code/model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import warnings\n",
    "from tqdm.notebook import trange as notebook_trange\n",
    "from tqdm import tqdm, trange\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import precision_recall_curve\n",
    "from sklearn.metrics import roc_curve, roc_auc_score\n",
    "\n",
    "from pycaret.classification import*\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size of Dataset: (70000, 13)\n"
     ]
    },
    {
     "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>id</th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>ap_hi</th>\n",
       "      <th>ap_lo</th>\n",
       "      <th>cholesterol</th>\n",
       "      <th>gluc</th>\n",
       "      <th>smoke</th>\n",
       "      <th>alco</th>\n",
       "      <th>active</th>\n",
       "      <th>cardio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>18393</td>\n",
       "      <td>2</td>\n",
       "      <td>168</td>\n",
       "      <td>62.0</td>\n",
       "      <td>110</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>20228</td>\n",
       "      <td>1</td>\n",
       "      <td>156</td>\n",
       "      <td>85.0</td>\n",
       "      <td>140</td>\n",
       "      <td>90</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>18857</td>\n",
       "      <td>1</td>\n",
       "      <td>165</td>\n",
       "      <td>64.0</td>\n",
       "      <td>130</td>\n",
       "      <td>70</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>17623</td>\n",
       "      <td>2</td>\n",
       "      <td>169</td>\n",
       "      <td>82.0</td>\n",
       "      <td>150</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>17474</td>\n",
       "      <td>1</td>\n",
       "      <td>156</td>\n",
       "      <td>56.0</td>\n",
       "      <td>100</td>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id    age  gender  height  weight  ap_hi  ap_lo  cholesterol  gluc  smoke  \\\n",
       "0   0  18393       2     168    62.0    110     80            1     1      0   \n",
       "1   1  20228       1     156    85.0    140     90            3     1      0   \n",
       "2   2  18857       1     165    64.0    130     70            3     1      0   \n",
       "3   3  17623       2     169    82.0    150    100            1     1      0   \n",
       "4   4  17474       1     156    56.0    100     60            1     1      0   \n",
       "\n",
       "   alco  active  cardio  \n",
       "0     0       1       0  \n",
       "1     0       1       1  \n",
       "2     0       0       1  \n",
       "3     0       1       1  \n",
       "4     0       0       0  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"cardio_train.csv\",sep=\";\")\n",
    "print(f\"Size of Dataset: {df.shape}\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, this Dataset contains the following Features :\n",
    "\n",
    "    id => It's just the Id no of the row. Not revelant\n",
    "    age => It's the age of a person in Days\n",
    "    gender => It's the gender of the person\n",
    "    height => It's the height of the person in cm\n",
    "    weight => It's the weight of the person in kg\n",
    "    ap_hi => It's the Systolic blood pressure i.e. Pressure exerted when Blood is ejected in arteries. Normal value : 120mmhg or Below\n",
    "    ap_low => It's the Diastolic blood pressure i.e. Pressure exerted when Blood exerts between arteries and heartbeats. Normal Value : 80mmhg or Below\n",
    "    cholesterol :> It's the Cholestreol value (Cholesterol is a type of fat found in your blood) of your blood. In Adults, 200 mg/dL is desired with 200 and 239 mg/dL as Boderline High. In Children, 170 mg/dL is desired with 170 and 199 mg/dL as Boderline High\n",
    "    gluc => It's the Glucose Level. They're less than 100 mg/dL after not eating (fasting) for at least 8 hours. And they're less than 140 mg/dL 2 hours after eating. For most people without diabetes, blood sugar levels before meals hover around 70 to 80 mg/dL\n",
    "    smoke => It contain Binary Values stating whether Person is a Smoker or not i.e. {0 : 'Not a Smoker', 1 : 'Smoker'}\n",
    "    alco => It contain Binary Values stating whether Person is an alchoalic or not i.e. {0 : 'Not a Alcholic', 1 : 'Alcholic'}\n",
    "    active => It contain Binary Values stating whether Person is involved in physical activites or not i.e. {0 : 'Not involved in Physical Activites', 1 : 'involved in physical activites'}\n",
    "    cardio => It's our Target Value Binary Values stating whether Person has Cardiovascular diseases (CVDs) or Not i.e. {0 : 'Not Have CVD', 1 : 'Have CVD'}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset contains no Duplicate Values\n"
     ]
    }
   ],
   "source": [
    "duplicate_sum = df.duplicated().sum()\n",
    "if duplicate_sum:\n",
    "    print('Duplicates Rows in Dataset are : {}'.format(duplicate_sum))\n",
    "else:\n",
    "    print('Dataset contains no Duplicate Values')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For the feature height, No of Outliers is 519\n",
      "For the feature weight, No of Outliers is 1819\n"
     ]
    }
   ],
   "source": [
    "# Seeking to outlier then destroy it!!!\n",
    "def outliers(df_out, drop = False):\n",
    "    for each_feature in df_out.columns:\n",
    "        feature_data = df_out[each_feature]\n",
    "        Q1 = np.percentile(feature_data, 25.) # 25th percentile of the data of the given feature\n",
    "        Q3 = np.percentile(feature_data, 75.) # 75th percentile of the data of the given feature\n",
    "        IQR = Q3-Q1 #Interquartile Range\n",
    "        outlier_step = IQR * 1.5 #That's we were talking about above\n",
    "        outliers = feature_data[~((feature_data >= Q1 - outlier_step) & (feature_data <= Q3 + outlier_step))].index.tolist()  \n",
    "        print('For the feature {}, No of Outliers is {}'.format(each_feature, len(outliers)))\n",
    "\n",
    "outliers(df[['height', 'weight']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For the feature height, No of Outliers is 484\n",
      "For the feature weight, No of Outliers is 1174\n",
      "For the feature height, No of Outliers is 0\n",
      "For the feature weight, No of Outliers is 349\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(68370, 13)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['height', 'weight']] = np.log(df[['height', 'weight']])\n",
    "outliers(df[['height', 'weight']])\n",
    "df = df[(df['weight'] > df['weight'].quantile(0.005)) & (df['weight'] < df['weight'].quantile(0.995))]\n",
    "df = df[(df['height'] > df['height'].quantile(0.005)) & (df['height'] < df['height'].quantile(0.995))]\n",
    "outliers(df[['height', 'weight']])\n",
    "df[\"height\"] = np.expm1(df[\"height\"])\n",
    "df[\"weight\"] = np.expm1(df[\"weight\"])\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(68370, 14)\n"
     ]
    },
    {
     "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>id</th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>ap_hi</th>\n",
       "      <th>ap_lo</th>\n",
       "      <th>cholesterol</th>\n",
       "      <th>gluc</th>\n",
       "      <th>smoke</th>\n",
       "      <th>alco</th>\n",
       "      <th>active</th>\n",
       "      <th>cardio</th>\n",
       "      <th>bmi</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>18393</td>\n",
       "      <td>2</td>\n",
       "      <td>167.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>110</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>21.87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>20228</td>\n",
       "      <td>1</td>\n",
       "      <td>155.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>140</td>\n",
       "      <td>90</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>18857</td>\n",
       "      <td>1</td>\n",
       "      <td>164.0</td>\n",
       "      <td>63.0</td>\n",
       "      <td>130</td>\n",
       "      <td>70</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>23.42</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id    age  gender  height  weight  ap_hi  ap_lo  cholesterol  gluc  smoke  \\\n",
       "0   0  18393       2   167.0    61.0    110     80            1     1      0   \n",
       "1   1  20228       1   155.0    84.0    140     90            3     1      0   \n",
       "2   2  18857       1   164.0    63.0    130     70            3     1      0   \n",
       "\n",
       "   alco  active  cardio    bmi  \n",
       "0     0       1       0  21.87  \n",
       "1     0       1       1  34.96  \n",
       "2     0       0       1  23.42  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# BMI formulation\n",
    "df[\"bmi\"] = np.round(df.weight/(df.height/100)**2,2)\n",
    "print(df.shape)\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If your BMI is:\n",
    "\n",
    "    below 18.5 – you're in the underweight range\n",
    "    between 18.5 and 24.9 – you're in the healthy weight range\n",
    "    between 25 and 29.9 – you're in the overweight range\n",
    "    between 30 and 39.9 – you're in the obese range\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[(df.bmi <= 18.4), 'bmicat'] = 1\n",
    "df.loc[(df.bmi > 18.4) & (df.bmi < 25), 'bmicat'] = 2\n",
    "df.loc[(df.bmi >= 25) & (df.bmi < 30), 'bmicat'] = 3\n",
    "df.loc[(df.bmi >= 30) & (df.bmi < 40), 'bmicat'] = 4\n",
    "df.loc[(df.bmi >= 40), 'bmicat'] = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(data=df,x=\"bmicat\",hue=\"cardio\",kde=True,bins=10)\n",
    "plt.show()\n",
    "\n",
    "# When the BMI category increases, you can see from the histplot that the risk of cardiovascular disease also increases within the category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfsAAAFYCAYAAABUA1WSAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABzT0lEQVR4nO3deXhb5Zn4/e/RLlvyvsVL9h0ngSSlIRBSCAWSdlp2aAltp8vLtLRhpp0ybYe2MLQdwlC68CtdSMt0gEIJe1soNYFAErIQZyHO5iWO90WSLVvWvpz3DyUmdhxHTmxLlu7PdXERHR09534kWfc559kUVVVVhBBCCJG0NPEOQAghhBBjS5K9EEIIkeQk2QshhBBJTpK9EEIIkeQk2QshhBBJThfvAEZLJBLB7Xaj1+tRFCXe4QghhBBjSlVVgsEg6enpaDTDX7snTbJ3u91UV1fHOwwhhBBiXM2ePRur1TrsPkmT7PV6PRCttMFgiHM0Y6uqqory8vJ4hxEXqVr3VK03SN2l7qkn1roHAgGqq6v7899wkibZn7x1bzAYMBqNcY5m7KVCHc8kVeueqvUGqXuqkrrHJpam6zHtoFddXc1VV13FU089NWD7li1bmDNnTv/jiooKbr31Vq677jqef/55AMLhMD/4wQ+47bbbuO2222hqahrLUIUQQoikNWbJ3uPx8MADD3DJJZcM2O73+/nd735Hfn4+AH19faxfv54NGzbwzDPPsGHDBtxuNy+//DKKovDss89y55138uijj45VqEIIIURSG7NkbzAYePzxxykoKBiw/Te/+Q233357f7v6gQMHWLBgAVarFbPZzOLFi9m9ezc7d+5k1apVAKxYsYJdu3aNVahCCCFEUhuzZK/T6TCZTAO21dfXU1NTw7XXXtu/zWazkZOT0/84NzcXu90+YLtOpyMcDhMOh8cqXCGEECJpjWsHvfXr13PvvfcO2Da4F6GqqiiKMmTvwlg6IVRVVZ1fkBNEZWVlvEOIm1Ste6rWG6TuqUrqPnrGLdl3dHRQV1fHN7/5TQA6OztZu3Yt//qv/4rD4ejfz263s2zZMvLz8/u3BwIB9Hr9WScNACgvL0/6HpyVlZUsWbIk3mHERarWPVXrDVJ3qXvqibXufr8/5gvccZsut7CwkIqKCp577jmee+45CgoKeOqpp1i4cCFHjx7F5XLhdrvZv38/S5cu5fLLL2fTpk0AbN68meXLl49XqCP2yiuvcMcddwBwzz338NZbb8U5IiGEEOJDY3ZlX1VVxfr162lpaUGn0/HGG2/w6KOPkpWVNWA/g8HAunXruP3229FoNNx1112YTCauuuoq3nrrLW644QbMZjM//elPxyrUEQmHw2i12jM+/9BDD41jNEIIIcTZjVmyLy8v58knnzzj86de/a5evZrVq1cPeF6r1bJ+/fqxCo/XXnuNxx57jEAgwIIFC/jJT37C3/72Nx5//HHC4TD5+fk89NBDlJSU8OKLL7Jp0yZ6e3uZP38+//Ef/8GPfvQj3nrrLfLy8vjIRz7SX+4dd9zBTTfdxKc//Wl27tzJf//3f+P1esnIyOCHP/xhys4IJYQQIn6SZga9kWhvb+eBBx7g5ZdfpqCggG984xv88Y9/5Je//CX/+Mc/KC4u5rvf/S6PPfYYP/7xjwF47733ePHFF5k2bRrvvPMO27Zt47XXXkOj0XDHHXecNvLA4/Fw99138+tf/5qLLrqIiooKvvWtb/H666/H1PdACCGEGC0pmXW2bNnChRdeSGFhIYqi8NOf/pQvfvGLVFZWUlxcDMDSpUsHzNo3depUpk2bBsD777/PihUrSEtLw2QynXZXAmD//v3k5eVx0UUXAXDVVVdhs9lobm4ehxoKIVJJOBymurp6wH8yVFmcKiWv7Lu6usjMzOx/bDQaCYVC/OxnP2Pr1q2EQiH6+vr6kzswYP+enp4BkwWd+txJDodjQP8ERVHIzMykq6uLyZMnj3KNhBCprK6ujoodL1BSVgRAS1M7cCOzZ8+Ob2AiYaRkss/JyWHPnj39j/v6+qioqOCdd97h6aefJicnh40bN/Lqq68O+fqMjAxcLlf/41OHDp6Um5tLd3d3/+NIJILT6SQ3N3cUayKEEFElZUVMmV4a7zBEgkrJ2/grVqxg7969NDc3o6oqP/zhD2lpaaG0tJTs7GwcDgevvvoqbrd7yNdfdNFFbNmyBZ/Ph9fr5e9///tp+1x44YV0dXWxf/9+AF5//XVKSkooLZU/RiGEEOMrJa/si4qKuO+++7jzzjsJBoMsWLCAG264gXfeeYeVK1cyY8YMvv3tb/PVr36V9evXM2vWrAGvv+KKK9i8eTPXXnstOTk5fOxjH2PHjh0D9jGbzfziF7/ghz/8IV6vl5ycHB555JGYZgEUQgghRlNKJnuANWvWsGbNmgHbNm7cOODxtm3b+v99ww039P9bq9Xyox/9aMC+69atAxgw3PDiiy/m5ZdfHq2QhRAiaYTDYerq6gZsmzFjxrDzmIhzl7LJXggxMUmSSA7SqXB8SbIXQkwodXV1PLBxE1lFJQA421v4/s1IkpiApFPh+JFkL4SYcLKKSsgrnRLvMFKK3FGZ2CTZCyGEOCu57T6xSbIXQggRE7ntPnGl5Dh7IYQQIpVIshdCCCGSXErdxh+qg8n5iqWDyi9+8Qu2b99OIBDg/vvvZ8GCBaMagxBCCDGclEr2g4fsnK9Yhvzs2LGDAwcO8Oyzz1JdXc3999/P008/PSrHF0IIIWKRUskexn/Izs6dO1m1ahUQPSno7OzE6/ViNpvHLQYhhBCpTdrsx5jNZiMnJ6f/cU5ODna7PY4RCSGESDWS7MeYXq8f8FhVVVkMRwghxLiSZD/G8vPzB6x339XVRV5eXhwjEkIIkWok2Y+xyy+/nE2bNgFw8OBBysrKMJlMcY5KCCFEKkm5DnrO9pZRLmvusPuUl5czd+5crr/+erRaLT/+8Y9H7fhCCCFELFIq2c+YMYPv3zyaJc5lxowZZ93r29/+9mgeVAghhBiRlEr2Wq1WFm0QQgiRcqTNXgghhEhykuyFEEKIJCfJXgghhEhykuyFEEKIJCfJXgghhEhyKdUbP15L3FZXV/O1r32NL3zhC6xdu3ZUjy+EEEKcTUol+7q6Oip2vEBJWdGolNfS1A7cOOxwPo/HwwMPPMAll1wyKscUQgghRiqlkj1ASVkRU6aXjtvxDAYDjz/+OI8//vi4HVMIIYQ4Vcol+/Gm0+nQ6eRtFkIIET9j2kGvurqaq666iqeeegqAjo4OvvjFL7J27Vo+97nP0dHRAUBFRQW33nor1113Hc8//zwQbV//wQ9+wG233cZtt91GU1PTWIYqhBBCJK0xS/ZDtVX//Oc/56abbuKpp57immuu4X//93/p6+tj/fr1bNiwgWeeeYYNGzbgdrt5+eWXURSFZ599ljvvvJNHH310rEIVQgghktqYJfuTbdUFBQX92+69916uvvpqALKzs+nr6+PAgQMsWLAAq9WK2Wxm8eLF7N69m507d7Jq1SoAVqxYwa5du8YqVCGEECKpjVlj8lBt1enp6UD0Fv2f/vQnvv71r9PZ2UlOTk7/Prm5udjtdmw2W/92nU5HOBwmHA6fdZhbVVXVGZ9raGjgYOMhbDbbuVZrgPZWG5HuTFwu1xn3OXbsGE8//TQ2mw2tVsvzzz/Pv/3bv2GxWM7r2JWVlef1+oksVeueqvWGgXVvaGigs9NBQKMHwNnZQVVVaNi/w4ksls+9oaEBR8iGwRz9fbTZbFR1V43qezLax4ilPPnOj55x7zkWDoe55557uPjii1m2bBmvv/76gOdVVUVRFPR6/WmvVRTlrOWXl5djNBqHfO7CCy+kvK783AIfssCzj7NfsmQJN988quvqUllZyZIlS0a1zIkiVeueqvWG0+tutVp5u/sIecXFABgiQcrL5yblipaxfu5Wq5VDLT1MmjQJgIA3zPyS8lF9T0b7GGcrT77zZ6+73+8f9gL3VOOe7L/73e9SUlLCunXrAMjPz8fhcPQ/b7fbWbZs2YDtgUAAvV6PRnN+rQ6yxK0QIh4GT+gVDocBBlwoxDJBVzILhyPU19f3P25oaODCCy9M6fdkNI1rsn/11VfRaDR885vf7N+2cOFCjh49isvlQqPRsH//fu677z7cbjebNm1i5cqVbN68meXLl49nqEIIMWrq6up4YOMmsopKAGis2gNGM5NnzQPA2d7C928mpS9G2ls76e2txWtoBeBg4yHK60b37kQqG7NkX1VVxfr162lpaUGn0/HGG2/gcDgwGo3ccccdQPRM9r777mPdunXcfvvtaDQa7rrrLkwmE1dddRVvvfUWN9xwA2azmZ/+9KdjFaoQQoy5rKIS8kqnANDd3oJiSut/LKImlRT0T3o2Wn2rxtJQU7An6h2aMUv25eXlPPnkkzHtu3r1alavXj1gm1arZf369WMRmhBCJJVwOHxi+u6olqZ25hSF4xhRahg8BXssU6jHi0ztJoQQE1xjYyMbdzsoclgBaK93MLOwkXnz5sU5suQ33lOwnytJ9kIIkQTSc/PJOjFCwdXrjnM0ItHIevZCCCFEkpMreyGEGGQidbwSIhaS7IUQYpDBQ+VkaJyY6CTZCyHEEE4dKifERCdt9kIIIUSSkyt7IYQYY+FwGKezG8UcXQDL5erFajDHOSqRSiTZCyHEGGtsbKSxvY5QdOFPWm2NFBvS4huUSCmS7IUQYhxYLOlkZkcnvUlLk6t6Mb6kzV4IIYRIcpLshRBCiCQnyV4IIYRIctJmL4QQ406lz+3G4XAA4HR2Ew7LKnVi7MiVvRBCjDOPx0eXqwNbbxO23iYa2+tobGyMd1giicmVvRBCxIHZbOrvnd9rSY9zNKcbvD5AfX09ql6NY0TifEiyF0IIcZq6ujoqdrxASVkRAHv2VzF9VilQFt/AxDmRZC+EEGJIJWVFTJleCkBLU3ucoxHnQ5K9EEKIpCPLFA8kyV4IIUTSGdwMEb0zcWPKLlMsyV4IIURSOrUZItXJ0DshhBAiyUmyF0IIIZKcJHshhBAiyUmyF0IIIZKcJHshhBAiyUmyF0IIIZKcJHshhBAiyck4eyGEGCQcDuN0dqOYLYAsQQsQiURob+0kK6sZiE5SM6cotd+TiUSSvRBCDNLY2Ehjex2hE4vRtbbX0dhYxLx58+IbWBzZOrt481iQfQE3AO31DmYWNqb0ezKRSLIXQoghWCzpCb0EbTyk5+SRVVwMgKvXHedoxEhIshdCTGiRSJj6+voB21J5wRMxdgYvrlNfX4+qV+MYUezGNNlXV1fzta99jS984QusXbsWh8PBPffcg8vloqioiIcffhiDwUBFRQUbNmzA7/ezdu1abrrpJsLhMPfffz/V1dUA/M///A9lZbKOshBioJ7Odh6r62Vysx8AZ3sL37+ZlF3wRIydwYvr7NlfxfRZpUDi56Yx643v8Xh44IEHuOSSS/q3PfTQQ9x4440899xzlJSU8Oqrr9LX18f69evZsGEDzzzzDBs2bMDtdvPyyy+jKArPPvssd955J48++uhYhSqEmOAy8ovIK51CXukUsopK4h1O3KmqSnNzM9XV1f3/pXoHw9FycnGdKdNLKSzKi3c4MRuzZG8wGHj88ccpKCjo37Zr1y6uvPJKAFatWsXWrVs5cOAACxYswGq1YjabWbx4Mbt372bnzp2sWrUKgBUrVrBr166xClUIIZKK2+3hUMMuDrVs41DLNip2vHDa2u4itYzZbXydTodON7B4t9uNyWQCICcnB7vdjs1mIycnp3+f3Nzc07brdDrC4TDhcFja4YQQIgb5hTmyvKvoN64d9PR6ff+/VVVFUZQB24bbDqAoylmPUVVVdf6BTgCVlZXxDiFuUrXuqVpvGFj3hoYGOjsdBDTR34iuLgeK0UdraysAzs4OqqpCuFyucz5eTU0NXp+Hvr4+ALw+DzU1NeTlndtt28HlBf0BVL0/pvJj+dwHlx/w+XE6e2lrawPAZrNR1V01ovekoaEBR8iGwRy9wOp1OvH6DaP2ngwuv7urC1/Q2B8zRH/Pz/VzHFz+ubwHI415NI5x0mj/vY9rsk9PT8fr9WI2m7Hb7RQUFJCfn4/D4ejfx263s2zZsgHbA4EAer0ejebsrQ7l5eUYjcYxq0MiqKysZMmSJfEOIy5Ste6pWm84ve5Wq5W3u4+Qd2IImLv1OIopjeITjw2RIOXlc8+rg57dbsfscGOxRCfVMZvSmDVr1jl/BoPL0xsN6I3Gs5Yf6+c+uHyDyUhWVhaTJk0CIOANM7+kfETvidVq5VBLT38ZGVlZmJ3mUXtPBpefndNCutXU/7ilsZ3y8pHFPFz55/IejDTm0TgGxP65+/3+mC9wx3W63BUrVrBp0yYAKioqWLlyJQsXLuTo0aO4XC7cbjf79+9n6dKlXH755f37bt68meXLl49nqEIIIUTSGLMr+6qqKtavX09LSws6nY433niDhx9+mH//93/niSeeYNq0aaxZswadTse6deu4/fbb0Wg03HXXXZhMJq666ireeustbrjhBsxmMz/96U/HKlQhhBAiqY1Zsi8vL+fJJ588bftQ21avXs3q1asHbNNqtaxfv36swhNCCCFShqx6J4QQQiQ5SfZCCCFEkpNkL4QQQiQ5SfZCCCFEkpNV74QQYoIZvPpac3MzMDFWXxPxIcleCCEmmNNWXzuyjYB+anyDEglNkr0QQkxAJ1dfA8jNzwZnfOMRiU3a7IUQQogkJ8leCCGESHKS7IUQQogkJ8leCCGESHKS7IUQQogkJ8leCCGESHIy9E4IkfKGmqRGVWWSGpE8JNkLIVJeXV0dD2zcRFZRCQCHdh4mlJkW56iEGD2S7IUQAsgqKiGvdAoAlpp8nGF3nCMSYvRIm70QQgiR5CTZCyGEEElObuOLlDK4IxbAjBkz0Gq1cYpICCHGniR7kVIGd8Rytrfw/Zth9uzZcY5MCCHGjiR7kXJO7YglhBCpQNrshRBCiCQnV/ZCCCFOEw6HaWlq73/ssHWj6jLiGJE4H5LshRBCnKaxsZGNux0UOawAHKr3YynxxTkqca4k2QshRJJRIxFsHV00HGsGoKWpnTlF4RGXk56bT1ZxMQDGzKzRDFGMM0n2QogJJRwO43R2o5gtALhcvVgN5jhHlVi8zm7etqm07IvOAthe72BmYSPz5s2Lc2QiXiTZCyEmlMbGRhrb6wilRx+32hopNsg89oOZsnP7r8pdvTL1b6qTZC+EmHAslnQys6NtyWlpclUvxNlIshdCCBF3kUiE9tZOsrKi/Qw6Wm2EF4y8n4EYmiR7IYQQcWfr7OLNY0H2BaJNDscPe/nIvNj7GQyeCru+vh5Vr45JrBORJHshRMob3Omvz9UHZiXOUaWe9Jy8/n4GaZ22Eb22rq6Oih0vUFJWBMCe/VVMn1UKlI12mBOSJHshRMob3OnP5mzDosuPb1BJbiwm7SkpK2LK9FKAAWULSfZCiAQXDoeprq7uf9zc3Iyqjv7t2VM7/RlNxlEvXwwkk/aMr3FN9m63m3vuuYfe3l78fj933XUX5eXl3HPPPbhcLoqKinj44YcxGAxUVFSwYcMG/H4/a9eu5aabbhrPUIUQCaK5uZlHKyr7Vyo8tPMwoUwZapcMZNKe8TOuyf6ll15i2rRp/Pu//zvt7e18/vOf58ILL+TGG29kzZo1rF+/nldffZVrr72W9evX89JLL6HT6bj++utZvXo16enp4xmuECJBnLpSoaUmH2dYxo0LMRLjuupddnY2DocDgJ6eHnJycti1axdXXnklAKtWrWLr1q0cOHCABQsWYLVaMZvNLF68mN27d49nqEIIIUTSGNdkv2bNGtrb27nmmmv4/Oc/zz333IPb7cZkMgGQk5OD3W7HZrORk5PT/7rc3Fzsdvt4hiqEEEIkjXG9jf/KK69QXFzME088wZEjR/je976HXq/vf15VVRRFGbDt1O2xqKqqGtWYE1VlZWW8Q4ib86l7Q0MDnZ0OAprod8zZ2UFVVQiXyzVa4Y2ZVP7MOzs7PvzMup0EDBH6+voA8Ad8OHt6aG1tjT5/Dp9pTU0NXp+nv8ygP4Cq9/c/9vo81NTUkJeXd07xn0/5Q33uDQ0NOEI2DGYtAL1OJ16/YVTjj/d7cnKfWMsf/J50d3XhCxppa2sDwGazUdVddV5/6+NxjJNG++99XJP93r17ufzyywGYO3cunZ2dGI1GvF4vZrMZu91OQUEB+fn5/bf7Aex2O8uWLYvpGOXl5RiNyd2TtrKykiVLlsQ7jLg437pbrVbe7j5C3olOQYZIkPLyucyePXu0QhwTqfyZNzQ0UFBQ2P+ZtWdn4Qy7sViiY+KNBhNZmZkUn8dnarfbMTs+LFNvNKA3Gvsfm01pzJo165w/g3Mt/0yfu9Vq5VBLD5MmTQIgIysLs9M8qvHH+z0BRlT+4PckO6eFdKup/7G3L0h6OB2r1dr/mhkzZqDVamOO+WzHCHjDzC8pP+/fk1j/3v1+f8wXuON6G3/y5Mn9gXV0dJCens4VV1zBpk2bAKioqGDlypUsXLiQo0eP4nK5cLvd7N+/n6VLl45nqEIIIZJIe2sne2o2c6hlG4datlGx44UBM+4lu3G9sr/tttv4zne+w9q1awkGg9x3333MmDGDb33rWzzxxBNMmzaNNWvWoNPpWLduHbfffjsajYa77rqrv11fiFRycgrQhoaG/iuSkV6NCCGiJpUU9E+6k2piSvZ1dXXMmDFjwLZ9+/Zx4YUXjuhg6enpPProo6dtf/LJJ0/btnr1alavXj2i8oVINnV1dTywcRMBjY63u4/gbG/h+zeT8M0OQojEMmyy7+3txel08r3vfY+HH364f7vP5+O73/0ur7/++pgHKESqyyoqIaDR97dZC5EIVFWlubl5wOyGctcpcQ2b7Pfu3csf//hHDh8+zOc///n+7RqNhssuu2zMgxNCiESgRiIJldjGYl75kXK7PRxq2EXu9GjXr2g8N8pdpwQ1bLJfuXIlK1eu5JlnnuEzn/nMeMUkhBAJxd3l4E92D/uILo4T7+aURJlXPr8wJ2XbwCeamNrsr7zySv74xz/S09MzYAGKu+++e8wCE0KMj8HrgIPcjh2KJbegf8reRCDzyouRiCnZf/WrX2Xu3Ln941iFEMnjZCfAkwvNxPuqVSSnwSeVzc3NwOivXiiGFlOyN5lM/OQnPxnrWIQQcXLqQjNCjIW6ujoqdrxASVkRAHuObCOgnxrfoFJITJPqLFmyhNra2rGORQghRBIrKStiyvRSpkwvJTc/O97hpJSYruzfeecdfv/735OTk4NOp+ufq37z5s1jHJ4QiU3au0XiUMkxdWHU+Zk/W09rt9wiFx+KKdn/6le/QqMZ15l1hZgQpL1bJIqlF8Ds3Ogd2CmXmvB44WiPD384sWcf1WmCpOs9zJslJyhjKaZkv3PnztO2qarKzTffPOoBCTHRSHu3iLclCw0sng/ekJHGnjJczQdZusjIPP1RDtrmxTu8M5qUF2Zx0T40isq8y0z0eaCmN/FPUCaimJL9qUvthUIhDh06xOzZsyXZCyESgEqf292/UqbT2U04HI5zTOPHYnBRvsSIyw01rrkEwkYq9wRQdEaWXOBnVk4tx85zGZRsUxefutpMWrqK3nyIeufU845bb4bLyv2AQnNvMbbG41xUbmC+4QiH7HPPu3wxUEzfgP/+7/8e8DgcDvPjH/94TAISQoiR8Hh8hPwd2HqbAGhtr6OxsYh58xL3inb0qEzOaAbgrZ1QOOfD5b0rD8G06VnkmJ2UFqSf8xHy02xMz6pHUXQEgmDQ93FB/mHa85VzLjMcCVI0T4deDzVd03F4c6l8/yj+sIFliwJMz67nwDmXLoZyTg3xWq2W1tbW0Y5FCCHOidlsIjPbSma2FYvl3BPbRJNp7CHD6OJ4U4gOx+Dkq9DQM5mIqrBwpheVkd/tmFzkZ0Z2PSFVy59fdfO/LyvUdM1Ao0T4+LIQ+rRza2Pv6D2O3qRQVavD4c3t3/5BtYLTl0mm0UVhTuScyhZDi+nK/rOf/SyK8uEXqauri1mzZo1ZUEIIIc5GpezEVf2OSj+gP20Pf9hEe18hxdZ23IHmEZUeVnwsnuMhFNFy2DYPu2M7egs4vLkEIzrm5R4lswwiahiNEvvok0DIi93VRNCncqBWz7QFA59v7i0hy9TDwlmhEcUrhhdTsv/Xf/3X/n8rikJGRgZz5swZq5iEEEKcRVFuCIvBg92Tg6Pbhd4y9H4trmJyjR0o+kb8IQ9GXVpM5fcYatDroK57Mp7QwNf0+jM5XK9h/vQIHT31TMqaGXPcbT3HUFHpbgwTiZzeFNAXtOD0ZTIprwe/tzvmcsXwYrqNf/HFFwNQVVXFwYMH8Xg8YxqUEBNVJBKmvr6e6urq/v9SqbOYGD+zyqIL37T2TRp2v7Cq40iDCVUJUdX8bkxlNzgO4tfZsXXrsHnyhtyn8pCWkB86XQ14Aq6Yyg3hw+lpx6y30mc/cxNAc290KKtb3xhTueLsYkr2P/vZz/if//kf7HY7HR0dPPDAA/z2t78d69iEmHB6Ott57O19PLrlCI9uOcIDGzedNumOEOcrqLgpyg3R67fiCZ69j0JtsxFNxMDh1m1nTczBkJ+dda+CqlB5JA0YuiNeKKzQe6JloLX7KOpZ5rlXVRWf1gZw1jsBfUELtm4Fv7YLt79n2H1FbGIeZ//ss8/2zwoWDAZZu3Ytd95555gGJ8RElJFfJOPuU9xYL/ri0UezbFtfUUz7RyIKluBUejXVHGh6m4/O+NQZ993T8AaeQA+W4FRcnh6yhik30KeQac6jx2sjTRm+ecAZaCak8WA15WI15Zw15tpmLfnZIeo697Cw7Iqz7i+GF3Nv/FOn/9Tr9QM67AkhhPjQyUVfDrVs41DLtuiiL/7AqJTtC7rx6Nrp82ro9mXF/Lq00CSsplyOtO3A7hq6s57d1cThtu1kmPOwBCfHVO6krFkoaPBqbYQjwSH3iagRGvp2gwqTMmNr3z/eqgFVQ03H7gFLq4tzE1OyLy8v58477+TJJ5/kySef5F/+5V8oLy8f69iEEGLCGqtFXw63vgdKhJomI2e6xT4UBQ3LZ16PSoStNRsJRwb2dveHvGypfg5QuWTG9SjE1sPeqDNTkDEZVQlxrG/HkPscatmCJ9yNQc3EbDhDT8JBgiEFczgfl89BR299TK8RZ3bWZN/U1MT3vvc9PvWpT9HU1ER1dTVLly7l3nvvHY/4hBBiWCrgR09bb5hwJLmvAINhP0fatqOoOupbjGd/wSCTsmYyp+ijOD0d7Gl4o/+KORIJs/nw0/R4bVxQcjmTsmaMqNyCjGloI0ZsvhrqOvcOeK6zt4HK42+g15gxhYfu7Hcm5mC082FNx+4RvU6cbthkv337dj7zmc/gdrv5xCc+wfe+9z1uv/12/vSnP3HggMxvJISIr9beMAcyFrNPN4u/HvHx8kEvveHzmxp2PKSZFWaUqUyytGHQxH57v7p9F/6Qh/RgKeEhhq3FYunUNViM2Rxs2ULFwT9Q0/4+/zj4e9p6ainLmc+SqdeOuEyNoiEtXIxW0bO97iWO2w+gqipdfa1sPvInQGV2xsfQjHDaXkMkC6sphwb7AQIh34jjEh8aNtk/+uij/OEPf8BqtfZvmzt3Lr/97W/52c9+NubBCSHEmfT5I2yq9RFWdOREepiRo6XLq7Ldnc/W9sRMDH3+CAfaumgqXEizcRK2PicLC/dTPuPsM9yFIyEOtmxBp9GTHiw95xj0OiNrFn6Vkuw5tDpr2Fb7Au09xyjKnM7lc25Do5zbCqdaDMywXkY4EmLzkad5fvd6Xt33SzyBHi6acjWZhuGHCA5FQWFmwRJCkSDH7R+cU1wiatjTLI1GM+RSnbNmzSIYHLojhhBCnGpwz3SAGTNmDOj0O1IRVaGi1o8vBGXeBooNXhbOLGamM8Sb1V5+WtXL9c0OFpfmnr2wcdIUymDzBx4iqhEFlQanmXeP5/CRkl6+vLSJLvUDgqFV6HVD354/0LwZT6CX+cWX0e06fba8kUgzZnDV/C9w3P4Bbn8PpTlzyUorOK8yAfJM0yiftYTdx1+n0XGQSVkzmV98GaXZc6ipqTmnMmcWLmFv45vUdOxmdtHF5x1jqho22Q83eU5fX9+oByOEmPgGJ/f6+nr+7/1aciaVAeBsb+H7NzPkhUSsWsnE7o4wK0+HpbYDDBkATM7SscjczV5vLjc8sZmd/7qGQqv5/Co0Ct735FETziXLFOTT8+w07GnGFs7FljWT91sy6PFM4ZsrGnnj4Aaumv8FTPqBY+cdfa3sb3qLNEMGiyavYnNNbJPjDEdRFKblLzrvcgbLMOdx5bw7iKiRc75LcKp0YxbFWbNodVbj9HSOyklJKhr2k5g+fTrPPffcadt///vfM3euLEEohDhdXV0dD2zc1D+x0GN/347GbCGvdAp5pVPIKio5r/IjKrSpWeg0cMlkw2n90Qv0fj43M50mp4e1T20lHBmNBVU+XEbX4XCMaBndvx728g9XKVZDiG9fdpx8Sz69PWEs+Fgzx0RJhpbqbgsv7J+O3dXE3/Y/Ro/H1v/6YNjP1pqNqGqE5bNuxKiL/8lLLEYj0Z80u2gpIB31zsewV/bf/e53ueuuu3jxxRcpLy8nEomwb98+dDodjz/++HjFKISYYLKKSvonFupubxnVsjtUE370zMvTYdQN3UntlmlpdBuzefVgMz+uOMAPrjm/K9hzXUb3cGeQn7zVg1kT4p4Vx9Hr0unpzex/Xq9VuHq2kY17eqhoMHF1+XLgPf62/1fMLFxCd8jDS5Vv4An0MqtwKaXZqbkmSVnOfEz6dGrad3Hh5FXotSMfiZDqhj31ys3N5dlnn+Xuu+9m8uTJlJWV8e1vf5tnn32WzMzM4V4qREIIh8MD5qmvr69HHZUrPREvxyPRW9zzC87cbq0oCr+/bTmTs9P5r4oPqDga+5LcvqCKS2vFpmTS5AwROjGcb6TL6LoC8B+vOQmG4Z8XNTHJ6u+f8/1UOo3CInM3Bg385C0vM4uuI0KEQ63baAvuxRf0sKhsFctmXBdzHZKNVqNj3qTlBMI+qtvfj3c4E1JM4yAuueQSLrnkkrGORYhRd/KW8slbx41Ve8gonUZ+nOMS56bT5cOhGsnEQ07a8Ak3J83IM3es4GO/+gc3/fEdNn31apaWnbnDnjus5d16PzX2EBHrBQAcq/aj00CmeTpT6Yo5zoiq8st9Glp7w3xuscKSKR4aWrV41aGnlLVqQ3xljpVfHXbxk7fdvPrF72NzNfLB0V1csvBqMsyJ09EwXuZMWsaB5s0cat3CvEmXoNGcewfPVDR6jSpCJKiTt5TzSqdgzSuMdzjiPOxviy55OklxxrT/sin5PHX7ZXgCYdb8bhN7mh2n7eNw+/n1YRdb3QUctYWwGBQKfS1MDbeysEhPml7BYSzgA+1MjnfHtsb6b/a0sqdTwyWTDVw7Mzo17YHa4XvQryk18fHZk3jjSCv/t7uBSVkzyNPNlkR/gkmfzqzCj+D291Bv3x/vcCYcSfZCiAlBVVWOdvagJUIOsS+zfdOiKfzm5o/i8Pj56M9f564XdvL3Iy38/UgLX31+B9N//CJ/afJi1oRZNdPIzQvNlPiaKFSdfHSygZsXmin2NhJCQ0WNn11NAYabqv21wy38v93N5JlV/vNKPd6gE48zQk/f8D+3iqLwu1suwWrU861Xd9PU7Y65jqlifsllKGjY2/CmTLIzQpLshRAJLRKJ4HR2U9vSQZcnQE7YjWaEK8h96aOzeP3/W8WsPCu/ea+aTzz+Fp94/C1+t72GLJOBL8+2cFl6J9NzdGgGLfKlURSK/K2Uh+vJMCrsbwvyvieXds/pvfGf3VvPjU9sRqdR+PaSMKFwtHNib3ts/UQmZ6fz008vodcX5LYn3yUYTtzpf9VIBFtHFw3Hmmk41kxLU3vMIxTOldWUQ3np5fT5u9hR9/LwC+QoKijSP+ekxJ9XUgiR0trb22ls78TpB9BgcDUTtIw8CV49p5j93/4nnt17nJYeN4FQhCVluVw7t5g3Kyr4Zfvwr0/Dz3UXmHnnmJ8Gp5GvvufgmHkfH59djIrKn/bU87vtNWSY9PzyqhlY+IBudzt6rRFPV+yTkH3x4pm8VdPOs3uP83NjmGfPMo+Mqqr06jIJKtn4mgLkpGmIjGCBnHPldXbztk2lZV/0DkR7vYOZhY1nHaFwvi6a/HHaeuo4ZttHQcYU5hQt61+FVSVCl7sVp6eTtCm9oIGDLVuwmHJAk9qJf9yT/V/+8heeeOIJVFXl7rvvZsGCBdxzzz24XC6Kiop4+OGHMRgMVFRUsGHDBvx+P2vXruWmm24a71CFEAnCYknnqE+HQoRcpQ84e2/4oei1Gu5YOv2c4zDqFD4+y8iuqhaOh3L4UcUBflTx4Tohk7PTeeWLV2Dqs7G3rYeIGqbAMgU4GvMxFEXhdzcvo6rNycbqbi7cdID/uLL8tGXFVRVqHSH2tQbotkQTbENb9KRCl3ERk1XbaWWPNlN2LlnFxQC4esen2UGj0bJyzmd4de8v2FH3Csds+yjJmk1z73F6dcfo6YreXVBDGhRVh5Km4PS0Yy5VwHV+KU9VVYKKi+P2A+RZzn3K4ngY12Tvdrt54okneOaZZ3A6nTz66KO8/vrr3HjjjaxZs4b169fz6quvcu2117J+/XpeeukldDod119/PatXryY9/dz+wIUQHxpqhrtEH47oi2iwuSMUZ2jQOcf2VvHZKIpCsd7Lfy7LRS2bz7b6TvyhMNeVT+bKWUXotRoOHWrDHbaj0YDHruCwdaPqMmI+RrpRz/NfWMnlv/gb//naPva3dvOj1RcxPddCu8vL35u9bHEX4KnzowA5ARs5Wh9zLphFQ3eIQ+0hjmlKKGwPUl50flPrJiKrKYdPLLqLPcf/TmPXITp7GwBQ0FJgnUKOpYTKbQdJt5qYN+MCOl3HaXMeg6xuPP4e0owjHzre5+umqvtvuHWd4IUebyf6TBNEJsaY/3FN9lu3bmXlypUYjUYKCwv50Y9+xBVXXMH9998PwKpVq3jqqacoKSlhwYIF/QvwLF68mN27d7Ny5crxDFeIpDQRhyN2hkwATM3W4WuMczAnWPUarlk0hZsXTTntudqWA2jMEdocZv5+zM+hej+WkpF1KJuVn8H/XjudH+3t5rl9DTy3r4FMk54eX/TqXUHL3HwdiybpqXm3Dr0lg+IMLcUZWsI1u6m1XsD2RlAUiP00Y+LISivgyvmfo6uvFU+gF0e7i2MdB5iUVTZgP0VRKMyYxvGj7ZgKPRx3HGBW4UdGdKxQOMhbh5/EFepEH7EwKX8Krc5ayPXBea5TMF7GNdm3tbXh9Xr5+te/TldXF9/4xjdwu92YTNE/5JycHOx2OzabjZycnP7X5ebmYrfbxzNUIZLaWM5wFw6HcTq7UcwWAFyuXqyG85vi1R6KXj2VZWk5t+VUxpdHH31PuyJTySq2YszMOqdy8sw6Nn314/x+Vy2bazvY19LFsqn5lIR7qGurY/a0odedN0d8zAs3UGOexfbGAB9NmxgJ6VzkWIrJoRiPvRplmL4KYY8e3BaClj6O2w+gJ7ZhuKqqsqPuZbrcrRSaZuN3KWSnFWExZHGwZRuK2YOqqqc1sySacU32gUCA5uZmfvGLX9DU1MQXvvCFAStfnXzD9PqBX8yRvJFVVVWjGnOiqqysjHcIcTOSujc0NNDZ6SCgiX6nurocKEYfra3RGdWcnR1UVYVwuVznFMt4lZ9VVEpra+t5lzceMW/fvp2a400UKNE23IbmWrK8QdJPzHEw0vLDKjhCBix6FU3QQ9AfQNX7+xfjGvzY6/NQU1NDXl5ezDHX1NTg9XnOWOZIjhFUvXi1nTh7Fdq6APrOq7wD+/dxsREuviAdLog2ZW7ffpi6QN+w8er0LpbNiPB2A+xzZ3LgaPzeE4h+7xwhGwZz9De/1+nE6zecsbyTMcQa8+Dyu7u68AWNtLW1nfLYQPHcDDyBHsI+P1VVVWf9HnaF6mkJVmJWsvE0ZtAVOtZ/DE9XhPQ8aGitxaixYLPZqOo+e5mxGO3f+HFN9vn5+Vx44YVotVqmTp2KxWJBo9Hg9Xoxm83Y7XYKCgrIz8/H4fhw8gu73c6yZctiOkZ5eTlG48RoQzlXlZWVLFmyJN5hxMVI6261Wnm7+wh5JzoRuVuPo5jSKD7x2BAJUl4+95xXYBuv8gNAcXHxeZc3HjHb7XZyHW6KS4oA6KnPJCsz85zLf+btXYTRUpqlw2Ixojca0BuNWCzROweDH5tNacyaNWtE3xO73Y7Z4T5jmSM5xr7GN1Ea4VirGcuJpshzLe9M3/dY451ZaKErEGB/m0JFMIN/j9N7AtHv3aGWHiZNiq5rn5GVhdlpPmN5wIg+x8HlZ+e0kG41nfZ4VvFcqtt3EbB4KSvLZUn5mcvv9Tp4de8r6LUm1lz0Fdoa7RxqcfWX2dDQALhRjR4m5c0i4A0zv6T8vP4+IfbfOb/fH/MF7riOs1++fDk7duxAVVUcDgdut5srrriCTZs2AVBRUcHKlStZuHAhR48exeVy4Xa72b9/P0uXLh3PUIUQCeKoN3pNUpKZ+NOj+kNeDrVuQ1F1NLYb4h0OAEtL9GRoArzZ6uPvR0a3yWYi0mp0TMmNTodc0/sOnkDvkPtF1DDvHn2WUCTA1PSP0tZoj3ZmPWVsf8Svg6COXq894Sf5GddkX1hYyMc//nE+97nP8ZWvfIV7772XO++8kz//+c/ceOONOJ1O1qxZg8FgYN26ddx+++3cfvvt3HXXXf3t+kKI1FLt0wEqxRmJn+wPtWwhEPJiCU4mFE6MNlyNRqHc7ESrwFef34nLF/uY/2SVZszEFMknEPGw6eAfCYb9A55XVZWddX/B3tdEyKXH5mjnUMs2tu3/Bz09PQML80bXO+h2t41X+Odk3MfZ33rrrdx6660Dtj355JOn7bd69WpWr149XmEJIRJQnz9IvU9LhiaI6QzL2SYKb6CPgy1bMektpLlLge54h9QvQxvi5mlpPHvMzXf/tof/d+NH4x1S3Bkj2WSkZ9Pprubtw09z+ZxbMenTUVWVXcf+wtH2HaTpstGZcpkyPTqmvqVpiJmXAtFm4z5/NzoKxrMKIyIz6AkhEtaWY52EUcjV+c++8yjT61QyjD2k6T3Mnamj0T78rH17G/5BKBJgydRraXSefwet0faZ6ekc8Oj59XvVrJ5XwifmT6xJYUabgsJ06yXozQot3Ud5YfdDlOXMx+5qotdnJyutkFlpV1Lj3TN8QaoGky4dd6CHjAQewCpz4wshEtammuit0bxxTvZ5ORpuXQ3z844yNbOJVSvMrP0nmJZVj14TOG3/gy1bqe7YRVZaIbOLzjK/bZzoNQpP3b4Co07DF599j7be2BcTSlYaRcOV8+7g4mmfRKvRc8y2F0+glym5C7im/MvoNbENGU03ZqGqEcJK4rbby5W9SGmRSJj6+voB22bMmDFgSKg4Xyp9bnf/CBunszvmBVM2VbejV1SytKcn2LFiNbi4bnUaBj209RXiCljoOn6I+bONFFps5JkdmH0Ggoobt7+HY7Z9VB7/O2aDlavmfwGtJnF/VhcWZ/PQJ5dw98vv89knt/C3r6wizRBbvIm7JM/50Wp0zC+5jFlFF9Pj7SQ7reiUzzC2dvh0UzYOdwshJXFPoBL3WynEOOjpbOexul4mN0evHJ3tLXz/Zs576Iz4kMfjI+TvwNbbBEBrex2NjUVnXTClw+Xlg7Zu5phCaMepuV6vCTAntxoFeGsnpJdFJx6q3B/gwDEjq1ZNpSyjmfnTfNjZxcb3d0VfpzVy1fx/xmLKHp9Az8Ndl83hnWMdvPhBI/+04S1e+dIVWIynT7rT6Q1T67ew77AXpzdCIPMjZKgeTLYgs/KSL3XotYZznu/eYswCIKR4RzGi0ZV8n5gQI5SRX9Q/m5wYG2aziczs6JjzXktsa1xsqol2hppjDhEYp+nwp2Y2otOE2fyen7oWMwtPmXlVVRU6PQXYvblEuo7wsblp5ORlU5Yzhym5C0gzToxJaRVF4enbL+N2tvLiB41c9ujf+Y8ry/mnC0pxB0K8e6yTP+yspeKoA5UMFH8Eq1FB9Qdwaq28Ux/geHeYLDQk77x8I6PXGjHozASCXlQ1MdeZkGQvxASSSs0Om6qjt1Bnm0NU9Z1l51FQmBMkN60PVyCdg0dd6C1Dt9dGVC1NHQayp5Vzdfk1Yx/YGDDotDyzdgVfe2Enf9hVy9qnt562z7xMHVq/nSUXFGPQKlT+YzthSy62nDk0OMN0WMuZrzbFIfrEZDFm0xVqpaaxCkX5sDtcovx9SrIXYgJJlWYHVVXZVNNGbpqREkMPYz0JtkqEi+Z4UFWod04FzrK4fRLQaTX87pZLuOfKC3hs21EOd/SSYdIzI9fCHUtn0LhvB7/c0YjhlDYUE0FWzzGxvSHAoc40atRSLoqoaDSJPSxyPKQbs+hyt1Ln2E1IFx2NER2qd2NC/H1KshdigkmFZodqWy9NTg83L5qCprdzzI/n1XViNUZo7yvAE0ytpbRn5mXwyKdPXwXuTIsLahSF5VMMtDZ34DTksqMpwPIpyT1FeSzSDdFlczPzTP3j8hOJJHshEsgZ15pPsUGym6qjV9arZk/Cv/uDMT2Wqkbo0zcQiUBr36QxPdYZY4hEaG5uprq6un/bjBlDr2iXCBRFYYqnDr8+nYMdJiZZtSmfTAw6M2oEFG0o3qEMKdU/HyESypnWmk/LSdyZucbCG0ejK/BdNauIv+0e22M1OA4S1nhoaDUQUONzherusrOh0c5mZ/QnudfWzoOf/1RcYomVlgizws0cNMxky3E/y00pdkY6iKIoRAJatMYQqhoZ0G6fCBIrGiFE/1rzeaVTsObFtuZ2MvEFw7xV28ac/Aym5VrH9FiqqnKgaTOocKQhfutveDw+fPoIoXQIpYMz7Kax8Uw30hOHmQAXlxrwh6DKlzVgkZh4i0QitLd20nCsmYZjzXS024mEx7anfCSgAQX8ocQbby9X9kKIhLLlWAeeQJjV80rG/Fgtzmoc7hZM4QL6PEHiOUr+XIYnJoILCnUc7w7R5jJR0erj2ngHdIKts4s3jwXZF3ADcOyojxvS7WN6zEhACwTxBvsw6S1n3X88yZW9ECKh/P1I9Bb+tXOLx/xYB5reBsASnDzmx0p04XCY6urq/v+am5uJZd48RVH42HQjOiL89kgfx7vGYZxkjNJz8sgqLiaruBhzTu6YHy+a7MF34gQjkUiyF0IklL8faSHNoOXyGWPbhNHRU09H73FKs+eij4xtc8FEUFdXR8WOFzjUso1DLdvYc2QbAX9s0xRbjBrmmnrwhlW+9Ox7RCKJczt/PEUC0ZTqCybOCc9JkuyFEAnjeFcfRzp7uWJmEUbd2E5E8kFz9Kp+YdkVY3qciaSkrIgp00uZMr2U3PyRNWqU6L0syzewua6DH76xb2wCTHQRDYQ1eBMw2UubvRCpTo3Q29N7TgvVjLbXj7QAsHru2LbXd/Qcp6W7msKMaRRkTAGOjOnxUoGiwN0XZGA74Ocnb1YxMy+Dz3/k3IcPqqo65HDERJiNblhhHcGwj3AksYbgSbIXIsW5+vpoczlI643O7R7rQjVj4ZUD0elXx7JznqpG2FX/FwCWTF09ZsdJRZkGDX/58hVc+su/c+fGHWg1CmuXTD+nsnxuP4cadpE7PXoDOpFmoxtWSAeGQMLdypfb+EIIzGnRnuCZ2VYsceoJbuvz8VZtOx8py2Vqztj1ZK7r3Iujr4Xp+RdSkCEd80bbnIJMXvjnj2HWa/n8n7bxlT9vp9l5bh3W8gtz+psVSsqKRjfQsRKKXkMn2q18ubIXQiSEFw80Eo6o3HLh1BG8SiUrU0N6porF4MIdGP5ExRd0s6fhDbQaHYunJMogseSzckYh7//bGm77vy38YVctf9hVy4L8dIwaDRmHu9FpFdp7JuMOQZorTKElia47w9G0Gr2yT5zhd5LshRAJYeO+4wDcvCi2ef+1SphpWfUsu+Fkgj9MIKwnzaclQvC0/YPhAG8e+l88gV4umnI1FlPW6AQuhjQzL4Ot37iWJ96v5cX9jWyuayeiaqDDf2KPHADqDvvIS9eQrzEnx5K5oWifAn/Qiy6Bkn0SnU4JISaq9l4v79R1snxqPmXZZ29GsKarLCioIi+tiw5bmH1HoKOvAI0SZsEMH51pO6hqfhdvIHortcdj4+3DT2F3NTEj/yIWln5sjGskAEx6LV9dPoeKr36cl1blc0nwAJ+a1MsnJ/Uys+VdZoabmZ6jxe6OcMS6AJuSGe+QR4EGvdaIP5RYY+3lyl4IEXcvfNBARFW55cKzX9UrClxxMZh0flpdk3jlb9Vo0zNYaJ1KY28ZRs8hFs0Isvv4a+w+/hpphgw8gV4AirNmc+msmxJu3vJUoNcoZObmUlgWXWyo9fAh9GovC2eamNkdYtNRN/WaYtpdYYqsCd7j/iyMujT6/N2ojO30vCMh33ghRFypqsrvd9ai1SjcFMMt/EUX6CnKA7snh8beMk6dvyWsajnaYKLAs4zFU66lOGs2ETVMcdYsLp99G6vmfw6NZmInkmQ0JVvHdE81KvBmrZ8+f+IkyXNh1KUBECG2SYnGg1zZCyHianNdB/tbu7l50RQmZaQNu69J56X8IiMeHxzvOfOJgQY9C8s+xsKyj41ytGKsWEO9TIm00xCcxLv1AcZ+ZYSxY9RHv8dh5fS+I/EiV/ZCiLj6xbuHAbj78rOP6y+ztqDTKWzbA6FIUnTnEqcoVLspydDQ0hvGqSRO57aR6r+yVxLnyl6SvRAibmrtvfz1UDMfnZzHJVPzh93XpPOSY+6i0x6mvmWcAhTjSgGWTTaiAE3aIsIJPsX+mZbRPZnswwmU7OU2vhAjEA6Hqaur639cX1+PGpnY7Yvx9LN3DqOqsO7yuWfdt9jShqJA5QcBYPjb/WLiyknTMCdfxxGbid2ubNbEO6BhnGkZ3Qt15YAibfZCJIKIqlLr09HUp1C59zhWk56puuHnhK+rq+OBjZvIKoq2KDZW7SGjdBrDX5OehxPz1ge1RoxGIy5XL1aDeayONq4OtHXz+I4aZuZZuXHh8B3zzMYIeWlOvEETxxpcJNhS4WKULSnRU90Z4F1nLqGwik6rxDukMzq5jC6AubUTCKIoGow6M4GgL77BnUKSvUhJDreflw400uoxAdBuiw7N2g00/+UwP70pm2VThk7hWUUl5JVGk1N3+9jeTz45b31RlhZbr59WWyPFhol/VauqKute3EU4ovLz6z6CXjt8i+KsMh8aRaWlbxJgG58gRdykGTTkRbrpDOXyZq2Pa+dMvBNcoy4Nf8hDMJIYCV/a7EXKCauwcX8Drb1ephiCXFcY4dtXzOezi6dRmq5jR2svl/7y79z2f+/yQWt3vMPFnGYiI9NCZraVtLSJ96M3lM3tft491smnLig9+6I3mgjTigMEwnocntzxCVDEXVHYjoLKk3vcqGqCN94P4WSPfG+4J86RRMmVvUg5VS6Fzj4fS0pzmO1pRNGmYdbrmJlnJcuXziUzi/nl3k427m9g4/4GLp9ewKrZkygvyqKjvZuangBHg504vQE6XSaCboWDVU3kphsxhxWSYQ6wseQOa3nssAuTTssjn1561v112R4MepXm3nxUuT5JGSaCXJDeS5Utk/ebAxTGO6AROtlJzxfqjXMkUZLsRUrpCmk41AeZJj1XzZ5E477G0/ZZOimD91Ys4bUjLfzy3cNsqmnn3WOdg/bynvh/9E+ovf8OQBplfshy+8hLN41ZPc5HOBzG6exGMUcbvvtcfWAenzbRQFhljzcHd0Tlfz+zjGm51mH3V1UVfV4fkQh0uAtiOoYaiUzMddDFaVZkOqhyZ/LUHjffWhDvaEbmZLKXK3sh4uADrwEVhX+6oBSj7sw//hqNwifnl/LJ+aW09XrY09zF4Y4e7HY7O453UlxUSJbZQMfhfRhMZvJnzKO118M7h47T5NPy+I5a/ml+CYm4KGdjYyON7XWETkxBb3O2YdGNWRfDfqGIyps1ftwRPddNMXPH0rOvc97eU4fWHKKxXU8wYojpOO4uB3+ye9h3otuks72F799M4q+DLk5TavKxaJKe9xoCfGYazI93QCNw8ja+L5zCV/Y+n49PfOIT3HXXXaxcuZJ77rkHl8tFUVERDz/8MAaDgYqKCjZs2IDf72ft2rXcdNNN8QhVJBGnN0BbUEueXmX6Wa4oTzUpI41PzE/jE/NLqa6uxt3bQ15B9GZ9r0ZF0UBuupHcdCPG5sM0kcb7vRpePNDEhbkGvp6A7Y0WSzqZ2dH3wGgyjvnxwii8Ue2jtTdCvs7Hl2bFdnJxuPU9AGqaTKQXx348S25BfyfKIeMZNISyubl5QrYLp4JbFqWxv62HNxo0XDUn3tHETqcxgKrgC7viHQoQp2T/61//mqysLAAeeughbrzxRtasWcP69et59dVXufbaa1m/fj0vvfQSOp2O66+/ntWrV5OefvbVsIQ4kz3NXYDCzPSxGxevKDDFBIvmz+S5fQ3sc/j5wbv1/GnWLLSac2tvjqgKPX7wEMaDkcwxzkmjfRs8jIYa7WRcvRGmZGmZE+5Cq5l81tf1+bpp6jpM2KOnq1c7omR/NoOHUB7aeZhQ5sQf5ZCMrpxhIi/dxVtNKu7gmYfGhsNhWpra+x87bN2ouoxRjeXkJDpZWc1nPYaiKGjQ4wv3oqoqihLf4YPjnuzr6uqoq6vjYx/7GAC7du3i/vvvB2DVqlU89dRTlJSUsGDBAqzW6JXH4sWL2b17NytXrhzvcEWSCEdU9rV0oVdUJo9Dh/a8dBOf/8gM/m/HEZ4/YkPz9Fb++JlLMQzTdDBYl8fPwXAGTWohoToF8IFhLmneML217Sw/y4xz52qkt8GHu0oOhFRqLfNwK+lMy9Fy5XQjLdVDFnOaI207UFEJ2sZmUP2pQygtNfk4w4m1JKmI0msVbihP43c7+3i12s5FFwy9X2NjIxt3OyhyRPPGoXo/lpLRHfY2eBKdsx1DoxoIqX34Q25McZ4cYtyT/UMPPcT3v/99XnrpJQDcbjcmU7QjU05ODna7HZvNRk5OTv9rcnNzsdvtMZVfVVU1+kEnoMrKyniHEDcjqXtDQwOdnQ6OuiL0BUJMxkVvV4DW1lYAurocKEZf/2NnZwdVVSFcrqFvvZ0sL6DRD/n6wY+Xp3k4akjjuX0NtHQ6+O/LSjHpznyF39DQQFuHg+22APvtXiKqBb0aZHq2BoMWOtu76NFls+VYJ5WNdhaYfcPGO5Samhq8Pg+Onj46+qBDyUEbNpLf1YfVAP6AD0xZ/XUMaHRUVVUN+55s2HYIa160v/Sx/bvR5WeR3tPH5gZw66xkh7q4uDAbjyeE1+ehpqaGvLy8M8YYUUMc8b2HFgMhZxpen52+vuja9EF/AFXvP+Njf8CHs6dn2M908Ofo7HYSMERiPsbZ6nDyPY61vOHKH+r7PtLyY3nPGxoacIRsGMzRE9JepxOv3xC39wTA6XTS1tbGZQXwewX+UNnAJ8rMQ14l19TUoElPR5cRvdLWmNII+D8sL+Dz43T20tbWBoDNZqOq+8zf6zO9J5r0gpiP4feE0Fpg1973SNeObNjoaP/Gj2uyf/nll1m6dCmlpaX92/T6DxezOHmr49Rtp26PRXl5OUbj2LdBxlNlZSVLliyJdxhxMdK6W61W3u4+wgFb9EqzPENHlsVC8YkZr9ytx1FMaf2PDZEg5eVzz3gVe7K8vDO8fqjyfrhsFt/Z1so/jrby7ztsPL12xRn7DHzgDLOp10tv0EuGUc+MUCfmiJ0pxTOxWCwcbvsAa7oWz6Q5bK3vZKfLwP91mPi/NQvIMMXWgW1rbRuHmwzYq5Xo8rCZ0bo21kKhRUO+MY/CzMwRvSczunX9V8m+ni5sIR+bm7R0+SLk+juZobWTYS0DwGxKY9asWcN+jtXtuwjXBllYegXv7avBbErDYoleGemNBvRG4xkfGw0mss4S/+DPsT07C2fYHfMxzlYHu92O2RF7eWcqHxjyGCMtP5b33Gq1cqilh0mTouvNZ2RlYXaa4/aeAGRlZTFp0iQmAcsntbKlVaU3s4QrZ00acfkGk7G/PICAN8z8kvJhO26O9D0ZfAyX344XHyVT85hRsPiMxxks1t85v98f8wXuuA5a3bx5M2+88Qa33HILGzdu5LHHHsNoNOL1Rocx2e12CgoKyM/Px+Fw9L/u5HYhzkU4onLM4SI3zUimbvznsTfrtbz8xY+xdsl0djU6WPzTv/GLdw9j74ve/lNVlbdq2rjuD2/z5deO4gpGWDYlj69eOpsSjRfNoPNcrQIrZxTy1eWzKTRr+UtttMxN1W3DxnGko4eb/vcdvrmrm86QmSyThqWleqa4a5kabqMkQ0NHX4Qq/QwO+s0Ew+f2XvlUDQfUEro8EeYV6JjsPcZIWivDkRAfNG1Go2iZM2nZOcUgks/qadHv46+2HY1zJLHTqNET8F6v4yx7jr1xvbL/+c9/3v/vRx99lJKSEg4ePMimTZv45Cc/SUVFBStXrmThwoUcPXoUl8uFRqNh//793HfffeMZqkgiLZ4QwYjKzHwruOLzR2fUafnfzyxn1ewivv7CLr75ym7u+UslxZlpOL0Ben3Rda8vKrQwKx3mzjx7b7ScNCPXTU0Hk4XH97Vx9W/f5PYl07h7xTwWl+agKAqqqrK3pYvfvFfN/75fRziiMjdTR0awnUXl0Svxyio7emOAhXNnYusL88ZBJ/UhE398v45bLpw6onr2+oJsD+XhQUd5oY5lkw3sibGN/qSajt30+buYN2k56UaZokhEzc6CC/LSebWqmYauPqbkJP4CCRo1epfa5UuxZD+UO++8k29961s88cQTTJs2jTVr1qDT6Vi3bh233347Go2Gu+66q79dX4iRaugLATArz0o8R8EoisLnls7g6tnFPLO3nmf21GN3+5mabWFhcTZfvXQ22b4u/t/W2K9ctIrCNy4u40srL+TOjTt4urKepyvrmZqTTpbJgMPjp8npAWBeYSY/Wn0hppZDPLrz9MmEAPItWi4MVtOcNoemXtiws4ZrimNrFnO4/Ty9px43OkroYtnk0hH3QA6Fg3zQ9BY6jZ4FZVeM6LUiuaiRCLYOJw3Hoj3fW5vb+ez8mfznu8f5zXvV/PcnY78tHi8a9CgouHxd8Q4lfsn+G9/4Rv+/n3zyydOeX716NatXrx7PkESSauwLYdBqmJKdzrH6eEcDRRlm/m3lfP5t5elThFRXn9tc/EvKctlx92peO9zCU5XH2FzbQZcngEGr4TMXTeWGhVP41AWl6LQa3mg9PGxZWlQWGrzMnT6VN6vbePl4iEtqHQw3J027J8TrNbV4g2Fma3rJVx0oStmI63GodSueQC8LSj9GmiH2uRDE8FRVnXCzCvp6nLztUmjZF+353l7v4FvXusm3GPndjhq+e1V5zP1U4kVBwai1pt5tfCHGW73TS08gwtyCjHMe5z5R6LQaPlVexqfKR55kB1MUuGRqPnnpRp7ff5x/31RLjUfD/dcuIt/y4V02XzDMz3c18fJxN6oCn5xfAjWtOIdfKXhINlcT+xrfxKS3UF5y+XnXYTjxnDI4HtxuD4cadpE7Pfo3EB2PfmPCzypoys7tXz7W1evGoFW4e8U87n19H7/eVs1/rCqPuazonYKu/jsFLU3tzCk6hy/qCJm0VpyBFgIhHwZd/O5QS7IXSe2dRicAs/JGd3KNVDErP4Mbp1nY6Qjx2+3V/GlPPZ8uL2NOQQbHHC7eONJKa68Xi17huoVTmZ5rZU/NyI/jD3l458ifiKgRLp9zW/9Uo2MlXlMGx1N+YQ5TppeefccE97VL5/A/bx/kZ+8e4hsr5pJmiC2NeZ3dvG1TB9wpmFnYyLx588YyXEzaDKAFl89BruUsKzyOIUn2Iqm91xxdhGJGntwSPlc5Ri2v3DSPzQ54oOIDnqo89uFzaQY+v6AIQ9DDpBFMQXwql8/B5iPP0OfvZlHZKoqzZo5W6MMa7ymDxejINBu467I5/OTNKn6/s4ZvrIg9WQ++UzAeoskeSfZCjJVwJMKedhdZBg0ZJv3ZXyDOSK/VcNdls7nzktnUd/Vx1NZLkdXMRSXZ1NXW8uiWIyMqT0Wl291OU9dhDjRvJhj2M7NgCYsmrxqjGohkcveKefz83cM8uOkgX/jITKwJ/Pdt0kZPKOPdbi/JXiSt/a3duIMR5mUl7g/BRKPTapiVn8Gs/JE1i+Rnq5Ram0k3uCm/xEW7+V1e2bs5WqZGz2WzbmZmYWpOFCVGLs9i4p4ryrnvjf38+M0DPJjAPfNPvbKPJ0n2ImltPbEG/aQ0+ZqPpeE6u5l1HtasMjNtMkB0+lqvX0EfsVBaNJ2SrNmU5szFpJdFrsZSvDqnjaV/v2I+T+yq5efvHuZLHx2fpp9zYdJaAEWu7IUYK1vqJdmPhzN1dssw9DIntxqtRkebDXo0s+gLWKg/dJx1yy5k5Zxr4ht4Comlc9p4rBo3msx6HQ/90xJu/b93ufvl91lXlphLFGsUHenGjLiPtZdfQZGUVFVl67FOCtL0ZOiTd0hVohjc2W3yJJW5edHJgd7Y7KXBZmbh8ux4hpjyztY5bTxWjRttNy6czMdnT+KNI61MURN3Rj2rKZf2nmOEwkF02vg0Kyb3wGORsmrsLjr7fCydlBH3daRTTW62hqsuAVVVOOqYTW19CEY0O76Il/TcfLKKi8kqLsaYmRXvcM5KURSe+MxyCiwm/lDdR084MfvnWE3RFe/6/PG7updkL5LSlmMdACyZJEPuxpNWCXPtFWZ0WqjtnkGPX+a2F2NrUkYa//fZSwmrsNeTTZ9//Be7OpsMczTZx7PdXpK9SEonO+ctLUq2ZK/S53bjcDhwOBw4nd2Ew4nS0UplWlY9WZka9h+Bbp/cthfj4+NzivnczHR8qo7XjvrwBhOr/f7klX08e+RLshdJaUeDnQyTnpk55niHMqo8Hh9drg5svU3YeptobK+jsXHoRW3GW665i7y0Lto6wuyKbYltIc5ZOBymurq6/7/l+m6mGVz0+FReO+IlqCTOLf1ESPbSQU8knW6Pn2pbL6tmFaFJwvZ6s9nU3xmu15IYQ9aMhghTMxsIRzRs2tKHqibbHRWRaOrq6qjY8QIlZUUA7D26jam6qZgLsjnUGaLPcgFz1eY4RxmVYTp5Gz9+bfaS7EXS2d0UPXu+eHJenCMZubFYoCUcDlNXV9f/uLm5GVUdvducKiqL53jQa0Mcd06mx3UQfeJ2jBZJpKSsqH++/9z8bBQnLJ9iwKRX2NMCh9SpTO8Lk2+J7+p+ep0Rk94iV/ZCjKb3TyT7j0zOA7zxDWaExmKBlrq6Oh7YuImsoui83Id2HiaUOXoLzfi0nZQWBOn1W2l3FwIHR61sIUZKURSWlBiw1RylyTyNvx7xcdXM+K99YDXlYHc1E4mE0WjG/+RDkr1IOrsa7QBcPDkXV3ti3MYbibFYoCWrqIS80inR8mvycYZHZxEQb8BFj7GGUBjquqchQ+zGxlB3ZyCxOqElmvxAJyajnmOaMt6o8TNZn0cRgbjFk2HKxeZqpM/v7O+dP54k2YukEgqF2H6sg6J0A672Zurr61EjYzgUR43Q29OLwxG9m9Db0wP+YP/jri7HgB9pgBkzZqDVxve24mhQVZXttS+hKkEO1JrxW+K3VneyG9w+vefINgL6qfENagLIUV3MnWvijWofDekzUcMdLIxTLFbzh530JNkLcZ62fXAYuzfIdKuOR7ccobFqDxml0xirlcpdfX20uRyk9UanFT3eUos23Up6cXR8ee2R3Tx8PIf5i4MAONtb+P7NMHv27DGKaPwcaH6Hxq5DGMJZ1DZD2dzYXqdGIjQ3N1NdXd2/LVlOgMbS4PZpnPGNZ6Iosmr5p3lmXv3ASaO2kD0tARaXGMY9jnj3yJdkL5LKgc7o7elpRXnklRbQ3d4y5sc0p5kG3HbXn9JbPi3NjCW7oP8WerJochxiT8MbpBkysTjnA4djfq27y8Gf7B72nTgFS6YTIJGYctI0zHYdpDajnMoTPwljdQFwJvGeWEfG2Yuk8kFnHwDFo9gBTQx0zLafd44+g1ajY9X8z6FVR96vwJIbPQHKK53S33FQiLFkVAPMCzdgNSpUtgSp9Y/vkJEMcx6KoiEQik+nYbmyF0mlP9lnJNdkOmMpK0NhzhyVsqx6tEqY9IUafEEPeqWH4Cnj5Z2eTo60vceRth3otAaumPNZci0lgMygIyYGI0E+OdfEX4/4qPVn8FRtH9eM0+KLRl0aqxf8C1ZTfGaWlGQvkkY4EuGgzU2OUYNRJ+2/Z2PWeZma2cCyG09e4dgAyJunAbqB11BVDcUztLxv30uwM3pFkmnO54p5d5CVVhCXuIU4Hxajhk/ONfHyBy7+dMzDjL/v54fXLByXBbMKMiaP+THORJK9SBqHO3rwhCJMtiTONJmJSWXm5DDl+QfRaiI0t4U42qAjd9oCIqqGjkN7KCnOYfqMHPSaHtSIBw1aynLmMS1vEZNz56PTjn8HJyFGi8Wo4eJ0O0fDxTxQ8QERVeX+axcl9QqZkuxF0tjVGO34UmCSq/ozU7lgrpfZM8OEIlqqHTOp+Pte9JYM0suiTR/tNhVfyEJm2TIA7M0NfGPF3FHsQPfhYj5Agi3mI1KFWRNh/eJs/utggB+/eYBjDhe/vXkZ6cbkvFiQDnoiabzfFJ1Mp9As57BDU8k27GH2TD9OFxzoLKfLlzPuUSTyYj4iteSbtGy+62oumZLPM3uPc+mjf2dbfWe8wxoTkuxF0ni/0YFBq5BjGuZrfcokOA6HA5erFzWSCjORqeQY3idDX01Pr4bXt+nxh8/Ui37sl9E9uZhPZrYVS4Is5iNSU3FmGm997eN8dflsDrQ5ufz/vcENT2zmr4eaCYTHcEKucSaXQCIpeIMhPmjrZkF+Otph2t0GT4LTamuk2JDcw/RUNcLUolqsehv+cDZbt0fw+c88/Mfj8RHyR6+8AVrb62hsLGLevHnjFfJ5kmYCMTIGnZb/d+NH+eziadzzlz28UtXEK1VNmHUaStK1zKl1Mi1HR6c/g5A6Mdv1JdmLpLC3uYtwRGVhgQXCw49jPXUSnLS05B6iFwj5ONzzJrmZNvzhXDp9HyMQfOusr0vEZXRjNfFPVkS8LJ9WwJZvXMOOBjsvftDIXw/Uc8zpobbHd2KP6Sio1B/1cVHJxGrbl2QvksLJxW8WFqRzrG1irXQXC40SIcvo5MJyPYpeJc9sxxUYfs14u6uJLdXP0ROw0ePOoocrUJlYP1DnaiKfrIj4UhSFS6bmc8nUfL4y18qBpm3o8oo45gjx8uYj7PNk09Sj0NQTJs88lal0xzvkmEiyF0nhZE/8BfkWjrXZ4xzN6NHr4dKLVOZMP4pOo0LeycVmjgEwO1ODU3eY6vZsrKYc9FojPV4bDfYqGrsOATDJfAGVRzPIK02NRC9ELFRVjWmNBq0GJmfpmJylo3t3O4aQl7Tp89ha78dOEYGIlQURFZ1GQY1EsHV00XAsutpmS1M7c4oSowlJkr1ICu832clJM1CWEf91q0dLhrGHz1yfjjUdvAEdnf5c9u2qR2cyM/uCyWQYXaTrnHj17bxX++Jpr8+3TmbxlKtxdUaAI+NfASESmNvt4VDDLnKnRzv0tjS1AzfGNMR0klXLDeVmXnivmR5DDptq/Xx8phGvs5u3bSot+6JrdLTXO5hZ2JgQTUiS7MWEZ+/zcczRxzVzi5NmUowcUxczc+pQVYXKg+DUzSTdYqW+sQa9RSFjahHt7iKajlTz1UtmMXvhZNz+bgIhH5lp+eSkF5NvnYyiKLg6q89+wAQyeO32zs5OVE1idaKUlfuSQ35hTv9KgiOl1ShM89RwTHcBjU4L7zcH0QGm7FyyiosBcPW6RzHa8zPuyf6RRx5h586dBINBvvKVr3DxxRdzzz334HK5KCoq4uGHH8ZgMFBRUcGGDRvw+/2sXbuWm266abxDFRPE+03RW/gXl+Wd/uSg9eb7XH1gTuwTgrLCALNyaomoGl75uxe7O53pC88Us4JetTC76CPjGuNYqqur44GNm/oXyNlR3UFaWWGcoxpIVu4TABpUZkeaqDbP50B7kNnadLLiHdQZjGuyf//99zl8+DB//vOfcTqdfOpTn+KSSy7hxhtvZM2aNaxfv55XX32Va6+9lvXr1/PSSy+h0+m4/vrrWb16Nenp0tFGnO79E53zPjI5F/AMeG7wUDubsw2LbrwXt4ydR9fGRy9wE1a1HLHPoa1jJ/rzXJwrHA7jdHajmKMFTYQTnqyikv5lgdMyx3/in1icXLlPpDYtKpdNM/LaER+NadPJUBviHdKQxnVSnYsuuoif//znAGRkZBAMBtmxYwdXXnklAKtWrWLr1q0cOHCABQsWYLVaMZvNLF68mN27d49nqGIC2XXyyn7yEFf2fDjULjPbitF0fm36JxPnyQln+lx9MEpz8hxt30mP4QjBkMJh+1z6gqOzBGdjYyON7XX9M9bZnG0Eg8FRKVuMj3A4TEtTOw3Hmmk41ozD1o2qpsJkUBNDSYaWOfk6vNp0OpTceIczpHG9stfpdOh00UNu3LiRlStX8tZbb2EyRXsY5+TkYLfbsdls5OR8eDafm5uL3Z48PazF6FFVlfcb7UzLsZBvMY35IJiTiTN04ibTaN0pONy6jZ3H/oIGPZv3mMgoje0uVqxtxxZLev9QtPM94RHjr7GxkY27HRQ5op/hoXo/lhLfWV4lxtNHywzUdvho1eQSDKvotYl19ywuHfTefPNNnnvuOZ544gm2bNnSv11VVRRFQa8fOETo5PZYVFWlxtralZWV8Q4hbk6te0tfALvbz0W5RiorK2loaKCz00FAE/0OObudBAwR+vqi69wH/QFUvb//sT/gw9nTQ2tra3T/zg6qqkK4XK4hj11TU4NWp6DVR7+PGkUh4Pefc/mqqmILHaUjdAAdJnqOZNJud6LJGlgeQF9f32nld7W38Nv6Dv7RGZ3W02Xv4MuXzmfKlA9vL9fU1OD1ec4Y49kee30eampqyMsb+s7JaJc/+DMECPhiLy+W+M835rN9rqP5nlRWVlJTU4MmPR1dRrQ5SmNKG/Z7F4/3ZLS/J8CAOg71PXGEbBjM0RPbXqcTr98Qc/kBnx+ns5e2tjYAbDYbVd1VA/72R3qMXE8znZap7G1yMz/v7O/JcEb7N37ck/2WLVt47LHH+P3vf09GRgbp6el4vV7MZjN2u52CggLy8/P7O1QB2O12li1bFlP55eXlGI3JfeVSWVnJkiVL4h1GXAyue83eeqCWay6cxZIl87FarbzdfYS8E71h27OzcIbdWCzRW+J6owG90dj/2GgwkZWZSfGJ/Q2RIOXlZ17hzW63Y3acubyRlB8I+dhW8wIdjgOkGTK4ZsFX2OHeg7l132nlAVgslqHLL5zC/MUXR+Nrbjgt/pHGPPix2ZTGrFmzzvidG+3yB3+GRwCDKfbyYon/fGM+2/dmtN4TgCVLlpx3eePxnoz29wTAcJbvyaGWHiZNmgRARlYWZqc55vINJiNZWVn9rw94w8wvKR/wt5OWlsZb7x8l4I2OlQ8GVUwG0xnLLArbcKhlHO3SclFZ2lnfkzOJ9Tfe7/fHfIE7rm32LpeLBx98kN/97ndkZ2cDsGLFCjZt2gRARUUFK1euZOHChRw9ehSXy4Xb7Wb//v0sXbp0PEMVE8T7JybT+chQPfETVFgNcbh1O6/s/RkNjgMUZkzlk4u+TqY5cTsOCpGKTjafPLvPzbP73Gyp9+PxnLn5RKeGKYo48IXgUGdi9YsZ1yv71157jZ6eHv7t3/6tf9uDDz7Id77zHZ544gmmTZvGmjVr0Ol0rFu3jttvvx2NRsNdd93V364vxKneb7Sj1ShcVJId71BI07tZME9PYZFKfk41GkVl0qUaFE0XOYYdgEruZAfv23YSsYXRavQsLLuCCydfhUaR8dliYot1RrqJJj03v3/cvDEz66z7F6lddGgKONgR4tIEusk8rsn+1ltv5dZbbz1t+5NPPnnattWrV7N69erxCEtMUP5QmN3NDhZOyibdGL+pYKeXqlyQfxCrwQ0FJ09KnQBklWoAL1APgKoDky6bGUWLmF98GWbD6PS4TyaDhwp6vF7SrDLVb6I7nxnpkomOCLPzdBzqDNGpTZyLVJlBT0xYlU0O/KEIl06Lz+1vk87LddeaKZkEquqmy5vFnt1tdLnNTFu4GBWFmh07yc0rYcHSaJt6R3MnX7/sAmZPTa0fwJEYPOLB5e1CH0zu1QmTxfnMSJdM5hfqOdQZojGQOHPDSLIXE9a2ehsAl04rGOcjqxSmdzAlsxGNouN4Czi1C/GFTRytbUJvUQir0T+tQBC8Ph0hNXqVqqqO4QoWJ5w6VNCU5B1uk0UiLwIz3rLNGkoyNLT0Gql3heIdDiDJXkxgW+s7gdFN9pFImPr6+gHbTm13jBBk+UI3JVlOgmEd/3jHRaPDzMLliXO7Toh4SORFYOLhgkI9Lb1+/trk4V/iHQyS7MUEFYmovHe8k2k5FkoyR2+RlJ7Odh6r62Vyc3Rs+6lznnf01GMzv09JepAev5XarhnUNWxBb0msyTOEiJdEXQQmHsqytJiUEJvb/PT5g1ji2K8IJNmLCepIZw9dngBr5p1v+6BKn9vdP6+Dy9WLNbdwwJznoUiAnXWvcqRtO6qiUlVnos80F5Akn7oGfm+czm7C4dS8ZS2GplEUSvRe6gI6Nu5v4J8vnhnfeOJ6dCHO0Ye38M+vc57H46PL1dE/b3yrrbF/Bi0NPopymtnb9QKH297DYsoh13cRh4+bkUSf2gZ/bxrb62hsbIx3WCLBlBo8KMDvd9TGOxS5shcT02h2zrNYjBTm6zBqA1jK0yjI76HQVIFR04WSHiGs6rhoytWUl1zOmxWbzvt4pxq8dntzc7MscDJBmM2m/k6EvZbE6XUtEodZE2ZxroHtDTYOtju5oCgrbrFIshcTjqqqbDnWQbbZwLyCzHMqI13vJi/Nzqzr0sjNBjgQfSJPC/Shqm4CkWw67JncdMGlzC8rH63wBxi8dvuhnYcJjWIfBCHE0MZr9MA1pSYqHQF+v7OGRz79kVEvP1aS7MWEU+dw0dDt5voFk9FoRnY7PStTw2XLVKYUHAQgGNLQ2gm69Hz8YQPHDzeiagqZOv9jqOixOxvQaQxjUY0PYzpl7XZLTT7OcGp3bBJiPIzX6IGP5hspsJh4cvcx/vsTizHq4jOjoLTZiwmnojq6StVVsyeN6HXZpi5u/qc0phRDr9/KYftsNjzdx1/fUTjmnEaLq4Ta4yqObiMqo9Vz9sOOXA6HQzpyCZFATo4eyCouJj13bCbn0msUPrd0Ol2eAC8faBqTY8RCkr2YcCqORpP91XNiT/aTLG3Mya1FAd7cDofsc+nxZxGJjFGQJ0hHLiHEl5ZFVzD8/c6auMUgyV5MKKGIytu17UzPtTA91xrTa2aWhZmS2YQ/ZOCFv3k41qwwnr3pT3bkysy2YpGOXEKknNn5GVw+vYBNNe0cc7jiEoMkezGhHHJ46fUFY76FX5Af5NILw4QiWg475uDoHuNLeSGEOOHUlQDXTIlOmf3EzvgMw5MOemJC2dke7Uzz8dnFZ91Xg5+lF7qJqHC0aza+kCymIkSyCIfDJ1bWi3LYulF1GXGM6HSnrgQ42QRT0iKEPb1xiUWSvZhQdrX3oVEUrpxVdNZ9c4y7MepU3q/SEs6O7Za/EGJ0DF7fvrm5GRi9OSQaGxvZuNtBkSP6t32o3o+lxDdq5Y+WU1cCfETbzPySkXUsHi2S7MWE0enycsDu5aOT88kyDz8czqxtIl3XSFe3loN1GuYuHacghRDA6evb7zmyjYB+6qgeIz03v38ufmNm1qiWnWwk2YsJ4+WqJiIq3Lho8rD7aTVBcg0HUFUNe/ano5J4Z/ujZ+znaFcjkQFXaDBwJUAhzuTUq9rc/GxwxjeeVCbJXkwYz+9vAODGhVOG3a+ssB6txkd3YBGuvuQe5ubx+Aj5o0P7AFrb62hsLBrViUHcXQ7+ZPewj+g45FNXAhRj4+QJltvtxmq1yjTK4rxJshcTgr3Px+a6Di7INTE5+8zD12zeenIz7Lj9GdR35NPnOgTmxF20JhwO43R2o5ijPXX7XH0jjnc85mi35BYMWAlQDDT47sf5Jmd3l50NjXZyJpWSU++idk8lhkl5oxWuSEGS7MWE8HJVE+GIyqrJZ+5t6w96qOvZRliFvY15uP0t2JxtWHRjMzPWaGhsbKSxvY7QiRyd6PGKoZ1Mzpud0Z/U803OHo+PkN6ANs9IKA3cBFCCwdEKd0Ia6w5/yU6SvZgQXvggejv+irIzJ/td9X8loglwqNaELi2PzDQwmozjFeI5s1jS+6/MJ0K84nQnk/PJk7bRSM5ms4mMTAsWi0W+F4xPh79kJsleJLxmp5tNNW18pCyXEsvQvfCbug5T17kHfdjK0UYtpXPGOcikNvadAJPBqc0pkpxPX1VuNMbBS4e/cyfJXiS8326vJhxR+cols4DTJ6Twhzxsr30JjaIl0z8XVR29GaoikTD19fX9j1Oxo9R4dAIUyWfwqnKJOg4+VUiyFwnNHwrz+I4ass0GPnPRNA4f2H/aPu8f+xueQC8XTf447YdDo3r8ns52HqvrZXKzH0jd9ebHoxOgSD4nV5UDGQcfbzI3vkhoG/c3YOvz88WPziTNcPq5aXPXUWo7K8lJL2ZB6cfGJIaM/CLySqeQVzoFS450nhNCxObUpoyGY820NLXHrQlMruxFwlJVlV9tPYKiwFeXnz6m2xd0817tCyiKhstm34xGMxqTvAxsn3a5erEaZE79U0mvaBEPY9EHYKwNbspor3cws7AxLk1gkuxFwnr1YDO7Gh18uryMaYOWs1VVla3VG/EEelk85Vpy0kdnvunT2qdtjRQbRu+2/WiPx44H6RUt4mGi9gE4tSnD1euOWxyS7EVCCoTC3POXSrQahR+vuei05w+1bqO5+wiTsmayoPTyUT32qe3TaWmje1U/eDa6idoHYDR7RY/1CdBQ0/1OxJMsIX0Azocke5GQfrXtKLV2F1+/bA7zCjMHPNfSXc3u+tcw6S2smH0rijKxup6cOhudpSYfZzh+Z/vnYrRvp472hDSnlz/wBAsm7knWaJEToNQjyV4knGMOFw/84wOyzQZ+cPWiAc95I042H3kVRdFwxby1pBlk6drxNtq3U8diQprBBk/3OxFPskaTnACdbnBfFEiu/iiS7EVCcfuD3PDEZnp8QTbcegm56R9OTtLtbue4fwsh/Kyc81kKM6bGL9AUN9q3U2VCmvGmgim9f00GAPTmZMlr52RwXxRIrv4okuxFwghHInzpz9s50ObkX5bP5p8vntn/XGdvA28efIIQPj46/Z+Ylr8wjpGej4G9/c9l4RsxUgPfc5D3fXBHVJB1GWBgXxRIrln6JNmLhODyBfnsU1t47XALl07N52efXgpARI1wqGULexr+gaqqlOovZl7xpeMY2egm58E/svIDO/YksQ3t1LspIHdUkl1CJ/tf/OIXbN++nUAgwP3338+CBQviHZIYA+/WdfCNF3dR1e7k47Mn8efPXY5Bp6XNWceehjewuRpPdMa7hfZjrnGNbSyS81jespaOV0OTxCZSXcIm+x07dnDgwAGeffZZqquruf/++3n66afjHZYYJb5gmNePtPCHnbW8drgFgLsuncOPVs+m1VnJ1uq92FzRle6m5C7gkpnXYdKn007luMc6kdqTB/dsh9Hv3S5EMho8ygQmxsQ9sUrYZL9z505WrVoFwOzZs+ns7MTr9WI2j89sZl19Hir2foACKACKSklxCTqtFuXEXVwFBUUBjRLdR1Gi2+h/rJzYV/3wcfRh/76KAurg5/vL4pRjRf+tqtDpdXHc0cHJ3jQnL9oip/SuOfVK7uQ/1VOej5xyoRfdV+XUi79TrwNPltX/enXg85H+5wdu8wTCuAMh3IEwLl+I1l4fjd1e9rf1cKCtD28wAsCiSTq+/JEwBWmbebHyxZPvAKXZc1g0+SryrWWcSTgcpq6urv9xMvWePReDe7aDrIUuRCwGjzKBiTNxTywSNtnbbDbmzp3b/zgnJwe73U5Z2dA//CcTUiAQGJXjX/e7FznmHPz2NIxK2aPiH+3xjuCcKajMyQkxN8/Nokkuiq1+FAUCfj3p4WIMkSyM4VzUVgP7Wo8AR/pfe/z4cbq6uvoft7e3s796B9nZ0bH4hw7U4E8rw3XipNAQ8KN19+FqaJrQj72trahGc0z7m40aNL3O/vfIqiEh6nCuj42hALpRLC8R6jQWn3uyvSej/bnH9NhoRhv8cO56q9k8qscId3eh0Wjw+/2cTSz7nMx3sTTTKWqCNubdf//9LFu2jGuuuQaAW265hUceeYTS0tIh93e5XAPaKYUQQohUMHv2bKzW4eccSdgr+/z8/AFDZbq6usjLO3O7Y3p6OrNnz0av16MoqTukRgghRGpQVZVgMEh6+tmXnU7YZH/55Zfzs5/9jM9+9rMcPHiQsrIyTCbTGffXaDRnPbMRQgghkslwefFUCZvsy8vLmTt3Ltdffz1arZYf//jH8Q5JCCGEmJASts1eCCGEEKNjYi0XJoQQQogRk2QvhBBCJDlJ9nFWXV3NVVddxVNPPQVAR0cHX/ziF1m7di2f+9zn6OjoAKCiooJbb72V6667jueffx6ITijzgx/8gNtuu43bbruNpqboWM76+nrWrl3LjTfeyA9/+MOEnSp1cN1P2rJlC3PmzOl/nAp1D4VCfPvb3+aWW27hc5/7HD09PUBq1H3Xrl3cdttt3HHHHXzlK1/B6XQCyVn3Rx55hFtvvZUbbriB119/HYfDwZe+9CVuueUW1q1b1z9uOtnqPrjeqfQ7N7juJ43775wq4sbtdqtr165V7733XvXJJ59UVVVVv/Od76h/+9vfVFVV1aeeekp98MEHVZfLpa5atUrt7e1VPR6Pes0116h9fX3q888/r/7gBz9QVVVV33rrLfXb3/62qqqqevvtt6v79u1TVVVVv/GNb6jvvfdeHGo3vKHqrqqq6vP51LVr16qXXnqpqqpqytT9+eefV3/5y1+qqqqqzz77rFpRUZEydb/++uvV2tpaVVVV9bHHHlN/85vfJGXdd+3apX75y19WVVVVu7u71RUrVqj33HNP/9/7gw8+qG7cuDHp6j5UvVPld26ouqtqfH7n5Mo+jgwGA48//jgFBQX92+69916uvvpqALKzs+nr6+PAgQMsWLAAq9WK2Wxm8eLF7N69e8CUwitWrGDXrl0EAgEaGhpYtGgRAFdeeSVbt24d/8qdxVB1B/jNb37D7bffjsFgAEiZuldUVHDdddcBcOutt3LVVVelTN2zs7P7Z0Xs7e0lJycnKet+0UUX8fOf/xyAjIwMgsEgO3bs4MorrwRg1apVbN26NenqPlS9//M//zMlfueGqnskEonL75wk+zjS6XSnjZFMT09Hp9MRDof505/+xCc+8QlsNhs5OTn9++Tm5mK32wdsP/ma7u5usrKyTts30QxV9/r6empqarj22mv7t6VK3dvb29m8eTP//M//zDe/+U2cTmfK1P0//uM/WLduHWvWrGH37t1cf/31SVv3k5OfbNy4kZUrV+L1evvfj5NTgidb3Yeqt8ViSZnfucF1b2hoiMvvnCT7BBQOh7nnnnu4+OKLWbZsGXq9fsDzqqqiKMpp2wG0Wu2Q+04E69ev5zvf+c6AbalSd7/fT2FhIU888QQzZszgt7/9bcrU/Uc/+hGPPvoor732GkuXLuXpp59O6rq/+eabPPfcc3zve98bUJ8z1TFZ6n5qvSG1fudOrXu8fuck2Seg7373u5SUlLBu3Trg9KmD7XY7BQUFA7YHAgH0ej05OTn09vaetm+i6+jooK6ujm9+85vccsstdHZ2snbt2pSoO0BeXh5Lly4ForNH1tbWpkzdq6ur++u+fPlyqqqqkrbuW7Zs4bHHHmPDhg1kZGSQnp6O1+sFhq7jmbZPtLoPrjekzu/cqXX3er1x+52TZJ9gXn31VTQaDd/85jf7ty1cuJCjR4/icrlwu93s37+fpUuXcvnll7Np0yYANm/ezPLly9FoNMybN4+9e/cC0bbglStXxqUuI1FYWEhFRQXPPfcczz33HAUFBTz11FMpUXeItsdt2bIFgP379zNt2rSUqXt+fj7Hjh0D4PDhw0yZMiUp6+5yuXjwwQf53e9+R3Z2NhD93E/W52TcyVb3oeqdKr9zg+sez985mUEvjqqqqli/fj0tLS3odDoKCwtxOBwYjUYsFgsAM2bM4L777uP111/n17/+NRqNhi9/+ct88pOfJBwO873vfY+amhrMZjM//elPKSoqora2lu9+97uEw2Euvvji024ZJYKh6v7oo4/2t0VdeeWVvPXWWwApUfeHH36Yhx56iM7OTvR6Pf/zP/9Dbm5uStR93bp1PPzwwxgMBiwWCw899BBWqzXp6v7nP/+ZRx99lGnTpvVve/DBB/nOd76Dx+Nh2rRpPPjgg+h0uqSq+1D1bm1tJSMjI+l/54aq+/r16ykuLgbG93dOkr0QQgiR5OQ2vhBCCJHkJNkLIYQQSU6SvRBCCJHkJNkLIYQQSU6SvRBCCJHkJNkLIYQQSU6SvRBCCJHkdPEOQAiR2FRV5f777+fIkSOEw2EWLVrEvffeyyOPPMLbb79NaWkpJSUlWK1W7r77bt577z1+9atfodVq0Wg03HfffUydOjXe1RAipUmyF0IMq7e3l8mTJ/PDH/4QgGuvvZYjR47wyiuv8MYbbxAOh/nUpz7Fddddh8/n47/+67/485//TGZmJps3b+bBBx/kN7/5TZxrIURqk2QvhBhWWloabW1tfPazn0Wn02G322loaGDBggX9y7NeeumlQHSZ4s7OTr7+9a8DEIlECIVCcYtdCBElyV4IMay//OUvHD58mD/+8Y8YDAY+/elPE4lE0Gg+7PJzcolNRVEoLi7mySefjFe4QoghSAc9IcSwenp6KCsrw2AwsH//fpqbm+ns7OTQoUMEg0E8Hg/bt28HYOrUqXR3d1NTUwNAZWUlTz/9dDzDF0IgC+EIIc6io6ODO++8k7S0NBYtWkRGRgZ//etfufjii9m9ezfTp08nLS2NsrIyvva1r7F9+3YeeeQRjEYj4XCYBx54gJkzZ8a7GkKkNEn2QogRC4VCvPTSS1x33XXo9XruvPNOPv3pT7NmzZp4hyaEGIK02QshRkyn09Hc3MzNN9+M0Whk2rRpXHPNNfEOSwhxBnJlL4QQQiQ56aAnhBBCJDlJ9kIIIUSSk2QvhBBCJDlJ9kIIIUSSk2QvhBBCJDlJ9kIIIUSS+/8B9J7vQMJB4TUAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# I handle to age now\n",
    "sns.histplot(data=df,x=\"age\",hue=\"cardio\",bins=100,kde=True)\n",
    "plt.show()\n",
    "# We can say with histplot; Cardiovascular diseases increases with age"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Finding the categorical years\n",
    "\n",
    "df[\"year\"] = np.round(df.age/365,2)\n",
    "df.loc[(df.year <= 40), 'agecat'] = 1\n",
    "df.loc[(df.year > 40)&(df.year<=45), 'agecat'] = 2\n",
    "df.loc[(df.year > 40)&(df.year<=45), 'agecat'] = 3\n",
    "df.loc[(df.year > 45)&(df.year<=50), 'agecat'] = 4\n",
    "df.loc[(df.year > 50)&(df.year<=55), 'agecat'] = 5\n",
    "df.loc[(df.year > 55)&(df.year<=60), 'agecat'] = 6\n",
    "df.loc[(df.year > 60)&(df.year<=65), 'agecat'] = 7\n",
    "df.loc[(df.year > 65)&(df.year<=70), 'agecat'] = 8\n",
    "df.loc[(df.year >= 70), 'agecat'] = 9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAFYCAYAAABZHSXVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACWa0lEQVR4nOz9d5hd5X2vjd+r7F5mTx/NqAsJCSRRhDFgQBhwEfab2GAHErDjk5zz45dDgvPG5QQfN0KasIntcBInsU6IDTEEN4wTMJbBYMAgQAghoT6SRqPps6fsXlZ5/1h77ZnRFE3ZdfTc1+XLaO1VnlmzZ33Wt0umaZoIBAKBQCBY9MjlXoBAIBAIBILSIERfIBAIBIJzBCH6AoFAIBCcIwjRFwgEAoHgHEGIvkAgEAgE5whquRdQTAzDIB6P43A4kCSp3MsRCAQCgaDomKZJNpvF5/MhyxNt+0Ut+vF4nCNHjpR7GQKBQCAQlJx169YRCAQmbFvUou9wOADrB3c6nQU77/79+9m4cWPBzlfNiHsxEXE/JiLuxxjiXkxE3I8xCn0vMpkMR44cyWvgeBa16NsufafTicvlKui5C32+akbci4mI+zERcT/GEPdiIuJ+jFGMezFVWFsk8gkEAoFAcI4gRF8gEAgEgnMEIfoCgUAgEJwjCNEXCAQCgeAcQYi+QCAQCATnCEL0BQKBQCA4RxCiLxAIBAJBmfjpT3/KfffdB8DnP/95nnvuuaJeb1HX6QsEAoFAUCnouo6iKNN+fv/99xd9DcLSFwgEAoFgDjz11FN8+MMf5v3vfz+f+cxnSKfT/PjHP2bbtm28//3v5/bbb6erqwuAH//4x9x111184hOf4P7778cwDP7iL/6C6667jo997GMcOnQof95PfOIT/PSnPwVg165dfOQjH+EDH/gAH//4x9m/f39B1i4sfYFAIBAIZklvby/33XcfTzzxBE1NTfzJn/wJ3/3ud/n7v/97fvGLX9Da2so999zDP/7jP/JXf/VXAPzmN7/hxz/+MatWreKFF17g5Zdf5qmnnkKWZT7xiU9MukYikeDTn/403/72t7nkkkvYuXMnn/nMZ3j66acnDdCZK8LSFwgEAkFJOdD9Mqczr2OaZrmXMmdefPFFLr74Ypqbm5EkiQceeIA/+IM/YPfu3bS2tgJw2WWX0dnZmT9m5cqVrFq1CoDXX3+da665Bq/Xi9vtZtu2bZOusXfvXhoaGrjkkksAuPHGGxkYGOD06dMLXr+w9AUCgUBQMsKxLl4//p+YmAzFu6n3t5V7SXNiaGiImpqa/L9dLheapvGNb3yDl156CU3TiMVieZEHJuw/OjpKU1PTlJ/ZhMNhQqFQ/t+SJFFTU8PQ0BDLly9f0PqFpS8QCASCkmCYBr859hNMLAu/I/xOmVc0d+rq6hgeHs7/OxaL8bOf/YwXXniB733ve/z85z/n05/+9LTHB4NBotFo/t/hcHjSPvX19ROuYRgGIyMj1NfXL3j9QvQFAoFAUBIO9bxCOHaaFfUXIiFzqgpF/5prrmHPnj2cPn0a0zT5yle+QldXF0uXLqW2tpZwOMyTTz5JPB6f8vhLLrmEF198kVQqRTKZ5Oc///mkfS6++GKGhobYu3cvAE8//TRtbW0sXbp0wesX7n2BQCAQlISjva+jyA6uWPNRhoaHGEn0EEkOEvQ0lHtps6alpYWvfvWr3HnnnWSzWTZt2sTNN9/MCy+8wNatW1mzZg2f+9zn+KM/+iO2b9/O2rVrJxz/3ve+l+eff54PfvCD1NXVcd111/HLX/5ywj4ej4dvfetbfOUrXyGZTFJXV8ff/d3fTTkqd64I0RcIBAJB0TFMndHkALW+JXicfoJKG1Gjh1Phd9i4dGu5lzcnbrrpJm666aYJ237wgx9M+PfLL7+c/++bb745/9+KovCXf/mXE/Z9z3veA8DDDz+c33b55ZfzxBNPFGrJeYR7XyAQCARFJ5YaxjB1Ql4riS2otCIhcSp8oMwrO7cQoi8QCASCojOS6AegxmOJviq5aAquoD96ioyWKufSzimE6AsEAoGg6NiiH/I25rfV+pYAJrHUUJlWde4hRF8gEAgERWc0aYt+c36b31ULQFSIfskQoi8QCASCojOa6EeWFPzu2vy2gLsOEKJfSoToCwQCgaComKbJSLKfoKcBWRqbMmeLfiw9PN2hggIjRF8gEAgERSWRGUXTMxNc+wB+YemXnKLW6R85coT/+T//J5/61Ke444478ttffPFF/vt//+8cPnwYgJ07d7Jjxw7S6TR33HEHH/vYx9B1nXvvvZcjR44A8LWvfY1ly5Zx4sQJvvSlL5FMJtm4cSNf/epXC9KwQCAQCATFYSxzv3HCdqfqxqV6qyKRT9d12tvbC3rONWvWoCjKjPt861vf4pVXXiGTyXDvvfeyadOmBV2zaKKfSCS47777uPLKKydsT6fT/Mu//AuNjdYvPxaLsX37dn7yk5+gqiof/ehH2bZtGz//+c+RJInHHnuMX/3qVzz44IPcf//9fOlLX+Jzn/scF110EXfffTevvvrqpGsIBAKBoHIYzWfuN036zO+uYzjei2kaSFLlOp/b29u57wfPEmopzICgkd4uvvRxWLdu3bT7vPrqq+zbt4/HHnuMI0eOcO+99/Lv//7vC7pu0UTf6XTyne98h+985zsTtv/TP/0Tt99+O/fffz8A+/btY9OmTQQCAQAuvfRS3njjDXbt2sWHP/xhwOp1fO+995LJZOjo6OCiiy4C4Prrr+ell14Soi8QCAQVzEguc79mCtEPuOsIx06TyETxuSZPnKskQi1tNCxdUbLr7dq1ixtuuAGwXg76+/tJJpN4PJ55n7Nor1WqquJ2uydsO3HiBEePHuWDH/xgftvAwAB1dXX5f9fX1zM4ODhhu6qq6LrO8PDwhHGD9r4CgUAgqFxGE/1ISFP22M8n86VEMt+ZnKmPdXV1C9a8kvbe3759O1/84hcnbHM4HBP+bZomkiRN2g5Min3Y+56N/fv3z2O1M7N79+6Cn7NaEfdiIuJ+TETcjzHO1XsxlOpHldzs3fP2hO27d+9mSLPGzO47tJvT6uQxs5VCR0cH/f1hMvJkbZoPI/197N+v5cfsTvXdGBkZ4fjx4/nPYrEY77zzDv39/fO+bslEv6+vj/b2dv7sz/4MgP7+fu644w7+9E//dMI84cHBQa644goaGxvz2zOZDA6Hg7q6OiKRyIR9m5omu4vOZOPGjbhcroL9LLt372bLli0FO181I+7FRMT9mIi4H2Ocq/fCNE3e+c2PqfO1suXisZ/fvh/dI0G69u+mviXIJSsq9/4EAgF+NXyIhtbWgpzPaWTZuHE969atm/a7sWvXLmpqavKfpdNprrvuukle9DNJp9PTGrsly5pobm5m586dPP744zz++OM0NTXxyCOPsHnzZg4fPkw0GiUej7N3714uu+wyrr32Wp599lkAnn/+ea666ipkWWbDhg3s2bMHsLL+t26trulMAoFAcC6R1hIYpo7HGZjyc79L1OpPx3gdfOedd1i2bNlZBf9sFM3S379/P9u3b6erqwtVVXnmmWd48MEHJ8TkwUr4u/vuu7n99tuRZZm77roLt9vNjTfeyHPPPcfNN9+Mx+PhgQceAOCzn/0s99xzD7quc/nll5+Tb84CgUBQLSQzlvvaO63oh5CQqqJWf6S3q8DnWj/jPhs3bmT9+vV89KMfRVEU/uqv/mrB1y2a6G/cuHHCbOAzee655/L/vW3bNrZt2zbhc0VR2L59+6TjzjvvvElziwUCgUBQmdiiP52lL8sKPleo4mv116xZw5c+XsgzrmfNmjVn3etzn/tcIS9a2kQ+gUAgEJxbJDJWHpbXGZx2H7+7lt7R42hGFrVAiXKFRlGUGWvqq4XK7YQgEAgEgqonmbUt/elF3+cKWftmItPuIygMQvQFAoFAUDQSZ4npA3gc1mfJTKwkazqXEaIvEAgEgqJhW+/TxfQB3A4fAKmsEP1iI0RfIBAIBEUjkYkiIeF2+Kfdx34hSArRLzpC9AUCgUBQNJKZKG6HH3mGYTp5S1+494uOyN4XCAQCQVEwTZNkJkLwjJG6Z+LJeQEq2dIv12hdmH5M/XwQoi8QCASCopDV02hGdsYkPgB37vNKjum3t7ez89Uf0baspSDn6+rsBW45axngdGPq54sQfYFAIBAUhbHGPNOX6wG4HV5Aqvjs/bZlLaxYvbSk15xuTP18EaIvEAgEgqIw1phnZktflhRcqreiLf1yoaoqqlo4qRaJfAKBQCAoCrNpzGPjcfoqOqa/WBCiLxAIBIKikEifvUbfxu3wk9GS6IZW7GWd0wjRFwgEAkFRsC39s7n3YawrXyobL+qaznVETF8gEAgERSGfyOc4u3t/fFc+n6umqOuaL1bGfeHOdUHb2feb7Zj62SJEXyAQCARFIZFvwTt9Nz6bSu/KZ43BvaVg57ugjVmN1j3bmPq5IkRfIBAIBEUhmYnhUr0o8tmlptK78onRugKBQCAQzEAqG5+x5/54qqEr32JAiL5AIBAICo5pGmS0BC6Hd1b7V0NXvsWAEH2BQCAQFJyMlsLExKXOUvRz7v1K78pX7QjRFwgEAkHBSWsJwG6xe3Zs976w9IuLEH2BQCAQFBy73t6l+ma1v6o4URWniOkXGSH6AoFAICg4c7X0wWrQU6nZ+8VCNzQieg+maZbkekL0BQKBQFBwxiz9uYi+n1Q2jmkaxVpWxfFmxzN0ZF5iINpZkusJ0RcIBAJBwUlnLUt/ttn7YCXzmRiktWSxllVRJDMxDvW8ikPyUO9vLck1hegLBAKBoODY7n2XY3YxfQC3007mOzf677/T9Wt0I0ujun5WDYwKgejIJxAIBIKCk7f05+Det/e1XxgWM6lsjEM9r+B1BqmVV5XsusLSFwgEAkHBSWuWte6eg6XvUj0AZLKLX/SP9e1GM7JsXLoVWVJKdl0h+gKBQCAoOKlsApBw5oR8NtiWfuocsPTD8W4AltVtKOl1hegLBAKBoOCktQRO1Y0szV5mnLmkv8w5kMg3Eu9DlZ34XaGSXleIvkAgEAgKTjqbwD3Lxjw27nMkpm8YOqPJAULeJqQ5vBQVgqJe7ciRI9x444088sgjAPT19fEHf/AH3HHHHXzyk5+kr68PgJ07d3LrrbfykY98hB/+8IcA6LrOl7/8ZW677TZuu+02OjutGsYTJ05wxx13cMstt/CVr3ylZA0NBAKBQDA7TNMkpcXnVK4H4LRFf5HH9COpMIapE/I2l/zaRRP9RCLBfffdx5VXXpnf9s1vfpOPfexjPPLII3zgAx/g3/7t34jFYmzfvp0dO3bw6KOPsmPHDuLxOE888QSSJPHYY49x55138uCDDwLwpS99ic997nP86Ec/Ynh4mFdffbVYP4JAIBAI5kFWT2Oaxpwy9wFcDiv+v9jr9EcSlsFb62sp+bWLJvpOp5PvfOc7NDU15bd98Ytf5P3vfz8AtbW1xGIx9u3bx6ZNmwgEAng8Hi699FLeeOMNdu3axQ033ADANddcw2uvvUYmk6Gjo4OLLroIgOuvv56XXnqpWD+CQCAQCObBWI3+HEX/HHHvD8d7Acpi6RetTl9VVVR14ul9Piu+o+s63//+9/njP/5j+vv7qaury+9TX1/P4OAgAwMD+e2qqqLrOsPDw4RCoQn7vvLKK2ddy/79+wvwE01k9+7dBT9ntSLuxUTE/ZiIuB9jnCv3ImEMATA6FJvxZ57qMxmF4dHBRX2vOtKHAOhs76dXigKl+26UvDmPrut8/vOf5/LLL+eKK67g6aefnvC5aZpIkoTD4Zh0rKIoU+57NjZu3IjL5VrYwsexe/dutmzZUrDzVTPiXkxE3I+JiPth8eLxPr7601f4vasu5A/fvbbcyyk6p4cP0/4OLG9bxeZlU//+p/tuHH/9FwCL+ntzavfzODNu3r3lPUiSVPC/k3Q6Pa2xW/Ls/XvuuYe2tjbuvvtuABobGwmHw/nPBwcHaWpqmrA9k8ngcDioq6sjEolM2lcgEAgqEd0wuPV7v+a6f/gFz5+O8uf/+SaprF7uZRWd+fTdt3Gp3kVdsqcZWSLJMCFf86yM1kJTUtF/8sknkWWZP/uzP8tv27x5M4cPHyYajRKPx9m7dy+XXXYZ1157Lc8++ywAzz//PFdddRWyLLNhwwb27NkDWFn/W7duLeWPIBAIBLPm9c4wP9zbwSVtddywPMhQIsMT+0+Ve1lFJ53rnT/Xkj2wRD+rp9ENrdDLqggiiQFMDELe0ifxQRHd+/v372f79u10dXWhqirPPPMM4XAYl8vFJz7xCQDWrFnDV7/6Ve6++25uv/12ZFnmrrvuwu12c+ONN/Lcc89x88034/F4eOCBBwD47Gc/yz333IOu61x++eWL2gUkEAiqm2cOWV3X/vf7NmEOnubZUxH+ddcxbrukdL3Wy8F8E/lgrGwvoyXxOAMFXVclMJLoB6DWWx4vddFEf+PGjTz88MOz2nfbtm1s27ZtwjZFUdi+ffukfc877zx+8IMfFGSNAoFAUEx+fqgLVZa4YW0LRzMDXLu6iWeP9nI8HGV1/eITNJtUftjOPCx9h53Bv0hFP2mV65Ujcx9ERz6BQCAoCoOxFK93hnnPqiaCbicAf5BL4vvXXcfKubSisxBL3x66s1gb9MTTowD43bVlub4QfYFAICgCO4/0YJrwgfNbAUgYYW7aUIfHofBfB7rKvLriMjZWd/bDdmwWe61+Iif6HmewLNcvecmeQCAQnAs8c9iK51+zKsCzB75LZ/og8eMn2bRkKXu6hsloOk61dCNVS0lKi+NQXCjy3CVmzL2/OEU/nhnF7fChypPL0kuBsPQFAoGgwBiGyS8Od9MScNPR/zCdQweRkOgdPcEFzX6yusHB/tFyL7NopLOJecXzYXG7903TJJEexecMlW0NQvQFAoGgwBwLR+mLprhqhZ+0FmNt87tocWwGTJYGLTHb2z1c3kUWkYyWnJdrH8bc+4uxVj+jJ9GMLF5XeVz7IERfIBAICs6B3hEA2mpSACyvv4CgvBSAGpcVz397kYq+YehoRgbnfEU/595PLUL3fiJtNZfzuWrKtgYh+gKBQFBgDuVc93Vuqwd9U2AFTtlLg38ZXvkEAHu7Fqfo2xPy5iv6Y3X6i0/04+kRALxOIfoCgUCwaDjUb1l0XqWLkLc5b72uaLgQt0NjWcjB3u5hTNMs5zKLQka3Rd89r+PHYvqLz70fz1gvg8LSFwgEgkXEob5RnIpErSdBU3BlfvuK+o3W/4cyhBNpuiOLT9jsWPx8Y/qKrKIqTtJavJDLqgjscj1h6QsEAsEiwTRNDvVHWB5SUWRoHif6QU8DHmeAJX5rnOpiTOZbqHsfrGS+9CJM5LMb8whLXyAQCBYJXaMJouksrcEsAE3BFRM+r/E00uwbAWBv11Cpl1d0MpqVvLhw0V98Mf1EJhf2KVNjHhCiLxAIBAXFjufXe0bwOoP4XRPbrdZ4GlmWy+pfjJZ+pkCWvqZnFt2kvXh6FJfqRVWcZVuDEH2BQCAoIIf6LBduozdKY2DFpJnpQU8j9d4sPqfMwb7F16BnoTF9AJcjl8y3yFz8icxoWa18EKIvEAgEBcXutLckkCHkbZz0eY23EUmC1qDMiaHYosvgL1RMHxZXV76MliKrp8sazwch+gKBQFBQDvWNIgHN/jR+d92kz2s81otAk08nntEIx9MlXmFxGXPvz69kzzrWemGwy/8WA4lcuZ5XiL5AIBAsHg71R2gJyrhUk8AUou9zhZAllTqPJWgnhmKlXmJRKYR7Py/6i8i9n8/cL2O5HgjRFwgEgoIxkszQG02yrMYAIOCun7SPLMnUeBoIuqyyvcUq+k5lIe79xSf6+Rp9YekLBALB4qB90BLyJl8aWVLwOgNT7hf0NOYt/Y6hxdWEJqOlUGUnsjz/scGL0tKvgG58IERfIBAICsapEUvAg64YfncdkjT1I7bG20iDNwMsPks/rSUXlMQHY/kAiyl7P5mxXgg9DpG9LxAIBIuCzmFL9GtcCYJTxPNtanJlewAnhxeX6Ge0xILi+bA4Lf1U1vpuuB2+sq5DiL5AIBAUCNvSr/Nkp8zct6nxNOJxGNS4JU6GF4/om6ZBRk8v2NLPD91ZhKJv9yAoF0L0BQKBoECcyln69d7slJn7NkFPA2CV7Z0cjmEYi6NWP6OnAXNB5XowlgRot/RdDKS1BE7VgyzNP9ehEAjRFwgEggLRORLHoUDApc0o+k7Vjdvho96bIa0Z9MUWh0VbiBa8AI7cS8Nic++X27UPQvQFAoGgYJwaTtDkk5Clqcv1xuN11lDrsTrOnVgkLv5C1OiDVdboUNyLRvRN0yCdTeBWhegLBALBoiCt6fRGkzR4dQD87toZ9/e5aqjzWO7rk8OLo2yvUJY+WC8Oi6UjX0ZPYWLgcnjLvRQh+gKBQFAITo9YVnvIk8bt8ONQXDPu73UGabQz+BdJ2V4h+u7bOFX3oknks2cICPe+QCAQLBLG1+jPFM+38bpqaPDlavWFe38STtWDpmcwDH3B5yo3Y5n7QvQFAoFgUWBn7td5MrMSfZ+zJl+r37FIavVn496PpbPc+r1f8wfPnCCR0abdz7WIhu7ka/RFTF8gEAgWB53javR9rtBZ9/e6anAqJvVemY5FE9O3chSm67vfH01yw7d38sO9HewPJ/m7Fw5Me66xBj3VX7aXzlv6izymf+TIEW688UYeeeQRAMLhMH/4h3/I7/zO73D33XeTyViurZ07d3LrrbfykY98hB/+8IcA6LrOl7/8ZW677TZuu+02Ojs7AThx4gR33HEHt9xyC1/5ylcW3SxqgUBQnYyv0fdM03N/PF5nMLe/SddoYlE8y84W0//k91/mjc4wn7hsNXVuhe3P7adrNDHlvs5F1KAnpdkxfX+ZV1JE0U8kEtx3331ceeWV+W33338/t9xyC48//jhtbW08+eSTxGIxtm/fzo4dO3j00UfZsWMH8XicJ554AkmSeOyxx7jzzjt58MEHAfjSl77E5z73OX70ox8xPDzMq6++WqwfQSAQCGbNmHs/mxf0mbBHrNZ6dJJZndFUtqjrKwUzxfSHE2meO9bLu5bV89BtV/FHFzWRyOh88ak9U55rMbXiHWvBW35LXy3WiZ1OJ9/5znf4zne+k9/22muvce+99wJwww038Mgjj9DW1samTZsIBKw340svvZQ33niDXbt28eEPfxiAa665hnvvvZdMJkNHRwcXXXQRANdffz0vvfTShBcLgUAgKAedI3Fq3BIu1ZyV6DtUFw7FTY0rDXjoHk0Q8jiLv9ACous67e3t+X8PRwYB6Dh5GkWaKC9Pt4fRDZMrWzwcPXqUTcoo59d5efiN49x5QYg6j2PC/iMJa0BNR+dx4gMLW+eaNWtQlPJ1wsu79ysgpl800VdVFVWdePp4PI7bbXVaqqurY3BwkIGBAerqxpJe6uvrJ21XVRVd1xkeHiYUCk3Y95VXXinWjyAQCASzwjRNTo3Eact59Wfj3gfwuYL4XUnAQ9doggtaQkVbYzFob29n56s/om1ZCwBRJQySxKHuV5GQJuz75CEZkFnmPsmBrpOM6ANc2tjE4SGFJw+8zuUtE8MbGSkCKnQOH6Tf6Jn3Grs6e4FbWLdu3bzPsVAqZdgOFFH0p8LhGHuTM00TSZImbJtpOzDpTc3e92zs379/niuent27dxf8nNWKuBcTEfdjIufC/RhJayQyOjXONACH9h+bssf6mfdCS0sEnZb7+jdvH6QuNn9xKwcdHR043QpOT+5n1QxkZFyeidJimPDWANS6TdY1K8gStC1vYZMbHj0MRyIyV6+aeG7TUMEA2WHilOdvpTvdCvv37ycajc77HAslnOpHQuLtt/ZPq1ml+jspqej7fD6SySQej4fBwUGamppobGwkHA7n9xkcHOSKK66YsD2TyeBwOKirqyMSiUzYt6mp6azX3bhxIy7XzI0y5sLu3bvZsmVLwc5XzYh7MRFxPyZyrtyPPaeHgCM0Bgycqod3XXb5pH2muhepoycIdR8CwFnbxJYtm0qx3IIRCAQ40DXKkiVLABjqOooqu/L/tjnQl2UkHeb/ucBDW6uVy9DT08O1F7SgvNjHsYiTJUsmti2Op72M9nfi9blZEpp4vrmQSepc0LaxrJb+yTeew234ueyyy6b8vNB/J+l0elpjt6Qle9dccw3PPvssYGXsb926lc2bN3P48GGi0SjxeJy9e/dy2WWXce211+b3ff7557nqqquQZZkNGzawZ8+eCecQCASCctIbtax1vyMxq3i+jddZQ8hjJfB1R6o7Yc00TXRDQ5En25Ivn7Q8IO9ZMdH4cjskNjQ6ONifJZWd6N63z6Mb09fyVwupbLwiavShiJb+/v372b59O11dXaiqyjPPPMPXv/51PvvZz/LQQw+xatUqbrrpJlRV5e677+b2229HlmXuuusu3G43N954I8899xw333wzHo+HBx54AIDPfvaz3HPPPei6zuWXX35OWBECgaCysUXf50zhdTbO+jivK0jIbYnadKVr1YJhGpiYKPLk0Owrp9IoErx7+eRExYtbHezvy/JOX5YtS8c+XyyirxsaWT1VEfF8KKLob9y4kYcffnjS9qm2bdu2jW3btk3YpigK27dvn7Tveeedxw9+8IPCLVQgEAgWSH/UaiBT49ZmncQHVtme36njUKAnUt2irxuWx+JMS98wTY4MaKyuVwm4JjuXL2p18sieBG/1ZM4QfceE81Yr6VyNfiW04AXRkU8gEAgWjG3pB13a3Nz7rhokCRq8El2j1e3ety3yM0W/e1QnpZmsqZ/axrxoiSXue7szE7bLkowsyVVv6Y8N2yl/jT4I0RcIBIIF0zfO0vfOwdK3XxBqPQa90SS6YRRlfaVAN21Lf6J7/1jYEu3zphH9Oq/C8pDC3p4sujExri/LjqoX/VQF1eiDEH2BQCBYMH05Sz/g0vDMwdJ3qV4UWSXksQSvP1a9febzlv4ZTXnah6zt01n6AJtaHMQzJp2jEyfqqZKaf5moViqpRh+E6AsEAsGC6YulqHFLqDJzcu9LkoTHESSYq+/vrmIX/5h7/wxLf/Dsor+qzvqsY3iiVa/IKrqhVfVcgrRWOWN1QYi+QCAQLJjeSJLaXLv5uSTyWfv7Cbosse+u4mS+6RL52sMaXofEksD0DXZW1tqiP9HSt18gDFOfdEy1ICx9gUAgWESkNZ3hZIYatyVMcxV9t8NPjdtKYqvmZD7b0lfHiX5WN+kY0VhTr87YPdUW/ZNTWPrWuavXxZ8XfVUk8gkEAkHVY5frBV0ZXKoXdYo69ZnwOP2E3JaoVXPZ3pilP/bzdwxr6Mb0SXw2S2sUFHl60deqOJnPzt4X7n2BQCBYBOQb8zhSc7byATwO/6Jo0DNVyZ6duT9TPB9AVSTagsoUMf3qr9VPacK9LxAIBIuGvlzGvd+ZmlMSn43bESDkscSumlvxTpXI136Wcr3xrKxVGU2ZjCTHyhYXQ1e+dDaOQ3FN2Z64HAjRFwgEggXQGxlrzDMvS9/pw60a+J0S3VVt6VvW+PjpgmOW/tlDHivrJsf1xyz96hX9VDZeMTX6IERfIBAIFoRdoz/XFrw2bocfgHqvTE+VW/qKPDFh78SQRq1HptZ7dqlZWWu9LJwcGi/61Z3IZ5omqWyiYrrxgRB9gUAgWBB9+UQ+HU9OwOeCx2G9KIQ8JuFEGk2vzq58uqmhSGMWvWaY9ER0loWmL9Ubz1QZ/HnRN6vT0teMDIapVUw8H4o4cEcgEAgqGV3XaW9vX/B5jvUMAJalPzwY5Uj8yJT7dXR0EAhM9gRohtWYx6emMU0Hr759gCbf2OCZNWvWoCizE85yohvZCW7s/piObkJbcHZrX5EX/bGafPslolot/XwLXiH6AoFAUF7a29u57wfPEmppW9B53u6NIWEScGk8fWCASGJqq7S/P8yvhg9N8YnJJWslVCkJOPiH3xylwW0J5UhvF1/6OKxbt25Bayw2hmlgmMaEZLWuXEvdtprZiX6NW6bOI0/I4K/2RL50hdXogxB9gUBwDhNqaaNh6YoFnSN98hA+p4Ysgb9hOU6jbsr9MrKDhtbWKT8z2EvIa4mkGmqkoWHuuQHlZKrM/bzoz9LSB1hRaw3eyWgmTlWq+kS+VIXV6IOI6QsEAsGCiGc0gm4rDq+brnmdQzfdhNyp/Pmqjala8HZFLNFvDc7etlweUjHMsWNlSUZCqmL3fgyonBp9EKIvEAgE8yarG6Q1g6DLEinddM/rPLrpzrfijaerT+CmtPQjc3PvA7Tm9u3JHStJUn7oTjWS1oSlLxAIBIuGWNoSoxqXhmGqwPwS7gzTnX9xiC0WS39UxyFDo2/2MtOaG8rTHR2XzCc7qtjSt2P6QvQFAoGg6olnLDGqcWfmbeWDZekHc614q9O9P7kFb3dEZ0lQQZGnH7RzJkty8f/eyHjRr97xunbffeHeFwgEgkWAbZXXuNIY84zng5ULUOOyzmV7D6qJM937iYzBcNKYUxIfkB+/O9HSVzExMc3q619gx/RdojmPQCAQVD/xnECHPNkFW/oOxcStVqmlb+bc+5Jl6XfPI54P0OCTUeSxmD5U99CdlJZAQsKpesq9lDxC9AUCgWCe2JZ+0KXNO3MfrJg+gN9VpaKfs/TVnHv/dL5Gf25V4Yos0eJX6Blv6UvVO143nY3jVL3IUuVIrajTFwgEgnliW/pBt7ZA975lCQZdBuG4hmGayNLsY+Hl5kz3/li53twTG5cEFd44nSGtmbjG1+rPsxWvrhucOHFiXsculHgqgkP2cOTI1F0abXRdn/HzQiJEXyAQCOZJzE7kc2lkF+jeB6t/v4lMIqPhd519Ml2lcGb2ft69P0/RB+iL6SwPqQseutPb3U8kcoyks3tex88XExNNTWPqMge6Xp52v67OXpb4z+fyyy8vybpmJfrt7e2sWbNmwra33nqLiy++uBhrEggEgqoglrY68fmcOkPpBbj3cWKaEkFXFnAQqzrRn5i9b3fjWzrHmD6MS+aLnCn683fvL2lrYsXqpfM+fj5oeobR7iP4vX5WNMx8bWO4RIviLDH9SCTCqVOn+MIXvkBnZ2f+f0ePHuWee+4p1RoFAoGgIolnNPxOkCXQmb+lD9IZDXqqK36tGxqypCDlYtddoxpBl4TfNfdYth0SsJP5qjWRT8t7Pyrr5W1GS3/Pnj1897vf5eDBg/z+7/9+frssy1x99dVFX5xAIBBUMrF0lia/VT++kEQ+AMN0VW0rXt3I5i1y0zTpiRqsqJ1foyLb0reT+ap16I5mWC9wquw8y56lZUbR37p1K1u3buXRRx/ld3/3d0u1JoFAIKh4MppO1jCpcVviZCwgpg+g46LGZTVzqbaufJqh4VStnz+SNklpJs3+eYp+0PIOjFn61Sn6um5Z+qpSRZa+zfXXX893v/tdRkdHJ3RF+vSnP120hQkEguqlazRBMqvRF89imiZSFWWizxZbmAO5pjoLt/TdBN0RoLr675uYGKaWF+f+mCXW8xX9Jr+CIo239IV7v5DMSvT/6I/+iPXr19M6zVhIgUAgsPnB3g5u+96v8//e4W7gv11+XhlXVBzsznkhVxbDdDDfvvs2uuki6Kq+VrwmVqc8W/T7cmLdHJhfbboqSzT65bylr1appW+LflW5923cbjd//dd/veCLxeNxPv/5zxOJREin09x1111s3LiRz3/+80SjUVpaWvj617+O0+lk586d7Nixg3Q6zR133MHHPvYxdF3n3nvvzdc8fu1rX2PZsmULXpdAICgMhmFy3y/2osgSn7xsNf/+xnH+auc+PrFlNapSOQ1KCoEtzEF3esFWPuT679uteKtK9HMWuWRZtL0x6yVgvpY+WIN39nRnyermmOjPs06/XGh6LqZfYe79Wf0VbtmyhWPHji34Yj/5yU9YtWoVDz/8MH//93/PX//1X3P//fdzyy238Pjjj9PW1saTTz5JLBZj+/bt7Nixg0cffZQdO3YQj8d54oknkCSJxx57jDvvvJMHH3xwwWsSCASF42cHTvNO7yi/e8kqdtx6Ff/PmhAnhmL8x1sny720ghPLueBr3akFteC1MUwXLtXEVWWteKe39Ocv+i1BBTN3LkmSUCS16tz7et7Sr0LRf+GFF/it3/otrr76aq677jq2bt3KddddN+eL1dbWEg6HARgdHaWuro7XXnuN66+/HoAbbriBl156iX379rFp0yYCgQAej4dLL72UN954g127dnHDDTcAcM011/Daa6/NeQ0CgaA4mKbJ3z67D4D/df2FANyxoR5Fltj+3H4Mo/qmpM1EftiOO7ugbnw29otDwFldQ3dMaWLCXV8upt/kn79nx/YS9MXGXiiq1b2vVKN7/x/+4R+Q5YW75m666SZ++MMf8oEPfIDR0VG+/e1vc+edd+J2W1/2uro6BgcHGRgYoK6uLn9cfX39pO2qqqLrOrquoygzv1Hu379/wWs/k927dxf8nNWKuBcTOVfvx+6+OK+dCnPd0gDJrnZ2d0Gb38n7lwd5+uQof/+zF7hmaaDcy8zT0dFBf3+YzDwtsf7hKGD13Y/Gdbr7z97xrbt7+n387ihNy8CraoQTcLqri0h/H/v3a0Sj0Xmtsdh0dHQwqo6g1kE8lqQn0cPpMICEER2gJzHz8T09PVNud+nWOY50hWlVwNBBJzPt/jMxPDREKuua17ELIanFAYm+3r4ZE1kHBgaoV2tK9tyYlejv2rVr0jbTNPn4xz8+p4v99Kc/pbW1lYceeohDhw7xhS98AYdj7A/OzvIdv22m7cCssoI3btyIy7XwN3Gb3bt3s2XLloKdr5oR92Ii5/L9+Pefvg7A//7wu9mydglg3Y8v//ZVPP2tpzmQdvGnFXRvAoEAvxo+RMM8E5TNvpNAmhq3hkbdWROdu7u7Z9xHlQLAW4S8cCoCtQ3NuE2NjRvXs27dunmtsdgEAgH2dHeQJEZtqI463xKGMwPUe02WL22a8dienh6WLFky5WfnJVKwZ4SsGmTJEh+J/h5i6RQtLc35BkCzpbauC1/APe21isVI93EcOM/6vcgkdYxhCvrcSKfT0xq7sxL98W8gmqZx4MAB1q1bN2fR37NnD9deey0A69evp7+/H5fLRTKZxOPxMDg4SFNTE42NjfkwAMDg4CBXXHHFhO2ZTAaHw1EQD4RAIFg4v27vx6XKXLVy4sN+y9I6gm4Hv27vK9PKpkbXdUZGhpE8/nkdPxJPokjgdRicHtEJR8Mz7j86Ojqj8SFLWdq84HdkAJWugUHkkeGSDmOZD6ZkueBV2YFpmvTHdFbXL2ysi50PMBCfGDrQDA2HUlnu8unQjCwu1VvuZUxiVr+Zv/mbv5nwb13X+au/+qs5X2z58uXs37+fD3zgA/T19eHz+bj66qt59tln+fCHP8zOnTvZunUrmzdv5vDhw0SjUWRZZu/evXz1q18lHo/z7LPPsnXrVp5//nmuuuqqOa9BIBAUnuFEmre6h7h2dTNux8RwmyLLXL2qiacOdtE1mqCtpjIehKdOneJUbzuab37HxzISXoeBJMFwLMpApHPG/RNGnIFIeoY9TIw2Cb8zBXjpGe1F723n1KkWNmzYML9FloB89r6sMpI0SesLy9wHaPRZxpxd82/XuhuGBlUg+oapY5h6xdXowzyn7CmKMmNsajpuu+02/vzP/5w77riDbDbLV7/6VdasWcNnPvMZHnroIVatWsVNN92Eqqrcfffd3H777ciyzF133YXb7ebGG2/kueee4+abb8bj8fDAAw/MZ/kCgaDAvHi8H9OErWuap/x865pmnjrYxQvtffzepatKvLrp8ft91NTOPc/ANE3SeoJmr5Wc6PAEqJFnPo/ikPD7Z/YqZE2VOq+VACa7PHj883wjKRG6rhOLx3AEoa8rTMdoDFDxGkk6jsdnPHZgYIBMcmovhmGCIimcDqfoOH6apJwEBU6f7kLNjSGeLX29g6z0lrbHjFah3fhglqL/e7/3exNi50NDQ6xdu3bOF/P5fFOW2T388MOTtm3bto1t27ZN2KYoCtu3b5/zdQUCQXH59XHLdX/tDKIP8OsKE/35ktFBNyHgtFzbWaMwU8o13UGtx6rvTmRNCpeJVBxOnTrF6ZEUq4IyT76T5kQEQOX0cJrH3srMeGwiAd6u6V8MnHKAzgg89lac1Ut01i2F59sTDEaMOa3x+OEUN/sG53TMQqnUcj2Ypej/6Z/+af6/JUkiGAxy/vnnF2tNAoGgynihvQ+nInPFioYpP7+krY6Ay8ELFRbXny/JrGXh2810snphHu5ZQyXkzuavUVuQsxYXj88JaHgb2zANHcIZ6htChM4S11djsRk9H4HhJAMJg5olS3D4FSBCTWMNmr9+buvr7gdKW+NfqcN2YJZ1+pdffjlglb698847JBJnqcMQCATnDCPJDG91DXPFigY8jqkf9Koi855VjRwZiNATqf7nhy36NTmB1gpk6WcNBzVubcI1Kh2HamKYEiYy8Yy1Zr9z4bMWfE4J04SkNnZ/Vbk6avVt975Sge79WYn+N77xDb72ta8xODhIX18f9913H//8z/9c7LUJBFXDW11DPPCrdzg+OlOi1uLkpRP9GKY5rWvfxnbxLwZrP5ET5JA7i2YomLN7lJ4VzVDz3oOkVh2i73SYeVGOZSzXu68Aou/NnSOeMdAMKzFQkSq7ksFGq3b3/q5du3jsscfyTXCy2Sx33HEHd955Z1EXJxBUA7F0lo/866/oHLEs2G/tj/DzO2/A76q8P/hi8FqHFS99z6qZ67Kvzn2+q2OQ2y6p7ri+bYWHPJmCxfPBChO4VBOHbFrXKNypi4ZDNdFzomxb+j5HYSx9gETGRK8yS1+vdvc+MKHrncPhWJSjMgWC+XDvM2/TOZLgdy5ewZYmL690DPDQawufVVEtvN0zDMBFrTNHoDctsT7f3zNS7CUVHVv069xptALF88Fy7wP4nGZVuPdNTJzqmKUfz5h4HRKyXADRd8j5c2qmpT+qXCWWvl6ZY3Vhlu+RGzdu5M477+Tqq68G4OWXX2bjxo1FXZhAUA3s7R7iWy8eZHW9n3+97Sp+89pufuvJdr7160P8z/ecj3IONI/a3zNCk99Nc2DmUqqA28GqOj/7eodLtLLikXfvezSyRuHK6mzx9LsMukZlCjDHp6iY6MgyaKaCaZrEMyb13sJ8521LP54de6lQqsTSz7v3qzGm39nZyRe+8AV+67d+i87OTo4cOcJll13GF7/4xVKsTyCoaL789F50w+T/3PxuPA6VkFvlk+9azYmhGE/sn7lZy2IgmspyYijGpiWhWe2/cUmIgViavmiyuAsrMuOz9wvt3gcIOHVMIGtW9kujKVkirBsqKc2qr/cWIJ4PY+ex3Ps5S79KYvqVXLI34zfqlVde4Xd/93eJx+N86EMf4gtf+AK333473//+99m3b1+p1igQVCSJjMYvj/SwsSXEB9aPNf/49DVW97RvvnCwXEsrGft7RwBLzGfD5pyL/+3u6rb2k1kTVTZxq0bBMvdhrN4/aE2cIV3hom9gib5mKCTsJL4CxPNhnKWfMTCR0Q25amL6mpFBkdQ5zwkoBTOu6MEHH+Rf//VfCQTGOk2tX7+ef/7nf+Yb3/hG0RcnEFQyvzrWS0rTuWlD24Tt65tr2Lahjd+cHOCdnCguVvbl4vl2vP5s2C8H+6v8viSzJj4HSFLhavRhLKYfzJXtZSpd9CVb9FXiOe9HITL3AVRZwqWQP69mqChVFNOvxHI9OIvoy7I85XSntWvXks2WttmBQFBpPHWwC4BtZ4g+wM2blgPwq6O9JV1TqbGT8mYr+vZ++6o4mc80TZKaid9V2G58AIYpY5hSvkFP2lxYD/tiY0rWOnVTIZHL3PcWyNIHy8Vvn1c3laqw9E3TRDOyFenah7OI/kxNeGKxWMEXIxBUC6Zp8tTBLkIeJ1etbJz0+XvPs2rSf9W+uEV/X88wsiRxQXPNrPZf2xDApcp5D0E1ks7Frv35FryFfLhLZHUHIbfV7yFjVJGlb5frOQu3Zp9TJqNDVreS+aw6/cquajBMDTArslwPziL6q1ev5vHHH5+0/f/+3//L+vXri7YogaDSead3hFPDcd5//hJUZfKf0co6P8trffy6vQ/DqOyH1HwxTZN9PSOc1xDA65ydtasqMhc0hzjQO4qmz62HeqWQOKMFbyFL9sDyHNS6rTrvSo/pm+Nj+rn7UqhEPhjzGsQzJpqhIEmV36CnkrvxwVlK9u655x7uuusufvzjH7Nx40YMw+Ctt95CVVW+853vlGqNAkHF8fRBa8rkTRuWTvm5JElct6aZ771xnH29w1zUWlfK5ZWE7kiS4WSG965tmdNxG5eE2NM1xLHBKOtn6SGoJOxOeXa73EK69wE0w0Gtx/KyZircvW+My94vZGMeG/+4sr2xBj06ul65XYsquRsfnEX06+vreeyxx3jllVc4evQouq7zvve9j3e/+93I50D9sUAwHb843I0kwQfXTz+yc+uaFr73xnFeONa3KEXfzsDfPMt4vo29/77ekeoU/Uxx+u7bZA2VoLtKsvdzMX3NVIhnTVQZHAV8Txkr2zPyDXoUWQO9cucPjpXrVaZ7f1bf1iuvvJIrr7yy2GsRCKoCwzB543SY8xuDNPqn755ynR3XP9bL3dduKNXySsY705TrJTJR9nQ8g2ZkGcmMcn5qNX732ItBPoO/Z5iPX7SiRKstHLalX+vOFLTvvk1Wd+BWDVTZrPiY/ph7XyWR0fA5pYJ2ax0r2zOrZuiOptsteCvT0q/sb5RAUIEcC0eJpLJcunTmEZ8r6/ysrPPx4vH+RRnXP9wfAWB900Rrfe+pZzna9wYnBvYyrJ9kz6mdEz7fkNvfPr7aGOvGly5ouZ6NLW4+Z+Vn79vu/Ywmk9IKm7kPY6GCeBU16LHd+5Ua0xeiLxDMkd2dYQAuW3b2ud7XrWlhOJlZFK1nz+ToYARZklhdPzYTPZWNcaz/DXyuEB9/1z24pADHB94inh7N79NW48WtKhwbjJZj2QvG7sZX70kXPJ4PY9UAfqdBxpQxzMp9YTQkDdOEaMaSkkLV6Nt4c5UAiSpqxatVuHtfiL5AMEd2n7ZEf8tZLH2Ay5Zb++w5vQhFfyDKilofLnXMGj3U8yq6oXFh69X4XDU0qOdjmgYHu3+T30eWJdY2Bjg6GMGsYEGbDlv0a9waWkHL9SzsFwm/08BEIlbBg3dMKUtGk4hbHu28SBcKtwqyVF1DdzRDuPcFgkXF7s4wkgQXt509ge3iXALf2z1DxV5WSYmkMvRGk6xtDOa3aXqWQz2v4FTcrG15FwAhZTluh5/Dva+S0VL5fc9rCBJLa/RWYQ/+RNbEIYNLNYti6dslgIFcK96RTOWWNhpoZLNSPuRRyMx9sKpgvA6pqsbr6nrlDtsBIfoCwZwwDJM3u4bY0FSD33X2P+pNS0JIEuztWlyWvu2aX9sw1qK7I7yPVDbO+UuuwKFY2dWypHBB63vI6mmOD7yV33ddo3Xc0YHqc/EnsyY+pyVyxYjp51vx5voADFey6EtaztIvfI2+jc9pvVRkdEuuKr5O38gCErJUmWWFQvQFgjlwZCBCLK2dNYnPxu9ycF59gL3dw1Xpyp6OIwNWEt66cZZ+98gxAFY1XjRh31WNmwHoGWnPbzuvITjhPNWCYZqkxol+ocv1YMy9b5cEjqQrU/R1QwPJsCz9ItTo2/icEiYQSVeHpa8ZGVTZUdAqhkIiRF8gmANvnLaT+GZfd7+5tZbhZIbOkenbWlcbtqV/XuOYpd8fOYlTcVPrbZ6wr99Vh9cZpC9yIv/iY78sVFsyX1qzmsAGXMVowWthmNZEuVCuK1+lWvppLZH7f4l4Njdhr0iWPkA0bclVpcf0dT1bsa59EKIvEMyJN+eQxGdzcZv1grC3e/HE9c+09BPpCNHUEE3BlZPGiUqSRHNwFalsjEhqEIC1uZeFarP07dh1wGkJTzFi+iChjWvFW6kx/XTWysfIZOW8pe8pgqXvdVjfp2gaDFOqaEvfNA10U0Op0CQ+EKIvEMyJN08PIUsSF7XOvgvd5ty+e6t8hvx4jg1EcSgyy0M+APoiJwBorlk15f7NNSut/UZPAtDkdxN0Ozg2WF2ib2fu26NvC9133yZrOKj1WkN3Klb0tTgAmawV0/eooMjFs/QTWavHfyXH9Cu9XA+E6AsEs8Y0TfbnBsz4ZpHEZ3PxIhT9IwMR1tT788OG8qIfXDnl/s3BVRP2kySJtQ0Bjg1Gq6pxUTLnxg65rId7cSx967z2eN3hCo3ppzXL0k9nrES7Qpfr2YzvyqcbakVb+vawnUot1wMh+gLBrOmPpRhOZtgwx37xbTVe6rzORZPBH46nGU5m8sl4YFnwiuyg3t825TEhbxNO1UN/5GR+29rGIGnNoHMkXuwlFwzbvV/jKU7ffRtNd+BRDWSMyrX0s1ZMP5ZW0IzixPNhrCLArtVX5Modr5vvuy9i+gJB9XOgz+oqN1fRlySJi1vraA9HiaayxVhaSbHj8HZcPq0lGE700RhYhiJPLYKSJNMUWEE0NZTvzrc299JwtIqS+Wz3frH67ttkDRVJAo9SuaKfySXyjeSy6gvdgtfGrghIZA00Q0WWTGSpMu+J3ZhHqWD3fmUWEgoEFcjB3vmJPlhx/eeO9bKvZ5irVjUVemkl5eigLfqWaPdHOgBzkmtf13U6OjoIBKyXAyVjtevde/gVGtyrCeiWhf/SO8dYTumF//Tp03Muo7RFv86TKkqNvo1dFeBRdYbTBqZpVlwJmO3ej6SttRbL0ldkCbc6uf9+JY4drvSxuiBEXyCYNQf7LdG/YB6if0GLdcyh/kj1i75t6eca8wxGOwFoDE6cmNfe3s6Olw+wZjg3QMaTYv1yePHkYU4PZOhLWrHZnx7oJjxU+tDHgV0H0Wq8czpmTPTTpHRfMZYFjBu6o2oMpl2MprKEPJVlPdrufbt+vliiD1Z730jKIKuP679vVNb9gMrvxgdlEP2f/exnPPTQQ5imyac//Wk2bdrE5z//eaLRKC0tLXz961/H6XSyc+dOduzYQTqd5o477uBjH/sYuq5z7733cuTIEQC+9rWvsWzZslL/CIJzlIN9I0jS5Klys+H8Rnuy3OhZ9qx8jodjAKypt0R/ONEHQJ1vyaR9Aw3NNCy1XgYkssB+gjUmDa4V+LIanDhASnHn9ykl/qONjOhzyydIZE2cCjhVk2i2iJa+bg/d0SEOfdFk5Yl+zr0fy1oWd7Hc+2C5+IcSEM9WdoOeanDvlzSmH4/Heeihh3j00Uf5p3/6J375y19y//33c8stt/D444/T1tbGk08+SSwWY/v27ezYsYNHH32UHTt2EI/HeeKJJ5Akiccee4w777yTBx98sJTLF5zjHOgbZWWtH69z7u/K65ssV/ihKh0nO56TQzFUWWJpyLKSh+O9uFQvHkdgxuNMHGiGF4c0AoBbVXCpMiPJTLGXXDCsFrzWfxcriQ/GqgLsVrx90dRMu5cFW/TjuiX6viJl71vntl4oRlLWy1ClNuipBvd+SUX/pZdeYuvWrbhcLpqbm/nLv/xLXnvtNa6//noAbrjhBl566SX27dvHpk2bCAQCeDweLr30Ut544w127drFDTfcAMA111zDa6+9VsrlC85hwvE0fdHUvOL5AA1+N/Ve16Kx9FfU+lFkGU3PEE0NEfI2zyrmnDFCqHIKmTSSJFHrcTKcSFdFi2LDMElp4HMUr+++jT29rybXD6AvVoGin00gmQopo/ju/bzoJyvc0hclexPp6ekhmUzyx3/8x/ze7/0er7zyCvF4HLfbDUBdXR2Dg4MMDAxQVzfW5rS+vn7SdlVV0XUdXa/MNz7B4uJg3/zj+Tbrm4IcH4qR1qr3OxtPZ+mPpVhVbyXljST6AZNaX8usjs8a1v1zyCMA1HpcZA2TeKYyH+LjSWqW2PvzLXiLb+mHPDnRj1TeNMK0lkQ2HaQNGUUGZxHz6rx5S7+yRV83ssiSgixXXpKhTUlj+plMhtOnT/Otb32Lzs5OPvWpT6EoYzfHzlB1OCa+JU23HZiVdbF///6FL/4Mdu/eXfBzVivnwr34+TEr0cyTHDnrzzvd5/VKFt0wefLXu1hd4yr4GktB+4hlcQaMJLt372ZYOwlAZDDN7pGJP3dHRwcA3d3d+W2ZgElNCySjp+gd1XHkYqBHO7tp9pbWOhoZHiHjNIjFYrPafygJIOFVrTXHEvqsj7WZy/66IVGTa8X71tET7PZWVj+DZCaGlpJJGjIep0k8Pvf1zfZ+WBovEY5ZL1ymlpj1sZlUmpGRCD09PXNe31yxKhrkOV1rYGCAerWmZM/Rkop+Y2MjF198MYqisHLlSvx+P7Isk0wm8Xg8DA4O0tTURGNjI+FwOH/c4OAgV1xxxYTtmUwGh8OBLJ/dWbFx40ZcrsI9ZHfv3s2WLVsKdr5q5ly5F/9++nWgh23v3syWFY3T7jfT/bg64ubJ9jdRGpayZfPyIq20uHS/0wkc5/LzV7Fly0ZeP9HL6S64aMPlk0r2AoEA/3H4BVpbW/PbHLIHOERDrYnsa2WZHubtwS4Ub5DWObQ2LgS9tSFG9Dh+v39W+w9rGpAml8qA4gjM+liwBG4u+2cNBw1+y6KV/aGK+jvTjCz7fvMDHI4gWRQaXAp+v2dO55jL/aiXDehMksxl7Hvd8qyPdbpdhEIhliyZnGhaSEzTZKDrEG7Vx5KW2V8rk9Qxhino7zedTk9r7JbUvX/VVVfx6quvYpom4XCYeDzOe9/7Xp599lkAdu7cydatW9m8eTOHDx8mGo0Sj8fZu3cvl112Gddee21+3+eff56rrrqqlMsXnMMcWECNvs35uWOPDFRvXP9ELnN/ZZ31wB2O9wJQ652tez+IaUo4ZOsehLzWQ3y4CpL57HK9mlxyXTHd+2DF9Rv8Voy40hL5MrlhO5GkC5Dy7vdiYTfoiaWt/69E975hGpimgapUbuY+lNjSb25u5n3vex+f/OQnicfjfPGLX2TTpk185jOf4aGHHmLVqlXcdNNNqKrK3Xffze23344sy9x111243W5uvPFGnnvuOW6++WY8Hg8PPPBAKZcvOIc5PBChNegh6J7/H/RiyOA/MWSJ/upcud5Iog+fqwan6p7lGRSyZgCnPAqY1ObK0IYTlS/6dgteuye+VoSxuuPJGiohl4EqQX+ssmL6dub+SMr6/fkcxbUfXSooEsQyEqZZmaKv50JVlZzEB2Wo07/11lu59dZbJ2x7+OGHJ+23bds2tm3bNmGboihs3769qOsTCHRdp729Pf/vtGbQORznsiWBfI+I6Rjfge5MNMPEIUvsPdV31vPMhjVr1kzIiSkFdo3+qjo/6WyCRCZCW+35czpH1qjBqUZQpAQ1bi+SBMPJdDGWW1DyLXg91lqLbelndQeSBCGXVHGWfl70k8XP3Acrd8vrlEjk+u9Xoujb5XqVPFYXREc+gWAS7e3t3PeDZwm1WMNjhtI6JhBLpnjwxUMzHtvfH+ZXw9PvE3BIHBqM8/e/PrigtqojvV186eOwbt26eZ9jPpwYihJ0O6jzOvMT82br2rfJGiGgE4c8im76qHE7q6JWf3zf/ayhAMUVunzZnhO6oqmKasVrd+MbTlprLLZ7H6zmP/0xg4yu4qhE0c934xPufYGg6gi1tOW7xIX7R4EYrY31NCyduYVuRnbQMC5x7UyaBzsY6h/F3dRGYA7jeSsB0zQ5MRRjXUMQSZIYjlud+Gp9zXM6TyZXtueUR0jprdR6nJwYipHVDRxK5c4AS4zru68VsUbfJl+25zJpj+hEUllqKqQrn913fzida8xTxG58Nj6nhAkMJ120BqJYk/Yq4yUIqse9X7l/YQJBhTCUizfXeRdeAVLvs84xGK98d/aZ9MdSJDI6K3M1+pHkAAA1nrnNEsjX6ktWboMd1690az+ZNXGp4FK1/ECcYmI3/wm5rTK1ngqq1bfd+0NJS3SL7d6HsTa/Q0knsmSiVNikvWpx7wvRFwjOwlDCEug678KtrPrcOexzVhP5JL46K2chkrLKZwOe+jmdRzN9mKaEKluT9aolgz+ZNfE6JCSp+PF8GGvzG3JZzZx6oxUo+ikJMIvad9/GbvM7lLS+L6pcWWOqtXkO2zkZgcc6MiUbuy1EXyA4C+ECWvq1uXMMVUG2+pmMT+IDiCbDuFQfLnVu9dmgoJk+HDnRr4YMft0wSevgc1ou/tK49ye24q0kS98u2RtKg1MykOUSWPr5Vrx2//3Kiutreff+7IwDwzR55M04/+slhR+dztIxPLdGT/NFiL5AcBaGE2kCLkdB4s31edGvPkv/ZM7SX1XvxzB1oukhgnO08m00I4AipZHIUJu39Cv3nthJfH5n8Vvw2uSH7rjtWv3KEf20Fsc0IZwycUulaStt5w1Uav99fY7Ddh5+M843X4riU+HPN7jYuKQ0zamE6AsEM6DpBqOpbEFc+wBepzVZbqgKY/rHw5ZlvqrOTzw9gmkaBNzzE/2saYUIHHK0Kiz9yaJfuph+0G19VyrJ0k9rSRJZmYwBLrk0sXXb0h/NJQ9WmujPJaafzBp8b3ecgEvigWt1ttSVLqdeiL5AMAN2nLkQrn2w6o3rPC6Gk5mqmCw3HtvSX1nnJ5K04vnztfSzRk70pSgeh4pbVSo6kc/O3A/m4uvFnLBnYyKT1caJfiVZ+tkEsYwV5im1pR/JiX6lle1pegZVdsyqrPJH+5KMpkxuu8hL7Wz7WhUIIfoCwQwUMonPps7nRDNMIunKSkQ6G8fDMVqDHtwOhaidxDdPS1/Lib46Lq5fyS9C+Ra87tK04LVJZ2SCbqsxT29FWfoJohlrCIG7RLPtVUXCqUC0QlvxWqJ/9udEWjN5ZE8cr0Pi1ot8JVjZRIToCwQzUMgkPptaT/Ul82V1g86RRL797oItfdNqSZxP5vNaL0KxCh2xm2/B67F+Z6Vw7wOksxKKmiHkcVZY9n6SaMZK4HSVyNIHqzSwEvvvG6aBbmqzaszz1KEkg3GDj232EvKUXoKF6AsEMzBcBEu/Gsv2Tg3HMUwzP2hnoZa+bnoxTRlVypXtVXhc3xb9+tyo21K498Gy9JFMWgKuionpa3oW3cgymuu77y5RTB/A65BJ65DWpIoSfV2ffWOenUctz82tm71FXdN0CNEXCGagKJZ+FZbtjQ3asRvzhHEqblzqfB9cElkzkLP0zXEZ/JV5TxKZXDc+bxrDlNDN0sw8SGctq7bJ72QokSGtlc6qno60FgdgJGUJXKktffvalST6WWN2LXijaYPdXRkuaHbQHCjt3AwbIfoCwQwUslzPphrL9vKZ+/V+DNMgmgoT8DQsqBe8ZgSQpSwyqXEZ/JV5TxJZE0WCGlc6Z+WXpv1rOmNdp95nff8qYfBOKmuJ/lDSEq1SxfRhLIM/nHBWVCKfZlv6ZxH933Sk0Q24dlXhjIi5IkRfIJiGQpfr2XidCk5FrlhX9lScGNeYJ5GOYJg6wXm69m3yGfxyrOJb8SYyJh6HhFPNliyeD5DKWI/oeq8ldj2RRMmuPR226IfjEi4ZVEqXfGlb+kMJV0VZ+rNtzPPr49ZL7dbVQvQFgopjqMDlejaSJFHvdTGUSFdstvqZjLn3A0RTg8Dc2++eiWbaGfwRatxOJGnsnlcSpmmSyJr4nCBLZsky92HMvV/rtr4nvRVh6VvfhYG4Qb1boZSD/+yyvXDSWVmiPwtLX9NNXj6ZZklA5rz68s26E6IvEExDMZL4bOxs9WiVlO2dGIrhUmWWBDz5nvsFs/SlKLIsEXI7GalA70dSs+a5+XMteEuVxAe5RD6gxlM5rXhT2TiaIRFO6DS4Sishftu9n3SgyAYSlTF0R8t345v+WfFmd4ZYxmTrandZRyQL0RcIpqEYSXw29jnDFShyU3EiHGNlrR9Zlogm5zdo50zGLH3Lcgx5nMQyGlm9Mh7kNsmMtZ6AKyf6JXTv25Z+0GV9TyqhVj+VjTOSsizVendpJcQeulNp/ffHLP3pvxsvnbCMiGvKGM8HIfoCwbQU09KvprK9SCpDOJEeG6lbIEtfN90YpoIqWaJfqRn88TJ047OxLX2/M9egpwJq9VPZWL7/fb2rtBnoLhUUCUZSldWKdzYx/T3dGRwyXNxa+OfJXChfYEEgqHCKaelXU9ne2Ejdsel6quLE7fDPeJyu68SiMcLh8LT7NDk8ONUI4fAgLtN6gJ/qG0RJz3Vy39yJRWPgObubNTGpG18pRT83r96RACqjVj+VjTOcK9erd8mUZjachSRJ+JwSo6mxVrzlvyOWpS9JMrI09UtQPGNweEBjU4sDl1o+1z4I0RcIpmWoCOV6NtVUtjd+pK5pmkRSYWpmUa536tQpwtFegpG6afdpC0FzUGckfhJ0FZDpHh3EXwIf5MBID3618az72TX6ody0u1Ja+oYpIZkKKjFcqqciJu2lsvG8e73eLdNR4uv7nBI9UQnNqJwGPZqRRZWd0/5N7OvJYphwSZmtfBCiLxBMiWaYRFJZVtQWpzd2NZXtjY3UDZDMRNGNLAF3w6yO9frd1NQGpv1cl/xAjKZ6BwnFA70psoqTmtrixz1d7tldw7b0a0vcgtdGNp2k9TgtgeUVY+nbffcbSuzeB7tsT2IkqeJQyp8Ia5ommpHBrU7/rNjTbX13Lm4t7XdnKkRMXyCYgkgueasYrn2orrK98ZZ+JFeuN9+e+2eS1q3761JTBHKZ4JFUZd0PW/Tr8qJfWltJNh2ksnFagh56o0kMo7z3J5WNE0lZv7dSJ/LBWDLfcNKBQy6/6BumjmkaM5brvdWdRQIuqgBLX4i+QDAFo1lb9Iv3R1otZXvjW/DmM/cXmMRnk9KsuaJuNY1LlXApVqvSSiKRMZElqHFbLXi1kou+E9M0aPFb35fBePlq9XVDI6unGE45kCSodZZeQuyyvaGUWhFd+fJJfNOIflY32d+b4bwGNf9iW07KvwKBoAIZSRfX0h9/7kov2zsRjlLrcVLjcY5l7hfI0k9p1j1wq5aQBdwy0bRZUd6PRNbE65BwKlpO8EubiKWY1j1q9Fuu9K7R8rn401mrI+BQUqbZ70GVS5+UZnflG044UCvAva/pM9foH+zPktbh4iXlt/JBiL5AMCWjmeJb+va5K7XfPIBhmJwciucH7Sx0ut6ZpHUXhinhVqx7EHRJ6OaYS73c2N34vA4Jh5wtaRKfjWxa35PGXMj49Gi85GuwSWZjmCYMxqGtpvgVFlORb8VbIe59u0bfMY2ln4/nt5U/ng9C9AWCKRnNWDXZ57ql3xtNktJ0VtVbyXiRZBhFduB1Tp+cNzckMroTl23p23H9dGWIfkoD0wSvExTZKHkSH4CSE/06j/WdLK+lHyeeVcjo0FpTntGwdkw/nHRWRCLfWI3+1N+NfT3WGi8Slr5AULmMZoyilevZVIOlf2a5XjQVJuCuQ5IKd19SmgunoiFLOkGXZcVFU5UR10/kcjsCZWjBayPn3PshjyUeXWW09FPZOEMJ6x4sDxWnsuVsuFWQJTuRrwJi+mfpu3+wP0u9V6a5FHWos6AyViEQVBApzSCmmUV17QP4nCpORa7oBj0n8uV6flLZOFk9veBOfGeST+ZT0hVn6ds1+gF3rhtfGS39Gpdl4ZfT0k9lYwzlavSXF6mc9WzYDXqGkyqqrJe9//5M3fgG4zp9MYMLmh1l7bc/HiH6AsEZdEYsV3MxXftQHWV7J8JRwLL08/F8z+xq9GfL+LK9oDtn6VdIBn/ijBa8mRJn7sNYTN/nsCz80yNltPS1OOGcpb+sTJY+WHH9aFpBMyi7i38mS/9An7W2C5oqI54PZWrOk0ql+NCHPsRdd93F1q1b+fznP080GqWlpYWvf/3rOJ1Odu7cyY4dO0in09xxxx187GMfQ9d17r33Xo4cOQLA1772NZYtW1aOH0GwiOkYtUW/+DG4Wq+TnmiSaDpL0F0ZMb/xjC/XiySPAgvvuX8m48v2fE4JSaocSz+esRvz5Frw6qX/HcmmEwkJw4xS562ju4wNelLZ+ARLP9JfnnX4nBImEqMpK5kvo0//gm4aBgN9Q3QcP12UtcSUGMjQ1dGPdEZlx28Oy4BMgzlKx/GRKY/v6uylyTNzS+tCUhbR//a3v00oFALg/vvv55ZbbuGmm25i+/btPPnkk3zwgx9k+/bt/OQnP0FVVT760Y+ybds2fv7znyNJEo899hi/+tWvePDBB7n//vvL8SMIFjGnIvagneJ3hRufzFepoi9JVvz2na7CTNc7k/Fle7IkEXBKRCtE9GM50W/wWt+JTBli+hISbqefZCbK0ppl+RexcpDKjln6y2t97C/TOqxkPp3h5Nlr9ZMjw/xqwKTrreJ4SK7emMGpyvzHW4lJn7004AVk9nXGOdI99Xe690SYj60rVGLs2Sm56Le3t9Pe3s51110HwGuvvca9994LwA033MAjjzxCW1sbmzZtIhCwbsSll17KG2+8wa5du/jwhz8MwDXXXJM/TiAoJKW09Mcn862qK93b/mw5EY6xLOTDqSoFm653JrZ73y7bC7gkuiIGWd3EoZQ3Dmpb+vW5bnyZMlj6AF5HkJFkP601Ht7uGSaSKs9Lop3I51BklgQ8ZRN9u0FPOOmgwXd29767tp5Qa2tR1uJ29ZDWXZPOb5omI70J/E6J5mVLpj0+GiltuKbkMf3777+fP//zP8//Ox6P43Zb7r26ujoGBwcZGBigrm5sSEd9ff2k7aqqous6uq6X9gcQLHpOjZYmpj/+GpVYtpfWdLoiifzLSDQZRpYUvK6agl7HMBUyumNS2V4lWPuxjIFTgYDLeiEpRyIfgMcZQDeytAat70u5kvlS2ThDKSdLa7zIZWjMY5MX/YSzrLX6sqSjyvqUYZ9YxiSlQWOFZO3blNTSf+KJJ7jssstYunRpfpvDMfZHZJomkiRN2DbTdmBWGZH79xf+fXT37t0FP2e1stjuxdHBKB7ZZKCvd17Hd3d3z3rfjGYlrHUPjdI9jftvKkb6+9i/XyMajc55fbOlI5LGNCFoptm9ezfDyT4ckpc9b+6Z1fFHj1o5ALHY2d3R8bSDkCdBPBbFLVkDVfpHEziLmM+XTWcwHekZ1xdPg9cBqpQmq8tEFjjlbjb3YjzJVIKjR4+yLGg9+8yE5W15/o23SLSU3jM0Gh9lJNnIqoDB7t27OXr0KMlUYs4/l818j5N1AIlwwoGpz3z92fye54vHYb2sJ9LSpPN3Rqw11qjajNdOphKAr2TP0ZKK/vPPP8/p06fZuXMnvb29OJ1OXC4XyWQSj8fD4OAgTU1NNDY2TpjBPTg4yBVXXDFheyaTweFwIMtnf4vauHEjLlfhrLbdu3ezZcuWgp2vmlls9yKZ1Qh//wCtXpXWebgDu7u753ScaZo4j46QNOQ5Hec0smzcuJ5169bNeY2zZeBQF9DOZetWsHHzeezb9QOWhNaw5cLZ/b4HBwehawC//+zilDW9SFKC+pBKQ8YB/Wmykgu/v3iWtcPlxOFyTbu+jG6SNRIE3Qpuh0ZGn37f2RCLxeZ8vMftZe3atTS1Sgx3nuCClY2wP4KncSlbtqyZ91rmg2HoPLfzCUwkLljWwpYtWxgcHMQTjs/rvsznftg43SaQIJxw4HVnZjzP2X7PCyHgjABgSN5J548OZYAsbXUe/P7ppxF63FaTo0I+R9Pp9LTGbklF/5vf/Gb+vx988EHa2tp45513ePbZZ/nwhz/Mzp072bp1K5s3b+bw4cNEo1FkWWbv3r189atfJR6P8+yzz7J161aef/55rrrqqlIuX3AOYDejqSnRIBFJkqjzOhmMp/MerUohX6Nf5y94z/0zGR/XD7gsV2mkzA16Yrnwgs8JqqwTy5Sv7MrjDAJjXfm6I5OTxopNSkuMy9wvTzc+G6cq4VQgnChvK15nrlxwqrDPYML6XTX4zmH3/lTceeedfOYzn+Ghhx5i1apV3HTTTaiqyt13383tt9+OLMvcdddduN1ubrzxRp577jluvvlmPB4PDzzwQLmXL1hkHB2w3txDztLNCa/zuuiNpiqubO9E2C7XCxBNdgEQcBe2Rt/GLttzqWmCuZyBcsf047n5CzVu6//LUa5n48m1Pa5xW7kFp0dKL/rJTKQiavRt/C6ZcMKJWkbRt184zqzqME2TcNwg6JJwljkZ9UzKJvp/8id/kv/vhx9+eNLn27ZtY9u2bRO2KYrC9u3bi742wblL+6AVIw/O0dIPqIfxqp24GlVkxSCpLz37QTnsZL6hCivbG9+Ct3dkECiepT++bM+ZG7EbKXODHjtzP+S2G/OUz9K3Zx0E8135Si/6iUx0QrleuQk4JYYSMhm9fN8Tp2LF9M98IYxlTNI6tNVUlpUPoiOfQDCBoznRD81B9Gsc+6hzvYlbGaAl1EOT+0V86olZH2+X7Q1VWA/+Y4MR/C6V5oCbaLKw0/XOZHwrXoBgBYzYjZ3RmKdc5XoAHofl3lelGF6nUhbRT2YiY+79irD0LQt6NCUB5fme2O79M18IB+PWi0i9t/IktvJWJBCUkWODlnt/tpZ+QD1EyLmfrOGjK/Fh9ndehGGq1DnfQJVml1lfiWV7pmlybDDKefUBJEkikgojSTJ+d6go19NNFc1QxpXtSRjmmLVdDuxr13ntGv1yxvStJLFkNkpb0FsmSz8yNmynAix9f+5vdCjhQC3T4B1H3tKf+N0IJyzRr7R4PgjRFwgmcGwwSrPPgWMWNcgyGULOfeimi77UDWhmgGgqRDj9LmRJo8H1G5jFMJBKnLbXE0mSzOqsabDcytFUGL+rFlkqXq5DSnPhVtOASbACavXtmH69J1ejX0ZLX5FVXKqXZCZKW42X/liKtFbaHiWJTJRw0kGtx4HfVf5e8gGX3aDHWbZpe045S1ZXMc+Q0jFLv3S5QbNFiL5AkCOR0egcSbCixj2r/QOOI8iSRiS7Ad0cs3wS+kpi2gpcyhBe5ez9vitx2p4d5ljbGCSjpUhl40WL59ukNTeyZOJUMvkHejnj+rGMiUsFv7O83fhsPM4AiUyUZTkru9TJfIm0ZekvK3Pmvo3t3h8sYwa/Q8lMmesRThj4nBIeR2Ul8YEQfYEgz9Gca39Vjees+0poBByH0U0H0ex5kz4fzWwEIOA4evZz5cr2Kmnanh3mWFMfGJuuV6R4vk1qXNle0F1eS980TeIZE79TxqlkMU3IlmHC3ni8ziBZPcWykPX9PFniHvx9sRhpXWF5qDLaRQdy7v1wwlGWSXtWNz5jkgcokTFIZM2KjOeDEH2BIM/h/pzoh85u6fvVYyhShmj2fEwmv+lrZpCk3oxb6UeVRs96vjqvC80wiabLOybU5tiAbekHiCSL03P/TMaX7ZXb0k/roBnWNDeHkiFrOCa5cEuNL1fKuCRg3ZuTw6UV/VPDlmdhVX3phsPMhEsFh2yWrVbfOU25Xj6eL0RfIKhsjgzYon82S98k4DiGYSpEs9N3xItl1wIQcBw767XHl+1VAsfCluif1zDO0vcUp0bfZnzZnj1iN5oqj6Vvx/P9TgmnnC1rEp+NzxUCoMlnxfI7hko3qMUwDU6PWHHzNfWVYelLkkTAZfXft0vnSomdxHdm2GcwJ/r1FZjEB0L0BYI8s7X0HdIIDjlKUm/FYPr2zgm9Dc3w4FdPIDGzJWIn84UrJJmvfTCKz6nSEvAQSVo1+gF33VmOWhj5sr1xI3bLZenbmft+p4kyhQu3HNiibycWltLST2Xj9MetF59KsfTBatCTyCoYZhks/Wm68YXjwtIXCKqCIwMRXKpMq3/mOQ0+tROAhLb8LGeUiWmrkaUsHqVnxj3rfbmyvXj5RT9frtdgl+sNIiEVXfSzhgPdkMeN2JVJaVYP/FJj1+iHPNYDvJyNeWxs977fGUeWpJJa+slMhIG49eKzpoJE35frnBkrw8vhdJb+QNzArVqhoUpEiL5AgCV0h/sjnNcQQJmxXM/Eq57CMBWS+tkH5CT0ZQB41Zmz+Btyoj9YAaLfG00Sz2j5cr1IchC/uw5FLnYim0RKc+NWU4BJ0G39HsqRzGdfs85jx20rx9LP6BGWhrwlTeRLZKIM5Cz9lXXlr9G3sUfsRsvwZzNVTD+lmcQyJg0+paLmaIxHiL5AgCV00XSWdY0zz4p3yLZrvw1zFl2ss0YIzfDhUbqZqWbf41DxOdWKEP2jdhJfQ4C0lsyV6xU3nm+T0twosoFDzuZr9csxeCeau2ajb+rmK+XA57S+m7H0CKvq/HRFEmRKVKufzEQYSDhp9qt4HGUf2ZInkPuOjKRkSt2VL9+Cd5wXqNJd+yBEXyAAxuL55zcFZ9zPp5wCIKEtm+WZJRJ6G7KUxS33z7hng8/FSDJDtoy9xMFqUASwpiFANGn33C+N6CdzcX2Pmspn8JfL0lckqPNYD3Z7CmA5URUnLtVHPD3CilofpgmdJarVH02OMpRwsKLu7OWspaQm5w0aiJe+K59dJjj+hXAgXpmT9cZTuSsTCErI4Vzm/rrGmUXfo56etWvfJqm15Y+dCdvFX+5kvvZ85n6Q0Zzo15TQ0gcrmS+YF/3SvwRF0gYBt5TrEAjpCnDvA/hdIeLpUVbkGvSUysV/PBzBRGJ1XeXE84G8N6g/XvoMfpeSIaNPLOUcrOD2uzaVuzKBoIQcmYWlr0hxnHKElN48K9e+TcpoQjcdeJUuZnJBNvgswSt3Mp89Xvi8hkA+c7+U7n2wRD+Qa9AzWuKyvbRmktEtQXHlhCStld/SB/C6atCNLG011ktIqTL47ZeLtY2hklxvtjhVCa/DoC/mzP+uSoPVOfLMl8HBuIFLGcs1qESE6AsEjFn6589g6dsZ+Cl9yRzPLpPUW1HlBA55eNq96iskme9wf4SAy0FrcKxcL+huLMm1x7v3nYrVxnS0xDF9u0ww4JJwqelJ1lw58eeS+Zr91hpLZemfHLa+k2sba0tyvblQ6zEJJ5woUulE3yFnkSWTzLiXwbRmEk2bNPjkik3iAyH6AgFgWbeNfhe13uktOrfSC0BSb5nz+VOaFQ7wKH3T7tOYF/3UnM9fKDTd4MhAhPVNQatcLzmIIqv4XDOHPQqFbqpkdTWXwQ8ht0QsY5ItYdmenUMQdElTWnPlxC7ba8hN/jtZorK9rlErfm1XdFQSQZeEbkoks6WL6bvyYZ+x58WYa7/yhuyMR4i+4JwnldU5MRTj/Bkz9w08Si+a4UMz5/7gSxnNALjl3mn3Cbqt6X7ltPRPDMXI6Abrm2swTZNIapCguwFJKt2jIqW5cw9Vk1qP7eIvnbVvVwvUeoxJ1ly5scv2Aq4kiizRUQJL3zANeixHGKvrKqMb33gCLktkIyUMA7mUybkeg1WQxAdC9AUCDg+MohsmF7RML/ouOYwsZUnqS4C5u+5000PGCOJSBoCpy6wkSaLe5yIcL9/gnYN91pyA9U1BktkYWT1d9Ol6Z5LSXciSiUtJE8qJ/kiy9JZ+vdeybishc9/GFv10dpSlNV5ODhff0re68TnxOKDRP7sJlKXE77Lya0ZK6CCbKtdjsArK9UCIvkDAO72W0G1sCU27z0Jc+zYpvRlZ0nHJ4Wn3afC50QyTkWR5Bu8c6rdFv4ZIcgCAoKc08XybZNYqC/OoKUK5ZL7hElr6drVAk6+yMvdhTPTj6VFW1vnpjiRIF7lWP54aZiDhoC1YmbHqYM7SH06Wbm2uKao6BuMGToV8qWmlIkRfcM7zTu8IABfMIPoepQfTlEjpzfO+Tir3wuCeIa7fUOa4vm3pb2iuKXnmvs34DP6Qx3qAjiRL6N5Pm3gcEn5nZWXuA3icASRk4ukRVtb5Mc3iJ/N1DA2Q1hSW11aelQ/kxzAPJkoXS89b+jkvUEYziVRBEh8I0RcI8qI/naUvkcUpD5E26jGZv9WX1pswTWmWol+euP6h/lEcisya+tKX69mMF32vQ8KhlE70DdMkljYJ5jL3obIsfVmS8bqCxDMjrGu0ckvs6ZDF4p1e63uwtgKT+ABcqoTXoTMQd6JIpelQ6FLSZA0Fw7ReNPJJfN7KTuIDIfoCAe/0jtDod00br3Qr/UiSSVpvWtB1DJxkjFpcchiJqTONmwLWGvpjpbf0TdPkYF+EtQ0BVEVmNOfeL1VjHpuU7sI0weNIIkkStW6Z0bSJYRQ/rh/LmJjkyvWUyRnalYDPFSKRjrDWFv3+4or+wVzIZ+OS0n4P5kKdx8h15SvFy3KuRn+qeH6FJ/EBc+gwIhAsQuLpLCeGYly3Zsxtr+s6IyPDSB4rU7kt1AFu6B/2EE1PH48HGB0dxeWaXiRcNUFaaoZIx9uJpiY/RE3Tav/aPRIjHJ7+WiMjw+h6Ya2a7og1f2B9s5XQOJLow+3w4XaUNmPbMBXSuguPmgQg5JHpjxtE0mbe3V8soim7XM9qzJPV1bw1Vyn4XSH6OcmyGuteHC6ypX9sMAEobF4y/3yWYlPrMTkdkUlr2fnk2c4JVdZQZIPM+HK9KsncByH6gnOcg/0RTBMuaA7lt506dYpTve1ouWFiaxr7MAyJk4MxDHPmXucJI85AZHprwzB0WmpAlU4yEElOuU/QKRFOZugb7WS6gX/dve2cOtXChg0bZlzPXMjH85tq0PQM0dQwLTWrCnb+uZDMeqj1jKDK2bzQDyeNfDZ/sYie0Zgnka2sXvMwFm5p9KaQJano7v3jQxqgsKm1ci19uwd/LK1R7AKDKcv1EgYOhXzr6EpGiL7gnMaO51+4JDRhu9/vo6Y2gCJpBN0popkAgdDME/gAFIeE3z+9ZWxIXgzzNI01KfoyU8dIG4fTDA9qSB4vNdOIXMRf+PGmh+3M/eaanGvfJOSdf+LiQkhoHmoZwaMmCbm9AIyUIIPfbvlb59GRJbPiXPsAAbdVQpnWhlhZ58sPiyoGpmlyelSm1mNQ46mc3IYzydfqp40SiP7EBM+MbjKaMlkSqPwkPhAxfcE5Tl70x1n64wm6okgSRNKFSWIyTIVYxo/PkUCRpo7r1+XqfMOJ0rafHV+jP5Kwkg3LJfp22Z7XkRxXq1/8+zGcu0aL3/rdVFLmvo1t6UeSg6xrDNIfSzGSLE4L2oHYCOGEg5W1lS0VAZc16W4oUfy8jzO78YWrKJ4PQvQF5zh50Z+mMU/QZVlRo+nCtaEdTQeRpLFzn0ldTuSGSlimBnCgbxRJsuYPDOdFf2HJi/Mlodm1+kkCLglFguESNOgZSRl4HBI1nsrL3LexmyVFkuH8gCjbS1No9nZ1AbCmvvJefsbjczqRMBmIFz//4kz3frW037URoi84pznQN0pr0DNtz/2gK4JhSsQyhUtmi+ReIGqmE/2cpT9cQkvfNE32dg+ztiGIz+VgJF5eSz+lufMZ/LIkUeuVGU4aRc3gz+rWwJSQu3Iz9wFcqheX6iWSCrM2NyCqWMl8+3utCo7zGyuzXM9GVWQavFl6YmrRu1k6z3Dv55P4KrwTn011rFIgKAIjyQynhuNcOE19vipn8TmSRNOBgk5Zi2V86IY8reh7HNZ0uVK6908NxxlJZrio1ZqiNpLox+3w43YUPndgNtgZ/N5cBn+9V8YwYaSI/dVH8z33ZTwO67qpCkzkAwh46omlhljXYL2MFqts71Au5HPhktK2Yp4PTX6deEYloxd38I5LTaMZMnquqqM/ZiXx2cmElY4QfcE5y56uIQAuXVo35edBZxQorGsfwEQmkgngcaRwyFPHYus81nS5TImmy73VbY38vbitlqyeIZYeKptr3yaR9eBQNFQ5S30+z6F4zVfs8EHII+NWUximRKoCLX2AoLsew9RZmivbK1YG/9EBq1rlorbKLdezafBZv79Yupi1+iYeNZVrICWRylqd+Jp8SlUk8UEZRP/v/u7vuPXWW7n55pt5+umnCYfD/OEf/iG/8zu/w913300mYz0Ed+7cya233spHPvIRfvjDHwJW/fSXv/xlbrvtNm677TY6OztLvXzBIuLNTqsO/pKlU1sxdsw9Mk2W/UKoNBf/3twL0EWtdYwm+oHyufZtkrm4vldNjhP94t0PO1Gw1iPlHuwuil70PU/sZD6fI4rfpRZN9I8PaXgcOmvqy/tdmA11udLO0VTxLH2XkkGWTFK572ZfzHoJbfZXj/1c0pW+/vrrHDx4kP/4j//gX//1X/mbv/kb7r//fm655RYef/xx2traePLJJ4nFYmzfvp0dO3bw6KOPsmPHDuLxOE888QSSJPHYY49x55138uCDD5Zy+YJFxps5odsyjaVf44qgGzLxTOFd3Lb34GzJfKVy8Y+39MuduW+TH7zjSJakosEe6tPo1VFlPf9gr0SCbkv0o6kh1jUGOToQLXi+Q1Y36I5AW1DDoVamx2M8NW6rAr2YCZ/uXLgpqdmdM63vTHNAiP6UXHLJJXzzm98EIBgMks1mefXVV7n++usBuOGGG3jppZfYt28fmzZtIhAI4PF4uPTSS3njjTfYtWsXN9xwAwDXXHMNr732WimXL1hkvHl6iJDHyaopZoS7nQYeh1WfX8h4vk0i60UzlGlFv943sad3sdnbPUST301LwMNIztKvLbPoj8/gdyoSQZdEOG4ULVFrJGngUqAul7lvP9grkUAugz+assr2UppOx3BhB+8cHRhFNyVWhCrT23EmXqcTRTIZLOK0YY9qtce2X0htS7+pSjL3ocTNeVRVRVWtS/7gBz9g69atPPfcc7jd1h9XXV0dg4ODDAwMUFc3Zn3V19dP2q6qKrquo+s6ijLzDd+/f3/Bf5bdu3cX/JzVSjXei1hW58hAhMuavbz55psTPjt69Cghv/XkGIi6iMXm9jCd7f5DcS9NgShaeohUdmJpmMMERYK+SJZYbPKY3WQqwdGjR2loWHiXtGhG5+RQnHe3+HjzzTc5mT4MwMmjvZyWhuZ93qNHjwKzvx9nkpAMzEZwyVFisRg1LuhMS/SPxPE55r2sPNl0BtORJhaLoRsQSUG9B2TDSl4biUnzXvt0zPV80/2eddMKg3b2HqfOWA/Aj198g+uWFS7/5D9PWC9/y/3GtH/jR48eJZlKzPs+FfL+Spg0+TP0x51EozHsEPv43/NCUX1Wnk84YhBNx+iPQY0LMqk48+2UkEwlAF/JnqNl6cj3y1/+kscff5yHHnqIF198Mb/dNE0kScLhmPgXPd12YFbJExs3bpyxH/pc2b17N1u2bCnY+aqZar0Xv27vAw6zdcPKSesfHBykf8R64KXMhhk77J1JLBab9f5xvQ6IsqRWZyAx+ZgGX5L+mIHL48WhTPyee9xe1q5dW5B7/0LuXly7YSWXXnoJR197Gp8jxLsvu3JB5x0cHISugTndvzNJah6C7jR+v4/mYJbOSJYUbpr9C390OVxOHC4Xfr+foYSBSZIGv0rIb3lXDCW0oLWfyVy+GzYz/Z7bX92J5Mjyocs38o97nyPirmXLlosLtFr4TvsvAbjqvJZpv2eDg4N4wvF53af53I+z0eLvpSfqRnJ68LssD9343/NCCXo0TBNkZx2ZrIluplgSVPH7568vnlzHyUI+R9Pp9LTGbskDES+++CL/+I//yI4dOwgGg/h8PpJJK04yODhIU1MTjY2NE4aNTLU9k8ngcDiQ5eqJpQgqhzdPW9+j6TL3m2o1NEMhni1eydroWZL5mvwyJsV38Y8l8dWSyERIZWM0+NuKes3ZEs96UWQDt5oqajKf3Ykv5JHzLtxUBbv3wS7bG+aiVquxlF2NUij2dA0jYfKu5a0FPW8xafJZv8eRIiXzudUkad2FiUyfHc+voiQ+KLHoR6NR/vZv/5Z/+Zd/obbWqge+5pprePbZZwErY3/r1q1s3ryZw4cPE41Gicfj7N27l8suu4xrr702v+/zzz/PVVddVcrlCxYRb+bL9SZn7mtSEr/XyLXeLV48M6W5yeiOXFx/cpy6MRcnHIgVV/THkvjqCMdOA1DvX1rUa84WO4nS74iXRvTdEm41SUZ3oJuVPZqkxt2AYer4HClagx7e6hou2LkNw+RAX4qWQJq20JKCnbfY1Ofe0UeTk0NiC0WRNJyKNi6JLxfP91dPPB9K7N5/6qmnGB0d5f/9f//f/La//du/5c///M956KGHWLVqFTfddBOqqnL33Xdz++23I8syd911F263mxtvvJHnnnuOm2++GY/HwwMPPFDK5ZPK6tz7zF7SI2GyDQNcvqwBeboxaIKK5s3TQwRcDs6rn1yOl1asF4LR9NkH7CwMiUg6SIM3jFtNTcoWb8xZEANxHShAEHsadneG8ToV1jYE2HfaartaXyGWfiznafE543gd9XgcEgPxwou+PQ+90SfhUjJFKdMsNLW+JTCwh6F4Dxe31fHUwS76o0maAguvOjgWjpLIwkUtKYLexgKstjQ05vrfF8M7ZjdsSmY9mKZJb9TApVZPUx6bkor+rbfeyq233jpp+8MPPzxp27Zt29i2bduEbYqisH379qKt72ycHo3zjV8fJKsbfGvPz7lyRSM7br0yP39cUB3E0lkO9Y9yzaqmKV/aMoplMY2miv97Hc2Jfo1rdJLoB5wSbnWsLKgYjCQzvNM3wtbVzaiKzGCsskQ/kfVimuBzJJAkiWa/zMlhnVjayMdsF4ppmgzEdfxOiTpPGkmq3E5846nzWRb4ULybS9qaeepgF3u6hvnA+oWvfXfnIADrGhVUuXgvnIXGqTgJurL0RgtvfY8P+0TTJrGMyara6mnKY1NdwYgyc15DkGNf+Aj3XdXGRzYt45WOAS79u//ke2+0l3tpgjnwyskBTBOuWDHZgjFMg7QyTDwpl6Qbmx3XD7knD0yRJIlGn0IsY5LMFqdM7dUO615ctaoR0zQJx7rwuUJla797JoapkNQ8+BxxwKQ550rtLeCLUDRjktIsz4rbMbEOu5KptUU/Zln6AG8VKK6/61QvAJuXVL7HYzxJzcuq2iSRtEwiU9iX5Xy5nuahO2K59luD1eXaByH6c2ZpyMcHVtbwo09dxw8/tRWfU+UPHvsNj755otxLE8ySl3KlSFevntxmNhzrwpQ0+oZUStGNLaO7SGTdBJ1RJCY/pCa6+AvPb05YA1WuWtlUcUl8NuOT+eykqb5o4e6HnTPR5FPyvf6rQfQ9Tj9eZ5CheA+XtFk5UoVK5nvj1AASJluWVY9rH6wBSStDljj3FzgMNL4xjxD9c5SPblrOM3feSNDl4PcffZmfvSPaAlcDLx3vR5LgPasmi373iFVb3jdUOpfmaLoGRTYIuKKTPmvKxSj7iuTi/81J6wXoihUNFZfEZ2Mn8/kccRp8stW/oID3w36havTJ+JxWf4ZiVm0UklrfEhKZUVoCEiGPsyCWvmGY7OuJ0uzPVFUSn4XEkoDlFRuIFfZF2eNIkjUUsrpCd0TH65CqLp4PQvQXzKVL63nq/3cDLlXm97//MsfDkx/cgsoho+m82jHIppZaQp7Js9J7Ro6BCf1DpUt3GcnlDoRck138zX4FCeiJFN7S13SDXacGubClhlqvi3CFxfNtbAH2O+MoskSDT2YoYRRsGNFA3EACGnwSPkectOZEM6ojjm3H9UcSvVzSVsvRwSiR1HzbxFgcH4oSy5isCCXL3pVxPthz7cOJwpXtKZKGR02TyPgYTkJSg9agXHXxfBCiXxCuWNHI/7n53Yymsvzewy+S0Yo3CUywMHafHiKl6VO69jNaiv5IBw4jQEYr3Z9GJB3EMCVqpojrO1WJRp9Mf9wgW+CJe3u7h0lkdK5cablwB/OWfmXVZcfzyXyWFd4SUDApTIKjiZW5H/JI+JxZnIqWrxioBsaS+Xry5aevnwrPdMhZ+XW75f1ZU5fKD/apJkw8tPjT9MfMgrVstj1Asayvql37IES/YPz+u9bwyctW83pnmC89/Va5lyOYhpeO5+L5U7r2j2CYOi69tLPDTWQi6QA+R3LKUbutQQXThN4CxrHBSmgEK55vmDr9kQ6CngbcjsJ2SVsohqkQz3rxO+NIGAWN6ydxoRlWTwT7paIYA5aKRZ3PekEbivdw7RrLKn++vXdB53zuaA8A71rmQJEru1fBVMSzVjJfWpcYSRVG9P2570YsU/2iX32/0TKj6zodHR0EApOzWj+9uY7nj3Txdy8c4Io6mQsby/PwWLNmzVnnEZyr/Pq4NUHumiks/VPhgwC49QZgstVdTEZSIULuCCH3KAOJiclTrUGFt3qydEV0loUK9yf7ci6ef9XKRoZi3WT1NKtqLirY+QtJJB3E70wQcMZo9lt/e70FiNnGJSthr9Ev4x9nzVULAU89iuxgKN7DdeubkCWJ54/1zft8pmny3LEegq4sG1tK+/JbKBJZL6tqe3mlM5RvoLNQ7O9GJOWjO6IRdEkEClQyWmqE6M+R9vZ2drx8gDXDU9+6i0IqpyJp/uipg9yyyodc4pjPSG8XX/o4rFu3rqTXrQYMw+Q3JwdYXe+ntcY78TNTp2v4MF5nDWq89JbuSCoEnKLOPTxJ9JsDVvJad6RwyWuGYfL8sT6aA27OawjwTtdbALTUrC7YNQpJJBOglV6CrgiRTJBaj0Rf1EDTTVRl/n9jUcn6HjT55Kq09GVJptbXwlCsG79L5rJldbx2apBYOovfNfe8hEP9Efqiad69NE6df0MRVlx8TGSW1ljx/J6ITiGKDn2OOBndwckRhayhsb62eo0qIfrzINDQTMPSFVN+1gCczJ5iX88IJww/715RfTGxxcqbXUOMJDP89sZlkz7rj3SQ1hKc3/JuIsOlT85J6W4SWQ817lFkSccwxx4qqizR5JfpiRqksiZux8LXt/t0mP5Yit9/1xokSaJ39DgALcHKFP1oOoBpQtAVhSgsD6nszXk/VtTO7zFmAiOSH7cK9V4JvzNOUnNVfPvdM6n3tTIY7bSs/TUtvHYqzMsnBvjA+rnnZvzqqBUaWN8Ypym4ssArLR21HpWAU6MrAucv8FwOOYNLzTCUDHFy2PIczPc7VwlUp3+iwnn/+a14HAq/OtbL6AIzaQWF478OWIlqN22YnJ3eOWS59pfVXVDSNY1nKFmLLJmE3COTPrPjhz0Fius/ddDK1L9pQxuGqdMXOUHQ3YDXVbjRrIVEN9VcXD+GhMHykHU/To3M/36kZA9ZyUFbjYJbzaDKelVZ+TbNNasA6B09znvXtgDwq2Pzi+s/lzvugsYkjYHJL8fVQlLzcWFzjEQWkrL37AfMgO3aj2Z8dIzouNXqG7IznupdeQXjc6q8b90SMrrB0we7C5ZBKlgY/3XgNA5F5v3nT6w9Nk2TzvBBVNlJS6h8lu5QymqwUueePDilLSf6naOFE31VlnjfuiUMxXrI6umKde3bRNJBZMkk4IzR5JdxKdA5os/77yviCAGwrEaZkJ1dbdi/t97R47xnZSMOReb5eYi+bhg8f6yXBm+G85sbcSjF70hZLBJZLxubYgBEHAtrp22HfdrDfpJZk+UhteRh20IiRL9IXNRay4paH0cGIhzqn3p0qqB0dI8m2H16iGtXNxF0T6zPH4p3E0kN0la7rqx9xhNZLynNRcg9Mqk7X6NfxuuQODmkYRgLe4nsjSR5ozPMNaubqPE4865922KsVOwhOEFXBFmSWBZSiGdNhuY5XCWihgBoq1EJOC2BiGcqq3JhNnidQYKeRvoiJ/A4ZC5fVs/u00OMJufmZXzz9BDDyUzOtT91+LJaiGb8bGjKWei53/N88ee+GwcGrJegFVUczwch+kVDkiQ+dEEbiiTx80PdpEXtfln5r5w7+8MXTO42d3zgLQBWN11SyiVNgcRQshZVNqhxTXxRlCWJVXUKaR26F+jif/qQ7dq37kXvqDU7otIt/Qlxfay4PkDHPFz8Wd0kpgbwmkm8DomQaxTdkIlVoXsfYEnNajQ9w2CsixvWLcEwzXwIZ7Y8/lYHABe1RGmu4ng+WGWesuRmeU2SmBpAn2dLbQkDvzNOIuvi+JCBIsPSKi3VsxGiX0QafG7es6qRaDo77xiboDDY8fwPnSH6hmlwfOAtnIqbpbULTflZOENJa3BKg3dw0mer6iyROz60MNG3xWDbhjayWprukXZC3mZ8rsqeFqmbKrGMn4AziipnWVpjdSucT1y/O6JjSjIhM4ZLSeNxpBhNBzGr9JHYUrMGsFz8t168EoB/n8M8EN0weHTPCfxOk03Nsaq39MFqb72xOYYpyUSk+b3M+Z0xVFlnX189oymTFSFlQdUilUB1fsOriKtXNVHvdfL6qTDdo4lyL+ecJJnVePZoDxuaa1jTMLGAp3f0OMlMlJWNmyuiEUks6yOZdVPnGUaRJrYRbRnv4p+nh38kmeHpg12c1xBgfVOQ08OHMEyNFfUbC7D64hNO1SJJVt6DS5Voq1EYiBtzdvHb3oEaI05Nrv3xaLqyX3pmoiWfzNfO+uYaLltWzy8Od9MXTc7q+F8d66MnkuTypRHqfLV4nZWZ0DkXIukgG5st1/ywNL/CvdpcUu1vTlnfjfWN1dGeeSaE6BcZVZG56YKlmMB/HehacDxWMHee2NdJIqNPWap3vH8PAGsay+3at5HoTzQiSyYN3ontVKVxLv7wPMf+PrrnBMmszn+73CrVOzm4H4AVDdUh+sM5T0idx0p2XN9ovagdGsjO+hwZ3eR4WMNppAmQyFdL2DMQqhGPM0DI20R/5CSGoXPHllXohsl/7Dk5q+O/n/MKXN42VNWleuOJZXysrk0RcmUIS8F5tbEOuUeIplUODcgEXBKtweqXzOr/CaqAVXV+Ni8J0RNN8nrnwvpiC+bOQ68dA6xWyePJ6mk6wvvxuUIV5c4cTNRjmtDoHZj02eqci78745nXuf911zEUWeL337UGTc/QNXyIgLueWm/LgtZcKtK6i1jGR9AVQZE0VoQUvA6Jo4ParB/q7WGNrAH16X4UyaTGFSGZdZPWK3+c7ky01KxBM7L0RiwXvyJLPLL7+FmPS2Q0fvz2KVqDCmvqE7SG1pZgtcXHRCaWDXDNqhEMSeHE8NwG8LiUNF5Hil+fbEIz4PxGtSoH7JyJEP0S8b5xtfsLnYIlmD0nwlGePdrLNaubWNc40WV5rG83WT3N2ubLkKTK+VPIGk6GUyH8zgTeXLmQTbNfps4j0at56EvOLZb95ukwb54e4kMb2lgS9NI1fATNyLKyYVNVPczsfga1nmFkWeL8RpWMDseHzv5QN02Tg/0aElCf6ae5ARTZYKSKXfs2Kxs2AdDev4emgIf3n9/K7tNDvN09uQR0PN99o51oOssVyyKossqyuursxDcVo+kg71k+AsCRgbmJfsg9gmnCb04FkYB1DeUP/xWCynnSLXJ8TpUb11q1+08d7BK1+yXi3163MtM/9a7zJmw3TIMD3S8jSyrnt1xRjqXNiN2Kt8XXP2G7JElsXuLAROKJjrnliPzfXZbH479fYVlyHeGca79K4vk2drJjvceaHb++UUUCDvRpZ/27GogbhBMGK2oVnGaWFbmmdaNV7Nq3aQ6uxOcK0RHej6Zn+KOrrFbc/+s/35z2vkRSGf7imbfxORWuXt7JsrrzcarV7fEYz0gqRKMvy8pgjJ6oQSQ1+9yPWvcIe3v99MYUVtYq+JyLQy4Xx09RJVzcZtfuR3mra+a3b8HC0Q2Df3utnYDLwccvWj7hs9NDB4mmwqxpugSPs/Jqs4dTIZJZNw3eQZxKesJna+pU3JLOM10phhPpac4wka7RBN97o52lNV4+cH4rqWyMk4P7CbjrqfdP7lBYyaR0N7GMj5BrFJeSxu+SWVWnMJgwODI4szW3t8eK/W9oUlEUWLcCsrrKaLr6E9ckSWZN0yVoeoZTQwe4aUMbN6xt4ReHu6ct37v/uXfoj6X4xKVuatw6Kxsqc+DSfElpHnr7dW5YOwLAwf7ZWfsOOYPfGeFH71hhry1LnWc5onoQoj8HEpkoe8I/4potp2h2P0vIsReXPADM7u1RkiQ+snEZLlXm54e7GZrlA1swP3649xSnRxPcdslKfGcMH3mn6yUALmi9uhxLmwUS3bElyJLJEv/Eck9ZlljhjJHSTR588dCszva/n9pDIqPz5Q9sRlVkDvfswjA1Lmi9qqpc+zY9sWYkCVpy9+bdy5yoMuzqzJDKTm3VnhjSODms0+yXaQsqrFmh4nZBf6Kxakv1zmR1LiG1vX8PkiTxd799GYos8ZmfvjGpV8iRgQjfeOEgrUEP71nWjio7WVq3vhzLLioHjmS4rC1Cjcvgnb4s0fTZn9fNvn52dwfojrpYW69S61kc3w8Qoj8nFEnBIXtwqDpupZ8a5wFaPL+k1fNf+NWjSJz9LbLG4+RDG9rI6gY/fvsUml64yWmCMTTd4N5n9qLIEp9774UTPuseOUZf5AStoXXU+prLtMKzM5ioJ605afb145AnZqcvcyYIOSX+5tn9vNM7MuN53ugM8/Abx7m4tZZPvWsNuqFxsOcVHIqb85ovK+JPUDyGknWkdQdN3gEUScPvktnS5iStwaunMpPc2SnN5OWODIoE165yIUkSF55vvQj2xxunukRVEvI2Ue9fSvfwURLpCBuX1PL/v3IdRwejfPCff0l/roRvz+khtv6fZ0hpOvfcsIKMFmZZ3QYcyuKxaG2OntDAMPnoBf3oJrzeOXNOlYRBjWuAH7/TjCzBpW3VX6Y3HiH6c8Dl8LKx9iae27WKU/GP0Z+6llh2FaqUoN71Bq2e/8KnnMCa3zU9G5fUclFrLd2RpIjvF4lHdp/g8ECE/3b5mgm1+Yap89rxnwESW1Z+oHwLnAUmct7abwtMdM+qksndFwTJ6Aa///2XyUzT8TGV1bn7x68B8MBvX4YiyxwfeItUNsb5LZdXbX91E5neWAuKbNDks6ocNrao1HtljoY1nj+eQc+Vx0bTBr84kiKZNbm0zUHII+NRE7S2qHT2UvVZ+2dyfsvlmBi81fksAH/74Uu5efNyfn28n0se+C+u+OZTXPsPP2cgnuLBj17OeaG3AVjb8q5yLrtoaBocOwXvWRFmScCgfUinb4aulvWeQb7/djODCScbmx0E3YtLJhfXT1NCTBwk9TbCmSs4nfgtRjMbUKQUDe5XaXb/Eoc0MuPxN21oozXo4a3uYV47Jcr4CklG07lv516ciswXb9w84bMjva8zkuhjbfOWqohl98cbSWbdNPv68TujEz67osnFp961hj1dQ3z6idcneY2yusGt3/s1u04N8nuXruK681rQDY39p19AQmZD61Wl/FEKTn+8Ed2QWeLvRZE0ZEnig+tcNPlkjoU1frAvyVOHUvxoX5K+mMGqOoXNSyyrrTXQA8CB9nL+BMXhvOYt1HiaONr7GsPxPrxOlcc/eS1/8cGLGE6mebtnmHqvi0c/cS0fuVCme+QoraG1tIbOO/vJq5SDx0GW4NaNVjjol8fSUyb1SRgcHUzw2ukaWvwS71q6uKx8gMVRg1BCdF0nFo0RDk8U6gGW4VQaaKs9Qq23jyWen9MXXUHv6GqMaeZz37giyA8OpnnmcDdGJsm6uoWNgAQYGRlG18/tPv9f+fleTg7Fufua9SyrHWu/mchE2dPxCxyKi0tXVLaVb2Mic3xkFRc2HmR16CT7+i+cEH/+xkcu47VTg/zLK0dpH4zyTx+/glV1ft7uGeZLT7/Ffx3o4sZ1S9hx65UA7D/9AqPJAc5vuQKfK1Smn6ow6KZKV7SV5TWnWVFziuMjq/E6ZT60wc3LJzMcH9KIpnWcCly32sl59VadddAZodEbZjCsc6pHYdPKcv8khUWWFC5bdRPPHvg33jj5X7zvwj9AkiT+9/s2c88Nm5BlK4fDNA3+861/AGDLyg+Wc8lFJzwi0R+vZ1PLIO8/r5ZfHPPwX4dS3LTeTU3OkjcMk8MDQ7zU0YjPYXD9eb78vVpMCNGfI6dOnSIc7SUYqZvy867hOhr8Tja09tISPEmNu4uDPc0MRKduA3n1Eni+S+KXJ4aIJ8MsnV+3yDzdve2cOtXChg2Lp9Z2Ljx3tIevPf8Oa+oD/MUHL85vNwydFw59n7SW4PJVH8bjXOCNLiHRTIC+eCPNvgGWBU9zKjJWiRB0O3npTz7IJ7//Mv954DRr//oJgm4HkZSVA3Dt6iZ+/KmtuFSF0cQAezufw+sMLpqHfE+shTrPEE2+QcLJOkbTIVRZYutqF1tXu9B0E0kCJffwljBYFTqJacKvfpPCNCuvcqMQLK09nyWh8+gaPsLB7pfZ0PoegAki9tapZwnHu1jVeFFVeL0WSsfockLuUT6+8QSR9AW82mnyg7eTLAspuBSJwUSW4aSHBm+G61b7F02J3pkI0Z8HXr+bmtrpRSNLgP2DTbQFulni7+XSFacJJ2vpGFlBxpiYKFMDbAvoPH0oxSu9Ele6nFzYPH+XUsRfnVPCCkFPJMHvf/9lFEnikTuuJuAeu4+7O56hL3KCFfUb8w/AauLU6DKCrgitgV7SuovOcZ/VeJz85L9dx3d2HeW5o73s7xnmypWN/M/3nM+29a0osoymZ3jp6A8wTJ13r/6tRVOLbSJzfHg1G5veYU3tCQ4OriepjXUrnDgcxWR5TSceR4qeWDP9g8dwLE7NR5IkrlzzEZ5++5/ZdfxnmMAFue+9aZq83fkcezufxe+q47KVN5V3sSVCN1VOjKzk/Pqj/P4lh6j3ncdrnc78wCZVNrlq+SgXLalBMxefW99GiH6RMEyFzsgyBhP1rAqdpN4zTMg1SmdkKb3xZhg36rHZr7BtvZudR1L8piPDUMLgiuVOHFU+zamUnB6Jc+O3d9IdSfI3H7qEy5c3AGMPuHe6fk3Q08B71n6sKkvUdFPl0OD5bGw8wMqaDvqaJ77cybLEnVeu484r1006VjOyPHfwYQaip1jdeHHV9NmfLQnNS8foclaFOrig4SAHw+eTyJ758muyKnSSZt8Ayayb05GlwLFyLLdkBD0NfGDT/+CZfd/hteM/43DPLpqDK+mNHCeSHMTvquWDm/5HxU9XLCTDqVqOD69kZaiDmzccZttaFz1RB141Q51H41RkJYPJRfommGNx+i8qiKTm5cDgBtqHV2EisTJ0is1N+6n3hBmf5d/sV/jtCz3UeSQODWj8aH+SrtFzOzY/W/Z2D/Hef/wFRwej/K/rL8yX6BmGlam/59ROfK4QN17wqaq2cNO6m0PhdRimzBUb44w6j6IbM5eJxtOjPHvgu3SPHGVZ3Qbes/ZjJVptaemLN3N8eCWqrLGx8QCrQ8cJOKN41QQNnkE2Nh6g2TdAPOPlncEN6GZ1z0SfLSFvE9s238mK+guJpYc40vcaifQoK+o38YFN/wO/u7bcSyw5/Ykm9vdfSCTtR5EMVtQkcKo+9g9uYjDZUO7lFZ2qtPS/9a1v8corr5DJZLj33nvZtGlTuZd0FiQGEo0Mp0IsD3bS6B1kbV07y7TT9McbGUg0kDWcBFwyv32hh91dWfb1ZHnqcIolAZlLWp20BuWqtFCLSSyd5YHnD/DXv9yHZph8+f2b+fL7NyNJEn2Rk7xy7CeMJPoIeZt534V/sCgsmnjWz76BC1nlewf8p/nx7q9zYevVrGq8eEJnwWgqzImBt3n79K/Q9AxLa89n6/rfq4jxwcWiP9FE1nCwPNhJk2+QJt9g/jPTtHr2tw+vQp8msXaxEvQ08N4NnyCrZxhJ9FLrXYKqLF739WxIaF4ODF5Q7mWUhar79r/66qvs27ePxx57jCNHjnDvvffy7//+7+Ve1qzQDAfHR1bTFW2lNdBDgyfM8prTLAueJprxM5IKEckEuGKZj9V1Cm+cznJ6VKfncAq/U2J1nUpbjUyzXzlnXf+abvB6Z5gfv32K/7vrKKOpLEtrvPzTx6/gveeFOD7wFkd6d9EXOQnAupZ3s2XlB3Gp85tKV4mkNA/Pvh7kjqv9pOQ+Xjvxn7x24j/xOmtwKC6yeopEJgKAU/Vw1Xm3sLZ5S0UNFSoWw6lahlMhat3DBJwxZMlAM1QGEo2k5zmOeLHgUJw0BpaffUfBoqbqRH/Xrl3ccMMNAKxbt47+/n6SySQeT/U81NO6mxP/X3t3H9tUvcdx/N2uD6Mbgz05MzbYRJfdCwYwhit4GUamRcwVWVI1CzNgIk0ICZFofMIgmRqHZiKLaCQ8qSBuRiAjmikCiYZ5A0SZlCG4K7vb3APb2m5ru3Ztf/ePsSFh5gKynY3zfSVbm+6s53NOe/o9p+d3fj9PNv/1ZpJi6yB5XCfjLd0kWHuA/qOSUMTCvMmx/NoRT3VDHKfbzNS09FHTAgYUE2MNJNnAZgGb2YDNbCDWZKALMyfdPsbV1RNrMmI1mbCajBgMl87lGAwGDIO3F1s1GwZ+Blo5X7qvFKiLpyKU6r+nLl7i+ruvh4kX2vofu/j3wenUxcdQg/932XNc/NX/9/7pwlGFPxShJxTGF4rQ1dtHS3cvDZ4Ata1dnGrupivYf9oj2RbDin/E86+/9RIKVLDn360MnDJJn3gHMyfnj6ohc2+kSNTAhNAdzP/ncs61HqPVe55O3+/09vkwx1jISMxlcvLfmZw8jViz3hp3GnD3JuHuHfoKGyH0bMwV/QsXLpCbe6l/6KSkJNrb28nMzLxi2oECFArduKFsjUYjdHXTXd/w/ye+Ch7gV+KxmGwkTYiQlBAmIS6KLTZC6rgwqRke5mR46IsY+I/bRl3HOOo9sbT2WPEHY7iy+/50Np0Osel09Q3Jd1UONo/IbAwopiSEmZrkJyfFT26KD1OMwt0FKCO2aBqWyARiI8kYmsdxsvkscPaa5+NyuYi4O6/rNe4NBlDW4d8Bjbg7cblc/e9HACaRyKXLrpQH6n/3Us/RYc/yZ1wu1w3dVm40SyhIjK9nxPJdz3vjytd5dBkL28pIv87XKuLuxJqeRDB448ZiGah5Q/X2alBjrA/Y9evXc88992C393eu8thjj1FaWkpGRsYV03Z3d3P27LV/6AshhBBjXU5ODuPHX355+Zg70k9NTb2sN7zOzk5SUoZucRkXF0dOTg5ms1kawQkhhNAFpRR9fX3ExV15am/MFf28vDzeeecdCgsLcblcZGZmEhs79GVYRqPxir0cIYQQ4mb3Z3VxzBX96dOnk5uby5IlS4iJieH111/XOpIQQggxJoy5c/pCCCGEuD6js0moEEIIIW44KfpCCCGETkjRv0Znz54lPz+fTz75ROsomistLeXxxx+noKCAr776Sus4mgkEAqxevZqlS5dSUFDAt99+q3WkUaG3t5cFCxbwxRdfaB1FU6dOnSIvL4+ioiKKioooLi7WOpKmKisrKSgoYMmSJRw5ckTrOJqqqKgYfF8UFRUxa9asYZ/nmGvIpyW/309xcTFz5szROormjh07Rm1tLZ999hkej4dHHnmEhx56SOtYmjh06BDTp0/n6aefpqmpiaeeemqw10g9e//995k4caLWMTTn9/ux2+28/PLLWkfRnM/nY/v27Xz66ad4PB7Kysq47777tI6lGYfDgcPhAOD48eNUVlYO+zyl6F8Di8XCli1b2LJli9ZRNDdr1iw2btwIQEJCAn19fUSj0VHbc9hwevjhhwfvt7S0kJaWpmGa0aGuro66ujpdf6AP8Pl8WkcYNb7//nvmz5+P1WolLS2N1157TetIo0ZZWRkbNmwY9vno7xP6LzCZTH967aPemEymwY4fKioqmD9/vi4L/h85HA6effZZXnnlFa2jaG7Dhg288MILWscYFfx+PydOnGD58uUsXbqU6uoR7CJ7lGlubiYQCLBq1SoKCwt1vS7+qKamhrS0tBE5YJAjffGXHDx4kPLycrZv3651FM1VVFTgcrlYs2YN+/fv1+1O0L59+7j77ruH7Bpbj3Jzc3E6ndjtdurr61m2bBlVVVVYLBato424UChEY2Mj7777Lg0NDSxbtoxDhw7pdlsZUF5ezqJFi0ZkXlL0xXX77rvv2Lx5M1u3biUhIUHrOJr5+eefSU5OJj09nWnTphGNRnG73SQnJ2sdTRNHjhyhsbGRb775hpaWFiwWC7feeitz587VOpompk6dytSpUwGYMmUKKSkptLa2DjlI2M0uNTWVmTNnEhMTQ1ZWFvHx8breVgYcO3ZsxL4h1Pfulbhu3d3dvPnmm3z44YckJiZqHUdTP/74Izt37gSgvb0dn8+n63WyceNGPv/8c8rLy3E4HKxcuVK3BR9g79697NixA4COjg46Ojp02+5j7ty5/PDDDyil6Ojo0P22AgzuGFut1hGZnxzpX4NTp05RUlJCU1MTJpOJqqoqysrKdNlC+csvv8Tr9fLMM88MPlZSUkJ6erqGqbTxxBNP8OKLL1JYWEgoFGLdunW6/7pSXJKfn89zzz3H119/TTgcZt26dbr8ah8gLS2NBx54gCeffBKfz8fatWt1v620tbVxyy23jNj8pBteIYQQQif0vYslhBBC6IgUfSGEEEInpOgLIYQQOiFFXwghhNAJKfpCCCGETkjRF0KMKvv379c6ghA3LSn6QohRIxKJsHnzZq1jCHHTks55hBBDUkqxfv16zpw5QyQSYcaMGaxdu5bS0lIOHz5MRkYGkyZNYvz48axevZqjR4/y3nvvERMTg9Fo5NVXXyUrK4uamhpKSkowGAzExcXx9ttvYzabef7552lvbycQCLBw4UJWrFjBSy+9NDg88bZt27ReBULcfJQQQgzB4/GorVu3qmg0qqLRqHrwwQdVbW2tysvLU4FAQPX09Kj7779fbdq0SQUCAWW325XH41FKKXX48GHldDqVUkotXLhQ/fLLL0oppT744AN14MAB1dDQoPbs2aOUUioYDKq77rpLdXd3q4aGBjVv3jxtFlgIHZAjfSHEkGw2G83NzRQWFmIymWhvb6e+vp4777xzcIjpe++9F4DffvuNtrY2Vq1aBUA0GiUcDtPV1UVnZyc5OTkAOJ1OoH/shp9++ol9+/ZhMpkIBoN4PJ6RX0ghdEaKvhBiSJWVldTW1rJz504sFguLFy8mGo1e1le6wWAYvE1PT+fjjz++7Dm8Xi/RaPSK596xYwehUIjdu3cDMHv27GFcEiHEAGnIJ4QYktfrJTMzE4vFwsmTJ2lsbKStrY3Tp0/T19eH3++nuroagKysLNxuN+fOnQPgxIkT7Nq1iwkTJpCSkkJNTQ0A27ZtY9euXXi9XrKzszEYDFRVVREKhQiFQhiNRoLBoGbLLMTNTgbcEUIMqbW1FafTic1mY8aMGSQkJHDgwAFmz57N8ePHue2227DZbGRmZrJy5Uqqq6spLS3FarUSiUQoLi7m9ttvp6amhjfeeAOTyUR8fDxvvfUWTU1NrFmzhsTERBYsWMD58+c5c+YMu3fvpqCgAIvFwkcffYTNZtN6NQhxU5GiL4S4auFwmL179/Loo49iNptxOp0sXryYRYsWaR1NCHEV5Jy+EOKqmUwmGhsbcTgcWK1WsrOzsdvtWscSQlwlOdIXQgghdEIa8gkhhBA6IUVfCCGE0Akp+kIIIYROSNEXQgghdEKKvhBCCKETUvSFEEIInfgfAXi9QAqlh4cAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "sns.histplot(data=df,x=\"agecat\",hue=\"cardio\",bins=10,kde=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# I choose the AHA 2020 hypertension guideline for the categories\n",
    "\n",
    "# We already know men have more uses smoke and alcohol rather than women\n",
    "fig, axes = plt.subplots(1, 2, figsize=(15, 5), sharey=True)\n",
    "fig.suptitle('Smoke and Alcohol using by gender')\n",
    "\n",
    "# Bulbasaur\n",
    "sns.countplot(ax=axes[0], x=df.smoke,hue=df.gender)\n",
    "axes[0].set_title(\"Smoke use for gender\")\n",
    "\n",
    "# Charmander\n",
    "sns.countplot(ax=axes[1], x=df.alco,hue=df.gender)\n",
    "axes[1].set_title(\"Alcohol use for gender\")\n",
    "plt.show()\n",
    "# Gender: 2 is must be men"
   ]
  },
  {
   "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>id</th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>ap_hi</th>\n",
       "      <th>ap_lo</th>\n",
       "      <th>cholesterol</th>\n",
       "      <th>gluc</th>\n",
       "      <th>smoke</th>\n",
       "      <th>alco</th>\n",
       "      <th>active</th>\n",
       "      <th>cardio</th>\n",
       "      <th>bmi</th>\n",
       "      <th>bmicat</th>\n",
       "      <th>year</th>\n",
       "      <th>agecat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>18393</td>\n",
       "      <td>male</td>\n",
       "      <td>167.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>110</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>21.87</td>\n",
       "      <td>2.0</td>\n",
       "      <td>50.39</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>20228</td>\n",
       "      <td>female</td>\n",
       "      <td>155.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>140</td>\n",
       "      <td>90</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34.96</td>\n",
       "      <td>4.0</td>\n",
       "      <td>55.42</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>18857</td>\n",
       "      <td>female</td>\n",
       "      <td>164.0</td>\n",
       "      <td>63.0</td>\n",
       "      <td>130</td>\n",
       "      <td>70</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>23.42</td>\n",
       "      <td>2.0</td>\n",
       "      <td>51.66</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id    age  gender  height  weight  ap_hi  ap_lo  cholesterol  gluc  smoke  \\\n",
       "0   0  18393    male   167.0    61.0    110     80            1     1      0   \n",
       "1   1  20228  female   155.0    84.0    140     90            3     1      0   \n",
       "2   2  18857  female   164.0    63.0    130     70            3     1      0   \n",
       "\n",
       "   alco  active  cardio    bmi  bmicat   year  agecat  \n",
       "0     0       1       0  21.87     2.0  50.39     5.0  \n",
       "1     0       1       1  34.96     4.0  55.42     6.0  \n",
       "2     0       0       1  23.42     2.0  51.66     5.0  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.gender.replace(1,\"female\",inplace=True)\n",
    "df.gender.replace(2,\"male\",inplace=True)\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(66826, 17)\n"
     ]
    }
   ],
   "source": [
    "# Some conditions are incompatible with normal life.\n",
    "\n",
    "out_filter = ((df[\"ap_hi\"]>180) | (df[\"ap_lo\"]>150))\n",
    "df = df[~out_filter]\n",
    "out_filter2 = ((df[\"ap_hi\"] < 60) | (df[\"ap_lo\"] < 30))\n",
    "df = df[~out_filter2]\n",
    "\n",
    "# Some cases diastolic pressure is higher than systolic, it is impossible with normal physiologic condition. \n",
    "df = df[~(df['ap_lo']> df['ap_hi'])]\n",
    "\n",
    "print(df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 66826/66826 [00:06<00:00, 10199.63it/s]\n",
      "100%|██████████| 66826/66826 [00:02<00:00, 24904.66it/s]\n"
     ]
    },
    {
     "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>id</th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "      <th>height</th>\n",
       "      <th>weight</th>\n",
       "      <th>ap_hi</th>\n",
       "      <th>ap_lo</th>\n",
       "      <th>cholesterol</th>\n",
       "      <th>gluc</th>\n",
       "      <th>smoke</th>\n",
       "      <th>alco</th>\n",
       "      <th>active</th>\n",
       "      <th>cardio</th>\n",
       "      <th>bmi</th>\n",
       "      <th>bmicat</th>\n",
       "      <th>year</th>\n",
       "      <th>agecat</th>\n",
       "      <th>risk_number</th>\n",
       "      <th>hyper_tension_gr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>18393</td>\n",
       "      <td>male</td>\n",
       "      <td>167.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>110</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>21.87</td>\n",
       "      <td>2.0</td>\n",
       "      <td>50.39</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>20228</td>\n",
       "      <td>female</td>\n",
       "      <td>155.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>140</td>\n",
       "      <td>90</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34.96</td>\n",
       "      <td>4.0</td>\n",
       "      <td>55.42</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>18857</td>\n",
       "      <td>female</td>\n",
       "      <td>164.0</td>\n",
       "      <td>63.0</td>\n",
       "      <td>130</td>\n",
       "      <td>70</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>23.42</td>\n",
       "      <td>2.0</td>\n",
       "      <td>51.66</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>17623</td>\n",
       "      <td>male</td>\n",
       "      <td>168.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>150</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>28.70</td>\n",
       "      <td>3.0</td>\n",
       "      <td>48.28</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>17474</td>\n",
       "      <td>female</td>\n",
       "      <td>155.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>100</td>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>22.89</td>\n",
       "      <td>2.0</td>\n",
       "      <td>47.87</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id    age  gender  height  weight  ap_hi  ap_lo  cholesterol  gluc  smoke  \\\n",
       "0   0  18393    male   167.0    61.0    110     80            1     1      0   \n",
       "1   1  20228  female   155.0    84.0    140     90            3     1      0   \n",
       "2   2  18857  female   164.0    63.0    130     70            3     1      0   \n",
       "3   3  17623    male   168.0    81.0    150    100            1     1      0   \n",
       "4   4  17474  female   155.0    55.0    100     60            1     1      0   \n",
       "\n",
       "   alco  active  cardio    bmi  bmicat   year  agecat  risk_number  \\\n",
       "0     0       1       0  21.87     2.0  50.39     5.0          0.0   \n",
