Solution/期末專題實作(二)/不動產時價登入分析_Soultion.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "哈摟，各位資料科學家大家好，我是來自於台北市\"頭期款不足\"的阿宏，我想買房子很久了，但是不知道要買哪個地方，想先從內政部不動產交易時價登入資料中先了解行情，我現在手邊蒐集了台北市Taipei(A_lvr_land_A.csv)、新北市New_Taipei(F_lvr_land_A.csv)、台中市Taichung(B_lvr_land_A.csv)、高雄市Kaohsiung(E_lvr_land_A.csv)共四份資料，請各位資料科學家幫幫忙分析了。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1.不動產資料檔案讀取並串接"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "資料分散在4個資料，先合併成一個資料方便後續分析，但是我發現有英文欄位跟中文欄位，留下中文刪除英文，並加入新的欄位用以分辨台北市Taipei、新北市New_Taipei、台中市Taichung、高雄市Kaohsiung"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. 利用Pandas中的pd.read_csv()分別讀取A_lvr_land_A.csv、B_lvr_land_A.csv、E_lvr_land_A.csv、F_lvr_land_A.csv\n",
    "2. 刪除csv中第二列英文翻譯，The villages and towns urban district,transaction sign,land sector position building sector house number plate,land shifting total area square meter,....\n",
    "3. 生成city欄位其中以地區分類台北市Taipei(A_lvr_land_A.csv)、新北市New_Taipei(F_lvr_land_A.csv)、台中市Taichung(B_lvr_land_A.csv)、高雄市Kaohsiung(E_lvr_land_A.csv)\n",
    "4. 將以上四份資料運用pd.concat()串接"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/hong/.local/lib/python3.6/site-packages/IPython/core/interactiveshell.py:3072: DtypeWarning: Columns (3,7,14,15,16,17,18,21,22,24,25) have mixed types.Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n"
     ]
    }
   ],
   "source": [
    "#1.利用Pandas中的pd.read_csv()分別讀取A_lvr_land_A.csv、B_lvr_land_A.csv、E_lvr_land_A.csv、F_lvr_land_A.csv\n",
    "data_a = pd.read_csv('A_lvr_land_A.csv')\n",
    "data_b = pd.read_csv('B_lvr_land_A.csv')\n",
    "data_e = pd.read_csv('E_lvr_land_A.csv')\n",
    "data_f = pd.read_csv('F_lvr_land_A.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "#2.刪除csv中第二列英文翻譯，The villages and towns urban district,transaction sign,land sector position building sector house number plate,land shifting total area square meter,....\n",
    "data_a = data_a.iloc[1:]\n",
    "data_b = data_b.iloc[1:]\n",
    "data_e = data_e.iloc[1:]\n",
    "data_f = data_f.iloc[1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "#3. 生成city欄位其中以地區分類台北市Taipei(A_lvr_land_A.csv)、新北市New_Taipei(F_lvr_land_A.csv)、台中市Taichung(B_lvr_land_A.csv)、高雄市Kaohsiung(E_lvr_land_A.csv)\n",
    "data_a['city'] = 'Taipei'\n",
    "data_b['city'] = 'Taichung'\n",
    "data_e['city'] = 'Kaohsiung'\n",
    "data_f['city'] = 'New_Taipei'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>鄉鎮市區</th>\n",
       "      <th>交易標的</th>\n",
       "      <th>土地區段位置建物區段門牌</th>\n",
       "      <th>土地移轉總面積平方公尺</th>\n",
       "      <th>都市土地使用分區</th>\n",
       "      <th>非都市土地使用分區</th>\n",
       "      <th>非都市土地使用編定</th>\n",
       "      <th>交易年月日</th>\n",
       "      <th>交易筆棟數</th>\n",
       "      <th>移轉層次</th>\n",
       "      <th>...</th>\n",
       "      <th>車位類別</th>\n",
       "      <th>車位移轉總面積(平方公尺)</th>\n",
       "      <th>車位總價元</th>\n",
       "      <th>備註</th>\n",
       "      <th>編號</th>\n",
       "      <th>主建物面積</th>\n",
       "      <th>附屬建物面積</th>\n",
       "      <th>陽台面積</th>\n",
       "      <th>電梯</th>\n",
       "      <th>city</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>文山區</td>\n",
       "      <td>房地(土地+建物)+車位</td>\n",
       "      <td>臺北市文山區忠順街二段85巷1~30號</td>\n",
       "      <td>26.97</td>\n",
       "      <td>住</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1090406</td>\n",
       "      <td>土地1建物1車位2</td>\n",
       "      <td>四層</td>\n",
       "      <td>...</td>\n",
       "      <td>坡道平面</td>\n",
       "      <td>40.7</td>\n",
       "      <td>0</td>\n",
       "      <td>含增建或未登記建物。;內含獨立產權車位，未包含在棟數內。</td>\n",
       "      <td>RPVOMLMKNHPFFAA18CA</td>\n",
       "      <td>110.91</td>\n",
       "      <td>9.63</td>\n",
       "      <td>11.09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>文山區</td>\n",
       "      <td>房地(土地+建物)</td>\n",
       "      <td>臺北市文山區福興路91~120號</td>\n",
       "      <td>10.32</td>\n",
       "      <td>住</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1090317</td>\n",
       "      <td>土地1建物1車位0</td>\n",
       "      <td>三層</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>含增建或未登記建物。;</td>\n",
       "      <td>RPPPMLMKNHPFFAA67CA</td>\n",
       "      <td>24.58</td>\n",
       "      <td>1.28</td>\n",
       "      <td>2.30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>文山區</td>\n",
       "      <td>房地(土地+建物)</td>\n",
       "      <td>臺北市文山區木柵路四段121~150號</td>\n",
       "      <td>39.64</td>\n",
       "      <td>其他</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1090418</td>\n",
       "      <td>土地2建物1車位0</td>\n",
       "      <td>六層</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>RPSPMLMKNHPFFAA97CA</td>\n",
       "      <td>78.80</td>\n",
       "      <td>4.15</td>\n",
       "      <td>8.76</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>中正區</td>\n",
       "      <td>房地(土地+建物)</td>\n",
       "      <td>臺北市中正區和平西路一段31~60號</td>\n",
       "      <td>4.26</td>\n",
       "      <td>商</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1090413</td>\n",
       "      <td>土地2建物1車位0</td>\n",
       "      <td>九層</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>含增建或未登記建物。;</td>\n",
       "      <td>RPPOMLMKNHPFFAA57CA</td>\n",
       "      <td>18.34</td>\n",
       "      <td>0.00</td>\n",
       "      <td>6.45</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>文山區</td>\n",
       "      <td>房地(土地+建物)</td>\n",
       "      <td>臺北市文山區樟新街56巷1~30號</td>\n",
       "      <td>44.57</td>\n",
       "      <td>住</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1090328</td>\n",
       "      <td>土地3建物1車位0</td>\n",
       "      <td>四層</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>含增建或未登記建物。;</td>\n",
       "      <td>RPWNMLNKNHPFFAA28CA</td>\n",
       "      <td>98.74</td>\n",
       "      <td>2.26</td>\n",
       "      <td>11.84</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19814</th>\n",
       "      <td>鶯歌區</td>\n",
       "      <td>房地(土地+建物)+車位</td>\n",
       "      <td>國中街91~120號</td>\n",
       "      <td>28.21</td>\n",
       "      <td>其他</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1080817</td>\n",
       "      <td>土地2建物1車位1</td>\n",
       "      <td>十三層</td>\n",
       "      <td>...</td>\n",
       "      <td>坡道平面</td>\n",
       "      <td>30.31</td>\n",
       "      <td>1100000</td>\n",
       "      <td>預售屋、或土地及建物分次登記案件;</td>\n",
       "      <td>RPXNMLOJRHPFFIF57DA</td>\n",
       "      <td>73.39</td>\n",
       "      <td>2.65</td>\n",
       "      <td>8.52</td>\n",
       "      <td>有</td>\n",
       "      <td>New_Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19815</th>\n",
       "      <td>三峽區</td>\n",
       "      <td>土地</td>\n",
       "      <td>成福段成福小段331~360地號</td>\n",
       "      <td>762</td>\n",
       "      <td>NaN</td>\n",
       "      <td>山坡地保育區</td>\n",
       "      <td>農牧用地</td>\n",
       "      <td>1090814</td>\n",
       "      <td>土地2建物0車位0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>親友、員工、共有人或其他特殊關係間之交易;</td>\n",
       "      <td>RPQNMLRLQHPFFIF27DA</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>無</td>\n",
       "      <td>New_Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19816</th>\n",
       "      <td>鶯歌區</td>\n",
       "      <td>房地(土地+建物)+車位</td>\n",
       "      <td>國中街91~120號</td>\n",
       "      <td>28.21</td>\n",
       "      <td>其他</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1080831</td>\n",
       "      <td>土地2建物1車位1</td>\n",
       "      <td>四層</td>\n",
       "      <td>...</td>\n",
       "      <td>坡道平面</td>\n",
       "      <td>30.31</td>\n",
       "      <td>1400000</td>\n",
       "      <td>預售屋、或土地及建物分次登記案件;</td>\n",
       "      <td>RPSPMLLLQHPFFIF07DA</td>\n",
       "      <td>73.40</td>\n",
       "      <td>2.64</td>\n",
       "      <td>8.52</td>\n",
       "      <td>有</td>\n",
       "      <td>New_Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19817</th>\n",
       "      <td>鶯歌區</td>\n",
       "      <td>房地(土地+建物)+車位</td>\n",
       "      <td>中山路301~330號</td>\n",
       "      <td>19.85</td>\n",
       "      <td>其他</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1090809</td>\n",
       "      <td>土地1建物1車位1</td>\n",
       "      <td>七層</td>\n",
       "      <td>...</td>\n",
       "      <td>坡道平面</td>\n",
       "      <td>24.14</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>RPUOMLRKQHPFFIF67DA</td>\n",
       "      <td>51.72</td>\n",
       "      <td>0.00</td>\n",
       "      <td>5.40</td>\n",
       "      <td>有</td>\n",
       "      <td>New_Taipei</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19818</th>\n",
       "      <td>鶯歌區</td>\n",
       "      <td>房地(土地+建物)+車位</td>\n",
       "      <td>國中街91~120號</td>\n",
       "      <td>26.94</td>\n",
       "      <td>其他</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1090225</td>\n",
       "      <td>土地2建物1車位1</td>\n",
       "      <td>十一層</td>\n",
       "      <td>...</td>\n",
       "      <td>坡道平面</td>\n",
       "      <td>33.05</td>\n",
       "      <td>1250000</td>\n",
       "      <td>預售屋、或土地及建物分次登記案件;</td>\n",
       "      <td>RPSOMLSJQHPFFIF47DA</td>\n",
       "      <td>69.20</td>\n",
       "      <td>2.42</td>\n",
       "      <td>8.52</td>\n",
       "      <td>有</td>\n",
       "      <td>New_Taipei</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>59476 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      鄉鎮市區          交易標的         土地區段位置建物區段門牌 土地移轉總面積平方公尺 都市土地使用分區 非都市土地使用分區  \\\n",
       "1      文山區  房地(土地+建物)+車位  臺北市文山區忠順街二段85巷1~30號       26.97        住       NaN   \n",
       "2      文山區     房地(土地+建物)     臺北市文山區福興路91~120號       10.32        住       NaN   \n",
       "3      文山區     房地(土地+建物)  臺北市文山區木柵路四段121~150號       39.64       其他       NaN   \n",
       "4      中正區     房地(土地+建物)   臺北市中正區和平西路一段31~60號        4.26        商       NaN   \n",
       "5      文山區     房地(土地+建物)    臺北市文山區樟新街56巷1~30號       44.57        住       NaN   \n",
       "...    ...           ...                  ...         ...      ...       ...   \n",
       "19814  鶯歌區  房地(土地+建物)+車位           國中街91~120號       28.21       其他       NaN   \n",
       "19815  三峽區            土地     成福段成福小段331~360地號         762      NaN    山坡地保育區   \n",
       "19816  鶯歌區  房地(土地+建物)+車位           國中街91~120號       28.21       其他       NaN   \n",
       "19817  鶯歌區  房地(土地+建物)+車位          中山路301~330號       19.85       其他       NaN   \n",
       "19818  鶯歌區  房地(土地+建物)+車位           國中街91~120號       26.94       其他       NaN   \n",
       "\n",
       "      非都市土地使用編定    交易年月日      交易筆棟數 移轉層次  ...  車位類別 車位移轉總面積(平方公尺)    車位總價元  \\\n",
       "1           NaN  1090406  土地1建物1車位2   四層  ...  坡道平面          40.7        0   \n",
