{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [教學目標]\n",
    "- 以下程式碼將示範如何繪製特定特徵與目標值之間的散佈圖, 更直覺地看出特徵與目標的關係  \n",
    "- 繪製前需要觀察資料, 將異常值排除, 並且轉換成適合的數值單位輔助觀察  \n",
    "- 好的圖可以讓你更快認識資料, 繪圖畫的好也是一種藝術"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [範例重點]\n",
    "- 直接列出的觀察方式 (In[3], Out[3])\n",
    "- 出現異常數值的資料調整方式 (In[4])\n",
    "- 散佈圖異常與其調整方式 (Out[5], In[6], Out[6])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 載入需要的套件\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# 設定 data_path\n",
    "dir_data = './data/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Path of read in data: ./data/application_train.csv\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SK_ID_CURR</th>\n",
       "      <th>TARGET</th>\n",
       "      <th>NAME_CONTRACT_TYPE</th>\n",
       "      <th>CODE_GENDER</th>\n",
       "      <th>FLAG_OWN_CAR</th>\n",
       "      <th>FLAG_OWN_REALTY</th>\n",
       "      <th>CNT_CHILDREN</th>\n",
       "      <th>AMT_INCOME_TOTAL</th>\n",
       "      <th>AMT_CREDIT</th>\n",
       "      <th>AMT_ANNUITY</th>\n",
       "      <th>...</th>\n",
       "      <th>FLAG_DOCUMENT_18</th>\n",
       "      <th>FLAG_DOCUMENT_19</th>\n",
       "      <th>FLAG_DOCUMENT_20</th>\n",
       "      <th>FLAG_DOCUMENT_21</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_HOUR</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_DAY</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_WEEK</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_MON</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_QRT</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_YEAR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>100002</td>\n",
       "      <td>1</td>\n",
       "      <td>Cash loans</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>202500.0</td>\n",
       "      <td>406597.5</td>\n",
       "      <td>24700.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>100003</td>\n",
       "      <td>0</td>\n",
       "      <td>Cash loans</td>\n",
       "      <td>F</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>270000.0</td>\n",
       "      <td>1293502.5</td>\n",
       "      <td>35698.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>100004</td>\n",
       "      <td>0</td>\n",
       "      <td>Revolving loans</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>67500.0</td>\n",
       "      <td>135000.0</td>\n",
       "      <td>6750.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>100006</td>\n",
       "      <td>0</td>\n",
       "      <td>Cash loans</td>\n",
       "      <td>F</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>135000.0</td>\n",
       "      <td>312682.5</td>\n",
       "      <td>29686.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>100007</td>\n",
       "      <td>0</td>\n",
       "      <td>Cash loans</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>121500.0</td>\n",
       "      <td>513000.0</td>\n",
       "      <td>21865.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 122 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SK_ID_CURR  TARGET NAME_CONTRACT_TYPE CODE_GENDER FLAG_OWN_CAR  \\\n",
       "0      100002       1         Cash loans           M            N   \n",
       "1      100003       0         Cash loans           F            N   \n",
       "2      100004       0    Revolving loans           M            Y   \n",
       "3      100006       0         Cash loans           F            N   \n",
       "4      100007       0         Cash loans           M            N   \n",
       "\n",
       "  FLAG_OWN_REALTY  CNT_CHILDREN  AMT_INCOME_TOTAL  AMT_CREDIT  AMT_ANNUITY  \\\n",
       "0               Y             0          202500.0    406597.5      24700.5   \n",
       "1               N             0          270000.0   1293502.5      35698.5   \n",
       "2               Y             0           67500.0    135000.0       6750.0   \n",
       "3               Y             0          135000.0    312682.5      29686.5   \n",
       "4               Y             0          121500.0    513000.0      21865.5   \n",
       "\n",
       "   ...  FLAG_DOCUMENT_18 FLAG_DOCUMENT_19 FLAG_DOCUMENT_20 FLAG_DOCUMENT_21  \\\n",
       "0  ...                 0                0                0                0   \n",
       "1  ...                 0                0                0                0   \n",
       "2  ...                 0                0                0                0   \n",
