week3_clustering_강미경.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import silhouette_score\n",
    "import numpy as np\n",
    "import warnings\n",
    "warnings.filterwarnings(action='ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('Mall_Customers.csv') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 그래프 한글 깨짐 방지 폰트 설정\n",
    "plt.rc('font', family='Malgun Gothic')\n",
    "plt.rc('axes', unicode_minus=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>CustomerID</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Male</td>\n",
       "      <td>21</td>\n",
       "      <td>15</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>Female</td>\n",
       "      <td>20</td>\n",
       "      <td>16</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>Female</td>\n",
       "      <td>23</td>\n",
       "      <td>16</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>Female</td>\n",
       "      <td>31</td>\n",
       "      <td>17</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerID  Gender  Age  Annual Income (k$)  Spending Score (1-100)\n",
       "0           1    Male   19                  15                      39\n",
       "1           2    Male   21                  15                      81\n",
       "2           3  Female   20                  16                       6\n",
       "3           4  Female   23                  16                      77\n",
       "4           5  Female   31                  17                      40"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Preprocessing & EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#의미없는 CustomerID 변수를 제거하고 바로 데이터에 적용함.\n",
    "data.drop('CustomerID', axis=1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def encoding_sex(x):\n",
    "    if x['Gender'] == \"Male\":\n",
    "        return 1\n",
    "    else:\n",
    "        return 0\n",
    "#Male = 1 / Female = 0으로 인코딩"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>15</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>20</td>\n",
       "      <td>16</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>23</td>\n",
       "      <td>16</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>17</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Gender  Age  Annual Income (k$)  Spending Score (1-100)\n",
       "0       1   19                  15                      39\n",
       "1       1   21                  15                      81\n",
       "2       0   20                  16                       6\n",
       "3       0   23                  16                      77\n",
       "4       0   31                  17                      40"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#gender변수를 0/1로 인코딩하고 데이터를 업데이트\n",
    "data['Gender'] = data.apply(encoding_sex, axis=1)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "#Scailing\n",
    "scaler = MinMaxScaler()\n",
    "data_sc = scaler.fit_transform(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.019231</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.387755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.057692</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.816327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.038462</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.051020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.096154</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.775510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.397959</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Gender       Age  Annual Income (k$)  Spending Score (1-100)\n",
       "0     1.0  0.019231            0.000000                0.387755\n",
       "1     1.0  0.057692            0.000000                0.816327\n",
       "2     0.0  0.038462            0.008197                0.051020\n",
       "3     0.0  0.096154            0.008197                0.775510\n",
       "4     0.0  0.250000            0.016393                0.397959"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#data를 스케일링한 데이터로 업데이트\n",
    "data = pd.DataFrame(scaler.fit_transform(data_sc), columns = data.columns)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Gender의 분포를 확인해보자 *잘 안 보이는데 하늘색(56.00%)이 여자입니다.\n",
    "sex_count = data['Gender'].value_counts()\n",
    "sex_count.plot.pie(explode=[0,0.1],autopct='%1.2f%%',shadow=True, colors =['#e1f2fb', '#e9b2bc']) \n",
    "plt.ylabel('성별')\n",
    "plt.title('성별 비율')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#연속형 변수의 분포를 확인해보자. 다들 적절하게 분포되어 있으므로 이상치를 고려할 필요는 없어 보인다.\n",
    "data.hist(bins=20)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAFgCAYAAAB670TrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOy9eZykZXnv/b2frZbeZ+nZWUYZEQTEjLKIOBLHuBB3/UCi4nljGE1OJJ64nRwlBpOjBHM8vIlRUD8uMS/EfQE1goCDCgiKIOgIOjAbM9MzPb1U1/Zs1/vH/VR3dXd1d/VSVb3c38+npqqeqnqee6pmnuu5tt+lRASDwWAwGOaC1eoFGAwGg2HpYoyIwWAwGOaMMSIGg8FgmDPGiBgMBoNhzhgjYjAYDIY547R6AfXw0pe+VL7//e+3ehkGg8EwV1SrF9AoloQncvz48VYvwWAwGAw1WBJGxGAwGAyLE2NEDAaDwTBnjBExGAwGw5wxRsRgMBgMc8YYEYPBYDDMGWNEDAaDwTBnjBExGAwGw5wxRsRgMBgMc8YYEYPBYDDMGWNEDAaDwTBnGqKdpZRaC/w1EIvIB6u2twOfBjYBJ4C3iMhwI9bQKO7a08cNu/dyYKDAlp4suy7eyo7Te1u9LIPBYGgJjfJE/hkoA+6E7e8CviMiFwO3Ae9o0PEbwl17+rj624/SlyvRnXHpy5W4+tuPcteevlYvzWAwGFpCQ4yIiLwF2F3jpUuArySPvwZc0IjjN4obdu/FtRVZz0Epfe/aiht27x19z117+rj8xnu56No7uPzGe42BMRgMy5pm50RSIhIkj/uBnqneqJS6Uin1gFLqgWPHjjVndTNwYKBAxrXHbcu4NgcHCoDxVAwGw8qj2UYkVkpVjtkDTGkdRORGEdkuItvXrl3bnNXNwJaeLMUgGretGERs7skC9XkqBoPBsJxothG5D3hV8vh1wO1NPv682HXxVoJIKPghIvo+iIRdF28FZvZUDIuUx26Dz18K//csff/Yba1e0cwsxTUbliVNMSJKqWuVUh7wEeBKpdRdwB8An2vG8ReKHaf3cs0rz6S3I81QMaC3I801rzxztDprJk/FsAh57Db43rshdxTSPfr+e+9e3Cflpbhmw7JFiUir1zAj27dvlwceeKDVy5iRSk7EtRUZ16YYRASRjDM0hkXG5y/VJ2GvytD7BehYB2+9pXXrmo6luGaDGY9rmJmZPBXDImRwH7iZ8dvcDAzub8166mEprtmwbGlIs+FKZsfpvcZoLCW6T558VR8Uofuk1q1pJpbimg3LFuOJGFY2F14Fsa/DQSL6Pvb19sXKUlyzYdliPBHD8uOx2+Cn1+uwT/fJ+uS6bWft927bCXwsef9+fTU/3ftnoCmyOAu8ZoNhPpjEumF5UalcsjydJwiK+ir9ZR9r+EnWFFYYpsEk1g2GJcFPr9cGxMuCUvre8vT2BmOaTQ0rERPOqhOj3rtEGNyneyeqaVLl0oGBAt2Z8ZqjptnUsNwxnkgdGE2sJUT3yTqEVU2TKpdMs6lhJWKMSB2YMMUSooWVSzPJ4hgMyxETzqoDE6ZYQtRTuTSb6q2EesKZO07v5Rr0RcfBgQKbTdjTsAIw1Vl1cPmN99KXK5H1xmxuwQ/p7Uhz05Xnt2xdhjkwh+otU3VlWABMddZKpqVhinrVWo2qa33MoXprTuFM83sYVgjGiNRByzSx6lVrNaqu9TMH3alZS/yb38OwgjA5kTppiSZW9VUz6Hs/2V4deqn3fYY56U5t6clOCmdOW3Vlfg/DCsJ4IouZeq+aV7qq62xCR3Oo3pp1OHOZ/h537enj8hvv5aJr7+DyG+81Je4GwBiRxU29PQ8t7I1oObMNHW3bqZPoHeugNKjvZ5BEmXU4cxn+HqZXyjAVJpy1mLnwKn1C9BlfSTTxqrne9y1H5hI62rZz1mGlWYUzl+HvUV1cAJD1HAp+yA2795oKtRWO8UQWM/VeNc/h6nrZsJCho4WqqJrL77HIq7lmXVxgWDEYT2SxU+9V8xyurpcFCzWgqbp/pDosxhyN8Wx+j4U+dgOYdXGBYcVgPBHD0mahZE5aqP7b0mPXiZF0MUyFMSKGpc1ChfIG95GLXPYez7PnSI69x/PkIrc5FVVLoJqrZb1ShkWPCWcZlj4LEMob8DYydOwAgUpjW4ogiukfGiRcu4WemT8+P5bIzPSW9EoZFj3GEzEYgBvCS/EIyVAGETKU8Qi5Iby08Qc3M9MNSxjjiRiWPfUo8N5SPJON3h/yGv9bZKRIUWX4hvcqbi2eyfsbvcBZzEw3w9EMiw1jRAzLmmoF3uomuWtg3Mn30syjvHjghwyoHo6o9aQo82L/hzzVcwZwSeMXWkdIrt6/i8HQTEw4y7CsqVeBd5dzCz4ORVKgFEVS+Djscm5p0conY4ajGRYjxhNpMSY80VimGyhW/d1/3X+CTLaH2I/wwxjPsVjd1k2Hf7hFK5+MGY5mWIwYT6SFGD2ixjPV3PM2zx733R+il2Ihx9r2FKev72DrmjY67GBRVUiZGe6GxYgxIi3EhCcaz1RNckqpcd/9NzOvxyNkaHhoQSukFkr59q49fQzkyzzZn+fxozmGi75p+DMsCowRaSFGj6jxTNUklyuH4777X6S286m2d3BUuhZMf2yhPM3KfoJY2NydAQUHB0t4tmUa/gwtx+REWojRI2oOtZrktuye/N3/RJ3L4+sv4KYrz1+Q4y6U8u3E/XRmPAp+SHfWMwbE0HKMJ9JCjB5R62jGd79QnuZc9mMGSBmahTEiLcToEbWOZnz3C5UIn+1+TMGGoZmYcFaLMXpEraPR3/2ui7dy9bcfpeDr/EsxiObk7cx2P2aAlKGZGE/EYGgQC+XtzHY/pmDD0EyMJ2IwNJCF8nZmsx9TsGFoJsYTMRiWGaZgw9BMjCdiMNRgKcvR7Di9l2vQuZGDAwU2L7H1N4zHbkuUkvfpGS5TKCUbZocSkVavYUa2b98uDzzwQKuXYVghVKvlVieyTeXcEqZ6jr2b0UO/Yn/eDaWzQDXjIK3AhLMMhgkYOZplyBKYY79UMeGsVmFc60WLUctdhgzug/SEQceLbI79UsV4Iq2g4lrnjup/2Lmj+vljt7V6ZQaMWu6ypPtkHcKqZhHOsV+KGCPSCoxrvagx1U3LEDPHvmGYcFY1zQoxzcK1XspVQksVU920DJnFHHvD7DDVWRWaWb3x+Ut1CMurCo/4BS0//taxcaymSshgWDaY6qxlTzNDTHW61qZKyGAwLHaMEakwuE97INU0qnpj207t4XSsm3YAktFAMhgMix2TE6nQffLkEFMjqze27ZwxTGY0kAwGw2KnYZ6IUurDSqkfKaV+opQ6s2q7p5T6nFLqDqXUd5VSXY1aw6yYbfXGY7fp3Mb/PUvfN6A811QJGZYDZkDW8qYhRkQp9QJgnYi8ENgFXFf18kuBQyJyCfB14G2NWMOsqTPEBDStz8MMrTIsdcyArOVPo8JZLwFuAhCRR5RSq6peywGV+tY1wFO1dqCUuhK4EuCkk5rUEFRHiAkYn4QHfe8n2xe4kssMrTIsZcyArOVPo4xIL3Cs6nmolLJEJAZ+DHxQKfVrIAIurLUDEbkRuBF0iW+D1jk3jITCssT05Cw8RkJm+dOonMgQY94GQJwYEID/DXxMRM4A3kxiKJYURkJh2WHCLo3BSMgsfxplRO4GXg+glDoDOFj12snAkeRxH7ClQWtoHEZCYdlhenIagykOWf40Kpx1K/BypdTd6BzILqXUtcAHk9u/KaUswAXe06A1zItpQxvbdvLQwQ9g3/svrAoOc8LdQHT+eznHSCi0lPmEo0zYpTEYCZnlj5E9qcFMciNGjmTxMd/f5PIb753Uk1PwQ3o70tx05fmNXLphZWBkT1YSM4U2TOhj8THf38SEXQyGuWE61mtQHdoYLgYcHynjRzEHB4rctafPhD4WIfP9TXac3svrDw7ymR8/Qd6PaPNs3nbRqcazNBhmwHgiNahUlAwXA54aKhJGgkL7o1d/+1E6Uo6pOFlkzLcK6K49fXz1F4dY25Himes7WNuR4qu/OGSqswyGGTBGpAaV0MbRXEkHMpM/1nelcW2FiJjQxyJjvuEoE6I0GOaGCWfVoFJRsutLPycWIWVbrO1I0ZF2ERGGigEfftWzFkXFiWmQ01RXAb3oyOf4E7mFLEWsb7bD+X8JO9437efnGw6b+Du8/+kHOGf/Fxo/4MxgaDHGiEzBjtN7ec5JPVOq6C4GOZLqiqTqBrlroOVrawU7Tu9lx5HPwZGv6pkwytU9PLv/KXnD1IZkPorJE3+Hk0/8hDV330C+vZ22tiptNRow4MxgaDEmnDUNi71ix4RganDvJwAFtgNWco9Ktk/NfH7rib/D5eE3CXA4WrIaP+DM0BTCKGaw4FP0o5nfvMIwnsg0LPZGKVMlVoPyCFjjvxOUrbdPQSUUlS8HBJHgORan9XbU/VtP/B3WxUfJqXaiMB57k9FWq5/Hbktmobc2FCgi5P2IkVJIwQ8BWNuRAuzpP7jCMEZkBhZD2GoqzNCqGqTadQir+p+2RHp7DapDURu6MqNNirO5WJj4Oxy11tEV9WM7TRpwtpyojFmwvPFjFpoYCiyHEblSSL4cEsWLvxm71ZhwVqOoDK36p6fBR0+C606bdnjVTIN7ar2+2MNtLeH8vwQEohDiyn0E6e6aA8SqQ1G5UsiRoRJPDRZ5580P1l3eO/F3uMl5NS4h69Kx0VabLdVjFpoYCoxiXTBzcKDAoYEiw8XAGJA6MUakEVSupo7/HsrD+iRSHIATe2sOr5pJQXaq1wEztGoiO94HF79Xn3ziAGwX0p1gp2oOEKvMsa/uCbItyPth3Sq+E4eH7Vv1fI6/4B9pW71x5gFnhvEM7tOhv2oaGAos+CF9wyX2nyjQP1LGrw5BGurCaGc1gs9fqk9WuacgCsCyII71Ca1joz6pvPWW0bfPpNtkdJ3mQeW38KpCS35h9DeofLdHhkqEkWBZilgEx9J9QeY7bjIz/F4LQRDF5EohI6WQMJ6d0aiU+s8Bo51l0NQ1L7pyNRX52iUHUJZ+XuOqqnI1XE11gnym1w3TUHVlmyuF7D2eZ09/SN+Bx8aFBMthDEqIRRDRJwvzHbeABo1ZiGMhVwp4arDIgRMFBgv+rA2IoTbGiMyCugcXVYZW2Z7+jwAgsX5eI8E6k2SHGewzD5LfIlcKOTRUJIhispbPIXrHhQSznk0Yaw9kY3eajrRrvuNWsG2nDv11rFuQUGApiDiWK7P/RIFjuTKlwJToLjTGiMyCuvsyKldTXiejSV6JId1V86pqpgS5SaDPg+S3GBoeQomQoYxLyDczrx/97Xac3sv/e9m5bOrOsr4rTXvKMd9xK9m2U4eu/vphfT9LA1Lp6ThwosBTg0VypYB4CYTtlyrGiMyCusNKlaupNU+DVJeO72Z6YNXWmldVExOzExPkM71uqM1de/q4/K4O/mbkTRwMOulghAFrFZ/KvoNfpLaP++3Md7y0ERHyZV1dt/9EgRN5nyBaBOGqoAT5/lavoqGYxPosMAnupcPEIVW/OzZCGAmbezKjiVHz2y19yqFuBhxpUk9HXYn1oKibW/2cLqgBWPN0k1g3mLDSUmJi6HFdRxqAI0Ml89stceKkp+PQYJFDA0WGWt3TEUfaaIwcg4EnYegQlIbGDMgyx3Ssz4LFLoNiGGOiFElnxgWEI8NlhoqB+e2WIEU/IlcKyPsRLY2gxDEEBQhLyb3furUsAowRmSWLWQbFMEYtSRjHtnjOST0mfLWECKJ4NFzVshyHxKiwhAqLqDgG21R4VWOMiGFZsuvirVz97Ucp+CEZ1x6niWVY3IgII2VtOFqimjtqNBLDEZVRJJ6P4+mmYcMoxogYliWLJvTYIkXapTisrBSMCR82uyR31GAEhfFGwzAjpjrLYGgU1Yq0bkZX7cR+w3W0JlamVbywxViyXA4j8uWIfLPDVSKosIgV5FFBHiX1eTw9bR7t3hw8EVOdZTAYZk2LFGkX+7CychhxIq+bAQ8NFBksNLGnI/KxCsdxhp7EyR/G8ofrNiCG2phwlsHQKAb3aeXgahqlSFsVNnv3cCffyr6eB3nu6Mut1gFrmccB2usI8ljlIayo1NxjrwCMETEYGkX3yZMVaRsxnGrCIKd1uRO8Pf9JbkDxi9R2oDVaa34Yky+3sLIqjrD8IaxyDiVh84+/QjDhLIOhUTRIkXYSE8JmXZ1d+Di8uvjVpjdWBlW6VQcHCgw0M1SVoMISdv4ozvA+7NKAMSANxngiBkOj2LYT+FgSZtqvPZBGVGdNCJt1pB3o6iab62tKY2WlJDdXClunkiuCCkawysMmZNVkpjUiSqlTgDcB5wJZ4CngNuArIiYbZTDMyLadjS/prRE267ADOrZs4+63XtKww7ayJHeUONSGwyTIW8aURkQp9UHgJOAm4F+BPLAR2Al8RSn1P0Xkt01Z5WKmRX0ABsMoF16lcyI+40uJGzDTPY6FET9kuBi0dJSsCoo63xHkW7YGg2bKPhGl1PNF5CdTvOYCZ4jIQ41cXIVF2yfSoj4Ag2ESoxczjQmb+WHMcClgpNRKryPC8nNYfg4Vt0avyvSJTGa6cNZDSqn3ABcC3cAx4HbgcyISAE0xIIua6oQm6Hs/2W6MiKGZNChsli+HDJeC1siPAERlrKCACgom17FImc6IfBGd/3g70A/0AK8DPgFc2filLQGa2QdQJ0tR7sKwuIiSeeTDxbD5c8hFtARJkMcKC6g4aO7xDbNmOiOyWkQ+WfX8GPAppdQdDV7TZBarNEuz+gDqpFruonoG/DVgDIlhRlomtS6iPY2KBAkrYw7HcmG6PpFDSqnLlFKj82CVUq8Gmu9TRr6+ul9sg16a1QdQJ4td7sKwuKj0kBzLldnXn+fwUJGRctg0A6LCInbhGM7wPpzCEawgZwzIEmQ6T+TPgXcBtymlPCAG7gPe3IyFTSL09eQwdRxSHZDqBDfdkqWM0qw+gDqZOIgJWi93YVhcaMOh5UcKftTcJLnEicdRQIUFU5K7TJjSiIhIXin1T8B30PmQY8BvRKS1lwoiUBrWN8fTxiTVCVaLmu+b0QdQJ7UGMbVC7sKw+CgF0agESVNHyUaBzm0EeZ3rMBLry47p+kTeCLwfuBc4gTYkZyql/k5EftSk9U1P6EN4HAr94GZ1bsJrB8ue+bPLEDOIyVAhinWoqhhEFP2oqYZDJ8aTHEeLSnENzWO6cNa7gItEZDQWopTKArcAjWuDnQsi4Of1jWO6QirVvuIMyqIZxGRoSRNqKTEYhSCi3Ez5EYl181+QN2GqFch0RiRAdz1UU0bLnyxugqK+qePaoHjt4LWtCINiZsAvAiao6pI7qp+zsE2ocSwUgkh7HE32NnT/RlFPAwyLJky1gpnOiPwzcKdS6nZ0PmQV8EfAp5uxsAWhUjHlF/RQICetjYnXDrbRnjQ0iAY3oZbDiOFi2NRKKkikRoKRWU0CNCx/pkusf0spdRtwPtALPAF8UkT6m7W4BUVkzEPJHwcnpQ1KqgPsOcgYGAxT0YAmVBEh70cMF4PmKeXOcYSsYWUx7eV4kg9pfnNhMwjL+lY4oY2I16b/o7vJKFODYa4sUBOqiFAMIgp+RKEcNb57XGJUWEZFRX1vwlSGOpiuOuslU70mIj9ozHJaRBRAcVDfKmGvikFpVC9KgxKvRvZkETAPVd0giin4OkFeDBrcOR6HiadRREVlU0llmBMzNRvmgH0TtguwvIxINdVhL05oo+JmxxL0C5FLaVDi1cieLBJm2YRaGSOb98PGy6tHfpLXKBpBQ8OCMN0Z8YPA+0Xk75u1mEVJdflwJZeSage3TTc7zoUGJV6rZU8Asp5DwQ+5YfdeY0SazQxNqJXO8aYo5I5KqI+g4nJjj2VYcUyXWN+jlLq6mYtZSAaLAUU/xHMsPMfCtSxsW+HOt7O9kkuhX3slXrvumJ+NQWmQ+u+0sieV8FnfbyAOwE7B2mc0VaZlJYfaRIRyGFP0I0phRCmIGxuqqiTF/VwiamhyG4bGMFNiffSsppTqEJFcvTtWSn0YuDg5xpUi8mjVa/8N2AVEwNUi8sPZLrwewjgm9GMK40K9CsdSuI6FrRSuo3CUhWMrnNkamCgcy6W4Gch015eYb5D671SyJ6/IPArf+wwEPpSHE1XkApzY25D+hVqsxFBbGMW6j6McUQqaoFM1muPQ8zeMmKGhGczmrPmtet+olHoBsE5EXog2FtdVvXYm8ALgQhF5fqMMyNQIYRxT9ENGygEDeZ9jIyUODxU5MFDgyHCR/nyZXCmgHM4izBAUYfiwPjEPH55ecbhB6r+7Lt5KEGm5i4pCaxAJu5xbdPjMHwaU9qCUpddoedpDaTArRWG4HEYMFnwODRbZf6LA8VyZgt+4aYAqqCjh7scd3odT6NM5D2NADE1iNlni2dS9vgQ9mx0ReUQptarqtT9DJ+vvUEr1AX8hIsdnse8GIgSR6AqZZItjWWQ8h6xn49l12NzqHIo6DukuSHePT8g3SP13KtmTnu8/pcNnka+NB+j7yK8ZRquEnV505HP8idxCliJWqh3O/0vY8b45ra0uheEpKtYWexgsjGJGyiG5UkgQNeHkHQVYwUgyJtYMbTK0ltkYkdn0i/Siu9wrhEopK1EAPg34vojsUEq9Afg74K8m7kApdSXJBMWTNm+cxaEXljCOyZV8ciXwHJu2lE3WdbDq6SUR0aGu0pAOc1X0vJRqmPpvTdmTe5Pwme3pcmalQGL9fEIYrRJ2emvwZd4Sf4UYKGPhlQvYu/8pOcjsDcmMCsNTVKw9dPADXP3AmkUXBqs0/+WaNDrWiBoaFit1h7NE5MOz2O8QWvW3QlwlIR8C300e3wKcMcXxbhSR7SKyfe3qVbXe0nT8MGIg7/PUoA551R3uqngnuaM63JU7og1L2KSTQSV85nUConM5EmsvaUIYrRJ2ekP4LWIgxkFhEWIBCu79xJyWMGWoraIwXF2xppS+tzzse/9l0YTByqGupurLldh/okDfcKlxBkQE5ed1qGroSZyRQ9jlAWNADIuORglI3Q28HrhbKXUGcLDqtXuAl6Nnte8AHm7QGhqGoE+CBT/EsSyynkPGs3FthZop6icC5RF9g6TCq0PLr8y1ZHgmqsNnfYE2HHYKVm2dFEarhJ0yUiSs+ucRC1rAsrLuWVAJR+XLAUEkeI7Fab0d48NSU1SsrQoOk2kfL5w5MQzWqHBXEMUUg4hS0vjXcIHD0dkbBdMtblgyTNex/mmgbeJmQETkT2bY763Ay5VSd6MbFncppa5F9578G/C5JJQ1BPw/c138YiCMY4ZLPsNJ35alFLalsC0LS4FtWdgW2KPb9W3U2EQhFAf0bXTIVsfCKw7XGT6rhJ2KKkNKSsTJPxFLARLpkNwsqK7K2tCVGTfjZNyJfoqKtRPuBopBNGUYbCGrvuJYy4yUEqmRhuc3JNZhqkpFlfEyDEuQ6TyR36BDT3VXZVVIQlfvmLC5Ekj3gTfMdp9LhViEOEnOT4drW3i2jecq0o6ty4urh2x5bWOKw03U8qoMtvqK8yreEvwnEBJhoVPiopPrs6DuBsgppEKi899L8IBMOWhrPg2WcSyUwijp3YibM4MjKo+V4EZl420YljzTGZEbgQ+LyETZE8MCEEQxQRSTTy4+Hcsi5di4jm6IdOMcdnkEVN9YUt5ta/gY4LEKrysIj8Szqs6qFVaqe+77FBVr52zbyTUb+8ZVnF2wdRU37N7LB771CMdyZdZ3pmbef0I5MRoFP6IcNrjhDyYo4RZQEjb2eAZDk1HNnEcwV7Y/+yx54PZvzOozg8WAXGlphwdc28JzbFxLgQJV0fFKtaPcNizbQSmwLYVbT/lxA6kOK1V7DG2ejR/F48JRBT+ktyPNTVeeP+/j/O7YCGEkbO7J0JF2J+3fD2PKSYd40W+CEi4kk/4KY4bD9GwsG3raPNq9OYyOWPP0ZSsNPl1O5LPAP4rI3gnbFfBmoCgiX2nw+lY0FW9lPGVgAEEhTobY60DcNizLIuVaeLZFytU9LZ7TPMMyVVhJREarshZi7vvE46zrSHNosMjhwSJtvTaFICIIhTdu38y+/nxzpv2J6NBUWEBVFHFNmMqwQpgunPUe4D1Kqc3ono8CsA5IA18Wke80YX2GKVAIKixghQUEi9jroJjqpGh7UNQNaEolEi+2lnWp6Ic5ltIeTJLsV0rVJU0/XRXUVGGroWLAh1/1rPrmvtexhv0n8lxiP8Trh77BuugIR6x1fCH7Sm4pPov+vM/6zgyXPXcL52zpbqgBqSTE9a1svA3DiqWucJZSqgs9W71fRJoeI1qp4ay5EDtZ4lQ3Yru6K13N7I1k99/J6h/9LWJ7iJPRMfzIZ+iSj+KfegmWUtzzu+Nc94PHRsNIpSAiiIWrX3EGFz9jLVd89mf0jYw1Eyp0WGldZ4abrjwfEUEELGsKr76q2VDcDHFQgNAnd8lHGDnpRcQxRCJ88Quf5m25fyNULmVSpCjjSMBnOv6CN7/lbQv4TU4gjrTRrpTfmil/KxITzppMXX0iIjKELsc1LHKsxDupJrbTiNdO7LbXLB3u/MW/IZaHOLpsVpwssUDm/n9laNMLAfj8T/dhKfBsiygWXNsijCP+7a7f8/R17bzm3E1cf8fjhFFA2rUoBTFhLLzq2Rt54nh+XALbthSWUliWQqd7FD27P46FQ2ynIRKwMigleD/7V/yNLxz97NvdWyngUCaFpaAoKVIIb3dvJc8CGpHIR0V+EqYqmdkbBsMUNKrZ0NBiHj4wxK2/OkzfSJne9hSvOGsDZ23p1lf6dnJz0oiTxh0+gGDh5H+PinzE9ojSa3CHDwCQ2XcHH+i/lk30ccRfx1e81/KA+wekXYsjw0UAnrd1FR/pO8Kqhz/FmuIRjrvrOXH22znp1PO47/f93Hz/AQ4PF9nQmeHcLV08eGBo9Pllz93C64b2I8rGGalaQ2bt6BoqrAoO43Z0cqKoGxddW9GV6aQjOEx+tl9SxVDEASoOYfQ+NDmNFU4QCSN+TM6PGfGFkXJMPoiJKbF9UxvPWmrgGvYAACAASURBVNegiadLEGNEliEPHxjiC/fsw7Gg3bMZLPh84Z59XAGcvaVLDyZKdPtEOUR2mtTQXkTZgIWKQ9z8Icrdp5HZdwdrdv8vQitmKGpnVTzAfy/dwL+yi7t5Nus7M4A2NM9//KNIm4c461gTFlCPf5SfANc/sh7HUnSmHQ4O5Hn40CCr21y6sx79+TLX3/E4f+Rl6Mrt1Z6S5aDiECd3kGDVaeP+bkHnFtrzfbT1jDUlqqBA0LZl5i8m8lFRoMN1YcGIFy5TRIRyJIyUhZEgZqQcM1IxBv74x7nEOIyUk9eSx+Vo6ouId1+EMSJV1GVElFIeWtr9wIxvNrScW391GMeClKNDVynHhjDi1l8dHn292kPZSaRHjChJxBkFiQUVB3Q9cD2CTVs2Qz7nUxQPlPC68te40z2Hy56rT95dD34SohC72D/qScReF6se/hSO8/dkXL2WvB9hKRgpR/Rkx8qBB/I+3UrPXo6jmEhiLIRDA0V+tvcEz9uq9dOGzn0Ha3b/LwgYzd+o2Gfo3Am9rXGkk99RKbk3FVNLBRGhGMioF5Af9QjGvIKJRkG/Z+z5Qk8ZthW0exadaYuMu2zTG3NiRiOilLoCXdLbq5R6HvA/ReTvGr4yw5zpGynT7o3PfXiOxaGBQk0PZYc1Am3rscsnUFGA2C5RZq2WGw9GEK+TdsBqsxgqBfiRzRaO8jcv2MBzTm4DEbwTj2GVhnQzZOJJ2IVjbJI86cxYcj+IYpRiXOly2rVIFwsEHZtQhWOI+IQ4DLCKtBS4/o7HuYrTeN7WVRRPvoTjF/8jXQ9+Enf4AEHHZobO+XNKG87DKg1qYxH5RkKkhUSxUAiqr/pjcsmVfj45+edGH489H/HjJGwkLHRhnWtBm2fRkbJo9xTtrn7c5lm0eYoOz6J9wmP9Xr0t4+gqxjkn1pcx9XgibxORFyil7hSRklLqvIavyjAvettTDBb8UU8EwA9jglhoq+GhHIh62OoViTqqJiuGRcKMLsO1i/3gZMh6jq6+CotEmc2cty6AnNbWVGEJ7UeoZHKiAgSPgFIQj3oirm0RRPG45shSEHPcXc8au8A+tZFQCZZSpKTEMasHT8V842ePc/7GZ6DiAH/16Ry/5P8kuQttjJzCkYZ9nyuNMJYaV/3jr/xzvg775JNwUS7xGEZ8bQQWmpSt6Egp2pMTvL4p2ion/ORx5cTf7qnEAFh0eBYpx3gPjaIeIyJKKRtGYwFdDVyPYQF4xVkb+MI9+yCM8BwLP9TuvVujAdFzLL4WvIj3xt9CQsBJQ1hCxQG5My4HoOf+j0/5WgVRLkiJKAqJUVgINuA6LmujI4SSxnFdet2Q/jBmleeQEcEPIjJxzNCZV6B+/3GcMAJSeOLjEfID5w851e4jPxxhl9Y05ftb6pTD8V5ArVxA7TxBTN4XiuHCG4E2t3LC11f6bZUTfGrseUdy1d8+4XG7Z+HaxggsVuoxItcBPwCerpT6DvDZxi7JMF/O3tLFFZzMrb86zLGRMmuT3Metvzpc00M51PUcBs7ZSsevb8LJHyZs20DujMspb9KyJAO8a8rXKgxmNiP+ftoo4hARYjNEBtV2En92zuZkLSOc2ZXimc/o4DdHchwbGWJTsrZTtpzCQPc7Kf34s/SEx+i31vA99yXscZ6BH0asbR+vj7VcERFKoUw6uVe8gurcQK1cQT6IWegRJ5bSRqDds2hPTT7Bt3vjPYSKx1AxDFlXN7Ualif1GJFfAq8FtgFPLJ5RtobpOHtLF2dvmew01vJQXnHWBsqbTp9kGCqUN50/5WsVbgp38Eb+PwbowcfDw8ch5MvhDl5dYy2vnuI4v7/wmaN5G73GaHSNS4FYknxAuRL2karY/2SvIFceSwxXXpumMGhO2IpxYZ62CSf/iifQ7qlxeYBKbiDjqvomeRpWJPUYkS+JyAuB+xu9GEPjqPSNlIKQMGkW3NSd4RVnbZh0gq/VY1LLIFXzI38bOfsNvDy4nXWqn6Oymu+6L+ZBf1tNgwGQOnTvJA/n7C3n8zf9x+j67c30lI8x4Kxl6BmXsXHL2Qv0TUxPGOuTfn5c8rcq8etXlYPW8Aryvix4DZhnM+5Kf3LyV41e9bdX5wZc/d50khQ2GBpBPUbkXqXUPwA/Rc8XQUR+0NBVGRaU6r6RVW3eOA+klgGp3WNy8rSGJOtY/HDoNO5kW6VKmDgUNnXVll1JHbpX51osF/E6sYv99Nz/cUb6X8ZznvweknHBWUNPWEQ9eSMDa9tm9IYA/Kg6zDPmBegr/rEKoZxfXS00ZiiKDUgKZxw1zguonPjH5QaqE8Epi3Z3zGMwSWHDYqYeI1LR0Hhuci/oHIlhiTBd38hEwzCb946jcqWrkhskxVq1T4Adv74JsVxwdLMiToY4APc33+CIt5GcdJPz0wzHaXKhxdB9v+bI086YV5PYXFAwGv+vPtlXewUVj6DNVVVhozGj4Zh8gGEZM6MREZG/B1BKrRaR/sYvybDQTNU3cmykPPq8EsL67dEcrqXozno8V+3hZcEPWBMf52h5NalDfz6lN1AIIi5JP84lwW4yqsi+eAM/dC/iN8Wt/Nfv8mPlosnJPjyyk2E6GJY0uThNLrkPmGYs8APDs/67V5rEZkoEj4WDql5P6aSwyQcYDFNTT7PhTuCfgCeVUr3AX4uIyY8sIabqG6lUPFVCWHYyE74YKU4e+T3PtO/nIbZwQp6JrzzkR4/S19vFkLdutBy0kisYGOniF3IRn+bisQMnQ/z+4UcDNVb17LrW7hLSqYp02AGZnnVzbhIzGAyNoZ5w1oeAHSIylEjCfwV4SUNXZZgTtZrEcn5MT89qfnUiGWSlLPwIglixWmW44utHeWq4TBB1EzGWv9gnvfxX/NzxBwiAfTAW4aym9j+lND7t2fSkJrHO4Bhrj91Hh+3TYUd0yjBdjGBvfjbrD99Bux3S6UIqGkHFAQPPfRflTesW6qsyGAwLRD1GpJxIwZMYEiPa2CDKYXUH8IRy0HLMSFBbN6jyvDRtk1h20pbBwQiIgKlnjmQp0UaRdlWmxy7RrUZwTnoe7ROaxDb/8uO4SlDBCOm4SLdTZmM6oFuNcOTVX66x59U89ctDugrLr6rCevYf8tQvu5Df3oyTP8YhZy2/XX85//VQN30/eajuajGDwdAc6jEIQ0qp54vIT5RSFwLFRi9qEoV+ePjL4KT0zfZ09/To89SE5x5Ic2eOi+hO33wl8Vtd+z+hSaxWeeiI3/gmsXZPJ3/DMOSJYyO4lpCyYaTkg0Ssyji0uQpHCe8s/ivr42OI7WEpiAU8ymS61lHc8dJJx1qz/5iWR2nPjG0MfcJM7f6Ohw8M8YXH1uLYf4WXSnpWHoMXxIe4+3dj24eLAUN7Q7ozRTrSbt3VYgbDUkUptQqdQjgDfZUH8EYROTzH/b0VSIvIpxZmheOpx4i8A7hOKXUdsB/Y1YiFTMvIUdh97aw+0g10WR7ieIidmnCrzNMY2xbbHr5KURaXEh5FcSmIRyF2GYldRiKXXHIbilwGA5fB0GHAd+gPXPp9h6K4jJUmzZ9aTWK1kr/VVUGV3EBH0iRWKx/wke/uwcqO5UgKdsTxfJmo7NOeSeOHMd+JLuAq52tEgC8eaXwcIm4Kd9Ts+8idcXld8igVpqoC+96jR+jOuKPbi0GUTEmM6Ex79VeLGQxLEKWUC3wb+ICI3JVsW9RyDfUYkQ7gLSJS0dDa3OA1TcZJQ8+pEJYhKuv7sARxOO3HVOyjfB8Yaez6FJCCWBRltBEq41LGw1ceofIIrRSR5RFbKWI7hXI8cFIoJ4XlprDdFLabxk2l8bwUrpfC9dLjDJ1+XBkqpbdhuVOW0U7FxGqtrGezBo8ThYC8ryVGfu0/i//wPF4W3Maa+DjHrTV8U13CnYOn8qMvTw4rlTedX5c8SvUaLrD28LLibWyMn0qkUhz2Si938lIe4VmAHg6kVX/HQnUTK8sMhmXEq4HbKwYEQETKAEqpDwEvQp9x3iUiP1dK3QV8D3gpkAH+WESOKaVeD7wHGASOAT9O9vHKZLsFfEZEPpfsNwtcALxJRPbNZsH1GJHPichFyV8mUkrdCPzRbA4yb1ZthT/96uTtcQSRrw1KWB41Mh/54QEePjRECp80AWl8Ukrfj96UT2r0tcr2YPS1ND4ZFZCu+pyXXI1PhaWEDD4ZJsiQC9opbcBYbkFNMDCpcUZGvzb++S5rhOGSTWylCJRLgEtBbJzuDK/evhWxA754/xH6S+18KXUZPh7DgcXhguCoiHYvXTOsVI88SoUXeo/xxvzN2ERkKaIVf8tssvp5U/lmvsTlPOKciWur0QmGFaorywyGZcbTgIdhNKz1dXRg5RtAt4i8MNn+ReDS5DO/FJFrlVL/A7hMKfXvwP9AF0QVlVKfSPbXDfw1cAm6dvJ2pdRNyT5CEXnBXBZcjxGZeLnfNpcDNQTLBisDbmbc5v7ODu49MDyuSay6Q7giINdRlSewUxYpz8L2LNJJf4BjKSIgn9wALT9ePX87KqPC8tjj0bncE55Xv17ztYmfK48dR6a2PgpBRSWYxQzwnZUHE3cbAj/SD99d2VY9/M9LPpa38HHxcQl/7NHZ1oY4KVTkYxVPoOIAcTIE3VsJ2zdOMHD69tb46/gEtFNAIQgKBXSqEfJxml2lT1NQGfrsVXwp2sHPozPZP1DAVoqMZ/On5y0NLS2DYZY8CTwdQEROADsST0GAP0w8D2BcQ9Xu5P43wPPQOof3i0glf/0AkEq2nwbclmxfA1RKHn861wXXK3vyEeA/gZcBv5vrwZrFP+xcx9++cDX5cgPGn1oOYjmIO7naqWHEYQ0DU0aFNQxU9WuxP/kzyWuFQp7hkRGsyCdjhXQ4EZ4Kx94/jQKUTUyGMhnKEAO5E5PfFJWxj/4Cjv5iVn9VF8hYOvzYTpleBvmot5dIFCU8bbzEo+3+LN6vMuONk5Oa7IlN2jbxtWrj5o37DKq5xRkGA3AL8EOl1PdF5OFkmw3sAb4sIh8GUEpVn4Ck6l4BB4HtSilHREJgB3AP8ATay7k0SU9kRaSQ5E2nzw1MQz0d6+9XSr0Z+DPgt8CVcz1Ys0g7FqVgGTWYjRquhXUCKz9+ObmNIsJ1332EfKFAux3hETA8ksOKfdqskHVtFp5og9XtCZeeuYr23/wnlj+CshyQGBBIpiT6a581ydtyRg6h4ggSL2umX8tWQhtl2iorLZ5oeJ2gWO6YcXFSiDW+GGOcYXLS0xqkmoZs3GspxHaN4VrhiMiIUuoy4GNKqbXo/5rDwGfQXsmPgRzwOaBW7Twi8pRS6uvA/UqpI8DjyfZjSqlvAvcopYaB+4APznfNSqT2FWfSnX5cROLk+dloV+hbiXVrGtuffZY8cPs3ZvWZwWJArrRyR6TORYm3mnd9+SFsYLgU6OQ2UBlou6UnMyrieMUFOiey/ptvRLDHj9hNrUIR1ewTSR26l+w9/4xfLtNOThseBRZCjGKIHsq4RLHgWUKGIl9M/SkeAa74xGGZNzx7beI5+ZM9sVHPrca22B/11BYbMouKwsr2uKZBGv9+EsMVT8iVYXmzLsxYycx5PO6apy/bL3k6T+R7wHaAZLb6B4GfoasA/rzxSzPMlbkq8VaTdSwODZWw0NpRIroXxrPVaAVXtWESN4szuC+ZsW6j4hAnf4Sw++Sa+78/fiYPha/llep2NkmIQ0ggDp6KKJFiRLUTixApwcbnKWsjDzpaKqUcRnR3ebz0mafP74uSeJr81nQ5rzreH0+R+1osFYWVrwA1IcQ3RXHGJO9puuKN6UKGKbAcY7iWEdMZkSEZc1P+F3CZiOSVUrdN8xnDImDOSrzV1FDlVUBvZ5p/fM2zJr1dRH9EKk+SXUzh6Oopi86ZPO7oOSEFX/eqnGft4Srn67hSoozHGi/GDyK+Zf0hgoyTsZ83ytJhKCc9/33VSxxNbWAmbZuq2GKG1yZum7EwIymdbxKirOnzVFN5YzVK3Gsbq8k5MSwjtNEopvtm/aQk7MXAbhGpFCi1N35ZhvlQj2rvxIFQpXXnkj764OjzbaXzKLedORrOcm1FZ9qjGNQ+IVlhgdDtxPYHR6utQq8bOyzgWBaxCL/cP8h3f3WYvhGfoYLP6jZPj01V0JFxsBQ8kH8md6rz+GO5kwxFVGAROBleyx18rQSHu5/DK8/ZxDkndRFFQhjHNdezaLFsxMogTmbm9y4UoxWFszE+EwxWrdcmeVxVn5GpfxclMSosQtg88QtR9jQFFzW8rlq5LCeFk81Cqm2CWkaq9vMVYrim+1teje6c/B3wNgCl1EnozL9hETOTau/EgVDO8EG6+x4izKxB0t3YxX6ulK/yWVE81jk2UbAcRqzOepOOZysLcbO4hWPJfxyFQnCDYWg/lQ1dGe7b289//Gw/rq3oybrkSj7HRnxsW9GW0jFmz7V43Zr9XCYPQdgBpRBE8KIyz8gO8Lf2t+GCZ8IpY0KMghDGQhyLDn+JEMcQRDFhLESREE1zQlsRNLuiUE8kq10dOFNFYa3tNT9XFVoMS1gzeFJKoqYbLixHGxU3De/d27zjNpkpjYiI/Ayqdb1BRPYDb2j0ogzz4xVnbZhyljpMHghlhSOIsrCCHFG6B5wMmXTMpcXb+Yh9Jp5jEUQxscAb/mAzPW0ejrKwbd1Lo1BgV3k+1XGsJCx28wMHcG1F2tX/5Fa3pTg6XOZ4rkzWcyiHEUEkXJa6E2IXgkQ+3rYhjsEfgbZeePALcMrzxw6FwrXUdBqSo4YmCLVhCaI4uQks+DBbA0qB7SK2u+AVhVMiAnFQw8BMNGLTFVvM/Jod+9pghXWEAONQ34L89O9b4qwMf2uFcfaWLq7gZG791WGOjZTHJcEfPjDEecf2MyQZHLtIV9qlMwpAWagoAAVKKTKZNp4uA6xtT9M3XGJ9d4bLtm/hvK2rax/UH4GODbr0NgrAdiGzCnz9H+jIkBZQrNCedgHh+IjPSClgXZfef9fdfZDq0vsYzctY+rmThuHZa9BVDI3rTbY0kQhRHBPH+nEsQhSjPRuEKDE+Cz853bCgKJVUoHkIHQ07zLjqLJEqGSZfN/xWlDMq0kyV15cxxogsU87e0jXl/PSTWUUPQ5SiFP15nzbbxYkDlOPhVeRFgjLO6i1c/5pz6ztg50bI90P3SWPbghK0aaOzvivDiXx51BMBcGyLMzZ28fE3Vg2oeijZj+2OGRKJ9fOwBJ0L26luK4Vt20w3UBEgiGP8UHsvfhQRhDHxVFUDhpWBUol6eOMKM055/60vRWtdnYpuFrzuyY++4vvz2adS6sPoKJMDXCkijybb24FPA5uAE2jNxBnHiZrOphXErb86jGPDbak/wlUxGeWjEHJRWneoe+06uhOUIA7g3Cvq3/m5V+jPBKWa+7hs+xaCSCgFISL6PoiEy7Zvqb0fN6nfiCJ9xee1z35NC4hrWbR5Dt0Zl972NJu6s2zqzrKuM01Pm0eb5+JY5r+TYeFIDMgngA3ok/oG4BPJ9jmhlHoBsE5EXohWZL+u6uV3Ad8RkYvR0ijvqGef9YzH3TZhUwk4IFN1KRoWFbaySLs2KdfiRN6nM+OyR53NzZZiZ1nPT3+KNax67hvh0AM6XNS5QZ+sq3IPM3LK8/nNkXdgP/QluoI+htxeonPexjOTfZy3dTVXoXMjR4eKo+GrSeGxU54PvE/nPvoDbThsD7q3zH5NDcZSCs+28Wx7VFcsFp1z8aOYKIYojolE8AMTEjPMmvegO9Yro0QLVdvn6o28BLgJQEQeScQcK1wCfDR5/DWgrvkj9YSz/gUt1PUIcD7we6BXKfV+Ebm9zoUbmojn2GRcm7Rr6RNcwobusZDSo95ZPOqdRSkIWdWW4uPPezbzUbS5b28/1z/Sg+u9k1TW1onyR4SrevtHDcV5W1dPnVOp5pTnLypjMRsspUg59rjKuAp+ktD3w5ggjk1IzDATp6I9kGoKyfa50ouWhq8QKqWsRJkkJSIVwcF+oKeeHdZjRPJowa4gEf36LHAZWijMGJFFgufYZD2brOdgT9ENfNn2LVx/x+NASMqxxyqiJoaU5sDE6it9H3LzAwfqMxwrAM+28GyLtqoq6UiEMNJGJYwgTIzLkut/MTSCJ9AhrELVtmyyfa4MMd44xBVpKyCuMig9jDc2U1KPEemtWKdE8fEkERlWSjVVP8swGc+xybo2ac/GrSMeX3dIaQ5MrL4C3Sl/dGgOdflP/kSHs4afgs6N/GbDa7hx/waODBVZ35Xh3C1dPHhgaPT5Qv0dWoGtFLZjk5qQ2a9052uPRfCjGD9swEAaw2LmOnROBLQhyaIl3a+b8hMzczfweuBupdQZjO/7uw94FXp2yeuo00mox4j8RCl1A/BddEa/YgXNpVLTUaRdm4xnkXbsOSVy6w4pzZJa1VflMGJd1yw7s5/8iR6FbLmQ6qIw2MeqQx9nk3cZ+czZHBwo8PDBIVa3uXRnPU7ky1x/x+NcBUvWkNRCkYTFGJ9vKQURfqTLksNYezEmJLY8efKjr/j+Ke+/9S9Z2OqsW4GXK6XuRqsB71JKXYvWRvwI8O9KqavQTeZ/Wc8Op1TxHfcmpV4MnIuesf5VdP1Nr4gcmcvfYrasZBVfSykyrkPa1Qlyq07hui/e8wRf+flBCn5M1rN4wx9s5i0XzC2Uet/efm5+4MC0V/737e3nf//X4wyWIYjBsmyyKZe/++Mzef5pa9ESvZY2DrYLlsNdvz3ODXc/wca+3fw39R22qGPY/jB5PEpON1nPYrDg48VlBlUn12XfxYm8PyrDsmWVbmQbzetUlwqvIGLR8i9+ZLyWRmNUfCdTb5/IYXSVgAKeLyK7gaYYkJWIa1ukXYeMa9VM0M7EF+95gi/csx9LgWNBKYj5wj37AWZnSCyLnz45wnV3HNT5jkwHB/Lw4TuP8P70ai56xvpkuqRDsaODQ2qYERVq5V2laFcOQXo1tK2ZtOu79vRx9Xd+zYXxL7gq/iyl2GZ/nOJ0q4RHiSNhiiN+BrDxVYbVDPD4sIsrMVkrGDdzfc5hs2VCrSqxSIRiEFIKYkp+ZCrDDA2jnhLf/0CPxP1NskkYG8doWCAcy6I95dSd35iOr/z8YGJA9MWPpSCMha/8/ODURsSytNaP5eqSWq8N3DT/8tV7GXZXk/UcPQPKhqIf8ol7j3PR2U8f/fgNdz9BV8ZlQ1X4quCH3LB7LztO7510uBt278W1FZf73yTAIS8eIJTFxVMBq2SQITKAIiUBB1nPoOokiGNsUWTtmC4rjS0RflBmfbcNjqcbFE14Rxtxz6XdA2nTYbBSoHMsptzYsJDU44lsFZELGr6SFYrn2HSkHLLewokHFPwYZ4IdshTkfYED98MvPgfDh6DrJLjwnfCMl0453+HAQIHuzHj3PePaHBwoTHrfJfZDvG7o66yLj3LUWsfX0q/lzoFzJu/0sdt495EPsYlj9MQnOK7WUAmr9kkXm+jHIwCEDD6eCvlMdCkgWiJJIJtJUyJFMYwIxOOqF50J3b3agISl5FaRnljZNSAKHRKt/hmD0SowGS07NhVhhrlQz5nrQaXUahHpb/hqVgSKlGuRcXUvRyO6nLOeRSmICZVDkTRlXIpi80Lv17D7Jj3NLrMa8sfhv96vPZBtO2vua0tPlr5caZyRKwYRm3vGK8JemnmUtwx8klC55GinJz7B2/OfJNvzV+gepoTHboPvvZt1KmIobqOLQXqlj0CtZVgyjJCln4CslOgiz1P08rXUa7mn9CyiMKYj7fC2i07lnr0nODhQYHNPll0Xbx3zdpQCN6NvFeJovFEJiiveW3Eta5KWWCy6IsyPYsphTDkwYTDDzNRjRJ4LPK6U+m3yXETkwgauadlR6RqfbXJ81iQn0FecfzYf332IONYzOuLk6v1vu27TBsRLDICXBR/46fVTGpFdF2/l6m8/SsEPybg2xUD3luy6eOv49zm3MIRDmRSWUhQlRQphl3ML8Fdjb/zp9WB5dHV6jAwVOS49rOcYvZxgmI1k8QlxeFf437kzfjaerdjW3sH6lD7uNa88kx2n9/LO2Xwvlq3Dc16iKBvH4OegnNPyLAZA51b0v9OxPFw5jPDDmHKkjYqpBDNMZEYjIiLPncuOpxL5qnp9HbpkbZWILNr/yXOdVa5QZFMObV7t7uUFQalkAE5m7OpbKd7+0o34Tgef+fET5P2INs/mbRedypZH+sCd0ITqZmBw/5SH2HF6L9egcxg1r/wTevyncLq6OZb38cMYz7FY3dZNhz9BdXdwH6R76FCKnsDj+IgQxqtZrwZYpfIcopcbokv5mfMc1noWw6WIoWIw7rh37enjht17OTBQYMsU65kWy4J0F3c9WebGH+3j+MAAT+tSXLG9l/NPratJd8VQ6b6v6OL6UYwfRZQDEwIzaKY0IkqpZyXaKi+Z+JqI/GC6nVaLfCmlnoVujnn5hLe9Hzg+hzU3jbnMKncsi46MQ9Z1GuNxKAVuFlLt4LbpE2IN3vnibbzzxRNkzw6eDLmjY54I6NBOtfJuDXac3jvzSbr7ZDpyR+lYUzU/wi9Ax0mT3kfuKLnYY6Dg41gK2/J4KNzGm8IPsLknQ0faZSs6Mf/03jQ3XXn+6Mfv2tPH1d9+FNdWdGdc+nIlrv72o1yTrLNeqveTyXbxWCHivXeO8A+ZdVx8apuWto+CmXe0wqh03bcnVWBhrENflTCYkc1fYD7UNUnFlw8NzVfFdy3w1+hu9Q9WbV9wFd+KBvgFNW4zMU7kC6gW+UIp9Rx0ldeiHvdVPau80vzlWHr7RDzHZlVbig1dGdo9d+EMiFJ6Mlp2lZZbX7VVCySmOqY0IFNy4VUQ+/rkLqLvY19vny/17jt539DwEEqEDGVcQr6Wfi0AR4ZKiAgFP6wZNqtUdWU9B6X0vWsrbtg9u39KU+3nkz99SsvX95ysXA3JLwAAIABJREFURR8zPTocZqiJk6gb92Q91nWk2dyTZV1nxigbLwTagExS8U22z4d/RrdsTGx4WVgVXxH59+T+7+ewyClFvhL9rY+iJyR+a6odKKWuJFEEPGnzxjksYfZMnDu+ceg8Cpmzx72nela5pRQZz6E9ZY8TOpwXtpvMKEiN3S+UQdq2E/iYzksM7tceyIVXTZkPaci+k/cdvflDbFLHOGqt4+vp17EntZ1Njs+R4fKk8FU19VaLzURd+6nMym5bDeURKA9r47gIqKcBtFVUvBWqvJVSqENgJZNXmQ2NUPFFRN6ilNoBTDRGC6viq5S6h8mzQwWIEks1HdOJfH0cuFZEhtQ0J0cRuRG4EXTH+gzHmzcT546PzhkvKx5LjRkSP4pZ35mmtyO9MLkOx9PhKSet8xONvurdtnNhjMZ89r1tJx9b3zGp6suxLZ5zUs+48NVE6q0Wm4lZ7yfVrm9RqENd/kjLkvL37e3n+jsex7UVHWl30Uu/OJZFuzcWAqvkVEqmAmwmGqHiOx1zUvGd0tcUkQuSKqy7gXcmjz+I1l6ZiYrIF9UiX0qpXuAPgD9XSt0MnAF8vp6FNprxc8dVMmc8zaXh7fhxhG0pIokBxZvOO3nuBsR2Id0JHetg1an6ir1tjT5BraCwya6LtxJEOmw1XfhqoT63YPuxHch0Q9dm6DlFhxnt5g4IrVZMVkrfu7bi5gcONHUdc8WzbTrSLmvbU0n4K013xiPjNSiPuHR5Ai26WM18VXynI1ZKVWxC3Sq+9QQsnyMiDwCIyB3ofMdM3Ap4icjXx4D