CJ_미래기술챌린지_Code.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "kw6GbP-1a4u7",
      "metadata": {
        "id": "kw6GbP-1a4u7"
      },
      "source": [
        "#필요한 모듈 불러오기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d50056e4",
      "metadata": {
        "id": "d50056e4"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import os\n",
        "import pickle\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "from minisom import MiniSom\n",
        "from tqdm import tqdm, tqdm_notebook, tnrange\n",
        "import time\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from multiprocessing import Pool, Manager\n",
        "from datetime import datetime as dt\n",
        "from lxml.html import fromstring, tostring\n",
        "from matplotlib import font_manager, rc\n",
        "from factor_analyzer.factor_analyzer import calculate_bartlett_sphericity\n",
        "from factor_analyzer import FactorAnalyzer\n",
        "import gc\n",
        "plt.rcParams['font.family'] = 'Malgun Gothic'\n",
        "%matplotlib inline  "
      ]
    },
    {
      "cell_type": "markdown",
      "id": "0a41ea72",
      "metadata": {
        "id": "0a41ea72"
      },
      "source": [
        "# 데이터프레임 결합"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d8967a93",
      "metadata": {
        "id": "d8967a93"
      },
      "outputs": [],
      "source": [
        "pickle_path = \"이커머스 FC주문 데이터.pickle\"\n",
        "month_3_path = \"1-1. 이커머스 FC주문 데이터_3월.csv\"\n",
        "month_4_path = \"1-2. 이커머스 FC주문 데이터_4월.csv\"\n",
        "month_5_path = \"1-3. 이커머스 FC주문 데이터_5월.csv\"\n",
        "month_6_path = \"1-4. 이커머스 FC주문 데이터_6월.csv\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "49461c28",
      "metadata": {
        "id": "49461c28"
      },
      "outputs": [],
      "source": [
        "df_M3 = pd.read_csv(month_3_path, low_memory=False, parse_dates=['BKG_DATE','ORDER_CRT_DATETIME'])\n",
        "df_M4 = pd.read_csv(month_4_path, low_memory=False, parse_dates=['BKG_DATE','ORDER_CRT_DATETIME'])\n",
        "df_M5 = pd.read_csv(month_5_path, low_memory=False, parse_dates=['BKG_DATE','ORDER_CRT_DATETIME'])\n",
        "df_M6 = pd.read_csv(month_6_path, low_memory=False, parse_dates=['BKG_DATE','ORDER_CRT_DATETIME'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fd9ca34c",
      "metadata": {
        "id": "fd9ca34c"
      },
      "outputs": [],
      "source": [
        "df=pd.concat([df_M3,df_M4,df_M5,df_M6],ignore_index=True)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5d7edeb3",
      "metadata": {
        "id": "5d7edeb3",
        "outputId": "715c0b2f-c2af-4582-a1b4-d8bc11fc1ea0"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "40"
            ]
          },
          "execution_count": 148,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "del [df_M3,df_M4,df_M5,df_M6]\n",
        "gc.collect()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3b734f9a",
      "metadata": {
        "id": "3b734f9a"
      },
      "outputs": [],
      "source": [
        "df['day']=df['BKG_DATE'].dt.dayofweek"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e4d51582",
      "metadata": {
        "id": "e4d51582"
      },
      "outputs": [],
      "source": [
        "fcols = df.select_dtypes('float').columns\n",
        "icols = df.select_dtypes('integer').columns\n",
        "df[fcols] = df[fcols].apply(pd.to_numeric, downcast='float')\n",
        "df[icols] = df[icols].apply(pd.to_numeric, downcast='integer')\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "9c68e93f",
      "metadata": {
        "id": "9c68e93f",
        "outputId": "2097cc65-5257-45e3-ffb4-c1692da43097"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 1786374 entries, 0 to 1786373\n",
            "Data columns (total 33 columns):\n",
            " #   Column              Dtype         \n",
            "---  ------              -----         \n",
            " 0   CORP_ID             category      \n",
            " 1   REF_ORD_NO          object        \n",
            " 2   BKG_NO              object        \n",
            " 3   BKG_TYP             int8          \n",
            " 4   BKG_DATE            datetime64[ns]\n",
            " 5   BKG_TIME            int32         \n",
            " 6   SHPR_CD             int32         \n",
            " 7   INV_AMT             float32       \n",
            " 8   ITEM_SEQ            int8          \n",
            " 9   ITEM_CD             object        \n",
            " 10  BRAND_NM            object        \n",
            " 11  ITEM_QTY            int32         \n",
            " 12  ITEM_AMT            float32       \n",
            " 13  IF_YN               object        \n",
            " 14  ORDER_CRT_DATETIME  datetime64[ns]\n",
            " 15  DLV_DV              float32       \n",
            " 16  REF_ITEM_SEQ        object        \n",
            " 17  ORDER_IDX           object        \n",
            " 18  ORDER_YN            object        \n",
            " 19  DLVPREARRBRANCD     object        \n",
            " 20  DLVPREARREMPNICKNM  object        \n",
            " 21  DLVCLSFCD           object        \n",
            " 22  DLVSUBCLSFCD        object        \n",
            " 23  INS_ID              object        \n",
            " 24  INS_DATE            int32         \n",
            " 25  INS_TIME            int32         \n",
            " 26  POST_ZONE           float32       \n",
            " 27  SPLIT_EXEC_YN       object        \n",
            " 28  SHPR_ADDR_1         object        \n",
            " 29  SHPR_ADDR_2         object        \n",
            " 30  CNEE_ADDR_1         object        \n",
            " 31  CNEE_ADDR_2         object        \n",
            " 32  day                 category      \n",
            "dtypes: category(2), datetime64[ns](2), float32(4), int32(5), int8(2), object(18)\n",
            "memory usage: 340.7+ MB\n"
          ]
        }
      ],
      "source": [
        "df.info()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e7fe9d3e",
      "metadata": {
        "id": "e7fe9d3e"
      },
      "outputs": [],
      "source": [
        "\n",
        "df['day']=df['BKG_DATE'].dt.dayofweek"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3b564fc5",
      "metadata": {
        "id": "3b564fc5"
      },
      "outputs": [],
      "source": [
        "with open(pickle_path, 'wb') as fw:\n",
        "    pickle.dump(df, fw)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "6b573f03",
      "metadata": {
        "id": "6b573f03"
      },
      "outputs": [],
      "source": [
        "with open(pickle_path, 'rb') as fr:\n",
        "    df = pickle.load(fr)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b5961c7d",
      "metadata": {
        "id": "b5961c7d"
      },
      "source": [
        "# 날씨데이터 매핑"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "6105aae8",
      "metadata": {
        "id": "6105aae8"
      },
      "source": [
        "- 송화인 주소 통일"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "dd67b626",
      "metadata": {
        "id": "dd67b626"
      },
      "outputs": [],
      "source": [
        "dic = {'강원도':['강원'],\n",
        "'경기도':['\\t경기도', '(12730)', '(13279)', '(13499)', '(14782)', '[16817]', '경기', '경기도수원시', '경기도시흥시', '곤지암', '평택시', '화성시'],\n",
        "'경상남도':['(53298)', '경남', '경삼남도'],\n",
        "'경상북도':['경북', '상주시', '영천시', '태종로711번길'],\n",
        "'광주광역시':['광주', '중구'],\n",
        "'대구광역시':['대구', '대구시'],\n",
        "'대전광역시':['대전'],\n",
        "'부산광역시':['부산', '부산시'],\n",
        "'서울특별시':['서울시', '道路名稱首爾特別市中區麻將路1街22號（STUDIO', '강서구', '마포구', '서울', '영등포구', '울특별시', '\\u200b서울시'],\n",
        "'세종특별자치시':['(30100)', '세종', '세종시', '종시'],\n",
        "'울산광역시':['울산'],\n",
        "'인천광역시':['(22738)', '인천', '인천광역시 서구 정서진로 47', '인천광역시미추홀구', '인천시', '인천광역시\\xa0서구\\xa0정서진로\\xa047'],\n",
        "'전라남도':['전남', '전남목포시'],\n",
        "'전라북도':['전북'],\n",
        "'제주도':['제주', '제주특별자치도'],\n",
        "'충청남도':['청남도', '충남', '충남금산군'],\n",
        "'충청북도':['청주시', '충북']}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2b6f2f1d",
      "metadata": {
        "id": "2b6f2f1d"
      },
      "outputs": [],
      "source": [
        "#각 지역별 대분류로 매핑\n",
        "for key, value in dic.items():\n",
        "    n = len(value)\n",
        "    for i in range(n):\n",
        "        df.loc[df['CNEE_ADDR_1'] == value[i],'CNEE_ADDR_1'] = key"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "03a81011",
      "metadata": {
        "id": "03a81011"
      },
      "outputs": [],
      "source": [
        "# CNEE_ADDR_1가 NaN이면 CNEE_ADDR_2의 값을 참조하여 채움\n",
        "df.loc[(df[\"CNEE_ADDR_1\"].isnull()) & (df[\"CNEE_ADDR_2\"]=='경상남도'), 'CNEE_ADDR_1'] = '경상남도'\n",
        "df.loc[(df[\"CNEE_ADDR_1\"].isnull()) & (df[\"CNEE_ADDR_2\"]=='충남공주시'), 'CNEE_ADDR_1'] = '충청남도'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "691bffb5",
      "metadata": {
        "id": "691bffb5"
      },
      "outputs": [],
      "source": [
        "# CNEE_ADDR_1과 CNEE_ADDR_2가 모두 NaN이어서 알 수 없으므로 drop\n",
        "df.dropna(subset = ['CNEE_ADDR_1'], axis=0, inplace=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "24a0568b",
      "metadata": {
        "id": "24a0568b"
      },
      "source": [
        "- 날씨데이터 지점명 통일"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0f33a671",
      "metadata": {
        "id": "0f33a671"
      },
      "outputs": [],
      "source": [
        "df_weather=pd.read_csv('날씨.csv') #날씨 Raw Data 가져오기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3bc57368",
      "metadata": {
        "id": "3bc57368"
      },
      "outputs": [],
      "source": [
        "\n",
        "dic={'강원도':['강릉','대관령','동해','북강릉','북춘천','삼척','속초','영월','원주','인제','정선군','철원','춘천','태백','홍천'],\n",
        "'경기도':['동두천','수원','양평','이천','파주'],\n",
        "'경상남도':['거제','거창','김해시','남해','밀양','북창원','산청','양산시','의령군','진주','창원','통영','함양군','합천'],\n",
        "'경상북도':['경주시','구미','문경','봉화','상주','안동','영덕','영주','영천','울릉도','울진','의성','청송군','포항'],\n",
        "'광주광역시':['광주'],\n",
        "'대구광역시':['대구','대구(기)'],\n",
        "'대전광역시':['대전'],\n",
        "'부산광역시':['부산'],\n",
        "'서울특별시':['관악산','서울'],\n",
        "'세종특별자치시':['세종'],\n",
        "'울산광역시':['울산'],\n",
        "'인천광역시':['강화','백령도','인천'],\n",
        "'전라남도':['강진군','고흥','광양시','목포','무안','보성군','순천','여수','영광군','완도','장흥','주암','진도(첨찰산)','진도군','해남','흑산도'],\n",
        "'전라북도':['고창','고창군','군산','남원','부안','순창군','임실','장수','전주','정읍'],\n",
        "'제주도':['고산','서귀포','성산','성산','성산포','제주'],\n",
        "'충청남도':['금산','보령','부여','서산','천안','홍성'],\n",
        "'충청북도':['보은','제천','청주','추풍령','충주']}\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "1f049ab0",
      "metadata": {
        "id": "1f049ab0"
      },
      "outputs": [],
      "source": [
        "#각 지역별 대분류로 매핑\n",
        "for key,value in dic.items():\n",
        "    n=len(value)\n",
        "    for i in range(n):\n",
        "        df_weather.loc[df_weather['지점명']==value[i],'지점명']=key"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "99c63ee1",
      "metadata": {
        "id": "99c63ee1"
      },
      "outputs": [],
      "source": [
        "df_weather['지점명'].value_counts().sum()\n",
        "df1=df_weather.groupby(['지점명','일시']).mean().reset_index()\n",
        "df1.drop('지점',axis=1,inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a526ce9e",
      "metadata": {
        "id": "a526ce9e",
        "outputId": "f9ff827f-06a1-4656-eb3a-0c6c771f165f"
      },
      "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>지점명</th>\n",
              "      <th>일시</th>\n",
              "      <th>평균기온(°C)</th>\n",
              "      <th>일강수량(mm)</th>\n",
              "      <th>평균 풍속(m/s)</th>\n",
              "      <th>평균 상대습도(%)</th>\n",
              "      <th>평균 전운량(1/10)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-01</td>\n",
              "      <td>2.692857</td>\n",
              "      <td>68.000000</td>\n",
              "      <td>2.135714</td>\n",
              "      <td>91.292857</td>\n",
              "      <td>9.915385</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-02</td>\n",
              "      <td>-0.735714</td>\n",
              "      <td>3.653846</td>\n",
              "      <td>1.564286</td>\n",
              "      <td>83.457143</td>\n",
              "      <td>6.892308</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>0.857143</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.200000</td>\n",
              "      <td>73.328571</td>\n",
              "      <td>4.438462</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-04</td>\n",
              "      <td>4.028571</td>\n",
              "      <td>0.100000</td>\n",
              "      <td>1.007143</td>\n",
              "      <td>72.564286</td>\n",
              "      <td>7.484615</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-05</td>\n",
              "      <td>6.571429</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.035714</td>\n",
              "      <td>70.171429</td>\n",
              "      <td>2.346154</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>2069</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-26</td>\n",
              "      <td>21.160000</td>\n",
              "      <td>7.960000</td>\n",
              "      <td>0.940000</td>\n",
              "      <td>83.640000</td>\n",
              "      <td>8.420000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2070</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-27</td>\n",
              "      <td>23.200000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.440000</td>\n",
              "      <td>72.140000</td>\n",
              "      <td>6.220000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2071</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-28</td>\n",
              "      <td>22.880000</td>\n",
              "      <td>1.900000</td>\n",
              "      <td>1.620000</td>\n",
              "      <td>73.660000</td>\n",
              "      <td>6.920000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2072</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-29</td>\n",
              "      <td>22.860000</td>\n",
              "      <td>12.620000</td>\n",
              "      <td>1.200000</td>\n",
              "      <td>78.040000</td>\n",
              "      <td>6.220000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2073</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-30</td>\n",
              "      <td>23.240000</td>\n",
              "      <td>0.350000</td>\n",
              "      <td>1.040000</td>\n",
              "      <td>81.260000</td>\n",
              "      <td>6.980000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2074 rows × 7 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       지점명          일시   평균기온(°C)   일강수량(mm)  평균 풍속(m/s)  평균 상대습도(%)  \\\n",
              "0      강원도  2021-03-01   2.692857  68.000000    2.135714   91.292857   \n",
              "1      강원도  2021-03-02  -0.735714   3.653846    1.564286   83.457143   \n",
              "2      강원도  2021-03-03   0.857143        NaN    1.200000   73.328571   \n",
              "3      강원도  2021-03-04   4.028571   0.100000    1.007143   72.564286   \n",
              "4      강원도  2021-03-05   6.571429        NaN    1.035714   70.171429   \n",
              "...    ...         ...        ...        ...         ...         ...   \n",
              "2069  충청북도  2021-06-26  21.160000   7.960000    0.940000   83.640000   \n",
