--- a
+++ b/Final_Project.ipynb
@@ -0,0 +1,2284 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "e8ccf29a",
+   "metadata": {},
+   "source": [
+    "## Predictive Modeling of Smoking Status Using Bio-Signals:\n",
+    "## Leveraging AI and Machine Learning for Improved Smoking Cessation Strategies\n",
+    "\n",
+    "\n",
+    "### AIM OF PROJECT:\n",
+    "The objective of this study is to develop a machine learning model that utilizes bio-signals to accurately predict the smoking status of an individual. By leveraging AI and machine learning techniques, aim to provide a reliable tool for identifying individuals who are smokers or non-smokers. This predictive model can aid in the development of effective smoking cessation strategies and interventions, ultimately contributing to improved public health outcomes. The model will consider various bio-signals, such as physiological, behavioral, or environmental factors, to achieve accurate and reliable predictions. The goal is to create a practical and accessible solution that can assist healthcare professionals in assessing an individual's smoking status and provide personalized recommendations for smoking cessation.\n",
+    "\n",
+    "\n",
+    "### Table of Contents:\n",
+    "\n",
+    "> - INTRODUCTION\n",
+    "> - DATA PREPROCESSING\n",
+    "> - DATA EXPLORATION\n",
+    "> - FEATURE SELECTION\n",
+    "> - TRAINING\n",
+    "> - EVALUATION \n",
+    "> - PARAMETER TUNING\n",
+    "> - CONCLUSIONS\n",
+    "> - REFERENCES\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "95dca433",
+   "metadata": {},
+   "source": [
+    "### 1. INTRODUCTION\n",
+    "\n",
+    "> Smoking is a well-known contributor to a number of health problems and has a major effect on public health globally. For the purpose of creating successful smoking cessation methods and interventions, it is essential to precisely determine each individual's smoking status. Self-reporting, which can be subjective and unreliable, is frequently used in traditional techniques of determining smoking status. The use of AI and machine learning approaches in predicting smoking status using biosignals has attracted attention as a solution to this problem.\n",
+    "\n",
+    "> In order to accurately anticipate a person's smoking status, the goal of this project is to develop a predictive modelling strategy that makes use of AI and machine learning algorithms. The model intends to provide a trustworthy tool for healthcare providers to assess smoking status and direct individualised smoking cessation efforts by analysing bio-signals, including physiological, behavioural, and environmental aspects.\n",
+    "\n",
+    "> The project's many phases, including data preprocessing, dataset exploration, feature selection, algorithm selection, training, assessment, parameter tweaking, and deployment, will be covered in the report. Each step will be carefully explained, showing the approaches used and the justification for the choices made.\n",
+    "\n",
+    "> A prediction model with high accuracy in predicting smoking status is the anticipated result of this investigation. The model's usefulness and accessibility will be emphasised to ensure that healthcare professionals may use it in practical contexts. The created model can help enhance smoking cessation tactics and ultimately lessen the harmful effects of smoking on public health by enabling precise identification of smoking status.\n",
+    "\n",
+    "> A summary of the report's conclusions, its caveats, and its suggestions for additional research are included at the end. A list of the cited sources and pertinent material used in the study is provided in the references section."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "21ea682a",
+   "metadata": {},
+   "source": [
+    "### 2. DATA PREPROCESSING"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 156,
+   "id": "714c7eac",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Importing the Library  \n",
+    "import pandas as pd;"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 157,
+   "id": "da685198",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>age</th>\n",
+       "      <th>height(cm)</th>\n",
+       "      <th>weight(kg)</th>\n",
+       "      <th>waist(cm)</th>\n",
+       "      <th>eyesight(left)</th>\n",
+       "      <th>eyesight(right)</th>\n",
+       "      <th>hearing(left)</th>\n",
+       "      <th>hearing(right)</th>\n",
+       "      <th>systolic</th>\n",
+       "      <th>relaxation</th>\n",
+       "      <th>...</th>\n",
+       "      <th>HDL</th>\n",
+       "      <th>LDL</th>\n",
+       "      <th>hemoglobin</th>\n",
+       "      <th>Urine protein</th>\n",
+       "      <th>serum creatinine</th>\n",
+       "      <th>AST</th>\n",
+       "      <th>ALT</th>\n",
+       "      <th>Gtp</th>\n",
+       "      <th>dental caries</th>\n",
+       "      <th>smoking</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>35</td>\n",
+       "      <td>170</td>\n",
+       "      <td>85</td>\n",
+       "      <td>97.0</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>118</td>\n",
+       "      <td>78</td>\n",
+       "      <td>...</td>\n",
+       "      <td>70</td>\n",
+       "      <td>142</td>\n",
+       "      <td>19.8</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>61</td>\n",
+       "      <td>115</td>\n",
+       "      <td>125</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>20</td>\n",
+       "      <td>175</td>\n",
+       "      <td>110</td>\n",
+       "      <td>110.0</td>\n",
+       "      <td>0.7</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>119</td>\n",
+       "      <td>79</td>\n",
+       "      <td>...</td>\n",
+       "      <td>71</td>\n",
+       "      <td>114</td>\n",
+       "      <td>15.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.1</td>\n",
+       "      <td>19</td>\n",
+       "      <td>25</td>\n",
+       "      <td>30</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>45</td>\n",
+       "      <td>155</td>\n",
+       "      <td>65</td>\n",
+       "      <td>86.0</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>110</td>\n",
+       "      <td>80</td>\n",
+       "      <td>...</td>\n",
+       "      <td>57</td>\n",
+       "      <td>112</td>\n",
+       "      <td>13.7</td>\n",
+       "      <td>3</td>\n",
+       "      <td>0.6</td>\n",
+       "      <td>1090</td>\n",
+       "      <td>1400</td>\n",
+       "      <td>276</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>45</td>\n",
+       "      <td>165</td>\n",
+       "      <td>80</td>\n",
+       "      <td>94.0</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>0.7</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>158</td>\n",
+       "      <td>88</td>\n",
+       "      <td>...</td>\n",
+       "      <td>46</td>\n",
+       "      <td>91</td>\n",
+       "      <td>16.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>32</td>\n",
+       "      <td>36</td>\n",
+       "      <td>36</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>20</td>\n",
+       "      <td>165</td>\n",
+       "      <td>60</td>\n",
+       "      <td>81.0</td>\n",
+       "      <td>1.5</td>\n",
+       "      <td>0.1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>109</td>\n",
+       "      <td>64</td>\n",
+       "      <td>...</td>\n",
+       "      <td>47</td>\n",
+       "      <td>92</td>\n",
+       "      <td>14.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.2</td>\n",
+       "      <td>26</td>\n",
+       "      <td>28</td>\n",
+       "      <td>15</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>38979</th>\n",
+       "      <td>40</td>\n",
+       "      <td>165</td>\n",
+       "      <td>60</td>\n",
+       "      <td>80.0</td>\n",
+       "      <td>0.4</td>\n",
+       "      <td>0.6</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>107</td>\n",
+       "      <td>60</td>\n",
+       "      <td>...</td>\n",
+       "      <td>61</td>\n",
+       "      <td>72</td>\n",
+       "      <td>12.3</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.5</td>\n",
+       "      <td>18</td>\n",
+       "      <td>18</td>\n",
+       "      <td>21</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>38980</th>\n",
+       "      <td>45</td>\n",
+       "      <td>155</td>\n",
+       "      <td>55</td>\n",
+       "      <td>75.0</td>\n",
+       "      <td>1.5</td>\n",
+       "      <td>1.2</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>126</td>\n",
+       "      <td>72</td>\n",
+       "      <td>...</td>\n",
+       "      <td>76</td>\n",
+       "      <td>131</td>\n",
+       "      <td>12.5</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0.6</td>\n",
+       "      <td>23</td>\n",
+       "      <td>11</td>\n",
+       "      <td>12</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>38981</th>\n",
+       "      <td>40</td>\n",
+       "      <td>170</td>\n",
+       "      <td>105</td>\n",
+       "      <td>124.0</td>\n",
+       "      <td>0.6</td>\n",
+       "      <td>0.5</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>141</td>\n",
+       "      <td>85</td>\n",
+       "      <td>...</td>\n",
+       "      <td>48</td>\n",
+       "      <td>138</td>\n",
+       "      <td>17.1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>24</td>\n",
+       "      <td>23</td>\n",
+       "      <td>35</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>38982</th>\n",
+       "      <td>40</td>\n",
+       "      <td>160</td>\n",
+       "      <td>55</td>\n",
+       "      <td>75.0</td>\n",
+       "      <td>1.5</td>\n",
+       "      <td>1.5</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>95</td>\n",
+       "      <td>69</td>\n",
+       "      <td>...</td>\n",
+       "      <td>79</td>\n",
+       "      <td>116</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.6</td>\n",
+       "      <td>24</td>\n",
+       "      <td>20</td>\n",
+       "      <td>17</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>38983</th>\n",
+       "      <td>55</td>\n",
+       "      <td>175</td>\n",
+       "      <td>60</td>\n",
+       "      <td>81.1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>114</td>\n",
+       "      <td>66</td>\n",
+       "      <td>...</td>\n",
+       "      <td>64</td>\n",
+       "      <td>137</td>\n",
+       "      <td>13.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>18</td>\n",
+       "      <td>12</td>\n",
+       "      <td>16</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>38984 rows × 23 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       age  height(cm)  weight(kg)  waist(cm)  eyesight(left)  \\\n",
+       "0       35         170          85       97.0             0.9   \n",
+       "1       20         175         110      110.0             0.7   \n",
+       "2       45         155          65       86.0             0.9   \n",
+       "3       45         165          80       94.0             0.8   \n",
+       "4       20         165          60       81.0             1.5   \n",
+       "...    ...         ...         ...        ...             ...   \n",
+       "38979   40         165          60       80.0             0.4   \n",
+       "38980   45         155          55       75.0             1.5   \n",
+       "38981   40         170         105      124.0             0.6   \n",
+       "38982   40         160          55       75.0             1.5   \n",
+       "38983   55         175          60       81.1             1.0   \n",
+       "\n",
+       "       eyesight(right)  hearing(left)  hearing(right)  systolic  relaxation  \\\n",
+       "0                  0.9              1               1       118          78   \n",
+       "1                  0.9              1               1       119          79   \n",
+       "2                  0.9              1               1       110          80   \n",
+       "3                  0.7              1               1       158          88   \n",
+       "4                  0.1              1               1       109          64   \n",
+       "...                ...            ...             ...       ...         ...   \n",
+       "38979              0.6              1               1       107          60   \n",
+       "38980              1.2              1               1       126          72   \n",
+       "38981              0.5              1               1       141          85   \n",
+       "38982              1.5              1               1        95          69   \n",
+       "38983              1.0              1               1       114          66   \n",
+       "\n",
+       "       ...  HDL  LDL  hemoglobin  Urine protein  serum creatinine   AST   ALT  \\\n",
+       "0      ...   70  142        19.8              1               1.0    61   115   \n",
+       "1      ...   71  114        15.9              1               1.1    19    25   \n",
+       "2      ...   57  112        13.7              3               0.6  1090  1400   \n",
+       "3      ...   46   91        16.9              1               0.9    32    36   \n",
+       "4      ...   47   92        14.9              1               1.2    26    28   \n",
+       "...    ...  ...  ...         ...            ...               ...   ...   ...   \n",
+       "38979  ...   61   72        12.3              1               0.5    18    18   \n",
+       "38980  ...   76  131        12.5              2               0.6    23    11   \n",
+       "38981  ...   48  138        17.1              1               0.8    24    23   \n",
+       "38982  ...   79  116        12.0              1               0.6    24    20   \n",
+       "38983  ...   64  137        13.9              1               1.0    18    12   \n",
+       "\n",
+       "       Gtp  dental caries  smoking  \n",
+       "0      125              1        1  \n",
+       "1       30              1        0  \n",
+       "2      276              0        0  \n",
+       "3       36              0        0  \n",
+       "4       15              0        0  \n",
+       "...    ...            ...      ...  \n",
+       "38979   21              1        0  \n",
+       "38980   12              0        0  \n",
+       "38981   35              1        1  \n",
+       "38982   17              0        1  \n",
+       "38983   16              0        1  \n",
+       "\n",
+       "[38984 rows x 23 columns]"
+      ]
+     },
+     "execution_count": 157,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Reading dataset using pandas library\n",
+    "train_data = pd.read_csv(\"E:/MA336 - AI and ML/Project_AI/train_dataset.csv\")\n",
+    "train_data"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8e1cd0b9",
+   "metadata": {},
+   "source": [
+    "It can be seen there are 22 predictor variables with 38984 observations.Last column smoking is a response variable.\n",
+    ">Description of Each variables in dataset.\n",
+    "> - ***age:***             Age of the individual in 5-year intervals.\n",
+    "> - ***height(cm):***      Height of the individual in centimeters.\n",
+    "> - ***weight(kg):***      Weight of the individual in kilograms.\n",
+    "> - ***waist(cm):***       Waist circumference length of the individual.\n",
+    "> - ***eyesight(left):***  Eyesight measurement of the left eye.\n",
+    "> - ***eyesight(right):*** Eyesight measurement of the right eye.\n",
+    "> - ***hearing(left):***   Hearing measurement of the left ear.\n",
+    "> - ***hearing(right):***  Hearing measurement of the right ear.\n",
+    "> - ***systolic:***        Systolic blood pressure measurement.\n",
+    "> - ***relaxation:***      Diastolic blood pressure measurement.\n",
+    "> - ***fasting blood sugar:***  Fasting blood sugar level.\n",
+    "> - ***Cholesterol:***     Total cholesterol level.\n",
+    "> - ***triglyceride:***    Triglyceride level.\n",
+    "> - ***HDL:***             High-density lipoprotein (HDL) cholesterol level.\n",
+    "> - ***LDL:***             Low-density lipoprotein (LDL) cholesterol level.\n",
+    "> - ***hemoglobin:***      Hemoglobin level in the blood.\n",
+    "> - ***Urine protein:***   Presence of protein in the urine.\n",
+    "> - ***serum creatinine:***  Serum creatinine level.\n",
+    "> - ***AST:***             Level of glutamic oxaloacetic transaminase (AST) enzyme.\n",
+    "> - ***ALT:***             Level of glutamic oxaloacetic transaminase (ALT) enzyme.\n",
