{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a200c150-c137-4519-8069-261814c4fad8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from matplotlib.pyplot import subplots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d0944276-c72e-4c95-a85e-77cab3ee9322",
   "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>family</th>\n",
       "      <th>father</th>\n",
       "      <th>mother</th>\n",
       "      <th>midparentHeight</th>\n",
       "      <th>children</th>\n",
       "      <th>childNum</th>\n",
       "      <th>sex</th>\n",
       "      <th>childHeight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>73.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>female</td>\n",
       "      <td>69.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>69.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  family  father  mother  midparentHeight  children  childNum     sex  \\\n",
       "0      1    78.5    67.0            75.43         4         1    male   \n",
       "1      1    78.5    67.0            75.43         4         2  female   \n",
       "2      1    78.5    67.0            75.43         4         3  female   \n",
       "\n",
       "   childHeight  \n",
       "0         73.2  \n",
       "1         69.2  \n",
       "2         69.0  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('family_heights.csv')\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "aa65c161-d953-4395-80eb-3a1c912e393e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.6373609 , 22.63624055])"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "coefficients = np.polyfit(data['midparentHeight'], data['childHeight'], 1)\n",
    "coefficients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "661ab22b-70b0-47b2-ba95-fac314b23d95",
   "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>family</th>\n",
       "      <th>father</th>\n",
       "      <th>mother</th>\n",
       "      <th>midparentHeight</th>\n",
       "      <th>children</th>\n",
       "      <th>childNum</th>\n",
       "      <th>sex</th>\n",
       "      <th>childHeight</th>\n",
       "      <th>regressionEstimate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>73.2</td>\n",
       "      <td>70.712373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>female</td>\n",
       "      <td>69.2</td>\n",
       "      <td>70.712373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>69.0</td>\n",
       "      <td>70.712373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>female</td>\n",
       "      <td>69.0</td>\n",
       "      <td>70.712373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>75.5</td>\n",
       "      <td>66.5</td>\n",
       "      <td>73.66</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>73.5</td>\n",
       "      <td>69.584244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2</td>\n",
       "      <td>75.5</td>\n",
       "      <td>66.5</td>\n",
       "      <td>73.66</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>male</td>\n",
       "      <td>72.5</td>\n",
       "      <td>69.584244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2</td>\n",
       "      <td>75.5</td>\n",
       "      <td>66.5</td>\n",
       "      <td>73.66</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>65.5</td>\n",
       "      <td>69.584244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2</td>\n",
       "      <td>75.5</td>\n",
       "      <td>66.5</td>\n",
       "      <td>73.66</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>female</td>\n",
       "      <td>65.5</td>\n",
       "      <td>69.584244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3</td>\n",
       "      <td>75.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>72.06</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>71.0</td>\n",
       "      <td>68.564467</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3</td>\n",
       "      <td>75.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>72.06</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>female</td>\n",
       "      <td>68.0</td>\n",
       "      <td>68.564467</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  family  father  mother  midparentHeight  children  childNum     sex  \\\n",
       "0      1    78.5    67.0            75.43         4         1    male   \n",
       "1      1    78.5    67.0            75.43         4         2  female   \n",
       "2      1    78.5    67.0            75.43         4         3  female   \n",
       "3      1    78.5    67.0            75.43         4         4  female   \n",
       "4      2    75.5    66.5            73.66         4         1    male   \n",
       "5      2    75.5    66.5            73.66         4         2    male   \n",
       "6      2    75.5    66.5            73.66         4         3  female   \n",
       "7      2    75.5    66.5            73.66         4         4  female   \n",
       "8      3    75.0    64.0            72.06         2         1    male   \n",
       "9      3    75.0    64.0            72.06         2         2  female   \n",
       "\n",
       "   childHeight  regressionEstimate  \n",
       "0         73.2           70.712373  \n",
       "1         69.2           70.712373  \n",
       "2         69.0           70.712373  \n",
       "3         69.0           70.712373  \n",
       "4         73.5           69.584244  \n",
       "5         72.5           69.584244  \n",
       "6         65.5           69.584244  \n",
       "7         65.5           69.584244  \n",
       "8         71.0           68.564467  \n",
       "9         68.0           68.564467  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1 = data.assign(regressionEstimate = coefficients[0]*data['midparentHeight'] + coefficients[1])\n",
    "data1.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e93481ab-1f55-4107-89ea-537a922629d9",
   "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>family</th>\n",
       "      <th>father</th>\n",
       "      <th>mother</th>\n",
       "      <th>midparentHeight</th>\n",
       "      <th>children</th>\n",
       "      <th>childNum</th>\n",
       "      <th>sex</th>\n",
       "      <th>childHeight</th>\n",
       "      <th>regressionEstimate</th>\n",
       "      <th>residuals</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>73.2</td>\n",
       "      <td>70.712373</td>\n",
       "      <td>2.487627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>female</td>\n",
       "      <td>69.2</td>\n",
       "      <td>70.712373</td>\n",
       "      <td>-1.512373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>69.0</td>\n",
       "      <td>70.712373</td>\n",
       "      <td>-1.712373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>78.5</td>\n",
       "      <td>67.0</td>\n",
       "      <td>75.43</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>female</td>\n",
       "      <td>69.0</td>\n",
       "      <td>70.712373</td>\n",
       "      <td>-1.712373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>75.5</td>\n",
       "      <td>66.5</td>\n",
       "      <td>73.66</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>73.5</td>\n",
       "      <td>69.584244</td>\n",
       "      <td>3.915756</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  family  father  mother  midparentHeight  children  childNum     sex  \\\n",
       "0      1    78.5    67.0            75.43         4         1    male   \n",
       "1      1    78.5    67.0            75.43         4         2  female   \n",
       "2      1    78.5    67.0            75.43         4         3  female   \n",
       "3      1    78.5    67.0            75.43         4         4  female   \n",
       "4      2    75.5    66.5            73.66         4         1    male   \n",
       "\n",
       "   childHeight  regressionEstimate  residuals  \n",
       "0         73.2           70.712373   2.487627  \n",
       "1         69.2           70.712373  -1.512373  \n",
       "2         69.0           70.712373  -1.712373  \n",
       "3         69.0           70.712373  -1.712373  \n",
       "4         73.5           69.584244   3.915756  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data2 = data1.assign(residuals = data1['childHeight'] - data1['regressionEstimate'])\n",
    "data2.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "74f95794-e9b8-476d-9c6c-123545de1e84",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = subplots(figsize = (5, 5))\n",
    "ax.scatter(data2['midparentHeight'], data2['residuals'])\n",
    "ax.axhline(y=0, color='red', linewidth = 3);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a81c96b8-4537-4a5f-8cf9-fe8ef84b8bc9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean of the residuals =  -7.323002542212767e-14\n"
     ]
    }
   ],
   "source": [
    "# We showed in class that the mean of the residuals equals 0. \n",
    "# Let us check this for the heights data\n",
    "print('Mean of the residuals = ', data2['residuals'].mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80c6bff5-a36f-4f84-a4f4-433dee8d20c6",
   "metadata": {},
   "source": [