              "2070  충청북도  2021-06-27  23.200000        NaN    1.440000   72.140000   \n",
              "2071  충청북도  2021-06-28  22.880000   1.900000    1.620000   73.660000   \n",
              "2072  충청북도  2021-06-29  22.860000  12.620000    1.200000   78.040000   \n",
              "2073  충청북도  2021-06-30  23.240000   0.350000    1.040000   81.260000   \n",
              "\n",
              "      평균 전운량(1/10)  \n",
              "0         9.915385  \n",
              "1         6.892308  \n",
              "2         4.438462  \n",
              "3         7.484615  \n",
              "4         2.346154  \n",
              "...            ...  \n",
              "2069      8.420000  \n",
              "2070      6.220000  \n",
              "2071      6.920000  \n",
              "2072      6.220000  \n",
              "2073      6.980000  \n",
              "\n",
              "[2074 rows x 7 columns]"
            ]
          },
          "execution_count": 167,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4a9ef44d",
      "metadata": {
        "id": "4a9ef44d"
      },
      "outputs": [],
      "source": [
        "df1['일시']=pd.to_datetime(df1['일시'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "dccbfe61",
      "metadata": {
        "id": "dccbfe61",
        "outputId": "7b63225d-daf8-48d5-eaca-3e69d18585e6"
      },
      "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>지점명</th>\n",
              "      <th>일시</th>\n",
              "      <th>평균기온(°C)</th>\n",
              "      <th>일강수량(mm)</th>\n",
              "      <th>평균 풍속(m/s)</th>\n",
              "      <th>평균 상대습도(%)</th>\n",
              "      <th>평균 전운량(1/10)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-01</td>\n",
              "      <td>2.69</td>\n",
              "      <td>68.00</td>\n",
              "      <td>2.14</td>\n",
              "      <td>91.29</td>\n",
              "      <td>9.92</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-02</td>\n",
              "      <td>-0.74</td>\n",
              "      <td>3.65</td>\n",
              "      <td>1.56</td>\n",
              "      <td>83.46</td>\n",
              "      <td>6.89</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>0.86</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.20</td>\n",
              "      <td>73.33</td>\n",
              "      <td>4.44</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-04</td>\n",
              "      <td>4.03</td>\n",
              "      <td>0.10</td>\n",
              "      <td>1.01</td>\n",
              "      <td>72.56</td>\n",
              "      <td>7.48</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>강원도</td>\n",
              "      <td>2021-03-05</td>\n",
              "      <td>6.57</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.04</td>\n",
              "      <td>70.17</td>\n",
              "      <td>2.35</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>2069</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-26</td>\n",
              "      <td>21.16</td>\n",
              "      <td>7.96</td>\n",
              "      <td>0.94</td>\n",
              "      <td>83.64</td>\n",
              "      <td>8.42</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2070</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-27</td>\n",
              "      <td>23.20</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.44</td>\n",
              "      <td>72.14</td>\n",
              "      <td>6.22</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2071</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-28</td>\n",
              "      <td>22.88</td>\n",
              "      <td>1.90</td>\n",
              "      <td>1.62</td>\n",
              "      <td>73.66</td>\n",
              "      <td>6.92</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2072</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-29</td>\n",
              "      <td>22.86</td>\n",
              "      <td>12.62</td>\n",
              "      <td>1.20</td>\n",
              "      <td>78.04</td>\n",
              "      <td>6.22</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2073</th>\n",
              "      <td>충청북도</td>\n",
              "      <td>2021-06-30</td>\n",
              "      <td>23.24</td>\n",
              "      <td>0.35</td>\n",
              "      <td>1.04</td>\n",
              "      <td>81.26</td>\n",
              "      <td>6.98</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2074 rows × 7 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       지점명         일시  평균기온(°C)  일강수량(mm)  평균 풍속(m/s)  평균 상대습도(%)  \\\n",
              "0      강원도 2021-03-01      2.69     68.00        2.14       91.29   \n",
              "1      강원도 2021-03-02     -0.74      3.65        1.56       83.46   \n",
              "2      강원도 2021-03-03      0.86       NaN        1.20       73.33   \n",
              "3      강원도 2021-03-04      4.03      0.10        1.01       72.56   \n",
              "4      강원도 2021-03-05      6.57       NaN        1.04       70.17   \n",
              "...    ...        ...       ...       ...         ...         ...   \n",
              "2069  충청북도 2021-06-26     21.16      7.96        0.94       83.64   \n",
              "2070  충청북도 2021-06-27     23.20       NaN        1.44       72.14   \n",
              "2071  충청북도 2021-06-28     22.88      1.90        1.62       73.66   \n",
              "2072  충청북도 2021-06-29     22.86     12.62        1.20       78.04   \n",
              "2073  충청북도 2021-06-30     23.24      0.35        1.04       81.26   \n",
              "\n",
              "      평균 전운량(1/10)  \n",
              "0             9.92  \n",
              "1             6.89  \n",
              "2             4.44  \n",
              "3             7.48  \n",
              "4             2.35  \n",
              "...            ...  \n",
              "2069          8.42  \n",
              "2070          6.22  \n",
              "2071          6.92  \n",
              "2072          6.22  \n",
              "2073          6.98  \n",
              "\n",
              "[2074 rows x 7 columns]"
            ]
          },
          "execution_count": 169,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# 날씨데이터 수치 소수점 둘 째 자리까지 반올림\n",
        "df1[['평균기온(°C)','일강수량(mm)','평균 풍속(m/s)','평균 상대습도(%)','평균 전운량(1/10)']]=\\\n",
        "df1[['평균기온(°C)','일강수량(mm)','평균 풍속(m/s)','평균 상대습도(%)','평균 전운량(1/10)']].round(2)\n",
        "df1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5f837ec2",
      "metadata": {
        "id": "5f837ec2"
      },
      "outputs": [],
      "source": [
        "#이커머스 데이터와 날씨 데이터 JOIN\n",
        "df_wea_order=pd.merge(df,df1,left_on=['BKG_DATE','CNEE_ADDR_1'],right_on=['일시','지점명'])"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "1acfd138",
      "metadata": {
        "id": "1acfd138"
      },
      "source": [
        "# 컬럼 정제"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0f45d05e",
      "metadata": {
        "id": "0f45d05e"
      },
      "outputs": [],
      "source": [
        "#결측치 0값으로 대체 \n",
        "df_wea_order['일강수량(mm)'].fillna(value=0,inplace=True)\n",
        "df_wea_order['평균 전운량(1/10)'].fillna(value=0,inplace=True)\n",
        "#'ITEM_CD' 일련화               \n",
        "df_wea_order['ITEM_CD'].replace('_\\d','',regex=True,inplace=True)\n",
        "df_wea_order['ITEM_CD'].replace('-\\d','',regex=True,inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "29e65f79",
      "metadata": {
        "id": "29e65f79"
      },
      "outputs": [],
      "source": [
        "df_wea_order=df_wea_order[df_wea_order['BKG_TYP']==7]\n",
        "df_wea_order.drop(\n",
        "    columns=['REF_ITEM_SEQ','SHPR_ADDR_2','CNEE_ADDR_2','지점명','일시',\\\n",
        "             'ITEM_AMT','INV_AMT','BKG_TYP','BKG_TIME','ITEM_SEQ',\\\n",
        "             'BRAND_NM','IF_YN', 'ORDER_CRT_DATETIME','DLV_DV','ORDER_YN',\\\n",
        "             'DLVPREARRBRANCD', 'DLVPREARREMPNICKNM','DLVSUBCLSFCD','INS_ID','INS_DATE', \\\n",
        "             'INS_TIME','SPLIT_EXEC_YN','BKG_NO','REF_ORD_NO','DLVCLSFCD','POST_ZONE'] \\\n",
        "              , axis=1,inplace=True)\n",
        "df_wea_order.reset_index(drop=True,inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2754d326",
      "metadata": {
        "id": "2754d326"
      },
      "outputs": [],
      "source": [
        "df_wea_order.rename(columns={'평균기온(°C)':'평균기온'\\\n",
        "                             ,'일강수량(mm)':'평균강수량'\\\n",
        "                             ,'평균 풍속(m/s)':'평균풍속'\\\n",
        "                             ,'평균 상대습도(%)':'평균습도'\\\n",
        "                             ,'평균 전운량(1/10)':'평균전운량'},inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "433e7e8c",
      "metadata": {
        "id": "433e7e8c"
      },
      "outputs": [],
      "source": [
        "def 체감온도(T,V):\n",
        "    V=V*3.6\n",
        "    return round(13.12 + 0.6215*T - 11.37 *V**0.16 + 0.3965 *V**0.16*T)\n",
        "\n",
        "def 열지수(df):\n",
        "    T=df.평균기온\n",
        "    A=df.평균습도\n",
        "    import math \n",
        "    F=1.8*T+32\n",
        "    #화씨 80 이하 \n",
        "    if (F<80):\n",
        "        hi=0.5 * (F + 61.0 + (F-68.0)*1.2 + (A*0.094))\n",
        "      #상대습도 85 이상 , 온도 80~87\n",
        "    elif (A>=85) & ((F>=80) & (F<87)):\n",
        "        ad=((A-85)/10) * ((87-F)/5)\n",
        "        hi=-42.379+2.049*F+10.143*A-0.225*F*A-0.007*(F**2)-0.055*(A**2)+0.001*(F**2)*A+0.0009*F*(A**2)-0.000002*(F**2)*(A**2)\n",
        "        hi=hi+ad\n",
        "    elif (A<=13) & ((F>=80) & (F<112)): \n",
        "      #상대습도가 13이하 , 화씨 80~112\n",
        "        ad=((13-A)/4)*math.sqrt((17-abs(F-95))/17)\n",
        "        hi=-42.379+2.049*F+10.143*A-0.225*F*A-0.007*(F**2)-0.055*(A**2)+0.001*(F**2)*A+0.0009*F*(A**2)-0.000002*(F**2)*(A**2)\n",
        "        hi=hi-ad\n",
        "    else:\n",
        "        hi=-42.379+2.049*F+10.143*A-0.225*F*A-0.007*(F**2)-0.055*(A**2)+0.001*(F**2)*A+0.0009*F*(A**2)-0.000002*(F**2)*(A**2)\n",
        "    heat_hi = (5/9)*(hi-32)\n",
        "  \n",
        "    return round(heat_hi,1)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "53ffdc89",
      "metadata": {
        "id": "53ffdc89",
        "outputId": "7d5246a7-8e89-4ed7-ac5b-636e9730b550"
      },
      "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>CORP_ID</th>\n",
              "      <th>BKG_DATE</th>\n",
              "      <th>SHPR_CD</th>\n",
              "      <th>ITEM_CD</th>\n",
              "      <th>ITEM_QTY</th>\n",
              "      <th>ORDER_IDX</th>\n",
              "      <th>SHPR_ADDR_1</th>\n",
              "      <th>CNEE_ADDR_1</th>\n",
              "      <th>day</th>\n",
              "      <th>평균기온</th>\n",
              "      <th>평균강수량</th>\n",
              "      <th>평균풍속</th>\n",
              "      <th>평균습도</th>\n",
              "      <th>평균전운량</th>\n",
              "      <th>날씨지수</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>90001302</td>\n",
              "      <td>134265392</td>\n",
              "      <td>1</td>\n",
              "      <td>2021030323885041</td>\n",
              "      <td>세종특별자치시</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>2</td>\n",
              "      <td>4.4</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.8</td>\n",
              "      <td>64.4</td>\n",
              "      <td>4.5</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>90001542</td>\n",
              "      <td>3858391637</td>\n",
              "      <td>1</td>\n",
              "      <td>2021030323895281</td>\n",
              "      <td>경기도</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>2</td>\n",
              "      <td>4.4</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.8</td>\n",
              "      <td>64.4</td>\n",
              "      <td>4.5</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>90001341</td>\n",
              "      <td>5161458796</td>\n",
              "      <td>1</td>\n",
              "      <td>2021030323912641</td>\n",
              "      <td>경기도</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>2</td>\n",
              "      <td>4.4</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.8</td>\n",
              "      <td>64.4</td>\n",
              "      <td>4.5</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>90001441</td>\n",
              "      <td>4624200932</td>\n",
              "      <td>1</td>\n",
              "      <td>2021030323949931</td>\n",
              "      <td>경기도</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>2</td>\n",
              "      <td>4.4</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.8</td>\n",
              "      <td>64.4</td>\n",
              "      <td>4.5</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>90001541</td>\n",
              "      <td>4929464978</td>\n",
              "      <td>1</td>\n",
              "      <td>2021030323878091</td>\n",
              "      <td>경기도</td>\n",
              "      <td>서울특별시</td>\n",
              "      <td>2</td>\n",
              "      <td>4.4</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.8</td>\n",
              "      <td>64.4</td>\n",
              "      <td>4.5</td>\n",
              "      <td>3.0</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",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1785320</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-06-29</td>\n",
              "      <td>90001622</td>\n",
              "      <td>5528524517</td>\n",
              "      <td>1</td>\n",
              "      <td>2021062911883661</td>\n",
              "      <td>경기도</td>\n",
              "      <td>세종특별자치시</td>\n",
              "      <td>1</td>\n",
              "      <td>23.2</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>5.1</td>\n",
              "      <td>23.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1785321</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-06-29</td>\n",
              "      <td>90001622</td>\n",
              "      <td>55194709829</td>\n",
              "      <td>1</td>\n",
              "      <td>2021062913232751</td>\n",
              "      <td>경기도</td>\n",
              "      <td>세종특별자치시</td>\n",
              "      <td>1</td>\n",
              "      <td>23.2</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>5.1</td>\n",
              "      <td>23.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1785322</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-06-29</td>\n",
              "      <td>90001682</td>\n",
              "      <td>558225707043</td>\n",
              "      <td>1</td>\n",
              "      <td>2021062976202651</td>\n",
              "      <td>경기도</td>\n",
              "      <td>세종특별자치시</td>\n",
              "      <td>1</td>\n",
              "      <td>23.2</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>5.1</td>\n",
              "      <td>23.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1785323</th>\n",
              "      <td>KX007</td>\n",
              "      <td>2021-06-29</td>\n",
              "      <td>90001341</td>\n",
              "      <td>5256729806</td>\n",
              "      <td>1</td>\n",
              "      <td>2021062910638861</td>\n",