       "1     0       1       1  34.96     4.0  55.42     6.0          2.0   \n",
       "2     0       0       1  23.42     2.0  51.66     5.0          2.0   \n",
       "3     0       1       1  28.70     3.0  48.28     4.0          1.0   \n",
       "4     0       0       0  22.89     2.0  47.87     4.0          1.0   \n",
       "\n",
       "   hyper_tension_gr  \n",
       "0                 1  \n",
       "1                 2  \n",
       "2                 1  \n",
       "3                 2  \n",
       "4                 1  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find the risk number\n",
    "risk_number = []\n",
    "for i in tqdm(range(len(df))):\n",
    "    total = []\n",
    "    if df.agecat.iloc[i]>8:\n",
    "        total.append(1)\n",
    "    if df.gluc.iloc[i]>1:\n",
    "        total.append(1)\n",
    "    if df.cholesterol.iloc[i]>1:\n",
    "        total.append(1)\n",
    "    if df.bmicat.iloc[i]>=3:\n",
    "        total.append(1)\n",
    "    if df.active.iloc[i]==0:\n",
    "        total.append(1)\n",
    "    if df.alco.iloc[i] == 1:\n",
    "        total.append(1)\n",
    "    if df.smoke.iloc[i] == 1:\n",
    "        total.append(1)\n",
    "    risk_number.append(np.sum(total))\n",
    "\n",
    "df[\"risk_number\"] = risk_number\n",
    "\n",
    "hypertension_grade = []\n",
    "for i in tqdm(range(len(df))):\n",
    "    if df.ap_hi.iloc[i]<140 and df.ap_lo.iloc[i]<90:\n",
    "        if df.risk_number.iloc[i] >=3:\n",
    "            hypertension_grade.append(3)\n",
    "        else:\n",
    "            hypertension_grade.append(1)\n",
    "    elif df.ap_hi.iloc[i]>=160 and df.ap_lo.iloc[i]>=100:\n",
    "        if df.risk_number.iloc[i]>0:\n",
    "            hypertension_grade.append(3)\n",
    "        else:\n",
    "            hypertension_grade.append(2)\n",
    "    else:\n",
    "        if df.risk_number.iloc[i] < 1:\n",
    "            hypertension_grade.append(1)\n",
    "        elif df.risk_number.iloc[i] >= 3:\n",
    "            hypertension_grade.append(3)\n",
    "        else:\n",
    "            hypertension_grade.append(2)\n",
    "\n",
    "\n",
    "df[\"hyper_tension_gr\"] = hypertension_grade\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>cholesterol</th>\n",
       "      <th>gluc</th>\n",
       "      <th>smoke</th>\n",
       "      <th>alco</th>\n",
       "      <th>active</th>\n",
       "      <th>cardio</th>\n",
       "      <th>bmicat</th>\n",
       "      <th>agecat</th>\n",
       "      <th>hyper_tension_gr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>male</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>female</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>female</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   gender  cholesterol  gluc  smoke  alco  active  cardio  bmicat  agecat  \\\n",
       "0    male            1     1      0     0       1       0     2.0     5.0   \n",
       "1  female            3     1      0     0       1       1     4.0     6.0   \n",
       "2  female            3     1      0     0       0       1     2.0     5.0   \n",
       "\n",
       "   hyper_tension_gr  \n",
       "0                 1  \n",
       "1                 2  \n",
       "2                 1  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's the drop dependent columns;\n",
    "df.drop(columns=[\"id\",\"age\",\"height\",\"weight\",\"ap_hi\",\"ap_lo\",\"bmi\",\"year\",\"risk_number\"],inplace=True)\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>cholesterol</th>\n",
       "      <th>gluc</th>\n",
       "      <th>smoke</th>\n",
       "      <th>alco</th>\n",
       "      <th>active</th>\n",
       "      <th>cardio</th>\n",
       "      <th>bmicat</th>\n",
       "      <th>agecat</th>\n",
       "      <th>hyper_tension_gr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   gender  cholesterol  gluc  smoke  alco  active  cardio  bmicat  agecat  \\\n",
       "0       1            1     1      0     0       1       0     2.0     5.0   \n",
       "1       0            3     1      0     0       1       1     4.0     6.0   \n",
       "2       0            3     1      0     0       0       1     2.0     5.0   \n",
       "\n",
       "   hyper_tension_gr  \n",
       "0                 1  \n",
       "1                 2  \n",
       "2                 1  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import preprocessing\n",
    "le = preprocessing.LabelEncoder()\n",
    "df.gender = le.fit_transform(df.gender)\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>cardio</th>\n",
       "      <th>gender_1</th>\n",
       "      <th>cholesterol_2</th>\n",
       "      <th>cholesterol_3</th>\n",
       "      <th>gluc_2</th>\n",
       "      <th>gluc_3</th>\n",
       "      <th>smoke_1</th>\n",
       "      <th>alco_1</th>\n",
       "      <th>active_1</th>\n",
       "      <th>bmicat_2.0</th>\n",
       "      <th>bmicat_3.0</th>\n",
       "      <th>bmicat_4.0</th>\n",
       "      <th>bmicat_5.0</th>\n",
       "      <th>agecat_3.0</th>\n",
       "      <th>agecat_4.0</th>\n",
       "      <th>agecat_5.0</th>\n",
       "      <th>agecat_6.0</th>\n",
       "      <th>agecat_7.0</th>\n",
       "      <th>hyper_tension_gr_2</th>\n",
       "      <th>hyper_tension_gr_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   cardio  gender_1  cholesterol_2  cholesterol_3  gluc_2  gluc_3  smoke_1  \\\n",
       "0       0         1              0              0       0       0        0   \n",
       "1       1         0              0              1       0       0        0   \n",
       "2       1         0              0              1       0       0        0   \n",
       "3       1         1              0              0       0       0        0   \n",
       "4       0         0              0              0       0       0        0   \n",
       "\n",
       "   alco_1  active_1  bmicat_2.0  bmicat_3.0  bmicat_4.0  bmicat_5.0  \\\n",
       "0       0         1           1           0           0           0   \n",
       "1       0         1           0           0           1           0   \n",
       "2       0         0           1           0           0           0   \n",
       "3       0         1           0           1           0           0   \n",
       "4       0         0           1           0           0           0   \n",
       "\n",
       "   agecat_3.0  agecat_4.0  agecat_5.0  agecat_6.0  agecat_7.0  \\\n",
       "0           0           0           1           0           0   \n",
       "1           0           0           0           1           0   \n",
       "2           0           0           1           0           0   \n",
       "3           0           1           0           0           0   \n",
       "4           0           1           0           0           0   \n",
       "\n",
       "   hyper_tension_gr_2  hyper_tension_gr_3  \n",
       "0                   0                   0  \n",
       "1                   1                   0  \n",
       "2                   0                   0  \n",
       "3                   1                   0  \n",
       "4                   0                   0  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.get_dummies(df,columns=['gender', 'cholesterol', 'gluc', 'smoke', 'alco', 'active', 'bmicat',\n",
    "       'agecat', 'hyper_tension_gr'],drop_first=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((66826, 19), (66826,))"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.copy()\n",
    "y = X.cardio\n",
    "X.drop(columns=\"cardio\",inplace=True)\n",
    "X.shape,y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_preprocessed_data = X.copy()\n",
    "all_preprocessed_data[\"target\"] = y\n",
    "all_preprocessed_data.to_csv(\"all_preprocessed_data.csv\",index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((33413, 19), (33413, 19), (33413,), (33413,))"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import StratifiedKFold, train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "y,\n",
    "test_size = 0.5,\n",
    "random_state=42,\n",
    "stratify=y)\n",
    "#train = pd.concat([X_train,y_train],axis=1)\n",
    "#test = pd.concat([X_test,y_test],axis=1)\n",
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(33413, 20)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = X_train.copy()\n",
    "data[\"target\"] = y_train\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_64526_row29_col1, #T_64526_row40_col1, #T_64526_row42_col1, #T_64526_row44_col1, #T_64526_row58_col1 {\n",
       "  background-color: lightgreen;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_64526\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_64526_level0_col0\" class=\"col_heading level0 col0\" >Description</th>\n",
       "      <th id=\"T_64526_level0_col1\" class=\"col_heading level0 col1\" >Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_64526_row0_col0\" class=\"data row0 col0\" >session_id</td>\n",
       "      <td id=\"T_64526_row0_col1\" class=\"data row0 col1\" >42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_64526_row1_col0\" class=\"data row1 col0\" >Target</td>\n",
       "      <td id=\"T_64526_row1_col1\" class=\"data row1 col1\" >target</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_64526_row2_col0\" class=\"data row2 col0\" >Target Type</td>\n",
       "      <td id=\"T_64526_row2_col1\" class=\"data row2 col1\" >Binary</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_64526_row3_col0\" class=\"data row3 col0\" >Label Encoded</td>\n",
       "      <td id=\"T_64526_row3_col1\" class=\"data row3 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_64526_row4_col0\" class=\"data row4 col0\" >Original Data</td>\n",
       "      <td id=\"T_64526_row4_col1\" class=\"data row4 col1\" >(33413, 20)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_64526_row5_col0\" class=\"data row5 col0\" >Missing Values</td>\n",
       "      <td id=\"T_64526_row5_col1\" class=\"data row5 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_64526_row6_col0\" class=\"data row6 col0\" >Numeric Features</td>\n",
       "      <td id=\"T_64526_row6_col1\" class=\"data row6 col1\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_64526_row7_col0\" class=\"data row7 col0\" >Categorical Features</td>\n",
       "      <td id=\"T_64526_row7_col1\" class=\"data row7 col1\" >19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_64526_row8_col0\" class=\"data row8 col0\" >Ordinal Features</td>\n",
       "      <td id=\"T_64526_row8_col1\" class=\"data row8 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_64526_row9_col0\" class=\"data row9 col0\" >High Cardinality Features</td>\n",
       "      <td id=\"T_64526_row9_col1\" class=\"data row9 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_64526_row10_col0\" class=\"data row10 col0\" >High Cardinality Method</td>\n",
       "      <td id=\"T_64526_row10_col1\" class=\"data row10 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_64526_row11_col0\" class=\"data row11 col0\" >Transformed Train Set</td>\n",
       "      <td id=\"T_64526_row11_col1\" class=\"data row11 col1\" >(28567, 19)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_64526_row12_col0\" class=\"data row12 col0\" >Transformed Test Set</td>\n",
       "      <td id=\"T_64526_row12_col1\" class=\"data row12 col1\" >(3342, 19)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_64526_row13_col0\" class=\"data row13 col0\" >Shuffle Train-Test</td>\n",
       "      <td id=\"T_64526_row13_col1\" class=\"data row13 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_64526_row14_col0\" class=\"data row14 col0\" >Stratify Train-Test</td>\n",
       "      <td id=\"T_64526_row14_col1\" class=\"data row14 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_64526_row15_col0\" class=\"data row15 col0\" >Fold Generator</td>\n",
       "      <td id=\"T_64526_row15_col1\" class=\"data row15 col1\" >StratifiedKFold</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_64526_row16_col0\" class=\"data row16 col0\" >Fold Number</td>\n",
       "      <td id=\"T_64526_row16_col1\" class=\"data row16 col1\" >10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_64526_row17_col0\" class=\"data row17 col0\" >CPU Jobs</td>\n",
       "      <td id=\"T_64526_row17_col1\" class=\"data row17 col1\" >-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "      <td id=\"T_64526_row18_col0\" class=\"data row18 col0\" >Use GPU</td>\n",
       "      <td id=\"T_64526_row18_col1\" class=\"data row18 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "      <td id=\"T_64526_row19_col0\" class=\"data row19 col0\" >Log Experiment</td>\n",
       "      <td id=\"T_64526_row19_col1\" class=\"data row19 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "      <td id=\"T_64526_row20_col0\" class=\"data row20 col0\" >Experiment Name</td>\n",
       "      <td id=\"T_64526_row20_col1\" class=\"data row20 col1\" >clf-default-name</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "      <td id=\"T_64526_row21_col0\" class=\"data row21 col0\" >USI</td>\n",
       "      <td id=\"T_64526_row21_col1\" class=\"data row21 col1\" >b3f1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "      <td id=\"T_64526_row22_col0\" class=\"data row22 col0\" >Imputation Type</td>\n",
       "      <td id=\"T_64526_row22_col1\" class=\"data row22 col1\" >simple</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row23\" class=\"row_heading level0 row23\" >23</th>\n",
       "      <td id=\"T_64526_row23_col0\" class=\"data row23 col0\" >Iterative Imputation Iteration</td>\n",
       "      <td id=\"T_64526_row23_col1\" class=\"data row23 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row24\" class=\"row_heading level0 row24\" >24</th>\n",
       "      <td id=\"T_64526_row24_col0\" class=\"data row24 col0\" >Numeric Imputer</td>\n",
       "      <td id=\"T_64526_row24_col1\" class=\"data row24 col1\" >mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row25\" class=\"row_heading level0 row25\" >25</th>\n",
       "      <td id=\"T_64526_row25_col0\" class=\"data row25 col0\" >Iterative Imputation Numeric Model</td>\n",
       "      <td id=\"T_64526_row25_col1\" class=\"data row25 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row26\" class=\"row_heading level0 row26\" >26</th>\n",
       "      <td id=\"T_64526_row26_col0\" class=\"data row26 col0\" >Categorical Imputer</td>\n",
       "      <td id=\"T_64526_row26_col1\" class=\"data row26 col1\" >constant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row27\" class=\"row_heading level0 row27\" >27</th>\n",
       "      <td id=\"T_64526_row27_col0\" class=\"data row27 col0\" >Iterative Imputation Categorical Model</td>\n",
       "      <td id=\"T_64526_row27_col1\" class=\"data row27 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row28\" class=\"row_heading level0 row28\" >28</th>\n",
       "      <td id=\"T_64526_row28_col0\" class=\"data row28 col0\" >Unknown Categoricals Handling</td>\n",
       "      <td id=\"T_64526_row28_col1\" class=\"data row28 col1\" >least_frequent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row29\" class=\"row_heading level0 row29\" >29</th>\n",
       "      <td id=\"T_64526_row29_col0\" class=\"data row29 col0\" >Normalize</td>\n",
       "      <td id=\"T_64526_row29_col1\" class=\"data row29 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row30\" class=\"row_heading level0 row30\" >30</th>\n",
       "      <td id=\"T_64526_row30_col0\" class=\"data row30 col0\" >Normalize Method</td>\n",
       "      <td id=\"T_64526_row30_col1\" class=\"data row30 col1\" >robust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row31\" class=\"row_heading level0 row31\" >31</th>\n",
       "      <td id=\"T_64526_row31_col0\" class=\"data row31 col0\" >Transformation</td>\n",
       "      <td id=\"T_64526_row31_col1\" class=\"data row31 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row32\" class=\"row_heading level0 row32\" >32</th>\n",
       "      <td id=\"T_64526_row32_col0\" class=\"data row32 col0\" >Transformation Method</td>\n",
       "      <td id=\"T_64526_row32_col1\" class=\"data row32 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row33\" class=\"row_heading level0 row33\" >33</th>\n",
       "      <td id=\"T_64526_row33_col0\" class=\"data row33 col0\" >PCA</td>\n",
       "      <td id=\"T_64526_row33_col1\" class=\"data row33 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row34\" class=\"row_heading level0 row34\" >34</th>\n",
       "      <td id=\"T_64526_row34_col0\" class=\"data row34 col0\" >PCA Method</td>\n",
       "      <td id=\"T_64526_row34_col1\" class=\"data row34 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row35\" class=\"row_heading level0 row35\" >35</th>\n",
       "      <td id=\"T_64526_row35_col0\" class=\"data row35 col0\" >PCA Components</td>\n",
       "      <td id=\"T_64526_row35_col1\" class=\"data row35 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row36\" class=\"row_heading level0 row36\" >36</th>\n",
       "      <td id=\"T_64526_row36_col0\" class=\"data row36 col0\" >Ignore Low Variance</td>\n",
       "      <td id=\"T_64526_row36_col1\" class=\"data row36 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row37\" class=\"row_heading level0 row37\" >37</th>\n",
       "      <td id=\"T_64526_row37_col0\" class=\"data row37 col0\" >Combine Rare Levels</td>\n",
       "      <td id=\"T_64526_row37_col1\" class=\"data row37 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row38\" class=\"row_heading level0 row38\" >38</th>\n",
       "      <td id=\"T_64526_row38_col0\" class=\"data row38 col0\" >Rare Level Threshold</td>\n",
       "      <td id=\"T_64526_row38_col1\" class=\"data row38 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row39\" class=\"row_heading level0 row39\" >39</th>\n",
       "      <td id=\"T_64526_row39_col0\" class=\"data row39 col0\" >Numeric Binning</td>\n",
       "      <td id=\"T_64526_row39_col1\" class=\"data row39 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row40\" class=\"row_heading level0 row40\" >40</th>\n",
       "      <td id=\"T_64526_row40_col0\" class=\"data row40 col0\" >Remove Outliers</td>\n",
       "      <td id=\"T_64526_row40_col1\" class=\"data row40 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row41\" class=\"row_heading level0 row41\" >41</th>\n",
       "      <td id=\"T_64526_row41_col0\" class=\"data row41 col0\" >Outliers Threshold</td>\n",
       "      <td id=\"T_64526_row41_col1\" class=\"data row41 col1\" >0.050000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row42\" class=\"row_heading level0 row42\" >42</th>\n",
       "      <td id=\"T_64526_row42_col0\" class=\"data row42 col0\" >Remove Multicollinearity</td>\n",
       "      <td id=\"T_64526_row42_col1\" class=\"data row42 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row43\" class=\"row_heading level0 row43\" >43</th>\n",
       "      <td id=\"T_64526_row43_col0\" class=\"data row43 col0\" >Multicollinearity Threshold</td>\n",
       "      <td id=\"T_64526_row43_col1\" class=\"data row43 col1\" >0.900000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row44\" class=\"row_heading level0 row44\" >44</th>\n",
       "      <td id=\"T_64526_row44_col0\" class=\"data row44 col0\" >Remove Perfect Collinearity</td>\n",
       "      <td id=\"T_64526_row44_col1\" class=\"data row44 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row45\" class=\"row_heading level0 row45\" >45</th>\n",
       "      <td id=\"T_64526_row45_col0\" class=\"data row45 col0\" >Clustering</td>\n",
       "      <td id=\"T_64526_row45_col1\" class=\"data row45 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row46\" class=\"row_heading level0 row46\" >46</th>\n",
       "      <td id=\"T_64526_row46_col0\" class=\"data row46 col0\" >Clustering Iteration</td>\n",
       "      <td id=\"T_64526_row46_col1\" class=\"data row46 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row47\" class=\"row_heading level0 row47\" >47</th>\n",
       "      <td id=\"T_64526_row47_col0\" class=\"data row47 col0\" >Polynomial Features</td>\n",
       "      <td id=\"T_64526_row47_col1\" class=\"data row47 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row48\" class=\"row_heading level0 row48\" >48</th>\n",
       "      <td id=\"T_64526_row48_col0\" class=\"data row48 col0\" >Polynomial Degree</td>\n",
       "      <td id=\"T_64526_row48_col1\" class=\"data row48 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row49\" class=\"row_heading level0 row49\" >49</th>\n",
       "      <td id=\"T_64526_row49_col0\" class=\"data row49 col0\" >Trignometry Features</td>\n",
       "      <td id=\"T_64526_row49_col1\" class=\"data row49 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row50\" class=\"row_heading level0 row50\" >50</th>\n",
       "      <td id=\"T_64526_row50_col0\" class=\"data row50 col0\" >Polynomial Threshold</td>\n",
       "      <td id=\"T_64526_row50_col1\" class=\"data row50 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row51\" class=\"row_heading level0 row51\" >51</th>\n",
       "      <td id=\"T_64526_row51_col0\" class=\"data row51 col0\" >Group Features</td>\n",
       "      <td id=\"T_64526_row51_col1\" class=\"data row51 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row52\" class=\"row_heading level0 row52\" >52</th>\n",
       "      <td id=\"T_64526_row52_col0\" class=\"data row52 col0\" >Feature Selection</td>\n",
