Used Car Price Prediction/2021-09-18_중고 자동차 가격 Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NqqLNJZjfi8U"
   },
   "source": [
    "#  중고 자동차 가격 예측\n",
    "----------\n",
    "    \n",
    "    - vehicles.csv\n",
    "    id : 중고차 거래의 아이디\n",
    "    url : 중고차 거래 페이지\n",
    "    region : 해당 거래의 관리 지점\n",
    "    region_url : 거래 관리 지점의 홈페이지\n",
    "    price : 기입된 자동차의 거래가\n",
    "    year : 거래가 기입된 년도\n",
    "    manufacturer : 자동차를 생산한 회사\n",
    "    model : 자동차 모델명\n",
    "    condition : 자동차의 상태\n",
    "    cylinders : 자동차의 기통 수\n",
    "    fuel : 자동차의 연료 타입\n",
    "    odometer : 자동차의 운행 마일 수\n",
    "    title_status : 자동차의 타이틀 상태 (소유주 등록 상태)\n",
    "    transmission : 자동차의 트랜스미션 종류\n",
    "    vin : 자동차의 식별 번호 (vehicle identification number)\n",
    "    drive : 자동차의 구동 타입\n",
    "    size : 자동차 크기\n",
    "    type : 자동차의 일반 타입 (세단, ...)\n",
    "    paint_color : 자동차 색상\n",
    "    image_url : 자동차 이미지\n",
    "    description : 세부 설명\n",
    "    county : 실수로 생성된 미사용 컬럼\n",
    "    state : 거래가 업로드된 미 주\n",
    "    lat : 거래가 업로드된 곳의 위도\n",
    "    long : 거래가 업로드된 곳의 경도\n",
    "    \n",
    "    \n",
    "- 데이터 출처: https://www.kaggle.com/austinreese/craigslist-carstrucks-data\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IRssoNrgP7So"
   },
   "source": [
    "## 1. 데이터 및 라이브러리 로드"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "executionInfo": {
     "elapsed": 702,
     "status": "ok",
     "timestamp": 1631959234426,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "g49RuFGrBvt7"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "executionInfo": {
     "elapsed": 16485,
     "status": "ok",
     "timestamp": 1631959270162,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "RnJv-4YwCMSx"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('C:\\\\Users\\\\korea\\\\OneDrive\\\\바탕 화면\\\\DA\\\\vehicles.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9L3BNVM7tHN5"
   },
   "source": [
    "##  2.EDA \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "o9VBgWeCEQZG"
   },
   "source": [
    "### 피쳐 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 257
    },
    "executionInfo": {
     "elapsed": 12,
     "status": "ok",
     "timestamp": 1631959270166,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "YcR9BX23DIFW",
    "outputId": "04aab1e5-cf2e-4d75-9a0b-5890c64a1edf"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>url</th>\n",
       "      <th>region</th>\n",
       "      <th>region_url</th>\n",
       "      <th>price</th>\n",
       "      <th>year</th>\n",
       "      <th>manufacturer</th>\n",
       "      <th>model</th>\n",
       "      <th>condition</th>\n",
       "      <th>...</th>\n",
       "      <th>drive</th>\n",
       "      <th>size</th>\n",
       "      <th>type</th>\n",
       "      <th>paint_color</th>\n",
       "      <th>image_url</th>\n",
       "      <th>description</th>\n",
       "      <th>state</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>posting_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>7240372487</td>\n",
       "      <td>https://auburn.craigslist.org/ctd/d/auburn-uni...</td>\n",
       "      <td>auburn</td>\n",
       "      <td>https://auburn.craigslist.org</td>\n",
       "      <td>35990</td>\n",
       "      <td>2010.0</td>\n",
       "      <td>chevrolet</td>\n",
       "      <td>corvette grand sport</td>\n",
       "      <td>good</td>\n",
       "      <td>...</td>\n",
       "      <td>rwd</td>\n",
       "      <td>NaN</td>\n",
       "      <td>other</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://images.craigslist.org/00N0N_ipkbHVZYf4...</td>\n",
       "      <td>Carvana is the safer way to buy a car During t...</td>\n",
       "      <td>al</td>\n",
       "      <td>32.590000</td>\n",
       "      <td>-85.480000</td>\n",
       "      <td>2020-12-02T08:11:30-0600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>7240309422</td>\n",
       "      <td>https://auburn.craigslist.org/cto/d/auburn-201...</td>\n",
       "      <td>auburn</td>\n",
       "      <td>https://auburn.craigslist.org</td>\n",
       "      <td>7500</td>\n",
       "      <td>2014.0</td>\n",
       "      <td>hyundai</td>\n",
       "      <td>sonata</td>\n",
       "      <td>excellent</td>\n",
       "      <td>...</td>\n",
       "      <td>fwd</td>\n",
       "      <td>NaN</td>\n",
       "      <td>sedan</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://images.craigslist.org/00s0s_gBHYmJ5o7y...</td>\n",
       "      <td>I'll move to another city and try to sell my c...</td>\n",
       "      <td>al</td>\n",
       "      <td>32.547500</td>\n",
       "      <td>-85.468200</td>\n",
       "      <td>2020-12-02T02:11:50-0600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>7240224296</td>\n",
       "      <td>https://auburn.craigslist.org/cto/d/auburn-200...</td>\n",
       "      <td>auburn</td>\n",
       "      <td>https://auburn.craigslist.org</td>\n",
       "      <td>4900</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>bmw</td>\n",
       "      <td>x3 3.0i</td>\n",
       "      <td>good</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>SUV</td>\n",
       "      <td>blue</td>\n",
       "      <td>https://images.craigslist.org/00B0B_5zgEGWPOrt...</td>\n",
       "      <td>Clean 2006 BMW X3 3.0I.  Beautiful and rare Bl...</td>\n",
       "      <td>al</td>\n",
       "      <td>32.616807</td>\n",
       "      <td>-85.464149</td>\n",
       "      <td>2020-12-01T19:50:41-0600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>7240103965</td>\n",
       "      <td>https://auburn.craigslist.org/cto/d/lanett-tru...</td>\n",
       "      <td>auburn</td>\n",
       "      <td>https://auburn.craigslist.org</td>\n",
       "      <td>2000</td>\n",
       "      <td>1974.0</td>\n",
       "      <td>chevrolet</td>\n",
       "      <td>c-10</td>\n",
       "      <td>good</td>\n",
       "      <td>...</td>\n",
       "      <td>rwd</td>\n",
       "      <td>full-size</td>\n",
       "      <td>pickup</td>\n",
       "      <td>blue</td>\n",
       "      <td>https://images.craigslist.org/00M0M_6o7KcDpArw...</td>\n",
       "      <td>1974 chev. truck (LONG BED) NEW starter front ...</td>\n",
       "      <td>al</td>\n",
       "      <td>32.861600</td>\n",
       "      <td>-85.216100</td>\n",
       "      <td>2020-12-01T15:54:45-0600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>7239983776</td>\n",
       "      <td>https://auburn.craigslist.org/cto/d/auburn-200...</td>\n",
       "      <td>auburn</td>\n",
       "      <td>https://auburn.craigslist.org</td>\n",
       "      <td>19500</td>\n",
       "      <td>2005.0</td>\n",
       "      <td>ford</td>\n",
       "      <td>f350 lariat</td>\n",
       "      <td>excellent</td>\n",
       "      <td>...</td>\n",
       "      <td>4wd</td>\n",
       "      <td>full-size</td>\n",
       "      <td>pickup</td>\n",
       "      <td>blue</td>\n",
       "      <td>https://images.craigslist.org/00p0p_b95l1EgUfl...</td>\n",
       "      <td>2005 Ford F350 Lariat (Bullet Proofed). This t...</td>\n",
       "      <td>al</td>\n",
       "      <td>32.547500</td>\n",
       "      <td>-85.468200</td>\n",
       "      <td>2020-12-01T12:53:56-0600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0          id                                                url  \\\n",
       "0           0  7240372487  https://auburn.craigslist.org/ctd/d/auburn-uni...   \n",
       "1           1  7240309422  https://auburn.craigslist.org/cto/d/auburn-201...   \n",
       "2           2  7240224296  https://auburn.craigslist.org/cto/d/auburn-200...   \n",
       "3           3  7240103965  https://auburn.craigslist.org/cto/d/lanett-tru...   \n",
       "4           4  7239983776  https://auburn.craigslist.org/cto/d/auburn-200...   \n",
       "\n",
       "   region                     region_url  price    year manufacturer  \\\n",
       "0  auburn  https://auburn.craigslist.org  35990  2010.0    chevrolet   \n",
       "1  auburn  https://auburn.craigslist.org   7500  2014.0      hyundai   \n",
       "2  auburn  https://auburn.craigslist.org   4900  2006.0          bmw   \n",
       "3  auburn  https://auburn.craigslist.org   2000  1974.0    chevrolet   \n",
       "4  auburn  https://auburn.craigslist.org  19500  2005.0         ford   \n",
       "\n",
       "                  model  condition  ... drive       size    type paint_color  \\\n",
       "0  corvette grand sport       good  ...   rwd        NaN   other         NaN   \n",
       "1                sonata  excellent  ...   fwd        NaN   sedan         NaN   \n",
       "2               x3 3.0i       good  ...   NaN        NaN     SUV        blue   \n",
       "3                  c-10       good  ...   rwd  full-size  pickup        blue   \n",
       "4           f350 lariat  excellent  ...   4wd  full-size  pickup        blue   \n",
       "\n",
       "                                           image_url  \\\n",
       "0  https://images.craigslist.org/00N0N_ipkbHVZYf4...   \n",
       "1  https://images.craigslist.org/00s0s_gBHYmJ5o7y...   \n",
       "2  https://images.craigslist.org/00B0B_5zgEGWPOrt...   \n",
       "3  https://images.craigslist.org/00M0M_6o7KcDpArw...   \n",
       "4  https://images.craigslist.org/00p0p_b95l1EgUfl...   \n",
       "\n",
       "                                         description state        lat  \\\n",
       "0  Carvana is the safer way to buy a car During t...    al  32.590000   \n",
       "1  I'll move to another city and try to sell my c...    al  32.547500   \n",
       "2  Clean 2006 BMW X3 3.0I.  Beautiful and rare Bl...    al  32.616807   \n",
       "3  1974 chev. truck (LONG BED) NEW starter front ...    al  32.861600   \n",
       "4  2005 Ford F350 Lariat (Bullet Proofed). This t...    al  32.547500   \n",
       "\n",
       "        long              posting_date  \n",
       "0 -85.480000  2020-12-02T08:11:30-0600  \n",
       "1 -85.468200  2020-12-02T02:11:50-0600  \n",
       "2 -85.464149  2020-12-01T19:50:41-0600  \n",
       "3 -85.216100  2020-12-01T15:54:45-0600  \n",
       "4 -85.468200  2020-12-01T12:53:56-0600  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 387,
     "status": "ok",
     "timestamp": 1631959270549,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "7TRe0sR9CoKw",
    "outputId": "db501e4a-b181-4ac9-83c9-9e39f6f61b63"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 458213 entries, 0 to 458212\n",
      "Data columns (total 26 columns):\n",
      " #   Column        Non-Null Count   Dtype  \n",
      "---  ------        --------------   -----  \n",
      " 0   Unnamed: 0    458213 non-null  int64  \n",
      " 1   id            458213 non-null  int64  \n",
      " 2   url           458213 non-null  object \n",
      " 3   region        458213 non-null  object \n",
      " 4   region_url    458213 non-null  object \n",
      " 5   price         458213 non-null  int64  \n",
      " 6   year          457163 non-null  float64\n",
      " 7   manufacturer  439993 non-null  object \n",
      " 8   model         453367 non-null  object \n",
      " 9   condition     265273 non-null  object \n",
      " 10  cylinders     287073 non-null  object \n",
      " 11  fuel          454976 non-null  object \n",
      " 12  odometer      402910 non-null  float64\n",
      " 13  title_status  455636 non-null  object \n",
      " 14  transmission  455771 non-null  object \n",
      " 15  VIN           270664 non-null  object \n",
      " 16  drive         324025 non-null  object \n",
      " 17  size          136865 non-null  object \n",
      " 18  type          345475 non-null  object \n",
      " 19  paint_color   317370 non-null  object \n",
      " 20  image_url     458185 non-null  object \n",
      " 21  description   458143 non-null  object \n",
      " 22  state         458213 non-null  object \n",
      " 23  lat           450765 non-null  float64\n",
      " 24  long          450765 non-null  float64\n",
      " 25  posting_date  458185 non-null  object \n",
      "dtypes: float64(4), int64(3), object(19)\n",
      "memory usage: 90.9+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 318,
     "status": "ok",
     "timestamp": 1631959270865,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "CpthOJohC8ce",
    "outputId": "ab2272c1-9b84-4092-c8f3-d4438c87c204"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unnamed: 0           0\n",
       "id                   0\n",
       "url                  0\n",
       "region               0\n",
       "region_url           0\n",
       "price                0\n",
       "year              1050\n",
       "manufacturer     18220\n",
       "model             4846\n",
       "condition       192940\n",
       "cylinders       171140\n",
       "fuel              3237\n",
       "odometer         55303\n",
       "title_status      2577\n",
       "transmission      2442\n",
       "VIN             187549\n",
       "drive           134188\n",
       "size            321348\n",
       "type            112738\n",
       "paint_color     140843\n",
       "image_url           28\n",
       "description         70\n",
       "state                0\n",
       "lat               7448\n",
       "long              7448\n",
       "posting_date        28\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "중요 정보들이 NaN값들이 많아보임"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>price</th>\n",
       "      <th>year</th>\n",
       "      <th>odometer</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>458213.000000</td>\n",
       "      <td>4.582130e+05</td>\n",
       "      <td>4.582130e+05</td>\n",
       "      <td>457163.000000</td>\n",
       "      <td>4.029100e+05</td>\n",
       "      <td>450765.000000</td>\n",
       "      <td>450765.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>229106.000000</td>\n",
       "      <td>7.235233e+09</td>\n",
       "      <td>4.042093e+04</td>\n",
       "      <td>2010.746067</td>\n",
       "      <td>1.016698e+05</td>\n",
       "      <td>38.531925</td>\n",
       "      <td>-94.375824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>132274.843786</td>\n",
       "      <td>4.594362e+06</td>\n",
       "      <td>8.194599e+06</td>\n",
       "      <td>8.868136</td>\n",
       "      <td>3.228623e+06</td>\n",
       "      <td>5.857378</td>\n",
       "      <td>18.076225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.208550e+09</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1900.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-82.607549</td>\n",
       "      <td>-164.091797</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>114553.000000</td>\n",
       "      <td>7.231953e+09</td>\n",
       "      <td>4.900000e+03</td>\n",
       "      <td>2008.000000</td>\n",
       "      <td>4.087700e+04</td>\n",
       "      <td>34.600000</td>\n",
       "      <td>-110.890427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>229106.000000</td>\n",
       "      <td>7.236409e+09</td>\n",
       "      <td>1.099500e+04</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>8.764100e+04</td>\n",
       "      <td>39.244500</td>\n",
       "      <td>-88.314889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>343659.000000</td>\n",
       "      <td>7.239321e+09</td>\n",
       "      <td>2.149500e+04</td>\n",
       "      <td>2016.000000</td>\n",
       "      <td>1.340000e+05</td>\n",
       "      <td>42.484503</td>\n",
       "      <td>-81.015022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>458212.000000</td>\n",
       "      <td>7.241019e+09</td>\n",
       "      <td>3.615215e+09</td>\n",
       "      <td>2021.000000</td>\n",
       "      <td>2.043756e+09</td>\n",
       "      <td>82.049255</td>\n",
       "      <td>150.898969</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Unnamed: 0            id         price           year      odometer  \\\n",
       "count  458213.000000  4.582130e+05  4.582130e+05  457163.000000  4.029100e+05   \n",
       "mean   229106.000000  7.235233e+09  4.042093e+04    2010.746067  1.016698e+05   \n",
       "std    132274.843786  4.594362e+06  8.194599e+06       8.868136  3.228623e+06   \n",
       "min         0.000000  7.208550e+09  0.000000e+00    1900.000000  0.000000e+00   \n",
       "25%    114553.000000  7.231953e+09  4.900000e+03    2008.000000  4.087700e+04   \n",
       "50%    229106.000000  7.236409e+09  1.099500e+04    2013.000000  8.764100e+04   \n",
       "75%    343659.000000  7.239321e+09  2.149500e+04    2016.000000  1.340000e+05   \n",
       "max    458212.000000  7.241019e+09  3.615215e+09    2021.000000  2.043756e+09   \n",
       "\n",
       "                 lat           long  \n",
       "count  450765.000000  450765.000000  \n",
       "mean       38.531925     -94.375824  \n",
       "std         5.857378      18.076225  \n",
       "min       -82.607549    -164.091797  \n",
       "25%        34.600000    -110.890427  \n",
       "50%        39.244500     -88.314889  \n",
       "75%        42.484503     -81.015022  \n",
       "max        82.049255     150.898969  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 가격: 최소값이 0인 부분 이상함, 최대가격이 너무 높아 평균치가 너무 높게 나타남\n",
    "- year: 최소값 1900 입력값이 이상해보임\n",
    "- odometer: max 값이 너무 큼"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 6,
     "status": "ok",
     "timestamp": 1631959270865,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "O76XgHQ6DLSQ",
    "outputId": "97e97db7-08d1-43a9-b3a4-b90dc2176aab"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0', 'id', 'url', 'region', 'region_url', 'price', 'year',\n",
       "       'manufacturer', 'model', 'condition', 'cylinders', 'fuel', 'odometer',\n",
       "       'title_status', 'transmission', 'VIN', 'drive', 'size', 'type',\n",
       "       'paint_color', 'image_url', 'description', 'state', 'lat', 'long',\n",
       "       'posting_date'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 불필요한 피쳐 제거"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "executionInfo": {
     "elapsed": 3,
     "status": "ok",
     "timestamp": 1631959270865,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "iP_JqJuKDNzx"
   },
   "outputs": [],
   "source": [
    "# 불필요한 피쳐 제거\n",
    "df.drop(['Unnamed: 0','id', 'url', 'region_url', 'VIN', 'image_url', 'description', 'lat', 'long', 'posting_date'],\n",