              "      <td>경기도</td>\n",
              "      <td>세종특별자치시</td>\n",
              "      <td>1</td>\n",
              "      <td>23.2</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>5.1</td>\n",
              "      <td>23.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1785324</th>\n",
              "      <td>GP001</td>\n",
              "      <td>2021-06-29</td>\n",
              "      <td>90001765</td>\n",
              "      <td>4772445613</td>\n",
              "      <td>2</td>\n",
              "      <td>2021062982271251</td>\n",
              "      <td>경기도</td>\n",
              "      <td>세종특별자치시</td>\n",
              "      <td>1</td>\n",
              "      <td>23.2</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>5.1</td>\n",
              "      <td>23.7</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1785325 rows × 15 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "        CORP_ID   BKG_DATE   SHPR_CD       ITEM_CD  ITEM_QTY  \\\n",
              "0         KX007 2021-03-03  90001302     134265392         1   \n",
              "1         KX007 2021-03-03  90001542    3858391637         1   \n",
              "2         KX007 2021-03-03  90001341    5161458796         1   \n",
              "3         KX007 2021-03-03  90001441    4624200932         1   \n",
              "4         KX007 2021-03-03  90001541    4929464978         1   \n",
              "...         ...        ...       ...           ...       ...   \n",
              "1785320   KX007 2021-06-29  90001622    5528524517         1   \n",
              "1785321   KX007 2021-06-29  90001622   55194709829         1   \n",
              "1785322   KX007 2021-06-29  90001682  558225707043         1   \n",
              "1785323   KX007 2021-06-29  90001341    5256729806         1   \n",
              "1785324   GP001 2021-06-29  90001765    4772445613         2   \n",
              "\n",
              "                ORDER_IDX SHPR_ADDR_1 CNEE_ADDR_1 day  평균기온  평균강수량  평균풍속  \\\n",
              "0        2021030323885041     세종특별자치시       서울특별시   2   4.4    0.0   1.8   \n",
              "1        2021030323895281         경기도       서울특별시   2   4.4    0.0   1.8   \n",
              "2        2021030323912641         경기도       서울특별시   2   4.4    0.0   1.8   \n",
              "3        2021030323949931         경기도       서울특별시   2   4.4    0.0   1.8   \n",
              "4        2021030323878091         경기도       서울특별시   2   4.4    0.0   1.8   \n",
              "...                   ...         ...         ...  ..   ...    ...   ...   \n",
              "1785320  2021062911883661         경기도     세종특별자치시   1  23.2    3.1   1.0   \n",
              "1785321  2021062913232751         경기도     세종특별자치시   1  23.2    3.1   1.0   \n",
              "1785322  2021062976202651         경기도     세종특별자치시   1  23.2    3.1   1.0   \n",
              "1785323  2021062910638861         경기도     세종특별자치시   1  23.2    3.1   1.0   \n",
              "1785324  2021062982271251         경기도     세종특별자치시   1  23.2    3.1   1.0   \n",
              "\n",
              "         평균습도  평균전운량  날씨지수  \n",
              "0        64.4    4.5   3.0  \n",
              "1        64.4    4.5   3.0  \n",
              "2        64.4    4.5   3.0  \n",
              "3        64.4    4.5   3.0  \n",
              "4        64.4    4.5   3.0  \n",
              "...       ...    ...   ...  \n",
              "1785320  80.0    5.1  23.7  \n",
              "1785321  80.0    5.1  23.7  \n",
              "1785322  80.0    5.1  23.7  \n",
              "1785323  80.0    5.1  23.7  \n",
              "1785324  80.0    5.1  23.7  \n",
              "\n",
              "[1785325 rows x 15 columns]"
            ]
          },
          "execution_count": 179,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df_wea_order['날씨지수'] = np.where((df_wea_order['평균기온']<=10) & (df_wea_order['평균풍속']>=1.3),\\\n",
        "                        체감온도(df_wea_order['평균기온'],df_wea_order['평균풍속']),df_wea_order.apply(열지수,axis=1))\n",
        "df_wea_order"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e78efb0d",
      "metadata": {
        "id": "e78efb0d"
      },
      "outputs": [],
      "source": [
        "df_wea_order['날씨지수'].replace(-18.1,28,inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ffb2fdfc",
      "metadata": {
        "id": "ffb2fdfc",
        "outputId": "d8aadcd2-8a45-4fef-99fc-40a341b35bdc"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 1785325 entries, 0 to 1785324\n",
            "Data columns (total 15 columns):\n",
            " #   Column       Dtype         \n",
            "---  ------       -----         \n",
            " 0   CORP_ID      category      \n",
            " 1   BKG_DATE     datetime64[ns]\n",
            " 2   SHPR_CD      int32         \n",
            " 3   ITEM_CD      object        \n",
            " 4   ITEM_QTY     int32         \n",
            " 5   ORDER_IDX    object        \n",
            " 6   SHPR_ADDR_1  object        \n",
            " 7   CNEE_ADDR_1  object        \n",
            " 8   day          category      \n",
            " 9   평균기온         float64       \n",
            " 10  평균강수량        float64       \n",
            " 11  평균풍속         float64       \n",
            " 12  평균습도         float64       \n",
            " 13  평균전운량        float64       \n",
            " 14  날씨지수         float64       \n",
            "dtypes: category(2), datetime64[ns](1), float64(6), int32(2), object(4)\n",
            "memory usage: 166.9+ MB\n"
          ]
        }
      ],
      "source": [
        "df_wea_order.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "0aa53370",
      "metadata": {
        "id": "0aa53370"
      },
      "source": [
        "# 요인분석 및 SOM"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "583a701e",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 205
        },
        "id": "583a701e",
        "outputId": "d451fbcd-b0e8-4290-b5fc-8b758a652985"
      },
      "outputs": [
        {
          "ename": "NameError",
          "evalue": "ignored",
          "output_type": "error",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-1-63d66767e4f9>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mdf_wea_order\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mdf_org2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_excel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'NAVER_DATERLAB_CATEGORY_1.xlsx'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mdf2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdf_org2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'df_wea_order' is not defined"
          ]
        }
      ],
      "source": [
        "df_wea_order\n",
        "df_org2 = pd.read_excel('NAVER_DATERLAB_CATEGORY_1.xlsx')\n",
        "df2 = df_org2.copy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "33c29bf4",
      "metadata": {
        "id": "33c29bf4",
        "outputId": "eb277f10-bdd7-40b5-80b6-a794b8936e78"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<AxesSubplot:>"
            ]
          },
          "execution_count": 186,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "sns.heatmap(data = df2.corr(), annot=True, \n",
        "fmt = '.3f', linewidths=.5, cmap='Blues')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "483385e8",
      "metadata": {
        "id": "483385e8",
        "outputId": "57953172-004e-4c01-e4dd-1f154e018459"
      },
      "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>화장품/미용</th>\n",
              "      <th>출산/육아</th>\n",
              "      <th>생활/건강</th>\n",
              "      <th>식품</th>\n",
              "      <th>패션의류</th>\n",
              "      <th>패션잡화</th>\n",
              "      <th>디지털/가전</th>\n",
              "      <th>가구/인테리어</th>\n",
              "      <th>스포츠/레저</th>\n",
              "      <th>여가/생활편의</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>날짜</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2021-03-01</th>\n",
              "      <td>21.0</td>\n",
              "      <td>55.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>39.0</td>\n",
              "      <td>86.0</td>\n",
              "      <td>83.0</td>\n",
              "      <td>75.0</td>\n",
              "      <td>51.0</td>\n",
              "      <td>45.0</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-02</th>\n",
              "      <td>19.0</td>\n",
              "      <td>49.0</td>\n",
              "      <td>91.0</td>\n",
              "      <td>37.0</td>\n",
              "      <td>61.0</td>\n",
              "      <td>62.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>39.0</td>\n",
              "      <td>36.0</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-03</th>\n",
              "      <td>18.0</td>\n",
              "      <td>47.0</td>\n",
              "      <td>87.0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>61.0</td>\n",
              "      <td>59.0</td>\n",
              "      <td>63.0</td>\n",
              "      <td>36.0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-04</th>\n",
              "      <td>18.0</td>\n",
              "      <td>47.0</td>\n",
              "      <td>85.0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>63.0</td>\n",
              "      <td>59.0</td>\n",
              "      <td>63.0</td>\n",
              "      <td>36.0</td>\n",
              "      <td>34.0</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-05</th>\n",
              "      <td>15.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>74.0</td>\n",
              "      <td>29.0</td>\n",
              "      <td>58.0</td>\n",
              "      <td>54.0</td>\n",
              "      <td>56.0</td>\n",
              "      <td>31.0</td>\n",
              "      <td>31.0</td>\n",
              "      <td>3.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            화장품/미용  출산/육아  생활/건강    식품  패션의류  패션잡화  디지털/가전  가구/인테리어  스포츠/레저  \\\n",
              "날짜                                                                            \n",
              "2021-03-01    21.0   55.0  100.0  39.0  86.0  83.0    75.0     51.0    45.0   \n",
              "2021-03-02    19.0   49.0   91.0  37.0  61.0  62.0    67.0     39.0    36.0   \n",
              "2021-03-03    18.0   47.0   87.0  35.0  61.0  59.0    63.0     36.0    35.0   \n",
              "2021-03-04    18.0   47.0   85.0  35.0  63.0  59.0    63.0     36.0    34.0   \n",
              "2021-03-05    15.0   40.0   74.0  29.0  58.0  54.0    56.0     31.0    31.0   \n",
              "\n",
              "            여가/생활편의  \n",
              "날짜                   \n",
              "2021-03-01      3.0  \n",
              "2021-03-02      3.0  \n",
              "2021-03-03      3.0  \n",
              "2021-03-04      3.0  \n",
              "2021-03-05      3.0  "
            ]
          },
          "execution_count": 187,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df2.set_index('날짜', inplace = True)\n",
        "df2.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3287f679",
      "metadata": {
        "id": "3287f679",
        "outputId": "236ff819-de21-4f39-bcb7-9d127d81b129"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(1700.6342003446177, 0.0)"
            ]
          },
          "execution_count": 188,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "chi_square_value, p_value = calculate_bartlett_sphericity(df2)\n",
        "chi_square_value, p_value"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "23783b9a",
      "metadata": {
        "id": "23783b9a",
        "outputId": "977a3753-df75-4140-94c4-dd66ac74a255"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "0.758555876916214"
            ]
          },
          "execution_count": 189,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "#KMO 검정 : 0.75로 우수한 데이터 \n",
        "from factor_analyzer.factor_analyzer import calculate_kmo\n",
        "kmo_all, kmo_model = calculate_kmo(df2)\n",
        "kmo_model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "43b4fba7",
      "metadata": {
        "id": "43b4fba7",
        "outputId": "7b7bd052-71b4-4f9e-d148-ff5637b9f398"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "FactorAnalyzer(n_factors=10, rotation=None, rotation_kwargs={})"
            ]
          },
          "execution_count": 190,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "#요인 수 선택\n",
        "fa = FactorAnalyzer(n_factors=10, rotation=None)\n",
        "fa.fit(df2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "db5f9806",
      "metadata": {
        "id": "db5f9806",
        "outputId": "245d668e-976f-43b4-8e55-d0d554b0b0a4"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array([5.76570573, 2.00204142, 1.25864029, 0.42098425, 0.20713398,\n",
              "       0.12415287, 0.10754132, 0.06707181, 0.03282592, 0.01390241])"
            ]
          },
          "execution_count": 191,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "ev, v = fa.get_eigenvalues()\n",
        "ev"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "496cfcc6",
      "metadata": {
        "id": "496cfcc6",
