16. Data_preprocessing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0f39a8d9",
   "metadata": {},
   "source": [
    "# Loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "03f4b6c8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "👽 Hello.\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "print(\"👽 Hello.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "388c9775",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "      <th>기타사항</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>중국</td>\n",
       "      <td>116932.0</td>\n",
       "      <td>12083947.0</td>\n",
       "      <td>334522.0</td>\n",
       "      <td>8143271.0</td>\n",
       "      <td>3940676.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>미국</td>\n",
       "      <td>65888.0</td>\n",
       "      <td>5561545.0</td>\n",
       "      <td>509564.0</td>\n",
       "      <td>3625062.0</td>\n",
       "      <td>1936484.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>일본</td>\n",
       "      <td>54017.0</td>\n",
       "      <td>2251307.0</td>\n",
       "      <td>82480.0</td>\n",
       "      <td>3827247.0</td>\n",
       "      <td>-1575940.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>86228.0</td>\n",
       "      <td>9927642.0</td>\n",
       "      <td>209100.0</td>\n",
       "      <td>6980874.0</td>\n",
       "      <td>2946768.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>미국</td>\n",
       "      <td>60225.0</td>\n",
       "      <td>5021264.0</td>\n",
       "      <td>428678.0</td>\n",
       "      <td>2998216.0</td>\n",
       "      <td>2023048.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          기간 국가명      수출건수        수출금액      수입건수       수입금액       무역수지  기타사항\n",
       "0  2015년 01월  중국  116932.0  12083947.0  334522.0  8143271.0  3940676.0   NaN\n",
       "1  2015년 01월  미국   65888.0   5561545.0  509564.0  3625062.0  1936484.0   NaN\n",
       "2  2015년 01월  일본   54017.0   2251307.0   82480.0  3827247.0 -1575940.0   NaN\n",
       "3  2015년 02월  중국   86228.0   9927642.0  209100.0  6980874.0  2946768.0   NaN\n",
       "4  2015년 02월  미국   60225.0   5021264.0  428678.0  2998216.0  2023048.0   NaN"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "csv_file_path = os.getenv('HOME')+'/desktop/Aiffel/Becoming_Kaggler/trade.csv'\n",
    "trade = pd.read_csv(csv_file_path) \n",
    "trade.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60504f20",
   "metadata": {},
   "source": [
    "## 16-2. 결측치(Missing Data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1f42745f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "전체 데이터 건수: 199\n"
     ]
    }
   ],
   "source": [
    "print('전체 데이터 건수:', len(trade))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e9913eed",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "컬럼별 결측치 개수\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "기간        0\n",
       "국가명       0\n",
       "수출건수      3\n",
       "수출금액      4\n",
       "수입건수      3\n",
       "수입금액      3\n",
       "무역수지      4\n",
       "기타사항    199\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('컬럼별 결측치 개수')\n",
    "len(trade) - trade.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b28e83ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "기간      199\n",
       "국가명     199\n",
       "수출건수    196\n",
       "수출금액    195\n",
       "수입건수    196\n",
       "수입금액    196\n",
       "무역수지    195\n",
       "기타사항      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9863de70",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<bound method DataFrame.count of             기간 국가명      수출건수        수출금액       수입건수       수입금액       무역수지  \\\n",
       "0    2015년 01월  중국  116932.0  12083947.0   334522.0  8143271.0  3940676.0   \n",
       "1    2015년 01월  미국   65888.0   5561545.0   509564.0  3625062.0  1936484.0   \n",
       "2    2015년 01월  일본   54017.0   2251307.0    82480.0  3827247.0 -1575940.0   \n",
       "3    2015년 02월  중국   86228.0   9927642.0   209100.0  6980874.0  2946768.0   \n",
       "4    2015년 02월  미국   60225.0   5021264.0   428678.0  2998216.0  2023048.0   \n",
       "..         ...  ..       ...         ...        ...        ...        ...   \n",
       "194  2020년 05월  미국  126598.0   4600726.0  1157163.0  4286873.0   313853.0   \n",
       "195  2020년 05월  일본  166568.0   1798128.0   133763.0  3102734.0 -1304606.0   \n",
       "196  2020년 06월  중국       NaN         NaN        NaN        NaN        NaN   \n",
       "197  2020년 06월  미국       NaN         NaN        NaN        NaN        NaN   \n",
       "198  2020년 06월  일본       NaN         NaN        NaN        NaN        NaN   \n",
       "\n",
       "     기타사항  \n",
       "0     NaN  \n",
       "1     NaN  \n",
       "2     NaN  \n",
       "3     NaN  \n",
       "4     NaN  \n",
       "..    ...  \n",
       "194   NaN  \n",
       "195   NaN  \n",
       "196   NaN  \n",
       "197   NaN  \n",
       "198   NaN  \n",
       "\n",
       "[199 rows x 8 columns]>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "058c0d9b",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>중국</td>\n",
       "      <td>116932.0</td>\n",
       "      <td>12083947.0</td>\n",
       "      <td>334522.0</td>\n",
       "      <td>8143271.0</td>\n",
       "      <td>3940676.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>미국</td>\n",
       "      <td>65888.0</td>\n",
       "      <td>5561545.0</td>\n",
       "      <td>509564.0</td>\n",
       "      <td>3625062.0</td>\n",
       "      <td>1936484.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>일본</td>\n",
       "      <td>54017.0</td>\n",
       "      <td>2251307.0</td>\n",
       "      <td>82480.0</td>\n",
       "      <td>3827247.0</td>\n",
       "      <td>-1575940.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>86228.0</td>\n",
       "      <td>9927642.0</td>\n",
       "      <td>209100.0</td>\n",
       "      <td>6980874.0</td>\n",
       "      <td>2946768.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>미국</td>\n",
       "      <td>60225.0</td>\n",
       "      <td>5021264.0</td>\n",
       "      <td>428678.0</td>\n",
       "      <td>2998216.0</td>\n",
       "      <td>2023048.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>미국</td>\n",
       "      <td>126598.0</td>\n",
       "      <td>4600726.0</td>\n",
       "      <td>1157163.0</td>\n",
       "      <td>4286873.0</td>\n",
       "      <td>313853.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>일본</td>\n",
       "      <td>166568.0</td>\n",
       "      <td>1798128.0</td>\n",
       "      <td>133763.0</td>\n",
       "      <td>3102734.0</td>\n",
       "      <td>-1304606.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>2020년 06월</td>\n",
       "      <td>중국</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>2020년 06월</td>\n",
       "      <td>미국</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>2020년 06월</td>\n",
       "      <td>일본</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>199 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액       수입건수       수입금액       무역수지\n",
       "0    2015년 01월  중국  116932.0  12083947.0   334522.0  8143271.0  3940676.0\n",
       "1    2015년 01월  미국   65888.0   5561545.0   509564.0  3625062.0  1936484.0\n",
       "2    2015년 01월  일본   54017.0   2251307.0    82480.0  3827247.0 -1575940.0\n",
       "3    2015년 02월  중국   86228.0   9927642.0   209100.0  6980874.0  2946768.0\n",
       "4    2015년 02월  미국   60225.0   5021264.0   428678.0  2998216.0  2023048.0\n",
       "..         ...  ..       ...         ...        ...        ...        ...\n",
       "194  2020년 05월  미국  126598.0   4600726.0  1157163.0  4286873.0   313853.0\n",
       "195  2020년 05월  일본  166568.0   1798128.0   133763.0  3102734.0 -1304606.0\n",
       "196  2020년 06월  중국       NaN         NaN        NaN        NaN        NaN\n",
       "197  2020년 06월  미국       NaN         NaN        NaN        NaN        NaN\n",
       "198  2020년 06월  일본       NaN         NaN        NaN        NaN        NaN\n",
       "\n",
       "[199 rows x 7 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade = trade.drop(\"기타사항\", axis=1)\n",
    "trade"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bc55aced",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>199 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        기간    국가명   수출건수   수출금액   수입건수   수입금액   무역수지\n",
       "0    False  False  False  False  False  False  False\n",
       "1    False  False  False  False  False  False  False\n",
       "2    False  False  False  False  False  False  False\n",
       "3    False  False  False  False  False  False  False\n",
       "4    False  False  False  False  False  False  False\n",
       "..     ...    ...    ...    ...    ...    ...    ...\n",
       "194  False  False  False  False  False  False  False\n",
       "195  False  False  False  False  False  False  False\n",
       "196  False  False   True   True   True   True   True\n",
       "197  False  False   True   True   True   True   True\n",
       "198  False  False   True   True   True   True   True\n",
       "\n",
       "[199 rows x 7 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# isnull(), any(axis=1)을 이용해서 missing data 가 있는 행 찾기\n",
    "\n",
    "trade.isnull()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "944ff806",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      False\n",
       "1      False\n",