3JSJfgyKyXUQuE5HLgF8Db61noY3GyR8eN+ZSWYpsto3TMsP0dmQo+CGr29Ncdclps7vaU0obiPZefdLpOVk/TnWsKKMxkR2n93LNK8+ktyPNUDGgtyM9Wr7biM81ZD+2o41Izyk6X+W1zfiRheDIUHHSRcxSln6pGJU1balkWqTOqWQ9Z6XnVK5Dq/ZWDMlCqPhOR0XFFxZYxTdQSp0vIvcqpc5CS6BMSxK6mpiUeV+N9+2oZ5HNIGzbgFPqR7lZbEuLhBGUsVdt5uOvmaWwn2XrEJWX1WNdzdVVTeqq+lrAzzVqP0DyW2d1uKs8rHtQGlTdtWCKyYuUWhVgpTDCD8e66ydH2pchHxr6Ph/qmqTiO9/qrIk0XMVXKbURuBZ4Gjo+97ci8vB8Fj1bGqniW1HJbT96H96Pr9PaUU5adzbHAVz8vpmn7FlWVa9GVuc5DIawPJaQjxdOVbc6J1I9XGzWXvISRZCxnErSDNkso2JUfCdTT7PhU8Cbm7CWpqFQZDw9CTDt6vJdtr5AG4MHvzDznPFqaQ03qyt4DIaJVFd3+Xntnfj5eUuuNHK42FJAUd1Z744O8CpXJFtMBVhTqUfF97XAe9GmXgCWouxJxXBkXJuMlxiOiUw121upseopN6v7NgyG2VCRXYkjbUzKwxDOfd5No4aLLUVGB3hV5YmCOPFSAsGPIoLIdNY3inpyItcALxKRujL1i42Ua9PmOWSnMhy1GGc0MvqxyWsYFgLL1tVdmW5tRMo5reO1wpWGF5pRgcmqqZDlJK9SjiIjh7+A1GNE9i5FA5JxLdpSmfpmc1Qbjcr9UjUaj92WNPzt0xIjC9VMuJxp1XfmeOCs1uGusKxDXf7IvDwUQ20quU/dX6pzGn4UjUm2GB2wOVOPETmglPoSuvejEs66saGrWgBm7ONwUkloKrO0jUY1icw6lgfpHq2T9b13Ax8zhmQqFst3VsmfZFeNeSgLnJA3jGd0GmSS0gzjeFxuZcVUgc2TeozIzxq+imZgu+ONxgL1aEypKNvkq9u79vSx5usfpt0PiF2Xte0RHemZpd6XJbP57hNp+ony+AO3f4y/uKtj7krB86HaQ/ELYx3y5kq5oTiWheNZZJMQWK2EvWEy/3979x8d51Xfefz9HWk0kuyxJP+QkjiOk5Q1bhwIBNENJA5qWqfAaUi6SU/r7paT0JKQw5YfLeDAoWxwgNaQ/shJaYvZcxKWkrCQ0hQCW2oIJiaJaY2BJmlMmpj1Or9sObFlyfo1I9394z5jj+SZ0cxonplnnvm8zrFnNDMa3Ueame9zv/fe711wii+Ama0BznbOPRp+k05XzRRfEokgaAT/QlhVPKcSbN5eG3cOHuGixz7hP5ySXb5S7uw0vCWcs9tcO/73xI2MWZrc+dPqni7SqTaYPAbvq+us7MbJ71mU87v/y1f5HkheT3R0IsPoyDC/u/Tzc/6u1SxqrBnnIDPupwxnTiigNIjDlT+2mi/GU3wXHDAws48AnwY+a2adZlZWUa6GCPaJYNlZ0HcepM/wJUZCKktRrBJs2+47T53dmvnLRIc/6w2xHYfbziDFFAnzL/PhsamySr3HSn7Popzffe9a/zvKMzJ6nEOJgUVXCq4pMz+7Kz3gX9vLzvKv7RauetAIVQWQmCunpsCVzrnNwEiweVRlRYrqoaP71Jtr6apTHyAhO3h0nK7k3DdxV7KN5ZkX5m7PCgtu/lSLdnyt81qSZEm5SRJAIlPDUu/N4tiBk7/70cks+4+cYN9LWQ4ffIqd+w6f/vgCJextZpp/6Lp2zsOqqRQcmlxwXNrv66/1nO1L1muRqzRAOUHEBZuVODNrh5ObnDVWe8fcukWpdN0Hx9f0dTMxL086kZnh5eSZp53dhtkjyLVjb2qQv+2+maOJ5Sx1o4x1rAwthRZZQc9idDLLcyMTZGZm6U5M8xz9fOzrT5weSNZt8r+j9IBP+6UH+OLyP+Bhe+2ch1VTKbhukp1+/KT3HP9+WNofag9cJF85QeTD+Nr1FwLfBT4ZaouKaUv6ufXLzvQbM/We05AKqvmKVYKdueQPwtv8aYF2/Kjjdfxh1238Vtd2jvyX+1orgEDRTa/u77queEpq3Sa4/gE/bnT9A1xy5W/VpFJwQ7S1+wCSK/jZe44PMHGZgSiRU9bAOpzcUvGIK/cbamhw8HVuz54f1fvHliU3O+u0/cdPzhCq8eZPlbajFT21gx99+VZWc2rTq72pQZxzjExk2LXligWfIpa/z9lZyE749SiZcS1wrKcYD6yXU4DxtfgCjEuA54APOucO1KFtJw0ODro9e/bU80dKk9u8ffdpm06NT2fpT3eW3PSqpWSn/UyvqVEtcAxbjINIOemsvwFuds5diq9j//lwmySyeLXavCrW2jv8gHzvOX4/+c4ezfaSipUzoHDCOfcMgHPuX0ttaSvlK7pIUWpiaH0/WyE2KanQXy/tKT+zkVWneijTJxq2BbA0j3LSWVuAJ4AdwC8Bv4Yvyohzri594Lils4otUmzoYjaJrIa+XmZnghXzJ/wMQ5VYr06Lp7PeDPwRfobWVuBS4NvB11KFYosUG7qYTSKroa+XRFveAt5zYclKbYUgc5SzKdUv16MhreTg0XF6fTlRjk9kODI2xfTMLM8enWDnvsPqjcgc+a+XnIYsfswvYz+T9SXsp8Z8BWJpWUV7ImZ2q5klg+tJM7vHzB42s4XnR0pJucWBxycyPD8yQXbGYfh93QsuiJOWVmxRa0MXP7a1B4Pya6BvrV+zpRXzLalUOmuTcy4TXP8o8AXgSuAjobcq5nIzhw6NTvpKPMF/Z/R0Kq0lp4n8TLO2pA8ivefkLQKuYh9yaUqlgsgEgJmtBC5wzn3bOXcC0BzARRpa38/Wt23AOb/jWnvCOKu3k3RnMlo1miQScq+X/nQnIxMZ+tOd0Z2EcbIc0VrfS+nqU/mVmCv11/2xmf0JcBHwQfBpLaCnHg2Lu6H1/Vx8Tt9pC+IanqaQSBpa3x/NoFFKbqOtJSv8VOHpMb+wURttxUqpILIFPzPrfznnngxuWw58IPRWNbMKNkS66fLz+djXn2B8Ojtn6mZk0hRNRmtvIizZGRSKXOlryU0d99OGNWW46ZVdO6uRmmadSKUbIhHTGk0NoLU3TWh2BiZH/L+4905ivE5EycpaKrLVaqntaZsyTRFB+WspALo72hmfzvK5h/br9xtViTY/ftLV51NdE8c0XbgJKYjU0rEDfqvVfCFuRlVIq6Z0IrOWQipn5vcDSqUhM8kjjz/NVx/9GS+OjHNGTxe/PbiG/3z+ika3UoooZ8W6lKvAVqv13J42l9I5PDpJb1eSw6OTLbPuJJJrKaRiO585zi07hnl8vBfr6mP4RIY7HvwPfrj/pUY3TYpQEKmlAlut1nN72lYupxL5tRRSltxruDOVYrStl5HU2Yy19/F3e15sdNOkCKWzamndJuD2um5Gla+VUzpRq9rbqmnFxTrtNWzGbMcyfjKagZ7Vforw9JjfYEsiQUGk1tZtatiWtGv6ult63UlUJinkzxTLTytuDdooxZV8DSe7/D+3yu/MODWqacIRoHRWjCilEw2tnFZcrLJew2bQsQTSZ8Dy8yE94IOLNIR6Ik1uftrkuotX8+j+lyOR0mkpeYtMP3B8Gf/YfR0/5vUn726VtOJiVZyWzJ/ZlZ2CiaO+srDUjYJIEyuUNrlv73NaYFdv+YtMO/sYGH2Zd534Gz6HsTc1CLRWWnGxqk5Ltqd876Q7EwSTUaW66kDprCamtElE5C8yNaNnWQ/TtHPNxH1KKzZCWxKW9vuJLZ09vrcioVFPpIm18mysSJm3yDTd2Q49vXSPHmZkIqO0YqO0Jf2+8d0r/AZakyN+/3ipKQWRJtbqs7Eio3ctjB46Ve4GSLdlSK9Zx67rtYdbwyUSvkfS2eOrCU8dV6qrhpTOamKajRURDV5kKhVIdvpUV995vpfSnmp0i5qeeiJNLGoL7FpWgxeZghY3Viy/d5Kdgsnjvoei3knFVApepMmpDH6NOOfTXFPHfdqrllQKXkSiqlAZ/OHRSd7z5R+zrCupnkm5zKBzmf+XnfID8Ro7WZDGRESa3MGj43Ql205+fXwiw0snpjkxnW25as41054Kxk7O9XueJPRRWYx+MyJNbn4Z/CNjfmOnzvY2rR9arNzGWbmB+Lbkwt/TYhRERAI79x1m8/bdXLbtQTZv3900Z+7zZ+lNZWfBwar0qZlHWj+0SGZ+EL5vLSw7U7W68iiIiNDcG3oNre9n69s20J/uZGQiQ3dHGyvTHaQ7T501a/1QDXUs8WXpe9f4ml0tviJeA+siNP8e7fn1pnIBcXw6O2e2ltYP1Vh7ylcQnlnhB+Enj7XkILx6IiKcPjgNzZsCmt8z6U93arpvmNraYckKP26yZIX/uoW01tGKFBG3EjJR2aCrpSQS0NXn/02NwsQxP1U45kLriZjZbWb2fTN72Mw25N3+ajP7ZzPbZWZfMbOOsNogUi6VkJGaSqX9mEnPaj+GEmOhBBEz2wgMOOfeBNwEfCbvbgdc5ZzbCBwArg6jDSKVGFrfz3UXr2Z4dIonXxxleHSK6y5eXfnZ/FM74O5fh798lb98akc4DZbmkOzys7liLKyeyJXAvQDOuceB5bk7nHOPOedyfbyjwIlCT2BmN5rZHjPbMzw8HFIzRbyd+w5z397nWJVO8YtnpFmVTnHf3ucqm52V25xq9JAvDT96yH+tQCIxFlYQ6QfyP/mzZjbnZ5nZpcAG4NuFnsA5t905N+icG1y1alVIzRTxarLB17zNqejo9l8/ckd4DRdpsLAG1keAvryvZ51zswBmZsAWIAm83Tk3U+D7ReqqJht8zducCvDpjGP/rwYtFImmsHoiu4DrAMzsAuDZvPveBbzgnLtNAUSiYn7pEKhidlbvWshMzL0tM+FLw4vEVFhB5JtAh5ntAm4HtpjZtmAm1lXATWa2M/j3hyG1QaRsNZmdpc2ppAVpPxGRQG5jp0Vt8PXUjoZuTiUlnPzbHPC9xvr+bWJbG0VBRETiLzdzLtHhx6kyE76X+Jbb6xVIYhtEVPZEROJPM+dCo7InTUJ7aNdBY9MdNaXXyzyaORca9USaQDOXKW8aMVooqNdLAZo5FxoFkSZQk4VwUtojd3Ai28b+4459h8bYf9xxItvWlOmOqL1eIrHZl2bOhUZBpAnEqUx5VE0M/5yDY47MzCxtCSMzM8vBMcfE8M8b3bSKRen1Eple0bpNfhA9PeD3/UgP1HNQPdY0JtIE4lamPIqeyaxgKUfImN/2NGFG0k3yTGYlFza4bZWK0uslUpt9rdukoBEC9USqVM8uusqUh+8udxUpmyHlJsE5Um6SlM1wl7tqcU9c66q+ZTxflF4vUeoVSTgURKpQ7y66dqoL3/OrNnJHx40cTSwnzRhHE8u5o+NGnl+1sfonrfVgfZnPF6XXS03KyUikabFhFTZv331aumB8Okt/upN7b7ykgS2TauVODJJtNmdf8kV9+N796/6DviPvA3N63Ofjr3+g8c9XB6H8XptTbBcbakykUk/t4AMv3spqhjk0McDXOq9lb2qw4i56ufP4Nd+/PobW97MVFl/2JF8laxPKWaPShGsdQvm9SqSoJ1KJIJ3w7PEZTswm6bJpkmT52+6b+YG9puyeSLlnZzqLa3Ll9hzKLcnRhD0ROSm2PRGNiVQiKJ3Qs6wHZ8YEKTK0c83EfRUNXJY7jz9q8/2lQuWuTSi3JIfWOkgEKZ1ViSCdkDbjnOksjB+m3WU4g0PcOXiEi8rsHZS7AVJNNkqSxgjSU1Pjo2Smp5hybbzQcS4zl3yIi6pNU63bBNyuKsESKQoilehd69MJsxm6J1+AhIFLQKKNix77BJzdW9Ybutx5/FGa7y8VCNJTJ7JtHJzuoZMpUokZ/i5xNY/sWcnWsw7PTUfmXlf5aapiJTm01kEiRumsSuTSCaMvcjLFacCyMyuqCFruPP4ozfeXCgTpqUOTCcyMTKKLDO1szt5fOB2pNJU0MQWRSuRKJ7hZ/68tCT1nQ2pZRbNkyp3HH6X5/rFW6wWBxw5Asovp7CwJ8ycbU6QYmD1UOB2pkhzSxJTOqtS6TbDml8pPPxQxtL6/rGBQ7uOkSvkzo/IX8LGID/EgPdXRniAz4wNJiikOJQaKpyOVppImpZ5INZR+iI8wNisKXh8DnbM450jOTpAky73t1ygdKbGjIFINpR+irZL0VJB6mmOxC/iC18eSFWextnuaseRKbm/7fQ4sv1TpSIkdLTaUeKl0L20t4JP6iO1iQ42JSLzkp6fAX05zKj01v7TIG9/rg840c4POYlOTMdpqV6QUpbMkXoqlpw7vK1wBF2qfmozRVrsiC1FPROKl2MK92WlI9BXuoVz/QG17CaV6Q+qNSMyoJyLx8sb3+h7F8M/g0BP+cvIYtKVqP4BeTBiD9SIRpSAiMTR/DNNgab/vkeSrcG1P2XrX1u9niTSYgojEyyN3QGcPrHolDGzwl509fj1Pvdb2aB2RtBAFEYmXYqmk6bH6re2J6zqiWpeHkVjQwLrES6mKuPUsLRK3MiZhlIeRWFBPROJFqaRwhFEeRmJBQUTipUAq6aev+iibd6a5bNuDbN6+m537Dje6lc1HM86kCKWzJH7yUkmn9qmfpLcryeHRST729SfYCqphVYlKNs6SlqKeiMSa9qmvEaUJpQj1RIrYue8wn3toPwePjrOmr5s3nL+cR/e/fPLrmy4/X2eyTUD71NeI9neXIhRECjiVAjF6u5L8/MgY//J/X6Y/3cGKJSmlRJqI9qmvobjNOJOaUDqrgPkpkNHJLAmD4xNZpUSajPapFwmXeiIFzE+BTM/MkjB/mVNuSmR+WkxpsPoaWt/Pnc9/n7bdd7J87AVeTp7J0694B597qIuP/uPj5f9NVNpdpCD1RApY09fNRGbm5NcdbQlmnb/MKSclkkuLHR6dOzNIU0zr6KkdXPTYJ7gwPcFZA2dyXmqM1z/5Kda+/HD5fxOVdhcpSkGkgPkpkHRnO7MOlnW1V5QSqWhmULGSEio1sTjzFskdmkyQoZ3N2fvLT01qoZ1IUQoiBQyt72fr2zbQn+5kZCLDeSuX8t4rXsG5K5YyMpGhP91Z1l7ZB4+O05Vsm3NbwTRYsTPdndt0BrxY8xbJTWdnyVgnA7OHTt62YGpSC+1EitKYSBFD6/tPCxLvqfA5yp4ZVGwTo92fhSUD2txoMeYtkutoT5DIjnOobeDkQxZMTWqhnUhR6olUqoL0Utkzg4qd6U6N6Qx4seYtkhvonCVJlnvbryk/NamFdiJFKYhUosIB1vlpsaJpsGKbGKWWanOjxZpXS2vJirM4svGTHFh+afmpybiWdhepAXPONboNCxocHHR79uxpdDN8z2N+WmN63H+oXP9A9c+bX2Y72XVqT/CLfgd+es/pt+sDTKTZzN9uMzbUE6lEWAOsxc50h7boDFhEIk0D65UIc4C1WEkJlZoQkQhTT6QSGmAVEZkjtCBiZreZ2ffN7GEz25B3+1Izu9fMHjKz+81sWVhtqDkNsMbbzm3wp+fAx5f7y53bGt0ikcgLJZ1lZhuBAefcm8zsQuAzwFuDu98PfMM5d4+ZvRu4GWied6vSS/G0cxs89GnAIJH0vcyHPu3vG9rS0KaJRFlYPZErgXsBnHOPA8vz7rsC+Gpw/e+BN4TUBpHy7f4sYNDWDongEgtuF5Fiwgoi/cBw3tdZM8v9rJRzLhNcfwnoK/QEZnajme0xsz3Dw8OFHiJSO1NjYHNL1GBt/nYRKSqsIDLC3OAw65zL1VGfzQsofcwNNic557Y75wadc4OrVq0KqZkigdRScDNzb3Mz/nYRKSqsILILuA7AzC4Ans2774fA1cH1a4HvhNQGkfJd8m7AwUwWZoNLXHC7iBQTVhD5JtBhZruA24EtZrbNzDqAPwFuNLOdwOuAu0Jqg0j5hrbA5R/ya4BmM/7y8g9pUF1kASp7IiISPpU9ERERmU9BREREqqYgIiIiVVMQERGRqimIiIhI1RRERESkagoiIiJSNQURERGpmoKIiIhUTUFERESqpiAiIiJVUxAREZGqNUUBRjMbBg7U8UeuBI7U8eeFRccRHXE4BojHcTTiGI44595c559ZF00RROrNzPY45wYb3Y7F0nFERxyOAeJxHHE4hihROktERKqmICIiIlVTEClse6MbUCM6juiIwzFAPI4jDscQGRoTERGRqqknIiIiVVMQERGRqrV8EDGz28zs+2b2sJltyLt9qZnda2YPmdn9Zraske1cSInjeLWZ/bOZ7TKzr5hZRyPbWUqxY8i7f8DMxs2ssxHtK1ep4zCzG8xsd3DfrzSqjeUo8ZrqMLO7zOxBM/uWmfU0sp2lmNkqM/ukmd027/amen9HWUsHETPbCAw4594E3AR8Ju/u9wPfcM5dDuwAbm5AE8uywHE44Crn3Eb8gs2rG9DEBS1wDDm3EPGFbqWOI/gg3gi80Tl3qXPuuw1q5oIW+Hu8GXjOOXcF8DXg9xvQxHL9GTAFJOfd3jTv76hr6SACXAncC+CcexxYnnffFcBXg+t/D7yhvk2rSNHjcM495pybCr48Cpyof/PKUupvgZldjA+I++vftIqUOo7fwwfyB4Ne4coGtK9cpY5jFOgLrq8EhuvbtPI5594OPFTgrmZ6f0daqweRfua+AbJmlvudpJxzmeD6S5x600RRqeMAwMwuBTYA365nwypQ9BjMrBv4U+DjjWhYhUr9Lf4TvvzFEP4D7H/UuW2VKHUcPwB+0cz+HfivwD/Uu3E10Ezv70hr9SAywtwXz6xzbjZ3Pe9N00eEz7YocRzm3YI/83q7c26mEQ0sQ6m/xV8A25xzI/VvVsVKHUcW+FZw/QHggno2rEKljuNTwO3OuQuA36U511000/s70lo9iOwCrgMwswuAZ/Pu+yGnxg+uBb5T36ZVpNRxvAt4wTl3W4QDCBQ5BjPrB14HvNPMvoz/4L27QW0sR6m/xaPAW4PrQ8C/1bVllSl1HGuBF4Prh4E19W1aTTTT+zvSWnqxYXAm8lngQnye9ybgvwN/DCwDvgh0AU8D784bW4iUBY7jfqAXmA4e/nXn3J83op2llDoG59x03uN2Am92zk02op0LWeBv0QHcBazCn+m/wzn3UoOaWtICx3Ee8Nf4k9Ak8EHn3KMNauqCzGwI/5q5xcy20WTv76hr6SAiIiKL0+rpLBERWQQFERERqZqCiIiIVE1BRCTCgkFhkchSEJHQmNm1Zvb0/IWPIf68u81sfYHbbzWzyO1vbWZbzOzSQu0zszYz+yLwpaDW1qvM7DIz+1iDmitSkIKIhOm/Ad/El9CQPEHJk3Odcw8XechbgeeBzzvnLgH2Oed+AKw1s/Pr1U6RhSiISCjM7Bz8+oI/x9eMyt1+q5l9xsz+j5k9YWaXBbffbWYfN7MdZvYTM3tlcPvuvO99l5ldH1z/cFBFdq+ZXVVBu3ab2SeCqsbfMbNUcPtmM/tBUNX1d4KewF+Z2c7gez4UPG7IzL5oZl8ys8fN7JfN7Gtm9lMze3/wmKVmdk/QvgfMbHmBplzDqdpNubadHfz884EX8GVqlgDklei4B19qRCQSFEQkLO8A7nLOHQC6zeyMvPuyzrm34EtmvD/v9uedc5uArcCNCzz//wyqyP4G8J4K2rUK+FJQ1fjfgSuDD+3fA341qOr65aD9h4M6V28ELjOzVwfPsRrfy/rtvMe+HnhncP8twFeC9v018O4C7bgQ2Jv3dT9+Jf4Nzrn9zrk9wF8BV+UCZ2Av8NoKjlckVAoiUnPBGMhvAu8zs/vxH9w35D0kV1X1SeZWhy12e6Hnf1+w+vidQLqC5h1xzj057+dcDHwrtwo+qBH1Gnx9q9zX3wNeGXzfvzq/SvdpfJrpWLCqfjy4/2Lgj4LV9R8pcizdeY8HHzi/4Jx7JneDc+6fgK8AF5lZLhCdIOidiESBgoiE4dfwZ+JXO+euAS4DfsPMLLjf5V1a3vcVun3GzLqC668ILl8LrHTObaHyCrJu3nUD/gP4FTNrBzCzJPAEft+MXNDayKlaV/OfY76ngI8454acc5fhA8l8LwJn5X29Bbgh2McDM+sLfu6JhMIWAAAA50lEQVQs8E/41BbB9zxf3qGKhE9BRMLwTvLy/cFZ+h7gV6t4rtuBL5jZHwNnB7ftA9ab2feqfM45nHM/Bb4LPGpm38UPan8eOM/MduF7Id90zv2szKf8FPBhM/uemX0D+IUCj9nJ3AkHI/hCgJ8ys8uBdfgiiDcAHwA+HTxuE34TJZFIUO0skQYIemVfAq7PLzBZ4HG3OuduDa534NNbv5k30C7SUAoiIg0SzEDrcs79pMzHvwaYyhvTEWk4BREREamaxkRERKRqCiIiIlI1BREREamagoiIiFRNQURERKqmICIiIlX7/8JOSUIln0KQAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 409.25x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#성별에 따라 본 Annual Income X Spending Score \n",
    "sns.lmplot(x='Annual Income (k$)',y='Spending Score (1-100)',hue='Gender',data=data)\n",
    "plt.show()\n",
    "#성별에 따른 두 변수의 값은 크게 차이가 없다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Modeling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2-1. Hierarchical Clustering\n",
    "계층적 할당 방법을 이용하여 군집을 몇 개로 나누면 좋을 지 봅시다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2880x1440 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Perform the necessary imports\n",
    "from scipy.cluster.hierarchy import linkage, dendrogram\n",
    "\n",
    "# Calculate the linkage: mergings\n",
    "# method = 'single','complete', 'average', 'weighted', 'centroid', 'median', 'ward'\n",
    "\n",
    "mergings = linkage(data, method='complete')\n",
    "# 완전결합군집 방법으로 Clustering\n",
    "\n",
    "# Plot the dendrogram, using varieties as labels\n",
    "plt.figure(figsize=(40,20))\n",
    "dendrogram(mergings,\n",
    "#            labels = labels.as_matrix(columns=['labels']),\n",
    "           leaf_rotation=90,\n",
    "           leaf_font_size=20,\n",
    ")\n",
    "plt.show()\n",
    "\n",
    "#정해진 분류가 없으므로 labels은 주석처리하였음."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "덴드로그램을 보아 크게는 2, 3, 5, 7개의 군집을 고려할 수 있음.  \n",
    "그 중에서도 3개의 군집으로 나누는 것을 먼저 선택하여 분석을 진행"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>195</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>196</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>197</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>198</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>199</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>200 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     predict\n",
       "0          3\n",
       "1          3\n",
       "2          1\n",
       "3          1\n",
       "4          1\n",
       "..       ...\n",
       "195        1\n",
       "196        1\n",
       "197        3\n",
       "198        3\n",
       "199        3\n",
       "\n",
       "[200 rows x 1 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from scipy.cluster.hierarchy import fcluster\n",
    "\n",
    "# fcluster(linkage함수, 클러스터링을 멈추도록하는 임계값, 거리를 기준으로)\n",
    "# 위의 덴드로그램에서 클러스터간의 거리가 1.4인 지점에서 자르면 3개의 cluster가 형성됨을 알 수 있다.\n",
    "# 따라서 임계값을 1.4로 정해서 분석을 실시하면\n",
    "predict = pd.DataFrame(fcluster(mergings,1.4,criterion='distance'))\n",
    "predict.columns=['predict'] # predict column에 예측 결과\n",
    "predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    112\n",
       "3     53\n",
       "2     35\n",
       "Name: predict, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#각각의 그룹에 할당된 개수\n",
    "predict['predict'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4395412725206968"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#실루엣 스코어 결과 값\n",
    "silhouette_score(data, predict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2    78\n",
       "3    35\n",
       "1    34\n",
       "5    30\n",
       "4    23\n",
       "Name: predict, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#마찬가지로 임계치 1.2인 지점에서 자르면 총 다섯개의 그룹이 생성된다.\n",