        "outputId": "5eae0ee6-140b-48b3-ee40-d972eb590dac"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAETCAYAAAAiZy2MAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAAlbUlEQVR4nO3deXxV9Z3/8dfn3uwJZIUoAQQSxA0ViSKJWrDTYl3R+ms7dXTaaaWdbmNtcXA6tp1HF62Mnda6VLo5jrWp0ypjq5XaQlwCLiAoUkVBEQhuBAIkZL/f3x/nZkMgNyQn5y7v5+ORR+4999xzPvkq73vu93zP95hzDhERSR6hoAsQEZHhpWAXEUkyCnYRkSSjYBcRSTIKdhGRJKNgFxFJMgp2ER+Y2bfN7Jqg65DUpGCXhGRmF5vZY2a23MzWmdmXA6hhjpntNLNaM3vSzB4xsylHsJ1vmVmuHzVKakoLugCRwTKz8cD3gVnOuebosvxBvN/c8F2Z95Rzbn50uxcDPwM+OMhtfBr48TDVI6IjdklIuUAmYN0LnHN7AMwsZGbXmdlT0aPoJdHltWZ2jZmtAK4yszwz+5mZ/cXMVpnZ1d3bMrN50W8Cy83sj2Z2dIx1rQDKD1xoZmlm9p1oPU+Y2R/MbGL0tT8CRwF/NLNFR9geIv3oiF0SjnNuo5ndAaw1sx8DS5xz7dGXvwicAsx1znUc0MUxwTk3F8DMfgo86Jx7xMzSgVVm9hjeh8W1wPnOuVYz+3/Ad4HPHK4mMwsD/wI8cJCXvw6UAh9wzkXM7BLg18DZzrkLzWwLcKFzrvEImkPkfRTskpCcc/9lZr8FrgNeMLOPOefWA58A/sk51xFdr7nP25b2eXwJcJyZXRd9ngVMBo4HKoBHzQy8fyMNhynlLDOrBbqAZcC/HmSdS4ErnHORaE3/Z2a3mtko59y+QfzZIjFRsEvCcs7tAK4xswuAnwLVQA7QcYi39A3RNGCec66t7wpmNh2occ59I8YyevrYDyMMRA4s/yDLRIaF+tgl4ZjZ0WY2ps+ivUD3kfnDwNfNLBRdt+AQm1mO13XSvc3K6MO/Ap8ws7HR5aPMbNoQS34oWpNFt3khsKHPt4kWYPQQ9yHSQ0fskoiKgP82sy5gD16ofy762neAHwBPm9l+4Fm87poDfQW4y8yeAdqAF4HVzrkNZvYfwDIz24d3VH2w7pXBuAlvFM8qM2sG3gI+1ef1u4CHzewe59ziIe5LBNN87CIiyUVdMSIiSUbBLiKSZBTsIiJJRsEuIpJkfBsVY2ZnAP+JN4b3/5xzNx9q3ZKSEjdp0iS/ShkRzc3N5OZqHidQWxxI7dGf2qPXUNtizZo1O51zYw5c7kuwRy/R/iZwiXNu90DrT5o0idWrV/tRyoipra1lzpw5QZcRF9QW/ak9+lN79BpqW5jZmwdd7sdwx+gsd/PwJkRKBxY6554/YJ0FwAKA0tLSmTU1NcNex0hqamoiLy8v6DLigtqiP7VHf2qPXkNti7lz565xzlUeuNyvrpipeBeRXAiMB34DzO67gnNuCbAEoLKy0iX6J7iOQnqpLfpTe/Sn9ujlV1v4dfK0E/izc67TObcFiHRfTi0iIv7yK9hXAR8BMLNSoGMYb2wgIiKH4UtXjHPuWTPbaGZ1eEfv1/qxHxEReT/fhjs6524AbvBr+0vX1rN42UZ2NLYwriCbhfOmMX9GmV+7ExFJGAk5u+PStfVc/8B6Wjq6AKhvbOH6B9YDKNxFJOUl5JWni5dt7An1bi0dXSxetjGgikRE4kdCBvuOxpZBLRcRSSUJGezjCrIHtVxEJJUkZLAvnDeN7PRwv2XZ6WEWzhvqHcxERBJfQp487T5B+oNHX+GtPa3kZabx3fkn6cSpiAgJesQOXrivuv6DnDmliGOKcxTqIiJRCRvs3arLS/jbW3vZ3dwedCkiInEh4YO9qqIE52DV6w1BlyIiEhcSPthPGZ9PXmYadZt2Bl2KiEhcSPhgTwuHmDW5iJWbdcQuIgJJEOwAs8uLeWNnsy5QEhEhSYK9uqIEQN0xIiIkSbBPKx1FcW6GumNEREiSYA+FjNnlxdRt2onu5yEiqS4pgh287ph397Wx+b2moEsREQlU8gR7udfPru4YEUl1SRPsE4tzGF+YrROoIpLykibYwTtqX7W5ga6I+tlFJHUlVbBXVRSzt7WTDTv2BF2KiEhgkivYy7vHs6ufXURSV1IF+5hRmUwrHcXKzepnF5HUlVTBDl53zHNbdtHW2TXwyiIiSSjpgr26vITWjgjPv9kYdCkiIoFIumCfNaWIcMjUHSMiKSvpgn1UVjonj8/XeHYRSVm+BbuZrTez2ujPJ/3az8FUlRfzwvY97GvtGMndiojEBT+P2N9xzs2J/tzn437ep7q8hK6I49k3do3kbkVE4oKfwR7xcduHddoxhWSmhTSeXURSkvkxza2Z5QIvAvXA28DXnHPbDlhnAbAAoLS0dGZNTc2w1nDzcy3sbXN896ycYd3uoTQ1NZGXlzci+4p3aov+1B79qT16DbUt5s6du8Y5V/m+F5xzvv4AHwLuP9w6M2fOdMPttuWvuWP+9Y/uvX2tw77tg1mxYsWI7CcRqC36U3v0p/boNdS2AFa7g2SqL10xZhbu8/Q9P/YxkO7b5a3SNL4ikmLSfNpuhZn9EmiP/vyzT/s5pOll+YzKSmPl5p1cdMq4kd69iEhgfAl259xGoNqPbccqHDLOnFKsE6giknKS7gKlvqrLi9m6az/bdu0PuhQRkRGT3MFe0X27PF2FKiKpI6mDvWJsHmNHZao7RkRSSlIHu5lRVV7Mys0N3UMvRUSSXlIHO0BVRQk7m9p49Z2moEsRERkRyR/s5cUAmu1RRFJG0gf7+MIcjinO0QlUEUkZSR/s4N3k+pnXd9HZFdi8ZCIiIyYlgr26oph9bZ28WL8n6FJERHyXEsE+e4rXz75S/ewikgJSItiL8zI5/ujRGs8uIikhJYIdvOkF1mzdTWtHV9CliIj4KnWCvaKE9s4Ia97cHXQpIiK+SplgP2NyEWkh03h2EUl6KRPsuZlpnDqhgDrdeENEklzKBDt40wus397InpaOoEsREfFNSgV7dXkxEQfPvK6jdhFJXikV7DMmFpKdHmalumNEJImlVLBnpIU4fXKRTqCKSFJLqWAHrzvmtXebeHdva9CliIj4IvWCved2eeqOEZHklHLBfvzRo8nPTld3jIgkrZQL9nDImD1Ft8sTkeSVcsEO3jS+9Y0tvNmwP+hSRESGXUoGe1W0n71Od1USkSSUksE+pSSXo0Zn6QSqiCQlX4PdzJ43s/P83MeRMDOqKopZtbmBSET97CKSXHwLdjO7HMj3a/tDVV1ewq7mdl55e1/QpYiIDCtfgt3MRgFXAr/2Y/vDoXc8u/rZRSS5mB9D/szsV8AdwAXA0865Rw+yzgJgAUBpaenMmpqaYa9jIIue3M/YnBDXzswa8raamprIy8sbhqoSn9qiP7VHf2qPXkNti7lz565xzlUeuDxtSFUdhJldAWx1zj1nZhccaj3n3BJgCUBlZaWbM2fOcJcyoA81vsQDz2+n+uxzSA8P7ctLbW0tQfwN8Uht0Z/aoz+1Ry+/2sKPrphPAieYWQ1wObDIzKb5sJ8hq64oprm9ixe2NQZdiojIsBn2I3bnXM9Rupl9G68rZuNw72c4nDmlGDOo29RA5aSioMsRERkWvg53dM59+2D96/GiICeDk8bl60IlEUkqKXmBUl9VFcWs3bqb/e2dQZciIjIsFOzlJXR0OZ7bsjvoUkREhkXKB/vpkwpJDxsrNY2viCSJlA/2nIw0ZkwsVD+7iCSNlA928KYX2LBjL43724MuRURkyBTseOPZnYOnX9dsjyKS+BTswCkTCsjNCFO3ScEuIolPwQ6kh0OcMblI/ewikhQU7FHVFSW8/l4zb+9pDboUEZEhUbBHVZVHb5enYY8ikuAU7FHHHTWKotwMdceISMKLOdjNLMPMJvhZTJBCIWN2eTErNzXgxxz1IiIjJaZgN7N/BB4BHjazLDP7D3/LCkZ1eQlv723l9Z3NQZciInLEYj1i/6xz7u+ABudcKzDLx5oCU11RDKDpBUQkocUa7M7MwkB3H0Xc3qR6KCYW5VBWkK3x7CKS0GIN9sXAn4EKM/sD8Av/SgqOmVFdUcyq1xvoiqifXUQSU0zB7pz7A3AZ8FHg0865n/taVYCqykvY09LB33bsDboUEZEjEtOt8czsqj5PjzcznHP3+FRToKrKvX72us07mT4+KXucRCTJxdoVk93nZzpwnm8VBWzs6Cymjs3ThUoikrBiOmJ3zt3V97mZfcOfcuJDdUUJNc9tpa2zi8y0cNDliIgMyqCvPDWzTLyj9qRVVV5Ma0eEdVsbgy5FRGTQYu1jX4U31NGATuAWP4sK2qwpxYQM6jY3MGtKcdDliIgMSqxdMbP9LiSe5GenM318ASs37eTaDx0bdDkiIoNy2GA3sxvpvSipH+fcv/lSUZyoLi9myROv09zWSW5mTJ9/IiJxYaDEenREqohD1RUl3FG7mWff2MXc48YGXY6ISMwOG+zOuce7H5vZ0cAo3yuKEzOPKSQjLUTdpp0KdhFJKLGePP05MAkojC7aDlziU01xISs9TOUxhdRt1rwxIpJYYh3ueGx0dsdlwBnAYa+3j87d/gczqzWzx82sbKiFBqG6ooSX39pLQ1Nb0KWIiMQs1mDvvhFoLt7J1JMHWL8T+Lhzbg7wM+Afj6i6gHVPL7DqdR21i0jiiHW4x/VmVoh3MnUl8OvDreyciwD7o0+nAqsPXMfMFgALAEpLS6mtrY2xlJHTFXFkp8HvnlhP3q5XD7tuU1NTXP4NQVBb9Kf26E/t0cu3tnDODfiDN03vzFjW7fOehcBrwMNA7uHWnTlzpotXn7n7WXfOzcsHXG/FihX+F5Mg1Bb9qT36U3v0GmpbAKvdQTI11q6YW4ErzexhM/tUdFqBgT4wFjvnpgK3AbcfwWdOXKgqL+HNhv1s371/4JVFROJArPOxv+CcuwZvJMwEYNPh1jezUWZm0adbgbyhFBmk6ooSAFbqrkoikiBivZn1ODNbiHcXpSLg/AHechzwlJktB27G65ZJSMeW5lGSl0ndZk3jKyKJIdaTp78Afgmc55xrH2hl59xzQPVQCosXZkZVeTErNzfgnKP3i4iISHyKtSvmI8DfgFlmdo6ZneNvWfGluqKY9/a1sendpqBLEREZUKxXnv4abwz7K3jj2B3whI91xZWqcq+fvW7TTqaWpsysCiKSoGLtipniUmzq3r4mFOUwsSiHus0NfKp6ctDliIgcVqzDHdeaWUrfcaK6opinX2+gsysSdCkiIocVa7CfDmwys1XRn5V+FhWPqspL2NfayUs7DjtNjohI4GK9g9LpfhcS77rnjanbtJNTJxQEW4yIyGHEOo69wMy+YWb/aWaZZnaC34XFm+K8TI47ahQrNZ5dROJcrF0x9wDPA2c459qAG/0rKX5VV5SwestuWju6gi5FROSQYg32HOfcn/Cm44UUupNSX9UVxbR1Rnj+zd1BlyIickixBvs7ZnYxEDazaqDFx5ri1hmTiwmHTNMLiEhcizXYF+CNjGkCPgp8yq+C4lleZhqnjM+nThOCiUgci3VUTDNwg8+1JITqihJuX7GJva0djM5KD7ocEZH3iXVUzHNm9rqZPW1mW82szsx+b2aTfK4v7lSVlxBx8Mzru4IuRUTkoGLtinkOOMc5dybwYaAO+B5wh1+FxavTjikgKz1E3Sb1s4tIfIo12Kc557YDOOdewRv2+DyQ5VtlcSozLczpk4pYtVn97CISn2IN9j1m9vdmlm9mFwHdk5Kn5OTkVeUlbHxnH+/tawu6FBGR94k12D8NnAjcB3wEuMLM0oDv+1VYPKuu8KYX0FWoIhKPYh0Vswf494O89NjwlpMYThyXz+isNFZuauCSU8uCLkdEpJ/DBruZ3eycu87MVuHdXKOHc67K18riWDhkzC4v1oVKIhKXDhvszrnror9T9iYbh1JdUcKyDe+wtWE/E4tzgi5HRKTHYfvYzexrfR5f2OfxbX4WlQh6bpeno3YRiTMDnTy9oM/ja/s8PtGHWhJK+ZhcSkdnajy7iMSdWEfFHMgNvEpyMzOqy0tYtbmBSCTlm0NE4shAo2JmRm+DZ8AJfR4f73tlCaCqooQH1taz8Z19QZciItJjoJOn+SNVSCLqe7u8ioBrERHpdqRdMQKMK8hmckkuKzW9gIjEEV+CPXqP1BozqzWzJ8xssh/7iQdH52ex4pV3Wbd9D9U3LWfp2vqgSxKRFOfXEXsOcK1zbg7wA+DrPu0nUEvX1rN6y24c8PZ+qG9s4foH1ivcRSRQvgS7c26Hc25H9OluoNmP/QRt8bKNtHdFAHji7TAALR1dLF62MciyRCTFmXP+DdUzszLgJ8CX+gR992sL8G65R2lp6cyamhrf6vDL+vo9ALzSaDxWHyJscP6ECJNGOaaXpe5556amJvLy8oIuI26oPfpTe/QaalvMnTt3jXOu8sDlvgV79ErVi4B/c84d9uxiZWWlW716tS91+Kn6puXUN3r39f7U1E7ufs0bZDQqM4113/ow4VBKzmpMbW0tc+bMCbqMuKH26E/t0WuobWFmBw12v06engxc5Jz73EChnsgWzptGdrrXBVMcveVI2Ix9bZ38w8+f4d19rQFWJyKpyq+Tp+cBZ0dHxdSa2T0+7SdQ82eUceNl0ykryAagrCCbWz52CosvP5m123Zz/o+f0pztIjLiYpqPfbCcczcDN/ux7Xgzf0YZ82eUUVtby5evmNOz/OTxBXzh12v4h58/w1f/7li+OLeCUIp2zYjIyNIFSj6ZdtQoHvrSWVx8yjhueexV/vFXz9LQpFvpiYj/FOw+ys1M478+fio3XjadZ97YxQW3PsVzW3YFXZaIJDkFu8/MjL8/YyIPfqGKrPQQn1jyNHfWbtaMkCLiGwX7CDlxXD5/+PJZnHfSUfzg0Vf47D2r2d3cHnRZIpKEFOwjaFRWOrf9/Qy+c8mJPPXaTi649Ume37o76LJEJMko2EeYmXHl7En8/p+rCIeNj/10FT9/8nX8vAJYRFKLgj0g08fn88cvn825x43luw+/zOfvXcOelo6gyxKRJKBgD1B+djp3XTmTGy48gb++/C4X/uRJXtzeGHRZIpLgFOwBMzM+c9Zk7v/8bCIRuPzOVdyzaou6ZkTkiCnY48RpEwt5+CtncfbUEr75fxv40n1r2deqrhkRGTwFexwpyMngZ1dVsugjx/Hohre56CdPsWHHnqDLEpEEo2CPM6GQ8fkPlFOz4ExaOyJcesdK7ntmq7pmRCRmCvY4dfqkIh7+ylmcOaWYf3twPV/97Tqa2zqDLktEEoCCPY4V52Vy96dO5+sfPpaHXtjBxbc9xca39wVdlojEOQV7nAuFjC+dO5V7PzuLva2dXHL7U/zv6m1BlyUicUzBniCqykt4+CtnMWNCIQt/9yJf/98XaGnvCrosEYlDCvYEMnZUFvd+dhZf+eBUfv/8dubfXsemd5uCLktE4owvd1AS/4RDxrUfOpbTJxVyTc06Lr7tKb5/6XQAFi/byI7GFsYVZLNw3jTmzygLuFoRCYKCPUGdPXUMj/zL2Xz5vrVc89t1hENGV3SO9/rGFq5/YD2Awl0kBakrJoGVjs7ivqtnkZeZ1hPq3Vo6uli8bGNAlYlIkBTsCS4tHDrk+PYdjS0jXI2IxAMFexIYV5A9qOUiktwU7Elg4bxpZKeH37f85PH5mopAJAUp2JPA/Bll3HjZdMoKsjFgXH4Wp08q5E8vvc3X7n+Btk6NdxdJJRoVkyTmzyjrNwLGOcftKzbxn39+le27W/jplTMpys0IsEIRGSk6Yk9SZt5UBLd9cgbrtjdy6R11bH5PFzOJpAJfgt3MxpjZ98zsO35sX2J34cnjqFlwJk2tnVx6ex0rN+8MuiQR8ZlfR+y3AG1Auk/bl0E4bWIhS79YTenoLK76xbPcr0nERJKaL8HunLsKeMKPbcuRmVCUw++/UMXs8mKu+92L/ODRV4hENGJGJBmZX8PhzGwOcJ5zbtEhXl8ALAAoLS2dWVNT40sdI6WpqYm8vLygyxhQV8Rx78vtrNjWSWVpmKtPziQzbMO6j0Rpi5Gi9uhP7dFrqG0xd+7cNc65ygOXBzYqxjm3BFgCUFlZ6ebMmRNUKcOitraWRPkbzp3r+MVTb/C9R16m4+V0fnZVJWNHZw3b9hOpLUaC2qM/tUcvv9pCo2JSkJnx2bOnsOTKSl57t4n5t9fxytt7gy5LRIaJgj2FfeiEUu7/3GwiDi6/cxUrNr4bdEkiMgx8C3bnXO2h+tclfpxUls/SL1ZzTHEOn7n7Of575ZagSxKRIdIRu3BUfhb3f2425x5Xyrce2sC3H9rwvmmARSRxKNgFgNzMNO66ciZXnz2Zu1du4ep7VtN0iOmARSS+KdilRzhkfOOCE/j+pdN5/NX3uPzOldRrTneRhKNgl/f55KyJ3P3p06nf3cL82+t4YVtj0CWJyCAo2OWgzp46hge+UEVmWoiPL1nFoy+9FXRJIhIjBbsc0tTSUSz9YjUnHD2az9/7PD99fLNu3CGSABTsclgleZncd/WZXHTKOG760yss+v162jsjQZclIoehG23IgLLSw/z446cyuTiHW5dvYuuu/fz0H2aSn6PJO0XikY7YJSahkHHth6fxw4+dwpo3d3PpnXW82dAcdFkichAKdhmUy04bz72fncXu5nbm317Hc1t2BV2SiBxAwS6DdsbkIh78QjWFORlc8bNneHDt9qBLEpE+FOxyRCaV5PLAF6qYeUwhX/3tC/zwsVc1YkYkTijY5YgV5GTw3/90Bh+rHM+tf32Nj965kqob/8r6+j1U37ScpWvrgy5RJCVpVIwMSUZaiB989GT2t3fxxxe9i5j2d0J9YwvXP7AegPkzyoIsUSTl6IhdhszMWLu1sef5/7wWBqClo4vvP/Ky7q0qMsJ0xC7DYkefycJKshxNTd59VN/d18Yp//FnTp1YwKkTCpgxsYBTJxRSlJsRVKkiSU/BLsNiXEF2z0yQH50c4Zb13pfBgux0Ljj5aNZubeSO2s0987wfU5zDjAndYV/I8UePJiNNXyBFhoOCXYbFwnnTuP6B9bR0dPUsy04P8+2LT+zpY9/f3sn67XtYu62RdVsbWbm5gaXrdgBeX/1J40YzY2Jhz5F9WUE2ZhbI3yOSyBTsMiy6w3vxso3APsoKslk4b1q/E6c5GWnMmlLMrCnFADjneGtPK+u2NbJ2627WbWvk3qff5BdPvQHAmFGZfbpvCjhlfAG5mfpfVmQg+lciw2b+jDLmzyijtraWL18xZ8D1zYxxBdmMK8jm/OlHA9DRFeGVt/axbttu1m5tZN22Rh772zsAhAyOLR3FjIkFzJhQyIyJBZSPySMU6n9Uv3RtPYuXbWRHYwvjDvIBI5LsFOwSV9LDIaaPz2f6+HyunO0t293czrrtXvfN2m2NPPziW/zm2W0AjMpM45QJvSdm32ps5XuPvNzTJaRhl5KKFOwS9wpzM5g7bSxzp40FIBJxvNHQzNqtvV04dz6++ZA34G7p6OJ7D7/M6ZOLKM7NICs9PJLli4w4BbsknFDIKB+TR/mYPC6fOR7wTsy+VL+Xj9216qDvea+pjeqblgOQmxGmOC+TotwMSvIyKM7NpCgvg+LcDIqjz7t/F+VmHPFone4uoU9M2Mc3blquLiEZMQp2SQo5GWmcMbmIsj7DLvsqys3gunnTaGhup6GpnYbmNnY1t1Pf2Mr6+j00NLXTeYgj/tFZaRTnZVKcm0FRbgbFeZmU5PU+7vuBUJiTTlo4xNK19b2jhCYE2yWkcw6pR8EuSeVQwy6/eeEJhw0z5xx7WzppaG7rH/5N7TQ0t7OzyfsgeLNhP89vbWRXcxsH+xww88bu72vt7PmgeHS7d8Tf0tHFNx5cz4YdewiHQqSFjLSwkRaygz/veWykh0OEQxZdJ9Rn+SHeG33+2IZ3+P4jL9MavetVPHzA6BuM/xTsklT6DrsczBGqmZGfk05+TjpTxgy8n0jEsaelg4bmNnY2tbOruZ2Gpt4Phf95+s2edbc29Y7aaW7v4r5nttIRcXRFf0ZaS0cX196/jh8+9io5GWGy0sPkZITJTg+THf2dkxEmKyNMTnoa2RkhsjPSyOl+vc86/d+TRmZa6H2jlIC4+QaTKt9eFOySdLqHXfopFDIKczMozM2gYuz7X1/+yrs9XUILjuvilvXeP7WygmzqFp3bs55zjs5owHdGHJ1dkZ7nHV2RPssdnZFI9Hd0/UOtG12vK+JYFA3PA0UcnDaxgP3tXbR0dNHS3sWelg5aos+7lx/J/W37hn12hvcB8Mpb+2jv8rb1yLbebzD/vvQltjQ0k5UeJist5P1OD5OVHiIzLUxmenRZmres7+tZaeGDfogcSr8PF5L724tvwW5m3wHOie5jgXNug1/7Eok3h+oSWjhvWr/1zLzuFL8G6vxk+aaDnnMoK8jmR5+YMeD7O7sitHZG2N/e2S/0W9u7+n0o9P0waGnv7F0v+rs71AHqm3vDuKmtkx/95bUj/vsywiEy00JkpvcN/lD0gyD6ARH9YPjTS2/1++8B3ofLtx7awL62TtKj3Vxe95aRFvIep4VDPa+lhY30aDdZethbx3vsdZV1v9a9XlDfXnwJdjM7Gyh1zn3AzE4CFgPn+7EvkXgUy5W4IyHWD5hDSQuHyAuHyBviFb/VNy3v+YC5+oBvME9cN5e2zi5aOyK0dnRFfyK0dnqP26LL2zr7vN7zuPs173Hf7exv72RXs7edto4I+9u7DlrbnpYOblj60pD+vkMJGf0+GNLDxq7m9p7zM796tXcm1MXLNg7b/x/mx11vokfry51zK6LPn3bOnXnAOguABQClpaUza2pqhr2OkdTU1EReXl7QZcQFtUV/QbdHY0sH7+xppb0rQkY4RGl+FgXZ6SNeQ/3uFiLOUZoN77RAyIyywuwRq2Xj273dQREHXc77HQ6FmDwml66It6znt3P9n0d6l3X2LDtgne73HbitiKMz+vrOpvae/WeE4O/KInQf2E8vyx/U3zR37tw1zrnKA5f71RUzFnivz/NOMws553q+jznnlgBLACorK92cOXN8KmVk1NbWkuh/w3BRW/Sn9vD07Veu2TZqxL/BNB7Qxw7et5cbL5s+onX0/fbytemd/b69xDIVRyz8mid1D1DY53mkb6iLSOqZP6OMukXnMr0sn7pF5454t9T8GWXceNl0b9ZQvCAd6VAHr3ss+4CTKoPpHouFX0fsTwKXA0+a2QmAbmMvIoEbiRFTsdQA/p5/8SvYHwbON7MngX3A53zaj4hIwhnsTKiD5UuwR7td/tmPbYuIyOHpXmQiIklGwS4ikmQU7CIiSUbBLiKSZHy58nTQRZi9B7w54IrxrQTYGXQRcUJt0Z/aoz+1R6+htsUxzrn3zUcaF8GeDMxs9cEu7U1Faov+1B79qT16+dUW6ooREUkyCnYRkSSjYB8+S4IuII6oLfpTe/Sn9ujlS1uoj11EJMnoiF1EJMko2EVEkoyCfQjMrMDMasys1syeMLPJQdcUL8zseTM7L+g6gmZmZ0T/36gzs+uCridIZnatmT0ebYuBb7iahMxsjJl9L3qXOcxsmpn9Ndomi4drP77dzDpF5ADXOud2mNkFwNeBLwZcU+DM7HJgcPf4SkJmlg58E7jEObc76HqCZGYFwMXAHKAc+C/gogBLCsotwCa87AD4EfAZ59wWM/tfM5vlnHtmqDvREfsQOOd2OOd2RJ/uBpqDrCcemNko4Erg10HXEgc+gndF9W+iR2WnBV1QgLrw8iYD72rL9w6/enJyzl0FPAFgZmlAlnNuS/Tl3wOzh2M/OmIfBmZWhne0/qWga4kDtwLfBS4IupA4MBUoAi4ExgO/YZj+4SYa59w+M3sCeBnIAz4YcEnxYAzQ0Od5A3D8cGxYR+xDZGYX4n3dvrrP0XtKMrMrgK3OueeCriVOdAJ/ds51Ro/KImZmAdcUiGhXZTpeN8xxwK3RrqpU1ggU9HleyDB9k1GwD4GZnQxc5Jz7nHOuYcA3JL9PAieYWQ3ePW8Xmdnw3aE38azC647BzEqBDpe6F44cA7wT/fv3AqOArGBLCpZzrgXIjH7jB7gM+OtwbFtdMUNzHnC2mdVGn2+N9qGlJOdcT/eLmX0beNo5tzG4ioLlnHvWzDaaWR3e0fu1QdcUoLuBX5rZ40AmcJdzbl+wJcWFa4HfmVkb8JBz7uXh2KiuPBURSTLqihERSTIKdhGRJKNgFxFJMgp2EZEko2AXEUkyCnZJema2NzpRW62ZLRzE+/LN7FQfSxPxhcaxSyr4m3NuzhG8bwbetQrrYlnZzCyFL0CSOKJgl5RjZpXAjUA28Kpz7p+il/rfBJyJd+n79Xgz7xWZ2Tjn3FVmdgPwYbxvus85564xs0nAj/EuQHrCzN4AFgER4Bbn3IMj+9eJKNglNZzQ5+rgHwJ1wDzAAX+JXtL9QQDn3AcAzCwEXAOc55xbZGYfAiYB5zjnnJndbmYXAeuBk4ATnXOtZvYgcKVzbnN0GyIjTsEuqaBfV4yZnY83h0sT3uyLo4AzgF92r+OcixwwX9epwMN9ulr+gjeZ1XpgrXOuNbr8GuBLZtaC9yHSOOx/jcgAdEQhqehbwFeBG/CO2gFexetPB3rmyu7Cm9cEYAPeUX63c4G10cedfZa/65xbiPet4IZhr1wkBgp2SUUPAs/jTUxVH112FzAteouyx4AJeEfjHzSzXzrnHgF2mtmqaLfOO865vxxk2z+Mzju+CO/GCSIjTpOAiYgkGR2xi4gkGQW7iEiSUbCLiCQZBbuISJJRsIuIJBkFu4hIklGwi4gkmf8PRYcEkyeharMAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.scatter(range(1, df2.shape[1]+1), ev)\n",
        "plt.plot(range(1, df2.shape[1]+1), ev)\n",
        "plt.title('Scree Plot')\n",
        "plt.xlabel('Factors')\n",
        "plt.ylabel('Eigenvalue')\n",
        "plt.grid()\n",
        "plt.show()\n",
        "#eigen값 1 이상일 때까지 or 기울기 완만하기 전까지 "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4bdf4392",
      "metadata": {
        "id": "4bdf4392",
        "outputId": "b2748aaa-2a2c-4196-9616-69bf9fd9ff7c"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "FactorAnalyzer(rotation=None, rotation_kwargs={})"
            ]
          },
          "execution_count": 193,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "#요인수 3개로 선택 \n",
        "fa = FactorAnalyzer(n_factors=3, rotation=None)\n",
        "fa.fit(df2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "772277a0",
      "metadata": {
        "id": "772277a0",
        "outputId": "3cdb87cf-5de4-43e4-d276-9bec799aa562"
      },
      "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",
              "      <th>2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>화장품/미용</th>\n",
              "      <td>0.845308</td>\n",
              "      <td>0.112996</td>\n",
              "      <td>0.317174</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>출산/육아</th>\n",
              "      <td>0.751863</td>\n",
              "      <td>-0.444661</td>\n",
              "      <td>0.008940</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>생활/건강</th>\n",
              "      <td>0.909406</td>\n",
              "      <td>-0.331056</td>\n",
              "      <td>0.255021</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>식품</th>\n",
              "      <td>0.738321</td>\n",
              "      <td>-0.093044</td>\n",
              "      <td>0.604290</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>패션의류</th>\n",
              "      <td>0.763690</td>\n",
              "      <td>0.174321</td>\n",
              "      <td>-0.546183</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>패션잡화</th>\n",
              "      <td>0.733410</td>\n",
              "      <td>0.553926</td>\n",
              "      <td>-0.356814</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>디지털/가전</th>\n",
              "      <td>0.882935</td>\n",
              "      <td>-0.104929</td>\n",
              "      <td>-0.094637</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>가구/인테리어</th>\n",
              "      <td>0.939209</td>\n",
              "      <td>-0.112387</td>\n",
              "      <td>-0.261677</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>스포츠/레저</th>\n",
              "      <td>0.440416</td>\n",
              "      <td>0.866030</td>\n",
              "      <td>0.232294</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>여가/생활편의</th>\n",
              "      <td>-0.182528</td>\n",
              "      <td>0.624705</td>\n",
              "      <td>0.257938</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                0         1         2\n",
              "화장품/미용   0.845308  0.112996  0.317174\n",
              "출산/육아    0.751863 -0.444661  0.008940\n",
              "생활/건강    0.909406 -0.331056  0.255021\n",
              "식품       0.738321 -0.093044  0.604290\n",
              "패션의류     0.763690  0.174321 -0.546183\n",
              "패션잡화     0.733410  0.553926 -0.356814\n",
              "디지털/가전   0.882935 -0.104929 -0.094637\n",
              "가구/인테리어  0.939209 -0.112387 -0.261677\n",
              "스포츠/레저   0.440416  0.866030  0.232294\n",
              "여가/생활편의 -0.182528  0.624705  0.257938"
            ]
          },
          "execution_count": 194,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "efa_result = pd.DataFrame(fa.loadings_, index=df2.columns)\n",
        "efa_result"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0c72cb74",
      "metadata": {
        "id": "0c72cb74",
        "outputId": "45dd6d41-3fc9-4e46-a9bf-1d901b51226b"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(10, 5))\n",
        "sns.heatmap(efa_result, cmap=\"Blues\", annot=True, fmt='.2f')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "de7802b5",
      "metadata": {
        "id": "de7802b5",
        "outputId": "de0cc816-899b-4fb8-da7e-99f74ddc4d82"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "FactorAnalyzer(rotation='varimax', rotation_kwargs={})"
            ]
          },
          "execution_count": 196,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "#최대우도 방법 선택 \n",
        "fa_1 = FactorAnalyzer(n_factors=3, rotation=\"varimax\")\n",
        "fa_1.fit(df2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "bcf2bf29",
      "metadata": {
        "id": "bcf2bf29",
        "outputId": "6610bd91-c409-41d9-8e76-6de5dced8a0b"
      },