       "2      False\n",
       "3      False\n",
       "4      False\n",
       "       ...  \n",
       "194    False\n",
       "195    False\n",
       "196     True\n",
       "197     True\n",
       "198     True\n",
       "Length: 199, dtype: bool"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.isnull().any(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4284f0f3",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>2020년 06월</td>\n",
       "      <td>중국</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>2020년 06월</td>\n",
       "      <td>미국</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>2020년 06월</td>\n",
       "      <td>일본</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수  수출금액       수입건수       수입금액  무역수지\n",
       "191  2020년 04월  미국  105360.0   NaN  1141118.0  5038739.0   NaN\n",
       "196  2020년 06월  중국       NaN   NaN        NaN        NaN   NaN\n",
       "197  2020년 06월  미국       NaN   NaN        NaN        NaN   NaN\n",
       "198  2020년 06월  일본       NaN   NaN        NaN        NaN   NaN"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade[trade.isnull().any(axis=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "590d30b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "trade.dropna(how='all',axis=0, subset=['수출건수','수출금액','수입건수','수입금액','무역수지'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1fdaca7c",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>일본</td>\n",
       "      <td>134118.0</td>\n",
       "      <td>1989323.0</td>\n",
       "      <td>141207.0</td>\n",
       "      <td>3989562.0</td>\n",
       "      <td>-2000239.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>중국</td>\n",
       "      <td>185320.0</td>\n",
       "      <td>10746069.0</td>\n",
       "      <td>349007.0</td>\n",
       "      <td>8989920.0</td>\n",
       "      <td>1756149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>미국</td>\n",
       "      <td>126598.0</td>\n",
       "      <td>4600726.0</td>\n",
       "      <td>1157163.0</td>\n",
       "      <td>4286873.0</td>\n",
       "      <td>313853.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>일본</td>\n",
       "      <td>166568.0</td>\n",
       "      <td>1798128.0</td>\n",
       "      <td>133763.0</td>\n",
       "      <td>3102734.0</td>\n",
       "      <td>-1304606.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액       수입건수       수입금액       무역수지\n",
       "191  2020년 04월  미국  105360.0         NaN  1141118.0  5038739.0        NaN\n",
       "192  2020년 04월  일본  134118.0   1989323.0   141207.0  3989562.0 -2000239.0\n",
       "193  2020년 05월  중국  185320.0  10746069.0   349007.0  8989920.0  1756149.0\n",
       "194  2020년 05월  미국  126598.0   4600726.0  1157163.0  4286873.0   313853.0\n",
       "195  2020년 05월  일본  166568.0   1798128.0   133763.0  3102734.0 -1304606.0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c1359312",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수  수출금액       수입건수       수입금액  무역수지\n",
       "191  2020년 04월  미국  105360.0   NaN  1141118.0  5038739.0   NaN"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade[trade.isnull().any(axis=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "bb0b767a",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>188</th>\n",
       "      <td>2020년 03월</td>\n",
       "      <td>미국</td>\n",
       "      <td>97117.0</td>\n",
       "      <td>7292838.0</td>\n",
       "      <td>1368345.0</td>\n",
       "      <td>5388338.0</td>\n",
       "      <td>1904500.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>미국</td>\n",
       "      <td>126598.0</td>\n",
       "      <td>4600726.0</td>\n",
       "      <td>1157163.0</td>\n",
       "      <td>4286873.0</td>\n",
       "      <td>313853.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수       수출금액       수입건수       수입금액       무역수지\n",
       "188  2020년 03월  미국   97117.0  7292838.0  1368345.0  5388338.0  1904500.0\n",
       "191  2020년 04월  미국  105360.0        NaN  1141118.0  5038739.0        NaN\n",
       "194  2020년 05월  미국  126598.0  4600726.0  1157163.0  4286873.0   313853.0"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 앞, 뒤 데이터의 평균으로 missing value 바꿔주기\n",
    "\n",
    "trade.loc[[188, 191, 194]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "b7b51d83",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>5946782.0</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수       수출금액       수입건수       수입금액  무역수지\n",
       "191  2020년 04월  미국  105360.0  5946782.0  1141118.0  5038739.0   NaN"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.loc[191, '수출금액'] = (trade.loc[188, '수출금액'] + trade.loc[194, '수출금액'] )/2\n",
    "trade.loc[[191]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "4988bd7b",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>5946782.0</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>908043.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수       수출금액       수입건수       수입금액      무역수지\n",
       "191  2020년 04월  미국  105360.0  5946782.0  1141118.0  5038739.0  908043.0"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.loc[191,'무역수지'] = trade.loc[191,'수출금액'] - trade.loc[191,'수입금액']\n",
    "trade.loc[[191]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "a9da9556",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5870371.685634511"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 다른 방법으로 채워보기(2020년 3월, 2020년 5월 데이터의 수출당 금액을 구하고 평균값을 구하기)\n",
    "\n",
    "x1 = trade.loc[188,'수출금액'] / trade.loc[188,'수출건수']\n",
    "x3 = trade.loc[194,'수출금액'] / trade.loc[194,'수출건수']\n",
    "x2 = (x1+x3) / 2\n",
    "\n",
    "y2 = trade.loc[191,'수출건수'] * x2\n",
    "\n",
    "y2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b900922a",
   "metadata": {},
   "source": [
    "## 16-3. 중복된 데이터"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "af42b855",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      False\n",
       "1      False\n",
       "2      False\n",
       "3      False\n",
       "4      False\n",
       "       ...  \n",
       "191    False\n",
       "192    False\n",
       "193    False\n",
       "194    False\n",
       "195    False\n",
       "Length: 196, dtype: bool"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# DataFrame.duplicated() 사용하기\n",
    "\n",
    "trade.duplicated()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "e8aeae57",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>187</th>\n",
       "      <td>2020년 03월</td>\n",
       "      <td>중국</td>\n",
       "      <td>248059.0</td>\n",
       "      <td>10658599.0</td>\n",
       "      <td>358234.0</td>\n",
       "      <td>8948918.0</td>\n",
       "      <td>1709682.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액      수입건수       수입금액       무역수지\n",
       "187  2020년 03월  중국  248059.0  10658599.0  358234.0  8948918.0  1709682.0"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade[trade.duplicated()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "111575c8",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>186</th>\n",
       "      <td>2020년 03월</td>\n",
       "      <td>중국</td>\n",
       "      <td>248059.0</td>\n",
       "      <td>10658599.0</td>\n",
       "      <td>358234.0</td>\n",
       "      <td>8948918.0</td>\n",
       "      <td>1709682.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187</th>\n",
       "      <td>2020년 03월</td>\n",
       "      <td>중국</td>\n",
       "      <td>248059.0</td>\n",
       "      <td>10658599.0</td>\n",
       "      <td>358234.0</td>\n",
       "      <td>8948918.0</td>\n",
       "      <td>1709682.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액      수입건수       수입금액       무역수지\n",
       "186  2020년 03월  중국  248059.0  10658599.0  358234.0  8948918.0  1709682.0\n",
       "187  2020년 03월  중국  248059.0  10658599.0  358234.0  8948918.0  1709682.0"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade[(trade['기간'] == '2020년 03월') & (trade['국가명'] == '중국') & (trade['무역수지'] == 1709682.0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "2423b008",
   "metadata": {},
   "outputs": [],
   "source": [
    "# DataFrame.drop_duplicates()를 통해 중복값 지우기\n",
    "trade.drop_duplicates(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "391ea7dd",
   "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>id</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>001</td>\n",
       "      <td>Park Yun</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>002</td>\n",
       "      <td>Kim Sung</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>003</td>\n",
       "      <td>Park Jin</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>004</td>\n",
       "      <td>Lee Han</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>002</td>\n",
       "      <td>Kim Min</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    id      name\n",
       "0  001  Park Yun\n",
       "1  002  Kim Sung\n",
       "2  003  Park Jin\n",
       "3  004   Lee Han\n",
       "4  002   Kim Min"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# drop_duplicates() 자세히 살펴보기\n",
    "\n",
    "df = pd.DataFrame({'id':['001', '002', '003', '004', '002'], \n",
    "                   'name':['Park Yun', 'Kim Sung', 'Park Jin', 'Lee Han', 'Kim Min']})\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "268b648a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop_duplicates(subset='id', keep='last', inplace=True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e8b020f8",
