아파트실거래가_예측.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "아파트실거래가 예측.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DzV7t814TNGz"
      },
      "source": [
        "transaction_id : 아파트 거래에 대한 유니크한 아이디\n",
        "\n",
        "apartment_id : 아파트 아이디\n",
        "\n",
        "city : 도시\n",
        "\n",
        "dong : 동\n",
        "\n",
        "jibun : 지번\n",
        "\n",
        "apt : 아파트단지 이름\n",
        "\n",
        "addr_kr : 주소\n",
        "\n",
        "exclusive_use_area : 전용면적\n",
        "\n",
        "year_of_completion : 설립일자\n",
        "\n",
        "transaction_year_month : 거래년월\n",
        "\n",
        "transaction_date : 거래날짜\n",
        "\n",
        "floor : 층\n",
        "\n",
        "transaction_real_price : 실거래가 (target variable)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hl0zgEPw4b9q"
      },
      "source": [
        "# 데이터 확인"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "USjcuKu4IWt4"
      },
      "source": [
        "# For example, here's several helpful packages to load in \n",
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "import matplotlib.pyplot as plt #data visualization\n",
        "import seaborn as sns\n",
        "import timeit # time measure\n",
        "import sklearn # modeling\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "import sys\n",
        "%matplotlib inline\n",
        "import re\n",
        "import warnings\n",
        "import pylab as pl\n",
        "from scipy.stats import mode\n",
        "\n",
        "test = pd.read_csv(\"test.csv\")\n",
        "train = pd.read_csv(\"train.csv\")\n",
        "submission = pd.read_csv(\"submission.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2om9pg15D5H1"
      },
      "source": [
        "pd.options.display.max_rows = 99"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Rqv2kH8HEbLn",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "cedbe7b7-1d9d-431b-adf7-1ad093266b24"
      },
      "source": [
        "display(train,test,submission)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "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>transaction_id</th>\n",
              "      <th>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>7622</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>신교동</td>\n",
              "      <td>6-13</td>\n",
              "      <td>신현(101동)</td>\n",
              "      <td>신교동 6-13 신현(101동)</td>\n",
              "      <td>84.8200</td>\n",
              "      <td>2002</td>\n",
              "      <td>200801</td>\n",
              "      <td>21~31</td>\n",
              "      <td>2</td>\n",
              "      <td>37500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>5399</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>필운동</td>\n",
              "      <td>142</td>\n",
              "      <td>사직파크맨션</td>\n",
              "      <td>필운동 142 사직파크맨션</td>\n",
              "      <td>99.1700</td>\n",
              "      <td>1973</td>\n",
              "      <td>200801</td>\n",
              "      <td>1~10</td>\n",
              "      <td>6</td>\n",
              "      <td>20000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>3578</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>필운동</td>\n",
              "      <td>174-1</td>\n",
              "      <td>두레엘리시안</td>\n",
              "      <td>필운동 174-1 두레엘리시안</td>\n",
              "      <td>84.7400</td>\n",
              "      <td>2007</td>\n",
              "      <td>200801</td>\n",
              "      <td>1~10</td>\n",
              "      <td>6</td>\n",
              "      <td>38500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>10957</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>내수동</td>\n",
              "      <td>95</td>\n",
              "      <td>파크팰리스</td>\n",
              "      <td>내수동 95 파크팰리스</td>\n",
              "      <td>146.3900</td>\n",
              "      <td>2003</td>\n",
              "      <td>200801</td>\n",
              "      <td>11~20</td>\n",
              "      <td>15</td>\n",
              "      <td>118000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>10639</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>내수동</td>\n",
              "      <td>110-15</td>\n",
              "      <td>킹스매너</td>\n",
              "      <td>내수동 110-15 킹스매너</td>\n",
              "      <td>194.4300</td>\n",
              "      <td>2004</td>\n",
              "      <td>200801</td>\n",
              "      <td>21~31</td>\n",
              "      <td>3</td>\n",
              "      <td>120000</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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1216548</th>\n",
              "      <td>1234823</td>\n",
              "      <td>9578</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>정관읍 용수리</td>\n",
              "      <td>1387</td>\n",
              "      <td>정관 동일스위트1차</td>\n",
              "      <td>정관읍 용수리 1387 정관 동일스위트1차</td>\n",
              "      <td>59.8245</td>\n",
              "      <td>2012</td>\n",
              "      <td>201711</td>\n",
              "      <td>11~20</td>\n",
              "      <td>3</td>\n",
              "      <td>22000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1216549</th>\n",
              "      <td>1234824</td>\n",
              "      <td>9578</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>정관읍 용수리</td>\n",
              "      <td>1387</td>\n",
              "      <td>정관 동일스위트1차</td>\n",
              "      <td>정관읍 용수리 1387 정관 동일스위트1차</td>\n",
              "      <td>84.9923</td>\n",
              "      <td>2012</td>\n",
              "      <td>201711</td>\n",
              "      <td>21~30</td>\n",
              "      <td>7</td>\n",
              "      <td>30750</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1216550</th>\n",
              "      <td>1234825</td>\n",
              "      <td>9578</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>정관읍 용수리</td>\n",
              "      <td>1387</td>\n",
              "      <td>정관 동일스위트1차</td>\n",
              "      <td>정관읍 용수리 1387 정관 동일스위트1차</td>\n",
              "      <td>84.9923</td>\n",
              "      <td>2012</td>\n",
              "      <td>201711</td>\n",
              "      <td>21~30</td>\n",
              "      <td>9</td>\n",
              "      <td>29500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1216551</th>\n",
              "      <td>1234826</td>\n",
              "      <td>9586</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>정관읍 용수리</td>\n",
              "      <td>1364</td>\n",
              "      <td>정관신도시롯데캐슬</td>\n",
              "      <td>정관읍 용수리 1364 정관신도시롯데캐슬</td>\n",
              "      <td>101.7400</td>\n",
              "      <td>2008</td>\n",
              "      <td>201711</td>\n",
              "      <td>21~30</td>\n",
              "      <td>7</td>\n",
              "      <td>26700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1216552</th>\n",
              "      <td>1234827</td>\n",
              "      <td>9578</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>정관읍 용수리</td>\n",
              "      <td>1387</td>\n",
              "      <td>정관 동일스위트1차</td>\n",
              "      <td>정관읍 용수리 1387 정관 동일스위트1차</td>\n",
              "      <td>84.9923</td>\n",
              "      <td>2012</td>\n",
              "      <td>201711</td>\n",
              "      <td>21~30</td>\n",
              "      <td>16</td>\n",
              "      <td>28800</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1216553 rows × 13 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         transaction_id  apartment_id  ... floor transaction_real_price\n",
              "0                     0          7622  ...     2                  37500\n",
              "1                     1          5399  ...     6                  20000\n",
              "2                     2          3578  ...     6                  38500\n",
              "3                     3         10957  ...    15                 118000\n",
              "4                     4         10639  ...     3                 120000\n",
              "...                 ...           ...  ...   ...                    ...\n",
              "1216548         1234823          9578  ...     3                  22000\n",
              "1216549         1234824          9578  ...     7                  30750\n",
              "1216550         1234825          9578  ...     9                  29500\n",
              "1216551         1234826          9586  ...     7                  26700\n",
              "1216552         1234827          9578  ...    16                  28800\n",
              "\n",
              "[1216553 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "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>transaction_id</th>\n",
              "      <th>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1145756</td>\n",
              "      <td>10453</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>목동</td>\n",
              "      <td>938</td>\n",
              "      <td>청학</td>\n",
              "      <td>목동 938 청학</td>\n",
              "      <td>35.5500</td>\n",
              "      <td>2002</td>\n",
              "      <td>201711</td>\n",
              "      <td>11~20</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1198704</td>\n",
              "      <td>989</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>초량동</td>\n",
              "      <td>1143-8</td>\n",
              "      <td>고관맨션</td>\n",
              "      <td>초량동 1143-8 고관맨션</td>\n",
              "      <td>68.7200</td>\n",
              "      <td>1977</td>\n",
              "      <td>201708</td>\n",
              "      <td>21~31</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1222384</td>\n",
              "      <td>8597</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>괴정동</td>\n",
              "      <td>447-13</td>\n",
              "      <td>우림그린</td>\n",
              "      <td>괴정동 447-13 우림그린</td>\n",
              "      <td>72.5400</td>\n",
              "      <td>1989</td>\n",
              "      <td>201710</td>\n",
              "      <td>11~20</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1179897</td>\n",
              "      <td>11086</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>대치동</td>\n",
              "      <td>1007-2</td>\n",
              "      <td>풍림아이원4차(1007-2)</td>\n",
              "      <td>대치동 1007-2 풍림아이원4차(1007-2)</td>\n",
              "      <td>111.5400</td>\n",
              "      <td>2004</td>\n",
              "      <td>201707</td>\n",
              "      <td>1~10</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1223091</td>\n",
              "      <td>2121</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>다대동</td>\n",
              "      <td>1670</td>\n",
              "      <td>다대롯데캐슬블루</td>\n",
              "      <td>다대동 1670 다대롯데캐슬블루</td>\n",
              "      <td>119.6398</td>\n",
              "      <td>2014</td>\n",
              "      <td>201712</td>\n",
              "      <td>11~20</td>\n",
              "      <td>21</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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5458</th>\n",
              "      <td>1174640</td>\n",
              "      <td>8965</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>양재동</td>\n",
              "      <td>15-9</td>\n",
              "      <td>위너스</td>\n",
              "      <td>양재동 15-9 위너스</td>\n",
              "      <td>84.0500</td>\n",
              "      <td>2003</td>\n",
              "      <td>201710</td>\n",
              "      <td>11~20</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5459</th>\n",
              "      <td>1175575</td>\n",
              "      <td>3831</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>반포동</td>\n",
              "      <td>18-1</td>\n",
              "      <td>래미안퍼스티지</td>\n",
              "      <td>반포동 18-1 래미안퍼스티지</td>\n",
              "      <td>84.9300</td>\n",
              "      <td>2009</td>\n",
              "      <td>201712</td>\n",
              "      <td>1~10</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5460</th>\n",
              "      <td>1157024</td>\n",
              "      <td>5550</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>구로동</td>\n",
              "      <td>256-1</td>\n",
              "      <td>삼성래미안</td>\n",
              "      <td>구로동 256-1 삼성래미안</td>\n",
              "      <td>110.1880</td>\n",
              "      <td>2004</td>\n",
              "      <td>201712</td>\n",
              "      <td>21~31</td>\n",
              "      <td>23</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5461</th>\n",
              "      <td>1136863</td>\n",
              "      <td>3371</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>홍제동</td>\n",
              "      <td>24-16</td>\n",
              "      <td>동양드림</td>\n",
              "      <td>홍제동 24-16 동양드림</td>\n",
              "      <td>92.4400</td>\n",
              "      <td>2005</td>\n",
              "      <td>201710</td>\n",
              "      <td>21~31</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5462</th>\n",
              "      <td>1218921</td>\n",
              "      <td>6825</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>재송동</td>\n",
              "      <td>1187</td>\n",
              "      <td>센텀그린타워</td>\n",
              "      <td>재송동 1187 센텀그린타워</td>\n",
              "      <td>84.6400</td>\n",
              "      <td>1993</td>\n",
              "      <td>201712</td>\n",
              "      <td>1~10</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5463 rows × 12 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      transaction_id  apartment_id  ... transaction_date floor\n",
              "0            1145756         10453  ...            11~20     2\n",
              "1            1198704           989  ...            21~31     2\n",
              "2            1222384          8597  ...            11~20     2\n",
              "3            1179897         11086  ...             1~10    10\n",
              "4            1223091          2121  ...            11~20    21\n",
              "...              ...           ...  ...              ...   ...\n",
              "5458         1174640          8965  ...            11~20     7\n",
              "5459         1175575          3831  ...             1~10     4\n",
              "5460         1157024          5550  ...            21~31    23\n",
              "5461         1136863          3371  ...            21~31     1\n",
              "5462         1218921          6825  ...             1~10     7\n",
              "\n",
              "[5463 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "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>transaction_id</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1145756</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1198704</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1222384</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1179897</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1223091</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5458</th>\n",
              "      <td>1174640</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5459</th>\n",
              "      <td>1175575</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5460</th>\n",
              "      <td>1157024</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5461</th>\n",
              "      <td>1136863</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5462</th>\n",
              "      <td>1218921</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5463 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      transaction_id  transaction_real_price\n",
              "0            1145756                       0\n",
              "1            1198704                       0\n",
              "2            1222384                       0\n",
              "3            1179897                       0\n",
              "4            1223091                       0\n",
              "...              ...                     ...\n",
              "5458         1174640                       0\n",
              "5459         1175575                       0\n",
              "5460         1157024                       0\n",
              "5461         1136863                       0\n",
              "5462         1218921                       0\n",
              "\n",
              "[5463 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oOvFz9soWOnz",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 417
        },
        "outputId": "0f59dd98-8d62-4807-9e6f-19337d24eeae"
      },
      "source": [
        "train_null = train.drop(\"transaction_real_price\", axis = 1).isnull().sum()/len(train.isnull().sum()) # 나중에 데이터프레임으로 묶어줘야되기에 train,test의 컬럼수가 동일해야되기때문에 train데이터에서 transaction_real_price컬럼을 뺀다.\n",
        "#train_null\n",
        "test_null = test.isnull().sum()/len(test.isnull().sum())\n",
        "#test_null\n",
        "pd.DataFrame({\"train_null\" : train_null, \"test_null\" : test_null}) # 데이터프레임 형태로 바꿔줌"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>train_null</th>\n",
              "      <th>test_null</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>transaction_id</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>apartment_id</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>city</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>dong</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>jibun</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>apt</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>addr_kr</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>year_of_completion</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>transaction_year_month</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>transaction_date</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>floor</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                        train_null  test_null\n",
              "transaction_id                 0.0        0.0\n",
              "apartment_id                   0.0        0.0\n",
              "city                           0.0        0.0\n",
              "dong                           0.0        0.0\n",
              "jibun                          0.0        0.0\n",
              "apt                            0.0        0.0\n",
              "addr_kr                        0.0        0.0\n",
              "exclusive_use_area             0.0        0.0\n",
              "year_of_completion             0.0        0.0\n",
              "transaction_year_month         0.0        0.0\n",
              "transaction_date               0.0        0.0\n",
              "floor                          0.0        0.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i31kBfxpYCPP"
      },
      "source": [
        "# Target variable\n",
        "## transaction_real_price : 실거래가"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MiIk5hnjX0NX",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9655b514-7a09-4384-b069-ccac6890de58"
      },
      "source": [
        "train['transaction_real_price'].describe()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "count    1.216553e+06\n",
              "mean     3.822769e+04\n",
              "std      3.104898e+04\n",
              "min      1.000000e+02\n",
              "25%      1.900000e+04\n",
              "50%      3.090000e+04\n",
              "75%      4.700000e+04\n",
              "max      8.200000e+05\n",
              "Name: transaction_real_price, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JseKm7vuMrMW",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "457be0d8-71ab-4eb6-b083-42aede69c1d8"
      },
      "source": [
        "f,ax = plt.subplots(figsize = (8,6))\n",
        "sns.distplot(train['transaction_real_price'])\n",
        "print(\"Skewness : %s , Kurtosis : %s\",(train['transaction_real_price'].skew(),train['transaction_real_price'].kurt())) # 왜도 - skew(), 첨도 - kurt()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Skewness : %s , Kurtosis : %s (3.4071692420405535, 24.839820787251245)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Gw2rIk0hOTuI"
      },
      "source": [
        "- 왜도(Skewness) : 왼쪽으로 치우쳐져 있을수록 값이크고, 오른쪽으로 치우쳐져 있을 수록 값이 작아진다. 즉, 0에 가까울수록 좋은 형태\n",
        "\n",
        "- 첨도(Kurtosis) : 첨도 값이 3에 가까울 경우 정규분포에 가까우며, 첨도 값이 클수록 뾰족하고 값이 작을 수록 완만해진다"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rE4O9LKqOMz4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 427
        },
        "outputId": "33fcc485-1551-4482-b9dc-d243a59613fd"
      },
      "source": [
        "train['transaction_real_price'] = np.log1p(train['transaction_real_price']) # np.log1p(z) = log(1+z)\n",
        "print(\"Skewness : %f , Kurtosis : %f\",(train['transaction_real_price'].skew(),train['transaction_real_price'].kurt()))\n",
        "f,ax = plt.subplots(figsize = (8,6))\n",
        "sns.distplot(train['transaction_real_price'])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Skewness : %f , Kurtosis : %f (-0.09493151731286435, 0.24886558181939034)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbe1b64400>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RwKBqVuTSXCt"
      },
      "source": [
        "# Columns 상관관계"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r09bC_ziT4CW"
      },
      "source": [
        "#cols = corrmat.nlargest(k, 'transaction_real_price')['transaction_real_price'].index\n",
        "#cols\n",
        "#cm = np.corrcoef(train[cols].values.T)\n",
        "#cm"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OGdXViZFSVOw",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 508
        },
        "outputId": "f495fd7f-827c-4852-d10d-1e80e2ba0ca0"
      },
      "source": [
        "k = train.shape[1]\n",
        "corrmat = train.corr()\n",
        "cols = corrmat.nlargest(k, 'transaction_real_price')['transaction_real_price'].index #nlargest(N,내부) - 내부에서 상위에 N개만 추출\n",
        "cm = np.corrcoef(train[cols].values.T) #상관계수 구하기\n",
        "f,ax = plt.subplots(figsize = (8,6))\n",
        "sns.heatmap(data = cm, annot=True, square= True, fmt= '.2f', linewidths= 5, cmap ='Reds', yticklabels= cols.values, xticklabels= cols.values)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbe1617e10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Po3gfV4ZWjmb"
      },
      "source": [
        "# Numeric Variable\n",
        "## transaction_id & apartment_id : 거래id & 아파트 id"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "E3wKlahcXA15",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a835a502-b2b8-4604-b135-8872ad5b7528"
      },
      "source": [
        "len(train['apartment_id'].unique()) # 12533개로 겹치는 부분이 상당히 많은것을 알 수 있다."
