{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The question\n",
    "How can the analysis of various socio-economic and environmental factors, like crime rate, land zoning, business proportion, pollution levels, housing features, and educational resources, accurately predict the market values of houses in Boston?\n",
    "\n",
    "This question seeks to understand the impact of a wide range of factors on house values, aiming to develop a comprehensive model that accounts for the multifaceted nature of real estate valuation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import necessary libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.utils import resample\n",
    "from numpy.linalg import LinAlgError"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set a seed for reproducibility of random operations like train-test split\n",
    "seed = 22980254\n",
    "\n",
    "# Load the dataset into a pandas DataFrame\n",
    "df = pd.read_csv('housing_data.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "# Perform Exploratory Data Analysis (EDA)\n",
    "\n",
    "Perform an exploratory analysis on a broader set of variables relevant to the research question."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 506 entries, 0 to 505\n",
      "Data columns (total 12 columns):\n",
      " #   Column   Non-Null Count  Dtype  \n",
      "---  ------   --------------  -----  \n",
      " 0   crim     506 non-null    float64\n",
      " 1   zn       506 non-null    float64\n",
      " 2   indus    506 non-null    float64\n",
      " 3   nox      506 non-null    float64\n",
      " 4   rm       506 non-null    float64\n",
      " 5   age      506 non-null    float64\n",
      " 6   dis      506 non-null    float64\n",
      " 7   rad      506 non-null    int64  \n",
      " 8   tax      506 non-null    int64  \n",
      " 9   ptratio  506 non-null    float64\n",
      " 10  lstat    506 non-null    float64\n",
      " 11  medv     506 non-null    float64\n",
      "dtypes: float64(10), int64(2)\n",
      "memory usage: 47.6 KB\n"
     ]
    }
   ],
   "source": [
    "# Basic information about the dataset\n",
    "data_info = df.info()\n",
    "\n",
    "# Initial exploration of specified variables\n",
    "exploration = df[['medv', 'rm', 'rad']].describe()\n",
    "\n",
    "# Checking for missing values\n",
    "missing_values = df[['medv', 'rm', 'rad']].isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualizing the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Histograms\n",
    "plt.figure(figsize=(15, 5))\n",
    "plt.subplot(1, 3, 1)\n",
    "sns.histplot(df['medv'], kde=True)\n",
    "plt.title('Distribution of MEDV')\n",
    "\n",
    "plt.subplot(1, 3, 2)\n",
    "sns.histplot(df['rm'], kde=True)\n",
    "plt.title('Distribution of RM')\n",
    "\n",
    "plt.subplot(1, 3, 3)\n",
    "sns.histplot(df['rad'], kde=True, bins=24)\n",
    "plt.title('Distribution of RAD')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Scatter plots\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.subplot(1, 2, 1)\n",
    "sns.scatterplot(x='rm', y='medv', data=df)\n",
    "plt.title('MEDV vs. RM')\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "sns.scatterplot(x='rad', y='medv', data=df)\n",
    "plt.title('MEDV vs. RAD')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          medv        rm       rad\n",
      "medv  1.000000  0.695360 -0.381626\n",
      "rm    0.695360  1.000000 -0.209847\n",
      "rad  -0.381626 -0.209847  1.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABBcAAANECAYAAADrAcfmAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAABcAElEQVR4nO3deXRU9fk/8GcSQkICYZGyqiCiIAVRsUVUClYUcflWrfsCUrW1lbaKS8WquFVaF75Yq2KtVu1P6v6tbbW4AV2UukCxYgU3FEVBFgHZEkju7w9rNBIUcsMkDK/XOXOOc+femedeOJG853nuJ5MkSRIAAAAAtZRX3wUAAAAAWzbhAgAAAJCKcAEAAABIRbgAAAAApCJcAAAAAFIRLgAAAACpCBcAAACAVIQLAAAAQCrCBQAAACAV4QIAm+SOO+6ITCYTb731Vp2951tvvRWZTCbuuOOOOnvPLd3AgQNj4MCB9V0GAMBGES4ANABvvPFGfO9734suXbpEUVFRlJaWxj777BPXX399rF69ur7LqzMTJkyIcePG1XcZ1ZxyyimRyWSitLS0xmv92muvRSaTiUwmE9dee+0mv/97770Xl156acyYMaMOqq29TCYTI0aMqPG1TwKjF154YbN9fkO5DgDA5tGovgsA2No98sgjcfTRR0dhYWEMHTo0evbsGeXl5fGPf/wjzjvvvHj55Zfj17/+dX2XWScmTJgQM2fOjLPOOqva9k6dOsXq1aujoKCgXupq1KhRrFq1Kv70pz/FMcccU+21u+++O4qKimLNmjW1eu/33nsvLrvssujcuXPstttuG33c448/XqvPa6hqex0AgC2DcAGgHs2ZMyeOO+646NSpU0yaNCnat29f9dqZZ54Zr7/+ejzyyCOpPydJklizZk00adJkvdfWrFkTjRs3jry8+mtmy2QyUVRUVG+fX1hYGPvss0/8/ve/Xy9cmDBhQhxyyCHx4IMPZqWWVatWRXFxcTRu3DgrnwcAUBeMRQDUo6uvvjpWrFgRt912W7Vg4RNdu3aNH//4x1XP161bF1dccUXsuOOOUVhYGJ07d44LL7wwysrKqh3XuXPnOPTQQ+Oxxx6LPffcM5o0aRK33HJLTJkyJTKZTNxzzz1x0UUXRceOHaO4uDiWL18eERHPPvtsHHTQQdG8efMoLi6OAQMGxNNPP/2l5/Hwww/HIYccEh06dIjCwsLYcccd44orroiKioqqfQYOHBiPPPJIvP3221VjBp07d46IDd9zYdKkSdG/f/8oKSmJFi1axLe+9a145ZVXqu1z6aWXRiaTiddffz1OOeWUaNGiRTRv3jyGDx8eq1at+tLaP3HCCSfEX/7yl1i6dGnVtueffz5ee+21OOGEE9bbf8mSJXHuuedGr169omnTplFaWhpDhgyJF198sWqfKVOmxNe+9rWIiBg+fHjVeX9yngMHDoyePXvGtGnT4hvf+EYUFxfHhRdeWPXaZ++5MGzYsCgqKlrv/AcPHhwtW7aM9957b6PPdWPNmjUrjjrqqGjVqlUUFRXFnnvuGX/84x8323X497//HQMGDIji4uLo2rVrPPDAAxER8de//jX69u0bTZo0iW7dusWTTz5ZrYa33347fvCDH0S3bt2iSZMmsc0228TRRx+93n1BPhn/+Nvf/hbf+973YptttonS0tIYOnRofPjhh3V89QBg66JzAaAe/elPf4ouXbrE3nvvvVH7n3baaXHnnXfGUUcdFeecc048++yzMWbMmHjllVfi//7v/6rtO3v27Dj++OPje9/7Xpx++unRrVu3qteuuOKKaNy4cZx77rlRVlYWjRs3jkmTJsWQIUOiT58+MXr06MjLy4vf/va38c1vfjP+/ve/x9e//vUN1nXHHXdE06ZNY+TIkdG0adOYNGlSXHLJJbF8+fK45pprIiLipz/9aSxbtizefffd+N///d+IiGjatOkG3/PJJ5+MIUOGRJcuXeLSSy+N1atXxw033BD77LNPTJ8+vSqY+MQxxxwTO+ywQ4wZMyamT58ev/nNb6JNmzbxi1/8YqOu7ZFHHhlnnHFGPPTQQ/Gd73wnIj7uWujevXvsscce6+3/5ptvxh/+8Ic4+uijY4cddogFCxbELbfcEgMGDIj//Oc/0aFDh9hll13i8ssvj0suuSS++93vRv/+/SMiqv15L168OIYMGRLHHXdcnHTSSdG2bdsa67v++utj0qRJMWzYsJg6dWrk5+fHLbfcEo8//nj87ne/iw4dOnzpOa5ZsyYWLVq03vYVK1ast+3ll1+OffbZJzp27BgXXHBBlJSUxH333ReHH354PPjgg3HEEUfU6XX48MMP49BDD43jjjsujj766Lj55pvjuOOOi7vvvjvOOuusOOOMM+KEE06Ia665Jo466qh45513olmzZhHxcQj0zDPPxHHHHRfbbrttvPXWW3HzzTfHwIED4z//+U8UFxdXO7cRI0ZEixYt4tJLL43Zs2fHzTffHG+//XZV+AYA1EICQL1YtmxZEhHJt771rY3af8aMGUlEJKeddlq17eeee24SEcmkSZOqtnXq1CmJiGTixInV9p08eXISEUmXLl2SVatWVW2vrKxMdtppp2Tw4MFJZWVl1fZVq1YlO+ywQ3LAAQdUbfvtb3+bREQyZ86cavt93ve+972kuLg4WbNmTdW2Qw45JOnUqdN6+86ZMyeJiOS3v/1t1bbddtstadOmTbJ48eKqbS+++GKSl5eXDB06tGrb6NGjk4hIvvOd71R7zyOOOCLZZptt1vuszxs2bFhSUlKSJEmSHHXUUcn++++fJEmSVFRUJO3atUsuu+yyqvquueaaquPWrFmTVFRUrHcehYWFyeWXX1617fnnn1/v3D4xYMCAJCKS8ePH1/jagAEDqm177LHHkohIrrzyyuTNN99MmjZtmhx++OFfeo5JkiQR8aWP559/vmr//fffP+nVq1e1P7/Kyspk7733TnbaaafNch0mTJhQtW3WrFlJRCR5eXnJP//5z/WuwWffp6a/f1OnTk0iIrnrrruqtn3yd7dPnz5JeXl51farr746iYjk4Ycf3tDlAwC+hLEIgHryySjCJ9++fplHH300IiJGjhxZbfs555wTEbHevRl22GGHGDx4cI3vNWzYsGr3X5gxY0ZV+//ixYtj0aJFsWjRoli5cmXsv//+8be//S0qKys3WNtn3+ujjz6KRYsWRf/+/WPVqlUxa9asjTq/z3r//fdjxowZccopp0SrVq2qtu+6665xwAEHVF2LzzrjjDOqPe/fv38sXry46jpvjBNOOCGmTJkS8+fPj0mTJsX8+fNrHImI+Pg+DZ/cp6KioiIWL14cTZs2jW7dusX06dM3+jMLCwtj+PDhG7XvgQceGN/73vfi8ssvjyOPPDKKiorilltu2ejP+ta3vhVPPPHEeo/zzjuv2n5LliyJSZMmxTHHHFP157lo0aJYvHhxDB48OF577bWYN29eVf11cR2aNm0axx13XNXzbt26RYsWLWKXXXaJvn37Vm3/5L/ffPPNqm2f/fu3du3aWLx4cXTt2jVatGhRYw3f/e53q9089Pvf/340atSoxr9XAMDGMRYBUE9KS0sj4uNfxjfG22+/HXl5edG1a9dq29u1axctWrSIt99+u9r2HXbYYYPv9fnXXnvttYj4OHTYkGXLlkXLli1rfO3ll1+Oiy66KCZNmrTeL/PLli3b4HtuyCfn8tlRjk/ssssu8dhjj8XKlSujpKSkavv2229fbb9Pav3www+rrvWXOfjgg6NZs2Zx7733xowZM+JrX/tadO3adb3Z/YiIysrKuP766+Omm26KOXPmVLu/xDbbbLNRnxcR0bFjx026eeO1114bDz/8cMyYMSMmTJgQbdq02ehjt9122xg0aNB62999991qz19//fVIkiQuvvjiuPjii2t8rw8++CA6duxYZ9dh2223XW8koXnz5rHddtutty0iqt0jYfXq1TFmzJj47W9/G/PmzYskSapeq+nv30477VTtedOmTaN9+/Y1/jkDABtHuABQT0pLS6NDhw4xc+bMTTpuY2fCa1oZYkOvfdKVcM0112xwmcAN3R9h6dKlMWDAgCgtLY3LL788dtxxxygqKorp06fHT37yky/seKhL+fn5NW7/7C+aX6awsDCOPPLIuPPOO+PNN9+MSy+9dIP7XnXVVXHxxRfHd77znbjiiiuiVatWkZeXF2edddYmnfMX/TnV5F//+ld88MEHERHx0ksvxfHHH79Jx2+MT+o/99xzN9j98knIVVfXYUN/fhvz5/rDH/4wfvvb38ZZZ50V/fr1i+bNm0cmk4njjjsua3//AGBrJ1wAqEeHHnpo/PrXv46pU6dGv379vnDfTp06RWVlZbz22muxyy67VG1fsGBBLF26NDp16lTrOnbccceI+DjwqOmb7S8yZcqUWLx4cTz00EPxjW98o2r7nDlz1tt3Y4ORT85l9uzZ6702a9asaN26dbWuhbp0wgknxO233x55eXnV2vQ/74EHHoj99tsvbrvttmrbly5dGq1bt656Xpc3CFy5cmUMHz48evToEXvvvXdcffXVccQRR1StxFBXunTpEhERBQUFX/r3oT6uQ001DBs2LK677rqqbWvWrKm28sdnvfbaa7HffvtVPV+xYkW8//77cfDBB2+2GgEg17nnAkA9Ov/886OkpCROO+20WLBgwXqvv/HGG3H99ddHRFT94jNu3Lhq+4wdOzYiIg455JBa19GnT5/Ycccd49prr61x5YCFCxdu8NhPvln+7DfJ5eXlcdNNN623b0lJyUaNSbRv3z522223uPPOO6v9gjhz5sx4/PHHN+svgfvtt19cccUV8atf/SratWu3wf3y8/PX64q4//77q+5F8IlPQpAN/aK7KX7yk5/E3Llz484774yxY8dG586dY9iwYestRZpWmzZtYuDAgXHLLbfE+++/v97rn/37UB/X4fNqquGGG26oNqLxWb/+9a9j7dq1Vc9vvvnmWLduXQwZMqTOawOArYXOBYB6tOOOO8aECRPi2GOPjV122SWGDh0aPXv2jPLy8njmmWfi/vvvj1NOOSUiInr37h3Dhg2LX//611WjCM8991zceeedcfjhh1f7JnZT5eXlxW9+85sYMmRIfPWrX43hw4dHx44dY968eTF58uQoLS2NP/3pTzUeu/fee0fLli1j2LBh8aMf/SgymUz87ne/q3EcoU+fPnHvvffGyJEj42tf+1o0bdo0DjvssBrf95prrokhQ4ZEv3794tRTT61airJ58+ZfOK6QVl5eXlx00UVfut+hhx4al19+eQwfPjz23nvveOmll+Luu++u+tb/EzvuuGO0aNEixo8fH82aNYuSkpLo27fvF94ToyaTJk2Km266KUaPHl21NOZvf/vbGDhwYFx88cVx9dVXb9L7fZkbb7wx9t133+jVq1ecfvrp0aVLl1iwYEFMnTo13n333XjxxRcjIvvXoSaHHnpo/O53v4vmzZtHjx49YurUqfHkk09u8J4P5eXlsf/++8cxxxwTs2fPjptuuin23Xff+J//+Z/UtQDAVqve1qkAoMqrr76anH766Unnzp2Txo0bJ82aNUv22Wef5IYbbqi2FODatWuTyy67LNlhhx2SgoKCZLvttktGjRpVbZ8k+XgpykMOOWS9z/lkKcr777+/xjr+9a9/JUceeWSyzTbbJIWFhUmnTp2SY445Jnnqqaeq9qlpKcqnn3462WuvvZImTZokHTp0SM4///yqJQMnT55ctd+KFSuSE044IWnRokUSEVXLUta0FGWSJMmTTz6Z7LPPPkmTJk2S0tLS5LDDDkv+85//VNvnk6UoFy5cWG17TXXW5LNLUW7IhpaiPOecc5L27dsnTZo0SfbZZ59k6tSpNS4h+fDDDyc9evRIGjVqVO08BwwYkHz1q1+t8TM/+z7Lly9POnXqlOyxxx7J2rVrq+139tlnJ3l5ecnUqVO/8BwiIjnzzDNrfO2Ta/XZpSiTJEneeOONZOjQoUm7du2SgoKCpGPHjsmhhx6aPPDAA1m5Dhv6e/z5c/nwww+T4cOHJ61bt06aNm2aDB48OJk1a1bSqVOnZNiwYeud51//+tfku9/9btKyZcukadOmyYknnlhtyVMAYNNlkmQT7nQFALCFuuOOO2L48OHx/PPPx5577lnf5QBATnHPBQAAACAV4QIAAACQinABAAAASEW4AABsFU455ZRIksT9FgDY4vztb3+Lww47LDp06BCZTCb+8Ic/fOkxU6ZMiT322CMKCwuja9eucccdd2zWGoULAAAA0ICtXLkyevfuHTfeeONG7T9nzpw45JBDYr/99osZM2bEWWedFaeddlo89thjm61Gq0UAAADAFiKTycT//d//xeGHH77BfX7yk5/EI488EjNnzqzadtxxx8XSpUtj4sSJm6UunQsAAACQRWVlZbF8+fJqj7Kysjp7/6lTp8agQYOqbRs8eHBMnTq1zj7j8xpttnfeRI8UdKvvEoCtWP9fDKnvEoCt2MGTj6jvEoCt3D/+NKC+S9gsGurvmc//9Pi47LLLqm0bPXp0XHrppXXy/vPnz4+2bdtW29a2bdtYvnx5rF69Opo0aVInn/NZDSZcAAAAgK3BqFGjYuTIkdW2FRYW1lM1dUO4AAAAAFlUWFi4WcOEdu3axYIFC6ptW7BgQZSWlm6WroUI4QIAAAA5KlOQqe8S6kW/fv3i0UcfrbbtiSeeiH79+m22z3RDRwAAAGjAVqxYETNmzIgZM2ZExMdLTc6YMSPmzp0bER+PWQwdOrRq/zPOOCPefPPNOP/882PWrFlx0003xX333Rdnn332ZqtRuAAAAAAN2AsvvBC777577L777hERMXLkyNh9993jkksuiYiI999/vypoiIjYYYcd4pFHHoknnngievfuHdddd1385je/icGDB2+2Go1FAAAAkJPyGuXGWMTAgQMjSZINvn7HHXfUeMy//vWvzVhVdToXAAAAgFSECwAAAEAqxiIAAADISZkC36dniysNAAAApCJcAAAAAFIxFgEAAEBOypXVIrYEOhcAAACAVIQLAAAAQCrGIgAAAMhJmQJjEdmicwEAAABIRbgAAAAApGIsAgAAgJxktYjs0bkAAAAApCJcAAAAAFIxFgEAAEBOslpE9uhcAAAAAFIRLgAAAACpGIsAAAAgJ1ktInt0LgAAAACpCBcAAACAVIxFAAAAkJMy+cYiskXnAgAAAJCKcAEAAABIxVgEAAAAOSnPWETW6FwAAAAAUhEuAAAAAKkYiwAAACAnZfKMRWSLzgUAAAAgFeECAAAAkIqxCAAAAHJSJt/36dniSgMAAACpCBcAAACAVIxFAAAAkJPy8q0WkS06FwAAAIBUhAsAAABAKsYiAAAAyEmZPGMR2aJzAQAAAEhFuAAAAACkYiwCAACAnGS1iOzRuQAAAACkIlwAAAAAUjEWAQAAQE7KGIvIGp0LAAAAQCrCBQAAACAVYxEAAADkpEye79OzxZUGAAAAUhEuAAAAAKkYiwAAACAnZfKsFpEtOhcAAACAVIQLAAAAQCrGIgAAAMhJefnGIrJF5wIAAACQinABAAAASMVYBAAAADnJahHZo3MBAAAASEW4AAAAAKRiLAIAAICclMnzfXq2uNIAAABAKsIFAAAAIBVjEQAAAOQkq0Vkj84FAAAAIBXhAgAAAJCKsQgAAAByUl6+sYhs0bkAAAAApCJcAAAAAFIxFgEAAEBOslpE9uhcAAAAAFIRLgAAAACpGIsAAAAgJ2XyfJ+eLa40AAAAkIpwAQAAAEjFWAQAAAA5yWoR2aNzAQAAAEhFuAAAAACkYiwCAACAnGQsInt0LgAAAACpCBcAAACAVIxFAAAAkJOMRWSPzgUAAAAgFeECAAAAkIpwAQAAAEjFPRcAAADISZk836dniysNAAAApCJcAAAAAFIxFgEAAEBOysu3FGW26FwAAAAAUhEuAAAAAKkYiwAAACAnZfKMRWSLzgUAAAAgFeECAAAAkIqxCAAAAHJSJs/36dniSgMAAACpCBcAAACAVIxFAAAAkJOsFpE9OhcAAACAVIQLAAAAQCrGIgAAAMhJxiKyR+cCAAAAkIpwAQAAAEjFWAQAAAA5KZPn+/RscaUBAACggbvxxhujc+fOUVRUFH379o3nnnvuC/cfN25cdOvWLZo0aRLbbbddnH322bFmzZrNVp9wAQAAABqwe++9N0aOHBmjR4+O6dOnR+/evWPw4MHxwQcf1Lj/hAkT4oILLojRo0fHK6+8Erfddlvce++9ceGFF262GoULAAAA5KRMXqZBPjbV2LFj4/TTT4/hw4dHjx49Yvz48VFcXBy33357jfs/88wzsc8++8QJJ5wQnTt3jgMPPDCOP/74L+12SEO4AAAAAA1UeXl5TJs2LQYNGlS1LS8vLwYNGhRTp06t8Zi99947pk2bVhUmvPnmm/Hoo4/GwQcfvNnqdENHAAAAyKKysrIoKyurtq2wsDAKCwvX23fRokVRUVERbdu2rba9bdu2MWvWrBrf/4QTTohFixbFvvvuG0mSxLp16+KMM85oeGMRV155ZcyZM6euawEAAIA6k8nLa5CPMWPGRPPmzas9xowZU2fnPWXKlLjqqqvipptuiunTp8dDDz0UjzzySFxxxRV19hmfV6tw4f7774+uXbvG3nvvHTfddFMsWrSorusCAACAnDRq1KhYtmxZtceoUaNq3Ld169aRn58fCxYsqLZ9wYIF0a5duxqPufjii+Pkk0+O0047LXr16hVHHHFEXHXVVTFmzJiorKys8/OJqGW48OKLL8a///3vGDhwYFx77bXRoUOHOOSQQ2LChAmxatWquq4RAAAAckZhYWGUlpZWe9Q0EhER0bhx4+jTp0889dRTVdsqKyvjqaeein79+tV4zKpVqyIvr/qv+/n5+RERkSRJHZ1FdbW+oeNXv/rVuOqqq+LNN9+MyZMnR+fOneOss87aYHICAAAAWZXJNMzHJho5cmTceuutceedd8Yrr7wS3//+92PlypUxfPjwiIgYOnRotc6Hww47LG6++ea45557Ys6cOfHEE0/ExRdfHIcddlhVyFDX6uSGjiUlJdGkSZNo3LhxfPTRR3XxlgAAAEBEHHvssbFw4cK45JJLYv78+bHbbrvFxIkTq27yOHfu3GqdChdddFFkMpm46KKLYt68efGVr3wlDjvssPjZz3622WrMJLXsiZgzZ05MmDAhJkyYELNnz44BAwbECSecEEcddVQ0b958k9/vkYJutSkDoE70/8WQ+i4B2IodPPmI+i4B2Mr9408D6ruEzeLdHx5T3yXUaNsb7qvvEupcrToX9tprr3j++edj1113jeHDh8fxxx8fHTt2rOvaAAAAoNYyeZs+gkDt1Cpc2H///eP222+PHj161HU9AAAAwBamVuHC5pzTAAAAALYsGx0ujBw5cqPfdOzYsbUqBgAAAOpKJq/WCySyiTY6XPjXv/5V7fn06dNj3bp10a3bxzdifPXVVyM/Pz/69OlTtxUCAAAADdpGhwuTJ0+u+u+xY8dGs2bN4s4774yWLVtGRMSHH34Yw4cPj/79+9d9lbABrfbdM7qcc2o036NnFHVoEy98+wex4I9P1XdZQA4q6L1vFO75zciUNIvKhe/F6skPRuX8uTXuW3z0iGi0Xdf1tq998+VY/YdbN3epQI449cTOcdiB7aJZSaN46ZXlce1Nr8W776/e4P6HD2kfhw/pEO3bFkVExJy5q+KOe96Of05bUrVPqxYF8YPv7Bhf261lFDfJj7nzVsVd982Nvz6zaLOfD5DbanXPheuuuy4ef/zxqmAhIqJly5Zx5ZVXxoEHHhjnnHNOnRUIXyS/pDiW/3t2vHPHg7HnAzfWdzlAjmq08+5RNODwWPPUfVHx/tvReI8BUXLkGbHit1dFsnrFevuv+tPtkcnLr3qeaVISJSefF+tefTGbZQNbsBO/vV0cdWjH+Nm4WfH+gjVx2omdY+zlveKkHzwf5WtrXkl+4aLyGH/nnHj3vdWRyUQM2b9tjPnpV+M7Z02LOXNXRUTERSO7R9OSRnHBFTNj2fK1ccCANnH5+T3itJHT47U31/95Bls6q0VkT60GUJYvXx4LFy5cb/vChQvjo48+Sl0UbKyFj/0tXh09LhY8/GR9lwLksMI+A2PtzKmx9uXnonLJgljz5P2RrCuPgp59az5gzapIVn1U9Wi0fbeItWtj7aszslo3sOU6+n86xl33vR3/eHZxvPHWyrjyf2fFNq0Ko/9erTd4zNPPL45/TlsS776/Ot55b3X8+ndvxeo1FdGjW2nVPj27N48H/zwvXnnto3hvwZq48765sWLluujWtWk2TgvIYbUKF4444ogYPnx4PPTQQ/Huu+/Gu+++Gw8++GCceuqpceSRR9Z1jQBQf/LyI6/ttrHu7Vc/szGJdW+/GvntO2/UWxT06htrZ0+PWFe+WUoEckuHtkXRulVhPD/jw6ptK1dVxH9eXR49u5d+wZGfysuL2L//V6KoKD9enrW8avvMWcvim/3bRLOmjSKT+Xifxo3z4l8vLa3r0wC2MrUaixg/fnyce+65ccIJJ8TatWs/fqNGjeLUU0+Na665pk4LBID6lGlSEpm8/EhWVe/MS1Z9FPmt2n7p8Xntto/81h1i9eP3bK4SgRzTqmXjiIj4cOnaats/XFpe9dqGdOlUEuOv2T0aN86L1asr4sKfvRxvvbOq6vVLfvGfuOz8HvGX3+8T69ZVxpqyyrjwqpdj3vtr6v5EoAGwWkT21CpcKC4ujptuuimuueaaeOONNyIiYscdd4ySkpKNOr6srCzKysqqbVubVEZBxh88ALmlcc+9omLhexu8+SPAAQPaxHln7lz1/PzLX6r1e82dtyqG//iFaFrcKAbu85X46dnd4oejXqwKGE47cYdoVtIofvzTF2PZ8rXRf6/Wcfn5PeLMC2bEm2+vTH0uwNarVuHCJ95///14//334xvf+EY0adIkkiSJTObLb5gxZsyYuOyyy6ptOz7TKk7M3/AMGQDUh2T1ykgqKyJT3Kza9kxxs6hcuXwDR/1Xo8ZR0G33KHvmL5uxQmBL94/nFsd/Xn2h6nnjgo+/cGvZoiAWf/jpOFXLFo3j9S+56eK6dUlVF8LsN1bELjs1i6P/p2Ncc+Nr0aFdURx1WMc4+cznq27w+PpbK6P3V5vHkYd0iGtveq2uTw3YitSqVWDx4sWx//77x8477xwHH3xwvP/++xERceqpp27UShGjRo2KZcuWVXsck9eqNqUAwOZVWRGVC96NRtvv9JmNmWi0/c5R8f5bX3howc67ReQ3irWvvPCF+wFbt9WrK2Le+2uqHnPmropFS8piz96frsxW3CQ/euxcGjNnfUmo+TmZTETBf8OKosKPV7GprKy+T0VlEm6oT67K5GUa5CMX1SpcOPvss6OgoCDmzp0bxcXFVduPPfbYmDhx4pceX1hYGKWlpdUeRiKojfyS4ijt3T1Ke3ePiIjiHbaN0t7do2i79vVcGZBLyqZNiYJe/aKgx9cir1XbKBp0dGQKGsfal5+NiIiig06Mwn0PXe+4gp59Y93rL0WyZtV6rwF8kfv/OC+GHbt97PP1baJLp5K4aGT3WLykLP7+z0VV+4y7ctc48pAOVc+/N3SH6P3V5tGuTWF06VQS3xu6Q+zeq0U8PuWDiIh4+91V8c57q+K8M3eKXXZqFh3aFcVxh28bX9utZfztn4uzfo5AbqnVWMTjjz8ejz32WGy77bbVtu+0007x9ttv10lhsDGa9+kZ/Z76XdXzHtdeGBER79z1UPz71FH1VRaQY9a9+q9YU1wShXsPiUxxaVQunBerHrolklUftyfnNWsZlUn1defzWraJRtvuGCsfuKk+Sga2cHc/+E4UFeXH+SN2jqYljeKl/yyLc0a/FOVrP/1Z07Fdk2hRWlD1vGXzgrjo7O6xTavGsXLlunjjrZUxcvRL8cJ/V52oqEjivEtnxhmn7BC/uLhnNGmSH/PeXx0/Gzcr/jltSdbPEcgtmST53L+GNkKzZs1i+vTpsdNOO0WzZs3ixRdfjC5dusQLL7wQgwcPjsWLNz35fKSg2yYfA1BX+v9iSH2XAGzFDp58RH2XAGzl/vGnAfVdwmbxwaih9V1CjdqMuau+S6hztZpF6N+/f9x116cXI5PJRGVlZVx99dWx33771VlxAAAAQMNXq7GIq6++Ovbff/944YUXory8PM4///x4+eWXY8mSJfH000/XdY0AAABAA1arcKFnz54xe/bsuPHGG6NZs2axYsWKOPLII+PMM8+M9u3dSA8AAIAGIM/CAdlSq3AhIqKoqCgOOOCA6N27d1T+dz2b559/PiIi/ud//qduqgMAAAAavFqFCxMnToyTTz45lixZEp+/H2Qmk4mKioo6KQ4AAABo+GrVI/LDH/4wjjnmmHjvvfeisrKy2kOwAAAAQEOQyWQa5CMX1SpcWLBgQYwcOTLatm1b1/UAAAAAW5hahQtHHXVUTJkypY5LAQAAALZEtbrnwq9+9as4+uij4+9//3v06tUrCgoKqr3+ox/9qE6KAwAAgNrKWC0ia2oVLvz+97+Pxx9/PIqKimLKlCnVZkYymYxwAQAAALYitQoXfvrTn8Zll10WF1xwQeRJggAAAGCrVqtwoby8PI499ljBAgAAAA1WJi83V2ZoiGqVDgwbNizuvffeuq4FAAAA2ALVqnOhoqIirr766njsscdi1113Xe+GjmPHjq2T4gAAAICGr1bhwksvvRS77757RETMnDmz2mufvbkjAAAA1Buj/FlTq3Bh8uTJdV0HAAAAsIUS4wAAAACp1KpzAQAAABo6q0Vkj84FAAAAIBXhAgAAAJCKsQgAAAByUibj+/RscaUBAACAVIQLAAAAQCrGIgAAAMhNVovIGp0LAAAAQCrCBQAAACAVYxEAAADkpEye79OzxZUGAAAAUhEuAAAAAKkYiwAAACAnZawWkTU6FwAAAIBUhAsAAABAKsYiAAAAyE0Z36dniysNAAAApCJcAAAAAFIxFgEAAEBOslpE9uhcAAAAAFIRLgAAAACpGIsAAAAgN+X5Pj1bXGkAAAAgFeECAAAAkIqxCAAAAHJSJmO1iGzRuQAAAACkIlwAAAAAUjEWAQAAQG6yWkTWuNIAAABAKsIFAAAAIBVjEQAAAOSkTJ7VIrJF5wIAAACQinABAAAASMVYBAAAALkp4/v0bHGlAQAAgFSECwAAAEAqxiIAAADITVaLyBqdCwAAAEAqwgUAAAAgFWMRAAAA5KSM1SKyxpUGAAAAUhEuAAAAAKkYiwAAACA3WS0ia3QuAAAAAKkIFwAAAIBUjEUAAACQkzJ5vk/PFlcaAAAASEW4AAAAAKRiLAIAAIDclLFaRLboXAAAAABSES4AAAAAqRiLAAAAIDdZLSJrXGkAAAAgFeECAAAAkIqxCAAAAHKT1SKyRucCAAAAkIpwAQAAAEjFWAQAAAA5KWO1iKxxpQEAAIBUhAsAAABAKsYiAAAAyE0Z36dniysNAAAApCJcAAAAAFIxFgEAAEBuysvUdwVbDZ0LAAAAQCrCBQAAACAVYxEAAADkpIzVIrLGlQYAAABSES4AAABAA3fjjTdG586do6ioKPr27RvPPffcF+6/dOnSOPPMM6N9+/ZRWFgYO++8czz66KObrT5jEQAAAOSmHFkt4t57742RI0fG+PHjo2/fvjFu3LgYPHhwzJ49O9q0abPe/uXl5XHAAQdEmzZt4oEHHoiOHTvG22+/HS1atNhsNQoXAAAAoAEbO3ZsnH766TF8+PCIiBg/fnw88sgjcfvtt8cFF1yw3v633357LFmyJJ555pkoKCiIiIjOnTtv1hqNRQAAAEAWlZWVxfLly6s9ysrKaty3vLw8pk2bFoMGDaralpeXF4MGDYqpU6fWeMwf//jH6NevX5x55pnRtm3b6NmzZ1x11VVRUVGxWc4nQrgAAABArsrkNcjHmDFjonnz5tUeY8aMqfEUFi1aFBUVFdG2bdtq29u2bRvz58+v8Zg333wzHnjggaioqIhHH300Lr744rjuuuviyiuvrPNL/AljEQAAAJBFo0aNipEjR1bbVlhYWGfvX1lZGW3atIlf//rXkZ+fH3369Il58+bFNddcE6NHj66zz/ks4QIAAABkUWFh4UaHCa1bt478/PxYsGBBte0LFiyIdu3a1XhM+/bto6CgIPLz86u27bLLLjF//vwoLy+Pxo0b1774DTAWAQAAQG7KZBrmYxM0btw4+vTpE0899VTVtsrKynjqqaeiX79+NR6zzz77xOuvvx6VlZVV21599dVo3779ZgkWIoQLAAAA0KCNHDkybr311rjzzjvjlVdeie9///uxcuXKqtUjhg4dGqNGjara//vf/34sWbIkfvzjH8err74ajzzySFx11VVx5plnbrYajUUAAABAA3bsscfGwoUL45JLLon58+fHbrvtFhMnTqy6yePcuXMjL+/T3oHtttsuHnvssTj77LNj1113jY4dO8aPf/zj+MlPfrLZahQuAAAAkJvycqdZf8SIETFixIgaX5syZcp62/r16xf//Oc/N3NVn8qdKw0AAADUC+ECAAAAkIqxCAAAAHJTxvfp2eJKAwAAAKkIFwAAAIBUjEUAAACQm/Iy9V3BVkPnAgAAAJCKcAEAAABIxVgEAAAAuclqEVnjSgMAAACpCBcAAACAVIxFAAAAkJsyVovIFp0LAAAAQCrCBQAAACAVYxEAAADkpjzfp2eLKw0AAACkIlwAAAAAUjEWAQAAQG6yWkTW6FwAAAAAUhEuAAAAAKkYiwAAACA3ZXyfni2uNAAAAJCKcAEAAABIxVgEAAAAuSnP9+nZ4koDAAAAqQgXAAAAgFSMRQAAAJCbMpn6rmCr0WDChf6/GFLfJQBbsb//5C/1XQKwFdv5ogvruwQASMVYBAAAAJBKg+lcAAAAgDqV8X16trjSAAAAQCrCBQAAACAVYxEAAADkJqtFZI3OBQAAACAV4QIAAACQirEIAAAAclOe79OzxZUGAAAAUhEuAAAAAKkYiwAAACAnJVaLyBqdCwAAAEAqwgUAAAAgFWMRAAAA5KaM79OzxZUGAAAAUhEuAAAAAKkYiwAAACA3GYvIGlcaAAAASEW4AAAAAKRiLAIAAICclGQy9V3CVkPnAgAAAJCKcAEAAABIxVgEAAAAuclqEVnjSgMAAACpCBcAAACAVIxFAAAAkJusFpE1OhcAAACAVIQLAAAAQCrGIgAAAMhNeb5PzxZXGgAAAEhFuAAAAACkYiwCAACAnJRYLSJrdC4AAAAAqQgXAAAAgFSMRQAAAJCbMr5PzxZXGgAAAEhFuAAAAACkYiwCAACAnJQYi8gaVxoAAABIRbgAAAAApGIsAgAAgNyUydR3BVsNnQsAAABAKsIFAAAAIBVjEQAAAOQkq0VkjysNAAAApCJcAAAAAFIxFgEAAEBuslpE1uhcAAAAAFIRLgAAAACpGIsAAAAgN1ktImtcaQAAACAV4QIAAACQirEIAAAAclJitYis0bkAAAAApCJcAAAAAFIxFgEAAEBuslpE1rjSAAAAQCrCBQAAACAVYxEAAADkpCSsFpEtOhcAAACAVIQLAAAAQCrGIgAAAMhJidUissaVBgAAAFIRLgAAAACpGIsAAAAgNxmLyBpXGgAAAEhFuAAAAACkYiwCAACAnJRkMvVdwlZD5wIAAACQinABAAAASMVYBAAAADkpsVpE1rjSAAAAQCrCBQAAAGjgbrzxxujcuXMUFRVF375947nnntuo4+65557IZDJx+OGHb9b6hAsAAADkpkymYT420b333hsjR46M0aNHx/Tp06N3794xePDg+OCDD77wuLfeeivOPffc6N+/f22v4EYTLgAAAEADNnbs2Dj99NNj+PDh0aNHjxg/fnwUFxfH7bffvsFjKioq4sQTT4zLLrssunTpstlrFC4AAABAA1VeXh7Tpk2LQYMGVW3Ly8uLQYMGxdSpUzd43OWXXx5t2rSJU089NRtlWi0CAACA3NRQV4soKyuLsrKyatsKCwujsLBwvX0XLVoUFRUV0bZt22rb27ZtG7Nmzarx/f/xj3/EbbfdFjNmzKizmr9Mw7zSAAAAkKPGjBkTzZs3r/YYM2ZMnbz3Rx99FCeffHLceuut0bp16zp5z42hcwEAAACyaNSoUTFy5Mhq22rqWoiIaN26deTn58eCBQuqbV+wYEG0a9duvf3feOONeOutt+Kwww6r2lZZWRkREY0aNYrZs2fHjjvumPYU1iNcAAAAICclsekrM2TDhkYgatK4cePo06dPPPXUU1XLSVZWVsZTTz0VI0aMWG//7t27x0svvVRt20UXXRQfffRRXH/99bHddtulrr8mwgUAAABowEaOHBnDhg2LPffcM77+9a/HuHHjYuXKlTF8+PCIiBg6dGh07NgxxowZE0VFRdGzZ89qx7do0SIiYr3tdUm4AAAAAA3YscceGwsXLoxLLrkk5s+fH7vttltMnDix6iaPc+fOjby8+r2lonABAACAnNRQV4uojREjRtQ4BhERMWXKlC889o477qj7gj4nd640AAAAUC+ECwAAAEAqxiIAAADITZmGuVpELtK5AAAAAKQiXAAAAABSMRYBAABATkp8n541rjQAAACQinABAAAASMVYBAAAADkpsVpE1uhcAAAAAFIRLgAAAACpGIsAAAAgJyUZ36dniysNAAAApCJcAAAAAFIxFgEAAEBOSsJqEdmicwEAAABIRbgAAAAApGIsAgAAgJxktYjscaUBAACAVIQLAAAAQCrGIgAAAMhJScZqEdmicwEAAABIRbgAAAAApGIsAgAAgJyUhLGIbNG5AAAAAKQiXAAAAABSMRYBAABATkoyvk/PFlcaAAAASEW4AAAAAKRiLAIAAICcZLWI7NG5AAAAAKQiXAAAAABSMRYBAABATrJaRPa40gAAAEAqwgUAAAAgFWMRAAAA5CSrRWSPzgUAAAAgFeECAAAAkIqxCLYIBb33jcI9vxmZkmZRufC9WD35waicP7fGfYuPHhGNtuu63va1b74cq/9w6+YuFdiKtNp3z+hyzqnRfI+eUdShTbzw7R/Egj8+Vd9lATni8IEl8Y09iqK4KC9ef2dt3PXIR/HBkooN7n/wvsXRp3thtG+dH+XrIl5/Z2088OSKmL/402MG7FEUfXsVRaf2jaJJYV6c+fOFsbosycbpQL2wWkT21CpcWLNmTdxwww0xefLk+OCDD6KysrLa69OnT6+T4iAiotHOu0fRgMNjzVP3RcX7b0fjPQZEyZFnxIrfXhXJ6hXr7b/qT7dHJi+/6nmmSUmUnHxerHv1xWyWDWwF8kuKY/m/Z8c7dzwYez5wY32XA+SQIfsUx6C+TeI3f1geiz6siCP2axrnnNQifnrj4li3gXyhW6eCmPT86pjz3trIz8vEkd8siZEntYiLbloc5Ws/3qdxQSZmvl4eM18vj6MGNc3eCQE5r1bhwqmnnhqPP/54HHXUUfH1r389Mhk3yWDzKewzMNbOnBprX34uIiLWPHl/NOrSIwp69o3y52v4hnDNqvhs/l7QbY+ItWtj7aszslIvsPVY+NjfYuFjf6vvMoAcdEDfJvGnv62MGbPLIyLiN39YHuPObR17dC+M514uq/GY/717WbXntz+8PK4/7yvRuX1BvDr343ThiWdXR8THQQRAXapVuPDnP/85Hn300dhnn33quh6oLi8/8tpuG2XPPfmZjUmse/vVyG/feaPeoqBX31g7e3rEuvLNUiIAQF36Sou8aNEsP/7z5tqqbavLknjz3bWx43YFGwwXPq9J4cft4CtXV37JnpC7rBaRPbUaQOnYsWM0a9asrmuB9WSalEQmLz+SVR9V256s+ijySkq/9Pi8dttHfusOUT7zn5urRACAOlXa9ON/oi9fWT0UWL6yMpqXbNw/3zMRcfxBTeO1ueUxb+GG79MAUFdq1blw3XXXxU9+8pMYP358dOrUaZOPLysri7Ky6olr2bp1UdjI/SWpW4177hUVC9/b4M0fAQDq2169CmPooZ9+cTduwrIv2HvjnHRI0+jYplGMuf3D1O8FsDFq9dv8nnvuGWvWrIkuXbpEcXFxFBRUn9lasmTJFx4/ZsyYuOyyy6ptu+DAvjFq8F61KYcclqxeGUllRWSKq3fKZIqbReXK5V98cKPGUdBt9yh75i+bsUIAgHRmzC6PN9/9NAT45Pu20pK8WLbi0+6F0pK8mLtg3Ze+34lDmkbvnQrj53d8GB9+ZCSCrVvi/oBZU6tw4fjjj4958+bFVVddFW3btt3kGzqOGjUqRo4cWW1b2fgLa1MKua6yIioXvBuNtt8p1r3x0n83ZqLR9jtH+Yy/f+GhBTvvFpHfKNa+8sJmLxMAoLbWlCexprz66MLSjyqiR5eCeOe/YUJR40x02bYgJr+w+gvf68QhTWOP7oXxizuXxqKlggUge2oVLjzzzDMxderU6N27d60+tLCwMAoLC6ttW24kgg0omzYlmhx0QlQseCcq5s+NxnsMiExB41j78rMREVF00ImRrFgWZf/4c7XjCnr2jXWvvxTJmlX1UTawFcgvKY6SrttXPS/eYdso7d09ypcsizXvvF+PlQFbuieeXR2H9i+JBYsrYuHSj5eiXPpRZUyf9elo8bknt4jps8pi0vMfBw4nHdw09upVFL+8Z1msKUui9L/3Z1hdVhlr/9vwUFqSF82b5kWbVh8v271t20axpiyJJcsqYuWaJABqq1a/0Xfv3j1Wr/7i1BTqyrpX/xVrikuicO8hkSkujcqF82LVQ7dEsmpFRETkNWsZlUn1/xnmtWwTjbbdMVY+cFN9lAxsJZr36Rn9nvpd1fMe137chffOXQ/Fv08dVV9lATngL0+visKCTAw7rFkUF+XFa3PXxtj/tzTWfabBoU2r/GhW/OkNHr/5teKIiLjglJbV3uu2PyyPp19cExER++3ZJL41sKTqtVHDW663D+SSJDEWkS2ZJEk2OaJ8/PHH47LLLouf/exn0atXr/XuuVBa+uV38f+85WPP2uRjAOrK33/i3hxA/Xnwoi8e9QPY3G4f3aa+S9gsXn9jTn2XUKOuO+5Q3yXUuVp1Lhx00EEREbH//vtX254kSWQymaiosNwNAAAAbC02OVxYu3ZtRESMHz8+unXrVucFAQAAQF1IIu/Ld6JObHK4UFBQENtss03st99+sdNOO22OmgAAAIAtSK1inJNOOiluu+22uq4FAAAA2ALV6p4L69ati9tvvz2efPLJ6NOnT5SUlFR7fezYsXVSHAAAANRWElaLyJZahQszZ86MPfbYIyIiXn311WqvZTL+8AAAAGBrUqtwYfLkyXVdBwAAALCFqlW4AAAAAA2dsYjssS4HAAAAkIpwAQAAAEjFWAQAAAA5yVhE9uhcAAAAAFIRLgAAAACpGIsAAAAgJxmLyB6dCwAAAEAqwgUAAAAgFWMRAAAA5KQkMRaRLToXAAAAgFSECwAAAEAqxiIAAADISVaLyB6dCwAAAEAqwgUAAAAgFWMRAAAA5CRjEdmjcwEAAABIRbgAAAAApGIsAgAAgJxkLCJ7dC4AAAAAqQgXAAAAgFSMRQAAAJCTksRYRLboXAAAAABSES4AAAAAqRiLAAAAICdVWi0ia3QuAAAAAKkIFwAAAIBUjEUAAACQkxJjEVmjcwEAAABIRbgAAAAApGIsAgAAgJyUJMYiskXnAgAAAJCKcAEAAABIxVgEAAAAOclqEdmjcwEAAABIRbgAAAAApGIsAgAAgJxktYjs0bkAAAAApCJcAAAAAFIxFgEAAEBOslpE9uhcAAAAAFIRLgAAAACpGIsAAAAgJ1ktInt0LgAAAACpCBcAAACAVIQLAAAA5KTKBvqojRtvvDE6d+4cRUVF0bdv33juuec2uO+tt94a/fv3j5YtW0bLli1j0KBBX7h/XRAuAAAAQAN27733xsiRI2P06NExffr06N27dwwePDg++OCDGvefMmVKHH/88TF58uSYOnVqbLfddnHggQfGvHnzNluNwgUAAABowMaOHRunn356DB8+PHr06BHjx4+P4uLiuP3222vc/+67744f/OAHsdtuu0X37t3jN7/5TVRWVsZTTz212Wq0WgQAAAA5qaGuFlFWVhZlZWXVthUWFkZhYeF6+5aXl8e0adNi1KhRVdvy8vJi0KBBMXXq1I36vFWrVsXatWujVatW6Qr/AjoXAAAAIIvGjBkTzZs3r/YYM2ZMjfsuWrQoKioqom3bttW2t23bNubPn79Rn/eTn/wkOnToEIMGDUpd+4boXAAAAIAsGjVqVIwcObLatpq6FurCz3/+87jnnntiypQpUVRUtFk+I0K4AAAAQI5KomGORWxoBKImrVu3jvz8/FiwYEG17QsWLIh27dp94bHXXntt/PznP48nn3wydt1111rXuzGMRQAAAEAD1bhx4+jTp0+1mzF+cnPGfv36bfC4q6++Oq644oqYOHFi7Lnnnpu9Tp0LAAAA0ICNHDkyhg0bFnvuuWd8/etfj3HjxsXKlStj+PDhERExdOjQ6NixY9V9G37xi1/EJZdcEhMmTIjOnTtX3ZuhadOm0bRp081So3ABAACAnNRQV4vYVMcee2wsXLgwLrnkkpg/f37stttuMXHixKqbPM6dOzfy8j4dTLj55pujvLw8jjrqqGrvM3r06Lj00ks3S43CBQAAAGjgRowYESNGjKjxtSlTplR7/tZbb23+gj7HPRcAAACAVHQuAAAAkJMa6moRuUjnAgAAAJCKcAEAAABIxVgEAAAAOakyqe8Kth46FwAAAIBUhAsAAABAKsYiAAAAyElWi8genQsAAABAKsIFAAAAIBVjEQAAAOSkJDEWkS06FwAAAIBUhAsAAABAKsYiAAAAyElJUt8VbD10LgAAAACpCBcAAACAVIxFAAAAkJMqw2oR2aJzAQAAAEhFuAAAAACkIlwAAAAAUnHPBQAAAHJSkrjnQrboXAAAAABSES4AAAAAqRiLAAAAICclSX1XsPXQuQAAAACkIlwAAAAAUjEWAQAAQE5KwmoR2aJzAQAAAEhFuAAAAACkYiwCAACAnFRptYis0bkAAAAApCJcAAAAAFIxFgEAAEBOShKrRWSLzgUAAAAgFeECAAAAkIqxCAAAAHJSYrWIrNG5AAAAAKQiXAAAAABSMRYBAABATqoMq0Vki84FAAAAIBXhAgAAAJCKsQgAAAByktUiskfnAgAAAJCKcAEAAABIxVgEAAAAOSlJrBaRLToXAAAAgFSECwAAAEAqxiIAAADISZVWi8ganQsAAABAKsIFAAAAIBVjEQAAAOSkxFhE1uhcAAAAAFIRLgAAAACpGIsAAAAgJyWRqe8Stho6FwAAAIBUhAsAAABAKsYiAAAAyEmVVovIGp0LAAAAQCrCBQAAACAVYxEAAADkpMRYRNboXAAAAABSaTCdCwdPPqK+SwC2YjtfdGF9lwBsxb59Zf/6LgHY2o2eXd8VsIVrMOECAAAA1CVjEdljLAIAAABIRbgAAAAApGIsAgAAgJxUmWTqu4Sths4FAAAAIBXhAgAAAJCKsQgAAAByktUiskfnAgAAAJCKcAEAAABIxVgEAAAAOclYRPboXAAAAABSES4AAAAAqRiLAAAAICdVGovIGp0LAAAAQCrCBQAAACAVYxEAAADkpCTJ1HcJWw2dCwAAAEAqwgUAAAAgFWMRAAAA5KTEahFZo3MBAAAASEW4AAAAAKRiLAIAAICcVGksImt0LgAAAACpCBcAAACAVIxFAAAAkJOsFpE9OhcAAACAVIQLAAAAQCrGIgAAAMhJxiKyR+cCAAAAkIpwAQAAAEjFWAQAAAA5qdJYRNboXAAAAABSES4AAAAAqRiLAAAAICdZLSJ7dC4AAAAAqQgXAAAAgFSMRQAAAJCTKivru4Kth84FAAAAIBXhAgAAADRwN954Y3Tu3DmKioqib9++8dxzz33h/vfff3907949ioqKolevXvHoo49u1vqECwAAAOSkJGmYj0117733xsiRI2P06NExffr06N27dwwePDg++OCDGvd/5pln4vjjj49TTz01/vWvf8Xhhx8ehx9+eMycOTPlFd0w4QIAAAA0YGPHjo3TTz89hg8fHj169Ijx48dHcXFx3H777TXuf/3118dBBx0U5513Xuyyyy5xxRVXxB577BG/+tWvNluNwgUAAABooMrLy2PatGkxaNCgqm15eXkxaNCgmDp1ao3HTJ06tdr+ERGDBw/e4P51wWoRAAAA5KTajCBkQ1lZWZSVlVXbVlhYGIWFhevtu2jRoqioqIi2bdtW2962bduYNWtWje8/f/78GvefP39+yso3TOcCAAAAZNGYMWOiefPm1R5jxoyp77JS0bkAAAAAWTRq1KgYOXJktW01dS1ERLRu3Try8/NjwYIF1bYvWLAg2rVrV+Mx7dq126T964LOBQAAAHJSZdIwH4WFhVFaWlrtsaFwoXHjxtGnT5946qmnPj2vysp46qmnol+/fjUe069fv2r7R0Q88cQTG9y/LuhcAAAAgAZs5MiRMWzYsNhzzz3j61//eowbNy5WrlwZw4cPj4iIoUOHRseOHatGK3784x/HgAED4rrrrotDDjkk7rnnnnjhhRfi17/+9WarUbgAAAAADdixxx4bCxcujEsuuSTmz58fu+22W0ycOLHqpo1z586NvLxPBxP23nvvmDBhQlx00UVx4YUXxk477RR/+MMfomfPnputxkySNIz7Z+572F/ruwRgK7bznrvUdwnAVuzbV/av7xKArdwha2fXdwmbxa8ebRC/7q5nxMGZ+i6hzrnnAgAAAJCKcAEAAABIxT0XAAAAyEkN4yYAWwedCwAAAEAqwgUAAAAgFWMRAAAA5KTKyvquYOuhcwEAAABIRbgAAAAApGIsAgAAgJxktYjs0bkAAAAApCJcAAAAAFIxFgEAAEBOqjQWkTU6FwAAAIBUhAsAAABAKsYiAAAAyElWi8genQsAAABAKsIFAAAAIBVjEQAAAOSkpMEuF5Gp7wLqnM4FAAAAIBXhAgAAAJCKsQgAAAByUoOdishBOhcAAACAVIQLAAAAQCrGIgAAAMhJibGIrNG5AAAAAKQiXAAAAABSMRYBAABATqq0XETW6FwAAAAAUhEuAAAAAKkYiwAAACAnWS0ie3QuAAAAAKkIFwAAAIBUjEUAAACQk4xFZI/OBQAAACAV4QIAAACQirEIAAAAclKluYis0bkAAAAApCJcAAAAAFIxFgEAAEBOSirru4Kth84FAAAAIBXhAgAAAJCKsQgAAAByUmK1iKzRuQAAAACkIlwAAAAAUjEWAQAAQE6qtFpE1uhcAAAAAFIRLgAAAACpGIsAAAAgJ1ktInt0LgAAAACpCBcAAACAVIxFAAAAkJMqTUVkjc4FAAAAIBXhAgAAAJCKsQgAAAByUmIuImt0LgAAAACpCBcAAACAVIxFAAAAkJMSUxFZo3MBAAAASEW4AAAAAKRiLAIAAICcVGm1iKzRuQAAAACkIlwAAAAAUjEWAQAAQE5KLBeRNToXAAAAgFSECwAAAEAqxiIAAADISUllfVew9dC5AAAAAKQiXAAAAABSMRYBAABATqq0WkTW6FwAAAAAUhEuAAAAAKls9FhEy5YtI5PJbNS+S5YsqXVBAAAAUBcSYxFZs9Hhwrhx46r+e/HixXHllVfG4MGDo1+/fhERMXXq1Hjsscfi4osvrvMi2TqdemLnOOzAdtGspFG89MryuPam1+Ld91dvcP/Dh7SPw4d0iPZtiyIiYs7cVXHHPW/HP6d9Gna1alEQP/jOjvG13VpGcZP8mDtvVdx139z46zOLNvv5AFuewweWxDf2KIriorx4/Z21cdcjH8UHSyo2uP/B+xZHn+6F0b51fpSvi3j9nbXxwJMrYv7iT48ZsEdR9O1VFJ3aN4omhXlx5s8Xxuoy//ABNl2rffeMLuecGs336BlFHdrEC9/+QSz441P1XRawldrocGHYsGFV//3tb387Lr/88hgxYkTVth/96Efxq1/9Kp588sk4++yz67ZKtjonfnu7OOrQjvGzcbPi/QVr4rQTO8fYy3vFST94PsrX1vyP8IWLymP8nXPi3fdWRyYTMWT/tjHmp1+N75w1LebMXRUREReN7B5NSxrFBVfMjGXL18YBA9rE5ef3iNNGTo/X3lyRzVMEGrgh+xTHoL5N4jd/WB6LPqyII/ZrGuec1CJ+euPiWLeBfKFbp4KY9PzqmPPe2sjPy8SR3yyJkSe1iItuWhzlaz/ep3FBJma+Xh4zXy+PowY1zd4JATknv6Q4lv97drxzx4Ox5wM31nc5wFauVvdceOyxx+Kggw5ab/tBBx0UTz75ZOqi4Oj/6Rh33fd2/OPZxfHGWyvjyv+dFdu0Koz+e7Xe4DFPP784/jltSbz7/up4573V8evfvRWr11REj26lVfv07N48HvzzvHjltY/ivQVr4s775saKleuiW1f/wAeqO6Bvk/jT31bGjNnl8e4HFfGbPyyPFs3yYo/uhRs85n/vXhZPv7gm3ltYEe8sWBe3P7w8WrfIj87tC6r2eeLZ1fHo06vijXfXZuM0gBy28LG/xaujx8WCh/37GzaksjJpkI9cVKtwYZtttomHH354ve0PP/xwbLPNNqmLYuvWoW1RtG5VGM/P+LBq28pVFfGfV5dHz+6lX3Dkp/LyIvbv/5UoKsqPl2ctr9o+c9ay+Gb/NtGsaaPIZD7ep3HjvPjXS0vr+jSALdhXWuRFi2b58Z83Pw0AVpcl8ea7a2PH7Qq+4MjqmhR+/L/Zlasr67xGAICGZKPHIj7rsssui9NOOy2mTJkSffv2jYiIZ599NiZOnBi33nprnRbI1qdVy8YREfHh0urf6n24tLzqtQ3p0qkkxl+zezRunBerV1fEhT97Od56Z1XV65f84j9x2fk94i+/3yfWrauMNWWVceFVL8e899fU/YkAW6zSph+HAstXVg8Flq+sjOYlG5fLZyLi+IOaxmtzy2Pewg3fpwEAIBfUKlw45ZRTYpdddolf/vKX8dBDD0VExC677BL/+Mc/qsKGL1JWVhZlZWXVtlVWlEde/hf/4khuOmBAmzjvzJ2rnp9/+Uu1fq+581bF8B+/EE2LG8XAfb4SPz27W/xw1ItVAcNpJ+4QzUoaxY9/+mIsW742+u/VOi4/v0ececGMePPtlanPBdgy7dWrMIYe2qzq+bgJy1K/50mHNI2ObRrFmNs//PKdAYDNwmIR2VOrcCEiom/fvnH33XfX6tgxY8bEZZddVm3bdjsNi+27Da9tOWzB/vHc4vjPqy9UPW9c8PG3gi1bFMTiD8urtrds0The/5KbLq5bl1R1Icx+Y0XsslOzOPp/OsY1N74WHdoVxVGHdYyTz3y+6gaPr7+1Mnp/tXkceUiHuPam1+r61IAtxIzZ5fHmu5+GAI3++3/H0pK8WLbi0+6F0pK8mLtg3Ze+34lDmkbvnQrj53d8GB9+ZCQCAMh9tbrnwmetWbMmli9fXu3xZUaNGhXLli2r9ti264lpS2ELtXp1Rcx7f03VY87cVbFoSVns2btl1T7FTfKjx86lMXPWl//9+qxMJqLgv2FFUWF+RERUfu7f+RWVSeRl0p0DsGVbU57EBx9WVD3eW1gRSz+qiB5dPr2/QlHjTHTZtiDeeOeLb8R44pCmsUf3wrj6rqWxaKlgAQDYOtSqc2HVqlVx/vnnx3333ReLFy9e7/WKii+eLS0sLIzCwup32zYSwWfd/8d5MezY7eOd91Z/vBTlSZ1j8ZKy+Ps/F1XtM+7KXeNvUxfFQ4+8FxER3xu6Q/xz2pJYsHBNFDdpFAcMaBO792oRI0d/PGbx9rur4p33VsV5Z+4UN97+Ziz7aG18Y6/W8bXdWsb5l8+sl/MEGq4nnl0dh/YviQWLK2Lh0o+Xolz6UWVMn/XpWN+5J7eI6bPKYtLzqyMi4qSDm8ZevYril/csizVlSZT+9/4Mq8sqY+1/Gx5KS/KiedO8aNPq48Bz27aNYk1ZEkuWVcTKNXo3gY2XX1IcJV23r3pevMO2Udq7e5QvWRZr3nm/HiuDhiPJ0ZUZGqJahQvnnXdeTJ48OW6++eY4+eST48Ybb4x58+bFLbfcEj//+c/ruka2Qnc/+E4UFeXH+SN2jqYljeKl/yyLc0a/FOVrP/3h0LFdk2hR+um3ii2bF8RFZ3ePbVo1jpUr18Ubb62MkaNfihf+u+pERUUS5106M844ZYf4xcU9o0mT/Jj3/ur42bhZ8c9pS7J+jkDD9penV0VhQSaGHdYsiovy4rW5a2Ps/1sa6z6Tn7dplR/Nij9tAvzm14ojIuKCU1pWe6/b/rA8nn7x45Gt/fZsEt8aWFL12qjhLdfbB2BjNO/TM/o99buq5z2uvTAiIt6566H496mj6qssYCuVSZJNv8XF9ttvH3fddVcMHDgwSktLY/r06dG1a9f43e9+F7///e/j0Ucf3eRC9j3sr5t8DEBd2XnPXeq7BGAr9u0r+9d3CcBW7pC1s+u7hM3ix9d/VN8l1Oj6Hzf78p22MLXqXFiyZEl06dIlIiJKS0tjyZKPv/Xdd9994/vf/37dVQcAAAC1VGm5iKyp1Q0du3TpEnPmzImIiO7du8d9990XERF/+tOfokWLFnVWHAAAANDw1SpcGD58eLz44osREXHBBRfEjTfeGEVFRXH22WfHeeedV6cFAgAAAA3bJo9FrF27Nv785z/H+PHjIyJi0KBBMWvWrJg2bVp07do1dt111zovEgAAADaV1SKyZ5PDhYKCgvj3v/9dbVunTp2iU6dOdVYUAAAAsOWo1VjESSedFLfddltd1wIAAABsgWq1WsS6devi9ttvjyeffDL69OkTJSUl1V4fO3ZsnRQHAAAAtWUsIntqFS7MnDkz9thjj4iIePXVV6u9lslk0lcFAAAAbDFqFS5Mnjy5rusAAAAAtlC1ChcAAACgoTMVkT21uqEjAAAAwCeECwAAAEAqxiIAAADISVaLyB6dCwAAAEAqwgUAAAAgFWMRAAAA5KQkMRaRLToXAAAAgFSECwAAAEAqwgUAAAByUmVl0iAfm9OSJUvixBNPjNLS0mjRokWceuqpsWLFii/c/4c//GF069YtmjRpEttvv3386Ec/imXLlm3S5woXAAAAIEeceOKJ8fLLL8cTTzwRf/7zn+Nvf/tbfPe7393g/u+991689957ce2118bMmTPjjjvuiIkTJ8app566SZ/rho4AAACQA1555ZWYOHFiPP/887HnnntGRMQNN9wQBx98cFx77bXRoUOH9Y7p2bNnPPjgg1XPd9xxx/jZz34WJ510Uqxbty4aNdq42EDnAgAAADkpSZIG+SgrK4vly5dXe5SVlaU+36lTp0aLFi2qgoWIiEGDBkVeXl48++yzG/0+y5Yti9LS0o0OFiKECwAAAJBVY8aMiebNm1d7jBkzJvX7zp8/P9q0aVNtW6NGjaJVq1Yxf/78jXqPRYsWxRVXXPGFoxQ1ES4AAABAFo0aNSqWLVtW7TFq1KgN7n/BBRdEJpP5wsesWbNS17V8+fI45JBDokePHnHppZdu0rHuuQAAAEBOSjbzygy1VVhYGIWFhRu9/znnnBOnnHLKF+7TpUuXaNeuXXzwwQfVtq9bty6WLFkS7dq1+8LjP/roozjooIOiWbNm8X//939RUFCw0fVFCBcAAACgQfvKV74SX/nKV750v379+sXSpUtj2rRp0adPn4iImDRpUlRWVkbfvn03eNzy5ctj8ODBUVhYGH/84x+jqKhok2s0FgEAAAA5YJdddomDDjooTj/99Hjuuefi6aefjhEjRsRxxx1XtVLEvHnzonv37vHcc89FxMfBwoEHHhgrV66M2267LZYvXx7z58+P+fPnR0VFxUZ/ts4FAAAAclJDHYvYnO6+++4YMWJE7L///pGXlxff/va345e//GXV62vXro3Zs2fHqlWrIiJi+vTpVStJdO3atdp7zZkzJzp37rxRnytcAAAAgBzRqlWrmDBhwgZf79y5cyTJp6HLwIEDqz2vLWMRAAAAQCo6FwAAAMhJlXXwjTwbR+cCAAAAkIpwAQAAAEjFWAQAAAA5aWtcLaK+6FwAAAAAUhEuAAAAAKkYiwAAACAnJVaLyBqdCwAAAEAqwgUAAAAgFWMRAAAA5KRKq0Vkjc4FAAAAIBXhAgAAAJCKsQgAAAByUmIsImt0LgAAAACpCBcAAACAVIxFAAAAkJOSxFhEtuhcAAAAAFIRLgAAAACpGIsAAAAgJyWVlfVdwlZD5wIAAACQinABAAAASMVYBAAAADmpstJqEdmicwEAAABIRbgAAAAApGIsAgAAgJyUJMYiskXnAgAAAJCKcAEAAABIxVgEAAAAOSmxWkTW6FwAAAAAUhEuAAAAAKkYiwAAACAnGYvIHp0LAAAAQCrCBQAAACAVYxEAAADkpMqksr5L2GroXAAAAABSES4AAAAAqRiLAAAAICdZLSJ7dC4AAAAAqQgXAAAAgFSMRQAAAJCTjEVkj84FAAAAIBXhAgAAAJCKsQgAAAByUpIYi8gWnQsAAABAKsIFAAAAIBVjEQAAAOSkysrK+i5hq6FzAQAAAEhFuAAAAACkYiwCAACAnJRUWi0iW3QuAAAAAKkIFwAAAIBUjEUAAACQk5LEahHZonMBAAAASEW4AAAAAKRiLAIAAICcZLWI7NG5AAAAAKQiXAAAAABSMRYBAABATjIWkT06FwAAAIBUhAsAAABAKsYiAAAAyEmVSWV9l7DV0LkAAAAApCJcAAAAAFIxFgEAAEBOslpE9uhcAAAAAFIRLgAAAACpGIsAAAAgJyWVVovIFp0LAAAAQCrCBQAAACAVYxEAAADkJKtFZI/OBQAAACAV4QIAAACQirEIAAAAclKSWC0iW3QuAAAAAKkIFwAAAIBUjEUAAACQkyqtFpE1OhcAAACAVIQLAAAAQCrGIgAAAMhJSaXVIrJF5wIAAACQinABAAAASMVYBAAAADkpsVpE1uhcAAAAAFIRLgAAAACpGIsAAAAgJyWJ1SKyRecCAAAAkIpwAQAAAEjFWAQAAAA5yWoR2aNzAQAAAEhFuAAAAACkYiwCAACAnJRUWi0iW3QuAAAAAKkIFwAAAIBUMkmSuH0mW7SysrIYM2ZMjBo1KgoLC+u7HGAr5OcQUJ/8DAIaAuECW7zly5dH8+bNY9myZVFaWlrf5QBbIT+HgPrkZxDQEBiLAAAAAFIRLgAAAACpCBcAAACAVIQLbPEKCwtj9OjRbmAE1Bs/h4D65GcQ0BC4oSMAAACQis4FAAAAIBXhAgAAAJCKcAEAAABIRbjAVmHgwIFx1lln1XcZAAD14o477ogWLVrUdxlADhMuAAAAAKkIFwBgE5WXl9d3CcBWys8foKESLlCvBg4cGD/84Q/jrLPOipYtW0bbtm3j1ltvjZUrV8bw4cOjWbNm0bVr1/jLX/5SdczMmTNjyJAh0bRp02jbtm2cfPLJsWjRoqrXV65cGUOHDo2mTZtG+/bt47rrrqv2mRdeeGH07dt3vVp69+4dl19++eY7WWCLNXDgwBgxYkScddZZ0bp16ygsLIxMJhOPPfZY7L777tGkSZP45je/GR988EH85S9/iV122SVKS0vjhBNOiFWrVtV3+cAW7PM/fwYPHhxjx46NXr16RUlJSWy33Xbxgx/8IFasWFHtuDvuuCO23377KC4ujiOOOCIWL15cT2cAbC2EC9S7O++8M1q3bh3PPfdc/PCHP4zvf//7cfTRR8fee+8d06dPjwMPPDBOPvnkWLVqVSxdujS++c1vxu677x4vvPBCTJw4MRYsWBDHHHNM1fudd9558de//jUefvjhePzxx2PKlCkxffr0qtdPPPHEeO655+KNN96o2vbyyy/Hv//97zjhhBOyeu7AluPOO++Mxo0bx9NPPx3jx4+PiIhLL700fvWrX8UzzzwT77zzThxzzDExbty4mDBhQjzyyCPx+OOPxw033FDPlQNbus///MnLy4tf/vKX8fLLL8edd94ZkyZNivPPP79q/2effTZOPfXUGDFiRMyYMSP222+/uPLKK+vxDICtQSZJkqS+i2DrNXDgwKioqIi///3vERFRUVERzZs3jyOPPDLuuuuuiIiYP39+tG/fPqZOnRpPPvlk/P3vf4/HHnus6j3efffd2G677WL27NnRoUOH2GabbeL//b//F0cffXRERCxZsiS23Xbb+O53vxvjxo2LiIjddtstvv3tb8fFF18cER93M0yaNCn++c9/ZvHsgS3FwIEDY/ny5VVB5ZQpU2K//faLJ598Mvbff/+IiPj5z38eo0aNijfeeCO6dOkSERFnnHFGvPXWWzFx4sR6qx3Ysn3+509NHnjggTjjjDOqOjlPOOGEWLZsWTzyyCNV+xx33HExceLEWLp06eYuGdhK6Vyg3u26665V/52fnx/bbLNN9OrVq2pb27ZtIyLigw8+iBdffDEmT54cTZs2rXp07949IiLeeOONeOONN6K8vLza2EOrVq2iW7du1T7zxBNPjAkTJkRERJIk8fvf/z5OPPHEzXaOwJavT58+62377M+vtm3bRnFxcVWw8Mm2Dz74ICv1Abnr8z9/Pgk2O3bsGM2aNYuTTz45Fi9eXDWG9corr6w3AtqvX7+s1QtsnYQL1LuCgoJqzzOZTLVtmUwmIiIqKytjxYoVcdhhh8WMGTOqPV577bX4xje+sdGfefzxx8fs2bNj+vTpVe3Mxx57bN2cEJCTSkpK1tv2+Z9VNf08q6ys3Oy1Abntsz9/3nrrrTj00ENj1113jQcffDCmTZsWN954Y0S42SNQvxrVdwGwKfbYY4948MEHo3PnztGo0fp/fXfccccoKCiIZ599NrbffvuIiPjwww/j1VdfjQEDBlTtt+2228aAAQPi7rvvjtWrV8cBBxwQbdq0ydp5AADUxrRp06KysjKuu+66yMv7+HvC++67r9o+u+yySzz77LPVthn9BDY3nQtsUc4888xYsmRJHH/88fH888/HG2+8EY899lgMHz48KioqomnTpnHqqafGeeedF5MmTYqZM2fGKaecUvU/38868cQT45577on777/fSAQAsEXo2rVrrF27Nm644YZ4880343e/+13VTWY/8aMf/SgmTpwY1157bbz22mvxq1/9yr1fgM1OuMAWpUOHDvH0009HRUVFHHjggdGrV68466yzokWLFlUBwjXXXBP9+/ePww47LAYNGhT77rtvjbPSRx11VNV84uGHH57lMwEA2HS9e/eOsWPHxi9+8Yvo2bNn3H333TFmzJhq++y1115x6623xvXXXx+9e/eOxx9/PC666KJ6qhjYWlgtAgAAAEhF5wIAAACQinABAAAASEW4AAAAAKQiXAAAAABSES4AAAAAqQgXAAAAgFSECwAAAEAqwgUAAAAgFeECAAAAkIpwAQAAAEhFuAAAAACkIlwAAAAAUvn/KV6sw1bTaPEAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1400x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate a correlation matrix heatmap to visualize the correlations between all variables\n",
    "plt.figure(figsize=(14, 10))\n",
    "correlation_matrix = df[['medv', 'rm', 'rad']].corr()\n",
    "sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm')\n",
    "plt.title('Correlation Matrix Heatmap')\n",
    "print(correlation_matrix)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             medv          rm         rad\n",
      "count  506.000000  506.000000  506.000000\n",
      "mean    22.532806    6.284634    9.549407\n",
      "std      9.197104    0.702617    8.707259\n",
      "min      5.000000    3.561000    1.000000\n",
      "25%     17.025000    5.885500    4.000000\n",
      "50%     21.200000    6.208500    5.000000\n",
      "75%     25.000000    6.623500   24.000000\n",
      "max     50.000000    8.780000   24.000000\n",
      "medv    0\n",
      "rm      0\n",
      "rad     0\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(exploration)\n",
    "print(missing_values)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Some comments and findings\n",
    "The whole dataset is composed by 506 entries spread alongside 12 columns, in this dataset there are no missing values. \n",
    "\n",
    "### The Statistics \n",
    "1) MEDV (Median Value of Houses):\n",
    "- Range: $5,000 - $50,000\n",
    "- Mean: $22,533\n",
    "- Standard Deviation: $9,197\n",
    "2) RM (Average Number of Rooms per Dwelling):\n",
    "- Range: 3.56 - 8.78 rooms\n",
    "- Mean: 6.28 rooms\n",
    "- Standard Deviation: 0.70 rooms\n",
    "3) RAD (Index of Accessibility to Radial Highways):\n",
    "- Range: 1 - 24\n",
    "- Mean: 9.55\n",
    "- Standard Deviation: 8.71\n",
    "\n",
    "### Visualizations\n",
    "1. Histograms:\n",
    "- MEDV: Appears to have a semi-normal distribution with a peak at around the values $20,000 - $25,000.\n",
    "- RM: Shows a normal distribution centered around 6 rooms.\n",
    "- RAD: Exhibits a bimodal distribution, indicating two groups of areas based on highway accessibility.\n",
    "\n",
    "2. Scatter Plots:\n",
    "MEDV vs. RM: \n",
    "- Shows a positive correlation, suggesting that houses with more rooms tend to have higher median values.\n",
    "MEDV vs. RAD: \n",
    "- The relationship is not as clear, but there seems to be a trend\n",
    "- where higher accessibility to highways correlates to lower house values, possibly due to noise or other factors.\n",
    "\n",
    "3. Correlation Matrix:\n",
    "- MEDV and RM: Strong positive correlation (0.70).\n",
    "- MEDV and RAD: Negative correlation (-0.38).\n",
    "\n",
    "## Initial Findings\n",
    "• The number of rooms (RM) has a significant positive impact on the median value of houses.\n",
    "\n",
    "• Accessibility to highways (RAD) appears to negatively impact house values, though this relationship is less clear and might be influenced by other factors like noise or pollution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Formal Analysis \n",
    "The proces is the folowwing:\n",
    "\n",
    "1. *Data Splitting*:\n",
    "- Create a function to split the dataset into a 90% training set and a 10% test set.\n",
    "2. *Model Fitting*:\n",
    "- Develop a function to fit a linear regression model using the training set.\n",
    "3. *Coefficient Interpretation*:\n",
    "- Interpret the coefficients for rad and crim.\n",
    "4. *Bootstrap Analysis*:\n",
    "- Implement a bootstrap method to test the significance of coefficients.\n",
    "5. *Hypothesis Testing and Model Refinement*:\n",
    "- Use hypothesis testing to identify significant variables.\n",
    "- Refine the model based on significant variables.\n",
    "- Evaluate the refined model on the test set using Mean Squared Error (MSE)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a custom function to split the data into training and testing sets\n",
    "# This function shuffles the indices and splits the data accordingly, ensuring reproducibility by setting a random seed\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "def custom_split(data, target_column, test_size=0.1, student_id=seed):\n",
    "    # Validate inputs\n",
    "    if len(data) == 0 or test_size <= 0 or test_size >= 1:\n",
    "        raise ValueError(\"Invalid data size or test size\")\n",
    "    \n",
    "    if target_column not in data.columns:\n",
    "        raise ValueError(\"Target column not found in data\")\n",
    "\n",
    "    # Set random seed using student ID for reproducibility\n",
    "    np.random.seed(student_id)\n",
    "\n",
    "    # Shuffle the dataset more thoroughly\n",
    "    data_shuffled = data.sample(frac=1, random_state=student_id).reset_index(drop=True)\n",
    "\n",
    "    # Determine the size of the test set\n",
    "    test_set_size = int(len(data) * test_size)\n",
    "    \n",
    "    # Ensure test_set_size is not larger than the dataset\n",
    "    test_set_size = min(test_set_size, len(data) - 1)\n",
    "\n",
    "    # Split the indices for the test and training sets\n",
    "    test_indices = np.arange(test_set_size)\n",
    "    train_indices = np.arange(test_set_size, len(data))\n",
    "\n",
    "    # Split the data into training and test sets\n",
    "    test_set = data_shuffled.iloc[test_indices.tolist()]\n",