+    "> - ***Gtp:***             Level of γ-GTP enzyme.\n",
+    "> - ***dental caries:***   Presence of dental caries (tooth decay).\n",
+    "> - ***smoking:***         Smoking status of the individual (1 for smoker, 0 for non-smoker).\n",
+    "\n",
+    "Aim is to find the status of smoking through bio-signals.\n",
+    "\n",
+    "Target variable:\n",
+    "\n",
+    "When ***smoking is yes*** = 1 , When ***no smoking*** = 0\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 158,
+   "id": "d326b597",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas.core.frame.DataFrame'>\n",
+      "RangeIndex: 38984 entries, 0 to 38983\n",
+      "Data columns (total 23 columns):\n",
+      " #   Column               Non-Null Count  Dtype  \n",
+      "---  ------               --------------  -----  \n",
+      " 0   age                  38984 non-null  int64  \n",
+      " 1   height(cm)           38984 non-null  int64  \n",
+      " 2   weight(kg)           38984 non-null  int64  \n",
+      " 3   waist(cm)            38984 non-null  float64\n",
+      " 4   eyesight(left)       38984 non-null  float64\n",
+      " 5   eyesight(right)      38984 non-null  float64\n",
+      " 6   hearing(left)        38984 non-null  int64  \n",
+      " 7   hearing(right)       38984 non-null  int64  \n",
+      " 8   systolic             38984 non-null  int64  \n",
+      " 9   relaxation           38984 non-null  int64  \n",
+      " 10  fasting blood sugar  38984 non-null  int64  \n",
+      " 11  Cholesterol          38984 non-null  int64  \n",
+      " 12  triglyceride         38984 non-null  int64  \n",
+      " 13  HDL                  38984 non-null  int64  \n",
+      " 14  LDL                  38984 non-null  int64  \n",
+      " 15  hemoglobin           38984 non-null  float64\n",
+      " 16  Urine protein        38984 non-null  int64  \n",
+      " 17  serum creatinine     38984 non-null  float64\n",
+      " 18  AST                  38984 non-null  int64  \n",
+      " 19  ALT                  38984 non-null  int64  \n",
+      " 20  Gtp                  38984 non-null  int64  \n",
+      " 21  dental caries        38984 non-null  int64  \n",
+      " 22  smoking              38984 non-null  int64  \n",
+      "dtypes: float64(5), int64(18)\n",
+      "memory usage: 6.8 MB\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Basic Dataset information\n",
+    "train_data.info()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 159,
+   "id": "b21183eb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Let create a copy of data set so that original data set is never altered.\n",
+    "train_copy = train_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 160,
+   "id": "f1ecf4d2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Column 'age' has 0 missing values.\n",
+      "Column 'height(cm)' has 0 missing values.\n",
+      "Column 'weight(kg)' has 0 missing values.\n",
+      "Column 'waist(cm)' has 0 missing values.\n",
+      "Column 'eyesight(left)' has 0 missing values.\n",
+      "Column 'eyesight(right)' has 0 missing values.\n",
+      "Column 'hearing(left)' has 0 missing values.\n",
+      "Column 'hearing(right)' has 0 missing values.\n",
+      "Column 'systolic' has 0 missing values.\n",
+      "Column 'relaxation' has 0 missing values.\n",
+      "Column 'fasting blood sugar' has 0 missing values.\n",
+      "Column 'Cholesterol' has 0 missing values.\n",
+      "Column 'triglyceride' has 0 missing values.\n",
+      "Column 'HDL' has 0 missing values.\n",
+      "Column 'LDL' has 0 missing values.\n",
+      "Column 'hemoglobin' has 0 missing values.\n",
+      "Column 'Urine protein' has 0 missing values.\n",
+      "Column 'serum creatinine' has 0 missing values.\n",
+      "Column 'AST' has 0 missing values.\n",
+      "Column 'ALT' has 0 missing values.\n",
+      "Column 'Gtp' has 0 missing values.\n",
+      "Column 'dental caries' has 0 missing values.\n",
+      "Column 'smoking' has 0 missing values.\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Check for missing values in each column\n",
+    "missing_values = train_data.isnull().sum()\n",
+    "\n",
+    "# Iterate over columns and print the number of missing values\n",
+    "for column, count in missing_values.iteritems():\n",
+    "    print(f\"Column '{column}' has {count} missing values.\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5dad71cb",
+   "metadata": {},
+   "source": [
+    "Hence the dataset has no missing values in it."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fd2ffc55",
+   "metadata": {},
+   "source": [
+    "#### Removing outliers using IQR Method\n",
+    "The IQR method is a robust approach that can be effective in identifying and removing outliers."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 161,
+   "id": "07ed30bc",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[IQR SELECTED] Number of rows in training data before removing outliers: 38984\n",
+      "[IQR SELECTED] Number of rows in training data after removing outliers: 36528\n"
+     ]
+    }
+   ],
+   "source": [
+    "from scipy.stats import iqr\n",
+    "\n",
+    "# Print the number of rows in the training data before removing outliers\n",
+    "print(f\"[IQR SELECTED] Number of rows in training data before removing outliers: {len(train_copy)}\")\n",
+    "\n",
+    "# Identify and remove outliers from the training set using IQR method\n",
+    "Q1 = train_copy.quantile(0.25)\n",
+    "Q3 = train_copy.quantile(0.75)\n",
+    "IQR = iqr(train_copy)\n",
+    "lower_bound = Q1 - 1.5 * IQR\n",
+    "upper_bound = Q3 + 1.5 * IQR\n",
+    "train_copy = train_copy[(train_copy >= lower_bound) & (train_copy <= upper_bound)].dropna()\n",
+    "\n",
+    "\n",
+    "# Print the number of rows in the training data after removing outliers\n",
+    "print(f\"[IQR SELECTED] Number of rows in training data after removing outliers: {len(train_copy)}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0d77d4b5",
+   "metadata": {},
+   "source": [
+    "#### Removing Duplicates"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 162,
+   "id": "fa0e984f",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Number of duplicates: 5159\n",
+      "Number of observations after removing duplicates: 31369\n"
+     ]
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "# Assuming train_copy is your DataFrame containing the dataset\n",
+    "# Check for duplicates\n",
+    "duplicates = train_copy.duplicated()\n",
+    "\n",
+    "# Count the number of duplicates\n",
+    "num_duplicates = duplicates.sum()\n",
+    "print(\"Number of duplicates:\", num_duplicates)\n",
+    "\n",
+    "# Remove duplicates from the DataFrame\n",
+    "train_copy = train_copy.drop_duplicates()\n",
+    "\n",
+    "# Check the number of observations after removing duplicates\n",
+    "num_observations = len(train_copy)\n",
+    "print(\"Number of observations after removing duplicates:\", num_observations)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "595badb7",
+   "metadata": {},
+   "source": [
+    "### 3. DATA EXPLORATION"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 163,
+   "id": "223b36a2",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<style type=\"text/css\">\n",
+       "#T_ca6ab_row0_col0, #T_ca6ab_row0_col1, #T_ca6ab_row0_col2, #T_ca6ab_row0_col3, #T_ca6ab_row0_col4, #T_ca6ab_row0_col5, #T_ca6ab_row0_col6, #T_ca6ab_row0_col7, #T_ca6ab_row1_col0, #T_ca6ab_row1_col1, #T_ca6ab_row1_col2, #T_ca6ab_row1_col3, #T_ca6ab_row1_col4, #T_ca6ab_row1_col5, #T_ca6ab_row1_col6, #T_ca6ab_row1_col7, #T_ca6ab_row2_col0, #T_ca6ab_row2_col1, #T_ca6ab_row2_col2, #T_ca6ab_row2_col3, #T_ca6ab_row2_col4, #T_ca6ab_row2_col5, #T_ca6ab_row2_col6, #T_ca6ab_row2_col7, #T_ca6ab_row3_col0, #T_ca6ab_row3_col1, #T_ca6ab_row3_col2, #T_ca6ab_row3_col3, #T_ca6ab_row3_col4, #T_ca6ab_row3_col5, #T_ca6ab_row3_col6, #T_ca6ab_row3_col7, #T_ca6ab_row4_col0, #T_ca6ab_row4_col1, #T_ca6ab_row4_col2, #T_ca6ab_row4_col3, #T_ca6ab_row4_col4, #T_ca6ab_row4_col5, #T_ca6ab_row4_col6, #T_ca6ab_row4_col7, #T_ca6ab_row5_col0, #T_ca6ab_row5_col1, #T_ca6ab_row5_col2, #T_ca6ab_row5_col3, #T_ca6ab_row5_col4, #T_ca6ab_row5_col5, #T_ca6ab_row5_col6, #T_ca6ab_row5_col7, #T_ca6ab_row6_col0, #T_ca6ab_row6_col1, #T_ca6ab_row6_col2, #T_ca6ab_row6_col3, #T_ca6ab_row6_col4, #T_ca6ab_row6_col5, #T_ca6ab_row6_col6, #T_ca6ab_row6_col7, #T_ca6ab_row7_col0, #T_ca6ab_row7_col1, #T_ca6ab_row7_col2, #T_ca6ab_row7_col3, #T_ca6ab_row7_col4, #T_ca6ab_row7_col5, #T_ca6ab_row7_col6, #T_ca6ab_row7_col7, #T_ca6ab_row8_col0, #T_ca6ab_row8_col1, #T_ca6ab_row8_col2, #T_ca6ab_row8_col3, #T_ca6ab_row8_col4, #T_ca6ab_row8_col5, #T_ca6ab_row8_col6, #T_ca6ab_row8_col7, #T_ca6ab_row9_col0, #T_ca6ab_row9_col1, #T_ca6ab_row9_col2, #T_ca6ab_row9_col3, #T_ca6ab_row9_col4, #T_ca6ab_row9_col5, #T_ca6ab_row9_col6, #T_ca6ab_row9_col7, #T_ca6ab_row10_col0, #T_ca6ab_row10_col1, #T_ca6ab_row10_col2, #T_ca6ab_row10_col3, #T_ca6ab_row10_col4, #T_ca6ab_row10_col5, #T_ca6ab_row10_col6, #T_ca6ab_row10_col7, #T_ca6ab_row11_col0, #T_ca6ab_row11_col1, #T_ca6ab_row11_col2, #T_ca6ab_row11_col3, #T_ca6ab_row11_col4, #T_ca6ab_row11_col5, #T_ca6ab_row11_col6, #T_ca6ab_row11_col7, #T_ca6ab_row12_col0, #T_ca6ab_row12_col1, #T_ca6ab_row12_col2, #T_ca6ab_row12_col3, #T_ca6ab_row12_col4, #T_ca6ab_row12_col5, #T_ca6ab_row12_col6, #T_ca6ab_row12_col7, #T_ca6ab_row13_col0, #T_ca6ab_row13_col1, #T_ca6ab_row13_col2, #T_ca6ab_row13_col3, #T_ca6ab_row13_col4, #T_ca6ab_row13_col5, #T_ca6ab_row13_col6, #T_ca6ab_row13_col7, #T_ca6ab_row14_col0, #T_ca6ab_row14_col1, #T_ca6ab_row14_col2, #T_ca6ab_row14_col3, #T_ca6ab_row14_col4, #T_ca6ab_row14_col5, #T_ca6ab_row14_col6, #T_ca6ab_row14_col7, #T_ca6ab_row15_col0, #T_ca6ab_row15_col1, #T_ca6ab_row15_col2, #T_ca6ab_row15_col3, #T_ca6ab_row15_col4, #T_ca6ab_row15_col5, #T_ca6ab_row15_col6, #T_ca6ab_row15_col7, #T_ca6ab_row16_col0, #T_ca6ab_row16_col1, #T_ca6ab_row16_col2, #T_ca6ab_row16_col3, #T_ca6ab_row16_col4, #T_ca6ab_row16_col5, #T_ca6ab_row16_col6, #T_ca6ab_row16_col7, #T_ca6ab_row17_col0, #T_ca6ab_row17_col1, #T_ca6ab_row17_col2, #T_ca6ab_row17_col3, #T_ca6ab_row17_col4, #T_ca6ab_row17_col5, #T_ca6ab_row17_col6, #T_ca6ab_row17_col7, #T_ca6ab_row18_col0, #T_ca6ab_row18_col1, #T_ca6ab_row18_col2, #T_ca6ab_row18_col3, #T_ca6ab_row18_col4, #T_ca6ab_row18_col5, #T_ca6ab_row18_col6, #T_ca6ab_row18_col7, #T_ca6ab_row19_col0, #T_ca6ab_row19_col1, #T_ca6ab_row19_col2, #T_ca6ab_row19_col3, #T_ca6ab_row19_col4, #T_ca6ab_row19_col5, #T_ca6ab_row19_col6, #T_ca6ab_row19_col7, #T_ca6ab_row20_col0, #T_ca6ab_row20_col1, #T_ca6ab_row20_col2, #T_ca6ab_row20_col3, #T_ca6ab_row20_col4, #T_ca6ab_row20_col5, #T_ca6ab_row20_col6, #T_ca6ab_row20_col7, #T_ca6ab_row21_col0, #T_ca6ab_row21_col1, #T_ca6ab_row21_col2, #T_ca6ab_row21_col3, #T_ca6ab_row21_col4, #T_ca6ab_row21_col5, #T_ca6ab_row21_col6, #T_ca6ab_row21_col7, #T_ca6ab_row22_col0, #T_ca6ab_row22_col1, #T_ca6ab_row22_col2, #T_ca6ab_row22_col3, #T_ca6ab_row22_col4, #T_ca6ab_row22_col5, #T_ca6ab_row22_col6, #T_ca6ab_row22_col7 {\n",
+       "  font-size: 15px;\n",
+       "  color: #000000;\n",
+       "  border: 1.5px solid black;\n",
+       "}\n",
+       "</style>\n",
+       "<table id=\"T_ca6ab\">\n",
+       "  <thead>\n",
+       "    <tr>\n",
+       "      <th class=\"blank level0\" >&nbsp;</th>\n",
+       "      <th id=\"T_ca6ab_level0_col0\" class=\"col_heading level0 col0\" >count</th>\n",
+       "      <th id=\"T_ca6ab_level0_col1\" class=\"col_heading level0 col1\" >mean</th>\n",
+       "      <th id=\"T_ca6ab_level0_col2\" class=\"col_heading level0 col2\" >std</th>\n",
+       "      <th id=\"T_ca6ab_level0_col3\" class=\"col_heading level0 col3\" >min</th>\n",
+       "      <th id=\"T_ca6ab_level0_col4\" class=\"col_heading level0 col4\" >25%</th>\n",
+       "      <th id=\"T_ca6ab_level0_col5\" class=\"col_heading level0 col5\" >50%</th>\n",
+       "      <th id=\"T_ca6ab_level0_col6\" class=\"col_heading level0 col6\" >75%</th>\n",
+       "      <th id=\"T_ca6ab_level0_col7\" class=\"col_heading level0 col7\" >max</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row0\" class=\"row_heading level0 row0\" >age</th>\n",
+       "      <td id=\"T_ca6ab_row0_col0\" class=\"data row0 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row0_col1\" class=\"data row0 col1\" >44.160318</td>\n",
+       "      <td id=\"T_ca6ab_row0_col2\" class=\"data row0 col2\" >12.149882</td>\n",
+       "      <td id=\"T_ca6ab_row0_col3\" class=\"data row0 col3\" >20.000000</td>\n",
+       "      <td id=\"T_ca6ab_row0_col4\" class=\"data row0 col4\" >40.000000</td>\n",
+       "      <td id=\"T_ca6ab_row0_col5\" class=\"data row0 col5\" >40.000000</td>\n",
+       "      <td id=\"T_ca6ab_row0_col6\" class=\"data row0 col6\" >55.000000</td>\n",
+       "      <td id=\"T_ca6ab_row0_col7\" class=\"data row0 col7\" >85.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row1\" class=\"row_heading level0 row1\" >height(cm)</th>\n",
+       "      <td id=\"T_ca6ab_row1_col0\" class=\"data row1 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row1_col1\" class=\"data row1 col1\" >164.505563</td>\n",
+       "      <td id=\"T_ca6ab_row1_col2\" class=\"data row1 col2\" >9.236820</td>\n",
+       "      <td id=\"T_ca6ab_row1_col3\" class=\"data row1 col3\" >130.000000</td>\n",
+       "      <td id=\"T_ca6ab_row1_col4\" class=\"data row1 col4\" >160.000000</td>\n",
+       "      <td id=\"T_ca6ab_row1_col5\" class=\"data row1 col5\" >165.000000</td>\n",
+       "      <td id=\"T_ca6ab_row1_col6\" class=\"data row1 col6\" >170.000000</td>\n",
+       "      <td id=\"T_ca6ab_row1_col7\" class=\"data row1 col7\" >190.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row2\" class=\"row_heading level0 row2\" >weight(kg)</th>\n",
+       "      <td id=\"T_ca6ab_row2_col0\" class=\"data row2 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row2_col1\" class=\"data row2 col1\" >65.471166</td>\n",
+       "      <td id=\"T_ca6ab_row2_col2\" class=\"data row2 col2\" >12.738713</td>\n",
+       "      <td id=\"T_ca6ab_row2_col3\" class=\"data row2 col3\" >30.000000</td>\n",
+       "      <td id=\"T_ca6ab_row2_col4\" class=\"data row2 col4\" >55.000000</td>\n",
+       "      <td id=\"T_ca6ab_row2_col5\" class=\"data row2 col5\" >65.000000</td>\n",
+       "      <td id=\"T_ca6ab_row2_col6\" class=\"data row2 col6\" >75.000000</td>\n",
+       "      <td id=\"T_ca6ab_row2_col7\" class=\"data row2 col7\" >135.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row3\" class=\"row_heading level0 row3\" >waist(cm)</th>\n",
+       "      <td id=\"T_ca6ab_row3_col0\" class=\"data row3 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row3_col1\" class=\"data row3 col1\" >81.705155</td>\n",
+       "      <td id=\"T_ca6ab_row3_col2\" class=\"data row3 col2\" >9.233237</td>\n",
+       "      <td id=\"T_ca6ab_row3_col3\" class=\"data row3 col3\" >54.000000</td>\n",
+       "      <td id=\"T_ca6ab_row3_col4\" class=\"data row3 col4\" >75.000000</td>\n",
+       "      <td id=\"T_ca6ab_row3_col5\" class=\"data row3 col5\" >81.300000</td>\n",
+       "      <td id=\"T_ca6ab_row3_col6\" class=\"data row3 col6\" >87.800000</td>\n",
+       "      <td id=\"T_ca6ab_row3_col7\" class=\"data row3 col7\" >129.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row4\" class=\"row_heading level0 row4\" >eyesight(left)</th>\n",
+       "      <td id=\"T_ca6ab_row4_col0\" class=\"data row4 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row4_col1\" class=\"data row4 col1\" >1.014855</td>\n",
+       "      <td id=\"T_ca6ab_row4_col2\" class=\"data row4 col2\" >0.503095</td>\n",
+       "      <td id=\"T_ca6ab_row4_col3\" class=\"data row4 col3\" >0.100000</td>\n",
+       "      <td id=\"T_ca6ab_row4_col4\" class=\"data row4 col4\" >0.800000</td>\n",
+       "      <td id=\"T_ca6ab_row4_col5\" class=\"data row4 col5\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row4_col6\" class=\"data row4 col6\" >1.200000</td>\n",
+       "      <td id=\"T_ca6ab_row4_col7\" class=\"data row4 col7\" >9.900000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row5\" class=\"row_heading level0 row5\" >eyesight(right)</th>\n",
+       "      <td id=\"T_ca6ab_row5_col0\" class=\"data row5 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row5_col1\" class=\"data row5 col1\" >1.008850</td>\n",
+       "      <td id=\"T_ca6ab_row5_col2\" class=\"data row5 col2\" >0.494946</td>\n",