      "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",
              "      <th>2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>화장품/미용</th>\n",
              "      <td>0.796619</td>\n",
              "      <td>0.367749</td>\n",
              "      <td>0.240981</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>출산/육아</th>\n",
              "      <td>0.688252</td>\n",
              "      <td>0.366489</td>\n",
              "      <td>-0.393823</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>생활/건강</th>\n",
              "      <td>0.925139</td>\n",
              "      <td>0.330179</td>\n",
              "      <td>-0.191713</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>식품</th>\n",
              "      <td>0.944502</td>\n",
              "      <td>0.039493</td>\n",
              "      <td>0.159050</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>패션의류</th>\n",
              "      <td>0.197243</td>\n",
              "      <td>0.933694</td>\n",
              "      <td>-0.035174</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>패션잡화</th>\n",
              "      <td>0.188126</td>\n",
              "      <td>0.886871</td>\n",
              "      <td>0.387439</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>디지털/가전</th>\n",
              "      <td>0.633877</td>\n",
              "      <td>0.619606</td>\n",
              "      <td>-0.117596</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>가구/인테리어</th>\n",
              "      <td>0.577174</td>\n",
              "      <td>0.771297</td>\n",
              "      <td>-0.187593</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>스포츠/레저</th>\n",
              "      <td>0.241192</td>\n",
              "      <td>0.368851</td>\n",
              "      <td>0.896499</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>여가/생활편의</th>\n",
              "      <td>-0.147131</td>\n",
              "      <td>-0.127130</td>\n",
              "      <td>0.672529</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                0         1         2\n",
              "화장품/미용   0.796619  0.367749  0.240981\n",
              "출산/육아    0.688252  0.366489 -0.393823\n",
              "생활/건강    0.925139  0.330179 -0.191713\n",
              "식품       0.944502  0.039493  0.159050\n",
              "패션의류     0.197243  0.933694 -0.035174\n",
              "패션잡화     0.188126  0.886871  0.387439\n",
              "디지털/가전   0.633877  0.619606 -0.117596\n",
              "가구/인테리어  0.577174  0.771297 -0.187593\n",
              "스포츠/레저   0.241192  0.368851  0.896499\n",
              "여가/생활편의 -0.147131 -0.127130  0.672529"
            ]
          },
          "execution_count": 197,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "efa_result_1 = pd.DataFrame(fa_1.loadings_, index=df2.columns)\n",
        "efa_result_1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0bf5010b",
      "metadata": {
        "id": "0bf5010b",
        "outputId": "5c53a436-692b-45e9-e8a5-686d1cea1ac6"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<AxesSubplot:>"
            ]
          },
          "execution_count": 198,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(10, 5))\n",
        "sns.heatmap(efa_result_1, cmap=\"Blues\", annot=True, fmt='.2f')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2f4417c4",
      "metadata": {
        "id": "2f4417c4",
        "outputId": "7f2125a1-0120-47a0-f474-7101b9fe3a97"
      },
      "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",
              "      <th>2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>SS Loadings</th>\n",
              "      <td>3.745304</td>\n",
              "      <td>3.169479</td>\n",
              "      <td>1.731591</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Proportion Var</th>\n",
              "      <td>0.374530</td>\n",
              "      <td>0.316948</td>\n",
              "      <td>0.173159</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Cumlative Var</th>\n",
              "      <td>0.374530</td>\n",
              "      <td>0.691478</td>\n",
              "      <td>0.864637</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                       0         1         2\n",
              "SS Loadings     3.745304  3.169479  1.731591\n",
              "Proportion Var  0.374530  0.316948  0.173159\n",
              "Cumlative Var   0.374530  0.691478  0.864637"
            ]
          },
          "execution_count": 199,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "fa.get_factor_variance()\n",
        "df2_1 = pd.DataFrame(fa_1.get_factor_variance())\n",
        "\n",
        "df2_1.index=['SS Loadings', 'Proportion Var', 'Cumlative Var']\n",
        "df2_1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0ed40458",
      "metadata": {
        "id": "0ed40458"
      },
      "outputs": [],
      "source": [
        "wth = df_wea_order.copy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "55ed37bf",
      "metadata": {
        "id": "55ed37bf"
      },
      "outputs": [],
      "source": [
        "shopping = df_org2.copy()\n",
        "\n",
        "fact1 = [\"화장품/미용\", \"생활/건강\",\"식품\",\"출산/육아\"]\n",
        "fact2 = [\"패션의류\", \"패션잡화\",\"가구인테리어\" ,\"디지털/가전\"]\n",
        "fact3 = [\"스포츠/레저\" , \"여가/생활편의\"]\n",
        "\n",
        "shop_factor = pd.DataFrame({\"BKG_DATE\" : shopping['날짜'],\n",
        "                            \"Healthcare\": shopping.filter(regex=\"|\".join(fact1)).sum(axis=1),\n",
        "                            \"Lifestyle\" : shopping.filter(regex=\"|\".join(fact2)).sum(axis=1),\n",
        "                            \"Leisure\" : shopping.filter(regex=\"|\".join(fact3)).sum(axis=1)})\n",
        "\n",
        "full = pd.merge(wth, shop_factor,on=['BKG_DATE'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "99390657",
      "metadata": {
        "id": "99390657"
      },
      "outputs": [],
      "source": [
        "cl = pd.DataFrame({\"BKG_DATE\" :full[\"BKG_DATE\"],\n",
        "                   \"ITEM_QTY\":full[\"ITEM_QTY\"],\n",
        "                   \"SHPR_CD\" : full['SHPR_CD'],\n",
        "                   \"ITEM_CD\" : full['ITEM_CD'],\n",
        "                   \"Weather\" : full['날씨지수'],\n",
        "                   \"Healthcare\": full['Healthcare'],\n",
        "                   \"Lifestyle\" : full['Lifestyle'],\n",
        "                   \"Leisure\" : full['Leisure'],\n",
        "                   \"Day\": full['day']})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a4bf7652",
      "metadata": {
        "id": "a4bf7652",
        "outputId": "661afccd-c1dc-4afd-aa63-f87e5c718166"
      },
      "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>BKG_DATE</th>\n",
              "      <th>ITEM_QTY</th>\n",
              "      <th>SHPR_CD</th>\n",
              "      <th>ITEM_CD</th>\n",
              "      <th>Weather</th>\n",
              "      <th>Healthcare</th>\n",
              "      <th>Lifestyle</th>\n",
              "      <th>Leisure</th>\n",
              "      <th>Day</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>1</td>\n",
              "      <td>90001302</td>\n",
              "      <td>134265392</td>\n",
              "      <td>3.0</td>\n",
              "      <td>187.0</td>\n",
              "      <td>183.0</td>\n",
              "      <td>38.0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>1</td>\n",
              "      <td>90001542</td>\n",
              "      <td>3858391637</td>\n",
              "      <td>3.0</td>\n",
              "      <td>187.0</td>\n",
              "      <td>183.0</td>\n",
              "      <td>38.0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>1</td>\n",
              "      <td>90001341</td>\n",
              "      <td>5161458796</td>\n",
              "      <td>3.0</td>\n",
              "      <td>187.0</td>\n",
              "      <td>183.0</td>\n",
              "      <td>38.0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>1</td>\n",
              "      <td>90001441</td>\n",
              "      <td>4624200932</td>\n",
              "      <td>3.0</td>\n",
              "      <td>187.0</td>\n",
              "      <td>183.0</td>\n",
              "      <td>38.0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2021-03-03</td>\n",
              "      <td>1</td>\n",
              "      <td>90001541</td>\n",
              "      <td>4929464978</td>\n",
              "      <td>3.0</td>\n",
              "      <td>187.0</td>\n",
              "      <td>183.0</td>\n",
              "      <td>38.0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    BKG_DATE  ITEM_QTY   SHPR_CD     ITEM_CD  Weather  Healthcare  Lifestyle  \\\n",
              "0 2021-03-03         1  90001302   134265392      3.0       187.0      183.0   \n",
              "1 2021-03-03         1  90001542  3858391637      3.0       187.0      183.0   \n",
              "2 2021-03-03         1  90001341  5161458796      3.0       187.0      183.0   \n",
              "3 2021-03-03         1  90001441  4624200932      3.0       187.0      183.0   \n",
              "4 2021-03-03         1  90001541  4929464978      3.0       187.0      183.0   \n",
              "\n",
              "   Leisure Day  \n",
              "0     38.0   2  \n",
              "1     38.0   2  \n",
              "2     38.0   2  \n",
              "3     38.0   2  \n",
              "4     38.0   2  "
            ]
          },
          "execution_count": 204,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "cl.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ce499575",
      "metadata": {
        "id": "ce499575"
      },
      "outputs": [],
      "source": [
        "df = cl[['Weather', 'Healthcare', 'Lifestyle', 'Leisure','Day']].copy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "155e9ddb",
      "metadata": {
        "id": "155e9ddb"
      },
      "outputs": [],
      "source": [
        "#표준정규분포를 활용한 Normalization 수행\n",
        "data = (df-np.mean(df,axis=0))/np.std(df,axis=0)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "98fcaab0",
      "metadata": {
        "id": "98fcaab0"
      },
      "outputs": [],
      "source": [
        "from minisom import MiniSom\n",
        "\n",
        "from tqdm import tqdm, tqdm_notebook, tnrange\n",
        "import time"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4468441f",
      "metadata": {
        "colab": {
          "referenced_widgets": [
            "5f7af9ee490e4f8dbe49f5d5e711e60c",
            "8ec63407da9c40b7ad161186a4911192",
            "03226b1da3444eef9ec5bfb10cecacbf",
            "1498d745e47747f6b440c095625c1601",
            "5e74a24233334c4aa522fb926a4b87b8",
            "be54d466b82e47df9c8a47e945dff23b",
            "4b405c1a93b04916801b3456e8bbfe3e",
            "f4643342aafe436180e3d2c64c6ab679",
            "1dd2e8dc72bd44b090f247e8dca00a6b"
          ]
        },
        "id": "4468441f",
        "outputId": "f2ac6b51-007f-41cb-83a4-cc49de970750",
        "scrolled": true
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "시작시간: 2021-10-28 10:19:54.352853 \n",
            "\n"
          ]
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "5f7af9ee490e4f8dbe49f5d5e711e60c",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "first:   0%|          | 0/4 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "8ec63407da9c40b7ad161186a4911192",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "second:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "03226b1da3444eef9ec5bfb10cecacbf",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "third:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "1498d745e47747f6b440c095625c1601",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "third:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "5e74a24233334c4aa522fb926a4b87b8",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "third:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "be54d466b82e47df9c8a47e945dff23b",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "third:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "4b405c1a93b04916801b3456e8bbfe3e",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "third:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "f4643342aafe436180e3d2c64c6ab679",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "third:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "1dd2e8dc72bd44b090f247e8dca00a6b",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "third:   0%|          | 0/9 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "ename": "KeyboardInterrupt",
          "evalue": "",
          "output_type": "error",
          "traceback": [
            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_4388/3836490970.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     19\u001b[0m                 \u001b[0mestimator\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mrandom_weights_init\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     20\u001b[0m                 \u001b[0mestimator\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m1000\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mrandom_order\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 21\u001b[1;33m                 \u001b[0mqe\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mestimator\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mquantization_error\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     22\u001b[0m                 \u001b[1;31m#te = estimator.topographic_error(data.values)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     23\u001b[0m                 \u001b[0mwinner_coordinates\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0marray\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mestimator\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwinner\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mT\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
            "\u001b[1;32mC:\\Anaconda3\\lib\\site-packages\\minisom.py\u001b[0m in \u001b[0;36mquantization_error\u001b[1;34m(self, data)\u001b[0m\n\u001b[0;32m    490\u001b[0m         distance between each input sample and its best matching unit.\"\"\"\n\u001b[0;32m    491\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_check_input_len\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 492\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mnorm\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m-\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mquantization\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmean\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    493\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    494\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0mtopographic_error\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
            "\u001b[1;32mC:\\Anaconda3\\lib\\site-packages\\minisom.py\u001b[0m in \u001b[0;36mquantization\u001b[1;34m(self, data)\u001b[0m\n\u001b[0;32m    334\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_check_input_len\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    335\u001b[0m         \u001b[0mwinners_coords\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0margmin\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_distance_from_weights\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 336\u001b[1;33m         return self._weights[unravel_index(winners_coords,\n\u001b[0m\u001b[0;32m    337\u001b[0m                                            self._weights.shape[:2])]\n\u001b[0;32m    338\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