    "    train_set = data_shuffled.iloc[train_indices.tolist()]\n",
    "\n",
    "    # Split features and target variable\n",
    "    X_train = train_set.drop(columns=target_column)\n",
    "    y_train = train_set[target_column]\n",
    "    X_test = test_set.drop(columns=target_column)\n",
    "    y_test = test_set[target_column]\n",
    "\n",
    "    return X_train, X_test, y_train, y_test\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Custom function to fit a linear regression model using the normal equation\n",
    "def custom_lr(X, y):\n",
    "    X_b = np.c_[np.ones((X.shape[0], 1)), X]  # Adding a column of ones for the intercept term\n",
    "    theta_best = np.linalg.inv(X_b.T.dot(X_b)).dot(X_b.T).dot(y)  # Calculating best fit parameters\n",
    "    return theta_best\n",
    "\n",
    "# Custom function to predict using the linear regression model\n",
    "def custom_predict(X, theta):\n",
    "    X_b = np.c_[np.ones((X.shape[0], 1)), X]\n",
    "    return X_b.dot(theta)\n",
    "\n",
    "# Custom function to calculate Mean Squared Error\n",
    "def custom_mean_squared_error(y_true, y_pred):\n",
    "    mse = np.mean((y_true - y_pred) ** 2)\n",
    "    return mse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Splitting the data\n",
    "X_train, X_test, y_train, y_test = custom_split(df, 'medv')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fitting the model using the custom function\n",
    "theta_best = custom_lr(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Coefficient Interpretation\n",
    "coefficients = pd.DataFrame([theta_best[1:]], columns=X_train.columns, index=['Coefficient']).T\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Bootstrap Analysis\n",
    "def bootstrap_analysis(data, n_bootstrap=1000):\n",
    "    bootstrap_coefs = []\n",
    "    for _ in range(n_bootstrap):\n",
    "        sample_data = data.sample(n=len(data), replace=True)\n",
    "        X_sample, y_sample = sample_data.drop(columns='medv'), sample_data['medv']\n",
    "        theta_sample = custom_lr(X_sample, y_sample)\n",
    "        bootstrap_coefs.append(theta_sample[1:])  # Exclude the intercept\n",
    "    return np.array(bootstrap_coefs)\n",
    "\n",
    "bootstrap_coefs = bootstrap_analysis(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculating statistics for bootstrap coefficients\n",
    "coef_std = np.std(bootstrap_coefs, axis=0)\n",
    "confidence_intervals = np.percentile(bootstrap_coefs, [2.5, 97.5], axis=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preparing results for display\n",
    "coef_analysis = pd.DataFrame({\n",
    "    'Coefficient Mean': np.mean(bootstrap_coefs, axis=0),\n",
    "    'Std Dev': coef_std,\n",
    "    '95% CI Lower': confidence_intervals[0],\n",
    "    '95% CI Upper': confidence_intervals[1]\n",
    "}, index=X_train.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check for significance\n",
    "coef_analysis['Significant'] = (coef_analysis['95% CI Lower'] > 0) | (coef_analysis['95% CI Upper'] < 0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Predicting and calculating MSE\n",
    "y_pred_custom = custom_predict(X_test, theta_best)\n",
    "mse_custom = custom_mean_squared_error(y_test, y_pred_custom)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Custom Coefficients:          Coefficient\n",
      "crim       -0.131907\n",
      "zn          0.048926\n",
      "indus       0.024283\n",
      "nox       -18.026122\n",
      "rm          3.520948\n",
      "age         0.003638\n",
      "dis        -1.535510\n",
      "rad         0.334010\n",
      "tax        -0.013442\n",
      "ptratio    -1.004608\n",
      "lstat      -0.583388\n",
      "Custom Mean Squared Error on Test Set: 21.57614518279428\n"
     ]
    }
   ],
   "source": [
    "# Displaying the results\n",
    "print(\"Custom Coefficients:\", coefficients)\n",
    "print(\"Custom Mean Squared Error on Test Set:\", mse_custom)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "crim: [-0.17961279 -0.05838793]\n",
      "zn: [0.01795932 0.07418449]\n",
      "indus: [-0.06823981  0.13497858]\n",
      "nox: [-25.5537351  -11.27688918]\n",
      "rm: [2.27561259 5.28224894]\n",
      "age: [-0.02491452  0.03740492]\n",
      "dis: [-1.93131905 -1.0892111 ]\n",
      "rad: [0.18975246 0.42646345]\n",
      "tax: [-0.01970752 -0.00916768]\n",
      "ptratio: [-1.20534711 -0.75727759]\n",
      "lstat: [-0.75317524 -0.37464871]\n"
     ]
    }
   ],
   "source": [
    "for i, feature in enumerate(X_train.columns):\n",
    "    print(f\"{feature}: {confidence_intervals[:, i]}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In our confindence list of the intervals the data that is considered statistically significant are: CRIM, ZN, NOX, RM, DIS, RAD, TAX, PTRATIO and LSTAT"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Coefficients \n",
    "- crim       -0.131907\n",
    "- zn          0.048926\n",
    "- indus       0.024283\n",
    "- nox       -18.026122\n",
    "- rm          3.520948\n",
    "- age         0.003638\n",
    "- dis        -1.535510\n",
    "- rad         0.334010\n",
    "- tax        -0.013442\n",
    "- ptratio    -1.004608\n",
    "- lstat      -0.583388\n",
    "\n",
    "## Interpreting the coefficients \n",
    "The coefficients of the linear regression model, which predicts the median value of houses (medv) using various features, are as follows:\n",
    "- CRIM: -0.132 (per capita crime rate by town)\n",
    "- ZN: 0.049 (proportion of residential land zoned for lots over 25,000 sq.ft.)\n",
    "- INDUS: 0.024 (proportion of non-retail business acres per town)\n",
    "- NOX: -18.03 (nitric oxides concentration)\n",
    "- RM: 3.52 (average number of rooms per dwelling)\n",
    "- AGE: 0.004 (proportion of owner-occupied units built prior to 1940)\n",
    "- DIS: -1.54 (weighted distances to five Boston employment centers)\n",
    "- RAD: 0.334 (index of accessibility to radial highways)\n",
    "- TAX: -0.013 (full-value property-tax rate per $10,000)\n",
    "- PTRATIO: -1.005 (pupil-teacher ratio by town)\n",
    "- LSTAT: -0.583 (% lower status of the population)\n",
    "\n",
    "_Values can vary because of the manual splits function_ \n",
    "\n",
    "*Interpretation of Key Coefficients*:\n",
    "- RM (Rooms): A coefficient of 3.52 suggests a strong positive relationship between the\n",
    "number of rooms and house value. Each additional room is associated with an increase in\n",
    "the median value by approximately $3521.\n",
    "- RAD (Accessibility to Highways): The coefficient of 0.334 indicates a positive\n",
    "relationship, meaning that increased accessibility to highways is associated with a higher\n",
    "median house value. However, the impact is relatively smaller compared to other\n",
    "variables like rm.\n",
    "*Mean Squared Error (MSE) on Test Set*:\n",
    "- The MSE of the model on the test set is 21.58. This value represents the average\n",
    "squared difference between the actual and predicted house values, providing a measure\n",
    "of the model's accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Custom Model - Bootstrap MSE Mean: 22.266506448229325, Std: 1.9577850287579381\n"
     ]
    }
   ],
   "source": [
    "# Perform bootstrap evaluation\n",
    "def custom_bootstrap_evaluation(X_train, y_train, X_test, y_test, n_iterations=1000):\n",
    "    mse_values = []\n",
    "    for _ in range(n_iterations):\n",
    "        boot_x, boot_y = resample(X_train, y_train)\n",
    "        theta_boot = custom_lr(boot_x, boot_y)\n",
    "        y_pred_boot = custom_predict(X_test, theta_boot)\n",
    "        mse_values.append(custom_mean_squared_error(y_test, y_pred_boot))\n",
    "    return np.mean(mse_values), np.std(mse_values)\n",
    "\n",
    "bootstrap_mse_mean, bootstrap_mse_std = custom_bootstrap_evaluation(X_train, y_train, X_test, y_test)\n",
    "print(f\"Custom Model - Bootstrap MSE Mean: {bootstrap_mse_mean}, Std: {bootstrap_mse_std}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bootstrap Analysis:\n",
    "- The bootstrap analysis performed on the linear regression model provides insightful statistics about the stability and significance of each coefficient. By calculating the standard deviation of coefficients across multiple bootstrap samples, we gain an understanding of their variability. This variability measurement is crucial for assessing the reliability of each feature's impact on the model. \n",
    "\n",
    "- Additionally, the 95% confidence intervals for each coefficient are calculated. These intervals give a range in which the true value of the coefficient is likely to fall, offering a degree of certainty about the estimations. This approach is particularly useful in determining the robustness of each feature's influence on the target variable.\n",
    "\n",
    "- Lastly, significance testing is conducted to determine whether the coefficients are significantly different from zero. A coefficient significantly different from zero suggests that the corresponding feature plays a meaningful role in predicting the target variable. This step is vital for feature selection and model refinement, as it helps in identifying the most impactful predictors among the available features.\n",
    "-Overall, the bootstrap analysis deepens our understanding of the model's behavior by providing a comprehensive view of the importance and reliability of each feature in the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Refined model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original Model MSE: 21.57614518279428\n",