+       "      <td id=\"T_ca6ab_row5_col3\" class=\"data row5 col3\" >0.100000</td>\n",
+       "      <td id=\"T_ca6ab_row5_col4\" class=\"data row5 col4\" >0.800000</td>\n",
+       "      <td id=\"T_ca6ab_row5_col5\" class=\"data row5 col5\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row5_col6\" class=\"data row5 col6\" >1.200000</td>\n",
+       "      <td id=\"T_ca6ab_row5_col7\" class=\"data row5 col7\" >9.900000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row6\" class=\"row_heading level0 row6\" >hearing(left)</th>\n",
+       "      <td id=\"T_ca6ab_row6_col0\" class=\"data row6 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row6_col1\" class=\"data row6 col1\" >1.024897</td>\n",
+       "      <td id=\"T_ca6ab_row6_col2\" class=\"data row6 col2\" >0.155814</td>\n",
+       "      <td id=\"T_ca6ab_row6_col3\" class=\"data row6 col3\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row6_col4\" class=\"data row6 col4\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row6_col5\" class=\"data row6 col5\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row6_col6\" class=\"data row6 col6\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row6_col7\" class=\"data row6 col7\" >2.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row7\" class=\"row_heading level0 row7\" >hearing(right)</th>\n",
+       "      <td id=\"T_ca6ab_row7_col0\" class=\"data row7 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row7_col1\" class=\"data row7 col1\" >1.025758</td>\n",
+       "      <td id=\"T_ca6ab_row7_col2\" class=\"data row7 col2\" >0.158415</td>\n",
+       "      <td id=\"T_ca6ab_row7_col3\" class=\"data row7 col3\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row7_col4\" class=\"data row7 col4\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row7_col5\" class=\"data row7 col5\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row7_col6\" class=\"data row7 col6\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row7_col7\" class=\"data row7 col7\" >2.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row8\" class=\"row_heading level0 row8\" >systolic</th>\n",
+       "      <td id=\"T_ca6ab_row8_col0\" class=\"data row8 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row8_col1\" class=\"data row8 col1\" >121.136664</td>\n",
+       "      <td id=\"T_ca6ab_row8_col2\" class=\"data row8 col2\" >13.569889</td>\n",
+       "      <td id=\"T_ca6ab_row8_col3\" class=\"data row8 col3\" >71.000000</td>\n",
+       "      <td id=\"T_ca6ab_row8_col4\" class=\"data row8 col4\" >111.000000</td>\n",
+       "      <td id=\"T_ca6ab_row8_col5\" class=\"data row8 col5\" >120.000000</td>\n",
+       "      <td id=\"T_ca6ab_row8_col6\" class=\"data row8 col6\" >130.000000</td>\n",
+       "      <td id=\"T_ca6ab_row8_col7\" class=\"data row8 col7\" >223.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row9\" class=\"row_heading level0 row9\" >relaxation</th>\n",
+       "      <td id=\"T_ca6ab_row9_col0\" class=\"data row9 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row9_col1\" class=\"data row9 col1\" >75.743218</td>\n",
+       "      <td id=\"T_ca6ab_row9_col2\" class=\"data row9 col2\" >9.593874</td>\n",
+       "      <td id=\"T_ca6ab_row9_col3\" class=\"data row9 col3\" >40.000000</td>\n",
+       "      <td id=\"T_ca6ab_row9_col4\" class=\"data row9 col4\" >70.000000</td>\n",
+       "      <td id=\"T_ca6ab_row9_col5\" class=\"data row9 col5\" >76.000000</td>\n",
+       "      <td id=\"T_ca6ab_row9_col6\" class=\"data row9 col6\" >81.000000</td>\n",
+       "      <td id=\"T_ca6ab_row9_col7\" class=\"data row9 col7\" >146.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row10\" class=\"row_heading level0 row10\" >fasting blood sugar</th>\n",
+       "      <td id=\"T_ca6ab_row10_col0\" class=\"data row10 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row10_col1\" class=\"data row10 col1\" >98.004654</td>\n",
+       "      <td id=\"T_ca6ab_row10_col2\" class=\"data row10 col2\" >16.464372</td>\n",
+       "      <td id=\"T_ca6ab_row10_col3\" class=\"data row10 col3\" >46.000000</td>\n",
+       "      <td id=\"T_ca6ab_row10_col4\" class=\"data row10 col4\" >89.000000</td>\n",
+       "      <td id=\"T_ca6ab_row10_col5\" class=\"data row10 col5\" >95.000000</td>\n",
+       "      <td id=\"T_ca6ab_row10_col6\" class=\"data row10 col6\" >103.000000</td>\n",
+       "      <td id=\"T_ca6ab_row10_col7\" class=\"data row10 col7\" >234.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row11\" class=\"row_heading level0 row11\" >Cholesterol</th>\n",
+       "      <td id=\"T_ca6ab_row11_col0\" class=\"data row11 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row11_col1\" class=\"data row11 col1\" >195.880806</td>\n",
+       "      <td id=\"T_ca6ab_row11_col2\" class=\"data row11 col2\" >35.519681</td>\n",
+       "      <td id=\"T_ca6ab_row11_col3\" class=\"data row11 col3\" >55.000000</td>\n",
+       "      <td id=\"T_ca6ab_row11_col4\" class=\"data row11 col4\" >171.000000</td>\n",
+       "      <td id=\"T_ca6ab_row11_col5\" class=\"data row11 col5\" >194.000000</td>\n",
+       "      <td id=\"T_ca6ab_row11_col6\" class=\"data row11 col6\" >218.000000</td>\n",
+       "      <td id=\"T_ca6ab_row11_col7\" class=\"data row11 col7\" >348.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row12\" class=\"row_heading level0 row12\" >triglyceride</th>\n",
+       "      <td id=\"T_ca6ab_row12_col0\" class=\"data row12 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row12_col1\" class=\"data row12 col1\" >117.028914</td>\n",
+       "      <td id=\"T_ca6ab_row12_col2\" class=\"data row12 col2\" >57.650741</td>\n",
+       "      <td id=\"T_ca6ab_row12_col3\" class=\"data row12 col3\" >8.000000</td>\n",
+       "      <td id=\"T_ca6ab_row12_col4\" class=\"data row12 col4\" >73.000000</td>\n",
+       "      <td id=\"T_ca6ab_row12_col5\" class=\"data row12 col5\" >104.000000</td>\n",
+       "      <td id=\"T_ca6ab_row12_col6\" class=\"data row12 col6\" >150.000000</td>\n",
+       "      <td id=\"T_ca6ab_row12_col7\" class=\"data row12 col7\" >290.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row13\" class=\"row_heading level0 row13\" >HDL</th>\n",
+       "      <td id=\"T_ca6ab_row13_col0\" class=\"data row13 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row13_col1\" class=\"data row13 col1\" >57.722560</td>\n",
+       "      <td id=\"T_ca6ab_row13_col2\" class=\"data row13 col2\" >14.412297</td>\n",
+       "      <td id=\"T_ca6ab_row13_col3\" class=\"data row13 col3\" >4.000000</td>\n",
+       "      <td id=\"T_ca6ab_row13_col4\" class=\"data row13 col4\" >47.000000</td>\n",
+       "      <td id=\"T_ca6ab_row13_col5\" class=\"data row13 col5\" >56.000000</td>\n",
+       "      <td id=\"T_ca6ab_row13_col6\" class=\"data row13 col6\" >66.000000</td>\n",
+       "      <td id=\"T_ca6ab_row13_col7\" class=\"data row13 col7\" >157.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row14\" class=\"row_heading level0 row14\" >LDL</th>\n",
+       "      <td id=\"T_ca6ab_row14_col0\" class=\"data row14 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row14_col1\" class=\"data row14 col1\" >114.941280</td>\n",
+       "      <td id=\"T_ca6ab_row14_col2\" class=\"data row14 col2\" >32.910135</td>\n",
+       "      <td id=\"T_ca6ab_row14_col3\" class=\"data row14 col3\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row14_col4\" class=\"data row14 col4\" >92.000000</td>\n",
+       "      <td id=\"T_ca6ab_row14_col5\" class=\"data row14 col5\" >113.000000</td>\n",
+       "      <td id=\"T_ca6ab_row14_col6\" class=\"data row14 col6\" >136.000000</td>\n",
+       "      <td id=\"T_ca6ab_row14_col7\" class=\"data row14 col7\" >265.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row15\" class=\"row_heading level0 row15\" >hemoglobin</th>\n",
+       "      <td id=\"T_ca6ab_row15_col0\" class=\"data row15 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row15_col1\" class=\"data row15 col1\" >14.572973</td>\n",
+       "      <td id=\"T_ca6ab_row15_col2\" class=\"data row15 col2\" >1.563132</td>\n",
+       "      <td id=\"T_ca6ab_row15_col3\" class=\"data row15 col3\" >4.900000</td>\n",
+       "      <td id=\"T_ca6ab_row15_col4\" class=\"data row15 col4\" >13.500000</td>\n",
+       "      <td id=\"T_ca6ab_row15_col5\" class=\"data row15 col5\" >14.700000</td>\n",
+       "      <td id=\"T_ca6ab_row15_col6\" class=\"data row15 col6\" >15.700000</td>\n",
+       "      <td id=\"T_ca6ab_row15_col7\" class=\"data row15 col7\" >21.100000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row16\" class=\"row_heading level0 row16\" >Urine protein</th>\n",
+       "      <td id=\"T_ca6ab_row16_col0\" class=\"data row16 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row16_col1\" class=\"data row16 col1\" >1.080238</td>\n",
+       "      <td id=\"T_ca6ab_row16_col2\" class=\"data row16 col2\" >0.383653</td>\n",
+       "      <td id=\"T_ca6ab_row16_col3\" class=\"data row16 col3\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row16_col4\" class=\"data row16 col4\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row16_col5\" class=\"data row16 col5\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row16_col6\" class=\"data row16 col6\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row16_col7\" class=\"data row16 col7\" >6.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row17\" class=\"row_heading level0 row17\" >serum creatinine</th>\n",
+       "      <td id=\"T_ca6ab_row17_col0\" class=\"data row17 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row17_col1\" class=\"data row17 col1\" >0.883216</td>\n",
+       "      <td id=\"T_ca6ab_row17_col2\" class=\"data row17 col2\" >0.217669</td>\n",
+       "      <td id=\"T_ca6ab_row17_col3\" class=\"data row17 col3\" >0.100000</td>\n",
+       "      <td id=\"T_ca6ab_row17_col4\" class=\"data row17 col4\" >0.700000</td>\n",
+       "      <td id=\"T_ca6ab_row17_col5\" class=\"data row17 col5\" >0.900000</td>\n",
+       "      <td id=\"T_ca6ab_row17_col6\" class=\"data row17 col6\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row17_col7\" class=\"data row17 col7\" >11.600000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row18\" class=\"row_heading level0 row18\" >AST</th>\n",
+       "      <td id=\"T_ca6ab_row18_col0\" class=\"data row18 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row18_col1\" class=\"data row18 col1\" >24.974625</td>\n",
+       "      <td id=\"T_ca6ab_row18_col2\" class=\"data row18 col2\" >10.147233</td>\n",
+       "      <td id=\"T_ca6ab_row18_col3\" class=\"data row18 col3\" >6.000000</td>\n",
+       "      <td id=\"T_ca6ab_row18_col4\" class=\"data row18 col4\" >19.000000</td>\n",
+       "      <td id=\"T_ca6ab_row18_col5\" class=\"data row18 col5\" >23.000000</td>\n",
+       "      <td id=\"T_ca6ab_row18_col6\" class=\"data row18 col6\" >28.000000</td>\n",
+       "      <td id=\"T_ca6ab_row18_col7\" class=\"data row18 col7\" >156.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row19\" class=\"row_heading level0 row19\" >ALT</th>\n",
+       "      <td id=\"T_ca6ab_row19_col0\" class=\"data row19 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row19_col1\" class=\"data row19 col1\" >25.212758</td>\n",
+       "      <td id=\"T_ca6ab_row19_col2\" class=\"data row19 col2\" >16.668417</td>\n",
+       "      <td id=\"T_ca6ab_row19_col3\" class=\"data row19 col3\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row19_col4\" class=\"data row19 col4\" >15.000000</td>\n",
+       "      <td id=\"T_ca6ab_row19_col5\" class=\"data row19 col5\" >20.000000</td>\n",
+       "      <td id=\"T_ca6ab_row19_col6\" class=\"data row19 col6\" >30.000000</td>\n",
+       "      <td id=\"T_ca6ab_row19_col7\" class=\"data row19 col7\" >161.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row20\" class=\"row_heading level0 row20\" >Gtp</th>\n",
+       "      <td id=\"T_ca6ab_row20_col0\" class=\"data row20 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row20_col1\" class=\"data row20 col1\" >33.777519</td>\n",
+       "      <td id=\"T_ca6ab_row20_col2\" class=\"data row20 col2\" >26.767146</td>\n",
+       "      <td id=\"T_ca6ab_row20_col3\" class=\"data row20 col3\" >3.000000</td>\n",
+       "      <td id=\"T_ca6ab_row20_col4\" class=\"data row20 col4\" >17.000000</td>\n",
+       "      <td id=\"T_ca6ab_row20_col5\" class=\"data row20 col5\" >25.000000</td>\n",
+       "      <td id=\"T_ca6ab_row20_col6\" class=\"data row20 col6\" >40.000000</td>\n",
+       "      <td id=\"T_ca6ab_row20_col7\" class=\"data row20 col7\" >174.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row21\" class=\"row_heading level0 row21\" >dental caries</th>\n",
+       "      <td id=\"T_ca6ab_row21_col0\" class=\"data row21 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row21_col1\" class=\"data row21 col1\" >0.212630</td>\n",
+       "      <td id=\"T_ca6ab_row21_col2\" class=\"data row21 col2\" >0.409175</td>\n",
+       "      <td id=\"T_ca6ab_row21_col3\" class=\"data row21 col3\" >0.000000</td>\n",
+       "      <td id=\"T_ca6ab_row21_col4\" class=\"data row21 col4\" >0.000000</td>\n",
+       "      <td id=\"T_ca6ab_row21_col5\" class=\"data row21 col5\" >0.000000</td>\n",
+       "      <td id=\"T_ca6ab_row21_col6\" class=\"data row21 col6\" >0.000000</td>\n",
+       "      <td id=\"T_ca6ab_row21_col7\" class=\"data row21 col7\" >1.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th id=\"T_ca6ab_level0_row22\" class=\"row_heading level0 row22\" >smoking</th>\n",
+       "      <td id=\"T_ca6ab_row22_col0\" class=\"data row22 col0\" >31369.000000</td>\n",
+       "      <td id=\"T_ca6ab_row22_col1\" class=\"data row22 col1\" >0.349262</td>\n",
+       "      <td id=\"T_ca6ab_row22_col2\" class=\"data row22 col2\" >0.476744</td>\n",
+       "      <td id=\"T_ca6ab_row22_col3\" class=\"data row22 col3\" >0.000000</td>\n",
+       "      <td id=\"T_ca6ab_row22_col4\" class=\"data row22 col4\" >0.000000</td>\n",
+       "      <td id=\"T_ca6ab_row22_col5\" class=\"data row22 col5\" >0.000000</td>\n",
+       "      <td id=\"T_ca6ab_row22_col6\" class=\"data row22 col6\" >1.000000</td>\n",
+       "      <td id=\"T_ca6ab_row22_col7\" class=\"data row22 col7\" >1.000000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n"
+      ],
+      "text/plain": [
+       "<pandas.io.formats.style.Styler at 0x21c824944f0>"
+      ]
+     },
+     "execution_count": 163,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#Descriptive statistics of all predictor variables\n",
+    "# Calculate descriptive statistics\n",
+    "descriptive_stats = train_copy.describe()\n",
+    "\n",
+    "# Apply style to the table\n",
+    "styled_table = descriptive_stats.T.style.set_properties(**{\n",
+    "    \"font-size\": \"15px\",\n",
+    "    \"color\": \"#000000\",\n",
+    "    \"border\": \"1.5px solid black\"\n",
+    "})\n",
+    "\n",
+    "# Display the styled table\n",
+    "styled_table"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f2614053",
+   "metadata": {},
+   "source": [
+    "From the descriptive statistics of the predictor variables in the train_data dataset, we can derive several insights and interpretations:\n",
+    "> - Age: The average age of the individuals in the dataset is approximately 44 years, with a standard deviation of 12.06. The minimum and maximum ages are 20 and 85 years, respectively.\n",
+    "> - Height and Weight: The average height is around 164.69 cm, with a standard deviation of 9.19. The average weight is approximately 65.94 kg, with a standard deviation of 12.90.\n",
+    "> - Waist Circumference: The average waist circumference is about 82.06 cm, with a standard deviation of 9.33. The minimum and maximum waist circumferences are 51 cm and 129 cm, respectively.\n",
+    "> - Eyesight and Hearing: The average eyesight values for both the left and right eyes are close to 1, indicating relatively good eyesight. The average hearing values for both the left and right ears are slightly above 1, suggesting normal hearing ability.\n",
+    "> - Blood Pressure: The average systolic blood pressure is 121.48, with a standard deviation of 13.64. The average relaxation (diastolic) blood pressure is 75.99, with a standard deviation of 9.66.\n",
+    "> - Cholesterol Levels: The dataset includes variables such as total cholesterol, triglyceride levels, HDL cholesterol, and LDL cholesterol. The mean values and standard deviations provide insights into the cholesterol distribution within the population.\n",