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "12533"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Apem_jW6XZMy"
      },
      "source": [
        "train_id = train['transaction_id']\n",
        "train = train.drop('transaction_id',axis =1)\n",
        "test_id = test['transaction_id']\n",
        "test = test.drop('transaction_id',axis =1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8YX3JeQKXvoC"
      },
      "source": [
        "# floor : 층"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O7LOrLkYXy3o",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 409
        },
        "outputId": "c415a8a9-cc16-4946-eb6c-76e5f18f6c9e"
      },
      "source": [
        "f, ax = plt.subplots(figsize = (8,6))\n",
        "sns.scatterplot(train['floor'],train['transaction_real_price']) # plt.scatter() -> sns.scatterplot()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbe168f6a0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c5u7IlPvfFSA"
      },
      "source": [
        "# train[(train['floor']<0)&(train['floor']<6)] #이상치 확인"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xzVzirGlYBPo",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "outputId": "7bc19931-2ebb-4df5-f53f-19b8309146e2"
      },
      "source": [
        "f, ax = plt.subplots(figsize = (8,6))\n",
        "sns.distplot(train['floor'])\n",
        "print(\"Skewness: %f, Kurtosis: %f\",(train['floor'].skew(),train['floor'].kurt()))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Skewness: %f, Kurtosis: %f (1.3247097550477565, 3.7966033670123065)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c0S-5caCYw6o"
      },
      "source": [
        "# exclusive_use_area : 전용 면적"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZN7IECisYpg4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8150bfff-2772-46a0-ce22-11625461e74b"
      },
      "source": [
        "train['exclusive_use_area'].describe()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "count    1.216553e+06\n",
              "mean     7.816549e+01\n",
              "std      2.915113e+01\n",
              "min      9.260000e+00\n",
              "25%      5.976000e+01\n",
              "50%      8.241000e+01\n",
              "75%      8.497000e+01\n",
              "max      4.243200e+02\n",
              "Name: exclusive_use_area, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "frxdZz234Hhr"
      },
      "source": [
        "# exclusive_use_area : 전용 면적(시각화)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jQRVDKJSY7_Y",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 410
        },
        "outputId": "dc601071-621f-441d-d845-1f44a468d253"
      },
      "source": [
        "f,ax = plt.subplots(figsize = (8,6))\n",
        "sns.scatterplot(train['exclusive_use_area'],train['transaction_real_price'])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbdfcb6630>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lyUGA7PSc7VU",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 203
        },
        "outputId": "c66e66ab-1501-48d1-aadb-d9e36e392ec2"
      },
      "source": [
        "train[train['exclusive_use_area']>400] # 위 시각화에서 가장 많이 벗어나 보이는 이상치를 탐색"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>563870</th>\n",
              "      <td>12633</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>도곡동</td>\n",
              "      <td>193-1</td>\n",
              "      <td>힐데스하임빌라</td>\n",
              "      <td>도곡동 193-1 힐데스하임빌라</td>\n",
              "      <td>424.32</td>\n",
              "      <td>1998</td>\n",
              "      <td>201604</td>\n",
              "      <td>11~20</td>\n",
              "      <td>10</td>\n",
              "      <td>13.118357</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        apartment_id   city  ... floor transaction_real_price\n",
              "563870         12633  서울특별시  ...    10              13.118357\n",
              "\n",
              "[1 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LljsUyC9evU8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "d4d054e6-bc93-4717-bcf1-310bb8b22137"
      },
      "source": [
        "train[(train['exclusive_use_area']<150)&(train['transaction_real_price']<6)] # 두 이상치 모두 아무 이상이 없어 보임"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>312119</th>\n",
              "      <td>10853</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>신월동</td>\n",
              "      <td>128</td>\n",
              "      <td>태헌(신월프라자)</td>\n",
              "      <td>신월동 128 태헌(신월프라자)</td>\n",
              "      <td>56.97</td>\n",
              "      <td>1999</td>\n",
              "      <td>201005</td>\n",
              "      <td>21~31</td>\n",
              "      <td>7</td>\n",
              "      <td>5.860786</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>722888</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>138.25</td>\n",
              "      <td>1974</td>\n",
              "      <td>201606</td>\n",
              "      <td>21~30</td>\n",
              "      <td>-1</td>\n",
              "      <td>4.615121</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        apartment_id   city  ... floor transaction_real_price\n",
              "312119         10853  서울특별시  ...     7               5.860786\n",
              "722888          6225  부산광역시  ...    -1               4.615121\n",
              "\n",
              "[2 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "krODkV4YflDM",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "outputId": "0fac76d5-7385-4226-8151-ae9233cfbd5f"
      },
      "source": [
        "f, ax = plt.subplots(figsize = (8,6))\n",
        "sns.distplot(train['exclusive_use_area'])\n",
        "print(\"Skewness: %f, Kurtosis: %f\",(train['exclusive_use_area'].skew(), train['exclusive_use_area'].kurt()))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Skewness: %f, Kurtosis: %f (1.2275090586886344, 3.1005174479571935)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfcAAAF0CAYAAAA+UXBRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZxcdZ3v/9enqro76ew7kIUEEpaAbAYQxdGRUWFcoldUcBmcyx3GGZmZq9ffDM7jDqM8vHOHmXtlFvEqCiOiCIpbrqKowFVcCIQ9IQRiyEr2rbN113I+vz/Oqerq6uol6XP6VKrez8cjj3SdOl31TSWPvPu7fb7m7oiIiEjzyKTdABEREYmXwl1ERKTJKNxFRESajMJdRESkySjcRUREmozCXUREpMnk0m5AXKZPn+7z589PuxkiIiKj5oknntjl7jNqrzdNuM+fP58VK1ak3QwREZFRY2Yb6l3XsLyIiEiTUbiLiIg0GYW7iIhIk0k03M3scjNbY2ZrzeyGOs93mNm90fPLzWx+dP2DZvZ01a/AzM5Lsq0iIiLNIrFwN7MscCtwBbAYuNrMFtfcdi2w190XArcANwO4+zfc/Tx3Pw/4MPCyuz+dVFtFRESaSZI994uAte6+zt3zwD3A0pp7lgJ3Rl/fB1xmZlZzz9XR94qIiMgwJBnus4FNVY83R9fq3uPuRWA/MK3mnvcD30yojSIiIk2noRfUmdnFwGF3XznA89eZ2QozW7Fz585Rbp2IiEhjSjLctwBzqx7Pia7VvcfMcsAkYHfV81cxSK/d3W9z9yXuvmTGjH4FekRERFpSkuH+OLDIzBaYWTthUC+ruWcZcE309ZXAQ+7uAGaWAd6H5ttFRESOSmLlZ929aGbXAw8AWeAOd19lZjcBK9x9GXA7cJeZrQX2EP4AUPZ7wCZ3X5dUG0VERJqRRR3l496SJUtcteVFRKSVmNkT7r6k9npDL6iT48u6nQdZsX5P2s0QEWl5CneJzecfWstff+fZtJshItLyFO4SmyOFEvlikHYzRERansJdYlMoOUHQHGs4RESOZwp3iU2hFKBsFxFJn8JdYlMoBZSaZPeFiMjxTOEusSmUAg3Li4g0AIW7xKZQcgL13EVEUqdwl9gUSgEl9dxFRFKncJfYaEGdiEhjULhLbIoalhcRaQgKd4lNXsPyIiINQeEusQmH5RXuIiJpU7hLbMJh+bRbISIiCneJjYblRUQag8JdYlMohYfGqJCNiEi6FO4Sm2IpDHXNu4uIpEvhLrEIAqcY9dhVX15EJF0Kd4lFIeg9xz3Qke4iIqlSuEssCqXe3rqG5UVE0qVwl1gUS73ddQ3Li4ikS+EusciXqoflFe4iImlSuEss+g7Lp9gQERFRuEs8+gzLK91FRFKlcJdYFKqH5TXnLiKSKoW7xCJf1Gp5EZFGoXCXWBQDDcuLiDQKhbvEos+wvIrYiIikSuEusdCwvIhI41C4SywKKmIjItIwFO4Si2KgIjYiIo1C4S6x6Dssn2JDRERE4S7xKKiIjYhIw1C4Syz6DMtrzl1EJFUKd4lFQavlRUQahsJdYpHXsLyISMNINNzN7HIzW2Nma83shjrPd5jZvdHzy81sftVz55jZb81slZk9Z2ZjkmyrjEyxT235FBsiIiLJhbuZZYFbgSuAxcDVZra45rZrgb3uvhC4Bbg5+t4c8HXgo+5+FvBGoJBUW2Xk+h75qnQXEUlTkj33i4C17r7O3fPAPcDSmnuWAndGX98HXGZmBrwFeNbdnwFw993uXkqwrTJCGpYXEWkcSYb7bGBT1ePN0bW697h7EdgPTANOA9zMHjCzJ83srxNsp8SgqJ67iEjDyKXdgAHkgEuBC4HDwINm9oS7P1h9k5ldB1wHMG/evFFvpPTSwTEiIo0jyZ77FmBu1eM50bW690Tz7JOA3YS9/F+6+y53PwzcD1xQ+wbufpu7L3H3JTNmzEjgjyDDpdryIiKNI8lwfxxYZGYLzKwduApYVnPPMuCa6OsrgYfc3YEHgFeZWWcU+m8Ank+wrTJC+ZKK2IiINIrEhuXdvWhm1xMGdRa4w91XmdlNwAp3XwbcDtxlZmuBPYQ/AODue83sc4Q/IDhwv7v/KKm2ysj1mXPXgjoRkVQlOufu7vcTDqlXX7ux6utu4L0DfO/XCbfDyXFAteVFRBqHKtRJLPIqYiMi0jAU7hILbYUTEWkcCneJRaEUYBZ+rWF5EZF0KdwlFoVSQEcu/OeknruISLoU7hKLQsnpyGUBhbuISNoU7hKLQilgTFv4z6mkCnUiIqlSuEsswmF59dxFRBqBwl1iUSh5peeuIjYiIulSuEssqnvuqi0vIpIuhbvEou9q+ZQbIyLS4hTuEotwWD6ac1e6i4ikSuEusajuuauIjYhIuhTuEotCKaCjTUVsREQagcJdYlEoOWO0FU5EpCEo3CUW1T13FbEREUmXwl1ioSI2IiKNQ+EusSiUvHfOXQvqRERSpXCXEQsCpxS4itiIiDQIhbuMWCEIJ9lVxEZEpDEo3GXECqUwzduyRsY0LC8ikjaFu4xYoRj23NuyGbIZ04I6EZGUKdxlxMrD8m3ZDGamOXcRkZQp3GXEysPy7dkMWTMNy4uIpEzhLiNWHpbPZS0alk+5QSIiLU7hLiNW7DMsr4NjRETSpnCXEcsXy6vltaBORKQRKNxlxAqlcs/dwjl3hbuISKoU7jJixdrV8jo4RkQkVQp3GbG+w/IqYiMikjaFu4yYhuVFRBqLwl1GrN+wvMJdRCRVCncZsX6r5TUsLyKSKoW7jFifYXkVsRERSZ3CXUasN9yjIjYalhcRSZXCXUasWD7yNafa8iIijUDhLiOW7zcsr3AXEUmTwl1GrDIsn1ERGxGRRpBouJvZ5Wa2xszWmtkNdZ7vMLN7o+eXm9n86Pp8MztiZk9Hv76YZDtlZPoMy2dQz11EJGW5pF7YzLLArcCbgc3A42a2zN2fr7rtWmCvuy80s6uAm4H3R8/9zt3PS6p9Ep/qYfmMitiIiKQuyZ77RcBad1/n7nngHmBpzT1LgTujr+8DLjMzS7BNkoDqYfmMmY58FRFJWZLhPhvYVPV4c3St7j3uXgT2A9Oi5xaY2VNm9gsze329NzCz68xshZmt2LlzZ7ytl2ErlpxsxshktKBORKQRNOqCuq3APHc/H/gEcLeZTay9yd1vc/cl7r5kxowZo95ICRVKAW3ZcMAlYxBoQZ2ISKqSDPctwNyqx3Oia3XvMbMcMAnY7e497r4bwN2fAH4HnJZgW2UE8qWAtkz4Tymj2vIiIqlLMtwfBxaZ2QIzaweuApbV3LMMuCb6+krgIXd3M5sRLcjDzE4BFgHrEmyrjECx5LTlwn9Kqi0vIpK+xFbLu3vRzK4HHgCywB3uvsrMbgJWuPsy4HbgLjNbC+wh/AEA4PeAm8ysAATAR919T1JtlZHpOyyvOXcRkbQlFu4A7n4/cH/NtRurvu4G3lvn+74DfCfJtkl88qWAXHlYPmOUlO0iIqlq1AV1chwplJz28rC8oWF5EZGUKdxlxIqlgFxGw/IiIo1C4S4jVig5uWzVsLx67iIiqVK4y4gF7kTZHh75qp67iEiqFO4yYqXAyUZVgzMZUMddRCRdCvcWsm1/dyKL3UpBWH4Wojl3pbuISKoU7i1i3+E8v/dPD/PT57fH/trV4a7a8iIi6VO4t4j9RwrkSwG7DvbE/toldzLW23NX+VkRkXQp3FtEIaosUz6eNU5B4OSqK9Tp4BgRkVQp3FtEMUrcJMK9GPT23LMZNCwvIpIyhXuLKFZ67vEHb7gVrmpYXgvqRERSpXBvEeUee74Yf8+971Y4LagTEUmbwr1FlHvTSQzL91ktb6Z97iIiKVO4t4gkF9T13eeOhuVFRFKmcG8RvQvqEihi404mUzUsr3AXEUmVwr1FlBfU5ZPaCtdnWF7hLiKSJoV7iygPxxcSWFBXrFlQpyI2IiLpUri3iGI0VF5MYMg8CKqG5VXERkQkdQr3FlHZCpfEgjrv7bmriI2ISPoU7i2iUsQmkX3ukM2qtryISKNQuLeIJMvPloKgd87dDHdwBbyISGoU7i2ikGD52dojXwEVshERSZHCvUUUE5xzD5yqI1/DaypkIyKSHoV7iygmWH62GAS9R75Weu4KdxGRtCjcW0Sy57n39tzLc+8KdxGR9CjcW0SpvKCumEz52Wz0L6kc8hqWFxFJj8K9RSTVc3f3aEFd+E+pMiyvQjYiIqlRuLeIyla4mFO33EHP1iyo07C8iEh6FO4toreITbyhWx5+Lw/Ll7fCqZCNiEh6FO4tIqlh+XIPvbq2PKBjX0VEUqRwbxHlYfm497mXt9jlasNd2S4ikhqFe4tIqudeHpbPVB0cAxqWFxFJk8K9RZQr1MVdfjaozLlrWF5EpFEo3FtEefi8FHise9DLPfT+w/IKdxGRtCjcW0T1cHycQ/OVYfmag2NUxEZEJD0K9xZRrBqOTyLcK/vcVVteRCR1iYa7mV1uZmvMbK2Z3VDn+Q4zuzd6frmZza95fp6ZHTSzTybZzlZQrCpeU4xx3r22595bxCa2txARkaOUWLibWRa4FbgCWAxcbWaLa267Ftjr7guBW4Cba57/HPDjpNrYSgoJ9dyDmjn3rGrLi4ikLsme+0XAWndf5+554B5gac09S4E7o6/vAy4zC9PBzN4FvAysSrCNLaM6bOPc616sXS2vYXkRkdQlGe6zgU1VjzdH1+re4+5FYD8wzczGA38DfCbB9rWUvgvq4gveoGafe+9WuNjeQkREjlKjLqj7NHCLux8c7CYzu87MVpjZip07d45Oy45TxSChBXXet+euIjYiIukbVrib2XfN7G1mdjQ/DGwB5lY9nhNdq3uPmeWAScBu4GLgn8xsPfBfgb81s+tr38Ddb3P3Je6+ZMaMGUfRtNZTrAr0fDHGYfnSAEVsFO4iIqkZblh/AfgA8JKZ/aOZnT6M73kcWGRmC8ysHbgKWFZzzzLgmujrK4GHPPR6d5/v7vOBfwH+wd0/P8y2Sh2FklcCOIkFddl+w/IKdxGRtAwr3N395+7+QeACYD3wczP7jZn9sZm1DfA9ReB64AFgNfAtd19lZjeZ2Tuj224nnGNfC3wC6LddTuJRDAI627JAvHPupaB2WF6r5UVE0pYb7o1mNg34EPBh4CngG8ClhD3vN9b7Hne/H7i/5tqNVV93A+8d7H3d/dPDbaMMrFhyxrZnOdBTTKbnrlPhREQaxrDC3cy+B5wO3AW8w923Rk/da2YrkmqcxKcQBHS2hz33WLfC9ZtzD69rzl1EJD3D7bl/OeqFV5hZh7v3uPuSBNolMSuWnPGdbZWv41JeFd975KuG5UVE0jbcBXWfrXPtt3E2RJJVKHml5x7rsHz0UipiIyLSOAbtuZvZCYSFZsaa2flANOjKRKAz4bZJjIpVw/Jxhnu5Zr22womINI6hhuXfCnyEcI/656quHwD+NqE2SQKKJWdstFo+zn3utQvqemvLx/YWIiJylAYNd3e/E7jTzN7j7t8ZpTZJAvr23OPcChf+3nvka/hYPXcRkfQMNSz/IXf/OjDfzD5R+7y7f67Ot0kDKm+Fg2TOcy+HuorYiIikb6hh+XHR7+OTbogkx90pBs7YtvCvO4lwz0XpXlktr567iEhqhhqW/1L0u05nO46VD41JYp9778Ex4WMVsRERSd9wi9j8E+F2uCPAT4BzgI9HQ/bS4Mr72ivD8sVkjny9e/lGdh3oAeBXL+3kYHexz70fuHhebO8rIiIDG+4+97e4exfwdsLa8guB/y+pRkm8CtF2tY5chmzGEh2Wt0qFutjeQkREjtJww73cw38b8G13359QeyQB1SVi27JWCfs4DLSgTlPuIiLpGW752R+a2QuEw/J/ZmYzgO7kmiVxKp/lnstmaMtmYh2WL9Xscy/33F3pLiKSmuEe+XoD8FpgibsXgEPA0iQbJvEpRL3rtoyF4Z7AsHx5n7tpQZ2ISOqGfeQrcAbhfvfq7/lazO2RBPTtuScz5157KpyjdBcRSctwV8vfBZwKPA2UosuOwv24UK5I15YNe+6xboULaofl1XMXEUnbcHvuS4DFronU41L5cJdcJkN7NhNr+dlymdnyaXDleR79UxERSc9wV8uvBE5IsiGSnPJq+VzUcy/EeHBM71a4vj13ZbuISHqG23OfDjxvZo8BPeWL7v7ORFolsSpXqGvLGm25eOfci1VFbKB6n7vSXUQkLcMN908n2QhJVmVBXSYT+5x70G9BnXruIiJpG1a4u/svzOxkYJG7/9zMOoFssk2TuBRqh+WTqC1v2ucuItIohjXnbmZ/AtwHfCm6NBv4flKNkniVF9S1ZcMFdcU4F9QFjlnvgrrKsHxs7yAiIkdruAvqPga8DugCcPeXgJlJNUriVVlQlzFyMe9zLwZe6bVD9alw6rmLiKRluOHe4+758oOokI3+9z5OlMO8LVuec4+3/Gy51w5Q/krZLiKSnuGG+y/M7G+BsWb2ZuDbwP9NrlkSp/KK9lzWon3u8S6oq+65mxmG5txFRNI03HC/AdgJPAf8KXA/8N+TapTEq1BZLW+xl58tBl7Z415mpgp1IiJpGu5q+cDMvg983913JtwmiVnvnHsm9iI2QdB3WB7C3rs67iIi6Rm0526hT5vZLmANsMbMdprZjaPTPIlDpfxs1mjLxT/nnq0J94xpWF5EJE1DDct/nHCV/IXuPtXdpwIXA68zs48n3jqJRe/BMZnY59xLAf3C3cy02lJEJEVDhfuHgavd/eXyBXdfB3wI+KMkGybxKSY4514Kgj4L6iBcMa+tcCIi6Rkq3NvcfVftxWjevS2ZJkncelfLZ8iNQs89Y6YFdSIiKRoq3PPH+Jw0kD4Hx0RHvsY1Jx64k6n5V2SacxcRSdVQq+XPNbOuOtcNGJNAeyQB1QfHtGfDXnYxcNqyNti3De+1AydXk+5aLS8ikq5Bw93ddThME+hdUBf23MNrQeXrkQgCp2ZUnoxpzl1EJE0j/99dGl4xCMhmDLOqcC/GE76loN5WOK2WFxFJk8K9BRRLvVXk2nLhX3lcZ7qH+9xrhuXRnLuISJoSDXczu9zM1pjZWjO7oc7zHWZ2b/T8cjObH12/yMyejn49Y2bvTrKdza5Q8kqPvTznHteK+bDn3veays+KiKQrsXA3syxwK3AFsBi42swW19x2LbDX3RcCtwA3R9dXAkvc/TzgcuBL0Ul0cgyKQUAuCvXqOfc4lGoOjoFoWF49dxGR1CTZc78IWOvu66LjYu8BltbcsxS4M/r6PuAyMzN3P+zuxej6GHS87IgUSr0r2uMO98Dr1ZZXz11EJE1JhvtsYFPV483Rtbr3RGG+H5gGYGYXm9kqwpPoPloV9nKUiqWgd8496sHnY1xQ1/9UOPXcRUTS1LAL6tx9ubufBVwIfMrM+u2rN7PrzGyFma3YuVOH1Q2kGHi/YfnyYTJxvHam37C8hlpERNKUZLhvAeZWPZ4TXat7TzSnPgnYXX2Du68GDgJn176Bu9/m7kvcfcmMGTNibHpzqd7THvuwfJ2tcIbKz4qIpCnJcH8cWGRmC8ysHbgKWFZzzzLgmujrK4GH3N2j78kBmNnJwBnA+gTb2tT6bIWLwj22YXkd+Soi0nASW4Hu7kUzux54AMgCd7j7KjO7CVjh7suA24G7zGwtsIfwBwCAS4EbzKwABMCf1zvARoYnHJaPtsLlktgKV2/OPZaXFxGRY5Do9jJ3vx+4v+bajVVfdwPvrfN9dwF3Jdm2VlIMgspCutHYCmcqPysikqqGXVAn8ak3LB9nuNduhcuo5y4ikiqFewsolILKsHxlzr0U35Gv/bfCQaD18iIiqVG4t4Dq413bKwfHxLgVrs5qefXcRUTSo3BvAWERm/CvOhdzbfmgbvlZzbmLiKRJ4d4CwoNjEppzr7sVTj13EZE0KdxbQDHo7bm3xzznXirV2wqnfe4iImlSuLeAYqm3/Gz592KcPfe6W+FieXkRETkGCvcWUAiSKz9bCqi/oE6r5UVEUqNwbwF997mXF9TFtxUuW/OvKCw/G8vLi4jIMVC4t4BC1bC8mZHLWGw99+qV+GVmptXyIiIpUri3gOoFdRAOzRdjmhQPnH5Hvpp67iIiqUq0trw0hlLUc797+UYAHOe5Lfsrj8s+cPG8o3/toN6wvI58FRFJk3ruLaB6QR1A1owgpvQN97nXDstrK5yISJoU7i2gekEdQDZjlOIK9wF67op2EZH0KNybnLv3Oc8dEgj32jl3VH5WRCRNCvcmV14411bbc48hfMtD+zryVUSksSjcm1wx2s9e3XPPWDw99/IPDvWOfNWcu4hIehTuTa4QhPvZy8VrIAzjOMK9PPTer+ce47C/iIgcPYV7k6v03KsCOK7wLb9G7Zx71uIZ9hcRkWOjcG9y5QNi+i2oiyF8y69ReypcNmME8RTAExGRY6Bwb3KF8oK6qmH5bExz7qVS/XDPqOcuIpIqhXuTq/TcM/FvhRu45x6upNeiOhGRdCjcm1yhslq+71a4OCrUVbbC1c65Z8IiNlpTJyKSDoV7kyuf/tZeM+cex8ExA22FKy+wUyEbEZF0KNybXL4YhXuub7jHEbylgYrYRI+1HU5EJB0K9yaXL9UJ95gW1JV/QOi3FS4K97gOpxERkaOjcG9ylZ57ArXlK/vc66yWB7RiXkQkJQr3JldvWD72IjY14Z7TsLyISKoU7k2uZ4A59ySL2GjOXUQkXQr3Jleec+9IYM69NNBWOA3Li4ikSuHe5Hrn3LOVa3HNuZdLzNZuhctUFtSN+C1EROQYKNyb3MBb4UZ+LGsxSu9+FerUcxcRSZXCvcnliyWgf7jDyMN3oCNfywvztRVORCQdCvcmN9A+dxj5sHn00nX2uYfvFUcVPBEROXoK9yY30D53GPlq9oGG5ctn1Kj8rIhIOhTuTa4c7m01B8dAbzgfq/K3Dzjnrp67iEgqFO5NrqcU0J7LYFVD570Hu4zstXv3ufe9rvKzIiLpSjTczexyM1tjZmvN7IY6z3eY2b3R88vNbH50/c1m9oSZPRf9/qYk29nM8sWAjpr0jWtYPqhUqOv7+io/KyKSrsTC3cyywK3AFcBi4GozW1xz27XAXndfCNwC3Bxd3wW8w91fBVwD3JVUO5tdvhj0WUwHcc65D35wjIblRUTSkWTP/SJgrbuvc/c8cA+wtOaepcCd0df3AZeZmbn7U+7+SnR9FTDWzDoSbGvTyhcD2mp67nGVh+098rXvdYW7iEi6kgz32cCmqsebo2t173H3IrAfmFZzz3uAJ929p/YNzOw6M1thZit27twZW8ObSb7Uv+eei3mf+0AL6rRaXkQkHQ29oM7MziIcqv/Tes+7+23uvsTdl8yYMWN0G3ecKNQJ90xMq9nLw/IDlZ8tqfysiEgqkgz3LcDcqsdzomt17zGzHDAJ2B09ngN8D/gjd/9dgu1savli0GePO8S/oK7fwTExjQyIiMixSTLcHwcWmdkCM2sHrgKW1dyzjHDBHMCVwEPu7mY2GfgRcIO7/zrBNja9ngQX1A10nntvBTyFu4hIGhIL92gO/XrgAWA18C13X2VmN5nZO6Pbbgemmdla4BNAebvc9cBC4EYzezr6NTOptjazJFfLl3vmA/bcFe4iIqnIJfni7n4/cH/NtRurvu4G3lvn+z4LfDbJtrWKfClgfEffv+ZKkZkRDpuXwzuXrZ1zj57XsLyISCoaekGdjFzdOXcrl5+NaVi+tueu8rMiIqlSuDe5wYblRzonPtCRr2ZGxjTnLiKSFoV7k6u3zz3uBXW1W+EgnIfXsLyISDoU7k0uya1wvRXq+od7NmPquYuIpETh3uTqDcuXs3ikPeuB5twhDPeRzumLiMixUbg3uXrhnouWs8e1Fa52nzuEga/ysyIi6VC4N7meJOfcSwOHeyZjKj8rIpIShXsTc/e657nHtQ+90nMfYFhePXcRkXQo3JtYIepZ9+u5x7QPPRhkQV3GTPvcRURSonBvYvloXLw23Mv70OOYc683JA+QzaiIjYhIWhTuTSxfjMI92/+vOZsZec+6GAwW7uq5i4ikReHexCrhnsv2ey6bGXmRmSDwuvPtoNXyIiJpUrg3sd5wr9Nzj2FOvBTUXykP5dXyCncRkTQo3JtYvlQCoC2bTAW5YLA5d5WfFRFJjcK9ifVEPfeOej33WObcg0Hn3FV+VkQkHQr3JjbQVjiIZ869FIRb3urRgjoRkfQo3JtY72r5/gvq4tiHHgROnYX4va+vYXkRkVQo3JvYYAvqcjFthSvXqa+VVflZEZHUKNybWHlBXb1wj2M1e+DOANmu8rMiIilSuDexpIvYlAbZ567ysyIi6VG4N7Geofa5x3BwjMrPiog0HoV7E8snvBWuVFL5WRGRRqRwb2IDHRwD8exDL7kPvBVO5WdFRFKjcG9iQ825F2PZCqfysyIijUbh3sQGrS0f21Y49dxFRBqNwr2JDXVwzEjDN18MaBugik0mYwSOAl5EJAUK9yaWLwWYUbd3HUfPff+RApPGttV9rjxcr/ryIiKjT+HexPLFgPZsBquz6C2OOfFBwz16T5WgFREZfQr3JtZTDOoOyUM8B8d0HSkwccie+4jeQkREjoHCvYnlS0HdPe4AuRFWkCsFzoGe4oA990xGPXcRkbQo3JtYeVi+npEOy3cdKQAMPSyvOXcRkVGncG9i+SGG5QMHP8ae9f6hwj16W4W7iMjoU7g3saHCHY592HyocC9XrtNqeRGR0adwb2L50iDhPsJh80q4dw6+oE5z7iIio0/h3sQKpYGLzIx0NfvQw/KacxcRSYvCvYn1DLKgrhy+xWNM967u4S2oU4U6EZHRl2i4m9nlZrbGzNaa2Q11nu8ws3uj55eb2fzo+jQze9jMDprZ55NsYzMbzpz7sXasyz33iWOG2AqnnruIyKhLLNzNLAvcClwBLAauNrPFNbddC+x194XALcDN0fVu4O+ATybVvmYy0KK1fHHgfe5xzLm3ZzOMaRtiwZ7CXURk1CXZc78IWOvu69w9D9wDLK25ZylwZ/T1fcBlZmbufsjdf0UY8jKIp5Q4Bs4AABncSURBVDft49zP/JTnNu/v99ygC+pGOiwfVaerV9oWVH5WRCRNSYb7bGBT1ePN0bW697h7EdgPTEuwTU3niQ17OdBT5MZlK/v14IcqYgMjW1A3aWxuwOd1cIyISHqO6wV1Znadma0wsxU7d+5MuzmpeHnXQQCe2riP7z61pc9zg82550Y4bD7YoTFQPed+TC8vIiIjkGS4bwHmVj2eE12re4+Z5YBJwO7hvoG73+buS9x9yYwZM0bY3OPTy7sOce6cSZw/bzL/+OMXONhTrDw32LB8ZoTD5kOFu4blRUTSk2S4Pw4sMrMFZtYOXAUsq7lnGXBN9PWVwEN+rPVQW9TLOw9xyozxfPwPTmPXwR6e2LC38lw4LJ+t+30jXfA2dM89/F3D8iIio2/gSdMRcveimV0PPABkgTvcfZWZ3QSscPdlwO3AXWa2FthD+AMAAGa2HpgItJvZu4C3uPvzSbX3eNRdKPHK/m4WTB/HolnjAdi893Dl+WGVnz3WcD88zJ67wl1EZNQlFu4A7n4/cH/NtRurvu4G3jvA985Psm3NYP3uQwAsmD6OmRPG0JY1Nu05AoQHwgxntfyxhG8wxHGvfV5fAzEiIqMu0XCXeN29fGOfxyu3hNvf1mw7wIHuIrMnj2VT1HPPRyvZBtznPoJwP9BdxB0mDifc1XMXERl1x/Vq+Va362APANPGtQMwd2onm/eGPfd8MQz3AcvPjmDB21B15atfX+VnRURGn8L9OLb7YJ4JY3J0tIWL5uZMGcvmPVHPvRzuCfTchxPuKj8rIpIehftxbNfBHqaP76g8njOlk92H8hzOFyvD8kPWlk8o3DUsLyKSHoX7cWzXwZ7KkDyEPXeAzXuPDD0sH0fPfYCz3GHk++hFROTYKdyPU0fyJQ7lS3167nOndgLhdrihhuWjbE9szj1jYCS7z93duevRDeyO1h6IiEhIq+Ub3AOrtvF331/JabMmML4jx6WLppMxqyym6xPuU8Jw37TnCJM7wx79QKvlc1GVmaTm3M2MTMYSLT/74vaD/N33V9J1pMDHfn9hcm8kInKcUbg3uIdf2MGB7iK7D+X51dpdFIOA3z99ZqUS3ayJveE+fXw7Y9oybN57mK4jBczgvHmT675u5VS4Y0jf/UcKtGWNsW31q99V3sMs0dXyz2zaB8CL2w8k9h4iIscjhXuDW73tAOfNnczdf3Ix7/7Cb3hw9Q52H8zz1KZ9vH7RdKZV9dzNjDlTOtm05wib9+3m/LmTmTlhTN3XLYfznsOFYbelvM/+iQ17ac9l+eZjmwa9P5NJdkHdM5vDcF+zTeEuIlJNc+4NLAicl7Yf4IwTJ2BmvOu82cyY0MFTm/Zx1kkTeetZJ/T7njlTxvLkxr2s3NLFW+o8X2ZmzJo4hu1d3UfdriOF0pC9dgh77kmG+7PRGfbrdh6ioOPnREQqFO4NbNPewxzOlzjjhAlAuDjuw685mTedMZP3LZlbWZFebe6UTnYcCOfj37J41qCvP2tiB9u7ujnas3oO9RQZ2zb0P51sxhJbLd9dKLF6axezJ48lXwrYEJXiFRERhXtDW701HG4+/YSJlWvTxnfwB2fOom2ALW5zp4bb4RbOHM8pM8YP+vqzJo6hpxjQ1V0c9L58MajMzT+6bjcv7zrEgumDvzaE4Z7UavnVW7soBs57Xj0HgBc0NC8iUqE59wa2ZtsBzOC0WUMHKYRz4i/vCivUzZ48tl8t+lqzJobz8du7uuuufF+/6xC3/+plvvvkZnLZDHOndrJqy37OOGECbx5iVADCve5J9dzLi+nec8Fsbn14LS9uOwDnJPJWIiLHHYV7A3thWxcnT+2ks334f01zp4xlxvgOzp9bf5V8tVkTwsV427u6OW3WhD7P7TmU511f+DWH8yXe/qoTceD+57Yyb2onV104r7LafjDZTHJz7s9u3s/MCR3Mm9rJ/GmdrNGKeRGRCoV7A1uz7QCnnzBh6BurTO5s5