0oHqPJTpWAvAUe5Fsne0s2/1psrz6q4mtKtq+g+nfm8bdv7q7UbFVjWHZPhaBP32W7f40Wbb702TZ7k+T5dVfc0hzLqiW1NuXGmO6u/+3AP4fzpX7AC8BlxljWrtX5A8AlkXv02v70nt++ju1KKmmqLOdZbs/TVayNUdDbFXgL31YtvvTZNnuT5Nluz9NVjpFSflRzdG+dIAx5vvuz7OBmQAi8qEx5jngI5wr/b8vImE48valsTpw4Atef+11Rg4dhSDkHZeXsiipIG7E3yyv/rzEVkUiYUwLw5RJUxEReuX18sVYeKnzMha2P5c0Wbb702TZ7k+TZbs/TVY6RUkFc04boQTtSx8QkePd5ecSFYgmIr8WkX4icoKIzI+6v0nblw4ZOphwJMwLr8xmyfL3OLBvf9rHXmiybPenyfLqz0tsVW7PXPIH5VOwehlLVrxH+cHyBiPQNI9L87HyMha2P5c0Wbb702TZ7k+TZbs/TVY6RUn5USpX69smP8ZeaLJs96fJ8uqv/6h+7N6yh4vumMJn67az+NH/NBhbFR2B1qlDx6Qi0Lx6tP2x8jIWtj+XNFm2+9Nk2e5Pk2W7P01WOkVJ+fGTU2s6RLlxU+uBje4ulojIdXH2mwk8C/R26y8Rkb2N8VZSXEKoZajOV4LpHnuhyfIyfpr+NFnJ1niJrWqKCDQbx8JLXRAH1/w1fmXZ7k+TZbs/TVY6RUn5UdZ0iHK1KWoKwGEnpq5+DrwmIgOA19yfG6VQKMTYcWfVfiVYVVXFnj17EtbYHnuhyfIyfpr+NFlH4q/RsVUeItA0j0v1sQri4HztT5Nluz9Nlu3+NFlpFSVljN6ipJR9cioixUCxe/sLY0x9HaIWALc3YtcXAGe6t58C3sC5SCppZffIZs+eUgCqq6upOFRBVpcuCWtsj73QZHkZP01/mixNf14i0DSPy/axsP25pMmy3Z8my3Z/mizb/Wmy0ilKyo9KeQg/1NshCup2iALoY4z5wBjzpjHm9Hp2l+2e+AKU4EwbaJTyB+VTuLWQKeddyFmnjaM6XM2lV1yWsMb22AtNlpfx0/SnydL0V7JtJ+07dwCgsqKKvTv3JZVvF4yFvj/bWbb702TZ7k+TZbs/TVYQJdW8SvkFUbEdoowx3wUeNMbcjpNtWtMhqhjIE5FSY8wI4AVjTL7bVSquRESMMVLf+hrFm6829JuDWfiPxc6k5wxYvvt9itpvrV0fO1/N9tiLVLK8jB80zRjaNhZN5S+ZKKTY8dtqinjxvbn8ZMKt1ESgDe43JO480+Y4rlQ+Vk0xFrY/lzRZtvvTZNnuT5Nlexyc5nEFUVLNq5R+clpfhygRGS8iI4B/AZvc+ytEpNS9vcK9//g4u91pjMlx958D7KqHfa0xpsAYUzDv6VfrrIuEIyxbUMBtM2/mvrl3E2qZwdKFBQmPxfbYC02Wl/HT9KfJ8urPSxSSlwg0zePSfKyCODh/+9Nk2e5Pk2V7HJzmcQVRUs0razpEGWO6GWMy3Nt9cTpEbY6z65dwOkNBkh2ioluXAqwv2Ehmdhe65mYRiQihVi0p21uW8Hhsj73QZHkZP01/miyv/qKjkA6VH/JFFFLgL31YtvvTZNnuT5PVXHFwyb4HemX5PkrKGLVFS9Z0iALOAP7PGFMFRIDrRORzAGPME8CfRKQA+C3wnDFmGlAIXNKQkcO+EtxbxK4tu3no+seIRCIMGzyMbt26JfxKsHRPKcNHDOP6a24gEolw0rBhScdeNLbONlZTjF8q/TUnK9maeFMjJl43nsmTLiBcHabvoD5JRSH5YSwCf83Pst2fJst2f5qsVPprivdAzePyOhaBmkapvFr/Hah3IsQDcbZ/HmcKQLx9TY+6XQokPgtqQLm5OeQPGkhR0Q4i4QgZoeQ+qq/96wGDyUj+LwgvdTazvI6flj+AqqoqNqzfQLfu3aiuDhOOVFvlDyAjI4Os7EzC1WH2797Pji3F5PZJHMRv8/MCoEdODx5+bAaXX/xtPvtsG1ndErdk1fZn+/hpsmz3p8my3Z8mS9Ofl/dAryzNsdCWCeac+kNdu3VlxYqVPPLYDOa8/DwfFKxMav5JY2v8yrLdHzjzmQaccDz/nvMcz81+ho/WrU+qvaWWv0g4wr8fnMN1v5nObX+9mcqKyqTmPds87t2zu1NSspOn//5P+vbrQ0VFBdndu1njz/bx02TZ7k+TZbs/TZamPy/vgZrH5XUsAjWNjsqTU9vn4tjOst0f6M3p9OrPj/Oe8wfls2XTFhYvXMzkC85n967djDnrTGv82T5+mizb/WmybPenydL0Z/u1C2k151Txn5ZS2b60DU7YfmuXM0tE7jDG9AGeAbKAFcCVIlJpjGmN0+50BFAKXCoiW+PsdwLOtIAM4AkR+W1jvdXMJblu+vWUFJfQqVOntG+1psmyzV+8uUyg094y2Zp483aLPt5RJwqpb7f+Dc57tmncn9o487C6ylAFa9au5YbrbqRNu9a8G3mLdze+Vbs+to1rKv01RY1fWbb702TZ7k+TlUp/TfEeqHlcQfvS5lUqPzmtAMaKyFDgJGCCMeZU4B7gfhHpD+wFprnbTwP2uvff725XR+7V/A8D5wIDgcuN0xK10eqR04NLLruYs8ePI6trZoPb295qTZNluz/Qa2/p1V+PHtkYQ20UUnnZwbQf901LN1NWepCrZ3yHC293LnIo/azUGn+2j58my3Z/mizb/WmyNP15eQ/UPK50al/qx6v1U3ZyKo5qPqNv6S4CjAVmufc/BUxxb1/g/oy7fpw5fCRGAZ+KyGYRqcT5BPaCxnrrnt2dwsJtvP3mO0y9aGpSc+OiW5m1bNXSU6u1ZOtsZ9nuD+q2t+zYpUNte0tb/PmR9fH7nxCuDjPrl7OZ/8BCqivDzPvDAmv82T5+mizb/WmybPenybLdnybLq79ATaNUh/BnuDFSu4BFOMH6+0Sk5tLp7UBP93ZP4DMAd/1+nK/+o1W7TZz6evXUxpl1lpWtClizdg05Y7uzoHA+RSU72Ndnb51tYmV7qzVNlm3+xuSMO2zptD+TfjkDGJMzjlO6jKaitIqux3Srs40fx0KTddUJV9dZpk2ZxuTJ5/P2m2/x4B8foP+A/oz9xrg62zTnWNg2fs3Jst2fJst2f5os2/1pstKpfWkw57SREpEwcJIxpjMwB/haKnnJauvKQrJ6ZfLyvfMQEdq0a03XvMSRN7a3naup+/rXT2XHtmJamdY8++xzvDz3Zav8abEOHPiC1197nZFDR9XOZ8rtmWuNP7+ydpbsZPLEKZSXlzsXFCQh29sR+pFluz9Nlu3+NFm2+9NkpVP7Uj9K5Wp9EdkHvA58HehsjKk5KT4WKHJvFwG9ANz1nXAujIpW7TZx6uvIRLUvffu5d+qs276+iF2bd9O20zG06dCGL8sOMeeulxIeg+1t58CZrnDFt7/N10efQveeXZnz4vM4/Qzs8KfJ0mpvmQ5jodnuryZ6Zf7CuSBinT+bx0+TZbs/TZbt/jRZtvvTZKVT+1KbPjk1xvQyxrxujPnIGPOhMeZHXo4ple1Lu7mfmGKMOQb4JrAe5yT1W+5mV/FV+9HotqTfAv4jh3/0shynw1QfY0wr4DK37jBFty89/ZLT6qzrP6ofeUN6ce1fpnHeT8+lS05nep6YOPjX9ggQcKJ8+vXtR+Fn26iqrGLe3Pl8ffQ3rPHnR5bt/jRZgb/0YdnuT5Nluz9Nlu3+NFnpFCVlmaqBn4jIQOBU4PteLlxP5df6OcBT7hX2LYDnROQVY8xHwDPGmLuAD4C/uNv/Bfi7MeZT4HOcE0+MMbk4kVETRaTaGHMjsAAnSupJEfmwISOxc93mfTKPBYWLeOXn8yn7ogxT0YKekbx658SBfREg33rlB4fV9enYi1P3f41H7nuIUCjEsNEjuePTR6jYWFm7zazzHlLx11R1trO8+ispLiHUMsSUSVMREXrl9Tpqx0LT30nDhjL1vAsREVq1asWqD1Zb40+TZbs/TZbt/jRZtvvTZKVTlJTmVfQNSUSKgWL39hfGmPU41wZ91Jj9pPJq/TUiMkxEhojIIBH5P/f+zSIySkT6i8jFIlLh3n/I/bm/u36ze/8OEZkYtd95InK8iPQTkV978RYKhTjtjNP495zneGn+i7Rt25Y9e/YkrLE9AgScB3PmjCcYfNNptP9BX1at+IAxrRMHz9seN2I7y6u/UCjE2HFnUbB6GUtWvEdVVVXaPwdt9weQk5vDm+++TsHqZdz8i59RXFzMmtVrrPEXPFb6LNv9abJs96fJSqcoKVtljOkNDAOWNrY2pRdE2arsHtns2eNMZ62urqbiUAVZXbokrImOlQA8xV4kW+eVVbhhKz2OzWF7aA+GFnQZ1oPdq3fAADv8+ZHl1Z8fn4O2++ue3Z2dO3fRtl1bAEp2FJPRIgMamEflx7Gw3Z8my3Z/mizb/WmyvPrzu4wx1wLXRt31uIg8Hme79sDzwI9F5EBjOUdl+9L8QfkUbi1kynkXctZp46gOV3PpFZclrLE99gJg7569dOzaiZ7tswE4Pq8/23Z8lrDGjxEgmiyv/vz4HLTdX/6gfLYVbqOwcBsXT7mEPz3yOKNP/wZDhg62wp8my3Z/mizb/WmybPenyQqipOL/i76ex13inZi2xDkxfVpEZns5Jmval0bVXYQTwn+yiBTE2W+j25fGa2/5rR9PYWCnIfTp3QeAdXtXUVS8tXZ9bA6mbbEXsXNHARa1XMTsWS+w495PiEQiHBrYjuMyc3nsvJ+r+ztaWMnWxHsODv3mYBb+Y7Ez8T4Dlu9+n6L2W2vX2/4cbIqaVLLijfmQswcx9Xxnzmn/oX1Zt2kd/3rvH+T2+eqCyKYYd9vGIt38abJs96fJst2fJiuIkvImt3nSX4D1IvIHr/uxqX0pxpgOwI+oZ35CU7YvnTLuQroNzmRjxloe+tcfWfbu8oTb2x57UVO3YsUKHn7sIWa/NIuCggKrIjb8yPLqLxKOsGxBAbfNvJn75t5NqGUGSxce9rdYk7BsHwstVuyYHzxQTo/e2axfvtEKf5os2/1psmz3p8my3Z8mK62ipOxqXzoauBIYa4xZ5S4TGyqKlU3tSwF+hXPyeqie3TZJ+9IvD1TQuWUXisq3UVlRxUfLN9Cpe8eGjsfq2AtNlu3+NFle/a0v2Ehmdhe65mYRiQihVi0p21uWsMavY6HFKtywjS7dO9O2Y1tCLUMMPX0IHy5ZT3av7lb402TZ7k+TZbs/TZbt/jRZQZSUN4nIOyJi3IvhT3KXeY3dT0oviHI/6VwB9Mf5xLPe9qXGmOFALxGZa4z5WT27jNe+9JSGfDy04oU6Px9b2ZUDH3zJY48+RiiUwbDRI3nvmK28vuLj2m3GnFf3qz3bYy80Wbb702QlWxP7VfGnuwvZvHYLPz33FwhCz5496dutf9y2qkfiL9XHZbO/2LGsXLOIdqYdt1/8fyDQqlUr+vTtzfcvPDyWTcNfc7Js96fJst2fJst2f5qstIqSUmwrqqWUXhAlImEROQmnk9Mo6mlfaoxpAfwB+Ekq/dSodddj+NvMpxh80+iURi55rbOdZbs/TZZXf716HcvE885l+aqlvL/8Xaqrq4MoKQV/n37yKS+8PJv3l79Lhw7tOTavV8IaTX+aLNv9abJs96fJst2fJiuIkmpeqURJicg+Y8zrRLUvdT89rWk/2gEYBLzhzmnoAbxkjJkcc1FUo9qX4sYdDL7+G+Sd89V5sVbkktc621m2+9NkBVFSgb90ZNnuT5Nluz9Nlu3+NFnpFCUVfHLaCDWmfamI7BeRriLSW0R6A0uA2BNT8Ni+NPrEFPQil7zW2c6y3Z8mK4iSCvylI8t2f5os2/1psmz3p8lKpygpP8qm9qVx1RTtS2Njl+Yxj3t/cx/dP+5OJFzNF8e14rjspo9c8lpnO8t2f5qsIEoq8JeOLNv9abJs96fJst2fJiudoqRsal/aVLKmfWlM7Zk1n5qmon1pbs9cBpwwgOfmPMOzz/+Ldes+JNLARGfbYy80Wbb702QFUVKBv3Rk2e5Pk2W7P02W7f40WekUJeVHHZXtSwcNHkTR9iK2by+iU4eO7Nu3j+EjEl8QZXvshSbLdn+arKaIkqqsqAqipAJ/qizb/WmybPenybLdnyYrnaKk/Djn9Kg4OY33lerE68YzedIFhKvD9B3Uh3YjWtXZLvYr1dI9pQwfMYzrr7mBSCTCScOGJR170dg621m2+9NkJVsT+3zaureIXVt289D1jxGJRBg2eBjdunVrMErKD2MR+Gt+lu3+NFm2+9Nk2e5Pk+XVX6CmUSoviGpjjFlmjFltjPnQGPO/7v19jDFLjTGfGmOedS9swhhzf1Q3gY+NMfvq2e8IY8xat/5B43GyRUZGBlnZmWRlZ7J/9352bClO5picBYPJSB7rpc52lu3+vNb1yOnBC6/Mpl27dnz22baUcXJzc8gfNJDKykoqKyrJCCX3dZHt4x74Sx+W7f40Wbb702TZ7k+T5dWftoziPy1Z075URG6q6SYAPATMrme/jwLX4AQ/DQAmNNZYJBzh3w/O4brfTOe2v95MZUVlUvP9VqxYySOPzWDOy8/zQcHKRrQUbVyd7Szb/Xmt657dnZKSnTz993/St18fKioqyO7ezRp/trMCf+nDst2fJst2f5os2/1psrz6C9Q0sq19aY0uB/4Ve6cxJgfoKCJLRESAv9VTn1B+bB2pybLdn9e6/EH5bNm0hcULFzP5gvPZvWs3Y8460xp/trMCf+nDst2fJst2f5os2/1pstJqzmnNJ7wKi5asaV8aVXMc0Af4T5xd9nRranRYfTwdDa0jNVm2+0u27qmNMw+rqwxVsGbtWm647kbatGvNu5G3eHfjW7XrrzrhajV/TVHjta6kuIRQyxBTJk1FROiV18sqf7aPn+0s2/1psmz3p8my3Z8mK53al/pRVrQvjdFlwCwRSdmzwI+tIzVZtvvzWrdp6WbKSg9y9YzvcOHtzsVypZ+VWuNPkxUKhRg77iwKVi9jyYr3qKqqSvvXiO3+NFm2+9Nk2e5Pk2W7P01WerUvNYqLjmxpXxqty4Dv17OrIremRkm1L53x6ENMu+a7tev82DpSk2W7P691H7//CeHqMLN+OZvqqjDVlWHm/WEBV95/hRX+NFl+fI3Y7k+TZbs/TZbt/jRZtvvTZKVT+1I/yor2pVE1XwO6AO/H26eIFAMHjDGnulfpfye6Pmbb2val/c87jjeLX6tdSrvuZOOmDYyfcA5njD6TLyvL6Te+d51tYmV7qzVNlu3+kq276oSr6yzTpkxj8uTzefvNt3jwjw/Qf0B/xn5jXJ1t/DoW0c99v75GbPenybLdnybLdn+aLNv9abKC9qXNK9val14GPONe7FQrY8wqd3oAwA3AX4FjgPnu0ih1aN2RF198kf17DiAi9MztybqS1eyj/q9wbW+1psmy3d+R1O0s2cnkiVMoLy8n5mnY7P50WRmcMmFUbXvVjFBGg+NxJP6GDB3MlElTycjI4LTTTwuet8FrOBiLZmDZ7k+TFbQvbV5Z1b5URO4UkcMa3EedmCIiBe7++onIjbEnssmovPogH5QtY3ObDWw5ZiNhCfPaf+Jdf/WVbG+1psmy3d+RsGqiQ+YvnAsi1vnTYmm2V+2e3Z3i4hLOHj+Ok085mYH5X0sqwsvm8bOdZbs/TZbt/jRZtvvTZAXtS5tXKb0gylZ9sa+M8rIvAejUogufbf+M1p1bJqyxPfZCk2W7P02W7f681mnGrXXt2pWS4hJOP+N0JBLh1fkLUhLh5dfHyo/+NFm2+9Nk2e5Pk5VWUVI+DOFP2df6xpg2wFtAa5czS0TuMMbcCPwY6Ad0E5E97vZfA2YCw4HbROS+evbbB3gGyMKJqbpSRCoTebnllSfq/BzZfYiKl4ugrJrMzpkcN6QfG046AK+8V7vNkmsOb19qc+yFJst2f5os2/0lWxevvWrRxzv4yYRbEYS84/KaJG4tXnzXy/fO5dQrRvH/fnE7EhG+cfmpKYnw8stjdTT402TZ7k+TZbs/TVYQJdW8ao4OUe8CZwOFMdt/DvwQiHtSGqW4HaYaI5PVGgO0++8BvPjqy6z9cB2R0oqENbbHXmiybPenybLdn9e6Hj2yMQZeeGU2S5a/R3nZwZT427x8C207tQUReg7MpU3HNpxy8ckpOSa/PlZ+9KfJst2fJst2f5qsdIqSCj45bYTcuaCHdYgSkQ/g8Am8IrIL2GWMmVTfPt0r9McCNdk+TwF34rQ0TVqRkkOYzq04Y8ipbNq/DQa0I7ypjBZZreutsT32QpNluz9Nlu3+NFleaoo27OCTJZ9ScbCSUKsQleWVzLt/ARNvOscKf35l2e5Pk2W7P02W7f40WUGUVPMqpXNOjTEZxphVwC5gkYgsPcJdZtFAh6mkVFaF6RBiQv8zePXTtzHtQ0hZVcIS22MvNFm2+9Nk2e5Pk+Wl5vQrR9PzxFwu+83FnHrpKNp0bNPgiammP7+ybPenybLdnybLdn+arHSKkgralzZSbpenk9y80znGmEEisi6VzHhacs2/6vy8aMEiZs+aw/3fv4twOMyZ+YPIzM/kF9ccFhRQK6+xEpFIGNPC1GkDebRGbPiRZbs/TZaXmpqv9cv3l7NsVgGHyg6x7PkCRl000gp/fmXZ7k+TZbs/TZbt/jRZ6RQl5UepXK0vIvtwwvcnHOGuSnE7TLk/J+wQZYwpMMYU/OXPT9ZZVxMZ9PBjDzHn5ef5oGBlymIlcnvmkj8ov7YNZPnBciINTKr2a8SGH1m2+9Nkeakp2rCDT5dt4oVfv4yI0CKjBcvnrGiwbawfx0+TZbs/TZbt/jRZtvvTZKVTlFQw57QRMsZ0A6rc1qU1HaLuOZJ9ioi4bVC/hXPF/lUk6BAFPA5wKFwuMevUYiUGDR5E0fYitm8volOHjuzbt4/hI4anhGV7xIYfWbb702R5qTn9ytH0O7kv7z+7lFEXjaTghZX0PDGHTcs2k9Urq9n9+ZVluz9Nlu3+NFm2+9NkpVOUlB/VHB2ifgjcDPQA1hhj5onIdGNMD6AA6AhEjDE/BgaKyAFjzDxguojsAG6h/g5TSSmVsRLx2joOHjuQSePPAyAzuwvtRrSqs11sPI9fIzb8yLLdnyYrmZp4LWAXbV3Eob6VLPnzcg6WHWTaxdNYu2Zd3G1rVFJcQqhlqM50mXQfP02W7f40Wbb702TZ7k+TlU5RUpqfaGqpOTpEPSgix4pISERyRWS6e3+Je39HEens3j7grpvonpgm7DCVrFRjL8IRli9ayS//fit3zbqDsv0HWfNu4mm3fo3Y8CPLdn+arCOJXvn0k00MGpxP/uD8pLYPhUKMHXdW7XSZqqoq9uzZkxJ/wWPlb5bt/jRZtvvTZKVTlJQfldILomyVZqxE4YZtdOuZRddc5yvK3gPzePeVJQwZPajJWbZHbPiRZbs/TZZXf6FQiG3btvHDm37A35/6B7tKdjbYvjS7RzZ79jjzUqurq6k4VEFWly4p8Rc8Vv5m2e5Pk2W7P01WOkVJaV5Fr6Wjsn2pZqxEybadtO/cAYDKiir27tzX4BPJrxEbfmTZ7k+T5dXfyy++QuvWrSktLU26fWn+oHwKtxYy5bwLOeu0cVSHq7n0istS4i94rPzNst2fJst2f5qsdIqS8qNsal/aCfgHkOduf5+IHNbr0BgzAvgrcAwwD/iRG/iftFIZ73RYG0hTxIvvza3TBnJwvyEJ20D6NWLDjyzb/WmykqmJbV+6efkW9oR2039cP35xy21IRDhuaN5h7Ut7d8w7jDf0m4NZ+I/FzoULGbB89/sUtd9auz72NWb7+GmybPenybLdnybLdn81dUOGDmbKpKlkZGRw2umnWTUWzaFgzmnj1Nj2pd8HPnK3PxP4vTGmVZz9PgpcAwxwl0bHU2nGOw0ZOphwJFzbBvLAvv1HbcSGH1m2+9NkHUmU1JJnl9K6XWsyWmZQ8unOBqOkIuEIyxYUcNvMm7lv7t2EWmawdGFBk/vzWmc7y3Z/mizb/WmybPcH0D27O8XFJZw9fhwnn3IyA/O/1uA0IL9HSflRqbwgSkQkbvtSEdkarwTo4LYobQ98DlRHb2CMyQE6isgS99PSvwFTGustOt7pUPkhX8Q7abJs96fJst2fJstrlNT5P5tE3pBenH/zRPKG9OLkqcPZtGxzwrr1BRvJzO5C19wsIhEh1KolZXvLEtbYPn6aLNv9abJs96fJst0fQNeuXSkpLuH0M05PehpQECWVfkrpBVFujNQKoD/wsCRuXzoDeAnYAXQALhWRSMw2PXFaltYoqfal8eKdJl43nsmTLiBcHabvoD5JxTsNHzGM66+5gUgkwknDhiUde9HYOttZtvvTZNnuT5OVTE19UVIV/ao4p9e5lLTfzTmDxjcYJbV1bxG7tuzmoesfIxKJMGzwMLp165Zwuozt46fJst2fJst2f5os2/zFTgMCePneuXz9v07h1/fcTUVZBad86+TDpgHFvndojkVzKLggqpESkbCInITTyWmUMab+S9ThHGAVkIszDWCGMaZjqrwN/no+v5/7G3J696Bk6052bClusKa2vywGk5H8k8FLne0s2/1psmz3p8ny6m9nyU5+deddbN28lTffeKvB7XNzc8gfNJDKykoqKyrJCCX3dZvt46fJst2fJst2f5osm/3VtDvuktvZOSEzYFrYNxaBjlw2tS+9GpjtTgf4FNgCfC1mmyKcE90aJdW+dN7Tr8YFvjH7bXL75NCuU1vWL9+Y8BhqWp4+8tiMpFueeq2znWW7P02W7f40WYG/9GHZ7k+TZbs/TZbt/qLbHVdXViMRSbrdsdZYNIeM4j8tpezk1BjTzRjT2b1d0750Q4KSbcA4d/ts4ASgzsQzESkGDhhjTnXnpn6HBO1LRWSkiIyc+O2658Rf7CujeGsJHy1dz8izR3Bw/0Gye3VPeDzpMBcnmJekz7LdnyYr8Jc+LNv9abJs96fJst2f1znqwZzT9JM17UuBXwF/NcasBQxwS1TM1Cp3egDADXwVJTXfXRJq64FtdX4u2bSTf/38OUwLw6drNtPymJa0z29fZ7sxOXX3YXurNU2W7f40Wbb702QF/tKHZbs/TZbt/jRZtvmrb476tnbbeevhd9m/bz/9uw0gMzMz4Rx1v7cvRfETTS3Z1L50h4iMF5HB7vb/iNrXSVG3C9z1/UTkRveq/Ubp4N6D5I89kR//+0am3j4ZIsKOjYnnnNreak2TZbs/TZbt/jRZgb/0YdnuT5Nluz9Nlu3+aupWr1rNk089wTtL3mLtqjVWjUWgptFR2b50x8Zitn5QyBPXzqSqspqKLyt5/5mlXHTHlHprbG+1psmy3Z8my3Z/mqzAX/qwbPenybLdnybLdn+arLRqX9rcBlKgo6J9ae+OeXWWK79/OXf96w46dOxA9ZdVZPXI4ie//2GdbWJle6s1TZbt/jRZtvvTZAX+0odluz9Nlu3+NFm2+9NkBe1Lm1fN0b70aWAkUAUsA74nIlXGmG8Dt+D8EfAFcL2IrI6z3z7AM0AWTobqlSJS2Vh/G1Z8TFVFFce0P4YDpQfYsaWY3D459W6fLm3dghZ3uizb/WmyAn/Nxxp60hCum349JcUldOrUyTp/NrNs96fJst2fJiut2pf68KKt5mhf+jRORNRgnIuaprvbbwHGiMhgnIujHq9nv/cA94tIf2AvMK2xxiLhCP9+cA7X/WY6d/7zNtq0ax20PvSRP02W7f40WYG/5mF1z+4OxnDJZRdz9