   "metadata": {},
   "source": [
    "## 16-4. 이상치(Outlier)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b78ee571",
   "metadata": {},
   "source": [
    "#### z-score method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "019382bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "👽 It's okay, no biggie.\n"
     ]
    }
   ],
   "source": [
    "# outlier 라는 함수 만들기\n",
    "\n",
    "def outlier(df, col, z):\n",
    "    return df[abs(df[col] - np.mean(df[col]))/np.std(df[col])>z].index\n",
    "print(\"👽 It's okay, no biggie.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "5c73bcb0",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2015년 03월</td>\n",
       "      <td>중국</td>\n",
       "      <td>117529.0</td>\n",
       "      <td>11868032.0</td>\n",
       "      <td>234321.0</td>\n",
       "      <td>7226911.0</td>\n",
       "      <td>4641121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>2017년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>159062.0</td>\n",
       "      <td>11118131.0</td>\n",
       "      <td>188555.0</td>\n",
       "      <td>6600637.0</td>\n",
       "      <td>4517495.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>2017년 03월</td>\n",
       "      <td>일본</td>\n",
       "      <td>65093.0</td>\n",
       "      <td>2395932.0</td>\n",
       "      <td>165734.0</td>\n",
       "      <td>5157589.0</td>\n",
       "      <td>-2761657.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>2017년 09월</td>\n",
       "      <td>중국</td>\n",
       "      <td>183442.0</td>\n",
       "      <td>13540683.0</td>\n",
       "      <td>295443.0</td>\n",
       "      <td>8443414.0</td>\n",
       "      <td>5097269.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>2017년 10월</td>\n",
       "      <td>중국</td>\n",
       "      <td>137873.0</td>\n",
       "      <td>12580474.0</td>\n",
       "      <td>244977.0</td>\n",
       "      <td>7932403.0</td>\n",
       "      <td>4648071.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101</th>\n",
       "      <td>2017년 10월</td>\n",
       "      <td>일본</td>\n",
       "      <td>63510.0</td>\n",
       "      <td>1847999.0</td>\n",
       "      <td>127696.0</td>\n",
       "      <td>4418583.0</td>\n",
       "      <td>-2570584.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>2017년 11월</td>\n",
       "      <td>중국</td>\n",
       "      <td>421194.0</td>\n",
       "      <td>14000887.0</td>\n",
       "      <td>307790.0</td>\n",
       "      <td>9253318.0</td>\n",
       "      <td>4747569.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>2017년 12월</td>\n",
       "      <td>중국</td>\n",
       "      <td>218114.0</td>\n",
       "      <td>13848364.0</td>\n",
       "      <td>290347.0</td>\n",
       "      <td>8600132.0</td>\n",
       "      <td>5248232.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>2018년 03월</td>\n",
       "      <td>중국</td>\n",
       "      <td>232396.0</td>\n",
       "      <td>13576667.0</td>\n",
       "      <td>267249.0</td>\n",
       "      <td>8412516.0</td>\n",
       "      <td>5164151.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>2018년 03월</td>\n",
       "      <td>일본</td>\n",
       "      <td>80142.0</td>\n",
       "      <td>2603450.0</td>\n",
       "      <td>159601.0</td>\n",
       "      <td>5226141.0</td>\n",
       "      <td>-2622691.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>2018년 05월</td>\n",
       "      <td>중국</td>\n",
       "      <td>214145.0</td>\n",
       "      <td>13851900.0</td>\n",
       "      <td>307183.0</td>\n",
       "      <td>9279720.0</td>\n",
       "      <td>4572180.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>2018년 06월</td>\n",
       "      <td>중국</td>\n",
       "      <td>257130.0</td>\n",
       "      <td>13814241.0</td>\n",
       "      <td>279023.0</td>\n",
       "      <td>8713018.0</td>\n",
       "      <td>5101223.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>2018년 07월</td>\n",
       "      <td>중국</td>\n",
       "      <td>181772.0</td>\n",
       "      <td>13721233.0</td>\n",
       "      <td>293164.0</td>\n",
       "      <td>8869278.0</td>\n",
       "      <td>4851955.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>2018년 08월</td>\n",
       "      <td>중국</td>\n",
       "      <td>199010.0</td>\n",
       "      <td>14401521.0</td>\n",
       "      <td>280033.0</td>\n",
       "      <td>8525532.0</td>\n",
       "      <td>5875989.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>2018년 09월</td>\n",
       "      <td>중국</td>\n",
       "      <td>171328.0</td>\n",
       "      <td>14590529.0</td>\n",
       "      <td>280337.0</td>\n",
       "      <td>7889890.0</td>\n",
       "      <td>6700639.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>2018년 10월</td>\n",
       "      <td>중국</td>\n",
       "      <td>169809.0</td>\n",
       "      <td>14767041.0</td>\n",
       "      <td>319876.0</td>\n",
       "      <td>9963108.0</td>\n",
       "      <td>4803932.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액      수입건수       수입금액       무역수지\n",
       "6    2015년 03월  중국  117529.0  11868032.0  234321.0  7226911.0  4641121.0\n",
       "75   2017년 02월  중국  159062.0  11118131.0  188555.0  6600637.0  4517495.0\n",
       "80   2017년 03월  일본   65093.0   2395932.0  165734.0  5157589.0 -2761657.0\n",
       "96   2017년 09월  중국  183442.0  13540683.0  295443.0  8443414.0  5097269.0\n",
       "99   2017년 10월  중국  137873.0  12580474.0  244977.0  7932403.0  4648071.0\n",
       "101  2017년 10월  일본   63510.0   1847999.0  127696.0  4418583.0 -2570584.0\n",
       "102  2017년 11월  중국  421194.0  14000887.0  307790.0  9253318.0  4747569.0\n",
       "105  2017년 12월  중국  218114.0  13848364.0  290347.0  8600132.0  5248232.0\n",
       "114  2018년 03월  중국  232396.0  13576667.0  267249.0  8412516.0  5164151.0\n",
       "116  2018년 03월  일본   80142.0   2603450.0  159601.0  5226141.0 -2622691.0\n",
       "120  2018년 05월  중국  214145.0  13851900.0  307183.0  9279720.0  4572180.0\n",
       "123  2018년 06월  중국  257130.0  13814241.0  279023.0  8713018.0  5101223.0\n",
       "126  2018년 07월  중국  181772.0  13721233.0  293164.0  8869278.0  4851955.0\n",
       "129  2018년 08월  중국  199010.0  14401521.0  280033.0  8525532.0  5875989.0\n",
       "132  2018년 09월  중국  171328.0  14590529.0  280337.0  7889890.0  6700639.0\n",
       "135  2018년 10월  중국  169809.0  14767041.0  319876.0  9963108.0  4803932.0"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.loc[outlier(trade, '무역수지', 1.5)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "d2ba84b8",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>2018년 08월</td>\n",
       "      <td>중국</td>\n",
       "      <td>199010.0</td>\n",
       "      <td>14401521.0</td>\n",
       "      <td>280033.0</td>\n",
       "      <td>8525532.0</td>\n",
       "      <td>5875989.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>2018년 09월</td>\n",
       "      <td>중국</td>\n",
       "      <td>171328.0</td>\n",
       "      <td>14590529.0</td>\n",
       "      <td>280337.0</td>\n",
       "      <td>7889890.0</td>\n",
       "      <td>6700639.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액      수입건수       수입금액       무역수지\n",
       "129  2018년 08월  중국  199010.0  14401521.0  280033.0  8525532.0  5875989.0\n",
       "132  2018년 09월  중국  171328.0  14590529.0  280337.0  7889890.0  6700639.0"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.loc[outlier(trade, '무역수지', 2)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "c1de5039",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [기간, 국가명, 수출건수, 수출금액, 수입건수, 수입금액, 무역수지]\n",
       "Index: []"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.loc[outlier(trade, '무역수지', 3)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "76b85af6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# z score라는 건 결국 평균으로부터 얼마나 그 값이 떨어져 있는지를 구하고,\n",
    "# 그 값이 표준편차의 몇 배인지,\n",
    "# 즉 평균과 떨어진 값에 비교해서 그것과 같은 수준인지 아님 2배인지를\n",
    "# 확인하는 것이다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "c1bda8f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "👽 It's okay, no biggie.\n"
     ]
    }
   ],
   "source": [
    "# not_outlier 함수를 만들어서 이상치가 아닌 값들만 갖고 오기\n",
    "\n",
    "def not_outlier(df, col, z):\n",
    "    return df[abs(df[col] - np.mean(df[col]))/np.std(df[col]) <= z].index\n",
    "print(\"👽 It's okay, no biggie.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "6a810a21",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>중국</td>\n",
       "      <td>116932.0</td>\n",
       "      <td>12083947.0</td>\n",
       "      <td>334522.0</td>\n",
       "      <td>8143271.0</td>\n",
       "      <td>3940676.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>미국</td>\n",
       "      <td>65888.0</td>\n",
       "      <td>5561545.0</td>\n",
       "      <td>509564.0</td>\n",
       "      <td>3625062.0</td>\n",
       "      <td>1936484.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>일본</td>\n",
       "      <td>54017.0</td>\n",
       "      <td>2251307.0</td>\n",
       "      <td>82480.0</td>\n",
       "      <td>3827247.0</td>\n",
       "      <td>-1575940.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>86228.0</td>\n",
       "      <td>9927642.0</td>\n",
       "      <td>209100.0</td>\n",
       "      <td>6980874.0</td>\n",
       "      <td>2946768.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>미국</td>\n",