            "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ],
      "source": [
        "s_time = pd.Timestamp.now()\n",
        "print('시작시간:',s_time,'\\n')\n",
        "\n",
        "#원하는 파라미터 조합 리스트화\n",
        "map_n= [n for n in range(2,6)]\n",
        "para_sigma= [np.round(sigma*0.1,2) for sigma in range(1,10)]\n",
        "para_learning_rate= [np.round(learning_rate*0.1,2) for learning_rate in range(1,10)]\n",
        "\n",
        "#결과 값을 담을 리스트 res 생성\n",
        "res = []\n",
        "#모든 조합에 대해 모델 생성 및 qe,te값 계산\n",
        "for n in tqdm_notebook(map_n, desc = 'first'):\n",
        "    for sigma in tqdm_notebook(para_sigma, desc = 'second'):\n",
        "        for lr in tqdm_notebook(para_learning_rate, desc='third'):\n",
        "            time.sleep(0.01)\n",
        "            try:\n",
        "                #랜덤으로 초기값을 설정하는 경우\n",
        "                estimator = MiniSom(n,n,5,sigma =sigma, learning_rate = lr, topology='hexagonal',random_seed=0)\n",
        "                estimator.random_weights_init(data.values)\n",
        "                estimator.train(data.values,1000,random_order=True)\n",
        "                qe = estimator.quantization_error(data.values)\n",
        "                #te = estimator.topographic_error(data.values)\n",
        "                winner_coordinates = np.array([estimator.winner(x) for x in data.values]).T\n",
        "                cluster_index = np.ravel_multi_index(winner_coordinates,(n,n))\n",
        "                \n",
        "                res.append([str(n)+'x'+str(n),sigma,lr,'random_init',qe,len(np.unique(cluster_index))])\n",
        "\n",
        "                #pca로 초기값을 설정하는 경우\n",
        "                estimator = MiniSom(n,n,5,sigma =sigma, learning_rate = lr,topology='hexagonal', random_seed=0)\n",
        "                estimator.pca_weights_init(data.values)\n",
        "                estimator.train(data.values,1000,random_order=True)\n",
        "                qe = estimator.quantization_error(data.values)\n",
        "                #te = estimator.topographic_error(data.values)\n",
        "                winner_coordinates = np.array([estimator.winner(x) for x in data.values]).T\n",
        "                cluster_index = np.ravel_multi_index(winner_coordinates,(n,n))\n",
        "                \n",
        "                res.append([str(n)+'x'+str(n),sigma,lr,'pca_init',qe,len(np.unique(cluster_index))])\n",
        "                \n",
        "            except ValueError as e:\n",
        "                print(e)\n",
        "            \n",
        "#결과 데이터프레임 생성 및 sorting \n",
        "df_res = pd.DataFrame(res,columns=['map_size','sigma','learning_rate','init_method','qe','n_cluster']) \n",
        "df_res.shape\n",
        "df_res.sort_values(by=['qe'],ascending=True,inplace=True,ignore_index=True)\n",
        "df_res.head(10)\n",
        "\n",
        "#시각화를 위한 lineplot 생성\n",
        "plt.figure(figsize=(20,10))\n",
        "sns.lineplot(data = df_res)\n",
        "\n",
        "e_time = pd.Timestamp.now()\n",
        "print('\\n종료시간:',e_time,'\\n총 소요시간:',e_time-s_time)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "319148d5",
      "metadata": {
        "id": "319148d5"
      },
      "outputs": [],
      "source": [
        "df_res.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "7dd12fbb",
      "metadata": {
        "id": "7dd12fbb"
      },
      "source": [
        "# 모델 파라미터 조정 "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "19b153e2",
      "metadata": {
        "id": "19b153e2"
      },
      "outputs": [],
      "source": [
        "#앞서 클러스터 개수가 16이면서 qe값이 가장 작게 나온 파라미터로 모델링 \n",
        "som_b2 = MiniSom(4,4,5,sigma=0.5,learning_rate=0.1,topology='hexagonal',neighborhood_function='gaussian',activation_distance='euclidean', random_seed=0)\n",
        "som_b2.random_weights_init(data.values)\n",
        "som_b2.train(data.values,1000,random_order=True)\n",
        "\n",
        "#평가\n",
        "som_b2.quantization_error(data.values)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "c4dce2a9",
      "metadata": {
        "id": "c4dce2a9"
      },
      "source": [
        "# 시각화"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "02069463",
      "metadata": {
        "id": "02069463"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.patches as patches\n",
        "from matplotlib import cm\n",
        "import matplotlib as mpl\n",
        "from mpl_toolkits.axes_grid1 import make_axes_locatable\n",
        "from matplotlib.patches import RegularPolygon"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fcfeb8d4",
      "metadata": {
        "id": "fcfeb8d4"
      },
      "outputs": [],
      "source": [
        "#지정한 파라미터로 모델 생성 \n",
        "estimator = MiniSom(4,4,5,sigma=0.5,learning_rate=0.1, topology='hexagonal',random_seed=0)\n",
        "estimator.random_weights_init(data.values)\n",
        "estimator.train(data.values,1000,random_order=True)\n",
        "qe = estimator.quantization_error(data.values)\n",
        "\n",
        "winner_coordinates = np.array([estimator.winner(x) for x in data.values]).T\n",
        "cluster_index = np.ravel_multi_index(winner_coordinates,(4,4))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4b640373",
      "metadata": {
        "id": "4b640373"
      },
      "outputs": [],
      "source": [
        "xx, yy = som_b2.get_euclidean_coordinates()\n",
        "umatrix = som_b2.distance_map()\n",
        "weights = som_b2.get_weights()\n",
        "\n",
        "f = plt.figure(figsize=(20,20))\n",
        "ax = f.add_subplot(111)\n",
        "ax.set_aspect('equal')\n",
        "\n",
        "# iteratively add hexagons\n",
        "# plotting the distance map as background\n",
        "# 해당 셀과 다른 이웃들 간 거리를 표현, 밝을수록 가깝고, 어두울수록 멀다\n",
        "for i in range(weights.shape[0]):\n",
        "    for j in range(weights.shape[1]):\n",
        "        wy = yy[(i, j)] * 2 / np.sqrt(4) * 4 / 5\n",
        "        hex = RegularPolygon((xx[(i, j)], wy), \n",
        "                             numVertices=6, \n",
        "                             radius=.95 / np.sqrt(3),\n",
        "                             facecolor=cm.Blues(umatrix[i, j]), \n",
        "                             alpha=.4, \n",
        "                             edgecolor='gray')\n",
        "        plot = ax.add_patch(hex)\n",
        "\n",
        "#output노드에 해당하는 클러스터 종류 및 밀도 확인         \n",
        "cnt=[]\n",
        "for c in tqdm_notebook(np.unique(cluster_index)):\n",
        "        time.sleep(0.01)\n",
        "        x_= [som_b2.convert_map_to_euclidean(som_b2.winner(x))[0] + (2*np.random.rand(1)[0]-1)*0.4 for x in data.values[cluster_index==c]]\n",
        "        y_= [som_b2.convert_map_to_euclidean(som_b2.winner(x))[1] + (2*np.random.rand(1)[0]-1)*0.4 for x in data.values[cluster_index==c]]\n",
        "        y_= [(i* 2 / np.sqrt(4) * 4 / 5) for i in y_]\n",
        "    \n",
        "        plot = sns.scatterplot( x = x_,  y= y_ ,label='cluster='+str(c),alpha=.5, palette=\"pastel\")\n",
        "        \n",
        "        #클러스터에 속한 데이터 개수 데이터프레임으로 출력\n",
        "        cnt.append([c,len(x_)])\n",
        "\n",
        "#클러스터별 개수를 표 형태로 출력 \n",
        "df_cnt = pd.DataFrame(cnt,columns=['cluster이름','개수'])\n",
        "df_cnt\n",
        "\n",
        "#x축,y축 간격 설정 \n",
        "xrange = np.arange(weights.shape[0])\n",
        "yrange = np.arange(weights.shape[1])\n",
        "plot = plt.xticks(xrange-.5, xrange)\n",
        "plot = plt.yticks(yrange * 2 / np.sqrt(4) * 4/ 5, yrange)\n",
        "\n",
        "#차트 우측에 color bar생성\n",
        "divider = make_axes_locatable(plt.gca())\n",
        "ax_cb = divider.new_horizontal(size=\"5%\", pad=0.05)    \n",
        "cb1 = mpl.colorbar.ColorbarBase(ax_cb, cmap=cm.Blues, \n",
        "                            orientation='vertical', alpha=.4)\n",
        "cb1.ax.get_yaxis().labelpad = 16\n",
        "plot = cb1.ax.set_ylabel('distance from neurons in the neighborhood',\n",
        "                  rotation=270, fontsize=16)\n",
        "plot = plt.gcf().add_axes(ax_cb)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "57048b66",
      "metadata": {
        "id": "57048b66"
      },
      "outputs": [],
      "source": [
        "df_cnt"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "145e8453",
      "metadata": {
        "id": "145e8453"
      },
      "outputs": [],
      "source": [
        "cluster_df = pd.DataFrame(cluster_index)\n",
        "cluster_df.columns=['cluster']\n",
        "\n",
        "cluster_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "53d039e6",
      "metadata": {
        "id": "53d039e6"
      },
      "outputs": [],
      "source": [
        "result = pd.concat([cl, cluster_df], axis=1)\n",
        "result"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e89efe10",
      "metadata": {
        "id": "e89efe10"
      },
      "outputs": [],
      "source": [
        "## 클러스터별 데이터프레임 생성 \n",
        "cl0 = result[result['cluster']==0]\n",
        "cl0 = cl0.drop(['cluster'], axis=1)\n",
        "\n",
        "cl1 = result[result['cluster']==1]\n",
        "cl1 = cl1.drop(['cluster'], axis=1)\n",
        "\n",
        "cl2 = result[result['cluster']==2]\n",
        "cl2 = cl2.drop(['cluster'], axis=1)\n",
        "\n",
        "cl3 = result[result['cluster']==3]\n",
        "cl3 = cl3.drop(['cluster'], axis=1)\n",
        "\n",
        "cl4 = result[result['cluster']==4]\n",
        "cl4 = cl4.drop(['cluster'], axis=1)\n",
        "\n",
        "cl5 = result[result['cluster']==5]\n",
        "cl5 = cl5.drop(['cluster'], axis=1)\n",
        "\n",
        "cl6 = result[result['cluster']==6]\n",
        "cl6 = cl6.drop(['cluster'], axis=1)\n",
        "\n",
        "cl7 = result[result['cluster']==7]\n",
        "cl7 = cl7.drop(['cluster'], axis=1)\n",
        "\n",
        "cl8 = result[result['cluster']==8]\n",
        "cl8 = cl8.drop(['cluster'], axis=1)\n",
        "\n",
        "cl9 = result[result['cluster']==9]\n",
        "cl9 = cl9.drop(['cluster'], axis=1)\n",
        "\n",
        "cl10 = result[result['cluster']==10]\n",
        "cl10 = cl10.drop(['cluster'], axis=1)\n",
        "\n",
        "cl11 = result[result['cluster']==11]\n",
        "cl11 = cl11.drop(['cluster'], axis=1)\n",
        "\n",
        "cl12 = result[result['cluster']==12]\n",
        "cl12 = cl12.drop(['cluster'], axis=1)\n",
        "\n",
        "cl13 = result[result['cluster']==13]\n",
        "cl13 = cl13.drop(['cluster'], axis=1)\n",
        "\n",
        "cl14 = result[result['cluster']==14]\n",
        "cl14 = cl14.drop(['cluster'], axis=1)\n",
        "\n",
        "cl15 = result[result['cluster']==15]\n",
        "cl15 = cl15.drop(['cluster'], axis=1)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "4c12f288",
      "metadata": {
        "id": "4c12f288"
      },
      "source": [
        "# XGBoost"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0de3d5cf",
      "metadata": {
        "id": "0de3d5cf"
      },
      "outputs": [],
      "source": [
        "from xgboost import XGBRegressor # 회귀트리 모델\n",
        "from xgboost import plot_importance # 중요변수 시각화\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import mean_absolute_error, mean_squared_error, explained_variance_score"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5a45dee8",
      "metadata": {
        "id": "5a45dee8"
      },
      "outputs": [],
      "source": [
        "#  'SHPR_CD'로 그룹화 \n",
        "grouped_0 = cl0.groupby('SHPR_CD').mean()\n",
        "grouped_1 = cl1.groupby('SHPR_CD').mean()\n",
        "grouped_2 = cl2.groupby('SHPR_CD').mean()\n",
        "grouped_3 = cl3.groupby('SHPR_CD').mean()\n",
        "grouped_4 = cl4.groupby('SHPR_CD').mean()\n",
        "grouped_5 = cl5.groupby('SHPR_CD').mean()\n",
        "grouped_6 = cl6.groupby('SHPR_CD').mean()\n",
        "grouped_7 = cl7.groupby('SHPR_CD').mean()\n",
        "grouped_8 = cl8.groupby('SHPR_CD').mean()\n",
        "grouped_9 = cl9.groupby('SHPR_CD').mean()\n",
        "grouped_10 = cl10.groupby('SHPR_CD').mean()\n",
        "grouped_11 = cl11.groupby('SHPR_CD').mean()\n",
        "grouped_12 = cl12.groupby('SHPR_CD').mean()\n",
        "grouped_13 = cl13.groupby('SHPR_CD').mean()\n",
        "grouped_14 = cl14.groupby('SHPR_CD').mean()\n",
        "grouped_15 = cl15.groupby('SHPR_CD').mean()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "713b9a51",
      "metadata": {
        "id": "713b9a51"
      },
      "source": [
        "### 훈련/테스트 구분 및 모델 생성"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "db0e02fe",
      "metadata": {
        "id": "db0e02fe"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_0['ITEM_QTY']\n",
        "X_features = grouped_0.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_0 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_0.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4119b5b6",
      "metadata": {
        "id": "4119b5b6"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_1['ITEM_QTY']\n",
        "X_features = grouped_1.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_1 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_1.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "8563afd3",
      "metadata": {
        "id": "8563afd3"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_2['ITEM_QTY']\n",
        "X_features = grouped_2.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_2 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_2.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "6332cf9f",
      "metadata": {
        "id": "6332cf9f"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_3['ITEM_QTY']\n",
        "X_features = grouped_3.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_3 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_3.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0e67cf72",
      "metadata": {
        "id": "0e67cf72"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_4['ITEM_QTY']\n",
        "X_features = grouped_4.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_4 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_4.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c549caf8",
      "metadata": {
        "id": "c549caf8"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_5['ITEM_QTY']\n",
        "X_features = grouped_5.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_5 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_5.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "af69c86a",
      "metadata": {
        "id": "af69c86a"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_6['ITEM_QTY']\n",
        "X_features = grouped_6.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_6 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_6.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "69810eec",
      "metadata": {
        "id": "69810eec"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_7['ITEM_QTY']\n",
        "X_features = grouped_7.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_7 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_7.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "9afac2ee",
      "metadata": {