vhxtO/QLiXtHL3W2c6y3Z8my3Z/mqygfWnzqjnal85z1wnOJ6fHutu/JyJ73e2XUDfPFAA3PmosMMu96yk8tC8NWh/6258my3Z/mqzAX/Ow8gfls2XTFhYvXMzkC85n967dKWnn6LXOdpbt/jRZtvvTZKVXlJRRXHTUbO1LjTEtgSuBH8UpnUb8iKgsYJ+IVLs/J9W+NLal4ae7C9m8dgs/PfcXCELPnj3p261/g60PvcRKlBSXEGoZYsqkqYgIvfJ6pV2cR7r502TZ7k+TFfjTYcVr6VgZqmDN2rXccN2NtGnXOql2jn4Yi6PBnybLdn+arPSKkvKfmrN96SPAWyLydnSNMeYsnJPTW1Llq1evY5l43rksX7WU95e/S3V1NXv27ElY4zVWIhQKMXbcWRSsXsaSFe9RVVWVMlYQN6LPst2fJivw1zysTUs3U1Z6kKtnfIcLb7+AcHW4wY45fh0LP/rTZNnuT5OVTlFS/vvcVClKSkT2GWNex2lfus4YcwfQDfhe9HbGmCHAE8C5IhLv3bUU6GyMCbmfniZsXwpcCzDj0YeYds13a9dl98hmzx5n99XV1VQcqiCrS5eEx+A1VkKTZXvEhh9ZtvvTZAX+mof18fufEK4OM+uXs6muClNdGWbeHxZw5f1XWOHPdpbt/jRZtvvTZKVTlJQfpd6+1BgzHTgHuFxEIlHb5wGzca6+/zjePt15qq8D33Lvuook2pdGn5iCM0ercGshU867kLNOG0d1uJpLr7gs4fF4jZXQZNkeseFHlu3+NFmBv+Zh9RvVlxPHnMD0x69m0k8mkNkrk9wTc63xZzvLdn+aLNv9abLSK0rKf5+dNkf70mqgEHjfjT+Y7XaP+iXOnNJH3PurRWQkgDFmHjBdRHbgfN3/jDHmLuAD4C8NGXmz+LXD7hv6zcEs/MdiZ9JzBizf/T5F7bfWro+df5psrIQmK1a2R2z4kWW7P01W4E+HFTt/dNHWRcx+bw7P3/Qi5eXlIPC1LicmbOfol7FIN3+RSBjTwtS5BuFoHQvbWekUJeVHNUf70pDbevQkd6m5f7qIdIm6f2TUvia6J6aIyGYRGSUi/UXkYhGpaKy3SDjCsgUF3DbzZu6bezehlhkpi5LSZNkeseFHlu3+NFmBv+ZjrVixkkcem8H8hXNBxDp/NrM0/eX2zCV/UH7tNQjlB8uJJHGRjR/HwnZWECXVvDoq25dGR0lVVlSlNEpKk2V7xIYfWbb702QF/tKHZbs/TZamv0GDB1G0vYjt24vo1KEj+/btY/iI4Slh2T4WtrPSKUrKjyH8R8XJaezX5lv3FrFry24euv4xIpEIwwYPo1u3bg1GSQ0fMYzrr7mBSCTCScOGxY2V0GQ1RZ1WjV9ZtvvTZAX+0odluz9NVir9xZvmNfG68Uye5KQq9B3Uh3YjWtXZLt7vBj+MRbqxvPoL1DRK5QVRbYwxy4wxq40xHxpj/te9/2ljzEZjzDpjzJNu3inGmJ8ZY1a5yzpjTNgYkxlnv32MMUuNMZ8aY541xrRqrLfc3BzyBw2ksrKSyopKMkLJfVRvjHEWDCYjub9UNFle67Rq/Mqy3Z8mK/CXPizb/WmyNP0N/no+v5/7G3J696Bk6052bClOGcv2sbCd5dVfoCOXNe1LReR3NfNNgVuBN0Xk8zj7PeL2pdFztOa8/DwfFKxs1LyuZGv8yrLdnybLdn+arMBf+rBs96fJ0vRXozdmv01unxzadWrL+uUbU8KyfSxsZx3J4xvoyGVN+9IYXQ78K/bOpmpf6sf5MZos2/1psmz3p8kK/KUPy3Z/mixNf1/sK6N4awkfLV3PyLNHcHD/QbJ7dU8Jy/axsJ2VVnNOFf9pybr2pcaYtjhh/TfG2aWn9qWxqplLct306ykpLqFTp05p32pNk2W7P02W7f40WUfib9jwk5gyaSoZGRmcdvpp1vnzG8t2f5qsVPrbemBbnZ93b9vDv37+HJHqCJ+s3kTrdq1pn9++znZjcprvuGx/rDRZQfvS5pV17UuB84F36/lKv8nUI6cHl1x2MWePH0dW18Omth4m21utabJs96fJst2fJutI2v2VlR3k7PHjOPmUkzlz3BhrjsmvLNv9abI0/R3YeYAWLVrw3zOu5MLbLyCSRKtZzeOy/bHSZKVX+1L/fXKa0pPTGonIPpzOThMAotqX/k+czS8jzlf6rmrbl7o/J2xfaowpMMYU/OXPT9ZZ1z27O4WF23j7zXeYetFUKioqyO7eLeExRLcya9mqpadWa8nW2c6y3Z8my3Z/miyv/kKhENu2bWPqRVMB2FWyMyWvR9vHT5Nluz9Nlqa/6Faz8x9YWNtqNhUs28fCdpZXf4GaRta0L3W36wSMof6WpIKH9qWhMwxPbZxZu6xsVcCatWvIGdudBYXzKSrZwb4+e+tsEyvbW61psmz3p8my3Z8mK5ma6NdYzTLj6YeJZIR5Ye1sPvtiG/984ZmUvB5tHz9Nlu3+NFmp9HfVCVfXWaZNmcbkyefz9ptv8eAfH6D/gP6M/ca4Otv4dSzSjZVe7Uv9J5valwJMBRaKyMHoHR1p+9JYbV1ZSK9Bx/LGzLf5YvcBMkIZdM3LSlhje6s1TZbt/jRZtvvTZHmp2bx8C+06t2XwOYOZ/8BCJCIcNzQvJa9H28dPk2W7vyNhDT1pSJ3rCWzyB7CzZCeTJ06hvLzcuegmCQXPC31W0L60eWVN+1J33V9F5LI4+2rS9qVFG3awe+tuDn1xCNOihfPVyv2Jv1qxvdWaJst2f5os2/1psrzUFG3YwafLNrHk2aW0bteajJYZlHy6s8F5eH4cP02W7f681nXP7g7G1F5P0L5DuwaniNjeavZIWDY/Vraz0ql9aW0eq8KiJZU5p7bp9CtHc/lvL6F7326c9l9fp03HNky86ZyENbbHXmiybPenybLdnybLS83pV47m/J9NIm9IL86/eSJ5Q3px8tThbFq22Qp/fmXZ7s9rXf6gfLZs2sLihYuZfMH57N61mzFnnWmNP9tZtvvTZKVTlJQflbKv9Y0xbYC3gNYuZ5aI3BG1/kHguyLS3v25NfA3YATOhU+XisjWOPudADwAZABPiMhvG+uturKav/34nxzT8RiWz1lJ1aGqBmtsj73QZNnuT5Nluz9NVjI18ebULdq6iG3ttvPWw++yf99++ncbQGZmZr3z7zSPya8s2/0lWxdvPnJlqII1a9dyw3U30qZda96NvMW7G9+qXR/7vPLLWBwN/jRZQZRU86o5OkRhjBkJdInZfhqwV5zOT/fjdIKqI3f+6sPAucBA4HJjzMDGGitctY0BX+/Hdx/5Duf+eDwZGS3YsTFxCznbYy80Wbb702TZ7k+TdST+Vq9azZNPPcE7S95i7ao11vnzG8t2f17rNi3dTFnpQa6e8R0uvN3pX9/QFBG/joUf/Wmy0ilKyo9K2Sen7pX1h3WIck8wfwdcgXMBVI0uAO50b88CZhhjjNSdMT4K+FRENgMYY55x6z5qjLcdG4vZ+kEhT1w7k6rKaiq+rOT9Z5Zy0R1T6q2JjpUAPMVeJFtnO8t2f5os2/1psgJ//7nRawAAIWlJREFU6cOy3Z/XuuiopuqqcG1U05X3X2GFP9tZtvvTZHn11xzSzB/VUkrnnBpjMowxq4BdwCK3Q9SNwEsiEvtRZU/gMwC3A9R+nI5QcbdxlVSHqN4d8+osV37/cu761x106NiB6i+ryOqRxU9+/8M628TK9tgLTZbt/jRZtvvTZAX+0odlu79k65oiqskvY3E0+NNkBVFSzauUti8VkTBwkpt3OscYcwZwMXBmKrnJaMOKj6mqqOKY9sdwoPQAO7YUk9snTt84V7bHXmiybPenybLdnyYr8Jc+LNv9HUkdQDgc5n9/+X98Xvo5w4YPs8af7Szb/Wmygiip5pV2h6izgP7Ap8aYrUBbY8yn7mZFQC8AtwNUJ5wLo6JVu42rpDpEzXv61TrrIuEI/35wDtf9Zjp3/vM22rRrzdKFBQmPwfbYC02W7f40Wbb702QF/tKHZbs/r3Xds7tTUrKTp//+TwYNzqdb967WRUnZzLLdnyYrnaKkwCguOtLuELVCRHqISG8R6Q2UuxdAAbyE0/EJnA5Q/4mZbwqwHBhgjOljjGmF0+r0pXj86A5RE789oc669QUbyczuQtfcLCIRIdSqJWV7y+LtJnp/VsdeaLJs96fJst2fJivwlz4s2/15rQuipPztT5MVREk1r9Q7RCXY/i/A391PUj/HOfHEGJOLExk1UUSqjTE3AgtwoqSeFJEPGzIyJmdcnZ+37i2i6OMd/GTCrQhC3nF59O3W/7DtomV77IUmy3Z/mqwj8Td8xLA6nWyO5rEI/OmybPeXbF0QJdX8NX5lpVOUlB9Pn9U7RMVs0z7q9iG341N/twPUZvf+HSIyMWq7eSJyvDhdpn7txVuPHtkYAy+8Mpsly9+jvOxg2sdeaLJs96fJOpK4kR45PWo72WR1zUzJMWkeV+AvfVi2+/NaF0RJ+dufJiuIkmpepfSCKFvlx9gLTZbt/jRZXv11z+5OYeE2irbvYPr3pnH3r+5ucG6cX8ci8KfPst2f17ogSsrf/jRZaRUl5cOpB0dl+1I/xl5osmz3p8ny6i9/UD4fffgRV1x5BZFwJKm5cX4di8CfPst2f17r+o3qy4ljTmD641cz6ScTyOyVSe6Judb4s51luz9NVhAl1bxSb19qjPkrMAYnxxTgv0VklTHmAuBXQASoBn4sIu/E2e8I4K/AMcA84EdxLpxKKD/GXmiybPenyUq2JnZu3OblW8gZ2IM77rqDL3YfICOU0eDcOL+MReCv+Vm2+0u2LvY1smjrIma/N4fnb3qR8vJyEPhalxMTtsL1y1gcDf40WekVJRV8ctoY1du+FPiZiJzkLqvc+14DhorIScB3gSfq2e+jwDXAAHeZUM929cqPsReaLNv9abK8+ivasIPdW3dz6ItDmBYtnK8f71+QEpbtYxH402fZ7u9IWCtWrOSRx2Ywf+FcELHOn80s2/1pstIrSsp/SuUFUSIih7UvTbB9WdQnoO3ibWuMyQE6isgSd9u/AVM8ePNd7IUmy3Z/miyv/k6/cjSX//YSuvftxmn/9XXadGzDxJvOSQnL9rEI/OmzbPenybLdnybLdn+arHSKkopNIk3loqWUXhDlxkitwAnef1hElhpjrgd+bYz5Jc6npT8XkQp3+6nAb4DuwKQ4u+yJ07K0Rkm1L41VTYzP9dfcQCQS4aRhw5KKlWhsjV9ZtvvTZHn1B/DGk29xxlWnsf2juH0kmoxl+1gE/vRZtvvTZNnuT5Nluz9N1pG8twc6cqX0gigRCbtf0x8LjDLGDAJuBb4GnAxkArdEbT9HRL6G82nor1LpzRjjLBhMRnJ/D3ip8SOrqqqKDes30KZNa1q1akU4Um2VP22Wl5rNy7e4VxK/SsELK2nTvnXKWF7rgsfK3yzb/WmybPenybLdnybLqz99+e+zU+32pRNEpNj9yr8CmAmMirP9W0BfY0zXmFVFOCe6NUqqfelf/vxknXXR85LmvPw8HxSsbNRcpmRr/MrK7ZnLgBOO599znuO52c/w0br1RJL4i9KPY+HV3/b1RWz9oJCqQ1VEwhH27tjHnLviNjs7YpbtYxH402fZ7k+TZbs/TZbt/jRZXv0Fahql8mr9bkCViOyLal96jzEmR0SKjRPMNQVY527fH9gkImKMGY5zlX+d9GS37oB7YdVS4DvAQ/H4IvI48DjAoXC5xKzz3fwYTdagwYMo2l7E9u1FdOrQkX379jF8xHBr/GmyvPrrP6ofu7fs4aI7pvDZuu0sfvQ/9DwxJyUs28ci8KfPst2fJst2f5os2/1pstJqzmkaeGys1NuXGmP+4564GmAVcJ27