       "      <td>60225.0</td>\n",
       "      <td>5021264.0</td>\n",
       "      <td>428678.0</td>\n",
       "      <td>2998216.0</td>\n",
       "      <td>2023048.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>5946782.0</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>908043.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>일본</td>\n",
       "      <td>134118.0</td>\n",
       "      <td>1989323.0</td>\n",
       "      <td>141207.0</td>\n",
       "      <td>3989562.0</td>\n",
       "      <td>-2000239.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>중국</td>\n",
       "      <td>185320.0</td>\n",
       "      <td>10746069.0</td>\n",
       "      <td>349007.0</td>\n",
       "      <td>8989920.0</td>\n",
       "      <td>1756149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>미국</td>\n",
       "      <td>126598.0</td>\n",
       "      <td>4600726.0</td>\n",
       "      <td>1157163.0</td>\n",
       "      <td>4286873.0</td>\n",
       "      <td>313853.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>일본</td>\n",
       "      <td>166568.0</td>\n",
       "      <td>1798128.0</td>\n",
       "      <td>133763.0</td>\n",
       "      <td>3102734.0</td>\n",
       "      <td>-1304606.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>179 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액       수입건수       수입금액       무역수지\n",
       "0    2015년 01월  중국  116932.0  12083947.0   334522.0  8143271.0  3940676.0\n",
       "1    2015년 01월  미국   65888.0   5561545.0   509564.0  3625062.0  1936484.0\n",
       "2    2015년 01월  일본   54017.0   2251307.0    82480.0  3827247.0 -1575940.0\n",
       "3    2015년 02월  중국   86228.0   9927642.0   209100.0  6980874.0  2946768.0\n",
       "4    2015년 02월  미국   60225.0   5021264.0   428678.0  2998216.0  2023048.0\n",
       "..         ...  ..       ...         ...        ...        ...        ...\n",
       "191  2020년 04월  미국  105360.0   5946782.0  1141118.0  5038739.0   908043.0\n",
       "192  2020년 04월  일본  134118.0   1989323.0   141207.0  3989562.0 -2000239.0\n",
       "193  2020년 05월  중국  185320.0  10746069.0   349007.0  8989920.0  1756149.0\n",
       "194  2020년 05월  미국  126598.0   4600726.0  1157163.0  4286873.0   313853.0\n",
       "195  2020년 05월  일본  166568.0   1798128.0   133763.0  3102734.0 -1304606.0\n",
       "\n",
       "[179 rows x 7 columns]"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.loc[not_outlier(trade, '무역수지', 1.5)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "d1585a88",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>중국</td>\n",
       "      <td>116932.0</td>\n",
       "      <td>12083947.0</td>\n",
       "      <td>334522.0</td>\n",
       "      <td>8143271.0</td>\n",
       "      <td>3940676.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>미국</td>\n",
       "      <td>65888.0</td>\n",
       "      <td>5561545.0</td>\n",
       "      <td>509564.0</td>\n",
       "      <td>3625062.0</td>\n",
       "      <td>1936484.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>일본</td>\n",
       "      <td>54017.0</td>\n",
       "      <td>2251307.0</td>\n",
       "      <td>82480.0</td>\n",
       "      <td>3827247.0</td>\n",
       "      <td>-1575940.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>86228.0</td>\n",
       "      <td>9927642.0</td>\n",
       "      <td>209100.0</td>\n",
       "      <td>6980874.0</td>\n",
       "      <td>2946768.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>미국</td>\n",
       "      <td>60225.0</td>\n",
       "      <td>5021264.0</td>\n",
       "      <td>428678.0</td>\n",
       "      <td>2998216.0</td>\n",
       "      <td>2023048.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>105360.0</td>\n",
       "      <td>5946782.0</td>\n",
       "      <td>1141118.0</td>\n",
       "      <td>5038739.0</td>\n",
       "      <td>908043.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>일본</td>\n",
       "      <td>134118.0</td>\n",
       "      <td>1989323.0</td>\n",
       "      <td>141207.0</td>\n",
       "      <td>3989562.0</td>\n",
       "      <td>-2000239.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>중국</td>\n",
       "      <td>185320.0</td>\n",
       "      <td>10746069.0</td>\n",
       "      <td>349007.0</td>\n",
       "      <td>8989920.0</td>\n",
       "      <td>1756149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>미국</td>\n",
       "      <td>126598.0</td>\n",
       "      <td>4600726.0</td>\n",
       "      <td>1157163.0</td>\n",
       "      <td>4286873.0</td>\n",
       "      <td>313853.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>일본</td>\n",
       "      <td>166568.0</td>\n",
       "      <td>1798128.0</td>\n",
       "      <td>133763.0</td>\n",
       "      <td>3102734.0</td>\n",
       "      <td>-1304606.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>193 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수        수출금액       수입건수       수입금액       무역수지\n",
       "0    2015년 01월  중국  116932.0  12083947.0   334522.0  8143271.0  3940676.0\n",
       "1    2015년 01월  미국   65888.0   5561545.0   509564.0  3625062.0  1936484.0\n",
       "2    2015년 01월  일본   54017.0   2251307.0    82480.0  3827247.0 -1575940.0\n",
       "3    2015년 02월  중국   86228.0   9927642.0   209100.0  6980874.0  2946768.0\n",
       "4    2015년 02월  미국   60225.0   5021264.0   428678.0  2998216.0  2023048.0\n",
       "..         ...  ..       ...         ...        ...        ...        ...\n",
       "191  2020년 04월  미국  105360.0   5946782.0  1141118.0  5038739.0   908043.0\n",
       "192  2020년 04월  일본  134118.0   1989323.0   141207.0  3989562.0 -2000239.0\n",
       "193  2020년 05월  중국  185320.0  10746069.0   349007.0  8989920.0  1756149.0\n",
       "194  2020년 05월  미국  126598.0   4600726.0  1157163.0  4286873.0   313853.0\n",
       "195  2020년 05월  일본  166568.0   1798128.0   133763.0  3102734.0 -1304606.0\n",
       "\n",
       "[193 rows x 7 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.loc[not_outlier(trade, '무역수지', 2)]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c06f0b5",
   "metadata": {},
   "source": [
    "## IQR method"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b043e9a4",
   "metadata": {},
   "source": [
    "- 한계점이 있는 z-score 대신에 IQR 사용하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "25a677bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1.76884571,  0.07555227, -1.1306297 , -0.65143017, -0.89311563,\n",
       "       -1.27410098, -0.06115443,  0.06451384,  0.41011295, -0.57288249,\n",
       "       -0.80133362,  1.31203519,  1.27469887, -1.2143576 ,  0.31371941,\n",
       "       -1.44482142, -0.3689613 , -0.76922658,  0.3926161 ,  0.05729383,\n",
       "        2.08997884,  0.04197131, -0.04834072, -0.51315392, -0.08458928,\n",
       "       -1.21545008, -1.41293073, -1.48691055,  0.38222486,  0.937673  ,\n",
       "        1.77267804,  0.87882801,  0.33171912, -0.30603567,  1.24026615,\n",
       "       -0.21562684,  0.15592948,  0.09805553,  0.83209585,  2.04520542,\n",
       "       -0.31681392, -1.31283291, -1.75445746,  0.10209408, -1.36150208,\n",
       "        0.48178488, -0.20832874, -0.09186351,  0.70268816,  0.10365506,\n",
       "        0.62123638,  0.95411497,  2.03781352, -0.48445122,  0.2071549 ,\n",
       "        1.64424216, -0.4882074 , -0.01782826,  0.46891556,  0.27987266,\n",
       "       -0.64664972, -0.54406002, -0.16008985,  0.03781172,  1.03383296,\n",
       "       -1.23096117, -1.24673665,  0.29572055,  2.1409624 , -0.92020227,\n",
       "       -0.06000238,  0.27978391, -1.53126966, -0.30293101, -0.14601413,\n",
       "        0.27746159, -0.13952066,  0.69515966, -0.11338746, -1.233267  ,\n",
       "       -0.79614131, -0.46739138,  0.65890607, -0.41063115,  0.17344356,\n",
       "        0.28946174,  1.03451736,  1.22661712,  1.71998252,  0.40806834,\n",
       "        0.32256894,  1.04722748, -1.8196003 , -0.42582157,  0.12454883,\n",
       "        2.31256634, -0.96557586, -0.34627486,  0.96668378, -0.92550192,\n",
       "        8.        , 10.        , -3.        , -5.        ])"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.seed(2020)\n",
    "data = np.random.randn(100)  # 평균 0, 표준편차 1의 분포에서 100개의 숫자를 샘플링한 데이터 생성\n",
    "data = np.concatenate((data, np.array([8, 10, -3, -5])))      # [8, 10, -3, -5])를 데이터 뒤에 추가함\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "b3e26758",
   "metadata": {},
   "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": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.boxplot(data)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "69b6e4f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.1644925829790964"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Q3, Q1 = np.percentile(data, [75 ,25])\n",
    "IQR = Q3 - Q1\n",
    "IQR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "29db016f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.1644925829790964"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Q3, Q1 = np.percentile(data, [75, 25])\n",
    "IQR = Q3 - Q1\n",
    "IQR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "260e3bae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 95, 100, 101, 102, 103]),)"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# DJ's practice to get indices of outliers\n",
    "\n",