    "        axis=1,\n",
    "        inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "description 피쳐: word embedding -> BoW방식처럼 feature vec로 바꿔주고 이를 자동차의 피쳐로 활용할 수 있을것 같기도 함"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "executionInfo": {
     "elapsed": 8,
     "status": "ok",
     "timestamp": 1631959271147,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "MHLHzwvQDrrN"
   },
   "outputs": [],
   "source": [
    "# 중고차 나이 피쳐 생성\n",
    "df['age'] = 2021 - df['year']\n",
    "df.drop(['year'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 257
    },
    "executionInfo": {
     "elapsed": 8,
     "status": "ok",
     "timestamp": 1631959271148,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "DaAA06fBEMmn",
    "outputId": "6cc7c63a-b7c6-4b7c-ebec-f97bed546cb6"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>region</th>\n",
       "      <th>price</th>\n",
       "      <th>manufacturer</th>\n",
       "      <th>model</th>\n",
       "      <th>condition</th>\n",
       "      <th>cylinders</th>\n",
       "      <th>fuel</th>\n",
       "      <th>odometer</th>\n",
       "      <th>title_status</th>\n",
       "      <th>transmission</th>\n",
       "      <th>drive</th>\n",
       "      <th>size</th>\n",
       "      <th>type</th>\n",
       "      <th>paint_color</th>\n",
       "      <th>state</th>\n",
       "      <th>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>auburn</td>\n",
       "      <td>35990</td>\n",
       "      <td>chevrolet</td>\n",
       "      <td>corvette grand sport</td>\n",
       "      <td>good</td>\n",
       "      <td>8 cylinders</td>\n",
       "      <td>gas</td>\n",
       "      <td>32742.0</td>\n",
       "      <td>clean</td>\n",
       "      <td>other</td>\n",
       "      <td>rwd</td>\n",
       "      <td>NaN</td>\n",
       "      <td>other</td>\n",
       "      <td>NaN</td>\n",
       "      <td>al</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>auburn</td>\n",
       "      <td>7500</td>\n",
       "      <td>hyundai</td>\n",
       "      <td>sonata</td>\n",
       "      <td>excellent</td>\n",
       "      <td>4 cylinders</td>\n",
       "      <td>gas</td>\n",
       "      <td>93600.0</td>\n",
       "      <td>clean</td>\n",
       "      <td>automatic</td>\n",
       "      <td>fwd</td>\n",
       "      <td>NaN</td>\n",
       "      <td>sedan</td>\n",
       "      <td>NaN</td>\n",
       "      <td>al</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>auburn</td>\n",
       "      <td>4900</td>\n",
       "      <td>bmw</td>\n",
       "      <td>x3 3.0i</td>\n",
       "      <td>good</td>\n",
       "      <td>6 cylinders</td>\n",
       "      <td>gas</td>\n",
       "      <td>87046.0</td>\n",
       "      <td>clean</td>\n",
       "      <td>automatic</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>SUV</td>\n",
       "      <td>blue</td>\n",
       "      <td>al</td>\n",
       "      <td>15.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>auburn</td>\n",
       "      <td>2000</td>\n",
       "      <td>chevrolet</td>\n",
       "      <td>c-10</td>\n",
       "      <td>good</td>\n",
       "      <td>4 cylinders</td>\n",
       "      <td>gas</td>\n",
       "      <td>190000.0</td>\n",
       "      <td>clean</td>\n",
       "      <td>automatic</td>\n",
       "      <td>rwd</td>\n",
       "      <td>full-size</td>\n",
       "      <td>pickup</td>\n",
       "      <td>blue</td>\n",
       "      <td>al</td>\n",
       "      <td>47.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>auburn</td>\n",
       "      <td>19500</td>\n",
       "      <td>ford</td>\n",
       "      <td>f350 lariat</td>\n",
       "      <td>excellent</td>\n",
       "      <td>8 cylinders</td>\n",
       "      <td>diesel</td>\n",
       "      <td>116000.0</td>\n",
       "      <td>lien</td>\n",
       "      <td>automatic</td>\n",
       "      <td>4wd</td>\n",
       "      <td>full-size</td>\n",
       "      <td>pickup</td>\n",
       "      <td>blue</td>\n",
       "      <td>al</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   region  price manufacturer                 model  condition    cylinders  \\\n",
       "0  auburn  35990    chevrolet  corvette grand sport       good  8 cylinders   \n",
       "1  auburn   7500      hyundai                sonata  excellent  4 cylinders   \n",
       "2  auburn   4900          bmw               x3 3.0i       good  6 cylinders   \n",
       "3  auburn   2000    chevrolet                  c-10       good  4 cylinders   \n",
       "4  auburn  19500         ford           f350 lariat  excellent  8 cylinders   \n",
       "\n",
       "     fuel  odometer title_status transmission drive       size    type  \\\n",
       "0     gas   32742.0        clean        other   rwd        NaN   other   \n",
       "1     gas   93600.0        clean    automatic   fwd        NaN   sedan   \n",
       "2     gas   87046.0        clean    automatic   NaN        NaN     SUV   \n",
       "3     gas  190000.0        clean    automatic   rwd  full-size  pickup   \n",
       "4  diesel  116000.0         lien    automatic   4wd  full-size  pickup   \n",
       "\n",
       "  paint_color state   age  \n",
       "0         NaN    al  11.0  \n",
       "1         NaN    al   7.0  \n",
       "2        blue    al  15.0  \n",
       "3        blue    al  47.0  \n",
       "4        blue    al  16.0  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HGHvPwkNtXgj"
   },
   "source": [
    "### 범주형 데이터 분석 및 시각화\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 8,
     "status": "ok",
     "timestamp": 1631959271148,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "b_zcrUDF7khF",
    "outputId": "ebd8e27f-5c24-4759-8f4b-fb4c9b55fce9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['region',\n",
       " 'manufacturer',\n",
       " 'model',\n",
       " 'condition',\n",
       " 'cylinders',\n",
       " 'fuel',\n",
       " 'title_status',\n",
       " 'transmission',\n",
       " 'drive',\n",
       " 'size',\n",
       " 'type',\n",
       " 'paint_color',\n",
       " 'state']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_columns = df.dtypes[df.dtypes == 'object'].index.tolist()\n",
    "\n",
    "cat_columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "31520"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 모델은 너무 많은 종류, 표현법도 다름\n",
    "len(df['model'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "executionInfo": {
     "elapsed": 6,
     "status": "ok",
     "timestamp": 1631959271149,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "sH5NWNNVHeXk"
   },
   "outputs": [],
   "source": [
    "cat_columns =['manufacturer','condition','cylinders', 'fuel','transmission','drive','size','type','paint_color']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "executionInfo": {
     "elapsed": 16468,
     "status": "ok",
     "timestamp": 1631959287611,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "F-rEjztfGYZ8",
    "outputId": "d69df2c2-7224-4211-c31c-b855bca33a81"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsgAAALICAYAAABiqwZ2AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAApB0lEQVR4nO3deZxldX3n//dHWgUFEQQVQWzcUHTcQCcquP+ImKCJEjXREZeRZJJoTKKTOCTGJDqJW+b3MJqYdsa4RI2j0VEziWJcUHGBBllERRExLgSCKCKu4Gf+uKfxS9lLNd1Vt6r6+Xw86lH3nnvOud/zfdwuXpw691Z1dwAAgJnrzXsAAACwkghkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZIA5qaoHVdVXh/vnVtWDtrL+P1fV8csxtpVm4VwBLKV18x4AADPdfZdNt6vq+Ulu391PHB4/Zh7jWgpV9dokX+3uP1hN+wZ2Dc4gAwDAQCADbEZV3bqq3l5V/15V36iqV0zLr1dVf1BVX66qS6rq9VW19/TY+qrqqjq+qv61qi6tqhOHfe5RVa+tqm9W1WeS3HvBc15YVQ+rqocn+W9JHldV36mqs6bHP1RV/3lHx7GZY92jql427evyqvpoVe0xPfbI6dKPb03Pf+dhu66q2w/3X1tVL5huP6iqvlpVvzuN76Kqesr02AlJnpDkv07H9+7h+J9bVZ+Z5uhvq2r3LYz5ztN4vjWN75Fb2zfA9hDIAAtU1W5J/jHJl5OsT3Jgkr+fHn7y9PXgJLdNsmeSVyzYxZFJDk3y0CTPG6Lyj5Lcbvr62SSbvZ64u9+T5L8neUt379ndd9/MajsyjoVemuTwJPdLsm+S/5rkx1V1xyRvTvKsJPsn+ack766qG2xhPwvdMsnemc3f05K8sqr26e4NSd6Y5MXT8R07bPOEzObmdknumOSnLpOoqusneXeSk5LcPMkzkryxqg7dxr4BFkUgA/y0+yS5VZLndPeV3f397v7o9NgTkvxFd1/Q3d9J8twkj6+q8T0df9zd3+vus5KclWRT4D42yQu7+7Lu/kqSl+/AGHdkHNeoqusleWqS3+rur3X31d39se7+QZLHJfm/3f2+7v5RZiG9R2YhvRg/SvIn3f2j7v6nJN/JLNi35hXd/ZXuvizJC5P88mbW+ZnM/ofgz7v7h939gcz+h2Zz6wJsN4EM8NNuneTL3X3VZh67VWZnljf5cmZveL7FsOzfhtvfzSzmNm37lQXbXlc7Mo7Rfkl2T/LFbT1Hd/84s/EfuMgxfmPBHG5pDKOF83OrLYzrK9N4xnUXOy6ArRLIAD/tK0kOXnA2dpOvJ7nNcP/gJFcluXgR+70os/get92S3sa+dmQco0uTfD+zSxq2+hxVVZmN/2vTou8mudGw/i2343m3dHwL5+frWxjXraez3+O6m8a1rbkD2CqBDPDTTs0sZv+8qm5cVbtX1f2nx96c5Ler6pCq2jM/uVZ4c2ebF/rfSZ5bVftU1UGZXTu7JRcnWb8gAkc7Mo5rTGdhX5PkL6rqVlW1W1Xdt6puOI3356rqodN1v7+b5AdJPjZtfmaSX5m2eXiSB27HU1+c2bXTC/1GVR1UVftm9kbFt2xmnU8muTKzN+Jdv2afHX1sfnKd+Jb2DbAoAhlgge6+OrPgun2Sf03y1cyux01mMfmGJB9O8qXMzr5uLXRHf5zZpQBfyuwNZm/Yyrpvnb5/o6rO2MzjOzKOhZ6d5JwkpyW5LMmLklyvu89L8sQkf5nZmeZjkxzb3T+ctvutadm3Mrsm+v9sx3P+rySHTZ9CMW73pszm5oLp6wULN5ye/5FJjpnG9VdJntTdn9vGvgEWpbr9JgqA+auqC5P85+7+l3mPBdi1OYMMAAADgQwAAAOXWAAAwMAZZAAAGGzuMz7ZjP3226/Xr18/72EAALAFp59++qXdvf+O7kcgL9L69euzcePGeQ8DAIAtqKod+Qul13CJBQAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAM/KGQRfrsV7+Rw5/z+nkPAwBgxTv9JU+a9xB2iDPIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMNilArmq7ltVr573OAAAWLl2qUBO8vAk75n3IAAAWLnWXCBX1fqq+mxVvbqqzq2qk6pqj+nhhyb5l2mdj1TVGdPX/eY5ZgAAVo41F8iTOyR5ZXffJcm3kjymqvZL8qPuvjzJJUn+v+6+V5LHJXn55nZSVSdU1caq2njVd69YpqEDADBP6+Y9gCXype4+c7p9epL1SY5OctK07PpJXlFV90hydZI7bm4n3b0hyYYkufEtD+mlGy4AACvFWg3kHwy3r06yR5JjkvzFtOy3k1yc5O6ZnUX//rKODgCAFWutXmKxUCW5W5Izp/t7J7mou3+c5D8l2W1O4wIAYIXZVQL58CSf6u5Nl0n8VZLjq+oTmV1eceXcRgYAwIqy5i6x6O4Lk9x1uP/Sqto9yfnDsi9kdkZ5k+cu2wABAFjR1lwgb053v2DeYwAAYHXYVS6xAACARRHIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADBYN+8BrBZ3Puhm2fiSJ817GAAALDFnkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGCwbt4DWC1+eNG5+dc/+Q/zHgbALuvg550z7yEAuwhnkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGCwSwdyVa2vqk/PexwAAKwcu3QgAwDAQuvmPYDtUVV/mOQJSb6S5NIkpyf5lySvSnKjJF9M8tTu/mZV3WMLyw9P8pok303y0WU/CAAAVrRVcwa5qo5I8pgk90zy6CRHTA+9PsnvdffdkpyT5I+2sfxvkzyzu++7iOc8oao2VtXGy668eucdDAAAK9aqCeQkRyZ5Z3d/r7uvSPLuJDdOctPuPnla53VJHlBVey9y+Ru29oTdvaG7j+juI/a98W47/YAAAFh5VlMg107aR++E/QAAsEatpkD+aJJjq2r3qtozyc8luTLJN6vqqGmd/5Tk5O6+fAvLv5Xk8qo6clr+hOUbPgAAq8GqeZNed59WVe9KclaSLyfZmOTyJMcneVVV3SjJBUmeMm2ypeVPSfKaqvpukvcu4yEAALAKVPfqueKgqvbs7u9M0fvhJCd09xnL8dx3O3CP/sdfvf1yPBUAm3Hw886Z9xCAFa6qTu/uI7a95tatmjPIkw1VdViS3ZO8brniGACAXceqCuTu/pV5jwEAgLVtNb1JDwAAlpxABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAwbp5D2C1uMEBd8nBz9s472EAALDEnEEGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAwbp5D2C1+Nwln8v9//L+8x4GwC7nlGecMu8hALsYZ5ABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgsCoDuapeW1XHTbc/VFVHXMf9PKiq7rdzRwcAwGq2KgN5J3pQEoEMAMA1ljSQq+qJVXVqVZ1ZVX9TVf+xqs6uqt2r6sZVdW5V3bWqdquql1bVOdPjz5i2P7yqTq6q06vqvVV1wDae7+iq+nhVnVFVb62qPaflF1bVH0/Lz6mqO1XV+iS/luS3p/EdtZRzAQDA6rBkgVxVd07yuCT37+57JLk6yaFJ3pXkBUlenOTvuvvTSU5IckiSe3b33ZK8saqun+QvkxzX3YcneU2SF27l+fZL8gdJHtbd90qyMcnvDKtcOi3/6yTP7u4Lk7wqyf/o7nt090c2s88TqmpjVW380Xd+tAOzAQDAarFuCff90CSHJzmtqpJkjySXJPmTJKcl+X6SZ07rPizJq7r7qiTp7suq6q5J7prkfdP2uyW5aCvP9zNJDktyyrT+DZJ8fHj87dP305M8ejEH0N0bkmxIkj0P3rMXsw0AAKvbUgZyJXlddz/3WgurbplkzyTXT7J7kiundRcGaCU5t7vvux3P977u/uUtPP6D6fvVWdrjBgBgFVvKa5Dfn+S4qrp5klTVvlV1m8zOyP5hkjcmedG07klJfq2q1m1aN8l5SfavqvtOy65fVXfZyvN9Isn9q+r20/o3qqo7bmOMVyTZ6zodHQAAa9KSBXJ3fyaza4JPqqqzk7wvyfFJruruNyX58yT3rqqHJPmfSf41ydlVdVaSX+nuHyY5LsmLpmVnZiufONHd/57kyUnePD3fJ5LcaRvDfHeSX/QmPQAANqlul9Yuxp4H79l3f87d5z0MgF3OKc84Zd5DAFaJqjq9u6/T38cY7eqfgwwAANcikAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGAgkAEAYCCQAQBgIJABAGCwbt4DWC3udPM75ZRnnDLvYQAAsMScQQYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAICBQAYAgIFABgCAgUAGAIDBunkPYLW44rzzcvIDHjjvYawKD/zwyfMeAgDAdeYMMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADNYtdsWqul+S9eM23f36JRgTAADMzaICuarekOR2Sc5McvW0uJMIZAAA1pTFnkE+Islh3d1LORgAAJi3xV6D/Okkt1zKgQAAwEqw2DPI+yX5TFWdmuQHmxZ29yOXZFQAADAniw3k5y/lIAAAYKVYVCB398lVdYsk954WndrdlyzdsAAAYD4WdQ1yVT02yalJfinJY5N8sqqOW8qBAQDAPCz2EosTk9x701njqto/yb8kedtSDQwAAOZhsZ9icb0Fl1R8Yzu2BQCAVWOxZ5DfU1XvTfLm6f7jkvzT0gwJAADmZ7Fv0ntOVT0myf2TVJIN3f2OJR0ZAADMwWLPIKe7/yHJPyzhWAAAYO62GshV9dHuPrKqrkgy/pnpStLdfZMlHR0AACyzrQZydx85fd9reYYDAADztdjPQX7DYpYBAMBqt9iParvLeKeq1iU5fOcPBwAA5murgVxVz52uP75bVX17+roiycVJ3rm9T1ZV35m+36qq3jbdfnJVveI6jB0AAHa6rQZyd//ZdP3xS7r7JtPXXt19s+5+7nV90u7+enf7U9UAAKw42zqDfKfp5lur6l4Lv67rk1bV+qr69GaW/1xVfbyq9quqo6fbZ1TVW6tqz82s/6GqelFVnVpVn6+qo6blu1XVS6rqtKo6u6p+dVr+V1X1yOn2O6rqNdPtp1XVC67r8QAAsHZs63OQfzfJ05O8bDOPdZKH7KyBVNUvJvmdJI9IsluSP0jysO6+sqp+b3rsTzaz6bruvk9VPSLJHyV5WJKnJbm8u+9dVTdMckpVnZTkw0mOSvKuJAcmOWDax5FJ/n4zYzohyQlJcosb3nBnHSoAACvYtj7m7enT9wcv8TgenOSIJEd397er6ueTHJZZ2CbJDZJ8fAvbvn36fnqS9dPtozO7bnrTZRx7J7lDko8keVZVHZbkM0n2qaoDktw3yTMX7ri7NyTZkCSH7rVXL3wcAIC1Z1t/KOTRW3u8u9++tce3wwVJbpvkjkk2ZvaHSN7X3b+8iG1/MH2/Oj85nkryjO5+78KVq2qfJA/P7Gzyvkkem+Q73X3FDh0BAABrwrYusTh2+n7zJPdL8oHp/oOTfCg/OXu7o76c5NlJ3lFVv5TkE0leWVW37+7zq+pGSQ7q7s8vcn/vTfJfquoD3f2jqrpjkq9195WZnYl+VmaXh9wsydumLwAA2OanWDylu5+S2fXGh3X3Y7r7MVnwucg7Q3efl+QJSd6a5CZJnpzkzVV1dmbBfKctb/1T/mdml1CcMb0Z8G/yk/8Z+Ehm1y2fn+SMzM4if2RnHAMAAKtfdW/70tqq+nR333W4f70kZ4/L1rpD99qrN9zzOn9wxy7lgR8+ed5DAAB2QVV1encfsaP72dYlFpt8qKrem+TNmZ1NfnySD+7okwMAwEqzqEDu7t+c3rB31LRoQ3e/Y+mGBQAA87HYM8ibPrFiZ70pDwAAVqStvklvk6p6dFV9oaour6pvV9UVVfXtpR4cAAAst8WeQX5xkmO7+7NLORgAAJi3RZ1BTnKxOAYAYFew2DPIG6vqLUn+T37yl+t25l/SAwCAFWGxgXyTJN9NcvSwrONNewAArDGL/Zi3pyz1QAAAYCVY7KdYHFRV76iqS6rq4qr6h6o6aKkHBwAAy22xb9L72yTvSnKrJAcmefe0DAAA1pTFBvL+3f233X3V9PXaJPsv4bgAAGAuFhvIl1bVE6tqt+nriUm+sZQDAwCAeVhsID81yWOT/FuSi5Icl8Qb9wAAWHMW+zFvf5rk+O7+ZpJU1b5JXppZOAMAwJqx2DPId9sUx0nS3ZcluefSDAkAAOZnsYF8varaZ9Od6QzyYs8+AwDAqrHYyH1Zko9V1dsy+wt6j03ywiUbFQAAzMli/5Le66tqY5KHJKkkj+7uzyzpyAAAYA4WfZnEFMSiGACANW2x1yADAMAuQSADAMBAIAMAwEAgAwDAQCADAMBAIAMAwEAgAwDAQCADAMBAIAMAwEAgAwDAQCADAMBg3bwHsFrsdeiheeCHT573MAAAWGLOIAMAwEAgAwDAQCADAMBAIAMAwEAgAwDAQCADAMBAIAMAwEAgAwDAQCADAMBAIAMAwEAgAwDAQCADAMBAIAMAwEAgAwDAQCADAMBAIAMAwEAgAwDAQCADAMBg3bwHsFpc8tXL84rfffeS7Ps3X3bskuwXAIDt5wwyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADNZUIFfVM6vqs1X1xi08fkRVvXy5xwUAwOqxbt4D2Ml+Pckx3f2lzT3Y3RuTbFy4vKrWdfdVSz04AABWvjVzBrmqXpXktkneVVW/V1Ufq6pPTd8PndZ5UFX943T7+VW1oapOSvL6OQ4dAIAVZM2cQe7uX6uqhyd5cJIfJnlZd19VVQ9L8t+TPGYzmx2e5Mju/t7m9llVJyQ5IUn22Wv/pRk4AAArypoJ5AX2TvK6qrpDkk5y/S2s964txXGSdPeGJBuS5OBb3qF3+igBAFhx1swlFgv8aZIPdvddkxybZPctrHfl8g0JAIDVYK0G8t5JvjbdfvIcxwEAwCqzVgP5xUn+rKpOSbLbvAcDAMDqsaauQe7u9dPNS5PccXjoD6fHP5TkQ9Pt5y/fyAAAWC3W6hlkAAC4TgQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADNbNewCrxc0P2ju/+bJj5z0MAACWmDPIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMFg37wGsFhd96Yt54ROPu+b+iX/3tjmOBgCApeIMMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMVnUgV9X6qvpsVb26qs6tqpOqao+qul1VvaeqTq+qj1TVnapqt6q6oGZuWlU/rqoHTPv5SFXdft7HAwDA/K3qQJ7cIckru/suSb6V5DFJNiR5RncfnuTZSf6qu69O8vkkhyU5MsnpSY6qqhsmOai7z1+446o6oao2VtXGK7//g+U5GgAA5mrdvAewE3ypu8+cbp+eZH2S+yV5a1VtWueG0/ePJHlAkkOS/FmSpyc5Oclpm9txd2/ILLZz4M326Z0/dAAAVpq1cAZ5PLV7dZJ9k3yru+8xfN15evwjSY5Kcp8k/5TkpkkelOTDyzdcAABWsrUQyAt9O8mXquqXkmS65vju02OfzOzs8o+7+/tJzkzyq5mFMwAArMlATpInJHlaVZ2V5Nwkj0qS7v5Bkq8k+cS03keS7JXknHkMEgCAlWdVX4Pc3Rcmuetw/6XDww/fwjZHDbfflORNSzU+AABWn7V6BhkAAK4TgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAACDdfMewGpxwCG3y4l/97Z5DwMAgCXmDDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBPIiff+iK/LZF35g3sMAAGCJCWQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYrPhArqrXVtVx8x4HAAC7hhUfyAAAsJzmEshVdeOq+r9VdVZVfbqqHldVz6uq06b7G6qqFmxzTFX97+H+g6rq3dPtv66qjVV1blX98bDOI6rqc1X10ap6eVX94/D8r5me71NV9ajlOnYAAFa2eZ1BfniSr3f33bv7rknek+QV3X3v6f4eSX5+wTbvS/IzVXXj6f7jkrxlun1idx+R5G5JHlhVd6uq3ZP8TZJjuvvIJPsP+zoxyQe6+95JHpzkJcN+r1FVJ0zhvfGyK7+1M44bAIAVbl6BfE6Sh1XVi6rqqO6+PMmDq+qTVXVOkockucu4QXdflVlIH1tV65L8XJJ3Tg8/tqrOSPKpabvDktwpyQXd/aVpnTcPuzs6ye9X1ZlJPpRk9yQHLxxkd2/o7iO6+4h9b3zTnXDYAACsdOvm8aTd/fmqOjzJI5L8WVWdlOQ3khzR3V+pqudnFq0LvWVa77Ikp3X3FVV1SJJnJ7l3d3+zql47bVub2X6TSvKY7j5vpx0UAABrwryuQb5Vku92998leWmSe00PXVpVeybZ0qdWfGha9+n5yeUVN0lyZZLLq+oWSY6Zln8uyW2rav10/3HDft6b5BmbrnOuqnvu6DEBALA2zOUMcpL/kNl1vz9O8qMk/yXJL2R26cWFSU7b3EbdffX0RrsnJzl+WnZWVX0qyblJLkhyyrT8e1X160neU1WXJjl12NWfJvn/k5w9RfKF+elrngEA2AVVd897DEumqvbs7u9MEfzKJF/o7v9xXfZ11wMP7bf++l/nzic+ZOcOEgCAnaKqTp8+uGGHrPXPQX769Ea8c5PsndmnWgAAwBbN6xKLZTGdLb5OZ4wBANg1rfUzyAAAsF0EMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADAQyAAAMBDIAAAwEMgAADATyIu1+wF6584kPmfcwAABYYgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAAAGAhkAAAYCGQAABtXd8x7DqlBVVyQ5b97jWEX2S3LpvAexypiz7WO+tp852z7ma/uZs+1jvrbftubsNt29/44+ybod3cEu5LzuPmLeg1gtqmqj+do+5mz7mK/tZ862j/nafuZs+5iv7bdcc+YSCwAAGAhkAAAYCOTF2zDvAawy5mv7mbPtY762nznbPuZr+5mz7WO+tt+yzJk36QEAwMAZZAAAGAhkAAAYCORtqKqHV9V5VXV+Vf3+vMeznKrq1lX1war6bFWdW1W/NS1/flV9rarOnL4eMWzz3Gmuzquqnx2WH15V50yPvbyqalp+w6p6y7T8k1W1ftkPdCerqgunYz2zqjZOy/atqvdV1Rem7/sM6++yc1ZVhw6vozOr6ttV9SyvsWurqtdU1SVV9elh2bK8pqrq+Ok5vlBVxy/TIe+QLczXS6rqc1V1dlW9o6puOi1fX1XfG15rrxq22SXmK9ninC3Lv8PVOGdbmK+3DHN1YVWdOS3f5V9jteWeWLk/x7rb1xa+kuyW5ItJbpvkBknOSnLYvMe1jMd/QJJ7Tbf3SvL5JIcleX6SZ29m/cOmObphkkOmudtteuzUJPdNUkn+Ockx0/JfT/Kq6fbjk7xl3se9E+btwiT7LVj24iS/P93+/SQvMmc/NW+7Jfm3JLfxGvup435Aknsl+fRyvqaS7Jvkgun7PtPtfeY9H9dxvo5Osm66/aJhvtaP6y3Yzy4xX1uZsyX/d7ha52xz87Xg8ZcleZ7X2DXHuaWeWLE/x5xB3rr7JDm/uy/o7h8m+fskj5rzmJZNd1/U3WdMt69I8tkkB25lk0cl+fvu/kF3fynJ+UnuU1UHJLlJd3+8Z6/W1yf5hWGb102335bkoZv+b3CNGY/zdbn28ZuzmYcm+WJ3f3kr6+yS89XdH05y2YLFy/Ga+tkk7+vuy7r7m0nel+ThO/v4drbNzVd3n9TdV013P5HkoK3tY1ear2SLr7Et8RrbynxNx/XYJG/e2j52sfnaUk+s2J9jAnnrDkzyleH+V7P1QFyzpl9V3DPJJ6dFv1mzX1W+ZviVyJbm68Dp9sLl19pm+o/X5UluthTHsIw6yUlVdXpVnTAtu0V3X5TMflAkufm03Jz9xONz7f+geI1t3XK8ptbqz8CnZnbmaZNDqupTVXVyVR01LTNfM0v973AtztlRSS7u7i8My7zGJgt6YsX+HBPIW7e5s0y73OfiVdWeSf4hybO6+9tJ/jrJ7ZLcI8lFmf0qKdnyfG1tHtfiHN+/u++V5Jgkv1FVD9jKuuYsSVXdIMkjk7x1WuQ1dt3tzDlac3NXVScmuSrJG6dFFyU5uLvvmeR3krypqm4S85Usz7/DtTZnSfLLufb/7HuNTTbTE1tcdTPLlvU1JpC37qtJbj3cPyjJ1+c0lrmoqutn9mJ+Y3e/PUm6++Luvrq7f5zk1ZldipJseb6+mmv/OnOcx2u2qap1SfbO4n/NtyJ199en75ckeUdm83Px9KuhTb9Wu2Ra3ZzNHJPkjO6+OPEaW6TleE2tqZ+B05tzfj7JE6Zfz2b6Fe43ptunZ3at4x1jvpbr3+GamrPp2B6d5C2blnmNzWyuJ7KCf44J5K07LckdquqQ6QzX45O8a85jWjbTtTv/K8lnu/svhuUHDKv9YpJN7+J9V5LHT+8kPSTJHZKcOv3a5Iqq+plpn09K8s5hm+On28cl+cCm/3CtRlV146raa9PtzN4Y9Olc+ziPz7WPf5ees8m1zrh4jS3Kcrym3pvk6KraZ/r1+tHTslWnqh6e5PeSPLK7vzss37+qdptu3zaz+bpgV5+vZNn+Ha6pOUvysCSf6+5rLgPwGttyT2Ql/xzrFfDuxpX8leQRmb3b8otJTpz3eJb52I/M7NcQZyc5c/p6RJI3JDlnWv6uJAcM25w4zdV5md5ZOi0/IrMfrl9M8or85K847p7Zr9XPz+ydqbed93Hv4JzdNrN33p6V5NxNr5nMroN6f5IvTN/3NWfXHOeNknwjyd7DMq+xa8/RmzP7Ne2PMjsb8rTlek1ldr3u+dPXU+Y9FzswX+dndh3ipp9lm97t/pjp3+pZSc5IcuyuNl9bmbNl+Xe4Gudsc/M1LX9tkl9bsO4u/xrLlntixf4c86emAQBg4BILAAAYCGQAABgIZAAAGAhkAAAYCGQAABgIZAC2qqqeVVU3mvc4AJaLj3kDYKuq6sIkR3T3pfMeC8BycAYZYA2oqidV1dlVdVZVvaGqblNV75+Wvb+qDp7We21VHTds953p+4Oq6kNV9baq+lxVvbFmnpnkVkk+WFUfnM/RASyvdfMeAAA7pqruktlfnbp/d19aVfsmeV2S13f366rqqUlenuQXtrGreya5S5KvJzll2t/Lq+p3kjzYGWRgV+EMMsDq95Akb9sUsN19WZL7JnnT9PgbMvtTr9tyand/tbt/nNmfgl2/84cKsPIJZIDVr5Js6w0lmx6/KtPP/qqqJDcY1vnBcPvq+C0jsIsSyACr3/uTPLaqbpYk0yUWH0vy+OnxJyT56HT7wiSHT7cfleT6i9j/FUn22lmDBVjpnB0AWOW6+9yqemGSk6vq6iSfSvLMJK+pquck+fckT5lWf3WSd1bVqZmF9ZWLeIoNSf65qi7q7gfv/CMAWFl8zBsAAAxcYgEAAAOBDAAAA4EMAAADgQwAAAOBDAAAA4EMAAADgQwAAIP/B8VU/bD5H+evAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for column in cat_columns:\n",