/UXAd4wxVcCXwKU1F0gZY1a50wMAbuCrKKn57tIo+bHVWipZbxa/dljd4LEDmTT+PAAys7vQbkSrw7aLbQfrxV9JcQmhliGmTJqKiNArr5dV455sTbzIm23tttdG3mRIiJ6RvISRN7Y9LwJ/6cuy3Z8my3Z/mizb/Wmy0ql9qR+l3r5URMaKyGD3vv+quaJfRO4RkXxx4qW+LlEZp1EnpohIgVvbT0RujLrCP2n5sdWaJisSjrB80Up++fdbuWvWHZTtP8iad9elxF8oFGLsuLMoWL2MJSveo6qqij179qSEpf1YrV61ujbyJpSRkf7Pi8Bf2rBs96fJst2fJst2f5qsoH1p8ypoX4o/Wq1psgo3bKNbzyy65mYB0HtgHu++soQhowc1OSu7RzZ79jizO6qrq6k4VEFWly4Ja7yy/PhYabICf+nDst2fJst2f5os2/1pstKpfakfFbQv9UmrtVSyxuSMq7N02p9Jv5wBjMkZxyldRlNRWkXXY7odtp0X1pvFr9VZSrvuZOOmDYyfcA5njD6TLyvL6Te+d51tNMeiKWr8ygr8pQ/Ldn+aLNv9abJs96fJCtqXNq+ao33pWOA+oBVOBuo0EamOqjsZeB+4TERmxdlv0L60mVkHDnzB66+9zsihoxCEvOPyyO2ZuH+1d38ZnDJhFAv/sRgRISOUQTIPt+0t7iKRMKaFqTOXNt2fF4G/9GHZ7k+TZbs/TZbt/jRZ6dS+1KjG4+tIu33pN4CncE48BwGFwFU1Be7FU/cACxPsN2hf2sysIUMHE46EeeGV2SxZ/h4H9u1Pmb9IOMKyBQXcNvNm7pt7N6GWGSxdWJASluZjldszl/xB+bVzacsPljcYx2X78yLwlz4s2/1psmz3p8my3Z8mK2hf2rzSbl8aBipF5GP3/kU4V+nX6AfA88CuePsM2pfawdL0t75gI5nZXeiam0UkIoRataRsb1mDdbaPRXQc16HyQ0nFcQXPC3/702TZ7k+TZbs/TZbt/jRZ6RQl5ccQftX2pcAyIGSMGSkiBcC3gF7utj2BqcBZON2j4qnJ2pf6LfZCk5VKf7FzVbfuLaLo4x38ZMKttVMI+nbrH3dOa3McV7I1XuK4miKKK9XHFfhLT5bt/jRZtvvTZB2Jv+EjhnHd9OspKS6hU6dOR+1YBGoaqbYvBfKBy4D7jTHLgC9wPk0F+CNwi4hEUukJ/Bl7ocnS9NejRzbGUDuFoLzsoD/GwkMcV/C88Lc/TZbt/jRZtvvTZB1JfFKPnB5cctnFnD1+HFldM1NyTJrHlU5RUv773LR52pe+LyKni8gonAumar7iHwk8Y4zZivOJ6iPGmCkxu2qS9qXREREtW7VsdKxEsjV+ZdnuT5Pl1V90HFfHLh1q47hSwbJ9LAJ/+izb/WmybPenyfLqr3t2dwoLt/H2m+8w9aKpVFRUkN29W0pYto9FoKZRyk5OjTHdjDGd3ds17Us3GGO6u/e1Bm4B/gQgIn1EpLeI9AZmATeIyAvR+xSRYuCAMeZUt/3pd4AX4/FF5HERGSkiI6dd89066/wYe6HJst2fJivZGi9xXH4di8Bf87Ns96fJst2fJivZmqc2zqyzrGxVwJq1a8gZ250FhfMpKtnBvj5762zj17GwQcYYtUVLzdG+9HfGmPPc+x4Vkf80tKOmbl/qx9gLTZbt/jRZXv15iePy61h4qfNjFJcmy3Z/mizb/WmyvPrburKQrF6ZvHzvPESENu1a0zUvKyUs28ciUNOoOdqX/kxEThSRE0Tkj/XU/nd0xmlTty/1Y+yFJst2f5osr/68xHH5dSy81PkxikuTZbs/TZbt/jRZXv1tX1/Ers27advpGNp0aMOXZYeYc9dLKWHZPhbNI//NOj0q25f6MfZCk2W7P02W7f40WZr+oqO4OnXo6IsoLk2W7f40Wbb702R59dd/VD92b9nDRXdM4bN121n86H/oeWJOSli2j0WgplHKT07dr/ULgCIROc8YcyPwY6Af0E1E9rjb/Qz4dpSvE931n8fsrw/wDJCFE1N1pYhUNsaTHyNANFma/kqKSwi1DNX5+vZoHQvbWan0dzREcWmybPenybLdnyYr2ZqrTri6zs+Lti5iW7vtPH/Ti5SXl5MhIXpG8g7b7kj9pfq4msJfc8iPp88aV+v/CFgf9fO7wNk43aFqJSK/E5GT3K/wbwXejD0xdXUPcL+I9Af2AtMaa8iPESCaLE1/oVCIsePOqv36tqqqij179qSEZftY2M5S9efDKC5Nlu3+NFm2+9NkHYm/1atW88hjM5i/cC6hjIyjdiwCNY1SHcJ/LDAJ+DXwPwAi8oG7LlHp5cC/4uzPAGOBK9y7ngLuxGlpmrSiIyKARsdKJFvjV5amv+we2ezZUwpAdXU1FYcqyOrSJSUs28fCdpamv+goLqA2imvI6EFW+LOdZbs/TZbt/jRZtvvTZHn11zzy32enqf7k9I/AzUDSwfrGmLbABJw2prHKAvaJSLX7s6cOUX6MANFkafrLH5RP4dZCppx3IWedNo7qcDWXXnFZSli2j4XtLE1/Jdt20r5zBwAqK6rYu3NfgzEnto+fJst2f5os2/1psmz3p8lKpygpPypln5y6cVG7RGSFMebMRpSeD7xbz1f6TSI/RoBoslLpL97cwqHfHMzCfyx2JqhnwPLd71PUfmvt+niZoH4Yi3RjeYl3SpZ1WFtbU8SL782t09Z2cL8hCdva2j5+mizb/WmybPenybLdnyYrnaKkNPNHtZTKT05HA5Pdjk/PAGONMf9Iou4y4nyl76oU6GyMqTmp9tQhyo8RIJosTX+RcIRlCwq4bebN3Df3bkItM1i6sCAlLNvHwnaWl3gnryw/RnFpsmz3p8my3Z8my3Z/mqz0ipLyn1L2yamI3IpzYRPuJ6c/FZH/SlRjjOkEjAHibiciYox5Hae96TPAVSToEAU8DnAoXC4x63wXAaLJ0vS3vmAjmdld6JqbRWVFFaFWLSnbW5YSlu1jYTvLS7yTpj/bx0+TZbs/TZbt/jRZtvvTZAVRUs0r9ZxTY8wPceah9gDWGGPmich0d/VUYKGIHIypmQdMF5EdOC1PnzHG3AV8APylsR5K95QyfMQwrr/mBiKRCCcNG5ZUrERja/zKSqW/w76+3VvEri27eej6x4hEIgwbPIxu3bol/PpW87hsf6xSyYo3BWPideOZPOkCwtVh+g7q02C8Uyr9NUWNX1m2+9Nk2e5Pk2W7P02WV3+BwBgzAXgAyACeEJHfNnYfGlFSiMgbInKee/tBETlWREIikht1YoqI/FVEDrvaRUQmuiemiMhmERklIv1F5GIRqfDiqUdOD16e/yJzF7zMmePGpKzGrywtf7m5OYw9+yxmvfAcs1+axdizz0oZS7PGr6zBX8/n/lfv4bcv/IoWLVqwY0uxVf5sHz9Nlu3+NFm2+9Nk2e5Pk+XV39EsN9v+YeBcYCBwuTFmYGP3o3Jyapu6Z3enpGRn7c+7SnaS3b1bk9f4lWW7P02W7f60WTVq2/4YBpzUj/XLN1rjLx3GLxgLfZbt/jRZtvvTZB3pe6CmjOK/JDQK+NT9ILESZwrmBY09pqPy5DR/UD7bCrfVRkS8On8BY846s8lr/Mqy3Z8my3Z/mqwv9pVRXvYl4MQ7bVzxCdm9ulvjz/bx02TZ7k+TZbs/TZbt/jRZXv0FoifwWdTPniI/jYg0vJUPdcIJJ0wE/hgOhztnZGQ8sHHjxl+noqamLhwOz8zIyCgDnrSRZbs/TZbt/jRZjak54YQThuA0xsgIh8NZGRkZj23cuPH/UumP4DXsmWW7vyNh2e5Pk2W7P02Wlj8/yxhzLXBt1F2Puxeg16z/FjChZsqmMeZK4BQRubFRIBE5qhegwNaawF8wFs3Nst1fMBbBWKSjv2As/D8WR+sCfB1YEPXzrcCtjd3PUfm1fqBAgQIFChQoUKAm13JggDGmjzGmFU52/UuN3Yl6lFSgQIECBQoUKFAg/0lEqo0xNwILcKKknhSRDxu7n+Dk1A3qt7RGk2W7P02W7f40Wbb702TZ7k+TFfhLH5bt/jRZmv6OWonIPGDekezjqL0gKlCgQIECBQoUKJB9CuacBgoUKFCgQIECBbJGwclpoECBAgUKFChQIGsUnJwGChQoUKBAgQIFskZH5cmpMSbTGJPZ3D5SJWPMcAVGR2PMCGNMl1SzXF7XRm7fxRjTMVV+AqWfjDHZxpjh7pJ9BPtp35S+6mE0+v3JGDNZg2WM6W+MuaihftnGmFDU7fbGmJHJsowx3Ywxw4wxQxoz3uny3q7xHu1yPL9PN+Y9V3vcvYxfss/bQJaouQNbFYNh83B6vO4GPgE+BXa59/X2uM+19dzfy93v28AvgJZR616op+ZrwHxgLtAP+CuwD1gGnJjAw/CYZQROu7BhwPB6ar4bdftY4DWX9R5wfD01/wC6urfPAbYBi4FC4OIE/j4HngDG4V6Al8S4ngtsAd5xj+NDYJN7XOMS1OUCfwP2A2HX4zbgzujHIEF9dtQ4Zh/Bc629wvM500PNZA0W0B+4CBjYwHah6DEDRibLArq5z40hDY03cBKwBFjvPmcXAxvc++K+RhrY37YE6wa7+/0M5wrfLlHrltVTM9r19iFwCrDIfb5/Bny9npoLY5aLgJKanxP4+39RtwcCH7uvta04XVzi1bwe9dq/0q15AlgL/KCemv8GSt1tzwU247zPfAZcnsDfQPfx+RSoBJa6/v4KdKqnxvb39ka/R7t1mu/TjX7P9TLuyuPX6Oet19dwsDT90uwG1A4U3gcuBTKi7svACYhdkqAu9pdA9C+D3fXULAKuw/ml+JD7ZpLlrvugnpq3gPOBy903kssA4973WgJ/EXf/r0ctX7r//6eempVRt5/DaUXWAphaH4uoN2uX19u93RVYncDfRuBG4F2gCHgAOLWBx2oVcCJOp4nSmu3d+1YmqPsPcGbU43Y/0A64C6fFWn11J6Fw8uL1TY/g5IUYT409eVkVzztwan3PXeB/6ll+AnyewN87wASgM/BT9zHr5677oJ6aZe5z4+vAHuA09/7hwLv11FQBrwBPAjPd5Qv3/ycT+It+7c8FznVvjwLeq6dmXdTt5Xz1XtYWWFNPzVqc94Y+wIGoMciur8ZdvwQ4IcrTU+7ta4BZ9dTY/t7e6PfoOI9Vqt+nV9HI91wv4648fo1+3np9DQdL0y/NbkDtQOETj+uqcH7xzYyzfFFPzaqYn/+r5gme4IX+QdTtT2PWJTohuwh4E/eXjHvflgbGYmUCrx/UU/Mh0NG9/Q7QInpdkqw84GZgJc7JyN1J1HyWaGxj1q2O+XlF1O0NCepWoXDy4vVNj+DkJbrOy8lLotf3p/Xcfwj4FXBHnGVfI56DZ+F8qnQqyb3219f3eMTcfzLOyfz1Ufdtqc9XPY/vB/X5iL0f6Onefh1o497OoJ7XPlGvU2BHzLpEj2/s+EX7XV9Pje3v7Y1+j45z7LHc+h6rpnifTuo918u4K49fo5+39TwHG3wNB0vTL0dTCP8KY8wjwFM4n86A8xXDVThP4vq0BrhPRNbFrjDGnF1PTUtjTBsROQQgIv8wxpTgdExoV09NRtTtP8Ssa1WfORF53hizAPiVMea7OCdHUt/2ro41xjyI88lsN2NMSxGpqvFeT83/Aq8bYx7G+RT038aYl3BeuK8mYJkor9uAe4F7jTFfw/mrO572GWO+B3QE9hpjbsL55OBsoCwBa7cx5r9w3oguxPm0D2OMIfH86nYisjT2ThFZYoyp7/G6G/gdUB1nXX2sDiJSM1b3GWNWAK8aY64k8WPWUkTWAhhjdovIO66/lcaYY+qp+QbwW2C5iDzq1p4pIlcn4MQqV0Tmu6xlCVhVxpieIlKE8/gcdO+voO7zOlphEdkD7DHGlInIJpez03m46tUxIrIxytOf3Nt/Nsb8Tz01840xc3GmfES/9r9D/c/dlThfM66IXWGMmZ7IoDGmk4jsd329boy5CHgeqG9OXvTz5daYdXFf+yKy3BjzTeAHxpjXgVto+HUP0Nd93Rqc94G2IlLurqvvtX8TsNAY8zzOicR/3Pec03BO5OJpmzHmN0AHYIMx5vfAbJzXcHECf5uMMbfjfAtyIc4fjhhjWlL/68rq93aP79Gg+z7t5T3Xy7hrjp+X5y3g6TUcqKnV3GfHWgvOm/z1OC/Qte4yH7gBaJ2g7nQgr551I+u5/yZgTJz7hwGL6qn5HnHmzuHM3/tjksc4DOfEbFcD210Vs3Rx7+9BPZ9mRnm5B5gDvAw8CpzTAOsPHh6rXsBj7v57uOO5DueTvETzb/Nw3lDX4cy9ynHvzwIuSlD3oLvvS3FO6r7h3p4LzKin5j1gRD3rPqvn/tXEfPWMM2fyE6A0gb/VUbenxKxbl6CuBfAj9zkxCticxNjvw+mD/DLOXLK2DbGAM3He/P8PmOGOzR04X+H9tJ6al4DfuNv/B/g9zvSFO4AFCfzNBm53t/097qfAOL+sNyaoOxf4k3tcL7u3JybY/gTcqQpx1tU7Hxm4gjjTVtzn5p/rqZkcPc5R9/cDbk7iMct1n/fJPL5jYpb2NccEfD9BXSec98/7cb6OvQX4WoLtO+KcaP8cZz7xRTif5D9c87qsp64zzh+wrwC/xvmDroYfdzoQlr+3x9muwfdod1vN9+lGv+d6GXfN8fPyvHVrGv0aDpamX4IOUT6T+ylhBxE50Nxe0knGmHOBC4Ce7l1FwEvitGGLt/0JOCeUe+KsyxaRnXHuvwLnBGJJzP15wO0ick09rMnAYvnqE66a+/vhnHTf28Cx5QJ/xPmF27eBbcfE3LVCRMrcq9u/JSIP11PXCedN/XictsjbgRdFZEM923cEvo/zCcgMnIs3rsaZb32XiMT9dM0Y0xnnQoqBOCf7vxWRL1z+ibFjGyiQbQreo49MwfgdJWrus2MbFuA8rTqtmsBf09QFi78X4FqNGk1W4K9OjdXvTbb7s30sjuCY1J6DweJtOSpzTuPoZMU6rRpNlu3+PNcZY661tcavLE1/RM2JTnGNJivw95Vsf2+y3Z8mS/OYNJ+DgTzoaLogCvcinHhf3d7R1HVaNYG/pqlLtEuLa/zKanJ/7vOiJ7BURKIv8ChsyhpNVuCvTs0oQMS5UGwgTirGhiTeLxpdp1WTDqw4+/ibiHzH0prTcOberxORxxJsdwpOMsQB9yLQn+Mko3yEcyFsIAUdNZ+cGmNuwQn/NTjRPMvc2/8yxvy8Keu0agJ/TVPXgCotrvErq0n9GWN+CLwI/ABYZ4y5IGp13F82Xmo0WYG/OjV34FzU+KhxEgJm4Fz5/XNjzG0JWI2u06pJE9ZLMcvLwIU1PzdnjVu3LOr2Ne4xdQDuaOD3wZNAzfz+B3AuqrrHvW9mgrpATanmnlegteCEfR/WJQjnisNEeW2NrtOqCfw1TV0Dz5t6uwE1d41fWU3tD+cq4pqr0nsDBcCP3J8/aKoaTVbg77CaDJxc3QN8lfN5DIkzVRtdp1WTJqyVOKkoZ+IkP5yJExM2hjhX5GvWxD5fcPKXu7m321FPBzB3/fpodsy6VfXVBUvTLkfT1/oRnMiV2K+Gctx1TVmnVRP4a4I6Y8ya+lbhROw0W41fWZr+cILIywBEZKsx5kxgljHmOOqfCuClRpMV+PtK1SISBsqNMZvEvYpbRL40xiR6v/BSp1WTDqyROFF1twE/E5FVxpgvReTNBBytGoAWxpguON8QGxHZ7R7TQWNMvHzqGq0zxlwtIjOB1caYkSJSYIw5HqdxQyAFHU0npz8GXjPGfMJXgcF5OJlwNzZxnVZN4K9p6rJxooz2xtxvcDI7m7PGryxNfzuNMSeJyCoAcaKxzsP5+m5wE9ZosgJ/X6nSfNVMYETNncaJF0t0QualTqvGepaIRID7jTH/dv/fSQPnFFo1rjoBK3DeG8QYkyMixcaY9iT+A2k68IAx5v/hdOR73xjzGc7vlIQNOAI1oZr7o1vNBecvqFNxAqEvcm9npKJOqybw1ySsv+C2BI2z7p/NWeNXlrK/Y4Ee9awb3VQ1mqzAX5376wt87woMTsBqdJ1WTTqw4mw7iQTNAZqzJqa+LdAnie06AkNxTtbrbbwRLKlZghD+QIECBQoUKFCgQNboqLlaP1CgQIECBQoUKJD9Ck5OAwUKFChQoECBAlmj4OQ0UKBAvpQxpqye+/9qjPlWE7P+2xgzoyn3GShQoEBHq4KT00CBAgUKFChQoEDWKDg5DRQoUNrLGPM/xph17vLjmHXGGDPDGLPRGLMY6B61bqsx5l5jzFpjzDJjTH/3/m7GmOeNMcvdZbR7/yhjzPvGmA+MMe8ZY06I42WSu03X1B51oECBAvlTR1POaaBAgXwoY8wI4GrgFJz8wqXGmOiA7qnACcBAnJzUj3ByM2u0X0QGG2O+A/wROA+nbeH9IvKOMSYPWACcCGwATheRamPM2TgtNS+K8jIV+B9goojEZrEGChQoUKAkFJycBgoUKN11GjBHRA4CGGNmA6dHrT8D+Jc4HXB2GGP+E1P/r6j/73dvnw0MNKY2q7ujG97dCXjKGDMAEKBl1H7G4nSzGS9uh51AgQIFCtR4BSengQIFOtolcW63AE4VkUPRG7oXPb0uIlONMb2BN6JWbwL6Asfj9IQPFChQoEAeFMw5DRQoULrrbWCKMaatMaYdztf4b0etfwu41BiTYYzJAc6Kqb806v/33dsLgR/UbGCMOcm92Qkocm//d8x+CnG+4v+bMSbf89EEChQo0FGu4OQ0UKBAaS0RWQn8FVgGLAWeEJEPojaZA3yCM9f0b3x1AlqjLsaYNcCPgJvc+34IjDTGrDHGfARc595/L/AbY8wHxPnmSUQ2AN8G/m2M6dcEhxcoUKBAR52C9qWBAgU6amWM2QqMFJE9ze0lUKBAgQI5Cj45DRQoUKBAgQIFCmSNgk9OAwUKFChQoECBAlmj4JPTQIECBQoUKFCgQNYoODkNFChQoECBAgUKZI2Ck9NAgQIFChQoUKBA1ig4OQ0UKFCgQIECBQpkjYKT00CBAgUKFChQoEDWKDg5DRQoUKBAgQIFCmSN/j8oerpsTlQCdgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# First:\n",
    "# Help from GPT\n",
    "df1 = pd.DataFrame({'Real': y_val, 'Predicted': predictions, 'chol': X_val[:, 0], 'oldpeak': X_val[:, 1]})\n",
    "df2 = df1[df1['Real'] == df1['Predicted']]\n",
    "heatmap = df2.groupby(['chol', 'oldpeak']).size().unstack(fill_value=0)\n",
    "plt.figure(figsize=(12, 10))\n",
    "sns.heatmap(heatmap, annot=True, cmap='Greens')\n",
    "plt.show()\n",
    "# Second:\n",
    "datas = df[['target' , 'oldpeak' , 'chol']]\n",
    "correlation_matrix = datas.corr()\n",
    "plt.figure(figsize=(12, 10))\n",
    "heatmap = sns.heatmap(correlation_matrix, annot=True, cmap=\"PiYG\")\n",
    "plt.show()\n",
    "# Third:\n",
    "columns = ['chol', 'oldpeak']\n",
    "correlation_matrix = df[columns].corr()\n",
    "plt.figure(figsize=(12, 10))\n",
    "sns.heatmap(correlation_matrix, annot=True, cmap='Blues')\n",
    "plt.show()\n",
    "# Fourth:\n",
    "predictions = model.predict(X)\n",
    "df3 = pd.DataFrame({'Predicted': predictions, 'chol': X[:, 10], 'oldpeak': X[:, 12]})\n",
    "df_correct = df3[df3['Predicted'] == y]\n",
    "heatmap = df_correct.groupby(['chol', 'oldpeak']).size().unstack(fill_value=0)\n",
    "plt.figure(figsize=(12, 10))\n",
    "sns.heatmap(heatmap, annot=True, cmap='Greens')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "N5E-qQQv2TgC"
   },
   "source": [
    "Train your model once by removing the chol feature and once by removing the oldpeak feature and calculate its accuracy. Compare the obtained results with the previous part."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1lKo-l15nI5E"
   },
   "source": [
    "Remove Chol and test model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "id": "Gb21jrDZjy2m"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  87.805 \n",
      "\n",
      "TP:  85\n",
      "FN:  14\n",
      "FP:  11\n",
      "TN:  95\n",
      "\n",
      "recall:  0.8586\n",
      "precision:  0.8854\n",
      "f1_scores:  0.8718\n"
     ]
    }
   ],
   "source": [
    "# print(X)\n",
    "idx = feature_list.index('chol')\n",
    "new_X = np.delete(X, idx, 1)\n",
    "# print(new_X)\n",
    "train_count = int(new_X.shape[0] * 0.8)\n",
    "X_train, X_val, y_train, y_val = new_X[:train_count], new_X[train_count:], y[:train_count], y[train_count:]\n",
    "\n",
    "model2 = MixedDataNB()\n",
    "model2.fit(X_train, y_train)\n",
    "predictions = model2.predict(X_val)\n",
    "\n",
    "print(\"Accuracy: \", round(accuracy_score(y_val, predictions) * 100, 3),\"\\n\")\n",
    "cm = confusion_matrix(y_val, predictions)\n",
    "print(\"\\nrecall: \", round(recall_score(cm), 4))\n",