+    "> - Hemoglobin: The average hemoglobin level is 14.62, with a standard deviation of 1.57. Hemoglobin levels can provide insights into the individual's blood health and oxygen-carrying capacity.\n",
+    "> - Urine Protein and Serum Creatinine: The average urine protein level is approximately 1.09, and the average serum creatinine level is around 0.89. These variables are indicators of kidney function and can be useful in assessing renal health.\n",
+    "> - AST and ALT: AST and ALT are liver enzyme levels. The mean AST value is 26.20, and the mean ALT value is 27.15. These values can provide insights into liver health.\n",
+    "> - Gtp: The average Gtp level is 39.91, with a standard deviation of 49.69. Elevated Gtp levels can indicate liver dysfunction or damage.\n",
+    "> - Dental Caries: The average dental caries value is 0.21, indicating a relatively low prevalence of dental caries in the population.\n",
+    "> - Smoking: The dataset includes a binary variable for smoking, with a mean value of 0.37. This indicates that approximately 37% of the individuals in the dataset are smokers, while the remaining 63% are non-smokers.\n",
+    "These descriptive statistics provide an overview of the distribution and range of each variable in the dataset, allowing us to identify any potential outliers, assess the variability of the data, and gain initial insights into the characteristics of the study population."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 164,
+   "id": "824b6e0d",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi4AAAJDCAYAAAAsIJ9bAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB3M0lEQVR4nO3dd5xcd33v/9eZurO9N6206rLVLMuyJNuyrGJZkhvuEGMDIdz8aDchEMrF9zqEYIMTIJhAgAChEwLENm6SrWZZsmzJkmX1Xrb3Ntumn98fu1pUtozk3anv5+Oxj5VmZ+Z8dmzNec+3fI7hdrtNREREROKAJdoFiIiIiIRLwUVERETihoKLiIiIxA0FFxEREYkbCi4iIiISNxRcREREJG4ouIiIiEjcUHARERGRuKHgIiIiInFDwUXkPG+//TYPP/wwM2fOJD8/nylTprBy5Uq+/OUvR6WeJ598kszMTFpaWoa938c//nFmz54doaouPXZmZubAV35+PvPnz+eJJ57A4/FEpabR9Morr/Dkk09GuwwR6afgItJv/fr1rFq1is7OTr761a/y3HPP8dRTT7F48WKeeeaZaJc3rC984Qv85je/idrxXS4XGzduZOPGjfz2t79lwYIFPPXUU3z84x+PWk2j5dVXX+Ub3/hGtMsQkX62aBcgEiuefvppysvLefbZZ7HZ/vxP44EHHuCf/umfoljZyCZPnhzV41ssFhYuXDjw99tuu43KykqeeeYZnnzySUpLS6/4uU3TxOPx4HK5RqNUEYlzGnER6dfa2kpeXt4FoeUci+XCfyqzZ8/mwQcfZN26dSxZsoTCwkIWLFjAunXrAPjNb37DggULKC4uZtmyZbzzzjuXPOfLL7/MypUrKSoqorS0lPe9733s3LlzxDqPHz/O3LlzWb58OU1NTcDgU0WZmZl87nOf47/+679YsGABRUVF3HjjjQM1nu+ll17ihhtuID8/n7lz5/Lv//7vA9NUV+r6668HoKqqCgC3281jjz3GnDlzyMvLY8aMGXzxi1+ku7t70Lp/+tOfsmDBAvLz8/ntb3878Lv/5V/+JVOmTCE/P5+ZM2fy13/913i93oHHNzQ08Ld/+7dcddVV5OXlMWfOHL7+9a8TCAQG7lNRUUFmZibf/e53+d73vsecOXMoKSlh5cqV7Nq1a+B+H//4x/nxj388UNe5r4qKCgD+4z/+gzVr1jB58mSKi4tZvHgx3/nOd/D7/Rf8TqZp8s1vfpNZs2ZRUFDALbfcwubNm7n99tu5/fbbL7hvuK+TSLLSiItIv4ULF/KLX/yCz3/+8zz00EPMmzcPu90+5P0PHjzIP/7jP/L3f//3ZGZm8o1vfINHHnmEz372s2zdupV/+Id/wDAMHn/8cR566CEOHDgwMGrw+9//no997GOsWLGC//zP/8Tr9fL0009zxx138MILL3DDDTcMeszt27fz8MMPc9NNN/HTn/6U1NTUYX+nV199lXfeeYfHHnuMtLQ0nn76aT74wQ+yZ88eJk2aBMCGDRv44Ac/yE033cTPf/5zAoEA3/3udwdC0ZU6ffo0AHl5efT09HD77bdTU1PD5z73OWbPns2RI0d48sknOXz4MM8//zyGYQw89qWXXuLNN9/ki1/8IkVFRRQUFHDgwAFWr15NXl4ejz32GFOmTKG+vp5169bh8/lwOp00NDSwfPlyLBYLX/ziF5k0aRK7du3iX/7lX6isrOQHP/jBBTX++Mc/Ztq0aQNTQV/72td44IEHOHDgAFlZWXzhC1+gp6eH5557jo0bNw48rri4GIAzZ87w4IMPUl5ejsPh4MCBA3zzm9/k+PHj/Pu///vA/b/61a/yrW99i7/8y7/krrvuoqamhv/9v/83fr+fqVOnDtzvcl8nkWSk4CLS7ytf+QrHjx/nRz/6ET/60Y+w2+3Mnz+ftWvX8td//dekp6dfcP/W1lY2bdo0MA1SUlLCTTfdxC9+8QvefffdgVBhGAZ/8Rd/wWuvvcbatWsJhUL8v//3/5g1axbPPPPMwGjObbfdxjXXXMPjjz/Ohg0bLqnvd7/7HZ/+9Kf56Ec/yje+8Y1LRoEG09vby/PPP09GRgYA8+bNY/r06Tz77LN89rOfBeCJJ56gtLSUZ599FofDAcCtt9562Yt9z41odHR08Pvf/54XX3yR+fPnM3XqVL797W9z8OBBNm/ezPz58wFYtmwZpaWlPProo2zYsIHbbrtt4Lm6u7t58803ycnJGbjtrrvuwmazsWXLFvLz8wduf//73z/w569//eu0t7ezc+dOxo8fP3Acl8vFY489NjASc056ejp/+MMfsFqtQF8gWb58ORs2bOCBBx5g8uTJFBQUAFwwFXb+8c4JhULceOON5Obm8slPfpInnniCnJwc2tra+N73vsd9993H008/PXD/q6++mltvvfWC4PLDH/7wsl4nkWSkqSKRfnl5ebzyyiu89tpr/OM//iN33HEHJ0+e5Ctf+Qo33HDDJTt75syZc8HajRkzZgCwZMmSC0ZCpk+fDkBlZSUAJ06coK6ujg984AMXhI/09HTuvvtu3n77bXp6ei441r/8y7/wiU98gq985Sv88z//c1ihBeDmm28eCC0AhYWFFBQUDEzfdHd3s3fvXu64446B0HKulrVr14Z1jHPPk5ubS25uLpMmTeJLX/oSq1atGpjiWb9+PTNnzmTu3LkEAoGBr5UrV2IYBtu3b7/g+ZYuXXpBaOnp6WH79u3ce++9F4SWi61fv56bb76ZkpKSC46zatUqgEuOs3r16oHQAgyEtXOvz0j27dvH+9//fsrLy8nOziY3N5f/7//7/wgGg5w8eRLo26nm9Xq57777LnjswoULKS8vv6T+y3mdRJKRRlxELjJ//vyBT7t+v5/HH3+c73//+3znO9+5YJHu+SdWYODEP9Tt59ZhtLa2AlBUVHTJsUtKSgiFQrS3t18Qfv77v/+b0tJSHnjggcv6XXJzcy+5zel00tvbC0B7ezumaVJYWHjJ/Qa7bSgul2tg7YzT6WT8+PEXrI9pbGzk9OnTg9YDXBIKL35t2tvbCQaDIy7ybWxsZN26dWEf5+L7OZ1OgIHXZzhVVVWsWbOGadOm8dRTT1FeXo7T6WTPnj187nOfG9gKfu6/92Cv57nRnPPrv5zXSSQZKbiIDMNut/OlL32J73//+xw+fHhUnvPcSamhoeGSn9XV1WGxWMjOzr7g9meeeYaPfOQjrF69mhdeeIEJEyaMSi3Z2dkYhkFjY+MlPxusvqFYLJaBsDeYvLw8XC4X3//+94f8+fkuXseRk5OD1WqltrZ22Dry8vKYNWsWjz/++KA/LykpGfbxl+PFF1+ku7ubX//61xf89zhw4MAF9zv333uw17ixsfGCx17u6ySSjDRVJNKvvr5+0NuPHTsGjN5Jb9q0aZSWlvKHP/wB0zQHbu/u7ub5559n4cKFlyy6nTBhAq+88gpOp5M1a9YMTEO8V2lpaVx77bW89NJL+Hy+gdu7urpYv379qBwDYM2aNZw5c4bc3NyBEa3zvy6eMrmYy+ViyZIlPPvss8OOOqxZs4YjR44wadKkQY9zJf8NhxqFOReuzv0c+nYP/fznP7/gfgsWLMDpdF7SC2jXrl0D04fn1/9eXieRZKARF5F+9957L6Wlpaxdu5bp06cTCoU4cOAA//Zv/0Z6ejqf+MQnRuU4FouFr371q3zsYx/jwQcf5KMf/ejArqKOjg7+8R//cdDHFRcXs27dOu69917Wrl3Ln/70J2bOnPme63nsscd48MEHuffee/nEJz5BMBjk6aefJj09nba2tvf8/ACf/OQnef7551m7di2f+tSnmDVrFqFQiOrqajZv3synP/3pge3TQ3nyySdZvXo1y5cv57Of/SyTJ0+msbGRl19+maeffpqMjAwee+wxtmzZwqpVq/j4xz/OtGnT8Hg8VFZW8uqrr/Kd73yHcePGXVbts2bNAuBf//VfWbVqFVarldmzZ7NixQocDgcf/ehH+du//Vu8Xi8/+clPaG9vv+Dxubm5fPrTn+Zb3/oW2dnZ3HnnndTW1vKNb3yD4uLiC9YrjcbrJJLoFFxE+n3+85/npZde4vvf/z4NDQ14vd6BPiyf+9znBhbfjoaHHnqItLQ0vvWtb/GRj3wEq9XK9ddfz0svvcSiRYuGfFxeXh4vvPACDz74IGvXruXZZ58ddoomHKtWreLXv/41TzzxBB/5yEcoKiriYx/7GHV1dfzud797T899TlpaGuvXr+fb3/42P/vZz6ioqMDlclFWVsayZcvCGkmYM2cOW7Zs4cknn+QrX/kKXV1dFBUVsXTp0oF1RMXFxWzdupWnnnqK7373u9TU1JCens7EiRNZuXLlJVNw4XjwwQd56623+MlPfsJTTz2FaZocOHCA6dOn86tf/Yqvfe1rPPLII+Tm5vLggw/y6U9/mvvvv/+C53j88cdJTU3lP//zP/n1r3/N9OnT+fa3v80//dM/kZWVNaqvk0iiM9xutzny3UQkmfj9fm666SZKSkr405/+FO1yEtLZs2dZsGABX/rSl/j7v//7aJcjEjc04iIifOpTn2L58uUUFxfT0NDAf/7nf3Ls2DGeeuqpaJeWEA4cOMAf//hHFi1aREZGBidOnOA73/kOGRkZfOhDH4p2eSJxRcFFROjq6uL//t//S3NzM3a7nWuuuYY//vGPLF++PNqlJYTU1FT27t3LL3/5Szo6OsjMzOTmm2/m8ccfv6xt5yKiqSIRERGJI9oOLSIiInFDwUVERETihoKLiIiIxA0FFxEREYkbCi4iIiISNxRcREREJG4ouIiIiEjcUHARERGRuKHgIiIiInFDwUVERETihoKLiIiIxA0FFxEREYkbCi4iIiISNxRcREREJG4ouIiIiEjcUHARERGRuKHgIiIiInFDwUVERETihoKLiIiIxA0FFxEREYkbCi4iIiISNxRcREREJG4ouIiIiEjcUHARERGRuKHgIiIiInFDwUVERETihoKLiIiIxA0FFxEREYkbCi4iIiISNxRcREREJG4ouIiIiEjcUHARERGRuKHgIiIiInFDwUVERETihoKLiIiIxA0FFxEREYkbCi4iIiISNxRcREREJG4ouIiIiEjcUHARERGRuKHgIiIiInFDwUVERETihoKLiIiIxA0FFxEREYkbtmgXICIymuq6LJiAxTj3ZWKc93eH1cSldz6RuKV/viKSUP59Xyq9AWPY+1gME5fNJM1ukmozSbXT/90krf976sDP+m5z2U2c1gj9EiIyJAUXEUk6IdOg22/Q7b+8xzksJnmuEPmuEAWuEPmpIQpcJvmuEBkOc2yKFZELKLiIiITJFzKo67ZS133p0EuKtS/AnPsqSP1zwEm1R6FYkQSl4CIiMgo8QYPqLivVXZeGmlSbSUFqiHHpQcoz+77yXRqhEbkSCi4iImOsJ2BQ4bZS4bayo7bvtjR7iAkZoYEgMz4jSIrekUVGZLjdbsV+EYkrpytqWP/amwAYFgsWw8BqtWIxDI4V3U/Q4oxyhZfPwKQorT/IZAQpzwxRmBrCGH6dsUjSUb4XkbhztrqOQ8fPkJ2ZgWn2ffYyTRMTCOaH4rJDlYlBfbeV+m4rO+v6bnPZTMZn9I3ITM0OMikriEVBRpKcgouIxCWnw8644oJLbj9rsRCKQj1joTdgcLzNxvE2Gxsq+tbKzMgNcHVugKtyA1r0K0lJwUVEJE70BAz2NtrZ22jHgkl5VpCZuUGuzgtQnJYocU1keAouIiJxKITBmQ4bZzpsvHTGSV5KiKvz+kZjpmQHscXhdJlIOBRcREQSQIvHwvYaB9trHDitJtNyAszMDXB1XlDN8SShKLiIiCQYb9DgYLOdg812DEymZgdZUOxnbn4Auy5bIHFOwUVErlwo1PdlmgNfRv9302YDu1aPRpuJwYl2GyfabTxrNbmm0M/1RX4mZmlNjMQnBRcRwfB4MHp6MHy+vi+/H/q/D3mb348RCAz5nN7Zs/HNmRPB30JG4gka7KxzsLPOQYEryILiAAuK/GQ5NZUk8UPBRSQJGB4PRnc3lu5uLF1dA38e+B4MRrtEibCmXivrzlhZf8bB9Jwg1xf7mZ0f0KJeiXkKLiKJwu/H2t6Opa0Ni9vdF0i6urD09Aw7MiLJzcTgWJuNY202XDaTeQV+ri/2MyFTU0kSmxRcROKQ0duLpa0Na1tb3/f2dozOTtRUVd6L3oDBm3UO3qxzUJQaZFGJn0Ulfpxa0CsxRMFFJJaZJpbOzr5RlHNBpb0di8cT7cokwTX0WHn+lJUNFU5uLPWxZJxf26olJii4iMQS08TS0YG1oQFrYyO2xkYMny/aVUkS6w0YbKp0srXawYIiP7eU+ShIVYCR6FFwEYkyS0cH1sbGgbBi8XqjXZLIJQIhg7fqHOysszM7P8Dy8T6tg5GoUHARiTCjsxNbQ8Ofg4qmfSSOmBgcaLZzoNnO5Ky+AHN1nnalSeQouIiMtWAQa0MDtupqbHV1WHp6ol2RyKg43WHjdIeNkrQgt5T5uLYwgFXbqWWMKbiIjIVAAFtdHbaqKmy1tX3N20QSVF23ld8dc7H+bIibx/lYXKqdSDJ2FFxERovPh62mZmBkRU3dJNm0ey28cDqFzVUOVpX7uKHUj1V79GWUKbiIvAeGx9MXVKqqsDY2YoS0WFGk22/huZMpvFHj4PbJXubkqwGijB4FF5HLFQhgq6zEfuYM1qamvosKisglmnot/OKQi0mZAe6c4qVcu5BkFCi4iITJ0tKC/dQp7JWVWrMichnOuG38214b1xT4uX2SlzyXwr5cOQUXkeH4fNjPnsV+6hTW9vZoVyMS1/Y12TnYbOPGUj+ryr2k2qNdkcQjBReRi5km1sZG7KdOYauu1iJbkVEUNA221TjY3WBn5QQvS8b5dUVquSwKLiL9jN5e7KdPYz99GktXV7TLEUlovQGDF0+n8Eatg9sneZlXEMDQDiQJg4KLJD1LayuOI0ewVVVpoa1IhLV5LPzmiIut1UHunerRAl4ZkYKLJC1rXV1fYGloiHYpIkmvutPK9/amckNp3wLeFJ2dZAj6X0OSSyiEraoKx5EjWNvaol2NiJzHxGBHrYNDLTbum+phVr7Wl8mlFFwkOQQC2M+cwXH0qNaviMS4Dq+Fnx1KZU6+n3unesl0agpX/kzBRRKb14vjxAnsx49j8XqjXY2IXIYDzXZOtNm4fbKXG0r8WrwrgIKLJCijpwfH0aPYT53CCKjduEi88gQNnjmRwv4mGw/N8JCbotGXZKfgIgnF8HpxHDqE/cQJXTdIJIGcbLfxrd1p3KHRl6Sn4CKJwe/HcfQojqNHNcIikqC8/aMvB5psPKjRl6Sl4CLxLRjEfuIEjsOHtYZFJEmc0OhLUlNwkfhkmtgqKnDu34+luzva1YhIhJ0bfTnSYuMvrurVdY+SiIKLxB1rYyPOd9/F2tIS7VJEJMqOtNr41z1pPDqzlwnqupsUFFwkbhidnTjffRd7dXW0SxGRGNLmtfD9d1O5c7KXm8v80S5HxpiCi8S+QADngQPYjx/XTiERGVTQNPjTqRTOuK08NN2jSwYkMP2nlZhmrakhZc8erWMRkbDsb7JT22XlQzN7KU3XB51EZIl2ASKDMXp7Sdm+ndTXX1doEZHL0txr4bt7U9lZpxW7iUgjLhJbTBP7iRM49+/H8GuuWkSuTCBk8IfjKZzusHL/NA8Oa7QrktGi4CIxw9LWRsrbb2u3kIiMmj0Ndqo7LXx4lofCVE0dJQJNFUn0BQI49+4l9ZVXFFpEZNQ19Fh5+p1U3mnQZ/VEoP+KElXW2lpSdu/WOhYRGVPeoMFvj7o46/Zxz1QvFnXbjVsKLhIdPh8pu3djr6iIdiUikkR21Dpo9Vh4dGYvTq17iUuaKpKIszY2krZunUKLiETF0VYbP3g3lU6fhl3ikYKLRE4ohGP/flybNmHp6Yl2NSKSxKq7rPzb3lQaexRe4o2Ci0SE0dWFa+NGnIcOobcJEYkFrR4L39ubxpkOzRnFEwUXGXO2M2dIXbcOm3YMiUiM6QkY/Gifi31NWvIZL/RfSsaOz4dz1y4cVVXRrkREZEgB0+DXh1Non+zllvFqfBnrFFxkTFibmnC88Qa23t5olyIiMiITgxdOp9DmtXD3FG2XjmUKLjK6QiEcBw/i0FoWEYlD22scdHgNHr7Kg11LX2KS1rjIqDE8HlI2bdICXBGJawea7fxwfyrdmjWKSQouMiosbW04X34Ze3NztEsREXnPKtxW/m1vGi29+hgWaxRc5D2znjlDyiuvYPd6o12KiMioae618IN9qbR6FF5iiYKLXDnTxNi1i9S33sJqmtGuRkRk1LV7LfxwXyptCi8xQ8FFrozfD+vWkX7qVLQrEREZU62evvDS7lV4iQUKLnLZQq2tGM8+S0ZHR7RLERGJiJb+8NKh8BJ1Ci5yWfynTuF65RXSg8FolyIiElHNvRZ+uM+FWxdnjCoFFwmbd+dOsnfuxBHtQkREoqSp18oP97l0ZekoUnCREZnBIP7168k/fRqLoX+sIpLcGnus/Gi/i26/3g+jQcFFhhX0ePD/6U/ktrVFuxQRkZhR39038qImdZGn4CJD8ra3E3ruOfLUn0VE5BJ13Vb+Y38qPQovEaXgIoNyV1djfeklctWfRURkSDVdfeGlNxDtSpKHgotcov7QITJee40ci/73EBEZSXV/ePEovESEzkwyIBQKcWr7dkr37iXTqsuiioiEq6rTyq8OuwhpkHrMKbgIAMFAgKPr13PV2bOkKbSIiFy2Y202njvpjHYZCU/BRfD7fBx74QXmt7WRotAiInLFdtQ62FZtj3YZCU3BJcl5ens59dxzLOjtxa41LSIi79nzp5wcbtGHwLGiM1US6+nqovK551gQCGBVYzkRkVFhYvCbIy5qu3SKHQt6VZNUZ3s7Z555husAQ6FFRGRUeYMGPz3owq2LMo46BZck1NbczMlnn2WRzRbtUkREElaH18J/HnTh0zVpR5WCS5JprK3l2DPPsMTp1EiLiMgYq+6y8tsjKdomPYoUXJJIzdmzHHnuOVakpeliiSIiEXKwxc5Lp7VNerQouCSJM8eOcfj551mTlYVVu4dERCJqa7WDt+q0TXo06AyWBE4dOcLhl1/mzrw8bAotIiJR8cwJJyfatE36vdJZLMGdPXGCQ+vXc09BAQ6FFhGRqAmZBr845KKxR1P174XOZAms6vRp9r34IvcVFKgjrohIDPAEDX592EUgFO1K4peCS4KqraxkzwsvcE9+Pi6FFhGRmFHbbeX5U1qse6UUXBJQQ00NO557jrtzckhXrxYRkZizo9bBwWa9P18JBZcE01Rfz2vPPMPq9HSy7FrBLiISq/77WAptHq13uVwKLgmktamJjc88wxKHg1KXK9rliIjIMHoDBr9Rc7rLpuCSINpbWtjwzDPMCgS4KjMz2uWIiEgYzrptvHLWEe0y4oqCSwJwt7Wx8dlnKejo4KaCgmiXIyIil2FzpUP9XS6Dgkuc63a72fjssxj19dxZVhbtckRE5DKZGPzX0RS6fFrvEg4Flzjm83p57aWXcFdW8oFJk7Dq+kMiInHJ7bPwu2MpmFrvMiIFlzgVCgbZsXEjFYcO8cjUqaSoK66ISFw72mpja7V2g45EZ7s4ZJome998kwNvvcXDU6eSowZzIiIJYd0ZJ5VunZqHo1cnDp08dIhdW7Zw14QJjHeq+6KISKIImga/OeLCE4h2JbFLwSXO1FZW8vq6dczPzGROenq0yxERkVHW4rHokgDDUHCJI+0tLbz2wgtk+Xws17ZnEZGEtavewal2LQMYjIJLnOjt6eG1F1+kva6OByZO1A4iEZEE98fjTl1FehAKLnEgEAiwff16zhw7xkPTppGhHUQiIgmvqdfKpkp11b2YzoAxzjRN3t66lUN79rB80iQmOfQ/sYhIsthc6aCxR6fq8+nViHGH9+5lz7ZtTC0s5Ma0tGiXIyIiERQ0Df543KnGdOdRcIlhNWfPsuPVV0l1OrkjP1/rWkREktDpDhu76tWY7hwFlxjV3dnJtvXr6e3u5q6yMjK1rkVEJGm9eNpJp65lBCi4xKRQMMiODRuoq6jglkmTmGyzRbskERGJot6Aod4u/RRcYtDBPXs4sncv08vKWJySEu1yREQkBuxttHOsVb1dFFxiTF1lJTs3byY9LY012dnYtK5FRET6PXMiBX8w2lVEl4JLDOnp7mb7K6/Q09XFmtJSsrWuRUREztPisfBqRXK3xdCZMUaEQiHe3LiR6jNnuG7iRGbYtYJcREQutbXaQW1X8p6+k/c3jzFH9u7l8J49lBQXc7PLFe1yREQkRoVMg2dPJO9CXQWXGNBQU8NbmzbhcDq5JSeHdE0RiYjIMM64bRxuSc6FujpDRllvTw/b1q+ny+1mTmkp07X1WUREwrDujJNQEnbUVXCJolAoxM7Nm6k8eZLxZWXcpOsQiYhImOq6rextTL4PuwouUXTq8GEO7t5NflERC1NTNUUkIiKX5ZWzTgKhaFcRWTpTRklnRwc7N2/GMAwmZmVxlaaIRETkMrV6LLxVl1y7UBVcoiAUCrHrtddoqq+nuKSEGx0ODDWaExGRK7CxwoE3iZrSKbhEwanDhzmydy/5RUVc43SSoykiERG5Ql1+C1urIrtG8o033uChhx5i+vTpZGZm8uKLL0bs2DpjRlhnRwc7t2zBMAxKMjKYq0ZzIiLyHm2tdtDtj9zIfXd3N7Nnz+ab3/xmxI55jhZWRJBpmrz92ms01dVRNnEiNzgcWDVFJCIi75E3aLCxwsH7pnojcrzbbruN2267LSLHuphGXCLozLFjHHn3XfIKC5nscFBsTc7mQSIiMvrerLXT5kn8D8MKLhHS093Nzi1bCJkmGRkZXKcpIhERGUUB0+CVs4l/KQAFlwh5d8cO6quqKCopYabNRoYW5IqIyCjb02Cjvjuxzy+J/dvFiNqKCvbv2kV2bi5pNpsW5IqIyJgwMVh3JrG7sCu4jDGf18tbmzbh7e0lIzuba+12HFqQKyIiY+RQi53arsQ9vSfubxYjjuzdS+WpUxSNG0eOxcI0dcgVEZEx9toY93Xp6upi//797N+/H4CzZ8+yf/9+qqqqxvS4oOAypjrb29m7YweutDTsDgcLHQ4sGm0REZEx9m6TbUx3GO3du5clS5awZMkSAL785S+zZMkSnnjiiTE75jn6+D+G9u3cSVtTE2WTJ1NmsVCq7c8iIhIBIdNgW7WDu8eor8vNN9+M2+0ek+ceiUZcxkhjTQ2H9uwhOy8Pq8XCAkdiL5YSEZHYsrPeTm8g2lWMPgWXMRAKhdjzxhv0dHeTkZ3NDJuNbG1/FhGRCPIGDd6sTbwPzTqbjoGKEyc4dfgwBUVFOA2Dedr+LCIiUbC9xk4gFO0qRpeCyyjz+3zs2b4d0zRxpaUxx24nRQtyRUQkCtw+C3sbE2s5q4LLKDu2fz/VZ85QUFyME7hK259FRCSKttck1nSRgsso6u7sZO8bb5CSkoLd4WCm3Y5doy0iIhJFNV1WznQkzq5WBZdRdODtt2luaCCvsBA7cLVGW0REJAa8UZM4ay0VXEZJS0MDB3btIjM7G4vVytU2m1r7i4hITNjfbKPDmxjnJAWXUWCaJu+88QZdbjdZubnYgJnaSSQiIjEiZBq8WZsY5yUFl1FQV1XFyUOHyCssxDAMZths2kkkIiIx5a26xNgareDyHpmmyaHdu/F6PKRlZGAFZmu0RUREYkyX38L+pvhfe6ng8h411NRw8vBhcvLzAZhms+HSaIuIiMSgPQ3x/8FaweU9ME2TQ3v24OnpIS0jAwswRzuJREQkRp1os9Lpi+8P1wou70FzfT0nDx0iJz8fwzCYarWSpmsSiYhIjAph8G6cd9LVWfY9OLRnDz1dXaRnZmIAc7S2RUREYtyexvg+Vym4XKGWhgaOHzhAdl4ehmEw0WolQ6MtIiIS46o7rTT1xO90kc60V+jwO+/Q7XaTkZUF6JpEIiISP+J5ka6CyxVoa27m6L59ZOXmYhgGOYZBkTVxrgMhIiKJbW8cTxcpuFyBI3v30uV2k5mTA8AMjbaIiEgcafFYONsRnxEgPquOoo7WVg6/8w6Z2dkYhoENmKLgIiIicSZeF+kquFymI+++i7u9naz+0ZYpNht2NZwTEZE4s6/RRjAOLwGg4HIZerq6OLJ3LxlZWRj9O4i0KFdEROJRT8DC0db4W5+p4HIZzhw7RkdrK1m5uQAUWizkaAu0iIjEqXicLtJZN0yhYJAj776LzW7H2r+DSKMtIiISzw632PAEol3F5VFwCVNtRQX1VVUDF1NMAcq1BVpEROJYIGSwvzm+PoQruITpxKFDBPx+UlwuoO8q0FYtyhURkTh3sDm+posUXMLgbmvj1OHDZGZnD9w2XdNEIiKSAE61W+Nqd5GCSxhOHz1KZ0cHGf3BZZzFousSiYhIQvAGDc50xM/SB519RxDw+zmydy8pLheW/rAyWaMtIiKSQI61KbgkjKrTp2muryc7Lw8AKzBBi3JFRCSBHGuNnw/kCi7DME2T4wcOEAqFcDidAJRZreqUKyIiCaWu24LbFx/nNgWXYbQ2NXH2+HGy+xvOAUzSaIuIiCQYE4PjcdJFN37GhqLg9JEj9HR2kldYCPS9WGUKLiKSxA6//AMOr/sBnQ1nAciZMIv5H3icCQvWDtynreoIO3/+ReoObgUzRM6EWdz6hd+TXjhh0Od84f8s67vvRcYvuJ21//ASAHv/8HXO7niG9pqjWB0uiq66kUUfeYrsshkD9/f3drHzF1+i4q3n8HS2kFE4kdl3/Q0zb//EwH3e/MlnOb7p59hS0ln0l//M1KUfGPjZqW2/58SWX7Hm8Rfe02sUr4612VhQHPvd6BRchuDzejm6bx+pGRkY/VND461WbJomEpEklpZfxsIPf4PMkqkAHN/0C1594n3c95295JbPwl13iue/uIQZq/6KBQ//I460LNqrjmB1pAz5nKu+/AyhgG/g7x53C//zN9cw+aYHB26rO7iVmXd8ioJp12OGArz9y8d4+fHbePDfD2NPSQPgzZ/8HbUHtrD8c78mo3Ai1XtfZfsPPklqbikTF7+Pil0vcHLrb7n9q6/SUXuCrU//JWXzVpGSmYe3q523f/UYd35t0xi9crHveJuVkAmWGD/NKbgMobaykvbmZgpKSgZu0zSRiCS78oV3XfD3hR96giPrfkDjsbfILZ/Frl89xvjrbmfxX/7zwH0yiycP+5wpGbkX/P3U67/D5kxl8pI/B5fb/3H9Bfe55TM/41ePFNJ8cg8ls5cC0HD0Taav+DClc5YBcPWav+bI+h/RdHI3Exe/j7aqI5TMWUbBtAUUTFvAmz/5DO7606Rk5rHz519g5u2fHHJUKBl0+y1Ud1qYkBnbTV20xmUIVSdPEgwGsTscANiBcQouIiIDQsEgJ1//HX5PN0VX3YAZClG1+yWyx03n5cdX88tHCnn2c4s4++Zzl/W8Rzf8lClLPzAwkjIYX3cHAM7zQk/xzCVU7Hye7pYaTNOkdv8WOmqPM/7a1QDkTbqG5pO78Xa10XRyDwFvL1mlU6k/tJ3mU+8w+66/ufwXIcEcb4v98YzYrzAKvB4Pp44eJS0jY+C2CVarWvyLiACtZw/w3OdvIOjzYHelc9tjz5IzYSY9bfX4e7t494/fYMEjX2PhR56ies96Xv36fdz5xBZK59wy4nM3Ht9FW8VBbvmbnw55H9M0efOnn6V45hJyy2cP3H7jX3+X17/3v/jNR8owrDYMw8LS//0TimctAWD8/NVMXfYIz372eqwOF8v+7hfYnGls+8EnWPaZn3N43Q849OK/kZKZz82f+g9yy2e99xcrzhxttXJrebSrGJ6CyyBqz57F3dpK0bhxA7dNUtM5EREAssbN4P6n38XX3c6ZHf/Da//6Ye76+lYcadkAlC96H3Pv+TsA8ifPo/7oDo6s/2FYweXoqz8lp3w2hdMXDnmfN374aVrP7ufup7ZfcPvBF75L47G3WP3/nie9oJy6Q6/zxg8/SWpuCWXzbgVgwcNfYcHDXxl4zO7ffoVx19yKxWpn739/jQe+d4DKt1/ktX/9EPd9Z89lvjLxr7LTSm8AXDF8ytNU0SAqTp4kFAphs/ddeMoBlKjFv4gIAFa7g6zSqRRMW8DCD3+dvEnXcOD5p0nJzMew2siZMPOC++eMv5qupsoRnzfg6eHUtt9x1W0fG/I+b/zof1Ox63nufGIL6fllf36st5e3f/Vlbvirb1O+8C7yJs1l9p2fZvKS97P/2W8O+lztVUc5+dpvuP6Rf6LuwGuUzFqKK6uAyUseovnUO/h63OG9IAkkZBqcjPHpIp2NL+Lp6eHMsWOkZ2YO3FauaSIRkSGZpknI78Vqd1A47Xraq49d8POOmuOkF4w8/3Bq++8J+b1MW/bIoMfY/sNPc2bHM9z5xGYyiydd8PNQ0E8o4AfjwtOaYbFihi5dbGqaJq9//69Z/Fffwu5KJxQKEgr6+54r0Pd9sMclg6Mx3v4/tmNVFNScPYu7vZ3isj8n+XJNE4mIALDrl19m/HVrSc8fj7+3k5Ov/466g6+x9it9u37m3vd5Nv3z+ymZvZTSOcupemc9Fbte4K4nXxt4ji3f/hBpeeNY+OGvX/Dcxzb8lPLF95CSmXfJcd/4wac4+fpvue2xP2F3ZdDTVg+AIzULm9OFIzWTktm3sPNnn8fmdPVNFR3cyoktv+SGv/r2Jc939JUf48oqZOKiuwEonnkTe/7rKzQcfYuqPevIGT8TZ3r2KL1q8SXWL7ioM/JFKk6cANPE1h9WrECxpolERADobW9gy7cfpae1DkdaFnkT57L2K+spu3YVAJNuuJcln/wh7/7h6+z4j78he9wMVv2f/xlYIAvQ1VSJcdHISHvNceoPb+f2r7466HEPr/sBAC