    "np.where((data > Q3 + IQR*1.5) | (data < Q1 - IQR*1.5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "30c493f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 2.31256634,  8.        , 10.        , -3.        , -5.        ])"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[(Q1-1.5*IQR > data)|(Q3+1.5*IQR < data)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "2bada73d",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [기간, 국가명, 수출건수, 수출금액, 수입건수, 수입금액, 무역수지]\n",
       "Index: []"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# DJ's version\n",
    "\n",
    "def outlier2(df, col):\n",
    "    Q3, Q1 = np.percentile(df[col], [75, 25])\n",
    "    IQR = Q3 - Q1\n",
    "    return df[(df[col] > Q3 + IQR*1.5) | (df[col] < Q1 - IQR*1.5)]\n",
    "\n",
    "outlier2(trade, '무역수지')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "ec33802c",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [기간, 국가명, 수출건수, 수출금액, 수입건수, 수입금액, 무역수지]\n",
       "Index: []"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def outlier2(df, col):\n",
    "    q1 = df[col].quantile(0.25)\n",
    "    q3 = df[col].quantile(0.75)\n",
    "    iqr = q3 - q1\n",
    "    return df[(df[col] < q1-1.5*iqr)|(df[col] > q3+1.5*iqr)]\n",
    "\n",
    "outlier2(trade, '무역수지')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5de3742f",
   "metadata": {},
   "source": [
    "## 17-2. 정규화(Normalization)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ef3b5ae",
   "metadata": {},
   "source": [
    "- standardization 데이터의 평균은 0, 분산은 1로 변환\n",
    "- min-max-scaling 데이터의 최솟값은 0, 최댓값은 1로 변환"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "a36dd12f",
   "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>A</th>\n",
       "      <th>B</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-3.075383</td>\n",
       "      <td>-0.448552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.302209</td>\n",
       "      <td>-2.151012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.522519</td>\n",
       "      <td>-1.245304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.394279</td>\n",
       "      <td>-1.169169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.427537</td>\n",
       "      <td>-0.458080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>13.250265</td>\n",
       "      <td>-0.734271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>0.137697</td>\n",
       "      <td>-2.041710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>2.614901</td>\n",
       "      <td>-3.057545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>7.866735</td>\n",
       "      <td>-2.140529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>0.297992</td>\n",
       "      <td>-0.503868</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            A         B\n",
       "0   -3.075383 -0.448552\n",
       "1    4.302209 -2.151012\n",
       "2   -0.522519 -1.245304\n",
       "3    1.394279 -1.169169\n",
       "4    0.427537 -0.458080\n",
       "..        ...       ...\n",
       "95  13.250265 -0.734271\n",
       "96   0.137697 -2.041710\n",
       "97   2.614901 -3.057545\n",
       "98   7.866735 -2.140529\n",
       "99   0.297992 -0.503868\n",
       "\n",
       "[100 rows x 2 columns]"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 정규화 방법 펼쳐보기\n",
    "\n",
    "# 정규분포를 따라 랜덤하게 데이터 x를 생성합니다. \n",
    "np.random.seed(2020)\n",
    "x = pd.DataFrame({'A': np.random.randn(100)*4+4,\n",
    "                 'B': np.random.randn(100)-1})\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "117a298f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "A    4.026154\n",
       "B   -1.011467\n",
       "dtype: float64"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "b7136714",
   "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>A</th>\n",
       "      <th>B</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-7.101537</td>\n",
       "      <td>0.562915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.276055</td>\n",
       "      <td>-1.139545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-4.548673</td>\n",
       "      <td>-0.233837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-2.631875</td>\n",
       "      <td>-0.157702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-3.598617</td>\n",
       "      <td>0.553387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>9.224111</td>\n",
       "      <td>0.277196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>-3.888458</td>\n",
       "      <td>-1.030243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>-1.411254</td>\n",
       "      <td>-2.046078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>3.840581</td>\n",
       "      <td>-1.129062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>-3.728162</td>\n",
       "      <td>0.507599</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           A         B\n",
       "0  -7.101537  0.562915\n",
       "1   0.276055 -1.139545\n",
       "2  -4.548673 -0.233837\n",
       "3  -2.631875 -0.157702\n",
       "4  -3.598617  0.553387\n",
       "..       ...       ...\n",
       "95  9.224111  0.277196\n",
       "96 -3.888458 -1.030243\n",
       "97 -1.411254 -2.046078\n",
       "98  3.840581 -1.129062\n",
       "99 -3.728162  0.507599\n",
       "\n",
       "[100 rows x 2 columns]"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x - x.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "647de48b",
   "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>A</th>\n",
       "      <th>B</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.862058</td>\n",
       "      <td>0.559307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.072383</td>\n",
       "      <td>-1.132240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.192684</td>\n",
       "      <td>-0.232338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.690090</td>\n",
       "      <td>-0.156691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.943575</td>\n",
       "      <td>0.549839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>2.418607</td>\n",
       "      <td>0.275419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>-1.019573</td>\n",
       "      <td>-1.023639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>-0.370038</td>\n",
       "      <td>-2.032961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>1.007019</td>\n",
       "      <td>-1.121824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>-0.977542</td>\n",
       "      <td>0.504345</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           A         B\n",
       "0  -1.862058  0.559307\n",
       "1   0.072383 -1.132240\n",
       "2  -1.192684 -0.232338\n",
       "3  -0.690090 -0.156691\n",
       "4  -0.943575  0.549839\n",
       "..       ...       ...\n",
       "95  2.418607  0.275419\n",
       "96 -1.019573 -1.023639\n",
       "97 -0.370038 -2.032961\n",
       "98  1.007019 -1.121824\n",
       "99 -0.977542  0.504345\n",
       "\n",
       "[100 rows x 2 columns]"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 데이터 x를 Standardization 기법으로 정규화합니다. \n",
    "x_standardization = (x - x.mean())/x.std()\n",
    "x_standardization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "e4a54afc",
   "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>A</th>\n",
       "      <th>B</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.012283</td>\n",
       "      <td>0.607216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.458634</td>\n",
       "      <td>0.267692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.166733</td>\n",
       "      <td>0.448318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.282702</td>\n",
       "      <td>0.463502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.224213</td>\n",
       "      <td>0.605316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.550235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>0.206677</td>\n",
       "      <td>0.289490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>0.356550</td>\n",
       "      <td>0.086901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>0.674291</td>\n",
       "      <td>0.269782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>0.216375</td>\n",
       "      <td>0.596184</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           A         B\n",
       "0   0.012283  0.607216\n",
       "1   0.458634  0.267692\n",
       "2   0.166733  0.448318\n",
       "3   0.282702  0.463502\n",
       "4   0.224213  0.605316\n",
       "..       ...       ...\n",
       "95  1.000000  0.550235\n",
       "96  0.206677  0.289490\n",
       "97  0.356550  0.086901\n",
       "98  0.674291  0.269782\n",
       "99  0.216375  0.596184\n",
       "\n",
       "[100 rows x 2 columns]"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 데이터 x를 min-max scaling 기법으로 정규화합니다. \n",
    "x_min_max = (x-x.min())/(x.max()-x.min())\n",
    "x_min_max"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "f2057051",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1,2, figsize=(12, 4),\n",
    "                        gridspec_kw={'width_ratios': [2, 1]})\n",
    "\n",
    "axs[0].scatter(x['A'], x['B'])\n",
    "axs[0].set_xlim(-5, 15)\n",
    "axs[0].set_ylim(-5, 5)\n",
    "axs[0].axvline(c='grey', lw=1)\n",
    "axs[0].axhline(c='grey', lw=1)\n",
    "axs[0].set_title('Original Data')\n",
    "\n",
    "axs[1].scatter(x_standardization['A'], x_standardization['B'])\n",
    "axs[1].set_xlim(-5, 5)\n",