    "print(\"precision: \", round(precision_score(cm), 4))\n",
    "print(\"f1_scores: \", round(f1_score(cm), 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1tb_XKwsnLYq"
   },
   "source": [
    "Remove oldpeak and test model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "id": "9IdEjL1GmDA3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  86.829 \n",
      "\n",
      "TP:  85\n",
      "FN:  14\n",
      "FP:  13\n",
      "TN:  93\n",
      "\n",
      "recall:  0.8586\n",
      "precision:  0.8673\n",
      "f1_scores:  0.8629\n"
     ]
    }
   ],
   "source": [
    "# print(X)\n",
    "idx = feature_list.index('oldpeak')\n",
    "new_X = np.delete(X, idx, 1)\n",
    "# print(new_X)\n",
    "train_count = int(new_X.shape[0] * 0.8)\n",
    "X_train, X_val, y_train, y_val = new_X[:train_count], new_X[train_count:], y[:train_count], y[train_count:]\n",
    "\n",
    "model2 = MixedDataNB()\n",
    "model2.fit(X_train, y_train)\n",
    "predictions = model2.predict(X_val)\n",
    "\n",
    "print(\"Accuracy: \", round(accuracy_score(y_val, predictions) * 100, 3),\"\\n\")\n",
    "cm = confusion_matrix(y_val, predictions)\n",
    "print(\"\\nrecall: \", round(recall_score(cm), 4))\n",
    "print(\"precision: \", round(precision_score(cm), 4))\n",
    "print(\"f1_scores: \", round(f1_score(cm), 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "c9kuak-Wchp8"
   },
   "source": [
    "Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tuVe6m8zcjb5"
   },
   "source": [
    "**Answer:**\n",
    "<font dir=\"rtl\" color=\"Salmon\">\n",
    "    <br>\n",
    "     همانطور که مشخص است از لحاظ دقت، با حذف هردو مقداری کم شده است اما این مقدار تغییر برای oldpeak\n",
    "    بیشتر بوده است.\n",
    "    <br>\n",
    "    اما به طور کلی هر دو تاثیر کمی داشته اند.\n",
    "</font>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RYDwVLDqkoXc"
   },
   "source": [
    "# Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PFGNM7y2krbl"
   },
   "source": [
    "Which model did better: the perceptron or the Naive Bayes model? What factors do you think influenced the difference in their performance?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Gof1rS14kuOG"
   },
   "source": [
    "### Answer:\n",
    "<font dir=\"rtl\" color=\"Salmon\">\n",
    "    <br>\n",
    "    دقت برای perceptron حدود 74 درصد و برای Naive Bayes حدود 88 درصد بوده است. بنابراین Naive Bayes عملکرد بهتری داشته است.<br>\n",
    "    لازم بود که فضای فرضیه پیچیده تری داشته باشیم و برای همین با پیچده تر کردن فضای فرضیه و استفاده از مدل\n",
    "    Naive Bayes\n",
    "   دیدیم ک دقت زیاد شد.\n",
    "    <br>\n",
    "    همچنین به کمک کوررلیشن دیدیم ک داده ها تقریبا و relativly مستقل هستند که این با منطق و عملکرد Naive Bayes بیشتر همخوانی دارد ولی بایاس بیشتری برای Perceptron ایجاد میکند\n",
    "    که باعث شد دقت کمتری داشته باشد.\n",
    "    <br>\n",
    "    دلیل دیگر این بود که 2 دسته داده داشتیم و طبیعتا Naive Bayes بهتر عمل می‌نماید.\n",
    "</font>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "q0Ogh-b03oUx"
   },
   "source": [
    "# Imbalanced Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mAgBI2ug4pYo"
   },
   "source": [
    "By using the model you designed and the functions you implemented, train your model on the three imbalanced data sets provided to you and calculate accuracy, recall, precision, and F1 score metrics for it each time. The confusion matrix should also be plotted each time. Finally, analyze based on the obtained results which accuracy and F1 score metrics are most appropriate for evaluating the model in imbalanced datasets. **Note:** In order to prevent code duplication, you are free to define functions and only use them with different datasets."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Wp4n9e5HaOVx"
   },
   "source": [
    "## First dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "J6PrUxRcS-Gx"
   },
   "source": [
    "### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "id": "Eiupha6eTf8c"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"Imbalanced_Datasets/1.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "uXddb_7dUG5u"
   },
   "source": [
    "### Find the distribution\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oFfEYF3LUm7e"
   },
   "source": [
    "Find the number of samples from each target label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "id": "UWKeH1VmUpB4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    499\n",
       "1    226\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['target'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>52</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>125</td>\n",
       "      <td>212</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>168</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>53</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>140</td>\n",
       "      <td>203</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>155</td>\n",
       "      <td>1</td>\n",
       "      <td>3.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>70</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>145</td>\n",
       "      <td>174</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>125</td>\n",
       "      <td>1</td>\n",
       "      <td>2.6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>61</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>148</td>\n",
       "      <td>203</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>161</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>62</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>138</td>\n",
       "      <td>294</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>106</td>\n",
       "      <td>0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  slope  \\\n",
       "0   52    1   0       125   212    0        1      168      0      1.0      2   \n",