    "Due to precision errors, the value does not come out to be exactly 0, but it is really close!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "63f67ea4-1f4f-4009-9cf8-4b0df0f03354",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation of midparentHeights and residuals =  -1.3549018592547941e-15\n"
     ]
    }
   ],
   "source": [
    "# We showed in class that the correlation between the predictor and the residuals is 0\n",
    "# Let us check for the heights data\n",
    "print('Correlation of midparentHeights and residuals = ', data2['midparentHeight'].corr(data2['residuals']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c56c4931-7d0c-4405-bdf5-dc8815e3a4d1",
   "metadata": {},
   "source": [
    "Again, due to precision errors, the value does not come out to be exactly 0, but it is really close!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "6b8cfec3-4aba-4e7f-ad79-e34e6f12be7a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(934, 10)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "71ba3007-6c70-4d47-99a7-2ae5918dfb75",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Birth Weight')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Let's study Birth Weight versus Gestational Days\n",
    "baby = pd.read_csv('baby.csv')\n",
    "fig, ax = subplots(figsize = (5, 5))\n",
    "ax.scatter(baby['Gestational Days'], baby['Birth Weight'])\n",
    "ax.set_xlabel('Gestational Days')\n",
    "ax.set_ylabel('Birth Weight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8b055ab1-7feb-484a-bc50-4d589b13bc4f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.float64(0.40754279338885113)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's calculate the correlation\n",
    "r = baby['Gestational Days'].corr(baby['Birth Weight'])\n",
    "r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "9b34b568-523a-4193-a096-80fca9a7f700",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The 95% confidence interval on the true slope is: ( 0.3845205020430253 , 0.5556891240274472 )\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# In this section, we use bootstrap to compute the 95% confidence interval on the true slope\n",
    "# Note that the bootstrap works only under the regression model given in the class slides\n",
    "num_repetitions = 5000\n",
    "\n",
    "slopes = []\n",
    "for i in range(num_repetitions):\n",
    "    sample = baby.sample(frac=1, replace = True)\n",
    "    coefficients = np.polyfit(sample['Gestational Days'], sample['Birth Weight'], 1)\n",
    "    slopes.append(coefficients[0])\n",
    "\n",
    "#plot the histogram of slopes and mark the 95% confidence interval\n",
    "slopes = np.array(slopes)\n",
    "left = np.percentile(slopes, 2.5)\n",
    "right = np.percentile(slopes, 97.5)\n",
    "fig, ax = subplots(figsize = (5, 5))\n",
    "ax.hist(slopes, density = True, bins=50)\n",
    "ax.hlines(0, left, right, color = 'yellow', linewidth = 7)\n",
    "print('The 95% confidence interval on the true slope is: (', left, ',', right, ')')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5173fa03-68f6-4bd5-b610-4bcf6dcd6946",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation =  0.0269829110029295\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can use 95% confidence interval idea to test the hypothesis that there is no linear association between two variables\n",
    "# Let's first check Birth weight versus maternal age which does not seem to be correlated\n",
    "fig, ax = subplots(figsize = (5, 5))\n",
    "ax.scatter(baby['Maternal Age'], baby['Birth Weight'])\n",
    "\n",
    "print('Correlation = ', baby['Maternal Age'].corr(baby['Birth Weight']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56e4a547-46f2-43f8-9522-3d311192ef36",
   "metadata": {},
   "source": [
    "We perform the following hypothesis test using 95% confidence intervals on slope:\n",
    " - Null hypothesis: The slope of the true line is 0\n",
    " - Alternate hypothesis: The slope of the true line is not 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "d859ee4c-b5f6-4c6b-b43d-ed0033886f3c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The 95% confidence interval on slope of true line is: ( -0.09600622904085086 ,  0.27336580374007435 )\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Estimating 95% confidence interval on the true slope using bootstrapping\n",
    "num_repetitions = 5000\n",
    "\n",
    "slopes = []\n",
    "for i in range(num_repetitions):\n",
    "    sample = baby.sample(frac = 1, replace=True)\n",
    "    coefficients = np.polyfit(sample['Maternal Age'], sample['Birth Weight'], 1)\n",
    "    slopes.append(coefficients[0])\n",
    "\n",
    "# plot a histogram of the slopes\n",
    "#calculate the percentiles\n",
    "slopes = np.array(slopes)\n",
    "left = np.percentile(slopes, 2.5)\n",
    "right = np.percentile(slopes, 97.5)\n",
    "fig, ax = subplots(figsize = (5, 5))\n",
    "ax.hist(slopes, density=True, bins=50);\n",
    "ax.hlines(y=0, xmin=left, xmax=right, color = 'yellow', linewidth = 7)\n",
    "print('The 95% confidence interval on slope of true line is: (', left, ', ', right, ')')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9d1c81c-ffb7-4ad4-aa83-6dbcd2ffa77d",
   "metadata": {},
   "source": [
    "Since the 95% confidence interval includes 0, we cannot reject the null hypothesis. So, conclusion is that guessing birth weight based on the maternal age is not a good idea."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "2a983604-6383-4e4b-ae51-496f0ce058a8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The 95% confidence interval on the prediction for Birth Weight at 300 gestational days: ( 127.34584318751573 , 131.36402218780412 )\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Suppose we want a prediction for a particular value of the predictor variable\n",
    "# We can bootstrap the prediction to find the 95% confidence interval on the prediction\n",
    "# Here we use the prediction of Birth Weight using the number of Gestational days\n",
    "\n",
    "num_repetitions = 5000\n",
    "x = 300 # The value of the predictor variable (gestational days)\n",
    "\n",
    "predictions = []\n",
    "for i in range(num_repetitions):\n",
    "    sample = baby.sample(frac = 1, replace=True)\n",
    "    coefficients = np.polyfit(sample['Gestational Days'], sample['Birth Weight'], 1)\n",
    "    y = coefficients[0]*x + coefficients[1]\n",
    "    # instead of collecting the slopes as we did earlier, we collect the predictions\n",
    "    predictions.append(y)\n",
    "\n",
    "# plot a histogram of the slopes\n",
    "#calculate the percentiles\n",
    "predictions = np.array(predictions)\n",
    "left = np.percentile(predictions, 2.5)\n",
    "right = np.percentile(predictions, 97.5)\n",
    "fig, ax = subplots(figsize = (5, 5))\n",
    "ax.hist(predictions, density=True, bins=50);\n",
    "ax.hlines(0, left, right, color = 'yellow', linewidth = 7)\n",
    "print('The 95% confidence interval on the prediction for Birth Weight at 300 gestational days:', '(', left, ',', right, ')')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5062b697-b856-447d-aaee-d4221f3bdf19",
   "metadata": {},
   "source": [
    "# Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "a4db7295-94b8-408c-b0b5-326040ccb8d6",
   "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>Blood Pressure</th>\n",
       "      <th>Specific Gravity</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Sugar</th>\n",
       "      <th>Red Blood Cells</th>\n",
       "      <th>Pus Cell</th>\n",
       "      <th>Pus Cell clumps</th>\n",
       "      <th>Bacteria</th>\n",
       "      <th>Blood Glucose Random</th>\n",
       "      <th>...</th>\n",
       "      <th>Packed Cell Volume</th>\n",
       "      <th>White Blood Cell Count</th>\n",
       "      <th>Red Blood Cell Count</th>\n",
       "      <th>Hypertension</th>\n",
       "      <th>Diabetes Mellitus</th>\n",
       "      <th>Coronary Artery Disease</th>\n",
       "      <th>Appetite</th>\n",
       "      <th>Pedal Edema</th>\n",
       "      <th>Anemia</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>48</td>\n",
       "      <td>70</td>\n",
       "      <td>1.005</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>normal</td>\n",
       "      <td>abnormal</td>\n",
       "      <td>present</td>\n",
       "      <td>notpresent</td>\n",
       "      <td>117</td>\n",
       "      <td>...</td>\n",
       "      <td>32</td>\n",
       "      <td>6700</td>\n",
       "      <td>3.9</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>poor</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>53</td>\n",
       "      <td>90</td>\n",
       "      <td>1.020</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>abnormal</td>\n",
       "      <td>abnormal</td>\n",
       "      <td>present</td>\n",
       "      <td>notpresent</td>\n",
       "      <td>70</td>\n",
       "      <td>...</td>\n",
       "      <td>29</td>\n",
       "      <td>12100</td>\n",
       "      <td>3.7</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>poor</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>63</td>\n",