+NvPm1Y93Z25JgwJld3Ud0/P7CGA91FfvSXl3JGNC1wwbwpmFF3rr+eJMP9mc37OGfOZMyM00+YUJnCEBERhXtDqR5GzxcDXt51iPnTxw05vD4SsyaMYXtXT5/337L3CPc8tpHXnjqNJzfs48kN4RD4cHrr1TIJ7XPv6i7wu52HeNd5swE4bdYEfrxyG0fyJca2D72KX0Sk2WlBXYPacaAbB06YWH+felxmTexgx4HuSggXSgE/eGYLnR05Ljtz6Hn1wSTVc38xWjx31uxwROGMEybgDmt3HIz9vUREjkfquTeo8lB58uE+hkLJ2Xsoz+TOdu55bCNb9h7h6ovmMWYYe9kHk425/Gx5ZOGxl/cAsPqVA2zb38OuaOvfC9u6eNWcSfG9oYjIcUo99wa1cc9hOnIZpo5vH/rmESivmH9y416++dhGVm87wNvPPYmzZ488JJMqP7v9QDft2UzlVLqpUdndFev3xv5eIiLHI/XcG5C7s2bbARbOHD/sxWvHauaEDgx4eM1OshnjrWedwCWnTIvltTMZKBXiD/cdXd3MnNhR+WwyZpw6YzyPvLQTd8cS/sxERBqdwr0Bbevqpqu7yOmzjm6l/LHoaMvy4decjJmxYPq4AY+IPRZJlZ/d0dXDoprPZuHM8fzg6VdYt+sQpw5RvEdEpNkp3BtQ+SCU045yG9yxOuPEiUPfdAwyCZSfPZwvcqCnyMwJHX2uL5oZflaPvLhT4S4iLU9z7g1ozbYDnDR5DBPHDHxe+vEgifKzO6Jte9VH3QJMHdfOydM6eeSlXbG+n4jI8Ujh3mAO54ts3HOY02cl05seTdmjLD+76pX9/M/7V3MkXxrwnu0Hwl0EM+vsInj9ouk8um43+aJOiBOR1qZwbzCrt3bhcNSV6RrRcHvu7s43H9vIu7/wG770y3Xc+/jARXt2dPXQnsswuU4t/EsXzuBQvsRTG49+1fwdv3qZrz+64ai/T0SkESncG8SeQ3m+99RmvvvkFqaNa2fOlLFpN2nEMhmjOIxw/+GzW/nUd5/j4gVTedXsSdzx6/UDLsTbfqA7XOFfZ0X8JadOoz2X4R9+/AK7D/YMu51f++16bvrh89z0w+fZdRTfJyLSqBTuDaBQCrj2zsd5YsNeXnvqNP78jQsT3wI3GtoyRqEUDHle/DeWb2D+tE6++scX8WdvPJWNew7z89Xb6967o6uHmRPqF/aZNLaNWz9wAS9s7eLKL/6W+57YzEMvbOffH3yJ933xt/z9D1ay/0ihz/f87PntfHrZKi6aP5V8MVDvXUSaglbLN4B/f/Alntq4j6sunMs5c4Y+ze14MXNCWP1u96E808d31L1n4+7DPLpuD598y2nc+/gmSoEzpbON/3n/anYfzPe590B3gYM9xX6L6crKFew+8tr53PXoBj757WcAMMIdAXc9uoEfPbeNv3/HYt5+zon8eu1uPnb3k7xqzmTu/M8Xcf3dT3LXbzfw0TecOuLqfCIiaVK4p+zx9Xv4/MNrec8Fc5oq2AHmTA2nFjbtOTxguN/35GbM4D9dMIf/FxXSee2p0/nRc1v53lNbePPiWYzvyLH/SIE7f7OejMEpQ2x1O3naOP7m8jPoOlLgUE+RKePa+dM3nMrKLfv51Hef4y+++RR3L9/IU5v2csr0cXz1Ixcytj3Lta9fwAe+vJwfPL2F9184L/bPQ0RktCjcU9RdKPHJbz/DnCmdfGbpWSx7+pW0mxSrWRPH0J7NsGnvYc6fN6Xf80HgfOeJzVy6cDonTe5dY3DxgqnsPZzn0XW7eXrTXiaNbeNwvkQpcK557XxmTx56PUJbNsO08R1Mi36oKPfq33/hXE6e1slPn9/OpLFt/KcL5vDjldv4wMXzuOSUaSw+cSJfeeRl3rdkrirdichxS+Geolt+/iIbdh/m7j+5mPEdzfdXkTFj9pSxbNpzpO7zD76wgy37jvA3V5zR53oum+Ht55zERQumsvzlPRzsLhK48/unz+zzQ8Cxtum1p07n/LlTyGasUpGvHP5nnTSRbz+xmc/83+c5raoK3gcuVk9eRI4fzZcoDa76zPQv/3IdS06ewvpdh1m/K7kz29M0d0onv1q7k0LN8XCrt3bxiW89zakzxvGWxfWPlp05YQzvOOekRNo10Lnvr5oziZ+s2sav1+7qE+4iIscTrZYfZfliwIOrt3PbI79jXEeOK84+Me0mJWre1LEEDq/s6+29b9h9iD+64zHGtef42rUXN9TitVwmwyWnTOOlHQfZFh27KyJyvFHPfZQEgfO9p7bwuZ+toau7yNknTeTys08csAfZLOZM7QRg094j3L18I6/sO8JXfxPuY7/u907hF2t2ptzC/i5aMJWH1+zgJyu38odnn1iphlcKnA27D/Hi9gOs2XaQfUfynD5rAq8+eUq/g2xERNKkcB8Fm/Yc5uP3Ps2KDXuZM2UsV104j/nTx6XdrFExcUwbk8a2sX7XIca2Zfnhs68wpi3Lf7l0Qd0Sso2gsz3HG06byYOrt/Mv21+isz3LP9y/mu5CqVKUx4Bc1iiUwsfvPPck/uaKM4a12E9EJGkK9xF4YVsX9zy2ic72LJcunM6r50+hI9fbE3d3vv/0Fm78wSpw+F/vPZfuQqkpCtQcjblTO1m5ZT/Pb+1i9uSxfPDieUzubE+7WYN60xkzWXLyFFa+sr9yWE17LsOsiWOYNbGDmRPGkMsaew/leXLjXh5YtY2frNzGW88+gXeffxIzJ4xhTFsG97BS35wpY+nIZTnYU2TF+j3cvXwjD72wg0tOncY1l8znTWfMJJNprX8XIpIcG6p62Ihe3Oxy4F+BLPAVd//Hmuc7gK8BrwZ2A+939/XRc58CrgVKwF+6+wODvdeSJUt8xYoVsf8Z6unqLvCJe5/m56t30JHLUAqcYuBMGtvG2885kXPnTiZjxjcf28gTG/Zy/rzJ/NtV5zN3amdlQV0reWnHAZav28OF86dy2qzxTbnFbO/hPL96aRdPbdpLd6H/wTUGjB+T40B3EYDO9ixnnjiRl7YfoKu7yLlzJ3Pj2xfz6pP7bxkUERmImT3h7kv6XU8q3M0sC7wIvBnYDDwOXO3uz1fd8+fAOe7+UTO7Cni3u7/fzBYD3wQuAk4Cfg6c5u4DHhc2WuH+lUfW8R+/Xs/W/Ue47MxZXLxgKlkz1u06xDOb97F6a1dlqHZ8R463LJ7FBSdPabneeqsqlAI27TlMTzEgXwowwrn6XQfz7D+SZ+q4dk6aNJaFM8eTy4Y/GD6zaR+PrN3J9q4e5k/r5Px5U5g2rp32XKby64SJY5g/fRxTO9sr6zSKgTMml2HS2DZyWa2NFWlFA4V7ksPyFwFr3X1d1IB7gKXA81X3LAU+HX19H/B5C7t1S4F73L0HeNnM1kav99sE2zuonmKJ+5/bypd+sY49h/N88OKTOfPE3mNZzzxxImeeOJF8MeBQT5FCEDB5bHtlH7W0hrZsZsgKetWyGeOCk6dw9uxJPL5+Dy/vOsSDq7fTXQwolXzYR+bOmtjBgunjmDFhDGNyGTraMnTksrRlM7RljWzGaMtmyGaMXPQrm83Qlgmfy2WNXCb8QWLCmByd7TkKpYB8MfxVCpzO9izjOnKM68gxtj1LT6HE4XyJ7kKJ7kJAJgPt2Qy56D3D986Qi+oJ5DJGWy5DWyZDLmsE7hRKTqEYhO9VCsLH0fsWosdb9x9hzbYDHCmUmNrZztypnZw3dzInT+tsylEgkTgkGe6zgU1VjzcDFw90j7sXzWw/MC26/mjN985Orql9bd57mA99ZTmFkkf/4QQczpfIFwOmj2/nmkvms3Bm/f/Aw55WY88nS+Npz2V43cLpvG7h9D7XA3dKgbPvcIFdB3s4UihVagZkzchH/zb3HS7wyr5u1u44SKHkFEsBxWi6KAic5CbfRkfGwh+ceopBn2vZjNHZnqv88FAd9YMFf+1T1Y/7vkrtcyIj85VrlrBwZvK7a47rBXVmdh1wXfTwoJmtGeFLTgd2DXbDBuCJEb6J1DXkZy+J0uefHn326Rn1z37RX8f+kifXu5hkuG8B5lY9nhNdq3fPZjPLAZMIF9YN53tx99uA2+JqsJmtqDd3IcnTZ58uff7p0Wefnmb+7JOcEH4cWGRmC8ysHbgKWFZzzzLgmujrK4GHPFzhtwy4ysw6zGwBsAh4LMG2ioiINI3Eeu7RHPr1wAOEW+HucPdVZnYTsMLdlwG3A3dFC+b2EP4AQHTftwgX3xWBjw22Ul5ERER6JbrP/XhjZtdFQ/0yyvTZp0uff3r02aenmT97hbuIiEiT0SZsERGRJqNwJyyTa2ZrzGytmd2QdnuakZndYWY7zGxl1bWpZvYzM3sp+n1KdN3M7N+iv49nzeyC9Fp+/DOzuWb2sJk9b2arzOyvouv6/BNmZmPM7DEzeyb67D8TXV9gZsujz/jeaNEx0SLie6Pry81sfprtbwZmljWzp8zsh9HjlvjsWz7cozK5twJXAIuBq6PytxKvrwKX11y7AXjQ3RcBD0aPIfy7WBT9ug74P6PUxmZVBP6buy8GXgN8LPo3rs8/eT3Am9z9XOA84HIzew1wM3CLuy8E9hKeo0H0+97o+i3RfTIyfwWsrnrcEp99y4c7VWVy3T0PlMvkSozc/ZeEOyKqLQXujL6+E3hX1fWveehRYLKZnTg6LW0+7r7V3Z+Mvj5A+B/dbPT5Jy76DA9GD9uiXw68ibDkNvT/7Mt/J/cBl5lq7B4zM5sDvA34SvTYaJHPXuFev0zuqJW6bXGz3H1r9PU2YFb0tf5OEhINNZ4PLEef/6iIhoWfBnYAPwN+B+xz92J0S/Xn26ckN1AuyS3H5l+AvwbKdYun0SKfvcJdGkJUvEhbNxJkZuOB7wD/1d27qp/T558cdy+5+3mElTYvAs5IuUktwczeDuxw95asGK5wH2apW0nE9vJwb/T7jui6/k5iZmZthMH+DXf/bnRZn/8ocvd9wMPAJYRTHeUiYtWfb+WzrynJLUfvdcA7zWw94XTrm4B/pUU+e4X78MrkSjKqyw9fA/yg6vofRau2XwPsrxo+lqMUzRveDqx2989VPaXPP2FmNsPMJkdfjwXeTLjm4WHCktvQ/7OvV5JbjpK7f8rd57j7fML/1x9y9w/SIp+9itgAZvaHhHMz5TK5/yPlJjUdM/sm8EbCU5i2A38PfB/4FjCP8MC997n7niiMPk+4uv4w8MfuviKNdjcDM7sUeAR4jt65x78lnHfX558gMzuHcJFWlrAz9S13v8nMTiHsTU4FngI+5O49ZjYGuItwXcQe4Cp3X5dO65uHmb0R+KS7v71VPnuFu4iISJPRsLyIiEiTUbiLiIg0GYW7iIhIk1G4i4iINBmFu4iISJNRuIuIiDQZhbtIkzCz/2dmS47h+5aY2b8l0SYRSUdu6FtEpJlFBWqapkiNmeWqDgYRaUnquYs0CDP7kJk9ZmZPm9mXzOxiM3vWzMaY2TgzW2VmZ0enjP0vM1sZPf8XdV7rYNXXV5rZV6Ov3xt93zNm9svo2hvN7IdmljGz9eVyqdFzL5nZrKiM6nfM7PHo1+sG+XN82sw+WfV4pZnNj/4MP4ree6WZvT96/tVm9gsze8LMHhjseFkz+5Po/Z+J2tMZXf+qmX3RzJYD/2Rmp5rZT6LXfMTMzojue4eZLTezp8zs52Y2a6D3Ejmeqecu0gDM7Ezg/cDr3L1gZl8ATiesd/1ZYCzwdXdfaWZ/BswHznP3oplNPYq3uhF4q7tvqQ5xAHcPzOwHwLuB/zCzi4EN7r7dzO4GbnH3X5nZPOAB4Myj/GNeDrzi7m+L/syTogNt/h1Y6u47o8D/H8B/HuA1vuvuX46+/7PAtdH3Q3gIyGvdvWRmDwIfdfeXoj/HFwgPDvkV8Bp3dzP7L4THgf63o/xziDQ8hbtIY7gMeDXweFjanbGEp7TdRHi4UTfwl9G9fwB8sTz07O57juJ9fg181cy+BXy3zvP3Ev4A8B+Eh23cW/Wei6O2AUw0s/HufrD/SwzoOeB/m9nNwA/d/REzOxs4G/hZ9NpZYLBDas6OQn0yMJ7wh4yyb0fBPh54LfDtqvZ2RL/PAe6NRgfagZePov0ixw2Fu0hjMOBOd/9Un4thCI0H2oAxwKFhvl71oRFjKhfdPxr1ZN8GPGFmr675vt8CC81sBvAuwlEDCKfwXuPu3cN47yJ9p/zGRO/9opldAPwh8Nmod/09YJW7XzLMP9dXgXe5+zNm9hHCw4jKyp9NBtgXnaFe69+Bz7n7sugwkU8P831FjiuacxdpDA8CV5rZTAAzm2pmJwNfAv4O+AZwc3Tvz4A/tehM6gGG5beb2ZlmliEcZie691R3X+7uNwI76XtuO9ERl98DPkd4RGz5POufAn9R9Tr1grNsPXBBdN8FwILo65OAw+7+deCfo3vWADPM7JLonjYzO2uQ154AbI2G8z9Y7wZ37wJeNrP3Rq9pZnZu9PQkes/vvqbe94s0A/XcRRqAuz9vZv8d+GkUyAXCc6YL7n63mWWB35jZm4CvAKcBz5pZAfgy4RGt1W4AfkgY4CsIe/8A/2xmiwhHCh4EngHeUPO99xJOBXyk6tpfArea2bOE/2/8EvjoAH+c7xCeB7+K8FjZF6Prr4reP4j+fH/m7nkzuxL4NzObFL32vwCrBnjtv4tec2f0+4QB7vsg8H+iz7SN8IjPZwh76t82s73AQ0Q/eIg0Gx35KiIi0mQ0LC8iItJkNCwvIsfEzP4Y+Kuay79294/F8Nq3ArV76f/V3f9jpK8t0go0LC8iItJkNCwvIiLSZBTuIiIiTUbhLiIi0mQU7iIiIk1G4S4iItJk/n+L8/ik0NnJqwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Av_bMnYTgNr2"
      },
      "source": [
        "# transaction_year_month : 거래 년, 월"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J0VV5DSzgUMJ"
      },
      "source": [
        "all_data = [train,test]\n",
        "\n",
        "for dataset in all_data:\n",
        "  dataset['transaction_year_month'] = dataset['transaction_year_month'].astype(str)\n",
        "  dataset['year'] = dataset['transaction_year_month'].str[:4].astype(int) # year - 0~3\n",
        "  dataset['month'] = dataset['transaction_year_month'].str[4:6].astype(int) # month - 4~6\n",
        "  dataset['transaction_year_month'] = dataset['transaction_year_month'].astype(int)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kfv3nNyFh_7L",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a9a7e87a-d4e7-46d3-d765-5ad2bbcd8523"
      },
      "source": [
        "train['year']"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0          2008\n",
              "1          2008\n",
              "2          2008\n",
              "3          2008\n",
              "4          2008\n",
              "           ... \n",
              "1216548    2017\n",
              "1216549    2017\n",
              "1216550    2017\n",
              "1216551    2017\n",
              "1216552    2017\n",
              "Name: year, Length: 1216553, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h8eSqc0CiQtJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 409
        },
        "outputId": "928073d3-523a-4d79-a039-d793f320aa95"
      },
      "source": [
        "f,ax = plt.subplots(figsize = (8,6))\n",
        "sns.boxplot(train['year'],train['transaction_real_price'])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbdb8dd0f0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Wo5vk5epic0z",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 134
        },
        "outputId": "5bd156b9-4659-49ed-e628-a382d4b021f1"
      },
      "source": [
        "train[(train['year']==2016) & (train['transaction_real_price']<6)] # 위 시각화에서 가장 멀리있는 이상치 확인\n",
        "# apt = '서면', floor에 -1이 적혀있다."
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "      <th>year</th>\n",
              "      <th>month</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>722888</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>138.25</td>\n",
              "      <td>1974</td>\n",
              "      <td>201606</td>\n",
              "      <td>21~30</td>\n",
              "      <td>-1</td>\n",
              "      <td>4.615121</td>\n",
              "      <td>2016</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        apartment_id   city dong  ... transaction_real_price  year month\n",
              "722888          6225  부산광역시  범전동  ...               4.615121  2016     6\n",
              "\n",
              "[1 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CZJqoMb0i0yS",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "cfee6d45-28c8-46f2-d4ee-0bd7b32c82dc"
      },
      "source": [
        "train[train['apt']=='서면']"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "      <th>year</th>\n",
              "      <th>month</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>695725</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201105</td>\n",
              "      <td>21~31</td>\n",
              "      <td>3</td>\n",
              "      <td>8.682877</td>\n",
              "      <td>2011</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>696834</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201108</td>\n",
              "      <td>21~31</td>\n",
              "      <td>1</td>\n",
              "      <td>8.922792</td>\n",
              "      <td>2011</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>700884</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>88.93</td>\n",
              "      <td>1974</td>\n",
              "      <td>201206</td>\n",
              "      <td>1~10</td>\n",
              "      <td>1</td>\n",
              "      <td>10.126671</td>\n",
              "      <td>2012</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>702589</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201212</td>\n",
              "      <td>21~31</td>\n",
              "      <td>5</td>\n",
              "      <td>8.699681</td>\n",
              "      <td>2012</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>702876</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201301</td>\n",
              "      <td>11~20</td>\n",
              "      <td>4</td>\n",
              "      <td>8.909370</td>\n",
              "      <td>2013</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>703979</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201304</td>\n",
              "      <td>1~10</td>\n",
              "      <td>1</td>\n",
              "      <td>8.853808</td>\n",
              "      <td>2013</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>711024</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201407</td>\n",
              "      <td>21~31</td>\n",
              "      <td>5</td>\n",
              "      <td>8.853808</td>\n",
              "      <td>2014</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>712441</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201410</td>\n",
              "      <td>21~31</td>\n",
              "      <td>3</td>\n",
              "      <td>9.082621</td>\n",
              "      <td>2014</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>713864</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201501</td>\n",
              "      <td>1~10</td>\n",
              "      <td>4</td>\n",
              "      <td>8.955577</td>\n",
              "      <td>2015</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>721684</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201604</td>\n",
              "      <td>1~10</td>\n",
              "      <td>1</td>\n",
              "      <td>9.059634</td>\n",
              "      <td>2016</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>722888</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>138.25</td>\n",
              "      <td>1974</td>\n",
              "      <td>201606</td>\n",
              "      <td>21~30</td>\n",
              "      <td>-1</td>\n",
              "      <td>4.615121</td>\n",
              "      <td>2016</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>723957</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>55.04</td>\n",
              "      <td>1974</td>\n",
              "      <td>201608</td>\n",
              "      <td>21~31</td>\n",
              "      <td>2</td>\n",
              "      <td>10.051951</td>\n",
              "      <td>2016</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>726597</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201612</td>\n",
              "      <td>11~20</td>\n",
              "      <td>4</td>\n",
              "      <td>8.802522</td>\n",
              "      <td>2016</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1186983</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>34.08</td>\n",
              "      <td>1974</td>\n",
              "      <td>201702</td>\n",
              "      <td>11~20</td>\n",
              "      <td>1</td>\n",
              "      <td>9.705098</td>\n",
              "      <td>2017</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1189112</th>\n",
              "      <td>6238</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>전포동</td>\n",
              "      <td>산99-46</td>\n",
              "      <td>서면</td>\n",
              "      <td>전포동 산99-46 서면</td>\n",
              "      <td>57.59</td>\n",
              "      <td>1976</td>\n",
              "      <td>201706</td>\n",
              "      <td>21~30</td>\n",
              "      <td>3</td>\n",
              "      <td>9.137877</td>\n",
              "      <td>2017</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1189212</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>54.15</td>\n",
              "      <td>1974</td>\n",
              "      <td>201706</td>\n",
              "      <td>1~10</td>\n",
              "      <td>5</td>\n",
              "      <td>10.173553</td>\n",
              "      <td>2017</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         apartment_id   city dong  ... transaction_real_price  year month\n",
              "695725           6238  부산광역시  전포동  ...               8.682877  2011     5\n",
              "696834           6238  부산광역시  전포동  ...               8.922792  2011     8\n",
              "700884           6225  부산광역시  범전동  ...              10.126671  2012     6\n",
              "702589           6238  부산광역시  전포동  ...               8.699681  2012    12\n",
              "702876           6238  부산광역시  전포동  ...               8.909370  2013     1\n",
              "703979           6238  부산광역시  전포동  ...               8.853808  2013     4\n",
              "711024           6238  부산광역시  전포동  ...               8.853808  2014     7\n",
              "712441           6238  부산광역시  전포동  ...               9.082621  2014    10\n",
              "713864           6238  부산광역시  전포동  ...               8.955577  2015     1\n",
              "721684           6238  부산광역시  전포동  ...               9.059634  2016     4\n",
              "722888           6225  부산광역시  범전동  ...               4.615121  2016     6\n",
              "723957           6225  부산광역시  범전동  ...              10.051951  2016     8\n",
              "726597           6238  부산광역시  전포동  ...               8.802522  2016    12\n",
              "1186983          6225  부산광역시  범전동  ...               9.705098  2017     2\n",
              "1189112          6238  부산광역시  전포동  ...               9.137877  2017     6\n",
              "1189212          6225  부산광역시  범전동  ...              10.173553  2017     6\n",
              "\n",
              "[16 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4TZkXpIFjA5W",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "407e6a42-cbf2-4348-8622-4a46a0f2461f"
      },
      "source": [
        "train[train['floor']==-1].sort_values(\"transaction_real_price\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "      <th>year</th>\n",
              "      <th>month</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>722888</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>138.250</td>\n",
              "      <td>1974</td>\n",
              "      <td>201606</td>\n",
              "      <td>21~30</td>\n",
              "      <td>-1</td>\n",
              "      <td>4.615121</td>\n",
              "      <td>2016</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>651052</th>\n",
              "      <td>1514</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>대청동4가</td>\n",
              "      <td>75-176</td>\n",
              "      <td>근영빌라2동</td>\n",
              "      <td>대청동4가 75-176 근영빌라2동</td>\n",
              "      <td>47.170</td>\n",
              "      <td>1996</td>\n",
              "      <td>200809</td>\n",
              "      <td>21~30</td>\n",
              "      <td>-1</td>\n",
              "      <td>7.170888</td>\n",
              "      <td>2008</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>651087</th>\n",
              "      <td>1513</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>대청동4가</td>\n",
              "      <td>75-181</td>\n",
              "      <td>근영빌라1동</td>\n",
              "      <td>대청동4가 75-181 근영빌라1동</td>\n",
              "      <td>59.890</td>\n",
              "      <td>1996</td>\n",
              "      <td>200810</td>\n",
              "      <td>11~20</td>\n",
              "      <td>-1</td>\n",
              "      <td>7.824446</td>\n",
              "      <td>2008</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>651641</th>\n",
              "      <td>1514</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>대청동4가</td>\n",
              "      <td>75-176</td>\n",
              "      <td>근영빌라2동</td>\n",