    "  fig = plt.figure(figsize=(10,10))\n",
    "  sns.countplot(y=column, data=df.fillna('n/a'), order=df.fillna('n/a')[column].value_counts().index)\n",
    "  plt.title(column + ' countplot')\n",
    "  plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "amkU5Y6g7Dp4"
   },
   "source": [
    "### 수치형 피쳐 분석 및 시각화"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 8,
     "status": "ok",
     "timestamp": 1631959287612,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "GZXSBFPyDh6R",
    "outputId": "70569529-5ac2-40df-8946-d8f6a36f418c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['price', 'odometer', 'age']"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_columns = df.dtypes[df.dtypes != 'object'].index.tolist()\n",
    "num_columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 357
    },
    "executionInfo": {
     "elapsed": 19569,
     "status": "ok",
     "timestamp": 1631959307178,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "eAaDEX03LYx9",
    "outputId": "a99dc24d-8982-42e9-dd00-7cdb6c51a7e4"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAewAAACqCAYAAABvXHylAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAN+UlEQVR4nO3df6xkZX3H8ffH5ZcFhOquCPuDRYumaAHpleCPtrTFCjRmqUqEGLQWumrValKbgm00tVptakyqYMmmIphSoVXaEroElGgsUn5HkHUBtwjdla0uIGBRatZ++8ec3VzHe/eOzOzeeWber2Sy58cz53kevnf57Jk595xUFZIkabw9bbEHIEmSFmZgS5LUAANbkqQGGNiSJDXAwJYkqQEGtiRJDTCwJQGQ5FeS3LPY45A0t/h72JIkjT/PsCWRZK/FHoOkXTOwpQmW5P4k5yX5RpLvJfl0kv2SnJhkS5I/SfLfwKd3bJv13pVJrkiyLcnDSc6fte/3kmzsjnlNksMXZYLSFDGwpcn3BuBVwPOA5wN/1m1/DvBM4HBg7ew3JFkCXAU8AKwGlgOXdftOA94LvAZYBvw78NndOwVJBrY0+c6vqs1V9QjwIeDMbvv/Ae+vqv+tqh/2ved44DDgj6vqiap6sqqu7/a9BfhwVW2squ3AXwLHepYt7V4GtjT5Ns9afoBeEANsq6on53nPSuCBLpD7HQ78TZJHkzwKPAKE3lm4pN3EC02kybdy1vIq4MFueVe/IrIZWJVkrzlCezPwoaq6dIRjlLQAz7Clyff2JCuSPJPed8+XD/Cem4GtwEeS7N9dqPbybt+FwHlJXgiQ5KAkp++WkUvaycCWJt8/ANcC93WvDy70hqr6MfBq4BeA/wK2AK/v9v0z8FfAZUkeB+4CTtktI5e0kzdOkSZYkvuBc6rqi4s9FknD8QxbkqQGGNiSJDXAj8QlSWqAZ9iSJDXAwJYkqQFjfeOUpUuX1urVq0d2vK9/+zEAnn3gvhzyjP0A+M7jT+5cnr1+37YneO6y/Xe+75eWHzSycWi89f9MSK2Z72fYn+3RG/V/09tuu+2hqlo2176x/g57Zmambr311pEdb/W5/7Zz+f6P/PbObTuWZ6/P3t7fRpPNeqt18/0M+7M9eqP+b5rktqqamWufH4lLktQAA1uSpAYY2JIkNcDAliSpAQa2JEkNMLAlSWrASAI7yclJ7kmyKcm5c+x/Q5I7u9cNSY4ZRb+SJE2LoQM7yRLgAnrPwz0KODPJUX3NvgX8WlUdDfwFsG7YfiVJmiajOMM+HthUVfdV1Y+Ay4A1sxtU1Q1V9b1u9UZgxQj6lSRpaowisJcDm2etb+m2zeds4OoR9CtJ0tQYxb3EM8e2Oe93muTX6QX2K+Y9WLIWWAuwatWqEQxPkqT2jeIMewuwctb6CuDB/kZJjgb+DlhTVQ/Pd7CqWldVM1U1s2zZnPc/lyRp6owisG8BjkxyRJJ9gDOAK2c3SLIKuAI4q6ruHUGfkiRNlaE/Eq+q7UneAVwDLAEuqqoNSd7a7b8QeB/wLOCTSQC2z/c0EkmS9NNG8jzsqloPrO/bduGs5XOAc0bRlyRJ08g7nUmS1AADW5KkBhjYkiQ1wMCWJKkBBrYkSQ0wsCVJaoCBLUlSAwxsSZIaYGBLktQAA1uSpAYY2JIkNcDAliSpAQa2JEkNMLAlSWqAgS1JUgMMbEmSGmBgS5LUAANbkqQGGNiSJDXAwJYkqQEGtiRJDTCwJUlqgIEtSVIDDGxJkhpgYEuS1AADW5KkBhjYkiQ1wMCWJKkBBrYkSQ0wsCVJaoCBLUlSAwxsSZIaYGBLktQAA1uSpAaMJLCTnJzkniSbkpw7x/4k+Xi3/84kx42iX0mSpsXQgZ1kCXABcApwFHBmkqP6mp0CHNm91gJ/O2y/kiRNk1GcYR8PbKqq+6rqR8BlwJq+NmuAz1TPjcDBSQ4dQd+SJE2FUQT2cmDzrPUt3baftQ0ASdYmuTXJrdu2bRvB8CRJat8oAjtzbKun0Ka3sWpdVc1U1cyyZcuGHpwkSZNgFIG9BVg5a30F8OBTaCNJkuYxisC+BTgyyRFJ9gHOAK7sa3Ml8MbuavETgMeqausI+pYkaSrsNewBqmp7kncA1wBLgIuqakOSt3b7LwTWA6cCm4AfAG8etl9JkqbJ0IENUFXr6YXy7G0Xzlou4O2j6EuSpGnknc4kSWqAgS1JUgMMbEmSGmBgS5LUAANbkqQGGNiSJDXAwJYkqQEGtiRJDTCwJUlqgIEtSVIDDGxJkhpgYEuS1AADW5KkBhjYkiQ1wMCWJKkBBrYkSQ0wsCVJaoCBLUlSAwxsSZIaYGBLktQAA1uSpAYY2JIkNcDAliSpAQa2JEkNMLAlSWqAgS1JUgMMbEmSGmBgS5LUAANbkqQGGNiSJDXAwJYkqQEGtiRJDTCwJUlqwFCBneSZSb6Q5Jvdnz8/R5uVSb6UZGOSDUneNUyfkiRNo2HPsM8FrquqI4HruvV+24E/qqpfBE4A3p7kqCH7lSRpqgwb2GuAS7rlS4DT+htU1daqur1b/j6wEVg+ZL+SJE2VYQP7kKraCr1gBp69q8ZJVgMvBm4asl9JkqbKXgs1SPJF4Dlz7PrTn6WjJAcAnwfeXVWP76LdWmAtwKpVq36WLiRJmlgLBnZVnTTfviTfSXJoVW1Ncijw3Xna7U0vrC+tqisW6G8dsA5gZmamFhqfJEnTYNiPxK8E3tQtvwn41/4GSQJ8CthYVR8bsj9JkqbSsIH9EeCVSb4JvLJbJ8lhSdZ3bV4OnAX8RpKvda9Th+xXkqSpsuBH4rtSVQ8DvznH9geBU7vl64EM048kSdPOO51JktQAA1uSpAYY2JIkNcDAliSpAQa2JEkNMLAlSWqAgS1JUgNSNb53/0yyDXhghIdcCjw0wuONG+fXNufXtkmfH0z+HMdhfodX1bK5dox1YI9akluramaxx7G7OL+2Ob+2Tfr8YPLnOO7z8yNxSZIaYGBLktSAaQvsdYs9gN3M+bXN+bVt0ucHkz/HsZ7fVH2HLUlSq6btDFuSpCZNXGAnOTnJPUk2JTl3jv1J8vFu/51JjluMcQ5jgDmemOSxWc8ff99ijPOpSHJRku8muWue/U3Xb4D5NVs7gCQrk3wpycYkG5K8a442zdZwwPk1W8Mk+yW5Ockd3fz+fI42LddvkPmNb/2qamJewBLgP4HnAvsAdwBH9bU5Fbia3jO6TwBuWuxx74Y5nghctdhjfYrz+1XgOOCuefa3Xr+F5tds7brxHwoc1y0fCNw7SX8HB5xfszXsanJAt7w3cBNwwgTVb5D5jW39Ju0M+3hgU1XdV1U/Ai4D1vS1WQN8pnpuBA5OcuieHugQBpljs6rqK8Aju2jSdP0GmF/TqmprVd3eLX8f2Ags72vWbA0HnF+zupr8T7e6d/fqv9Cp5foNMr+xNWmBvRzYPGt9Cz/9l2mQNuNs0PG/tPvY5+okL9wzQ9sjWq/fICaidklWAy+mdxYz20TUcBfzg4ZrmGRJkq8B3wW+UFUTVb8B5gdjWr9JC+zMsa3/X0+DtBlng4z/dnq3tzsG+ATwL7t7UHtQ6/VbyETULskBwOeBd1fV4/2753hLUzVcYH5N17CqflxVxwIrgOOTvKivSdP1G2B+Y1u/SQvsLcDKWesrgAefQptxtuD4q+rxHR/7VNV6YO8kS/fcEHer1uu3S5NQuyR70wuzS6vqijmaNF3DheY3CTUEqKpHgS8DJ/ftarp+O8w3v3Gu36QF9i3AkUmOSLIPcAZwZV+bK4E3dlc6ngA8VlVb9/RAh7DgHJM8J0m65ePp1fnhPT7S3aP1+u1S67Xrxv4pYGNVfWyeZs3WcJD5tVzDJMuSHNwtPx04Cbi7r1nL9VtwfuNcv70WewCjVFXbk7wDuIbe1dQXVdWGJG/t9l8IrKd3leMm4AfAmxdrvE/FgHN8HfC2JNuBHwJnVHf547hL8ll6V2kuTbIFeD+9C0Mmon4DzK/Z2nVeDpwFfL37nhDgvcAqmIgaDjK/lmt4KHBJkiX0guofq+qqCfp/6CDzG9v6eaczSZIaMGkfiUuSNJEMbEmSGmBgS5LUAANbkqQGGNiSJA0pCzzYp6/t4Umu6x6e8uUkKwbpw8CWtFOSDyQ5abHHITXoYn76JjPz+Si9+7EfDXwA+PAgb/LXuiQBvXssV9WPF3scUqu6+8tfVVUv6tafB1wALKP3O+u/X1V3J9kAvKqqtnQ3aXmsqp6x0PE9w5amQJLVSe5Ockn3MdznkvxckvuTvC/J9cDpSS5O8rruPS9JckP3EISbkxzYPTjhr5Pc0h3nLYs8NWmcrQPeWVW/DLwH+GS3/Q7gtd3y7wAHJnnWQgebqDudSdqlFwBnV9VXk1wE/EG3/cmqegVAkpO7P/cBLgdeX1W3JHkGvbs+nU3vbOAlSfYFvprk2qr61h6fjTTGugfEvAz4p+5OpwD7dn++Bzg/ye8CXwG+DWxf6JgGtjQ9NlfVV7vlvwf+sFu+fI62LwC2VtUt0HsgAkCS3wKO3nEWDhwEHAkY2NJPehrwaPdksJ9QVQ8Cr4Gdwf7aqnpskANKmg79F6zsWH9ijraZo/2O7e+sqmO71xFVde0oBylNgu4fud9Kcjr0HhyT5JhueWmSHfl7HnDRIMc0sKXpsSrJS7vlM4Hrd9H2buCwJC8B6L6/3oveQ2felt4jJkny/CT7785BSy3oHuzzH8ALkmxJcjbwBuDsJHcAG4A1XfMTgXuS3AscAnxooD68SlyafN3Vq+vpfV/2MuCb9J469Q1gpqoe6tpdTO8q1891Yf0J4On0vr8+id6Vrh8EXk3vbHsbcNogH+dJGo6BLU2B/l83kdQePxKXJKkBnmFLktQAz7AlSWqAgS1JUgMMbEmSGmBgS5LUAANbkqQGGNiSJDXg/wG5hcf5Z9dvDwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for column in ['price', 'odometer']:\n",
    "  fig = plt.figure(figsize=(8,2))\n",
    "  sns.rugplot(x=column, data=df, height=1)\n",
    "  plt.title(column)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 301
    },
    "executionInfo": {
     "elapsed": 1316,
     "status": "ok",
     "timestamp": 1631959308492,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "yIcfA0L3Me1Y",
    "outputId": "c5f0b0fc-c38a-4818-93c9-4668e9efa76c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='age', ylabel='Count'>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEGCAYAAABYV4NmAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAqP0lEQVR4nO3de5SddX3v8fd3X2bPfTKTTC5kAklICDcBNUVutbZYoLYV7dGueFTSHnpiLbZau3oO6FrFowuXrrZabYUWJYLWiohS8QZEoFqRBoJS7iEhCSSQyyQTMpfMZV++54/n2ZOdYTKZZPaz9352Pq+19trP/j2X/fvNJM93fpfn9zN3R0REpNwS1c6AiIjUJwUYERGJhAKMiIhEQgFGREQioQAjIiKRSFU7A7Vizpw5vnjx4mpnQ0QkVh577LG97t492T4FmNDixYvZsGFDtbMhIhIrZvbikfapiUxERCKhACMiIpFQgBERkUgowIiISCQUYEREJBIKMCIiEgkFGBERiURkAcbMFpnZg2b2rJk9bWYfDtM/YWYvm9nj4ettJedcZ2abzWyjmV1ekv5GM3sy3PdFM7MwPWNm3wrT15vZ4pJzVpvZpvC1OqpyiojI5KKsweSAv3L3M4ALgGvM7Mxw3+fd/bzw9SOAcN8q4CzgCuBGM0uGx98ErAGWh68rwvSrgf3uvgz4PPDZ8FpdwPXAm4DzgevNrDPCsk7LL17Yy/k3/IQDw9lqZ0VEJHKRBRh33+nuvwy3B4BngYVTnHIlcLu7j7r7VmAzcL6ZLQDa3f1hD1ZH+xrwjpJzbgu37wQuDWs3lwPr3L3P3fcD6zgUlKrmv7b0sWdglG17h6qdFRGRyFWkDyZsuno9sD5M+pCZPWFma0tqFguB7SWn7QjTFobbE9MPO8fdc8ABYPYU15qYrzVmtsHMNvT29h5/AadpaxhYdvePRP5dIiLVFnmAMbNW4DvAR9y9n6C561TgPGAn8PfFQyc53adIP95zDiW43+zuK919ZXf3pHO1ldWW3kEA9gyMRv5dIiLVFmmAMbM0QXD5hrt/F8Ddd7t73t0LwJcJ+kggqGUsKjm9B3glTO+ZJP2wc8wsBXQAfVNcq2rcfbwGs0c1GBE5AUQ5isyAW4Bn3f1zJekLSg57J/BUuH03sCocGbaEoDP/EXffCQyY2QXhNa8CvldyTnGE2LuAB8J+mnuBy8ysM2yCuyxMq5rd/aMcHMuPb4uI1Lsop+u/GHg/8KSZPR6mfQx4j5mdR9BktQ34AIC7P21mdwDPEIxAu8bd8+F5HwRuBZqAH4cvCALY181sM0HNZVV4rT4z+xTwaHjcJ929L5JSTtOWvYPj27sHVIMRkfoXWYBx958zeV/Ij6Y45wbghknSNwBnT5I+Arz7CNdaC6ydbn6jVmweO2NBu2owInJC0JP8FbKld4jGdIJzezroVQ1GRE4ACjAVsnXvEItntzC/o5G9g2Nk84VqZ0lEJFIKMBWyde8QS7tbmNfeCECvhiqLSJ1TgKmQfYOjzG1rZF57BtDDliJS/xRgKsDdGRrL05JJMrctqMGoo19E6p0CTAWM5grkC05LJsXcsAajjn4RqXcKMBUwNJoDoKUhxeyWDMmEqQYjInVPAaYChkaD50VbMimSCaO7NcMu9cGISJ1TgKmAobFiDSZY3qans4mX+g5WM0siIpFTgKmA8SayTDBxwpI5LeNP9ouI1CsFmAoYGis2kQU1mKXdrfQOjDIwopUtRaR+KcBUwGQ1GEC1GBGpawowFVA6igxgafdrA4y786Mnd/LT56NfWVNEpBKinK5fQhNrMCd3NWMWTIAJMDia48//7Zc8uLGXtsYU/3XdpePHiojElWowFTCxD6YxnaSns2m8BnPbL7bx4MZe/uiixQyM5Pj2hu1Vy6uISLkowFTA0GiOVMJoSB76cS+Z08qWvYOMZPN89aFtvPm0bj7x9rN4w8mzWPvQNvIFr2KORURmTgGmAg6O5WnJpAhWfA4sndPC1t4h7vrVy+wdHOUDb14KwNWXLOWlvoP84oW91cquiEhZKMBUwOBobvwhy6Ilc1oYGsvzsbue5HULO7jo1NkAXLJ8DgBPvnyg4vkUESkn9SRXwNBo7jWd9r++fA7nLprFJctms/rCxeO1m46mNAtnNbFx10A1sioiUjYKMBUwNJaneUKAWdrdyveuuXjS40+f38ZzOxVgRCTe1ERWAUOjOVozyaMfGFoxv40XegcZy2lZZRGJLwWYChgazY0/ZDkdpy9oJ1dwXugdjDBXIiLRUoCpgKGx1/bBTOWM+W0APLerP6osiYhETgGmAg6O5scfspyOJXNaaEgm1A8jIrGmAFMBg8fYRJZKJlg2t5VnNZJMRGJMASZiuXyB0VzhmOcWWz6vlS3qgxGRGFOAiVhxHrLmhuk3kQEs6Ghid/8IBU0ZIyIxpQATsYPhcsmtx1iDOWlWI9m8s29oLIpsiYhETgEmYsWp+ic+aHk089sbAdh5YLjseRIRqQQFmIgNjgZNZMfyoCUETWQAOw+MlD1PIiKVoAATsYMTVrOcrgWzghrMLgUYEYkpBZiIDU5YzXK6upobaEgmeEVNZCISU5EFGDNbZGYPmtmzZva0mX04TO8ys3Vmtil87yw55zoz22xmG83s8pL0N5rZk+G+L1o49bCZZczsW2H6ejNbXHLO6vA7NpnZ6qjKeTQHx1ezPLYAk0gY8zsaVYMRkdiKsgaTA/7K3c8ALgCuMbMzgWuB+919OXB/+Jlw3yrgLOAK4EYzK3Zc3ASsAZaHryvC9KuB/e6+DPg88NnwWl3A9cCbgPOB60sDWSWN12COcZgywPyORna+qgAjIvEUWYBx953u/stwewB4FlgIXAncFh52G/COcPtK4HZ3H3X3rcBm4HwzWwC0u/vD7u7A1yacU7zWncClYe3mcmCdu/e5+35gHYeCUkUNHWcTGcBJHY3s7FcTmYjEU0X6YMKmq9cD64F57r4TgiAEzA0PWwhsLzltR5i2MNyemH7YOe6eAw4As6e41sR8rTGzDWa2obe3dwYlPLL+kSzJhB3zg5YA8zua2HVAD1uKSDxFHmDMrBX4DvARd59qemCbJM2nSD/ecw4luN/s7ivdfWV3d/cUWTt+AyM52hpT4ytWHgs9bCkicRZpgDGzNEFw+Ya7fzdM3h02exG+7wnTdwCLSk7vAV4J03smST/sHDNLAR1A3xTXqrhigDkeethSROIsylFkBtwCPOvunyvZdTdQHNW1GvheSfqqcGTYEoLO/EfCZrQBM7sgvOZVE84pXutdwANhP829wGVm1hl27l8WplXcwEiWtkz6uM4tPmz5ijr6RSSGju9P6+m5GHg/8KSZPR6mfQz4DHCHmV0NvAS8G8DdnzazO4BnCEagXePu+fC8DwK3Ak3Aj8MXBAHs62a2maDmsiq8Vp+ZfQp4NDzuk+7eF1E5p9Q/gxrM3PYMAHsHR8uZJRGRiogswLj7z5m8LwTg0iOccwNwwyTpG4CzJ0kfIQxQk+xbC6ydbn6j0j+cZVFX83Gd29ncAMC+QfXBiEj86En+iM2kD6YhlaCjKU3fkGowIhI/CjARGxjJ0t54fH0wALNbGtirUWQiEkMKMBFydwZHj78GAzC7tYE+NZGJSAwpwERoaCxPwZlRgOlqaWCfmshEJIYUYCLUP5wFmFkTWWuGPjWRiUgMKcBEaGAkmIesbYZ9MH1DY5ouRkRiRwEmQgMjQQ1mRn0wLQ0UHF4Na0MiInGhABOhQzWYGfTBtAYPW+7Tw5YiEjMKMBHqH6/BHH8T2ZyW8GFL9cOISMwowESoP6zBtM+oBqOn+UUknhRgIlTsg2lvmkknf9BEpqf5RSRuFGAiNDCSI500Mqnj/zF3NgfBaa9qMCISMwowERoYydLWmD6uxcaKUskEnc1pPQsjIrGjABOhY53octHJp2Bmr3nt2b6Ff771G5PuK30tOvmUCEsjInJsolwP5oR3rAFmx/aX+Nx9G1+TfudjO+DU0/jYNX885fkfvWzFMedRRCQqqsFEqH94ZjMpFzWlkwyP5Y9+oIhIDVGAidBM1oIp1dSQZDirACMi8aIAE6FiJ/9MFQNMwTUfmYjEhwJMhMpVg2lOJwEYUS1GRGJEASYihYIzOJYrWw0G4KD6YUQkRhRgymTiEONUUwvu8ImPX3vU4cXF15E0hwFGHf0iEicaplwmE4cYD4xkWfvQNt79Z9dx9sLPTOsaRxpm3BQ2kamjX0TiRDWYiIzlCgA0zGCamKIm1WBEJIYUYCIyli9fgGlMqw9GROJHASYi4zWY5Mx/xAmz4GFLNZGJSIwowESknE1kEDSTHRzLleVaIiKVoAATkXI2kUHwLIxqMCISJwowESnWYDJlaCKD8Gl+9cGISIwowESkGGDSZWwiU4ARkThRgInIaL5AOmkkZrDYWKmmdJKRXIF8QfORiUg8KMBEZCxXKMsIsqLi0/yaj0xE4kIBJiJjuULZOvjh0NP8ehZGROIisgBjZmvNbI+ZPVWS9gkze9nMHg9fbyvZd52ZbTazjWZ2eUn6G83syXDfFy2ctMvMMmb2rTB9vZktLjlntZltCl+royrjVMby5Q0wzQ3BrD4aSSYicRFlDeZW4IpJ0j/v7ueFrx8BmNmZwCrgrPCcG80sGR5/E7AGWB6+ite8Gtjv7suAzwOfDa/VBVwPvAk4H7jezDrLX7yplb0Go+liRCRmIgsw7v4zoG+ah18J3O7uo+6+FdgMnG9mC4B2d3/Y3R34GvCOknNuC7fvBC4NazeXA+vcvc/d9wPrmDzQRarcfTCHpuzXw5YiEg/V6IP5kJk9ETahFWsWC4HtJcfsCNMWhtsT0w87x91zwAFg9hTXeg0zW2NmG8xsQ29v78xKNUG5m8gaUwmSZgypBiMiMVHpAHMTcCpwHrAT+PswfbKxvD5F+vGec3ii+83uvtLdV3Z3d0+R7WM3liuQSSaPfuA0mRktmSRDo6rBiEg8VDTAuPtud8+7ewH4MkEfCQS1jEUlh/YAr4TpPZOkH3aOmaWADoImuSNdq2Lcvex9MAAtmZQCjIjExrTugGZ28XTSpnGdBSUf3wkUR5jdDawKR4YtIejMf8TddwIDZnZB2L9yFfC9knOKI8TeBTwQ9tPcC1xmZp1hE9xlYVrFZPOOU755yIpaGlIMjaqJTETiYborWv4j8IZppI0zs28CbwHmmNkOgpFdbzGz8wiarLYBHwBw96fN7A7gGSAHXOPuxTvpBwlGpDUBPw5fALcAXzezzQQ1l1XhtfrM7FPAo+Fxn3T36Q42KIvxiS7L2MkP0JJJ8tJ+1WBEJB6mDDBmdiFwEdBtZh8t2dUOTNnB4O7vmST5limOvwG4YZL0DcDZk6SPAO8+wrXWAmunyl+Uyj1Vf1FLJsVYrkA2XyBd5uAlIlJuR7tLNQCtBIGoreTVT9AsJZOIKsC0ZoK/B9QPIyJxMGUNxt1/CvzUzG519xcrlKfYK/daMEUt4wEmz6zmsl5aRKTsptsHkzGzm4HFpee4+29Fkam4K+dyyaVawoctB1WDEZEYmG6A+Tbwz8BXAA1jOorxxcaiqsHoaX4RiYHpBpicu98UaU7qSFRNZJlUgmTC1AcjIrEw3Tvg983sz8xsgZl1FV+R5izGxlezLHMTmZnRmtGzMCISD9OtwRQfaPzrkjQHlpY3O/Uhmy+QNCOZKM9qlqVaGjRdjIjEw7QCjLsviToj9SSXd1LJ8gcXCPphegdHI7m2iEg5TSvAmNlVk6W7+9fKm536kC1E9yBkSybF1r1DuDvh2msA9A2NMevX30eh4CQiqDmJiByr6TaR/VrJdiNwKfBLgvVZZIJsvkAqopt8R1OaXMEZzubHV7kE+I+Ne+i4aBVPv9LP63o6IvluEZFjMd0msj8v/WxmHcDXI8lRHYiyiay9KfiVHRjOjgeYHfsPsn3/MAA/37xXAUZEasLxtuMcJJjxWCaRK3hkTWQdjWkgCDBF67f00dyQJLtvOw9t3hvJ94qIHKvp9sF8n0OLdiWBM4A7ospU3EXdRAbQPxyMJBvN5dnx6jBvWtLFuoce45F5pzCSzdOYLt9iZyIix2O6fTB/V7KdA1509x1HOvhElyt4ZDf4VDJBSybJq8NjQNC5DzC3LcPwtscZ+7V38NiL+7l42ZxIvl9EZLqm1Y4TTnr5HMFMyp3AWJSZirtcvkA6wpFcHY3p8RrMvsHgVzG7NcPo9qdIJ42fq5lMRGrAdFe0/EPgEYL1V/4QWG9mmq7/CLJ5JxXhei0dTenxPph9Q2OkEkZ7YwrPjrBifhtPvXwgsu8WEZmu6TaRfRz4NXffA2Bm3cBPgDujylic5fIF0hGNIoMgwDy7a4BcvsC+wVFmtzaMPxNz2rw2dfSLSE2Y7p/ZiWJwCe07hnNPONlC9DUYgP6RHPuGxpjdkhnft2JeG7v7R3n1oFoxRaS6pnsXvMfM7jWzPzKzPwJ+CPwoumzFl7uTL3hko8gA2sMAs6t/hINjeWa3NozvO21+GwDP7x6M7PtFRKZjygBjZsvM7GJ3/2vgX4BzgHOBh4GbK5C/2MkVgtHcUT0HA4dqMMW+ltkthwLMinlBgNm4eyCy7xcRmY6j3QX/ARgAcPfvuvtH3f0vCWov/xBt1uIpG64FE2UNprkhyWnzWtl5YAQIRpAVLehopK0xxcZd/ZF9v4jIdBytk3+xuz8xMdHdN5jZ4miyFG+5fFCDiWqqGAjWhbnirPmcMnuAvqGx8aWUi/tWzGvj+V1qIhOR6jpagGmcYl9TOTNSL4o1mCibyCAIJGcuaJ9032nz2/jhEztfM+OyiEglHe0u+KiZ/e+JiWZ2NfBYNFmKt2IfTJRNZEezYl4bB4az7BnQujEiUj1Hq8F8BLjLzN7LoYCyEmgA3hlhvmKr2EQWdQ1mKqcVO/p3DTCvfapKqIhIdKYMMO6+G7jIzH4TODtM/qG7PxB5zmIqWwg7+SPsgzmaFfMPBZg3n9ZdtXyIyIltuuvBPAg8GHFe6sJ4J3+iejWYrpYGutsyGqosIlWlp/HLLDfeyV/dzvUV89p4XgFGRKpIAabMssVO/ir2wUDQD/P87gEKBT/6wSIiEVCAKbPxYcpVHEUGsGJ+KyPZAi/1HaxqPkTkxKUAU2aHHrSs7o92xfzgGRn1w4hItSjAlFmuUMAMqlyBYfncVgCe36UAIyLVEVmAMbO1ZrbHzJ4qSesys3Vmtil87yzZd52ZbTazjWZ2eUn6G83syXDfFy18NN3MMmb2rTB9fenUNWa2OvyOTWa2OqoyTiabd9KJRNWfoG/JpFjU1cRzqsGISJVEWYO5FbhiQtq1wP3uvhy4P/yMmZ0JrALOCs+50cyKE2zdBKwBloev4jWvBva7+zLg88Bnw2t1AdcDbwLOB64vDWRRy+ULVX0GptRpc9vYrGn7RaRKIgsw7v4zoG9C8pXAbeH2bcA7StJvd/dRd98KbAbON7MFQLu7P+zuDnxtwjnFa90JXBrWbi4H1rl7n7vvB9bx2kAXmVzBq/oUf6ll81rZsndwfOi0iEglVfpOOM/ddwKE73PD9IXA9pLjdoRpC8PtiemHnePuOeAAMHuKa72Gma0xsw1mtqG3t3cGxTokmy9UdR6yUqfNbSObd17USDIRqYLa+FMbJrsj+xTpx3vO4YnuN7v7Sndf2d1dnilVcnmvmSay5fOCjv5N6ocRkSqodIDZHTZ7Eb7vCdN3AItKjusBXgnTeyZJP+wcM0sBHQRNcke6VkVkCwXSVZwmptSp3cUAo34YEam8St8J7waKo7pWA98rSV8VjgxbQtCZ/0jYjDZgZheE/StXTTineK13AQ+E/TT3ApeZWWfYuX9ZmFYRtVSDacmk6Ols4vk9CjAiUnnTmuzyeJjZN4G3AHPMbAfByK7PAHeE68m8BLwbwN2fNrM7gGeAHHCNu+fDS32QYERaE/Dj8AVwC/B1M9tMUHNZFV6rz8w+BTwaHvdJd5842CAyQYCpjRoMBM/DqIlMRKohsgDj7u85wq5Lj3D8DcANk6Rv4NBSAaXpI4QBapJ9a4G1085sGQVNZLVRg4FgTrKHNu8Lh0/XTuATkfqnO06Z1VoNZtncVsbyBY0kE5GKq507YZ3IFQpVn6q/1JknBXOSPfNKf5VzIiInGgWYMnJ3snmv6mJjEy2f20ZDMsFTLx+odlZE5ARTO3fCOpAfXwumdmowDakEK+a38dQrCjAiUlkKMGWUDafqr5WpYorOXtjBUy/3E4ziFhGpjNq6E8ZctkaWS57o7IXtHBjOsmP/cLWzIiInEAWYMjoUYGrrx/q6hR0A6ocRkYqqrTthzGVrsA8GgmdhUgnjSQUYEakgBZgyKk6LXytzkRU1ppMsn9emACMiFVVbd8KYq9VOfoDzFs3i8e2vUiioo19EKqP27oQxVqud/ABvOHkWAyM5tuzVxJciUhkKMGVU9U5+S2Bmk77+5+UXAXDeW//giMeYGYtOPqU6eReRuhPZZJcnoly+yp38XuBz922cfJc7//KzLbzlT/6GS8/40hEv8dHLVkSVOxE5wagGU0ZVr8FMwcyY397Irv6RamdFRE4QtXcnjLFiJ3+qhqbrLzW/o5F9g2OM5QrVzoqInAAUYMooG86kHCy+WXsWdDTiwG7VYkSkAhRgyiibL9TUTMoTzW1vBBRgRKQyavduGEO5vNfkEOWipnSSjqY0u/tHq50VETkBKMCUUTZfqMkO/lLz2jPsHlANRkSiV9t3w5jJ5j0GAaaRgZEcQ6O5amdFROpcbd8NYyabL9TcRJcTzWsL+2FUixGRiCnAlFEuBjWYue0ZDNQPIyKRq+27YcwEfTC1XYNJJxN0tTZoJJmIRE4BpoyC52Bq/0c6r62R3f0jWkJZRCJV+3fDGMnmvebWgpnMvPYMI9kC/SPq6BeR6NT+3TBGcjHo5AeYrwcuRaQCFGDKJZGi4LU50eVEs1szJBOmACMikar9u2FMJNIZoDYXG5somTC6WzMaSSYikVKAKRNrCJqd4lCDgaAfZs/ACAV19ItIROJxN4wBSwcBJg59MBA80Z/NO/uHxqqdFRGpUwowZZJIx6sGM78jyO8rr6ofRkSiEY+7YQxYzALMrKY0rZkUO/YfrHZWRKROVeVuaGbbzOxJM3vczDaEaV1mts7MNoXvnSXHX2dmm81so5ldXpL+xvA6m83sixau9GVmGTP7Vpi+3swWR16mGHXyQ7CEck9nE9v3D+uBSxGJRDX/3P5Ndz/P3VeGn68F7nf35cD94WfM7ExgFXAWcAVwo5klw3NuAtYAy8PXFWH61cB+d18GfB74bNSFKTaR1fKCYxP1dDYxnM2zT/0wIhKBWrobXgncFm7fBryjJP12dx91963AZuB8M1sAtLv7wx78Cf61CecUr3UncKlFvI7xoSayeNRgABZ1NgOwY/9wlXMiIvWoWgHGgfvM7DEzWxOmzXP3nQDh+9wwfSGwveTcHWHawnB7Yvph57h7DjgAzI6gHOMONZHVUsyeWntTmo6mNNv71A8jIuWXqtL3Xuzur5jZXGCdmT03xbGTVQl8ivSpzjn8wkFwWwNw8sknT53jo0jE7DmYop7OJjbtGaTgTiLaSp6InGCqcjd091fC9z3AXcD5wO6w2YvwfU94+A5gUcnpPcArYXrPJOmHnWNmKaAD6JskHze7+0p3X9nd3T2jMsXtOZiiRZ3NjOUK9A7oqX4RKa+KBxgzazGztuI2cBnwFHA3sDo8bDXwvXD7bmBVODJsCUFn/iNhM9qAmV0Q9q9cNeGc4rXeBTzgEQ+VsnQjyYTFrhbQ09kEwHYNVxaRMqtGE9k84K6wzz0F/Ju732NmjwJ3mNnVwEvAuwHc/WkzuwN4BsgB17h7PrzWB4FbgSbgx+EL4Bbg62a2maDmsirqQiUamsmk4tU8BtCSSdHV0sCO/cOsPKXauRGRelLxAOPuW4BzJ0nfB1x6hHNuAG6YJH0DcPYk6SOEAapSEo0tsQwwAIs6m3hmZz/5gp6HEZHyiecdsQYlGlvJpJJHP7AG9XQ2k807uzR9v4iUkQJMmSQyLWTS8fxx9nQ2kTDYtneo2lkRkToSzztiDQpqMPH8cTamkyzsbGJz72C1syIidSSed8QalMi0xLaJDODU7lZePZglPXvR0Q8WEZkGBZgycPdYd/IDnDqnFYCm0y6sck5EpF5U60n+ujKSLWDJdGz7YABaG1PMb29kbMUluDsRT90mIieA+N4Ra8iB4SxArJvIAE5f0EbDvKVseHF/tbMiInVAAaYM+keKASbeP84zF7STH+7n5p9tqXZWRKQOxPuOWCP6h+sjwKSTCQZ+9SN+8uxuXtCIMhGZoXjfEWvEeBNZOt5NZAADj/2AhmSCGx98odpZEZGYU4Apg3ppIgMoHHyV911wCnf9agdbVIsRkRmI/x2xBvQP5wBojHknf9Gf/sapZFJJvnD/pmpnRURiTAGmDIpNZA11UIMB6G7LcNWFp/D9/35Fq12KyHGrjztilfUPZymMDZNM1M+zI3988RISZnz1oW3VzoqIxJQCTBn0j2QpjNbXRJHzOxr53XMWcMeG7QyEfUwiIsdCAaYM+odzFEbqr0P86kuWMDia444NO6qdFRGJIQWYMjgwnKUwUl81GIBzembxhpNn8Y31LxLxitMiUocUYMqgHpvIit77plPY0jvEwy/sq3ZWRCRmFGDKoH8kW5dNZAC/e84CZjWn+cb6l6qdFRGJGQWYMjhwsH4DTGM6ybvf2MO9T+/SkGUROSYKMDNUKDgDo7m6bSIDuPqSpSTM+NKDm6udFRGJEQWYGRocy+FOXQeY+R2NvOf8Rdz52A7VYkRk2hRgZqhQcH7n7Plk99Z3H8UH37KMRML42F1Pki9oRJmIHJ0CzAzNam7gpve9kZGtv6x2VsrDEpjZa14LZjWx8wdf4D837WXOW/9k0mOKr0Unn1LtUohIDdCSyXI4L/C5+zYecfeDG/fwBH/A7713DW84uXPSYz562YqociciMaIAI8fkN5Z3MzyW5z837SWVMM7pmVXtLIlIjVKAkWOSSBiXnzWfXGEnD27sJZ1McMaC9mpnS0RqkPpg5JglE8bbzp7Poq4m7ntmN/c9s2t8yQIRkSLVYOS4pJIJfv+ck1i/tY/HX3qVZ3cOsKCjkXN7ZkGiPhZeE5GZUQ1Gjls6meCSZXNYfdEpXHzqbIazee55ehcLP/Blbv7ZC+NLSYvIiUkBRmasrTHNysVdXHXBKfz+uQvI7d/Jp3/0HBd++n4++f1n9HCmyAlKTWRSNmbG0jmt7L794zy541Vu+flWvvbwNr76i62c2zOLC5bOZtncVrpa0rRm0rRmUsGrMXivlyWnRSSgACPlZwleFw5fTrbOpvWc32b9qSv51bZlWPLI/+QK2VF8bBj3AsmE0XPSSZjB7JYGFnU1s6irmZO7mjmlq5nTF7TT1dJQoQLNTDZf4MBwloOjeRrTCTpbGkgnFUyl/tV1gDGzK4AvAEngK+7+mSpn6cRwhIc18wWnfyTLaLbAWL7AWK7kPVdgNJcnm3fcnYd//G0u/PWrKRSc3sFRnnr5APc8tYtcyTQ1CzoaOeukds48qYOzTmrn9PltJBM2ft2h0Ry9A6O8uO8g2/YdpHdglFyhQEtDis6WNF3NDcxqbqCrpYHOlgY6m9N0NgfbB0dzbN9/kO19w7zQO8jzuwfYtGeQvQOjZPNOU0OSlkySloYUY/kCI2N5DmbzDI/lyeYLpJMJ0skE7s7QWP6wn0PC4KRZTSzqDALmybOb6elsCoJoZzNzWhsws8h/TSJRq9sAY2ZJ4EvAbwM7gEfN7G53f6a6OTtxJRNGZ/P0ah0//MiX+Pt7/vHwREuQbJ1NuuskGuYtZWjuqbw4bynruhZiRxm5lj94gPxgH57PkmhoItHUTqKpDbOj1yS8kCe3fyfZvS+RG9iLF3IkUg1YQzOJhiY8n8Wzo0ENLDsChRwkUuO1NR87SG7oAJ4dxlIZkm2z6e+Yx+ZZ80l1zCPV2nXY9xWyo+T795Ddv5PcqzvJD71KR0cH1117LQkDMxjLFRjJFhjJ5tk9MBoGvgLZgpPNFYLtfAEHWjMpOprStDemaW9Khe9pOprSNKQS5ApOoeDkC07BnVzBMYKlGjKpBJl0gsZUkkw6QSaVpDGdoLkhaNZsa0zRkklNWiPL5gsMZ/OMjOUZyQbbw2EQHs3laUonaQmbSZOJQwG1NLYWA21puC3ud4dc3hkLy1rczuULJBJGMmEkLXhPJY1UwkglEqSSNv4HQCpppBMJ0snguBMpsLs7BWd8tdpUBLXqug0wwPnAZnffAmBmtwNXAgowcXCUKWtKZfMF9g6O0jc0BjB+Y/nq36zhw3/7VWY1pcmkXxuACu6M5oLax3A2z0jxBpjNk04kaG9Kc9Ofv51P/+s6UsnTj7soH71sxZRlyeYL9A9n6R/Jhe9ZDgx3cWD4VA4MZ8nmgxvAZ+957rDz3At4doz80H7yQ/vxXBYKWTyfw/O5INA5WKaZRKaFZFPb+Haioem4yzOZhmQCx4OZxcMbVz2YGG8mhp/JAtJrj5m4f+qLThbijnaNifvdwQl/DyXb7o6H+0udt2gW/37NxZN888xYva61bmbvAq5w9z8JP78feJO7f6jkmDXAmvDjCmB6d7TJzQH2zuD8WqFy1J56KYvKUXvKUZZT3L17sh31XIOZ7A+Bw6Kpu98M3FyWLzPb4O4ry3GtalI5ak+9lEXlqD1Rl6Weh7LsABaVfO4BXqlSXkRETjj1HGAeBZab2RIzawBWAXdXOU8iIieMum0ic/ecmX0IuJdgmPJad386wq8sS1NbDVA5ak+9lEXlqD2RlqVuO/lFRKS66rmJTEREqkgBRkREIqEAM0NmdoWZbTSzzWZ2bbXzM11mtsjMHjSzZ83saTP7cJjeZWbrzGxT+N5Z7bxOh5klzexXZvaD8HNcyzHLzO40s+fC382FcSyLmf1l+O/qKTP7ppk1xqUcZrbWzPaY2VMlaUfMu5ldF/7/32hml1cn1691hHL8bfhv6wkzu8vMZpXsK3s5FGBmoGQ6mt8BzgTeY2ZnVjdX05YD/srdzwAuAK4J834tcL+7LwfuDz/HwYeBZ0s+x7UcXwDucffTgXMJyhSrspjZQuAvgJXufjbBIJtVxKcctwJXTEibNO/h/5lVwFnhOTeG94VacCuvLcc64Gx3Pwd4HrgOoiuHAszMjE9H4+5jQHE6mprn7jvd/Zfh9gDBjWwhQf5vCw+7DXhHVTJ4DMysB/hd4CslyXEsRzvwZuAWAHcfc/dXiWFZCEaoNplZCmgmeAYtFuVw958BfROSj5T3K4Hb3X3U3bcCmwnuC1U3WTnc/T53z4Uf/4vg+UCIqBwKMDOzENhe8nlHmBYrZrYYeD2wHpjn7jshCELA3Cpmbbr+Afg/QKEkLY7lWAr0Al8Nm/u+YmYtxKws7v4y8HfAS8BO4IC730fMyjHBkfIe53vA/wJ+HG5HUg4FmJk56nQ0tc7MWoHvAB9x9/5q5+dYmdnvAXvc/bFq56UMUsAbgJvc/fXAELXbjHREYf/ElcAS4CSgxczeV91cRSaW9wAz+zhBM/k3ikmTHDbjcijAzEysp6MxszRBcPmGu383TN5tZgvC/QuAPdXK3zRdDLzdzLYRNFH+lpn9K/ErBwT/nna4+/rw850EASduZXkrsNXde909C3wXuIj4laPUkfIeu3uAma0Gfg94rx96EDKScijAzExsp6OxYJ7xW4Bn3f1zJbvuBlaH26uB71U6b8fC3a9z9x53X0zw83/A3d9HzMoB4O67gO1mtiJMupRgeYm4leUl4AIzaw7/nV1K0McXt3KUOlLe7wZWmVnGzJYAy4FHqpC/abFgEcb/C7zd3Q+W7IqmHO6u1wxewNsIRmO8AHy82vk5hnxfQlAFfgJ4PHy9DZhNMEpmU/jeVe28HkOZ3gL8INyOZTmA84AN4e/l34HOOJYF+H/Ac8BTwNeBTFzKAXyToO8oS/CX/dVT5R34ePj/fyPwO9XO/1HKsZmgr6X4f/6foyyHpooREZFIqIlMREQioQAjIiKRUIAREZFIKMCIiEgkFGBERCQSCjAiIhIJBRgREYmEAoxIDTCzfzezx8I1VNaEaVeb2fNm9h9m9mUz+6cwvdvMvmNmj4avi6ube5HJ6UFLkRpgZl3u3mdmTQRTEF0OPEQwF9kA8ADw3+7+ITP7N+BGd/+5mZ0M3OvBuj4iNSVV7QyICAB/YWbvDLcXAe8HfurufQBm9m3gtHD/W4Ezg2m+AGg3szYP1vURqRkKMCJVZmZvIQgaF7r7QTP7D4L5oI5UK0mExw5XJIMix0l9MCLV1wHsD4PL6QRLWDcDv2FmneGqkP+j5Pj7gA8VP5jZeZXMrMh0KcCIVN89QMrMngA+RbCU7cvApwlWGf0JwbT9B8Lj/wJYaWZPmNkzwJ9WPssiR6dOfpEaZWat7j4Y1mDuAta6+13VzpfIdKkGI1K7PmFmjxOsqbKVYH0YkdhQDUZERCKhGoyIiERCAUZERCKhACMiIpFQgBERkUgowIiISCT+P9OfzJWRtoQZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(data=df, x='age', bins=18, kde=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "izztXKpllMss"
   },
   "source": [
    "## 3.데이터 클리닝"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KywGh9sjtYl9"
   },
   "source": [
    "### 범주형 데이터"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 310
    },
    "executionInfo": {
     "elapsed": 1865,
     "status": "ok",
     "timestamp": 1631959310353,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "TqXkwW8JDxhH",
    "outputId": "8dc86646-5a53-4ae7-d4de-f9e5476b8fda"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20,5))\n",
    "sns.boxplot(x='manufacturer', y='price', data=df.fillna('n/a'))\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 308
    },
    "executionInfo": {
     "elapsed": 978,
     "status": "ok",
     "timestamp": 1631959311329,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "JFw4s19CNIPE",
    "outputId": "c8ab5271-0a3c-4dae-cafd-0041ce4ac136"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='fuel', ylabel='price'>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(data=df.fillna('n/a'), x='fuel', y='price')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "데이터를 클리닝하고 다시 boxplot 확인 필요해보임"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* manufacturer "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcbd51310>]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'manufacturer'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 10번째 이후는 다 others로 통합\n",
    "n_categorical = 10\n",
    "\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in counts.index[n_categorical:] else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "others       134392\n",
       "ford          79666\n",
       "chevrolet     64977\n",
       "toyota        38577\n",
       "honda         25868\n",
       "nissan        23654\n",
       "jeep          21165\n",
       "ram           17697\n",
       "gmc           17267\n",
       "dodge         16730\n",
       "Name: manufacturer, dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* region"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcc03f8b0>]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'region'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 5\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in counts.index[n_categorical:] else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "others          441016\n",
       "columbus          3818\n",
       "rochester         3592\n",
       "jacksonville      3504\n",
       "springfield       3293\n",
       "fayetteville      2990\n",
       "Name: region, dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcad9ffd0>]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'model'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcd1f59a0>]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.grid()\n",
    "plt.plot(range(len(counts[:20])), counts[:20])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 10\n",
    "others = counts.index[n_categorical:]\n",
    "\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* condition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfc3be8190>]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAD6CAYAAAC/KwBlAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAsO0lEQVR4nO3deXxW9Zn//9eVDQhrCBACCYvssmlBxD2KAlbcqk7x96s6U2cY/Wqn1plW6Warper4mzp1OtoyStVO6/J1FwWKQnTsIIoL+xYQJRBZDAIBAiS5fn/cJ/Qm3iQhd8K57zvv5+NxP3Lu65zPyXUZyZVzPuc+x9wdERGRY0kLOwEREUlsahQiIlIvNQoREamXGoWIiNRLjUJEROqlRiEiIvVqsFGYWaGZLTSz1Wa20sy+G8S7mtl8M1sffM2JGjPdzErMbK2ZTYqKjzGz5cG6h8zMgngbM3smiC82s35RY24Ivsd6M7uhWasXEZEGWUOfozCzfCDf3T80s47AB8AVwN8C5e5+n5ndCeS4+x1mdjLwFDAO6AW8AQx292ozew/4LvAu8DrwkLvPMbP/A4xy95vMbCpwpbt/08y6AkuAsYAH33uMu+86Vr7dunXzfv36NfW/B/v27aN9+/ZNHp8oUqUOUC2JKlVqSZU6IL5aPvjgg53u3j3mSnc/rhfwMnARsJZIAwHIB9YGy9OB6VHbzwPOCLZZExW/Fvhd9DbBcgawE7DobYJ1vwOurS+/MWPGeDwWLlwY1/hEkSp1uKuWRJUqtaRKHe7x1QIs8WP8Xj2uOYrglNCpwGIgz93LgmZTBvQINusNbI4aVhrEegfLdeNHjXH3KmA3kFvPvkRE5ATJaOyGZtYBeB64zd33BNMLMTeNEfN64k0dE53bNGAaQF5eHsXFxcfKrUEVFRVxjU8UqVIHqJZElSq1pEod0HK1NKpRmFkmkSbxR3d/IQhvM7N8dy8L5jG2B/FSoDBqeAGwNYgXxIhHjyk1swygM1AexIvqjCmum5+7zwRmAowdO9aLiorqbtJoxcXFxDM+UaRKHaBaElWq1JIqdUDL1dKYq54MeAxY7e6/ilr1ClB7FdINROYuauNTgyuZ+gODgPeC01N7zWx8sM/r64yp3dfVwILgnNk8YKKZ5QRXVU0MYiIicoI05ojiLOA6YLmZfRzEfgjcBzxrZjcCnwHXALj7SjN7FlgFVAG3uHt1MO5m4HGgHTAneEGkEf3BzEqIHElMDfZVbmb3AO8H293t7uVNK1VERJqiwUbh7u8Qe64AYMIxxswAZsSILwFGxIhXEjSaGOtmAbMaylNERFqGPpktIiL1avRVT6nuwKFqfv3megZ4TdipiIgkFB1RBHbtP8QfFm3iiZWHaj/cJyIiqFEc0atLO34weSgrvqjmxY+2hJ2OiEjCUKOI8q3xfRnQOY17Zq/ii4qDYacjIpIQ1CiipKcZ3x7RhoqDVdwze1XY6YiIJAQ1ijp6d0zj5vMG8NLHWyleu73hASIiKU6NIoZbLhjIgO7t+dGLK9h3sCrsdEREQqVGEUObjHTuu2oUW748wK/mrws7HRGRUKlRHMNp/bry/57eh9//5ROWbv4y7HREREKjRlGPOy4eSveObbjj+WUcrtYH8USkdVKjqEentpncffkI1ny+l5lvbww7HRGRUKhRNGDS8J5MHt6TX7+5nk927gs7HRGRE06NohF+fvlw2mSkMf2FZbq9h4i0OmoUjZDXqS3TLx7GuxvLeXbJ5oYHiIikEDWKRpp6WiHj+ndlxmur2b63Mux0REROGDWKRkpLM+79xkgqq2r4+Su6vYeItB6NeWb2LDPbbmYromLPmNnHwWtT7SNSzayfmR2IWvfbqDFjzGy5mZWY2UPBc7MJnq39TBBfbGb9osbcYGbrg9cNhGxA9w780wUDeW15GfNXbQs7HRGRE6IxRxSPA5OjA+7+TXc/xd1PAZ4HXohavaF2nbvfFBV/BJgGDApetfu8Edjl7gOBB4H7AcysK3AXcDowDrjLzHKOr7zmN+3cAQzJ68hPXlrB3srDYacjItLiGmwU7v42UB5rXXBU8DfAU/Xtw8zygU7uvsgjlw09CVwRrL4ceCJYfg6YEOx3EjDf3cvdfRcwnzoNKwxZGWncd9VItu2t5F/nrg07HRGRFhfvHMU5wDZ3Xx8V629mH5nZW2Z2ThDrDZRGbVMaxGrXbQZw9ypgN5AbHY8xJlSn9snhhjP68d+LP2XJppg9VEQkZcT7zOxrOfpoogzo4+5fmNkY4CUzGw5YjLG1H0g41rr6xhzFzKYROa1FXl4excXFjcs+hoqKikaNPz3bebWN8U9/eJefn9WOzLRY6YansXUkA9WSmFKlllSpA1quliY3CjPLAL4BjKmNuftB4GCw/IGZbQAGEzkaKIgaXgBsDZZLgUKgNNhnZyKnukqBojpjimPl4u4zgZkAY8eO9aKiolibNUpxcTGNHd+2YDt/9/j7rKzpzW0XDG7y92wJx1NHolMtiSlVakmVOqDlaonn1NOFwBp3P3JKycy6m1l6sHwSkUnrje5eBuw1s/HB/MP1wMvBsFeA2iuargYWBPMY84CJZpYTTGJPDGIJ4/yhPbhsdC/+c2EJ67ftDTsdEZEW0ZjLY58CFgFDzKzUzG4MVk3lq5PY5wLLzGwpkYnpm9y99iT+zcCjQAmwAZgTxB8Dcs2sBLgduBMgGHcP8H7wujtqXwnjp5eeTPs2Gdz5wnJqanR7DxFJPQ2eenL3a48R/9sYseeJXC4ba/slwIgY8UrgmmOMmQXMaijHMHXr0IYfX3Iy//J/l/LHxZ9y3Rn9wk5JRKRZ6ZPZzeCqr/Xm7IHduH/uWsp2Hwg7HRGRZqVG0QzMjBlXjqCqpoafvLRSd5gVkZSiRtFM+ua253sXDuaN1duYs+LzsNMREWk2ahTN6Maz+zO8Vyd++vJKdu/X7T1EJDWoUTSjjPQ07r9qFLv2H+LeOavDTkdEpFmoUTSzEb078/dn9+fp9zezaMMXYacjIhI3NYoWcNuFg+nTNZsfvricysPVYacjIhIXNYoW0C4rnV9eOZJPdu7joTfXNzxARCSBqVG0kLMHdeOqrxUw8+2NrNq6J+x0RESaTI2iBf34kmF0bpfJ9BeWUa3be4hIklKjaEE57bP46aUns7R0N4//76aw0xERaRI1ihZ22ehenD+kO//fvLVsLt8fdjoiIsdNjaKFmRm/uHIkZvCjl1bo9h4iknTUKE6A3l3a8f1JQ3h73Q5e/nhrwwNERBKIGsUJcv0Z/TilsAt3z15F+b5DYacjItJoahQnSHqacd9VI9lz4DC/mL0q7HRERBpNjeIEGtqzEzedN4AXPtrC2+t2hJ2OiEijNOZRqLPMbLuZrYiK/czMtpjZx8Hr61HrpptZiZmtNbNJUfExZrY8WPdQ8OxszKyNmT0TxBebWb+oMTeY2frgVftc7aR26wUDOalbe3744nL2H6oKOx0RkQY15ojicWByjPiD7n5K8HodwMxOJvIs7eHBmIfNLD3Y/hFgGjAoeNXu80Zgl7sPBB4E7g/21RW4CzgdGAfcZWY5x11hgmmbmc693xhJ6a4DPDh/XdjpiIg0qMFG4e5vA+WN3N/lwNPuftDdPwFKgHFmlg90cvdFHrk+9EngiqgxTwTLzwETgqONScB8dy93913AfGI3rKRz+km5XDuuD4+98wnLSr8MOx0RkXplxDH2VjO7HlgC/HPwy7w38G7UNqVB7HCwXDdO8HUzgLtXmdluIDc6HmPMUcxsGpGjFfLy8iguLm5yURUVFXGNb6yzOzpzsoxbn1jET89oS0aaNev+T1QdJ4JqSUypUkuq1AEtV0tTG8UjwD2AB1//Dfg2EOu3ndcTp4ljjg66zwRmAowdO9aLiorqSb1+xcXFxDP+eKTnl3HTf3/I+rQ+3Fw0oFn3fSLraGmqJTGlSi2pUge0XC1NuurJ3be5e7W71wD/RWQOASJ/9RdGbVoAbA3iBTHiR40xswygM5FTXcfaV8qYPCKfiSfn8e9vrGPTzn1hpyMiElOTGkUw51DrSqD2iqhXgKnBlUz9iUxav+fuZcBeMxsfzD9cD7wcNab2iqargQXBPMY8YKKZ5QST2BODWEq5+/IRZKWn8cMXl+v2HiKSkBpzeexTwCJgiJmVmtmNwL8Gl7ouA84Hvgfg7iuBZ4FVwFzgFnevfcTbzcCjRCa4NwBzgvhjQK6ZlQC3A3cG+yonclrr/eB1dxBLKT07t+WOi4fyvxu+4P9+UNrwABGRE6zBOQp3vzZG+LF6tp8BzIgRXwKMiBGvBK45xr5mAbMayjHZ/T/j+vDyx1uY8dpqzh/Sg+4d24SdkojIEfpkdgJISzPu/cYoDhyq5uevrgw7HRGRo6hRJIiBPTpw6wUDmb2sjDdXbws7HRGRI9QoEshN5w1gcF4HfvzSCioO6vYeIpIY1CgSSFZGGvd+YxSf76nkgblrwk5HRARQo0g4Y/rmcP34vjz57qd88OmusNMREVGjSETfnzyUnp3aMv2FZRyqqgk7HRFp5dQoElCHNhn84ooRrNtWwW/f2hB2OiLSyqlRJKgJw/KYMiqf3ywooWT73rDTEZFWTI0igd116XDaZaUz/YXl1NTo9h4iEg41igTWvWMbfnTJMN7ftIs/vfdZ2OmISCulRpHgrhlTwJkDcrl/zho+310Zdjoi0gqpUSQ4M+OXV47kUHUNP315RcMDRESamRpFEujXrT23XTiYP6/axtwVZWGnIyKtjBpFkvj7c/pzcn4nfvLySnYfOBx2OiLSiqhRJInM9DTuv2oUX1Qc5L45ur2HiJw4ahRJZGRBZ248uz9PvfcZ7278Iux0RKSVaMwT7maZ2XYzWxEVe8DM1pjZMjN70cy6BPF+ZnbAzD4OXr+NGjMmeCpeiZk9FDwSleCxqc8E8cVm1i9qzA1mtj543YDwvYsGU5DTjh++sJzKw9UNDxARiVNjjigeBybXic0HRrj7KGAdMD1q3QZ3PyV43RQVfwSYRuQ52oOi9nkjsMvdBwIPAvcDmFlX4C7gdGAccFfw7OxWLTsrg19eOZKNO/fxmwUlYacjIq1Ag43C3d8GyuvE/uzutQ9MeBcoqG8fZpYPdHL3Re7uwJPAFcHqy4EnguXngAnB0cYkYL67l7v7LiLNqW7DapXOHdydb5zam9++tYE1n+8JOx0RSXHNMUfxbWBO1Pv+ZvaRmb1lZucEsd5AadQ2pUGsdt1mgKD57AZyo+MxxrR6P55yMp3aZXLH88up1u09RKQFZcQz2Mx+BFQBfwxCZUAfd//CzMYAL5nZcMBiDK/97XasdfWNqZvHNCKntcjLy6O4uLjRNdRVUVER1/gT6eoBxsxlX/LTJ9/gon6ZR61LpjoaoloSU6rUkip1QMvV0uRGEUwuTwEmBKeTcPeDwMFg+QMz2wAMJnI0EH16qgDYGiyXAoVAqZllAJ2JnOoqBYrqjCmOlYu7zwRmAowdO9aLiopibdYoxcXFxDP+RDrPnbWV7/PixnJuvvwsCnKyj6xLpjoaoloSU6rUkip1QMvV0qRTT2Y2GbgDuMzd90fFu5tZerB8EpFJ643uXgbsNbPxwfzD9cDLwbBXgNormq4GFgSNZx4w0cxygknsiUFMAmbGL64YgTv85KUVBP1aRKRZNeby2KeARcAQMys1sxuB3wAdgfl1LoM9F1hmZkuJTEzf5O61E+E3A48CJcAG/jqv8RiQa2YlwO3AnQDBuHuA94PX3VH7kkBh12z+ZdIQFq7dwStLtzY8QETkODV46sndr40RfuwY2z4PPH+MdUuAETHilcA1xxgzC5jVUI6t3d+e2Y9XPt7C3a+u4txB3clpnxV2SiKSQvTJ7BSQnmbcd9Uodh84zC9eWx12OiKSYtQoUsSw/E7843kn8fyHpbyzfmfY6YhIClGjSCHfuWAQ/bu154cvLudgtSa2RaR5qFGkkLaZ6dz7jZF8Vr6fF9YdonzfIb7cf4g9lYepOFjFgUPVVB6u5nB1DTU1rqukRKRR4vrAnSSe8SflMvW0Qp5+fzPz7pnf4PZpBmlmpKUZaQbpZke/TzPMLIgTxC2I17c9X4mn1W5bd79pHFkXa7+dK6uO+kCNiJxYahQp6GeXDaf9gW0U9h9IjUONe/CC6uBIoromOh55H4n7UWOOvK8J3rvjwX6OjK8hiB97fNXhmiM5xP6ef10XyTHyPQ4crmbPgcNcccEehvbsFPZ/WpFWSY0iBbXNTOecgkyKzuofdipx+3L/Ic785Xzun7OG3//duLDTEWmVNEchCa1LdhZTBmSycO0OFm3Qw5pEwqBGIQnvwj6Z9OrclnvnrKZGd8oVOeHUKCThZaUbt08cwrLS3by2vCzsdERaHTUKSQpXntqboT078sC8tRyqqgk7HZFWRY1CkkJ6mnHnxUP5rHw/f1r8adjpiLQqahSSNM4b3J0zB+Ty0IIS9lYeDjsdkVZDjUKShpkx/eJhlO87xO/e2hh2OiKthhqFJJWRBZ25bHQvHn1nI9v2VIadjkiroEYhSef7k4ZQXeM8OH9d2KmItApqFJJ0Crtmc934fjy7ZDPrt+0NOx2RlNeYR6HOMrPtZrYiKtbVzOab2frga07UuulmVmJma81sUlR8jJktD9Y9FDw7GzNrY2bPBPHFZtYvaswNwfdYb2a1z9UW4dYLBtI+K4P7564NOxWRlNeYI4rHgcl1YncCb7r7IODN4D1mdjIwFRgejHnYzNKDMY8A04BBwat2nzcCu9x9IPAgcH+wr67AXcDpwDjgruiGJK1b1/ZZ3Hz+AN5YvY33PtGj1EVaUoONwt3fBur+S7wceCJYfgK4Iir+tLsfdPdPgBJgnJnlA53cfZFHHoLwZJ0xtft6DpgQHG1MAua7e7m77wLm89WGJa3Yt8/qT89OkVt76NkaIi2nqXMUee5eBhB87RHEewObo7YrDWK9g+W68aPGuHsVsBvIrWdfIkDkLrm3XzSYjz77krkrPg87HZGU1dy3GbcYMa8n3tQxR39Ts2lETmuRl5dHcXFxg4keS0VFRVzjE0Wq1AH119LNnd4djJ+9+BGZO9aQkRbrf5vE0Vp+LskkVeqAlqulqY1im5nlu3tZcFppexAvBQqjtisAtgbxghjx6DGlZpYBdCZyqqsUjnqwWQFQHCsZd58JzAQYO3asFxUVxdqsUYqLi4lnfKJIlTqgEbXkb+Pbjy+hrF1/rjuj34lKq0la1c8lSaRKHdBytTT11NMrQO1VSDcAL0fFpwZXMvUnMmn9XnB6aq+ZjQ/mH66vM6Z2X1cDC4J5jHnARDPLCSaxJwYxkaOcP6QHp/fvyq/fXE/Fwaqw0xFJOY25PPYpYBEwxMxKzexG4D7gIjNbD1wUvMfdVwLPAquAucAt7l4d7Opm4FEiE9wbgDlB/DEg18xKgNsJrqBy93LgHuD94HV3EBM5ipkx/evD2FlxiJlv69YeIs2twVNP7n7tMVZNOMb2M4AZMeJLgBEx4pXANcfY1yxgVkM5ipxS2IVLRuXz6P9s5Fvj+9CjY9uwUxJJGfpktqSM708cwqGqGn79xvqwUxFJKWoUkjL6dWvPt8b35en3N7NhR0XY6YikDDUKSSnfuWAg7TLT+de5a8JORSRlqFFISsnt0IZ/PPck5q3cxgef6toHkeagRiEp58Zz+tOjYxt++foa3dpDpBmoUUjKyc7K4HsXDeaDT3fx51Xbwk5HJOmpUUhKumZMAQO6t+f+uWuoqq4JOx2RpKZGISkpIz2NOy8exsYd+3hmyeaGB4jIMalRSMq6cFgPTuuXw7+/sZ79h3RrD5GmUqOQlGVm3HnxMHbsPcij//NJ2OmIJC01CklpY/rmcPGInvzurQ3srDgYdjoiSUmNQlLe9ycNobKqhofe1K09RJpCjUJS3kndO3DtuEL+tPgzPtm5L+x0RJKOGoW0Ct+dMJisjDQemKdbe4gcLzUKaRW6d2zDtHNP4vXln/PRZ7vCTkckqahRSKvxD+ecRLcObbhXt/YQOS5qFNJqtG+TwW0XDuK9TeW8uXp7wwNEBIijUZjZEDP7OOq1x8xuM7OfmdmWqPjXo8ZMN7MSM1trZpOi4mPMbHmw7qHgudoEz95+JogvNrN+cVUrrd43TyvkpG66tYfI8Whyo3D3te5+irufAowB9gMvBqsfrF3n7q8DmNnJwFRgODAZeNjM0oPtHwGmAYOC1+QgfiOwy90HAg8C9zc1XxGAzPQ0fjB5COu3V/DcB6VhpyOSFJrr1NMEYIO7f1rPNpcDT7v7QXf/BCgBxplZPtDJ3Rd55MTxk8AVUWOeCJafAybUHm2INNWk4T35Wp8uPPjGOg4cqg47HZGEl9FM+5kKPBX1/lYzux5YAvyzu+8CegPvRm1TGsQOB8t14wRfNwO4e5WZ7QZygZ3R39zMphE5IiEvL4/i4uImF1JRURHX+ESRKnVAy9RycX41Mz47yI//8CaXDshq1n3XRz+XxJMqdUAL1uLucb2ALCK/uPOC93lAOpGjlRnArCD+n8C3osY9BlwFnAa8ERU/B3g1WF4JFESt2wDk1pfPmDFjPB4LFy6Ma3yiSJU63Fuulr9/4n0f/tO5vnNvZYvsPxb9XBJPqtThHl8twBI/xu/V5jj1dDHwobtvCxrPNnevdvca4L+AccF2pUBh1LgCYGsQL4gRP2qMmWUAnQE931KaxR2Th3LgcDX/saAk7FREElpzNIpriTrtFMw51LoSWBEsvwJMDa5k6k9k0vo9dy8D9prZ+GD+4Xrg5agxNwTLVwMLgs4nEreBPTrwN2ML+ePiT/n0C93aQ+RY4moUZpYNXAS8EBX+1+BS12XA+cD3ANx9JfAssAqYC9zi7rUziTcDjxKZ4N4AzAnijwG5ZlYC3A7cGU++InV978JBZKSl8cC8tWGnIpKw4prMdvf9RCaXo2PX1bP9DCLzFnXjS4ARMeKVwDXx5ChSnx6d2vIP5/TnoQUl/MM5XzK6sEvYKYkkHH0yW1q9aecNILd9FvfOWa1be4jEoEYhrV6HNhn804RBvLuxnOK1O8JORyThqFGIANeO60O/3Gzum7OG6hodVYhEU6MQAbIy0vj+pKGs3baXFz7UrT1EoqlRiAS+PrInowu78Kv566g8rFt7iNRSoxAJmBnTLx5K2e5Kfv+XTWGnI5Iw1ChEoow/KZcJQ3vwcHEJu/YdCjsdkYSgRiFSxx0XD2XfwSp+s1C39hABNQqRrxic15FrxhTyh0Wfsrl8f9jpiIROjUIkhu9dNJi0NPi3P+vWHiJqFCIx9OzclhvP7s9LH29lxZbdYacjEio1CpFj+MfzBpCTncl9c9aEnYpIqNQoRI6hU9tMvnPBIN4p2cnb63RrD2m91ChE6vGt8X3p0zWbe+esoUa39pBWSo1CpB5ZGWn8y6QhrC7bw0sfbwk7HZFQqFGINGDKyHxG9u7Mv/1Zt/aQ1ineJ9xtCp5m97GZLQliXc1svpmtD77mRG0/3cxKzGytmU2Kio8J9lNiZg8Fj0QleGzqM0F8sZn1iydfkaZISzOmf30oW748wJOLNoWdjsgJ1xxHFOe7+ynuPjZ4fyfwprsPAt4M3mNmJwNTgeHAZOBhM0sPxjwCTCPyHO1BwXqAG4Fd7j4QeBC4vxnyFTluZw7oRtGQ7vznwg3s3n847HRETqiWOPV0OfBEsPwEcEVU/Gl3P+junxB5PvY4M8sHOrn7Io88XuzJOmNq9/UcMKH2aEPkRLtj8lD2VB7m4WLd2kNal3gbhQN/NrMPzGxaEMtz9zKA4GuPIN4b2Bw1tjSI9Q6W68aPGuPuVcBu6jyjW+REGZbfiau+VsDv/3cTW748EHY6IidMRpzjz3L3rWbWA5hvZvV9MinWkYDXE69vzNE7jjSpaQB5eXkUFxfXm3R9Kioq4hqfKFKlDkisWs7oUMPLNTX84Mm3+YdRbY57fCLVEq9UqSVV6oCWqyWuRuHuW4Ov283sRWAcsM3M8t29LDittD3YvBQojBpeAGwN4gUx4tFjSs0sA+gMlMfIYyYwE2Ds2LFeVFTU5JqKi4uJZ3yiSJU6IPFqWcdqZr69kR9dfQYn9+p0XGMTrZZ4pEotqVIHtFwtTT71ZGbtzaxj7TIwEVgBvALcEGx2A/BysPwKMDW4kqk/kUnr94LTU3vNbHww/3B9nTG1+7oaWBDMY4iE5v8UDaRzu0zum6tbe0jrEM8cRR7wjpktBd4DXnP3ucB9wEVmth64KHiPu68EngVWAXOBW9y99qL0m4FHiUxwbwDmBPHHgFwzKwFuJ7iCSiRMndtlcuv5A3l73Q7+UrIz7HREWlyTTz25+0ZgdIz4F8CEY4yZAcyIEV8CjIgRrwSuaWqOIi3lujP68vu/bOLeOat55ZazSUvTxXiSuvTJbJEmaJORzvcnDWHFlj28umxrwwNEkpgahUgTXTa6F8N7deKBeWs5WKVbe0jqUqMQaaK0NOPOi4dSuusA//3uZ2GnI9Ji1ChE4nDOoO6cM6gb/7FgPbsP6NYekprUKETidOfFQ9l94DC/fWtD2KmItAg1CpE4De/VmStO6c2sdz5hq27tISlIjUKkGfzzxMG4w4Pz14WdikizU6MQaQYFOdnccGZfnv+wlDWf7wk7HZFmpUYh0kxuOX8gHdpkcP8c3dpDUosahUgz6ZKdxS3nD2Th2h0s2vBF2OmINBs1CpFmdMOZ/ejVuS33zllNTY3uXympQY1CpBm1zUzn9olDWFa6m9eWl4WdjkizUKMQaWZXntqboT078sC8tRyqqgk7HZG4qVGINLP04NYen5Xv50+LPw07HZG4qVGItIDzBnfnzAG5PLSghL2VurWHJDc1CpEWYGZMv3gY5fsO8bu3Noadjkhc1ChEWsjIgs5cNroXj76zkW17KsNOR6TJ4nlmdqGZLTSz1Wa20sy+G8R/ZmZbzOzj4PX1qDHTzazEzNaa2aSo+BgzWx6seyh4djbB87WfCeKLzaxfHLWKnHDfnzSE6hrXrT0kqcVzRFEF/LO7DwPGA7eY2cnBugfd/ZTg9TpAsG4qMByYDDxsZunB9o8A04BBwWtyEL8R2OXuA4EHgfvjyFfkhCvsms114/vx7JLNrN+2N+x0RJqkyY3C3cvc/cNgeS+wGuhdz5DLgafd/aC7fwKUAOPMLB/o5O6L3N2BJ4ErosY8ESw/B0yoPdoQSRa3XjCQ9lkZ3D93bdipiDRJs8xRBKeETgUWB6FbzWyZmc0ys5wg1hvYHDWsNIj1Dpbrxo8a4+5VwG4gtzlyFjlRurbP4qaiAbyxehtLPq8i8veQSPLIiHcHZtYBeB64zd33mNkjwD2AB1//Dfg2EOtIwOuJ08C66BymETl1RV5eHsXFxcdZxV9VVFTENT5RpEodkBq1DKxxurczfvPxQZ5aM4dx+Rmc3jOdvp3SSNaD5FT4uUDq1AEtV0tcjcLMMok0iT+6+wsA7r4tav1/AbODt6VAYdTwAmBrEC+IEY8eU2pmGUBnoLxuHu4+E5gJMHbsWC8qKmpyTcXFxcQzPlGkSh2QOrWcedZhHnq+mJJDnZm/fidzPjlMv9xspozKZ8rofIbkdUyqppEqP5dUqQNarpYmN4pgruAxYLW7/yoqnu/utTe5uRJYESy/AvzJzH4F9CIyaf2eu1eb2V4zG0/k1NX1wH9EjbkBWARcDSxwHbdLkurYNpOzemfyo6JxfLn/EPNWfs6rS8t4uLiE3ywsYWCPDlw6qhdTRuczoHuHsNMVOSKeI4qzgOuA5Wb2cRD7IXCtmZ1C5BTRJuAfAdx9pZk9C6wicsXULe5eHYy7GXgcaAfMCV4QaUR/MLMSIkcSU+PIVyRhdMnO4pun9eGbp/VhZ8VB5qz4nNlLt/Lvb67jwTfWMSy/E1NG5XPpqF70yc0OO11p5ZrcKNz9HWLPIbxez5gZwIwY8SXAiBjxSuCapuYokgy6dWjDdeP7ct34vmzbU8lry8qYvWwrD8xbywPz1jK6oDNTRvXiklH59OrSLux0pRWKezJbRJpPXqe2fPvs/nz77P6U7trP68vLeHVpGTNeX82M11cztm8OU0bl8/VR+fTo2DbsdKWVUKMQSVAFOdlMO3cA084dwKad+3hteRmvLt3Kz15dxc9nr+L0/l25dHQvLh6RT9f2WWGnKylMjUIkCfTr1p5bzh/ILecPpGT7Xl5dGjk99aMXV/DTl1dy5oBcLh3Vi0nDe9I5OzPsdCXFqFGIJJmBPTryvYs6ctuFg1jz+V5mL9vKq0vL+MHzy/jRS8s5d1B3pozO58JheXRsq6Yh8VOjEElSZsaw/E4My+/Ev0wcwvItu5m9rIzZS7fy5prtZGWkccGQHkwZnc8FQ3uQnaV/7tI0+j9HJAWYGaMKujCqoAt3Th7KR5t38erSMl5fXsbclZ/TLjOdCcN6MGVUL4qGdKdtZnrDOxUJqFGIpJi0NGNM366M6duVn0w5mfc+KWf2sq2Rz2osK6NDmwwmnpzHlNH5nD2wO1kZeiyN1E+NQiSFpacZZwzI5YwBufz8suEs2vgFs5eWMWdFGS98tIXO7TKZPLwnU0bnc8ZJuWSkq2nIV6lRiLQSGelpnDOoO+cM6s49V4zgnZIdzF5axmvLy3hmyWZy22cxeURPLh3di9P6dSU9LXnuOyUtS41CpBXKykjjgqF5XDA0j8rD1RSv3cHsZVt54cMt/HHxZ/To2Iavj8zn0tG9+FqfLkl1s0JpfmoUIq1c28x0Jo/oyeQRPdl/qIoFa7bz6tKt/Om9z3j8fzfRu0s7LhmVz5RR+Yzs3VlNoxVSoxCRI7KzMpgyqhdTRvVib+Vh3li9jdlLy/j9Xz5h5tsb6ZubzSUj87lkVD7VNbqRc2uhRiEiMXVsm8mVpxZw5akF7N5/OHJb9GVb+d3bG3m4eANpBr3eX0BBTjsKc7Ip7JpNYdd2FORkU5iTTY+ObUjTPEdKUKMQkQZ1zs7kb04r5G9OK+SLioMUr91B8YerSO+Yw+ZdB3hr3Q627z141JisjDQKurSjd067SBPJyY40la7ZFOa0o2v7LJ3GShJqFCJyXHI7tOGqMQXk7i2hqOjUI/HKw9Vs+fIAm8v3s3nXAUp37ae0/ACbd+1nxfIydu0/fNR+srPSjzoaKcgJjka6RppJJ91+JGGoUYhIs2ibmc6A7h2O+XS+ioNVlO7az+byA0e+bt61n83l+1n8STkVB6uO2r5zu8yoRtLuSDOJHJlk0y5Lny4/UdQoROSE6NAmg6E9OzG0Z6evrHN3dh84fKR5RDeS9dv3snDtdg5W1Rw1pluHNkedyjpyNJKTTa8u7fSJ82aUFI3CzCYDvwbSgUfd/b6QUxKRZmRmdMnOokt2FiMLOn9lvbuzo+LgkaOR0l21p7j2s3Tzl8xZXkZV1FVYZtCzU9vI0UfXdkfPj3TNpmentvpA4XFI+EZhZunAfwIXAaXA+2b2iruvCjczETlRzIweHdvSo2NbxvTN+cr66hrn8z2VkeZRHjSSYI7k3Q1f8OKeLXjU1bwZaUavLu0o7NoO9h9k7hfLSE8zMtKM9LQ0MtIt6n1UvPZ9+jHiR62PEU9LizE+Kp5mpKcfHU8zQp/0T/hGAYwDStx9I4CZPQ1cDqhRiAgQuadV7y7t6N2lHeNPyv3K+kNVNWz9sva01l8n3CONpZqSiu1U1zhVNU51dfC1xqmqqSERPi6S/pWGE7txdU+vpKio+b9/MjSK3sDmqPelwOkh5SIiSSgrI41+3drTr1v7r6wrLi6mqJ7frjU1TrV7nUZS89f3R77WUFXjVFV7nXU1x2xCX9lHdU2dfR6976PHR+LVNRxZT8WhFvnvZ+4J0C7rYWbXAJPc/e+D99cB49z9O1HbTAOmAeTl5Y15+umnm/z9Kioq6NAh9lUbySRV6gDVkqhSpZZUqQPiq+X888//wN3HxlqXDEcUpUBh1PsCYGv0Bu4+E5gJMHbsWK/vr4OGNPTXRbJIlTpAtSSqVKklVeqAlqslGa4fex8YZGb9zSwLmAq8EnJOIiKtRsIfUbh7lZndCswjcnnsLHdfGXJaIiKtRsI3CgB3fx14Pew8RERao2Q49SQiIiFSoxARkXqpUYiISL3UKEREpF4J/4G742VmO4BP49hFN2BnM6UTplSpA1RLokqVWlKlDoivlr7u3j3WipRrFPEysyXH+nRiMkmVOkC1JKpUqSVV6oCWq0WnnkREpF5qFCIiUi81iq+aGXYCzSRV6gDVkqhSpZZUqQNaqBbNUYiISL10RCEiIvVSowiY2WQzW2tmJWZ2Z9j5NJWZzTKz7Wa2Iuxc4mVmhWa20MxWm9lKM/tu2Dk1hZm1NbP3zGxpUMfPw84pXmaWbmYfmdnssHOJh5ltMrPlZvaxmS0JO594mFkXM3vOzNYE/2bOaLZ969TTkedyryPqudzAtcn4XG4zOxeoAJ509xFh5xMPM8sH8t39QzPrCHwAXJFsPxeLPPC4vbtXmFkm8A7wXXd/N+TUmszMbgfGAp3cfUrY+TSVmW0Cxrp70n+OwsyeAP7H3R8NHsmQ7e5fNse+dUQRceS53O5+CKh9LnfScfe3gfKw82gO7l7m7h8Gy3uB1UQejZtUPKIieJsZvJL2LzQzKwAuAR4NOxeJMLNOwLnAYwDufqi5mgSoUdSK9VzupPuFlMrMrB9wKrA45FSaJDhV8zGwHZjv7klZR+DfgR8ANSHn0Rwc+LOZfRA8UjlZnQTsAH4fnBJ81My++oDwJlKjiLAYsaT9iy/VmFkH4HngNnffE3Y+TeHu1e5+CpFH+Y4zs6Q8LWhmU4Dt7v5B2Lk0k7Pc/WvAxcAtwanbZJQBfA14xN1PBfYBzTbXqkYR0eBzuSUcwTn954E/uvsLYecTr+B0QDEwOdxMmuws4LLg3P7TwAVm9t/hptR07r41+LodeJHIaehkVAqURh2pPkekcTQLNYoIPZc7AQWTwI8Bq939V2Hn01Rm1t3MugTL7YALgTWhJtVE7j7d3QvcvR+RfycL3P1bIafVJGbWPrhIguA0zUQgKa8WdPfPgc1mNiQITQCa7aKPpHgUaktLpedym9lTQBHQzcxKgbvc/bFws2qys4DrgOXB+X2AHwaPxk0m+cATwdV1acCz7p7Ul5WmiDzgxcjfI2QAf3L3ueGmFJfvAH8M/tjdCPxdc+1Yl8eKiEi9dOpJRETqpUYhIiL1UqMQEZF6qVGIiEi91ChERKReahQiIlIvNQoREamXGoWIiNTr/wdLEbLHj/EgHwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'condition'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 3\n",
    "others = counts.index[n_categorical:]\n",
    "\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "good         115685\n",
       "excellent    115206\n",
       "others        34382\n",
       "Name: condition, dtype: int64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* cylinders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcaf08f70>]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAD4CAYAAADy46FuAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAApAklEQVR4nO3deXxV9Z3/8dcn92ZhX0NYAmF1AVQ0CMEFQVzQtmpbF6iAbZ3SWjttx3nMT512fnbq2LGd34zW2jq1ytStIONSaQsqCqlL2REFBCQgS1hlEYkKkuTz++Oe6CVmu8kN5yb3/Xw87iPnfs753rwTMZ98z/fcE3N3REREapMRdgAREUltahQiIlInNQoREamTGoWIiNRJjUJEROoUDTtAsnXv3t379+/f6PEffvgh7dq1S16gJFGuxChXYpQrMa0x14oVK/a5e26NO929VT0KCwu9KRYuXNik8c1FuRKjXIlRrsS0xlzAcq/l56pOPYmISJ3qbRRmNsPM9prZmrjak2a2KnhsMbNVQb2/mX0ct++/48YUmtlqMysxs/vMzIJ6dvB6JWa2xMz6x425wcw2Bo8bkvmFi4hIwzRkjeL3wP3Ao1UFd7+uatvM/hM4FHf8JncfUcPrPABMBxYDc4GJwDzgRuCguw82s0nAz4HrzKwrcAcwEnBghZnNcfeDDf7qRESkyeqdUbj7K8CBmvYFs4JrgZl1vYaZ9QI6uvui4FzYo8BVwe4rgUeC7aeACcHrXgrMd/cDQXOYT6y5iIjICdTUq57OB/a4+8a42gAzewP4APixu78K9AFK444pDWoEH7cDuHu5mR0CusXXaxhzHDObTmy2Ql5eHsXFxY3+gsrKypo0vrkoV2KUKzHKlZh0y9XURjGZ42cTu4B+7r7fzAqBP5rZMMBqGFt1N8La9tU15vii+4PAgwAjR470cePGNSx9DYqLi2nK+OaiXIlRrsQoV2LSLVejr3oysyjwFeDJqpq7H3X3/cH2CmATcBKx2UB+3PB8YGewXQr0jXvNTsROdX1ar2GMiIicIE25PPYiYL27f3pKycxyzSwSbA8EhgCb3X0XcNjMioL1h2nAc8GwOUDVFU1XAwuCdYwXgEvMrIuZdQEuCWrN4oMjx/jPFzew+8PK5voUIiItUkMuj50JLAJONrNSM7sx2DWJzy9ijwXeMrM3iS1Mf8fdqxbCbwIeAkqIzTTmBfWHgW5mVgLcAtwGEIy7E1gWPH4a91pJd/RYJb97dTNzNh1rrk8hItIi1btG4e6Ta6l/vYba08DTtRy/HBheQ/0IcE0tY2YAM+rLmAy5HbKZNqY/v3tlM5veK2NQbvsT8WlFRFKe3pkdZ/rYgWRG4Fcvb6z/YBGRNKFGEad7+2wm9Mtkzps7KdlbFnYcEZGUoEZRzWUDMsmORvjVAs0qRERAjeJzOmYZ084pCGYVh8OOIyISOjWKGnx77CDaZEa47+WSsKOIiIROjaIGXdtlccM5/fnTWzvZuEezChFJb2oUtfjW+QNpmxnhl7oCSkTSnBpFLapmFX9ZvYt3NKsQkTSmRlEHzSpERNQo6tSlXRZfP7c/c1fvYsNuzSpEJD2pUdTjW+cPpF1WlPs0qxCRNKVGUY/ObbP4xrmxtYr1uz8IO46IyAmnRtEAN543gA7ZUX75kmYVIpJ+1CgaoGpWMW/Nbtbt0qxCRNKLGkUD3XjeQDrkaFYhIulHjaKBOrXN5JvnDuD5tbtZu/NQ2HFERE4YNYoEfPO8AZpViEjaUaNIQKc2mdx43gBefHsPa3ZoViEi6UGNIkHfOHcAHXOiere2iKQNNYoExWYVA5mvWYWIpIl6G4WZzTCzvWa2Jq72EzPbYWargsflcftuN7MSM9tgZpfG1QvNbHWw7z4zs6CebWZPBvUlZtY/bswNZrYxeNyQtK+6ib5xXn865kS5V2sVIpIGGjKj+D0wsYb6Pe4+InjMBTCzocAkYFgw5jdmFgmOfwCYDgwJHlWveSNw0N0HA/cAPw9eqytwBzAaGAXcYWZdEv4Km0HHnEy+df5AXlq3h9WlmlWISOtWb6Nw91eAAw18vSuBWe5+1N3fBUqAUWbWC+jo7ovc3YFHgavixjwSbD8FTAhmG5cC8939gLsfBOZTc8MKxdfP7U+nNpnc+9I7YUcREWlW0SaM/Z6ZTQOWA/8Y/DDvAyyOO6Y0qB0LtqvXCT5uB3D3cjM7BHSLr9cw5jhmNp3YbIW8vDyKi4sb/UWVlZU1ePyEfHhm/V7+57mXGdApUv+AJkgk14mkXIlRrsQoV2KaK1djG8UDwJ2ABx//E/gmYDUc63XUaeSY44vuDwIPAowcOdLHjRtXR/S6FRcX09DxhUXHWPCLhbx6sCPfuPLsRn/OZOc6kZQrMcqVGOVKTHPlatRVT+6+x90r3L0S+B2xNQSI/dbfN+7QfGBnUM+voX7cGDOLAp2Ineqq7bVSRodgrWLB+r2s2v5+2HFERJpFoxpFsOZQ5ctA1RVRc4BJwZVMA4gtWi91913AYTMrCtYfpgHPxY2puqLpamBBsI7xAnCJmXUJFrEvCWop5YZz+tOlbSa/1FqFiLRS9Z56MrOZwDigu5mVErsSaZyZjSB2KmgL8G0Ad19rZrOBt4Fy4GZ3rwhe6iZiV1C1AeYFD4CHgcfMrITYTGJS8FoHzOxOYFlw3E/dvaGL6idM++wo3xo7kF88v4E3th3kzH4pcWGWiEjS1Nso3H1yDeWH6zj+LuCuGurLgeE11I8A19TyWjOAGfVlDNu0Mf353SubufeljTzyzVH1DxARaUH0zuwkaJ8dZfrYQfz1nfdYue1g2HFERJJKjSJJpo0poGu7LL1bW0RaHTWKJGmXHWX62IG88s57rNiqWYWItB5qFEk0bUwB3dpl6d3aItKqqFEkUdusKN++YCCvbtzHiq0pd4GWiEijqFEk2ZSiArq3z+Ke+VqrEJHWQY0iydpmRfn22EG8VrKPZVs0qxCRlk+NohlUzSq0ViEirYEaRTNokxXhOxcM4vWS/Sx9V7MKEWnZ1CiayfWjC+jePpt75mtWISItmxpFM2mTFeGmcYNYtHk/izfvDzuOiEijqVE0o+tH9yO3Q7bWKkSkRVOjaEY5mRFuumAQizcfYNEmzSpEpGVSo2hmXxvdjx6aVYhIC6ZG0cxyMmNrFUvePcDfNu0LO46ISMLUKE6AyaP6kdcxm3vnbyT2x/tERFoONYoTICczwnfHDWbpFq1ViEjLo0Zxglx3dl96dszhnpfe0axCRFoUNYoTJCczwnfHD2LZloO8XqJZhYi0HPU2CjObYWZ7zWxNXO0/zGy9mb1lZs+aWeeg3t/MPjazVcHjv+PGFJrZajMrMbP7zMyCeraZPRnUl5hZ/7gxN5jZxuBxQzK/8DBUzSru1axCRFqQhswofg9MrFabDwx399OBd4Db4/ZtcvcRweM7cfUHgOnAkOBR9Zo3AgfdfTBwD/BzADPrCtwBjAZGAXeYWZcEvraUkx2NcPP4QSzfepDXSnQFlIi0DPU2Cnd/BThQrfaiu5cHTxcD+XW9hpn1Ajq6+yKP/Sr9KHBVsPtK4JFg+ylgQjDbuBSY7+4H3P0gseZUvWG1ONee3ZfenXK4Z75mFSLSMiRjjeKbwLy45wPM7A0z+6uZnR/U+gClcceUBrWqfdsBguZzCOgWX69hTIuVHY3w3fGDWbntfV7dqFmFiKS+aFMGm9mPgHLgiaC0C+jn7vvNrBD4o5kNA6yG4VW/Tte2r64x1XNMJ3Zai7y8PIqLixv8NVRXVlbWpPENkVfpdM0x/vXp5fy4KIdguSb0XI2hXIlRrsQoV2KaLZe71/sA+gNrqtVuABYBbesYVwyMBHoB6+Pqk4HfBtsvAGOC7Siwj1iT+PSYYN9vgcn1ZS0sLPSmWLhwYZPGN9Tji7d4wa1/9oXr9zTo+BOVK1HKlRjlSoxyJaYpuYDlXsvP1UadejKzicCtwBXu/lFcPdfMIsH2QGKL1pvdfRdw2MyKgvWHacBzwbA5QdMBuBpYEIR+AbjEzLoEi9iXBLVW4ZrCvvTp3IZ7X9K7tUUktTXk8tiZxGYOJ5tZqZndCNwPdADmV7sMdizwlpm9SWxh+jvuXrUQfhPwEFACbOKzdY2HgW5mVgLcAtwGEIy7E1gWPH4a91otXlY0g+9dOJhV29+n+J33wo4jIlKretco3H1yDeWHazn2aeDpWvYtB4bXUD8CXFPLmBnAjPoytlRfPSuf+xeUcO/8dxh3Um6D1ipERE40vTM7RFnRDP7+wsG8WXqIhRv2hh1HRKRGahQh+2phPvldtFYhIqlLjSJkmZHYrOKt0kMsWK9ZhYikHjWKFPCVs/Lp17WtZhUikpLUKFJAZiR2BdTqHYd4eZ1mFSKSWtQoUsSXz+wTm1W8rHtAiUhqUaNIEVVrFWt2fMD8t/eEHUdE5FNqFCnky2f2oaCb1ipEJLWoUaSQaCSDv79wCG/v+oAXNasQkRShRpFirhrRmwHd23HvSxuprNSsQkTCp0aRYqLBWsU6zSpEJEWoUaSgK86omlW8o1mFiIROjSIFRSMZfH/CYNbvPswLa3eHHUdE0pwaRYq64ow+DOzejl++rLUKEQmXGkWKimQY358whPW7D/O8ZhUiEiI1ihT2pTN6Myi3Hb98aSOVel+FiIREjSKFVc0qNuw5zO9WH+X1kn2UV1SGHUtE0ky9f+FOwvXF03uzePMBnlmxjesfWkLXdllcOqwnl5/WkzEDuxGNqNeLSPNSo0hxkQzj379yGuM67cPzTuEvq3fz3KodzFy6jS5tM4Om0Ysxg7qRqaYhIs1AjaKFyI4Y44b3YuLwXhw5VkHxhveYt2YXf3pzJ7OWbadz20wuHdqTy0/vxTlqGiKSRPU2CjObAXwR2Ovuw4NaV+BJoD+wBbjW3Q8G+24HbgQqgO+7+wtBvRD4PdAGmAv8wN3dzLKBR4FCYD9wnbtvCcbcAPw4iPJv7v5Ik7/iViAnM8LE4T2ZOLwnR45V8Mo77zF39S7+snoXTy6PNY1LhuZx2Wm9OHdQd7Kiahoi0ngNmVH8Hrif2A/zKrcBL7v73WZ2W/D8VjMbCkwChgG9gZfM7CR3rwAeAKYDi4k1ionAPGJN5aC7DzazScDPgeuCZnQHMBJwYIWZzalqSBKTkxnhkmE9uWRYrGm8unEfc1fvYt7q3cxeXkqnNplcPDSPL5zWi3MHq2mISOLqbRTu/oqZ9a9WvhIYF2w/AhQDtwb1We5+FHjXzEqAUWa2Bejo7osAzOxR4CpijeJK4CfBaz0F3G9mBlwKzHf3A8GY+cSay8zEv8z0kJMZ4eKheVw8NI+j5RW8+k6sabywZjdPrSilY06Ui4f25Aun9+S8wblqGiLSII1do8hz910A7r7LzHoE9T7EZgxVSoPasWC7er1qzPbgtcrN7BDQLb5ewxipR3Y0wkVD87goaBqvbdzH3NW7efHt3Ty9spQOOVEuHprH5cN7cf5J3cmORsKOLCIpKtmL2VZDzeuoN3bM8Z/UbDqx01rk5eVRXFxcb9DalJWVNWl8c2lqrgjwpR5wWfcs1u6PsGx3Bc+/tYNnVu6gTRRG9IgwqmeUYd0iZEVq+tY3T67molyJUa7EpFuuxjaKPWbWK5hN9AL2BvVSoG/ccfnAzqCeX0M9fkypmUWBTsCBoD6u2pjimsK4+4PAgwAjR470cePG1XRYgxQXF9OU8c0lmbkuCj5+Ul7J65v2MfetXbz49h4W7TxK++woF53ag8tP68XYk3LJyax7ppEO369kUq7EKFdimitXYxvFHOAG4O7g43Nx9T+Y2X8RW8weAix19wozO2xmRcASYBrwq2qvtQi4GlgQXA31AvAzM+sSHHcJcHsj80oNsqIZjD+5B+NP7sHPKip5vWQf81bv5oW3d/PHVTtpnx1lQtA0LmhA0xCR1qkhl8fOJPabfXczKyV2JdLdwGwzuxHYBlwD4O5rzWw28DZQDtwcXPEEcBOfXR47L3gAPAw8Fix8HyB21RTufsDM7gSWBcf9tGphW5IvM5LBuJN7MO7kHvxbxXAWbdofWwhfu5vnVu2kXVaECafmcflpPRl3cg81DZE00pCrnibXsmtCLcffBdxVQ305MLyG+hGCRlPDvhnAjPoySnJlRjIYe1IuY0/K5c6rhrN4c1XT2MOcN3fSNivChaf04Aun9SJSoZsVirR2eme21CkzksH5Q3I5f0gud15ZyeLNB/hLMNP481u7aJ8J1x9dx9dG96OgW7uw44pIM1CjkAaLRjI4b0h3zhvSnTuvHMaizfv55Z9X8NBr7/Lgq5sZOySXqUUFjD+lB5GMhl85JSKpTY1CGiUazDQqzszhlDOLmLVsGzOXbuPvHl1On85t+Nroflw7si+5HbLDjioiTaS35kqT9eyUww8vOonXbr2QB64/i/7d2/IfL2zgnLtf5vsz32Dpuwdw/eElkRZLMwpJmsxIBped1ovLTutFyd4ynliyladWlDLnzZ2cnNeBKWMK+PKZfWifrX92Ii2JZhTSLAb3aM8dXxrGkn+ewN1fOY1oxPiXP65h9F0v8eM/rmb97g/CjigiDaRf7aRZtc2KMmlUP647uy+rtr/PY4u3Mnt5KY8v3sao/l25vqgflw3vpRsUiqQwNQo5IcyMM/t14cx+XfiXLwzlf1ds5/HF2/jBrFXc2f5trju7L5NH9SO/S9uwo4pINWoUcsJ1aZfF9LGD+LvzBvJqyT4eW7SVB4o38UDxJi48pQdTigoYOySXDF1iK5IS1CgkNBkZxgUn5XLBSbmUHvyImUu38eSy7by0bi/9urbl+tH9uGZkX7q2ywo7qkha04lhSQn5XdryT5eewt9um8B9k8+kZ8cc/n3eeor+/WVumb2KldsO6hJbkZBoRiEpJSuawRVn9OaKM3qzYfdhHl+8lWdWlvLMyh0M692RqUUFXDGiN22z9E9X5ETRjEJS1sk9O3DnVcNZ8qOLuPOq4VRUOrc9s5rRP3uZn8xZS8nesrAjiqQF/VomKa99dpSpRQVMGd2P5VsP8tiirTyxZCu//9sWzhnUjSlFBVw8NI/MiH7vEWkOahTSYpgZZ/fvytn9u/Le4aHMXr6dPyzZxnefWEmPDtlMHtWPyaP60bNTTthRRVoVNQppkXI7ZHPz+MF854JBFG/Yy2OLt3Lfgo3cv7CEi0/N48JulWFHFGk11CikRYtkGBNOzWPCqXls3f8hf1iyjccXb2XnHrj28rDTibQOOqkrrUZBt3bcfvmpfGvsQFbvq2Dr/g/DjiTSKqhRSKszeVQ/zOCJJdvCjiLSKqhRSKuT1zGHs3pEmL18O0eOVYQdR6TFa3SjMLOTzWxV3OMDM/uhmf3EzHbE1S+PG3O7mZWY2QYzuzSuXmhmq4N995mZBfVsM3syqC8xs/5N+molbUzol8n7Hx3jT2/uDDuKSIvX6Ebh7hvcfYS7jwAKgY+AZ4Pd91Ttc/e5AGY2FJgEDAMmAr8xs0hw/APAdGBI8JgY1G8EDrr7YOAe4OeNzSvp5ZSuGQzp0Z7HFm8NO4pIi5esU08TgE3uXtf/lVcCs9z9qLu/C5QAo8ysF9DR3Rd57GY+jwJXxY15JNh+CphQNdsQqYuZMXVMAW+VHuLN7e+HHUekRUvW5bGTgJlxz79nZtOA5cA/uvtBoA+wOO6Y0qB2LNiuXif4uB3A3cvN7BDQDdgX/8nNbDqxGQl5eXkUFxc3+gspKytr0vjmolyJKSsrI7f8XXIi8Itnl/Ct07PDjgSk9vdLuRou7XK5e5MeQBaxH9x5wfM8IEJstnIXMCOo/xqYEjfuYeCrwNnAS3H184E/Bdtrgfy4fZuAbnXlKSws9KZYuHBhk8Y3F+VKTFWuHz37lg/50Vw/UHY03ECBVP9+pRrlSkxTcgHLvZafq8k49XQZsNLd9wSNZ4+7V7h7JfA7YFRwXCnQN25cPrAzqOfXUD9ujJlFgU7AgSRkljQxtag/n5RXMnv59rCjiLRYyWgUk4k77RSsOVT5MrAm2J4DTAquZBpAbNF6qbvvAg6bWVGw/jANeC5uzA3B9tXAgqDziTTIyT07MGpAVx5fspXKSv3TEWmMJjUKM2sLXAw8E1f+RXCp61vAeOAfANx9LTAbeBt4HrjZ3asucr8JeIjYAvcmYF5QfxjoZmYlwC3AbU3JK+lpalEB2w98zF/feS/sKCItUpMWs939I2KLy/G1qXUcfxexdYvq9eXA8BrqR4BrmpJR5NJhPcntkM1ji7cy/pQeYccRaXH0zmxp9bKiGUw+uy8LN+xl+4GPwo4j0uKoUUhamDy6HxlmPL5Eb8ATSZQahaSFXp3acPGpecxepvs/iSRKjULSxtQxBRz86Bh/eWtX2FFEWhQ1Ckkb5wzqxsDcdrr/k0iC1CgkbZgZU4sKWLX9fVaXHgo7jkiLoUYhaeWrhfm0yYzw2OItYUcRaTHUKCStdMzJ5Koz+/Dcqp28/9EnYccRaRHUKCTtTC0q4Gh5JU+tKK3/YBFRo5D0M7R3R0YWdOGxxbr/k0hDqFFIWpo6poCt+z/i1ZJ99R8skubUKCQtTRzek+7ts3hs0Zawo4ikPDUKSUvZ0QiTzu7Hy+t1/yeR+qhRSNqaPLofBvxh6bawo4ikNDUKSVt9Ordhwql5PLlsO0fLdf8nkdqoUUhamzamgAMffsK81bvDjiKSstQoJK2dO6g7A7q341EtaovUSo1C0lpGhjGlqICV295nzQ7d/0mkJmoUkvauPiufnMwMHtddZUVqpEYhaa9T20yuGtGHP67awaGPj4UdRyTlNKlRmNkWM1ttZqvMbHlQ62pm881sY/CxS9zxt5tZiZltMLNL4+qFweuUmNl9ZmZBPdvMngzqS8ysf1PyitRmSlEBR47p/k8iNUnGjGK8u49w95HB89uAl919CPBy8BwzGwpMAoYBE4HfmFkkGPMAMB0YEjwmBvUbgYPuPhi4B/h5EvKKfM7wPp04q19nHtf9n0Q+pzlOPV0JPBJsPwJcFVef5e5H3f1doAQYZWa9gI7uvsjdHXi02piq13oKmFA12xBJtqljCnh334e8vkn3fxKJZ7GfzY0cbPYucBBw4Lfu/qCZve/uneOOOejuXczsfmCxuz8e1B8G5gFbgLvd/aKgfj5wq7t/0czWABPdvTTYtwkY7e7H/Z9sZtOJzUjIy8srnDVrVqO/prKyMtq3b9/o8c1FuRLTmFzHKp1bFn7EkC4Rvn9WTsrkOhGUKzGtMdf48eNXxJ0ZOk60SangXHffaWY9gPlmtr6OY2uaCXgd9brGHF9wfxB4EGDkyJE+bty4OkPXpbi4mKaMby7KlZjG5pryyXp++9dNDBkxmj6d26RMruamXIlJt1xNOvXk7juDj3uBZ4FRwJ7gdBLBx73B4aVA37jh+cDOoJ5fQ/24MWYWBToBB5qSWaQu14/uhwMzl+j+TyJVGt0ozKydmXWo2gYuAdYAc4AbgsNuAJ4LtucAk4IrmQYQW7Re6u67gMNmVhSsP0yrNqbqta4GFnhTzpWJ1CO/S1smnNKDWcu26f5PIoGmzCjygNfM7E1gKfAXd38euBu42Mw2AhcHz3H3tcBs4G3geeBmd6/6P/Em4CFiC9ybiK1dADwMdDOzEuAWgiuoRJrTlKIC9pV9wvNrdP8nEWjCGoW7bwbOqKG+H5hQy5i7gLtqqC8HhtdQPwJc09iMIo0xdkguBd3a8vjirVw5ok/YcURCp3dmi1STkWFMGV3Asi0HWbfrg7DjiIROjUKkBteMzCc7msFjuv+TiBqFSE06t83iijN688c3dvDBEd3/SdKbGoVILaaN6c9Hn1TwjO7/JGlOjUKkFqfld+KMvp15bPFWdFW2pDM1CpE6TCsqYNN7H7Jo0/6wo4iERo1CpA5fOL0XXdpm8ugiLWpL+lKjEKlDTmaEa8/uy/x1e9h16OOw44iEQo1CpB5TRhdQ6a77P0naUqMQqUffrm0Zf3IPZi7bzifllWHHETnh1ChEGmBqUQHvHT7KC2t1/ydJP2oUIg0w9qRc+nZto3dqS1pSoxBpgEhw/6el7x5g/W7d/0nSixqFSANdM7IvWdEMHtesQtKMGoVIA3Vtl8WXTu/Nsyt3cFj3f5I0okYhkoCpYwr48JMKnn1jR9hRRE4YNQqRBIzo25nT8zvx2CLd/0nShxqFSIKmFBWwcW8ZizcfCDuKyAmhRiGSoCvO6E2nNpla1Ja0oUYhkqCczAjXjsznhbW72fPBkbDjiDS7RjcKM+trZgvNbJ2ZrTWzHwT1n5jZDjNbFTwujxtzu5mVmNkGM7s0rl5oZquDffeZmQX1bDN7MqgvMbP+TfhaRZJmSlEB5ZXOzKW6/5O0fk2ZUZQD/+jupwJFwM1mNjTYd4+7jwgecwGCfZOAYcBE4DdmFgmOfwCYDgwJHhOD+o3AQXcfDNwD/LwJeUWSpqBbOy44KZc/LNnGsQrd/0lat0Y3Cnff5e4rg+3DwDqgTx1DrgRmuftRd38XKAFGmVkvoKO7L/LYZSSPAlfFjXkk2H4KmFA12xAJ29SiAvYePsr8t/eEHUWkWVkyLvELTgm9AgwHbgG+DnwALCc26zhoZvcDi9398WDMw8A8YAtwt7tfFNTPB2519y+a2RpgoruXBvs2AaPdfV+1zz+d2IyEvLy8wlmzZjX6aykrK6N9+/aNHt9clCsxJyJXpTv/9NePyW1r3DaqTcrkagzlSkxrzDV+/PgV7j6yxp3u3qQH0B5YAXwleJ4HRIjNVu4CZgT1XwNT4sY9DHwVOBt4Ka5+PvCnYHstkB+3bxPQra48hYWF3hQLFy5s0vjmolyJOVG5fr1woxfc+md/Z/cHDTo+3b9fiVKuxDQlF7Dca/m52qSrnswsE3gaeMLdnwkazx53r3D3SuB3wKjg8FKgb9zwfGBnUM+voX7cGDOLAp0AXbwuKeO6kX3JimTorrLSqjXlqicjNitY5+7/FVfvFXfYl4E1wfYcYFJwJdMAYovWS919F3DYzIqC15wGPBc35oZg+2pgQdD5RFJCt/bZfOH0XjyzcgdlR8vDjiPSLJoyozgXmApcWO1S2F8El7q+BYwH/gHA3dcCs4G3geeBm929Initm4CHiC1wbyK2dgGxRtTNzEqIrX3c1oS8Is1i6pgCyo6W6/5P0mpFGzvQ3V8DaroCaW4dY+4itm5Rvb6c2EJ49foR4JrGZhQ5Ec7s25lhvTvy+KKtTBndD12YJ62N3pkt0kRmxrQxBWzYc5hlWw6GHUck6dQoRJLgijP60DEnyqOLtoQdRSTp1ChEkqBNVoRrRvbl+TW72XtY93+S1kWNQiRJqu7/NGvp9rCjiCSVGoVIkgzo3o7zh3TnD0u2Ua77P0krokYhkkRTiwrY/cERXlqn+z9J66FGIZJEF57Sg96dcvRObWlV1ChEkigayeD6ogJeL9lPyd6ysOOIJIUahUiSXTuyL5kR059KlVZDjUIkyXI7ZHP5ab14ekUpH+r+T9IKqFGINIOpRQUcPlrOc6t21n+wSIpToxBpBoUFXTi1V0ceXbQF3fBYWjo1CpFmYGZMLSpg/e7DrNiq+z9Jy6ZGIdJMrjqzNx2yo7pUVlo8NQqRZtI2K8pXC/OZu3oX7x0+GnYckUZToxBpRlPHFHCswpm9XPd/kpZLjUKkGQ3Kbc+5g7vxxOKtuv+TtFhqFCLNbGpRf3YeOsKC9XvDjiLSKGoUIs3solN70LOj7v8kLZcahUgzi0Yy+Nrofry6cR+7P9TpJ2l5omEHaAgzmwj8EogAD7n73SFHEknIpFF9ue/ljdz+6sf837/NI5JhRDOMSCT4mGFEMzI+q1fVIkYkIyPumPiPGTW8Rg3HR2qpf7o/g03bj7F32XYyMoxIBkQyMohY3HYGZFgsY0YGQdbja1Wvl2Gf5f/0Ua0WGxfbNrOw//NIPVK+UZhZBPg1cDFQCiwzsznu/na4yUQarkeHHO7/2pn8ZdFq8vv2o6LSKa9wKiorKa/02PPjPlYG++PrlRyrqOTjYx43Plb//HinvOLz9TqtfevEfDOqMePTBhPNsKBZxbbLjx2j3eIFnzbNaNBQYw3UyKxqrtWa4afHRj5rjpmRzxpxfAOu/3WPHxfNyGDd/grabN4fa3pxjTDDLNY0rVo9w8iw4+tVX2eG8VlzDeqpJuUbBTAKKHH3zQBmNgu4ElCjkBZl4vBe5OzbwLhxp4Ty+d2dSuf4xlLhHKus5PXX/8aoojFUVDgVHmssVY9K/6zZVK9VxjWhymrjKirrfq3KuHE1vVZ5pVO6Yye5Pbp92lDLK+IaaVyzPFJeQUWlc6zi8/vKg6Zb9XmPxTXQJlm2ODn/YWpQWwOpXo+fvZlBbuQI48YlP4+l+n1ozOxqYKK7/13wfCow2t2/F3fMdGA6QF5eXuGsWbMa/fnKyspo375900I3A+VKjHIlJh1zVTXOiuDx6XZlXL0yVi/342sffvQxOW3aUBnUqx4Oxz2v2l+97g6VVG0Hr0G1/Z/WvMbX97jPUVXvHC3na8Mb9/0aP378CncfWes3K5UfwDXE1iWqnk8FflXb8YWFhd4UCxcubNL45qJciVGuxChXYlpjLmC51/JztSVc9VQK9I17ng/o3s0iIidIS2gUy4AhZjbAzLKAScCckDOJiKSNlF/MdvdyM/se8AKxy2NnuPvakGOJiKSNlG8UAO4+F5gbdg4RkXTUEk49iYhIiNQoRESkTmoUIiJSJzUKERGpU8q/MztRZvYe0JT7OXcH9iUpTjIpV2KUKzHKlZjWmKvA3XNr2tHqGkVTmdlyr+1t7CFSrsQoV2KUKzHplkunnkREpE5qFCIiUic1is97MOwAtVCuxChXYpQrMWmVS2sUIiJSJ80oRESkTmoUIiJSJzWKgJlNNLMNZlZiZreFnaeKmc0ws71mtibsLFXMrK+ZLTSzdWa21sx+EHYmADPLMbOlZvZmkOtfw84Uz8wiZvaGmf057CzxzGyLma02s1VmtjzsPFXMrLOZPWVm64N/a2NSINPJwfep6vGBmf0wBXL9Q/Bvfo2ZzTSznKS+vtYoYv8DA+8AFxP7Q0nLgMnuHvrf5TazsUAZ8Ki7Dw87D4CZ9QJ6uftKM+sArACuCvv7ZWYGtHP3MjPLBF4DfuDuzffHjRNgZrcAI4GO7v7FsPNUMbMtwEh3T6k3kJnZI8Cr7v5Q8Ldo2rr7+yHH+lTwc2MHsT/N3JQ3+TY1Rx9i/9aHuvvHZjYbmOvuv0/W59CMImYUUOLum939E2AWcGXImQBw91eAA2HniOfuu9x9ZbB9GFgH9Ak3FQR/0bEseJoZPFLiNyEzywe+ADwUdpaWwMw6AmOBhwHc/ZNUahKBCcCmMJtEnCjQxsyiQFuS/FdA1Shi+gDb456XkgI/+FoCM+sPnAksCTkK8OnpnVXAXmC+u6dELuBe4P8AlSHnqIkDL5rZCjObHnaYwEDgPeB/gtN1D5lZu7BDVTMJmBl2CHffAfw/YBuwCzjk7i8m83OoUcRYDbWU+E00lZlZe+Bp4Ifu/kHYeQDcvcLdRxD72+qjzCz003Vm9kVgr7uvCDtLLc5197OAy4Cbg9OdYYsCZwEPuPuZwIdAKq0dZgFXAP+bAlm6EDsDMgDoDbQzsynJ/BxqFDGlQN+45/kkeerW2gRrAE8DT7j7M2HnqS44TVEMTAw3CQDnAlcEawGzgAvN7PFwI33G3XcGH/cCzxI7FRu2UqA0bkb4FLHGkSouA1a6+56wgwAXAe+6+3vufgx4BjgnmZ9AjSJmGTDEzAYEvylMAuaEnCllBYvGDwPr3P2/ws5TxcxyzaxzsN2G2P9A60MNBbj77e6e7+79if3bWuDuSf2Nr7HMrF1wQQLBqZ1LgNCvsHP33cB2Mzs5KE0AQr+4JM5kUuC0U2AbUGRmbYP/NycQWzdMmhbxN7Obm7uXm9n3gBeACDDD3deGHAsAM5sJjAO6m1kpcIe7PxxuKs4FpgKrg/UAgH8O/rZ5mHoBjwRXo2QAs909pS5FTUF5wLOxny9EgT+4+/PhRvrU3wNPBL+8bQa+EXIeAMysLbErJL8ddhYAd19iZk8BK4Fy4A2SfCsPXR4rIiJ10qknERGpkxqFiIjUSY1CRETqpEYhIiJ1UqMQEZE6qVGIiEid1ChERKRO/x9/uAf0ku79/wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'cylinders'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 4\n",
    "others = counts.index[n_categorical:]\n",
    "\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6 cylinders    105677\n",
       "4 cylinders     94767\n",
       "8 cylinders     81179\n",
       "others           5450\n",
       "Name: cylinders, dtype: int64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* fuel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcdd46a00>]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'fuel'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 2\n",