       "1   53    1   0       140   203    1        0      155      1      3.1      0   \n",
       "2   70    1   0       145   174    0        1      125      1      2.6      0   \n",
       "3   61    1   0       148   203    0        1      161      0      0.0      2   \n",
       "4   62    0   0       138   294    1        1      106      0      1.9      1   \n",
       "\n",
       "   ca  thal  target  \n",
       "0   2     3       0  \n",
       "1   0     3       0  \n",
       "2   0     3       0  \n",
       "3   1     3       0  \n",
       "4   3     2       0  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mcHFkGt6UCzM"
   },
   "source": [
    "### Split data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "id": "DRodEXYEUE5K"
   },
   "outputs": [],
   "source": [
    "X = df[['sex', 'cp', 'fbs', 'restecg', 'exang', 'slope', 'ca', 'thal', 'age', 'trestbps', 'chol', 'thalach', 'oldpeak']].to_numpy()\n",
    "y = df.loc[:, \"target\"].to_numpy()\n",
    "y = np.array([1 if i == 1 else -1 for i in y])\n",
    "train_count = int(X.shape[0] * 0.8)\n",
    "X_train, X_val, y_train, y_val = X[:train_count], X[train_count:], y[:train_count], y[train_count:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OVPEziogTMqP"
   },
   "source": [
    "### Train model and make predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "id": "UGX1KttKTgPz"
   },
   "outputs": [],
   "source": [
    "model.fit(X_train, y_train)\n",
    "predictions = model.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5QS80Z9fTSJH"
   },
   "source": [
    "### Plot confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "id": "72m9_9HvTgn7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TP:  37\n",
      "FN:  6\n",
      "FP:  5\n",
      "TN:  97\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_val, predictions)\n",
    "plot_confusion_matrix(cm, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MBQxSo1oTZkB"
   },
   "source": [
    "### Calculate metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "id": "U6vQI8fItr3y"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  92.414 \n",
      "\n",
      "recall:  0.8605\n",
      "precision:  0.881\n",
      "f1_scores:  0.8706\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy: \", round(accuracy_score(y_val, predictions) * 100, 3),\"\\n\")\n",
    "print(\"recall: \", round(recall_score(cm), 4))\n",
    "print(\"precision: \", round(precision_score(cm), 4))\n",
    "print(\"f1_scores: \", round(f1_score(cm), 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CM968e-qah0g"
   },
   "source": [
    "## Second dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "i5EY_UPRah0n"
   },
   "source": [
    "### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "WCJt4rb8ah0n"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"Imbalanced_Datasets/2.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WT-pidZNah0n"
   },
   "source": [
    "### Find the distribution\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TvMMX9Paah0n"
   },
   "source": [
    "Find the number of samples from each target label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "id": "B_zQTT3nah0n"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    499\n",
       "1    126\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['target'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UtAGBCz5ah0n"
   },
   "source": [
    "### Split data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "id": "oWSWhuqkah0n"
   },
   "outputs": [],
   "source": [
    "X = df[['sex', 'cp', 'fbs', 'restecg', 'exang', 'slope', 'ca', 'thal', 'age', 'trestbps', 'chol', 'thalach']].to_numpy()\n",
    "y = df.loc[:, \"target\"].to_numpy()\n",
    "y = np.array([1 if i == 1 else -1 for i in y])\n",
    "train_count = int(X.shape[0] * 0.8)\n",
    "X_train, X_val, y_train, y_val = X[:train_count], X[train_count:], y[:train_count], y[train_count:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UgU9C0lcah0n"
   },
   "source": [
    "### Train model and make predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "id": "R_0U_L9Rah0n"
   },
   "outputs": [],
   "source": [
    "model.fit(X_train, y_train)\n",
    "predictions = model.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_L6TQSzHah0n"
   },
   "source": [
    "### Plot confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "id": "MYBVkHo4ah0n"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TP:  20\n",
      "FN:  8\n",
      "FP:  4\n",
      "TN:  93\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_val, predictions)\n",
    "plot_confusion_matrix(cm, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "c0XAiLBUah0o"
   },
   "source": [
    "### Calculate metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "id": "riRcbrlqah0o"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  90.4 \n",
      "\n",
      "recall:  0.7143\n",
      "precision:  0.8333\n",
      "f1_scores:  0.7692\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy: \", round(accuracy_score(y_val, predictions) * 100, 3),\"\\n\")\n",
    "print(\"recall: \", round(recall_score(cm), 4))\n",
    "print(\"precision: \", round(precision_score(cm), 4))\n",
    "print(\"f1_scores: \", round(f1_score(cm), 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZNu2VoZsak76"
   },
   "source": [
    "## Third dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nXzFkXaLak8B"
   },
   "source": [
    "### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "id": "HfD4l1G6ak8B"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"Imbalanced_Datasets/3.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tRg7HDJ-ak8B"
   },
   "source": [
    "### Find the distribution\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Lji-2CRSak8B"
   },
   "source": [