              "      <td>대청동4가 75-176 근영빌라2동</td>\n",
              "      <td>47.170</td>\n",
              "      <td>1996</td>\n",
              "      <td>201101</td>\n",
              "      <td>21~31</td>\n",
              "      <td>-1</td>\n",
              "      <td>8.071219</td>\n",
              "      <td>2011</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>674249</th>\n",
              "      <td>9764</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>동삼동</td>\n",
              "      <td>213-19</td>\n",
              "      <td>조은아크로빌</td>\n",
              "      <td>동삼동 213-19 조은아크로빌</td>\n",
              "      <td>46.800</td>\n",
              "      <td>2001</td>\n",
              "      <td>201506</td>\n",
              "      <td>11~20</td>\n",
              "      <td>-1</td>\n",
              "      <td>8.160804</td>\n",
              "      <td>2015</td>\n",
              "      <td>6</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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>537384</th>\n",
              "      <td>10310</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>청담동</td>\n",
              "      <td>102-13</td>\n",
              "      <td>청담파라곤Ⅱ 2단지</td>\n",
              "      <td>청담동 102-13 청담파라곤Ⅱ 2단지</td>\n",
              "      <td>241.880</td>\n",
              "      <td>2010</td>\n",
              "      <td>201012</td>\n",
              "      <td>11~20</td>\n",
              "      <td>-1</td>\n",
              "      <td>12.577640</td>\n",
              "      <td>2010</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1095417</th>\n",
              "      <td>11320</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>한남동</td>\n",
              "      <td>810</td>\n",
              "      <td>한남더힐</td>\n",
              "      <td>한남동 810 한남더힐</td>\n",
              "      <td>212.524</td>\n",
              "      <td>2011</td>\n",
              "      <td>201705</td>\n",
              "      <td>21~31</td>\n",
              "      <td>-1</td>\n",
              "      <td>13.081543</td>\n",
              "      <td>2017</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1095416</th>\n",
              "      <td>11320</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>한남동</td>\n",
              "      <td>810</td>\n",
              "      <td>한남더힐</td>\n",
              "      <td>한남동 810 한남더힐</td>\n",
              "      <td>212.524</td>\n",
              "      <td>2011</td>\n",
              "      <td>201705</td>\n",
              "      <td>21~31</td>\n",
              "      <td>-1</td>\n",
              "      <td>13.102163</td>\n",
              "      <td>2017</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23585</th>\n",
              "      <td>11320</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>한남동</td>\n",
              "      <td>810</td>\n",
              "      <td>한남더힐</td>\n",
              "      <td>한남동 810 한남더힐</td>\n",
              "      <td>240.230</td>\n",
              "      <td>2011</td>\n",
              "      <td>201612</td>\n",
              "      <td>1~10</td>\n",
              "      <td>-1</td>\n",
              "      <td>13.204866</td>\n",
              "      <td>2016</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23376</th>\n",
              "      <td>11320</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>한남동</td>\n",
              "      <td>810</td>\n",
              "      <td>한남더힐</td>\n",
              "      <td>한남동 810 한남더힐</td>\n",
              "      <td>240.230</td>\n",
              "      <td>2011</td>\n",
              "      <td>201610</td>\n",
              "      <td>11~20</td>\n",
              "      <td>-1</td>\n",
              "      <td>13.262127</td>\n",
              "      <td>2016</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>140 rows × 14 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         apartment_id   city   dong  ... transaction_real_price  year month\n",
              "722888           6225  부산광역시    범전동  ...               4.615121  2016     6\n",
              "651052           1514  부산광역시  대청동4가  ...               7.170888  2008     9\n",
              "651087           1513  부산광역시  대청동4가  ...               7.824446  2008    10\n",
              "651641           1514  부산광역시  대청동4가  ...               8.071219  2011     1\n",
              "674249           9764  부산광역시    동삼동  ...               8.160804  2015     6\n",
              "...               ...    ...    ...  ...                    ...   ...   ...\n",
              "537384          10310  서울특별시    청담동  ...              12.577640  2010    12\n",
              "1095417         11320  서울특별시    한남동  ...              13.081543  2017     5\n",
              "1095416         11320  서울특별시    한남동  ...              13.102163  2017     5\n",
              "23585           11320  서울특별시    한남동  ...              13.204866  2016    12\n",
              "23376           11320  서울특별시    한남동  ...              13.262127  2016    10\n",
              "\n",
              "[140 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nGGhfDUUj0LT"
      },
      "source": [
        "# year_of_completion : 설립일자"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RSGB8cCrjZTq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 375
        },
        "outputId": "0453fbfc-7250-4d79-aa4f-881ddb2cb9bc"
      },
      "source": [
        "f,ax = plt.subplots(figsize = (22,6))\n",
        "sns.boxplot(train['year_of_completion'],train['transaction_real_price'])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbdfabceb8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fG1z3JG7kJx9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 134
        },
        "outputId": "92e8e0c5-cf63-4448-dd63-532e780ea7fa"
      },
      "source": [
        "train[train['transaction_real_price']<5]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>addr_kr</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "      <th>year</th>\n",
              "      <th>month</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>722888</th>\n",
              "      <td>6225</td>\n",
              "      <td>부산광역시</td>\n",
              "      <td>범전동</td>\n",
              "      <td>263-5</td>\n",
              "      <td>서면</td>\n",
              "      <td>범전동 263-5 서면</td>\n",
              "      <td>138.25</td>\n",
              "      <td>1974</td>\n",
              "      <td>201606</td>\n",
              "      <td>21~30</td>\n",
              "      <td>-1</td>\n",
              "      <td>4.615121</td>\n",
              "      <td>2016</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        apartment_id   city dong  ... transaction_real_price  year month\n",
              "722888          6225  부산광역시  범전동  ...               4.615121  2016     6\n",
              "\n",
              "[1 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "POdxv6Wokje1"
      },
      "source": [
        "# Catergorical Variable\n",
        "## city : 도시"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QR644dQLlOj4"
      },
      "source": [
        "# 라벨링\n",
        "replace_name = {'서울특별시' : 0, '부산광역시' : 1 }\n",
        "train = train.replace({'city' : replace_name})\n",
        "test = test.replace({'city' : replace_name})"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7YNxD7MilCU_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 409
        },
        "outputId": "c4086ee0-2a0d-4a8b-e232-b7b5df1c4b19"
      },
      "source": [
        "f, ax = plt.subplots(figsize = (8,6))\n",
        "sns.boxplot(train['city'], train['transaction_real_price'])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbe573ceb8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Quxf2lBok7Jo"
      },
      "source": [
        "## addr_kr : 주소\n",
        "\n",
        "- addr_kr 은 동 지번 아파트단지를 모두 합친 것이므로 불필요 삭제"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0O_SpXtXlAVC"
      },
      "source": [
        "train = train.drop('addr_kr', axis=1)\n",
        "test = test.drop('addr_kr', axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "E0R84cesly_q"
      },
      "source": [
        "# dong : 동"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J-ob3bKRlrB4"
      },
      "source": [
        "# 한강 유무에 따른 컬럼 생성\n",
        "train['hangang']=train['dong'].isin(['성수동1가','삼성동','이촌동','공덕동','서교동','한강로3가','목동']).astype(int)\n",
        "test['hangang']=test['dong'].isin(['성수동1가','삼성동','이촌동','공덕동','서교동','한강로3가','목동']).astype(int)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qmxU5m5ol7J3",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9dbdfb5f-ab0d-420a-9323-2a3ebecd8601"
      },
      "source": [
        "len(train['dong'].unique())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "473"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BO_e5mLomEWL"
      },
      "source": [
        "- 동은 그 지역을 그룹화 한것으로 지역에 따라 아파트 가격차이가 날 수 있다.\n",
        "따라서 아프트 실거래가의 평균순으로 데이터를 라벨링 한다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SOah6X6rmD4Y",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 202
        },
        "outputId": "76276221-e98a-431d-af18-822746c0ba4c"
      },
      "source": [
        "group_dong = train[['transaction_real_price','dong']].groupby('dong').mean().sort_values('transaction_real_price').reset_index()\n",
        "group_dong.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>dong</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>신선동3가</td>\n",
              "      <td>8.543619</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>봉래동5가</td>\n",
              "      <td>8.590871</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>장안읍 명례리</td>\n",
              "      <td>8.597728</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>신창동2가</td>\n",
              "      <td>8.601466</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>신선동2가</td>\n",
              "      <td>8.622581</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "      dong  transaction_real_price\n",
              "0    신선동3가                8.543619\n",
              "1    봉래동5가                8.590871\n",
              "2  장안읍 명례리                8.597728\n",
              "3    신창동2가                8.601466\n",
              "4    신선동2가                8.622581"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QDxX6DTwmpoz",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7a4bdf56-d636-4c50-d3e7-ce0f7d335b84"
      },
      "source": [
        "dong_num = { }\n",
        "for i in range(len(group_dong)):\n",
        "  dong = group_dong['dong'].iloc[i]\n",
        "  dong_num[dong] = i\n",
        "\n",
        "dong_num"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'가락동': 391,\n",
              " '가리봉동': 159,\n",
              " '가산동': 167,\n",
              " '가야동': 94,\n",
              " '가양동': 210,\n",
              " '갈월동': 362,\n",
              " '갈현동': 169,\n",
              " '감만동': 29,\n",
              " '감전동': 47,\n",
              " '감천동': 46,\n",
              " '강일동': 311,\n",
              " '개금동': 70,\n",
              " '개봉동': 242,\n",
              " '개포동': 432,\n",
              " '거여동': 281,\n",
              " '거제동': 162,\n",
              " '견지동': 453,\n",
              " '고덕동': 388,\n",
              " '고척동': 222,\n",
              " '공덕동': 401,\n",
              " '공릉동': 192,\n",
              " '공항동': 279,\n",
              " '광안동': 84,\n",
              " '광장동': 415,\n",
              " '괘법동': 115,\n",
              " '괴정동': 66,\n",
              " '교남동': 465,\n",
              " '교북동': 280,\n",
              " '구기동': 327,\n",
              " '구로동': 194,\n",
              " '구산동': 191,\n",
              " '구서동': 144,\n",
              " '구수동': 321,\n",
              " '구의동': 387,\n",
              " '구평동': 35,\n",
              " '구포동': 65,\n",
              " '군자동': 316,\n",
              " '궁동': 204,\n",
              " '금곡동': 37,\n",
              " '금사동': 49,\n",
              " '금호동1가': 332,\n",
              " '금호동2가': 426,\n",
              " '금호동3가': 355,\n",
              " '금호동4가': 404,\n",
              " '기장읍 교리': 58,\n",
              " '기장읍 내리': 160,\n",
              " '기장읍 대라리': 42,\n",
              " '기장읍 대변리': 13,\n",
              " '기장읍 동부리': 50,\n",
              " '기장읍 서부리': 17,\n",
              " '기장읍 청강리': 41,\n",
              " '길동': 243,\n",
              " '길음동': 319,\n",
              " '낙민동': 136,\n",
              " '낙원동': 174,\n",
              " '남가좌동': 324,\n",
              " '남대문로5가': 466,\n",
              " '남부민동': 82,\n",
              " '남산동': 71,\n",
              " '남산동2가': 250,\n",
              " '남창동': 110,\n",
              " '남천동': 215,\n",
              " '남항동1가': 131,\n",
              " '남항동2가': 122,\n",
              " '남항동3가': 15,\n",
              " '남현동': 317,\n",
              " '내곡동': 443,\n",
              " '내발산동': 304,\n",
              " '내수동': 462,\n",
              " '냉천동': 383,\n",
              " '노고산동': 129,\n",
              " '노량진동': 312,\n",
              " '녹번동': 251,\n",
              " '녹산동': 128,\n",
              " '논현동': 413,\n",
              " '누상동': 21,\n",
              " '능동': 233,\n",
              " '다대동': 55,\n",
              " '답십리동': 260,\n",
              " '당감동': 67,\n",
              " '당리동': 72,\n",
              " '당산동': 310,\n",
              " '당산동1가': 298,\n",
              " '당산동2가': 197,\n",
              " '당산동3가': 339,\n",
              " '당산동4가': 352,\n",
              " '당산동5가': 406,\n",
              " '당산동6가': 73,\n",
              " '당주동': 363,\n",
              " '대교동1가': 335,\n",
              " '대교동2가': 137,\n",
              " '대림동': 249,\n",
              " '대방동': 328,\n",
              " '대연동': 100,\n",
              " '대저1동': 195,\n",
              " '대조동': 112,\n",
              " '대창동1가': 59,\n",
              " '대청동1가': 16,\n",
              " '대청동2가': 164,\n",
              " '대청동4가': 23,\n",
              " '대치동': 461,\n",
              " '대평동1가': 90,\n",
              " '대현동': 289,\n",
              " '대흥동': 369,\n",
              " '덕천동': 26,\n",
              " '덕포동': 52,\n",
              " '도곡동': 456,\n",
              " '도림동': 299,\n",
              " '도봉동': 185,\n",
              " '도선동': 333,\n",
              " '도원동': 379,\n",
              " '도화동': 358,\n",
              " '독산동': 182,\n",
              " '돈암동': 272,\n",
              " '동광동5가': 12,\n",
              " '동교동': 374,\n",
              " '동대신동1가': 96,\n",
              " '동대신동2가': 153,\n",
              " '동대신동3가': 33,\n",
              " '동빙고동': 457,\n",
              " '동삼동': 18,\n",
              " '동선동1가': 376,\n",
              " '동선동3가': 265,\n",
              " '동선동4가': 187,\n",
              " '동선동5가': 259,\n",
              " '동소문동4가': 220,\n",
              " '동소문동5가': 371,\n",
              " '동소문동6가': 150,\n",
              " '동소문동7가': 309,\n",
              " '동숭동': 104,\n",
              " '동자동': 454,\n",
              " '동작동': 427,\n",
              " '둔촌동': 408,\n",
              " '등촌동': 261,\n",
              " '마곡동': 381,\n",
              " '마장동': 291,\n",
              " '마천동': 315,\n",
              " '마포동': 237,\n",
              " '만덕동': 80,\n",
              " '만리동1가': 392,\n",
              " '만리동2가': 135,\n",
              " '망미동': 134,\n",
              " '망우동': 196,\n",
              " '망원동': 286,\n",
              " '면목동': 201,\n",
              " '명륜1가': 270,\n",
              " '명륜2가': 372,\n",
              " '명륜3가': 125,\n",
              " '명륜동': 254,\n",
              " '명일동': 341,\n",
              " '명장동': 54,\n",
              " '명지동': 240,\n",
              " '모라동': 31,\n",
              " '목동': 419,\n",
              " '무악동': 410,\n",
              " '묵동': 231,\n",
              " '묵정동': 158,\n",
              " '문래동2가': 207,\n",
              " '문래동3가': 407,\n",
              " '문래동4가': 300,\n",
              " '문래동5가': 257,\n",
              " '문래동6가': 224,\n",
              " '문배동': 396,\n",
              " '문정동': 398,\n",
              " '문현동': 107,\n",
              " '미근동': 45,\n",
              " '미아동': 267,\n",
              " '민락동': 211,\n",
              " '반송동': 19,\n",
              " '반여동': 105,\n",
              " '반포동': 467,\n",
              " '방배동': 442,\n",
              " '방이동': 434,\n",
              " '방학동': 190,\n",
              " '방화동': 202,\n",
              " '번동': 181,\n",
              " '범일동': 166,\n",
              " '범전동': 118,\n",
              " '범천동': 63,\n",
              " '보광동': 425,\n",
              " '보문동2가': 217,\n",
              " '보문동3가': 331,\n",
              " '보문동4가': 348,\n",
              " '보문동6가': 212,\n",
              " '보수동1가': 5,\n",
              " '보수동2가': 20,\n",
              " '보수동3가': 91,\n",
              " '복천동': 141,\n",
              " '본동': 361,\n",
              " '봉래동1가': 93,\n",
              " '봉래동2가': 155,\n",
              " '봉래동3가': 22,\n",
              " '봉래동4가': 28,\n",
              " '봉래동5가': 1,\n",
              " '봉천동': 282,\n",
              " '부곡동': 114,\n",
              " '부민동1가': 121,\n",
              " '부민동3가': 140,\n",
              " '부암동': 143,\n",
              " '부용동1가': 101,\n",
              " '부전동': 111,\n",
              " '부평동2가': 8,\n",
              " '부평동4가': 124,\n",
              " '북가좌동': 330,\n",
              " '북아현동': 306,\n",
              " '불광동': 290,\n",
              " '사근동': 230,\n",
              " '사당동': 346,\n",
              " '사직동': 156,\n",
              " '산림동': 206,\n",
              " '산천동': 347,\n",
              " '삼락동': 86,\n",
              " '삼선동2가': 314,\n",
              " '삼선동3가': 385,\n",
              " '삼선동4가': 285,\n",
              " '삼선동5가': 157,\n",
              " '삼성동': 452,\n",
              " '삼전동': 226,\n",
              " '상계동': 172,\n",
              " '상도1동': 203,\n",
              " '상도동': 384,\n",
              " '상봉동': 248,\n",
              " '상수동': 438,\n",
              " '상암동': 412,\n",
              " '상왕십리동': 440,\n",
              " '상월곡동': 238,\n",
              " '상일동': 365,\n",
              " '서교동': 424,\n",
              " '서대신동1가': 56,\n",
              " '서대신동2가': 99,\n",
              " '서대신동3가': 138,\n",
              " '서동': 24,\n",
              " '서빙고동': 464,\n",
              " '서초동': 433,\n",
              " '석관동': 269,\n",
              " '석촌동': 386,\n",
              " '성내동': 275,\n",
              " '성북동': 83,\n",
              " '성북동1가': 318,\n",
              " '성산동': 303,\n",
              " '성수동1가': 400,\n",
              " '성수동2가': 399,\n",
              " '세곡동': 430,\n",
              " '송정동': 147,\n",
              " '송파동': 420,\n",
              " '수색동': 263,\n",
              " '수서동': 380,\n",
              " '수송동': 241,\n",
              " '수안동': 88,\n",
              " '수영동': 168,\n",
              " '수유동': 178,\n",
              " '수정동': 9,\n",