    "axs[1].set_ylim(-5, 5)\n",
    "axs[1].axvline(c='grey', lw=1)\n",
    "axs[1].axhline(c='grey', lw=1)\n",
    "axs[1].set_title('Data after standardization')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "8e363ec6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1,2, figsize=(12, 4),\n",
    "                        gridspec_kw={'width_ratios': [2, 1]})\n",
    "\n",
    "axs[0].scatter(x['A'], x['B'])\n",
    "axs[0].set_xlim(-5, 15)\n",
    "axs[0].set_ylim(-5, 5)\n",
    "axs[0].axvline(c='grey', lw=1)\n",
    "axs[0].axhline(c='grey', lw=1)\n",
    "axs[0].set_title('Original Data')\n",
    "\n",
    "axs[1].scatter(x_min_max['A'], x_min_max['B'])\n",
    "axs[1].set_xlim(-5, 5)\n",
    "axs[1].set_ylim(-5, 5)\n",
    "axs[1].axvline(c='grey', lw=1)\n",
    "axs[1].axhline(c='grey', lw=1)\n",
    "axs[1].set_title('Data after min-max scaling')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "221bb5f2",
   "metadata": {},
   "source": [
    "### Standardization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "bf2a20d8",
   "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>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.007488</td>\n",
       "      <td>1.398931</td>\n",
       "      <td>-0.163593</td>\n",
       "      <td>1.283660</td>\n",
       "      <td>1.256342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.689278</td>\n",
       "      <td>-0.252848</td>\n",
       "      <td>0.412529</td>\n",
       "      <td>-0.964444</td>\n",
       "      <td>0.401088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.847838</td>\n",
       "      <td>-1.091156</td>\n",
       "      <td>-0.993148</td>\n",
       "      <td>-0.863844</td>\n",
       "      <td>-1.097779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.417598</td>\n",
       "      <td>0.852853</td>\n",
       "      <td>-0.576399</td>\n",
       "      <td>0.705292</td>\n",
       "      <td>0.832209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.764918</td>\n",
       "      <td>-0.389673</td>\n",
       "      <td>0.146306</td>\n",
       "      <td>-1.276341</td>\n",
       "      <td>0.438027</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       수출건수      수출금액      수입건수      수입금액      무역수지\n",
       "0 -0.007488  1.398931 -0.163593  1.283660  1.256342\n",
       "1 -0.689278 -0.252848  0.412529 -0.964444  0.401088\n",
       "2 -0.847838 -1.091156 -0.993148 -0.863844 -1.097779\n",
       "3 -0.417598  0.852853 -0.576399  0.705292  0.832209\n",
       "4 -0.764918 -0.389673  0.146306 -1.276341  0.438027"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# trade 데이터를 Standardization 기법으로 정규화합니다. \n",
    "cols = ['수출건수', '수출금액', '수입건수', '수입금액', '무역수지']\n",
    "trade_Standardization= (trade[cols]-trade[cols].mean())/trade[cols].std()\n",
    "trade_Standardization.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "8074d5ff",
   "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>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.950000e+02</td>\n",
       "      <td>1.950000e+02</td>\n",
       "      <td>1.950000e+02</td>\n",
       "      <td>1.950000e+02</td>\n",
       "      <td>1.950000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-1.019128e-16</td>\n",
       "      <td>5.921189e-17</td>\n",
       "      <td>6.091993e-17</td>\n",
       "      <td>-8.312439e-17</td>\n",
       "      <td>3.074464e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-9.194976e-01</td>\n",
       "      <td>-1.231761e+00</td>\n",
       "      <td>-9.984408e-01</td>\n",
       "      <td>-1.276341e+00</td>\n",
       "      <td>-1.603764e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-5.937426e-01</td>\n",
       "      <td>-1.041338e+00</td>\n",
       "      <td>-7.673625e-01</td>\n",
       "      <td>-7.911669e-01</td>\n",
       "      <td>-1.116765e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-4.373265e-01</td>\n",
       "      <td>-1.564700e-01</td>\n",
       "      <td>-3.429346e-01</td>\n",
       "      <td>-4.137392e-01</td>\n",
       "      <td>1.426824e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.420459e-01</td>\n",
       "      <td>1.037200e+00</td>\n",
       "      <td>3.927781e-01</td>\n",
       "      <td>8.827841e-01</td>\n",
       "      <td>7.461637e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.486317e+00</td>\n",
       "      <td>2.078416e+00</td>\n",
       "      <td>3.239068e+00</td>\n",
       "      <td>2.376092e+00</td>\n",
       "      <td>2.434109e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               수출건수          수출금액          수입건수          수입금액          무역수지\n",
       "count  1.950000e+02  1.950000e+02  1.950000e+02  1.950000e+02  1.950000e+02\n",
       "mean  -1.019128e-16  5.921189e-17  6.091993e-17 -8.312439e-17  3.074464e-17\n",
       "std    1.000000e+00  1.000000e+00  1.000000e+00  1.000000e+00  1.000000e+00\n",
       "min   -9.194976e-01 -1.231761e+00 -9.984408e-01 -1.276341e+00 -1.603764e+00\n",
       "25%   -5.937426e-01 -1.041338e+00 -7.673625e-01 -7.911669e-01 -1.116765e+00\n",
       "50%   -4.373265e-01 -1.564700e-01 -3.429346e-01 -4.137392e-01  1.426824e-01\n",
       "75%    4.420459e-01  1.037200e+00  3.927781e-01  8.827841e-01  7.461637e-01\n",
       "max    5.486317e+00  2.078416e+00  3.239068e+00  2.376092e+00  2.434109e+00"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade_Standardization.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b31a103",
   "metadata": {},
   "source": [
    "### min-max-scaling으로 정규화(Nomarlisation)하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "2d00721f",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>중국</td>\n",
       "      <td>0.142372</td>\n",
       "      <td>0.794728</td>\n",
       "      <td>0.197014</td>\n",
       "      <td>0.700903</td>\n",
       "      <td>0.708320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.035939</td>\n",
       "      <td>0.295728</td>\n",
       "      <td>0.332972</td>\n",
       "      <td>0.085394</td>\n",
       "      <td>0.496512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>일본</td>\n",
       "      <td>0.011187</td>\n",
       "      <td>0.042477</td>\n",
       "      <td>0.001249</td>\n",
       "      <td>0.112938</td>\n",
       "      <td>0.125310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>0.078351</td>\n",
       "      <td>0.629759</td>\n",
       "      <td>0.099597</td>\n",
       "      <td>0.542551</td>\n",
       "      <td>0.603281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.024131</td>\n",
       "      <td>0.254394</td>\n",
       "      <td>0.270146</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.505660</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          기간 국가명      수출건수      수출금액      수입건수      수입금액      무역수지\n",
       "0  2015년 01월  중국  0.142372  0.794728  0.197014  0.700903  0.708320\n",
       "1  2015년 01월  미국  0.035939  0.295728  0.332972  0.085394  0.496512\n",
       "2  2015년 01월  일본  0.011187  0.042477  0.001249  0.112938  0.125310\n",
       "3  2015년 02월  중국  0.078351  0.629759  0.099597  0.542551  0.603281\n",
       "4  2015년 02월  미국  0.024131  0.254394  0.270146  0.000000  0.505660"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# trade 데이터를 min-max scaling 기법으로 정규화합니다. \n",
    "trade[cols] = (trade[cols]-trade[cols].min())/(trade[cols].max()-trade[cols].min())\n",
    "trade.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "253d79fc",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>중국</td>\n",
       "      <td>0.142372</td>\n",
       "      <td>0.794728</td>\n",
       "      <td>0.197014</td>\n",
       "      <td>0.700903</td>\n",
       "      <td>0.708320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.035939</td>\n",
       "      <td>0.295728</td>\n",
       "      <td>0.332972</td>\n",
       "      <td>0.085394</td>\n",
       "      <td>0.496512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>일본</td>\n",
       "      <td>0.011187</td>\n",
       "      <td>0.042477</td>\n",
       "      <td>0.001249</td>\n",
       "      <td>0.112938</td>\n",
       "      <td>0.125310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>0.078351</td>\n",
       "      <td>0.629759</td>\n",
       "      <td>0.099597</td>\n",
       "      <td>0.542551</td>\n",
       "      <td>0.603281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.024131</td>\n",
       "      <td>0.254394</td>\n",
       "      <td>0.270146</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.505660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.118243</td>\n",
       "      <td>0.325201</td>\n",
       "      <td>0.823509</td>\n",
       "      <td>0.277977</td>\n",
       "      <td>0.387823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>2020년 04월</td>\n",
       "      <td>일본</td>\n",
       "      <td>0.178207</td>\n",
       "      <td>0.022433</td>\n",
       "      <td>0.046863</td>\n",
       "      <td>0.135050</td>\n",
       "      <td>0.080469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>중국</td>\n",
       "      <td>0.284970</td>\n",
       "      <td>0.692373</td>\n",
       "      <td>0.208265</td>\n",
       "      <td>0.816241</td>\n",
       "      <td>0.477453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.162527</td>\n",
       "      <td>0.222220</td>\n",
       "      <td>0.835972</td>\n",
       "      <td>0.175552</td>\n",