    "Find the number of samples from each target label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "id": "impG1HQFak8B"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    499\n",
       "1     26\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['target'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3pe45UaBak8C"
   },
   "source": [
    "### Split data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "id": "ze3YZgfmak8C"
   },
   "outputs": [],
   "source": [
    "X = df[['sex', 'cp', 'fbs', 'restecg', 'exang', 'slope', 'ca', 'thal', 'age', 'trestbps', 'chol', 'thalach']].to_numpy()\n",
    "y = df.loc[:, \"target\"].to_numpy()\n",
    "y = np.array([1 if i == 1 else -1 for i in y])\n",
    "train_count = int(X.shape[0] * 0.8)\n",
    "X_train, X_val, y_train, y_val = X[:train_count], X[train_count:], y[:train_count], y[train_count:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "N64rNbxmak8C"
   },
   "source": [
    "### Train model and make predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "id": "gvA0wDxYak8C"
   },
   "outputs": [],
   "source": [
    "model.fit(X_train, y_train)\n",
    "predictions = model.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "r4kXMUclak8C"
   },
   "source": [
    "### Plot confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "id": "ZEaCqPgrak8C"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TP:  0\n",
      "FN:  3\n",
      "FP:  0\n",
      "TN:  102\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_val, predictions)\n",
    "plot_confusion_matrix(cm, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "EFjScq2yak8D"
   },
   "source": [
    "### Calculate metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "id": "x8GSV1_5ak8D"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  97.143 \n",
      "\n",
      "recall:  0.0\n",
      "precision:  nan\n",
      "f1_scores:  nan\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\AMIRRE~1\\AppData\\Local\\Temp/ipykernel_4272/1178440399.py:2: RuntimeWarning: invalid value encountered in long_scalars\n",
      "  return cm[1][1] / (cm[1][1] + cm[0][1])\n",
      "C:\\Users\\AMIRRE~1\\AppData\\Local\\Temp/ipykernel_4272/1232598744.py:3: RuntimeWarning: invalid value encountered in long_scalars\n",
      "  precision = cm[1][1] / (cm[1][1] + cm[0][1])\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy: \", round(accuracy_score(y_val, predictions) * 100, 3),\"\\n\")\n",
    "print(\"recall: \", round(recall_score(cm), 4))\n",
    "print(\"precision: \", round(precision_score(cm), 4))\n",
    "print(\"f1_scores: \", round(f1_score(cm), 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5NDXj0qEZ3u_"
   },
   "source": [
    "## Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HvHYNnEEZ5to"
   },
   "source": [
    "Based on the obtained results which accuracy and F1 score metrics are most appropriate for evaluating the model in imbalanced datasets?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LqGtCJLtaKhN"
   },
   "source": [
    "**Answer:**\n",
    "<font dir=\"rtl\" color=\"Salmon\">\n",
    "    <br>\n",
    "     به دلیل اینکه منظور سوال را به طور درست متوجه نشده‌ام، یکبار جواب‌های 3تا دیتاست را مقایسه می‌کنم و بار دیگر تفاوت میان f1_scores و accuracy\n",
    "</font>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font dir=\"rtl\" color=\"SeaGreen\">\n",
    "    <br>\n",
    "    در دیتاست آخر، به دلیل کم بودن اطلاعات و مقدار زیاد imbalanced بودن آن، طبیعتا به نتیجه قابل توجه و درستی نرسیده‌‍ایم\n",
    "    .<br>\n",
    "    دیتاست اول و دوم بهتر عمل کرده‌اند اما با توجه به مقادیر f1_scores می‌توان گفت دیتاست اول به دلیل داشتن سمپل بیشتر، عملکرد بهتری داشته است.\n",
    "    <br>\n",
    "    \n",
    "</font>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font dir=\"rtl\" color=\"LightBlue\">\n",
    "    <br>\n",
    "     در اینجا به تفاوت میان f1_scores و accuracy می‌پردازیم.\n",
    "    :<br>\n",
    "    به طور کلی، مقدار f1_scores مخصوصا برای توزیع کلاس نامتعادل مفیدتر و قابل اعتناتر می‌باشد.<br>\n",
    "    accuracy \n",
    " ممکن است مناسب ترین متریک برای ارزیابی عملکرد مدل نباشد. زیرا که یک طبقه بندی کننده می تواند با پیش بینی کلاس اکثریت برای همه نمونه ها، به دقت بالایی برسد ، که لزوما نشان دهنده توانایی آن در طبقه بندی صحیح نمونه های کلاس اقلیت نیست.\n",
    "    به مثال زیر توجه فرمایید:\n",
    "    <br>\n",
    "</font>\n",
    "<font dir=\"ltr\" color=\"LightBlue\">\n",
    "    Imagine we have a classification model for detecting a specific disease, where only 10% of the population actually has the disease and the rest are healthy.\n",
    "\n",
    "If we have a very simple model that classifies everyone as healthy, our accuracy will be very high (around 90%) because most of the data is healthy. However, this model is not useful at all since it fails to detect any cases of the disease, so its F1 score will be very low.\n",
    "\n",
    "On the other hand, if we have a more sophisticated model that is powerful enough to detect many cases of the disease, but may misclassify some healthy individuals as diseased, it might have slightly lower accuracy (say 75%), but its F1 score will be much higher because it detects more cases of the disease and also correctly identifies a large number of healthy individuals.\n",
    "\n",
    "Therefore, the F1 score helps us evaluate a model's performance better in situations where classes are imbalanced and helps us consider how well precision and recall are balanced. In cases where precision and recall have equal importance, accuracy as a general metric may be sufficient.\n",
    "</font>  "
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [
    "utNwgYmlKvfm",
    "bjntS4nxH4Hs",
    "TNo6F7XEH7p3",
    "phw5_4uEIgyY",
    "NOo1N_ZaKHPg",
    "js3fx5A2lnR6",
    "GoPThTyLnd0N",
    "54V2QMyQoAQr",
    "0u9hzezSo48b",
    "RYDwVLDqkoXc",
    "q0Ogh-b03oUx"
   ],
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}