       "      <td id=\"T_64526_row52_col1\" class=\"data row52 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row53\" class=\"row_heading level0 row53\" >53</th>\n",
       "      <td id=\"T_64526_row53_col0\" class=\"data row53 col0\" >Feature Selection Method</td>\n",
       "      <td id=\"T_64526_row53_col1\" class=\"data row53 col1\" >classic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row54\" class=\"row_heading level0 row54\" >54</th>\n",
       "      <td id=\"T_64526_row54_col0\" class=\"data row54 col0\" >Features Selection Threshold</td>\n",
       "      <td id=\"T_64526_row54_col1\" class=\"data row54 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row55\" class=\"row_heading level0 row55\" >55</th>\n",
       "      <td id=\"T_64526_row55_col0\" class=\"data row55 col0\" >Feature Interaction</td>\n",
       "      <td id=\"T_64526_row55_col1\" class=\"data row55 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row56\" class=\"row_heading level0 row56\" >56</th>\n",
       "      <td id=\"T_64526_row56_col0\" class=\"data row56 col0\" >Feature Ratio</td>\n",
       "      <td id=\"T_64526_row56_col1\" class=\"data row56 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row57\" class=\"row_heading level0 row57\" >57</th>\n",
       "      <td id=\"T_64526_row57_col0\" class=\"data row57 col0\" >Interaction Threshold</td>\n",
       "      <td id=\"T_64526_row57_col1\" class=\"data row57 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row58\" class=\"row_heading level0 row58\" >58</th>\n",
       "      <td id=\"T_64526_row58_col0\" class=\"data row58 col0\" >Fix Imbalance</td>\n",
       "      <td id=\"T_64526_row58_col1\" class=\"data row58 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_64526_level0_row59\" class=\"row_heading level0 row59\" >59</th>\n",
       "      <td id=\"T_64526_row59_col0\" class=\"data row59 col0\" >Fix Imbalance Method</td>\n",
       "      <td id=\"T_64526_row59_col1\" class=\"data row59 col1\" >SMOTE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fdfbd7f4be0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "s = setup(data,\n",
    "target='target',\n",
    "categorical_features = ['gender_1', 'cholesterol_2', 'cholesterol_3', 'gluc_2', 'gluc_3',\n",
    "       'smoke_1', 'alco_1', 'active_1', 'bmicat_2.0', 'bmicat_3.0',\n",
    "       'bmicat_4.0', 'bmicat_5.0', 'agecat_3.0', 'agecat_4.0', 'agecat_5.0',\n",
    "       'agecat_6.0', 'agecat_7.0', 'hyper_tension_gr_2', 'hyper_tension_gr_3'],\n",
    "session_id=42,\n",
    "train_size = 0.9, \n",
    "data_split_shuffle=True,\n",
    "remove_multicollinearity=True,\n",
    "remove_outliers=True,\n",
    "normalize=True,\n",
    "normalize_method=\"robust\",\n",
    "use_gpu=False,\n",
    "fix_imbalance=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_2a5ed th {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_2a5ed_row0_col0, #T_2a5ed_row0_col2, #T_2a5ed_row0_col3, #T_2a5ed_row0_col4, #T_2a5ed_row1_col0, #T_2a5ed_row1_col1, #T_2a5ed_row1_col4, #T_2a5ed_row1_col5, #T_2a5ed_row1_col6, #T_2a5ed_row1_col7, #T_2a5ed_row2_col0, #T_2a5ed_row2_col1, #T_2a5ed_row2_col2, #T_2a5ed_row2_col3, #T_2a5ed_row2_col4, #T_2a5ed_row2_col5, #T_2a5ed_row2_col6, #T_2a5ed_row2_col7, #T_2a5ed_row3_col0, #T_2a5ed_row3_col1, #T_2a5ed_row3_col2, #T_2a5ed_row3_col3, #T_2a5ed_row3_col4, #T_2a5ed_row3_col5, #T_2a5ed_row3_col6, #T_2a5ed_row3_col7, #T_2a5ed_row4_col0, #T_2a5ed_row4_col1, #T_2a5ed_row4_col2, #T_2a5ed_row4_col3, #T_2a5ed_row4_col4, #T_2a5ed_row4_col5, #T_2a5ed_row4_col6, #T_2a5ed_row4_col7, #T_2a5ed_row5_col0, #T_2a5ed_row5_col1, #T_2a5ed_row5_col2, #T_2a5ed_row5_col3, #T_2a5ed_row5_col4, #T_2a5ed_row5_col5, #T_2a5ed_row5_col6, #T_2a5ed_row5_col7, #T_2a5ed_row6_col0, #T_2a5ed_row6_col1, #T_2a5ed_row6_col2, #T_2a5ed_row6_col3, #T_2a5ed_row6_col4, #T_2a5ed_row6_col5, #T_2a5ed_row6_col6, #T_2a5ed_row6_col7, #T_2a5ed_row7_col0, #T_2a5ed_row7_col1, #T_2a5ed_row7_col2, #T_2a5ed_row7_col3, #T_2a5ed_row7_col4, #T_2a5ed_row7_col5, #T_2a5ed_row7_col6, #T_2a5ed_row7_col7, #T_2a5ed_row8_col0, #T_2a5ed_row8_col1, #T_2a5ed_row8_col2, #T_2a5ed_row8_col3, #T_2a5ed_row8_col4, #T_2a5ed_row8_col5, #T_2a5ed_row8_col6, #T_2a5ed_row8_col7, #T_2a5ed_row9_col0, #T_2a5ed_row9_col1, #T_2a5ed_row9_col2, #T_2a5ed_row9_col3, #T_2a5ed_row9_col4, #T_2a5ed_row9_col5, #T_2a5ed_row9_col6, #T_2a5ed_row9_col7, #T_2a5ed_row10_col0, #T_2a5ed_row10_col1, #T_2a5ed_row10_col2, #T_2a5ed_row10_col3, #T_2a5ed_row10_col5, #T_2a5ed_row10_col6, #T_2a5ed_row10_col7, #T_2a5ed_row11_col0, #T_2a5ed_row11_col1, #T_2a5ed_row11_col2, #T_2a5ed_row11_col3, #T_2a5ed_row11_col4, #T_2a5ed_row11_col5, #T_2a5ed_row11_col6, #T_2a5ed_row11_col7, #T_2a5ed_row12_col0, #T_2a5ed_row12_col1, #T_2a5ed_row12_col2, #T_2a5ed_row12_col3, #T_2a5ed_row12_col4, #T_2a5ed_row12_col5, #T_2a5ed_row12_col6, #T_2a5ed_row12_col7, #T_2a5ed_row13_col0, #T_2a5ed_row13_col1, #T_2a5ed_row13_col2, #T_2a5ed_row13_col3, #T_2a5ed_row13_col4, #T_2a5ed_row13_col5, #T_2a5ed_row13_col6, #T_2a5ed_row13_col7, #T_2a5ed_row14_col0, #T_2a5ed_row14_col1, #T_2a5ed_row14_col2, #T_2a5ed_row14_col3, #T_2a5ed_row14_col4, #T_2a5ed_row14_col5, #T_2a5ed_row14_col6, #T_2a5ed_row14_col7, #T_2a5ed_row15_col0, #T_2a5ed_row15_col1, #T_2a5ed_row15_col2, #T_2a5ed_row15_col3, #T_2a5ed_row15_col4, #T_2a5ed_row15_col5, #T_2a5ed_row15_col6, #T_2a5ed_row15_col7 {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_2a5ed_row0_col1, #T_2a5ed_row0_col5, #T_2a5ed_row0_col6, #T_2a5ed_row0_col7, #T_2a5ed_row1_col2, #T_2a5ed_row1_col3, #T_2a5ed_row10_col4 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "}\n",
       "#T_2a5ed_row0_col8, #T_2a5ed_row1_col8, #T_2a5ed_row2_col8, #T_2a5ed_row3_col8, #T_2a5ed_row4_col8, #T_2a5ed_row6_col8, #T_2a5ed_row7_col8, #T_2a5ed_row8_col8, #T_2a5ed_row9_col8, #T_2a5ed_row10_col8, #T_2a5ed_row11_col8, #T_2a5ed_row12_col8, #T_2a5ed_row13_col8, #T_2a5ed_row14_col8, #T_2a5ed_row15_col8 {\n",
       "  text-align: left;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "#T_2a5ed_row5_col8 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_2a5ed\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_2a5ed_level0_col0\" class=\"col_heading level0 col0\" >Model</th>\n",
       "      <th id=\"T_2a5ed_level0_col1\" class=\"col_heading level0 col1\" >Accuracy</th>\n",
       "      <th id=\"T_2a5ed_level0_col2\" class=\"col_heading level0 col2\" >AUC</th>\n",
       "      <th id=\"T_2a5ed_level0_col3\" class=\"col_heading level0 col3\" >Recall</th>\n",
       "      <th id=\"T_2a5ed_level0_col4\" class=\"col_heading level0 col4\" >Prec.</th>\n",
       "      <th id=\"T_2a5ed_level0_col5\" class=\"col_heading level0 col5\" >F1</th>\n",
       "      <th id=\"T_2a5ed_level0_col6\" class=\"col_heading level0 col6\" >Kappa</th>\n",
       "      <th id=\"T_2a5ed_level0_col7\" class=\"col_heading level0 col7\" >MCC</th>\n",
       "      <th id=\"T_2a5ed_level0_col8\" class=\"col_heading level0 col8\" >TT (Sec)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row0\" class=\"row_heading level0 row0\" >gbc</th>\n",
       "      <td id=\"T_2a5ed_row0_col0\" class=\"data row0 col0\" >Gradient Boosting Classifier</td>\n",
       "      <td id=\"T_2a5ed_row0_col1\" class=\"data row0 col1\" >0.6998</td>\n",
       "      <td id=\"T_2a5ed_row0_col2\" class=\"data row0 col2\" >0.7527</td>\n",
       "      <td id=\"T_2a5ed_row0_col3\" class=\"data row0 col3\" >0.6561</td>\n",
       "      <td id=\"T_2a5ed_row0_col4\" class=\"data row0 col4\" >0.7053</td>\n",
       "      <td id=\"T_2a5ed_row0_col5\" class=\"data row0 col5\" >0.6798</td>\n",
       "      <td id=\"T_2a5ed_row0_col6\" class=\"data row0 col6\" >0.3979</td>\n",
       "      <td id=\"T_2a5ed_row0_col7\" class=\"data row0 col7\" >0.3989</td>\n",
       "      <td id=\"T_2a5ed_row0_col8\" class=\"data row0 col8\" >2.3200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row1\" class=\"row_heading level0 row1\" >lightgbm</th>\n",
       "      <td id=\"T_2a5ed_row1_col0\" class=\"data row1 col0\" >Light Gradient Boosting Machine</td>\n",
       "      <td id=\"T_2a5ed_row1_col1\" class=\"data row1 col1\" >0.6974</td>\n",
       "      <td id=\"T_2a5ed_row1_col2\" class=\"data row1 col2\" >0.7536</td>\n",
       "      <td id=\"T_2a5ed_row1_col3\" class=\"data row1 col3\" >0.6579</td>\n",
       "      <td id=\"T_2a5ed_row1_col4\" class=\"data row1 col4\" >0.7007</td>\n",
       "      <td id=\"T_2a5ed_row1_col5\" class=\"data row1 col5\" >0.6786</td>\n",
       "      <td id=\"T_2a5ed_row1_col6\" class=\"data row1 col6\" >0.3932</td>\n",
       "      <td id=\"T_2a5ed_row1_col7\" class=\"data row1 col7\" >0.3939</td>\n",
       "      <td id=\"T_2a5ed_row1_col8\" class=\"data row1 col8\" >1.7100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row2\" class=\"row_heading level0 row2\" >lr</th>\n",
       "      <td id=\"T_2a5ed_row2_col0\" class=\"data row2 col0\" >Logistic Regression</td>\n",
       "      <td id=\"T_2a5ed_row2_col1\" class=\"data row2 col1\" >0.6966</td>\n",
       "      <td id=\"T_2a5ed_row2_col2\" class=\"data row2 col2\" >0.7520</td>\n",
       "      <td id=\"T_2a5ed_row2_col3\" class=\"data row2 col3\" >0.6162</td>\n",
       "      <td id=\"T_2a5ed_row2_col4\" class=\"data row2 col4\" >0.7191</td>\n",
       "      <td id=\"T_2a5ed_row2_col5\" class=\"data row2 col5\" >0.6635</td>\n",
       "      <td id=\"T_2a5ed_row2_col6\" class=\"data row2 col6\" >0.3903</td>\n",
       "      <td id=\"T_2a5ed_row2_col7\" class=\"data row2 col7\" >0.3943</td>\n",
       "      <td id=\"T_2a5ed_row2_col8\" class=\"data row2 col8\" >2.6420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row3\" class=\"row_heading level0 row3\" >ridge</th>\n",
       "      <td id=\"T_2a5ed_row3_col0\" class=\"data row3 col0\" >Ridge Classifier</td>\n",
       "      <td id=\"T_2a5ed_row3_col1\" class=\"data row3 col1\" >0.6965</td>\n",
       "      <td id=\"T_2a5ed_row3_col2\" class=\"data row3 col2\" >0.0000</td>\n",
       "      <td id=\"T_2a5ed_row3_col3\" class=\"data row3 col3\" >0.5972</td>\n",
       "      <td id=\"T_2a5ed_row3_col4\" class=\"data row3 col4\" >0.7290</td>\n",
       "      <td id=\"T_2a5ed_row3_col5\" class=\"data row3 col5\" >0.6564</td>\n",
       "      <td id=\"T_2a5ed_row3_col6\" class=\"data row3 col6\" >0.3893</td>\n",
       "      <td id=\"T_2a5ed_row3_col7\" class=\"data row3 col7\" >0.3957</td>\n",
       "      <td id=\"T_2a5ed_row3_col8\" class=\"data row3 col8\" >1.5940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row4\" class=\"row_heading level0 row4\" >ada</th>\n",
       "      <td id=\"T_2a5ed_row4_col0\" class=\"data row4 col0\" >Ada Boost Classifier</td>\n",
       "      <td id=\"T_2a5ed_row4_col1\" class=\"data row4 col1\" >0.6965</td>\n",
       "      <td id=\"T_2a5ed_row4_col2\" class=\"data row4 col2\" >0.7512</td>\n",
       "      <td id=\"T_2a5ed_row4_col3\" class=\"data row4 col3\" >0.6009</td>\n",
       "      <td id=\"T_2a5ed_row4_col4\" class=\"data row4 col4\" >0.7268</td>\n",
       "      <td id=\"T_2a5ed_row4_col5\" class=\"data row4 col5\" >0.6578</td>\n",
       "      <td id=\"T_2a5ed_row4_col6\" class=\"data row4 col6\" >0.3895</td>\n",
       "      <td id=\"T_2a5ed_row4_col7\" class=\"data row4 col7\" >0.3952</td>\n",
       "      <td id=\"T_2a5ed_row4_col8\" class=\"data row4 col8\" >1.8960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row5\" class=\"row_heading level0 row5\" >lda</th>\n",
       "      <td id=\"T_2a5ed_row5_col0\" class=\"data row5 col0\" >Linear Discriminant Analysis</td>\n",
       "      <td id=\"T_2a5ed_row5_col1\" class=\"data row5 col1\" >0.6965</td>\n",
       "      <td id=\"T_2a5ed_row5_col2\" class=\"data row5 col2\" >0.7516</td>\n",
       "      <td id=\"T_2a5ed_row5_col3\" class=\"data row5 col3\" >0.5972</td>\n",
       "      <td id=\"T_2a5ed_row5_col4\" class=\"data row5 col4\" >0.7290</td>\n",
       "      <td id=\"T_2a5ed_row5_col5\" class=\"data row5 col5\" >0.6564</td>\n",
       "      <td id=\"T_2a5ed_row5_col6\" class=\"data row5 col6\" >0.3893</td>\n",
       "      <td id=\"T_2a5ed_row5_col7\" class=\"data row5 col7\" >0.3957</td>\n",
       "      <td id=\"T_2a5ed_row5_col8\" class=\"data row5 col8\" >1.5440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row6\" class=\"row_heading level0 row6\" >catboost</th>\n",
       "      <td id=\"T_2a5ed_row6_col0\" class=\"data row6 col0\" >CatBoost Classifier</td>\n",
       "      <td id=\"T_2a5ed_row6_col1\" class=\"data row6 col1\" >0.6965</td>\n",
       "      <td id=\"T_2a5ed_row6_col2\" class=\"data row6 col2\" >0.7491</td>\n",
       "      <td id=\"T_2a5ed_row6_col3\" class=\"data row6 col3\" >0.6486</td>\n",
       "      <td id=\"T_2a5ed_row6_col4\" class=\"data row6 col4\" >0.7035</td>\n",
       "      <td id=\"T_2a5ed_row6_col5\" class=\"data row6 col5\" >0.6749</td>\n",
       "      <td id=\"T_2a5ed_row6_col6\" class=\"data row6 col6\" >0.3912</td>\n",
       "      <td id=\"T_2a5ed_row6_col7\" class=\"data row6 col7\" >0.3924</td>\n",
       "      <td id=\"T_2a5ed_row6_col8\" class=\"data row6 col8\" >10.4660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row7\" class=\"row_heading level0 row7\" >xgboost</th>\n",
       "      <td id=\"T_2a5ed_row7_col0\" class=\"data row7 col0\" >Extreme Gradient Boosting</td>\n",
       "      <td id=\"T_2a5ed_row7_col1\" class=\"data row7 col1\" >0.6951</td>\n",
       "      <td id=\"T_2a5ed_row7_col2\" class=\"data row7 col2\" >0.7475</td>\n",
       "      <td id=\"T_2a5ed_row7_col3\" class=\"data row7 col3\" >0.6515</td>\n",
       "      <td id=\"T_2a5ed_row7_col4\" class=\"data row7 col4\" >0.7000</td>\n",
       "      <td id=\"T_2a5ed_row7_col5\" class=\"data row7 col5\" >0.6748</td>\n",
       "      <td id=\"T_2a5ed_row7_col6\" class=\"data row7 col6\" >0.3885</td>\n",
       "      <td id=\"T_2a5ed_row7_col7\" class=\"data row7 col7\" >0.3894</td>\n",
       "      <td id=\"T_2a5ed_row7_col8\" class=\"data row7 col8\" >3.1800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row8\" class=\"row_heading level0 row8\" >rf</th>\n",
       "      <td id=\"T_2a5ed_row8_col0\" class=\"data row8 col0\" >Random Forest Classifier</td>\n",
       "      <td id=\"T_2a5ed_row8_col1\" class=\"data row8 col1\" >0.6895</td>\n",
       "      <td id=\"T_2a5ed_row8_col2\" class=\"data row8 col2\" >0.7405</td>\n",
       "      <td id=\"T_2a5ed_row8_col3\" class=\"data row8 col3\" >0.6369</td>\n",
       "      <td id=\"T_2a5ed_row8_col4\" class=\"data row8 col4\" >0.6976</td>\n",
       "      <td id=\"T_2a5ed_row8_col5\" class=\"data row8 col5\" >0.6658</td>\n",
       "      <td id=\"T_2a5ed_row8_col6\" class=\"data row8 col6\" >0.3770</td>\n",
       "      <td id=\"T_2a5ed_row8_col7\" class=\"data row8 col7\" >0.3784</td>\n",
       "      <td id=\"T_2a5ed_row8_col8\" class=\"data row8 col8\" >2.4160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row9\" class=\"row_heading level0 row9\" >et</th>\n",
       "      <td id=\"T_2a5ed_row9_col0\" class=\"data row9 col0\" >Extra Trees Classifier</td>\n",
       "      <td id=\"T_2a5ed_row9_col1\" class=\"data row9 col1\" >0.6886</td>\n",
       "      <td id=\"T_2a5ed_row9_col2\" class=\"data row9 col2\" >0.7387</td>\n",
       "      <td id=\"T_2a5ed_row9_col3\" class=\"data row9 col3\" >0.6270</td>\n",
       "      <td id=\"T_2a5ed_row9_col4\" class=\"data row9 col4\" >0.7004</td>\n",
       "      <td id=\"T_2a5ed_row9_col5\" class=\"data row9 col5\" >0.6616</td>\n",
       "      <td id=\"T_2a5ed_row9_col6\" class=\"data row9 col6\" >0.3749</td>\n",
       "      <td id=\"T_2a5ed_row9_col7\" class=\"data row9 col7\" >0.3769</td>\n",
       "      <td id=\"T_2a5ed_row9_col8\" class=\"data row9 col8\" >2.2660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row10\" class=\"row_heading level0 row10\" >svm</th>\n",
       "      <td id=\"T_2a5ed_row10_col0\" class=\"data row10 col0\" >SVM - Linear Kernel</td>\n",
       "      <td id=\"T_2a5ed_row10_col1\" class=\"data row10 col1\" >0.6885</td>\n",
       "      <td id=\"T_2a5ed_row10_col2\" class=\"data row10 col2\" >0.0000</td>\n",
       "      <td id=\"T_2a5ed_row10_col3\" class=\"data row10 col3\" >0.5464</td>\n",
       "      <td id=\"T_2a5ed_row10_col4\" class=\"data row10 col4\" >0.7441</td>\n",
       "      <td id=\"T_2a5ed_row10_col5\" class=\"data row10 col5\" >0.6300</td>\n",
       "      <td id=\"T_2a5ed_row10_col6\" class=\"data row10 col6\" >0.3718</td>\n",
       "      <td id=\"T_2a5ed_row10_col7\" class=\"data row10 col7\" >0.3851</td>\n",
       "      <td id=\"T_2a5ed_row10_col8\" class=\"data row10 col8\" >1.8200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row11\" class=\"row_heading level0 row11\" >dt</th>\n",
       "      <td id=\"T_2a5ed_row11_col0\" class=\"data row11 col0\" >Decision Tree Classifier</td>\n",
       "      <td id=\"T_2a5ed_row11_col1\" class=\"data row11 col1\" >0.6880</td>\n",
       "      <td id=\"T_2a5ed_row11_col2\" class=\"data row11 col2\" >0.7373</td>\n",
       "      <td id=\"T_2a5ed_row11_col3\" class=\"data row11 col3\" >0.6241</td>\n",
       "      <td id=\"T_2a5ed_row11_col4\" class=\"data row11 col4\" >0.7007</td>\n",
       "      <td id=\"T_2a5ed_row11_col5\" class=\"data row11 col5\" >0.6601</td>\n",
       "      <td id=\"T_2a5ed_row11_col6\" class=\"data row11 col6\" >0.3735</td>\n",
       "      <td id=\"T_2a5ed_row11_col7\" class=\"data row11 col7\" >0.3757</td>\n",
       "      <td id=\"T_2a5ed_row11_col8\" class=\"data row11 col8\" >1.5780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row12\" class=\"row_heading level0 row12\" >nb</th>\n",
       "      <td id=\"T_2a5ed_row12_col0\" class=\"data row12 col0\" >Naive Bayes</td>\n",
       "      <td id=\"T_2a5ed_row12_col1\" class=\"data row12 col1\" >0.6796</td>\n",
       "      <td id=\"T_2a5ed_row12_col2\" class=\"data row12 col2\" >0.7250</td>\n",
       "      <td id=\"T_2a5ed_row12_col3\" class=\"data row12 col3\" >0.6022</td>\n",
       "      <td id=\"T_2a5ed_row12_col4\" class=\"data row12 col4\" >0.6968</td>\n",
       "      <td id=\"T_2a5ed_row12_col5\" class=\"data row12 col5\" >0.6461</td>\n",
       "      <td id=\"T_2a5ed_row12_col6\" class=\"data row12 col6\" >0.3562</td>\n",
       "      <td id=\"T_2a5ed_row12_col7\" class=\"data row12 col7\" >0.3594</td>\n",
       "      <td id=\"T_2a5ed_row12_col8\" class=\"data row12 col8\" >1.5780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row13\" class=\"row_heading level0 row13\" >knn</th>\n",
       "      <td id=\"T_2a5ed_row13_col0\" class=\"data row13 col0\" >K Neighbors Classifier</td>\n",
       "      <td id=\"T_2a5ed_row13_col1\" class=\"data row13 col1\" >0.6601</td>\n",
       "      <td id=\"T_2a5ed_row13_col2\" class=\"data row13 col2\" >0.7017</td>\n",
       "      <td id=\"T_2a5ed_row13_col3\" class=\"data row13 col3\" >0.6244</td>\n",
       "      <td id=\"T_2a5ed_row13_col4\" class=\"data row13 col4\" >0.6583</td>\n",
       "      <td id=\"T_2a5ed_row13_col5\" class=\"data row13 col5\" >0.6408</td>\n",
       "      <td id=\"T_2a5ed_row13_col6\" class=\"data row13 col6\" >0.3187</td>\n",
       "      <td id=\"T_2a5ed_row13_col7\" class=\"data row13 col7\" >0.3192</td>\n",
       "      <td id=\"T_2a5ed_row13_col8\" class=\"data row13 col8\" >4.9280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row14\" class=\"row_heading level0 row14\" >qda</th>\n",
       "      <td id=\"T_2a5ed_row14_col0\" class=\"data row14 col0\" >Quadratic Discriminant Analysis</td>\n",
       "      <td id=\"T_2a5ed_row14_col1\" class=\"data row14 col1\" >0.5472</td>\n",
       "      <td id=\"T_2a5ed_row14_col2\" class=\"data row14 col2\" >0.5652</td>\n",
       "      <td id=\"T_2a5ed_row14_col3\" class=\"data row14 col3\" >0.6450</td>\n",
       "      <td id=\"T_2a5ed_row14_col4\" class=\"data row14 col4\" >0.5710</td>\n",
       "      <td id=\"T_2a5ed_row14_col5\" class=\"data row14 col5\" >0.5418</td>\n",
       "      <td id=\"T_2a5ed_row14_col6\" class=\"data row14 col6\" >0.1001</td>\n",
       "      <td id=\"T_2a5ed_row14_col7\" class=\"data row14 col7\" >0.1221</td>\n",
       "      <td id=\"T_2a5ed_row14_col8\" class=\"data row14 col8\" >1.6380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_2a5ed_level0_row15\" class=\"row_heading level0 row15\" >dummy</th>\n",
       "      <td id=\"T_2a5ed_row15_col0\" class=\"data row15 col0\" >Dummy Classifier</td>\n",
       "      <td id=\"T_2a5ed_row15_col1\" class=\"data row15 col1\" >0.5144</td>\n",
       "      <td id=\"T_2a5ed_row15_col2\" class=\"data row15 col2\" >0.5000</td>\n",
       "      <td id=\"T_2a5ed_row15_col3\" class=\"data row15 col3\" >0.0000</td>\n",
       "      <td id=\"T_2a5ed_row15_col4\" class=\"data row15 col4\" >0.0000</td>\n",
       "      <td id=\"T_2a5ed_row15_col5\" class=\"data row15 col5\" >0.0000</td>\n",
       "      <td id=\"T_2a5ed_row15_col6\" class=\"data row15 col6\" >0.0000</td>\n",
       "      <td id=\"T_2a5ed_row15_col7\" class=\"data row15 col7\" >0.0000</td>\n",
       "      <td id=\"T_2a5ed_row15_col8\" class=\"data row15 col8\" >1.6260</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fdfbd7f4430>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best_models = compare_models(cross_validation=True,\n",
    "                            fold=5,\n",
    "                            sort=\"Accuracy\",\n",
    "                            n_select=1,\n",
    "                            )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_fba92_row5_col0, #T_fba92_row5_col1, #T_fba92_row5_col2, #T_fba92_row5_col3, #T_fba92_row5_col4, #T_fba92_row5_col5, #T_fba92_row5_col6 {\n",