      "Refined Model MSE: 21.698343879488906\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0kAAAHDCAYAAADxzVHXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAABC4UlEQVR4nO3deXhM5///8ddEZENiF0uEEEppailSVNVWal+bUoKipVW1tKi1to8q1RbVaiWorZTSTSkVtbZ2RYmtqD2aRIJYcn5/+GW+Z5pgJrKMeD6ua6527nPPOe8zZnLnlXPOfSyGYRgCAAAAAEiSXDK7AAAAAABwJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABNCEgAAAACYEJIAAAAAwISQBAAAAAAmhCQAmWLUqFGyWCypem14eLgsFotOnDiRtkWZnDhxQhaLReHh4em2jcwUGRmphg0bysfHRxaLRd9++22GvK938zC83yVKlFBoaGiqXmuxWDRq1Kg0rQeZY/369bJYLFq/fn1mlwIgHRGSADhk//796tSpk4oWLSp3d3cVKVJEHTt21P79+zO7tCwpKbgkPVxdXVW0aFGFhobqn3/+SfV6u3Tpon379mncuHGaN2+eqlatmoZVp5+kX1AtFou++uqrFPvUrFlTFotFFSpUyODqHkxSULRYLBo7dmyKfTp27CiLxaKcOXPatCcmJmru3LmqXr268ubNq1y5cqlMmTLq3Lmztm7dau1nfv9SeixatMiuWtevX6/WrVvL19dXbm5uKliwoJo1a6Zly5al/g0AACfimtkFAHh4LFu2TCEhIcqbN6+6d++ukiVL6sSJE/ryyy+1dOlSLVq0SK1atbJrXcOGDdPgwYNTVcfLL7+sF198Ue7u7ql6/cPovffeU8mSJXX9+nVt3bpV4eHh2rhxo/788095eHg4tK5r165py5Ytevfdd/X6669b2x+m99XDw0MLFixQp06dbNpPnDihzZs3O/yeOBMPDw8tXLhQw4YNs2mPj4/XihUrUty3vn37avr06WrRooU6duwoV1dXHTp0SD/99JMCAgJUo0aNZP2feuqpZOsJDg6+b30jR47Ue++9p8DAQPXq1Uv+/v6KiorSjz/+qDZt2mj+/Pl66aWXHNzrh8czzzyja9euyc3NLbNLAZCOCEkA7HL06FG9/PLLCggI0IYNG1SgQAHrsjfffFO1a9fWyy+/rL179yogIOCu64mPj1eOHDnk6uoqV9fU/QjKli2bsmXLlqrXPqwaN25sPdrzyiuvKH/+/Jo4caJWrlyp9u3bO7SuixcvSpJy585t0/4wva9NmjTRypUrdenSJeXPn9/avmDBAhUqVEiBgYH6999/M7HC1GvSpImWLVumPXv2KCgoyNq+YsUK3bhxQ88//7zWrVtnbT9//rxmzJihHj166PPPP7dZ19SpU63/3ma1a9dW27ZtHa5t6dKleu+999S2bVstWLBA2bNnty4bNGiQfv75Z928edPh9T4Mrl+/Ljc3N7m4uDzUIRyAfTjdDoBdJk2apKtXr+rzzz+3CUiSlD9/fn322WeKj4/X+++/b21Puu7owIEDeumll5QnTx7VqlXLZpnZtWvX1LdvX+XPn1+5cuVS8+bN9c8//yS7niOla2dKlCihpk2bauPGjapWrZo8PDwUEBCguXPn2mzj8uXLGjhwoCpWrKicOXPK29tbjRs31p49exx+T7Zv3y6LxaI5c+YkW/bzzz/LYrHo+++/lyRduXJF/fr1U4kSJeTu7q6CBQuqQYMG2rlzp8Pble78kivdCa9mf/31l9q2bau8efPKw8NDVatW1cqVK63LR40aJX9/f0l3fqm1WCwqUaKEpAd7XyUpOjpa/fr1k5+fn9zd3VW6dGlNnDhRiYmJyfqFhobKx8dHuXPnVpcuXRQdHe3Q/rdo0ULu7u5asmSJTfuCBQvUvn37FMPerVu3NGbMGJUqVUru7u4qUaKEhg4dqoSEBJt+hmFo7NixKlasmLy8vFS3bt27nk5q7z47Ijg4WCVLltSCBQts2ufPn6/nn39eefPmtWk/fvy4DMNQzZo1k63LYrGoYMGCqa7lv4YPH668efNq9uzZNgEpSaNGjdS0aVPr8wsXLqh79+4qVKiQPDw8FBQUlOz7knSa4QcffKDp06crICBAXl5eatiwoU6dOiXDMDRmzBgVK1ZMnp6eatGihS5fvmyzjqTP6erVq/Xkk0/Kw8ND5cuXT3b6n73f/6TTEhctWqRhw4apaNGi8vLyUmxsbIrXJEVGRqpNmzby9fWVh4eHihUrphdffFExMTHWPvZ+/hz5zgFIP4QkAHb57rvvVKJECesv5//1zDPPqESJEvrhhx+SLWvXrp2uXr2q8ePHq0ePHnfdRmhoqD755BM1adJEEydOlKenp1544QW7azxy5Ijatm2rBg0aaPLkycqTJ49CQ0NtfsE9duyYvv32WzVt2lRTpkzRoEGDtG/fPtWpU0dnzpyxe1uSVLVqVQUEBOjrr79Otmzx4sXKkyePGjVqJEl69dVX9emnn6pNmzaaMWOGBg4cKE9PTx08eNChbSZJCjJ58uSxtu3fv181atTQwYMHNXjwYE2ePFk5cuRQy5YttXz5cklS69at9eGHH0qSQkJCNG/ePE2dOvWe27Lnfb169arq1Kmjr776Sp07d9bHH3+smjVrasiQIerfv7+1n2EYatGihebNm6dOnTpp7NixOn36tLp06eLQ/nt5ealFixZauHChtW3Pnj3av3//XU/1euWVVzRixAhVrlxZH374oerUqaMJEyboxRdftOk3YsQIDR8+XEFBQZo0aZICAgLUsGFDxcfH2/Szd59TIyQkRIsWLZJhGJKkS5cuafXq1SnuW1LoXbJkia5evWrX+q9cuaJLly4leyRtLyWRkZH666+/1LJlS+XKleu+27h27ZqeffZZzZs3Tx07dtSkSZPk4+Oj0NBQffTRR8n6z58/XzNmzNAbb7yhAQMGKCIiQu3bt9ewYcO0atUqvfPOO+rZs6e+++47DRw4MMX6OnTooMaNG2vChAlydXVVu3bttGbNGmsfR7//Y8aM0Q8//KCBAwdq/PjxKZ5id+PGDTVq1Ehbt27VG2+8oenTp6tnz546duyYTfi39/Mn2fedA5DODAC4j+joaEOS0aJFi3v2a968uSHJiI2NNQzDMEaOHGlIMkJCQpL1TVqWZMeOHYYko1+/fjb9QkNDDUnGyJEjrW1hYWGGJOP48ePWNn9/f0OSsWHDBmvbhQsXDHd3d2PAgAHWtuvXrxu3b9+22cbx48cNd3d347333rNpk2SEhYXdc5+HDBliZM+e3bh8+bK1LSEhwcidO7fRrVs3a5uPj4/Rp0+fe64rJUn7+ssvvxgXL140Tp06ZSxdutQoUKCA4e7ubpw6dcrat169ekbFihWN69evW9sSExONp59+2ggMDEy2b5MmTUpxW6l5X8eMGWPkyJHDOHz4sM06Bw8ebGTLls04efKkYRiG8e233xqSjPfff9/a59atW0bt2rXter9//fVXQ5KxZMkS4/vvvzcsFot13YMGDTICAgIMwzCMOnXqGI8//rj1dbt37zYkGa+88orN+gYOHGhIMtatW2fdNzc3N+OFF14wEhMTrf2GDh1qSDK6dOni8D4bhpHsM5wS87/Ln3/+aUgyfvvtN8MwDGP69OlGzpw5jfj4eKNLly5Gjhw5bF7buXNnQ5KRJ08eo1WrVsYHH3xgHDx48K7v390eZ8+evWt9K1asMCQZH3744T33I8nUqVMNScZXX31lbbtx44YRHBxs5MyZ0/pzImm/CxQoYERHR1v7DhkyxJBkBAUFGTdv3rS2h4SEGG5ubjaf86TP6TfffGNti4mJMQoXLmxUqlTJ2mbv9z/pfQoICDCuXr1q0z9p2a+//moYhmHs2rXL+pm8G3s/f+Z9ud93DkD64kgSgPu6cuWKJN33r8dJy2NjY23aX3311ftuY9WqVZKk3r1727S/8cYbdtdZvnx5myNdBQoUUNmyZXXs2DFrm7u7u1xc7vzou337tqKiopQzZ06VLVs2Vae+dejQQTdv3rQ5rWf16tWKjo5Whw4drG25c+fWtm3bHD5alaR+/foqUKCA/Pz81LZtW+XIkUMrV65UsWLFJN05jWjdunVq3769zVGCqKgoNWrUSJGRkameDc+e93XJkiWqXbu28uTJY3Nkon79+rp9+7Y2bNggSfrxxx/l6uqq1157zfrabNmyOfTvnKRhw4bKmzev9YjLokWLFBISkmLfH3/8UZKSHeEZMGCAJFmPgP7yyy+6ceOG3njjDZvTQfv165dsnfbuc2o8/vjjeuKJJ6xHyhYsWKAWLVrIy8srxf5hYWGaNm2aSpYsqeXLl2vgwIEqV66c6tWrl+K/+4gRI7RmzZpkj/+eymeW9L225yiSdOc99/X1tfk3yZ49u/r27au4uDhFRETY9G/Xrp18fHysz6tXry5J6tSpk831i9WrV9eNGzeS7VeRIkVsJo7x9vZW586dtWvXLp07d06S49//Ll26yNPT8577mVTzzz//fNcjefZ+/pLY850DkL6YuAHAfSX9UpQUlu7mbmGqZMmS993G33//LRcXl2R9S5cubXedxYsXT9aWJ08emwv4ExMT9dFHH2nGjBk6fvy4bt++bV2WL18+u7eVJCgoSI899pgWL16s7t27S7pzql3+/Pn13HPPWfu9//776tKli/z8/FSlShU1adJEnTt3vuckF2bTp09XmTJlFBMTo9mzZ2vDhg02s9AdOXJEhmFo+PDhGj58eIrruHDhgooWLerwPtrzvkZGRmrv3r3Jrlczb1u68+9cuHDhZFNYly1b1uG6smfPrnbt2mnBggWqVq2aTp06dddT7ZI+X//9PPn6+ip37tz6+++/rf0kKTAw0KZfgQIFbE5tlOzf59R66aWXNHnyZL311lvavHmzhg4dete+Li4u6tOnj/r06aOoqCht2rRJM2fO1E8//aQXX3xRv/32m03/ihUrqn79+g7V4+3tLen+PweS/P333woMDLSGkiTlypWzLjf77+csKXz4+fml2P7fiTlKly6d7DrHMmXKSLpzeqqvr6/D3397fnaVLFlS/fv315QpUzR//nzVrl1bzZs3V6dOnay12vv5S2LPdw5A+iIkAbgvHx8fFS5cWHv37r1nv71796po0aLWX6aS3O8vsWnlbjOzGabrLMaPH6/hw4erW7duGjNmjPLmzSsXFxf169cv1Rfbd+jQQePGjdOlS5eUK1curVy5UiEhITZ//W7fvr1q166t5cuXa/Xq1Zo0aZImTpyoZcuWqXHjxvfdRrVq1ayz27Vs2VK1atXSSy+9pEOHDilnzpzW2gcOHGi9Duq/HAmcZva8r4mJiWrQoIHefvvtFPsm/bKa1l566SXNnDlTo0aNUlBQkMqXL3/P/qm9gXFK0nufQ0JCNGTIEPXo0UP58uVTw4YN7Xpdvnz51Lx5czVv3lzPPvusIiIi9Pfff1uvXUqtxx57TJK0b9++B1rP3dztc2bP589ejn7/7f3ZNXnyZIWGhmrFihVavXq1+vbtqwkTJmjr1q3Wo72S/Z+/tNxnAKlDSAJgl6ZNm2rWrFnauHGjdYY6s99++00nTpxQr169UrV+f39/JSYm6vjx4zZ/xT9y5Eiqa07J0qVLVbduXX355Zc27dHR0TZTSTuiQ4cOGj16tL755hsVKlRIsbGxKV6MXbhwYfXu3Vu9e/fWhQsXVLlyZY0bN86ukGSWLVs2TZgwQXXr1tW0adM0ePBg6xGp7NmzO3yEIC2UKlVKcXFx9922v7+/1q5dq7i4OJujSYcOHUrVdmvVqqXixYtr/fr1mjhx4j23m5iYqMjISOuRDOnO9NnR0dHWAJH038jISJujfBcvXkz2V3x79zm1ihcvrpo1a2r9+vV67bXXUjVlftWqVRUREaGzZ88+cEgqU6aMypYtqxUrVuijjz5KdjTwv/z9/bV3714lJibaHE3666+/rMvTUtLRVHMQOXz4sCRZZ3BMj+9/kooVK6pixYoaNmyYNm/erJo1a2rmzJkaO3as3Z8/AM6Da5IA2GXQoEHy9PRUr169FBUVZbPs8uXLevXVV+Xl5aVBgwalav1JRz9mzJhh0/7JJ5+kruC7yJYtW7K/xi5ZsiTV1+tId04fqlixohYvXqzFixercOHCeuaZZ6zLb9++bTMVsCQVLFhQRYoUSTb9r72effZZVatWTVOnTtX169dVsGBBPfvss/rss8909uzZZP1TuldOWmrfvr22bNmin3/+Odmy6Oho3bp1S9KdewDdunVLn376qXX57du3U/3vbLFY9PHHH2vkyJF6+eWX79qvSZMmkpRsJr8pU6ZIknUWxfr16yt79uz65JNPbD4nKc0AaO8+P4ixY8dq5MiR97xm69y5czpw4ECy9hs3bmjt2rUpnuaVWqNHj1ZUVJReeeWVFPdv9erV1mnvmzRponPnzmnx4sXW5bdu3dInn3yinDlzqk6dOmlSU5IzZ85YZ3GU7lxDNXfuXD355JPy9fWVlD7f/9jY2GTvRcWKFeXi4mL9ftv7+QPgPDiSBMAugYGBmjNnjjp27KiKFSuqe/fuKlmypE6cOKEvv/xSly5d0sKFC1WqVKlUrb9KlSpq06aNpk6dqqioKNWoUUMRERHWvwSn1WlSTZs21XvvvaeuXbvq6aef1r59+zR//ny7rw26mw4dOmjEiBHy8PBQ9+7dbf5yfuXKFRUrVkxt27ZVUFCQcubMqV9++UV//PGHJk+enOptDho0SO3atVN4eLheffVVTZ8+XbVq1VLFihXVo0cPBQQE6Pz589qyZYtOnz6dqntBOVLLypUr1bRpU4WGhqpKlSqKj4/Xvn37tHTpUp04cUL58+dXs2bNVLNmTQ0ePFgnTpyw3svmvyHSES1atFCLFi3u2ScoKEhdunTR559/rujoaNWpU0e///675syZo5YtW6pu3bqS7lx7NHDgQE2YMEFNmzZVkyZNtGvXLv3000/JjjTYu88Pok6dOvcNE6dPn1a1atX03HPPqV69evL19dWFCxe0cOFC7dmzR/369UtWx2+//abr168nW9cTTzyhJ5544q7b6tChg/bt26dx48Zp165dCgkJkb+/v6KiorRq1SqtXbvWen+nnj176rPPPlNoaKh27NihEiVKaOnSpdq0aZOmTp1q9wQQ9ipTpoy6d++uP/74Q4UKFdLs2bN1/vx5hYWFWfukx/d/3bp1ev3119WuXTuVKVNGt27d0rx585QtWza1adNGkv2fPwDOg5AEwG7t2rXTY489pgkTJliDUb58+VS3bl0NHTpUFSpUeKD1z507V76+vlq4cKGWL1+u+vXra/HixSpbtmya3eF+6NChio+P14IFC7R48WJVrlxZP/zwgwYPHvxA6+3QoYOGDRumq1ev2sxqJ925p0/v3r21evVqLVu2TImJiSpdurRmzJhhM8ubo1q3bq1SpUrpgw8+UI8ePVS+fHlt375do0ePVnh4uKKiolSwYEFVqlRJI0aMeKD9ux8vLy9FRERo/PjxWrJkiebOnStvb2+VKVNGo0ePtl7A7uLiopUrV6pfv3766quvZLFY1Lx5c02ePFmVKlVK1xq/+OILBQQEKDw8XMuXL5evr6+GDBmikSNH2vQbO3asPDw8NHPmTP3666+qXr26Vq9eneyv/fbuc3orW7aspk6dqh9//FEzZszQ+fPn5eHhoQoVKmjWrFnWCUXMPv744xTXNXLkyHuGJOnO+/Pcc8/p448/1qeffqrLly8rT548qlGjhlasWKHmzZtLunM9z/r16zV48GDNmTNHsbGxKlu2rMLCwhQaGvrA+/1fgYGB+uSTTzRo0CAdOnRIJUuW1OLFi22u0UuP739QUJAaNWqk7777Tv/884+8vLwUFBSkn376STVq1LD2s/fzB8A5WAyuAgTgxHbv3q1KlSrpq6++UseOHTO7HABOqESJEqpQoYL1VD8AeFBckwTAaVy7di1Z29SpU+Xi4mJzjQ8AAEB64nQ7AE7j/fff144dO1S3bl25urrqp59+0k8//aSePXsmu1cKAABAeiEkAXAaTz/9tNasWaMxY8YoLi5OxYsX16hRo/Tuu+9mdmkAAOARwjVJAAAAAGDCNUkAAAAAYEJIAgAAAACTLH9NUmJios6cOaNcuXKl2c0oAQAAADx8DMPQlStXVKRIEZsbv/9Xlg9JZ86cYVYsAAAAAFanTp1SsWLF7ro8y4ekXLlySbrzRnh7e2dyNQAAAAAyS2xsrPz8/KwZ4W6yfEhKOsXO29ubkAQAAADgvpfhMHEDAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABPXzC4AAABIslgyuwIASD+GkdkVOIQjSQAAAABgwpGkDMYfCgFkZQ/ZHwoBAEgRR5IAAAAAwISQBAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwyNSRNmDBBTz31lHLlyqWCBQuqZcuWOnTokE2f69evq0+fPsqXL59y5sypNm3a6Pz585lUMQAAAICsLlNDUkREhPr06aOtW7dqzZo1unnzpho2bKj4+Hhrn7feekvfffedlixZooiICJ05c0atW7fOxKoBAAAAZGUWwzCMzC4iycWLF1WwYEFFRETomWeeUUxMjAoUKKAFCxaobdu2kqS//vpL5cqV05YtW1SjRo37rjM2NlY+Pj6KiYmRt7d3eu/CfVksmV0BAKQf5xlRHkIMEACyMicZIOzNBk51TVJMTIwkKW/evJKkHTt26ObNm6pfv761z2OPPabixYtry5YtKa4jISFBsbGxNg8AAAAAsJfThKTExET169dPNWvWVIUKFSRJ586dk5ubm3Lnzm3Tt1ChQjp37lyK65kwYYJ8fHysDz8/v/QuHQAAAEAW4jQhqU+fPvrzzz+1aNGiB1rPkCFDFBMTY32cOnUqjSoEAAAA8ChwzewCJOn111/X999/rw0bNqhYsWLWdl9fX924cUPR0dE2R5POnz8vX1/fFNfl7u4ud3f39C4ZAAAAQBaVqUeSDMPQ66+/ruXLl2vdunUqWbKkzfIqVaooe/bsWrt2rbXt0KFDOnnypIKDgzO6XAAAAACPgEw9ktSnTx8tWLBAK1asUK5cuazXGfn4+MjT01M+Pj7q3r27+vfvr7x588rb21tvvPGGgoOD7ZrZDgAAAAAclalTgFvuMt1pWFiYQkNDJd25meyAAQO0cOFCJSQkqFGjRpoxY8ZdT7f7L6YAB4CM4yQzvD6cGCAAZGVOMkDYmw2c6j5J6YGQBAAZJ2uPKOmMAQJAVuYkA8RDeZ8kAAAAAMhshCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABNCEgAAAACYEJIAAAAAwISQBAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABNCEgAAAACYEJIAAAAAwISQBAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABOHQtKtW7f03nvv6fTp0+lVDwAAAABkKodCkqurqyZNmqRbt26lVz0AAAAAkKkcPt3uueeeU0RERHrUAgAAAACZzuGQ1LhxYw0ePFgDBw7UwoULtXLlSpuHIzZs2KBmzZqpSJEislgs+vbbb22Wh4aGymKx2Dyef/55R0sGAAAAALu5OvqC3r17S5KmTJmSbJnFYtHt27ftXld8fLyCgoLUrVs3tW7dOsU+zz//vMLCwqzP3d3dHawYAAAAAOzncEhKTExMs403btxYjRs3vmcfd3d3+fr6ptk2AQAAAOBenH4K8PXr16tgwYIqW7asXnvtNUVFRWV2SQAAAACysFSFpIiICDVr1kylS5dW6dKl1bx5c/32229pXZuef/55zZ07V2vXrtXEiRMVERGhxo0b3/OUvoSEBMXGxto8AAAAAMBeDoekr776SvXr15eXl5f69u2rvn37ytPTU/Xq1dOCBQvStLgXX3xRzZs3V8WKFdWyZUt9//33+uOPP7R+/fq7vmbChAny8fGxPvz8/NK0JgAAAABZm8UwDMORF5QrV049e/bUW2+9ZdM+ZcoUzZo1SwcPHkxdIRaLli9frpYtW96zX4ECBTR27Fj16tUrxeUJCQlKSEiwPo+NjZWfn59iYmLk7e2dqtrSksWS2RUAQPpxbESBDQYIAFmZkwwQsbGx8vHxuW82cPhI0rFjx9SsWbNk7c2bN9fx48cdXZ1DTp8+raioKBUuXPiufdzd3eXt7W3zAAAAAAB7ORyS/Pz8tHbt2mTtv/zyi8OntsXFxWn37t3avXu3JOn48ePavXu3Tp48qbi4OA0aNEhbt27ViRMntHbtWrVo0UKlS5dWo0aNHC0bAAAAAOzi8BTgAwYMUN++fbV79249/fTTkqRNmzYpPDxcH330kUPr2r59u+rWrWt93r9/f0lSly5d9Omnn2rv3