       "3  ...                 0                0                0                0   \n",
       "4  ...                 0                0                0                0   \n",
       "\n",
       "  AMT_REQ_CREDIT_BUREAU_HOUR AMT_REQ_CREDIT_BUREAU_DAY  \\\n",
       "0                        0.0                       0.0   \n",
       "1                        0.0                       0.0   \n",
       "2                        0.0                       0.0   \n",
       "3                        NaN                       NaN   \n",
       "4                        0.0                       0.0   \n",
       "\n",
       "   AMT_REQ_CREDIT_BUREAU_WEEK  AMT_REQ_CREDIT_BUREAU_MON  \\\n",
       "0                         0.0                        0.0   \n",
       "1                         0.0                        0.0   \n",
       "2                         0.0                        0.0   \n",
       "3                         NaN                        NaN   \n",
       "4                         0.0                        0.0   \n",
       "\n",
       "   AMT_REQ_CREDIT_BUREAU_QRT  AMT_REQ_CREDIT_BUREAU_YEAR  \n",
       "0                        0.0                         1.0  \n",
       "1                        0.0                         0.0  \n",
       "2                        0.0                         0.0  \n",
       "3                        NaN                         NaN  \n",
       "4                        0.0                         0.0  \n",
       "\n",
       "[5 rows x 122 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 讀取 csv 資料檔, 並觀察前幾筆資料 ( .head() )\n",
    "f_app = os.path.join(dir_data, 'application_train.csv')\n",
    "print('Path of read in data: %s' % (f_app))\n",
    "app_train = pd.read_csv(f_app)\n",
    "app_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0           -637\n",
       "1          -1188\n",
       "2           -225\n",
       "3          -3039\n",
       "4          -3038\n",
       "5          -1588\n",
       "6          -3130\n",
       "7           -449\n",
       "8         365243\n",
       "9          -2019\n",
       "10          -679\n",
       "11        365243\n",
       "12         -2717\n",
       "13         -3028\n",
       "14          -203\n",
       "15         -1157\n",
       "16         -1317\n",
       "17          -191\n",
       "18         -7804\n",
       "19         -2038\n",
       "20         -4286\n",
       "21         -1652\n",
       "22         -4306\n",
       "23        365243\n",
       "24          -746\n",
       "25         -3494\n",
       "26         -2628\n",
       "27         -1234\n",
       "28         -1796\n",
       "29         -1010\n",
       "           ...  \n",
       "307481     -3147\n",
       "307482      -226\n",
       "307483    365243\n",
       "307484      -328\n",
       "307485      -670\n",
       "307486     -1185\n",
       "307487    365243\n",
       "307488     -1218\n",
       "307489      -286\n",
       "307490     -1928\n",
       "307491     -1953\n",
       "307492     -1618\n",
       "307493     -2306\n",
       "307494     -6573\n",
       "307495     -7438\n",
       "307496     -2178\n",
       "307497     -1222\n",
       "307498     -3689\n",
       "307499     -8694\n",
       "307500     -5326\n",
       "307501     -1046\n",
       "307502     -8736\n",
       "307503      -399\n",
       "307504     -7258\n",
       "307505    365243\n",
       "307506      -236\n",
       "307507    365243\n",
       "307508     -7921\n",
       "307509     -4786\n",
       "307510     -1262\n",
       "Name: DAYS_EMPLOYED, Length: 307511, dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 觀察聘雇天數 'DAYS_EMPLOYED' 內的數值\n",
    "app_train['DAYS_EMPLOYED']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 由於其他天數都是負值, 且聘僱日數不太可能是 365243 (大約 1000年), 算是異常數字 \n",
    "# 因此我們推斷這份資料中, DAYS_EMPLOYED 的欄位如果是 365243, 應該是對應到空缺值, 繪圖時應該予以忽略\n",
    "sub_df = app_train[app_train['DAYS_EMPLOYED'] != 365243]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation: 0.0130\n"
     ]
    }
   ],
   "source": [
    "# 如果直接畫散布圖 - 看不出任何趨勢或形態\n",
    "plt.plot(sub_df['DAYS_EMPLOYED'] / (-365), sub_df['AMT_INCOME_TOTAL'], '.')\n",
    "plt.xlabel('Days of employed (year)')\n",
    "plt.ylabel('AMT_INCOME_TOTAL (raw)')\n",
    "plt.show()\n",
    "corr = np.corrcoef(sub_df['DAYS_EMPLOYED'] / (-365), sub_df['AMT_INCOME_TOTAL'])\n",