       "      <td>70</td>\n",
       "      <td>1.010</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>abnormal</td>\n",
       "      <td>abnormal</td>\n",
       "      <td>present</td>\n",
       "      <td>notpresent</td>\n",
       "      <td>380</td>\n",
       "      <td>...</td>\n",
       "      <td>32</td>\n",
       "      <td>4500</td>\n",
       "      <td>3.8</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>poor</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Age  Blood Pressure  Specific Gravity  Albumin  Sugar Red Blood Cells  \\\n",
       "0   48              70             1.005        4      0          normal   \n",
       "1   53              90             1.020        2      0        abnormal   \n",
       "2   63              70             1.010        3      0        abnormal   \n",
       "\n",
       "   Pus Cell Pus Cell clumps    Bacteria  Blood Glucose Random  ...  \\\n",
       "0  abnormal         present  notpresent                   117  ...   \n",
       "1  abnormal         present  notpresent                    70  ...   \n",
       "2  abnormal         present  notpresent                   380  ...   \n",
       "\n",
       "   Packed Cell Volume  White Blood Cell Count  Red Blood Cell Count  \\\n",
       "0                  32                    6700                   3.9   \n",
       "1                  29                   12100                   3.7   \n",
       "2                  32                    4500                   3.8   \n",
       "\n",
       "   Hypertension  Diabetes Mellitus  Coronary Artery Disease  Appetite  \\\n",
       "0           yes                 no                       no      poor   \n",
       "1           yes                yes                       no      poor   \n",
       "2           yes                yes                       no      poor   \n",
       "\n",
       "   Pedal Edema Anemia Class  \n",
       "0          yes    yes     1  \n",
       "1           no    yes     1  \n",
       "2          yes     no     1  \n",
       "\n",
       "[3 rows x 25 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We build the nearest neighbor classifier, which is the simplest classification alorithm\n",
    "# We will work with the data on Chronic Kidney Disease (CKD)\n",
    "ckd = pd.read_csv('ckd.csv')\n",
    "ckd.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "8341e73c-c9a0-44bd-a15c-4d17ce2e74ba",
   "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>Glucose</th>\n",
       "      <th>Hemoglobin</th>\n",
       "      <th>WBCC</th>\n",
       "      <th>Class</th>\n",
       "      <th>Glucose_in_std</th>\n",
       "      <th>WBCC_in_std</th>\n",
       "      <th>Hemoglobin_in_std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>117</td>\n",
       "      <td>11.2</td>\n",
       "      <td>6700</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.220847</td>\n",
       "      <td>-0.567962</td>\n",
       "      <td>-0.863000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>70</td>\n",
       "      <td>9.5</td>\n",
       "      <td>12100</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.944594</td>\n",
       "      <td>1.158999</td>\n",
       "      <td>-1.452826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>380</td>\n",
       "      <td>10.8</td>\n",
       "      <td>4500</td>\n",
       "      <td>1</td>\n",
       "      <td>3.829056</td>\n",
       "      <td>-1.271539</td>\n",
       "      <td>-1.001783</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Glucose  Hemoglobin   WBCC  Class  Glucose_in_std  WBCC_in_std  \\\n",
       "0      117        11.2   6700      1       -0.220847    -0.567962   \n",
       "1       70         9.5  12100      1       -0.944594     1.158999   \n",
       "2      380        10.8   4500      1        3.829056    -1.271539   \n",
       "\n",
       "   Hemoglobin_in_std  \n",
       "0          -0.863000  \n",
       "1          -1.452826  \n",
       "2          -1.001783  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Rename the columns so that it becomes easier to handle later\n",
    "ckd = ckd.rename(columns = {'Blood Glucose Random': 'Glucose'})\n",
    "ckd = ckd.rename(columns = {'White Blood Cell Count': 'WBCC'})\n",
    "ckd_test = ckd[['Glucose', 'Hemoglobin', 'WBCC', 'Class']]\n",
    "\n",
    "# Add columns for readings in standard units\n",
    "ckd_test = ckd_test.assign(Glucose_in_std = (ckd_test['Glucose'] - ckd_test['Glucose'].mean())/ckd_test['Glucose'].std())\n",
    "ckd_test = ckd_test.assign(WBCC_in_std = (ckd_test['WBCC'] - ckd_test['WBCC'].mean())/ckd_test['WBCC'].std())\n",
    "ckd_test = ckd_test.assign(Hemoglobin_in_std = (ckd_test['Hemoglobin'] - ckd_test['Hemoglobin'].mean())/ckd_test['Hemoglobin'].std())\n",
    "\n",
    "ckd_test.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c7e8ff9a-6d85-419b-897f-342dc36de10f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAckAAAHACAYAAADJMJO5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8ekN5oAAAACXBIWXMAAA9hAAAPYQGoP6dpAACKYUlEQVR4nO2dB3hb5dn+76Pp7cSJsx1n773DDgmEGfYuUEoZLVD4uujm66T9aAv/AmW0tNCWDWXvkRAI2XvvvRPHe2i9/+t+juTIthQybMuynl8vNZZ0dM6rY3Pu82zLGGOgKIqiKEoDHA1fUhRFURSFqEgqiqIoShxUJBVFURQlDiqSiqIoihIHFUlFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocXAhhQiFQti1axeys7NhWVail6MoiqIkAPbQKSsrQ5cuXeBwHNlWTCmRpEAWFBQkehmKoihKC2D79u3o1q3bEbdJKZGkBRk5MTk5OYlejqIoipIASktLxWCKaMKRSCmRjLhYKZAqkoqiKKmNdRRhN03cURRFUZQ4qEgqiqIoShxUJBVFURQlDikVk1QURWkqgsEg/H5/opehhHG73XA6nThRVCQVRVFOkPLycuzYsUPq75SWk5TD8o6srKwT2o+KpKIoyglakBTIjIwM5Ofna6OSFgBvVvbv3y+/l759+56QRakiqSiKcgLQxcqLMgUyPT090ctRwvD3sWXLFvn9nIhIauKOoihKI6AWZOv8fahIKoqiKEoc1N2qKEqjYULlQGA1ECoGLA/g6gs4OquVpSQtKpKKojQKxrcUpvpNILgv/EoIsLIBz2ggfRosy5vgFaYOvCl57bXXcPHFFyd6KUmPulsVRTlhTGADTOWLQLAYcPa0LUhnP8BKB2pmwFS9m+gltir27NmDu+66C7169YLX65Vm3RdeeCE++eSTRC+t1aGWpKIoJwQzO03Nl4ApA5x9aMbYb/Bfq61tUfoWwHhPgeXMT/Rykx5mbJ588slo06YNHnjgAQwdOlQyOD/44APccccdWLNmTaKX2KpQS1JRlBPDMA65DnC0OyyQ0VAoTQkQ2JSI1bU6vv3tb4s7dd68ebjsssvQr18/DB48GN/97ncxZ86cmJ+59957ZTvWctL6/PnPf16nO9DSpUsxadIkGR3FCUmjR4/GggUL5L2tW7eKldq2bVtkZmbKsd5997BnYMWKFTj33HOlaL9jx464/vrrceDAgdr3X3nlFRFylse0a9cOU6ZMQUVFBZIFtSQVRTlB/IAJAlZa7Lct3otTPAPNvbBWR1FREd5//3389re/FcGqD63LWFD8nn76aXTp0gXLly/HLbfcIq/98Ic/lPevu+46jBw5Eo899pjUFC5ZskTauhFapz6fDzNnzpRjrlq1qraLTXFxMc4880x885vfxIMPPoiqqioR5CuvvBKffvopdu/ejWuuuQb/93//h0suuQRlZWX4/PPPk6ozkYqkoignBpNznHlAcD+ntTZ831TZlxpamsoJsWHDBhGYAQMGHNPnfvazn9X+3KNHD3z/+9/HCy+8UCuS27Ztww9+8IPa/bJLTYRt27aJxUprkNASjfDII4+IuP7ud7+rfe0f//iHxEjXrVsn7foCgQAuvfRSFBYWyvuR/SQL6m5VFOWEsCw3LM84wFTaj2hMCAjuAFyFgKtPopbYajheC+zFF1+UOGanTp3ECqRoUvwi0FVLa5Cu0N///vfYuHFj7Xvf+c538Jvf/EY+f99992HZsmV13LTTp0+XfUYeEaHlPoYPH47JkyeLMF5xxRX429/+hkOHDiGZUJFUFOXE8UwAPGOB4E4gsBUIFQHBPUBwPeDsCCv9YliWOq5OFFp4jEceS3LO7NmzxZ163nnn4e2338bixYvx05/+VFyoEf73f/8XK1euxPnnny9u0kGDBkkJCaF4btq0SWKNdNWOGTMGDz/8sLxHS5HxSrpnox/r16/HaaedJq7bjz76CO+9957sk5/r378/Nm/ejKTBpBAlJSW8DZN/FUVpXEKhahOqnmWCpQ+ZYPHPTbDkVyZY+aYJBfaY1kxVVZVZtWqV/NscnHPOOaZr166mvLy8wXuHDh2Sf3mde+211+TnP/7xj6ZXr151trv55ptNbm5u3GNcffXV5sILL4z53o9+9CMzdOhQ+fknP/mJ6d+/v/H7/Ue19kAgIGv/05/+ZBL5ezkWLVBLUlGURoHNAizvSbCyvgMr5yewsn8MR/qFsJwdE720VsWjjz4qk0fGjRuHV199Vay21atX4y9/+QsmTpwY0/qka5UxSLpAuV3ESiRMtrnzzjsxY8YMyWSdNWsW5s+fj4EDB8r799xzj5SX0PpbtGiRuFcj7zGph8lETM7hZ7h/bnvTTTfJGufOnSvxSmbKcg3//e9/ZTpH5PNJgUkh1JJUFCXZLUmya9cuc8cdd5jCwkLj8XjEOps2bZqZPn16A0uS/OAHPzDt2rUzWVlZ5qqrrjIPPvhgrSVZU1MjlmNBQYHsq0uXLubOO++s/T533nmn6d27t/F6vSY/P99cf/315sCBA7X7XrdunbnkkktMmzZtTHp6uhkwYIC55557TCgUkvMydepU+Rw/369fP/Pwww83yzlqLEvS4v8hRSgtLUVubi5KSkqkFkhRFOVEqa6uFiurZ8+eSEuLUwajtKjfy7FogbpbFUVRFCUOKpKKoiiKEgcVSUVRFEWJg4qkoiiKosRBRVJRFEVR4qAiqSiKoihxUJFUFEVRlDioSCqKoihKHFQkFUVRFCUOKpKKoigtgFAohD1b9mHz8q3yL583JV//+tdloghHY0Xz+uuvy+snis/nk2HLHJeVkZGB9u3by7itf/7zn/D7/bVruPjii+t87pVXXpEOOX/605/qrJMPDoLu2LEjzjrrLJlb2dTniOjsGkVRlASzdfUOzHp9Hrav2QlflQ+edA8KBnTFyRePQ+HAbk12XIrRH/7wB9x2221o27Zto+3X5/Nh6tSpMm/y17/+tYgj27/NmTMHf/zjH2VQ84gRIxp87u9//7s0TX/88celSXqEc845R8SVTdP37t2L999/H3fffbcI6ptvvgmXq+mkTC1JRVGUBAvka395FxsWbUZOu2x0699V/uVzvs73mwoOWeYg5vvvv/+I27366qsYPHgwvF4vevToUWvlxeOhhx7CzJkz8cknn4joURB79eqFa6+9ViaDcDJJfWh13nXXXTKtJFogCY/LdXbt2hWjRo3CT37yE7zxxhsyp/Lpp59GU6IiqSiKkiDoLqQFWXqgDIWDuiEzJwNOp0P+5XO+/uUb85rMrcihyBxlxWHIO3bEFuOFCxfiyiuvxNVXXy1Dlzmg+ec///kRxenZZ58VAabFWB+6TDMzM+u8du+994rFyaHQl1xyyVGt/cwzzxRXLsdvNSUqkoqiKAli37YD4mLNL2jXIA7I5/nd2mHb6p2yXVNBUaKld99998V8/89//jMmT54swtivXz+JEXL+5AMPPBB3n5xxOWDAgKM6Pq1BWpG0DHmcY4HH2LJlC5oSFUlFUZQEUVVWJTHItMzYI7bSMr3wVftlu6aEcclnnnlGhjfXZ/Xq1RJTjIbPKYSMEcbiWCYwDhs2TFy4FOny8vJjWjeP0xhJRkdCRVJRFCVBpGenS5JOdUV1zPerK2rgSXPLdk3JaaedJok2P/7xjxtlf/369cOaNWuOalvGGWfMmIGdO3dKgk5ZWdlRH4cCznmRTUnSiCT94JE04MjjaM15RVGUlkiH7u0li3X/9oMNrC8+37/jILoP7CrbNTUsBXnrrbcwe/bsOq8PHDgQs2bNqvMan1MIGdOMBRN0Pv74YyxevLjBeyz/qKioqPNaYWEhPvvsM+zZs+eohfLTTz+VGOlll12GpiRpRJIwu2r37t21jy+++CLRS1IURTluHA6HlHnktM/G1lU7UFFSiWAgKP/yeW5+Nk66aJxs19QMHToU1113Hf7yl7/Uef173/ueZKkysWbdunXiln3kkUfw/e9/P+6+7rnnHnHJMsb46KOPSinIpk2b8NJLL2HChAniqq1PQUGBWJT79u0Tq7a0tLT2vZqaGhFQWpuLFi2SZKOLLroIF1xwAW644QY0JUlVJ8laGKYBK4qitBZYB3nJd86rrZM8sKtIXKx9R/cUgWzKOsn6/OpXv8KLL75Y57VRo0aJuP3iF78QoezcubNsxwSeeLBk46OPPsKDDz6IJ554QgSVDQVolX7nO9/BkCFDYn6uW7duIpSTJk0Sofzggw/kddZF8rjUANZzMquVYn7jjTc2+Q2EZY4lwppgdyuzqXJzc6UAduLEiVLb071796PeB+9M+PmSkhIpbFUURTlRqqursXnzZomN8dp0vLDMg1msTNJhDJIu1uawIFPx91J6DFqQNJbk+PHjpS6nf//+4mr95S9/iVNPPRUrVqxAdnZ2zM/QROcjQrT5riiK0pKgIHbq0SHRy1CSVSTPPffcOinDFE0Ge+kGuPnmm2N+hpYmxVRRFEVRjoekteXbtGkj2VUbNmyIuw3TmWlORx7bt29v1jUqiqIoyU3SiiSLTjdu3CjB3CMFj+lvjn4oiqIoSqsTSWZHsY6GLYi+/PJLaaXEGp1rrrkm0UtTFEVRWilJE5Nk810K4sGDB5Gfn49TTjlFxq7wZ0VRFEVJaZHk+BRFURRFaU6Sxt2qKIqiKM2NiqSiKIqixEFFUlEURVHioCKpKIrSAjAmBBPYAeNfa/9rQs1yXDYOv+uuu9CrVy8pm2Oj8QsvvFCamhPOenzooYcQgZ1MWW3Akjr2WY1sE5nOlJ6eLs+vvPJKmdSR7CRN4o6iKEprxQQ2wFR/BAQ2AaYGsLyAqxeQdhYsV58mOy5L6jitg81Z2Bubk0A4yoqNxe+4444GMyE5ZPmWW27B22+/jenTp2P06NG177HpOd/z+Xyy3//85z+YMmWKNEX/6U9/imRFRVJRFCXRAlnxLyB0CHB0BhzpgKkC/KtggruBzBuaTCi//e1vi/U3b948ZGZm1hlL+I1vfKPOtuyDzTK8BQsW4PPPP5c+2tGwh3ZkShMHT3CQM5u9cHrI5Zdf3mD7ZEHdrYqiKIl0sdKCpEA6+wCOLMBy2v/yeegQTPXHTeJ6LSoqkhFUtBijBTICrcvoDmfnn38+Vq1aJQOXj1bw7r77bnHPvvHGG0hW1JJUFEVJFMFdtouVFqRl1X2Pzx2dgMBGeztX486VZN9rCtiAAQO+cttf//rXYimuXr36mBq45OXloUOHDuJ+TVbUklQURUkUpiIcg0yP/b6VARifvV1jH/oYRgmfffbZqKiowO9+97vjOg5dusmKiqSiKEqisDLtJB3GIGNhKgHLY2/XyPTt21fEq35yTiwmT54sLtPHH39cXKhHC9uI7t+/XwYfJysqkoqiKInC2cXOYg3tpslV9z0+D+0BXL3t7RoZukKnTp2KRx99VKzE+hQXFzewJt966y387W9/w3e+852jOsb/+3//T4ZJX3zxxUhWVCQVRVEShGU5YKWdBTjaAsENQKgMMEH7Xz535MFKmyLbNQUUSJZ1jBs3Dq+++irWr18vcce//OUvmDhxYoPtWdLB8o+nnnoKd955Z533ysrKpOaSc3tnzpyJW2+9Fb/5zW/w29/+Fn36NF0ZS1OjiTuKoigJRMo7Mm84XCcZ2me7WN2DbYFswjpJNhBYtGiRCNn3vvc97N69WxJzWP/42GOPxfzMmWeeiXfeeQcXXHCBxBsfeeQReZ2lHnx4PB4pBZkwYYI0JJg0aRKSGcscS/Q2ySktLUVubi5KSkp0ALOiKI1CdXU1Nm/eLHG3tLS0496PlHkwi5VJOoxBOrs0mQWZ6r+X0mPQArUkFUVRWgAiiI1c5qGcOCqSiqIkHYZxO6kf3MFngLMz4GK2pjvRS1NaGSqSiqIkFSZ4EKbqJcC/HoDPftFyAc6eQMYVsJogE1RJXdThrShK0mBMNUzVc4B/BeDsALj62Q9HFyCwHqbyWZhQaaKXqbQiVCQVRUke/KttC5JWI7vRRLDSAGdvILAF8C9P5AqVVoaKpKIoSYMJrLN/YIlEfdgYHF4YWpkJIIUKBZKCxvp9qEgqipI8sM8pKIZxYGxStmk+nE57PZyjqLQcIr+PyO/neNHEHUVRkgdOy8Bcu2VbrKbZrDF0Nm8ZhcvlQkZGhvQodbvd0oZNSSyhUEh+H/y98PdzIqhIKoqSNFieoTC+GXav0/pZrMEDUoRveYY375osS4YLs3B969atzXpsJT68WeHw5xOdQKIiqShK0mA5OwFpF8BUvS7ZrLDa2BZlqMR2tXrPAZy9mn1dbMXGqRrqcm058HfSGFa9iqSiKEmF5Z0IONrB+ObZQslmAu4RsLxjAdfghM0u5AX5RNrSKS0TFUlFUZIOy91PHoYDiQV3Ug/2VVouKpKKoiQtVqxSEEVpRDQNS1EURVHioCKpKIqiKHFQkVQURVGUOKhIKoqiKEocVCQVRVEUJQ4qkoqiKIoSBxVJRVEURYmDiqSiKIqixEFFUlEURVHioCKpKIqiKHFQkVQURVGUOKhIKoqiKEocVCQVRVEUJQ4qkoqiKIoSBxVJRVEURYmDiqSiKIqitDaR/P3vfy+TyO+5555EL0VRFEVppSSlSM6fPx9PPPEEhg0bluilKIqiKK2YpBPJ8vJyXHfddfjb3/6Gtm3bJno5iqIoSism6UTyjjvuwPnnn48pU6YkeimKoihKK8eFJOKFF17AokWLxN16NNTU1MgjQmlpaROuTlEURWltJI0luX37dtx999149tlnkZaWdlSfuf/++5Gbm1v7KCgoaPJ1KoqiKK0HyxhjkAS8/vrruOSSS+B0OmtfCwaDkuHqcDjEYox+L54lSaEsKSlBTk5Os65fURRFaRlQC2g4HY0WJI27dfLkyVi+fHmd12666SYMGDAA9957bwOBJF6vVx6KoiiKcjwkjUhmZ2djyJAhdV7LzMxEu3btGryuKIqiKCklkoqiNB0SdQluhvEtBgIbAcsBuAbB8oyE5eyc6OUpSsJIapGcMWNGopegKK1DIH2zYKreBkw5YGUDCAGBd2B8c4GMq2C5ByV6mYqSEJImu1VRlCYiuMUWSF4OXP0AWo7OroCzH2DKYKpegQkVJ3qVipIQVCQVJcURFystSGenum9YFuAoAIJ7Af+KRC1PURKKiqSipDrBzYCVGfs9i1njTpjgruZelaK0CFQkFSXloRCGjvA+S6n1UqGkJvqXryipjmswYCqZwdPwPeOnOQnL1SsRK1OUhKMiqSgpjuUZATg7AqFNgAkefsP4gOAmwNUTcA9M5BIVJWEkdQmIoignjuXMB9KvlixWBDeK5VjrYnX1gZVxDSwrPdHLVJSEoCKpKAosN0s/viNZrHaSjhOWqxBwD4BlHd1AAUVpjahIKooiWI4swDtB7EhFUWw0JqkoiqIocVCRVBRFUZQ4qEgqiqIoShxUJBVFURQlDiqSiqIoihIHFUlFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoShxUJBVFURQlDiqSiqIoihIHFUlFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoShxUJBVFURQlDiqSiqIoihIHFUlFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoShxUJBVFURQlDiqSiqIoihIHFUlFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoSrKL5GOPPYZhw4YhJydHHhMnTsR7772X6GUpiqIorZikEclu3brh97//PRYuXIgFCxbgzDPPxEUXXYSVK1cmemmKoihKK8UyxhgkKXl5eXjggQdw8803H9X2paWlyM3NRUlJiVijiqIoSupRegxa4EISEgwG8fLLL6OiokLcrvGoqamRR/SJURRFUZRW524ly5cvR1ZWFrxeL26//Xa89tprGDRoUNzt77//frlbiDwKCgqadb2KoihKcpNU7lafz4dt27aJifzKK6/g73//Oz777LO4QhnLkqRQqrtVURQldSk9BndrUolkfaZMmYLevXvjiSeeOKrtNSapKIqilB6DFiSVu7U+oVCojqWoKIqiKI1J0iTu/PjHP8a5556L7t27o6ysDM899xxmzJiBDz74INFLUxRFUVopSSOS+/btww033IDdu3eLmczGAhTIs846K9FLUxRFUVopSSOSTz31VKKXoCiKoqQYSR2TVBRFUZSmREVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoShxUJBVFURSlsUUyEAjg448/lpZwLO4nu3btQnl5+fHuUlEURVGSv05y69atOOecc6TZONvCsaA/Ozsbf/jDH+T5448/3vgrVRTluDDGDwTWwvjXAKYScOTDcg8BnN1gWVail6corU8k7777bowZMwZLly5Fu3btal+/5JJLcMsttzTm+hRFOQFMqAym8kUgsJxPwv/J+2BqPgO8U4C0ySqUitLYIvn555/jyy+/hMfjqfN6jx49sHPnzuPZpaIojQwH/JiqNwD/YsDZHbDSI28AoQNAzbuAsx3gGZnopSpK64pJcvpGMBhs8PqOHTvE7aooSgsgtBvwrwAcnQ4LJKHl6MynWsL4ZsOIhakoSqOJ5Nlnn42HHnqo9jndNUzYue+++3Deeecdzy4VRWlsAtsBUw5YceblWXnhbYqbe2WK0rrdrX/6058wdepUDBo0CNXV1bj22muxfv16tG/fHs8//3zjr1JRlOPgq+apayxSUZpEJLt16yZJOy+++KL8Syvy5ptvxnXXXYf09Ci3jqIoMTHGB5hqwPLCsrxNcxBnZ9vNKtZkjDBI6BDgKgCs3KY5vqK0AizD6H6KUFpaKrMoS0pKkJMTxwWlKE2ICZXC1MwB/AuAUAXg8ADukbA8E2FJnLCRE3cqngon7vQCLPfhN0MlkrxjZVwDyzuhUY+rKK1JC44rJvnMM8/gnXfeqX3+wx/+EG3atMFJJ50kNZSKojTEhIphKp8Bql8HQqWAlQaEfED1+yJmJrirUY/HXAEr/RLAPQAIbgECm4HgDiCw3rYivWcAnrGNekxFaW0cl0j+7ne/q3Wrzp49G4888gj+7//+T2KS//M//9PYa1SUVoGpng74VwPO3oCzC+DIBZwdAWdfILgdpuodsf4aE8vZDlbGN2BlfA1wDwKcXQHvybCyboGVfhEsy9mox1OU1sZxxSS3b9+OPn36yM+vv/46Lr/8ctx66604+eSTccYZZzT2GhWlVbhZxe1ptavr9iQUKkcX28Kjpcc4YSNiOTIB73hY3vGNul9FSQWOy5LMysrCwYMH5ecPP/xQ2tKRtLQ0VFVVNe4KFaU1ECq2E2gc8coxsuyWcaGi5l6ZoiiNbUlSFL/5zW9i5MiRWLduXW1t5MqVK6XrjqK0dMStGdoLmCqpI6RbskkR69EFsI9qzGzWANNRAatuFytFUZJQJB999FH87Gc/E7frq6++Wtu/deHChbjmmmsae42K0qiYwAY7PhjYCLAUw0qDcQ+CxT6mLJtoChwdAVeh7VJFVsP3Q/sAZwfApTeZitKS0BIQJaXgJAxT+R+7BILCxQxTU2FblZyKkXlTkwml8S+Hqfi3XeTvYA0jLcsgENoPoBJIuwyOtFOb5NiKohyfFhyXJUmKi4vx1FNPYfXq1fJ88ODB+MY3viEHVpSWiDEBmOoPAFMGOPvYPUwJXZwsqA+ug6mZIbWDTYJrCKyMK2Cq37dLMiIdcRxtAe80WN6Tm+a4iqI0ryW5YMECaUvHMpBx48bJa/Pnz5ekHSbyjBo1Ci0RtSRTG3Gzlv3VLruIbvgdgUkzxg8r53uwHHlNtw42EQistcWa63D1g+Vo02THUxSlmS1J1kJOmzYNf/vb3+By2bsIBAKSzHPPPfdg5syZx7NbRWlaKE5g4kyc1olWBmD2AiFmoTadSEpJhqdl3kgqitIIIklLMlogZUcul3Te4TBmRWmRMP7IDFJTEzvDlL1UQddrRiJWpyhKC+S46iRpnm7btq3B68x21XmSSovF1RNwdQWCuxu+JyUh+8T1CUcTl4MozYYJlcH45sFUfwRTMxMmuKfRuxoprZvjsiSvuuoqmfrxxz/+Ufq1klmzZuEHP/iBloAoLRaLCTres2CCz9t9TCWL1Rsu4t8FONrDSpskPU+V5Mf45sNUvwsED4RfCdmzNT0TgPTzYNXvfKQojSWSFEdeSG644QaJRRK3241vfetb+P3vf388u1SU5sE9HFaGBVPzMRDYEY5RpkkTcCvtHFhap9gqMP5VMJWv2MIoE1DoZjeAKQJqPoaxPLDSz030MpXWXidZWVmJjRs3ys+9e/dGRkbLjuVodqsSwbDzTXBrbccdOAtgWccVfVBaGPaIsL8D/uWAy+4xXYfgPmZPwcr+rmYVpyilTZ3dyh0Hg0Hk5eVh6NChta8XFRVJAo8KkNLSEVdbrAuokvywlCewCXDEmc/JmHNwk+1y94xs7tUpScZx3TpfffXVeOGFFxq8/tJLL8l7iqIoicNPczG+DVA7HswOFSlKo4vk3LlzMWnSpAavc0wW31OU1jLeytTMQqjyOYQqnpWfZeSV0rKhC5WzOk1J7PdZB8skLs1iVo6C43K31tTU1CbsROP3+3VUltJ6uvNUvgQEd9m1lZIt+aU9tDjjSljqqm2xWNKwfhxQ/bpd+yr1sWGkV+6u8ABqTdJSmsiSZCu6J598ssHrjz/+OEaPHn08u1SUFoMJFcNUvhhuet4bcIUf7Pca2gtT+QJM6FCil6kcAemD6x4JBLcDga12nJL1scH1diP79Is0UUtpOkvyN7/5DaZMmYKlS5di8uTJ8tonn3wi/VvZu1VRkhr/MtuCdPaVLMg6sSxHT/tCy228pydylcoRsBwZQOZ1gG8gjG8uwJsaRxbgPhWWZxwsZ/tEL1FpzSJ58sknY/bs2XjggQckWYeNzocNGyZTQfr27dv4q1SUZsT4N4Tb08WwNOQ1j2xjqUi2aCz26KVF6ZkIwMciWVi1STuKcnQc96isESNG4Nlnnz3ejytKCybES+wR3neEsyeVZMB2q0bFJRWlqUUyVt/WaLp37348u1WUltPj1b/Y7tBSv0Wd9N6otru4KIrS6jkukezRo8cR+1uy0YCiJCuWexhMzed2Rx5n4WGhpEAGt9k9Xj3DEr3MlMUwCSdUYk9ycXTSBByl5Ynk4sWLG5R+8LU///nP+O1vf9tYa1OUhGA584GMy+zen8G1gJVpv8FG6I52sDIug+XskOhlphwmuB+m5lM7aSrEdoIu2+r3ng6LJR2K0lJEcvjw4Q1e4xzJLl26SDLPpZde2hhrU5SEYbmHAln5ML6lQGA9L9EyRsvyDIfl7JTo5aUcJngQpvKZw+3mnF3suaCBNTCBbUDG1fK7UZQWk7gTi/79+0sZiKK0BiiGVjoFcWqil5LyiPubAillOeEMVXbNMVniFjfV7wPugfY4NEVpRBzH20E9+sGG52vWrMHPfvazJisBuf/++zF27FgZ6tyhQwdcfPHFWLt2bZMcS1GUloMJVQD+JYCVF9V3NQzjxbQq2UVHLH5FaQGWZJs2bRok7nA8TUFBQczG543BZ599hjvuuEOEki3xfvKTn+Dss8/GqlWrkJkZjhkpitL6MBXhkWa5sd8XizJk92RVlJYgktOnT6/z3OFwID8/H3369JFRWU3B+++/X+f5008/LRblwoULcdpppzXJMRVFaQGwKYC4UasBZDV8n7NBWdfqSE/E6pRWznEp2umnJ77TCF28hDMtj9SInY8IdA0ripJcWI5sGCZS1cwIu1zrRYmkx24Hu7euoiRKJN98882j3um0adPQlIRCIdxzzz3SHm/IkCFHjGP+8pe/bNK1KInD0L3mXw4T3Cw1jJarEHAP0WnzrRDLewpMYB0Q3AA4OgMWLUofENwLWAaWd7Ldr1VRGhnLMJh4FNClelQ7tKwmbybwrW99C++99x6++OILdOvW7ZgsScZNaYXm5OQ06RqVpsUEtsg0DgR3hlvI8REAJCP1Slju/oleYovFmIBd82mxl2l8F6VhHJCuTCsDFmsSEwxLPUz1e0Bggx2j5D2+qwss7yTAPeaIDU4UJRpqQW5u7lFpgetYrLeWwJ133om3334bM2fOPKJAEq/XKw+ldWFCZWGB5KSOXoczHmVW4FaYqhcBx7d10kM9DGcr+ubB+OYBwUNy3gwtb88EWK7DrSRNYAeMbzbgX8EngLMN4BkLeMaLqMp9dWi/nVDDRguO/GYRKFlj5q32+CvD0IkXcBVq2YfSpBzT7WF1dTU+/vhjXHDBBfL8xz/+cR1LjUk7v/rVr5CW1vjNhPkf5l133YXXXnsNM2bMQM+ePRv9GEqS4F8OBHfYsx6jSwJklFWPw6OsnGcmcpUtTiBNxfOAf5E9hNjKsa3Ems9g/CuBjGtguQfYw6YrngVC+8LxvzSAnW4qXwb8G2C8pwI1XwJ0fbKYn63hXP2AtMl1hLapEDFuhuMoynGJJDNK33nnnVqRfOSRRzB48GAZlUVYK9mpUyd897vfRWPD8o/nnnsOb7zxhtRK7tmzR16nyRw5vpIaGBaVw9mwZo6EJz6YwFpYUJGshdYjBdLZzc4WjWDaAaFNMFVvwvC9qjeA0EHA2S+quXsuQCu0Zhbgn2ePEXN0lBZ94vb0L4QJbQcyboDl6pGob6goiW8mwNFYt956a53XKFwsCeGDLelefvllNAWPPfaY+I/POOMMdO7cufbx4osvNsnxlCQeZUWhZN2cUhuDFBerWJD1bigphI4CILQTqPnEdmWKkNY7v/xsqBgIbAYcvQBHrl2WwX/ZBSe4D6b6I9sVqyipaklu2LABQ4cOrX1Ot2p0Qs+4cePE4msK9D8+JQKzWGXaPIWwfjkA/04YK3PpKKtaaO1JDDL7yMX4jPEyrktBbLCPMtua5CXD8te9dIjQdrITakK77Q44ipKKlmRxcXGdGOT+/ftlbFZ0ck/0+4rSJLBmztlRknTs+Y5hJKFkhz2pwzMikStsWTAzlQ8puo+BnEM+wuIYywo3PrvkQpJkYrm5M20RDZU18uIVJYlEktmkK1asiPv+smXLvjLjVFFOFMuRByv9crtNWZC1c9tsN6GMtfLCSr8UlrNzopfZYpAyD/dgwBTVvamIwNeZyOM9yXafMibZYBsKI8tBcsJCWf99to3zNHTnKkoqieR5552HX/ziF5LlWp+qqiop3D///PMbc32KEhPOD7Syvg2kXQw4ewDO7kDaBfKa5RmZ6OW1OFjmwWHRTNKxrcKI5X3QfnjGwsHaUs9EwBQDoQOHLUpuTyFloo7D01BoxcW9R8oxJJ6pKKnYTIDs3bsXI0aMgMfjkXrFfv36