r2aM2eOoqOjVaRIETVs2FBjxozhXkkAAAAA0o3D1yRJ0vLlyzV58mTr9UflypXToEGD1KJFizQv8EHZe95hRuGUcwBZmZOccv5wYoAAkJU5yQBhbzZw6EjSrVu3NH78eHXr1k0bN2584CIBAAAAwNk4dE2Sq6ur3n//fd26dSu96gEAAACATOXwxA316tVTREREetQCAAAAAJnO4YkbGjdurMGDB2vfvn2qUqWKcuTIYbO8efPmaVYcAAAAAGQ0hyducHG5+8Eni8Wi27dvP3BRaYmJGwAg4zjJdbkPJwYIAFmZkwwQ6TJxgyQlJiY+UGEAAAAA4Mwcuibp5s2bcnV11Z9//ple9QAAAABApnIoJGXPnl3Fixd3ulPqAAAAACCtODy73bvvvquhQ4fq8uXL6VEPAAAAAGQqh69JmjZtmo4cOaIiRYrI398/2ex2O3fuTLPiAAAAACCjORySWrZsmQ5lAAAAAIBzcHgK8IcNU4ADQMbJ2iNKOmOAAJCVOckAYW82sPuapN9///2eEzYkJCTo66+/dqxKAAAAAHAydoek4OBgRUVFWZ97e3vr2LFj1ufR0dEKCQlJ2+oAAAAAIIPZHZL+e1ZeSmfpZfEz9wAAAAA8AhyeAvxeLJxPDQAAAOAhl6YhCQAAAAAedg5NAX7gwAGdO3dO0p1T6/766y/FxcVJki5dupT21QEAAABABrN7CnAXFxdZLJYUrztKardYLPecAS8zMAU4AGQcLk19AAwQALIyJxkg7M0Gdh9JOn78eJoUBgAAAADOzO6Q5O/vn551AAAAAIBTYOIGAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABO7ZrerVKmSLHbev2Hnzp0PVBAAAAAAZCa7QlLLli2t/3/9+nXNmDFD5cuXV3BwsCRp69at2r9/v3r37p0uRQIAAABARrErJI0cOdL6/6+88or69u2rMWPGJOtz6tSptK0OAAAAADKYxTAMw5EX+Pj4aPv27QoMDLRpj4yMVNWqVRUTE5OmBT6o2NhY+fj4KCYmRt7e3pldjuw8axEAHkqOjSiwwQABICtzkgHC3mzg8MQNnp6e2rRpU7L2TZs2ycPDw9HVAQAAAIBTset0O7N+/frptdde086dO1WtWjVJ0rZt2zR79mwNHz48zQsEAAAAgIzkcEgaPHiwAgIC9NFHH+mrr76SJJUrV05hYWFq3759mhcIAAAAABnJ4WuSHjZckwQAGSdrjyjpjAECQFbmJANEul2TJEnR0dH64osvNHToUF2+fFnSnfsj/fPPP6mrFgAAAACchMOn2+3du1f169eXj4+PTpw4oVdeeUV58+bVsmXLdPLkSc2dOzc96gQAAACADOHwkaT+/fsrNDRUkZGRNrPZNWnSRBs2bEjT4gAAAAAgozkckv744w/16tUrWXvRokV17ty5NCkKAAAAADKLwyHJ3d1dsbGxydoPHz6sAgUKpElRAAAAAJBZHA5JzZs313vvvaebN29KkiwWi06ePKl33nlHbdq0SfMCAQAAACAjORySJk+erLi4OBUsWFDXrl1TnTp1VLp0aeXKlUvjxo1LjxoBAAAAIMM4PLudj4+P1qxZo02bNmnPnj2Ki4tT5cqVVb9+/fSoDwAAAAAylEMh6ebNm/L09NTu3btVs2ZN1axZM73qAgAAAIBM4dDpdtmzZ1fx4sV1+/bt9KoHAAAAADKVw9ckvfvuuxo6dKguX76cHvUAAAAAQKZy+JqkadOm6ciRIypSpIj8/f2VI0cOm+U7d+5Ms+IAAAAAIKM5HJJatmyZDmUAAAAAgHOwGIZhZHYR6Sk2NlY+Pj6KiYmRt7d3ZpcjiyWzKwCA9JO1R5R0xgABICtzkgHC3mzg8DVJAAAAAJCVOXy63e3bt/Xhhx/q66+/1smTJ3Xjxg2b5UzoAAAAAOBh5vCRpNGjR2vKlCnq0KGDYmJi1L9/f7Vu3VouLi4aNWpUOpQIAAAAABnH4ZA0f/58zZo1SwMGDJCrq6tCQkL0xRdfaMSIEdq6dWt61AgAAAAAGcbhkHTu3DlVrFhRkpQzZ07FxMRIkpo2baoffvghbasDAAAAgAzmcEgqVqyYzp49K0kqVaqUVq9eLUn6448/5O7unrbVAQAAAEAGczgktWrVSmvXrpUkvfHGGxo+fLgCAwPVuXNndevWLc0LBAAAAICM9MD3SdqyZYu2bNmiwMBANWvWLK3qSjPcJwkAMo6T3Abj4cQAASArc5IBwt5s4PAU4P8VHBys4ODgB10NAAAAADgFh0PS3Llz77m8c+fOqS4GAAAAADKbw6fb5cmTx+b5zZs3dfXqVbm5ucnLy8vpbibL6XYAkHGc5GyKhxMDBICszEkGCHuzgcMTN/z77782j7i4OB06dEi1atXSwoULH6hoAAAAAMhsDoeklAQGBup///uf3nzzzbRYHQAAAABkmjQJSZLk6uqqM2fOpNXqAAAAACBTODxxw8qVK22eG4ahs2fPatq0aapZs2aaFQYAAAAAmcHhkNSyZUub5xaLRQUKFNBzzz2nyZMnp1VdAAAAAJApHA5JiYmJ6VEHAAAAADiFNLsmCQAAAACyAoePJPXv39/uvlOmTHF09QAAAACQqRwOSbt27dKuXbt08+ZNlS1bVpJ0+PBhZcuWTZUrV7b2s3BTPAAAAAAPIYdDUrNmzZQrVy7NmTNHefLkkXTnBrNdu3ZV7dq1NWDAgDQvEgAAAAAyisUwDMORFxQtWlSrV6/W448/btP+559/qmHDhk53r6TY2Fj5+PgoJiZG3t7emV2OOMAGICtzbESBDQYIAFmZkwwQ9mYDhyduiI2N1cWLF5O1X7x4UVeuXHF0dQAAAADgVBwOSa1atVLXrl21bNkynT59WqdPn9Y333yj7t27q3Xr1ulRIwAAAABkGIevSZo5c6YGDhyol156STdv3ryzEldXde/eXZMmTUrzAgEAAAAgIzl8TVKS+Ph4HT16VJJUqlQp5ciRI00LSytckwQAGcdJTjl/ODFAAMjKnGSASLdrkpLkyJFDTzzxhHx8fPT3338rMTHR4XVs2LBBzZo1U5EiRWSxWPTtt9/aLDcMQyNGjFDhwoXl6emp+vXrKzIyMrUlAwAAAMB92R2SZs+enezmsD179lRAQIAqVqyoChUq6NSpUw5tPD4+XkFBQZo+fXqKy99//319/PHHmjlzprZt26YcOXKoUaNGun79ukPbAQAAAAB72R2SPv/8c+t9kSRp1apVCgsL09y5c/XHH38od+7cGj16tEMbb9y4scaOHatWrVolW2YYhqZOnaphw4apRYsWeuKJJzR37lydOXMm2REnAAAAAEgrdoekyMhIVa1a1fp8xYoVatGihTp27KjKlStr/PjxWrt2bZoVdvz4cZ07d07169e3tvn4+Kh69erasmXLXV+XkJCg2NhYmwcAAAAA2MvukHTt2jWbi5s2b96sZ555xvo8ICBA586dS7PCktZVqFAhm/ZChQrdczsTJkyQj4+P9eHn55dmNQEAAADI+uwOSf7+/tqxY4ck6dKlS9q/f79q1qxpXX7u3Dn5+PikfYUOGjJkiGJiYqwPR6+TAgAAAPBos/s+SV26dFGfPn20f/9+rVu3To899piqVKliXb5582ZVqFAhzQrz9fWVJJ0/f16FCxe2tp8/f15PPvnkXV/n7u4ud3f3NKsDAAAAwKPF7iNJb7/9tnr06KFly5bJw8NDS5YssVm+adMmhYSEpFlhJUuWlK+vr811TrGxsdq2bZuCg4PTbDsAAAAAYJbqm8mmhbi4OB05ckSSVKlSJU2ZMkV169ZV3rx5Vbx4cU2cOFH/+9//NGfOHJUsWVLDhw/X3r17deDAAXl4eNi1DW4mCwAZx0nuFfhwYoAAkJU5yQBhbzaw+3S79LB9+3bVrVvX+rx///6S7pzaFx4errffflvx8fHq2bOnoqOjVatWLa1atcrugAQAAAAAjsrUI0kZgSNJAJBxsvaIks4YIABkZU4yQNibDey+JgkAAAAAHgWEJAAAAAAwISQBAAAAgInDEzfcvn1b4eHhWrt2rS5cuKDExESb5evWrUuz4gAAAAAgozkckt58802Fh4frhRdeUIUKFWThQlMAAAAAWYjDIWnRokX6+uuv1aRJk/SoBwAAAAAylcPXJLm5ual06dLpUQsAAAAAZDqHQ9KAAQP00UcfKYvfXgkAAADAI8rh0+02btyoX3/9VT/99JMef/xxZc+e3Wb5smXL0qw4AAAAAMhoDoek3Llzq1WrVulRCwAAAABkOodDUlhYWHrUAQAAAABOgZvJAgAAAICJw0eSJGnp0qX6+uuvdfLkSd24ccNm2c6dO9OkMAAAAADIDA4fSfr444/VtWtXFSpUSLt27VK1atWUL18+HTt2TI0bN06PGgEAAAAgwzgckmbMmKHPP/9cn3zyidzc3PT2229rzZo16tu3r2JiYtKjRgAAAADIMA6HpJMnT+rpp5+WJHl6eurKlSuSpJdfflkLFy5M2+oAAAAAIIM5HJJ8fX11+fJlSVLx4sW1detWSdLx48e5wSwAAACAh57DIem5557TypUrJUldu3bVW2+9pQYNGqhDhw7cPwkAAADAQ89iOHj4JzExUYmJiXJ1vTMx3qJFi7R582YFBgaqV69ecnNzS5dCUys2NlY+Pj6KiYmRt7d3ZpcjiyWzKwCA9MMJBQ+AAQJAVuYkA4S92cDhkPSwISQBQMbJ2iNKOmOAAJCVOckAYW82SNXNZH/77Td16tRJwcHB+ueffyRJ8+bN08aNG1NXLQAAAAA4CYdD0jfffKNGjRrJ09NTu3btUkJCgiQpJiZG48ePT/MCAQAAACAjORySxo4dq5kzZ2rWrFnKnj27tb1mzZrauXNnmhYHAAAAABnN4ZB06NAhPfPMM8nafXx8FB0dnRY1AQAAAECmSdV9ko4cOZKsfePGjQoICEiTogAAAAAgszgcknr06KE333xT27Ztk8Vi0ZkzZzR//nwNHDhQr732WnrUCAAAAAAZxtXRFwwePFiJiYmqV6+erl69qmeeeUbu7u4aOHCg3njjjfSoEQAAAAAyTKrvk3Tjxg0dOXJEcXFxKl++vHLmzJnWtaUJ7pMEABnHSW6D8XBigACQlTnJAGFvNnD4SFISNzc3lS9fPrUvBwAAAACnZHdI6tatm139Zs+enepiAAAAACCz2R2SwsPD5e/vr0qVKimVZ+gBAAAAgNOzOyS99tprWrhwoY4fP66uXbuqU6dOyps3b3rWBgAAAAAZzu4pwKdPn66zZ8/q7bff1nfffSc/Pz+1b99eP//8M0eWAAAAAGQZDt0nyd3dXSEhIVqzZo0OHDigxx9/XL1791aJEiUUFxeXXjUCAAAAQIZx+Gay1he6uMhiscgwDN2+fTstawIAAACATONQSEpISNDChQvVoEEDlSlTRvv27dO0adN08uRJp71PEgAAAAA4wu6JG3r37q1FixbJz89P3bp108KFC5U/f/70rA0AAAAAMpzFsHPWBRcXFxUvXlyVKlWS5R53BV+2bFmaFZcW7L2rbkbhhuoAsjLm8XkADBAAsjInGSDszQZ2H0nq3LnzPcMRAAAAAGQFDt1MFgAAAACyulTPbgcAAAAAWREhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABNCEgAAAACYEJIAAAAAwISQBAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACAiVOHpFGjRslisdg8HnvsscwuCwAAAEAW5prZBdzP448/rl9++cX63NXV6UsGAAAA8BBz+sTh6uoqX1/fzC4DAAAAwCPCqU+3k6TIyEgVKVJEAQEB6tixo06ePJnZJQEAAADIwpz6SFL16tUVHh6usmXL6uzZsxo9erRq166tP//8U7ly5UrxNQkJCUpISLA+j42NzahyAQAAAGQBFsMwjMwuwl7R0dHy9/fXlClT1L179xT7jBo1SqNHj07WHhMTI29v7/Qu8b4slsyuAADSz8MzojghBggAWZmTDBCxsbHy8fG5bzZw+tPtzHLnzq0yZcroyJEjd+0zZMgQxcTEWB+nTp3KwAoBAAAAPOweqpAUFxeno0ePqnDhwnft4+7uLm9vb5sHAAAAANjLqUPSwIEDFRERoRMnTmjz5s1q1aqVsmXLppCQkMwuDQAAAEAW5dQTN5w+fVohISGKiopSgQIFVKtWLW3dulUFChTI7NIAAAAAZFFOHZIWLVqU2SUAAAAAeMQ49el2AAAAAJDRCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABNCEgAAAACYEJIAAAAAwISQBAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABNCEgAAAACYEJIAAAAAwISQBAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYEJIAgAAAAATQhIAAAAAmBCSAAAAAMCEkAQAAAAAJoQkAAAAADAhJAEAAACACSEJAAAAAEwISQAAAABgQkgCAAAAABNCEgAAAACYEJIAAAAAwISQBAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACaEJAAAAAAwISQBAAAAgAkhCQAAAABMCEkAAAAAYPJQhKTp06erRIkS8vDwUPXq1fX7779ndkkAAAAAsiinD0mLFy9W//79NXLkSO3cuVNBQUFq1KiRLly4kNmlAQAAAMiCnD4kTZkyRT169FDXrl1Vvnx5zZw5U15eXpo9e3ZmlwYAAAAgC3LN7ALu5caNG9qxY4eGDBlibXNxcVH9+vW1ZcuWFF+TkJCghIQE6/OYmBhJUmxsbPoWCwAQP2oBAClykgEiKRMYhnHPfk4dki5duqTbt2+rUKFCNu2FChXSX3/9leJrJkyYoNGjRydr9/PzS5caAQD/x8cnsysAADglJxsgrly5Ip971OTUISk1hgwZov79+1ufJyYm6vLly8qXL58sFksmVgZkvNjYWPn5+enUqVPy9vbO7HIAAE6AsQGPMsMwdOXKFRUpUuSe/Zw6JOXPn1/ZsmXT+fPnbdrPnz8vX1/fFF/j7u4ud3d3m7bcuXOnV4nAQ8Hb25uBEABgg7EBj6p7HUFK4tQTN7i5ualKlSpau3attS0xMVFr165VcHBwJlYGAAAAIKty6iNJktS/f3916dJFVatWVbVq1TR16lTFx8era9eumV0aAAAAgCzI6UNShw4ddPHiRY0YMULnzp3Tk08+qVWrViWbzAFAcu7u7ho5cmSyU1ABAI8uxgbg/izG/ea/AwAAAIBHiFNfkwQAAAAAGY2QBAAAAAAmhCQAAAAAMCEkAenkxIkTslgs2r17t92vCQ8PT/P7eqWmjvSQmn2zWCz69ttv06UeAHAGmzZtUsWKFZU9e3a1bNlS69evl8ViUXR0dLpul7EBuDdCEnAPp06dUrdu3VSkSBG5ubnJ399fb775pqKiou77Wj8/P509e1YVKlSwe3sdOnTQ4cOHH6TkVHn22WdlsVj0v//9L9myF154QRaLRaNGjcrwugDAWYWGhspischisSh79uwqWbKk3n77bV2/ft2h9fTv319PPvmkjh8/rvDwcD399NM6e/asXTe7TG+MDXiUEZKAuzh27JiqVq2qyMhILVy4UEeOHNHMmTOtNzO+fPnyXV9748YNZcuWTb6+vnJ1tX+mfU9PTxUsWDAtyneYn5+fwsPDbdr++ecfrV27VoULF86UmgDAmT3//PM6e/asjh07pg8//FCfffaZRo4c6dA6jh49queee07FihVT7ty55ebmJl9fX1kslnSq2jGMDXhUEZKAu+jTp4/c3Ny0evVq1alTR8WLF1fjxo31yy+/6J9//tG7775r7VuiRAmNGTNGnTt3lre3t3r27JniqQwrV65UYGCgPDw8VLduXc2ZM8fmtIr/nnYwatQoPfnkk5o3b55KlCghHx8fvfjii7py5Yq1z6pVq1SrVi3lzp1b+fLlU9OmTXX06FGH97dp06a6dOmSNm3aZG2bM2eOGjZsmCy4/fvvv+rcubPy5MkjLy8vNW7cWJGRkTZ9wsPDVbx4cXl5ealVq1YpHn1bsWKFKleuLA8PDwUEBGj06NG6deuWw7UDQGZwd3eXr6+v/Pz81LJlS9WvX19r1qyxLk9MTNSECRNUsmRJeXp6KigoSEuXLpX0f6e7RUVFqVu3brJYLAoPD092ul3SuPDzzz+rXLlyypkzpzWcmX3xxRcqV66cPDw89Nhjj2nGjBk2y3///XdVqlRJHh4eqlq1qnbt2mXXPjI24FFFSAJScPnyZf3888/q3bu3PD09bZb5+vqqY8eOWrx4scy3Gfvggw8UFBSkXbt2afjw4cnWefz4cbVt21YtW7bUnj171KtXL5ugdTdHjx7Vt99+q++//17ff/+9IiIibE59iI+PV//+/bV9+3atXbtWLi4uatWqlRITEx3aZzc3N3Xs2FFhYWHWtvDwcHXr1i1Z39DQUG3fvl0rV67Uli1bZBiGmjRpops3b0qStm3bpu7du+v111/X7t27VbduXY0dO9ZmHb/99ps6d+6sN998UwcOHNBnn32m8PBwjRs3zqG6AcAZ/Pnnn9q8ebPc3NysbRMmTNDcuXM1c+ZM7d+/X2+99ZY6deqkiIgI6ynZ3t7emjp1qs6ePasOHTqkuO6rV6/qgw8+0Lx587RhwwadPHlSAwcOtC6fP3++RowYoXHjxungwYMaP368hg8frjlz5kiS4uLi1LRpU5UvX147duzQqFGjbF5/L4wNeGQZAJLZunWrIclYvnx5isunTJliSDLOnz9vGIZh+Pv7Gy1btrTpc/z4cUOSsWvXLsMwDOOdd94xKlSoYNPn3XffNSQZ//77r2EYhhEWFmb4+PhYl48cOdLw8vIyYmNjrW2DBg0yqlevftfaL168aEgy9u3bl2IdKalTp47x5ptvGrt37zZy5cplxMXFGREREUbBggWNmzdvGkFBQcbIkSMNwzCMw4cPG5KMTZs2WV9/6dIlw9PT0/j6668NwzCMkJAQo0mTJjbb6NChg82+1atXzxg/frxNn3nz5hmFCxe2Pr/XvwEAZKYuXboY2bJlM3LkyGG4u7sbkgwXFxdj6dKlhmEYxvXr1w0vLy9j8+bNNq/r3r27ERISYn3u4+NjhIWFWZ//+uuvycYFScaRI0esfaZPn24UKlTI+rxUqVLGggULbLYzZswYIzg42DAMw/jss8+MfPnyGdeuXbMu//TTTxkbgHuw/2IJ4BFkmI4U3U/VqlXvufzQoUN66qmnbNqqVat23/WWKFFCuXLlsj4vXLiwLly4YH0eGRmpESNGaNu2bbp06ZL1CNLJkycdmjRCkoKCghQYGKilS5fq119/1csvv5zsmqqDBw/K1dVV1atXt7bly5dPZcuW1cGDB619WrVqZfO64OBgrVq1yvp8z5492rRpk81fB2/fvq3r16/r6tWr8vLycqh2AMhodevW1aeffqr4+Hh9+OGHcnV1VZs2bSRJR44c0dWrV9WgQQOb19y4cUOVKlVyaDteXl4qVaqU9bl5HIiPj9fRo0fVvXt39ejRw9rn1q1b1skfDh48qCeeeEIeHh7W5cHBwXZvn7EBjyJCEpCC0qVLy2KxpPgDXbrzgz5PnjwqUKCAtS1HjhzpUkv27NltnlssFptT6Zo1ayZ/f3/NmjVLRYoUUWJioipUqKAbN26kanvdunXT9OnTdeDAAf3+++8PVPu9xMXFafTo0WrdunWyZeaBHACcVY4cOVS6dGlJ0uzZsxUUFKQvv/xS3bt3V1xcnCTphx9+UNGiRW1e5+7u7tB2UhoHkv6Il7SdWbNm2QQUScqWLZtD27kXxgY8argmCUhBvnz51KBBA82YMUPXrl2zWXbu3DnNnz9fHTp0cGj2obJly2r79u02bX/88ccD1RkVFaVDhw5p2LBhqlevnsqVK6d///33gdb50ksvad++fapQoYLKly+fbHm5cuV069Ytbdu2LVkdSf3LlStns1yStm7davO8cuXKOnTokEqXLp3s4eLCjyYADxcXFxcNHTpUw4YN07Vr11S+fHm5u7vr5MmTyX7G+fn5pdl2CxUqpCJFiujYsWPJtlOyZElJd34m792712Z68v/+TL4fxgY8avi0AXcxbdo0JSQkqFGjRtqwYYNOnTqlVatWqUGDBipatKjDF5H26tVLf/31l9555x0dPnxYX3/9tXVa1dRO9ZonTx7ly5dPn3/+uY4cOaJ169apf//+qVqXeZ1nz57V2rVrU1weGBioFi1aqEePHtq4caP27NmjTp06qWjRomrRooUkqW/fvlq1apU++OADRUZGatq0aTanU0jSiBEjNHfuXI0ePVr79+/XwYMHtWjRIg0bNuyB6geAzNKuXTtly5ZN06dPV65cuTRw4EC99dZbmjNnjo4ePaqdO3fqk08+sU6okFZGjx6tCRMm6OOPP9bhw4e1b98+hYWFacqUKZLuBByLxaIePXrowIED+vHHH/XBBx84tA3GBjxqCEnAXQQGBmr79u0KCAhQ+/btVapUKfXs2VN169bVli1blDdvXofWV7JkSS1dulTLli3TE088oU8//dQ6u52jp14kcXFx0aJFi7Rjxw5VqFBBb731liZNmpSqdZnlzp37nqcPhoWFqUqVKmratKmCg4NlGIZ+/PFH6ykhNWrU0KxZs/TRRx8pKChIq1evTjbANWrUSN9//71Wr16tp556SjVq1NCHH34of3//B64fADKDq6urXn/9db3//vuKj4/XmDFjNHz4cE2YMEHlypXT888/rx9++MF6hCetvPLKK/riiy8UFhamihUrqk6dOgoPD7duJ2fOnPruu++0b98+VapUSe+++64mTpzo8HYYG/AosRiOXJkOIE2NGzdOM2fO1KlTpzK7FAAAAPx/TNwAZKAZM2boqaeeUr58+bRp0yZNmjRJr7/+emaXBQAAABNCEpCBIiMjNXbsWF2+fFnFixfXgAEDNGTIkMwuCwAAACacbgcAAAAAJkzcAAAAAAAmhCQAAAAAMCEkAQAAAIAJIQkAAAAATAhJAAAAAGBCSAIAAAAAE0ISAAAAAJgQkgAAAADAhJAEAAAAACb/DxncqdJQhWxlAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Splitting the data using custom_split\n",