9+edkFt9/ytz9jxq0fAWDlF37Hrl/8HzZ/84N4u1pJLyjn+kef4Oq1H7/gMT1tDez9w5O87593DNxWOH0hc+/5HOu/egeurEKW/d0vruj1SQRNPRa8QXDGaH4x3G63Ge0iYkVPdze/+/d/xzTNgYsqllgsrE4ZunGSiAzOO3s2vjlzxuS5N7+xmxc3bmfqxPGX/Ozs+L8gZHWOyXFFksUnrulhSnYw2mUMSkMJ56k5c4bOjg4ysrIGbitV7xYREUky1Z2xGw9it7IoOHv8OIZhYD1vTYuazomISLKp7ordc5+CS7/uzk4qTp68YDdRCpCj3UQiIpJkqjoVXGJebUUF3W73JdNEhoKLiIgkmZZeg94YvVC0gku/+upqTNPEct7UkNa3iIhIMjIxqI7RURcFFyAYCFB58iSpaRde0EvBRUREkpWCSwxrbWrC3dZ2wUUVcwyDVE0TiYhIkqrqis2IEJtVRVhDTQ3e3l6cLtfAbRptERGRZKYRlxhWc/YsFpvtgoW4Ci4iIpLMWj0WevzRruJSSR9cPL291Jw9e8E0kRUoUpt/ERFJcrG4LTrpz85NdXV0u90XBJd8iwWb1reIiEiSi8XpoqQPLg01NQQCAex2+8Bt+RptERERickFurFXUQSZpknlyZM4L7qIooKLiIgI1HdrxCWmdLa301xff8E0ESi4iIiIALR5DEJmtKu4UFKfoRtqaujt7iY1PX3gNieQoeAiIiJC0DRw+2JrzWdSn6Hrq6r62vyfF1Q02iIiIvJnLb2xdV6MrWoiyDRNqs+cwZWaesHtCi4iIiJ/1urRiEtM6Gxvp7OjA9dF1yfKV+M5ERGRAa2e2IoKsVVNBLU1N9Pb00PKeW3+QSMuIiIi51NwiRFtzc2YoRBWm23gtlTDwKXGcyIiIgM0VRQjmurqLliUC1Cg0RYREZELaMQlBoSCQeqrq7UwV0REZARur0EgFO0q/iwpz9QdbW10d3aSclFwyVNwERERuYCJQVsMTRcl5Zm6rbkZb2/vJcElV8FFRETkErE0XRQ7lURQW3PzJY3nnECKFuaKiIhcokXBJboaqquxnbebCCBToy0iIiKDiqWdRUl3tg74/TTW1l4yTZSh0RYREZFBaaooitpbWujt7r5kR5FGXERERAbXpuASPW3NzXg9HpwpKRfcnqkRFxERkUF5ArFzjky64NLR2gqAcdEIi4KLiIjI4DyBaFfwZ0kXXNpbWi7pmAuQoakiERGRQXmCsfPhPunO1q1NTTgumiZyAk6NuIiIiAzKHzIIxkj33KQKLn6fjy63G4fDccHtWpgrIiIyPE8w2hX0Saozdpfbjc/jweF0XnC7tkKLiIgML1YW6CZfcPH5LgkuGnEREREZXqysc0mqM3aX200oFMJ6cddcjbiIiIgMSyMuUdDV0QGmecntmioSEREZXqxsiU6q4NLR2orFar3k9lQFFxERkWF5NVUUea1NTZesbwFdFVpERGQkvZoqiiyf10tXR8clW6EdgFXBRUREZFgacYmw7s5OfF7vJc3nXAotIiIiI9Ialwgb2Ap90YiLgouIiMjItB06wobaCq31LSIiIiPTdugI8/b2MthLruAiIiIystCl3USiImmCi6enB3OQHi6X7jESERGRi1li5HN+0gSX7q4uLIO09tdVoUVEREam4BJh3W43Nrv9ktsVXEREREZmMWJjrihpgktXZye2ixbmAjgUXEQSitPXEu0SRBKSRlwiKBAI4O3txTrYiEsU6hGRsVPUuInUnspolyGScBRcIsjn8RAIBAYdcdFUkUhisZhBihq3kNF5PNqliCQUBZcI8nm9BAMBrINcYPHSKCMi8c7ApKBlB9nt+6JdikjCUHCJIJ/XSygYvKT5HIBFIy4iCSu3fS95LW+BGYp2KSJxT4tzI2i4ERfFFpHEltV5lMKm18EMRrsUkbhmjZETZlIEF7/XSzAYxDJIcEmKF0AkwdjtNgzDoMPdFdb903vOUtKwESPkG+PKRBKXpooiyOf1gmFgDDItFCP/HUTkMiyeP5tlN8ynub2DlraOsB7j8tRRWr8ea7B3jKsTSUwKLhHk8/lgkHb/kCQvgEiCsdts3LHyJu5YcSNd3T3UN4XXu8Xpa6W07mVsfvcYVyiSeBRcIigYCAw62gJJ8gKIJCCLxcLyG6/j3rXLCIZCVNU2DHo9sovZA52U1q/D4VWjOpHLocW5ERQKhQZ9QzNgyEAjIrHPMAwWz5/NQ3euxOl0cLamjlAYl7C1BXsprV9PSm9dBKoUSQxanBtBpmlqfYtIArtm5jQevuc2crMyOV1VQzA08vZni+mnpGEDad1nx75AkQSg4BJBoeDg2yCT4pcXSRLTJ0/gkfvWMK6ogFMVNQQCI29/NghR2PQame6jEahQJL6l2TVVFDGmaQ46VZQUv7xIlFi6wtuqPJrGlxbxyP1rmDJxHKcra/D6/CM+xgDyW98ip23v2BcoEsfSHQouEWMOMWwcI6NeIgnJfvYszrffHnJH31gpys/l0XvXMGvGZM5W19Hr8Yb1uJyOfeQ371CXXZEhpGvEJXJC2gotEhWOkydJeeMNGGK6dqxkZ2Xw8D23sWDuVVTVNtDV3RPW4zK7jlPU9BqGuuyKXCLDERuhPinO3UONuKAdRSJjzl5Vheu118A/8rTNaEpLdfHQXStZsvAa6ptaaHd3hve4nkqK61/Foi67IgMshklqjFyVOCmCSzAQGHRaKBjhIWyRZGVrbCR140aM3sh2rXU6HNyz5hZuXbKQ1vZOmlvbw3qcy9tASd06rIHwRmpEEl263YyZz/rJEVxCIQzLpb9qZD//iSQ3a3t7X3jpDG/kY7TYrFbWLF/MHStvpLvXQ11jS1iN6pz+Nkrr1WVXBGJnfQskSXAZaju0iUZdRCLJ0tVF6oYNWFpbI3tci4VlN8zn/tuXYWJSVdcYZpfdLsbVvYzD2xyBKkViV6zsKIIkCi5DdcgNRLgWkWRn8XpJ3bQJa319RI9rGAYL583i/XfdSmqKkzPV4XXZtYY8lNavx9VbG4EqRWKTRlwiLBgMDrkQN6ARF5GIMwIBXFu3YquoiPix51w1hYfvXU1eTn+X3WA4XXYDFDdsJK3rdAQqFIk9GnGJgqHWFGnERSQ6jFCIlB07sB8/HvFjT51YxqP3raWsuJBTlTX4AyO/ExiEKGx+nUz34QhUKBJbMjTiEll2h4PQEFui/RpxEYkaA0jZswfHvn0RP3ZZSSGP3r+GaRPLOF1Vi9c38vbnvi67u8hp2zP2BYrEkPQY6eECCi4acRGJAc7Dh3Hu2gVhXBxxNBXk5fDIfWuYM2MKZ6vr6en1hPW4nI4D5De/oS67kjS0xiXCHA7HkD9TcBGJDY5Tp0jZvj3iXXazMtN5+J7bWDhvJtX1TXR2hdtl9wRFjVswQnoXkcSXoTUukWW124dcnKupIpHYYa+pwbVlC4QxbTOaUl0pPHjHCpYumkdDSyttHWF22e2toqThVSzB8K6HJBKvNOISYTabTYtzReKErampr1FdT2S71jocdu6+7WZW3byQdncnTS1tYT0uxdtIaf06rIHuMa5QJDqshkmGU8Elomw225DNprQdWiT2WDs6+sKLO7Jda21WK6tvWcSdt96Ex+ujtqE5rEZ1Dn87pXUvY/d3RKBKkcjKTTGxxki7f0iS4GK124dsQKe2/yKxydLdTerGjVhaWiJ7XIuFpYuu5b7bl2OxGFTVNYTXZTfYTWndyzi9TRGoUiRyClJjaxF6UgQXm83GUG87Ho24iMQsi9dL6ubNWOvqInpcwzC4/pqref/dq0hzuThTVTvkzsTzWUNeSupfwdVTHYEqRSKjwKXgEnE2mw1Mc9BPTT0KLiIxzQgEcL3+OrazZyN+7FnTJ/HBe1dTkJfDqcqavi7cI7CYAYobN5HedSoCFYqMPQWXKLDa7RgWy6DBpVvBRSTmGaEQKW++if3o0Ygfe3L5OB69fw3l44o5WVGD3x9Ol12TguZtZHUcjECFImMrX1NFkWez2bBYLINeJVojLiLxwQBS9u7F8e67ET92aVEBj9y/lhmTJ/R12fWG12U3r203ua1vg95nJI5pxCUKBoLLIHPUPUNMIYlIbHIeOULKW29FvMtufk4WH7xvDfNmTqWiJvwuu9nuQxQ0b1eXXYlLTqtJVgxthYYkCS5Wux2L1TpocDGB8N5+RCRW2M+cwbVtG4RxccTRlJWRxvvvXsXCa2dRU9+Euyu83i0Z3acobtyMEdI+RokvhTE2TQRJElxSUlKw2WwE/IO/aWidi0j8sdXWkrplC3gj27U21ZXCA7cvZ+nia2lsbqO1PbxeM6m91f1ddvVRSeJHSVpkL8ERjuQILqmp2Oz2IYNLT4SHnEVkdFibm/sa1XVHtmutw2Hn7lU3s2bZYjo6u2gMu8tuU3+X3a4xrlBkdBSnxd75MSmCi9VmIzU9fejgohEXkbhldbv7GtV1RLZrrdVqYdXShdx92814fX5q6pvC7LLbwbi6l7H7wgs7ItFUouASPRnZ2fgVXEQSkqWnpy+8NDdH9LiGYbDk+mt48I4VWK0WKmvrwwovtmAPpfXrcHoaIlClyJXTiEsUZWZnExxiIZ/WuIjEP8Pn6+uyW1MT2eMaBvPnzOAv3ncbGelpnA67y66PkoZXSe2pikCVIpcvzR4iwxF758ekCS6paWlDfhLSiItIYjCCQVzbtmE7fTrix7562kQ+eO9qivJzOVVZQyCsLrtBiho3k955IgIVilyeWBxtgSQKLq60tCF/puAikjgM08S1cyeOw4cjfuxJ40t59L41lJcVc7qiBt8Q09PnMzApbHmDrI4DEahQJHylCi7R5UpLA8PAHGQIt8s0CSm8iCQU5759ON95J+Jda0uK8nn0vrVcNXUiZ6rq8ITRZRcgr20Pea271GVXYsbErNjbCg1JFFxSUlOx2+2DLtAN0hdeRCSxOI4dI+XNNyPeZTcvJ4sP3ruaa2dNp7Kmnu6e3rAel+U+TGHz62DG5glDkstkBZfocqWlDdvLpU29XEQSkr2iAtfWrRHvspuRnsr7776VG66bTU1DM+7O8HrNpHefobhhk7rsSlTlu2JzYS4kU3AZZsQFoF0jLiIJy1ZfT+rmzRgR7rLrSnFy39rlrLjpOppa22lpC6/XTKqnlpL69eqyK1EzKUZHWyCJgovd4cDpcg054tKuEReRhGZtacG1YUPEu+za7TbuWHETa5YtprO7h4am1rAel+JrobTuZWzqsitRMCkrsiOUlyNpgothGGRkZWmqSCSJWTs7Sd2wAUt7e2SPa7Vw683Xc8/qpfiDAarrGsPrshtwU1r3Eg5feGFHZLTE6voWSKLgApCTn4/fN/gK/w7TJKjpIpGEZ+ntJXXjRqxNTRE9rmEY3LhgLg/duRK73UZFTV2YXXZ7KalfT4qnPgJVikCGI0S+K3bPh0kVXLLy8oZ8ozABt4KLSFIw/H5cW7Zgq66O+LHnzZrOw/fcRlZmBqcrawmG2WW3uGEDqT2VEahQkt2kzNgdbYEkCy6Z2dkYhkFoiI6WWucikjyMYJCU7duxnzoV8WPPmFLOI/euobgwj9MV1QQC4XbZ3UJG57EIVCjJLJYX5kKyBZecHJwpKXiH2Fmg4CKSXAzTJGXXLhyHDkX82OVlxTx6/xomTRjH6crwu+wWtLxJdvu+CFQoyUrBJYZkZmfjdLnw9g7eDEpbokWSk3P/fpx79kS8a21xQR6P3LeGmdMncaaqjl5PeNu1c9v3ktfyFpj6sCWjy2k1KU2P7f+vkiq42Ox2cvLz8XoG742gnUUiyctx/DgpO3ZAGBdHHE252Zk8fM9tXDdnBlW1DXSF22W38yiFTeqyK6OrPDOIxYh2FcNLquACkF9cjG+IqaJO7SwSSWr2ysq+LrthTNuMpvS0VB6661Zuun4udY3NdLjD692S3nOWkoaNGKHwrockMpJY3gZ9TtIFl+zcXAxj8DhpolEXkWRna2jo67I7xMjsWElxOrhnzS2suGkBzW0dYXfZdXnqKK1fjzUY3kiNyHBi9cKK50u64JKZk4NhGASHuG5Jo4KLSNKztraSumEDRldku9babTbuWHEjd6y8ka7uHuqbWsJ6nNPX2tdl1+8e4wolkdkMk/IMBZeYk5mTgyMlZch1Lg0KLiICWLq6+rrstrVF9rgWC8tvvI571y4jGApRVdsQVqM6e6CT0vp1OLzhhR2Ri03NCWK3RruKkSVdcMnIyiLF5RoyuDRGeGGeiMQui8dD6qZNWBsaInpcwzBYPH82D925EqfTwdmaOkKh8LrsltavJ6W3LgJVSqKZkx+71yc6ny3aBUSa1WYjt7CQqiGaTvUC7lCITEvSZToAatra+OIzz7Du0CF6fT6mFxXx0w99iOvKywH4ygsv8Lu336aqrQ2HzcZ1EybwxD33sGjSpCGf88fbtvHLt97iYG0tANdNmMCT99zDwvMeM/HLX6ai5dJPip+85Ra+//DD+INB/u9zz/HywYOcbm4my+Xi1quv5hv33ktpdvbA/T/7+9/z8zffJN3p5J/vv58PXH/9wM9+v3s3v3rrLV749Kff68skScTw+3G99hqeG28kMH58RI99zcxpuFKc/M/LWzhdVcOk8aVYR3hvsph+Sho20FiwlO60iZEpVOKegcnMvPgILobb7U66bTRvbtzIzi1bGD958qA/X+JwMNWWdJmOtu5urn3iCZZPn84nbrmFwowMTjU1MTE/nykFBQD8dtcuCjMymJyfT6/fz79u3Mgf9uzh5Ne+RkFGxqDP+8Gf/pSbpkzhxilTSLHb+edXXuGZvXs59A//wLicHACaOjsvaH1+sLaWVd/5Dls++1mWzZhBR28vD/zoR/yvJUu4pqyMtp4ePvP73xMIBtn92GMAvLBvH//r17/mxU99ihONjXz0l7+k+hvfIC89nfaeHq7/+tfZ9Hd/x4Tc3DF+JSURmYaB97rr8E+bFvFjV9U28PsXNlFV18ik8SXYw3h/MoGW3MW4M68a+wIl7k3KDPCpa+NjgXfynZ2B7Lw8AEzTHHSHUUMwmJTB5alXXmF8Tg4/+8hHBm6bmJ9/wX0eXrjwgr9/+8EH+ekbb7C/upqVV1896PP+5q/+6oK///jRR/njO++w6ehRPnTDDQCXhJ5vrF/PlIICbpk+HYAsl4sNn/nMBff5tw98gIVf/zqVra1MyM3lSH09y6ZPZ8HEiSyYOJHP/P73nG5uJi89nS/8z//wyVtuUWiRK2aYJim7d2N4PPjmzInosceXFvHI/Wv4/YubOHWmmvKyEpwO+7CPMYD81rewBntpy7k2MoVK3JoVJ9NEkIRrXADyCgtxOJ1D9nNJ1gW6z+/fz4Lych780Y8o/Pu/59qvfY0fb9s25P19gQD/sW0bWS4X11zGEHqPz4c/GCQ3LW3I5/31zp189MYbh9y6DtDR24thGGS7XABcU1bG7ooK2rq72VNRQa/fz9SCArafPMk7lZX8zYoVYdcoMhTnwYM433474l12i/JzefTeNcyaMZmz1eF32c3p2Ed+8w512ZVhxcv6FkjSEZfcwkLSMjLo6erCmZJyyc/dpkmvaeIa5qSZiE43NfGDrVv57K238uW1a9l19ix/89//jdNmGxgZAXhx/34+8JOf0OPzUZKVxYbPfIb89PSwj/OlZ55hXHY2tw4xQvPcu+/S3tvLR268ccjn8Pj9fOmZZ3j4+uvJ7A8uq2fN4pFFi7j+61/HZbfzi498hDSnk0/85jf8/CMf4Qdbt/JvW7aQn57OfzzyCLNKS8OuWeR8jpMnMbxePDfcANbIbcPIzsrg4Xtu45l1r7Fn/1FKi/JJT0sd8XGZXcexhjw0FtyCacTBthGJqJK0IHmu+Fk1kpRrXABe+eMfObZ/P+P6F51ebLnDQXmSTRc5PvlJFpSXs+OLXxy47W9+9zvePnuWN7/0pYHbur1e6jo6aO7q4sfbt7P56FF2fulLFGZmjniMf37lFb6xfj2vfe5zzC0rG/Q+q59+GofVOuQiWn8wyIM/+hGVra289rnPDQSXwXzlhRfo6O3lL2+8kduefpoDjz/Oi/v3873XXmNP/9oYkSsVKCykd+lSsA8/bTPavD4fL2zYzhu791OQm0125uDryy7W6yyioWglIYtjjCuUeLKq3MvqifHTfTkpp4oAisvKCA2z9TkZG9GVZGUxs6TkgtuuLimh8qI+FmlOJ1MLC1k8eTI//dCHsFmt/PSNN0Z8/m+++ipPrlvHq3/7t0OGloqWFjYeOcLHliwZ9Of+YJCH/uM/ONPSwobPfGbY0HK0vp7f7NrFP919N68dP87SadMoyMjgoQULeKeyEvcQF9sUCZetsZHUjRsxIvz/ktPR12V31ZKFtLa7aW5tD+txLm8DJXXrsAZ6xrZAiSuz42Q30TlJG1zyi4ux2mz4fYOnzGRc53LTlCkcu6hfxfGGBspHWNBqmibeIToRn/Mvr7zCP730Euv/5m9YMHHikPf72Y4dFGZkcMcgix/PhZYTjY1s/MxnyBtmeso0Tf76V7/iWw88QHpKCsFQCH9/UD33PaTrUskosLa394WXzs6IHtdmtbJm+WLuWHkT3b0e6hpbwmpU5/S3UVr/MnZ/eJcUkMSWkxJiXEZ8ne+SNrjkFRaSmp5Ob3f3oD9vCYXwJ9mJ7e9uvZW3Tp/myZdf5mRjI7/dtYv/2LaNTy1bBvRNEX352Wd56/RpKlpaeKeyko/98pdUt7Xx4HXXDTzPh372M/7Ps88O/P2fX3mF//v88/znhz/MxLw86js6qO/ooOuiJoChUIif7djBh2+4AdtF6wYCwSAP/OhH7K6o4Dcf/SjBUGjgeXyDhKYfb9tGYWYmd19zDdAXyjYfPcpbp0/zrxs3MrOkhOzUkdcGiIRjoMtua2tkj2uxsOyG+dx/+zJMTKrqGsPssttFad3LOL3NEahSYlm8jbZAki7OBUhJTaWgpISqU6fI7O8lcj6TvumicRFceBdt10+cyLOf+AT/59ln+epLLzEpP5/vPPQQH1y0CACrxcLR+np+8dZbNHd1kZeWxvUTJ7Lt85+/YKFrZWsrlvMWNv/71q34AgEe+NGPLjjeP9x5J1+5666Bv288epTK1lY+etNNl9RW3dbG8/v2ATDva1+74Gfner2c0+B28+T69ez4whcGbls4aRKfW7WKO773PQozMvjFeVu+RUaDxeslddMmem++mWBxccSOaxgGC+fNIjUlhWfWvcaZ6jomjivBYhl+c4E15KWkfj0NhSvodWmherKaHUe7ic5J2sW5ALu3bWP7+vVDNqK7ymZjsUOL2EQkfKbFgmfxYgJDLPwfS6cqqvn9i5toamlnUlkpVuvIg+omFhrzl9CdPvj7oCSuNHuIf7ihmxEybsxJ2qkigPyior4rRQ+xSLda1y0SkctkhEKk7NiB/fjxiB97SnkZj963lrLiQk5V1uD3j/xp2iBEYfPrZLoPR6BCiSUz84JxF1ogyYNLXlERrrS0Ide5dJkmbUm4SFdE3hsDSNmzB0f/9GYklZUU8uj9a5g2qYzTVbV4h9iAcL6+Lru7yGnbM/YFSsyYm++PdglXJKmDS3pmJjn5+fR0dQ15nyqNuojIFXIePoxz1y6I8AeggrwcHrl3DXOumsLZ6np6ej0jPwjI6ThAfvMb6rKbBLIcIWbkxuf5LamDi2EYjJs4Ea9n6H/UCi4i8l44Tp0iZft2iPB7SVZmOg/fcxsL582kuq6Rzq7werdkdp2gqHELRij+Fm1K+K4v9sflNBEkeXABKCgpwTCMIZvRNYVC9CbZtmgRGV32mhpcW7ZAGNM2oynVlcKDd6xg6aJrqW9upa0jvF4zab1VlDS8iiUY3vWQJL4YmCwsic9pIlBwoWTCBFIzMugepnmUFumKyHtla2rqa1TXE9mutQ6HnbtX38xtSxfS7u6kqaVt5AcBKd5GSuvXYQ0MvgZQ4tf0nCC5KfH7gTzpg0t6ZiYl48fT2TF0F0lNF4nIaLB2dPSFF7c7ose1Wa2svmURd926BI/XR21Dc1iN6hz+dsbVvYzd1z72RUrELI7j0RZQcAGgfNo0AsHgkP+Qa4NBgpouEpFRYOnuJnXjRiwtLZE9rsXCzYvmcf8dy7FYDKpqG8IKL7ZgN6X163B6myJQpYy1DHuImXHYLfd8Ci5AyfjxpLhc9A4xhBsA6rQtWkRGicXrJXXzZqx1dRE9rmEYLJh7Ne+/exVpqS5OV9USCuO9ra/L7iu4eqojUKWMpQXFfsLoSxjT4rz80ZFbWEheYSFdmi4SkQgxAgFcr7+O7ezZiB971vRJfPDe1RTm5XCqsmbIJpzns5gBihs3kd51KgIVylgwMFkU59NEoOAC9A2hTpwxY9ht0VqgKyKjzQiFSHnzTexHj0b82JPLx/Ho/WsoH1fMyYpwu+yaFDRvI6vjYAQqlNE2OTtIviv+lz0ouPQrHT8eq82Gzzv49r9u06RJ4UVERpkBpOzdi+PddyN+7NKiAh65fy0zJk/o67LrDa/Lbl7bbnJb3wat/Ysr8b4o9xwFl35FZWVk5uTQNcxq/5MKLiIyRpxHjpDy1lsR77Kbn5PFB+9bw7yZUzlbUxd2l91s9yEKmrery26cSLWFmBOHV4IejIJLP7vDQfm0acP2czkTCGh3kYiMGfuZM7i2bYNAZE8wWRlpvP/uVSy6djbV9U24u8Lr3ZLRfYrixk0YocT4JJ/IFhQFsCXIGT9Bfo3RUTZx4rBXi/ahRboiMrZstbWkbtkCQ0xbj5VUVwoP3L6cWxZfS2NzG63t4fWaSe2t6e+yG95IjURHIizKPUfB5Twl48eTNkIXXU0XichYszY39zWqG+LK9WPF4bBz96qbWbNsMR2dXTSG3WW3qb/L7tAXrJXomZgZoCgtcab0FFzOk5aZSUl5+bBddGuCQV27SETGnNXt7mtUN8z70Zgc12ph1dKFvO+2pXh9fmrqm8LsstvR32U3vLAjkbNsfOKMtoCCyyUmTp9OKBgcsimTSd9aFxGRsWbp6ekLL83NET2uYRjcdP1cHrxjBVarhcra+jC77Pb0ddn1NESgSglHcVqQWXHeKfdiCi4XmTBlCumZmcOOupxUcBGRCDF8vr4uuzU1kT2uYTB/zgz+4n23kZGedhlddn2UNLxKak9VBKqUkawY78Mwol3F6FJwuUhGVhYTp0/H3Tb0cGeradKqSwCISIQYwSCubduwnT4d8WNfPW0iH7x3NUX5uZyqrCEQVpfdIEWNm0nvPBGBCmUoeSkh5hUm3gdtBZdBTLn6aiwWC37/0POCpzTqIiIRZJgmrp07cRw+HPFjTxpfyqP3raG8rJhTFTX4hnlvPMfApLDlDbI6DkSgQhnMigk+LAk22gIKLoMaN2kSOfn5uFtbh7zP6UCAkBbpikiEOfftw/nOOxHvWltSlM+j963l6qkTOVNVhyfM7dp5bXvIa92lLrsRlu0MsaAosRblnqPgMgiH08nU2bPp7uoackFaL1Cr6SIRiQLHsWOkvPlmxLvs5uVk8cF7V3PtrOlU1jTQ3dMb1uOy3IcpbH4dTLWTiJRl431xfxXooSTor/XeTZo+HWdKCp7eof9hapGuiESLvaIC19atEe+ym5GeyvvvvpUbrptNTUMz7s7wes2kd5+huEFddiMh3R5iUXHivs4KLkMoHDeOwnHj6BhmuqgiGKRboy4iEiW2+npSN2/GiHCXXVeKk/vWLmfFTdfR1NpOS1t4vWZSPbWU1K9Xl90xtrTMj90a7SrGjoLLECwWC9PnzMHn9WIO09PlmEZdRCSKrC0tuDZsiHiXXbvdxh0rbmLNssV0dvfQ0DT0h7zzpfhaKK17GZu67I4Jl83kxtKRr/IdzxRchlE+dSppGRl0DXMJgOO68KKIRJm1s5PUDRuwtLdH9rhWC7fefD33rF6KPxiguq4xvC67ATeldS/h8IUXdiR8S8b5SLFFu4qxpeAyjKzcXCZMmULHMD1dPMAZXb9IRKLM0ttL6saNWJuaInpcwzC4ccFcHrpzJXa7jYqaujC77PZSUr+eFE99BKpMDk6ryc3jEnu0BRRcRjRl5kwwTQLDTAkdDqOngYjIWDP8flxbtmCrro74sefNms7D99xGVmYGpyprCIbZZbe4YQOp3RURqDDx3VDiJ9Ue7SrGnoLLCMZPmUJWXt6InXQbNOoiIjHACAZJ2b4d+6lTET/2jCnlPHLvGkoK8zldUU0gEGaX3abXyOg8FoEKE5fNYrJ0fOKPtoCCy4hSXC5mzJ1Ll9s97PDnIS3SFZEYYZgmKbt24Th0KOLHLi8r5tH71zB5wjhOV9bg84XXZbeg5U2y2/dFoMLEdEuZj0xHcqy3VHAJw/TZs0lNT6fb7R7yPpXBIB3aGi0iMcS5fz/O3bsj3rW2uCCPR+5fy8zpkzhTXUuvJ7zt2rnte8lreQtMvZdejkxHiBUTkmO0BRRcwpJbWMjE6dNpH6anC8BhjbqISIxxnDhByo4dEOHp7JysDB6+5zaum3MVVbUNdIXbZbfzKIVN6rJ7OdZO8uJM4L4tF0uK4PLjH/+YOXPmUFBQwNKlS9mxY8dlPd4wDK665hosVitez9CNk04GAvRqa7SIxBh7ZWVfl90IbyRIT0vlobtu5abr51LX2EyHO7zeLek9Zylp2IgRSp5RhCs1PiPIgqLk+tCc8MHlf/7nf/jSl77E3//937N9+3ZuuOEG7r//fqqqqi7recomTaJk/HjahtlqGASOaoeRiMQgW0NDX5fdYT58jYUUp4N71tzCipsW0NzWEXaXXZenjtL69ViD4Y3UJKv3TfFgJOAVoIeT8MHle9/7Hh/60If48Ic/zIwZM3jqqacYN24cP/3pTy/reaw2G1fPn4/f7yc4zJTQ0UAAv0ZdRCQGWVtbSd2wAaMrsl1r7TYbd6y4kTtW3khXdw/1TS1hPc7pa+3rsusfen1hMptX4GdiVvKtB0ro4OLz+Xj33XdZsWLFBbevWLGCnTt3XvbzTbnqKnIKCmhvGfofnRddBkBEYpelq6uvy+4wLR7G5LgWC8tvvI571y4jGApRVdsQVqM6e6CT0rqXcXjDCzvJwm4xuXNyZK9RFSsSOri0tLQQDAYpLCy84PbCwkIaGhou+/lSUlOZdd11dHd1DXn9IoADfj8+jbqISIyyeDykbtqE9QreB98LwzBYPH82D925EqfTwdnqOkKhMLrshjyU1q8npbcuAlXGh2XjfWSnJOd5JqGDy1BM08S4wknB6bNnk5GdPexlALxoh5GIxDbD78f12mvYLnO932i4ZuY0PnjvanKzMzldWUMwOPJ0h8X0U9KwgbTus2NfYIzLcoRYniTN5gaT0MElLy8Pq9VKY2PjBbc3NTVdMgoTrsycHGbMnYu7vX34hnR+P16NuohIDDNCIVLeeAP7iRMRP/a0SeN55L41jCsu4FRlNf4wPuwZhChseo1M99EIVBi7bp/sxZFE258vltDBxeFwMG/ePDZv3nzB7Vu2bGHRokVX/LxXzZvX15BumKtG+4GD2mEkIjHOME1Sdu/GceBAxI89vrSIR+9fy5SJZZyprMUbVpddyG99i5y2vWNfYAyakBFkfmFyj+gndHAB+PSnP80vf/lLfvWrX3Hs2DG+9KUvUV1dzUc/+tErfs78oiImX3XVsIt0oW+6qEejLiISB5wHD+J8++2Id9ktzM/h0XvXMGvGZM5W14XdZTenYx/5zTuSqsuugcn7pibf9ueL2aJdwFi7//77aW1t5amnnqK+vp6ZM2fyxz/+kQkTJlzxcxqGwazrruPkoUN0d3aSlpEx6P2CwH6/n8UOxxUfS0QkUhwnT2J4vXhuuAGskZuLyO7vsvvMutfYs/8opUX5pKeljvi4zK7jWEMeGgtuwTQSf+7k2sIA5ZnJE9SGYrjdbg0JXAHTNNnwzDMcfucdyiZNGnKxrwW4LyWFdEvCD26JSIIIFBbSu3Qp2O0RPa7X5+OFDdt5Y/d+CnKzyc4c/EPhxXqdRTQUrSRkSdwPiSlWk89f302WU6dsnU2vkGEYzF20iJTU1GEvvhgC9mmti4jEEVtjI6kbN2L0RrZrrdPR12V31ZKFtLa7aW5tD+txLm8DJXXrsAZ6xrbAKHrfVI9CSz8Fl/egaNw4ps+eTWtz87A7jE7qytEiEmes7e194WWYTQhjwWa1smb5Yu5YeRPdvR7qGod/fz3H6W+jtP5l7P7wLikQT2bmBbi+OLkX5J5PweU9MAyDOYsWkZaRQWd7+5D3M4F3NeoiInFmoMtua2tkj2uxsOyG+dx/+zJMoKquMcwuu12U1r2M09s89kVGSKrN5IFpkb2+VKxTcHmPCoqLueqaa2hraRn2H9aZYJBWjbqISJyxeL19XXbr6yN6XMMwWDhvFh+461ZcKU7OhNll1xryUlK/HldvbQSqHHv3TPWQqSmiCyi4jII5Cxf2ddMd4VPJLl/ydjoUkfhlBAK4tm7FVlER8WPPvmoKj9y3mrycTE5XhdtlN0Bxw0bSuk5HoMKxMzvfz/wiTRFdTMFlFOTk5zNr/nzcbW3DXsOoPhTijC4FICJxyAiFSNmxA/vx4xE/9pTyMh69by1lxYWcqqzB7w+zy27z62S6D0egwtGXZgtx/7TkvIjiSBRcRsmsBQvIzM2lbYSmdLv8fvxqSiciccgAUvbswbFvX8SPXVZSyKP3r2HapDJOV9XiDWMEu6/L7i5y2vaMfYGj7N5pXjIcOlcMRsFllGTl5DB7wQK63G5Cw4y69JqmFuqKSFxzHj6Mc9cuiPC6vYK8HB65dw1zrprC2ep6enrDW7Sa03GA/OY34qbL7twCP/OSvK3/cBRcRtGs664jJz+ftqamYe93OBCgTQt1RSSOOU6dImX7dggGI3rcrMx0Hr7nNhbOm0l1XSOdXeH1bsnsOkFR4xaMUGwHgnS7pohGouAyitIzM5mzcCHdXV0Eh1nLYgI7tVBXROKcvaYG15YtEOH3s1RXCg/esYKli66lvrmVto7wes2k9VZR0vAqlmDsBoP7p3lJs2uKaDgKLqNs5vz5FI0bR9MIWwfrQyFOa6GuiMQ5W1NTX6O6nsh2rXU47Ny9+mZuW7qQto5OmlrawnpcireR0vp1WAPdY1zh5bu20M+cAp0XRqLgMspcqaksWLqUUDCIZ4R/yG/7/fi0UFdE4py1o6MvvAxz+ZOxYLNaWX3LIu5etQSP10dtQ3hddh3+dsbVvYzd1z72RYYpwxHi3qlqNBcOBZcxMHXmTKbMnEljff2w/4h6TVPXMRKRhGDp7iZ140YsI+ysHPXjWizcvGge99+xHIvFoKq2IazwYgt2U1q/Dqd3+DWJkWBg8hczPKRG9pqWcUvBZQxYrFauv+UW0tLTR2xKp4W6IpIoLF4vqZs3Y62ri+hxDcNgwdyref/dq0hLdXG6qnbY3Z3n9HXZfQVXT3UEqhza6ok+pudGdpFzPFNwGSMFJSVcs2gR7vb2ERfqvqWFuiKSIIxAANfrr2M7ezbix541fRIfvHc1hXk5nKqsIRjGjieLGaC4cRPpXaciUOGlZub6WTlB54DLoeAyhuYuXkxxWRlNI3z6aAiFOKGFuiKSIIxQiJQ338R+9GjEjz25fByP3r+G8nHFnKwIt8uuSUHzNrI6Dkagwj/Lcfj5i6s9GEZEDxv3FFzG0MBC3VCI3hEW6u7y+ejSlJGIJAgDSNm7F8e770b82KVFBTxy/1qumlLe12XXG16X3by23eS2vg0R2DRhI8hH5/pw2cb8UAlHwWWMTbn6aqbMnElTXd2wC8b8wHafL6xFZSIi8cJ55Agpb70V8S67+TlZfPC+1cybOZWzNXVhd9nNdh+ioHn7mHfZfWiGl5I0fVi9EgouY8xitXL9smWkZ2aOuFC3PhTisKaMRCTB2M+cwbVtG0T4/S0zPY33372KRdfOprq+CXdneL1bMrpPUdy4CSM0Nrs+F+R3Mr9Yi3GvlIJLBBQUFzM3jIW6AO/4/bRrykhEEoyttpbULVvAG9mutee67N6y+FoaW9pobQ+v10xqb01/l93R7a1S7OzmwZmj+pRJR8ElQq5ZtIjisjIaR1ioGwS2eb0ENWUkIgnG2tzc16iuO7Jda+12G3evupk1yxbT0dlFY9hddpv6u+x2jUodTrz89XwTqxbjvicKLhGSkprK9bfcAkB35/DX1WgxTfarMZ2IJCCr293XqK6jI7LHtVpYtXQh77ttKV6fn5r6pjC77Hb0d9kNL+wMxTBDfHSun0yHPpS+VwouETT56quZOX8+LQ0NI/YX2B8I0BThq66KiESCpaenL7w0N0f0uIZhcNP1c3nwjhVYrRYqa4fvbn6OLdjT12XX03DFx14zoYspOQoto0HBJYIsFguLli+nqKyMhpqaYe9rAtt8PgKaMhKRBGT4fH1ddkd4Lxz14xoG8+fM4C/edxsZ6WmX0WXXR0nDq6T2VF32Ma/K6GTlZM0PjRYFlwhLy8jgxlWrsNpsdLa3D3tft2myW1NGIpKgjGAQ17Zt2E6fjvixr542kQ/eu5qi/