yeucxsFMVzYe5/Dljh07ItkLSBWltWH8a2Eke3VnWACNbRl6xsJKP1/qDY3x22OnaubYYin30Zy00VW6GUn5B8rtrjeg1VhlxyGRBSvzeikjUZSWzrFowTGJJNm8ebN0vPnoo8OZbKxdOuuss/DXv/4VvXq13IQJFUklmZD/vtgZh9m80vXG2Qj79AGBtUCoiMFdwNUHFvue1t8ueCA8eopxyDaAewAsywvjXw1T/aHMcBQLky5WZyGstLPEulcUpLpIRigqKpJsV8LpH0dqNN5SUJFUkgXjXw/jmxMu2g9JopLlGQd4RsOy3IldGzNgRSTDHXcoko0g4IqS1G3p6kNRZMmHoiiNhzEhuz8p3aLMFpWaxTSAvWoDG+0kJSYmJbCXqgiiltgoKULiuxYriiKY4C6YyhcBiiRbvjHmZ8rtpCR2sglVAL4vAVdfQJOTFKXlZbcqitI0MAZoKv5lt4+je9XRBXDk2cX9jCHSinTYzQCMb2Gil6soKYOKpKK0AOzJHNsAq224+N8R7mTDuYl0t24PJ/FkyXgw7UClKM2DiqSiJBiWXdjTOdgP1W03v4nG4jDh6nCRPwv6tWBfUZoLFUlFSTQUSRk7lWa7WKWFXLjgn0ivcQsIsai/HHCP1AHDitJMqEgqSqLhTEYH5zuW2zWJHEMVaSgutZJ8BAFzQLJdLc+oRK9YUVIGFUlFSTAsqZAaSIk5+gD3QMBZELYwi+wOOfTBukfAyrgOlrNdopesKCmDloAoSkvAPQZwrwb8y2xr0tkbsDgQebvdGSf9AljpFyW8kYCipBoqkorSArAcmUDG12BqZgL+BXaSDov2vafC4oNTPBRFaXZUJBWlhWA5smClnwfjPSPcXNwFOPI1SUdREoiKpKK0MCwHSz74UBQl0WjijqIoiqLEQS1JRZG2cAdhfIuBwDK7RtFZYJdauPrDYvcbRVFSEhVJJeUxga0wlc9Kuze7u40LYLNxdsHxngmknatCqSgpioqkktJwCLGpehkI7gacfe2eqRE4mLjmE3tclWd4IpepKEqC0NtjJbXxrwaC22VwcB2BJGwRhyCMb742FFeUFEUtSSW1Ce2zW75Zntjvs+k4RZSNxRFnm2RooB5YC+NfY3f1YVmJe4jd4k7LSxTliKhIKinOVzhTONuRDceT1OliQmX2IOfAcvu7yH/yPpiazwDvFCBtsgqlohwBFUkltXHRzeoBQhUAu95EI43FiwHPabBEKJMLuohN1RuAfzHg7H54xBa/V+gAUPMuwD6wnpGJXqqitFiS8/ZYURoLZ0/ANdDukWqqDr9Oqyu0A7ByYHnGIikJ7Qb8KwBHp7ozKGk5OvOlabrxzYYRC1NRlFgk3+2xojTyBA5kXA5TGQQCjNkFwgMcQ3bsLn0aLFfPJl2DJAWF9gKBdfZcSUeuCLflyD6xHQco/OWAo3Ps96288DbF9s911rMLCGywa0YdbcLryTqx9ShKEqIiqaQ8lqMtkHkzEFgP498gSTqWswPgHgxLMlybNqnGVL0L+OYApvSwc8fZHkg77wSt2K/KyLXilMS8AfgW2jMta9fTAUibBktLYZQUQ0VSUcSidAPuQbDcg5r1uKb6Y6DmY7vcxNHPdoUy2za0C6byFcDKPP41OTvbblYZ5hzDKg0dAlwFdgZvZD1V7wA1MwBHB9sClfUEpNGCJAA5MmG5+pzAN1aU5EJjkoqSIEyoGPDNtkXK0c4WJEIXsAxdroKpmXX8NZpM1nH1t12nLAOJJlRCfywszwTb5Syt+fYDvgWAoz1A67p2PS57X6YMpuZLrRltpZhQpd19KrBNPAqKjVqSipIoAltsa87ZK/b7jnwgsMnu/MMs1GNESjvSL4FBFeBfZ//nTotZEpTcAEdyRbtzg5sBUwI4YliK3BeFXOKmcSxTJSkxphqm+jPAP8/+e6Qbnp4E78mAZ2LtTVSqoiKpKAkjbN3FvQjxP8/g4e2OA4vimvENyXI1fjZvrwScHWG5hzds3i7WptWw81D0ephYdALrUVoWEoOufAnwzZVMbsmEZiw7dNB+PVQEpF2Y0rW0KpKKkijExeqNXaNJxKqjK7bNCR3G4r6942F5x3/1ekQIq+qWjESvhwk8akW2rraM/oWAo1vdv0H2K6ZVWfMFIDdUhUhVNCapKInC2QNw9QJCO+1knWhMtZ1d6h4Hy0prnvUwIYcXw+COcHee6PVUinhaHq7H3TzrUZocw0YTJNZNmtUGMBUw/pVIZdSSVJQEIa7O9ItgQqVAcL2dwCOWJUsvfIB7JCzGhZptPS47hln576j1eKLWMwbwTGi29SjNQOgggBheAyIuVrddR5vCqEgqSgKxnF2BzG/C+ObZ7eNoQbq6isUGz2hYsdyeTbkeWpKZt4TXs8SOU7q62/Wash5vs65HaWKk/IcN/OPhT3n3uoqkoiQYy9keVvp5MGlTa6eNJHLIs8XEnvQLYdLOlTIRwJvSiRutGTaHMP6ldkJW/RsgejisNFjugUhlNCapKC0Eptoz/phIgazvfrXXowLZanEPBTg2LchypIN2LJrx8eBeu/evZ2z8EqUUQS1JRVGUFEWSwjKuhanOA3xLgNCGcJ1kHuA9D1balBZz05YoVCQVRVFSGDbStzKugPGeCYT22CLJgdyOnEQvrUWgIqkoiqLYjSeOo7NTaye17WhFURRFaS0iOXPmTFx44YXo0qWLJBO8/vrriV6SoiiK0opJKpGsqKjA8OHD8eijjyZ6KYqiKEoKkFQxyXPPPVceiqIoitIcJJVIHis1NTXyiFBaysnviqIoitIK3a3Hyv3334/c3NzaR0FBQaKXpCiKoiQRrVokf/zjH6OkpKT2sX37kXoUKoqiKEoKuVu9Xq88FEVRFOV4aNWWpKIoiqKkjCVZXl6ODRvYW9Bm8+bNWLJkCfLy8tC9e/eErk1RFEVpfSSVSC5YsACTJk2qff7d735X/r3xxhvx9NNPJ3BliqIoSmskqUTyjDPOgDEm0ctQFEVRUgSNSSqKoihKHFQkFUVRFCUOKpKKoiiKEgcVSUVRFEWJg4qkoiiKorSG7NZEYkLlgH8pjH8xwJ8d+bA8owD3YFiWJ9HLUxRFUZoAFcmjwISKYCr+AwTWAkgDLC8Q2gPjXwZ4xgIZV8Dia4qiKEqrQkXyK2Bdpql6yxZIZy/Ach9+M1QB+GbDODvDSpucyGUqR8CYABDcAoRKACsNcPWCZaUnelmKoiQBKpJfRWgX4F8NODrVFUjiyARMFuCbB+M9Ra3JFogJbICpegcIbAXgBywH4OgAeCcBnomwLCvRS1QUpQWjIvlVBPcBpgJwdI79vqMtEDoIhIoAZ5xtlIRgAltgKv4NhA4Bzq4ArUfjB0J7Yapegcij96REL1NRlBaMZrce9SmK0w7PhMLbqEXS4tzkNTPtGxhnb1sgCb0Bzm5yf2hqPoUxVYleqqIoLRgVya/CVQA4cm1LMRahA7aV4shv7pUpR8IcAvxr7d9LLJeqo6PtJQhsTMTqFEVJElQkvwLLkQe4xwCGLtXSw2+w0TovspaB5WVsy5nIZSr1MTUAfHYmciwkvmwAtSQVRTkCGpM8Cqz0c2BQBfgWAoHdACiIAcBqA3jPs0VUaVlY2YCVYde0OjMavm+q7T9/KzcRq1MUJUlQkTwKLJYNpF8FeMbB0IVnKsIW5iDJetUMyZaH5ciCcY8Aaj4CTB5guep6AUI7AVehlIMoiqLEQ0XyKLFYOuDqDcvVO9FLUY4SK+10mOAmILABsNoBjizbDRvaZ2cle88CTCUM0pKma5JhopiptJ9YmXqDpihNjIqk0moRaz/zJpjqGdJSUJKswOzWAYDDC1T/Fybkk3pX4x4LyzseliMHLVYc/UtgfHOB4A47m9rZA/BOAFxsjahiqShNgYqk0uqF0sq4FCZEq7EUJngIYAclNohgTJmudCZkVb8GE1wDZNwAi9nMLa2cpfo9oOZjO9nIamv/618GE1gDpF0AK+2MRC9TUVolmt2qpASWIxtwdAF8X9rxSGcfwNnRLu9xdrFrKf2rYaqno8VBd3HNdDvJiK0R6Sqmlezqw24IQPWHMMFdiV6lorRKVCSV1CG4HQist7sn1S/ZYUmIlQf4F8OEytCSMP4ldjYuhbE+bLFnSmB8KxKxNEVp9ahIKkeEHWlMqNRuEp7ssD0dxcbKiv0+45Gm3N6uJUErkeUssZBYpEem0iiK0vhoTFKJiQlsh/HNAfwr+ARwtrHHgnnGJ+8EDbEWeV9Iwa/XrJ6wr6vUTrawTNdIz9m4sGY3rRkXpCipg1qSSgOMfz1MxVNAzQy7Ny0vwMH9MJUvw1Q8m7z9TpkNyjZ1LAGJhdln107ShdmCsNxD7Akmsax5sYwdsFz9E7E0RWn1qEgqdTDGB1P9ht371NkPcOaHk1u6Ac5CidmhZg6SEcuRAXhOBVADBPcAJmi/IfMmt8tAbct7ul0T25JwD7OTdIIb7RmmtQ0Ryuw5ma4BgHtgolepKK0SdbcqdWFiCwXD0a1hY3BalFYmjG8eIPMzY7gsWziW91QYuidpJbPRQARnR1hp58ByD0ZL7B6EjGthql61M10DO8NvpAPu0bAyLkuaZgiKkmyoSCoxkltCR2gMngOESuwEF6nXSy5oJVppk2E8o4HAOrvBOfu8uvrD4hDtForl7ABk3goENwPB3eFmArZ1r40EFKXpUJFU6hG2DkUoY7kdfeHyieSzIqOxHExEGodkQibN0O0q9ZGKojQHLSz4oiQcXoBZChFrfqbEwYqksbu4ABVFUVo5KpJKHSxnO8AzwU7cYa9TWpTE+IDQZilotzwTE71MRVGUZkHdrUoDrLSp7AxqZ7GG1ofvpRgD6wor/UJYrh6JXqKiKEqzoCKpNICZkhRDQ4uS2ZQSh2wDuAfAipfQoyiK0gpRkVTiYrFGkg9FUZQURWOSiqIoihIHFUlFURRFiYOKpKIoiqLEQWOSiqIoLYRgMIiqsmo43U44XU6sW7ARq75ci0N7i5HdNgsDJ/bDgHF9kJ51dJN4yosrsHrueqyesw5VZVVo360dBp/UH31H9ZL9H2kdG5dswcov12LftgNIy/BiwPi+GDihL3LyspFKWMawQjw1KC0tRW5uLkpKSpCTk5Po5SiKogi+Gj+WfbYKyz5bieJ9JVKCVVFSKYKZlZuBtKw01FT64Pf50XtED0z79tSvFKuDuw/hjUffx7bVO5CW7oU7zY3K0ipYTgsjzxyCqV+fBJe7oZ0UDATxybMzMf+DpfJzZk4G/DUBVFdUo2vfTrjojnPQoXt+ymiBWpKKoihxRGbv1v3SaapDYT7ad8n76u237JPBAB0L89Guc9ujFsh3//axCKQnzYOstlnYvmYHtqzcgdz22ehQ0B7tutj78tf4sWHRJnz20mxcePvZcfcZCoXw8b8/w7ZVO9B9YDe43IetRorvwg+XoWNhB4w5e3iDz66YtQbz3l2Mtp3aiPUaIRgMyf4+eHoGrv3ppXA641uirQkVSUVRlCgqy6ow/YUvsGbuBlQUV8LAIDM3Q9yck64+GZm5dRvhV5RWYvrzX2Dt/A0oL66U17LacPu+OIPb52Qc8XgrZ63BspmrRLTSs9Lg9wdQUVqFNh1yEAqGsGn5VuS0z4bb44Lb60a7LnnihqUoxxPiXRv3YvOKbejYI7+OQBJ+l5IDZVg6fQVGTBpcx5qkuFKsHU5HHYEkTqcDnXt1xPY1O7Ft9U70HNIdqYAm7iiKooShO/OdJz8SS8rjdaNwcDf0GFwAb7oX8z9Ygref+Egsvwj8ma/Nf38JPGle2ZYPWoR8rf729WG0a9nM1XC7XSKQpLq8GjWVNUjLTENWbiYqSytxaE9x7Wdy2mVLrPHAzhj9lcMc3Fkk7tmM7NixS1qoRXuLUVZUXud1xi33bz8ox4hFWqYXAX/wiMdubaglqSitGMPeu4aDmi2ZBapjtWx81T74qv3wZnjg9hyeaLNp6VasmbcBXft2lmSVCLQMaV2tmb8BGxZvxqAJ/eR1JrfQquvWtzO8GV4R2f07DuLAjoOoLKvG7k17kZOXhSnXn1bnONHrOLTnEDLb2NZpKGTg8/lFPPlwhJNraqpqaj/Dbfh7pGUXD8tx5N8zLVTug9+p7ucc8lkeIxaG6woZOL5i/60JFUlFaa3i6F8M45sLBHeGe+/2BLwTANeglBVLuiiXTF+BVbPXikjS0hpyykBJZMlqk4n1izaJEEQEkq7X3Zv3Yc+mvQj4A/KZ9//xKQoHdhW36/pFG2V7CmR1ZTXWzNsoiTdOl0OyR0uLyvHmYx+guqIGo84eJnG8th1yal223MblcaGqvAa7Nu6RGGhVeTVKD5Sh7FAF2ndua4ul47CY0aqkK7Zzrw61r3FttAAZN8zr1AZd+nQSYS89WIbc9g0TUw7tK0HvEYXIzqvrUqU1yxgms1rb5Df8HN3J6dlpsv9EUHKgVKxfT5pbMnWjz0tToSKpKK1QIE31u0DNJ/YL7LvLfEn/UpjAGljp0wDvqUg19m3bj9cefg+7NuwR4WDGJ2OOn/xnJjYv34qL7zpXsj8jFh9dmktmrEDRrkMiPkzICfoCmPPWAhHX6352GSpLqsQty7jl2gWbJHGHVmFa2EJldmjQH8Sbj72PL16bg7xOFD2goH8XjJoyDP3G9EKfET3x3//3jrgxXS4nPOkeZOSko2hPMXZW+tCmfTZy83NELLmmskPlOOOqk0Ro7RjiKiz8cKlYsHzOWOLgkwegY48OWPTxcolbtu+WJ8JP0dy79QCCgYCUdNQXGd48jZg0RCxkCnZ+Qftaq5HizXM4cvJQdOpxWKCbA5bAfPnmfKydv1Hc0byxoLU//rxR6DOyZ5MeO+lKQB599FE88MAD2LNnD4YPH46HH34Y48Yd3fBcLQFRUgHjXwtT8SRg5QCOeokdwT1iVVrZd8FyJsYaSAS8zL30xzewes569BhUUMfNyJjh+oUb0XNIIaora6SmsHPvjti6ajsO7iqGy+MUIaSgVJZTRF0ikhd/5zz5d9br80RA6IblvQj3TUuSQsf98TW31yVxSlphTJqhRUo3LBNnKKqfPveF7J/WG61Lrpeid2hviQju4JP7w7IcSMvyYthpgzDl+tNlTTz29BdnibgyG5UW5dZVO7Br/W44XA64PG4pI0EoZE/2sQCXy4XsdllicfYa1gMnTRuD3iN7yv4i54rW9owXv0TJ/lL5LrxJYNJQvzG9cd43J39lMlJjQqv3vw+9LclCFPyMXJak+CXuyhjp+bedjYHj+x7TPlttCciLL76I7373u3j88ccxfvx4PPTQQ5g6dSrWrl2LDh2a985GaV0YUw34V8ME1tmzMx2dYXmGwnJ2bKT9VwH+VTCcqmJ8sJxdATf33/j1Zsa/BDA1gDNG5qOjIxBcC+NbDis9dURyz5Z92LJiOzp2z68jkH5fAJuWbcWeLfuxZeU2ESK6OXdt2iMXYlpWLk+6uElZYM9MULr5aJF9/upc3PTrq6TYnmJGEzE9J523IGIV8rWAL4AOhe1FILldbiBbrEkr7DpkmUfx/lIRR4fTifJD5eF4JJCZk45OPfJRXelDwYCu6Duqt9RIdundUQT1wM6DmPP2QhGsvM5tJKN168rtIq41VT4R57xOueKW3bN5H6pKq5DXpY1Y0WVFFdiycjs2Lt2CpZ+txLDTB+Gcm86U2Cq/88gzh0oC0rqFm1C8t0RqLHsOKRBX7JGaEDQFCz9aKgLJ9USOTUHn9965fjdmvjIbvYcXyjluCpJKJP/85z/jlltuwU033STPKZbvvPMO/vGPf+BHP/pRopenJCkmeACm6nnAz9mZISBUDoT2w8CCcQ+RsWFwD4flyDrO/e+DqXweCGy03Z5wwmAOUDMdSL8IlmdM436h4C5J0omJxCI9QGgfUonNy7Zh25qddrmDMWjTIVfEjok1jDfSWqqu8CGvc1tJ5jmwowihQAhOjws1FTXiMqWLL6ddlsTtTDAkIkVh4j5pyVWHDEJB/n4NQoFgbfINL96MozFZhjFAbkvojvVX++GrqoHTYWHMuSNRdrBcivYp5MxAzcjOEPEeNKE/Jk4bIwJItyMTjDYt3YKdG/ai/7je2Ltlvzynhcrvl9M2SyxkiiXdyxS+bv27YO/WffBV+ZHXsQ1y22VLzJXfj4L51mMf4Op7L0bbjnTPQ/4df96ohP7eqiqqpeNQxMKuT4fu7bFr015Zf7/RvVNbJH0+HxYuXIgf//jHta/xbmrKlCmYPXt2zM/U1NTII9rEVpRojAnAVL0M+NcAzu5AYL0tIBKFCAC+eTDB3YBnNJD5NViOvGPcvx+m6kWAFioTZ6zw3S73H9oJU/lfwNEOlqsR4ypWGs3JI2wQBFJoLijbsn30789wcNchidcxK3TH2l3YsW6XdJKh8B3cXSyuSCbv0DVKcSnaVQxDN6XTKQJHYaWw0FK0nE7J8qSA5Re0Q8fC9lLK4avyycWclhcfjB9WSuecKumaE53hStFldiuFkLHGikMVDeoeKbI8NpN6nvrpc1jw/hKUFVcgm9mwliWW5+rZ68RipTtUslYdljzEYja2O5k/0xXLshC2tOO2hIJPlyq7+lBombBzyiXj0VKooohX+pDVNvZNn3znUEgaJCDV6yQPHDgg7o6OHeu6v/ic8clY3H///eJ3jjwKCgqaabVK0kDrzk8B627H64I7ACsLcOYBzg5hMXECgbUwVW/JRevY9r/OPoaz8LBARiw6R1fAlML4FjTqV7Lcw3iLCJhAwzfp9rUcsNz9kQow4ePDp6dLQX37rnmShMJieibCUNDoNqVrlBfjDBEPl4hgm/xc6Z/KG3EKK7Mp6eKLZAVT+LLzMkX0vOkeFA4qkL6m7JbD7jidCjvIcbgvxiX5Z0MrMjqrmNYprUxaQ7QA2USgPqVFdpYrE4jmv7tYxICxTLqJy4rK7BijxyUuW5Zt2KUd9mWdPzMuSdEOBIKSpESi85plPRYQChiJr66dxyHr8fH7/OKeZlceumqja0ApVtvX7hShXbtgI6rKq47691S8vwSr5qyTrON92w/U/nfGWlG5aan0xfwcf3c8x9yuqUgaS/J4oNXJGGa0JalCqdQhuN22GClgUirhococfp8iaWhlFErMEqHdgLPLMex/h23V0bqrj1ygcoHAGrkoNFpZBkXSNTcszgWAlWFbroZu5J3iOoYrNUSS2ZCMDfYc2l1qDjcu2