    "predict = pd.DataFrame(fcluster(mergings,1.2,criterion='distance'))\n",
    "predict.columns=['predict'] # predict column에 예측 결과\n",
    "predict['predict'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.30316439293019537"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#실루엣 스코어 결과 값 :: 아까 3으로 나누었을 때보다 더 작아졌다.\n",
    "silhouette_score(data, predict)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2-2. K-means Clustering\n",
    "위의 분석 결과를 바탕으로 비계층적 군집분석 방법인 K-means clustering을 진행해 봅시다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "      <th>predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.019231</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.387755</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.057692</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.816327</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.038462</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.051020</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.096154</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.775510</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.397959</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Gender       Age  Annual Income (k$)  Spending Score (1-100)  predict\n",
       "0     1.0  0.019231            0.000000                0.387755        0\n",
       "1     1.0  0.057692            0.000000                0.816327        0\n",
       "2     0.0  0.038462            0.008197                0.051020        1\n",
       "3     0.0  0.096154            0.008197                0.775510        1\n",
       "4     0.0  0.250000            0.016393                0.397959        1"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "\n",
    "# 비계층적 방법은 사전에 군집의 수를 정의해 주어야 합니다. 위의 결과를 바탕으로 가장 좋은 결과를 낼 수 있을 거라 예상하는 3을 넣어보았습니다.\n",
    "model = KMeans(n_clusters=3,algorithm='auto')\n",
    "model.fit(data) #우리는 target value가 없기 때문에 그냥 data자체가 feature만을 담고 있는 변수입니다.\n",
    "\n",
    "# k-means 방법을 사용하여 data를 세 개의 군집으로 나누고 이를 predict라는 이름으로 저장\n",
    "predict = pd.DataFrame(model.predict(data))\n",
    "predict.columns=['predict']\n",
    "\n",
    "# 이를 data와 합친 새로운 데이터 프레임 r을 생성하고 출력\n",
    "r = pd.concat([data,predict],axis=1)\n",
    "r.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    112\n",
       "2     48\n",
       "0     40\n",
       "Name: predict, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r['predict'].value_counts()\n",
    "#계층적 방법으로 진행한 결과와 사뭇 다르긴 하네요"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.45835255811254244"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#실루엣 스코어 :: 계층적 군집 시 3개로 나누었던 결과보다 더 좋은 점수를 얻었다.\n",
    "silhouette_score(data, predict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#이를 시각적으로 표현하면 \n",
    "centers = pd.DataFrame(model.cluster_centers_,columns=['Gender','Age','Annual Income (k$)','Spending Score (1-100)'])\n",
    "center_x = centers['Age']\n",
    "center_y = centers['Annual Income (k$)']\n",
    "center_z = centers['Spending Score (1-100)']\n",
    "\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "# scatter plot\n",
    "fig = plt.figure( figsize=(6,6))\n",
    "ax = Axes3D(fig, rect=[0, 0, .95, 1], elev=48, azim=134)\n",
    "ax.scatter(r['Age'],r['Annual Income (k$)'],r['Spending Score (1-100)'],c=r['predict'],alpha=0.5)\n",
    "ax.scatter(center_x,center_y,center_z,s=50,marker='D',c='r')\n",
    "ax.set_xlabel('Age')\n",
    "ax.set_ylabel('Annual Income (k$)')\n",
    "ax.set_zlabel('Spending Score (1-100)')\n",
    "plt.show()\n",
    "#군집의 중심점이 빨간색으로 표현되어 있다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Elbow Point"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Elbow curve')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "distorsions = []\n",
    "\n",
    "# 군집 수 K의 변화에 따라\n",
    "# Inertia Value: 군집화가 된 후에, 각 중심점에서 군집내의 데이터간의 거리를 합산한 것\n",
    "# 즉, 군집의 응집도를 나타내는 값\n",
    "# 이 값을 구해 그래프로 그린다!\n",
    "for k in range(2, 20):\n",
    "    kmeans = KMeans(n_clusters=k)\n",
    "    kmeans.fit(data)\n",
    "    distorsions.append(kmeans.inertia_)\n",
    "\n",
    "fig = plt.figure(figsize=(15, 5))\n",
    "plt.plot(range(2, 20), distorsions)\n",
    "plt.grid(True)\n",
    "plt.title('Elbow curve')\n",
    "\n",
    "# 급격하게 하강하다가 변화의 폭이 크지 않는 지점 Elbow Point를 찾는다.\n",
    "# 군집수 3~4정도가 elbow point로 보이는데, 이외에도 10정도까지 고려해볼 만한 가치가 있다고 생각"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2-3. DBSCAN\n",
    "k-means 가 거리기반이었다면, 밀도 기반의 계산 방식을 사용한 dbscan을 사용해봅시다\n",
    "\n",
    "- dbscan 모델에는 두 가지 하이퍼 파라미터가 존재합니다\n",
    "   1.  eps : 군집의 기준이 되는 원과 점 사이의 반지름\n",
    "   2. min_samples : core point라고 분류할 점의 개수\n",
    "   \n",
    "   \n",
    "- 먼저 초기값을 지정하고, 그 뒤에 적절히 튜닝하자!\n",
    "-그렇다면 초기값을 지정하는 기준은?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import NearestNeighbors\n",
    "#k-dist를 이용하자!\n",
    "neigh = NearestNeighbors(n_neighbors=5)\n",
    "nbrs = neigh.fit(data)\n",
    "distances, indices = nbrs.kneighbors(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x22251260148>]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "distances = np.sort(distances, axis=0)\n",
    "distances = distances[:,1]\n",
    "plt.plot(distances)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "급격하게 변하는 지점이 0.15정도 이고, 이때의 minpts가 5이다.  \n",
    "(그런데 minpts는 어떤 값으로 지정하면 좋을 지 어떻게 아나요?)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "#위의 결과를 바탕으로 DBSCAN 모델을 생성해봅시다\n",
    "from sklearn.cluster import DBSCAN\n",
    "\n",
    "#모델을 생성하고 파라미터를 각각 대입\n",
    "model = DBSCAN(eps=0.15,min_samples=3)\n",
    "\n",
    "#모델을 적합시키고 이를 바탕으로 분류한 결과를 predict에 데이터 프레임 형태로 저장\n",
    "predict = pd.DataFrame(model.fit_predict(data))\n",
    "predict.columns=['predict']\n",
    "\n",
    "# 결과를 기존 data와 합쳐 r이라는 새로운 데이터 프레임에 저장\n",
    "r = pd.concat([data,predict],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "      <th>predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.019231</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.387755</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.057692</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.816327</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.038462</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.051020</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.096154</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.775510</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.397959</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Gender       Age  Annual Income (k$)  Spending Score (1-100)  predict\n",
       "0     1.0  0.019231            0.000000                0.387755       -1\n",
       "1     1.0  0.057692            0.000000                0.816327        0\n",
       "2     0.0  0.038462            0.008197                0.051020       -1\n",