FxOVdYQCKvLbpCixs2kd54I+zi5th4+dM17qVQupuASBeXTpjF34ULaWloIjBBMjgYC1GjKSEQSlGGauHbuxHH4cMSPPWl8KR+6fy0Ty0o4VVGDL4wPigYmhS1vkN2+f8T7ppi9fGpBEId1NKqVcxRcosAwDBbcfDNlkybRUFMz4hzrGz4fvZoyEpEE5ty3D+c770Ska+35igvzeOS+NVw9dSJnqurwhLldO7f9HfJadw1ZrzXk5ZPzfWQ5R7NaAQWXqElJTeXGVatwulwjNqbrMU22er2EFF5EJIE5jh0j5c03I95lNy8niw/eu5prZ02nsqaB7p7esB6X5T5MYfPrYF40Kh4K8MgMN6WZOsWOBb2qUTRu4kSuvekmOjs68I2Q8utDIfZovYuIJDh7RQWurVsj3mU3Iz2V9999KzdcN5uahuawu+ymd5+huOG8LrtmiDVF9cwpTRnDapObgkuUzVu8mInTp4c1ZXQoEOCsLgkgIgnOVl9P6ubNGBHusutKcXLf2uWsuOk6mlrbaWkLr9dMqqeWzDPPga+Ha5wV3Doza2wLTXIKLlHmcDq5cdUq0jMzaWtqGvH+230+XRJARBKetaUF14YNUemye8eKm1izbDGd3T00NA0/lQ/Q6/HSXHGEhcFtfHBRbgSqTG4KLjGgsLSU62+5hZ7ubrye4a+LEQA2e734tN5FRBKctbOT1A0bsIzQOmLUj2u1cOvN13PP6qX4gwGq6xqHHBH3+wNU1tQzf84M7l6+AItFp9Wxplc4RsxasIBps2fTUF1NaITtz27TZLvPF6HKRESix9LbS+rGjVjDGJEeTYZhcOOCuTx050rsdhsVNXWXhJdgMMSZ6lqumjaR+9YsI8XpiGiNyUrBJUbYbDZuXrOGorIy6qqrR1zvUhkMckCLdUUkCRh+P64tW7BVV0f82PNmTefhe1eTlZnBqcoagv1T9aGQyZmqGiaUFvPgHSvISE+NeG3JSsElhmRkZ7P09ttJcbloDePTxTt+P7VqTiciScAIBknZvh37qVMRP/aMyRN45N41lBTmc7qimkAgSGVNHfm52Tx450ryc7MjXlMyU3CJMWWTJnHDrbfi7e0d8SrSJrDV66VLi3VFJAkYpknKrl04Dh6M+LHLy4p59P41TJ4wjqOnzuJypXD/HSsYX1oY8VqSneF2u7XKM8aEQiFef/ll9u7YQcn48dgdw8+b5lks3O50YjV0ES8RSQ6+adPwXncdRPh9r62jk03b32bapPFcM3NaRI8tfRRcYpSnt5d1//3fnD1+nLJJk0ZcqV5utXKLw4FF4UVEkoR/wgQ8ixeDVRcDSiaaKopRKS4Xt9xxB7mFhTTU1o54/4pgkLe1WFdEkoi9srKvy67e+5KKgksMyy0oYOmaNVitVtpbWka8/5FAgIP6BywiScTW0NDXZXeEHliSOBRcYtzEGTO4fulSutxuent6Rrz/br+fU7osgIgkEWtrK6kbNmB0dUW7FIkABZcYZxgG1954I1dfey1NtbUEwgglb/h82iYtIknF0tXV12W3rS3apcgYU3CJA1abjZtuu41xkyZRV1mJOcL25xCwxeulVdukRSSJWDweUjdtwtrQEO1SZAwpuMSJtIwMlt15JzkFBWF11vUDGzwe9XgRkaRi+P2kvPUWaNQ5YSm4xJHC0lKW33UXrtRUmurqRrx/L7DB68WrCzKKSJIw7XZ6ly7VFukEpuASZyZMmcLS228HCOuyAB2mySavl4DCi4gkONNqpfeWWwjl5ES7FBlDCi5xaNrs2dy4ahWe3l7cYVzuvTEU4nWfj5DCi4gkKNNioXfJEoIFBdEuRcaYgkscMgyDuYsWsWDpUtxtbfSEsQWwMhhkm8KLiCQgE/AsWkSwtDTapUgEKLjEKcMwWHjLLcxZuJDm+nq8YTRfOqPwIiIJxgS8119PYOLEaJciEaLgEsesNhtLVq9mxjXX0FBdjT+MrrlngkFNG4lIQgiZJocyMvBPnRrtUiSCFFzinMPpZNmddzJxxgzqKisJhrEF8GwwyFaFFxGJYyHT5HWvl9wlS6JdikSYgksCSE1PZ8Xdd1MyYQK1FRUjNqiDvosybvX5CCq8iEicCZomm7q6mHzXXWRkZ0e7HIkwBZcEkZWby8r3vY+cggJqq6pGbFAHCi8iEn8CoRCvdHQw+c47ycnPj3Y5EgUKLgmkoKSEW++5h8zsbGorK8MKL5XBIK8pvIhIHPCHQrzU2sqU22+nUDuIkpaCS4IZN3Eiq+6778/hJYxpoyqFFxGJcf5QiOebm5m8Zg2l5eXRLkeiSMElAY2bOJHb7r+frJycvmmjMMPLFq9X4UVEYo43GOS55mamr13LxGnTol2ORJmCS4IqLS9n1f33k5WbG/bIS3UoxCavF5/Ci4jECE8wyJ9aW5l5xx1MnD492uVIDFBwSWClEyZw2333kZWXF3Z4qQ2FWO/x0K2rSotIlPUEAjzf1sacO+9kgnq1SD8FlwRX0h9esvPyqAkzvLSaJi95vbQpvIhIlLR6vTzT3s41d99N2eTJ0S5HYoiCSxIomTCB2+6/n9z8fGoqKwmFEUh6TJOXPR5qw2hoJyIymiq7u3nW7WbRPfdoIa5cwnC73VrQkCTqq6vZ8D//Q2tTE6Xl5VgsI+dWA7jJ4WCqzTb2BYpI0jvQ3s6OQICV991H0bhx0S5HYpBGXJJIcVkZtz3wALmFhdRWVIQ18mIC230+9oVxHSQRkffi9cZGdoRCrHrgAYUWGZJGXJJQQ00NG599lqa6OkomTMAW5mjKNKuVGxwOLIYxxhWKSDIJmiYv1NTQmJbGqvvuI7+4ONolSQxTcElSrU1NbH7+eSpPnqRk/HgcTmdYjxtnsXCL04lD4UVERoEnFOK/z57Fl5fXtxavoCDaJUmMU3BJYt1uN5tffJGTBw+SX1xMalpaWI/LNQxudTpJDWONjIjIUDqCQX518iSZEyey/K67yM7Li3ZJEgcUXJKc1+Nh+yuvcGDXLrJyc8kM80qrqYbBcoeDAqt1bAsUkYRU5/Pxm5MnmXjNNdy8dm3YH5xEFFyEYCDArq1b2f3666S4XGEP1VqARXY7M+z2sS1QRBLKoa4uXqyuZu6NN7J4xQpseg+Ry6DgIgCYpsmBt99mx4YNhIJBCktLMcJcxzLNamWxw4FV615EZBgB0+TVxkYOdnVx4623MmfhwrDaMoicT8FFLnDq8GFee+klujs7KRk/Puw3lTyLheUOB+l6ExKRQXSEQvz3mTP0Op0svf12ps6aFe2SJE4puMglaisr2fynP9FUX0/pZWyXdgJLnU7Gad2LiJznpM/HH06cILekhOV33UXJhAnRLknimIKLDKq1qYnNf/oTVadOUVRWhjMlJezHzrXZmGe3q9+LSJILmiZvdHez5fRpJk6bxvK77yYnPz/aZUmcU3CRIXW73bz28sucOHCA7NxcMsLccQRQbLGw1OHQlmmRJOUOhXi5tZUT9fVcNW8eS2+/XTuHZFQouMiw/D4fu7du5Z0dO7BYLBSUlIS9aDeFvqmjUk0diSSVs4EAf6qqotvrZd7ixSxeuRK7wxHtsiRBKLjIiEzT5Pj+/bzx6qt0dnRQPH582OteAOb0Tx1p15FIYguaJrs8HjadPk16VhY3rFzJVfPmaeeQjCoFFwlbY00NW19+merTpykoLcWVmhr2Y7MNg5udTvL0BiaSkDpCITa0t3O4pobxkyezdO1aCnWhRBkDCi5yWbo7O3nj1Vc5sncvqenp5OTnhz11ZADX2O3Mtdm0cFckQYRMk0N+Pxtra+nt7WXWggXccOutl/XBRuRyKLjIZQsGAuzbuZO3t27F6/FQXFaG9TLWseQZBkucTnI0+iIS19pDIbb29nLg7FnSMjJYvHIlM+fP19SQjCkFF7liVadPs339euqrqigcN44Ulyvsx1qAa+12Zmn0RSTuhEyTQ4EAO9rbqa+ro2zSJJauXUvx+PHRLk2SgIKLvCedHR288corHNu/n7SMDLLz8sKeOgIosFhY4nCQpU9oInGhLRRiu9fL8YYGPD09zLzuOm689VZS09OjXZokCQUXec8CgQD73nqL3a+/jqenh6KyMuyXcdE0K3Cd3c7VNttlhR4RiZyQaXIgEOCd3l5qq6txpaWxaMUKZl93HRa1PJAIUnCRUVNbWcmODRuoOnWKrJwcMnNyLiuIFFks3ORwkKnRF5GY0hoK8YbXy9n2dtqamymZMIGla9dSWl4e7dIkCSm4yKjyejzs3bGDd998E5/XS1Fp6WVdst4KzLLZmGu3Y9Poi0hUBU2Tg+dGWWpqsNlszF6wgAVLl2pqSKJGwUXGRPXp07y5aRPVZ86QnZdH5mVcLgAgzTBYYLcz6TIa3YnI6KkKBnnb56PmvFGWxStXUj51qqZ0JaoUXGTMeHp62LN9O/t37SLg81E0bhzWywwixRYLixwObZ0WiZD2UIi3fT4qfD4aa2v7Rlmuv57rbr5Z1xqSmKDgImPKNE2qTp1ix8aN1FVWkpOfT0ZW1mU9hwFcZbNxrd2OQ5/0RMaE1zTZ5/dzJBCgo72d9pYWSiZM4IaVK5mgURaJIQouEhE93d3sef11Drz9NsFgkKLS0ssefXEC1zkcTLNa9SYqMkpCpsmJQIB3/H66AwEaamqw2+3MXbiQ+TffrA64EnMUXCRiTNPk7PHjvLVpE/XV1WTn5pKRnX3ZISTPYmGx3U6BtmCKvCd1wSC7fD7aTBN3/yjLuIkTWbxiBeOnTNEHBIlJCi4Scd2dnezdsYNDe/bg6ekhv7j4srrunjPVauVau500rX8RuSydoRC7/X4qgkF8Xi9NdXXYnU7mLlzIdUuWkKJRFolhCi4SNfVVVby9bRtnjh7FarVSUFx82dNHFmC6zcZcm41UBRiRYXn6L4h4OBDAFwzS0tCAz+tlwtSpLFi6lLJJkzTKIjFPwUWiKhQMcuLQIfZs20ZDTQ0ZWVmXfdkA6Ov/MsNmY47djktvvCIXOBdYjgQC+E2TjrY23G1t5BcXM3/JEmbMmXNZ/ZZEoknBRWJCb08P+3fuZP/OnXS73eQVF1/R1ksrcLXNxmy7nRQFGEly5weWAH2L5Jvr60nLyGDOwoXMXbiQtIyMaJcpclkUXCSmNNfXs3vbNk4ePIgJFJaUXNEnQRt9AWaWAowkIY9pctDv52h/YPH7fDTV12MYBtNmz2b+kiUUFBdHu0yRK6LgIjEnFApx9vhxdr/+OrUVFaSmp5Obn49xBWtY7MDM/gCjHjCS6Hr7R1jOBZZQKERrYyO9PT2UTZrEdTffTPm0aVi0HkzimIKLxCyvx8OhPXt4d8cOOtrayMzJIesyL9x4jgOYabdzlc2mERhJOL39IyzH+gOLaZp0dnTQ3txMbmEh1950E1ddcw0OpzPapYq8ZwouEvM6Wls5uHs3h/fupau9nazc3Mu+8vQ5VmCy1cpMu12XEZC45w6FOBoIcCwQIEhfYOlyu2lrbiY1PZ2Z8+cz74YbLrtbtUgsU3CRuNHW3MzB3bs5+u67dLndV9zA7pwSi4WrbTbKrFYsGoWROGGaJrWhEEf8fqpDoYHbujo6aGtpITU9nWmzZzN7wQIKS0ujXK3I6FNwkbjT2tTEgbff5ti+fXR3dpKdl0dGVtYVB5gMw+Aqm41pNpvWwUjM8pkmJwMBjgQCdJp9b9sXB5bpc+Ywe8ECCkpKolytyNhRcJG41dLQwIHduzm2bx89XV3k5OeTnpl5xQHGBky12bjaZiNL00gSI9r6p4NO9a9fgfPWsPQHlhlz5jBLgUWShIKLxL2m+noOvv02xw8coKeri9z8fNLeQ4ABGGexMNNup9RiUSdRibiQaVIZDHI0EKC+fzoILgwsaRkZTD8XWLS1WZKIgoskBNM0aaqr48Dbb3Pi4EF6u7vJzM4mMyfnPW39TDcMJlutTNEojERAdyjEyWCQY4EAPeaf35pN06SzvZ2O1lZSMzKYPncus6+7jnwFFklCCi6SUEzTpKm2lmP793P8wAHc7e2kuFzkFBRgf48tzfMsFqZYrUyy2XRZARk1vv7RlVOBAHXnja4ABAIBOlpa6O7qIi0jgxnXXMOs+fMVWCSpKbhIwurs6ODU4cMcfucdmurrsRgG2fn5V3QpgfMZQKnFwhSbjQlWKzaFGLlMof6dQacCASqDQYIX/by3p4e25mZCwSC5BQVcNW8eU2fNIic/Pyr1isQSBRdJeD6vl4oTJziydy/VZ87g9XjIyskhMzv7irrxns8GlPdPJRVbLNpWLUMKmSYNoRBng0HOBgJ4L/q5GQrhbm/H3d6Ow+mktLycq+fNY+KMGThTUqJSs0gsUnCRpBEKhaivrub4vn2cOHyYro4OUtPSyM7LG5Ur47r618OMt1opVIgR+qYuG0MhzgSDVAQC9A5yH7/PR3tLC709PWRkZTF19mymz55N8fjxas0vMggFF0lKHa2tnOyfRmptbMSwWMjKySEtI2NUdhE5gFKrlTKrlXFWq9bEJBGvaVIXDFITDFIdCtFrXvoWa5omvd3dtLW0gGmSX1zMzPnzmXzVVWTm5EShapH4oeAiSc3r8XD22DGOHzxIbUUFPV1dOFNSyMrJwelyjdpW6HyLhbL+IJNnGNpinUBM06QlFKImFKImGKQpFGKoN1Wf14u7rY2e7m5SUlMZP3kyV8+bx4SpU7E7HBGtWyReKbiI0HfyaWtupvLkSY4fOEBzfT1ej4fU9HSycnJG9aTigoEQU2K1qltvHOo1TWr7R1VqgsFL1qucLxgI4G5vp8vtxmqzkVtQwLTZs5k4bRoFpaUKsSKXScFF5CKhYJCGmhoqTp7k5MGDfbs7QiHSMzPJyMrCarON2rEsQKHF0vdltVJgseDUiSzmBEyT5lCoL6yEQrRctG35YqFgkC63G3dHB5gmmTk5TL76aiZNn05pefmorKkSSVYKLiLD8Pt81FRUcPboUU4fO0ZnezuGxUJmdjZpGRljsngyyzAGgkyhxUKmppYiyjRN3KZJUyhEUyhEczBIq2kOOf1zTigUosvtprOjg1AwSHpmJuOnTGHyVVdRNnkyrtTUiNQvkugUXETC1NPdTdWpU5w6fJjqM2fo6erCsFhIS08nPTNzzNYoOOgblSnoDzL5Fgt2BZlR4+kPKc3961OaQyF8YT42EAjQ3dlJl9tNKBgkLTOT8ZMmMXHGDMomTSI9M3NMaxdJRgouIlego62N+spKqs+epfLkSbrcboJ+P06Xi/TMTFxpaWM2SmIAOYZBjsVCjsVCtsVCjmGQpq2zwzJNk+7+0ZT2/oDSFAoNXGk53Ofo7emh2+3G09uLxWIhLSODkvJyJk2fTtmkSWRkZ4/dLyEiCi4i75XP66Whpoa6ykrOHj9Oa1MTvT09WK1W0jMySM/MHNV1MUOxAzkWC1mGQZbFQlb/NFOGYSRVTxmfadIRCuEe5PvFHWrDEfD76XK76e7sJBgK9V1CIi+P8unTKRk/nsJx495zN2YRCZ+Ci8goCoVCtDU1UV9dTdXp09ScPUu3200oFMKVmkpqejopqakRbSxmABmGQabFQpphkGoYuPq/zv05BeIi3JimiZe+XT2e/q9u08QdCtHR/32wJm+XdYxQiN6eHrrcbrweDxarlfTMTMZNnEjZpEkUl5WRU1Cg5nAiUaLgIjKGent6qK+q6huNOXGCzvZ2eru7AbA7HKSmpeFKS4t6Dw8DSIELwsy5PzsNA5thYKXvEgdWw+j7ftGfwwk+wf5Rj9BFfz73dx9/DiWXfKdvPcpov2EF/H56e3ro7e7G6/GAYeBKTSW3oIDyadMoLiujcNw4UlyuUT6yiFwJBReRCAmFQrjb2mhpbKS1sZGas2dpaWykt6uLQCCAxWLB1R9kUlJS3vN1lCLNoC/Y2OgLMSaXhpNoC4VCeHp66O3pwdPTQygUwmq14uq/9ENJeTn5hYUUjhtHdl6ednOJxCAFF5Eo8vT00NLYSEtjI401NdRVVfVNUfT2TXg4U1Jwulx9353OuAsz0WSaJj6vty+kdHfj9/sBSElNJT0jg+IJEygsKSG3oICcggJS09MVVETigIKLSAwJBgK0t7bS2thIc0MDdRUVtLe04PV4BqYxLIaBMyUFR0oKDqcTh9OZ1OstAoEAPo8Hn9c78GX27xSyOxy40tIoKCmhuKyM3MJCcgsKyMrJwWK1RrlyEbkSCi4iMc7n9dLR1oa7/6utqYnGujq63W68Xi/+i07UDqcTu8OBzW7v+7LZ4nokwTRNQqEQAb8f77mA4vEQDAb7gpzF0hfknE6ycnPJLyoiMyenr9NxdjY5+fk4nM5o/xoiMkoUXETikGmaeHt76XS76ero6Gsv39ZGa1MT7c3NeD0eAn5/31cggAGYhgGmidVmw2az/TnY9Icbm92OEaEuvaZpEgwGCQYCAzVe/GfO1WGaWCwWbP2hLC09ndyiInLz80nPyiIjM5P0rCzSs7KwRWDbuYhEl4KLSIIJhUJ4PR48/QtQBxaj9vYObPPt6uigu7MTv9dL4LzAYPZfg+dceDFNs+/PF+3msVgsfSHHYgHTJGSafd8vuobPxSHo3POZponVasVmtw8EKYfTSWpGBmn9X660tL61PSkppKSmkpKaSkZm5qhetVtE4o+Ci0iSMk2zb+1Mf6Dxeb0E+0c7zo2GhIJBAv3fB0ZIAgECPh8+n49AIIDVah0IH+e+W6xWLBYLFqsV67k/9//dYrVidzj6AonLhdPlIsXlGhjxEREZjoKLiIiIxI3k3YogIiIicUfBRUREROKGgouIiIjEDQUXERERiRsKLiIiIhI3FFxEREQkbii4iIiISNxQcBEREZG4oeAiIiIicUPBRUREROKGgouIiIjEDQUXERERiRsKLiIiIhI3FFxEREQkbii4iIiISNxQcBEREZG4oeAiIiIicUPBRUREROKGgouIiIjEDQUXERERiRsKLiIiIhI3FFxEREQkbii4iIiISNxQcBEREZG4oeAiIiIicUPBRUREROKGgouIiIjEDQUXERERiRsKLiIiIhI3FFxEREQkbii4iIiISNxQcBEREZG4oeAiIiIicUPBRUREROKGgouIiIjEDQUXERERiRsKLiIiIhI3FFxEREQkbii4iIiISNxQcBEREZG4oeAiIiIicUPBRUREROKGgouIiIjEDQUXERERiRsKLiIiIhI3FFxEREQkbii4iIiISNxQcBEREZG4oeAiIiIicUPBRUREROKGgouIiIjEDQUXERERiRsKLiIiIhI3FFxEREQkbii4iIiISNxQcBEREZG48f8DAUCKXrztZVMAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 700x700 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "#Data visualisation using pie chart\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Set custom colors\n",
+    "colors = ['#ff9999', '#66b3ff']\n",
+    "\n",
+    "# Create the figure and axes\n",
+    "fig, ax = plt.subplots(figsize=[7, 7], facecolor='#f2f2f2')\n",
+    "\n",
+    "# Plot the pie chart\n",
+    "explode = [0, 0.15]\n",
+    "labels = train_data[\"smoking\"].value_counts().index\n",
+    "sizes = train_data[\"smoking\"].value_counts().values\n",
+    "ax.pie(sizes, explode=explode, labels=labels, autopct='%1.3f%%', shadow=True, startangle=90, colors=colors)\n",
+    "\n",
+    "# Add a title\n",
+    "ax.set_title(\"Smoking Percentage\")\n",
+    "\n",
+    "# Display the pie chart\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "025dd96f",
+   "metadata": {},
+   "source": [
+    "As it can be seen above visually and also in descriptive statistics that approximately 37% of the individuals in the dataset are smokers, while the remaining 63% are non-smokers. \n",
+    "This means there is a class imbalance in dataset.Lets deal with this in training section of this report."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c18a9c81",
+   "metadata": {},
+   "source": [
+    "#### Correlation Between Smoking and Input Features"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 165,
+   "id": "636e512f",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Assuming your DataFrame is named 'train_copy'\n",
+    "correlations = train_copy.corr()['smoking'].drop('smoking')\n",
+    "\n",
+    "# Sort the correlations in descending order\n",
+    "correlations_sorted = correlations.sort_values(ascending=False)\n",
+    "\n",
+    "# Plot the correlations\n",
+    "plt.figure(figsize=(10, 6))\n",
+    "correlations_sorted.plot(kind='bar')\n",
+    "plt.xlabel('Input Features')\n",
+    "plt.ylabel('Correlation')\n",
+    "plt.title('Correlation between Smoking and Input Features')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "b87b5bb3",
+   "metadata": {},
+   "source": [
+    "Based on the correlations between the \"smoking\" variable and the other input features, here are some insights and interpretations:\n",
+    "\n",
+    "***Positive Correlations:***\n",
+    "\n",
+    "Height (0.394), weight (0.297), waist (0.217), triglyceride (0.232), hemoglobin (0.398), AST (0.073), ALT (0.151), and Gtp (0.304) show positive correlations with smoking. This suggests that there might be a tendency for individuals who smoke to have higher values in these variables.\n",
+    "Dental caries (0.109) also shows a positive correlation, indicating a possible association between smoking and dental caries.\n",
+    "\n",
+    "***Negative Correlations:***\n",
+    "\n",
+    "Age (-0.171), HDL (-0.177), LDL (-0.047), and Cholesterol (-0.044) show negative correlations with smoking. This suggests that smoking might be associated with lower values in these variables.\n",
+    "Urine protein (0.006), eyesight (both left and right) (0.061 and 0.066), hearing (both left and right) (-0.022 and -0.019), systolic (0.061), relaxation (0.094), fasting blood sugar (0.081), and serum creatinine (0.218) also show weak correlations with smoking.\n",
+    "\n",
+    "As these values are not very close to 1 or -1 ,further analysis and domain knowledge are necessary to draw meaningful conclusions about the relationship between smoking and these variables."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d76b55d6",
+   "metadata": {},
+   "source": [
+    "### 4. FEATURE SELECTION"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "3dd29098",
+   "metadata": {},
+   "source": [
+    "#### Method 1. Random Forest classifier\n",
+    "Implementation of a feature importance analysis using the ***Random Forest classifier*** in scikit-learn.As this method is not sensitive to the scale of features, no scaling is performed before feature selection"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 166,
+   "id": "983f6aea",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 800x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "X = train_copy.iloc[:, :-1]\n",
+    "y = train_copy.smoking\n",
+    "\n",
+    "# Create a Random Forest classifier\n",
+    "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n",
+    "\n",
+    "# Fit the model to the data\n",
+    "rf.fit(X, y)\n",
+    "\n",
+    "feature_names = X.columns \n",
+    "\n",
+    "# Get feature importance scores\n",
+    "importance_scores = rf.feature_importances_\n",
+    "\n",
+    "# Sort features by importance in descending order\n",
+    "feature_importance = sorted(zip(importance_scores, feature_names), reverse=True)\n",
+    "\n",
+    "# Reverse the feature importance list to have the most important features at the top\n",
+    "feature_importance.reverse()\n",
+    "\n",
+    "# Extract the feature names and importance scores\n",
+    "features = [feature_name for importance, feature_name in feature_importance]\n",
+    "importances = [importance for importance, feature_name in feature_importance]\n",
+    "\n",
+    "# Plot the feature importances\n",
+    "plt.figure(figsize=(8, 5))\n",
+    "plt.barh(features, importances)\n",
+    "plt.xlabel('Importance Score')\n",
+    "plt.ylabel('Features')\n",
+    "plt.title('Feature Importances')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5a63a93d",
+   "metadata": {},
+   "source": [
+    "#### Method 2. Lasso regression model classifier\n",
+    "Implementation of a feature importance analysis using the ***Lasso regression model classifier*** in scikit-learn.\n",
+    "This method  L1 regularization, work better when the features are on a similar scale.Hence scaling is performed before doing feature selection."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 99,
+   "id": "a7522491",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from sklearn.linear_model import Lasso\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Create a scaler object\n",
+    "scaler = StandardScaler()\n",
+    "\n",
+    "# Scale the features\n",
+    "X_scaled = scaler.fit_transform(train_copy.iloc[:, :-1])\n",
+    "\n",
+    "# Create a Lasso model\n",
+    "lasso = Lasso(alpha=0.1)\n",
+    "\n",
+    "# Fit the model to the scaled data\n",
+    "lasso.fit(X_scaled, train_copy.smoking)\n",
+    "\n",
+    "# Get the coefficients and corresponding feature names\n",
+    "coefficients = lasso.coef_\n",
+    "feature_names = train_copy.columns[:-1]\n",
+    "\n",
+    "    # Create a bar plot for feature importance\n",
+    "plt.figure(figsize=(8, 4))\n",
+    "plt.bar(feature_names, coefficients)\n",
+    "plt.xticks(rotation=90)\n",
+    "plt.xlabel('Features')\n",
+    "plt.ylabel('Coefficient')\n",
+    "plt.title('Feature Importance based on Lasso Coefficients')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2bd40058",
+   "metadata": {},
+   "source": [
+    "From Both Model1 and Model2,It can be seen that ***height(cm), hemoglobin, Gtp*** are the most important features for this dataset."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1e85e1b0",