              " '순화동': 338,\n",
              " '숭인동': 228,\n",
              " '시흥동': 198,\n",
              " '신계동': 451,\n",
              " '신공덕동': 397,\n",
              " '신교동': 351,\n",
              " '신길동': 262,\n",
              " '신내동': 193,\n",
              " '신당동': 360,\n",
              " '신대방동': 356,\n",
              " '신도림동': 336,\n",
              " '신림동': 219,\n",
              " '신문로1가': 343,\n",
              " '신사동': 225,\n",
              " '신선동2가': 4,\n",
              " '신선동3가': 0,\n",
              " '신설동': 163,\n",
              " '신수동': 337,\n",
              " '신영동': 177,\n",
              " '신원동': 437,\n",
              " '신월동': 175,\n",
              " '신정동': 342,\n",
              " '신창동': 353,\n",
              " '신창동1가': 127,\n",
              " '신창동2가': 3,\n",
              " '신천동': 448,\n",
              " '신평동': 62,\n",
              " '신호동': 117,\n",
              " '쌍문동': 152,\n",
              " '아미동2가': 11,\n",
              " '아현동': 423,\n",
              " '안락동': 133,\n",
              " '안암동1가': 308,\n",
              " '안암동3가': 109,\n",
              " '안암동4가': 68,\n",
              " '암남동': 78,\n",
              " '암사동': 370,\n",
              " '압구정동': 470,\n",
              " '양재동': 377,\n",
              " '양정동': 139,\n",
              " '양평동1가': 245,\n",
              " '양평동2가': 213,\n",
              " '양평동3가': 301,\n",
              " '양평동4가': 273,\n",
              " '양평동5가': 284,\n",
              " '양평동6가': 247,\n",
              " '엄궁동': 79,\n",
              " '여의도동': 455,\n",
              " '역삼동': 411,\n",
              " '역촌동': 208,\n",
              " '연건동': 106,\n",
              " '연남동': 296,\n",
              " '연산동': 116,\n",
              " '연지동': 142,\n",
              " '연희동': 232,\n",
              " '염리동': 350,\n",
              " '염창동': 292,\n",
              " '영등포동': 313,\n",
              " '영등포동1가': 43,\n",
              " '영등포동2가': 44,\n",
              " '영등포동3가': 30,\n",
              " '영등포동4가': 60,\n",
              " '영등포동5가': 151,\n",
              " '영등포동7가': 252,\n",
              " '영등포동8가': 325,\n",
              " '영선동1가': 126,\n",
              " '영선동2가': 180,\n",
              " '영선동3가': 97,\n",
              " '영선동4가': 81,\n",
              " '영주동': 38,\n",
              " '영천동': 345,\n",
              " '예장동': 239,\n",
              " '오금동': 375,\n",
              " '오류동': 199,\n",
              " '오장동': 57,\n",
              " '옥수동': 417,\n",
              " '옥인동': 428,\n",
              " '온수동': 295,\n",
              " '온천동': 130,\n",
              " '용강동': 421,\n",
              " '용답동': 268,\n",
              " '용당동': 108,\n",
              " '용두동': 297,\n",
              " '용문동': 395,\n",
              " '용산동2가': 344,\n",
              " '용산동5가': 469,\n",
              " '용호동': 161,\n",
              " '우동': 277,\n",
              " '우면동': 429,\n",
              " '우암동': 64,\n",
              " '우이동': 227,\n",
              " '원효로1가': 444,\n",
              " '원효로2가': 149,\n",
              " '원효로3가': 154,\n",
              " '원효로4가': 405,\n",
              " '월계동': 179,\n",
              " '율현동': 436,\n",
              " '을지로5가': 287,\n",
              " '응봉동': 349,\n",
              " '응암동': 258,\n",
              " '이문동': 283,\n",
              " '이촌동': 449,\n",
              " '이태원동': 431,\n",
              " '이화동': 205,\n",
              " '익선동': 145,\n",
              " '인의동': 389,\n",
              " '인현동2가': 119,\n",
              " '일광면 삼성리': 27,\n",
              " '일광면 이천리': 25,\n",
              " '일원동': 435,\n",
              " '자곡동': 450,\n",
              " '자양동': 393,\n",
              " '잠실동': 459,\n",
              " '잠원동': 447,\n",
              " '장림동': 39,\n",
              " '장안동': 276,\n",
              " '장안읍 명례리': 2,\n",
              " '장안읍 좌천리': 103,\n",
              " '장위동': 278,\n",
              " '장전동': 176,\n",
              " '장지동': 409,\n",
              " '장충동1가': 472,\n",
              " '장충동2가': 6,\n",
              " '재송동': 95,\n",
              " '전농동': 294,\n",
              " '전포동': 48,\n",
              " '정관읍 달산리': 98,\n",
              " '정관읍 매학리': 148,\n",
              " '정관읍 모전리': 170,\n",
              " '정관읍 방곡리': 102,\n",
              " '정관읍 용수리': 184,\n",
              " '정동': 340,\n",
              " '정릉동': 223,\n",
              " '제기동': 271,\n",
              " '종암동': 293,\n",
              " '좌동': 146,\n",
              " '좌천동': 36,\n",
              " '주례동': 74,\n",
              " '주성동': 463,\n",
              " '중계동': 216,\n",
              " '중곡동': 200,\n",
              " '중동': 186,\n",
              " '중림동': 394,\n",
              " '중앙동4가': 7,\n",
              " '중화동': 209,\n",
              " '증산동': 229,\n",
              " '지사동': 87,\n",
              " '진관동': 373,\n",
              " '창동': 189,\n",
              " '창신동': 234,\n",
              " '창전동': 382,\n",
              " '창천동': 188,\n",
              " '천연동': 264,\n",
              " '천왕동': 302,\n",
              " '천호동': 256,\n",
              " '철마면 고촌리': 165,\n",
              " '청담동': 458,\n",
              " '청량리동': 307,\n",
              " '청룡동': 113,\n",
              " '청암동': 471,\n",
              " '청운동': 422,\n",
              " '청파동1가': 255,\n",
              " '청파동3가': 320,\n",
              " '청학동': 76,\n",
              " '초량동': 53,\n",
              " '초읍동': 92,\n",
              " '초장동': 14,\n",
              " '충무동1가': 274,\n",
              " '충무동2가': 77,\n",
              " '충무로4가': 364,\n",
              " '충무로5가': 173,\n",
              " '충신동': 89,\n",
              " '충정로2가': 288,\n",
              " '충정로3가': 218,\n",
              " '칠산동': 51,\n",
              " '토성동1가': 183,\n",
              " '토성동2가': 40,\n",
              " '토성동3가': 123,\n",
              " '토성동5가': 69,\n",
              " '토정동': 359,\n",
              " '통인동': 75,\n",
              " '평창동': 378,\n",
              " '풍납동': 329,\n",
              " '필동1가': 10,\n",
              " '필동3가': 323,\n",
              " '필운동': 414,\n",
              " '하계동': 214,\n",
              " '하단동': 85,\n",
              " '하왕십리동': 334,\n",
              " '하월곡동': 305,\n",
              " '하중동': 460,\n",
              " '학장동': 34,\n",
              " '한강로1가': 439,\n",
              " '한강로2가': 402,\n",
              " '한강로3가': 445,\n",
              " '한남동': 441,\n",
              " '합동': 390,\n",
              " '합정동': 366,\n",
              " '항동': 244,\n",
              " '행당동': 368,\n",
              " '행촌동': 32,\n",
              " '현석동': 403,\n",
              " '현저동': 322,\n",
              " '혜화동': 326,\n",
              " '홍은동': 235,\n",
              " '홍익동': 236,\n",
              " '홍제동': 253,\n",
              " '화곡동': 266,\n",
              " '화명동': 132,\n",
              " '화양동': 171,\n",
              " '황학동': 357,\n",
              " '회기동': 221,\n",
              " '회동동': 61,\n",
              " '회현동1가': 446,\n",
              " '회현동2가': 468,\n",
              " '효제동': 120,\n",
              " '효창동': 367,\n",
              " '후암동': 354,\n",
              " '휘경동': 246,\n",
              " '흑석동': 416,\n",
              " '흥인동': 418}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Wv7qXBVXm-5T",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 378
        },
        "outputId": "5144eae2-626c-4960-eaa7-de4c736d351e"
      },
      "source": [
        "train = train.replace({'dong' : dong_num}) # replace()로 라벨링\n",
        "test = test.replace({'dong' : dong_num}) # 동일하게\n",
        "train.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "      <th>year</th>\n",
              "      <th>month</th>\n",
              "      <th>hangang</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>7622</td>\n",
              "      <td>0</td>\n",
              "      <td>351</td>\n",
              "      <td>6-13</td>\n",
              "      <td>신현(101동)</td>\n",
              "      <td>84.82</td>\n",
              "      <td>2002</td>\n",
              "      <td>200801</td>\n",
              "      <td>21~31</td>\n",
              "      <td>2</td>\n",
              "      <td>10.532123</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5399</td>\n",
              "      <td>0</td>\n",
              "      <td>414</td>\n",
              "      <td>142</td>\n",
              "      <td>사직파크맨션</td>\n",
              "      <td>99.17</td>\n",
              "      <td>1973</td>\n",
              "      <td>200801</td>\n",
              "      <td>1~10</td>\n",
              "      <td>6</td>\n",
              "      <td>9.903538</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3578</td>\n",
              "      <td>0</td>\n",
              "      <td>414</td>\n",
              "      <td>174-1</td>\n",
              "      <td>두레엘리시안</td>\n",
              "      <td>84.74</td>\n",
              "      <td>2007</td>\n",
              "      <td>200801</td>\n",
              "      <td>1~10</td>\n",
              "      <td>6</td>\n",
              "      <td>10.558439</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>10957</td>\n",
              "      <td>0</td>\n",
              "      <td>462</td>\n",
              "      <td>95</td>\n",
              "      <td>파크팰리스</td>\n",
              "      <td>146.39</td>\n",
              "      <td>2003</td>\n",
              "      <td>200801</td>\n",
              "      <td>11~20</td>\n",
              "      <td>15</td>\n",
              "      <td>11.678448</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>10639</td>\n",
              "      <td>0</td>\n",
              "      <td>462</td>\n",
              "      <td>110-15</td>\n",
              "      <td>킹스매너</td>\n",
              "      <td>194.43</td>\n",
              "      <td>2004</td>\n",
              "      <td>200801</td>\n",
              "      <td>21~31</td>\n",
              "      <td>3</td>\n",
              "      <td>11.695255</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   apartment_id  city  dong  ...  year month  hangang\n",
              "0          7622     0   351  ...  2008     1        0\n",
              "1          5399     0   414  ...  2008     1        0\n",
              "2          3578     0   414  ...  2008     1        0\n",
              "3         10957     0   462  ...  2008     1        0\n",
              "4         10639     0   462  ...  2008     1        0\n",
              "\n",
              "[5 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ghVfDB-wnSBm"
      },
      "source": [
        "# Jibun : 지번"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8jGKx-OZnQkf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "23a9d1c2-017e-495b-cf5a-5a79643a8c56"
      },
      "source": [
        "len(train['jibun'].unique())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "8961"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NPMDbretndjJ"
      },
      "source": [
        "- 지번은 unique가 동 과는 다르게 너무 많아서 라벨링을 시키면 unique가 너무 많아져서 하지않는다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b3FRD4vVnah5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 202
        },
        "outputId": "4dd09471-72e2-4b58-c7c5-6fdc6e211826"
      },
      "source": [
        "group_jibun = train[['transaction_real_price','jibun']].groupby('jibun').mean().sort_values('transaction_real_price').reset_index()\n",
        "group_jibun.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>jibun</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>산3-148</td>\n",
              "      <td>7.225286</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>737-1</td>\n",
              "      <td>7.494417</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>741-1</td>\n",
              "      <td>7.498174</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1181-4</td>\n",
              "      <td>7.585491</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>11-174</td>\n",
              "      <td>7.601402</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    jibun  transaction_real_price\n",
              "0  산3-148                7.225286\n",
              "1   737-1                7.494417\n",
              "2   741-1                7.498174\n",
              "3  1181-4                7.585491\n",
              "4  11-174                7.601402"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "14R3EmTCnz5v",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d40f2a90-df53-413e-fe93-13d0ec979f3b"
      },
      "source": [
        "jibun_num = { }\n",
        "for i in range(len(group_jibun)):\n",
        "  jibun = group_jibun['jibun'].iloc[i]\n",
        "  jibun_num[jibun] = i\n",
        "\n",
        "jibun_num"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'산3-148': 0,\n",
              " '737-1': 1,\n",
              " '741-1': 2,\n",
              " '1181-4': 3,\n",
              " '11-174': 4,\n",
              " '1175-1': 5,\n",
              " '1161-7': 6,\n",
              " '1186-1': 7,\n",
              " '1173-3': 8,\n",
              " '745-70': 9,\n",
              " '13-1105': 10,\n",
              " '92-326': 11,\n",
              " '1005-1': 12,\n",
              " '293-7': 13,\n",
              " '728-3': 14,\n",
              " '201-10': 15,\n",
              " '460-2': 16,\n",
              " '92-316': 17,\n",
              " '1056-7': 18,\n",
              " '591-4': 19,\n",
              " '132-3': 20,\n",
              " '532-9': 21,\n",
              " '734-4': 22,\n",
              " '74-18': 23,\n",
              " '75-176': 24,\n",
              " '62-1': 25,\n",
              " '865-32': 26,\n",
              " '716-12': 27,\n",
              " '116-15': 28,\n",
              " '243-1': 29,\n",
              " '1594-1': 30,\n",
              " '16-290': 31,\n",
              " '산20-1': 32,\n",
              " '515-27': 33,\n",
              " '177-74': 34,\n",
              " '96-24': 35,\n",
              " '250-1995': 36,\n",
              " '1585-1': 37,\n",
              " '1781-80': 38,\n",
              " '1293-11': 39,\n",
              " '54-11': 40,\n",
              " '1299-8': 41,\n",
              " '515-29': 42,\n",
              " '410-5': 43,\n",
              " '775-4': 44,\n",
              " '865-42': 45,\n",
              " '797-38': 46,\n",
              " '91-1': 47,\n",
              " '595-40': 48,\n",
              " '189-77': 49,\n",
              " '169-7': 50,\n",
              " '717-2': 51,\n",
              " '47-6': 52,\n",
              " '20-125': 53,\n",
              " '252-3': 54,\n",
              " '36-4': 55,\n",
              " '524-1': 56,\n",
              " '409-1': 57,\n",
              " '1405-10': 58,\n",
              " '75-181': 59,\n",
              " '199-24': 60,\n",
              " '1073-1': 61,\n",
              " '산96-6': 62,\n",
              " '796-27': 63,\n",
              " '44-151': 64,\n",
              " '1051-1': 65,\n",
              " '177-183': 66,\n",
              " '737-7': 67,\n",
              " '89-16': 68,\n",
              " '1495-4': 69,\n",
              " '926-1': 70,\n",
              " '3-320': 71,\n",
              " '1259-2': 72,\n",
              " '1646-33': 73,\n",
              " '164-66': 74,\n",
              " '1642-28': 75,\n",
              " '515-5': 76,\n",
              " '185-16': 77,\n",
              " '424-26': 78,\n",
              " '1481-51': 79,\n",
              " '1183-3': 80,\n",
              " '385-2': 81,\n",
              " '1135-127': 82,\n",
              " '635-1': 83,\n",
              " '842-10': 84,\n",
              " '924-12': 85,\n",
              " '776-8': 86,\n",
              " '409-2': 87,\n",
              " '784-3': 88,\n",
              " '산20-55': 89,\n",
              " '94-5': 90,\n",
              " '743-6': 91,\n",
              " '537-11': 92,\n",
              " '828-31': 93,\n",
              " '11-28': 94,\n",
              " '374-19': 95,\n",
              " '1423-18': 96,\n",
              " '1291-1405': 97,\n",
              " '1493-12': 98,\n",
              " '75-167': 99,\n",
              " '1485-9': 100,\n",
              " '782-25': 101,\n",
              " '1485-19': 102,\n",
              " '249-20': 103,\n",
              " '1076-8': 104,\n",
              " '875-24': 105,\n",
              " '520-95': 106,\n",
              " '7-13': 107,\n",
              " '1070-20': 108,\n",
              " '1120-4': 109,\n",
              " '1516-2': 110,\n",
              " '1495-13': 111,\n",
              " '1554': 112,\n",
              " '11-41': 113,\n",
              " '478-19': 114,\n",
              " '627-1': 115,\n",
              " '527-2': 116,\n",
              " '221-56': 117,\n",
              " '65-17': 118,\n",
              " '775-3': 119,\n",
              " '395-3': 120,\n",
              " '1057-57': 121,\n",
              " '75-171': 122,\n",
              " '1481-13': 123,\n",
              " '870-30': 124,\n",
              " '181-71': 125,\n",
              " '87-8': 126,\n",
              " '1060-390': 127,\n",
              " '22-32': 128,\n",
              " '629-1': 129,\n",
              " '1235-20': 130,\n",
              " '176-24': 131,\n",
              " '796-37': 132,\n",
              " '1287-2': 133,\n",
              " '1533-1': 134,\n",
              " '208-2': 135,\n",
              " '879-2': 136,\n",
              " '215-1': 137,\n",
              " '93': 138,\n",
              " '1077-18': 139,\n",
              " '878-26': 140,\n",
              " '859-16': 141,\n",
              " '986-5': 142,\n",
              " '4-10': 143,\n",
              " '258-29': 144,\n",
              " '365-6': 145,\n",
              " '1070-2': 146,\n",
              " '417-8': 147,\n",
              " '364-3': 148,\n",
              " '193-29': 149,\n",
              " '649-3': 150,\n",
              " '263-6': 151,\n",
              " '126-3': 152,\n",
              " '40-15': 153,\n",
              " '576-1': 154,\n",
              " '109-8': 155,\n",
              " '산62-8': 156,\n",
              " '828-27': 157,\n",
              " '1027-2': 158,\n",
              " '574-23': 159,\n",
              " '349-14': 160,\n",
              " '1029-56': 161,\n",
              " '662-14': 162,\n",
              " '80-142': 163,\n",
              " '1076-19': 164,\n",
              " '442-5': 165,\n",
              " '176-28': 166,\n",
              " '325-8': 167,\n",
              " '278-3': 168,\n",
              " '485-6': 169,\n",
              " '628-1': 170,\n",
              " '16-939': 171,\n",
              " '451-3': 172,\n",
              " '1600-1': 173,\n",
              " '692-36': 174,\n",
              " '186-62': 175,\n",
              " '890-37': 176,\n",
              " '122-8': 177,\n",
              " '16-936': 178,\n",
              " '226-6': 179,\n",
              " '782-30': 180,\n",
              " '1786-1': 181,\n",
              " '948-1': 182,\n",