    "print(\"Correlation: %.4f\" % (corr[0][1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation: 0.0380\n"
     ]
    }
   ],
   "source": [
    "# 通常可以對數值範圍較大的取 log: 發現雖然沒有相關，但是受雇越久的人，AMT_INCOME_TOTAL 的 variance 越小\n",
    "plt.plot(sub_df['DAYS_EMPLOYED'] / (-365), np.log10(sub_df['AMT_INCOME_TOTAL'] ), '.')\n",
    "plt.xlabel('Days of employed (year)')\n",
    "plt.ylabel('AMT_INCOME_TOTAL (log-scale)')\n",
    "plt.show()\n",
    "corr = np.corrcoef(sub_df['DAYS_EMPLOYED'] / (-365), np.log10(sub_df['AMT_INCOME_TOTAL']))\n",
    "print(\"Correlation: %.4f\" % (corr[0][1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# [作業目標]\n",
    "- 請同學試著使用 pandas.corr() 這個函數來顯示相關係數並加以觀察結果 \n",
    "- 思考1 : 使用 pandas 有沒有什麼寫法, 可以顯示欄位中最大的幾筆, 以及最小幾筆呢? (Hint: 排序後列出前幾筆/後幾筆)\n",
    "- 思考2 : 試著使用散佈圖, 顯示相關度最大/最小的特徵與目標值的關係, 如果圖形不明顯, 是否有調整的方法?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [作業重點]\n",
    "- 綜合前幾單元的作法, 試試看是否能夠用繪圖顯示出特徵與目標的相關性"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 載入需要的套件\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# 設定 data_path\n",
    "dir_data = './data/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(307511, 122)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 讀取資料檔\n",
    "f_app_train = os.path.join(dir_data, 'application_train.csv')\n",
    "app_train = pd.read_csv(f_app_train)\n",
    "app_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(307511, 122)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SK_ID_CURR</th>\n",
       "      <th>TARGET</th>\n",
       "      <th>NAME_CONTRACT_TYPE</th>\n",
       "      <th>CODE_GENDER</th>\n",
       "      <th>FLAG_OWN_CAR</th>\n",
       "      <th>FLAG_OWN_REALTY</th>\n",
       "      <th>CNT_CHILDREN</th>\n",
       "      <th>AMT_INCOME_TOTAL</th>\n",
       "      <th>AMT_CREDIT</th>\n",
       "      <th>AMT_ANNUITY</th>\n",
       "      <th>...</th>\n",
       "      <th>FLAG_DOCUMENT_18</th>\n",
       "      <th>FLAG_DOCUMENT_19</th>\n",
       "      <th>FLAG_DOCUMENT_20</th>\n",
       "      <th>FLAG_DOCUMENT_21</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_HOUR</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_DAY</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_WEEK</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_MON</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_QRT</th>\n",
       "      <th>AMT_REQ_CREDIT_BUREAU_YEAR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>100002</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>202500.0</td>\n",
       "      <td>406597.5</td>\n",
       "      <td>24700.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>100003</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>270000.0</td>\n",
       "      <td>1293502.5</td>\n",
       "      <td>35698.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>100004</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>67500.0</td>\n",
       "      <td>135000.0</td>\n",
       "      <td>6750.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>100006</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>135000.0</td>\n",
       "      <td>312682.5</td>\n",
       "      <td>29686.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>100007</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>121500.0</td>\n",
       "      <td>513000.0</td>\n",
       "      <td>21865.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 122 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SK_ID_CURR  TARGET  NAME_CONTRACT_TYPE CODE_GENDER  FLAG_OWN_CAR  \\\n",
       "0      100002       1                   0           M             0   \n",
       "1      100003       0                   0           F             0   \n",
       "2      100004       0                   1           M             1   \n",
       "3      100006       0                   0           F             0   \n",
       "4      100007       0                   0           M             0   \n",
       "\n",
       "   FLAG_OWN_REALTY  CNT_CHILDREN  AMT_INCOME_TOTAL  AMT_CREDIT  AMT_ANNUITY  \\\n",
       "0                1             0          202500.0    406597.5      24700.5   \n",
       "1                0             0          270000.0   1293502.5      35698.5   \n",
       "2                1             0           67500.0    135000.0       6750.0   \n",
       "3                1             0          135000.0    312682.5      29686.5   \n",
       "4                1             0          121500.0    513000.0      21865.5   \n",