+   "metadata": {},
+   "source": [
+    "### 5. MODEL TRAINING"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "9d6d31b7",
+   "metadata": {},
+   "source": [
+    "#### Step1 Standardization process\n",
+    "It is important to standardize features around the center and 0, with a standard deviation of 1, when comparing measurements that have different units. This helps address the issue of variables being measured on different scales, which can lead to unequal contributions in the analysis and potential bias. For instance, a variable with a range of 0 to 1000 would overshadow a variable with a range of 0 to 1 if not standardized. By transforming the data to comparable scales, can avoid this problem. Common data standardization techniques aim to equalize the range and/or variability of the data."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 167,
+   "id": "de2a405c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "scaler = StandardScaler()\n",
+    "scaler.fit(train_copy.iloc[:, :-1])\n",
+    "X_scaled = scaler.transform(train_copy.iloc[:, :-1])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "04b3377d",
+   "metadata": {},
+   "source": [
+    "#### Step2 Balancing Dataset\n",
+    "As the dataset is imbalanced, before performing Training dataset needs to be balanced.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 168,
+   "id": "175c9a28",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Training Set Class before Balance: \n",
+      " 0    20413\n",
+      "1    10956\n",
+      "Name: smoking, dtype: int64\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Show the classes balance in the training set\n",
+    "print('Training Set Class before Balance: \\n', train_copy.smoking.value_counts())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1a555364",
+   "metadata": {},
+   "source": [
+    "RandomOverSampler is a technique used for handling imbalanced datasets by randomly oversampling the minority class samples. The sampling_strategy parameter determines the ratio of the majority class to the minority class after resampling. In this case, \"auto\" indicates that the ratio should be automatically determined."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 169,
+   "id": "ae232d8e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from imblearn.over_sampling import RandomOverSampler"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 170,
+   "id": "cdfd672a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Create an instance of RandomOverSampler\n",
+    "ros = RandomOverSampler(sampling_strategy=\"auto\", random_state=11)\n",
+    "\n",
+    "# Resample the training data\n",
+    "x_rovs, y_rovs = ros.fit_resample(X_scaled, train_copy[\"smoking\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 171,
+   "id": "26406207",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Training Set Class after Balance: \n",
+      " 1    20413\n",
+      "0    20413\n",
+      "Name: smoking, dtype: int64\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Show the classes balance in the training set\n",
+    "print('Training Set Class after Balance: \\n', y_rovs.value_counts())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f163df40",
+   "metadata": {},
+   "source": [
+    "#### Step3 Splitting the dataset as training set and testing set."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 172,
+   "id": "6c206f15",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.model_selection import train_test_split\n",
+    "import numpy as np\n",
+    "\n",
+    "# Set the random seed for reproducibility\n",
+    "random_seed = 42\n",
+    "np.random.seed(random_seed)\n",
+    "\n",
+    "# Splitting the dataset into train and test sets\n",
+    "X_train, X_test, y_train, y_test = train_test_split(x_rovs, y_rovs, test_size=0.2, random_state=42)\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "409199ab",
+   "metadata": {},
+   "source": [
+    "#### Step4 Training Model "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "91c32798",
+   "metadata": {},
+   "source": [
+    "With a dataset size of 38,984 rows and 23 columns, and all input features being numeric, This is a suitable dataset for applying various binary classification models. Here are a few modeling techniques commonly used for binary classification tasks with numerical input features:\n",
+    "\n",
+    "> ***Logistic Regression:*** This is a popular and interpretable model that estimates the probabilities of belonging to each class using a logistic function.\n",
+    "\n",
+    "> ***Support Vector Machines (SVM):*** SVMs aim to find a hyperplane that maximally separates the data points of different classes in a high-dimensional space.\n",
+    "\n",
+    "> ***Random Forest:*** This ensemble model consists of multiple decision trees and can handle both numerical and categorical features. It provides good predictive performance and feature importance rankings.\n",
+    "\n",
+    "> ***Gradient Boosting Algorithms:*** Models like XGBoost or LightGBM are gradient boosting algorithms that sequentially train weak classifiers and combine their predictions to improve overall accuracy.\n",
+    "\n",
+    "It's a good idea to try out different models and evaluate their performance using appropriate evaluation metrics like accuracy, precision, recall, and F1 score. Additionally,Lets consider using techniques like cross-validation to assess the generalization ability of the models and tune hyperparameters for optimal performance."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1dfbf937",
+   "metadata": {},
+   "source": [
+    "### 6. EVALUATION"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "74837a36",
+   "metadata": {},
+   "source": [
+    "In the code below,  first initialize the models - Logistic Regression, Support Vector Machines (SVM), Random Forest, and XGBoost - with their respective parameter settings. Then, iterate over the models and use the cross_val_score() function to perform cross-validation with 5 folds (cv=5) and evaluate the models based on accuracy (scoring='accuracy'). The mean accuracy across all cross-validation folds is calculated and printed for each model."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 140,
+   "id": "75236aa4",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Logistic Regression Cross-Validation Accuracy: 0.7368952847519902\n",
+      "Support Vector Machines Cross-Validation Accuracy: 0.7616962645437845\n",
+      "Random Forest Cross-Validation Accuracy: 0.7847826086956522\n",
+      "XGBoost Cross-Validation Accuracy: 0.78162890385793\n",
+      "Logistic Regression Test Accuracy: 0.7384276267450404\n",
+      "Support Vector Machines Test Accuracy: 0.7647563066372766\n",
+      "Random Forest Test Accuracy: 0.7893705608621112\n",
+      "XGBoost Test Accuracy: 0.7871662992897379\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.model_selection import cross_val_score\n",
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn.svm import SVC\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from xgboost import XGBClassifier\n",
+    "\n",
+    "# Initialize the models\n",
+    "logreg = LogisticRegression(max_iter=10000, solver='sag')\n",
+    "svm = SVC()\n",
+    "random_forest = RandomForestClassifier(n_estimators=150, max_depth=15, min_samples_split=2, min_samples_leaf=5)\n",
+    "xgb = XGBClassifier()\n",
+    "\n",
+    "# Perform cross-validation and evaluate models\n",
+    "models = [logreg, svm, random_forest, xgb]\n",
+    "model_names = [\"Logistic Regression\", \"Support Vector Machines\", \"Random Forest\", \"XGBoost\"]\n",
+    "\n",
+    "for model, name in zip(models, model_names):\n",
+    "    scores = cross_val_score(model, X_train, y_train, cv=5, scoring='accuracy')\n",
+    "    mean_accuracy = scores.mean()\n",
+    "    print(f\"{name} Cross-Validation Accuracy: {mean_accuracy}\")\n",
+    "\n",
+    "# Train and evaluate the models on the test set\n",
+    "for model, name in zip(models, model_names):\n",
+    "    model.fit(X_train, y_train)\n",
+    "    test_accuracy = model.score(X_test, y_test)\n",
+    "    print(f\"{name} Test Accuracy: {test_accuracy}\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 144,
+   "id": "60d0a343",
+   "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>Model</th>\n",
+       "      <th>Cross-Validation Accuracy</th>\n",
+       "      <th>Test Accuracy</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Logistic Regression</td>\n",
+       "      <td>0.736895</td>\n",
+       "      <td>0.738428</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Support Vector Machines</td>\n",
+       "      <td>0.761696</td>\n",
+       "      <td>0.764756</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Random Forest</td>\n",
+       "      <td>0.784783</td>\n",
+       "      <td>0.789371</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>XGBoost</td>\n",
+       "      <td>0.781629</td>\n",
+       "      <td>0.787166</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                     Model  Cross-Validation Accuracy  Test Accuracy\n",
+       "0      Logistic Regression                   0.736895       0.738428\n",
+       "1  Support Vector Machines                   0.761696       0.764756\n",
+       "2            Random Forest                   0.784783       0.789371\n",
+       "3                  XGBoost                   0.781629       0.787166"
+      ]
+     },
+     "execution_count": 144,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "# Define the accuracy values\n",
+    "data = {\n",
+    "    'Model': ['Logistic Regression', 'Support Vector Machines', 'Random Forest', 'XGBoost'],\n",
+    "    'Cross-Validation Accuracy': [0.7368952847519902, 0.7616962645437845, 0.7847826086956522, 0.78162890385793],\n",
+    "    'Test Accuracy': [0.7384276267450404, 0.7647563066372766, 0.7893705608621112, 0.7871662992897379]\n",
+    "}\n",
+    "\n",
+    "# Create a DataFrame from the data\n",
+    "df = pd.DataFrame(data)\n",
+    "\n",
+    "# Display the DataFrame\n",
+    "df\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "4aea77cd",
+   "metadata": {},
+   "source": [
+    "From the provided accuracy values,Lets draw several insights about the performance of the models:\n",
+    "\n",
+    "***Cross-Validation Accuracy:***\n",
+    "\n",
+    "Logistic Regression: 0.74\n",
+    "Support Vector Machines: 0.76\n",
+    "Random Forest: 0.784\n",
+    "XGBoost: 0.781\n",
+    "The cross-validation accuracy gives us an estimate of how well the models perform on unseen data. Based on these values, It can be observed that the Random Forest model has the highest cross-validation accuracy (0.784), indicating good generalization performance. The XGBoost model also performs well with a cross-validation accuracy of 0.781. The Logistic Regression and Support Vector Machines models have lower cross-validation accuracies, but still show reasonable performance.\n",
+    "\n",
+    "***Test Accuracy:***\n",
+    "\n",
+    "Logistic Regression: 0.74\n",
+    "Support Vector Machines: 0.76\n",
+    "Random Forest: 0.789\n",
+    "XGBoost: 0.787\n",
+    "The test accuracy represents the performance of the models on a separate, independent dataset. It provides a measure of how well the models generalize to unseen data. From the test accuracy values, It can observed that the Random Forest model achieves the highest accuracy (0.789), indicating good performance on new data. The Support Vector Machines and XGBoost models also show competitive test accuracies of 0.76 and 0.787, respectively. The Logistic Regression model has the lowest test accuracy but still performs reasonably well at 0.74.\n",
+    "\n",
+    "Based on these insights, It can concluded that the ***Random Forest model*** appears to be the most promising in terms of both cross-validation and test accuracy. "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "dc020a50",
+   "metadata": {},
+   "source": [
+    "### 7. PARAMETER TUNING"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "cfde57dd",
+   "metadata": {},
+   "source": [
+    "To tune hyperparameters, Lets use techniques such as grid search or randomized search in combination with cross-validation. These techniques involve systematically searching through different combinations of hyperparameter values and evaluating the model's performance using cross-validation.\n",
+    "As Random Forest accuracy was highest in model training ,So using parameter tuning to increase its performance and accoracy."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 130,
+   "id": "bf54236e",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Best Hyperparameters: {'random_forest__max_depth': None, 'random_forest__min_samples_leaf': 1, 'random_forest__min_samples_split': 2, 'random_forest__n_estimators': 200}\n",
+      "Best Cross-Validation Score: 0.8164740150642356\n",
+      "Test Accuracy: 0.8269654665686995\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.model_selection import GridSearchCV\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from imblearn.over_sampling import RandomOverSampler\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "from sklearn.pipeline import Pipeline\n",
+    "\n",
+    "# Define the parameter grid for the Random Forest classifier\n",
+    "param_grid = {\n",
+    "    'random_forest__n_estimators': [100, 200, 300],\n",
+    "    'random_forest__max_depth': [None, 5, 10, 15],\n",
+    "    'random_forest__min_samples_split': [2, 5, 10],\n",
+    "    'random_forest__min_samples_leaf': [1, 3, 5]\n",
+    "}\n",
+    "\n",
+    "# Initialize the Random Forest classifier\n",
+    "random_forest = RandomForestClassifier()\n",
+    "\n",
+    "# Create a pipeline with scaling\n",
+    "pipeline = Pipeline([\n",
+    "    ('scaler', StandardScaler()),\n",
+    "    ('random_forest', random_forest)\n",
+    "])\n",
+    "\n",
+    "# Perform oversampling on the training data\n",
+    "oversampler = RandomOverSampler(random_state=42)\n",
+    "X_train_oversampled, y_train_oversampled = oversampler.fit_resample(X_train, y_train)\n",
+    "\n",
+    "# Perform grid search with cross-validation\n",
+    "grid_search = GridSearchCV(pipeline, param_grid, cv=5, scoring='accuracy', n_jobs=-1)\n",
+    "grid_search.fit(X_train_oversampled, y_train_oversampled)\n",
+    "\n",
+    "# Print the best hyperparameters and the corresponding cross-validation score\n",
+    "print(\"Best Hyperparameters:\", grid_search.best_params_)\n",
+    "print(\"Best Cross-Validation Score:\", grid_search.best_score_)\n",
+    "\n",
+    "# Evaluate the model on the test set\n",
+    "test_accuracy = grid_search.score(X_test, y_test)\n",
+    "print(\"Test Accuracy:\", test_accuracy)\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "42e3bc9c",
+   "metadata": {},
+   "source": [
+    "Now lets Predict the Target variable output for ***test data set***"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 173,
+   "id": "1a1244a1",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>age</th>\n",
+       "      <th>height(cm)</th>\n",
+       "      <th>weight(kg)</th>\n",
+       "      <th>waist(cm)</th>\n",
+       "      <th>eyesight(left)</th>\n",
+       "      <th>eyesight(right)</th>\n",
+       "      <th>hearing(left)</th>\n",
+       "      <th>hearing(right)</th>\n",
+       "      <th>systolic</th>\n",
+       "      <th>relaxation</th>\n",
+       "      <th>...</th>\n",
+       "      <th>triglyceride</th>\n",
+       "      <th>HDL</th>\n",
+       "      <th>LDL</th>\n",
+       "      <th>hemoglobin</th>\n",
+       "      <th>Urine protein</th>\n",
+       "      <th>serum creatinine</th>\n",
+       "      <th>AST</th>\n",
+       "      <th>ALT</th>\n",
+       "      <th>Gtp</th>\n",
+       "      <th>dental caries</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>40</td>\n",
+       "      <td>170</td>\n",
+       "      <td>65</td>\n",
+       "      <td>75.1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>120</td>\n",
+       "      <td>70</td>\n",
+       "      <td>...</td>\n",
+       "      <td>260</td>\n",
+       "      <td>41</td>\n",
+       "      <td>132</td>\n",
+       "      <td>15.7</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>24</td>\n",
+       "      <td>26</td>\n",
+       "      <td>32</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>45</td>\n",
+       "      <td>170</td>\n",
+       "      <td>75</td>\n",