       "2           NaN  1090317  土地1建物1車位0   三層  ...   NaN           0.0        0   \n",
       "3           NaN  1090418  土地2建物1車位0   六層  ...   NaN           0.0        0   \n",
       "4           NaN  1090413  土地2建物1車位0   九層  ...   NaN           0.0        0   \n",
       "5           NaN  1090328  土地3建物1車位0   四層  ...   NaN           0.0        0   \n",
       "...         ...      ...        ...  ...  ...   ...           ...      ...   \n",
       "19814       NaN  1080817  土地2建物1車位1  十三層  ...  坡道平面         30.31  1100000   \n",
       "19815      農牧用地  1090814  土地2建物0車位0  NaN  ...   NaN             0        0   \n",
       "19816       NaN  1080831  土地2建物1車位1   四層  ...  坡道平面         30.31  1400000   \n",
       "19817       NaN  1090809  土地1建物1車位1   七層  ...  坡道平面         24.14        0   \n",
       "19818       NaN  1090225  土地2建物1車位1  十一層  ...  坡道平面         33.05  1250000   \n",
       "\n",
       "                                 備註                   編號   主建物面積 附屬建物面積  \\\n",
       "1      含增建或未登記建物。;內含獨立產權車位，未包含在棟數內。  RPVOMLMKNHPFFAA18CA  110.91   9.63   \n",
       "2                       含增建或未登記建物。;  RPPPMLMKNHPFFAA67CA   24.58   1.28   \n",
       "3                               NaN  RPSPMLMKNHPFFAA97CA   78.80   4.15   \n",
       "4                       含增建或未登記建物。;  RPPOMLMKNHPFFAA57CA   18.34   0.00   \n",
       "5                       含增建或未登記建物。;  RPWNMLNKNHPFFAA28CA   98.74   2.26   \n",
       "...                             ...                  ...     ...    ...   \n",
       "19814             預售屋、或土地及建物分次登記案件;  RPXNMLOJRHPFFIF57DA   73.39   2.65   \n",
       "19815         親友、員工、共有人或其他特殊關係間之交易;  RPQNMLRLQHPFFIF27DA    0.00   0.00   \n",
       "19816             預售屋、或土地及建物分次登記案件;  RPSPMLLLQHPFFIF07DA   73.40   2.64   \n",
       "19817                           NaN  RPUOMLRKQHPFFIF67DA   51.72   0.00   \n",
       "19818             預售屋、或土地及建物分次登記案件;  RPSOMLSJQHPFFIF47DA   69.20   2.42   \n",
       "\n",
       "        陽台面積   電梯        city  \n",
       "1      11.09  NaN      Taipei  \n",
       "2       2.30  NaN      Taipei  \n",
       "3       8.76  NaN      Taipei  \n",
       "4       6.45  NaN      Taipei  \n",
       "5      11.84  NaN      Taipei  \n",
       "...      ...  ...         ...  \n",
       "19814   8.52    有  New_Taipei  \n",
       "19815   0.00    無  New_Taipei  \n",
       "19816   8.52    有  New_Taipei  \n",
       "19817   5.40    有  New_Taipei  \n",
       "19818   8.52    有  New_Taipei  \n",
       "\n",
       "[59476 rows x 33 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#4. 將以上四份資料運用pd.concat()串接\n",
    "all_data = pd.concat([data_a,data_b,data_e,data_f],axis=0)\n",
    "all_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. 資料清理與切片"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "因為我是想買來住的房子，所以幫忙刪除住宿用途以外的資料，並且限制\n",
    "1. 交易年月日，限制在109年\n",
    "2. 建物現況格局-房，1~5間 \n",
    "3. 建物現況格局-廳，1~2廳"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "以下columns_mapping、analysis_columns、columns_type提供給資料科學家配合使用\n",
    "1. 利用.rename()並配合以下columns_mapping，將原中文欄位名稱改為英文方便之後分析\n",
    "2. 取出主要用途(main_use)為'住家用'以及都市土地使用分區(use_zoning)為'住'的資料並針對以下欄位analysis_columns做分析並去除na值\n",
    "   (提示:先取完之後再.dropna())\n",
    "3. 觀察欄位資料型態，並利用.astype()搭配以下提供的columns_type做欄位型態轉換\n",
    "4. 做資料切片將<br>\n",
    "    新增欄位交易年月日(tx_dt_year)，從交易年月日(tx_dt)萃取出年份<br>\n",
    "    1.交易年月日(tx_dt_year)，限制在109年<br>\n",
    "    2.建物現況格局-房(room_number)，限制在1到5間<br>\n",
    "    3.建物現況格局-廳(hall_number)，限制在1到2廳<br>\n",
    "    4.最後運用.reset_index()重新定義索引"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "columns_mapping = {'鄉鎮市區':'towns',\n",
    "'交易標的':'transaction_sign',\n",
    "'土地區段位置建物區段門牌':'house_number',\n",
    "'土地移轉總面積平方公尺':'land_area_square_meter', \n",
    "'都市土地使用分區':'use_zoning', \n",
    "'非都市土地使用分區':'land_use_district',\n",
    "'非都市土地使用編定':'land_use',\n",
    "'交易年月日':'tx_dt', \n",
    " '交易筆棟數':'transaction_pen_number', \n",
    " '移轉層次':'shifting_level', \n",
    " '總樓層數':'total_floor_number', \n",
    " '建物型態':'building_state', \n",
    " '主要用途':'main_use', \n",
    " '主要建材':'main_materials',\n",
    " '建築完成年月':'complete_date', \n",
    " '建物移轉總面積平方公尺':'building_area_square_meter', \n",
    " '建物現況格局-房':'room_number', \n",
    " '建物現況格局-廳':'hall_number', \n",
    " '建物現況格局-衛':'health_number', \n",
    "'建物現況格局-隔間':'compartmented_number', \n",
    " '有無管理組織':'manages', \n",
    " '總價元':'total_price', \n",
    " '單價元平方公尺':'unit_price', \n",
    " '車位類別':'berth_category', \n",
    " '車位移轉總面積(平方公尺)':'berth_area_square_meter',\n",
    "'車位總價元':'berth_price', \n",
    " '備註':'note', \n",
    " '編號':'serial_number', \n",
    " '主建物面積':'main_building_area', \n",
    " '附屬建物面積':'auxiliary_building_area', \n",
    " '陽台面積':'balcony_area', \n",
    " '電梯':'elevator'\n",
    "                  }\n",
    "analysis_columns = ['city','towns','main_use','use_zoning','total_price','building_area_square_meter',\n",
    "                                     'main_building_area',\n",
    "                                     'tx_dt','unit_price','room_number','hall_number','health_number']\n",
    "columns_type = {'total_price': 'int','unit_price':'float','building_area_square_meter':'float',\n",
    "                                      'main_building_area': 'float',\n",
    "                                      'room_number': 'int','hall_number': 'int','health_number': 'int'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#1. 利用.rename()並配合以下columns_mapping，將原中文欄位名稱改為英文方便之後分析\n",
    "analysis_data = all_data.rename(columns=columns_mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "#2. 取出主要用途(main_use)為'住家用'以及都市土地使用分區(use_zoning)為'住'的資料並針對以下欄位analysis_columns做分析並去除na值 (提示:先取完之後再.dropna())\n",
    "#   (提示:先取完之後再.dropna())\n",
    "analysis_data = analysis_data.loc[(analysis_data.main_use=='住家用')&(analysis_data.use_zoning=='住'),analysis_columns].dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 7802 entries, 1 to 11512\n",
      "Data columns (total 12 columns):\n",
      " #   Column                      Non-Null Count  Dtype  \n",
      "---  ------                      --------------  -----  \n",
      " 0   city                        7802 non-null   object \n",
      " 1   towns                       7802 non-null   object \n",
      " 2   main_use                    7802 non-null   object \n",
      " 3   use_zoning                  7802 non-null   object \n",
      " 4   total_price                 7802 non-null   object \n",
      " 5   building_area_square_meter  7802 non-null   object \n",
      " 6   main_building_area          7802 non-null   float64\n",
      " 7   tx_dt                       7802 non-null   object \n",
      " 8   unit_price                  7802 non-null   object \n",
      " 9   room_number                 7802 non-null   object \n",
      " 10  hall_number                 7802 non-null   object \n",
      " 11  health_number               7802 non-null   object \n",
      "dtypes: float64(1), object(11)\n",
      "memory usage: 792.4+ KB\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 7802 entries, 1 to 11512\n",
      "Data columns (total 12 columns):\n",
      " #   Column                      Non-Null Count  Dtype  \n",
      "---  ------                      --------------  -----  \n",
      " 0   city                        7802 non-null   object \n",
      " 1   towns                       7802 non-null   object \n",
      " 2   main_use                    7802 non-null   object \n",
      " 3   use_zoning                  7802 non-null   object \n",
      " 4   total_price                 7802 non-null   int64  \n",
      " 5   building_area_square_meter  7802 non-null   float64\n",
      " 6   main_building_area          7802 non-null   float64\n",
      " 7   tx_dt                       7802 non-null   object \n",
      " 8   unit_price                  7802 non-null   float64\n",
      " 9   room_number                 7802 non-null   int64  \n",
      " 10  hall_number                 7802 non-null   int64  \n",
      " 11  health_number               7802 non-null   int64  \n",
      "dtypes: float64(3), int64(4), object(5)\n",
      "memory usage: 792.4+ KB\n"
     ]
    }
   ],
   "source": [
    "#3. 觀察欄位資料型態，並利用.astype()搭配以下提供的columns_type做欄位型態轉換\n",
    "analysis_data.info()\n",
    "analysis_data = analysis_data.astype(columns_type)\n",
    "analysis_data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#4. 做資料切片將\n",
    "#     新增欄位交易年月日(tx_dt_year)，從交易年月日(tx_dt)萃取出年份\n",
    "#     1.交易年月日(tx_dt_year)，限制在109年\n",
    "#     2.建物現況格局-房(room_number)，限制在1到5間\n",
    "#     3.建物現況格局-廳(hall_number)，限制在1到2廳\n",
    "#     4.最後運用.reset_index()重新定義索引\n",
    "analysis_data['tx_dt_year'] = analysis_data['tx_dt'].apply(lambda x: int(x[:-4]))\n",
    "analysis_data = analysis_data.loc[(analysis_data.tx_dt_year>=109)&\n",
    "                                  (analysis_data.room_number>=1)&\n",
    "                                  (analysis_data.room_number<=5)&\n",
    "                                  (analysis_data.hall_number>=1)&\n",
    "                                  (analysis_data.hall_number<=2)]\n",
    "analysis_data = analysis_data.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. 建立自定義特徵加入分析\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "以台灣來說大家都是以坪為單位計算使用面積，應該是不會問說你家有幾平方公尺吧?，但是偏偏資料中沒有以坪為單位計算，所以接下來請各位幫忙產生新的欄位以坪為單位計算面積，轉換公式我也幫你找好了，在下面的定義。<br>\n",
    "定義 : 1平方公尺相當於0.3025坪 <br>\n",
    "建立新特徵\n",
    "1. 建物移轉總面積坪(building_area_square_feet) : 建物移轉總面積平方公尺\\*0.3025\n",
    "2. 主建物面積坪(main_building_area_square_feet) : 主建物面積\\*0.3025\n",
    "3. 單價元坪(unit_price_square_feet) : 單價元平方公尺/0.3025"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "可以利用.describe()做一下資料觀察，是否有奇怪的資料?<br>\n",
    "如果有請將資料移除，並說明為什麼移除此資料?<br>\n",
    "移除後再用.describe()觀察是否還有奇怪的資料?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 建立新特徵\n",
    "# 1. 建物移轉總面積坪(building_area_square_feet) : 建物移轉總面積平方公尺*0.3025\n",
    "# 2. 主建物面積坪(main_building_area_square_feet) : 主建物面積*0.3025\n",
    "# 3. 單價元坪(unit_price_square_feet) : 單價元平方公尺/0.3025\n",
    "analysis_data['building_area_square_feet'] = analysis_data.building_area_square_meter*0.3025\n",
    "analysis_data['main_building_area_square_feet'] = analysis_data.main_building_area*0.3025\n",
    "analysis_data['unit_price_square_feet'] = analysis_data.unit_price/0.3025"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "scrolled": true
   },
   "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>total_price</th>\n",
       "      <th>building_area_square_meter</th>\n",
       "      <th>main_building_area</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>room_number</th>\n",
       "      <th>hall_number</th>\n",