       "      <td>0.325028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>2020년 05월</td>\n",
       "      <td>일본</td>\n",
       "      <td>0.245869</td>\n",
       "      <td>0.007806</td>\n",
       "      <td>0.041081</td>\n",
       "      <td>0.014238</td>\n",
       "      <td>0.153985</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>195 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            기간 국가명      수출건수      수출금액      수입건수      수입금액      무역수지\n",
       "0    2015년 01월  중국  0.142372  0.794728  0.197014  0.700903  0.708320\n",
       "1    2015년 01월  미국  0.035939  0.295728  0.332972  0.085394  0.496512\n",
       "2    2015년 01월  일본  0.011187  0.042477  0.001249  0.112938  0.125310\n",
       "3    2015년 02월  중국  0.078351  0.629759  0.099597  0.542551  0.603281\n",
       "4    2015년 02월  미국  0.024131  0.254394  0.270146  0.000000  0.505660\n",
       "..         ...  ..       ...       ...       ...       ...       ...\n",
       "191  2020년 04월  미국  0.118243  0.325201  0.823509  0.277977  0.387823\n",
       "192  2020년 04월  일본  0.178207  0.022433  0.046863  0.135050  0.080469\n",
       "193  2020년 05월  중국  0.284970  0.692373  0.208265  0.816241  0.477453\n",
       "194  2020년 05월  미국  0.162527  0.222220  0.835972  0.175552  0.325028\n",
       "195  2020년 05월  일본  0.245869  0.007806  0.041081  0.014238  0.153985\n",
       "\n",
       "[195 rows x 7 columns]"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "549660cb",
   "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>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.143541</td>\n",
       "      <td>0.372113</td>\n",
       "      <td>0.235620</td>\n",
       "      <td>0.349450</td>\n",
       "      <td>0.397180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.156108</td>\n",
       "      <td>0.302099</td>\n",
       "      <td>0.235988</td>\n",
       "      <td>0.273790</td>\n",
       "      <td>0.247655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.050853</td>\n",
       "      <td>0.057527</td>\n",
       "      <td>0.054532</td>\n",
       "      <td>0.132836</td>\n",
       "      <td>0.120608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.075271</td>\n",
       "      <td>0.324844</td>\n",
       "      <td>0.154691</td>\n",
       "      <td>0.236172</td>\n",
       "      <td>0.432516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.212548</td>\n",
       "      <td>0.685450</td>\n",
       "      <td>0.328311</td>\n",
       "      <td>0.591147</td>\n",
       "      <td>0.581972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             수출건수        수출금액        수입건수        수입금액        무역수지\n",
       "count  195.000000  195.000000  195.000000  195.000000  195.000000\n",
       "mean     0.143541    0.372113    0.235620    0.349450    0.397180\n",
       "std      0.156108    0.302099    0.235988    0.273790    0.247655\n",
       "min      0.000000    0.000000    0.000000    0.000000    0.000000\n",
       "25%      0.050853    0.057527    0.054532    0.132836    0.120608\n",
       "50%      0.075271    0.324844    0.154691    0.236172    0.432516\n",
       "75%      0.212548    0.685450    0.328311    0.591147    0.581972\n",
       "max      1.000000    1.000000    1.000000    1.000000    1.000000"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ccbfc321",
   "metadata": {},
   "source": [
    "#### train, test 데이터가 따로 있을 경우 둘 다 같은 방법으로 normalisation 해줘야 함"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "c1869490",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "👽 It's okay, no biggie.\n"
     ]
    }
   ],
   "source": [
    "train = pd.DataFrame([[10, -10], [30, 10], [50, 0]])\n",
    "test = pd.DataFrame([[0, 1], [10, 10]])\n",
    "print(\"👽 It's okay, no biggie.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "e0e21a17",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "💫 It's okay, no biggie...\n"
     ]
    }
   ],
   "source": [
    "train_min = train.min()\n",
    "train_max = train.max()\n",
    "\n",
    "train_min_max = (train - train_min)/(train_max - train_min)\n",
    "test_min_max =  (test - train_min)/(train_max - train_min)    # test를 min-max scaling할 때도 train 정규화 기준으로 수행\n",
    "print(\"💫 It's okay, no biggie...\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "c0d9659c",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     0    1\n",
       "0  0.0  0.0\n",
       "1  0.5  1.0\n",
       "2  1.0  0.5"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_min_max"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "e9bd3fda",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.25</td>\n",
       "      <td>0.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      0     1\n",
       "0 -0.25  0.55\n",
       "1  0.00  1.00"
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_min_max"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "301898cb",
   "metadata": {},
   "source": [
    "#### scikit-learn의 StandardScaler, MinMaxScaler를 사용하는 방법"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "cb94b8e1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "👽 It's okay, no biggie.\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "train = [[10, -10], [30, 10], [50, 0]]\n",
    "test = [[0, 1]]\n",
    "scaler = MinMaxScaler()\n",
    "print(\"👽 It's okay, no biggie.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "1a7e6c82",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MinMaxScaler()"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "d932c4dd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0. , 0. ],\n",
       "       [0.5, 1. ],\n",
       "       [1. , 0.5]])"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler.fit_transform(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "f6bbdf27",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.25,  0.55]])"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler.transform(test) # fit_transform 이 아닌 것에 주의"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33b3500d",
   "metadata": {},
   "source": [
    "## 17-3. 원-핫 인코딩(One-Hot Encoding)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "021b1b99",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    중국\n",
      "1    미국\n",
      "2    일본\n",
      "3    중국\n",
      "4    미국\n",
      "Name: 국가명, dtype: object\n"
     ]
    },
    {
     "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>미국</th>\n",
       "      <th>일본</th>\n",
       "      <th>중국</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   미국  일본  중국\n",
       "0   0   0   1\n",
       "1   1   0   0\n",
       "2   0   1   0\n",
       "3   0   0   1\n",
       "4   1   0   0"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#trade 데이터의 국가명 컬럼 원본\n",
    "print(trade['국가명'].head())  \n",
    "\n",
    "# get_dummies를 통해 국가명 원-핫 인코딩\n",
    "country = pd.get_dummies(trade['국가명'])\n",
    "country.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3da5d487",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "65f58271",
   "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>기간</th>\n",
       "      <th>국가명</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "      <th>미국</th>\n",
       "      <th>일본</th>\n",
       "      <th>중국</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>중국</td>\n",
       "      <td>0.142372</td>\n",
       "      <td>0.794728</td>\n",
       "      <td>0.197014</td>\n",
       "      <td>0.700903</td>\n",
       "      <td>0.708320</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.035939</td>\n",
       "      <td>0.295728</td>\n",
       "      <td>0.332972</td>\n",
       "      <td>0.085394</td>\n",
       "      <td>0.496512</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>일본</td>\n",
       "      <td>0.011187</td>\n",
       "      <td>0.042477</td>\n",
       "      <td>0.001249</td>\n",
       "      <td>0.112938</td>\n",
       "      <td>0.125310</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>중국</td>\n",
       "      <td>0.078351</td>\n",
       "      <td>0.629759</td>\n",
       "      <td>0.099597</td>\n",
       "      <td>0.542551</td>\n",
       "      <td>0.603281</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>미국</td>\n",
       "      <td>0.024131</td>\n",
       "      <td>0.254394</td>\n",
       "      <td>0.270146</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.505660</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          기간 국가명      수출건수      수출금액      수입건수      수입금액      무역수지  미국  일본  중국\n",
       "0  2015년 01월  중국  0.142372  0.794728  0.197014  0.700903  0.708320   0   0   1\n",
       "1  2015년 01월  미국  0.035939  0.295728  0.332972  0.085394  0.496512   1   0   0\n",
       "2  2015년 01월  일본  0.011187  0.042477  0.001249  0.112938  0.125310   0   1   0\n",
       "3  2015년 02월  중국  0.078351  0.629759  0.099597  0.542551  0.603281   0   0   1\n",
       "4  2015년 02월  미국  0.024131  0.254394  0.270146  0.000000  0.505660   1   0   0"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade = pd.concat([trade, country], axis=1)\n",
    "trade.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "7baf6163",
   "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>기간</th>\n",
       "      <th>수출건수</th>\n",
       "      <th>수출금액</th>\n",
       "      <th>수입건수</th>\n",
       "      <th>수입금액</th>\n",
       "      <th>무역수지</th>\n",
       "      <th>미국</th>\n",
       "      <th>일본</th>\n",
       "      <th>중국</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>0.142372</td>\n",