    "others = counts.index[n_categorical:]\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gas       399399\n",
       "diesel     31077\n",
       "others     24500\n",
       "Name: fuel, dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* title_status"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfca9c5760>]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'title_status'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "executionInfo": {
     "elapsed": 8,
     "status": "ok",
     "timestamp": 1631959311330,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "3_MSXXfjNazv"
   },
   "outputs": [],
   "source": [
    "# title_status 0이 대부분, 칼럼 삭제\n",
    "df.drop('title_status', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* transmission"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfd0f98190>]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'transmission'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 3\n",
    "others = counts.index[n_categorical:]\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "automatic    396322\n",
       "other         32186\n",
       "manual        27263\n",
       "Name: transmission, dtype: int64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* drive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcae5f370>]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'drive'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 따로 제한을 둘 필요는 없어보임, fillna만 inplace적용\n",
    "df[col].fillna('others', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4wd       139693\n",
       "others    134188\n",
       "fwd       122602\n",
       "rwd        61730\n",
       "Name: drive, dtype: int64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcad151f0>]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'size'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 2\n",
    "others = counts.index[n_categorical:]\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "full-size    69736\n",
       "others       67129\n",
       "Name: size, dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcae65fa0>]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAD4CAYAAADy46FuAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAnLUlEQVR4nO3deXhU5d3/8fc3ewIkISwBk7AjGHaSImDVUESxtWURLK0LtlgerV3t83Pp8thNi7bVrtrygIpLi4hQ0YpKgdTqE1A2gbAIyBb2NRICZOH+/TEnGDAJYRJyZpLP67pyzcw95x4+4YJ8cs6ZuY855xAREalOhN8BREQktKkoRESkRioKERGpkYpCRERqpKIQEZEaRfkdoL61bt3aderUKej5x48fp1mzZvUXqIGEa25Qdr8ouz9CNfvy5csPOufaVPVcoyuKTp06sWzZsqDn5+bmkpOTU3+BGki45gZl94uy+yNUs5vZ9uqe06EnERGpkYpCRERqpKIQEZEaqShERKRGKgoREamRikJERGqkohARkRo1us9RBOtESTm/X7iJru6031FEREKK9ig8h4tLeC5vG8/kn0LX6BAR+YSKwpOWHM/91/ck/9BpXlpe4HccEZGQoaKo5ObLO9KjZQS/fG0d+z8+6XccEZGQoKKoJCLC+FrvWE6VnebH/1irQ1AiIqgoPqVdswjuGXEpb63bxz/X7PE7joiI71QUVZj02c70TU/iwVfyOXy8xO84IiK+UlFUISoygkfH9aXwRCk/fzXf7zgiIr5SUVSjZ7tEvjmsG/9YtZtFG/b5HUdExDcqihp8a1g3Lk1tzo/mruXYyVK/44iI+EJFUYOYqAgeHdePfR+f5FfzN/gdR0TEFyqK8+ifkcykz3bmb0t3kLflkN9xREQanIqiFu4Z0YNOrRK4f85qTpSU+x1HRKRBqShqIT4mkl+N7cv2Q8X89q2NfscREWlQKopaGtK1FV+9vANPvbuVlTuO+B1HRKTBqCguwAPX9yQ1MY77Xl7NqTIdghKRpkFFcQFaxEXz8Jg+fLiviD8v3uJ3HBGRBqGiuEDDerZlzIA0nli8mfV7PvY7jojIRaeiCML/3JBJUnw0985eTVm5rognIo2biiIILZvF8LNRvVizq5Bp72z1O46IyEWlogjSF/q059rMVB5f8CEfHSjyO46IyEVz3qIws6fMbL+Zra00lmJmC8xsk3fbstJzD5jZZjPbaGbXVRrPMrM13nN/MDPzxmPN7EVvfKmZdao0Z6L3Z2wys4n19l3XAzPjl6N7ExsVwf0vr+H0aV3kSEQap9rsUTwDjDxn7H5goXOuO7DQe4yZZQITgF7enCfMLNKb8yQwGejufVW85iTgiHOuG/A48Ij3WinAg8DlwCDgwcqFFAraJsbx4xsyeW/bYV5Yut3vOCIiF8V5i8I59zZw+JzhUcAM7/4MYHSl8ZnOuVPOua3AZmCQmbUHEp1zeS5wfdFnz5lT8VqzgeHe3sZ1wALn3GHn3BFgAZ8uLN+Nz0rnyu6tmTJ/AwVHiv2OIyJS76KCnJfqnNsD4JzbY2ZtvfE0YEml7Qq8sVLv/rnjFXN2eq9VZmaFQKvK41XMOYuZTSawt0Jqaiq5ublBfltQVFR0wfO/1P40731Uzp3T3+YHWbF4R9UaVDC5Q4Wy+0PZ/RGO2YMtiupU9RPS1TAe7JyzB52bCkwFyM7Odjk5OecNWp3c3FyCmX88aSs/fXUdhxK7My4rPeg/P1jB5g4Fyu4PZfdHOGYP9l1P+7zDSXi3+73xAiCj0nbpwG5vPL2K8bPmmFkUkETgUFd1rxWSbhvSieyOLfnFa+vYf+yk33FEROpNsEUxD6h4F9JE4JVK4xO8dzJ1JnDS+j3vMNUxMxvsnX+47Zw5Fa81Dljkncd4E7jWzFp6J7Gv9cZCUkSE8ci4vpwoLefBV3SdbRFpPGrz9ti/A3lADzMrMLNJwBRghJltAkZ4j3HO5QOzgHXAG8DdzrmK1fPuAqYROMG9BZjvjU8HWpnZZuAevHdQOecOA78A3ve+fu6NhayubZrzvWu6M3/tXuav2eN3HBGRenHecxTOua9U89TwarZ/CHioivFlQO8qxk8C46t5raeAp86XMZRMvrILr6/Zw09eyWdI11YkJ8T4HUlEpE70yex6FhUZwaM39uNocQk/f22d33FEROpMRXERZF6SyJ1Xd2XOil3kbtx//gkiIiFMRXGRfHt4N7q1bc6P5q6l6FSZ33FERIKmorhIYqMieeTGvuwuPMEj8zf4HUdEJGgqiosoq2NLvja0M88t2c7Sjw75HUdEJCgqiovsv6+7lA4pCdw/Zw0nS3WdbREJPyqKiywhJoopY/uw9eBxHn1jo99xREQumIqiAQzt1prbhnTkqXe38sgbGwh88FxEJDzU96KAUo0Hv9iL0nLHk7lbKDpZxs++1IuIiIZfZVZE5EKpKBpIZITx8JjeJMZF8de3P6LoVBm/HteXqEjt1IlIaFNRNCAz4/7re5IYH82v39xI0aky/viVAcRFR55/soiIT/TrbAMzM+4e1o2ffakXC9btY9KM9zmuD+SJSAhTUfhk4tBO/HZ8P/K2HOKW6UspLC71O5KISJVUFD66MSudJ27OIn/Xx3x5ah4Hjp3yO5KIyKeoKHw2snc7pt+ezfZDxdz01zwKjhT7HUlE5CwqihBwZfc2PH/HIA4WneKmv+Sx5UCR35FERM5QUYSIrI4pzJw8mFNlp7npL3nk7y70O5KICKCiCCm9Lkli1p1DiI2KYMLUJSzfHtJXfhWRJkJFEWK6tmnOS3cNpXXzWG6Z9h7/2XTA70gi0sSpKEJQWnI8s/5rCB1bJTDpmWW8sXav35FEpAlTUYSoNi1ieXHyEHqlJXL331bw8vICvyOJSBOloghhSQnRPD/pcgZ3SeEHL33AjP/b5nckEWmCVBQhrllsFNMnfoYRmak8OC+fPy3apGXKRaRBqSjCQFx0JE/cPJAxA9L4zVsfMmW+rmkhIg1Hq8eGiejICH47vh/NYwPLlB87VcYvRvUmUte0EJGLTEURRiIijJ+P6kWLuCie8C6A9Nub+hGta1qIyEVUp58wZvZ9M8s3s7Vm9nczizOzFDNbYGabvNuWlbZ/wMw2m9lGM7uu0niWma3xnvuDmZk3HmtmL3rjS82sU13yNgZmxr0je3LfyJ7M+2A3//Xcck6WlvsdS0QasaCLwszSgO8A2c653kAkMAG4H1jonOsOLPQeY2aZ3vO9gJHAE2ZWccWeJ4HJQHfva6Q3Pgk44pzrBjwOPBJs3sbmrpyu/GJ0bxZv3M/tT7/HiTKdsxCRi6OuxyyigHgziwISgN3AKGCG9/wMYLR3fxQw0zl3yjm3FdgMDDKz9kCicy7PBc7QPnvOnIrXmg0Mr9jbELh1cEcev6k/7287wqPvn9QFkETkogj6HIVzbpeZ/QbYAZwA3nLOvWVmqc65Pd42e8ysrTclDVhS6SUKvLFS7/654xVzdnqvVWZmhUAr4GDlLGY2mcAeCampqeTm5gb7bVFUVFSn+Q0tGbizbwx/XnWKR15czOc6RPsd6YKF2995ZcruD2VvWEEXhXfuYRTQGTgKvGRmt9Q0pYoxV8N4TXPOHnBuKjAVIDs72+Xk5NQQo2a5ubnUZb4frnaOuZvnk3+8GT/PGep3nAsWjn/nFZTdH8resOpy6OkaYKtz7oBzrhSYAwwF9nmHk/Bu93vbFwAZleanEzhUVeDdP3f8rDne4a0kQEuqnsPMGHpJFMu3H2H7oeN+xxGRRqYuRbEDGGxmCd55g+HAemAeMNHbZiLwind/HjDBeydTZwInrd/zDlMdM7PB3uvcds6citcaByxy+qRZlYa0j8IM5q7c5XcUEWlkgi4K59xSAieYVwBrvNeaCkwBRpjZJmCE9xjnXD4wC1gHvAHc7ZyreF/nXcA0Aie4twDzvfHpQCsz2wzcg/cOKvm0VvERDO7cirkrd+lT2yJSr+r0gTvn3IPAg+cMnyKwd1HV9g8BD1UxvgzoXcX4SWB8XTI2JWMGpnHv7NWs2HGUrI4tzz9BRKQW9JHeRuT63u2IjYpg7kotSS4i9UdF0Yi0iIvm2l7teG31HkrKTvsdR0QaCRVFIzN2QBpHi0tZvHH/+TcWEakFFUUjc2X31rRuHsPcFXr3k4jUDxVFIxMVGcEX+13Cog37KSwu9TuOiDQCKopGaOyAdErKT/Pamt3n31hE5DxUFI1Q77REurVtrsNPIlIvVBSNkJkxZkAay7YfYcehYr/jiEiYU1E0UqMHBBbg1ZIeIlJXKopGKi05nsFdUpi7skBLeohInagoGrGxA9LZdqiYlTuP+h1FRMKYiqIRu76Pt6SHTmqLSB2oKBqxFnHRjMhM5dXVu7Wkh4gETUXRyI0dGFjSI1dLeohIkFQUjdyV3dvQqlmM3v0kIkFTUTRy0d6SHgvXa0kPEQmOiqIJGDswjZLy0/xzzR6/o4hIGFJRNAF90pLo2qaZLmgkIkFRUTQBZsbYgem8v+0IOw9rSQ8RuTAqiiZiVP9LAC3pISIXTkXRRKS3TODyzinMXblLS3qIyAVRUTQhYwemsfXgcVZpSQ8RuQAqiibk+j7tA0t66PCTiFwAFUUTkhgXzTWZqbz6gZb0EJHaU1E0MWMHpHGkuJR/f3jA7ygiEiZUFE3MVZdWLOmhz1SISO3UqSjMLNnMZpvZBjNbb2ZDzCzFzBaY2SbvtmWl7R8ws81mttHMrqs0nmVma7zn/mBm5o3HmtmL3vhSM+tUl7zyyZIe/1q/n8ITWtJDRM6vrnsUvwfecM71BPoB64H7gYXOue7AQu8xZpYJTAB6ASOBJ8ws0nudJ4HJQHfva6Q3Pgk44pzrBjwOPFLHvAKMGZBGSdlpXteSHiJSC0EXhZklAlcB0wGccyXOuaPAKGCGt9kMYLR3fxQw0zl3yjm3FdgMDDKz9kCicy7PBd7g/+w5cypeazYwvGJvQ4LXNz2JLm2a6YJGIlIrUXWY2wU4ADxtZv2A5cB3gVTn3B4A59weM2vrbZ8GLKk0v8AbK/XunzteMWen91plZlYItAIOVg5iZpMJ7JGQmppKbm5u0N9UUVFRneb75UJz908uYc6m47z0+iLaJPh7qipc/85B2f2i7A2rLkURBQwEvu2cW2pmv8c7zFSNqvYEXA3jNc05e8C5qcBUgOzsbJeTk1NDjJrl5uZSl/l+udDcXfsWM+fRxeyNzWB8TveLF6wWwvXvHJTdL8resOryq2QBUOCcW+o9nk2gOPZ5h5PwbvdX2j6j0vx0YLc3nl7F+FlzzCwKSAIO1yGzeDJSEhikJT1EpBaCLgrn3F5gp5n18IaGA+uAecBEb2wi8Ip3fx4wwXsnU2cCJ63f8w5THTOzwd75h9vOmVPxWuOARU4/1erN2AFpfHTwOB8UFPodRURCWF0PTn8beMHMVgP9gYeBKcAIM9sEjPAe45zLB2YRKJM3gLudc+Xe69wFTCNwgnsLMN8bnw60MrPNwD3UfGhLLtD1fdoTExXB3BX6TIWIVK8u5yhwzq0Csqt4ang12z8EPFTF+DKgdxXjJ4Hxdcko1UuKj2bEZam8unoPP74hk+hIff5SRD5NPxmauDED0jh8vIR/b9SSHiJSNRVFE3d1jzakNIvRirIiUi0VRRMXHRnBF/u2Z8H6fVrSQ0SqpKIQxgxMp6TsNPO1pIeIVEFFIfRLT6JL62bM0eEnEamCikIwM8YOTOO9rYfZebjY7zgiEmJUFALAqP6B5bVeWaW9ChE5m4pCgE+W9JijJT1E5BwqCjlj7IA0PjpwnNVa0kNEKlFRyBlnlvTQSW0RqURFIWecWdLjg92Ulp/2O46IhAgVhZxlzIA0Dh0v4e0PtaSHiASoKOQsFUt66DMVIlJBRSFnObOkx7p9fHxSS3qIiIpCqqAlPUSkMhWFfMqZJT1W6PCTiKgopApmxpgBaSzdepiCI1rSQ6SpU1FIlUYPqFjSY7fPSUTEbyoKqVJGSgKDOqUwZ0WBlvQQaeJUFFKtMQPT2HLgOGt2aUkPkaZMRSHV+ry3pIdOaos0bSoKqVZSfDTXXNZWS3qINHEqCqnRmAHpHDpewn82aUkPkaZKRSE1uvrSwJIeP/lHPvPX7NGJbZEmSEUhNYqJiuB/b8uiRVwUd72wgglTl5C/Wye3RZoSFYWcV1bHFF779mf5xejefLjvGDf88R0emLOGQ0Wn/I4mIg2gzkVhZpFmttLMXvMep5jZAjPb5N22rLTtA2a22cw2mtl1lcazzGyN99wfzMy88Vgze9EbX2pmneqaV4ITFRnBrYM7kvvfw/ja0M68tGwnOb/JZdp/PqKkTCe6RRqz+tij+C6wvtLj+4GFzrnuwELvMWaWCUwAegEjgSfMLNKb8yQwGejufY30xicBR5xz3YDHgUfqIa/UQVJCNP/zxUze+N5VZHVsyS//uZ6Rv3ubRRv26fyFSCNVp6Iws3TgC8C0SsOjgBne/RnA6ErjM51zp5xzW4HNwCAzaw8kOufyXOAnzbPnzKl4rdnA8Iq9DfFXt7bNeeZrg3j69s+AwdefWcbtT7/P5v3H/I4mIvWsrnsUvwPuBSofe0h1zu0B8G7beuNpwM5K2xV4Y2ne/XPHz5rjnCsDCoFWdcws9WhYz7a88d2r+PEXLmPFjiNc97v/8LNX8yks1rUsRBqLqGAnmtkNwH7n3HIzy6nNlCrGXA3jNc05N8tkAoeuSE1NJTc3txZxqlZUVFSn+X7xO3c34KGhMczdVMIz727jpfe2MaZ7DDnpUURG1LwT6Hf2ulB2fyh7A3POBfUF/IrAb//bgL1AMfA8sBFo723THtjo3X8AeKDS/DeBId42GyqNfwX4a+VtvPtRwEHAasqVlZXl6mLx4sV1mu+XUMq9bnehm/DXPNfxvtfctY/9272z6UCN24dS9gul7P5Q9voHLHPV/FwN+tCTc+4B51y6c64TgZPUi5xztwDzgIneZhOBV7z784AJ3juZOhM4af2eCxyeOmZmg73zD7edM6fitcZ5f4bOmIa4y9on8rdvXM5fbhlIcWkZN09byuRnl7H90HG/o4lIEII+9FSDKcAsM5sE7ADGAzjn8s1sFrAOKAPuds6Ve3PuAp4B4oH53hfAdOA5M9sMHCZQSBIGzIyRvduT06Mt09/Zyp8Xb2bEY2/z9c925luf60bz2IvxT09ELoZ6+d/qnMsFcr37h4Dh1Wz3EPBQFePLgN5VjJ/EKxoJT3HRkdw9rBvjs9J59M2N/OXfW5i9vIB7R/Zg3MB0Is5z/kJE/KdPZkuDaJsYx2/G9+OVu6+gQ0o8985ezag/v8uybYf9jiYi56GikAbVLyOZl+8ayu8n9Odg0SnG/SWPv35wkqJTZX5HE5FqqCikwZkZo/qnsfAHV/Odz3Vj6d5ybvpLHnsLT/odTUSqoKIQ3yTERHHPtT34/sBYth86zpgn3mXD3o/9jiUi51BRiO/6tInipTuH4hyMezJPF0kSCTEqCgkJmZckMvfuoaS3jOdrT7/PrPd3nn+SiDQIFYWEjPZJ8bx05xCGdmvNvS+v5jdvbtSKtCIhQEUhIaVFXDTTJ2Yz4TMZ/GnxZr7/4ipOlZWff6KIXDT6eKyEnOjICH41tg8ZKQn8+s2N7Ck8ydRbs0lKiPY7mkiTpD0KCUlmxt3DuvH7Cf1ZueMoY598l52Hi/2OJdIkqSgkpI3qn8ZzkwZxsKiEMU+8y6qdR/2OJNLkqCgk5F3epRVzvjmU+JhIJkzN4838vX5HEmlSVBQSFrq2ac7cb15Bz3aJ3Pn8cp56Z6vfkUSaDBWFhI3WzWP5+zcGc21mKj9/bR0/ezWf8tN6+6zIxaaikLASHxPJEzdnMemznXn63W3c9fxyTpTo7bMiF5OKQsJOZITxkxsy+ekXM1mwfh8TpuZx4Ngpv2OJNFoqCglbt1/Rmb/eksXGfccY++S7bN5f5HckkUZJRSFh7dpe7Xhx8hBOlJxm7BPvsuSjQ35HEml0VBQS9vplJDP3m0NpmxjHrdOX8o+Vu/yOJNKoqCikUchISeDlO4eS1bEl33txFX9atEkLCorUExWFNBpJCdHM+PogxgxI4zdvfch9L6+mtPy037FEwp4WBZRGJTYqksdu6kdGy3j+sGgzewpP8viX+9O6eazf0UTClopCGh0z455re5CeksAP56xh8MMLGX5ZW8ZnZZDTow1RkdqRFrkQKgpptG7KzmBgh5a8+P4O5q7cxZv5+2jTIpaxA9IYn51Ot7Yt/I4oEhZUFNKodWvbnB99IZN7R/Zk8Yb9vLS8gGnvbOWvb3/EgA7JjM/K4IZ+7UmM07UuRKqjopAmIToygmt7tePaXu04cOwU/1i5i1nLdvLDuWv4+Wv5XN+7PeOz0xncuRUREeZ3XJGQoqKQJqdNi1i+cVUX7riyMx8UFDJr2U5eXbWbuSt3kZESz7iBGdyYlUZ6ywS/o4qEhKDP6plZhpktNrP1ZpZvZt/1xlPMbIGZbfJuW1aa84CZbTazjWZ2XaXxLDNb4z33BzMzbzzWzF70xpeaWac6fK8iZzEz+mck8/CYPrz3o2v43Zf70yElgcf/9SFXPrqYm6ct4ZVVuzhZqkUHpWmryx5FGfAD59wKM2sBLDezBcDtwELn3BQzux+4H7jPzDKBCUAv4BLgX2Z2qXOuHHgSmAwsAV4HRgLzgUnAEedcNzObADwCfLkOmUWqFB8TyegBaYwekMbOw8W8vKKAl5YV8N2Zq2gRF8WX+l3C+OwM+qUn4f0eI9JkBF0Uzrk9wB7v/jEzWw+kAaOAHG+zGUAucJ83PtM5dwrYamabgUFmtg1IdM7lAZjZs8BoAkUxCvip91qzgT+ZmTl95FYuooyUBL53zaV853PdWfLRIWYt28ns5QW8sHQHl6Y2Z3xWBqMHpNGmhT6bIU2D1cfPXO+Q0NtAb2CHcy650nNHnHMtzexPwBLn3PPe+HQCZbANmOKcu8YbvxK4zzl3g5mtBUY65wq857YAlzvnDp7z508msEdCampq1syZM4P+XoqKimjevHnQ8/0SrrkhPLIXlzqW7injP7vK+KjwNJEGfdtE0q9lGb1SE2gdb2G3pxEOf+/VUfb6N2zYsOXOueyqnqvzyWwzaw68DHzPOfdxDf9ZqnrC1TBe05yzB5ybCkwFyM7Odjk5OedJXb3c3FzqMt8v4Zobwif7573bTfuO8dLyAuas2MXK/eWw8QQtE6Lpk55Mv/Qk+qQl0Tc9mXZJcb7mPZ9w+XuvirI3rDoVhZlFEyiJF5xzc7zhfWbW3jm3x8zaA/u98QIgo9L0dGC3N55exXjlOQVmFgUkAYfrklmkrrqntuCHn7+M/3ddD154bTFRqd1YU1DIBwVHeSL34JnLs7ZtEUvf9EBp9ElPom9aEq20lIiEoaCLwntn0nRgvXPusUpPzQMmAlO821cqjf/NzB4jcDK7O/Cec67czI6Z2WBgKXAb8MdzXisPGAcs0vkJCRXRkRF0SookZ3DHM2MnSspZt+djVhccPVMeCzfsp+JfbVpy/Jny6JueRO+0JJLi9WE/CW112aO4ArgVWGNmq7yxHxIoiFlmNgnYAYwHcM7lm9ksYB2Bd0zd7b3jCeAu4BkgnsB5i/ne+HTgOe/E92EC75oSCVnxMZFkdWxJVscz7wrn2MlS1u76mDW7jrK6oJDVBYXMX7v3zPOdWzfzDlcFCqR3WiIJMfqIk4SOurzr6R2qPocAMLyaOQ8BD1UxvozAifBzx0/iFY1IuGoRF82Qrq0Y0rXVmbGjxSWsLihkza5CVhcc5f1th5n3QeCIa4QFlh7pl57MjVnpXN45JexOlEvjol9bRHyQnBDDVZe24apL25wZ23/sJGu8PY7VBUd5M38vLy0voHdaInd8tgtf6NueaK18Kz5QUYiEiLYt4hh+WRzDL0sF4GRpOXNW7GL6Ox/xvRdXMWX+Bm6/ohNf+UwHkhJ0XkMajn49EQlRcdGRfPXyDiz4/tU8fftn6Nq2GVPmb2DIlIX8dF4+2w8d9zuiNBHaoxAJcRERxrCebRnWsy3rdn/M9He28sLS7czI28a1manccWUXsju21HkMuWhUFCJhJPOSRH57Uz/uG9mDZ/O28/zS7byZv49+6UlMurIL1/dup/MYUu/0L0okDLVNjOO/r+tB3v3D+eXo3hw7WcZ3/r6Sqx9dzNS3t1B4otTviNKIqChEwlh8TCS3DO7Iv+65mmm3ZdOhVQIPv76Bob9ayM9ezWfn4WK/I0ojoENPIo1ARIRxTWYq12SmsnZXIU+9s5Xn8rYz4/+2cV2vdtxxZWcGdtB5DAmOikKkkemdlsRjX+7PvSN7MiNvGy8s2c78tXvpn5HMHVd2ZmSvdkTpPIZcABWFSCPVLimO+0b25FvDuvHyigKeemcr3/rbStKS4/naFZ1ILD5N4YlSWsRG6TrhUiMVhUgj1yw2ituGdOLmyzuycP0+pr2zlV/+cz0A9779FhEWWGYkOSGapPhPf30yHvOpsYSYSB3OagJUFCJNRGSEcW2vdlzbqx1rdxUyd/F7tO/YlcITpWe+jhYHbncdOcFRb6xi2fSqREUYyQnRJFYUiHfbqnksme0T6ZueRJc2zYnUHktYU1GINEG905I4mB5NzpVdatzOOcfxknKOFpd8UijFlYqlUskUFpdysKiELQeOs//YSU6WngagWUwkvdIC1+Pok55Ev/RkOrZK0J5IGFFRiEi1zIzmsVE0j40iveX5t69Qftqx5UBRYIXcgqOs3lXIs0u2U1IWKI/EuKjAxZzSk88USFpyvMojRKkoRKTeRUYYl6a24NLUFozLClzAsrT8NB/uOxZYIddbXv1/3/6IMu/QVqtmMWeuBFhxWdm2iaF9OdmmQkUhIg0iOjKCXpck0euSpDNXIDtZWs7GvccCxbHzKGt2FfL2hweoOC2SmhhLn7Rk76JOgeuR63KyDU9FISK+iYuOpF9GMv0yksG7pGzgcrKFfLDzkws7Ldyw76zLycZziulblpIQE0mzmCgSYiNJiIkiISbS+4o6c9ss9tNjCTGRxEdH6m3BtaSiEJGQEricbApZHVPOjFW+nOyaXR+zpWAvx06Wsf/jUxwvKeNESTnHS8rOnECv9Z8VHekVySclk9YygQmfyWBo11Y6Z+JRUYhIyDv3crK5ubnk5Fzxqe3KTztOlJZTfKqMYq88AiXyyVhxScVz5ZwoKfNuyzl+qowTpeW8s+kAr36wmy5tmnHr4I7cmJVOYlzTvlCUikJEGo3IiE/epRWsk6Xl/HP1Hp5bsp2fvbqOX7+5kdED0rhtSEd6tkusx7ThQ0UhIlJJXHQkN2alc2NWOmsKCnk2bxsvLy/gb0t3MKhTCrcO6cjIJnbdDxWFiEg1+qQn8evx/fjh5y/jpeU7eX7JDr7995W0aRHLVwZ14KuDOtAuqfG/hVdFISJyHi2bxTD5qq7c8dku/PvDAzybt40/LtrEnxdv5rpeqdw6uBODu6Q02pPfKgoRkVqqfP3yHYeKeX7pdmYt28nra/ZyaWpzbh3ckTED0+t0jiQUNZ2DbCIi9ahDqwR++PnLWPLAcB4d15fYqEh+8ko+gx9eyP+8spZN+475HbHeNK7aExFpYHHRkdyUncH4rHRW7TzKc3nbmfneTp7N286QLq24bUhHRmSmhvXFosKiKMxsJPB7IBKY5pyb4nMkEZGzmBkDOrRkQIeW/OgLlzFrWQHPL9nOXS+soF1iHF+9vAMTBmX4HTMoIV8UZhYJ/BkYARQA75vZPOfcOn+TiYhUrVXzWO7K6crkq7qweMN+nl2ynccWfMjv/vUhUREQvegNzAwDzAIlYwYRZ8YCjw1vzLt/1naVnse7n3lJEn/8yoB6/35CviiAQcBm59xHAGY2ExgFqChEJKRFRhjXZKZyTWYqWw8eZ96q3azf/BHp6RmcduBwOBe47ocDTjvvMd6YA+e8ce95vOcrb1sx3iEl/qJ8H+Zc9VevCgVmNg4Y6Zy7w3t8K3C5c+5blbaZDEwGSE1NzZo5c2bQf15RURHNmzevW2gfhGtuUHa/KLs/QjX7sGHDljvnsqt6Lhz2KKp6Y/JZ7eacmwpMBcjOznY5OTlB/2GBNWSCn++XcM0Nyu4XZfdHOGYPh9PwBUDlM0DpwG6fsoiINDnhUBTvA93NrLOZxQATgHk+ZxIRaTJC/tCTc67MzL4FvEng7bFPOefyfY4lItJkhHxRADjnXgde9zuHiEhTFA6HnkRExEcqChERqZGKQkREahTyH7i7UGZ2ANheh5doDRyspzgNKVxzg7L7Rdn9EarZOzrn2lT1RKMriroys2XVfToxlIVrblB2vyi7P8Ixuw49iYhIjVQUIiJSIxXFp031O0CQwjU3KLtflN0fYZdd5yhERKRG2qMQEZEaqShERKRGKgqPmY00s41mttnM7vc7T22ZWYaZLTaz9WaWb2bf9TvThTKzSDNbaWav+Z3lQphZspnNNrMN3t//EL8z1YaZfd/7t7LWzP5uZnF+Z6qJmT1lZvvNbG2lsRQzW2Bmm7zbln5mrEo1uX/t/XtZbWZzzSzZx4i1pqLgrOtyXw9kAl8xs0x/U9VaGfAD59xlwGDg7jDKXuG7wHq/QwTh98AbzrmeQD/C4HswszTgO0C2c643gRWZJ/ib6ryeAUaeM3Y/sNA51x1Y6D0ONc/w6dwLgN7Oub7Ah8ADDR0qGCqKgDPX5XbOlQAV1+UOec65Pc65Fd79YwR+WKX5m6r2zCwd+AIwze8sF8LMEoGrgOkAzrkS59xRX0PVXhQQb2ZRQAIhfiEw59zbwOFzhkcBM7z7M4DRDZmpNqrK7Zx7yzlX5j1cQuBCbCFPRRGQBuys9LiAMPphW8HMOgEDgKU+R7kQvwPuBU77nONCdQEOAE97h82mmVkzv0Odj3NuF/AbYAewByh0zr3lb6qgpDrn9kDglyWgrc95gvF1YL7fIWpDRRFw3utyhzozaw68DHzPOfex33lqw8xuAPY755b7nSUIUcBA4Enn3ADgOKF5+OMs3rH8UUBn4BKgmZnd4m+qpsfMfkTgsPELfmepDRVFQFhfl9vMogmUxAvOuTl+57kAVwBfMrNtBA73fc7Mnvc3Uq0VAAXOuYq9t9kEiiPUXQNsdc4dcM6VAnOAoT5nCsY+M2sP4N3u9zlPrZnZROAG4GYXJh9kU1EEhO11uc3MCBwnX++ce8zvPBfCOfeAcy7dOdeJwN/5IudcWPx265zbC+w0sx7e0HBgnY+RamsHMNjMErx/O8MJg5PwVZgHTPTuTwRe8TFLrZnZSOA+4EvOuWK/89SWioLAdbmBiutyrwdmhdF1ua8AbiXw2/gq7+vzfodqIr4NvGBmq4H+wMP+xjk/bw9oNrACWEPgZ0BILylhZn8H8oAeZlZgZpOAKcAIM9sEjPAeh5Rqcv8JaAEs8P6v/sXXkLWkJTxERKRG2qMQEZEaqShERKRGKgoREamRikJERGqkohARkRqpKEREpEYqChERqdH/B1sg+UDeWZNGAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "col = 'type'\n",
    "counts = df[col].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_categorical = 8\n",
    "others = counts.index[n_categorical:]\n",
    "df[col] = df[col].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sedan        92364\n",
       "SUV          79882\n",
       "pickup       43343\n",
       "truck        35694\n",
       "others       34538\n",
       "other        24160\n",
       "coupe        19106\n",
       "hatchback    16388\n",
       "Name: type, dtype: int64"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[df[col] == 'other', col] = 'others'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sedan        92364\n",
       "SUV          79882\n",
       "others       58698\n",
       "pickup       43343\n",
       "truck        35694\n",
       "coupe        19106\n",
       "hatchback    16388\n",
       "Name: type, dtype: int64"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[col].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* paint_color"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 282
    },
    "executionInfo": {
     "elapsed": 554,
     "status": "ok",
     "timestamp": 1631959311877,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "MMKF_fe_Nyn9",
    "outputId": "8786e9d3-2c5e-4bdd-efb3-861ed33952f9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfcd1a8f10>]"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "counts = df['paint_color'].fillna('others').value_counts()\n",
    "plt.grid()\n",
    "plt.plot(range(len(counts)), counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "executionInfo": {
     "elapsed": 284,
     "status": "ok",
     "timestamp": 1631959312153,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "pNDhBAcjOLlD"
   },
   "outputs": [],
   "source": [
    "n_categorical = 7\n",
    "others = counts.index[n_categorical:]\n",
    "df['paint_color'] = df['paint_color'].apply(lambda x: x if str(x) not in others else 'others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 20,
     "status": "ok",
     "timestamp": 1631959312153,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "jncdoeFsOrGZ",
    "outputId": "7f2c186d-45b1-44c9-b478-58f2eea3a76c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "white     82786\n",
       "black     64145\n",
       "silver    46722\n",
       "red       33274\n",
       "blue      32746\n",
       "grey      30455\n",
       "others    27242\n",
       "Name: paint_color, dtype: int64"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['paint_color'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xsd_T_h_lgtY"
   },
   "source": [
    "### 수치형 데이터"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nKAQKekDPzT5"
   },
   "source": [
    "* price 피쳐"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 9,