              " '227-233': 183,\n",
              " '1250-2': 184,\n",
              " '1075-3': 185,\n",
              " '328-4': 186,\n",
              " '822-4': 187,\n",
              " '산54-4': 188,\n",
              " '172-2': 189,\n",
              " '238-5': 190,\n",
              " '362-62': 191,\n",
              " '177-197': 192,\n",
              " '145-4': 193,\n",
              " '514-8': 194,\n",
              " '552-18': 195,\n",
              " '1057-2': 196,\n",
              " '54-67': 197,\n",
              " '497-11': 198,\n",
              " '508-22': 199,\n",
              " '75-128': 200,\n",
              " '423-4': 201,\n",
              " '327-22': 202,\n",
              " '317-40': 203,\n",
              " '113': 204,\n",
              " '1151-2': 205,\n",
              " '668-1': 206,\n",
              " '453-3': 207,\n",
              " '1150-7': 208,\n",
              " '1127': 209,\n",
              " '345-9': 210,\n",
              " '534-5': 211,\n",
              " '146-16': 212,\n",
              " '326-1': 213,\n",
              " '1160-12': 214,\n",
              " '1581-22': 215,\n",
              " '72-23': 216,\n",
              " '1402-5': 217,\n",
              " '189-10': 218,\n",
              " '1070-6': 219,\n",
              " '171-2': 220,\n",
              " '967-8': 221,\n",
              " '656-3': 222,\n",
              " '34-9': 223,\n",
              " '186-8': 224,\n",
              " '118-3': 225,\n",
              " '1480-35': 226,\n",
              " '260-3': 227,\n",
              " '1093-39': 228,\n",
              " '285-2': 229,\n",
              " '655-7': 230,\n",
              " '141-113': 231,\n",
              " '481-4': 232,\n",
              " '510-37': 233,\n",
              " '743-27': 234,\n",
              " '683-11': 235,\n",
              " '536-18': 236,\n",
              " '115-49': 237,\n",
              " '892-21': 238,\n",
              " '94-2': 239,\n",
              " '488-1': 240,\n",
              " '352-2': 241,\n",
              " '381-15': 242,\n",
              " '303-7': 243,\n",
              " '산14-6': 244,\n",
              " '871-1': 245,\n",
              " '261-128': 246,\n",
              " '752-15': 247,\n",
              " '3-495': 248,\n",
              " '1077-16': 249,\n",
              " '409-18': 250,\n",
              " '1829-3': 251,\n",
              " '265-44': 252,\n",
              " '150-8': 253,\n",
              " '485-8': 254,\n",
              " '413-13': 255,\n",
              " '55-139': 256,\n",
              " '산99-46': 257,\n",
              " '871-18': 258,\n",
              " '1078-2': 259,\n",
              " '78-76': 260,\n",
              " '63-245': 261,\n",
              " '51-11': 262,\n",
              " '892-10': 263,\n",
              " '752-13': 264,\n",
              " '95-3': 265,\n",
              " '325-19': 266,\n",
              " '329-1': 267,\n",
              " '771-11': 268,\n",
              " '782-29': 269,\n",
              " '936-18': 270,\n",
              " '75-120': 271,\n",
              " '482-8': 272,\n",
              " '445-8': 273,\n",
              " '1581-9': 274,\n",
              " '401': 275,\n",
              " '296-3': 276,\n",
              " '30-11': 277,\n",
              " '377-2': 278,\n",
              " '263-5': 279,\n",
              " '716-18': 280,\n",
              " '551-33': 281,\n",
              " '798-55': 282,\n",
              " '306-5': 283,\n",
              " '62-31': 284,\n",
              " '1261-32': 285,\n",
              " '538-40': 286,\n",
              " '166-4': 287,\n",
              " '36-3': 288,\n",
              " '172-35': 289,\n",
              " '574-13': 290,\n",
              " '674-1': 291,\n",
              " '545-1': 292,\n",
              " '980-7': 293,\n",
              " '1-2': 294,\n",
              " '1270-6': 295,\n",
              " '744-32': 296,\n",
              " '741-3': 297,\n",
              " '1081-1': 298,\n",
              " '763-28': 299,\n",
              " '1000-24': 300,\n",
              " '1048-4': 301,\n",
              " '356-20': 302,\n",
              " '1011-51': 303,\n",
              " '683-12': 304,\n",
              " '336-10': 305,\n",
              " '865-166': 306,\n",
              " '1378-24': 307,\n",
              " '676-329': 308,\n",
              " '128-11': 309,\n",
              " '31-3': 310,\n",
              " '81-19': 311,\n",
              " '92-2': 312,\n",
              " '1094-12': 313,\n",
              " '415-61': 314,\n",
              " '537-13': 315,\n",
              " '865-31': 316,\n",
              " '540-4': 317,\n",
              " '1146-1': 318,\n",
              " '122-1': 319,\n",
              " '605-13': 320,\n",
              " '191-2': 321,\n",
              " '1344-1': 322,\n",
              " '353-22': 323,\n",
              " '350-6': 324,\n",
              " '571-2': 325,\n",
              " '273-3': 326,\n",
              " '1995-5': 327,\n",
              " '481-8': 328,\n",
              " '163-26': 329,\n",
              " '797-26': 330,\n",
              " '1581-6': 331,\n",
              " '800-2': 332,\n",
              " '72-4': 333,\n",
              " '250-9': 334,\n",
              " '895-17': 335,\n",
              " '775-5': 336,\n",
              " '608-31': 337,\n",
              " '33-454': 338,\n",
              " '497-36': 339,\n",
              " '177-143': 340,\n",
              " '605-14': 341,\n",
              " '203-57': 342,\n",
              " '1246-5': 343,\n",
              " '118-8': 344,\n",
              " '596-1': 345,\n",
              " '50-18': 346,\n",
              " '638-5': 347,\n",
              " '91-12': 348,\n",
              " '141-5': 349,\n",
              " '354-8': 350,\n",
              " '152-20': 351,\n",
              " '1310-13': 352,\n",
              " '994-428': 353,\n",
              " '994-5': 354,\n",
              " '796-3': 355,\n",
              " '273-9': 356,\n",
              " '1112-2': 357,\n",
              " '770-39': 358,\n",
              " '162-9': 359,\n",
              " '168-8': 360,\n",
              " '57-1': 361,\n",
              " '390-26': 362,\n",
              " '345-12': 363,\n",
              " '967-1': 364,\n",
              " '1025-1': 365,\n",
              " '342-41': 366,\n",
              " '363-38': 367,\n",
              " '458-2': 368,\n",
              " '407-31': 369,\n",
              " '1092': 370,\n",
              " '172-12': 371,\n",
              " '35-1': 372,\n",
              " '1026-1': 373,\n",
              " '112-1': 374,\n",
              " '648-1': 375,\n",
              " '261-126': 376,\n",
              " '1074-2': 377,\n",
              " '129-9': 378,\n",
              " '327-84': 379,\n",
              " '583-2': 380,\n",
              " '1135-85': 381,\n",
              " '114-5': 382,\n",
              " '987-13': 383,\n",
              " '300-58': 384,\n",
              " '23-29': 385,\n",
              " '566-12': 386,\n",
              " '976-1': 387,\n",
              " '865-171': 388,\n",
              " '788-1': 389,\n",
              " '315-2': 390,\n",
              " '420-5': 391,\n",
              " '1498-3': 392,\n",
              " '447-13': 393,\n",
              " '195-9': 394,\n",
              " '568-10': 395,\n",
              " '331-4': 396,\n",
              " '419-12': 397,\n",
              " '1548-13': 398,\n",
              " '11-2': 399,\n",
              " '1144-6': 400,\n",
              " '740-33': 401,\n",
              " '109-11': 402,\n",
              " '333-4': 403,\n",
              " '812-1': 404,\n",
              " '1056-42': 405,\n",
              " '143-2': 406,\n",
              " '566-58': 407,\n",
              " '296-7': 408,\n",
              " '971': 409,\n",
              " '841-303': 410,\n",
              " '740-1': 411,\n",
              " '363-47': 412,\n",
              " '409-14': 413,\n",
              " '1068-49': 414,\n",
              " '718-11': 415,\n",
              " '60-34': 416,\n",
              " '237-33': 417,\n",
              " '596-3': 418,\n",
              " '509-7': 419,\n",
              " '793-16': 420,\n",
              " '산176-13': 421,\n",
              " '1041-9': 422,\n",
              " '452-35': 423,\n",
              " '745-57': 424,\n",
              " '1294-3': 425,\n",
              " '37-3': 426,\n",
              " '177-1': 427,\n",
              " '327-1': 428,\n",
              " '300-55': 429,\n",
              " '736-106': 430,\n",
              " '973-50': 431,\n",
              " '433-27': 432,\n",
              " '337-2': 433,\n",
              " '20-3': 434,\n",
              " '777-5': 435,\n",
              " '923-119': 436,\n",
              " '213-19': 437,\n",
              " '1-92': 438,\n",
              " '509-6': 439,\n",
              " '891-47': 440,\n",
              " '208-3': 441,\n",
              " '1261-6': 442,\n",
              " '1162-1': 443,\n",
              " '10-2': 444,\n",
              " '770': 445,\n",
              " '185-6': 446,\n",
              " '740-21': 447,\n",
              " '472-2': 448,\n",
              " '269-6': 449,\n",
              " '177-326': 450,\n",
              " '926-5': 451,\n",
              " '544-2': 452,\n",
              " '33-3': 453,\n",
              " '776-9': 454,\n",
              " '459-16': 455,\n",
              " '60-20': 456,\n",
              " '290-2': 457,\n",
              " '산26-2': 458,\n",
              " '106-9': 459,\n",
              " '1589-27': 460,\n",
              " '792-103': 461,\n",
              " '420-15': 462,\n",
              " '479-8': 463,\n",
              " '1157-1': 464,\n",
              " '750-7': 465,\n",
              " '817-29': 466,\n",
              " '70-3': 467,\n",
              " '156-2': 468,\n",
              " '717-11': 469,\n",
              " '1143-15': 470,\n",
              " '30-19': 471,\n",
              " '368-1': 472,\n",
              " '532-4': 473,\n",
              " '738-39': 474,\n",
              " '414-1': 475,\n",
              " '1550': 476,\n",
              " '792-102': 477,\n",
              " '29-3': 478,\n",
              " '320-31': 479,\n",
              " '1873-70': 480,\n",
              " '693-2': 481,\n",
              " '345-10': 482,\n",
              " '958': 483,\n",
              " '248-2': 484,\n",
              " '369-4': 485,\n",
              " '60-39': 486,\n",
              " '69-4': 487,\n",
              " '632-1': 488,\n",
              " '614-2': 489,\n",
              " '532': 490,\n",
              " '34-11': 491,\n",
              " '625-3': 492,\n",
              " '792-98': 493,\n",
              " '1465-5': 494,\n",
              " '89-38': 495,\n",
              " '458-4': 496,\n",
              " '99-15': 497,\n",
              " '627-120': 498,\n",
              " '752-10': 499,\n",
              " '744-47': 500,\n",
              " '716-23': 501,\n",
              " '166-6': 502,\n",
              " '742-58': 503,\n",
              " '497-76': 504,\n",
              " '183-10': 505,\n",
              " '1829-11': 506,\n",
              " '167-13': 507,\n",
              " '1187': 508,\n",
              " '1593-19': 509,\n",
              " '516-31': 510,\n",
              " '510-2': 511,\n",
              " '335-12': 512,\n",
              " '886': 513,\n",
              " '566-1': 514,\n",
              " '967-2': 515,\n",
              " '288-2': 516,\n",
              " '1161': 517,\n",
              " '399-5': 518,\n",
              " '1102-133': 519,\n",
              " '80-14': 520,\n",
              " '1143-8': 521,\n",
              " '1135-25': 522,\n",
              " '81-48': 523,\n",
              " '973-3': 524,\n",
              " '237-1': 525,\n",
              " '524-4': 526,\n",
              " '432-1970': 527,\n",
              " '574-82': 528,\n",
              " '771-30': 529,\n",
              " '155-13': 530,\n",
              " '857-10': 531,\n",
              " '1567-1': 532,\n",
              " '780-50': 533,\n",
              " '1165-1': 534,\n",
              " '190-19': 535,\n",
              " '741-7': 536,\n",
              " '285-3': 537,\n",
              " '248-1': 538,\n",
              " '363-46': 539,\n",
              " '81-12': 540,\n",
              " '419-20': 541,\n",
              " '342-2': 542,\n",
              " '29-104': 543,\n",
              " '477-9': 544,\n",
              " '1492-16': 545,\n",
              " '405-4': 546,\n",
              " '497-2': 547,\n",
              " '1262-8': 548,\n",
              " '669-6': 549,\n",
              " '430-30': 550,\n",
              " '178-10': 551,\n",
              " '814-4': 552,\n",
              " '12-108': 553,\n",
              " '320-54': 554,\n",
              " '793-54': 555,\n",
              " '919-2': 556,\n",
              " '418-5': 557,\n",
              " '171-51': 558,\n",
              " '552': 559,\n",
              " '993-38': 560,\n",
              " '200-27': 561,\n",
              " '351-7': 562,\n",
              " '296-44': 563,\n",
              " '896-4': 564,\n",
              " '1073-2': 565,\n",
              " '740-7': 566,\n",
              " '1836-19': 567,\n",
              " '194-4': 568,\n",
              " '158-3': 569,\n",
              " '580-10': 570,\n",
              " '174-12': 571,\n",
              " '177-8': 572,\n",
              " '172-4': 573,\n",
              " '652-45': 574,\n",
              " '416-10': 575,\n",
              " '165-18': 576,\n",
              " '1050-1': 577,\n",
              " '871-19': 578,\n",
              " '533-9': 579,\n",
              " '641-2': 580,\n",
              " '456-10': 581,\n",
              " '749-1': 582,\n",
              " '1825-2': 583,\n",
              " '1781-66': 584,\n",
              " '269-8': 585,\n",
              " '1165-11': 586,\n",
              " '1246-4': 587,\n",
              " '377-1': 588,\n",
              " '419-10': 589,\n",
              " '728-1': 590,\n",
              " '1849-6': 591,\n",
              " '744-21': 592,\n",
              " '396-8': 593,\n",
              " '44-2': 594,\n",
              " '1310-10': 595,\n",
              " '75-20': 596,\n",
              " '81-89': 597,\n",
              " '1121-7': 598,\n",
              " '458-45': 599,\n",
              " '127-27': 600,\n",
              " '1159-24': 601,\n",
              " '800-29': 602,\n",
              " '676-145': 603,\n",
              " '60-42': 604,\n",
              " '797-19': 605,\n",
              " '1093-10': 606,\n",
              " '302-3': 607,\n",
              " '159-1': 608,\n",
              " '146-8': 609,\n",
              " '774-3': 610,\n",
              " '759-11': 611,\n",
              " '75': 612,\n",
              " '972-10': 613,\n",
              " '484-12': 614,\n",
              " '740-24': 615,\n",
              " '61-15': 616,\n",
              " '802': 617,\n",
              " '791-21': 618,\n",
              " '1123': 619,\n",
              " '363-40': 620,\n",
              " '11-18': 621,\n",
              " '506-5': 622,\n",
              " '382-85': 623,\n",
              " '210-1': 624,\n",
              " '344-2': 625,\n",
              " '384-18': 626,\n",
              " '328-3': 627,\n",
              " '717-39': 628,\n",
              " '770-23': 629,\n",
              " '167-7': 630,\n",
              " '799-11': 631,\n",
              " '1-7': 632,\n",
              " '806-59': 633,\n",
              " '117-35': 634,\n",
              " '705-17': 635,\n",
              " '418-1': 636,\n",
              " '1247-10': 637,\n",
              " '10-14': 638,\n",
              " '17-3': 639,\n",
              " '492-10': 640,\n",
              " '121-60': 641,\n",
              " '1284-7': 642,\n",
              " '2121-2': 643,\n",
              " '989-2': 644,\n",
              " '178-3': 645,\n",
              " '159-14': 646,\n",
              " '978-7': 647,\n",
              " '617-1': 648,\n",
              " '186-9': 649,\n",
              " '489-16': 650,\n",
              " '744-45': 651,\n",
              " '878-1': 652,\n",
              " '254-40': 653,\n",
              " '506-8': 654,\n",
              " '394-2': 655,\n",
              " '54-3': 656,\n",
              " '568-6': 657,\n",
              " '779-13': 658,\n",
              " '297-1': 659,\n",
              " '657-50': 660,\n",
              " '15-92': 661,\n",
              " '807-29': 662,\n",
              " '23-4': 663,\n",
              " '917-14': 664,\n",
              " '969-2': 665,\n",
              " '1370-1': 666,\n",
              " '250-4': 667,\n",
              " '330-4': 668,\n",
              " '881-5': 669,\n",
              " '495-2': 670,\n",
              " '332-6': 671,\n",
              " '888-52': 672,\n",
              " '1645-2': 673,\n",
              " '327-3': 674,\n",
              " '384-26': 675,\n",
              " '1386-25': 676,\n",
              " '1386-2': 677,\n",
              " '797-5': 678,\n",
              " '1476-90': 679,\n",
              " '300-89': 680,\n",
              " '749-96': 681,\n",
              " '365-11': 682,\n",
              " '215-22': 683,\n",
              " '1147-3': 684,\n",
              " '705-2': 685,\n",
              " '882-4': 686,\n",
              " '1261-14': 687,\n",
              " '156-15': 688,\n",
              " '1425-4': 689,\n",
              " '585-19': 690,\n",
              " '1242-6': 691,\n",
              " '120-296': 692,\n",
              " '252-7': 693,\n",
              " '613-1': 694,\n",
              " '186-44': 695,\n",
              " '469-20': 696,\n",
              " '1248-7': 697,\n",
              " '655-36': 698,\n",
              " '793-24': 699,\n",
              " '969-20': 700,\n",
              " '799-18': 701,\n",
              " '184-8': 702,\n",
              " '35-8': 703,\n",
              " '20-7': 704,\n",
              " '97-15': 705,\n",
              " '1341': 706,\n",
              " '1832': 707,\n",
              " '740-6': 708,\n",
              " '343-2': 709,\n",
              " '725-5': 710,\n",
              " '481-6': 711,\n",
              " '297-13': 712,\n",
              " '743-18': 713,\n",
              " '806-14': 714,\n",
              " '807-16': 715,\n",
              " '486-41': 716,\n",
              " '808-76': 717,\n",
              " '882-61': 718,\n",
              " '539-7': 719,\n",
              " '3-7': 720,\n",
              " '1520-8': 721,\n",
              " '1448-3': 722,\n",
              " '362-3': 723,\n",
              " '510-11': 724,\n",
              " '1243-9': 725,\n",
              " '417-21': 726,\n",
              " '36-7': 727,\n",
              " '792-54': 728,\n",
              " '859-10': 729,\n",
              " '556-1': 730,\n",
              " '1056-30': 731,\n",
              " '925-11': 732,\n",
              " '898': 733,\n",
              " '873-62': 734,\n",