        "id": "9afac2ee"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_8['ITEM_QTY']\n",
        "X_features = grouped_8.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_8 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_8.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "50fa4047",
      "metadata": {
        "id": "50fa4047"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_9['ITEM_QTY']\n",
        "X_features = grouped_9.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_9 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_9.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0804a5f6",
      "metadata": {
        "id": "0804a5f6"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_10['ITEM_QTY']\n",
        "X_features = grouped_10.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_10 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_10.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "39ff2e3b",
      "metadata": {
        "id": "39ff2e3b"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_11['ITEM_QTY']\n",
        "X_features = grouped_11.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_11 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_11.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c5965790",
      "metadata": {
        "id": "c5965790"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_12['ITEM_QTY']\n",
        "X_features = grouped_12.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_12 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_12.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3bdb68c6",
      "metadata": {
        "id": "3bdb68c6"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_13['ITEM_QTY']\n",
        "X_features = grouped_13.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_13 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_13.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "88917124",
      "metadata": {
        "id": "88917124"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_14['ITEM_QTY']\n",
        "X_features = grouped_14.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_14 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_14.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "975b2aa7",
      "metadata": {
        "id": "975b2aa7"
      },
      "outputs": [],
      "source": [
        "y_target = grouped_15['ITEM_QTY']\n",
        "X_features = grouped_15.drop('ITEM_QTY', axis=1, inplace=False)\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_features, y_target, test_size=0.3,random_state=0)\n",
        "xgb_model_15 = XGBRegressor(n_estimators=100, learning_rate=0.08, gamma=0, subsample=0.75, colsample_bytree=1, max_depth=7)\n",
        "xgb_model_15.fit(X_train,y_train)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b81c1640",
      "metadata": {
        "id": "b81c1640"
      },
      "source": [
        "### RMSE: 예측성능 평가"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4e8af822",
      "metadata": {
        "id": "4e8af822"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_0.predict(X_test)\n",
        "rmse_0 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_0))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0cba4977",
      "metadata": {
        "id": "0cba4977"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_1.predict(X_test)\n",
        "rmse_1 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_1))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fdb7f0f6",
      "metadata": {
        "id": "fdb7f0f6"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_2.predict(X_test)\n",
        "rmse_2 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_2))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fbc5e59f",
      "metadata": {
        "id": "fbc5e59f"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_3.predict(X_test)\n",
        "rmse_3 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_3))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3e856cc9",
      "metadata": {
        "id": "3e856cc9"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_4.predict(X_test)\n",
        "rmse_4 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_4))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "b3f49ad6",
      "metadata": {
        "id": "b3f49ad6"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_5.predict(X_test)\n",
        "rmse_5 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_5))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "31e09322",
      "metadata": {
        "id": "31e09322"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_6.predict(X_test)\n",
        "rmse_6 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_6))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "25eb7bab",
      "metadata": {
        "id": "25eb7bab"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_7.predict(X_test)\n",
        "rmse_7 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_7))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fc9a7930",
      "metadata": {
        "id": "fc9a7930"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_8.predict(X_test)\n",
        "rmse_8 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_8))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5215fd55",
      "metadata": {
        "id": "5215fd55"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_9.predict(X_test)\n",
        "rmse_9 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_9))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d2fa8b22",
      "metadata": {
        "id": "d2fa8b22"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_10.predict(X_test)\n",
        "rmse_10 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_10))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "da99bf7d",
      "metadata": {
        "id": "da99bf7d"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_11.predict(X_test)\n",
        "rmse_11 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_11))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c3fbb30d",
      "metadata": {
        "id": "c3fbb30d"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_12.predict(X_test)\n",
        "rmse_12 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_12))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a28e6d5a",
      "metadata": {
        "id": "a28e6d5a"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_13.predict(X_test)\n",
        "rmse_13 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_13))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "69259c04",
      "metadata": {
        "id": "69259c04"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_14.predict(X_test)\n",
        "rmse_14 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_14))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "8134d9f3",
      "metadata": {
        "id": "8134d9f3"
      },
      "outputs": [],
      "source": [
        "y_pred = xgb_model_15.predict(X_test)\n",
        "rmse_15 = np.sqrt(mean_squared_error(y_test, y_pred))\n",
        "print(\"RMSE: %f\" % (rmse_15))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a45343ad",
      "metadata": {
        "id": "a45343ad"
      },
      "outputs": [],
      "source": [
        "RMSE = pd.DataFrame(\n",
        "                {'cluster' : [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15],\n",
        "                 'rmse' : [rmse_0,rmse_1, rmse_2, rmse_3, rmse_4, rmse_5,rmse_6,rmse_7,rmse_8,rmse_9,rmse_10,rmse_11,rmse_12,rmse_13, rmse_14,rmse_15]})\n",
        "RMSE.set_index('cluster')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "a9a42dbb",
      "metadata": {
        "id": "a9a42dbb"
      },
      "source": [
        "### 변수 중요도 (Feature Importance)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "b185ffe6",
      "metadata": {
        "id": "b185ffe6"
      },
      "outputs": [],
      "source": [
        "fig, axes = plt.subplots(nrows=4, ncols=4, figsize=(25,25))\n",
        "axes = [ax for row_axes in axes for ax in row_axes]\n",
        "plot_importance(xgb_model_0, ax = axes[0], title='cluster0',height=0.5)\n",
        "plot_importance(xgb_model_1, ax = axes[1], title='cluster1',height=0.5)\n",
        "plot_importance(xgb_model_2, ax = axes[2], title='cluster2',height=0.5)\n",
        "plot_importance(xgb_model_3, ax = axes[3], title='cluster3',height=0.5)\n",
        "plot_importance(xgb_model_4, ax = axes[4], title='cluster4',height=0.5)\n",
        "plot_importance(xgb_model_5, ax = axes[5], title='cluster5',height=0.5)\n",
        "plot_importance(xgb_model_6, ax = axes[6], title='cluster6',height=0.5)\n",
        "plot_importance(xgb_model_7, ax = axes[7], title='cluster7',height=0.5)\n",
        "plot_importance(xgb_model_8, ax = axes[8], title='cluster8',height=0.5)\n",
        "plot_importance(xgb_model_9, ax = axes[9], title='cluster9',height=0.5)\n",
        "plot_importance(xgb_model_10, ax = axes[10], title='cluster10',height=0.5)\n",
        "plot_importance(xgb_model_11, ax = axes[11], title='cluster11',height=0.5)\n",
        "plot_importance(xgb_model_12, ax = axes[12], title='cluster12',height=0.5)\n",
        "plot_importance(xgb_model_13, ax = axes[13], title='cluster13',height=0.5)\n",
        "plot_importance(xgb_model_14, ax = axes[14], title='cluster14',height=0.5)\n",
        "plot_importance(xgb_model_15, ax = axes[15], title='cluster15',height=0.5)\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "f3417d03",
      "metadata": {
        "id": "f3417d03"
      },
      "source": [
        "# Prophet 시계열 예측"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "ae3f1892",
      "metadata": {
        "id": "ae3f1892"
      },
      "source": [
        "### 1) 클러스터별 수량 예측"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "1b425577",
      "metadata": {
        "id": "1b425577"
      },
      "outputs": [],
      "source": [
        "cl0_fc = cl0[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl0_fc.columns = ['ds', 'y']\n",
        "cl0_fc['ds'] = pd.to_datetime(cl0_fc['ds'])\n",
        "\n",
        "cl1_fc = cl1[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl1_fc.columns = ['ds', 'y']\n",
        "cl1_fc['ds'] = pd.to_datetime(cl1_fc['ds'])\n",
        "\n",
        "cl2_fc = cl2[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl2_fc.columns = ['ds', 'y']\n",
        "cl2_fc['ds'] = pd.to_datetime(cl2_fc['ds'])\n",
        "\n",
        "cl3_fc = cl3[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl3_fc.columns = ['ds', 'y']\n",
        "cl3_fc['ds'] = pd.to_datetime(cl3_fc['ds'])\n",
        "\n",
        "cl4_fc = cl4[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl4_fc.columns = ['ds', 'y']\n",
        "cl4_fc['ds'] = pd.to_datetime(cl4_fc['ds'])\n",
        "\n",
        "cl5_fc = cl5[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl5_fc.columns = ['ds', 'y']\n",
        "cl5_fc['ds'] = pd.to_datetime(cl5_fc['ds'])\n",
        "\n",
        "cl6_fc = cl6[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl6_fc.columns = ['ds', 'y']\n",
        "cl6_fc['ds'] = pd.to_datetime(cl6_fc['ds'])\n",
        "\n",
        "cl7_fc = cl7[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl7_fc.columns = ['ds', 'y']\n",
        "cl7_fc['ds'] = pd.to_datetime(cl7_fc['ds'])\n",
        "\n",
        "cl8_fc = cl8[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl8_fc.columns = ['ds', 'y']\n",
        "cl8_fc['ds'] = pd.to_datetime(cl8_fc['ds'])\n",
        "\n",
        "cl9_fc = cl9[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl9_fc.columns = ['ds', 'y']\n",
        "cl9_fc['ds'] = pd.to_datetime(cl9_fc['ds'])\n",
        "\n",
        "cl10_fc = cl10[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl10_fc.columns = ['ds', 'y']\n",
        "cl10_fc['ds'] = pd.to_datetime(cl10_fc['ds'])\n",
        "\n",
        "cl11_fc = cl11[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl11_fc.columns = ['ds', 'y']\n",
        "cl11_fc['ds'] = pd.to_datetime(cl11_fc['ds'])\n",
        "\n",
        "cl12_fc = cl12[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl12_fc.columns = ['ds', 'y']\n",
        "cl12_fc['ds'] = pd.to_datetime(cl12_fc['ds'])\n",
        "\n",
        "cl13_fc = cl13[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl13_fc.columns = ['ds', 'y']\n",
        "cl13_fc['ds'] = pd.to_datetime(cl13_fc['ds'])\n",
        "\n",
        "cl14_fc = cl14[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl14_fc.columns = ['ds', 'y']\n",
        "cl14_fc['ds'] = pd.to_datetime(cl14_fc['ds'])\n",
        "\n",
        "cl15_fc = cl15[['BKG_DATE', 'ITEM_QTY']].copy()\n",
        "cl15_fc.columns = ['ds', 'y']\n",
        "cl15_fc['ds'] = pd.to_datetime(cl15_fc['ds'])"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "f6a79357",
      "metadata": {
        "id": "f6a79357"
      },
      "source": [
        "#### 클러스터별 Prophet 모델 생성"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3653f319",
      "metadata": {
        "id": "3653f319"
      },
      "outputs": [],
      "source": [
        "m0 = Prophet()\n",
        "m0.fit(cl0_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "32bb0e4e",
      "metadata": {
        "id": "32bb0e4e"
      },
      "outputs": [],
      "source": [
        "m1 = Prophet()\n",
        "m1.fit(cl1_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2936bb9b",
      "metadata": {
        "id": "2936bb9b"
      },
      "outputs": [],
      "source": [
        "m2 = Prophet()\n",
        "m2.fit(cl2_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3ce36528",
      "metadata": {
        "id": "3ce36528"
      },
      "outputs": [],
      "source": [
        "m3 = Prophet()\n",
        "m3.fit(cl3_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "50e8c496",
      "metadata": {
        "id": "50e8c496"
      },
      "outputs": [],
      "source": [
        "m4 = Prophet()\n",
        "m4.fit(cl4_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "289b11d6",
      "metadata": {
        "id": "289b11d6"
      },
      "outputs": [],
      "source": [
        "m5 = Prophet()\n",
        "m5.fit(cl5_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "686e715d",
      "metadata": {
        "id": "686e715d"
      },
      "outputs": [],
      "source": [
        "m6 = Prophet()\n",
        "m6.fit(cl6_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "f6b3ca9c",
      "metadata": {
        "id": "f6b3ca9c"
      },
      "outputs": [],
      "source": [
        "m7 = Prophet()\n",
        "m7.fit(cl7_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "7dd2e797",
      "metadata": {
        "id": "7dd2e797"
      },
      "outputs": [],
      "source": [
        "m8 = Prophet()\n",
        "m8.fit(cl8_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "cf846d1b",
      "metadata": {
        "id": "cf846d1b"
      },
      "outputs": [],
      "source": [
        "m9 = Prophet()\n",
        "m9.fit(cl9_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4b141777",
      "metadata": {
        "id": "4b141777"
      },
      "outputs": [],
      "source": [
        "m10 = Prophet()\n",
        "m10.fit(cl10_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a33d507d",
      "metadata": {
        "id": "a33d507d"
      },
      "outputs": [],
      "source": [
        "m11 = Prophet()\n",
        "m11.fit(cl11_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fcb1fb76",
      "metadata": {
        "id": "fcb1fb76"
      },
      "outputs": [],