       "3     0.0  0.096154            0.008197                0.775510        1\n",
       "4     0.0  0.250000            0.016393                0.397959        3"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1     60\n",
       " 4     23\n",
       "-1     21\n",
       " 7     20\n",
       " 6     16\n",
       " 3     11\n",
       " 5     10\n",
       " 0     10\n",
       " 10     8\n",
       " 11     6\n",
       " 9      4\n",
       " 8      4\n",
       " 2      4\n",
       " 12     3\n",
       "Name: predict, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r['predict'].value_counts()\n",
    "# -1값은 노이즈 값, 총 군집은 13개로 나누어졌고 노이즈 포인트는 21개가 있다. 하이퍼 파라미터가 별로 좋지 않았던 듯함."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "\n",
    "fig = plt.figure(figsize=(6,6))\n",
    "ax = Axes3D(fig, rect=[0, 0, .95, 1], elev=48, azim=134)\n",
    "ax.scatter(r['Age'],r['Annual Income (k$)'],r['Spending Score (1-100)'],c=r['predict'],alpha=0.5)\n",
    "ax.set_xlabel('Age')\n",
    "ax.set_ylabel('Annual Income (k$)')\n",
    "ax.set_zlabel('Spending Score (1-100)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.21915318713343657"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "silhouette_score(data, predict)\n",
    "#분석 중 최악의 스코어가 나왔다. 하이퍼 파라미터를 적절히 조정하여 군집 수를 줄여보자."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    101\n",
       " 2     38\n",
       " 3     26\n",
       "-1     25\n",
       " 0     10\n",
       "Name: predict, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#min_samples를 8로, eps를 0.23으로 바꾸어보자. (좀 더 넓은 범위의 원을 그리겠다는 느낌)\n",
    "model = DBSCAN(eps=0.23,min_samples=8)\n",
    "predict = pd.DataFrame(model.fit_predict(data))\n",
    "predict.columns=['predict']\n",
    "r = pd.concat([data,predict],axis=1)\n",
    "r['predict'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.33131121449424966"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "silhouette_score(data, predict)\n",
    "#썩 좋지 않다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### GridSearchCV 으로 DBSCAN 하이퍼 파라미터를 조정해보자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param = {'eps' : np.arange(0.1, 0.3, 0.02), 'min_samples' : np.arange(1, 10)}\n",
    "gs = GridSearchCV(DBSCAN(), param, cv=5, scoring = \"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "# gs.fit(data)\n",
    "#fit가 되지 않는다...왜일까...... "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[결론] 이 데이터는 밀도 기반 방식 보다는 거리 기반 방식에 더 적합한 데이터이다. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2-4. Spectral Clustering\n",
    "거리기반의 클러스터링 기법"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "      <th>predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.019231</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.387755</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.057692</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.816327</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.038462</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.051020</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.096154</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.775510</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.397959</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Gender       Age  Annual Income (k$)  Spending Score (1-100)  predict\n",
       "0     1.0  0.019231            0.000000                0.387755        0\n",
       "1     1.0  0.057692            0.000000                0.816327        0\n",
       "2     0.0  0.038462            0.008197                0.051020        1\n",
       "3     0.0  0.096154            0.008197                0.775510        1\n",
       "4     0.0  0.250000            0.016393                0.397959        1"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import SpectralClustering\n",
    "\n",
    "#Spectral Clustering 모델 생성\n",
    "model = SpectralClustering(n_clusters = 3)\n",
    "\n",
    "#예측 값 생성하고 이를 data에 합친 후 출력\n",
    "predict = pd.DataFrame(model.fit_predict(data))\n",
    "predict.columns = ['predict']\n",
    "r = pd.concat([data, predict], axis=1)\n",
    "r.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    112\n",
       "2     47\n",
       "0     41\n",
       "Name: predict, dtype: int64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r['predict'].value_counts()\n",
    "#n_cluster을 3으로 지정해주었기 때문에 군집을 3으로 분류하였다. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4573965431493555"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "silhouette_score(data, predict)\n",
    "#오!! 좀만 더 하면 좋은 결과가 나올 것 같은 예감,,,ㅎ"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "      <th>predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.019231</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.387755</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.057692</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.816327</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.038462</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.051020</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.096154</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.775510</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.397959</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Gender       Age  Annual Income (k$)  Spending Score (1-100)  predict\n",
       "0     1.0  0.019231            0.000000                0.387755        1\n",
       "1     1.0  0.057692            0.000000                0.816327        1\n",
       "2     0.0  0.038462            0.008197                0.051020        0\n",
       "3     0.0  0.096154            0.008197                0.775510        0\n",
       "4     0.0  0.250000            0.016393                0.397959        0"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#군집을 2개로 나누어 봅시다\n",
    "model = SpectralClustering(n_clusters = 2)\n",
    "predict = pd.DataFrame(model.fit_predict(data))\n",
    "predict.columns = ['predict']\n",
    "r = pd.concat([data, predict], axis=1)\n",
    "r.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    112\n",
       "1     88\n",
       "Name: predict, dtype: int64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r['predict'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5194624832665107"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "silhouette_score(data, predict)\n",
    "#군집을 두 개로 나누었을 때 0.5이상의 값이 나왔다. 군집 결과가 타당하다고 볼 수 있다!!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 아쉬운 점\n",
    "\n",
    "1. Spectral Clustering에 대한 자료를 읽어보았지만 솔직히 잘 이해가 가지 않는다. 그저 거리 기반의 클러스터링 방식인 것만 안다....\n",
    "\n",
    "2. 가장 좋은 결과가 나온 spectral clustering으로 그리드 서치를 이용해서 다양한 파라미터 조합에 대해 최적의 결과를 보고 싶었으나 안 되는 듯 하다 ㅠ (DBSCAN도 결국 fit 부분에서 실패했다)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}