       "  background: yellow;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_fba92\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_fba92_level0_col0\" class=\"col_heading level0 col0\" >Accuracy</th>\n",
       "      <th id=\"T_fba92_level0_col1\" class=\"col_heading level0 col1\" >AUC</th>\n",
       "      <th id=\"T_fba92_level0_col2\" class=\"col_heading level0 col2\" >Recall</th>\n",
       "      <th id=\"T_fba92_level0_col3\" class=\"col_heading level0 col3\" >Prec.</th>\n",
       "      <th id=\"T_fba92_level0_col4\" class=\"col_heading level0 col4\" >F1</th>\n",
       "      <th id=\"T_fba92_level0_col5\" class=\"col_heading level0 col5\" >Kappa</th>\n",
       "      <th id=\"T_fba92_level0_col6\" class=\"col_heading level0 col6\" >MCC</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th class=\"index_name level0\" >Fold</th>\n",
       "      <th class=\"blank col0\" >&nbsp;</th>\n",
       "      <th class=\"blank col1\" >&nbsp;</th>\n",
       "      <th class=\"blank col2\" >&nbsp;</th>\n",
       "      <th class=\"blank col3\" >&nbsp;</th>\n",
       "      <th class=\"blank col4\" >&nbsp;</th>\n",
       "      <th class=\"blank col5\" >&nbsp;</th>\n",
       "      <th class=\"blank col6\" >&nbsp;</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_fba92_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_fba92_row0_col0\" class=\"data row0 col0\" >0.6990</td>\n",
       "      <td id=\"T_fba92_row0_col1\" class=\"data row0 col1\" >0.7527</td>\n",
       "      <td id=\"T_fba92_row0_col2\" class=\"data row0 col2\" >0.6595</td>\n",
       "      <td id=\"T_fba92_row0_col3\" class=\"data row0 col3\" >0.7025</td>\n",
       "      <td id=\"T_fba92_row0_col4\" class=\"data row0 col4\" >0.6803</td>\n",
       "      <td id=\"T_fba92_row0_col5\" class=\"data row0 col5\" >0.3964</td>\n",
       "      <td id=\"T_fba92_row0_col6\" class=\"data row0 col6\" >0.3971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fba92_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_fba92_row1_col0\" class=\"data row1 col0\" >0.6918</td>\n",
       "      <td id=\"T_fba92_row1_col1\" class=\"data row1 col1\" >0.7464</td>\n",
       "      <td id=\"T_fba92_row1_col2\" class=\"data row1 col2\" >0.6537</td>\n",
       "      <td id=\"T_fba92_row1_col3\" class=\"data row1 col3\" >0.6940</td>\n",
       "      <td id=\"T_fba92_row1_col4\" class=\"data row1 col4\" >0.6732</td>\n",
       "      <td id=\"T_fba92_row1_col5\" class=\"data row1 col5\" >0.3821</td>\n",
       "      <td id=\"T_fba92_row1_col6\" class=\"data row1 col6\" >0.3827</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fba92_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_fba92_row2_col0\" class=\"data row2 col0\" >0.6970</td>\n",
       "      <td id=\"T_fba92_row2_col1\" class=\"data row2 col1\" >0.7551</td>\n",
       "      <td id=\"T_fba92_row2_col2\" class=\"data row2 col2\" >0.6565</td>\n",
       "      <td id=\"T_fba92_row2_col3\" class=\"data row2 col3\" >0.7007</td>\n",
       "      <td id=\"T_fba92_row2_col4\" class=\"data row2 col4\" >0.6778</td>\n",
       "      <td id=\"T_fba92_row2_col5\" class=\"data row2 col5\" >0.3924</td>\n",
       "      <td id=\"T_fba92_row2_col6\" class=\"data row2 col6\" >0.3932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fba92_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_fba92_row3_col0\" class=\"data row3 col0\" >0.7077</td>\n",
       "      <td id=\"T_fba92_row3_col1\" class=\"data row3 col1\" >0.7580</td>\n",
       "      <td id=\"T_fba92_row3_col2\" class=\"data row3 col2\" >0.6539</td>\n",
       "      <td id=\"T_fba92_row3_col3\" class=\"data row3 col3\" >0.7187</td>\n",
       "      <td id=\"T_fba92_row3_col4\" class=\"data row3 col4\" >0.6848</td>\n",
       "      <td id=\"T_fba92_row3_col5\" class=\"data row3 col5\" >0.4134</td>\n",
       "      <td id=\"T_fba92_row3_col6\" class=\"data row3 col6\" >0.4150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fba92_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_fba92_row4_col0\" class=\"data row4 col0\" >0.7035</td>\n",
       "      <td id=\"T_fba92_row4_col1\" class=\"data row4 col1\" >0.7515</td>\n",
       "      <td id=\"T_fba92_row4_col2\" class=\"data row4 col2\" >0.6568</td>\n",
       "      <td id=\"T_fba92_row4_col3\" class=\"data row4 col3\" >0.7106</td>\n",
       "      <td id=\"T_fba92_row4_col4\" class=\"data row4 col4\" >0.6827</td>\n",
       "      <td id=\"T_fba92_row4_col5\" class=\"data row4 col5\" >0.4052</td>\n",
       "      <td id=\"T_fba92_row4_col6\" class=\"data row4 col6\" >0.4063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fba92_level0_row5\" class=\"row_heading level0 row5\" >Mean</th>\n",
       "      <td id=\"T_fba92_row5_col0\" class=\"data row5 col0\" >0.6998</td>\n",
       "      <td id=\"T_fba92_row5_col1\" class=\"data row5 col1\" >0.7527</td>\n",
       "      <td id=\"T_fba92_row5_col2\" class=\"data row5 col2\" >0.6561</td>\n",
       "      <td id=\"T_fba92_row5_col3\" class=\"data row5 col3\" >0.7053</td>\n",
       "      <td id=\"T_fba92_row5_col4\" class=\"data row5 col4\" >0.6798</td>\n",
       "      <td id=\"T_fba92_row5_col5\" class=\"data row5 col5\" >0.3979</td>\n",
       "      <td id=\"T_fba92_row5_col6\" class=\"data row5 col6\" >0.3989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fba92_level0_row6\" class=\"row_heading level0 row6\" >Std</th>\n",
       "      <td id=\"T_fba92_row6_col0\" class=\"data row6 col0\" >0.0054</td>\n",
       "      <td id=\"T_fba92_row6_col1\" class=\"data row6 col1\" >0.0039</td>\n",
       "      <td id=\"T_fba92_row6_col2\" class=\"data row6 col2\" >0.0021</td>\n",
       "      <td id=\"T_fba92_row6_col3\" class=\"data row6 col3\" >0.0086</td>\n",
       "      <td id=\"T_fba92_row6_col4\" class=\"data row6 col4\" >0.0040</td>\n",
       "      <td id=\"T_fba92_row6_col5\" class=\"data row6 col5\" >0.0107</td>\n",
       "      <td id=\"T_fba92_row6_col6\" class=\"data row6 col6\" >0.0111</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fdfbe2a17c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best_model = create_model(estimator=\"gbc\",\n",
    "            fold=5,\n",
    "            cross_validation=True,\n",
    "            )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_27852_row10_col0, #T_27852_row10_col1, #T_27852_row10_col2, #T_27852_row10_col3, #T_27852_row10_col4, #T_27852_row10_col5, #T_27852_row10_col6 {\n",
       "  background: yellow;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_27852\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_27852_level0_col0\" class=\"col_heading level0 col0\" >Accuracy</th>\n",
       "      <th id=\"T_27852_level0_col1\" class=\"col_heading level0 col1\" >AUC</th>\n",
       "      <th id=\"T_27852_level0_col2\" class=\"col_heading level0 col2\" >Recall</th>\n",
       "      <th id=\"T_27852_level0_col3\" class=\"col_heading level0 col3\" >Prec.</th>\n",
       "      <th id=\"T_27852_level0_col4\" class=\"col_heading level0 col4\" >F1</th>\n",
       "      <th id=\"T_27852_level0_col5\" class=\"col_heading level0 col5\" >Kappa</th>\n",
       "      <th id=\"T_27852_level0_col6\" class=\"col_heading level0 col6\" >MCC</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th class=\"index_name level0\" >Fold</th>\n",
       "      <th class=\"blank col0\" >&nbsp;</th>\n",
       "      <th class=\"blank col1\" >&nbsp;</th>\n",
       "      <th class=\"blank col2\" >&nbsp;</th>\n",
       "      <th class=\"blank col3\" >&nbsp;</th>\n",
       "      <th class=\"blank col4\" >&nbsp;</th>\n",
       "      <th class=\"blank col5\" >&nbsp;</th>\n",
       "      <th class=\"blank col6\" >&nbsp;</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_27852_row0_col0\" class=\"data row0 col0\" >0.6975848792</td>\n",
       "      <td id=\"T_27852_row0_col1\" class=\"data row0 col1\" >0.7557416188</td>\n",
       "      <td id=\"T_27852_row0_col2\" class=\"data row0 col2\" >0.6469740634</td>\n",
       "      <td id=\"T_27852_row0_col3\" class=\"data row0 col3\" >0.7059748428</td>\n",
       "      <td id=\"T_27852_row0_col4\" class=\"data row0 col4\" >0.6751879699</td>\n",
       "      <td id=\"T_27852_row0_col5\" class=\"data row0 col5\" >0.3932852703</td>\n",
       "      <td id=\"T_27852_row0_col6\" class=\"data row0 col6\" >0.3945965735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_27852_row1_col0\" class=\"data row1 col0\" >0.6989849492</td>\n",
       "      <td id=\"T_27852_row1_col1\" class=\"data row1 col1\" >0.7503680776</td>\n",
       "      <td id=\"T_27852_row1_col2\" class=\"data row1 col2\" >0.6707492795</td>\n",
       "      <td id=\"T_27852_row1_col3\" class=\"data row1 col3\" >0.6979010495</td>\n",
       "      <td id=\"T_27852_row1_col4\" class=\"data row1 col4\" >0.6840558413</td>\n",
       "      <td id=\"T_27852_row1_col5\" class=\"data row1 col5\" >0.3968386452</td>\n",
       "      <td id=\"T_27852_row1_col6\" class=\"data row1 col6\" >0.3971235561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_27852_row2_col0\" class=\"data row2 col0\" >0.7003850193</td>\n",
       "      <td id=\"T_27852_row2_col1\" class=\"data row2 col1\" >0.7556305146</td>\n",
       "      <td id=\"T_27852_row2_col2\" class=\"data row2 col2\" >0.6611391492</td>\n",
       "      <td id=\"T_27852_row2_col3\" class=\"data row2 col3\" >0.7037605526</td>\n",
       "      <td id=\"T_27852_row2_col4\" class=\"data row2 col4\" >0.6817843866</td>\n",
       "      <td id=\"T_27852_row2_col5\" class=\"data row2 col5\" >0.3992367117</td>\n",
       "      <td id=\"T_27852_row2_col6\" class=\"data row2 col6\" >0.3999322998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_27852_row3_col0\" class=\"data row3 col0\" >0.6800840042</td>\n",
       "      <td id=\"T_27852_row3_col1\" class=\"data row3 col1\" >0.7377580448</td>\n",
       "      <td id=\"T_27852_row3_col2\" class=\"data row3 col2\" >0.6387887527</td>\n",
       "      <td id=\"T_27852_row3_col3\" class=\"data row3 col3\" >0.6820631255</td>\n",
       "      <td id=\"T_27852_row3_col4\" class=\"data row3 col4\" >0.6597170514</td>\n",
       "      <td id=\"T_27852_row3_col5\" class=\"data row3 col5\" >0.3584784679</td>\n",
       "      <td id=\"T_27852_row3_col6\" class=\"data row3 col6\" >0.3591642297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_27852_row4_col0\" class=\"data row4 col0\" >0.7038851943</td>\n",
       "      <td id=\"T_27852_row4_col1\" class=\"data row4 col1\" >0.7666063397</td>\n",
       "      <td id=\"T_27852_row4_col2\" class=\"data row4 col2\" >0.6625811103</td>\n",
       "      <td id=\"T_27852_row4_col3\" class=\"data row4 col3\" >0.7085582113</td>\n",
       "      <td id=\"T_27852_row4_col4\" class=\"data row4 col4\" >0.6847988077</td>\n",
       "      <td id=\"T_27852_row4_col5\" class=\"data row4 col5\" >0.4061823301</td>\n",
       "      <td id=\"T_27852_row4_col6\" class=\"data row4 col6\" >0.4069952426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_27852_row5_col0\" class=\"data row5 col0\" >0.6881344067</td>\n",
       "      <td id=\"T_27852_row5_col1\" class=\"data row5 col1\" >0.7458239532</td>\n",
       "      <td id=\"T_27852_row5_col2\" class=\"data row5 col2\" >0.6647440519</td>\n",
       "      <td id=\"T_27852_row5_col3\" class=\"data row5 col3\" >0.6839762611</td>\n",
       "      <td id=\"T_27852_row5_col4\" class=\"data row5 col4\" >0.6742230347</td>\n",
       "      <td id=\"T_27852_row5_col5\" class=\"data row5 col5\" >0.3752460092</td>\n",
       "      <td id=\"T_27852_row5_col6\" class=\"data row5 col6\" >0.3753863945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_27852_row6_col0\" class=\"data row6 col0\" >0.7143857193</td>\n",
       "      <td id=\"T_27852_row6_col1\" class=\"data row6 col1\" >0.7655773485</td>\n",
       "      <td id=\"T_27852_row6_col2\" class=\"data row6 col2\" >0.6791636626</td>\n",
       "      <td id=\"T_27852_row6_col3\" class=\"data row6 col3\" >0.7174409749</td>\n",
       "      <td id=\"T_27852_row6_col4\" class=\"data row6 col4\" >0.6977777778</td>\n",
       "      <td id=\"T_27852_row6_col5\" class=\"data row6 col5\" >0.4274265050</td>\n",
       "      <td id=\"T_27852_row6_col6\" class=\"data row6 col6\" >0.4280038778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_27852_row7_col0\" class=\"data row7 col0\" >0.7013305322</td>\n",
       "      <td id=\"T_27852_row7_col1\" class=\"data row7 col1\" >0.7499849080</td>\n",
       "      <td id=\"T_27852_row7_col2\" class=\"data row7 col2\" >0.6467195386</td>\n",
       "      <td id=\"T_27852_row7_col3\" class=\"data row7 col3\" >0.7119047619</td>\n",
       "      <td id=\"T_27852_row7_col4\" class=\"data row7 col4\" >0.6777483944</td>\n",
       "      <td id=\"T_27852_row7_col5\" class=\"data row7 col5\" >0.4006365215</td>\n",
       "      <td id=\"T_27852_row7_col6\" class=\"data row7 col6\" >0.4022412951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_27852_row8_col0\" class=\"data row8 col0\" >0.7013305322</td>\n",
       "      <td id=\"T_27852_row8_col1\" class=\"data row8 col1\" >0.7499586504</td>\n",
       "      <td id=\"T_27852_row8_col2\" class=\"data row8 col2\" >0.6496034607</td>\n",
       "      <td id=\"T_27852_row8_col3\" class=\"data row8 col3\" >0.7105678233</td>\n",
       "      <td id=\"T_27852_row8_col4\" class=\"data row8 col4\" >0.6787193974</td>\n",
       "      <td id=\"T_27852_row8_col5\" class=\"data row8 col5\" >0.4007332394</td>\n",
       "      <td id=\"T_27852_row8_col6\" class=\"data row8 col6\" >0.4021410541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_27852_row9_col0\" class=\"data row9 col0\" >0.7023809524</td>\n",
       "      <td id=\"T_27852_row9_col1\" class=\"data row9 col1\" >0.7533667435</td>\n",
       "      <td id=\"T_27852_row9_col2\" class=\"data row9 col2\" >0.6705118962</td>\n",
       "      <td id=\"T_27852_row9_col3\" class=\"data row9 col3\" >0.7029478458</td>\n",
       "      <td id=\"T_27852_row9_col4\" class=\"data row9 col4\" >0.6863468635</td>\n",
       "      <td id=\"T_27852_row9_col5\" class=\"data row9 col5\" >0.4035026161</td>\n",
       "      <td id=\"T_27852_row9_col6\" class=\"data row9 col6\" >0.4039101964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row10\" class=\"row_heading level0 row10\" >Mean</th>\n",
       "      <td id=\"T_27852_row10_col0\" class=\"data row10 col0\" >0.6988486189</td>\n",
       "      <td id=\"T_27852_row10_col1\" class=\"data row10 col1\" >0.7530816199</td>\n",
       "      <td id=\"T_27852_row10_col2\" class=\"data row10 col2\" >0.6590974965</td>\n",
       "      <td id=\"T_27852_row10_col3\" class=\"data row10 col3\" >0.7025095449</td>\n",
       "      <td id=\"T_27852_row10_col4\" class=\"data row10 col4\" >0.6800359525</td>\n",
       "      <td id=\"T_27852_row10_col5\" class=\"data row10 col5\" >0.3961566316</td>\n",
       "      <td id=\"T_27852_row10_col6\" class=\"data row10 col6\" >0.3969494720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_27852_level0_row11\" class=\"row_heading level0 row11\" >Std</th>\n",
       "      <td id=\"T_27852_row11_col0\" class=\"data row11 col0\" >0.0087318211</td>\n",
       "      <td id=\"T_27852_row11_col1\" class=\"data row11 col1\" >0.0081664349</td>\n",
       "      <td id=\"T_27852_row11_col2\" class=\"data row11 col2\" >0.0123274667</td>\n",
       "      <td id=\"T_27852_row11_col3\" class=\"data row11 col3\" >0.0109967370</td>\n",
       "      <td id=\"T_27852_row11_col4\" class=\"data row11 col4\" >0.0093467546</td>\n",
       "      <td id=\"T_27852_row11_col5\" class=\"data row11 col5\" >0.0174596471</td>\n",
       "      <td id=\"T_27852_row11_col6\" class=\"data row11 col6\" >0.0175381178</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7fdfbd3b36d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tuned_gbc = tune_model(estimator=best_model,\n",
    "                       round=10,\n",
    "                       fold=10,\n",
    "                       optimize=\"AUC\",\n",
    "                       search_library=\"optuna\",\n",
    "                       search_algorithm=\"tpe\",\n",
    "                       choose_better=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = predict_model(tuned_gbc,X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report: \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.70      0.72      0.71     16943\n",
      "           1       0.70      0.68      0.69     16470\n",
      "\n",
      "    accuracy                           0.70     33413\n",
      "   macro avg       0.70      0.70      0.70     33413\n",
      "weighted avg       0.70      0.70      0.70     33413\n",
      "\n",
      "Train Accuracy score: 0.70\n",
      "Train ROC AUC score: 0.70\n"
     ]
    }
   ],
   "source": [
    "def model_eval(prediction,test):\n",
    "    print(\"Classification Report: \")\n",
    "    print(classification_report(test,prediction))\n",
    "    print('Train Accuracy score: {0:0.2f}'.format(accuracy_score(test, prediction)))\n",
    "    print('Train ROC AUC score: {0:0.2f}'.format(roc_auc_score(test, prediction)))\n",
    "\n",
    "\n",
    "model_eval(prediction.Label,y_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "evaluate_model(tuned_gbc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transformation Pipeline and Model Successfully Saved\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(Pipeline(memory=None,\n",
       "          steps=[('dtypes',\n",
       "                  DataTypes_Auto_infer(categorical_features=['gender_1',\n",
       "                                                             'cholesterol_2',\n",
       "                                                             'cholesterol_3',\n",
       "                                                             'gluc_2', 'gluc_3',\n",
       "                                                             'smoke_1', 'alco_1',\n",
       "                                                             'active_1',\n",
       "                                                             'bmicat_2.0',\n",
       "                                                             'bmicat_3.0',\n",
       "                                                             'bmicat_4.0',\n",
       "                                                             'bmicat_5.0',\n",
       "                                                             'agecat_3.0',\n",
       "                                                             'agecat_4.0',\n",
       "                                                             'agecat_5.0',\n",
       "                                                             'agecat_6.0',\n",
       "                                                             'agecat_7.0',\n",
       "                                                             'hyper_tension_gr_2',\n",
       "                                                             'hyper_tension_gr_3'],\n",
       "                                       display_types=True...\n",
       "                                             max_features=0.6299166796003268,\n",
       "                                             max_leaf_nodes=None,\n",
       "                                             min_impurity_decrease=6.022651773824478e-09,\n",
       "                                             min_impurity_split=None,\n",
       "                                             min_samples_leaf=1,\n",
       "                                             min_samples_split=8,\n",
       "                                             min_weight_fraction_leaf=0.0,\n",
       "                                             n_estimators=184,\n",
       "                                             n_iter_no_change=None,\n",
       "                                             presort='deprecated',\n",
       "                                             random_state=42,\n",
       "                                             subsample=0.3537661849547683,\n",
       "                                             tol=0.0001, validation_fraction=0.1,\n",
       "                                             verbose=0, warm_start=False)]],\n",
       "          verbose=False),\n",
       " '/home/izzet/Desktop/stream-medical-data/model.pkl')"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "save_model(tuned_gbc,\n",
    "           \"code\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transformation Pipeline and Model Successfully Loaded\n"
     ]
    }
   ],
   "source": [
    "model = load_model(\"code/model.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report: \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.70      0.72      0.71     16943\n",
      "           1       0.70      0.68      0.69     16470\n",
      "\n",
      "    accuracy                           0.70     33413\n",
      "   macro avg       0.70      0.70      0.70     33413\n",
      "weighted avg       0.70      0.70      0.70     33413\n",
      "\n",
      "Train Accuracy score: 0.70\n",
      "Train ROC AUC score: 0.70\n"
     ]
    }
   ],
   "source": [
    "prediction = predict_model(model,X_test)\n",
    "model_eval(prediction.Label,y_test)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.13 ('datascience')",
   "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": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "d5e3d835a3425d0426acedd99e77a951ca3551842067eb24f0bf8de3598cac8b"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}