       "      <th>health_number</th>\n",
       "      <th>tx_dt_year</th>\n",
       "      <th>building_area_square_feet</th>\n",
       "      <th>main_building_area_square_feet</th>\n",
       "      <th>unit_price_square_feet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>6.894000e+03</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6894.0</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6894.000000</td>\n",
       "      <td>6.894000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.138712e+07</td>\n",
       "      <td>121.712089</td>\n",
       "      <td>82.086667</td>\n",
       "      <td>94364.671163</td>\n",
       "      <td>2.850885</td>\n",
       "      <td>1.793299</td>\n",
       "      <td>1.733536</td>\n",
       "      <td>109.0</td>\n",
       "      <td>36.817907</td>\n",
       "      <td>24.831217</td>\n",
       "      <td>3.119493e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.159445e+07</td>\n",
       "      <td>62.871712</td>\n",
       "      <td>37.363166</td>\n",
       "      <td>58317.311227</td>\n",
       "      <td>0.898498</td>\n",
       "      <td>0.404969</td>\n",
       "      <td>0.681085</td>\n",
       "      <td>0.0</td>\n",
       "      <td>19.018693</td>\n",
       "      <td>11.302358</td>\n",
       "      <td>1.927845e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.970000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>0.293425</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.580000e+06</td>\n",
       "      <td>82.965000</td>\n",
       "      <td>60.682500</td>\n",
       "      <td>53532.250000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>25.096912</td>\n",
       "      <td>18.356456</td>\n",
       "      <td>1.769661e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>8.550000e+06</td>\n",
       "      <td>109.390000</td>\n",
       "      <td>78.990000</td>\n",
       "      <td>79610.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>33.090475</td>\n",
       "      <td>23.894475</td>\n",
       "      <td>2.631736e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.329500e+07</td>\n",
       "      <td>147.517500</td>\n",
       "      <td>95.797500</td>\n",
       "      <td>118728.250000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>44.624044</td>\n",
       "      <td>28.978744</td>\n",
       "      <td>3.924901e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.250000e+08</td>\n",
       "      <td>1444.950000</td>\n",
       "      <td>335.760000</td>\n",
       "      <td>610104.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>437.097375</td>\n",
       "      <td>101.567400</td>\n",
       "      <td>2.016873e+06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        total_price  building_area_square_meter  main_building_area  \\\n",
       "count  6.894000e+03                 6894.000000         6894.000000   \n",
       "mean   1.138712e+07                  121.712089           82.086667   \n",
       "std    1.159445e+07                   62.871712           37.363166   \n",
       "min    0.000000e+00                    0.970000            0.000000   \n",
       "25%    5.580000e+06                   82.965000           60.682500   \n",
       "50%    8.550000e+06                  109.390000           78.990000   \n",
       "75%    1.329500e+07                  147.517500           95.797500   \n",
       "max    2.250000e+08                 1444.950000          335.760000   \n",
       "\n",
       "          unit_price  room_number  hall_number  health_number  tx_dt_year  \\\n",
       "count    6894.000000  6894.000000  6894.000000    6894.000000      6894.0   \n",
       "mean    94364.671163     2.850885     1.793299       1.733536       109.0   \n",
       "std     58317.311227     0.898498     0.404969       0.681085         0.0   \n",
       "min         0.000000     1.000000     1.000000       0.000000       109.0   \n",
       "25%     53532.250000     2.000000     2.000000       1.000000       109.0   \n",
       "50%     79610.000000     3.000000     2.000000       2.000000       109.0   \n",
       "75%    118728.250000     3.000000     2.000000       2.000000       109.0   \n",
       "max    610104.000000     5.000000     2.000000       6.000000       109.0   \n",
       "\n",
       "       building_area_square_feet  main_building_area_square_feet  \\\n",
       "count                6894.000000                     6894.000000   \n",
       "mean                   36.817907                       24.831217   \n",
       "std                    19.018693                       11.302358   \n",
       "min                     0.293425                        0.000000   \n",
       "25%                    25.096912                       18.356456   \n",
       "50%                    33.090475                       23.894475   \n",
       "75%                    44.624044                       28.978744   \n",
       "max                   437.097375                      101.567400   \n",
       "\n",
       "       unit_price_square_feet  \n",
       "count            6.894000e+03  \n",
       "mean             3.119493e+05  \n",
       "std              1.927845e+05  \n",
       "min              0.000000e+00  \n",
       "25%              1.769661e+05  \n",
       "50%              2.631736e+05  \n",
       "75%              3.924901e+05  \n",
       "max              2.016873e+06  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 可以利用.describe()做一下資料觀察，是否有奇怪的資料?\n",
    "# 如果有請將資料移除，並說明為什麼移除此資料?\n",
    "# 觀察到有總價total_price最小值為0還有主建物面積main_building_area最小為0的，進一步將其拿掉\n",
    "\n",
    "analysis_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>total_price</th>\n",
       "      <th>building_area_square_meter</th>\n",
       "      <th>main_building_area</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>room_number</th>\n",
       "      <th>hall_number</th>\n",
       "      <th>health_number</th>\n",
       "      <th>tx_dt_year</th>\n",
       "      <th>building_area_square_feet</th>\n",
       "      <th>main_building_area_square_feet</th>\n",
       "      <th>unit_price_square_feet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>6.854000e+03</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6854.0</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6854.000000</td>\n",
       "      <td>6.854000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.141669e+07</td>\n",
       "      <td>121.768404</td>\n",
       "      <td>82.565725</td>\n",
       "      <td>94548.453020</td>\n",
       "      <td>2.849869</td>\n",
       "      <td>1.793114</td>\n",
       "      <td>1.733149</td>\n",
       "      <td>109.0</td>\n",
       "      <td>36.834942</td>\n",
       "      <td>24.976132</td>\n",
       "      <td>3.125569e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.161669e+07</td>\n",
       "      <td>62.849105</td>\n",
       "      <td>36.940424</td>\n",
       "      <td>58374.690321</td>\n",
       "      <td>0.898343</td>\n",
       "      <td>0.405103</td>\n",
       "      <td>0.680869</td>\n",
       "      <td>0.0</td>\n",
       "      <td>19.011854</td>\n",
       "      <td>11.174478</td>\n",
       "      <td>1.929742e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.100000e+05</td>\n",
       "      <td>7.440000</td>\n",
       "      <td>7.440000</td>\n",
       "      <td>3239.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>2.250600</td>\n",
       "      <td>2.250600</td>\n",
       "      <td>1.070744e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.600000e+06</td>\n",
       "      <td>83.012500</td>\n",
       "      <td>61.062500</td>\n",
       "      <td>53597.250000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>25.111281</td>\n",
       "      <td>18.471406</td>\n",
       "      <td>1.771810e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>8.600000e+06</td>\n",
       "      <td>109.405000</td>\n",
       "      <td>79.110000</td>\n",
       "      <td>79742.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>33.095012</td>\n",
       "      <td>23.930775</td>\n",
       "      <td>2.636116e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.330000e+07</td>\n",
       "      <td>147.445000</td>\n",
       "      <td>95.970000</td>\n",
       "      <td>118880.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>44.602112</td>\n",
       "      <td>29.030925</td>\n",
       "      <td>3.929917e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.250000e+08</td>\n",
       "      <td>1444.950000</td>\n",
       "      <td>335.760000</td>\n",
       "      <td>610104.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>109.0</td>\n",
       "      <td>437.097375</td>\n",
       "      <td>101.567400</td>\n",
       "      <td>2.016873e+06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        total_price  building_area_square_meter  main_building_area  \\\n",
       "count  6.854000e+03                 6854.000000         6854.000000   \n",
       "mean   1.141669e+07                  121.768404           82.565725   \n",
       "std    1.161669e+07                   62.849105           36.940424   \n",
       "min    4.100000e+05                    7.440000            7.440000   \n",
       "25%    5.600000e+06                   83.012500           61.062500   \n",
       "50%    8.600000e+06                  109.405000           79.110000   \n",
       "75%    1.330000e+07                  147.445000           95.970000   \n",
       "max    2.250000e+08                 1444.950000          335.760000   \n",
       "\n",
       "          unit_price  room_number  hall_number  health_number  tx_dt_year  \\\n",
       "count    6854.000000  6854.000000  6854.000000    6854.000000      6854.0   \n",
       "mean    94548.453020     2.849869     1.793114       1.733149       109.0   \n",
       "std     58374.690321     0.898343     0.405103       0.680869         0.0   \n",
       "min      3239.000000     1.000000     1.000000       0.000000       109.0   \n",
       "25%     53597.250000     2.000000     2.000000       1.000000       109.0   \n",
       "50%     79742.500000     3.000000     2.000000       2.000000       109.0   \n",
       "75%    118880.000000     3.000000     2.000000       2.000000       109.0   \n",
       "max    610104.000000     5.000000     2.000000       6.000000       109.0   \n",
       "\n",
       "       building_area_square_feet  main_building_area_square_feet  \\\n",
       "count                6854.000000                     6854.000000   \n",
       "mean                   36.834942                       24.976132   \n",
       "std                    19.011854                       11.174478   \n",