      "source": [
        "m12 = Prophet()\n",
        "m12.fit(cl12_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c57cc17b",
      "metadata": {
        "id": "c57cc17b"
      },
      "outputs": [],
      "source": [
        "m13 = Prophet()\n",
        "m13.fit(cl13_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "090d315a",
      "metadata": {
        "id": "090d315a"
      },
      "outputs": [],
      "source": [
        "m14 = Prophet()\n",
        "m14.fit(cl14_fc)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "029282a8",
      "metadata": {
        "id": "029282a8"
      },
      "outputs": [],
      "source": [
        "m15 = Prophet()\n",
        "m15.fit(cl15_fc)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "c88616d9",
      "metadata": {
        "id": "c88616d9"
      },
      "source": [
        "#### Forecasting (향후 100일)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4383e122",
      "metadata": {
        "id": "4383e122"
      },
      "outputs": [],
      "source": [
        "future0 = m0.make_future_dataframe(periods=100)\n",
        "forecast0 = m0.predict(future0)\n",
        "\n",
        "future1 = m1.make_future_dataframe(periods=100)\n",
        "forecast1 = m1.predict(future1)\n",
        "\n",
        "future2 = m2.make_future_dataframe(periods=100)\n",
        "forecast2 = m2.predict(future2)\n",
        "\n",
        "future3 = m3.make_future_dataframe(periods=100)\n",
        "forecast3 = m3.predict(future3)\n",
        "\n",
        "future4 = m4.make_future_dataframe(periods=100)\n",
        "forecast4 = m4.predict(future4)\n",
        "\n",
        "future5 = m5.make_future_dataframe(periods=100)\n",
        "forecast5 = m5.predict(future5)\n",
        "\n",
        "future6 = m6.make_future_dataframe(periods=100)\n",
        "forecast6 = m6.predict(future6)\n",
        "\n",
        "future7 = m7.make_future_dataframe(periods=100)\n",
        "forecast7 = m7.predict(future7)\n",
        "\n",
        "future8 = m8.make_future_dataframe(periods=100)\n",
        "forecast8 = m8.predict(future8)\n",
        "\n",
        "future9 = m9.make_future_dataframe(periods=100)\n",
        "forecast9 = m9.predict(future9)\n",
        "\n",
        "future10 = m10.make_future_dataframe(periods=100)\n",
        "forecast10 = m10.predict(future10)\n",
        "\n",
        "future11 = m11.make_future_dataframe(periods=100)\n",
        "forecast11 = m11.predict(future11)\n",
        "\n",
        "future12 = m12.make_future_dataframe(periods=100)\n",
        "forecast12 = m12.predict(future12)\n",
        "\n",
        "future13 = m13.make_future_dataframe(periods=100)\n",
        "forecast13 = m13.predict(future13)\n",
        "\n",
        "future14 = m14.make_future_dataframe(periods=100)\n",
        "forecast14 = m14.predict(future14)\n",
        "\n",
        "future15 = m15.make_future_dataframe(periods=100)\n",
        "forecast15 = m15.predict(future15)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b3b50c82",
      "metadata": {
        "id": "b3b50c82"
      },
      "source": [
        "#### 시각화 "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "50dceeb4",
      "metadata": {
        "id": "50dceeb4"
      },
      "outputs": [],
      "source": [
        "fig, axes = plt.subplots(nrows=4, ncols=4, figsize=(25,25))\n",
        "axes = [ax for row_axes in axes for ax in row_axes]\n",
        "\n",
        "m0.plot(forecast0, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[0])\n",
        "m1.plot(forecast1, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[1])\n",
        "m2.plot(forecast2, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[2])\n",
        "m3.plot(forecast3, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[3])\n",
        "m4.plot(forecast4, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[4])\n",
        "m5.plot(forecast5, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[5])\n",
        "m6.plot(forecast6, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[6])\n",
        "m7.plot(forecast7, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[7])\n",
        "m8.plot(forecast8, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[8])\n",
        "m9.plot(forecast9, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[9])\n",
        "m10.plot(forecast10, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[10])\n",
        "m11.plot(forecast11, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[11])\n",
        "m12.plot(forecast12, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[12])\n",
        "m13.plot(forecast13, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[13])\n",
        "m14.plot(forecast14, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[14])\n",
        "m15.plot(forecast15, xlabel='Date', ylabel='ITME_QUANTITIY', ax = axes[15])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2c7fdb34",
      "metadata": {
        "id": "2c7fdb34"
      },
      "outputs": [],
      "source": [
        "m0.plot_components(forecast0)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "fd8b52da",
      "metadata": {
        "id": "fd8b52da"
      },
      "source": [
        "### 2) 쇼핑몰별 수량 예측"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "bbdfa280",
      "metadata": {
        "id": "bbdfa280"
      },
      "outputs": [],
      "source": [
        "result_df = result[result['SHPR_CD']==90001302]\n",
        "result_df = result_df.sort_values(\"BKG_DATE\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "dcb08ac3",
      "metadata": {
        "id": "dcb08ac3"
      },
      "outputs": [],
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "plt.plot(result_df['BKG_DATE'], result_df['ITEM_QTY'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "8452b717",
      "metadata": {
        "id": "8452b717"
      },
      "outputs": [],
      "source": [
        "result_df_sample = result_df.rename(columns={'BKG_DATE':'ds', 'ITEM_QTY':'y'})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5abad83f",
      "metadata": {
        "id": "5abad83f"
      },
      "outputs": [],
      "source": [
        "m = Prophet()\n",
        "m.fit(result_df_sample)\n",
        "# Forcasting into the future\n",
        "future = m.make_future_dataframe(periods=365)\n",
        "forecast = m.predict(future)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "ec4bba2e",
      "metadata": {
        "id": "ec4bba2e"
      },
      "source": [
        "#### 시각화"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "dd5cbd91",
      "metadata": {
        "id": "dd5cbd91"
      },
      "outputs": [],
      "source": [
        "figure = m.plot(forecast, xlabel='Date', ylabel='ITEM_QTY')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4492c008",
      "metadata": {
        "id": "4492c008"
      },
      "outputs": [],
      "source": [
        "figure3 = m.plot_components(forecast)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "XdOv-FaiXWNV",
      "metadata": {
        "id": "XdOv-FaiXWNV"
      },
      "source": [
        "#스크래핑"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "mcM9v-ASXS17",
      "metadata": {
        "id": "mcM9v-ASXS17"
      },
      "outputs": [],
      "source": [
        "cj_data_path = \"/CJ 2021년 미래기술 챌린지 이커머스 FC주문 데이터.pickle\"\n",
        "connection_failed_path = \"/바탕화면/connection_failed.txt\"\n",
        "ITEM_NM_dict_path = \"/바탕화면/ITEM_NM_dict.pickle\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4a5whMujXeIN",
      "metadata": {
        "id": "4a5whMujXeIN"
      },
      "outputs": [],
      "source": [
        "manager = Manager()\n",
        "ITEM_NM_dict = manager.dict(zip([], [[]]))\n",
        "not_operational_list = manager.list()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "_zKcDCs_Xo_3",
      "metadata": {
        "id": "_zKcDCs_Xo_3"
      },
      "outputs": [],
      "source": [
        "cj_data_df = pd.read_pickle(cj_data_path)\n",
        "len(cj_data_df)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "4xvff46OXxJj",
      "metadata": {
        "id": "4xvff46OXxJj"
      },
      "source": [
        "- 스크래핑 함수"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "spkxuLVoXpB0",
      "metadata": {
        "id": "spkxuLVoXpB0"
      },
      "outputs": [],
      "source": [
        "cj_data_gen = (i for i in cj_data_df[\"ITEM_CD\"])def scrape(url):\n",
        "    try:\n",
        "        headers = {\n",
        "            \"User-Agent\": \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.54 Safari/537.36\"\n",
        "        }\n",
        "        req = requests.get(url, headers=headers, timeout=5, allow_redirects=True).text\n",
        "        raw_title = fromstring(req).xpath(\"./head/title\")\n",
        "        title = tostring(raw_title[0], encoding=\"unicode\")\n",
        "        if \"상품이 존재하지 않습니다.\" in title or \"현재 운영되고 있지 않습니다.\" in title:\n",
        "            not_operational_list.append(f\"{url} {str(dt.now())}\")\n",
        "        else:\n",
        "            title_tag = str(title)[7:-13]\n",
        "            ITEM_CD = url[38:-1]\n",
        "            if ITEM_CD not in ITEM_NM_dict:\n",
        "                ITEM_NM_dict[ITEM_CD] = title_tag\n",
        "        return False\n",
        "    except Exception:\n",
        "        return url"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "3UcuePYCX1rx",
      "metadata": {
        "id": "3UcuePYCX1rx"
      },
      "source": [
        "- 실패 항목 재 스크래핑 함수"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "dKIjREHmXpEL",
      "metadata": {
        "id": "dKIjREHmXpEL"
      },
      "outputs": [],
      "source": [
        "def excep_scrape(excep_url):\n",
        "    try:\n",
        "        headers = {\n",
        "            \"User-Agent\": \"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.54 Safari/537.36\"\n",
        "        }\n",
        "        req = requests.get(excep_url, headers=headers, timeout=5, allow_redirects=True).text\n",
        "        raw_title = fromstring(req).xpath(\"./head/title\")\n",
        "        title = tostring(raw_title[0], encoding=\"unicode\")\n",
        "        if \"상품이 존재하지 않습니다.\" in title or \"현재 운영되고 있지 않습니다.\" in title:\n",
        "            not_operational_list.append(f\"{excep_url} {str(dt.now())}\")\n",
        "        else:\n",
        "            title_tag = str(title)[7:-13]\n",
        "            ITEM_CD = excep_url[38:-1]\n",
        "            if ITEM_CD not in ITEM_NM_dict:\n",
        "                ITEM_NM_dict[ITEM_CD] = title_tag\n",
        "        return False\n",
        "    except Exception:\n",
        "        with open(connection_failed_path, mode=\"at\", encoding=\"utf-8-sig\") as f:\n",
        "            f.write(f\"{excep_url} {str(dt.now())}\\n\")\n",
        "        return excep_url"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "i2pOcCKKX8Ls",
      "metadata": {
        "id": "i2pOcCKKX8Ls"
      },
      "source": [
        "- 스크래핑 결과 요약 함수"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2BHCjU9rXpG3",
      "metadata": {
        "id": "2BHCjU9rXpG3"
      },
      "outputs": [],
      "source": [
        "def summary(total, excep, failed):\n",
        "    m, s = divmod(time.time() - start_time, 60)\n",
        "    h, m = divmod(m, 60)\n",
        "    print(\"\\n####################SUMMARY####################\")\n",
        "    print(f\"  scraping finished at {str(dt.now())} properly.\")\n",
        "    print(f\"  Total {total - failed} urls successfully passed,\"\n",
        "          f\"\\n  {excep} exception(s) occured,\"\n",
        "          f\"\\n  and finally {failed} url(s) unabled to scrape.\")\n",
        "    print(f\"           --- total progress time {int(h)}h {int(m)}m {int(s)}s ---\")\n",
        "    print(\"###############################################\\n\")"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "f-t1Au2YYDl-",
      "metadata": {
        "id": "f-t1Au2YYDl-"
      },
      "source": [
        "- 스크래핑 본체 함수"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "EtX-UTkzXpMD",
      "metadata": {
        "id": "EtX-UTkzXpMD"
      },
      "outputs": [],
      "source": [
        "def run(proc, total, re):\n",
        "    proc, total, excep, failed = proc, total, 0, 0\n",
        "\n",
        "    print(f\"\\nbrand.naver scraping started at {str(dt.now())}\\n\")\n",
        "    pool = Pool(processes=proc)\n",
        "    if re == 'n':\n",
        "        urls = (f\"https://brand.naver.com/main/products/{str(url)}/\" for url in cj_data_gen)\n",
        "        excep_urls = [i for i in\n",
        "                      tqdm(pool.imap_unordered(scrape, urls), total=total, desc=\"scraping\", leave=False, ascii=True, ncols=100)\n",
        "                      if i is not False]\n",
        "        failed_urls = [i for i in tqdm(pool.imap_unordered(excep_scrape, excep_urls), total=len(excep_urls),\n",
        "                                        desc=\"exception\", leave=False, ascii=True, ncols=100) if i is not False]\n",
        "        write_pickle(ITEM_NM_dict_path, ITEM_NM_dict)\n",
        "        summary(total, len(excep_urls), len(failed_urls))\n",
        "        return\n",
        "    else:\n",
        "        with open(connection_failed_path, mode=\"r\", encoding=\"utf-8-sig\") as f:\n",
        "            urls = [i.split()[0] for i in f.readlines()]\n",
        "            print(f\"{len(urls)} urls are going to rescrape now...\")\n",
        "            excep_urls = [i for i in\n",
        "                          tqdm(pool.imap_unordered(scrape, urls), total=len(urls), desc=\"scraping\", leave=False, ascii=True, ncols=100)\n",
        "                          if i is not False]\n",
        "            failed_urls = [i for i in\n",
        "                            tqdm(pool.imap_unordered(excep_scrape, excep_urls), total=len(excep_urls),\n",
        "                                desc=\"exception\", leave=False, ascii=True, ncols=100) if i is not False]\n",
        "            write_pickle(ITEM_NM_dict_path, ITEM_NM_dict)\n",
        "            summary(len(urls), len(excep_urls), len(failed_urls))\n",
        "        if len(failed_urls) == 0 and os.path.isfile(connection_failed_path):\n",
        "            os.remove(connection_failed_path)\n",
        "            print(\"connection_failed.txt has been deleted.\\n\")\n",
        "        return"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "DPmzew1yXpOW",
      "metadata": {
        "id": "DPmzew1yXpOW"
      },
      "outputs": [],
      "source": [
        "start_time = time.time()\n",
        "\n",
        "#프로세스 수, 스크래핑 길이, 재탐색 여부\n",
        "run(150, len(cj_data_df), 'n')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "yl6LFOnEXpQ1",
      "metadata": {
        "id": "yl6LFOnEXpQ1"
      },
      "outputs": [],
      "source": [
        "ITEM_NM_dict = pd.read_pickle(ITEM_NM_dict_path)\n",
        "print(len(ITEM_NM_dict))\n",
        "print(ITEM_NM_dict)"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [
        "kw6GbP-1a4u7",
        "0a41ea72",
        "b5961c7d",
        "1acfd138",
        "9f59ddc8",
        "d877e40c",
        "0aa53370",
        "7dd12fbb",
        "c4dce2a9",
        "4c12f288",
        "f3417d03",
        "XdOv-FaiXWNV"
      ],
      "name": "CJ_미래기술챌린지_Code.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.8"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}