              " '807-17': 735,\n",
              " '1121-13': 736,\n",
              " '16-11': 737,\n",
              " '213-1': 738,\n",
              " '1030-3': 739,\n",
              " '620-5': 740,\n",
              " '133-8': 741,\n",
              " '870-22': 742,\n",
              " '466-5': 743,\n",
              " '718-4': 744,\n",
              " '1444-4': 745,\n",
              " '13-4': 746,\n",
              " '744-2': 747,\n",
              " '713-3': 748,\n",
              " '190-10': 749,\n",
              " '188-41': 750,\n",
              " '540-51': 751,\n",
              " '25-7': 752,\n",
              " '807-26': 753,\n",
              " '925-26': 754,\n",
              " '1392': 755,\n",
              " '1291-1460': 756,\n",
              " '198': 757,\n",
              " '752-6': 758,\n",
              " '100-9': 759,\n",
              " '29-101': 760,\n",
              " '951-4': 761,\n",
              " '746-17': 762,\n",
              " '357-1': 763,\n",
              " '1291-1414': 764,\n",
              " '740-5': 765,\n",
              " '206-1': 766,\n",
              " '894-1': 767,\n",
              " '191-5': 768,\n",
              " '1293-2': 769,\n",
              " '29-105': 770,\n",
              " '1026-5': 771,\n",
              " '832-1': 772,\n",
              " '543-9': 773,\n",
              " '39-36': 774,\n",
              " '807-19': 775,\n",
              " '112-2': 776,\n",
              " '1310-19': 777,\n",
              " '491-2': 778,\n",
              " '141-3': 779,\n",
              " '792-20': 780,\n",
              " '84-7': 781,\n",
              " '497-30': 782,\n",
              " '749-30': 783,\n",
              " '177-23': 784,\n",
              " '794-17': 785,\n",
              " '899-12': 786,\n",
              " '1197-8': 787,\n",
              " '9-45': 788,\n",
              " '563-50': 789,\n",
              " '661-1': 790,\n",
              " '870-19': 791,\n",
              " '743-8': 792,\n",
              " '396-35': 793,\n",
              " '281-52': 794,\n",
              " '113-11': 795,\n",
              " '981-9': 796,\n",
              " '387-9': 797,\n",
              " '1472-14': 798,\n",
              " '33-433': 799,\n",
              " '1256-5': 800,\n",
              " '824-152': 801,\n",
              " '291-4': 802,\n",
              " '797-8': 803,\n",
              " '752-17': 804,\n",
              " '497-38': 805,\n",
              " '805-4': 806,\n",
              " '823-1': 807,\n",
              " '1045-2': 808,\n",
              " '634-1': 809,\n",
              " '1331-19': 810,\n",
              " '780-31': 811,\n",
              " '780-86': 812,\n",
              " '748-14': 813,\n",
              " '614-9': 814,\n",
              " '194-11': 815,\n",
              " '767': 816,\n",
              " '산52-6': 817,\n",
              " '328-1': 818,\n",
              " '1058-3': 819,\n",
              " '378-3': 820,\n",
              " '773-1': 821,\n",
              " '1051-2': 822,\n",
              " '743-10': 823,\n",
              " '566-7': 824,\n",
              " '86-73': 825,\n",
              " '9-50': 826,\n",
              " '42-2': 827,\n",
              " '95-5': 828,\n",
              " '288-1': 829,\n",
              " '1063-2': 830,\n",
              " '392-4': 831,\n",
              " '967-7': 832,\n",
              " '1126-7': 833,\n",
              " '465-1': 834,\n",
              " '791-62': 835,\n",
              " '414-22': 836,\n",
              " '1144-2': 837,\n",
              " '740-107': 838,\n",
              " '484-60': 839,\n",
              " '959-2': 840,\n",
              " '480-9': 841,\n",
              " '1231-12': 842,\n",
              " '1224': 843,\n",
              " '508-6': 844,\n",
              " '산32-1': 845,\n",
              " '86-3': 846,\n",
              " '519-6': 847,\n",
              " '743-5': 848,\n",
              " '225-2': 849,\n",
              " '145-7': 850,\n",
              " '1087-2': 851,\n",
              " '530-1': 852,\n",
              " '386-13': 853,\n",
              " '791-19': 854,\n",
              " '372-5': 855,\n",
              " '148-6': 856,\n",
              " '414-4': 857,\n",
              " '233-1': 858,\n",
              " '53-1': 859,\n",
              " '1995-7': 860,\n",
              " '60-14': 861,\n",
              " '492-5': 862,\n",
              " '951-22': 863,\n",
              " '458-53': 864,\n",
              " '876-13': 865,\n",
              " '487-6': 866,\n",
              " '2140-4': 867,\n",
              " '5-56': 868,\n",
              " '530-8': 869,\n",
              " '238-233': 870,\n",
              " '792-62': 871,\n",
              " '51-19': 872,\n",
              " '1182-2': 873,\n",
              " '643-1': 874,\n",
              " '1060-3': 875,\n",
              " '183-23': 876,\n",
              " '189-2172': 877,\n",
              " '521-8': 878,\n",
              " '497-44': 879,\n",
              " '1289-32': 880,\n",
              " '870-243': 881,\n",
              " '857-5': 882,\n",
              " '870-311': 883,\n",
              " '1165-14': 884,\n",
              " '601-3': 885,\n",
              " '374-11': 886,\n",
              " '335-13': 887,\n",
              " '450-25': 888,\n",
              " '64-38': 889,\n",
              " '1293-10': 890,\n",
              " '181-7': 891,\n",
              " '13-28': 892,\n",
              " '494-3': 893,\n",
              " '970-10': 894,\n",
              " '456-17': 895,\n",
              " '281-11': 896,\n",
              " '409-13': 897,\n",
              " '592-9': 898,\n",
              " '117-2': 899,\n",
              " '1111-3': 900,\n",
              " '16': 901,\n",
              " '50-4': 902,\n",
              " '408-1': 903,\n",
              " '1995-6': 904,\n",
              " '457-2': 905,\n",
              " '159-2': 906,\n",
              " '203-25': 907,\n",
              " '1284-1': 908,\n",
              " '201-11': 909,\n",
              " '4-14': 910,\n",
              " '1486-17': 911,\n",
              " '1495-22': 912,\n",
              " '1375-1': 913,\n",
              " '921': 914,\n",
              " '1025-6': 915,\n",
              " '770-52': 916,\n",
              " '799-7': 917,\n",
              " '401-8': 918,\n",
              " '2121-12': 919,\n",
              " '793-44': 920,\n",
              " '401-7': 921,\n",
              " '295-3': 922,\n",
              " '744-48': 923,\n",
              " '1373-61': 924,\n",
              " '893-3': 925,\n",
              " '191-236': 926,\n",
              " '338-3': 927,\n",
              " '44-5': 928,\n",
              " '59-23': 929,\n",
              " '1643-2': 930,\n",
              " '458-12': 931,\n",
              " '1995-3': 932,\n",
              " '743-31': 933,\n",
              " '251-3': 934,\n",
              " '419-15': 935,\n",
              " '305-40': 936,\n",
              " '1182-1': 937,\n",
              " '634-7': 938,\n",
              " '287': 939,\n",
              " '781-6': 940,\n",
              " '261-22': 941,\n",
              " '1405-16': 942,\n",
              " '36-13': 943,\n",
              " '494-34': 944,\n",
              " '488-18': 945,\n",
              " '645-1': 946,\n",
              " '1288-1': 947,\n",
              " '244-6': 948,\n",
              " '479-2': 949,\n",
              " '798-22': 950,\n",
              " '742-64': 951,\n",
              " '51-5': 952,\n",
              " '619-7': 953,\n",
              " '1121-2': 954,\n",
              " '29-102': 955,\n",
              " '495-9': 956,\n",
              " '716-6': 957,\n",
              " '740-29': 958,\n",
              " '517-2': 959,\n",
              " '397-68': 960,\n",
              " '797-25': 961,\n",
              " '173-16': 962,\n",
              " '335-9': 963,\n",
              " '380-7': 964,\n",
              " '749-46': 965,\n",
              " '761-1': 966,\n",
              " '1116-1': 967,\n",
              " '616-7': 968,\n",
              " '920-16': 969,\n",
              " '807-23': 970,\n",
              " '208-6': 971,\n",
              " '579-9': 972,\n",
              " '74-8': 973,\n",
              " '350-72': 974,\n",
              " '500-5': 975,\n",
              " '793-46': 976,\n",
              " '1252-125': 977,\n",
              " '1043-38': 978,\n",
              " '1604-8': 979,\n",
              " '432-2006': 980,\n",
              " '304-151': 981,\n",
              " '781-26': 982,\n",
              " '5-89': 983,\n",
              " '160-7': 984,\n",
              " '112-8': 985,\n",
              " '469-6': 986,\n",
              " '264-1': 987,\n",
              " '237-2': 988,\n",
              " '491-3': 989,\n",
              " '87-2': 990,\n",
              " '666-4': 991,\n",
              " '636-1': 992,\n",
              " '1486-16': 993,\n",
              " '1057-14': 994,\n",
              " '185-29': 995,\n",
              " '346-1': 996,\n",
              " '596-17': 997,\n",
              " '1046-4': 998,\n",
              " '513-2': 999,\n",
              " ...}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ff3s4yCBok5e"
      },
      "source": [
        "# apt : 아파트\n",
        "- 아파트의 경우는 중복데이터가 너무 많기에 그룹화로 라벨링을 시키면 중복데이터가 너무 많아지므로 하지않는다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AoQNscM6oSu0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "48e5b493-8e72-4093-8d85-b70362e4ceb5"
      },
      "source": [
        "len(train['apt'].unique())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10440"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YF-zFtpqoTIR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 202
        },
        "outputId": "969fc5cc-6f7a-46ca-aab4-e5cfa7719e9b"
      },
      "source": [
        "group_apt = train[['transaction_real_price','apt']].groupby('apt').mean().sort_values('transaction_real_price').reset_index()\n",
        "group_apt.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apt</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>좌천시민(737-1)</td>\n",
              "      <td>7.494417</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>좌천시민(741-1)</td>\n",
              "      <td>7.498174</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>수정(1181-4)</td>\n",
              "      <td>7.585491</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>수정(1175-1)</td>\n",
              "      <td>7.626830</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>수정(1186-1)</td>\n",
              "      <td>7.685623</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "           apt  transaction_real_price\n",
              "0  좌천시민(737-1)                7.494417\n",
              "1  좌천시민(741-1)                7.498174\n",
              "2   수정(1181-4)                7.585491\n",
              "3   수정(1175-1)                7.626830\n",
              "4   수정(1186-1)                7.685623"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jqQcWdRFpTia"
      },
      "source": [
        "# transaction_date : 거래 일자"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CKr__90VpjTx",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "348fda78-5c16-4775-a98e-688d89858974"
      },
      "source": [
        "train['transaction_date']"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0          21~31\n",
              "1           1~10\n",
              "2           1~10\n",
              "3          11~20\n",
              "4          21~31\n",
              "           ...  \n",
              "1216548    11~20\n",
              "1216549    21~30\n",
              "1216550    21~30\n",
              "1216551    21~30\n",
              "1216552    21~30\n",
              "Name: transaction_date, Length: 1216553, dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7Yy_biHdqYYA"
      },
      "source": [
        "# 처음 거래시작한 날과 거래가 끝난 날의 차이를 거래길이를 나타내는 컬럼을 만들어줌\n",
        "train['day_diff'] = train['transaction_date'].str.extract('(~\\d+)')[0].str[1:].astype(int) - train['transaction_date'].str.extract('(\\d+~)')[0].str[:-1].astype(int)\n",
        "test['day_diff'] = test['transaction_date'].str.extract('(~\\d+)')[0].str[1:].astype(int) - test['transaction_date'].str.extract('(\\d+~)')[0].str[:-1].astype(int)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CTJRsw4irDgb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 202
        },
        "outputId": "f71d5d2c-02dd-4bb5-bbba-c2886c0af715"
      },
      "source": [
        "group_date = train[['transaction_real_price','transaction_date']].groupby('transaction_date').mean().sort_values('transaction_real_price').reset_index()\n",
        "group_date.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>transaction_date</th>\n",
              "      <th>transaction_real_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>21~29</td>\n",
              "      <td>10.176453</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>21~28</td>\n",
              "      <td>10.286294</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1~10</td>\n",
              "      <td>10.293390</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>11~20</td>\n",
              "      <td>10.306731</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>21~31</td>\n",
              "      <td>10.318796</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  transaction_date  transaction_real_price\n",
              "0            21~29               10.176453\n",
              "1            21~28               10.286294\n",
              "2             1~10               10.293390\n",
              "3            11~20               10.306731\n",
              "4            21~31               10.318796"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XUMjENjEsdG6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "49600600-3ae6-47e4-954e-6df06df558d4"
      },
      "source": [
        "date_num = {}\n",
        "for i in range(len(group_date)):\n",
        "    date = group_date['transaction_date'].iloc[i]\n",
        "    date_num[date] = i\n",
        "    \n",
        "date_num"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'11~20': 3, '1~10': 2, '21~28': 1, '21~29': 0, '21~30': 5, '21~31': 4}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gcAugtidsdYd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 378
        },
        "outputId": "583808d4-5a67-4567-ba76-07ab5c5270dc"
      },
      "source": [
        "train = train.replace({'transaction_date' : date_num})\n",
        "test = test.replace({'transaction_date' : date_num})\n",
        "train.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>jibun</th>\n",
              "      <th>apt</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_year_month</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "      <th>year</th>\n",
              "      <th>month</th>\n",
              "      <th>hangang</th>\n",
              "      <th>day_diff</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>7622</td>\n",
              "      <td>0</td>\n",
              "      <td>351</td>\n",
              "      <td>6-13</td>\n",
              "      <td>신현(101동)</td>\n",
              "      <td>84.82</td>\n",
              "      <td>2002</td>\n",
              "      <td>200801</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>10.532123</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5399</td>\n",
              "      <td>0</td>\n",
              "      <td>414</td>\n",
              "      <td>142</td>\n",
              "      <td>사직파크맨션</td>\n",
              "      <td>99.17</td>\n",
              "      <td>1973</td>\n",
              "      <td>200801</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>9.903538</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3578</td>\n",
              "      <td>0</td>\n",
              "      <td>414</td>\n",
              "      <td>174-1</td>\n",
              "      <td>두레엘리시안</td>\n",
              "      <td>84.74</td>\n",
              "      <td>2007</td>\n",
              "      <td>200801</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>10.558439</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>10957</td>\n",
              "      <td>0</td>\n",
              "      <td>462</td>\n",
              "      <td>95</td>\n",
              "      <td>파크팰리스</td>\n",
              "      <td>146.39</td>\n",
              "      <td>2003</td>\n",
              "      <td>200801</td>\n",
              "      <td>3</td>\n",
              "      <td>15</td>\n",
              "      <td>11.678448</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>10639</td>\n",
              "      <td>0</td>\n",
              "      <td>462</td>\n",
              "      <td>110-15</td>\n",
              "      <td>킹스매너</td>\n",
              "      <td>194.43</td>\n",
              "      <td>2004</td>\n",
              "      <td>200801</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>11.695255</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   apartment_id  city  dong   jibun  ...  year  month  hangang  day_diff\n",
              "0          7622     0   351    6-13  ...  2008      1        0        10\n",
              "1          5399     0   414     142  ...  2008      1        0         9\n",
              "2          3578     0   414   174-1  ...  2008      1        0         9\n",
              "3         10957     0   462      95  ...  2008      1        0         9\n",
              "4         10639     0   462  110-15  ...  2008      1        0        10\n",
              "\n",
              "[5 rows x 15 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tMwy2cB4sm4y"
      },
      "source": [
        "# Preprocessing\n",
        "- floor 같은 경우 -값이 존재하기 때문에 log를 취하기 전에 각 값에 +5를 하면서 모든 값을 양수로 만들어줍니다.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ox0H2Mh1sv6i",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4165906b-7f28-4a43-cb46-feea7a5b255c"
      },
      "source": [
        "train['floor'].value_counts() # -1 부터 -4 까지의 값이 있기에 5를 더해줘 모든 값을 양수로 바꿔줌"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              " 5     88676\n",
              " 4     88399\n",
              " 3     85549\n",
              " 2     82589\n",
              " 1     74927\n",
              " 6     74236\n",
              " 7     70113\n",
              " 8     66771\n",
              " 9     65272\n",
              " 10    63494\n",
              " 11    60065\n",
              " 12    57746\n",
              " 13    52240\n",
              " 14    49041\n",
              " 15    45782\n",
              " 16    27472\n",
              " 17    26231\n",
              " 18    24446\n",
              " 19    21476\n",
              " 20    18859\n",
              " 21    13654\n",
              " 22    12080\n",
              " 23    10377\n",
              " 24     8730\n",