       "min                     2.250600                        2.250600   \n",
       "25%                    25.111281                       18.471406   \n",
       "50%                    33.095012                       23.930775   \n",
       "75%                    44.602112                       29.030925   \n",
       "max                   437.097375                      101.567400   \n",
       "\n",
       "       unit_price_square_feet  \n",
       "count            6.854000e+03  \n",
       "mean             3.125569e+05  \n",
       "std              1.929742e+05  \n",
       "min              1.070744e+04  \n",
       "25%              1.771810e+05  \n",
       "50%              2.636116e+05  \n",
       "75%              3.929917e+05  \n",
       "max              2.016873e+06  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 移除後再用.describe()觀察是否還有奇怪的資料?\n",
    "# 否\n",
    "analysis_data = analysis_data.loc[(analysis_data.total_price!=0)&(analysis_data.main_building_area!=0)]\n",
    "analysis_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. 找出台北市時價登入總價高度相關的變數"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "阿宏我是台北人他想找出影響台北市總價、單價元坪的因子<br>\n",
    "相關係數0.3以下為低相關，0.3\\~0.7為中等相關，0.7以上為高度相關<br>\n",
    "1. 資料切片切出city欄位為台北市的資料，並找出時價登入總價(total_price)高度相關的變數\n",
    "2. 資料切片切出city欄位為台北市的資料，找出單價元坪(unit_price_square_feet)高度相關的變數"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>total_price</th>\n",
       "      <th>unit_price_square_feet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>total_price</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.581290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>building_area_square_meter</th>\n",
       "      <td>0.864627</td>\n",
       "      <td>0.218185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>main_building_area</th>\n",
       "      <td>0.710515</td>\n",
       "      <td>0.112488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unit_price</th>\n",
       "      <td>0.581290</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>room_number</th>\n",
       "      <td>0.326441</td>\n",
       "      <td>-0.100062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hall_number</th>\n",
       "      <td>0.270787</td>\n",
       "      <td>0.004883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>health_number</th>\n",
       "      <td>0.598832</td>\n",
       "      <td>0.128217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tx_dt_year</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>building_area_square_feet</th>\n",
       "      <td>0.864627</td>\n",
       "      <td>0.218185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>main_building_area_square_feet</th>\n",
       "      <td>0.710515</td>\n",
       "      <td>0.112488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unit_price_square_feet</th>\n",
       "      <td>0.581290</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                total_price  unit_price_square_feet\n",
       "total_price                        1.000000                0.581290\n",
       "building_area_square_meter         0.864627                0.218185\n",
       "main_building_area                 0.710515                0.112488\n",
       "unit_price                         0.581290                1.000000\n",
       "room_number                        0.326441               -0.100062\n",
       "hall_number                        0.270787                0.004883\n",
       "health_number                      0.598832                0.128217\n",
       "tx_dt_year                              NaN                     NaN\n",
       "building_area_square_feet          0.864627                0.218185\n",
       "main_building_area_square_feet     0.710515                0.112488\n",
       "unit_price_square_feet             0.581290                1.000000"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 阿宏我是台北人他想找出影響台北市總價、單價元坪的因子\n",
    "# 1. 資料切片切出city欄位為台北市的資料，並找出時價登入總價(total_price)高度相關的變數\n",
    "# 建物移轉總面積平方公尺、主建物面積、建物移轉總面積坪、主建物面積坪\n",
    "# 2. 資料切片切出city欄位為台北市的資料，找出單價元坪(unit_price_square_feet)高度相關的變數\n",
    "# 單價元平方公尺\n",
    "analysis_data.loc[analysis_data.city=='Taipei'].corr()[['total_price','unit_price_square_feet']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. 資料視覺化並解釋\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "我想要以視覺化的方式來看房價資料，並且請各位資料科學家解釋圖表給阿宏我知道\n",
    "1. 以城市(city)為x軸，以單價元坪(unit_price_square_feet)為y軸畫出boxplot，並找出單價元坪(unit_price_square_feet)中位數最高的地區\n",
    "2. 進一步對台北市的資料做圖，以建物現況格局-房(room_number)為x軸，以總價元(total_price)為y軸畫出boxplot，並找出總價元(total_price)中位數最高的房間數。hint:資料切片找出city欄位為台北市的資料，再進一步畫圖\n",
    "3. 對台北市的資料做圖，先將地區(twons)做編碼在進行，再以地區(twon)為x軸，以單價元坪(unit_price_square_feet)為y軸畫出boxplot，並找出單價元坪(unit_price_square_feet)中位數最高的地區。hint:運用LabelEncoder()對地區(twons)做編碼，運用.inverse_transform()反查編碼的地區"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'unit_price_square_feet'}, xlabel='city'>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA8UAAAGUCAYAAAAcf4AXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABErUlEQVR4nO3de5xdVXn4/8+TmYQI4Vo0QiKXFttOGOst1UqnX2eMclELWqsw4IVmhNKWlFYlotNvvXWUy8/6VbBF6KRAK0fUVqQGBEpmakdq5VK0kKktQpCJiNwhISSZyfP74+xJT4YkM4GZ7DNzPu/X67zm7LUv6znnZGef56y114rMRJIkSZKkRjSr7AAkSZIkSSqLSbEkSZIkqWGZFEuSJEmSGpZJsSRJkiSpYZkUS5IkSZIalkmxJEmSJKlhmRRLkrQDEZERcUTZcZQpItojYmgn66f8PYqIuyKifSrrkCQ1LpNiSVLdi4g1EbEhItZFxGMRsTIiXlJ2XKMi4tSIGCg7jpkqM4/MzH6AiPh4RPx9ySFJkmYQk2JJ0nTx25k5DzgIeBC4sOR4pkxENJcdgyRJjcKkWJI0rWTmM8DXgUWjZRGxb0RcEREPRcR9EfFnETErIg6IiKGI+O1iu3kRcXdEvLdYviwiLo6IGyPiqYj4l4g4dHv17qSOFuBi4HVFS/bjO9j/8Ij4TlHPP0fEF0dbPCPisKIbcldE/ARYVRz7z4q6fl7UvW+x/bO6NBet6W8snn88Ir4eEVcV9d0eES+v2fbgiPiH4rXcGxF/XLPuBcX78lhErAZ+fQIfy5sj4p6IeDgiLihinxMRj0bEy2qO/aKIeDoiXriD9+i0iBgsYl4dEa+qfW0RcSzwUeDE4r3+QUS8MyJuG3OcD0TENycQtyRJJsWSpOklIvYETgS+V1N8IbAv8IvA64H3Ar+XmY8CS4FLI+JFwOeAOzLzipp9TwE+BRwI3AF8eQdV76iOQeAM4N8yc15m7reD/a8Evg/8AvBx4D3b2eb1QAtwDHBq8ego6pwHXLSDY2/PCcDXgAOKuq+OiNkRMQv4J+AHwAJgCfAnEXFMsd/HgF8qHscA75tAXW8HFgOvKupdmpmbgK8A767ZrhO4KTMfGnuAiHgn1fflvcA+wPHAI7XbZOa3gU8DVxXv9cuBa4DDix8nRr0HqP2MJUnaIZNiSdJ0cXXRCvsE8CbgAoCIaAJOAj6SmU9l5hrgsxRJZ2beQDU5vAl4M/D7Y467MjO/k5kbgW6qLb7b3K88Xh3jiYhDqLa4/nlmbsrMAarJ3Fgfz8z1mbmBarL+l5l5T2auAz4CnLQLXatvy8yvZ+Zm4C+BucBvFHG8MDM/WcRyD3Bp8foA3gX0ZOajmXk/8IUJ1HVesf1PgP9HNfkFuBzojIgolt8D/N0OjvF+4PzMvCWr7s7M+8aruPjcrqJIviPiSOAw4FsTiFuSJJNiSdK08baiFXYucCbwLxHxYqotvLOB2gTqPqqtoKMuAVqByzJzm9ZH4P7RJ0Xy+Shw8JhtJlLHzhwMPJqZT2+v3h2UHbyd+pqB+ROss/Z1bQGGimMeChwcEY+PPqh2SR497sFj4hg3Md3O9gcX9f478DTQHhG/ChzB9n8MAHgJ8OMJ1LU9lwMnF8n3e4CvFsmyJEnjMimWJE0rmTmSmf8IjABtwMPAZqrJ3qhDgLWwtZX3Eqrdaf8wnj190NZW4YiYR7W78U/HbLPTOoAcJ+wHgAOKrt/Pqrf25dU8/+l26humOsjYemDrsYrXOPY+3drXNQtYWBzzfuDezNyv5rF3Zr65Jtba2A4Z57WNfS2HsO37dznVVtz3AF8v7gnfnvupdtkez7Pe68z8HrAJ+C3gZHbcGi1J0rOYFEuSppWoOgHYHxjMzBHgq0BPROxdDJT1AWB02p6PUk2kllLtcn1FkUSOenNEtEXEHKr3Fn+v6Da81QTqeBBYWBzjWYpuwLcCHy8GoHod8NvjvNQK8KfFAF3z+N97aYeB/wbmRsRbImI28GfAHmP2f3VE/E7R3fpPgI1U78P+PvBURHy4GFSrKSJaI2J0QK2vAh+JiP0jYiGwbJw4Ac4utn8JcBbV7syj/p7qPcfvZuf3+f4N8KGIeHXxGR8R2x/07EHgsCLRr3UF1XuuNxfd0yVJmhCTYknSdPFPEbEOeBLoAd6XmXcV65ZRbT29BxigOrDUioh4NdXk9b1FYnse1QT5nJrjXkl1cKlHgVez7cBQtbZbR7FuFXAX8LOIeHgH+58CvI7q4FF/QTVx3FkX3xVUWzy/A9wLPFPEQGY+Afwh1URybRHX0Jj9v0l1QLLHqLbS/k5mbi7eh7cCryiO+3BxnH2L/T5BtQv0vcANTKzV9ZvAbVQHKlsJ9I6uKH5guJ3q+/6vOzpAZn6N6ud6JfAUcDXVVvuxvlb8fSQibq8p/zuqXeSdw1iStEsic7weX5IkzUwRcRkwlJl/VkLdVwH/lZkfm4Jjfxw4IjN3lODvVhGxAvjpVL7PEfEC4OfAqzLzf6aqHknSzDPRESwlSdLzUHRPfpRqC+zRVKcuOrfUoHaDiDgM+B3glVNc1R8At5gQS5J2lUmxJEm7x4uBf6Q6T/EQ8AeZ+R/lhjS1IuJTwJ8Cn8nMe6ewnjVAAG+bqjokSTOX3aclSZIkSQ3LgbYkSZIkSQ3LpFiSJEmS1LBMiiVJkiRJDcukWJIkSZLUsEyKJUmSJEkNy6RYktQwIuLiiPi/u6muuyKifXfUNd1ExF9ExMMR8bOyY5EkySmZJEkNqUhY/z4zF5YcSkOJiEOAHwGHZubPn+exTgXen5ltkxGbJKkx2VIsSdIkiojmsmPYHZ7H6zwEeOT5JsSSJE0Wk2JJ0rQSERkRR9QsXxYRf1E8b4+IoYj4YET8PCIeiIjfG7ttROwFXAccHBHrisfBO6nz4xHx9Yi4KiKeiojbI+LlNevXRMSHI+KHwPqIaC7K3lisb4qIj0bEj4v9b4uIlxTrfjUiboyIRyPiRxHxrgm8B2+OiNXFsdZGxIdq1p1dvO6fRsTS2vcrIvoj4v01254aEQM1y5+PiPsj4skixt/aznvw9xHxJHBqROwbEb1FfWuL97ZpJ3G/Ebix5n2/rCj/jYi4OSIej4gf1HY731EdEdECXAy8rjjW4+O9b5IkbY9JsSRppnkxsC+wAOgCvhgR+9dukJnrgeOAn2bmvOLx03GOewLwNeAA4Erg6oiYXbO+E3gLsF9mDo/Z9wPF+jcD+wBLgaeL5PzG4ngvAk4C/ioiFo0TSy/w+5m5N9AKrAKIiGOBDwFvAl4KvHGc44x1C/CKmtf4tYiYW7P+BODrwH7Al4HLgGHgCOCVwNHA+9mBzPxntn3fT42IBcBK4C+Kej8E/ENEvLDYbbt1ZOYgcAbwb8Wx9tvF1ypJElDHSXFErCh+5b9zgtu/q/jV/K6IuHKq45Mk1a3NwCczc3NmXgusA35lEo57W2Z+PTM3A38JzAV+o2b9FzLz/szcsJ193w/8WWb+KKt+kJmPAG8F1mTm32bmcGb+B/APwDvHiWUzsCgi9snMxzLz9qL8XcDfZuadReL/8V15gZn595n5SBHLZ4E92Pa9+7fMvDozt1BN7t8M/Elmri+6Q3+OamK/K94NXJuZ12bmlsy8EbgVeHNEzJ+kOiRJ2qG6TYqp/jJ87EQ2jIiXAh8BfjMzjwT+ZOrCkiTVuUfGtNQ+DcybhOPeP/qkSAqHgIO3t347XgL8eDvlhwKvLboNP150AT6Famv3zryDarJ4X0T8S0S8rig/eEwc941znG1ExIciYjAinihi2Rc4sGaT2mMfCswGHqiJ/UtUW7x3xaHAO8e8B23AQZNYhyRJO1S3g4Fk5nci4rDasoj4JeCLwAupfsk5LTP/CzgN+GJmPlbs6+AdkjRzPQ3sWbP8YqoJ6q7a1ekXXjL6JCJmAQuB2i7XOzve/cAvAWN7P90P/EtmvmlXAsnMW4ATiu7bZwJfLeJ7oDZOqoNa1VrPs987AIr7h5cDS4C7MnNLRDwGRG3VY2LfCBy4ne7iu+J+4O8y87SxKyLioHHqcAoNSdLzVs8txdtzCbAsM19N9Z6jvyrKfxn45Yj4bkR8r7inSpI0M90BnFwMtnQs8PrneJwHgV+IiH0nuP2rI+J3ojrq8p9QTda+N8F9/wb4VES8NKp+LSJ+AfgW1evXeyJidvH49WIQqe2KiDkRcUpE7Ft05X4S2FKs/irVAbAWRcSewMfG7H4H8DsRsWcx+FZXzbq9qd67+xDQHBF/TrWL9HZl5gPADcBnI2KfiJgVEb8UEbv6efw98NsRcUzxmc6N6oBpCydQx4PAwoiYs4t1SpK01bRJiiNiHnAU1UE/7qDafeqgYnUz1QFF2qkOZHJpROy3+6OUJO0GZwG/DTxOtavx1c/lIEVPowpwT9E1d4ejTxe+CZwIPAa8B/idIimdiL+kmrDeQDWJ7QVekJlPUR046iSqrc4/A86jei/vzrwHWFOMAn0G1feBzLwO+H9UB966u/hb63PAJqrJ5OVUB8sadT3wbeC/qXa7foaddwkHeC8wB1hN9X35Ov97bZ6QzLyf6gBeH6WakN8PnM3/fkfZWR2rgLuAn0XEw7tSryRJoyKzfnseFd2nv5WZrRGxD/CjzHzWxTYiLgb+PTP/tli+CTin6F4mSdLzEhEfB47IzHeXHcuuiogEXpqZd5cdiyRJ9WjatBRn5pPAvRHxToCi+9noHJFXU20lJiIOpNqd+p4SwpQkSZIkTSN1mxRHRAX4N+BXImIoIrqodg/riogfUO0udUKx+fXAIxGxGugDzi6mupAkaUIi4rqIWLedx0dLiueuHcRzShnx7IqIuHgHsV9cdmySJI1V192nJUmSJEmaSnXbUixJkiRJ0lSry3mKDzzwwDzssMPKDkOTYP369ey1115lhyGphuelVJ88N6X643k5c9x2220PZ+YLt7euLpPiww47jFtvvbXsMDQJ+vv7aW9vLzsMSTU8L6X65Lkp1R/Py5kjIu7b0Tq7T0uSJEmSGpZJsSRJkiSpYZkUS5IkSZIalkmxJEmSJKlhmRRLkiRJkhqWSbEkSZIkqWGZFEuSJEmSGpZJsSRJkiSpYZkUS5IkSZIalkmxJEmSJKlhmRRLkiSVqFKp0NraypIlS2htbaVSqZQdkiQ1lOayA5AkSWpUlUqF7u5uent7GRkZoampia6uLgA6OztLjk6SGsO4LcUR8ZKI6IuI1RFxV0SctZ1tIiK+EBF3R8QPI+JVNeveFxH/UzzeN9kvQJIkabrq6emht7eXjo4Ompub6ejooLe3l56enrJDk6SGMZGW4mHgg5l5e0TsDdwWETdm5uqabY4DXlo8Xgv8NfDaiDgA+BiwGMhi32sy87FJfRWSJEnT0ODgIG1tbduUtbW1MTg4WFJEktR4xm0pzswHMvP24vlTwCCwYMxmJwBXZNX3gP0i4iDgGODGzHy0SIRvBI6d1FcgSZI0TbW0tDAwMLBN2cDAAC0tLSVFJEmNZ5fuKY6Iw4BXAv8+ZtUC4P6a5aGibEfl2zv26cDpAPPnz6e/v39XQlOdWrdunZ+lVGc8L6X68fa3v51TTjmFs88+m8MPP5zPfe5zXHDBBXR1dXmeSnXAa2ZjmHBSHBHzgH8A/iQzn5zsQDLzEuASgMWLF2d7e/tkV6ES9Pf342cp1RfPS6l+tLe3s2jRInp6ehgcHKSlpYXPfvazDrIl1QmvmY1hQklxRMymmhB/OTP/cTubrAVeUrO8sChbC7SPKe9/LoFKkiTNRJ2dnXR2dvrlW5JKMpHRpwPoBQYz8y93sNk1wHuLUah/A3giMx8ArgeOjoj9I2J/4OiiTJIkSZKk0k2kpfg3gfcA/xkRdxRlHwUOAcjMi4FrgTcDdwNPA79XrHs0Ij4F3FLs98nMfHTSopckSZIk6XkYNynOzAEgxtkmgT/awboVwIrnFJ0kSZIkSVNo3O7TkiRJkiTNVCbFkiRJkqSGZVIsSZIkSWpYJsWSJEmSpIZlUixJkiRJalgmxZIkSZKkhmVSLEkNolKp0NraypIlS2htbaVSqZQdkiRJUunGnadYkjT9VSoVuru76e3tZWRkhKamJrq6ugDo7OwsOTpJkqTy2FIsSQ2gp6eH3t5eOjo6aG5upqOjg97eXnp6esoOTZIkqVQmxZLUAAYHB2lra9umrK2tjcHBwZIikiRJqg8mxZLUAFpaWhgYGNimbGBggJaWlpIikiRJqg8mxZLUALq7u+nq6qKvr4/h4WH6+vro6uqiu7u77NAkSZJK5UBbktQARgfTWrZsGYODg7S0tNDT0+MgW5IkqeHZUixJDaKzs5M777yTm266iTvvvNOEWKoTTpcmSeWypViSJKkkTpcmSeWzpViSJKkkTpcmSeUzKZYkSSqJ06VJUvlMiiVJkkridGmSVD6TYkmSpJI4XZoklc+BtiRJkkridGmSVD6TYkmSpBJ1dnbS2dlJf38/7e3tZYcjSQ3H7tOSJEmSpIZlUixJkiRJalgmxZIkSZKkhmVSLEmSJElqWCbFkiRJkqSGZVIsSQ2iUqnQ2trKkiVLaG1tpVKplB2SJElS6ZySSZIaQKVSobu7m97eXkZGRmhqaqKrqwvA+VAlSVJDG7elOCJWRMTPI+LOHaw/OyLuKB53RsRIRBxQrFsTEf9ZrLt1soOXJE1MT08Pvb29dHR00NzcTEdHB729vfT09JQdmiRJUqkm0n36MuDYHa3MzAsy8xWZ+QrgI8C/ZOajNZt0FOsXP69IJUnP2eDgIG1tbduUtbW1MTg4WFJEkiRJ9WHcpDgzvwM8Ot52hU7Am9Qkqc60tLQwMDCwTdnAwAAtLS0lRSRJklQfIjPH3yjiMOBbmdm6k232BIaAI0ZbiiPiXuAxIIEvZeYlO9n/dOB0gPnz57/6K1/5yi68DNWrdevWMW/evLLDkBreTTfdRG9vL2effTaHH3449957LxdccAFdXV0sWbKk7PAk4TVTqkeelzNHR0fHbTvqvTyZA239NvDdMV2n2zJzbUS8CLgxIv6raHl+liJhvgRg8eLF2d7ePomhqSz9/f34WUrla29vZ9GiRfT09DA4OEhLSwuf/exnHWRLqiNeM6X643nZGCYzKT6JMV2nM3Nt8ffnEfEN4DXAdpNiSdLU6uzspLOz0wu8JElSjUmZpzgi9gVeD3yzpmyviNh79DlwNLDdEawlSZIkSSrDuC3FEVEB2oEDI2II+BgwGyAzLy42eztwQ2aur9l1PvCNiBit58rM/PbkhS5JkiRJ0vMzblKcmePecJaZl1Gduqm27B7g5c81MEmSJEmSptqkdJ+WJEmSJGk6MimWJEmSJDUsk2JJkiRJUsMyKZYkSZIkNSyTYkmSJElSwzIpliRJkiQ1LJNiSZIkSVLDMimWJEmSJDUsk2JJkiRJUsMyKZYkSZIkNSyTYkmSJElSwzIpliRJkiQ1LJNiSZIkSVLDMimWJEmSJDUsk2JJkiRJUsMyKZYkSZIkNSyTYkmSJElSwzIpliRJkiQ1LJNiSZIkSVLDMimWJEmSJDUsk2JJkiRJUsMyKZYkSZKkGpVKhdbWVpYsWUJrayuVSqXskDSFTIolSZIkqVCpVDjrrLNYv349mcn69es566yzTIxnMJNiSZIkSSosX76cpqYmVqxYwQ033MCKFStoampi+fLlZYemKWJSLEmSJEmFoaEhrrjiCjo6Omhubqajo4MrrriCoaGhskPTFDEplqQG4f1RkiRJz9ZcdgCSpKlXqVTo7u6mt7eXkZERmpqa6OrqAqCzs7Pk6CRJqh8LFy7kfe97H1/+8pcZGRmhr6+P973vfSxcuLDs0DRFxm0pjogVEfHziLhzB+vbI+KJiLijePx5zbpjI+JHEXF3RJwzmYFLkiaup6eH3t7ebbqC9fb20tPTU3ZokiTVlfPPP5/h4WGWLl3KMcccw9KlSxkeHub8888vOzRNkYl0n74MOHacbf41M19RPD4JEBFNwBeB44BFQGdELHo+wUqSnpvBwUHa2tq2KWtra2NwcLCkiCRJqk+dnZ2ceOKJPPDAA2zZsoUHHniAE0880Z5VM9i4SXFmfgd49Dkc+zXA3Zl5T2ZuAr4CnPAcjiNJep5aWloYGBjYpmxgYICWlpaSIpIkqT5VKhWuuuoqDjroICKCgw46iKuuusqxOGawyRpo63UR8YOIuC4ijizKFgD312wzVJRJknaz7u5uurq66OvrY3h4mL6+Prq6uuju7i47NEmS6opTMjWeyRho63bg0MxcFxFvBq4GXrqrB4mI04HTAebPn09/f/8khKayrVu3zs9SqgMHHXQQp5xyCkuXLuUnP/kJhxxyCO9+97s56KCDPEelOuE1U6oPQ0NDnHzyydtcM9/whjdw5ZVXeo7OUJGZ428UcRjwrcxsncC2a4DFVBPjj2fmMUX5RwAy8zPjHWPx4sV56623jhuX6l9/fz/t7e1lhyGphuelVJ88N6X6EBHMnz+fSqWydcaGzs5OHnzwQSaSO6k+RcRtmbl4e+ued/fpiHhxRETx/DXFMR8BbgFeGhGHR8Qc4CTgmudbnyRJkiRNlebmZjZv3rxN2ebNm2ludjbbmWrcTzYiKkA7cGBEDAEfA2YDZObFwO8CfxARw8AG4KSs/oQyHBFnAtcDTcCKzLxrSl6FJEmSJE2C0dbh2u7TTU1NjIyMlB2apsi4SXFm7nTs8cy8CLhoB+uuBa59bqFJkiRJ0u61aNEi3va2t3H11VcDsNdee3HKKadsXdbMYx8ASZIkSSp0d3fT3d1Nb2/v1lbjrq4uenp6yg5NU8SkWJIkSZIKnZ2d3HzzzRx33HFs3LiRPfbYg9NOO43Ozp12oNU0ZlIsSZIkSYVKpcLKlSu57rrrtmkpPuqoo0yMZ6jnPfq0JEmSJM0UPT099Pb20tHRQXNzMx0dHfT29tp9egYzKZYkSZKkwuDgIG1tbduUtbW1MTg4WFJEmmomxZIkSZJUaGlpYWBgYJuygYEBWlpaSopIU82kWJIkSZIK3d3ddHV10dfXx/DwMH19fXR1ddHd3V12aJoiDrQlSZIkSYXRwbSWLVvG4OAgLS0t9PT0OMjWDGZLsSRJkiSpYZkUS5IkSVKhUqlw1llnsX79ejKT9evXc9ZZZ1GpVMoOTVPEpFiSJEmSCsuXL6epqYkVK1Zwww03sGLFCpqamli+fHnZoWmKmBRLkiRJUmFoaIgrrrhim3mKr7jiCoaGhsoOTVPEpFiSJEmSaqxatYrW1laWLFlCa2srq1atKjskTSFHn5YkSZKkwgEHHMAFF1zA+eefz6JFi1i9ejXLly/ngAMOKDs0TRGTYkmSJEkq7LnnnmzZsoULL7yQn/zkJxxyyCHsvffe7LnnnmWHpili92lJkiRJKvz0pz/lC1/4AnvttRcAe+21F1/4whf46U9/WnJkmiomxZIkSZJUaGlpYeHChdx5553cdNNN3HnnnSxcuJCWlpayQ9MUMSmWJEmSpEJ3dzddXV309fUxPDxMX18fXV1ddHd3lx2apoj3FEuSJElSobOzk5tvvpnjjjuOjRs3sscee3DaaafR2dlZdmiaIibFkiRJklSoVCqsXLmS6667jpGREZqamujq6uKoo44yMZ6h7D4tSZIkSYWenh56e3vp6OigubmZjo4Oent76enpKTs0TRGTYklqEJVKhdbWVpYsWUJrayuVSqXskCRJqjuDg4O0tbVtU9bW1sbg4GBJEWmq2X1akhpApVKhu7ub3t7ebbqCAXYFkySpRktLC5/4xCe4+uqrGRwcpKWlhbe97W2OPj2D2VIsSQ3ArmCSJE1MR0cH5513HkuXLmXlypUsXbqU8847j46OjrJD0xSxpViSGoBdwSRJmpi+vj7e+ta38tGPfnTr6NNvfetb6evrKzs0TRGTYklqAC0tLQwMDGzzK/fAwIBdwSRJGmP16tU8/fTTzxp9es2aNWWHpili92lJagDd3d10dXXR19fH8PAwfX19dHV10d3dXXZokiTVlTlz5nDmmWduc8vRmWeeyZw5c8oOTVPElmJJagCjg2ktW7Zs66AhPT09DrIlSdIYmzZt4jOf+QwXXngh9913H4ceeijr1q1j06ZNZYemKTJuUhwRK4C3Aj/PzNbtrD8F+DAQwFPAH2TmD4p1a4qyEWA4MxdPXuiSpF3R2dlJZ2cn/f39tLe3lx2OJEl1acGCBTzwwAM8/PDDAKxZs4ampiYWLFhQcmSaKhPpPn0ZcOxO1t8LvD4zXwZ8CrhkzPqOzHyFCbEkSZKkevfQQw8xMjLC8ccfzze+8Q2OP/54RkZGeOihh8oOTVNk3KQ4M78DPLqT9Tdn5mPF4veAhZMUmyRJkiTtVhs3buQNb3gDP/7xj3nHO97Bj3/8Y97whjewcePGskPTFJnse4q7gOtqlhO4ISIS+FJmjm1F3ioiTgdOB5g/fz79/f2THJrKsG7dOj9Lqc54Xkr1yXNTqh9LlizhqKOOYt26dcybN4+bb76ZVatWeY7OUJGZ428UcRjwre3dU1yzTQfwV0BbZj5SlC3IzLUR8SLgRmBZ0fK8U4sXL85bb711gi9B9cx7F6X643kp1SfPTak+RAR777033/zmN7dOyXTCCSfw1FNPMZHcSfUpIm7b0S29k9JSHBG/BvwNcNxoQgyQmWuLvz+PiG8ArwHGTYolSZIkqQxHH300N9xwA2984xvZsmULs2bNYsuWLRx99NFlh6Yp8rznKY6IQ4B/BN6Tmf9dU75XROw9+hw4Grjz+dYnSZIkSVPl1FNPZfbs2WzZsgWALVu2MHv2bE499dRyA9OUGTcpjogK8G/Ar0TEUER0RcQZEXFGscmfA78A/FVE3BERo/2e5wMDEfED4PvAysz89hS8BkmSJEmaFMuXL6e5uZnZs2cDMHv2bJqbm1m+fHnJkWmqjNt9OjM7x1n/fuD92ym/B3j5cw9NkiRJknavoaEhmpqaOP/881m0aBGrV69m+fLlDA0NlR2apshkjz4tSZIkSdPa61//elasWMHg4CAtLS28/vWvZ9WqVWWHpSliUixJkiRJNVatWsWLX/xiAB555BHuuuuukiPSVHreA21JkiRJ0kzz0EMPsWXLFh566KGyQ9EUMymWJEmSpDH22Wefbf5q5jIpliRJkqQaL37xi3nssccAeOyxx7Z2pdbMZFIsSZIkSTV+9rOfMWtWNVWaNWsWP/vZz0qOSFPJpFiSJEmSxtiyZcs2fzVzmRRLUoOoVCq0trayZMkSWltbqVQqZYckSZJUOqdkkqQGUKlU6O7upre3l5GREZqamujq6gKgs7Oz5OgkSao/L37xi/n5z3/Oi170IrtPz3C2FEtSA+jp6aG3t5eOjg6am5vp6Oigt7eXnp6eskOTJKkuOSVT4zAplqQGMDg4SFtb2zZlbW1tDA4OlhSRJEn1bWRkZJu/mrlMiiWpAbS0tDAwMLBN2cDAAC0tLSVFJEmSVB9MiiWpAXR3d9PV1UVfXx/Dw8P09fXR1dVFd3d32aFJkiSVyoG2JKkBjA6mtWzZMgYHB2lpaaGnp8dBtiRJUsMzKZakBtHZ2UlnZyf9/f20t7eXHY4kSXVt3rx5rFu3butfzVx2n5YkSZKkMUYTYRPimc+kWJIkSZLGaGpq2uavZi6TYkmSJEkaY5999mHWrFnss88+ZYeiKWZSLEmSJEljPPnkk2zZsoUnn3yy7FA0xUyKJUmSJKnGnnvuyaxZ1VRp1qxZ7LnnniVHpKlkUixJkiRJhYULFzJnzhwWLFjArFmzWLBgAXPmzGHhwoVlh6Yp4pRMkiRJkqaViJjyOh5//HEA1qxZs3V5KuvNzCk7tnbOlmJJkiRJ00pmTunjyiuv5Mgjj4SYxZFHHsmVV1455XWqPLYUS5IkSVKNzs5OOjs7Oeycldx57lvKDkdTzJZiSZIkSVLDMimWJEmSJDUsk2JJkiRJUsMyKZYkSZIkNawJJcURsSIifh4Rd+5gfUTEFyLi7oj4YUS8qmbd+yLif4rH+yYrcEmSJEmSnq+JthRfBhy7k/XHAS8tHqcDfw0QEQcAHwNeC7wG+FhE7P9cg5UkSZIkaTJNKCnOzO8Aj+5kkxOAK7Lqe8B+EXEQcAxwY2Y+mpmPATey8+RakiRJkqTdZrLmKV4A3F+zPFSU7aj8WSLidKqtzMyfP5/+/v5JCk1lWrdunZ+lVGc8L6X65Lkp1SfPy5lvspLi5y0zLwEuAVi8eHG2t7eXG5AmRX9/P36WUn3xvJTqk+emVIe+vdLzsgFM1ujTa4GX1CwvLMp2VC5JkiRJUukmKym+BnhvMQr1bwBPZOYDwPXA0RGxfzHA1tFFmSRJkiRJpZtQ9+mIqADtwIERMUR1ROnZAJl5MXAt8GbgbuBp4PeKdY9GxKeAW4pDfTIzdzZglyRJkiRJu82EkuLM7BxnfQJ/tIN1K4AVux6aJEmSJElTa7K6T0uSJEmSNO2YFEuSJEmSGpZJsSRJkiSpYZkUS5IkSZIalkmxJEmSJKlhmRRLkiRJkhqWSbEkSZIkqWGZFEuSJEmSGpZJsSRJkiSpYZkUS1KDqFQqtLa2smTJElpbW6lUKmWHJEmSVLrmsgOQJE29SqVCd3c3vb29jIyM0NTURFdXFwCdnZ0lRydJklQeW4olqQH09PTQ29tLR0cHzc3NdHR00NvbS09PT9mhSZIklcqkWJIawODgIG1tbduUtbW1MTg4WFJEkiRJ9cGkWJIaQEtLCwMDA9uUDQwM0NLSUlJEkiRJ9cGkWJIaQHd3N11dXfT19TE8PExfXx9dXV10d3eXHZokSVKpHGhLkhrA6GBay5YtY3BwkJaWFnp6ehxkS5IkNTyTYklqEJ2dnXR2dtLf3097e3vZ4UiSJNUFu09LkiSVyDnEJalcJsWSJEklqVQqnHXWWaxfvx6A9evXc9ZZZ5kYS9JuZFIsSZJUkuXLl9Pc3MyKFSu4/vrrWbFiBc3NzSxfvrzs0CSpYZgUS5IklWRoaIjLL7+cjo4Ompub6ejo4PLLL2doaKjs0CSpYZgUS5IkSZIalkmxJElSSRYuXMh73/vebeYQf+9738vChQvLDk2SGoZJsSQ1CEe4lerP+eefz8jICEuXLuXoo49m6dKljIyMcP7555cdmiQ1DOcplqQGUKlU6O7upre3l5GREZqamujq6gKq8xdLKsfo+dfT00NEsNdee/HpT3/a81KSdiOTYklqAD09PZx88sksW7aMwcFBWlpaOPnkk+np6fHLt1Syzs5OOjs76e/vp729vexwJKnhmBRLUgNYvXo1Tz/99LNaitesWVN2aJIkSaWa0D3FEXFsRPwoIu6OiHO2s/5zEXFH8fjviHi8Zt1IzbprJjF2SdIEzZkzh6OOOoply5ZxzDHHsGzZMo466ijmzJlTdmhSw1u2bBlz586lo6ODuXPnsmzZsrJDkqSGMm5LcUQ0AV8E3gQMAbdExDWZuXp0m8z805rtlwGvrDnEhsx8xaRFLEnaZRs3buSqq67ivPPOY9GiRaxevZoPf/jDDA8Plx2a1NCWLVvGxRdf/KxzE+DCCy8sOTpJagwTaSl+DXB3Zt6TmZuArwAn7GT7TsAhTSWpjuyxxx6ceOKJrFixgre85S2sWLGCE088kT322KPs0KSGdumll3LeeefxgQ98gLlz5/KBD3yA8847j0svvbTs0CSpYUzknuIFwP01y0PAa7e3YUQcChwOrKopnhsRtwLDwLmZefVzC1WS9Fxt2rSJ7373u6xYsWLrPcVLly5l06ZNZYcmNbSNGzdyxhlnbFN2xhln8MEPfrCkiCSp8Uz2QFsnAV/PzJGaskMzc21E/CKwKiL+MzN/PHbHiDgdOB1g/vz59Pf3T3JoKsO6dev8LKU6cOihh9LW1sbSpUv5yU9+wiGHHEJbWxuA56hUotmzZ/OhD32Id73rXVuvmV/96leZPXu256ZUJzwXZ76JJMVrgZfULC8syrbnJOCPagsyc23x956I6Kd6v/GzkuLMvAS4BGDx4sXplAQzg9NLSPXh05/+NGeddRZ77bUXmQnA9ddfz+c//3nPUalEv//7v8/FF1/MEUccwaJFi7j99tu59NJLOeOMMzw3pXrw7ZWeiw1gIknxLcBLI+JwqsnwScDJYzeKiF8F9gf+raZsf+DpzNwYEQcCvwmcPxmBq75VKhV6enq2zofa3d3tXKhSyZ555hkef/xxMpO1a9cyd+7cskOSGt7oYFof/ehH2bhxI3vssQdnnHGGg2xJ0m40blKcmcMRcSZwPdAErMjMuyLik8CtmTk6zdJJwFdytAmiqgX4UkRsoTqo17m1o1ZrZqpUKnR3dz9rPlTAxFgqyfLly2lqamLBggXcd999LFiwgMcff5zly5d7XkolO+qoo+jr62NwcJAjjjiCo446quyQJKmhTOie4sy8Frh2TNmfj1n++Hb2uxl42fOIT9NQT08Pvb29dHR0bO0+3dvby7Jly/zyLZVkaGiIfffdd5uyzGRoaKikiCSBPyRLUj2YyJRM0i4ZHBzcOoDPqLa2NgYHB0uKSBLAhg0bWLt27dbu0xs2bCg7JKnh1f6Q3NzcTEdHB729vfT09JQdmiQ1DJNiTbqWlhYGBga2KRsYGKClpaWkiCRBdVqmzZs3A7B582anY5LqgD8kS1L5TIo16bq7u+nq6qKvr4/h4WH6+vro6uqiu7u77NAkSaor/pAsSeWb7HmKpa33QC1btmzr6NM9PT3eGyXVgf3335/HHnts619J5Rr9IXn0nuLRH5LtPi1Ju09sO1h0fVi8eHHeeuutZYehSeA8xVJ9iAhmzZpFU1MTmzdvZvbs2YyMjLBlyxbq8TogNRKnMdRM8vJP3MATGzaXHYZ2YN8XzOYHHzu67DBKERG3Zebi7a2zpViSGsSWLVvYsmULwNZ7iyWVr7Ozk87OTn9I1ozwxIbNrDn3LWWHMWlm2nl52Dkryw6hLnlPsSRJkiSpYZkUS5IkSZIalkmxJEmSJKlhmRRLUgNpamra5q+k8lUqFVpbW1myZAmtra1UKpWyQ5KkhuJAW5LUQEZHmnbEaak+VCoVuru7t07J1NTURFdXF4AjUEvSbmJLsSQ1kNHRp0f/SipXT08Pvb29dHR00NzcTEdHB729vc5TLEm7kUmxJElSSQYHBxkaGtqm+/TQ0BCDg4NlhyZJDcPu05IkSSU5+OCDWb58OVdeeeXW7tMnn3wyBx98cNmhSVLDMCmWpAYSEWTm1r+SyvfMM8+wdOlS7rvvPg499FCeeeYZ5s2bV3ZYktQw7D4tSQ3Egbak+rJ27Vqam6ttFBEBQHNzM2vXri0zLElqKCbFkiRJJZkzZw4f+chHuPfee7npppu49957+chHPsKcOXPKDk2SGobdpyVJkkqyadMmLrroIl75ylcyMjJCX18fF110EZs2bSo7NElqGCbFkiRJJVm0aBFve9vbWLZsGYODg7S0tHDyySdz9dVXlx2aJDUMk2JJkqSSdHd3093dTW9v79bRp7u6upynWJJ2I5NiSZKkknR2dgJs01Lc09OztVySNPVMiiVJkkrU2dlJZ2cn/f39tLe3lx2OJDUcR5+WJEkqUaVSobW1lSVLltDa2kqlUik7JElqKLYUS1KdGJ2jdCbV63zI0s5VKpXt3lMM2IVaknYTW4olqU5k5pQ9DjjggO3WecABB0xpvZJ2rqenh97eXjo6Omhubqajo4Pe3l4H2pKk3cikWJIawEUXXcTee+/N7NmzAZg9ezZ77703F110UcmRSY1tcHCQoaGhbbpPDw0NMTg4WHZoktQw7D4tSQ1gtBtmT08Pd60e5Jd/+Zfp7u62e6ZUsoMPPpjly5dz5ZVXbu0+ffLJJ3PwwQeXHZr0nOzdcg4vu/ycssOYXJeXHcDk2bsF4C1lh1F3TIolqUGMjnB72DkrufNcL4hSvXj00Ud5wxvesHV5zpw5vOhFLyoxIum5e2rwXNbMoGvMTBsV/rBzVpYdQl2aUPfpiDg2In4UEXdHxLN++omIUyPioYi4o3i8v2bd+yLif4rH+yYzeEmSpOlsaGiITZs2sf/++xMR7L///mzatImhoaGyQ5OkhjFuS3FENAFfBN4EDAG3RMQ1mbl6zKZXZeaZY/Y9APgYsBhI4LZi38cmJXpJkqRp7lWvehUbN27kiSee4OCDD+bwww/n9ttvLzssSWoYE2kpfg1wd2bek5mbgK8AJ0zw+McAN2bmo0UifCNw7HMLVZIkaea54447ePjhh8lMHn74Ye64446yQ5KkhjKRe4oXAPfXLA8Br93Odu+IiP8D/Dfwp5l5/w72XbC9SiLidOB0gPnz59Pf3z+B0FTv1q1b52cp1SHPS6l+bNmyhQcffBBg61/wPNX0NZP+7c7E77Iz7fVMhskaaOufgEpmboyI36c6RtsbxtlnG5l5CXAJwOLFi3Mm3dDeyGba4ATSjPDtlZ6XUp1pamraOvr0yMgIgOeppqcZdo2Zcd9lZ9jnM1km0n16LfCSmuWFRdlWmflIZm4sFv8GePVE95UkSWp0o4nw6F9J0u4zkaT4FuClEXF4RMwBTgKuqd0gIg6qWTweGJ1x/nrg6IjYPyL2B44uyiRJkgTsv//+rFq1ihtvvJFVq1ax//77lx2SJDWUcbtPZ+ZwRJxJNZltAlZk5l0R8Ung1sy8BvjjiDgeGAYeBU4t9n00Ij5FNbEG+GRmPjoFr0OSJGlaevLJJznmmGPYvHkzs2fPZsuWLWWHJEkNZUL3FGfmtcC1Y8r+vOb5R4CP7GDfFcCK5xGjJEnSjNTc3Mzw8PDWRHh4eJjMpLl5soZ9kSSNZyLdpyVJkjQF9tlnHwAyc5u/o+WSpKlnUixJklSSRx+t3lU22jI8+ne0XJI09UyKNSUqlQqtra0sWbKE1tZWKpVK2SFJklSX5s2bxw033MCNN97IDTfcwLx588oOSZIaijesaNJVKhW6u7vp7e3dOudiV1cXAJ2dnSVHJ0lSfdm8efM2A23NmmWbhSTtTibFmnQ9PT309vbS0dGxdcLz3t5eli1bZlIsSdIYGzdu3JoIj4yMsHnz5pIjkp6fw85ZWXYIk+vbM+f17PuC2WWHUJdMijXpBgcHaWtr26asra2NwcHBHewhSVJjGx192umYNN2tOfctZYcwqQ47Z+WMe016NvvnaNK1tLQwMDCwTdnAwAAtLS0lRSRJkiRJ22dSrEnX3d1NV1cXfX19DA8P09fXR1dXF93d3WWHJkmSJEnbsPu0Jl1nZyc333wzxx13HBs3bmSPPfbgtNNO835iSZIkSXXHpFiTrlKpsHLlSq677rptRp8+6qijTIwlSZIk1RW7T2vS1Y4+3dzcTEdHB729vfT09JQdmiRJkiRtw6RYk25wcJCvfe1rzJ07l46ODubOncvXvvY1R5+WJEmSVHfsPq1Jt99++/GlL32JCy64gEWLFrF69WrOPvts9ttvv7JDkyRJkqRtmBRr0j355JPMmjWLD37wg1vLmpubefLJJ0uMSpIkSZKezaRYk254eBiA/fffn8cff5z99tuPxx57rOSopOfu5Z+4gSc2bC47jEl12Dkryw5h0uz7gtn84GNHlx2GJEmapkyKNSXmz5/P448/Tmby9NNPM3/+fB588MGyw5Kekyc2bGbNuW8pO4xJ09/fT3t7e9lhTJqZlOBLkqTdz6RYU6I2Ad64caMJsSRJkqS65OjTkiRJJZs3bx4Rwbx588oORZIajkmxJElSydavX09msn79+rJDkaSGY1IsSZJUsszc5q8kaffxnmJJkqSdiIgZVa+JtyRty5ZiSZKkncjMKXuceeaZ263zzDPPnLI6JUnbsqVYkiSpJBdeeCEAl156KRs3bmSPPfbgtNNO21ouSZp6thRryox2+yqr25kkSdPBhRdeyDPPPMOhH/4WzzzzjAmxJO1mthQ3qN2RqG5v0JCprNcuYZoqe7ecw8suP6fsMCbX5WUHMHn2bgF4S9lhSJKkacqkuEFNZQI5a9as7R4/ItiyZcuU1StNlacGz2XNuTMn6erv76e9vb3sMCbNYeesLDsESZI0jdl9WpPuj/7oj3apXJIkSZLKYkuxJp2DhkiSJEmaLibUUhwRx0bEjyLi7oh41o11EfGBiFgdET+MiJsi4tCadSMRcUfxuGYyg1f9ctAQSZIkSdPBuC3FEdEEfBF4EzAE3BIR12Tm6prN/gNYnJlPR8QfAOcDJxbrNmTmKyY3bEmSJEmSnr+JtBS/Brg7M+/JzE3AV4ATajfIzL7MfLpY/B6wcHLDlCRJkiRp8k3knuIFwP01y0PAa3eyfRdwXc3y3Ii4FRgGzs3Mq7e3U0ScDpwOMH/+fPr7+ycQmqYDP0vNBDPp3/G6detm1OuBmfX5qLH5b1mqP56XM9+kDrQVEe8GFgOvryk+NDPXRsQvAqsi4j8z88dj983MS4BLABYvXpwzabqQhvbtlTNq6hc1qBn273imTck00z4fNTD/LUv1x/OyIUwkKV4LvKRmeWFRto2IeCPQDbw+MzeOlmfm2uLvPRHRD7wSeFZSLEn1bMbNhfvtmfN69n3B7LJDkCRJ09hEkuJbgJdGxOFUk+GTgJNrN4iIVwJfAo7NzJ/XlO8PPJ2ZGyPiQOA3qQ7CJUnTxppz31J2CJPqsHNWzrjXJEmS9FyNmxRn5nBEnAlcDzQBKzLzroj4JHBrZl4DXADMA74WEQA/yczjgRbgSxGxheqgXueOGbVakiTpOXn5J27giQ2byw5jUs2UXin7vmA2P/jY0WWHIUkTMqF7ijPzWuDaMWV/XvP8jTvY72bgZc8nQEmSpO15YsPmGdXrYSbd7z9TkntJjWEiUzJJkiRJkjQjmRRLkiRJkhqWSbEkSZIkqWFN6jzFev4cNKS+OXCIJEmSNLOYFNcZBw2pbzMpwZek6W7vlnN42eXnlB3G5Lq87AAmx94tADPn+4ykmc2kWJIkTUtPDZ7rD8l1yh+RJU0n3lMsSZIkSWpYJsWSJEmSpIZlUixJkiRJaljeU1xnHDSkvjlwiCTVlxl37+q3Z8br2fcFs8sOQZImzKS4zjhoSH2bcV++JGkam0nXS6heY2baa5Kk6cDu05IkSZKkhmVSLEmSJElqWCbFkiRJkqSGZVIsSZIkSWpYDrRVh2bcYE4zZCRNcDRNSZIkaaYxKa4zM23USUfSlCRJ0mSLiN1X13m7p57M3D0V6VnsPi1JkiRpWsnM3fLo6+vbbXWpPCbFkiRJkqSGZVIsSZIkSWpYJsWSJEmSpIZlUixJklSiSqVCa2sr951/PK2trVQqlbJDkqSGYlIsSZJUkkqlwllnncX69eshk/Xr13PWWWeZGEvSbuSUTJJUJ5xeQqpPu+PcfOihhwBYs2YNACeffDInn3zylNTleSlJ2zIplqQ6sbu+qPb399Pe3r5b6pJmgqk8NyOCF7zgBQwPD7N582Zmz55Nc3MzGzZsMHmVpN3EpFiSJKlEGzdu5IILLmDRokWsXr2as88+u+yQJKmhmBRLkiSVaM6cOVx44YX85Cc/4ZBDDmHOnDk888wzZYclSQ3DpLhBee+iJEn14ZlnntnaXXrDhg0mxJK0m01o9OmIODYifhQRd0fEOdtZv0dEXFWs//eIOKxm3UeK8h9FxDGTGLueh8zcLY++vr7dVpckSdPR3LlzecELXrD1/uK5c+eWHZIkNZRxk+KIaAK+CBwHLAI6I2LRmM26gMcy8wjgc8B5xb6LgJOAI4Fjgb8qjidJkiRg06ZNbNiwgS1btrBhwwY2bdpUdkiS1FAm0lL8GuDuzLwnMzcBXwFOGLPNCcDlxfOvA0ui2j/3BOArmbkxM+8F7i6OJ0mS1PCOPPJIjj/+eB5//HEAHn/8cY4//niOPPLIcgOTpAYykXuKFwD31ywPAa/d0TaZORwRTwC/UJR/b8y+C7ZXSUScDpwOMH/+fPr7+ycQmurdunXr/CylOuN5KdWPt7/97fT29vKZz3yGww8/nHvvvZcLLriArq4uz1OpDnjNbAx1M9BWZl4CXAKwePHidA7NmcH5UKX643kp1Y/29nYWLVpET08Pg4ODtLS08NnPfpbOzs6yQ5OE18xGMZGkeC3wkprlhUXZ9rYZiohmYF/gkQnuK0mS1LA6Ozvp7Oz0y7cklWQi9xTfArw0Ig6PiDlUB866Zsw21wDvK57/LrAqq8MBXwOcVIxOfTjwUuD7kxO6JEmSJEnPz7gtxcU9wmcC1wNNwIrMvCsiPgncmpnXAL3A30XE3cCjVBNniu2+CqwGhoE/ysyRKXotkiRJkiTtkgndU5yZ1wLXjin785rnzwDv3MG+PUDP84hRkiRJkqQpMZHu05IkSZIkzUgmxZIkSZKkhmVSLEmSJElqWCbFkiRJkqSGZVIsSZIkSWpYJsWSJEmSpIZlUixJkiRJaliRmWXH8CwR8RBwX9lxaFIcCDxcdhCStuF5KdUnz02p/nhezhyHZuYLt7eiLpNizRwRcWtmLi47Dkn/y/NSqk+em1L98bxsDHafliRJkiQ1LJNiSZIkSVLDMinWVLuk7AAkPYvnpVSfPDel+uN52QC8p1iSJEmS1LBsKZYkSZIkNSyTYkmSJElSwzIpbmARsa7m+Zsj4r8j4tDncJzLIuJ3d2H7ayNiv12tR5KkskXEL0TEHcXjZxGxtmZ5zphtz4iI9+7kWO0R8a2pj1rSZJ6749SzOCK+MDlRa3dpLjsAlS8ilgBfAI7JzPumur7MfPNU1yFNNxGRwF9m5geL5Q8B8zLz45NYxzHAecXiEcBaYAPww8x81sU/Iv6miGn1c6jrDODpzLzieYQs1Z3MfAR4BUBEfBxYl5n/3w62vXj3RSZpZ3bXuZuZtwK3Ptf9VQ5bihtcRPwf4FLgrZn546LstIi4JSJ+EBH/EBF7FuWHRcSqiPhhRNwUEYfUHOr/RMTNEXHPaKtxRBwUEd8pfoG7MyJ+qyhfExEHFse7syaWDxX/SRER/RFxXkR8v2jBHt13z4j4akSsjohvRMS/R4QTqmsm2Aj8TkQcOFUVZOb1mfmKzHwF1Qv2KcXydn8Nz8z3P5eEuNj3YhNiNYqdXDc/XvzARUQcERH/XGxze0T8UrH7vIj4ekT8V0R8OSKi2H7N6P8HRctTf80xVxTXyXsi4o9r4vi/EfGjiBiIiMpo3ZK2b4Lnbn9EfL7m++xrivK9inPx+xHxHxFxQlFuD5BpyKS4se0BXA28LTP/q6b8HzPz1zPz5cAg0FWUXwhcnpm/BnyZauvyqIOANuCtwLlF2cnA9cUX8JcDd+xifM2Z+RrgT4CPFWV/CDyWmYuA/wu8ehePKdWrYarTPvzp2BUR8cLiYn1L8fjNovw/I2K/qHpktKtXRFwREW+aaMUR8dcRcWtE3BURn6gp7x/90Ski1kXE54ptboqIFxblvxQR346I2yLiXyPiV4vyrV8opAawo+tmrS8DXyy2OQp4oCh/JdXr3CLgF4HfnEB9vwocA7wG+FhEzI6IXwfeQfV6exzgD8bS+CZy7gLsWXyf/UNgRVHWDawqvqt2ABdExF5THbCmhklxY9sM3Myz/wNoLb7c/idwCnBkUf464Mri+d9RTYJHXZ2ZW4pWpflF2S3A7xWtvy/LzKd2Mb5/LP7eBhxWPG8DvgKQmXcCP9zFY0r17IvAKRGx75jyzwOfy8zRL71/U5R/l+oX6COBe4DfKspfR/XcnqjuzFwM/Brw+oj4te1ssxdwa2YeCfwL//tD1SXAssx8NfAh4K92oV5pptjRdROAiNgbWJCZ3wDIzGcy8+li9fczcygzt1D98fiwCdS3MjM3ZubDwM+pXnd/E/hmceyngH+ajBcmzXA7PXdrVAAy8zvAPlEdG+do4JyIuAPoB+YCh+xgf9U57ylubFuAdwE3RcRHM/PTRfllVFuPfxARpwLtEzjWxprnAdX/OKLaPfstwGUR8ZdjulMOs+0PM3N3cMwR/LeqBpCZT0bEFcAfU73Xd9QbgUVFr0qoXpDnAf8K/B/gPuCvgdMjYgHV3hTrd6Hqd0XE6VTPs4OotliN/cFpC3BV8fzvgX8sYjgK+FpNbHvsQr3STHEZu37dHFV7/ay93tVeI3d0fRy7j6RdcxkTO3dzO8sBvCMzf1S7IiLmo2nHluIGV/xS/RaqrVOjLcZ7Aw9ExGyqv5qNuhk4qXh+CtUv5DsU1ZGsH8zMS6m2bL1qzCYPAi+K6miAe1Dtej2e71JN5ImIRcDLJrCPNJ38P6q9N2q7YM0CfmP0fuDMXJCZ64DvUG0d/i2qv1I/BPwu45ybtSLicKotvEuKWyNW8uwv4NuTRVyP18T1isxsmWjd0gyyo+smAEXL7VBEvA0gIvYYvXdxJ9bwv7cIvWMCMXwX+O2ImFv8YDWRa6rU6HZ67tY4ESAi2oAnMvMJ4HpgWc04AK+c6mA1dUyKRWY+ChwL/FlEHE/1Xt1/p3qBrb3XeBnV7tA/BN4DnDXOoduBH0TEf1D9z+TzY+rdDHwS+D5w45i6duSvgBdGxGrgL4C7gCcmsJ80LRTn41fZ9raGG6iefwBExCuKbe8HDgRempn3AANUE9zv7EKV+wDrgSeKX7eP28F2s6gm3FAdL2AgM58E7o2IdxZxRUS8fBfqlmaKHV03a70H+OPiGnoz8OJxjvkJ4PMRcSvV1uCdysxbgGuo9vK4DvhPvD5K45nIuQvwTPF99mL+9/r8KWA28MOIuKtY1jQVmWN7A0j1KyKagNmZ+UxUR+78Z+BXMnNTyaFJz0tErMvMecXz+cC9wPmZ+fGojkD7RaCFajfJ72TmGcW2fwc0ZebJEXEU1cT4hcXUEzurrx/4UGbeGhGXUe0GfT/VL9HXZOZlY7ZZR/X+4aOp3sN4YmY+VLQ0/zXVbtezga9k5idjnOkuJE2+iJiXmeuKVujvAKdn5u1lxyVNZ7XXwrJj0dQxKda0UgxW0kf1y3cAH87M68qNSpr5apN2SfUpIq6kOibAXKqzRXym5JCkac+kuDGYFEuSxmVSLEmSZiqTYkmagSLiGOC8McX3Zubby4hHkiSpXpkUS5IkSZIalqNPS5IkSZIalkmxJEmSJKlhmRRLkjSNRMQZEfHe4vmpEXFw2TFJkjSdeU+xJEnTlFOFSJL0/JkUS5JUx4pW4Q8BCfwQ+DGwDlgDXAasBTYA3cBpmfm2Yr83AX/oiOOSJO2c3aclSapTEXEk8GfAGzLz5cBZo+sy8+vArcApmfkK4FrgVyPihcUmvwes2L0RS5I0/ZgUS5JUv94AfC0zHwbIzEd3tGFWu379HfDuiNgPeB1w3e4IUpKk6ay57AAkSdKk+Vvgn4BnqCbTwyXHI0lS3bOlWJKk+rUKeGdE/AJARBwwZv1TwN6jC5n5U+CnVLtc/+3uClKSpOnMlmJJkupUZt4VET3Av0TECPAfVAfYGnUZcHFEbABel5kbgC8DL8zMwd0dryRJ05GjT0uSNINExEXAf2Rmb9mxSJI0HZgUS5I0Q0TEbcB64E2ZubHseCRJmg5MiiVJkiRJDcuBtiRJkiRJDcukWJIkSZLUsEyKJUmSJEkNy6RYkiRJktSwTIolSZIkSQ3r/wfxr17ywT7CyAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#1. 以城市(city)為x軸，以單價元坪(unit_price_square_feet)為y軸畫出boxplot，並找出單價元坪(unit_price_square_feet)中位數最高的地區\n",
    "#台北市\n",
    "analysis_data.boxplot(column=['unit_price_square_feet'],by='city',figsize=(16,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'total_price'}, xlabel='room_number'>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#2. 進一步對台北市的資料做圖，以建物現況格局-房(room_number)為x軸，以總價元(total_price)為y軸畫出boxplot，並找出總價元(total_price)中位數最高的房間數。hint:資料切片找出city欄位為台北市的資料，再進一步畫圖\n",
    "#5\n",
    "analysis_data.loc[analysis_data.city=='Taipei'].boxplot(column=['total_price'],by='room_number',figsize=(16,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "#3. 對台北市的資料做圖，先將地區(twons)做編碼在進行，再以地區(twon)為x軸，以單價元坪(unit_price_square_feet)為y軸畫出boxplot，並找出單價元坪(unit_price_square_feet)中位數最高的地區。\n",
    "# hint:運用LabelEncoder()對地區(twons)做編碼，運用.inverse_transform()反查編碼的地區\n",
    "#大安區\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder().fit(analysis_data['towns'].values)\n",
    "analysis_data['towns_2'] = le.transform(analysis_data['towns'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'unit_price_square_feet'}, xlabel='towns_2'>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA74AAAGUCAYAAAAS1v54AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABFEElEQVR4nO3de5xkdXng/89jN6AOCDEYIo6Amxi3hk4MkcT4oxOnHUXwSm5qDd4yJSy7Sy+uq0CsrLekDBdJgqOJS+yJuIYCNFliHEdhh261o8a77jAVLxFGBi/EK86oDN08vz+qeuwZeqarZqr7VFV/3q9Xvabq1Olznnqmqruec77f50RmIkmSJEnSoHpQ0QFIkiRJkrSULHwlSZIkSQPNwleSJEmSNNAsfCVJkiRJA83CV5IkSZI00Cx8JUmSJEkDzcJXkjQwIiIj4heLjqNIEbE2InYe5PkVnyNJ0spj4StJ6rqIuCMifhwRuyLiexGxOSIeXXRccyLipRExXXQc/axfchgRL4mIT0fEPRGxMyKuiIjhouOSJC0vC19J0lJ5dmYeDTwS+BawseB4loyFVE97KPBy4HjgicA64JVFBiRJWn4WvpKkJZWZPwHeA6yZWxYRx0bEOyPi3yNiR0T8cUQ8KCIe3jor9+zWekdHxFci4sWtx++IiLdFxC0R8cOI+FBEnLzQfg+yjxLwNuBJrTPS3z/Azz8mIj7c2s//jYi3RsS7Ws+d0hoyXImIrwG3trb9x6193d3a97Gt9R8w/Lh1Vvyprfuvi4j3RMQNrf19JiIeP2/dEyPi71uv5faI+G/znntIKy/fi4jtwK+38d/yjIj4akR8OyKubMV+ZER8NyJ+ed62fy4ifhQRj9gv9gVzeKCct57bERFPaN0/t5W/U1uPKxFx07xc3Njazg8j4raIOH3evi+JiLtaz30xItYd7IVm5l9n5kcyc09m3gX8HXBGGzmSJA0QC19J0pKKiIcCzwc+Pm/xRuBY4D8ATwZeDPxhZn4X2AD8TUT8HPAXwOcy853zfvZc4E9onsH7HM1CZiEH2kcDuAD4WGYenZnHHeDnrwM+Afws8DrgRQus82SgBDwdeGnrNtba59HAWw6w7YU8F3g38PDWvm+KiCNaheM/AZ8HHkXzjOXLI+LprZ97LfALrdvTgZe0sa/fAU4Hfq213w2ZuQe4HnjhvPXKwNbM/Pf5P3yQHC6Y89ZzHwLWtu4/Gfgq8NvzHn9o3i6e04rlOOC9tPIYEY8DLgR+PTOPab3eO9p4vfP9NnBbhz8jSepzFr6SpKVyU+tM4A+ApwFXAkTEEPAC4I8y84eZeQdwFa3CMjNvplkAbgWeAfyn/ba7OTM/nJn3AlWaZx33mT+82D4WExEn0Txz+prWmcJpmgXY/l6Xmbsz88c0C/I/z8yvZuYu4I+AF3QwDPrTmfmezLwP+HPgwcBvtuJ4RGa+oRXLV4G/ab0+gOcBtcz8bmbeCby5jX1d3lr/a8Bf0ixwAa4FyhERrccvAv53O8G3kfMP0SxwAX4L+LN5j/cvfKcz8/2ZOdva/9zZ71ngKGBNRByRmXdk5r+1E18rxg00C/43tfszkqTBYOErSVoq57TOBD6Y5lm6D0XEz9M8U3sEsGPeujtons2ccw0wArwjM7+z33bvnLvTKjC/C5y43zrt7ONgTgS+m5k/Wmi/B1h24gL7GwZOaHOf81/X/cDO1jZPBk6MiO/P3YBXz9vuifvFMT+GRffVWv/E1n7/BfgRsDYi/iPwiyxc8C9ksZx/CPitiHgkMATcCJwREafQPEv8uXk/9815938EPDgihjPzKzTn674OuDsiro+I/f/vFxQR59Asts/OzG+3+ZokSQPCwleStKQyczYz/4Hm2bpR4NvAfTQLujknAXfB3jOH1wDvBP5LPPDSO3vP7kbE0TSHBn99v3UOug8gFwn7G8DDW8O0H7Df+S9v3v2vL7C/GZqNvXbTbLI0F/cQsM+8WfZ9XQ8CVre2eSdwe2YeN+92TGY+Y16s82M7aZHXtv9rOYl983ctzeHOLwLe05qjvZD9c3jQnLeK1h8B48CHM/MemgXu+TTP8N7fRtxk5nWZOdraTwKXL/YzEXEWzbPkz87M/9fOfiRJg8XCV5K0pKLpucDPAI3W8NUbgVpEHBPN5lSvAN7V+pFX0yxoNtAcHv3OVqE45xkRMRoRR9Kc6/vx1hDfvdrYx7eA1a1tPEBm7gA+Bbyu1fTpScCzF3mpdeC/t5piHQ28EbghM2eAL9E8a/nMiDgC+GOaQ3bne0JE/G5raPTLgXtpzov+BPDDVlOnh0TEUESMRMRcE6sbgT+KiJ+JiNU0C8vFvKq1/qOBi4Ab5j33LppzgF9I8+DDgeyTwzZyDs2zvhfy02HNU/s9PqiIeFxEPCUijgJ+AvwYOGjBHBFPoTkP/Pcy8xPt7EeSNHgsfCVJS+WfImIXcA9QA16SmXNNhcZpngX9KjBNs5nTplbX31cAL24VUpfTLIIvnbfd62g2dPou8AT2bcY034L7aD13K80GR9+MiAMNez0XeBLwHeBPaRaH9x7k9W6iOR/1w8DtNAuzcYDM/AHwX4C30zwDupvmUOb5/pFmE7Dv0Tzb+ruZeV8rD88CfrW13W+3tnNs6+deT3NI8e3AzbQ3J/cfgU/THF68GZiYe6J1EOEzNPP+kYNsY6EcHizn0Cxwj6GZo4UeL+Yo4DKaOfgm8HM051IfzP+kmav3tzpQ74qILW3uT5I0ICJzsdFekiT1hoh4B7AzM/+4gH3fAPxrZr52Cbb9OuAXM/NARfyyiohNwNeLyLMkSUuh3U6TkiStKK2hxN+leSb1TJqX/bms0KCWQavZ1O8CpxUciiRJXeNQZ0mSFvbzNOeg7qJ5iaD/nJmfLTSiJRYRfwJsA67MzNuLjqddEbFl3jDm+bdXFx2bJKk3ONRZkiRJkjTQPOMrSZIkSRpoFr6SJEmSpIFm4StJkiRJGmgWvpIkSZKkgWbhK0mSJEkaaBa+kqSBExFvi4j/uUz7ui0i1i7HvvpNRPxpRHw7Ir5ZdCySpJXNyxlJkgZaqyh9V2auLjiUFSUiTgK+CJycmXcf5rZeCrwsM0e7EZskaeXxjK8kSYcgIoaLjmE5HMbrPAn4zuEWvZIkdYOFrySpJ0VERsQvznv8joj409b9tRGxMyL+R0TcHRHfiIg/3H/diFgFbAFOjIhdrduJB9nn6yLiPRFxQ0T8MCI+ExGPn/f8HRFxSUR8AdgdEcOtZU9tPT8UEa+OiH9r/fynI+LRref+Y0TcEhHfjYgvRsTz2sjBMyJie2tbd0XEK+c996rW6/56RGyYn6+ImIqIl81b96URMT3v8dURcWdE3NOK8bcWyMG7IuIe4KURcWxETLT2d1crt0MHifupwC3z8v6O1vLfjIiPRsT3I+Lz84eIH2gfEVEC3gY8qbWt7y+WN0mS9mfhK0nqVz8PHAs8CqgAb42In5m/QmbuBs4Gvp6ZR7duX19ku88F3g08HLgOuCkijpj3fBl4JnBcZs7s97OvaD3/DOBhwAbgR60C/JbW9n4OeAHwVxGxZpFYJoD/lJnHACPArQARcRbwSuBpwGOBpy6ynf19EvjVea/x3RHx4HnPPxd4D3Ac8HfAO4AZ4BeB04AzgZdxAJn5f9k37y+NiEcBm4E/be33lcDfR8QjWj+24D4yswFcAHysta3jOnytkiQVW/hGxKbWkfptba7/vNaR79si4rqljk+S1NPuA96Qmfdl5vuBXcDjurDdT2fmezLzPuDPgQcDvznv+Tdn5p2Z+eMFfvZlwB9n5hez6fOZ+R3gWcAdmfm3mTmTmZ8F/h74g0ViuQ9YExEPy8zvZeZnWsufB/xtZm5rFfev6+QFZua7MvM7rViuAo5i39x9LDNvysz7aRbwzwBenpm7W0OX/4Jm8d6JFwLvz8z3Z+b9mXkL8CngGRFxQpf2IUnSgoo+4/sO4Kx2VoyIxwJ/BJyRmacCL1+6sCRJfeA7+51x/RFwdBe2e+fcnVbhtxM4caHnF/Bo4N8WWH4y8MTWEN/vt4brnkvzrPXB/B7NgnBHRHwoIp7UWn7ifnHsWGQ7+4iIV0ZEIyJ+0IrlWOD4eavM3/bJwBHAN+bF/r9onrnuxMnAH+yXg1HgkV3chyRJCyq0MUdmfjgiTpm/LCJ+AXgr8AiaX2LOy8x/Bc4D3pqZ32v9rM0yJGmw/Qh46LzHP0+zCO1Up5cvePTcnYh4ELAamD88+mDbuxP4BWD/kUx3Ah/KzKd1EkhmfhJ4bmuo9YXAja34vjE/TpqNpObbzQNzB0BrPu/FwDrgtsy8PyK+B8T8Xe8X+73A8QsM7e7EncD/zszz9n8iIh65yD68BIUk6bAUfcZ3IdcA45n5BJrzf/6qtfyXgF+KiH+OiI+35jdJkgbX54D1rQZHZwFPPsTtfAv42Yg4ts31nxARvxvNbsYvp1mQfbzNn3078CcR8dho+pWI+FngfTT/hr0oIo5o3X691bhpQRFxZEScGxHHtoZd3wPc33r6RppNp9ZExEOB1+73458DfjciHtpqeFWZ99wxNOfS/jswHBGvoTmceUGZ+Q3gZuCqiHhYRDwoIn4hIjr9/3gX8OyIeHrr//TB0WxStrqNfXwLWB0RR3a4T0mSgB4rfCPiaOD/o9lk43M0hzk9svX0MM0GHmtpNg75m4g4bvmjlCQtk4uAZwPfpzks+KZD2Uhr1FAd+GprGO0Buzq3/CPwfOB7wIuA320Vnu34c5pF6c00C9UJ4CGZ+UOazZpeQPPs8TeBy2nOrT2YFwF3tLorX0AzD2TmFuAvaTa7+krr3/n+AthDs2C8lmaDqjkfBD4AfInmEOmfcPDh2wAvBo4EttPMy3v46d/ntmTmnTSbZr2aZtF9J/Aqfvpd5GD7uBW4DfhmRHy7k/1KkgQQmcWOHmoNdX5fZo5ExMOAL2bmA/6YRsTbgH/JzL9tPd4KXNoaBiZJ0mGLiNcBv5iZLyw6lk5FRAKPzcyvFB2LJEm9pqfO+GbmPcDtEfEHAK1hYnPXT7yJ5tleIuJ4mkOfv1pAmJIkSZKkPlL05YzqwMeAx0XEzoio0BzGVYmIz9Mc1vTc1uofBL4TEduBSeBVrUtESJLUtojYEhG7Fri9uqB4bjtAPOcWEU8nIuJtB4j9bUXHJknSfIUPdZYkSZIkaSn11FBnSZIkSZK6rbDr+B5//PF5yimnFLX7g9q9ezerVq0qOoy+Yb46Z846Z846Z846Z846Y746Z846Z846Z846Z84608v5+vSnP/3tzHzE/ssXLXwjYhPwLODuzBw5wDpraV5W4Qjg25m56LX9TjnlFD71qU8ttlohpqamWLt2bdFh9A3z1Tlz1jlz1jlz1jlz1hnz1Tlz1jlz1jlz1jlz1plezldE7FhoeTtDnd8BnHWQDR8H/BXwnMw8FfiDQ4hPkiRJkqQlsWjhm5kfBr57kFXWA/+QmV9rrX93l2KTJEmSJOmwtdXVOSJOAd630FDniPhLmkOcTwWOAa7OzHceYDvnA+cDnHDCCU+4/vrrDznwpbRr1y6OPvroosPoG+arc+asc+asc+asc+asM+arc+asc+asc+asc+asM72cr7GxsU9n5un7L+9Gc6th4AnAOuAhwMci4uOZ+aX9V8zMa4BrAE4//fTs1XHhvTxmvReZr86Zs86Zs86Zs86Zs86Yr86Zs86Zs86Zs86Zs870Y766UfjuBL6TmbuB3RHxYeDxwAMKX0mSJEmSlls3ruP7j8BoRAxHxEOBJwKNLmxXkiRJkqTD1s7ljOrAWuD4iNgJvJbmnF4y822Z2YiIDwBfAO4H3p6Z25YuZEmSJEmS2rdo4ZuZ5TbWuRK4sisRSZIkSZLURd0Y6ixJkiRJUs+y8JUkSZIkDTQL33nq9TojIyOsW7eOkZER6vV60SFJkiRJkg5TNy5nNBDq9TrVapWJiQlmZ2cZGhqiUqkAUC4vOs1ZkiRJktSjLHxbarUa69evZ3x8nEajQalUYv369dRqNQtfSZIkSepjFr4t27dvZ/fu3WzatGnvGd8NGzawY8eOokOTJEmSJB0G5/i2HHnkkYyPjzM2Nsbw8DBjY2OMj49z5JFHFh2aJEmSJOkweMa3Zc+ePbzlLW/htNNOY3Z2lsnJSd7ylrewZ8+eokOTJEmSJB0GC9+WNWvWcM455zxgju9NN91UdGiSJEmSpMNg4dtSrVYX7Opcq9WKDk2SJEmSdBgsfFvmOjfPP+NrR2dJkiRJ6n8WvvOUy2XK5TJTU1OsXbu26HAkSZIkSV1gV2dpGdXrdUZGRli3bh0jIyPU6/WiQ5IkSZIGnoWvtEzq9ToXXXQRu3fvJjPZvXs3F110kcWvJEmStMQsfKVlcvHFFzM0NMSmTZu4+eab2bRpE0NDQ1x88cVFhyZJkiQNNAtfaZns3LmTd77znYyNjTE8PMzY2BjvfOc72blzZ9GhSZIkSQPNwleSJEmSNNAsfOex8ZCW0urVq3nJS17C5OQkMzMzTE5O8pKXvITVq1cXHZokSZI00LycUUu9XqdarTIxMcHs7CxDQ0NUKhUAr+Wrrrjiiiu46KKL2LBhA1/72tc46aSTmJmZ4aqrrio6NEmSJGmgeca3pVarMTExsc/8y4mJCWq1WtGhaUCUy2WuvvpqVq1aBcCqVau4+uqrPbAiSZIkLTHP+LY0Gg1GR0f3WTY6Okqj0SgoIg2icrlMuVxmamqKtWvXFh2OJEmStCJ4xrelVCoxPT29z7Lp6WlKpVJBEUmSJEmSusHCt6VarVKpVPZpPFSpVKhWq0WHJkmSJEk6DA51bpmbZzk+Pk6j0aBUKlGr1Zx/KUmSJEl9zsJ3HudfSpIkSdLgcaizpJ7m9bUlSZJ0uDzjK6lneX1tSZIkdYNnfCX1LK+vLUmSpG6w8JXUs7y+tiRJkrrBwldSz/L62pIkSeoGC19JPcvra0uSJKkbFm1uFRGbgGcBd2fmyEHW+3XgY8ALMvM93QtR0krl9bUlSZLUDe2c8X0HcNbBVoiIIeBy4OYuxCRJe5XLZbZt28bWrVvZtm2bRa8kSZI6tmjhm5kfBr67yGrjwN8Dd3cjKEmSJEmSuiUyc/GVIk4B3rfQUOeIeBRwHTAGbGqtt+BQ54g4Hzgf4IQTTnjC9ddff+iRL6Fdu3Zx9NFHFx1G3zBfnTNnnTNnnTNnnTNnnTFfnTNnnTNnnTNnnTNnnenlfI2NjX06M0/ff/mic3zb8JfAJZl5f0QcdMXMvAa4BuD000/PtWvXdmH33Tc1NUWvxtaLzFf76vU6tVpt73zVarXq0N02+T7rnDnrnDnrjPnqnDnrnDnrnDnrnDnrTD/mqxuF7+nA9a2i93jgGRExk5k3dWHb0sCo1+tUq1UmJiaYnZ1laGiISqUCYPErqa94EE+S1G8Ou/DNzMfM3Y+Id9Ac6nzT4W5XGjS1Wo2JiQnGxsb2HiWbmJhgfHzcL4yS+oYH8SRJ/WjR5lYRUad5maLHRcTOiKhExAURccHShycNjkajwejo6D7LRkdHaTQaBUUkSZ2bfxBveHiYsbExJiYmqNVqRYcmSdIBtdPVuZyZj8zMIzJzdWZOZObbMvNtC6z7Uq/hKy2sVCoxPT29z7Lp6WlKpVJBEWlQ1et1RkZGWLduHSMjI9Tr9aJD0gDxIJ4kqR91Y46vpDZUq1Uqlcre4YGTk5NUKhXPkqirHIaqpTZ3EG9sbGzvMg/iSZJ6nYWvtEzmio7x8fG9DWFqtZrFiLrKueRaah7EkyT1IwtfaRmVy2XK5XJftoBXf3AYqpaaB/EkSf1o0Tm+kqT+4VxyLYdyucy2bdvYunUr27Zts+iVJPU8C19JGiBzw1AnJyeZmZnZOwy1Wq0WHZokSVJhHOosSQPEYaiSJEkPZOErSQPGueSSJEn7cqizJEmSJGmgWfhK0oCp1+uMjIywbt06RkZGqNfrRYckSZJUKIc6S9IAqdfrVKvVvddYHRoaolKpADjPV5IkrVie8ZWkAVKr1ZiYmGBsbIzh4WHGxsaYmJigVqsVHZokSVJhLHwlaYA0Gg1GR0f3WTY6Okqj0SgoIkmSpOJZ+ErSACmVSkxPT++zbHp6mlKpVFBEkiRJxbPwlaQBUq1WqVQqTE5OMjMzw+TkJJVKhWq1WnRokiRJhbG5lSQNkLkGVuPj4zQaDUqlErVazcZWkvpOvV6nVqvt/V1WrVb9XSbpkFn4StKAKZfLlMtlpqamWLt2bdHhSFLH7FAvqdsc6iypp3lNWklaeexQL6nbPOMrqWd5xF+SViY71EvqNs/4SupZHvGXpJXJDvWSus3CV1LP8oi/JK1MdqiX1G0OdZbUs+aO+I+Nje1d5hF/SRp85XKZj370o5x99tnce++9HHXUUZx33nlOc5F0yDzjK6lnecRf6k02ndNSq9frbN68mS1btnDLLbewZcsWNm/e7HtN0iHzjK+knuU1aaXeY9M5LYdarcbjH//4fc74nn322f4NkHTILHwl9TSvSSv1lvlN5+Y+lxMTE4yPj1uQqGtuu+02vvjFL3L55ZezZs0atm/fziWXXMLMzEzRoUnqUw51liRJbbPpnJZDRHDeeefxile8ggc/+MG84hWv4LzzziMiig5NUp+y8JUkSW3zMjNaDpnJli1b9unxsGXLFjKz6NAk9SmHOkuSpLbNNZ2bm+M713TO62urm4466ijOOOOMfXo8nHHGGXzjG98oOjRJfcrCV5Iktc2mc1oO5513Hm9729seMMf3ggsuKDo0SX3KwleSJEk9ZePGjQC8+tWv3tvV+YILLti7XJI6ZeErSZLa5uWMtFw2btzIxo0b7eovqStsbiVJkto2/3JGw8PDjI2NMTEx4RxfdV29XmdkZIR169YxMjJCvV4vOiRJfcwzvpIkqW2NRoN3v/vdnH322XuHoG7YsMHLGamrHFkgqdsWPeMbEZsi4u6I2HaA58+NiC9ExP+LiI9GxOO7H6YkSeoFxx13HNdccw1vfOMb2bJlC2984xu55pprOO6444oOTQPEkQWSuq2doc7vAM46yPO3A0/OzF8G/gS4pgtxSZKkHnTPPfdw7LHHctpppzE8PMxpp53Gscceyz333FN0aBogjUaD0dHRfZaNjo46skDSIVt0qHNmfjgiTjnI8x+d9/DjwOouxCVJknrQzMwMb3rTm/a5nNGb3vQmNmzYUHRoGiClUonXv/713HTTTXvfZ+eccw6lUqno0CT1qcjMxVdqFr7vy8yRRdZ7JfAfM/NlB3j+fOB8gBNOOOEJ119/fccBL4ddu3Zx9NFHFx1G3zBfnTNnnTNnnTNnnTNnizvzzDN52ctexvOe97y9+brxxht5+9vfzs0331x0eD3P91h7rr76av7pn/6J888/n6c85SnceuutXHPNNTz72c/moosuKjq8nuf7rHPmrDO9nK+xsbFPZ+bpD3giMxe9AacA2xZZZwxoAD/bzjaf8IQnZK+anJwsOoS+Yr7ad9111+Wpp56aD3rQg/LUU0/N6667ruiQ+obvs86Zs86Zs8VdeOGFOTw8nFdddVVu2bIlr7rqqhweHs4LL7yw6ND6gu+x9px66qlZrVb3+Zs591iL833WOXPWmV7OF/CpXKD+7EpX54j4FeDtwNmZ+Z1ubFMaNHaolDQINm7cyJe+9CVe+cpXkplEBE972tPYuHFj0aFpgDQaDT772c/yp3/6p3uv43vffffxZ3/2Z0WHJqlPHfZ1fCPiJOAfgBdl5pcOPyRpMNmhUtIgqNfrfPnLX2br1q3ccsstbN26lS9/+cteY1VdVSqVmJ6e3mfZ9PS0c3wlHbJ2LmdUBz4GPC4idkZEJSIuiIgLWqu8BvhZ4K8i4nMR8akljFc9xAvLd6bRaLBz5859crZz5047VErqKx7E03KoVqtUKhUmJyeZmZlhcnKSSqVCtVotOjRJfaqdrs4HHYOZzUZWCzaz0uBy2G7nTjzxRC6++GKuu+66vTlbv349J554YtGhSVLbvMyMlsPcd4n53cNrtZrfMSQdssMe6qyVySP+h+YnP/kJGzZs4Mwzz2TDhg385Cc/KTokSeqIQ1C1XMrlMtu2bWPr1q1s27bNolfSYbHw1SHxiH/n7rrrLo444ggAIgKAI444grvuuqvIsCSpIw5BlST1IwtfHRKP+HfuyCOP5NJLL+X2229n69at3H777Vx66aUceeSRRYcmSW0rl8vUajXGx8d5+tOfzvj4uENQJfUte9asHF25nJFWnrkj/nNzfOeO+DvU+cD27NnDxo0bOe200/bmbOPGjezZs6fo0HpavV6nVqvtneNVrVb9gi0VrFwuUy6X915mRjoUc6Ofuql5CU+pPfasWVksfHVIbDrRuTVr1nDOOefsk7Nzzz2Xm266qejQepZ/kCRpcLVbpJ5y6WbuuOyZSxyNVqL5PWvmDuRNTEwwPj7u94wBZOGrQ+YR/85Uq9UFizjPkh+Yf5AkSdJSsWfNymLhKy2TcrnMRz/6Uc4++2zuvfdejjrqKM477zwLuIPwD5IkSVoqcz1rxsbG9i6zZ83gsrmVtEzq9TqbN29my5Yt3HLLLWzZsoXNmzfbROEgbKIm9SabwUgaBHapX1k84ystE4ftdq5arfL85z+fVatW8bWvfY2TTjqJ3bt3c/XVVxcdmrRiOfde0qCwZ83K4hlfaZk4bPfw2KlT6g3zD+INDw8zNjbGxMSE/Qok9aVyucy2bdvYunUr27Zts+gdYBa+0jJx2G7narUa559/PqtWrSIiWLVqFeeff75fsKUCeRBPktSPHOosLROvfdy57du3s3v3bjZt2rR3SOWGDRvYsWNH0aFJK5bNYCRJ/cjCV1omziPp3JFHHskZZ5yxT87OOOMMvvGNbxQdmrRieRBPktSPLHylZeS1jztz7733csMNN3D55ZezZs0atm/fziWXXMLMzEzRoUkrlgfxJEn9yMJXUs866qij+P3f/302bdq09wv285//fN7znvcUHZq0onkQT5LUbyx8dVAR0dXt2ZlXndizZw///M///IA5vnv27Ck6NEmSNADq9Tq1Wm3vAfZqteoIlgFl4auDaqdQPeXSzdxx2TOXIRqtNGvWrOGcc87ZZ0jlueeey0033VR0aJIkqc95XfKVZUUVvp69lPpLtVpd8A+STXQkSdLhmn9d8rmpGxMTE4yPj1v4DqAVVfi2W6h6BlPqDTbRkSRJS8Xrkq8sDyo6AEk6mHK5zLZt29i6dSvbtm2z6JUkSV0xd13y+bwu+eCy8JUkSZK04sxdl3xycpKZmZm91yWvVqtFh6YlsKKGOkuSJEkSOKVqpbHwlSRJkrQieV3ylcOhzpIkSZKkgWbhK0mSJEkaaBa+kiRJkqSBZuErSZIkSRpoFr6SJKkj9XqdkZER1q1bx8jICPV6veiQJEk6KLs6S5KkttXrdarVKhMTE8zOzjI0NESlUgHwEiCSek5EdHV7mdnV7Wn5eMZXkiS1rVarMTExwdjYGMPDw4yNjTExMUGtVis6NEl6gMxs63byJe9raz31LwtfSZLUtkajwejo6D7LRkdHaTQaBUUkSdLiLHwlSVLbSqUS09PT+yybnp6mVCoVFJEkSYtbdI5vRGwCngXcnZkjCzwfwNXAM4AfAS/NzM90O1BJklS8arXK85//fFatWsWOHTs4+eST2b17N1dffXXRoUmSdEDtNLd6B/AW4J0HeP5s4LGt2xOBv279K61INlGQtFJ0+/edJElLZdGhzpn5YeC7B1nlucA7s+njwHER8chuBSj1m242ULDoldRrarUaN9xwA7fffjtbt27l9ttv54YbbrC5lSSpp3XjckaPAu6c93hna9k39l8xIs4Hzgc44YQTmJqa6sLul0Yvx9aLzFfnzFlndu3aZc46ZM46Z84W12g0mJ2dZWpqam++ZmdnaTQa5q4NvscOjTnrjO+zQ2PO2teP77FlvY5vZl4DXANw+umn59q1a5dz9+37wGZ6NrZeZL46Z846NjU1Zc46ZM46Z84WVyqVGBoaYu3atXvzNTk5SalUMndt8D12CPyb2THfZ4fA91lH+vE91o2uzncBj573eHVrmSRJGjDVapVKpcLk5CQzMzNMTk5SqVSoVqtFhyZJ0gF144zve4ELI+J6mk2tfpCZDxjmLEmS+l+5XAZgfHycRqNBqVSiVqvtXS5JUi9q53JGdWAtcHxE7AReCxwBkJlvA95P81JGX6F5OaM/XKpgJUlS8crlMuVyuS+HukmSVqZFC9/MPOgh3Gy2nf2vXYtIkiRJkqQu6sYcX0mSJEmSepaFryRJ6ki9XmdkZIR169YxMjJCvV4vOiRJkg5qWS9nJEmS+lu9XqdarTIxMcHs7CxDQ0NUKhUAG1xJknqWZ3wlSVLbarUaExMTjI2NMTw8zNjYGBMTE9RqtaJDkyTpgCx8JUlS2xqNBqOjo/ssGx0dpdFoFBSRJEmLs/CVJEltK5VKTE9P77NsenqaUqlUUESSJC3OwleSJLWtWq1SqVSYnJxkZmaGyclJKpUK1Wq16NAkSTogm1tJkqS2zTWwGh8fp9FoUCqVqNVqNraSJPU0C19JktSRcrlMuVxmamqKtWvXFh2OJEmLcqizJEmSJGmgWfhKkiRJkgaaha8kDZh6vc7IyAjr1q1jZGSEer1edEiSJEmFco6vJA2Qer1OtVplYmKC2dlZhoaGqFQqADYfkiRJK5ZnfCVpgNRqNSYmJhgbG2N4eJixsTEmJiao1WpFhyZJklQYz/hK0gBpNBqMjo7us2x0dJRGo1FQRNLKERFd3V5mdnV7krSSecZXkgZIqVRienp6n2XT09OUSqWCIpJWjsxc9HbyJe9raz2LXknqLgtfSRog1WqVSqXC5OQkMzMzTE5OUqlUqFarRYcmSZJUGIc6S9IAmWtgNT4+TqPRoFQqUavVbGyljjlsV5I0SCx8JRWq21+uwS/Y5XKZcrnM1NQUa9euLToc9al2PkenXLqZOy575jJEI0nS4bHwlVSodotUv2BLUv94/Otv5gc/vq9r2zvl0s1d2c6xDzmCz7/2zK5sS1J/sfCVJElSV/3gx/d17WBlN0evdKuAltR/bG4lSZIkSRpoFr6SJEmSpIFm4StJkiRJGmgWvpIkSZKkgWbhK0mSJEkaaBa+kiRJ0gCo1+uMjIywbt06RkZGqNfrRYck9QwvZyRJkiT1uXq9TrVaZWJigtnZWYaGhqhUKgCUy+WCo5OK5xlfSdLAi4i2bmNjY22tJ0m9plarMTExwdjYGMPDw4yNjTExMUGtVis6NKknWPhKkgZeZrZ1O/mS97W1niT1mkajwejo6D7LRkdHaTQaBUUk9RYLX0mSJKnPlUolpqen91k2PT1NqVQqKCKpt7RV+EbEWRHxxYj4SkRcusDzJ0XEZER8NiK+EBHP6H6okiRJkhZSrVapVCpMTk4yMzPD5OQklUqFarVadGhST1i0uVVEDAFvBZ4G7AQ+GRHvzczt81b7Y+DGzPzriFgDvB84ZQnilSRJkrSfuQZW4+PjNBoNSqUStVrNxlZSSztdnX8D+EpmfhUgIq4HngvML3wTeFjr/rHA17sZpCTpp7rdXMk5q5I0GMrlMuVymampKdauXVt0OFJPaafwfRRw57zHO4En7rfO64CbI2IcWAU8daENRcT5wPkAJ5xwAlNTUx2Gu7D/unU3u+/ryqb2OuXSzV3Zzqoj4K3rVnVlW72sW/+XK4k565w5a5qcnGxrvZd+YDfvOGvx3z/mdV/mozPmq3MrJWfdep27du3qas5WQv67nbOVwpy1rx/fY926jm8ZeEdmXhURTwL+d0SMZOb981fKzGuAawBOP/307NaRqN0f2Mwdlz2zK9sCunqU7JRLNw/+EbcPrIDX2G3mrHPmrHPmrHPmrDPmq3MrJWddfJ1dPXu5QvLvGd9DsELeG93Sj++xdppb3QU8et7j1a1l81WAGwEy82PAg4HjuxGgJEmSJEmHo53C95PAYyPiMRFxJPAC4L37rfM1YB1ARJRoFr7/3s1AJUmSJEk6FIsWvpk5A1wIfBBo0OzefFtEvCEintNa7X8A50XE54E68NK0W4okSZIkqQe0Ncc3M99P8xJF85e9Zt797cAZ3Q1NkiRJ/eiY0qX88rWXdm+D13ZnM8eUALrXF6bX1Ot1arXa3ssZVatVL2cktXSruZUkSZIEwA8bl3Wt8Wi3m44Oqnq9TrVaZWJigtnZWYaGhqhUKgAWvxIWvivW419/Mz/4cfeuAdXNPyTHPuQIPv/aM7u2PUmSpEFXq9WYmJhgbGxs78GCiYkJxsfHLXwlLHxXrB/8+L6ePBILg300VpIkaSk0Gg1GR0f3WTY6Okqj0SgoIqm3tNPVWZIkSVIPK5VKTE9P77NsenqaUqlUUERSb7HwlSRJkvpctVqlUqkwOTnJzMwMk5OTVCoVqtVq0aFJPcGhzlKbenVetHOiJUnS3Dze8fHxvV2da7XawM7v7fb3MvC72aCz8JXa1Kvzop0TLUmSoFn8lsvlrvdf6UXd/F4GfjdbCRzqLEmSJEkaaBa+kiRJkqSBZuErSZIkDYB6vc7IyAjr1q1jZGSEer1edEhSz3COryRJ0kHY3FD9oF6vc9FFF7Fq1SoAdu/ezUUXXQQwsA2upE5Y+EqSJB2EzQ3VDy6++GKGh4fZtGkTs7OzDA0Nce6553LxxRdb+Eo41FmSJEnqezt37uTaa69lbGyM4eFhxsbGuPbaa9m5c2fRoUk9wcJXkiRJGgC33nrrPnN8b7311qJDknqGQ50lSZKkPvfwhz+cK664giuvvJI1a9awfft2XvWqV/Hwhz+86NCknmDhK0mSJPW5hz70odx///1s3LiRHTt2cPLJJ/Owhz2Mhz70oUWHJvUEhzpLkiRJfe7rX/86b37zm1m1ahURwapVq3jzm9/M17/+9aJDk3qCha8kSZLU50qlEqtXr2bbtm1s3bqVbdu2sXr1akqlUtGhST3BwleSJEnqc9VqlUqlwuTkJDMzM0xOTlKpVKhWq0WHJvUE5/hKkiRJfW7uWr3j4+M0Gg1KpRK1Ws1r+EotFr6SJEnqulMu3dy9jX2gO9s69iFHdGU7vapcLlMul5mammLt2rVFhyP1FAtfSZIkddUdlz2za9s65dLNXd2epJXJOb6SJEmSpIHmGd8V6pjSpfzytZd2b4PXdm9Tx5QAPLIrSeoNvfo307+XWsm6/rkEP5sDzsJ3hfph47KuDRvq9jySrs4JkiTpMPXq30z/Xmol6+bnEvxsrgQWvpIkSVIfiIiubi8zu7o9qZc5x1eSJEnqA5nZ1u3kS97X1nrSSuIZX0mSJEkSMLgjCyx8JUl96/Gvv5kf/Pi+rm6zW3Ozjn3IEXz+tWd2ZVuSJC2XdgrVfrzMmIWvpCXT7aLEgkT7+8GP77O5iSRJWtRAFL62M5d6UzeLEgsSSZIkHaqBKHxtZy5JkiRJOpC2Ct+IOAu4GhgC3p6Zly2wzvOA1wEJfD4z13cxTkmS1AW9OgUBnIYgSVo6ixa+ETEEvBV4GrAT+GREvDczt89b57HAHwFnZOb3IuLnlipgSZJ06Hp1CgI4SkqStHTauY7vbwBfycyvZuYe4Hrgufutcx7w1sz8HkBm3t3dMCVJkiRJOjTtDHV+FHDnvMc7gSfut84vAUTEP9McDv26zPzA/huKiPOB8wFOOOEEpqamDiHkhXVzW7t27erZ2LqpW3F1O1/QmznrehO1LjZQm5pa1Z2NLYFefZ/14ntsKayE1+nv/8716ucSzFmnejVf3bZSXmc3rYSc+fu/WH33GjPzoDfg92nO6517/CLgLfut8z7g/wBHAI+hWSgfd7DtPuEJT8huOfmS93VtW5mZk5OTXdtWt2Prlm7G1c18ZZqzTvVqvjLNWdFWwuv093/nevVzmWnOOtWr+eq2lfI6u2kl5Mzf/8Xq5dcIfCoXqD/bGep8F/DoeY9Xt5bNtxN4b2bel5m3A18CHnuoxbgkSZIkSd3STuH7SeCxEfGYiDgSeAHw3v3WuQlYCxARx9Mc+vzV7oUpSZIkSdKhWbTwzcwZ4ELgg0ADuDEzb4uIN0TEc1qrfRD4TkRsByaBV2Xmd5YqaEmSJEmS2tXWdXwz8/3A+/db9pp59xN4ResmSTpEvXqNVa+vKkmS+llbha8kaXn06jVWvb6qJEnqZ+3M8ZUkSZIkqW9Z+EqSJEmSBppDnSVJkiT1na5Pw/lA9/piqPdY+EqSJEnqK93qhzHnlEs3d32b6i0OdZYkSZIkDTTP+EqSJC2iq0MqHU4pSctuYApfx/hL0spzTOlSfvnaS7u70Wu7s5ljSgAOmxsE3Rz+6HBKSSrGQBS+jvE/NL149Bo8WCCpfT9sXNbV39de+1iSpME0EIWvOufRa0mSJEkrhc2tJEmSJEkDzcJXkiRJkjTQHOosacl0vfGQTYekw9arn0vwsylJWjoWvlIHerEhWC83A+tm4yGbDknd0aufS/CzKUlaOha+UptsCCZJkiT1JwtfSZIkqWCPf/3N/ODH93Vte90aQXHsQ47g8689syvbkopk4StJkiQV7Ac/vq8npyE4BUGDwq7OkiRJkqSBZuErSZIkSRpoFr6SJEmSpIHmHF9JkiRJGnC92kANlqeJmoWvJEmSJA24Xm2gBsvTRM3CV5LU17r+x/ID3bsEiCRJ6g0WvpKkvtWtI9dzTrl0c9e3KUmSimfhK0k95JjSpfzytZd2b4PXdmczx5QALAglSVJ/svCVpB7yw8ZlPTn/Zjnm3kiSJC0VC19JS6qrBZNzLyVJA8oRP9LSsvCVtGS6OVfSuZeSpEHmiB9paT2o6AAkSZIkSVpKFr6SJEmSpIHmUGdJkiQtu4hof93L21svMw8xGkmDzsJXkiRJy67dIrWb81WllaxXG6jB8jRRa6vwjYizgKuBIeDtmXnZAdb7PeA9wK9n5qe6FqUkSeqaXuy2DnZcl6Sl1KsN1GB5mqgtWvhGxBDwVuBpwE7gkxHx3szcvt96xwAXAf+yFIFKkqTDZ7d1SdJK1E5zq98AvpKZX83MPcD1wHMXWO9PgMuBn3QxPkmSJEmSDks7Q50fBdw57/FO4InzV4iIXwMenZmbI+JVB9pQRJwPnA9wwgknMDU11XHAy6WXY+tF5qtz5qxzKyVn3Xqdu3bt6mrOzL8WYr46Z8460+3fZb3M3//FWgmvs1ffY7D0+T/s5lYR8SDgz4GXLrZuZl4DXANw+umnZ882KvjAZpsodMJ8dc6cdW6l5KyLr7Or82/MvxZivjpnzjq2Yppb+fu/WCvhdfbqewyWJf/tDHW+C3j0vMerW8vmHAOMAFMRcQfwm8B7I+L0bgUpSZIkSdKhaueM7yeBx0bEY2gWvC8A1s89mZk/AI6fexwRU8Ar7eosSZIkta8XO67bbV2DYtHCNzNnIuJC4IM0L2e0KTNvi4g3AJ/KzPcudZCSJEnSILPjurS02prjm5nvB96/37LXHGDdtYcfliStXB7xlyRJ6q7Dbm4lSeoej/hLkiR1XzvNrSRJkiRJ6lsWvpIkSZKkgWbhK0mSJEkaaBa+kiRJkqSBZnMrSZIkSVoBevHKEbA8V4+w8JUkSZKkAbfSrxzhUGdJkiRJ0kCz8JUkSZIkDTQLX0mSJEnSQLPwlSRJkiQNNAtfSZIkSdJAs/CVJEmSJA00L2ckSZIkaSBFRPvrXr74Opl5GNGoSJ7xlSRJkjSQMrOt2+TkZFvrqX9Z+EqSJEmSBpqFryRJkiRpoFn4SpIkSZIGmoWvJEmSpBWpXq8zMjLCunXrGBkZoV6vFx2SlohdnSVJkrqg3e6x7XSOBbvHSkutXq9TrVaZmJhgdnaWoaEhKpUKAOVyueDo1G2e8ZUkSeqCbnaOtej1TJyWXq1WY/369YyPj/P0pz+d8fFx1q9fT61WKzo0LQHP+EqSJKmneCZOy2H79u3s3r2bTZs27X2fbdiwgR07dhQdmpaAZ3wlSZLUU2q1GhMTE4yNjTE8PMzY2BgTExOeiVNXHXnkkYyPj+/zPhsfH+fII48sOjQtAc/4SpIkqac0Gg1GR0f3WTY6Okqj0SgoIg2iPXv28Ja3vIXTTjuN2dlZJicnectb3sKePXuKDk1LwMJXkiRJPaVUKjE9Pc3Y2NjeZdPT05RKpQKj0qBZs2YN55xzDuPj4zQaDUqlEuvXr+emm24qOjQtgRU11Dki2rrtuPxZba0nSZLUDhs1daZarVKpVJicnGRmZobJyUkqlQrVarXo0DRAqtUq1113HRs3buSDH/wgGzdu5LrrrvN9NqBW1BnfdjskTk1NsXbt2qUNRpIkrQg2aurcXF7mn4mr1WrmS13l+2xlWVFnfCVJkpabjZoOTblcZtu2bWzdupVt27ZZjGhJ+D5bOSx8JUmSlpCNmiSpeBa+kiRJS2iuUdN8NmqSpOVl4SupUN1uOmfjOUm9xkZNWi5zTdR2XPEcm6hJ+2mruVVEnAVcDQwBb8/My/Z7/hXAy4AZ4N+BDZm5o8uxSn2h3cIrLm9ve+02ZetXNp2TNOhsoKPlML+J2kvffw8bn/Ewm6hJ8yxa+EbEEPBW4GnATuCTEfHezNw+b7XPAqdn5o8i4j8DVwDPX4qApV7XTiFnESdJK0u5XKZcLvv7X4elnYPrT3nKU5r/vqn5eP369axfv37BdQf94Lo0XztnfH8D+EpmfhUgIq4HngvsLXwzc3Le+h8HXtjNIFUcz15KGgSdDIFv5/eZv8skFeFgv3uGhob4yU9+whFHHLH3AMt9993Hgx/8YGZnZ5cxSqk3tTPH91HAnfMe72wtO5AKsOVwglLvyMxFb5OTk22t5xdFSUVp93dUu7/PJKnX2ERNOri25vi2KyJeCJwOPPkAz58PnA9wwgknMDU11c3dd82uXbt6NrZeZL46Z846Z84OjTnrjO+zzpmvzvge65w5a8/v/M7vcO655/KqV72KxzzmMfzFX/wFV155JZVKxfy1wfdZ5/otX+0UvncBj573eHVr2T4i4qlAFXhyZt670IYy8xrgGoDTTz89e3WOi/NvOmO+OmfOOmfODsEHNpuzDvk+65DvsY75HuucOWvP2rVrWbNmDbVabW8TtauuusrGVm3yfdahPvz9307h+0ngsRHxGJoF7wuAfWbIR8RpwP8CzsrMu7sepSRJkqSDsomadGCLzvHNzBngQuCDQAO4MTNvi4g3RMRzWqtdCRwNvDsiPhcR712yiCVJkiRJ6kBbc3wz8/3A+/db9pp595/a5bjUB+r1+j7DaarVqsNpJGlA2NVfkjRI2unqLD3A3EXSN27cyAc/+EE2btxItVqlXq8XHZoGTL1eZ2RkhHXr1jEyMuJ7TFomdvWXJA0SC18dklqtxsTEBGNjYwwPDzM2NsbExAS1Wq3o0DRA6vU6F110Ebt37wZg9+7dXHTRRRa/kvqOB/EkqVhdvZyRVo5Go8Ho6Og+y0ZHR2k0GgVFpEF08cUXMzw8zKZNm5idnWVoaIhzzz2Xiy++2GH1kvrG3CipiYmJvb/LKpUKgL/LJPWVuamOO7Y3GHlff0119IyvDokXSddy2LlzJ9dee+0+IwuuvfZadu7cWXRoktQ2R0lJ6icRccDb+vXrue222yDv57bbbmP9+vUHXb/dfhHLwcJXh6RarVKpVJicnGRmZobJyUkqlQrVarXo0CRJ6imOkpLUTw7Uq+HUU0/l1ltv3afHw6233sqpp57aFz0eHOqsQzI3pGF8fHxvV+dardY3Qx3UH1avXs2LX/xirrvuOmZnZ5mcnOTFL34xq1evLjo0SWrb3CipsbGxvcscJSWp3/T7QTzP+OqQlctltm3bxtatW9m2bZtFr7ruiiuuYHZ2lg0bNnDmmWeyYcMGZmdnueKKK4oOTZLa5igpSYOg36c6esZXUs+aO5hSq9WICFatWsUb3/hGD7JI6iuOkpI0COYO4s016ps7iNcv/QosfCX1tHK5TLlcZmpqirVr1xYdjiQdEn+XSep3/X4Qz8JXkiRJkrSofj6I5xxfSZIkSdJAs/CVJEmSBkC9XmdkZIR169YxMjJCvV4vOiSpZzjUWZIkSepz9XqdarW6t/HQ0NAQlUoFoG/mYEpLyTO+kiRJUp+r1WpMTEwwNjbG8PAwY2NjTExM9E3HXWmpWfhKkiRJfa7RaDA6OrrPstHRURqNRkERSb3FwleS+kxEtHXbcfmz2lpPktT/SqUS09PT+yybnp6mVCoVFJHUWyx8JanPZGZbt8nJybbWkyT1v2q1SqVSYXJykpmZGSYnJ6lUKlSr1aJDk3qCza0kSZKkPjfXwGp8fJxGo0GpVKJWq9nYSmqx8JUkSZIGQLlcplwuMzU1xdq1a4sOR+opDnWWJEmSJA00C19JkiRJ0kCz8JUkSZIkDTQLX0mSJEnSQLPwlSRJkiQNNAtfSZIkSdJAs/CVJEmSJA00C19JkiRJ0kCz8JUkSZIkDTQLX0mSJEnSQLPwlSRJkiQNtMjMYnYc8e/AjkJ2vrjjgW8XHUQfMV+dM2edM2edM2edM2edMV+dM2edM2edM2edM2ed6eV8nZyZj9h/YWGFby+LiE9l5ulFx9EvzFfnzFnnzFnnzFnnzFlnzFfnzFnnzFnnzFnnzFln+jFfDnWWJEmSJA00C19JkiRJ0kCz8F3YNUUH0GfMV+fMWefMWefMWefMWWfMV+fMWefMWefMWefMWWf6Ll/O8ZUkSZIkDTTP+EqSJEmSBpqFryRJkiRpoFn4zhMRx0XEeyLiXyOiERFPKjqmXhYRj4uIz8273RMRLy86rl4TEZsi4u6I2DZv2cMj4paI+HLr358pMsZec4Cc/UFE3BYR90dEX7XPXw4HyNkN8z6fd0TE5woMsadExKMjYjIitrfeVxe1lvvZPICD5OxPIuILrffZzRFxYtGx9oqD5Ox1EXHXvM/nM4qOtRccKF+t58Zb389ui4grioyzl0TEgyPiExHx+VZuXt9a/pSI+ExEbIuIayNiuOhYe01EDEXEZyPifa3HH5n3mfx6RNxUcIg940Df+SPiytbn8gsR8X8i4riiYz0Y5/jOExHXAh/JzLdHxJHAQzPz+wWH1RciYgi4C3hiZu4oOp5eEhG/DewC3pmZI61lVwDfzczLIuJS4Gcy85Ii4+wlB8hZCbgf+F/AKzPzUwWG2HMWytl+z18F/CAz37DswfWgiHgk8MjM/ExEHAN8GjgHeCl+Nhd0kJztzMx7Wuv8N2BNZl5QXKS94yA5ex6wKzPfVGR8veYg+ToBqALPzMx7I+LnMvPuAkPtGRERwKrM3BURRwDTwH8HbgDWZeaXIuINwI7MnCgy1l4TEa8ATgcelpnP2u+5vwf+MTPfWUhwPWz+d37gccCtmTkTEZcD9PLfTM/4tkTEscBvAxMAmbnHorcj64B/s+h9oMz8MPDd/RY/F7i2df9amn/Y1bJQzjKzkZlfLCiknneA9xmw94vR84D6sgbVwzLzG5n5mdb9HwIN4FH42TygA+VsruhtWQV4RL3lIO8zLeAg+frPwGWZeW/rOYvelmza1Xp4ROs2C+zJzC+1lt8C/F4R8fWqiFgNPBN4+wLPPQx4CnDTMofVL/Z+58/MmzNzprX848DqAuNalIXvTz0G+Hfgb1vDHt4eEauKDqqPvAC/VHfihMz8Ruv+N2kezZaWym8B38rMLxcdSC+KiFOA04B/wc9mW/bLGRFRi4g7gXOB1xQYWs/aP2fAha3hgZscUv9A++Xrl4Dfioh/iYgPRcSvFxpcj2kN2f0ccDfNIvcTwHD8dFrQ7wOPLii8XvWXwMU0R5Lt7xxg634H9fRTB/rOvwHYssyxdMTC96eGgV8D/jozTwN2A5cWG1J/aA0Lfw7w7qJj6UfZnG/gGRItpTIemFpQRBwN/D3w8v2/5PjZXNhCOcvMamY+Gvg74MIi4+tFC+Tsr4FfAH4V+AZwVXHR9Z4F8jUMPBz4TeBVwI2tkSwCMnM2M3+V5tm23wBOpVmc/EVEfAL4Ic2zwAIi4lnA3Zn56QOs4t/MAzjQd/6IqAIzNP8G9CwL35/aSXOe0tyR2PfQLIS1uLOBz2Tmt4oOpI98qzWXaW5Ok8O2tCRaDU1+l+Z8L83Tmg/398DfZeY/tBb72TyIA+Rsvr/DIZX7WChnmfmtVrFyP/A3NIsVccD32E7gH1rDej9B8yzd8UXF2KtaU/QmgbMy82OZ+VuZ+RvAh4EvHfSHV5YzgOdExB3A9cBTIuJdABFxPM3P4+biwutpD/jOHxEvBZ4FnJs93jzKwrclM78J3BkRj2stWgdsLzCkfuKRsc69F3hJ6/5LgH8sMBYNtqcC/5qZO4sOpJe0zhZNAI3M/PN5T/nZPIAD5SwiHjtvtecC/7rcsfWqg+TskfNW+x1g2/4/uxId5HN5EzDWWueXgCOBby97gD0oIh4x10k3Ih4CPA3414j4udayo4BLgLcVFmSPycw/yszVmXkKzTPjt2bmC1tP/z7wvsz8SWEB9rZ9vvNHxFk0h4w/JzN/VFhUbbKr8zwR8as0J7kfCXwV+MPM/F6hQfW41jzorwH/ITN/UHQ8vSgi6sBamkenvwW8luYf8RuBk4AdwPMyc8HGRCvRAXL2XWAj8Ajg+8DnMvPpBYXYcxbKWWZORMQ7gI9npl965omIUeAjwP/jp3O8Xk1zPqGfzQUcJGcVmp0976eZswsy865CguwxB8lZmeYw5wTuAP7TvLnlK9ZB8vV/gU00c7aHZmf/W4uIsddExK/QbMQ3RPOE1o2Z+YaIuJLmWbgH0ZzG95fFRdm7ImItzffTs1qPp2g2UvtAgWH1pIW+80fEV4CjgO+0Vvt4L3f1t/CVJEmSJA00hzpLkiRJkgaaha8kSZIkaaBZ+EqSJEmSBpqFryRJkiRpoFn4SpIkSZIGmoWvJEmSJGmgWfhKktRlEXFcRPyXHojj3Ij4QkT8v4j4aEQ8vuiYJEkqgoWvJEnddxxQeOEL3A48OTN/GfgT4JqC45EkqRAWvpIkdd9lwC9ExOci4srWbVvrzOvzASLirRHxnNb9/xMRm1r3N0RELSJOiYhGRPxNRNwWETdHxENa6/y3iNjeOpt7/YGCyMyPZub3Wg8/Dqxe0lctSVKPsvCVJKn7LgX+LTN/lWbB+avA44GnAldGxCOBjwC/1Vr/UcCa1v3fAj7cuv9Y4K2ZeSrwfeD35m3/tMz8FeCCNmOqAFsO7eVIktTfLHwlSVpao0A9M2cz81vAh4Bfp1X4RsQaYDvwrVZB/CTgo62fvT0zP9e6/2nglNb9LwB/FxEvBGYWCyAixmgWvpd05RVJktRnLHwlSSpAZt5Fcy7wWTTP8H4EeB6wKzN/2Frt3nk/MgsMt+4/E3gr8GvAJyNimAOIiF8B3g48NzO/083XIElSv7DwlSSp+34IHNO6/xHg+RExFBGPAH4b+ETruY8DL+enhe8rW/8eUEQ8CHh0Zk7SPIN7LHD0AdY9CfgH4EWZ+aXDeUGSJPWzAx4hliRJhyYzvxMR/xwR22jOq/0C8HkggYsz85utVT8CnJmZX4mIHcDDWaTwBYaAd0XEsUAAb87M7x9g3dcAPwv8VUQAzGTm6Yfx0iRJ6kuRmUXHIEmSJEnSknGosyRJkiRpoDnUWZKkPhcRfwhctN/if87M/1pEPJIk9RqHOkuSJEmSBppDnSVJkiRJA83CV5IkSZI00Cx8JUmSJEkDzcJXkiRJkjTQ/n9FYI36uS7TXAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "analysis_data.loc[analysis_data.city=='Taipei'].boxplot(column=['unit_price_square_feet'],by='towns_2',figsize=(16,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['大安區'], dtype=object)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "le.inverse_transform([26])"
   ]
  },
  {
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}