+       "      <td>89.0</td>\n",
+       "      <td>0.7</td>\n",
+       "      <td>1.2</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>100</td>\n",
+       "      <td>67</td>\n",
+       "      <td>...</td>\n",
+       "      <td>345</td>\n",
+       "      <td>49</td>\n",
+       "      <td>140</td>\n",
+       "      <td>15.7</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.1</td>\n",
+       "      <td>26</td>\n",
+       "      <td>28</td>\n",
+       "      <td>138</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>30</td>\n",
+       "      <td>180</td>\n",
+       "      <td>90</td>\n",
+       "      <td>94.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>115</td>\n",
+       "      <td>72</td>\n",
+       "      <td>...</td>\n",
+       "      <td>103</td>\n",
+       "      <td>53</td>\n",
+       "      <td>103</td>\n",
+       "      <td>13.5</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>19</td>\n",
+       "      <td>29</td>\n",
+       "      <td>30</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>60</td>\n",
+       "      <td>170</td>\n",
+       "      <td>50</td>\n",
+       "      <td>73.0</td>\n",
+       "      <td>0.5</td>\n",
+       "      <td>0.7</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>118</td>\n",
+       "      <td>78</td>\n",
+       "      <td>...</td>\n",
+       "      <td>70</td>\n",
+       "      <td>65</td>\n",
+       "      <td>108</td>\n",
+       "      <td>14.1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.3</td>\n",
+       "      <td>31</td>\n",
+       "      <td>28</td>\n",
+       "      <td>33</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>30</td>\n",
+       "      <td>170</td>\n",
+       "      <td>65</td>\n",
+       "      <td>78.0</td>\n",
+       "      <td>1.5</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>110</td>\n",
+       "      <td>70</td>\n",
+       "      <td>...</td>\n",
+       "      <td>210</td>\n",
+       "      <td>45</td>\n",
+       "      <td>103</td>\n",
+       "      <td>14.7</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>21</td>\n",
+       "      <td>21</td>\n",
+       "      <td>19</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16703</th>\n",
+       "      <td>60</td>\n",
+       "      <td>165</td>\n",
+       "      <td>65</td>\n",
+       "      <td>82.0</td>\n",
+       "      <td>0.7</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>101</td>\n",
+       "      <td>68</td>\n",
+       "      <td>...</td>\n",
+       "      <td>131</td>\n",
+       "      <td>41</td>\n",
+       "      <td>110</td>\n",
+       "      <td>13.5</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>78</td>\n",
+       "      <td>75</td>\n",
+       "      <td>33</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16704</th>\n",
+       "      <td>60</td>\n",
+       "      <td>155</td>\n",
+       "      <td>70</td>\n",
+       "      <td>93.0</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>134</td>\n",
+       "      <td>70</td>\n",
+       "      <td>...</td>\n",
+       "      <td>259</td>\n",
+       "      <td>53</td>\n",
+       "      <td>60</td>\n",
+       "      <td>13.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.7</td>\n",
+       "      <td>19</td>\n",
+       "      <td>28</td>\n",
+       "      <td>28</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16705</th>\n",
+       "      <td>40</td>\n",
+       "      <td>155</td>\n",
+       "      <td>50</td>\n",
+       "      <td>67.2</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>0.8</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>134</td>\n",
+       "      <td>80</td>\n",
+       "      <td>...</td>\n",
+       "      <td>50</td>\n",
+       "      <td>64</td>\n",
+       "      <td>131</td>\n",
+       "      <td>13.4</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0.7</td>\n",
+       "      <td>16</td>\n",
+       "      <td>10</td>\n",
+       "      <td>14</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16706</th>\n",
+       "      <td>35</td>\n",
+       "      <td>165</td>\n",
+       "      <td>70</td>\n",
+       "      <td>76.1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>114</td>\n",
+       "      <td>68</td>\n",
+       "      <td>...</td>\n",
+       "      <td>43</td>\n",
+       "      <td>74</td>\n",
+       "      <td>118</td>\n",
+       "      <td>14.3</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.2</td>\n",
+       "      <td>19</td>\n",
+       "      <td>28</td>\n",
+       "      <td>30</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16707</th>\n",
+       "      <td>25</td>\n",
+       "      <td>180</td>\n",
+       "      <td>80</td>\n",
+       "      <td>87.0</td>\n",
+       "      <td>1.2</td>\n",
+       "      <td>0.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>110</td>\n",
+       "      <td>70</td>\n",
+       "      <td>...</td>\n",
+       "      <td>107</td>\n",
+       "      <td>53</td>\n",
+       "      <td>86</td>\n",
+       "      <td>15.9</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1.1</td>\n",
+       "      <td>23</td>\n",
+       "      <td>27</td>\n",
+       "      <td>23</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>16708 rows × 22 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       age  height(cm)  weight(kg)  waist(cm)  eyesight(left)  \\\n",
+       "0       40         170          65       75.1             1.0   \n",
+       "1       45         170          75       89.0             0.7   \n",
+       "2       30         180          90       94.0             1.0   \n",
+       "3       60         170          50       73.0             0.5   \n",
+       "4       30         170          65       78.0             1.5   \n",
+       "...    ...         ...         ...        ...             ...   \n",
+       "16703   60         165          65       82.0             0.7   \n",
+       "16704   60         155          70       93.0             0.8   \n",
+       "16705   40         155          50       67.2             0.9   \n",
+       "16706   35         165          70       76.1             1.0   \n",
+       "16707   25         180          80       87.0             1.2   \n",
+       "\n",
+       "       eyesight(right)  hearing(left)  hearing(right)  systolic  relaxation  \\\n",
+       "0                  0.9              1               1       120          70   \n",
+       "1                  1.2              1               1       100          67   \n",
+       "2                  0.8              1               1       115          72   \n",
+       "3                  0.7              1               1       118          78   \n",
+       "4                  1.0              1               1       110          70   \n",
+       "...                ...            ...             ...       ...         ...   \n",
+       "16703              1.0              1               1       101          68   \n",
+       "16704              1.0              1               1       134          70   \n",
+       "16705              0.8              1               1       134          80   \n",
+       "16706              1.0              1               1       114          68   \n",
+       "16707              0.9              1               1       110          70   \n",
+       "\n",
+       "       ...  triglyceride  HDL  LDL  hemoglobin  Urine protein  \\\n",
+       "0      ...           260   41  132        15.7              1   \n",
+       "1      ...           345   49  140        15.7              1   \n",
+       "2      ...           103   53  103        13.5              1   \n",
+       "3      ...            70   65  108        14.1              1   \n",
+       "4      ...           210   45  103        14.7              1   \n",
+       "...    ...           ...  ...  ...         ...            ...   \n",
+       "16703  ...           131   41  110        13.5              1   \n",
+       "16704  ...           259   53   60        13.9              1   \n",
+       "16705  ...            50   64  131        13.4              1   \n",
+       "16706  ...            43   74  118        14.3              1   \n",
+       "16707  ...           107   53   86        15.9              1   \n",
+       "\n",
+       "       serum creatinine  AST  ALT  Gtp  dental caries  \n",
+       "0                   0.8   24   26   32              0  \n",
+       "1                   1.1   26   28  138              0  \n",
+       "2                   1.0   19   29   30              0  \n",
+       "3                   1.3   31   28   33              0  \n",
+       "4                   0.8   21   21   19              0  \n",
+       "...                 ...  ...  ...  ...            ...  \n",
+       "16703               0.8   78   75   33              0  \n",
+       "16704               0.7   19   28   28              1  \n",
+       "16705               0.7   16   10   14              0  \n",
+       "16706               1.2   19   28   30              1  \n",
+       "16707               1.1   23   27   23              0  \n",
+       "\n",
+       "[16708 rows x 22 columns]"
+      ]
+     },
+     "execution_count": 173,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Reading dataset using pandas library\n",
+    "test_data = pd.read_csv(\"E:/MA336 - AI and ML/Project_AI/test_dataset.csv\")\n",
+    "test_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 174,
+   "id": "b4f3c39a",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Random Forest Accuracy: 0.8268430075924565\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Random Forest\n",
+    "rf = RandomForestClassifier(n_estimators=200, max_depth=None, min_samples_split=2, min_samples_leaf=1)\n",
+    "rf.fit(X_train, y_train)\n",
+    "rf_predictions = rf.predict(X_test)\n",
+    "rf_accuracy = accuracy_score(y_test, rf_predictions)\n",
+    "print(\"Random Forest Accuracy:\", rf_accuracy)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 182,
+   "id": "e486cd54",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Input Features: [ 40.  170.   65.   75.1   1.    0.9   1.    1.  120.   70.  102.  225.\n",
+      " 260.   41.  132.   15.7   1.    0.8  24.   26.   32.    0. ]\n",
+      "Predicted Value: 1\n",
+      "\n",
+      "Input Features: [ 45.  170.   75.   89.    0.7   1.2   1.    1.  100.   67.   96.  258.\n",
+      " 345.   49.  140.   15.7   1.    1.1  26.   28.  138.    0. ]\n",
+      "Predicted Value: 1\n",
+      "\n",
+      "Input Features: [ 30.  180.   90.   94.    1.    0.8   1.    1.  115.   72.   88.  177.\n",
+      " 103.   53.  103.   13.5   1.    1.   19.   29.   30.    0. ]\n",
+      "Predicted Value: 0\n",
+      "\n",
+      "Input Features: [ 60.  170.   50.   73.    0.5   0.7   1.    1.  118.   78.   86.  187.\n",
+      "  70.   65.  108.   14.1   1.    1.3  31.   28.   33.    0. ]\n",
+      "Predicted Value: 0\n",
+      "\n",
+      "Input Features: [ 30.  170.   65.   78.    1.5   1.    1.    1.  110.   70.   87.  190.\n",
+      " 210.   45.  103.   14.7   1.    0.8  21.   21.   19.    0. ]\n",
+      "Predicted Value: 0\n",
+      "\n",
+      "Input Features: [ 55.  175.   60.   75.    1.    1.    1.    1.  100.   64.   93.  186.\n",
+      "  80.   86.   84.   15.4   3.    1.   39.   20.   35.    0. ]\n",
+      "Predicted Value: 1\n",
+      "\n",
+      "Input Features: [ 40.  160.   55.   69.    1.5   1.5   1.    1.  112.   78.   90.  177.\n",
+      "  68.   78.   85.   12.4   1.    0.5  15.    9.   14.    0. ]\n",
+      "Predicted Value: 0\n",
+      "\n",
+      "Input Features: [ 55.  175.   60.   80.    1.2   1.5   1.    1.  137.   89.   80.  199.\n",
+      "  35.   68.  124.   16.    1.    1.1  23.   19.   17.    0. ]\n",
+      "Predicted Value: 1\n",
+      "\n",
+      "Input Features: [ 55.  160.   50.   68.    0.8   0.5   1.    1.  137.   87.   90.  176.\n",
+      "  36.   67.  102.   13.6   1.    0.7  15.   14.   13.    0. ]\n",
+      "Predicted Value: 0\n",
+      "\n",
+      "Input Features: [ 75.  145.   50.   81.    0.5   0.5   2.    2.  148.   86.  121.  192.\n",
+      " 109.   81.   89.   14.    1.    0.6  28.   24.   17.    1. ]\n",
+      "Predicted Value: 0\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "import numpy as np\n",
+    "\n",
+    "# Training data\n",
+    "X_tr = train_data.iloc[:, :-1]\n",
+    "y_tr = train_data.smoking\n",
+    "\n",
+    "# Test data\n",
+    "X_test = test_data\n",
+    "\n",
+    "# Create a Random Forest classifier\n",
+    "rf = RandomForestClassifier(n_estimators=200, max_depth=None, min_samples_split=2, min_samples_leaf=1)\n",
+    "\n",
+    "# Train the model\n",
+    "rf.fit(X_tr, y_tr)\n",
+    "\n",
+    "# Make predictions on the test data\n",
+    "predictions = rf.predict(X_test)\n",
+    "\n",
+    "# Set the number of rows to print\n",
+    "num_rows = 10\n",
+    "\n",
+    "# Print the predicted values along with the input features of the test data\n",
+    "for features, prediction in zip(test_data.values[:num_rows], predictions[:num_rows]):\n",
+    "    print(\"Input Features:\", features)\n",
+    "    print(\"Predicted Value:\", prediction)\n",
+    "    print()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "78de9440",
+   "metadata": {},
+   "source": [
+    "From above it can be seen preicted values for each observation. Only 10 rows are printed to reduce scrolling pages. By increasing num_rows , more predictions can be seen."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "b9e6a181",
+   "metadata": {},
+   "source": [
+    "***Results***\n",
+    "\n",
+    ">The developed predictive model underwent rigorous evaluation, validation, and parameter tuning. Employed various machine learning algorithms, including Logistic Regression, Support Vector Machines, Random Forest, and XGBoost, to predict smoking status based on bio-signals. Through cross-validation, obtained reliable estimates of the models' performance on unseen data. The Random Forest model initially demonstrated the highest accuracy, but through parameter tuning using techniques such as grid search, further optimized its performance.\n",
+    "\n",
+    ">After fine-tuning the hyperparameters, the Random Forest model achieved even higher accuracy. The best hyperparameters for the model were found to be 'max_depth': None, 'min_samples_leaf': 1, 'min_samples_split': 2, and 'n_estimators': 200. With these optimized hyperparameters, the model achieved a cross-validation accuracy of 81.6% and a test accuracy of 82.7%.\n",
+    "\n",
+    ">The parameter tuning process played a crucial role in improving the model's performance by finding the optimal combination of hyperparameters. By fine-tuning the Random Forest model,  were able to enhance its predictive power, resulting in higher accuracy and improved reliability for identifying smoking status based on bio-signals."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "b7ebaa85",
+   "metadata": {},
+   "source": [
+    "### 8. CONCLUSIONS"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5b4fbed4",
+   "metadata": {},
+   "source": [
+    "In conclusion, study successfully developed a machine learning model that utilizes bio-signals to predict an individual's smoking status. The integration of AI and machine learning techniques, coupled with parameter tuning, has significantly enhanced the model's performance and accuracy. The Random Forest model, after careful hyperparameter optimization, emerged as the most promising model, achieving a cross-validation accuracy of 81.6% and a test accuracy of 82.7%. These results demonstrate the effectiveness of  predictive model in accurately identifying smoking status and provide valuable insights for healthcare professionals in assessing and addressing smoking behaviors. By leveraging bio-signals and advanced machine learning techniques,  contribute to the development of reliable tools for smoking cessation strategies, ultimately leading to improved public health outcomes and a reduction in the harmful effects of smoking"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2c15aa32",
+   "metadata": {},
+   "source": [
+    "### 9. REFERENCES"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "41ee617e",
+   "metadata": {},
+   "source": [
+    "https://www.kaggle.com/datasets/gauravduttakiit/smoker-status-prediction"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}