     "status": "ok",
     "timestamp": 1631959312155,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "PRYiAxnKlgtY",
    "outputId": "d4197177-e133-4733-d6db-56f1e56ada0f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59900.0 651.0\n"
     ]
    }
   ],
   "source": [
    "# quantile() 메소드를 이용하여 outlier 제거\n",
    "p1 = df['price'].quantile(0.99)  # 상위1%\n",
    "p2 = df['price'].quantile(0.1)   # 하위 1%\n",
    "\n",
    "print(p1,p2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "executionInfo": {
     "elapsed": 6,
     "status": "ok",
     "timestamp": 1631959312155,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "3emahh_-Pcwc"
   },
   "outputs": [],
   "source": [
    "df = df[(p1 > df['price']) & (p2 < df['price'])]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "I_J40UeeP1Tm"
   },
   "source": [
    "* odometer 피쳐"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 6,
     "status": "ok",
     "timestamp": 1631959312155,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "EhBzabWGP3Ze",
    "outputId": "70f99901-2b10-4723-c07b-784f383d9239"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "272803.10999999824 18593.0\n"
     ]
    }
   ],
   "source": [
    "o1 = df['odometer'].quantile(0.99)\n",
    "o2 = df['odometer'].quantile(0.1) # 하위 10%\n",
    "\n",
    "print(o1, o2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "executionInfo": {
     "elapsed": 275,
     "status": "ok",
     "timestamp": 1631959312426,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "7MNd_ZlIQBmv"
   },
   "outputs": [],
   "source": [
    "df = df[(o1 > df['odometer']) & (o2 < df['odometer'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>odometer</th>\n",
       "      <th>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>322166.000000</td>\n",
       "      <td>322166.000000</td>\n",
       "      <td>321666.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>15211.239072</td>\n",
       "      <td>103341.828089</td>\n",
       "      <td>10.220235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>11248.543565</td>\n",
       "      <td>55049.524247</td>\n",
       "      <td>7.071633</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>652.000000</td>\n",
       "      <td>18596.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6500.000000</td>\n",
       "      <td>57454.000000</td>\n",
       "      <td>6.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12000.000000</td>\n",
       "      <td>98850.000000</td>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>20990.000000</td>\n",
       "      <td>141000.000000</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>59895.000000</td>\n",
       "      <td>272742.000000</td>\n",
       "      <td>121.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               price       odometer            age\n",
       "count  322166.000000  322166.000000  321666.000000\n",
       "mean    15211.239072  103341.828089      10.220235\n",
       "std     11248.543565   55049.524247       7.071633\n",
       "min       652.000000   18596.000000       0.000000\n",
       "25%      6500.000000   57454.000000       6.000000\n",
       "50%     12000.000000   98850.000000       9.000000\n",
       "75%     20990.000000  141000.000000      13.000000\n",
       "max     59895.000000  272742.000000     121.000000"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. 클리닝 이후 다시 시각화"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 301
    },
    "executionInfo": {
     "elapsed": 1251,
     "status": "ok",
     "timestamp": 1631959313676,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "lsABlg2zQW7I",
    "outputId": "1c498111-7ac6-41e2-ea7b-f91618de717a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJ8AAAFhCAYAAADeLQu/AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABB2klEQVR4nO3de5hdZXn///edA4ckHJNAAkOImohFCwiRUg+IxSBRwUM9YK2mlkp/fJVg8VBprSKForZoHf3qF5XiQG0REStYIokIoqLEcD5ZM9YII4FkQgg5ADndvz/2GpgJyWRPnDVr1uz367rmmv08e6+1PzsLZvbc+zlEZiJJkiRJkiSVYVTVASRJkiRJkjRyWXySJEmSJElSaSw+SZIkSZIkqTQWnyRJkiRJklQai0+SJEmSJEkqzZiqAwy1SZMm5fTp06uOIUmSJEmSNGLceuut3Zk5eVv3tVzxafr06SxevLjqGJIkSZIkSSNGRPx2e/c57U6SJEmSJEmlsfgkSZIkSZKk0lh8kiRJkiRJUmksPkmSJEmSJKk0Fp8kSZIkSZJUGotPkiRJkiRJKo3FJ0mSJEmSJJWm1OJTROwdEVdGxC8j4v6I+OOI2DciFkbEkuL7Pr0ef3ZEdEbE/0TEa3r1HxURdxf3tUdEFP27RsQ3i/5bImJ6ma9nuOvu7uaMM85g5cqVVUfRAF1wwQUce+yx/PM//3PVUbQTFi1axHHHHcett95adRRJkiRJGnbKHvn0eeD7mfkC4HDgfuCjwPWZORO4vmgTEYcCpwAvBE4EvhQRo4vzfBk4DZhZfJ1Y9J8KrMrMGcDngE+X/HqGtYsuuog777yTiy66qOooGqD58+cDcM0111ScRDvj7LPPZsuWLXzkIx+pOookSVK//MBaUhVKKz5FxJ7AscDFAJm5ITMfA94AdBQP6wDeWNx+A3B5Zj6Vmb8BOoGjI2IqsGdm/iwzE7h0q2N6znUlcHzPqKhW093dzcKFCwFYsGCBv0xq5IILLujTdvRTvSxatIiNGzcCsHHjRkc/SZKkYa2jo4O77rqLjo6OHT9YkgbJmBLP/VxgBXBJRBwO3AqcCeyfmcsAMnNZROxXPP5A4Oe9ju8q+jYWt7fu7znmweJcmyJiNTAR6O4dJCJOozFyimnTpg3W6xtWLrroIrZs2QLAli1buOiii/i7v/u7ilOpGT2jnnpcc801fPjDH64ojQbq7LPP7tP+yEc+wvXXX19RGu2M7u5uPvnJT3LOOecwceLEquNIw057ezudnZ2Dft6ursbbu7a2tkE/94wZM5g3b96gn1equ+7ubq699loyk2uvvZa5c+f6u68EZf3cBH92qr7KnHY3BjgS+HJmvhhYRzHFbju2NWIp++nv75i+HZlfycxZmTlr8uTJ/aeuqa3/2P3BD35QURKptfSMetpeW8OfnwBL1XjiiSd44oknqo4htZSOjo6n36ts2LDB33015M9O1VWZI5+6gK7MvKVoX0mj+PRIREwtRj1NBZb3evxBvY5vAx4q+tu20d/7mK6IGAPsBTxaxosZ7jZt2tRvW5L0bN3d3cyfP5/MZP78+X4CLG1DWZ+C95y3vb29lPNLerYFCxb0aV933XWcddZZFaUZucocPeTPTtVVaSOfMvNh4MGIOKToOh64D7gamFv0zQW+W9y+Gjil2MHuOTQWFl9UTNFbExHHFOs5vXurY3rO9Rbgh8W6UJIk7VBHRwc9vza2bNniJ8A146K5kjQwY8aM6bctSWUpe7e7M4BvRMRdwBHAPwGfAmZHxBJgdtEmM+8FrqBRoPo+8L7M3Fyc53TgazQWIf810LNIzsXAxIjoBM6i/2l9kiT1sXDhwj4Lxm/9ibCGN6dMStLArFmzpt+2JJWl1FJ3Zt4BzNrGXcdv5/HnA+dvo38x8KJt9D8JvPX3SylJalWzZ8/mv//7v9m0aRNjxozhhBNOqDqSmuSUSUmSpPooe+STJEnD1ty5c/vsFDp37twdHKHhwimTkiRJ9WHxSZIk1Y5TJiVp4F71qlf1ac+ePbuiJJJajcUnSVLL6ujoYNSoxq/CUaNGOXqmRmbPns3YsWMBGDt2rFMmJakJt9xyS5/2T3/604qSSGo1Fp8kSS1r4cKFbNq0CYBNmzY5eqZGek+ZzEynTEpSE9avX99vW5LKYvFJktSyHD1TX5MmTSIinm672LgkSdLwZfFJktSy5s6d+3QBY9SoUY6eqZFFixb1GbV26623VpxIkiRJ22PxSZLUsiZNmsScOXOICObMmePomRo555xz+rT/4R/+oZogklQjo0eP7rctSWWx+CRJamknnXQS48aN4+STT646igZg7dq1/bYlSc+2efPmftuSVBaLT5KklnbNNdewfv16rr766qqjaAAmTJjQb1uS9Gz+7JRUFYtPkqSW1d3dzfz588lM5s+fz8qVK6uOpCadddZZfdof/vCHK0oiSfVx2GGH9WkfeeSRFSWR1GosPkmSWlZHRweZCcCWLVvo6OioOJGadeedd/Zp33777RUlkaT6uPnmm/u0b7rppoqSSGo1Fp8kSS1r4cKFbNy4EYCNGzeyYMGCihOpWdddd12f9ve///2KkkiSJGlHLD5JklrW7Nmz+7RPOOGEipJooMaMGdNvW5IkScOHxSdJUst6xSte0af9yle+sqIkGih3u5MkSaoPi0+SpJb1r//6r33aF154YTVBNGCOfJIkSaoPi0+SpJbV1dXVb1vD16ZNm/ptS5Ikafiw+CRJalkR0W9bw9f06dP7bUuSJA0n3d3dnHHGGaxcubLqKJWw+CRJallbr/F03HHHVRNEA/b+97+/T/vMM8+sKIkk1ceoUaP6bUsqT0dHB3fddRcdHR1VR6mEP20kSS1r3rx5T492igjmzZtXcSI1a+HChX3a1113XUVJJKk+xo0b129bUjm6u7uZP38+mcn8+fNbcvSTxSdJUsuaNGkSRx55JACzZs1i4sSJFSdSs7YuPi1YsKCiJJJUH+4UKlWjo6ODzARgy5YtLTn6yeKTJKmlPfTQQ4CLjdfNli1b+m1Lkp7N9fKkaixcuJCNGzcCsHHjxpb80MzikySpZf3qV79i2bJlACxbtozOzs6KE0mSVJ6Pfexjfdof//jHK0oitZbZs2czduxYAMaOHcsJJ5xQcaKhZ/FJktSyzjnnnD5t34TXx/jx4/ttS5Ke7bHHHuvTXr16dTVBpBYzd+7cp9cZHTVqFHPnzq040dCz+CRJallbT7Vz6l19bN68ud+2JOnZPvShD/Vp/83f/E1FSaTWMmnSJObMmUNEMGfOnJZcZ9TikySpZfV8ArW9toav17zmNX3aJ554YkVJJEmSdmzu3LkcdthhLTnqCSw+SZJa2JQpU/q0p06dWlESDdTWb9xa9Y2cJEmqh0mTJvGFL3yhJUc9gcUnSVIL61lsvEfPznca/h599NE+7VWrVlWURJIkSTti8UmSJNXOeeed16d97rnnVpREkiRJO2LxSZIk1c7SpUv7bUuSJGn4GFN1AEmSNLK1t7fT2dk5qOfcddddeeqpp/q0582bN6jPMWPGjEE/pyRJUity5JMkSaqdgw8+uN+2JEmShg9HPklqGWWMvthjjz1Ys2ZNn/ZgjpRw5EW5IoLM7NPW4Cvrv+HZs2fz1FNPMX36dL72ta+V8hySNJKMGzeO9evXP90eP358hWkktRJHPknS72HatGn9tjW8feADH+jT/uAHP1hNEO2Ugw8+mFGjRvHxj3+86iiSVAu9C08A69atqyiJ1Hp+9atfMWfOnEH/MLwuSh35FBFLgTXAZmBTZs6KiH2BbwLTgaXA2zJzVfH4s4FTi8fPy8zriv6jgK8DuwPXAmdmZkbErsClwFHASuDtmbm0zNckDTZHXgydskZfvO51r2PNmjW89KUv5VOf+lQpz6FyvOlNb+Jzn/vc0+2TTz65wjQaqHHjxnHYYYcxY8aMqqNIUi3sv//+PPLII0+3p0yZUmEaqbWcd955rFu3jnPPPZdLL7206jhDbihGPr0qM4/IzFlF+6PA9Zk5E7i+aBMRhwKnAC8ETgS+FBGji2O+DJwGzCy+Tiz6TwVWZeYM4HPAp4fg9UiDqnfhaVttDX/Tpk1j/PjxfPjDH646igborLPO6tP+yEc+UlESSZLK17vwBPDwww9XlERqLb/61a+e3pl36dKlLTn6qYppd28AOorbHcAbe/VfnplPZeZvgE7g6IiYCuyZmT/Lxl/ll251TM+5rgSOD4eNSBpiY8eOZebMmUycOLHqKBqgxYsX92n//Oc/ryiJJEmSRqrzzjuvT/vcc8+tKEl1yi4+JbAgIm6NiNOKvv0zcxlA8X2/ov9A4MFex3YVfQcWt7fu73NMZm4CVgPP+usvIk6LiMURsXjFihWD8sIkSZIkSZJ2pGfU0/baraDs4tPLMvNIYA7wvog4tp/HbmvEUvbT398xfTsyv5KZszJz1uTJk3eUWZIkSZIkaVBMnz6933YrKHXB8cx8qPi+PCK+AxwNPBIRUzNzWTGlbnnx8C7goF6HtwEPFf1t2+jvfUxXRIwB9gIeLev1DJYytnvfFrd7l6T+7bXXXqxevfrp9j777FNhGkmSJI1E73//+/nQhz70dPvMM8+sME01Shv5FBHjI2KPntvACcA9wNXA3OJhc4HvFrevBk6JiF0j4jk0FhZfVEzNWxMRxxTrOb17q2N6zvUW4Ifpas2SpCbtueeefdoTJkyoKIkkSeXbenlcl8uVhsaPf/zjPu0f/ehHFSWpTpkjn/YHvlP8QBsD/Edmfj8ifgFcERGnAg8AbwXIzHsj4grgPmAT8L7M3Fyc63Tg68DuwPziC+Bi4LKI6KQx4umUEl/PoCljBNEFF1zA/Pnzn26fdNJJ7rwlSTvw4IMP9tuWJGkkcZdlqRoLFy7s016wYMGzdl0e6Uob+ZSZ/5uZhxdfL8zM84v+lZl5fGbOLL4/2uuY8zPzeZl5SGbO79W/ODNfVNz3/p7RTZn5ZGa+NTNnZObRmfm/Zb2e4e7ss8/u07bwJEmSpDJ0d3dzxhlnsHLlyqqjaIDGjBnTb1tSOWbPns3YsWOBxk7ZJ5xwQsWJhl7ZC45rCO29995AY9STJEmSVIaOjg7uuusuOjo6qo6iATr66KP7tF/60pdWlERqLXPnzn16muuoUaOYO3fuDo4YeSw+jSDTp0/niCOOcNSTJEmSStHd3c38+fPJTObPn+/op5q5+eab+7RvuummipJIrWXSpEnMmTOHiGDOnDlMnDix6khDzuKTJEmSpKZ0dHQ8vU7Qli1bHP0kSU2aO3cuhx12WEuOegKLT5IkSZKatHDhQjZu3AjAxo0bWbBgQcWJJEl1YPFJkiRJUlNcNLfe3vnOd/Zpt+oIDKkKF110EXfeeScXXXRR1VEqYfFJkiRJUlN6L5obERYvaubuu+/uty2pHN3d3SxcuBCABQsWtOR6eRafJEmSJDVl0qRJHHDAAQAccMABLblobp3dddddfdq33XZbRUmk1nLRRRexZcsWoLFeXiuOfrL4JEmSJKkp3d3d/O53vwPgoYceaslP7yVpoH7wgx/0afeMgmolFp8kSZIkNaX37naZ6W53ktSEnunK22u3AotPkiRJkpribnf1tvU0yf3226+iJFJrefnLX96n/YpXvKKiJNWx+CRJalnjx4/vty1J6svd7upt62mSy5cvryiJ1Fp23XXXftutwOKTJKllrVu3rt+2JKkvd7uTpIH78Y9/3Kd90003VZSkOhafJEmSJDXF3e4kaeBmz57dp92Ko0YtPkmSJElqirvdSdLATZgwoU97r732qihJdSw+SZIkSWqKu91J0sB94xvf6NNuxZ+dFp8kSZIkNcXd7iRJO8PikyRJkqSmuNudJGlnWHySJEmS1JTeu92NGjXK3e4kqQk9RfvttVuBxSdJkiRJTZk0aRJz5swhIpgzZ4673UlSE1772tf2ab/uda+rKEl1LD5JkiRJatpJJ53EuHHjOPnkk6uOIkm18IpXvKJP+5WvfGVFSapj8UmSJElS06655hrWr1/P1VdfXXUUSaqFL37xi33an//85ytKUh2LT5IkSZKa0t3dzbXXXktmcu2117Jy5cqqI2kA9tprrz7tffbZp6IkUmtZunRpv+1WYPFJkiRJUlM6OjrYtGkTABs3bqSjo6PiRBqI1atX92mvWrWqoiSSWo3FJ0mSJElNWbBgAZkJQGZy3XXXVZxIklQHFp8kSZIkNWX//ffvty1J0rZYfJIkSZLUlEceeaTftiRJ22LxSZIkSVJTTjjhBCICgIjgNa95TcWJJGn4Gz16dL/tVmDxSZIkSVJT5s6dy9ixYwEYO3Ysc+fOrTiRJA1/mzdv7rfdCiw+SZIkSWrKpEmTmDNnDhHBa1/7WiZOnFh1JElSDYypOoAkSZKk+pg7dy5Lly511JMkqWmOfJIkSZLUtEmTJvGFL3zBUU81tN9++/VpT5kypaIkUms57LDD+rQPP/zwipJUx+KTJEmSJLWA5cuX92k//PDDFSWRWsuvf/3rPu3Ozs6KklTH4pMkSZKkpnV3d3PGGWewcuXKqqNIUi2sW7eu33YrKH3Np4gYDSwGfpeZr4+IfYFvAtOBpcDbMnNV8dizgVOBzcC8zLyu6D8K+DqwO3AtcGZmZkTsClwKHAWsBN6emUvLfk2SpKHV3t4+ZJ8QzZs3b9DONWPGjEE9nyQNB+3t7dx55520t7fzyU9+suo4kjTsjR8/vk/Bafz48RWmqcZQjHw6E7i/V/ujwPWZORO4vmgTEYcCpwAvBE4EvlQUrgC+DJwGzCy+Tiz6TwVWZeYM4HPAp8t9KZIkSVLr6u7u5kc/+hEAN954o6OfJKkJ++yzT592K66ZV+rIp4hoA14HnA+cVXS/ATiuuN0B3Aj8bdF/eWY+BfwmIjqBoyNiKbBnZv6sOOelwBuB+cUx5xTnuhL4YkREZmaZr0uSNLTKGj101llnsXjx4qfbxxxzDJ/5zGdKeS5JGgna29vpeaudmY5+kqQmdHV19Wk/8MADFSWpTtkjn/4V+AiwpVff/pm5DKD43rPlwoHAg70e11X0HVjc3rq/zzGZuQlYDTyrhBgRp0XE4ohYvGLFit/zJUmSRorPfvazfdoWniSpfz2jnnrceOON1QSRJNVKacWniHg9sDwzb232kG30ZT/9/R3TtyPzK5k5KzNnTZ48uck4kqRWMG7cOKAx6kmS1L+tJxg44UCS1Iwyp929DDg5Il4L7AbsGRH/DjwSEVMzc1lETAV69vvsAg7qdXwb8FDR37aN/t7HdEXEGGAv4NGyXpAkaeR5/vOfDzjqSZKasd9++7F8+fI+bUmSdqS0kU+ZeXZmtmXmdBoLif8wM/8cuBqYWzxsLvDd4vbVwCkRsWtEPIfGwuKLiql5ayLimIgI4N1bHdNzrrcUz+HHL5IkSVIJRo8e3ac9Zkzpm2dLkkaAKn5bfAq4IiJOBR4A3gqQmfdGxBXAfcAm4H2Zubk45nTg68DuNBYan1/0XwxcVixO/iiNIpckSZKkEixbtqxP+6GHHtrOIyVJesaQFJ8y80Yau9qRmSuB47fzuPNp7Iy3df9i4EXb6H+SonglSZIkqVxTp07tU4A64IADKkwjSaqLsne7kyRJkjRCbNy4sU97w4YNFSWRpPporCD0jFGjWq8U03qvWJIkSdJO6e7u7rctSXq2rZem3rJlS0VJqmPxSZIkSZIkSaWx+CRJkiRJkqTSWHySJEmS1JR99923T3vixIkVJZEk1YnFJ0mSJElNWbduXZ/22rVrK0oiSaoTi0+SJEmSmvLUU0/125YkaVssPkmSJEmSJKk0Fp8kSZIkSZJUmjFVB5DqpL29nc7OztKfZ968eYN6vhkzZgz6OSVJkiRJaoYjnyRJkiRJklQaRz5JA1DG6KFjjz32WX3t7e2D/jySJEmSJFXBkU+SJEmSJEkqjcUnqWI33XRTv21JkiRJkurMaXeSJEk1N1QbYgymJUuWAOVMaS+Tm3hIkjRwFp+kYeCII44AXOtJkrRzOjs7uefOO9ljl/q8tdu0aTMAv73/3oqTNG/Nhk1VR5AkqZbq8w5FkiRJ27XHLmM4ev99qo4xoi16ZFXVESRJqiWLT5IkSZKaMnbsWDZu3Ph0e5dddqkwzcg2VNNpB3MaqdNSJW2PC45LkiRJakrvwhPAhg0bKkoiSaoTRz5JkiRJ0jBTxgiiyy67jK9+9atPt08//XTe8Y53DPrzSNLWHPkkSZIkSS3gXe96V5+2hSdJQ8XikyRJkiS1iClTpgCNUU+SNFScdidJkiRJLWLKlClMmTLFUU+ShpQjnyRJkiRJklQai0+SJEmSJEkqTdPFp4g4OCJeXdzePSL2KC+WJEmSJEmSRoKmik8R8V7gSuCioqsN+K+SMkmSJEmSJGmEaHbk0/uAlwGPA2TmEmC/skJJkiRJkiRpZGi2+PRUZm7oaUTEGCDLiSRJkiRJkqSRotni048i4u+A3SNiNvAt4JryYkmSJEmSJGkkGNPk4z4KnArcDfw1cC3wtbJCSZIkSZJUpvb2djo7O6uOMSBLliwBYN68eRUnad6MGTNqlVflaLb4tDvwb5n5VYCIGF30rS8rmCRJkiRJZens7OS++25n0uQ6rSgTACxfcVvFOZrTvSKqjqBhotni0/XAq4G1RXt3YAHw0u0dEBG7ATcBuxbPc2VmfiIi9gW+CUwHlgJvy8xVxTFn0xhhtRmYl5nXFf1HAV8vnvda4MzMzIjYFbgUOApYCbw9M5c2+ZokSZIkSS1s0uTkzW/esOMHaqdcddUuVUfQMNHsmk+7ZWZP4Yni9rgdHPMU8CeZeThwBHBiRBxDYwrf9Zk5k0ZR66MAEXEocArwQuBE4EvFCCuALwOnATOLrxOL/lOBVZk5A/gc8OkmX48kSZIkSZKGQLPFp3URcWRPoxiJ9ER/B2RDT8FqbPGVwBuAjqK/A3hjcfsNwOWZ+VRm/gboBI6OiKnAnpn5s8xMGiOdeh/Tc64rgeMjwnF9kiRJkiRJw0Sz0+4+AHwrIh4q2lOBt+/ooGLk0q3ADOD/ZuYtEbF/Zi4DyMxlEbFf8fADgZ/3Oryr6NtY3N66v+eYB4tzbYqI1cBEoHurHKfRGDnFtGnTmnm9kiRJkiRJGgRNFZ8y8xcR8QLgEBornP0yMzc2cdxm4IiI2Bv4TkS8qJ+Hb2vEUvbT398xW+f4CvAVgFmzZtVpNTlJkiRJkqRa67f4FBF/kpk/jIg3b3XXzIggM69q5kky87GIuJHGWk2PRMTUYtTTVGB58bAu4KBeh7UBDxX9bdvo731MV0SMAfYCHm0mkyRJkiRJksq3ozWfXll8P2kbX6/v78CImFyMeCIidqexW94vgauBucXD5gLfLW5fDZwSEbtGxHNoLCy+qJiityYijinWc3r3Vsf0nOstwA+LdaEkSZIkSZI0DPQ78ikzPxERo4D5mXnFAM89Fego1n0aBVyRmd+LiJ8BV0TEqcADwFuL57o3Iq4A7gM2Ae8rpu0BnA58HdgdmF98AVwMXBYRnTRGPJ0ywIySJEmSJEkq0Q7XfMrMLRHxfmBAxafMvAt48Tb6VwLHb+eY84Hzt9G/GHjWelGZ+SRF8UqSJEmSJEnDz46m3fVYGBEfioiDImLfnq9Sk0mSJEmSJKn2mtrtDvhLGrvI/Z+t+p87uHEkSZIkSZI0kjRbfDqURuHp5TSKUD8G/l9ZoSRJkiRJkoZae3s7nZ2dpT/PvHnzBvV8M2bMGPRzDqZmp911AH8AtANfKG53lBVKkiRJkiRpJNh3376rFk2cOLGiJNVpduTTIZl5eK/2DRFxZxmBJEmSJP3+/PRekgaurJ8/xx577NO3v/Od75TyHMNZsyOfbo+IY3oaEfFHwE/LiSRJkiRJkjRy9Ix+etvb3lZxkmo0O/Lpj4B3R8QDRXsacH9E3A1kZh5WSjpJkiRJO6WMT+97f3Lfo729fdCfR5JGmmnTpjFt2jTe//73Vx2lEs0Wn04sNYUkSZKkYe+9730vX/3qV59un3766RWmkSTVRVPT7jLzt/19lR1SkiRJUvXe9a539Wm/4x3vqCiJJKlOml3zSZIkSZKYMmUK4KgnSVLzmp12J0mSRrih2hlrsCxZsgQob1easpSxk1dXVxdrNmxi0SOrBvW86mvNhk10dXVVHaNyU6ZMYcqUKY56kiQ1zeKTJEkCoLOzk3vvvp+9x+1XdZSmbNkQAPzu1ysrTtK8x9YvrzqCJEnSkLP4JEmSnrb3uP141QtOqTrGiHXDLy8v5bxtbW1sXrOao/ffp5Tzq2HRI6toa2urOoYkSbXjmk+SJEmSJEkqjcUnSZIkSZIklcZpd5KGlboteAwueixJkiRJ/bH4JGlY6ezs5PZ7b4e9q04yAFsa327/3e3V5hiIx6oOIEmSJKlVWHySNPzsDVuO21J1ihFt1I3OupYkSZI0NPzrQ5IkSZIkSaVx5NN2uO7M0HHdGUmSJEmSRi6LT9vR2dnJ7Xffx5Zx+1YdpWmxIQG49dcPV5ykeaPWP1p1BEmSJEmSVCKLT/3YMm5fnjz09VXHGNF2u+97VUeQJEmSJEklcs0nSZIkSZIklcbikyRJkiRJkkpj8UmSJEmSJEmlsfgkSZIkSZKk0lh8kiRJkiRJUmksPkmSJEmSJKk0Fp8kSZIkSZJUGotPkiRJkiRJKo3FJ0mSJEmSJJXG4pMkSZIkSZJKY/FJkiRJkiRJpSmt+BQRB0XEDRFxf0TcGxFnFv37RsTCiFhSfN+n1zFnR0RnRPxPRLymV/9REXF3cV97RETRv2tEfLPovyUippf1eiRJkiRJkjRwZY582gR8MDP/ADgGeF9EHAp8FLg+M2cC1xdtivtOAV4InAh8KSJGF+f6MnAaMLP4OrHoPxVYlZkzgM8Bny7x9UiSJEmSJGmAxpR14sxcBiwrbq+JiPuBA4E3AMcVD+sAbgT+tui/PDOfAn4TEZ3A0RGxFNgzM38GEBGXAm8E5hfHnFOc60rgixERmZllvS5J0va1t7fT2dlZdYwBWbJkCQDz5s2rOEnzZsyYUau8kiRJam2lFZ96K6bDvRi4Bdi/KEyRmcsiYr/iYQcCP+91WFfRt7G4vXV/zzEPFufaFBGrgYlA91bPfxqNkVNMmzZt0F6XJKmvzs5OfnnHHUypOsgA9AwBfuyOO6qM0bSHqw4gSZIkDVDpxaeImAB8G/hAZj5eLNe0zYduoy/76e/vmL4dmV8BvgIwa9YsR0VJUommAKdu88ezBsPFz/41J0mSJA1rpe52FxFjaRSevpGZVxXdj0TE1OL+qcDyor8LOKjX4W3AQ0V/2zb6+xwTEWOAvYBHB/+VSJIkSZIkaWeUudtdABcD92fmZ3vddTUwt7g9F/hur/5Tih3snkNjYfFFxRS9NRFxTHHOd291TM+53gL80PWeJEmSJEmSho8yp929DHgXcHdE3FH0/R3wKeCKiDgVeAB4K0Bm3hsRVwD30dgp732Zubk47nTg68DuNBYan1/0XwxcVixO/iiN3fIkSZIkSZI0TJS5291P2PaaTADHb+eY84Hzt9G/GHjRNvqfpCheSZIkSZIkafgpdc0nSZIkSZIktbbSd7uTqtDe3k5nZ2fVMZq2ZMkSAObNm1dxkoGZMWNG7TJLkiRJkoaWxSeNSJ2dnfzqntuYNmHzjh88DOyysTEI8cmlv6g4SfMeWDu66giSJEnSTuvq6uLxx4Orrtql6igjVveKYMNTXVXH0DBg8Ukj1rQJm/nYrLVVxxixzls8oeoIkiRJkqQasPgkSZIkSWo5bW1tLF+xnDe/eUPVUUasq67ahf0mt1UdQ8OAC45LkiRJkiSpNI58kiRJGgHWbNjEokdWVR2jaes3NdZlHDemPmsIrtmwqeoIGmbqtskNuNGNpGpYfJIkSaq5GTNmVB1hwHr+AD545syKkwxMHf+tVZ7Ozk7uueceJkyoz1qYGzduBGDp0qXVBhmAtWtdx1WqO4tPkiQJaOz6s3r9Gm745eVVRxmxHlu/nOx6YtDPW8fRAD2Z29vbK04i/X4mTJjAkUceWXWMEe22226rOoKk35NrPkmSJEmSJKk0jnySJElAY9efeGolr3rBKVVHGbFu+OXlHNg2seoYkiRJQ8qRT5IkSZIkSSqNxSdJkiRJkiSVxuKTJEmSJEmSSuOaT5IkSZIkqVba29vp7OysOkbTlixZAtRvh9oZM2YMSmaLT5IkSZIkqVY6Ozu5/b772Tx5/6qjNGVUMfFs8YpHK07SvNErHhm0c1l8kiRJkiRJtbN58v6s+9N3VR1jxBr/7csG7Vyu+SRJkiRJkqTSOPJJ0rDS1dUFq2HUjdbGS/UYdGVX1SkkSbhuyVAZrHVLJEkDZ/FJkiRJqlBnZyf/c8/9HLTHlKqjNGXspsYHROt/u6riJM17cM3DVUeQpJZm8Wk7urq6GLV+Nbvd972qo4xoo9avpKtrU9UxNIy0tbWxIlaw5bgtVUcZ0UbdOIq2A9uqjiFJKhy0xxQ+ePR7qo4xYl246JKqI0hSS3NeiyRJkiRJkkrjyKftaGtr45GnxvDkoa+vOsqIttt936OtrR5DzCVJkiRJ0sA58kmSJEmSJEmlsfgkSZIkSZKk0lh8kiRJkiRJUmksPkmSJEmSJKk0Fp8kSZIkSZJUGotPkiRJkiRJKo3FJ0mSJEmSJJVmTNUBJEkjR1dXF2uAi8mqo4xYy4C1XV1Vx5AkSZKaZvFJI1JXVxfr1ozmvMUTqo4yYv12zWjG+wewJEmSJGkHLD5JkgZNW1sbj3V3cypRdZQR62KSvdvaqo4hSZIkNa204lNE/BvwemB5Zr6o6NsX+CYwHVgKvC0zVxX3nQ2cCmwG5mXmdUX/UcDXgd2Ba4EzMzMjYlfgUuAoYCXw9sxcWtbrUb20tbXx5KZlfGzW2qqjjFjnLZ7Abv4BLEmSJEnagTIXHP86cOJWfR8Frs/MmcD1RZuIOBQ4BXhhccyXImJ0ccyXgdOAmcVXzzlPBVZl5gzgc8CnS3slkiRJkiRJ2imlFZ8y8ybg0a263wB0FLc7gDf26r88M5/KzN8AncDRETEV2DMzf5aZSWOk0xu3ca4rgeMjwnkekiRJkiRJw8hQr/m0f2YuA8jMZRGxX9F/IPDzXo/rKvo2Fre37u855sHiXJsiYjUwEeje+kkj4jQao6eYNm3aoL0YSZIkSa2rq6uLNWvWcNttt1UdZURbs2YNXW50I9VamdPuBmJbI5ayn/7+jnl2Z+ZXMnNWZs6aPHnyTkaUJEmSJEnSQA31yKdHImJqMeppKrC86O8CDur1uDbgoaK/bRv9vY/piogxwF48e5qfJEmSJJWira2NTZs2ceSRR1YdZUS77bbbaCtpo5vuFcFVV+1SyrnLsPqxxhiMvfbe5riLYad7RbCf4z/E0BefrgbmAp8qvn+3V/9/RMRngQNoLCy+KDM3R8SaiDgGuAV4N/CFrc71M+AtwA+LdaEkSdJOemz9cm745eVVx2jK2idXATBht30qTtK8x9Yv50AmVh1DkgTMmDGj6ggDtvqxJQDsN3lmxUmas9/k8v6du7q6GP34GsZ/+7JSzi8YveIRup5aPyjnKq34FBH/CRwHTIqILuATNIpOV0TEqcADwFsBMvPeiLgCuA/YBLwvMzcXpzqdxs55uwPziy+Ai4HLIqKTxoinU8p6LZIktYK6vQlfsqQx4PnA59WnmHMgE2v37yxJI9W8efOqjjBgPZnb29srTiINTGnFp8x8x3buOn47jz8fOH8b/YuBF22j/0mK4pUkSfr91e1NuG/ANVJ0dXWxbs0aLlx0SdVRRqwH1zzM+K51VceQNIja2tp4eMWjrPvTd1UdZcQa/+3LaJu876Cca7gsOC5JkiRJkqQRaKjXfJIkSZLUS1tbG+s3r+KDR7+n6igj1oWLLmFcW33Wh5OkkcaRT5IkSZIkSSqNxSdJkiRJkiSVxuKTJEmSJEmSSuOaT/0Ytf5Rdrvve1XHaFo8+TgAudueFSdp3qj1jwJTqo6h4eYxGHVjjWrja4vvEypNMTCPAQdWHUKSJElSK7D4tB0zZsyoOsKALVmyBoCZz6tTMWdKLf+tVZ46/vewZMkSAGYeOLPiJANwYD3/rSVJkiTVj8Wn7Zg3b17VEQasJ3N7e3vFSaSd5/97kiRJkjSyWHySJA2qh4GLyapjNG1l8X1ipSma9zCwd9UhJEmSpAGw+CRJGjR1nMq3opg2uffMekyb3Jt6/jtLkiSpdVl80oj1wNrRnLe4HitAP7K+sbj2/uO2VJykeQ+sHc3zqw6hYcdpk5IkSRoqo1c8wvhvX1Z1jKaMemwVAFv23qfiJM0bveIRmLzvoJzL4pNGpLqNCthQjLzYbXo9Rl4APJ/6/TtLkiRJGhnq9rfIkscaiz3MHKRizpCYvO+g/TtbfNKIVLfRF468kCRJqqe1a9dy2223VR2jaevXrwdg3LhxFSdp3tq1a6uOoGHIv/nqxeKTJEmSJO2Euo28AFhSjLifPn16tUEGqI7/1pKeYfFJkiRJqtiDax7mwkWXVB2jKcvXPwrAfuPqM3XkwTUPcwiDv85K3UZegKMvJFXD4pMkSZJUobqN6Ni4pBuAcQfXZ9HcQ9indv/OkjSSWHySJEmSKlS30TOOnJEkDdSoqgNIkiRJkiRp5LL4JEmSJEmSpNJYfJIkSZIkSVJpLD5JkiRJkiSpNBafJEmSJEmSVBqLT5IkSZIkSSqNxSdJkiRJkiSVxuKTJEmSJEmSSmPxSZIkSZIkSaWx+CRJkiRJkqTSWHySJEmSJElSaSw+SZIkSZIkqTQWnyRJkiRJklQai0+SJEmSJEkqjcUnSZIkSZIklSYys+oMv5eIOBH4PDAa+Fpmfqq/x8+aNSsXL148JNm2p729nc7OzkE/75IlSwCYOXPmoJ97xowZzJs3b9DPWzd1vHbg9etRx+vntWso69qB128o1PH/PfD69ajj9fPaNdTx2oHXr0cdr5/XrsH3LfVWx//3YHhcv4i4NTNnbeu+MUMdZjBFxGjg/wKzgS7gFxFxdWbeV22yauy+++5VR9BO8trVm9ev3rx+9eW1qzevX3157erN61dvXr/6avVrV+uRTxHxx8A5mfmaon02QGZesL1jhsPIJ0mSJEmSpJGkv5FPdV/z6UDgwV7trqJPkiRJkiRJw0Ddi0+xjb5nDeWKiNMiYnFELF6xYsUQxJIkSZIkSRLUv/jUBRzUq90GPLT1gzLzK5k5KzNnTZ48ecjCSZIkSZIktbq6F59+AcyMiOdExC7AKcDVFWeSJEmSJElSoda73WXmpoh4P3AdMBr4t8y8t+JYkiRJkiRJKtS6+ASQmdcC11adQ5IkSZIkSc9W92l3kiRJkiRJGsYsPkmSJEmSJKk0Fp8kSZIkSZJUGotPkiRJkiRJKk1kZtUZhlRErAB+W3WOEk0CuqsOoZ3itas3r1+9ef3qy2tXb16/+vLa1ZvXr968fvU10q/dwZk5eVt3tFzxaaSLiMWZOavqHBo4r129ef3qzetXX167evP61ZfXrt68fvXm9auvVr52TruTJEmSJElSaSw+SZIkSZIkqTQWn0aer1QdQDvNa1dvXr968/rVl9eu3rx+9eW1qzevX715/eqrZa+daz5JkiRJkiSpNI58kiRJkiRJUmksPkmSJEmSJKk0Fp8kSZIkSZJUGotPkiSpNiJiYkR8ISJui4hbI+LzETGx6lySJEnaPhccr7mIeGtmfmtHfRr+ImIf4KDMvKvqLNq+iNi3v/sz89GhyiK1oohYCNwE/HvR9U7guMx8dXWptDMiYk9gTE/bn5/1EBEvAg4Fduvpy8xLq0ukZkXEOOCDwLTMfG9EzAQOyczvVRxNGvEi4qXAdPr+3mupn50Wn2ouIm7LzCN31KfhKSJuBE6m8UPoDmAF8KPMPKvCWOpHRPwGSCCAacCq4vbewAOZ+Zzq0qlZxRvuC3j2H1DPrSyUmhIRt2bmUVv1Lc7MWVVl0sBExF8D5wJP0Ph5CpD+/zf8RcQngONo/Oy8FpgD/CQz31JlLjUnIr4J3Aq8OzNfFBG7Az/LzCOqTaYdiYgO4MzMfKxo7wNcmJl/WWkwNSUiLgOeR+Pvvc1Fd2bmvMpCVWDMjh+i4Sgi5gCvBQ6MiPZed+0JbKomlXbCXpn5eET8FXBJZn4iIhz5NIz1FJci4v8BV2fmtUV7DuDIi/q4BPgE8DngVcB7aBQRNfzdEBGnAFcU7bcA/11hHg3ch4AXZmZ31UE0YG8BDgduz8z3RMT+wNcqzqTmPS8z3x4R7wDIzCciwt999XBYT+EJIDNXRcSLK8yjgZkFHJotPvLHNZ/q6yFgMfAkjU8wer6uBl5TYS4NzJiImAq8DXDIc728pKfwBJCZ84FXVphHA7N7Zl5PYwTwbzPzHOBPKs6k5vw18B/ABuAp4HLgrIhYExGPV5pMzfo1sL7qENopT2TmFmBTMW1yOeCItfrYUIx2SoCIeB6Nn6Ma/kYVo52Ap5eBcCBJfdwDTKk6RNX8D7amMvNO4M6I+A8a13FaZv5PxbE0cJ8ErqMxZP0XEfFcYEnFmdSc7oj4GI11ZxL4c2BltZE0AE9GxChgSUS8H/gdsF/FmdSEzNyj6gz6vZ0N3BwRt9DrD99Wm35QU4sjYm/gqzQ+9FwLLKo0kQbiE8D3gYMi4hvAy4C/qDSRmnUhjZ+bVxbttwLnV5hHAzMJuC8iFtH3997J1UUaeq75VHMRcRLwL8AumfmciDgCOLfV/kOuo4gYDczLzM9VnUUDV3zi9AngWBrFp5to/L/ngrk1EBEvAe6nsVbXP9KYsvyZzLylylzasWKKyDuB52TmP0bEQcDUzPQP4Joo3nz/BLgb2NLTn5kdlYXSgEXEdGBPN0qpl2J30GNoTDX/udNf6yMiDqUxSjuA6zPzvoojqUkRsc3ZEZn5o6HOUiWLTzUXEbfS+CF0Y2a+uOi7KzMPqzaZmhERN2Tmq6rOoYEpCocdmfnnVWfRznGn0PqKiC/TKFj8SWb+QTENYUFmvqTiaGpSRNycmS+tOoeaFxH9bmSTmbcNVRbtvO1cx9XAbzPTNWOHuZ6dsem7W5r/76k2nHZXf5syc7VrBdbWzRHxReCbwLqeTn+RDG+ZuTkiJkfELpm5oeo82ilnA1sXmrbVp+HnjzLzyIi4HZ5edHWXqkNpQG6IiNOAa+g7/cCRo8PXhcX33WgsnHsnjdEXhwG3AC+vKJcG5kvAkcBdNK7fi4rbEyPi/8vMBVWG0/ZFxD/SmCL5a3rtEorrVdZCRBwDfAH4A2AXYDSwLjP3rDTYELP4VH/3RMSfAaOLrcPnATdXnEnN6/nk99xeff4iqYelwE8j4mr6Fg4/W1ki7ZA7hY4IG4vRhz0L5k6m19Qt1cKfFd/P7tWXuHD1sNUzSjsiLgdOy8y7i/aLaOxeqHpYCpyamffC09O4Pkxj+vlVgMWn4ettNHYr9EPPevoicAqNDzlnAe8GZlaaqAIWn+rvDODvaXxy+J80Fq/+x0oTqWlOuau1h4qvUYALINdHz06hJ9NYLLfHGuBvKkmkgWoHvgPsHxHn09j6/WPVRtJAZOZzqs6gnfaCnsITQGbeU6w3qnp4QU/hCSAz74uIF2fm/zqLYti7h8Y6lcsrzqGdlJmdETE6MzcDl0REyw0Ycc0nqUIRsT/wT8ABmTmn+ATqjzPz4oqjqUkRsQeQmbm26ixqXkSMpTHl4PlF1/9k5sYKI2kAIuIFwPE8s+jq/RVH0gAVI2YOpTGNC4DMvLS6RGpGRPwnjdG+vXd6nZCZ76g0mJoSEd8EHgUuL7reTmMXrnfR2HnZtfOGqYiYBXyXRhGqZXdLq6uIuAl4NfA14GFgGfAXmXl4pcGGmMWnmoqIa3hmvu+z+IOoHiJiPnAJ8PeZeXhEjAFuz8w/rDiadqD4w+kyYN+iqxt4d+9PFDV8FbuOXEpjCkLQWMBzbmbeVGUuNSciXg7MzMxLiml3EzLzN1XnUnMi4hPAcTSKT9cCc2j84fuWKnNpxyJiN+B0Gju9QmOn1y9n5pPVpVKzImJ34P/QWKMraOw6+SXgSWCcH6QNXxFxL3ARz94ltKV2S6uriDgYeITGek9/A+wFfCkzOysNNsQsPtXU9rZr7OEPonqIiF9k5ksi4vZeuxXekZlHVBxNO1AMlf37zLyhaB8H/JM7ONVDsVPon2Xm/xTt5wP/mZlHVZtMO1IULmYBh2Tm8yPiAOBbmfmyiqOpSRFxN3A4jQ9bDi9GAX8tM0+qOJokDUsR8aPM7PfvP9VDz66FmXlX1VmGmms+1VTv4lKxy49TR+ppXURM5JmFc4+hseWthr/xPYUngMy8MSLGVxlIAzK2p/AEkJm/Kqbiafh7E/Bi4DaAzHyomP6q+ngyM7dExKaI2JPGGiYuNl4DEfEy4BzgYPpu9+71G8aKgm9/MyYOG8I42jm3RsQFwNX0nXbnDtk1EBE30lhvdAxwB7CiKCieVWWuoWbxqeaK0RYd9Jo6EhFOHamPs2j8EnleRPwUmExj8VwNf/8bEf9AY+odNNa9cNpPfSyOiIt55vq9k74LkGv42pCZGRE9RXuLvjUSjVWN74qIvYGv0vj/bi2wqMpcatrFNKaM3ApsrjiLmvf64vv7iu+9f/etH/o42gkvLr7/UfE9cIfsOtkrMx+PiL8CLsnMT0REy418ctpdzTl1pP6KdZ4OofFLxJFrw1xEXJaZ74qIs4DpPLNuwo+AT2bmqirzqTkRsSuNN+E91+8mGnPvn+r3QFUuIj5EY3vi2cAFwKnAf2Rme6XB1LSIuLXnfUpETAf2bMXpB3UUEbdk5h/t+JEajiLip1tPUd5Wn4afiPggjWJTz7aECTwOLM7MO6rKpeYUow9PoDFo5O8z8xcRcVerjTp05FP9OXWk/o6mUcQYAxwZEe74M7wdVSwaOBd4Fc988gTPvCHQ8Dcb+NfM/GzVQTQwmfkvETGbxpvu5wMfy8wfVBxLA/PziHhJZv4iM5dWHUYDckNE/DNwFU79qaPxEfHyzPwJQES8FHD0aD0cRWO9w6tpvN98HfAL4K8j4luZ+Zkqw2mHzgWuo7G5xi8i4rnAkoozDTlHPtVcRFxCY8eD3sNnx2Tme6pLpWZFxGXA82jM/e0Zvp6ZOa+yUOpXRMyjsdPPc4Hf9b6LxrVz3YsaiIh/B/4Y+DaN4c/3VxxJOxARP8nMl0fEGvp++guN34OPAv+cmV+qJKCaFhH30Sgc/hZYxzM/P1vqE+A6iogbttGdmenUnxqIiKOAf6Ox0xbAY8BfWjwc/iLiOuBPe3YkjIgJwJU01kG8NTMPrTKf1AyLTzXn1JF6i4j7gUPT/xFrJyK+nJmnV51DO69Y6PgdwHtoFDMuoTFteU2lwbRTis0bbs7MQ6rOov4Vo0efJTN/O9RZpFZU/P6LzHSTm5oo/mY4PDM3FO1dgTsy8w9675qt4SkiPgOcBzwBfJ/Gjq8fyMx/rzTYELP4VGMRMQq4KzNfVHUW7ZyI+BYwLzOXVZ1FakURMYnGYvEfAO4HZgDtmfmFKnNp50TEVH+eSuWJiL2ATwDHFl0/As61iFEPXr/6Kja5eRPw3aLrJBpT8C4EvpKZ76wqm3YsIu7IzCMi4k3AG2ls3HBDZh5ebbKhZfGp5iLiG8DZmflA1VnUvIi4hsZIiz2AI2js8tN77YSTq0kmtYaIOAn4SxrTXi8DOjJzeUSMA+7PzG2OzJCkVhYR3wbuobFoLsC7aIzGeHN1qdQsr1+9FdMme2a7/CQzF1ccSU2KiHsz84UR8VXg25n5/Yi40+KTaiUifgi8hEbxYl1Pv8WL4S0iXtnf/Zn5o6HKIrWiiLgU+Fpm3rSN+47PzOsriCVJw1rPp/c76tPw5PWTqhERn6Ix4ukJGptN7Q18r9V2D3W3u/r7ZNUBNHA9xaWI+HRm/m3v+yLi0zSGQUsqSWa+OyL2j4jXF12LMnN5cZ+FJ0natie22i3tZTT+mFI9eP2kCmTmR4u/8R7PzM0RsQ54Q9W5hpojn2ouIv4G+FZmdlWdRQMXEbdl5pFb9d3ljj9SuSLircC/ADfSGL7+CuDDmXlllbkkaTiLiCNoTNnq2S1tFTA3M++qLJSaFhGHA5fi9ZOGRET8SWb+MCK2ObU1M68a6kxVcuRT/e0JXBcRjwKXA1dm5iMVZ9IORMTpwP8BnhcRvX/h7wHcXE0qqaV8DHhJz2iniJgM/IDGtsWSpG27H/gMjfXy9gZW05hKYvFiGIuIs3o1LwXGF7fXAa/G6yeV5ZXAD2ksEA+NNX+h8cFnAi1VfHLk0wgREYcBbwf+FOjKzFdXHEn9KHYb2Qe4APgUz+w68pPMvL2yYFKLiIi7M/MPe7VHAXf27pMk9RUR3wceA24DNvf0Z+aFVWXSjkXEJ4qbh9BYK/a7NP74PQm4KTP/qqpsUiuIiN1o/J0+nWcGAGVmnltZqAo48mnkWA48DKwE9qs4i3ag2NJ2dUT8HPh3GlXvADoi4qtu8y6Vbn5EXAf8Z9F+O3BthXkkqQ7aMvPEqkNoYDLzkwARsQA4MjPXFO1zgG9VGE1qFf/FM4X7J4u+lhsFZPGp5orpW28HJtOYLvLezLyv2lQagFOBYzJzHTy92PjPAItPUrkSuIhntiz+CnBMpYkkafi7OSL+MDPvrjqIdso0YEOv9gYaIzEklcvCPRafRoKDgQ9k5h1VB9FOCXoNWy9uR0VZpFYyu9hp8um59hHxSeBvt3+IJLWmiLibRtF+DPCeiPhf4CmKdUvcKKU2LgMWRcR3aFzPN9FYQF5SuSzc45pPI0JEvByYmZmXFIvmTsjM31SdSztWLAA5F/hO0fVG4OuZ+a9VZZJGsl6L/T8X+HWvu/YAfpqZf15JMEkaxiLi4P7uz8zfDlUW/X4i4kgaO7xCY70n1xqVShYR9wEzgN/QwoV7i081VywgOAs4JDOfHxEHAN/KzJdVHE1NKt4E9Ez98U2AVKKtFvv/aK+71mTmo9WkkiRJ0ki1vQJ+qxXuLT7VXETcAbwYuC0zX1z03dVqVVRJkiRJkjQ8jao6gH5vG7JRQUyAiBhfcR5JkiRJkqSnWXyqvysi4iJg74h4L/AD4KsVZ5IkSZIkSQKcdjciRMRs4AQaawZdl5kLK44kSZIkSZIEWHySJEmSJElSiZx2V3MR8eaIWBIRqyPi8YhYExGPV51LkiRJkiQJHPlUexHRCZyUmfdXnUWSJEmSJGlrjnyqv0csPEmSJEmSpOHKkU81FRFvLm6+EpgC/BfwVM/9mXlVBbEkSZIkSZL6sPhUUxFxSXEzaexy11tm5l8OcSRJkiRJkqRnGVN1AO2czHwPQER0AGdm5mNFex/gwgqjSZIkSZIkPc01n+rvsJ7CE0BmrgJeXF0cSZIkSZKkZ1h8qr9RxWgnACJiXxzRJkmSJEmShgmLFPV3IXBzRFxJY/2ntwHnVxtJkiRJkiSpwQXHR4CIOBT4ExoLj1+fmfdVHEmSJEmSJAmw+CRJkiRJkqQSueaTJEmSJEmSSmPxSZIkSZIkSaWx+CRJkjQMRMTkiLglIm6PiFcM8NgjIuK1ZWWTJEn6fVh8kiRJGh6OB36ZmS/OzB8P8NgjgAEVn6Jhh+8Fm32cJEnS9vhGQpIkaRsiYnpE/DIivhYR90TENyLi1RHx04hYEhFHF183F6OVbo6IQ4pj/yIiroqI7xeP/Uyv867tdfstEfH1iDgC+Azw2oi4IyJ2j4gvR8TiiLg3Ij7Z65iXFM91Z0Qsioi9gHOBtxfHvj0izomID/U65p7i9UyPiPsj4kvAbcBBEfHhiPhFRNzV8zzbelyp/9iSJGlEs/gkSZK0fTOAzwOHAS8A/gx4OfAh4O+AXwLHZuaLgY8D/9Tr2COAtwN/SKMwtN0CTmbeURz/zcw8IjOfAP4+M2cVz/3KiDgsInYBvgmcmZmHA68G1m117Dd38JoOAS4tMh8CzASOLvIeFRHHbv24zPztDs4pSZK0XWOqDiBJkjSM/SYz7waIiHuB6zMzI+JuYDqwF9ARETOBBMb2Ovb6zFxdHHsfcDDw4ACe+20RcRqN92tTgUOL51iWmb8AyMzHi/MP5DX9NjN/Xtw+ofi6vWhPoFGMemCrx0mSJO00i0+SJEnb91Sv21t6tbfQeB/1j8ANmfmmiJgO3LidYzfzzPuu7NW/27aeNCKeQ2N01Usyc1VEfL14bGx1/PZsou8I997Ps673UwEXZOZFWz3/9K0eJ0mStNOcdidJkrTz9gJ+V9z+iyaPeSQi/qBYxPtN23nMnjSKP6sjYn9gTtH/S+CAiHgJQETsERFjgDXAHr2OXwocWTzmSOA523me64C/jIgJxWMPjIj9mnwdkiRJTbH4JEmStPM+A1wQET8FRjd5zEeB7wE/BJZt6wGZeSeNqXD3Av8G/LTo30BjHakvRMSdwEIao5puAA7tWXAc+Dawb0TcAZwO/Go7z7MA+A/gZ8VUwivpW8SSJEn6vUVmMyO3JUmSJEmSpIFz5JMkSZIkSZJKY/FJkiRJkiRJpbH4JEmSJEmSpNJYfJIkSZIkSVJpLD5JkiRJkiSpNBafJEmSJEmSVBqLT5IkSZIkSSrN/w+XXJ4sN+S3uQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20,5))\n",
    "sns.boxplot(x='manufacturer', y='price', data=df)\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "평균에는 차이가 존재하나, 값의 범위가 조금 비슷함"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20,5))\n",
    "sns.boxplot(x='model', y='price', data=df)\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NgOuPSTKSJz0"
   },
   "source": [
    "### 피쳐 간 corr Heatmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 286
    },
    "executionInfo": {
     "elapsed": 554,
     "status": "ok",
     "timestamp": 1631959314217,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "wVmEa1ChlrTc",
    "outputId": "c8d893ff-1042-4968-b439-0f1b57747182"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr_columns = ['price', 'odometer', 'age']\n",
    "corr = df[corr_columns].corr()\n",
    "mask = np.triu(np.ones_like(corr))\n",
    "\n",
    "sns.heatmap(corr, annot=True, cmap='coolwarm', vmax=1, vmin=-1, mask=mask)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FRfd7ABjepBS"
   },
   "source": [
    "## 4. 모델 학습을 위한 데이터 전처리 "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Dev2yLeMta85"
   },
   "source": [
    "### StandardScaler를 이용해 수치형 데이터 표준화\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "executionInfo": {
     "elapsed": 5,
     "status": "ok",
     "timestamp": 1631959314218,
     "user": {
      "displayName": "김별희/산업공학전공",
      "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s64",
      "userId": "07357944114157865332"
     },
     "user_tz": -540
    },
    "id": "_k_SDCh5xMgD"
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "region           object\n",
       "price             int64\n",
       "manufacturer     object\n",
       "model            object\n",
       "condition        object\n",
       "cylinders        object\n",
       "fuel             object\n",
       "odometer        float64\n",
       "transmission     object\n",
       "drive            object\n",
       "size             object\n",
       "type             object\n",
       "paint_color      object\n",
       "state            object\n",
       "age             float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "id": "W3EO22NCE3wG"
   },
   "outputs": [],
   "source": [
    "# StandardScaler를 이용해 수치형 데이터를 표준화\n",
    "X_num = df[['odometer', 'age']]\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(X_num)\n",
    "X_scaled = pd.DataFrame(X_scaled, index=X_num.index, columns=X_num.columns)\n",
    "\n",
    "# get_dummies를 이용해 범주형 데이터를 one-hot 벡터로 변환 \n",
    "X_cat = df.drop(['price','odometer','age'], axis=1)\n",
    "X_cat = pd.get_dummies(X_cat)\n",
    "\n",
    "# 입출력 데이터 통합하기\n",
    "X = pd.concat([X_scaled, X_cat], axis=1)\n",
    "y = df['price']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>odometer</th>\n",
       "      <th>age</th>\n",
       "      <th>region_columbus</th>\n",
       "      <th>region_fayetteville</th>\n",
       "      <th>region_jacksonville</th>\n",
       "      <th>region_others</th>\n",
       "      <th>region_rochester</th>\n",
       "      <th>region_springfield</th>\n",
       "      <th>manufacturer_chevrolet</th>\n",
       "      <th>manufacturer_dodge</th>\n",
       "      <th>...</th>\n",
       "      <th>state_sd</th>\n",
       "      <th>state_tn</th>\n",
       "      <th>state_tx</th>\n",
       "      <th>state_ut</th>\n",
       "      <th>state_va</th>\n",
       "      <th>state_vt</th>\n",
       "      <th>state_wa</th>\n",
       "      <th>state_wi</th>\n",
       "      <th>state_wv</th>\n",
       "      <th>state_wy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.282480</td>\n",
       "      <td>0.110267</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.176965</td>\n",
       "      <td>-0.455374</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.296022</td>\n",
       "      <td>0.675908</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.574188</td>\n",
       "      <td>5.201037</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.229942</td>\n",
       "      <td>0.817318</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 112 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   odometer       age  region_columbus  region_fayetteville  \\\n",
       "0 -1.282480  0.110267                0                    0   \n",
       "1 -0.176965 -0.455374                0                    0   \n",
       "2 -0.296022  0.675908                0                    0   \n",
       "3  1.574188  5.201037                0                    0   \n",
       "4  0.229942  0.817318                0                    0   \n",
       "\n",
       "   region_jacksonville  region_others  region_rochester  region_springfield  \\\n",
       "0                    0              1                 0                   0   \n",
       "1                    0              1                 0                   0   \n",
       "2                    0              1                 0                   0   \n",
       "3                    0              1                 0                   0   \n",
       "4                    0              1                 0                   0   \n",
       "\n",
       "   manufacturer_chevrolet  manufacturer_dodge  ...  state_sd  state_tn  \\\n",
       "0                       1                   0  ...         0         0   \n",
       "1                       0                   0  ...         0         0   \n",
       "2                       0                   0  ...         0         0   \n",
       "3                       1                   0  ...         0         0   \n",
       "4                       0                   0  ...         0         0   \n",
       "\n",
       "   state_tx  state_ut  state_va  state_vt  state_wa  state_wi  state_wv  \\\n",
       "0         0         0         0         0         0         0         0   \n",
       "1         0         0         0         0         0         0         0   \n",
       "2         0         0         0         0         0         0         0   \n",
       "3         0         0         0         0         0         0         0   \n",
       "4         0         0         0         0         0         0         0   \n",
       "\n",
       "   state_wy  \n",
       "0         0  \n",
       "1         0  \n",
       "2         0  \n",
       "3         0  \n",
       "4         0  \n",
       "\n",
       "[5 rows x 112 columns]"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "odometer                 0\n",
       "age                    500\n",
       "region_columbus          0\n",
       "region_fayetteville      0\n",
       "region_jacksonville      0\n",
       "                      ... \n",
       "state_vt                 0\n",
       "state_wa                 0\n",
       "state_wi                 0\n",
       "state_wv                 0\n",
       "state_wy                 0\n",
       "Length: 112, dtype: int64"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "age가 500개 na값임"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1.6971181595987114e-15"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X['age'].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "age 평균값 0에 가까움"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "id": "B6hPWyzPTg8a"
   },
   "outputs": [],
   "source": [
    "X.fillna(0.0, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "x93Tb6lptcA2"
   },
   "source": [
    "### 학습 데이터, 테스트 데이터 분리\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "id": "C9kMQI8SEche"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "id": "F07QjOFwFNEw"
   },
   "outputs": [],
   "source": [
    "# train_test_split() 함수로 학습 데이터와 테스트 데이터 분리하기\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RrWrE8Z4exup"
   },
   "source": [
    "##  5.Regression 모델 학습"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NypEtNhdfCow"
   },
   "source": [
    "###  XGBoost Regression 모델"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "id": "ow28ZRL4F7D5"
   },
   "outputs": [],
   "source": [
    "from xgboost import XGBRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "id": "LSSNqFUrGM6R"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "             importance_type='gain', interaction_constraints='',\n",
       "             learning_rate=0.300000012, max_delta_step=0, max_depth=6,\n",
       "             min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "             n_estimators=100, n_jobs=8, num_parallel_tree=1, random_state=0,\n",
       "             reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "             tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# XGBRegressor 모델 생성/학습\n",
    "xgb_reg = XGBRegressor()\n",
    "xgb_reg.fit(X_train, y_train) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gUo8NmHkfIpf"
   },
   "source": [
    "### 모델 학습 결과 평가"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "id": "rxvf1vgmfLs-"
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import mean_absolute_error, mean_squared_error\n",
    "from math import sqrt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "id": "KDVy7fFGfUP1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3073.144876308058\n",
      "4700.191209868938\n"
     ]
    }
   ],
   "source": [
    "pred = xgb_reg.predict(X_test)\n",
    "\n",
    "# MAE\n",
    "print(mean_absolute_error(y_test, pred))\n",
    "\n",
    "# MSE\n",
    "print(sqrt(mean_squared_error(y_test, pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "오차가 크게 나타남"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DTqb-HqPtc4I"
   },
   "source": [
    "## 6. 모델 학습 결과 분석\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "id": "kKEP06-OmrBs"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1bfca3e4c70>]"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 실제 값과 예측 값 간의 시각화\n",
    "sns.histplot(x=y_test, y=pred)\n",
    "plt.plot([0,60000], [0,60000], 'r-')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "id": "WLnyYNJwGRgd"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 오차율 시각화\n",
    "\n",
    "err = (pred - y_test) / y_test * 100\n",
    "sns.histplot(err)\n",
    "plt.xlabel('error (%)')\n",
    "plt.xlim(-100, 100)\n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "id": "Qb02ebbDV__k"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 오차 시각화\n",
    "\n",
    "err = (pred - y_test)\n",
    "sns.histplot(err)\n",
    "plt.xlabel('error ($)')\n",
    "plt.grid()"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "2021-09-18_중고 자동차 데이터 Regression.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}