SwdcigInnQvQgdKJTFErEiKWFhCmGDjSXfDV+lDKEjXKcRypPCUl1aIVTZ80hAphdi+dpd8hh1o6KJl7LH4QKmd2cqsWNgCyYSfCNwPBaewe4Eci4+inUVo0z5XRIHizTgq6y25jzYdc+VvhOJNwaagl4RLWtIy0uRvqbKsUkQ3VFkjgsn1shlBeUmFWKwUMVqWrnrr4Jfnd+Xfn7hr47Bm3np8/t+5kvVKgedNRMfu7XHyJePFNTz9hVnSjcdXE5A/bbp0x54zAmPOGRG3hR1vID576UsR1khjAyY8MWN18rWnSlOD/mP7YO7bC6XspX42Ll3evCnpMbgbkOoi2b59eznRe/furfM6n3fqFDsBwev1ykNR4mPHkMTCEiurfoeSsHBZ2UDoABDce2wiyRhnnXv3+li8+obX0TgiaTmygYzrYKpesYVSXK/G/m7usbAyLoUVbdW2YuimpEuOAtClV0cRBSZ70PJiLDDIGGLQiJszujjfYTnEYitDuQgHRZIixFpI/rr6juqJC249S9y1iz9ZLvWQ3L7v6F7SCo7CTKGjFdemQzaC/pBktbKeMlDtR3lJpYhh8YESiZV2H9AVnXt2wI51OyXGVn6oovamKRAIYOvKHWL50h0cgRmqFBaui1Ua4pZsY7/GMgmKDR+8KaCos3MPvXGSPSsRdyt8w5Am8cddG3aj1/DCmOdx/aJNdvegKp9YyRRqnpf92w/g1T+/JU0IeG7Zs5YuXN4g0L1NNzdF/qRpYxvsk2v54J+fYvEnKyQzl9Y4vzK/+5LpK8Vqvux/LsDos4dj8/JtkrXLPrb8TjzvFEjeDEycNvaop6K0apH0eDwYPXo0PvnkE1x88cXyGu+M+PzOO+9M9PKUZMXZGbCcdjaoUM+dytcdncRFab93jBEKBwWVsSFfXXdr7f5LAPeEWhdZYyFZuZm3A4FNtvXLNdCqdHZvFY0ETHAnjG+RWOHye3H2geUZBcvVo852Igi0oMKuxYJ+XdChoJ1Yj7zgsySEIlEwsJtciCliTKyhANDS7De2E3Zv2Cv3LxSS9Ox0ybJkD1M2+vZX2zFINiTn+4xbUpRp0TGL9OwbTkdm20yprVzxxRrs3rhXxJEiwuxZxiwpODWV2ZL0k5HLGss0dO3dWdywG5ZslpghBY6lHLn52fB4PbVCLolAHidKDpSj9KAt/CyLoKVHN/C+7QfFLUthj7gsaQky5kqLkm7owsEF4m52ul0YcvKAmGI25+0FIrzMbo3AdbNWcd77i6Re9JTLJtQmJfGcMJOW9ZbP/vZVzHjhCxE2HpMJU6zRzGmbiZWz1krNJYUvAr8TvwMTkdbN34jhZwzGJXefJ03dWZZDdyzXnd+tHcZfMFpKYpqSpBFJQtfpjTfeiDFjxkhtJEtAKioqarNdFeWYcfUDnD0A/wY7FmlKo5JrKsPi0gUIFdk1h65jdNe7BwAuJgRtBZy9bEGuTdzhxIg0WEwKagIsHsvN+rFjqyFr6Rj/MpjKl+3fCX9nVLDApzC++Xa2MLsKhSkc1E1q/ihKkexI9mGlNcSbbFp3FDQKpsfjElcp7yHoqrSY4LNul5Qa9B/bF0NO6S8X5rnvLhLrkYJJUWCNJAWObeik+59lIb9bHkacOUQK83ncUZOHSgIR+7lScMQVmpeJdp3aisuWjQHWzF+P/VsPyOfoVq2p9olVyKYCTMRhpufuTfukBCSSAEQXLEWHRf/sxOPN9CK7bTb8NT6p26RVxngdLTwKD2N/B3YewsFdRVIOQ5FjNyCK5oTzR0uJSX32bT2Anev3yHHqw9iovzogMU8KeWR8F2OVbNdHS7PkYJlYlXYrPQvZ7bMlcYr9aWnPduvXRYSYjRz4fXheeaPhdLnknFEkO/fsiKt+eJG4oHlO+D7XHqntbEqSSiSvuuoq7N+/H7/4xS8kWWfEiBF4//33GyTzKMrRIm7H9CtgzHOA/xAQqgCCtPp4QXWLhQIwLrkH8Jx1zNmtFmOa6VfCVD4HBJkUkRa2XCvtC3za+SkTH2wMTPCgnREcqgSc/cIlLWGCu2Cq3gScXWGFb2bo/qSoMFbGEojIxAuK5vZ1uyRW2XtYIWa9MV9EifFHlkjQHZndNhNd+3aVEVksy2CDc2aRUvS69OmI8kOVIrR8nxYkM03PuOpk9B7eA2075taZrkF34KjJw6Tkg2uhJcoLfXR/2MzsDGwtrZKyDLosKRAUOQpImscr1hOfs6aTokf3cKRlHdc0+frTEKgJiDuZws14ZLf+XcWVy5ght6OoUSA3Ldsin6HVRjf00NMGyZSTWG3eaqo4xzIglml9WOoiNwbhRKQItKwZT+X3o8h727jRvms7EdOyg2VwOC1pcVdRWoFta7KkpR6/N6E1z+9Hi7uihH2HbSiyFEs+mpOkEklC16q6V5XGxHJ1A7Jug/EtBXiRZUYqLUgRRLqoSgDvKbDSzznO/XcHsm4HaAH5V9quV7o9PSPEik1296dkIoZ2AeFGCbbFPRCWI/OrP8ua1AATopgc5bEte0en+OcksFxqWOHsW1cgiaMzEFwnzRQiIkkr8ILbz8Y7T34s4iHNwcOx304981EwsCsWfbJcXJzjzxuNyvJKLJ2xCkFfUAYL9x3Zs7ZcglmdjM1169sFSzfvkyJ/unrtuKOdcMMkHlpk8di8crusqf4YqohwM3bIpgcSN91XinZd28LabYm1R/Fwuuzhx3yfJSZsOEDhZG3n0ukrRcTpxuW5YckJY6dM8onA5CG6N2m5UoCu++llYh0f6W8wq02GWKG09Or3eqX7NJIhS/drpBk7RY8WLGOvtJozwh16aHnT2t2/owhtu7SR+Ouq2eskK1gmlliWWJhsE0gLn+tPNEknkorSFFiOXFhpp8Gwp2loj90dR9x5mXa5hLPnCcUNLUcbwHsaLO9pjbpuE9wj3XNkzZYXFq1S98AmS8yRxunBTTC+VXbWLzLtYwe3hqeNhM8RY6Lp08LlKHH25VsMU/02EIw0NmB/1GzAMw5IvyDmdzABZiO7wzHiesjFOgMIbLH3Fgph+czV0rqNRfTBQAjtC9pJd5bCgQXSzea5+18T1yXjfyQS76IlSPcl3a9068lXcjsl9sbBw0w0YeYlDxlJInF7PdiyfJsIXbQVGWv6RqxSlR20ACuqpWEArTLWNtKyY6yTblK6GV3h9naMhfYZ3gN5XdqKKzO/a57dW3XrflSUVEmsk9afr8aHwoHdxIpmDDMCS2HowqVoRq+HGbeMzfKGgPvL69wW/Ub3Qq+hhVj++WpkchpKlLXJhB0KJeOk7rClSbdzTbVfLPGKbRUinhTZCLQu+V2cDoeIOhN7pO4zjCsnQ9bEWGxm7vE18GhMVCQVJQq5YDCZx9m5kXJNm9B6882CqXoPMIdslzACMDVf2HHQjGtg1e/besLH9MFUvQH45oQTndy2IJmDdh2oeyTNZrs+M7gDpuJFICsTlqthJxTjXwdT+VLYqmaTBbsTjghvzacwlstuxF4fcYOHjrBICq1LMkKf+tGzthu1skbiW3Rb8uLPhuUTLxwrbk9eiCkiEWgtyWQOl0su/kxyYczQLo/wiQDyZiR6TiNFwOXJxb7t+2VG45GssoL+XSUZJbrkh3WQrCukgFK07BINI2tgkwFCQea66Kal65Sx1HNuPhNfvjEfXXt3EkHjkGVWADBTl2LGeCpFffOK7fLZrn1ssSfM7qUlHT03kpbl209+LH1seUNBoffX+DH/3UUYdFJ/dCjMFyuXPV9pJbJ8o2j3ISnXYGxw++qdkqzE88zzxDVKXWpOXWGlZSkWY1WNJB1Z4XrQdNY6hi1JimduhxxxYScaFUlFSUYCq23Bkrhp/8OuR4qXfyVM5atA5jcaNWvWVH8G1HxmZ/s6c+z2fdhkN1KnCzS4DXAxOclli2ZwPUzNnAYiKe3aambZdZuM+UbWLpZgW5rHdqcjurjrxYAtVx+Ymtm2EPM4dXZM8ayG5RqAtx77EJ+9PFtie6yji1hRFJO18zfg7z96FqdeaneWie7lSmvRbi1nt41jlqstGE6pgeQFni3s6sPdi9c5QJGG3T+1skaal7ftkFu73cDxfbDwwyXSDo+CR6Gk9cfMUcYY23XJk/NzcFe1/O6435IDJUxllazankMLw27ZoFi5zHiNZMESNjCIWGsUHX5fCjrb5zHOR+GjFUcRiyQVRX4nH/17prSAi56laYyRmCJLMk6/cqKMAWPtadHeQ2LRjpk6HGPPGSnZqfPfXywZvNXF1XK+0tLTpGRj/44DdW4KaC0zo5jnkmJLy57fgyUfkriTlYZOg+0m9Oxvm2hUJBUlybBFZq4tiPWzbZkoxE4+LI0IbrFFqzGOyYQm/xzbHeoIT00Qd2sNYOXZrlY2TpDmBe5wR6F2dqOEUHndvrfMIGb80tG+YVyRUBiZ5MQaT0+9RCnXYMBVaJe2iAUazm6kaIaYQdwZPgzG568+KhdlZolGwwswMzDXzF0nHXJYPcI4ZWQcFK0kjruKWEOMA0ZElBYee7lSDJmAEil3iIywYlyQHW6ev/81qc9kEg3jb/1G9ZK+q0xc4WPqTWdKfaBdcuLB1tU7xM1JK6zvqF4S2+ToK1p7tmXpRqfCfBFIrnPfriLpHsR2drTmtq/dLRYd43q1fwaWJfujhcbvcWBXkViaEUt1/Hkj65ROUHDXLdwoghudTGRZllimjLWyTpF1i2y2XrKvVMpRBozvV5theuY1p2LCBWPE+lv62Spxc9st98qljpSt5Zi4RCuxa99O2Lt5v4glR4XR9Vtdzs5EFE7GXp1yzFix2+ZGRVJRkg02PQhuajgGKwIFKbDL7ibUSCIptZZBulWjOpvQ4iMidOm2+Bm2SosIG12ozFg8nPVofy4Qfi3O5UesPtZhNGzTJslAGdfAVL5gl9VEu16Z1ZpxJXauqpQYYf0kkwgsvWC9ImsD2VqO8cBIz1RO02AMkGOY6P7k3EcKEzMvadWwiwwfpYfKxOKUcpGw9cOLO8WK7dryu7aTuZCcvLHgw2WSoEKBYYYoLcx+Y/pIlmp1WaVYYTntO8gIr4gbl9mmq+dyIg2zRu3+slRnJrywvIIuTpahULgoeqUHSmVwM7ezY6WWrIuuYlqQGxZtlu/F+OLACf3sDkRRLlBatrTmWBYSizb5OTJC7O/3/kdqTCMdd+a+s0g67gya0E+24/r5OPOaU8SNOuedhdJIgTWcPDf8/oxzwljoN7aPnHvJxG2TWadWkq5tWsxca6JRkVSUpCPS8OArOvnUb4xwQocMdyaKPiaTZNjXVsTMEbtRAmdW1p9Iwo5AFPjQQZaOx/gchdVtW6IxkDFjWd8OjzajUBo7Q9k1SCzWAK1MCXsdOapMd+TYc0fK3ERaN7x4003IkhFacU5jJKuUFhjjjiPPHIK92w6Im7D7oK52v9RgSOr6mGXKzjEUWYodj01hpSXH2NuiT5ZJSQSFoPRgOM5mjDQaYAkGT2t0nJMu4n6jekuWJ2OWMnbLH5TWbONPHyVF+hRtNianFVZT4RExpYBxfdwvLWSWS1DA+4/tjet/cYVkssb+9dp/K0yioZVI8Spnp54av7idabEykYk3DAX9u9XpuPPOEx+Jq5jt42p/Rw5L5j7yPDALlwLNBCeeJ8Y36brtNawQH//ncyz4YKncODBDmIlAFPyDew5hwLi+MpQ60ahIKkqyQXGiWzOwkrZPHJFx2SURjQXFjhm6oUN24/eIW5TxSLpdjQdwMPEibL2xDtTUwPKMh1UvdihZq55xdtu8+q0AJXt2B+Dud0Qr2OJn2GHH03CUU/cBXcTy4YU+WngiMPuTVh6tsQET+kpsje5NukgpehSyc28+E/3H9EZVhd2JidZYt36dsXTGSnzw9HRpFk5RdVh2kT9dg+x/yn1SGFj0TnHlGCq6YCl0LMhnRu3os4ZLMguFieInllkgiHad82pjiixToeuTQtupZ0eMOmuYuFgHn9QfCz5cKpZf35G9pJSDQspZlBQh1h9yf0xc6ty7E9p2ysWOtbsxfNJg+TyPQ8uN1me0W5XNCBhrXT5zlbTTYwITv1fIH4LltOdj8v0KKfgP1em4w6bmnGvJ7x6JcTKG+enzX0j5Cb8vxZXZxszEZYyTtagsuzn7xjOkxIRlNzyHPCfMih1//micfvkEafyQaFQkFSXJEAuJIiP1hQfrWlySWbpNykAQI6v0uI/pyIbh3MtqZtNm2UItHX0GAP7FduKO1dNOxgky2SIAeMYCnjhjl7wT7cxY/yK7qTytS8ZYxfrsBivtogbierTw4jth2hi89dcPpLk3i+oj0FXJ5JNBE/ti4MR+4nLkzEQOHaYLlvE7WjTxageZ7EJL8ON/f4b1H2+WuZJsjN5zcIGIDsWP5Rp0uRI2JqdVy6QUdpthZx+2Wxt88gA5NmOgLNNgV5u92/ZLfI+uR0nM2b5fRJQZppuWb5V44uCT+4uA0aXKWCljexQrZrLu3LBHLDIejz1ceaNAgew+sCsmXjgGCz5YgmWfrZLm64yn9hnVU4Yrd+ndSaxnWoacPcmesLSsuRZXtkuSirjf3PZtJTOY341u0Ei9KTvx8AZj9+Z9MrSZ53jB+4vh5oDqrodv4vh988PbsjEDvwuPccaVJ2PM2SPkxoJWMPcXnewUDa1nJh7xO3O7WM0PGhsVSUX5ioHJMsRY+oN2hRWxohIN6w+9Z0u5BAIHw5M+2MjcB7h6wkq//LhFJh6Wd7J0vLGFjS8wDlkFODoAzt52LJQxQlcPWCKQo+PWa1qcipJ5LeDrD+ObF277lwm4T5HPnuh5vvTu86TgfvGnK6TMg9YiBZBxRLaqu+X319e5wNL1SvdfNMwYZWs1Wjd9R/eWBBxmxjI7lfG4geP6iZVF4du34yDKi8pRMICDjQ+IVRe50Pt8fumbyjrK7LxsaRLAi33bDvZwY8YvGR9llijjdgd2HZLuPyw/YaMAJr8wgYhiOeu1+WLtDjtjkC3ktDgL2qN9tzxpKUeXLuOUjE0yaWnopQPRf1wffPbyl1jx+RoRcoowaygXvL8U6xdtxoW3ny2v2cKTF16/X85ZTaXP7ljscspNAy1dWoIUqkgLOi/PLWsyq+zesHs278OBnUW1taf1YfyXblrOr2RLOsLjM4EnHvQKcNg0M2urw5YwrVje4NCCbUpUJBUlBszIlEJ33zI7GUWCXDkwnmGw2EqObkUW0AeYeh8AHPliVcnFvxmQ0o60c6Q3K4vyRcjZB9Y9RARUJoE09jEdGbaw+UdKVxtxvTrawnKPsC1XiUsyPuk9qi5C0rLPexLgYa9VXmBZ2N44lyRak3c/fitmvTYPs9+Yj/07i5CZnYZRZw3HpGtOrhWoWDAT9G/3/kdmKzImyd89RarPyB5SssCsVY7Fiv6OLN6npcZ+qbQa0zIO3xyw1R1Fh5+l6FRVVIlwR9ZgZ89aItLn3jxZtv3Lt58UtzCtvOjGBKyR3LB0iyTRRGKfsg/Lgdx2OWI10/V6we1n2SUeTqe4ZymQjE9GF/WzYcKOdbvxyX9mYuSUoVLqMuSUASgvXiSCx3W5w80L2MqOnX5keLMvKOIdEUm6hJnFS7cpYWxUeuXGaahAgeOxaC0fDXQf//f/vYNtq3eKwLJek+do87Kt2L1xD86/7WwMHN90/YlVJBUlVtE8syfpRqSV5Aj3ipRC989ta4qC4V91uAk6XU9MKEm/FJY0FW965ALp6iO1g02NlIAEVsJIUkzQnrbhvjBGL9tjv6TYtZyNf3Ph8bgx6aqT5XG0VJZX4aHbnsD6hZuQ3S4bnTvZ1iBFgYk5TOyZfN2pDW4CWKrAuB5duex1SouNFiiTduhKpSjw50P7iu0L/PLtaNOBXXuypIaxaM8hzHt3kbgcaYGumrNeutDU79zDmB/fp2W1c+MedItqEBAp5WDSD5NeKJC0SGe8OMsWucqa2rpRWm78Dp16dpARWVwDcbndEsflsXljUFviUlKJyuKK2gSfSEoYk4PYpYixUn5/Qnc1k3ZoWXO4dX1YDkKLlElIR8PCj5aKQLJONBLzpJuWrm16C2a+MltqLSnmTYGKpKLUh8OV2SPU2b1uUgkL3Xkxr/7ArhV0DQAc3ewSCHaOCW23G5ln3QqLXXtaCTKWqvJ5u+RCvqsDxvcl4JwRvikYitbCF6/OwcalWyXeFZ3YQquJZQkHdhSJ9cV6xvrQ6uPw5IriCuzasFdcmrQqQxyW3CFHEmFk8LIxkljDWk0KEWN5LFfZtnaX1E5WlldLtxo2G4hF285txF27Y81OcXVSoOlKpquWCUmnX3GSCAhnRb7+8LuSbGRn21ZLQ3IKCzv/0DXM9XB9ZQfLZYDzns17UXKwXMaLsabTE57F6U1zI5SdLhm9fM6bBf7MY9Ktesql42tvHJjJyqQnZq2y80+00POc0F075uzh4m799LkvxPXKDGFagxyhFV0byXPGBgeMr0YEMhoem+U1W1ZuR7/RjReDj0ZFUlHqYQK0EMNDihtQbSeXIPdwUb1A1xGL4LfB1MyHlRGjpVoSYkxVuCZxS3jUlysqC3WbPbIqq31S3RTYHW3oPqwRi4YX9Qh0TfJ3Hy2QEdiflBYUE09iiSSFqmP39jj9+9OkoQDjkkzsYTkGRYXxQib9UMhYusGxV2xawIbjHAdFceDamK3qq+SYrAp4urjFLUlxo6tTWtaFIKLCZgB8j1YlM2h7j+wp4sO1cT+fPDsTa+ZtlGxdlmhwEgmnm7Bon6JC1yvjmsV7S7D8i1VS00nXKbctYe/XGr+d9ep0ikBR+BgnZZarne0aqO24w+zfaCiazGSl6PP40sauolqSjnoM7ib7fuXPb4klSsuVCUpsGbhs5ipcdMc5ItCEsVoZJt02drN8Np9n1ixvRpoKFUlFqU+IMcg4c+o4rUJcTpE2cJxisQkI7rfjcbQoq55DyNUfDk8rGIFFl7LMwgz3V41AF6m0nlsnw4+t9PORDLC7DcsV2KScF2p2vGHiB7M/OfWDpR1MUokF424UkMhIp2h4sWeZx+izh2PQxP4475tTJI5Gq5CiyOxQl9eN/IJ2cKd5pLi+urJajkVxZUIQxZLF/z2GdMeKWWtRtKdILDqKi0wvcVgi3uz2k9cxF9f+7DIRCY6eYvyPmbIRa45t8dgjlqLNV2j9Uqj4Pl2u/O5MsKF1TMuOVlivoT2wdsEGESwKG0dZ7VizE9ntctC+S1v5kx9//