              " 25     7098\n",
              " 26     3052\n",
              " 27     2719\n",
              " 28     2212\n",
              " 29     1792\n",
              " 30     1461\n",
              " 31     1148\n",
              " 32     1045\n",
              " 33      984\n",
              " 34      891\n",
              " 35      791\n",
              " 36      629\n",
              " 37      537\n",
              " 38      447\n",
              " 39      388\n",
              " 40      338\n",
              " 41      289\n",
              " 42      287\n",
              " 43      251\n",
              " 44      208\n",
              " 45      204\n",
              " 46      198\n",
              " 47      164\n",
              " 48      147\n",
              "-1       140\n",
              " 49      113\n",
              " 50       92\n",
              " 51       88\n",
              " 52       52\n",
              " 54       49\n",
              " 58       43\n",
              " 57       41\n",
              " 53       39\n",
              " 55       34\n",
              " 60       28\n",
              " 64       26\n",
              " 59       23\n",
              " 70       23\n",
              " 62       22\n",
              " 66       21\n",
              " 56       20\n",
              " 65       20\n",
              " 68       19\n",
              " 61       18\n",
              " 69       17\n",
              "-2        17\n",
              " 63       17\n",
              " 67       13\n",
              " 72       11\n",
              " 74       10\n",
              " 73        9\n",
              " 71        7\n",
              " 75        7\n",
              "-3         7\n",
              " 77        5\n",
              " 78        4\n",
              " 80        4\n",
              "-4         4\n",
              " 76        3\n",
              " 79        2\n",
              "Name: floor, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aBl1TCbas8LC",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "outputId": "976cfe89-feb6-43b6-cc4a-be6ab540fb45"
      },
      "source": [
        "train['floor'] = np.log(train['floor']+5) # log를 취하기전 5를 더해줌\n",
        "test['floor'] = np.log(test['floor']+5) # 동일하게\n",
        "f, ax = plt.subplots(figsize = (8,6))\n",
        "sns.distplot(train['floor'])\n",
        "print(\"Skewness: %f, Kurtosis: %f\" %  (train['floor'].skew(),train['floor'].kurt()))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Skewness: 0.089636, Kurtosis: -0.640885\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GBYw5Vq1tI6i"
      },
      "source": [
        "drop_columns = ['apt', 'jibun', 'transaction_year_month']\n",
        "\n",
        "train = train.drop(drop_columns, axis=1)\n",
        "test = test.drop(drop_columns, axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P-ClBpOxtNQM",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 222
        },
        "outputId": "d56192aa-7403-4d7a-9e2f-bfaf202c4a87"
      },
      "source": [
        "train.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>apartment_id</th>\n",
              "      <th>city</th>\n",
              "      <th>dong</th>\n",
              "      <th>exclusive_use_area</th>\n",
              "      <th>year_of_completion</th>\n",
              "      <th>transaction_date</th>\n",
              "      <th>floor</th>\n",
              "      <th>transaction_real_price</th>\n",
              "      <th>year</th>\n",
              "      <th>month</th>\n",
              "      <th>hangang</th>\n",
              "      <th>day_diff</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>7622</td>\n",
              "      <td>0</td>\n",
              "      <td>351</td>\n",
              "      <td>84.82</td>\n",
              "      <td>2002</td>\n",
              "      <td>4</td>\n",
              "      <td>1.945910</td>\n",
              "      <td>10.532123</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5399</td>\n",
              "      <td>0</td>\n",
              "      <td>414</td>\n",
              "      <td>99.17</td>\n",
              "      <td>1973</td>\n",
              "      <td>2</td>\n",
              "      <td>2.397895</td>\n",
              "      <td>9.903538</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3578</td>\n",
              "      <td>0</td>\n",
              "      <td>414</td>\n",
              "      <td>84.74</td>\n",
              "      <td>2007</td>\n",
              "      <td>2</td>\n",
              "      <td>2.397895</td>\n",
              "      <td>10.558439</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>10957</td>\n",
              "      <td>0</td>\n",
              "      <td>462</td>\n",
              "      <td>146.39</td>\n",
              "      <td>2003</td>\n",
              "      <td>3</td>\n",
              "      <td>2.995732</td>\n",
              "      <td>11.678448</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>10639</td>\n",
              "      <td>0</td>\n",
              "      <td>462</td>\n",
              "      <td>194.43</td>\n",
              "      <td>2004</td>\n",
              "      <td>4</td>\n",
              "      <td>2.079442</td>\n",
              "      <td>11.695255</td>\n",
              "      <td>2008</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   apartment_id  city  dong  exclusive_use_area  ...  year  month  hangang  day_diff\n",
              "0          7622     0   351               84.82  ...  2008      1        0        10\n",
              "1          5399     0   414               99.17  ...  2008      1        0         9\n",
              "2          3578     0   414               84.74  ...  2008      1        0         9\n",
              "3         10957     0   462              146.39  ...  2008      1        0         9\n",
              "4         10639     0   462              194.43  ...  2008      1        0        10\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B9MoauEQtSCe"
      },
      "source": [
        "# Feature Engineering"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1P4BFNMau7M_"
      },
      "source": [
        "- 재건축 유뮤를 판단하는 컬럼 생성"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Is-4XPvctRcC"
      },
      "source": [
        "all_data = [train,test]\n",
        "for dataset in all_data:\n",
        "  dataset['age'] = dataset['year'] - dataset['year_of_completion'] # 거래년도 - 설립연도\n",
        "  dataset['is_rebuild'] = (dataset['age']>=30).astype(int) # 30년이상인것은 일반적으로 재건축건물"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DGaxoEPMvYHN"
      },
      "source": [
        "- 정규분포모형을 가질 수 있도록 왜도가 1 이상 & 첨도가 3 이상인것은 조정 할 필요가 있다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SHSslBNCvJ8L",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a8007374-76ff-42f5-eac3-5708bd5d25e3"
      },
      "source": [
        "train_columns = []\n",
        "for column in train.columns[:]:\n",
        "    if train[column].skew() >= 1:\n",
        "        print(\"%s -> Skewness: %f, Kurtosis: %f\" %  (column,train[column].skew(),train[column].kurt()))\n",
        "        train_columns.append(column)\n",
        "    elif train[column].kurt() >= 3:\n",
        "        print(\"%s -> Skewness: %f, Kurtosis: %f\" %  (column,train[column].skew(),train[column].kurt()))\n",
        "        train_columns.append(column)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "exclusive_use_area -> Skewness: 1.227509, Kurtosis: 3.100517\n",
            "hangang -> Skewness: 6.358349, Kurtosis: 38.428662\n",
            "day_diff -> Skewness: -0.589751, Kurtosis: 4.573129\n",
            "is_rebuild -> Skewness: 3.324832, Kurtosis: 9.054522\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GZxFSgcewNIk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "822c5e4c-80ea-4c00-acab-279d9d4d243e"
      },
      "source": [
        "#정규분포모형을 가질 수 있도록 첨도와 왜도를 조정\n",
        "#조정하는 방법에는 square root, quarter root, log 등이 있다.\n",
        "#log에서 0의 값이 들어왔을 때 무한으로 가는 것을 방지하도록 1 더해주는 log1p를 사용\n",
        "\n",
        "for column in train_columns :\n",
        "    train[column] = np.log1p(train[column])\n",
        "    test[column] = np.log1p(test[column])\n",
        "    print(\"%s -> Skewness: %f, Kurtosis: %f\" %  (column,train[column].skew(),train[column].kurt()))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "exclusive_use_area -> Skewness: -0.438156, Kurtosis: 1.744119\n",
            "hangang -> Skewness: 6.358349, Kurtosis: 38.428662\n",
            "day_diff -> Skewness: -1.128137, Kurtosis: 6.788569\n",
            "is_rebuild -> Skewness: 3.324832, Kurtosis: 9.054522\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xw7GWciUweVV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 497
        },
        "outputId": "660ccd31-3712-4a26-cf9e-4c1d2dde8b23"
      },
      "source": [
        "#상관관계 확인\n",
        "k=train.shape[1] #히트맵 변수 갯수\n",
        "corrmat = train.corr() #변수간의 상관관계\n",
        "cols = corrmat.nlargest(k, 'transaction_real_price')['transaction_real_price'].index #price기준으로 제일 큰순서대로 20개를 뽑아냄\n",
        "cm = np.corrcoef(train[cols].values.T)\n",
        "f, ax = plt.subplots(figsize=(20, 6))\n",
        "sns.heatmap(data = cm, annot=True, square=True, fmt = '.2f', linewidths=.5, cmap='Reds',yticklabels = cols.values, xticklabels = cols.values)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efbdfbbac50>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 52
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CZ7qIDUsjXdo"
      },
      "source": [
        "# LabelEncoder 라벨링\n",
        "for col in train.columns:\n",
        "  if train[col].dtype == \"object\":\n",
        "    label = LabelEncoder()\n",
        "    train[col] = label.fit_transform(train[col])\n",
        "\n",
        "for col in test.columns:\n",
        "  if test[col].dtype == \"object\":\n",
        "    label = LabelEncoder()\n",
        "    test[col] = label.fit_transform(test[col])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tcn9-8iHJbjK"
      },
      "source": [
        "target = train['transaction_real_price']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lstayZB2JwZm"
      },
      "source": [
        "train = train.drop('transaction_real_price',axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MWuNVGZdyH7f",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ed3a26d4-5f12-4b35-fc6e-0859665503ec"
      },
      "source": [
        "train.shape, test.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((1216553, 13), (5463, 13))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gJ-v-O0pidJE"
      },
      "source": [
        "# Modeling"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LtXpGOm8ih_r"
      },
      "source": [
        "from sklearn.linear_model import ElasticNet, Lasso\n",
        "from sklearn.ensemble import GradientBoostingRegressor, RandomForestRegressor\n",
        "from sklearn.kernel_ridge import KernelRidge\n",
        "from sklearn.pipeline import make_pipeline\n",
        "from sklearn.preprocessing import RobustScaler\n",
        "from sklearn.model_selection import KFold, cross_val_score, train_test_split\n",
        "from sklearn.metrics import mean_squared_error\n",
        "import xgboost as xgb\n",
        "import lightgbm as lgb"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "36P3OsZSjxFh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d59b9284-6a25-4c58-a8be-3584dc38ed6e"
      },
      "source": [
        "pip install catboost"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting catboost\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/52/39/128fff65072c8327371e3c594f3c826d29c85b21cb6485980353b168e0e4/catboost-0.24.2-cp36-none-manylinux1_x86_64.whl (66.1MB)\n",
            "\u001b[K     |████████████████████████████████| 66.2MB 56kB/s \n",
            "\u001b[?25hRequirement already satisfied: graphviz in /usr/local/lib/python3.6/dist-packages (from catboost) (0.10.1)\n",
            "Requirement already satisfied: pandas>=0.24.0 in /usr/local/lib/python3.6/dist-packages (from catboost) (1.1.4)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from catboost) (1.4.1)\n",
            "Requirement already satisfied: plotly in /usr/local/lib/python3.6/dist-packages (from catboost) (4.4.1)\n",
            "Requirement already satisfied: numpy>=1.16.0 in /usr/local/lib/python3.6/dist-packages (from catboost) (1.18.5)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from catboost) (1.15.0)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.6/dist-packages (from catboost) (3.2.2)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.24.0->catboost) (2.8.1)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.24.0->catboost) (2018.9)\n",
            "Requirement already satisfied: retrying>=1.3.3 in /usr/local/lib/python3.6/dist-packages (from plotly->catboost) (1.3.3)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->catboost) (1.3.1)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib->catboost) (0.10.0)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib->catboost) (2.4.7)\n",
            "Installing collected packages: catboost\n",
            "Successfully installed catboost-0.24.2\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PDEyt-AWj0VH"
      },
      "source": [
        "from catboost import CatBoostRegressor\n",
        "from catboost import CatBoostClassifier"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J6lGjD7Uu9O3"
      },
      "source": [
        "#cross validation score\n",
        "n_folds = 2\n",
        "\n",
        "def cv_score(models):\n",
        "    kfold = KFold(n_splits=n_folds, shuffle=True ,random_state=42).get_n_splits(train.values)\n",
        "    for m in models:\n",
        "        cvs = np.mean(cross_val_score(m['model'], train.values, target, cv=kfold))\n",
        "        rmse = np.mean(np.sqrt(-cross_val_score(m['model'], train.values, np.expm1(target), scoring = \"neg_mean_squared_error\", cv = kfold)))\n",
        "        print(\"Model {} CV score : {:.4f}\".format(m['name'], cvs))\n",
        "        print(\"RMSE : {:.4f}\".format(rmse))\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NNfQfv0Sio9f"
      },
      "source": [
        "lasso = make_pipeline(RobustScaler(), Lasso(alpha = 0.0005, random_state=42))\n",
        "ENet = make_pipeline(RobustScaler(), ElasticNet(alpha=0.0005, l1_ratio=.9, random_state=42))\n",
        "gboost = GradientBoostingRegressor(random_state=42)\n",
        "forest = RandomForestRegressor(n_estimators = 100, n_jobs = -1, random_state=42)\n",
        "xgboost = xgb.XGBRegressor(random_state=42)\n",
        "lightgbm = lgb.LGBMRegressor(random_state=42, num_leaves = 100, min_data_in_leaf = 15, max_depth=6,\n",
        "                            learning_rate = 0.1, min_child_samples = 30, feature_fraction=0.9, bagging_freq= 1,\n",
        "                            bagging_fraction = 0.9, bagging_seed = 11, lambda_l1 = 0.1, verbosity = -1 )\n",
        "catboost = CatBoostClassifier(bootstrap_type='Bernoulli', random_seed = 42, n_estimators = 100, learning_rate = 0.1, subsample = 0.8)\n",
        "\n",
        "models = [{'model': gboost, 'name':'GradientBoosting'}, {'model': xgboost, 'name':'XGBoost'},\n",
        "          {'model': lightgbm, 'name':'LightGBM'}, {'model' : lasso, 'name' : 'LASSO Regression'}, \n",
        "          {'model' : ENet, 'name' : 'Elastic Net Regression'}, {'model' : forest, 'name' : 'RandomForset'}]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9NbtigEiisZA",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "27921823-ffb6-4490-bcd3-54a77f7d3618"
      },
      "source": [
        "start = timeit.default_timer() # start\n",
        "cv_score(models) # 총 모델들 교차검증\n",
        "stop = timeit.default_timer() # stop\n",
        "print('불러오는데 걸린 시간 : {}초'.format(stop - start)) # 모델 작동 걸린 시간(stop-start) 확인"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model GradientBoosting CV score : 0.5817\n",
            "RMSE : 15799.0451\n",
            "[05:02:25] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[05:03:00] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[05:03:35] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "[05:04:10] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "Model XGBoost CV score : 0.5947\n",
            "RMSE : 15868.0485\n",
            "Model LightGBM CV score : 0.6131\n",
            "RMSE : 14330.6681\n",
            "Model LASSO Regression CV score : 0.6842\n",
            "RMSE : 19852.5360\n",
            "Model Elastic Net Regression CV score : 0.6841\n",
            "RMSE : 19848.9725\n",
            "Model RandomForset CV score : 0.5636\n",
            "RMSE : 15188.2729\n",
            "불러오는데 걸린 시간 : 1876.740792484초\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0agJVYVUjDGa"
      },
      "source": [
        "- xgboost, lightgbm, randomforest catboost총 4개의 모델을 stacking하여 활용"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wtXfLSY7iuLc"
      },
      "source": [
        "#여러개의 모델로 만들어진 predict 데이터들의 평균을 구한다.\n",
        "\n",
        "models = [{'model':xgboost, 'name':'XGBoost'},\n",
        "          {'model':lightgbm, 'name':'LightGBM'},\n",
        "         {'model':forest, 'name' : 'RandomForest'}]\n",
        "\n",
        "def AveragingBlending(models, x, y, sub_x):\n",
        "    for m in models : \n",
        "        m['model'].fit(x.values, y)\n",
        "    \n",
        "    predictions = np.column_stack([m['model'].predict(sub_x.values) for m in models])\n",
        "    return predictions"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z4_hvQkIiwdT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a1cb7bca-4846-4055-f9fc-572e55a2815c"
      },
      "source": [
        "start = timeit.default_timer()\n",
        "\n",
        "y_test_pred = AveragingBlending(models, train, target, test)\n",
        "y_test_pred = (y_test_pred[:, 0]*0.05 + y_test_pred[:, 1]*0.1 + y_test_pred[:, 2]*0.85)\n",
        "predictions = y_test_pred\n",
        "\n",
        "stop = timeit.default_timer()\n",
        "print('불러오는데 걸린 시간 : {}초'.format(stop - start))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[05:26:20] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "불러오는데 걸린 시간 : 641.3480590150002초\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ncZ-CIqzJW7q"
      },
      "source": [
        "sub = pd.read_csv('submission.csv')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0P50Zsdvy6Ur"
      },
      "source": [
        "sub['transaction_real_price'] = np.expm1(predictions)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M4OUgkWHy8Tb"
      },
      "source": [
        "sub.to_csv('project_submission.csv', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IUXgMrmpmtHm"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}