       "      <td>0.794728</td>\n",
       "      <td>0.197014</td>\n",
       "      <td>0.700903</td>\n",
       "      <td>0.708320</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>0.035939</td>\n",
       "      <td>0.295728</td>\n",
       "      <td>0.332972</td>\n",
       "      <td>0.085394</td>\n",
       "      <td>0.496512</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015년 01월</td>\n",
       "      <td>0.011187</td>\n",
       "      <td>0.042477</td>\n",
       "      <td>0.001249</td>\n",
       "      <td>0.112938</td>\n",
       "      <td>0.125310</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>0.078351</td>\n",
       "      <td>0.629759</td>\n",
       "      <td>0.099597</td>\n",
       "      <td>0.542551</td>\n",
       "      <td>0.603281</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015년 02월</td>\n",
       "      <td>0.024131</td>\n",
       "      <td>0.254394</td>\n",
       "      <td>0.270146</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.505660</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          기간      수출건수      수출금액      수입건수      수입금액      무역수지  미국  일본  중국\n",
       "0  2015년 01월  0.142372  0.794728  0.197014  0.700903  0.708320   0   0   1\n",
       "1  2015년 01월  0.035939  0.295728  0.332972  0.085394  0.496512   1   0   0\n",
       "2  2015년 01월  0.011187  0.042477  0.001249  0.112938  0.125310   0   1   0\n",
       "3  2015년 02월  0.078351  0.629759  0.099597  0.542551  0.603281   0   0   1\n",
       "4  2015년 02월  0.024131  0.254394  0.270146  0.000000  0.505660   1   0   0"
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trade.drop('국가명', axis=1, inplace=True)\n",
    "trade.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3af3df58",
   "metadata": {},
   "source": [
    "## 17-4. 구간화(Binning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "a21781ff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "👽 Almost there..\n"
     ]
    }
   ],
   "source": [
    "# salary 라는 데이터 넣어보기\n",
    "\n",
    "salary = pd.Series([4300, 8370, 1750, 3830, 1840, 4220, 3020, 2290, 4740, 4600, \n",
    "                    2860, 3400, 4800, 4470, 2440, 4530, 4850, 4850, 4760, 4500, \n",
    "                    4640, 3000, 1880, 4880, 2240, 4750, 2750, 2810, 3100, 4290, \n",
    "                    1540, 2870, 1780, 4670, 4150, 2010, 3580, 1610, 2930, 4300, \n",
    "                    2740, 1680, 3490, 4350, 1680, 6420, 8740, 8980, 9080, 3990, \n",
    "                    4960, 3700, 9600, 9330, 5600, 4100, 1770, 8280, 3120, 1950, \n",
    "                    4210, 2020, 3820, 3170, 6330, 2570, 6940, 8610, 5060, 6370,\n",
    "                    9080, 3760, 8060, 2500, 4660, 1770, 9220, 3380, 2490, 3450, \n",
    "                    1960, 7210, 5810, 9450, 8910, 3470, 7350, 8410, 7520, 9610, \n",
    "                    5150, 2630, 5610, 2750, 7050, 3350, 9450, 7140, 4170, 3090])\n",
    "print(\"👽 Almost there..\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35d76811",
   "metadata": {},
   "source": [
    "#### pandas 'cut', 'qcut'을 통해 수치형 데이터를 범주형 데이터로 변환"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "26edb0df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "👽 Almost there..\n"
     ]
    }
   ],
   "source": [
    "bins = [0, 2000, 4000, 6000, 8000, 10000]\n",
    "print(\"👽 Almost there..\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "b59baeff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      (4000, 6000]\n",
       "1     (8000, 10000]\n",
       "2         (0, 2000]\n",
       "3      (2000, 4000]\n",
       "4         (0, 2000]\n",
       "          ...      \n",
       "95     (2000, 4000]\n",
       "96    (8000, 10000]\n",
       "97     (6000, 8000]\n",
       "98     (4000, 6000]\n",
       "99     (2000, 4000]\n",
       "Length: 100, dtype: category\n",
       "Categories (5, interval[int64, right]): [(0, 2000] < (2000, 4000] < (4000, 6000] < (6000, 8000] < (8000, 10000]]"
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ctg = pd.cut(salary, bins=bins)\n",
    "ctg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "id": "998b2003",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "salary[0]: 4300\n",
      "salary[0]가 속한 카테고리: (4000, 6000]\n"
     ]
    }
   ],
   "source": [
    "print('salary[0]:', salary[0])\n",
    "print('salary[0]가 속한 카테고리:', ctg[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "id": "00fb351c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 2000]        12\n",
       "(2000, 4000]     34\n",
       "(4000, 6000]     29\n",
       "(6000, 8000]      9\n",
       "(8000, 10000]    16\n",
       "dtype: int64"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ctg.value_counts().sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "id": "94b0b012",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      (4230.0, 5575.0]\n",
       "1      (8265.0, 9610.0]\n",
       "2     (1531.93, 2885.0]\n",
       "3      (2885.0, 4230.0]\n",
       "4     (1531.93, 2885.0]\n",
       "            ...        \n",
       "95     (2885.0, 4230.0]\n",
       "96     (8265.0, 9610.0]\n",
       "97     (6920.0, 8265.0]\n",
       "98     (2885.0, 4230.0]\n",
       "99     (2885.0, 4230.0]\n",
       "Length: 100, dtype: category\n",
       "Categories (6, interval[float64, right]): [(1531.93, 2885.0] < (2885.0, 4230.0] < (4230.0, 5575.0] < (5575.0, 6920.0] < (6920.0, 8265.0] < (8265.0, 9610.0]]"
      ]
     },
     "execution_count": 184,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# bins 에 그냥 정수형으로 넣어주면 그만큼 균등하게 나눠줌\n",
    "\n",
    "ctg = pd.cut(salary, bins=6)\n",
    "ctg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "id": "2b3d9b01",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      (4230.0, 5575.0]\n",
       "1      (8265.0, 9610.0]\n",
       "2     (1531.93, 2885.0]\n",
       "3      (2885.0, 4230.0]\n",
       "4     (1531.93, 2885.0]\n",
       "            ...        \n",
       "95     (2885.0, 4230.0]\n",
       "96     (8265.0, 9610.0]\n",
       "97     (6920.0, 8265.0]\n",
       "98     (2885.0, 4230.0]\n",
       "99     (2885.0, 4230.0]\n",
       "Length: 100, dtype: category\n",
       "Categories (6, interval[float64, right]): [(1531.93, 2885.0] < (2885.0, 4230.0] < (4230.0, 5575.0] < (5575.0, 6920.0] < (6920.0, 8265.0] < (8265.0, 9610.0]]"
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ctg = pd.cut(salary, bins=6)\n",
    "ctg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "id": "5194168c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       (3544.0, 4648.0]\n",
       "1       (7068.0, 9610.0]\n",
       "2     (1539.999, 2618.0]\n",
       "3       (3544.0, 4648.0]\n",
       "4     (1539.999, 2618.0]\n",
       "             ...        \n",
       "95      (2618.0, 3544.0]\n",
       "96      (7068.0, 9610.0]\n",
       "97      (7068.0, 9610.0]\n",
       "98      (3544.0, 4648.0]\n",
       "99      (2618.0, 3544.0]\n",
       "Length: 100, dtype: category\n",
       "Categories (5, interval[float64, right]): [(1539.999, 2618.0] < (2618.0, 3544.0] < (3544.0, 4648.0] < (4648.0, 7068.0] < (7068.0, 9610.0]]"
      ]
     },
     "execution_count": 188,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# qcut은 값(구간)을 균등하게 나누는 게 아니라 분포를 균등하게 나눔\n",
    "\n",
    "ctg = pd.qcut(salary, q=5)\n",
    "ctg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "id": "0802ce22",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1539.999, 2618.0]    20\n",
      "(2618.0, 3544.0]      20\n",
      "(3544.0, 4648.0]      20\n",
      "(4648.0, 7068.0]      20\n",
      "(7068.0, 9610.0]      20\n",
      "dtype: int64\n",
      ".\n",
      ".\n",
      "🛸 Well done!\n"
     ]
    }
   ],
   "source": [
    "print(ctg.value_counts().sort_index())\n",
    "print(\".\\n.\\n🛸 Well done!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "id": "825bdba4",
   "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>bin_10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(3544.0, 4648.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(7068.0, 9610.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(1539.999, 2618.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(3544.0, 4648.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>(1539.999, 2618.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>(2618.0, 3544.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>(7068.0, 9610.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>(7068.0, 9610.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>(3544.0, 4648.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>(2618.0, 3544.0]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                bin_10\n",
       "0     (3544.0, 4648.0]\n",
       "1     (7068.0, 9610.0]\n",
       "2   (1539.999, 2618.0]\n",
       "3     (3544.0, 4648.0]\n",
       "4   (1539.999, 2618.0]\n",
       "..                 ...\n",
       "95    (2618.0, 3544.0]\n",
       "96    (7068.0, 9610.0]\n",
       "97    (7068.0, 9610.0]\n",
       "98    (3544.0, 4648.0]\n",
       "99    (2618.0, 3544.0]\n",
       "\n",
       "[100 rows x 1 columns]"
      ]
     },
     "execution_count": 214,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ctg_df = pd.DataFrame(ctg)\n",
    "ctg_df.rename(columns={0:'bin_10'}, inplace=True)\n",
    "ctg_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0685b539",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}