ijprEOBpdUciUvWh0lDbJ7ApugsK+E5oxv2tMsnSi3mp89+LpmvkZIXwjgmk44+/NdnuPreiyWximtmXJRCSeu4PhRrJi9xu6ZCRVJR6sN2aVgW501ajOzPmWkLpG+JbVlKfSAtzzJbMCufgbFugMUSiSTGBLaFv2+MmwbptcruPmsAtHyRZN3iG4+8J65MWlDSEaesWmKNjMldds/54mbdsMQu36gPsz9pRbE2kvtgPJFlE/u27ZcmA3QJdg03Q2cbOs6rpAvTbmHnkLmOLNlgfI0iGalVZHyQIsAGAYwbch4lXaGrZ6+TJCImxPDY7HLDJBaK3imXjKsVKGkMXg8KIGsaaQmyQJ9WHd2jXAcF1+8PSM9YNjugILIGlOsefNIA6TZ0YNdBEf2SfaUoHMyZlN3s2ZO9O8kaO/fuGHfgMa1jDlXmOWVJx2lXTJQpJBRnmuL/+uVL4uaNFkhCa5UxUgolRZ3ni1YmmxTMfXuhuKzrl3ywsxJjrLROmwoVSUWph+UZZsfcaDWygL7BlAnOmmwfnoBRCljtohqM26UiFFBT/b6dWNPIpRjNS+goOvscYSpHAqHLkWJB64hixXmLFKMB4/uIC5HwQs/WdJxuwRZqEy4cI4ki0nHH7ZDtCdup0a2XkZuJceePQnVFlQwuptuWblXObszISsNbj32AUVOGScxvw6JNUrhPEfOHjCTlUBzpRk3PTEeFzIH0SkcbPphkwyYBtLDoWuV7FDgmDYn337LXwdpLumspRpHv0eC7c9JGGBbnd+yRj5VfrpH1RPerbe91o8fAw1myMm+yZwd5kA1LN8tIKyYkUVQp5FKf2acTpt40qc5kEcJGBO899Ym4Tpkhy/3SFcztzvnGmSLIFOyc9rEtUIoiXbosIaFIsuaS4siEqtlvLZCbGB6brm8KJL/HxGljxZXcVCTzf72K0jQw9uY5BaiZHhZBZnAyYaUoXAfYx67rC+4NF9VHTeCQz3e1xZVWmDQZb/rs06aCLeAYx5I+rTKmKgoZbcVJHnEaBiQQCtoXr80V1yrHRfFiy/hYWqZH7nGkkbjHtoRondAaYc3gSRePldjbvPcWyz4iA5dZB8gLPt2p0+6Yio+fnilNvHmRzi/MR49B3aQlHS2vj/41UwSM+6cLlK5NxtRYOrJz/V75c+E8SgogRS8Ck3AYs2MiD7NUaWFSKNkwnGthFjCtL8br2J2Ghf8U41jQrUuXqCQOBUMivox50uXJ53R/Umh5XijeXcOWZJ1zyM9t3CuNB4ZPGlLbvYjWMy29N//6Aa758SW1TeTZpP3tJz+S0hkmHUVvz/W++df3cf5tZ4nbmPuORSR7lm5tCu7bT3woiVRM/uFbLPvYumI7OhS2FwtzwgWj6zRqbwpUJBUlVklC+gWAMx/GN/vwUGBnR1jpE2E4CaPyaSDAOY65UQJJK7K7XTIiZQ3MeE38PLwTgnWXzi5AaAvgYFwy7O6SuVC7ZSpIrNZwiYYjmz5/ZY4IDZt203rZu3mfjK5isTwna7A5QAS6AunKnP7cF9i5Ybdd4+cLiLCxcQDFic0A6Cr9xYV/kP6iDqcLbTvloOJQudQtshk63Zt7Nu+R43XoYSez0D3JrE7bLWqkEXlaebUIWSBSUxiOg9KdS6FlEg17ntI6je4+Q1GjlUUx5SMeFKlew3tg5RerUVPll7IUuoMphPyZ4s736TamFc34Iy3OaHic8uJKDD11QIP5meziw8+tmr0OJ00bK6/LsOalW6RBABOVWDvKPxOeW1qFW1ZtFzcsXbvrF2yqM8syApsUsL9rp14dpP3f2gWbpIsPbyakFOVQhXQm4k0D45tNWR8ZQUVSUWIgLlLvydL+zW7ETZODrcrcYQfjjUBgT3jyhbFbw1FMOGtRhgtSNF1xMmSPD7nLDu0FgrvDdZndYDljdzVpLGTEVcaVdglIcJ0dd6VFydFYVq4MRrY4uqoFQSuOnXZ4sxPJXOVF23I6RCR4wWemKS++kYs/L76MobEsoexghQwyZpkG43LpGR5bXLfsx/6dB+zZlJyP6GBmaYlkdlLkNi3fJiJatKdERMgEjQgs750Yx+QPkoRS5ZckHdb20Xpl43B27qHrVDrZcMCyxCuZtOKWNnK0ZHmMnLwsO1nnULmIazz4Pc+6/jSUFZVh+vOz5E+SViVFme7Sbn3t6SCsY6RwsdXc0FMG1Io1RZ0x3LQMr6y1PrS+KX5r5m2oFUnGErlO3iDs3rRH4r30QrD/Ko9Fy5odc+jS3rxsG/ZvPygdfiIWLK3b/TsPYuzUEeKqpWu3c48OtdY2t2NclA8mYK36cg0GjGM4o2nHo6tIKsoRoChKc+/6r7uHwaRPBTjXkS3ZKKrR/7FSyFxdpUVcY2BCJTBVbwH+FeHsW/boNDCu7oD3DFhsARc9s7ERkWYFmZy6sQyG2a5MXmLrO/cIWPXnWbYA2M2G8TN2lIlA4WECDLNX2ZuUreE4A5IiyZsPtqBjp5jstjl2MwDpTGP3YmVcc++2g+Im5P9cHnswMxt9M2uVrlEm2tAtSmchG537qu3OMvw8xYQWG12EtN4WfbTMzg7NSsegif1EVMQ6LKsSN+xZ15+OD/81Q14/uNuOeUYmsFBY0rK84iqNHhYdgS5KCh6tTB7j9CsmYtNSThIpFwHk9430RpV4ocMh54WiRPcxLWZainyd569Nh5wGDQ0i8HW6kiMwU/bg7mJUle2WtTH7lcegJc2yloycDLTr2lbGdU25/nRpAkDrk9Yzz6c7zSU9dNmsgZ4A/h6ix2dFw1mZ29bskszfpoxHyvds0r0rSitF7l69Z8CwLZ24IpnAQHdVjS2QlgeWd0rc3qXHXqv4HOBfDlj5djZtaF+4/GQt4FtgNx9Pv0DEsimwnLSUJ8FKm4SWjlhxtPSiblp40Wb8joOFKSQsQ6DLjsLEeF11pQ+FA7tKK7j6dgkTagI+vyTnBHwUFicshGCMPbORF2qKLrvq8IIvyTTGXysufE6L7MCOQ2KV0RKi5cr6RlpyzLRlwgwbozP2x1rHj/8z03ZXhsWD+2Acj6LEhuj2iKzDIsl1sMaTg455k2C/CImF0hLLyMlHt752n1T7LYMd63dj66rtKGXjgt4dJdmnqrwKXft2FFcmazo/fGaGnKf6WaUy7HndLuR1ysVTP3lW1sb5mHRVt+t0uKMP4Y0C4617tx+Q/fB1fkfOtGRyE88dM435nDcb/K7h0GRc7H0b+V03NSqSinKciJsx8wY7i5W1kiE2QneLBUmBhHt44xyI1iMtONYqsmaRokw3rkULN2BblsEddmG/lZn0ZScnCmNv0kR7R1EdSyQy2YNWY3VFmRTRU9g4AYNCwJ6gLH/Yu21frVuS2I27ZbqyiBEtSRiniCOtS9bq0RKkyLDu0uVlkTs9C4fXROuMgsxOPEw06TG0QFrf0UXJ6z0t3VMunSBF+FyLxDwtB9yZ7HMatOsBeV+WYTdqZ2s3WsQRVs5ag4/+9Zkcp3BgN1k7BYRCfGhfqewzr1PbWvcyW+UxO7emwidda4aeMlBE99CeQyjaXYzqCp89luuDJeJmZrZrRPRo3a76cp0MnuY5XD13g3z3UCgoo8HkXMBIiUrkMyyDYTJOdMIOf0ftzo99Uyf1pB6XdNyJVeLC89NrRGHMUWiNjYqkopwAFmceZt4OBLeHM2HTpche3LSNhPHRgmSsKDxrkY3Ba2Od7vDFmO3OKmFqZgGu/k0ep2nJ0IKjRfbu3z4WlyqtmugG4e265mH0WcNwyiXjJdaYnuXFP3/2gpRHsL8qk1V48WfRf0Z2mm2t2Bop4kO3JS/4vICzMw6hdcdfD5uZM2ZYMKCrWGjs6sOWbtVVPpleQcvu3G9OllpItpajq5a/K7piI3WP6xZsQE21Hx6vS6xXIs0E6B51WtIxh2KzackWdOnZUUSUo6f4G+dEkAgs8aAQcY37dx2ULFaWvLCvKm8UaDW369xGpm9EYpHtuuSJe5eTToadNhCTrzsN7//jE0nSYdkGXaCMBVK02VCdMcgOBe3kvHBUl8RgLfaQPYhQICTr52e4ls69OsR0EceCiUHdB3XDxsWbUTCgm8SUI0hrPIeF4acPbpa/cxVJRWmMbNimTF6RKSQeu26T7lywh2w04UxaunwDGwFzKFy2krqMmDRY6ugWfrRMLBvG53jxZ0E+SzymfWtqbXYlrcNOhfmY8dJsBAMBZGanozxkxIKktUhXKwWS2aYyBcTnh7/G1EmckYu1g4lO9ugs7rP/WLZxOyDxQX+1X+ozr7/vSnQJ92RlXI2P+pQeqhArSsS1okasXVFAEWlL+qGyrKKy1G7FxgbjtPY4BDoW7Qvaobq6BuPPHSmdeBjLo3D1Gloomab1i/rzurTF3i37JJFp8En9pWkC28UxkYaWJhuJ9x3dTco6+FlaoFynxGpLDdp3yZUkIdY2sksRu/UwcYe9Y5l4dDTQ5XrOTZPwZpUP29bslBsGxnh5I0P3MUt12Oe1OVCRVJSWDmcrUvxkDFe4qjyChGRYw5gR/s85aNc0pji0Js+64XRpObd6znppJsBRWf3G9kG/Mb3FhUercdOybZJdunvzXpQdKpNhxTn5WSKgvNAzu5XWEqFrj4LBjFYKLy0liiIFjDWILNdgGzoK4k42QR/ZCwMntMWezfvh9jpx2f9cWCuQR4Kt5CgE4mLlaKpqnyS2MEOW7lZafW6vPZyYUPgZY422tuqeCzvDtveInph605liRb7w+9ele440XI+xPYU5chPQtW9nsWRz2uXItBK6pV0ul5wXWrSMrzIblv1leQNBFy/dpRTPEacPRlp2GvzVdqYwhY0WLbOAGV+lZc85mLGSb1gPetW9F0vdJWPJjBvz3HDgM2eCNpe3REVSUVo4lnu4PZhYshnspJDDbeJYK+e2R3oxoYdNDByxp7qnGrRG6EqMNcyXJQQfPj1dCvcZ49uyfJu4BOlClUQVy4KxINmYvUf2kMkZFAR2y+F2EauO7sM27XPFXUrxYlIQ6y3ZEo61k6ybZJnDGVeeJC7Wo4EJRkyk4fSO9Iy08IxIxhjtxB0KNEWCmaCEyT+sLaQIsXlBvNpDNhKgsLBulC5oZuNGN3c/vH2FxHK5fWlRGd7/x6diRdJC3rqSPW8DkrFLceQ5ke5A6R7ZN8WSVirjwbIWhwMl+8vEMmWj97cf/wD7th8Ut2xO2yypR+X5oVuXFn59mPTDDkZ8JAoVSUVp6bj6A56JQPVMWyQ57NiiEFaHOwBxOgfdseWA+6xGyahtzdAN+9ZfPxAxY7s0li6wswzjk0yusSeD5KK8qEJ6nFaWVIoo9BzWHZXFlZLFySxWul5ZspHb7nCCCt2nFJid6/eIxXPGVSeh57DCmMkn8aCrlo3OaQFXV9WIWEesO9ZY0u3YqWd+bes5WrFsqs7MVjYEiHQJsjNva2S9o84aKu3uCK3kgRP6YvabC5DVNquONUnrla32OISZVt4rf34La+auR6eedgcd7pMTRGgxl7GEhg3IMw4fj65YurZ5s8DYLuslKaTMhmUfWpbCMLmIXXy4flqRxXtL8c6TH0n8N14HoUSiIqkoLRyLSTvpl0i9pqn+CPDNB8x+WyidbOzssRN63KNgpR39gOFUZfkXa7B/ZxF6DukuViEtSUnGcTmRm5eNQ/tLZUQTDXfGwnwhIxd91lhOvfEMuNI8ePx/nhZRaBMjpsh4IYVywoWjRbyOFYoU45hjzxkh2a9sfGAnvzgkE7XX0O6yHVvD8TsQNlRnbHLD4s32WKryauzdus/Ogm2TKdbk+kWbpB0fBf2ki8ZK0hBrKCl+kmBTUSMzKGmlsgSEyTobmRzUp3Pt6DDpKLRlP4IBewAzLUdveEIJLU2WctCy5VrZL5Y1j1++Pl8scyZRMZGI+6JVTMHk8YecMkDim0umr1SRVBTl+JBsWe9pgGcCTGAd4FsGBDawPgRw5MFiZyA2FJC4pRIPXthpGVE4xG3KMdqZXsncZPkHxY2jp2hNMtmEgkLhYFIMrcl57y/BdT+7TOY/zn9vsdQVRkZQEVqlFCTW+x1vyzTG3uysUbu2sbykQkST8UhmlEasObs0BbXW5CXfOQ/LZq4Wq4yjwCj6hYMKJIN126qdYi2fee2pGHfuSMmk5dQTFu0zKYfx15z8bJx8yTj0GdVTBiGzoXjJwXLp6xp9nF7DCrFi1hoRPpnlWFopa6JF6vHYsUpO/aAg8uaCLmyJS9YEasddUURp0Yp7em8J2nZog41LNqPkYJncFDBBhy5i7ifSYzdRqEgqShJBV6rFfqruITCGCSV8MB50dKn1qY7E9XyBOkku7KLDNnQs2Ti4q1ja2lGQCN2vvLj3GtZdXLPsEMM5jeffNkUK51lvWFleY9cAhuxkF7opL7jt7OPuBJOZmy7JOHaXGntgMQUnUqYRyfCsP4iYAtSpR76Iy6gpQ2VaiLTQC8PM1s9fnYPug7rCV+WXBul0hbI2s1v/rtL8ffHHK/DyH99E0e5DEq8t3lcinYhosUaygdkQgTcDFFa6TXkcwjXR3cyEG77PGw1au1yXPfcx0gTAhueM1jqbJrCGde+Wg/j3L1+2j8nxYm6nHOuUyybEjFc2FyqSipKk2CO49D/hY03moZBQ6KJLJpjAQlfkgg+XiHBwRBbr8SiWPYcW1E7JoKuQXWo4NYRZstyOCSwcB5Wenib7Pv2qk8X9eTRQDNgjlW5NCjJbudHi43zK/dsOiPuSIs3sWMYFKdSsr2TWLpuY12f13PXi7q0/woowJkmX67/+9yU7vllZE669dEiiUOeeHaTnLbvjdO1rW3AVJZWSrcptB53UX5KICGOmdO1SAMedN6qO+PEmZPuaYunRumbuOnvQc6ZXMnWjm7lHoBXKxCi7VCdNGhewnpNCS9cux4857zo3ZgJWc6D/hSmKklIMOWWgdLuhBUP3YbSIsMyBsFRBZjd2yK2Nx5Gqyhos/2K1NDNnhiszMxm3pEXEeCGtO7pzKSqDJ/aPKWS1+yqvwrtPfYI1c9aLpUUB37bGnshBEfJmeqRlG+N9laWMi5ZLjJKN2SdcMKpBqzjCnrUy3DgGFCNak4xFjpo8rLYbESeLbFu9A/PeXYQegwvEzRypl8xqkyklLRRLtu9jWQthfJai7U13y5p402CLaoXcXDAhZ+SZQ1BVVokv31wocU66uBkjpaVN4bTHf9nlOmyyzhsVTlKJwAQkntNdG3bLyDO6eWN956ZGRVJRlJSCRf5Mipn33hLprJOVlyU1foyPte/aVoYwU4jqWzxM8NmxdqcIR79RvWuzSDkkmDG6PVv3izXKvqpfrJ+LZTNW4uyvT8KQk2O3CZz+4pdY9tkqdOnVSYSN4kKx5f5pTVJE6OKkZchOPrTQ2PeV2aEsxeg9vKFlRUGP1HXWhz1d+R0Zp4xu10erjccs2VuKss7lElMVN2+G1+6vungzAuVBsfZ408AxXrScp379DOmMw4YNrHss91eIJXzaFRMl7sl9DjppgMRJGe9lli/rHelOZRZsRVmVuFRpidOCpHVM+L0p5DyvjMXyBoJJSXS5jj1nJJobFUlFUVIKWmyTv3aaZG0yaYUXZI6mGn/BaLlQf/rcF5JtyW40kSblvFCvXbBRXLEysDkskLyIb1q2FWnpXnFbMn5HS4xWG6eHsJ8qXbCM00XDovzVs9ehfee8WsuP/UgZi2R2KCeAsA6zc8+OckzGSdmOLj07Xfa/7PPVGHvuyDqWMKELmOvkeus3Fziw65Bt8Ua1rot8N66VCTbrFm7CgZ2HpByDVnAnsRa92LZmhwgdt+07pheGnTZYbjZo2fUa1kOEzx4unVWnnypvNpjh+ulzn0ucM79bexxyHkLR3kMiwkNPHShW5/z3lkgMl00LeD4ZL6W1GYkNF+0pxttPfCTuZ7aja05UJJWkQGYpsj9qcJNdTM/5ja4BsBx1W2opytEKJS08tl3jxZ0CFxFExtnefvIjSTqRaSJMObFsK61jjw51hgVz/iFdq7SwHBXVEr+LTM1gbG3T8q0yc/GUS+qK5O5N+6Qkgu7NCHaPWLsgX+Z0hGzrlW5MWrUBX1BiejwWW7VR2OqLZP9xfbBkxkpxn1JgKcD8DhTzg7sOipUbPUKMnXo2LN4kNwrcjueBAklB5qzL6kofeg4pQKfCDjI4+eu/vkq6EkXHIPlzrKYEEUaeOVQScFbNWSediPqMKESPoYXoM7KHCDYtxpWz1krZDZOS2A6PdaUURMLOSEwIYtu76S/MEtdtm/zma5ihIqm0eGRUVNXrgG+xPexXepVy6HAnIP1iWG6784iiHCu8wNP1Fw0tqK/9/HJprr1r0167uL9XJymL+Py/c+tsS9ckxZX7ETGtt29amLSKYt701cOeYckm6bHHP1HM2Zw9MiIr1pgoCvy0b0+VbkJbpDuOX2Kn3He/0X0kOSZa4GjdUSApvBRSvsduOIytsgHAro170LZDjnTemXT1yTJQeu67i+V7s3EBrWo2c+fnuJ91CzbKlBBvhlcsTGYFU+R5TuPFZ2kZFw7uhjVzN0h/V2lKEBbIiMB37NlBykHoZqZ4czBzc6EiqbRo+B+JqXoDqPkCcHQCHF3t3qW0JkMcD/UCkPlNWBw+rCiNBC0ZJvjwEYFuTFpajBEyLkiYVBMRPFo8TDSJTi6hVSmTMerByRl0SzIexzgj4Sgrlm/QmhIL1uGo/SyTjMpLKuX4c95eYM+TjNO7tH2XPOl5ymQYtoDjdpzAQeH7130vSe9UChZfp2gKxkgMkcdj7JKuT3YVKtlfKXHE0WePkA5Fs3/3X/gq2czcLVbu/PeXYNgZg6V+k+tiqzu31yON4hd8sBR9R/WUcpjoSSz14TpOvXQCDuwowtx3FklHIe6bFnVkckr3/l3gdDjl3NK125yoSCotG85opAXp6Fi3JymL6x09gOA6GN9cFUmlyaFVxBgbXZ3dB3aTOr+2HXKxf/tBmcjhcDlrM0Mjrky/L1jbFScadq5h0f7S6StkUgaTZyi8zOBcOXudxA6lkQGnaZRWSsYrs21pPZYXl4kj5Z0nP0To5skYMK5vTHcyE3T4iIax2PeeOjz6ioLD1nW88ew1vFBcsXR3UihD1fYQ6cIh3SVBZ9Zr8yQDOLs355ja0GU8/bkvJGbavX9X9BxaWCvedEOztR6twkvuOu+IDcnZbP2y/7lAXNy0XnmjQMHu0reTlLNEmhAwPstz1ZyoSCotGxn9VAE4DqeG18L/6Ky2gH8ljLlIe5YqTQrdhpwF+eZfP5BaSdYTRhoIlB6qkexL9nIlTHDZsW6nWHBMpqkPBWPytaeKCG5aslUEgYOaKVhM9GG8juOvZNRWWSXSMtKkON+EQujWrwv6ju6JgzsO4cNnZogwHykmGM2gCf1kXNVSjr5avFnifGxe0G9MHxFArouuV66DGb+7Nu3FsFMHyiQOClX9GCjHdlHQaBEPOWlAHSGky5Vr27Bos8RPj1QOQxi3pKuYCTo8b/w8z3EE1pF60t0yZ7I5UZFUWjbSVYZiGOcuVKZhBMKTMTytIv4K/2ogdND+z9PVG3AWxL0LN6FKIMDti+xz4eT2zTdGKBXgmC122qFlRKG68Ftni8WzYdEmGY7MjFg+p1W1dcUO+0OWHds8/9az4roamQB0+XcvlDje2nnrZXII6w0Hju8nzdQZT3zpgTfxxWtzRaAoGhSySPedTr06YsuKbVg7fyMmXDD6qL8PRZYPv8+P5Z+vxluPfVhbu2gvncX/afJ9MrLT0SY/RyxOJi3VhzWUHI/F70vBp8BG43Q7xDL9+D8zJUmKmayxZmhGGDSxH5bPXCXdfiKiStcz18KYJwdB06ptTlQklZYNs1hlLLwvtgiGSgBXD8A6vhZgLQnjX20nKAV3hwdF0ueWAbhH2AlK9TJ5jX9VePs9dbf3jALSaVkn/zlJJLQQZ746B4s+WiYxwYiI0CV56mUTcPWPLql9jeUTtJg4l5JwbmTvkT2/cvoHM2aHnTZIHvVxZtgxuN7DesS0wth7lqUoLPI/XsuYwkWRZqcexkQpYBSl4r3F0rd1zNThIo4R92/MnIHwy9G5SHx9x/rd2LFul8zrZC3lpqVbkNM+BxMvHINx542M2RiAx7/oznPx/j8/lZgqLXL+WbP+ctRZwzDla6eJK7k5UZFUWjbuAfaki9B2wMGRUFH/oXI0FALS3DvZe5eawDaYyucBUwY4C22rkFcdUwr4voDh906/uvZCZQJbw9uXN9y+5nOwcAHpV6pFeQIwieSLV+cgOy9b4oo8l4wzsgH3B09Pl/KKSDyQYsiaPz4aExbtMw4XDyk3cR3/3z5rIC+8/WwRrzVz1okbmd+TtZqnXzkRJ180VhJoOGyadZztutjjtiIwM1hEy9g9ZyMwOYi9Z7kvJiZxliabNLD+85PnPpds2xGThsRcE120X/vZ5dLJhxNRmKzEwc+RhvPNjYqk0qKxLK+MiTKVz0mSDqw2AHuWhkrtWYreibbllOTIUGW6WJ39Dt8ISMw17L7yLQE8pwIuOx7DZCVxscbc3tjJTt5TAWeMWK7ylTD+tejjpcjIzqgzyJjJOrxYs9nAgg+WSLyxKVulUVzYZzZSe1nf0mWMske95JxjJTM3E+fdPFksPN4AMGOX8cGIm9iT5hHx/+yl2cjIzahjHbN3LQWPZTCMY3JbJvFEyl7YHMFu92fXVrKRORNz5n+wBINP7h93wgfdyUxi4iPRJI1I/va3v8U777yDJUuWwOPxoLi4ONFLUpoJy90XyLoFpmYeEFhmxyndvWB5xgPukfYYqSTG0JXsX2EnIcW6U7ZygNAeILhRRNKYaklWgpUXZ/tcILQXCGxSkTxO6CZkTWB0L9FoKJyMtdEy4oW/qRgwrg8WfbxMMmoL+nWpbZVH645rZOZn39F2P9UThZm6fMRi4rSxKNpbjFVfrrPrLjO8MhCa3ouTpo2VTjtLPl0u8VtaviwtoSuY56n3iJ51LEDGVFlzysQkxkZbOkkjkj6fD1dccQUmTpyIp556KtHLUZoZy9kFVsbFMOYC3p+Gx0NZrSg5KWhbyLGQ7+kATLDe9u6v2N7fdGtu5TAWFulAEwu2fKPrldZcU8IY3QW3nYV3//4Jtq/ZZYfnWUNpWTKZ5IJbz6rTBq6pSMvwYtq3pmLQhP4y2YMDk5ns039cX0lc4vkYMLY3Vs1dj/ULNorFSSuwQ7d2tS38IrAOklNIJN6YBCSNSP7yl7+Uf59++ulEL0VJIK1yPBQHJTs7AP5NMkC5AbQ0a5OYuH0G4MgHAtsAR9vY2zNG62w6C6e1w+QcxhyZsRkrO7XsUIUU9Ee3qGsqCvp3xQ33XSF9VRnro0CyrpDiFD2hpKlxe9wySDreMGnWSPJx6MIx+MfPnofH424gkISTQJgl3KZD/CzXlkQru9rUpaamRh4RSksZx1KUloUkHbnHwfjX28lIjqhaNCbjhHbYyUvu/oe394yDYQ1pzO2Z5NRNetsqxwdjcj2GFIh7kU3FGYuMwI47nKZx+lUnNYsVR9gBh429m7u59/HAmk3WjLLjDusoIz1xI25iNisfd+6IZu2/eiK0apG8//77ay1QRWnRMPkouBmomQ2E9gOOHNutaorFarTSWW4QVU7gGQ0ENgOSwFN/+w6wMrh98teNJgpaa2dec4q4FVmLyCJ6do6RJtwV1TIBg+OglNiccul4ideyvlTGbmWmobqiWvqw9hraXdrQJQuWidVpt5n40Y9+hD/84Q9H3Gb16tUYMODwHTHdrffcc89RJe7EsiQLCgpQUlKCnJzkMPWV1MEwhuhfbGe6BvfZMUrXUFjecbCcXWMn/PgWw/i5/f7w9sPC27f8hIhkoHh/CZZ/vgYrvlgtzQRYzsCaxiGnDBDrTokPGwCwWQGbAzBbOCMno/bcHamXa3NALcjNzT0qLUioSO7fvx8HD7KzSHx69eol2azHI5IncmIUJVHY/0kyDuk4qsxde3veDHIaRat2DiUMljVwmDFjbE1Z8tGaz53L42r2RgCNoQUJ/S8qPz9fHoqiHMbO2vUe4/ZH7uyinBi8uDvTW8YFPtlwJvm5S5rbzm3btqGoqEj+5Z0J6yVJnz59kJVVt+muoiiKoqSUSP7iF7/AM888U/t85Eg7aD59+nScccYZCVyZoiiK0lpJaEyyudGYpKIoilJ6DFqgEWhFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoShxUJBVFURQl2ZsJNAaRklAdmaUoipK6lIY14GjaBKSUSJaVlcm/nASiKIqipDZlZWXSVOBIpFTHnVAohF27diE7OzvcFDq1iIwK2759u3YcCqPnpCF6Thqi56R1nRPKHgWyS5cuXznVJaUsSZ6Mbt26IdXhH3Sy/VE3NXpOGqLnpCF6TlrPOfkqCzKCJu4oiqIoShxUJBVFURQlDiqSKYTX68V9990n/yo2ek4aouekIXpOUvecpFTijqIoiqIcC2pJKoqiKEocVCQVRVEUJQ4qkoqiKIoSBxVJRVEURYmDimSKMm3aNHTv3h1paWno3Lkzrr/+eulGlKps2bIFN998M3r27In09HT07t1bMvd8Ph9Smd/+9rc46aSTkJGRgTZt2iBVefTRR9GjRw/572X8+PGYN28eUpWZM2fiwgsvlG417Fz2+uuvozWjIpmiTJo0CS+99BLWrl2LV199FRs3bsTll1+OVGXNmjXStvCJJ57AypUr8eCDD+Lxxx/HT37yE6QyvEm44oor8K1vfQupyosvvojvfve7ctO0aNEiDB8+HFOnTsW+ffuQilRUVMg54I1DKqAlIIrw5ptv4uKLL0ZNTQ3cbneil9MieOCBB/DYY49h06ZNSHWefvpp3HPPPSguLkaqQctx7NixeOSRR+Q5b6bYs/Suu+7Cj370I6QylmXhtddek2tHa0UtSQVFRUV49tlnxa2mAnmYkpIS5OXlJXoZSoIt6YULF2LKlCl1ekDz+ezZsxO6NqV5UJFMYe69915kZmaiXbt22LZtG954441EL6nFsGHDBjz88MO47bbbEr0UJYEcOHAAwWAQHTt2rPM6n+/Zsydh61KaDxXJVgRdP3R/HOnB2FuEH/zgB1i8eDE+/PBDOJ1O3HDDDUc1hLQ1nxOyc+dOnHPOORKLu+WWW9DaOJ5zoiipSkqNymrtfO9738PXv/71I27Tq1ev2p/bt28vj379+mHgwIESZ5kzZw4mTpyIVD0nzPBlUhNdz08++SRaI8d6TlIZ/vfBG8i9e/fWeZ3PO3XqlLB1Kc2HimQrIj8/Xx7HA5MRCBN3UvWc0IKkQI4ePRr//Oc/v3IYayr+naQaHo9H/h4++eST2uQU/rfC53feeWeil6c0AyqSKcjcuXMxf/58nHLKKWjbtq2Uf/z85z+X2sDWZEUeCxTIM844A4WFhfjjH/+I/fv3176XyhYDY9VM7OK/jM0tWbJEXu/Tpw+ysrKQCrD848Ybb8SYMWMwbtw4PPTQQ1IGcdNNNyEVKS8vl5h9hM2bN8vfBZPcWHvd6mAJiJJaLFu2zEyaNMnk5eUZr9drevToYW6//XazY8cOk6r885//ZDA25iOVufHGG2Oek+nTp5tU4uGHHzbdu3c3Ho/HjBs3zsyZM8ekKtOnT4/5N8G/ldaI1kkqiqIoShxaZ9BFURRFURoBFUlFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoShxUJBVFacCxTpxnL9ivminIjkacSakoyYS2pVOURoaCweHE9UVmxowZ0hv20KFDaNOmDVKN//73vzqvVEk6VCQVRWkWdIC1koyou1VREsQXX3yBU089Fenp6TKm7Dvf+Y40zo7Qo0cP/OY3v5E5n2wmzubrb775pjRfv+iii+S1YcOGYcGCBXX2++qrr2Lw4MHwer2yjz/96U913t+9ezfOP/98OW7Pnj3x3HPPyXZs3B2P5cuX48wzz5TPcEj3rbfeKo2u6/PLX/5SJozk5OTg9ttvh8/ni+tu5TF/97vf4Rvf+Aays7OlOXZrHU+mJC8qkoqSADh5hYOdL7vsMixbtgwvvviiiGb98UsPPvggTj75ZBmOTWG7/vrrRTS/9rWvYdGiRTK5JXpY9sKFC3HllVfi6quvFmH73//9X5nw8vTTT9fuk9tzbibdvxRUCtO+ffvirpXCPXXqVJkYw+kxL7/8Mj7++OMGa+X4qNWrV8t+n3/+eXGvUjSPBAWc0zX4/b797W/jW9/6FtauXXucZ1VRmoBEd1hXlNYGpyE4nU6TmZlZ55GWlibTEg4dOmRuvvlmc+utt9b53Oeff24cDoepqqqS54WFheZrX/ta7fu7d++Wz//85z+vfW327NnyGt8j1157rTnrrLPq7PcHP/iBGTRokPy8evVq2X7+/Pm1769fv15ee/DBB2tf4/PXXntNfn7yySdN27ZtTXl5ee3777zzjqx1z549td+ZU2UqKipqt3nsscdMVlaWCQaD8vz00083d999d+379b9fKBQyHTp0kM8pSktBLUlFaQKYoMMZe9GPv//977XvL126VKw7ukwjD1prHOjL+XwR6E6N0LFjR/l36NChDV6LWIK05Gh5RsPn69evl3mQtNJcLhdGjRpV+z5nQ9JKjAf3OXz4cGRmZtbZJ9cabfVxm4yMjNrnnE1Kl+z27dvj7jv6+zGjlrM7j2TVKkpzo4k7itIEUFAoPtHs2LGj9meKx2233SZxyPpED66NzgaliMR7jYKVjNTPduX3SdbvorROVCQVJQHQklu1alUDIT1RBg4ciFmzZtV5jc/79esHp9OJ/v37IxAISAxw9OjR8j6nzLMs5Uj7pNXL2GTEmuQ+HQ6H7C/aOq6qqpLkHjJnzhyxkJmUpCjJirpbFSUB3Hvvvfjyyy8l+YWuWLpD33jjjQbJMMfK9773PUmg+fWvf41169bhmWeewSOPPILvf//78v6AAQMwZcoUyU6dN2+eiCV/prBFrNL6XHfddUhLS8ONN96IFStWYPr06bjrrrskiSji7iXMZL355ptF/N99913cd9998n0opoqSrOhfr6IkAMbiPvvsMxEyloGMHDkSv/jFL9ClS5cTtlBfeuklvPDCCxgyZIjs81e/+pU0OIjwr3/9S8TttNNOwyWXXIJbbrlFSjAohLFgnPGDDz5AUVERxo4di8svvxyTJ08W8Y2Gr/Xt21f2e9VVV2HatGmSXasoyYzF7J1EL0JRlMTBWCldoizroNApinIYFUlFSTE+/fRTSRxiliwbC/zwhz/Ezp07xarVtnGKUhdN3FGUFMPv9+MnP/kJNm3aJG7Wk046Cc8++6wKpKLEQC1JRVEURYmDJu4oiqIoShxUJBVFURQlDiqSiqIoihIHFUlFURRFiYOKpKIoiqLEQUVSURRFUeKgIqkoiqIocVCRVBRFUZQ4qEgqiqIoCmLz/wGYqMDA/UZmcgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Let us see if we can make a prediction using the Glucose and Hemoglobin readings only\n",
    "# First, we do a scatter plot where we color every data point with the class (ckd/no ckd)\n",
    "# to see if there is a pattern that points to the success of the nearest neighbor classifier\n",
    "fig, ax = subplots(figsize = (5, 5))\n",
    "scatter = ax.scatter(ckd_test['Hemoglobin_in_std'], ckd_test['Glucose_in_std'], c = ckd_test['Class'], alpha = 0.5)\n",
    "ax.set_xlabel('Hemoglobin')\n",
    "ax.set_ylabel('Glucose')\n",
    "handles, labels = scatter.legend_elements()\n",
    "# You can define custom labels here if \"1\", \"2\", etc. are not descriptive enough\n",
    "custom_labels = [\"No CKD\", \"CKD\"] \n",
    "# Add the legend using the generated handles and custom labels\n",
    "ax.legend(handles = handles, labels=custom_labels, loc=\"upper right\", title=\"Classes\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78dcd976-d59b-449f-890d-448507244449",
   "metadata": {},
   "source": [
    "For a new observation, the nearest neighbor classifier based on Glucose and Hemoglobin, assigns the same label as that of its nearest neighbor (where the similarity is based on distance in the above 2-D graph) in the training set. Given this, a **decision boundary** gets defined in the graph, which is a boundary across which the class/label changes when classifying based on nearest neighbor. Try drawing the decision boundary for the above graph.\n",
    "\n",
    "Let's try to see what decision boundary we get when classifying based on Glucose and WBCC. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "752e3421-f718-4b6a-b91b-a298b3d0bb53",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = subplots(figsize = (5, 5))\n",
    "scatter = ax.scatter(ckd_test['WBCC_in_std'], ckd_test['Glucose_in_std'], c = ckd_test['Class'], alpha = 0.5)\n",
    "ax.set_xlabel('WBCC')\n",
    "ax.set_ylabel('Glucose')\n",
    "handles, labels = scatter.legend_elements()\n",
    "# You can define custom labels here if \"1\", \"2\", etc. are not descriptive enough\n",
    "custom_labels = [\"No CKD\", \"CKD\"] \n",
    "# Add the legend using the generated handles and custom labels\n",
    "ax.legend(handles = handles, labels=custom_labels, loc=\"upper right\", title=\"Classes\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afdb7122-f06f-4f40-a8ff-59a68468dd05",
   "metadata": {},
   "source": [
    "We see that there is a region of overlaps, which causes the decision boundary to be not as neat as when classifying based on Glucose and Hemoglobin. In such cases, the KNN classifier, where we take the majority vote from the k nearest neighbors, helps. More discussion on this to follow in the next class. Stay tuned..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "85e1e9d9-9b79-4e94-98d8-d9be566db627",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}