    "X_train, X_test, y_train, y_test = custom_split(df, 'medv')\n",
    "\n",
    "# Fitting the original model with all features\n",
    "theta_best = custom_lr(X_train, y_train)\n",
    "\n",
    "# Predicting on the test set with the original model\n",
    "y_pred_original = custom_predict(X_test, theta_best)\n",
    "mse_original = custom_mean_squared_error(y_test, y_pred_original)\n",
    "\n",
    "# Refining the model to include only 'rm' and 'rad'\n",
    "X_train_refined = X_train[['crim', 'zn', 'nox', 'rm', 'dis', 'rad', 'tax', 'ptratio', 'lstat']]\n",
    "X_test_refined = X_test[['crim', 'zn', 'nox', 'rm', 'dis', 'rad', 'tax', 'ptratio', 'lstat']]\n",
    "theta_refined = custom_lr(X_train_refined, y_train)\n",
    "\n",
    "# Predicting on the test set with the refined model\n",
    "y_pred_refined = custom_predict(X_test_refined, theta_refined)\n",
    "mse_refined = custom_mean_squared_error(y_test, y_pred_refined)\n",
    "\n",
    "# Coefficients of the refined model\n",
    "refined_coefficients = pd.DataFrame([theta_refined[1:]], columns=X_train_refined.columns, index=['Coefficient']).T\n",
    "\n",
    "print(\"Original Model MSE:\", mse_original)\n",
    "print(\"Refined Model MSE:\", mse_refined)\n",
    "# Plotting the results\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.bar(['Original Model', 'Refined Model'], [mse_original, mse_refined], color=['blue', 'red'])\n",
    "plt.ylabel('Mean Squared Error')\n",
    "plt.title('Original vs Refined Model MSE Comparison')\n",
    "plt.show()\n",
    "\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.bar(refined_coefficients.index, refined_coefficients['Coefficient'], color='blue')\n",
    "plt.ylabel('Coefficient Value')\n",
    "plt.title('Coefficients of the Refined Model')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bar charts are used to compare the MSE of the original and refined models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Splitting the data\n",
    "X_train, X_test, y_train, y_test = custom_split(df, 'medv')\n",
    "\n",
    "# Fitting the original model\n",
    "theta_best = custom_lr(X_train, y_train)\n",
    "\n",
    "# Predicting on the test set with the original model\n",
    "y_pred_original = custom_predict(X_test, theta_best)\n",
    "\n",
    "# Refining the model to include only 'rm' and 'rad'\n",
    "X_train_refined = X_train[['rm', 'rad']]\n",
    "X_test_refined = X_test[['rm', 'rad']]\n",
    "theta_refined = custom_lr(X_train_refined, y_train)\n",
    "\n",
    "# Predicting on the test set with the refined model\n",
    "y_pred_refined = custom_predict(X_test_refined, theta_refined)\n",
    "\n",
    "# Preparing data for violin plot\n",
    "plot_data = pd.DataFrame({\n",
    "    'Actual Values': y_test,\n",
    "    'Original Model Predictions': y_pred_original,\n",
    "    'Refined Model Predictions': y_pred_refined\n",
    "})\n",
    "\n",
    "# Melting the DataFrame for use with seaborn\n",
    "plot_data_melted = plot_data.melt(var_name='Group', value_name='MEDV')\n",
    "\n",
    "# Plotting violin plots\n",
    "plt.figure(figsize=(12, 6))\n",
    "sns.violinplot(x='Group', y='MEDV', data=plot_data_melted)\n",
    "plt.title('Comparison of Prediction Distributions')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Violin plots are drawn to display the distribution of predictions from both models against the actual values, offering a comprehensive view of the model's performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results of the refined model\n",
    "\n",
    "*Mean Squared Error (MSE) Comparison*:\n",
    "- The MSE of the original model was 21.576.\n",
    "- After refining the model to focus on the variables RM (number of rooms) and RAD\n",
    "(accessibility to highways), the MSE increased to 21.69. This increase suggests that while these two variables are important, other variables might be needed to be included to have a better model and predictions\n",
    "\n",
    "*Coefficients of the Refined Model*:\n",
    "- RM (Rooms): Coefficient of 8.341. This larger coefficient in the refined model\n",
    "underscores the strong positive impact of the number of rooms on house value.\n",
    "- RAD (Accessibility to Highways): Coefficient of -0.266. In the refined model, this variable shows a negative impact, contrary to the original model. This suggests that when considered alone with RM, highway accessibility might be inversely related to house values.\n",
    "\n",
    "*Visualizations*:\n",
    "1) MSE Comparison: The bar chart compares the MSEs of the original and refined models, illustrating the change in prediction accuracy.\n",
    "2) Coefficients of the Refined Model: The bar chart displays the coefficients of RM and RAD in the refined model, highlighting their relative impacts on house value."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Some Considerations \n",
    "Since the model was so little and the split function made by myself was not randomized enough it means that the MSE and the impact of the coefficients are not as accurate as the pre defined one like from the 'sklearn' library or the 'xgboost' ones below we can see how close or far we got from these"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from xgboost import XGBRegressor\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Assuming df is your DataFrame and 'medv' is the target column\n",
    "# Splitting the data using custom_split\n",
    "X_train_custom, X_test_custom, y_train_custom, y_test_custom = custom_split(df, 'medv')\n",
    "\n",
    "# Splitting the data using sklearn's train_test_split\n",
    "X_train_sklearn, X_test_sklearn, y_train_sklearn, y_test_sklearn = train_test_split(df.drop(columns='medv'), df['medv'], test_size=0.1, random_state=42)\n",
    "\n",
    "# Initialize lists to store MSEs\n",
    "mse_custom_split = []\n",
    "mse_sklearn_split = []\n",
    "\n",
    "# Custom Linear Regression Model\n",
    "theta_custom = custom_lr(X_train_custom, y_train_custom)\n",
    "y_pred_custom = custom_predict(X_test_custom, theta_custom)\n",
    "mse_custom_split.append(custom_mean_squared_error(y_test_custom, y_pred_custom))\n",
    "\n",
    "theta_sklearn = custom_lr(X_train_sklearn, y_train_sklearn)\n",
    "y_pred_sklearn = custom_predict(X_test_sklearn, theta_sklearn)\n",
    "mse_sklearn_split.append(custom_mean_squared_error(y_test_sklearn, y_pred_sklearn))\n",
    "\n",
    "# Sklearn Linear Regression Model\n",
    "lr_model_custom = LinearRegression()\n",
    "lr_model_custom.fit(X_train_custom, y_train_custom)\n",
    "y_pred_lr_custom = lr_model_custom.predict(X_test_custom)\n",
    "mse_custom_split.append(mean_squared_error(y_test_custom, y_pred_lr_custom))\n",
    "\n",
    "lr_model_sklearn = LinearRegression()\n",
    "lr_model_sklearn.fit(X_train_sklearn, y_train_sklearn)\n",
    "y_pred_lr_sklearn = lr_model_sklearn.predict(X_test_sklearn)\n",
    "mse_sklearn_split.append(mean_squared_error(y_test_sklearn, y_pred_lr_sklearn))\n",
    "\n",
    "# XGBoost Model\n",
    "xgb_model_custom = XGBRegressor()\n",
    "xgb_model_custom.fit(X_train_custom, y_train_custom)\n",
    "y_pred_xgb_custom = xgb_model_custom.predict(X_test_custom)\n",
    "mse_custom_split.append(mean_squared_error(y_test_custom, y_pred_xgb_custom))\n",
    "\n",
    "xgb_model_sklearn = XGBRegressor()\n",
    "xgb_model_sklearn.fit(X_train_sklearn, y_train_sklearn)\n",
    "y_pred_xgb_sklearn = xgb_model_sklearn.predict(X_test_sklearn)\n",
    "mse_sklearn_split.append(mean_squared_error(y_test_sklearn, y_pred_xgb_sklearn))\n",
    "\n",
    "# Plotting the results\n",
    "plt.figure(figsize=(12, 6))\n",
    "model_names = ['Custom Linear Regression', 'Sklearn Linear Regression', 'XGBoost']\n",
    "bar_width = 0.35\n",
    "index = np.arange(len(model_names))\n",
    "\n",
    "plt.bar(index, mse_custom_split, bar_width, label='Custom Split')\n",
    "plt.bar(index + bar_width, mse_sklearn_split, bar_width, label='Sklearn Split')\n",
    "\n",
    "plt.xlabel('Model')\n",
    "plt.ylabel('Mean Squared Error')\n",
    "plt.title('MSE Comparison by Split Method')\n",
    "plt.xticks(index + bar_width / 2, model_names)\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see if we had used the already existing split functions the model would have been much better "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Conclusion\n",
    "\n",
    "In conclusion, this analysis employs a data-driven approach to unravel the intricate factors influencing house values in Boston. By considering a broad spectrum of socio-economic and environmental variables, the study provides valuable insights into the housing market, highlighting the significance of certain features while also emphasizing the complexity and multifaceted nature of real estate valuation.\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}