       "\n",
       "   ...  FLAG_DOCUMENT_18 FLAG_DOCUMENT_19 FLAG_DOCUMENT_20 FLAG_DOCUMENT_21  \\\n",
       "0  ...                 0                0                0                0   \n",
       "1  ...                 0                0                0                0   \n",
       "2  ...                 0                0                0                0   \n",
       "3  ...                 0                0                0                0   \n",
       "4  ...                 0                0                0                0   \n",
       "\n",
       "  AMT_REQ_CREDIT_BUREAU_HOUR AMT_REQ_CREDIT_BUREAU_DAY  \\\n",
       "0                        0.0                       0.0   \n",
       "1                        0.0                       0.0   \n",
       "2                        0.0                       0.0   \n",
       "3                        NaN                       NaN   \n",
       "4                        0.0                       0.0   \n",
       "\n",
       "   AMT_REQ_CREDIT_BUREAU_WEEK  AMT_REQ_CREDIT_BUREAU_MON  \\\n",
       "0                         0.0                        0.0   \n",
       "1                         0.0                        0.0   \n",
       "2                         0.0                        0.0   \n",
       "3                         NaN                        NaN   \n",
       "4                         0.0                        0.0   \n",
       "\n",
       "   AMT_REQ_CREDIT_BUREAU_QRT  AMT_REQ_CREDIT_BUREAU_YEAR  \n",
       "0                        0.0                         1.0  \n",
       "1                        0.0                         0.0  \n",
       "2                        0.0                         0.0  \n",
       "3                        NaN                         NaN  \n",
       "4                        0.0                         0.0  \n",
       "\n",
       "[5 rows x 122 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 將只有兩種值的類別型欄位, 做 Label Encoder, 計算相關係數時讓這些欄位可以被包含在內\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder()\n",
    "\n",
    "# 檢查每一個 column\n",
    "for col in app_train:\n",
    "    if app_train[col].dtype == 'object':\n",
    "        # 如果只有兩種值的類別型欄位\n",
    "        if len(list(app_train[col].unique())) <= 2:\n",
    "            # 就做 Label Encoder, 以加入相關係數檢查\n",
    "            app_train[col] = le.fit_transform(app_train[col])            \n",
    "print(app_train.shape)\n",
    "app_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 受雇日數為異常值的資料, 另外設一個欄位記錄, 並將異常的日數轉成空值 (np.nan)\n",
    "app_train['DAYS_EMPLOYED_ANOM'] = app_train[\"DAYS_EMPLOYED\"] == 365243\n",
    "app_train['DAYS_EMPLOYED'].replace({365243: np.nan}, inplace = True)\n",
    "\n",
    "# 出生日數 (DAYS_BIRTH) 取絕對值 \n",
    "app_train['DAYS_BIRTH'] = abs(app_train['DAYS_BIRTH'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 相關係數\n",
    "一樣，pandas 很貼心地讓我們可以非常容易計算相關係數"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SK_ID_CURR                    -0.002108\n",
       "TARGET                         1.000000\n",
       "NAME_CONTRACT_TYPE            -0.030896\n",
       "FLAG_OWN_CAR                  -0.021851\n",
       "FLAG_OWN_REALTY               -0.006148\n",
       "CNT_CHILDREN                   0.019187\n",
       "AMT_INCOME_TOTAL              -0.003982\n",
       "AMT_CREDIT                    -0.030369\n",
       "AMT_ANNUITY                   -0.012817\n",
       "AMT_GOODS_PRICE               -0.039645\n",
       "REGION_POPULATION_RELATIVE    -0.037227\n",
       "DAYS_BIRTH                    -0.078239\n",
       "DAYS_EMPLOYED                  0.074958\n",
       "DAYS_REGISTRATION              0.041975\n",
       "DAYS_ID_PUBLISH                0.051457\n",
       "OWN_CAR_AGE                    0.037612\n",
       "FLAG_MOBIL                     0.000534\n",
       "FLAG_EMP_PHONE                 0.045982\n",
       "FLAG_WORK_PHONE                0.028524\n",
       "FLAG_CONT_MOBILE               0.000370\n",
       "FLAG_PHONE                    -0.023806\n",
       "FLAG_EMAIL                    -0.001758\n",
       "CNT_FAM_MEMBERS                0.009308\n",
       "REGION_RATING_CLIENT           0.058899\n",
       "REGION_RATING_CLIENT_W_CITY    0.060893\n",
       "HOUR_APPR_PROCESS_START       -0.024166\n",
       "REG_REGION_NOT_LIVE_REGION     0.005576\n",
       "REG_REGION_NOT_WORK_REGION     0.006942\n",
       "LIVE_REGION_NOT_WORK_REGION    0.002819\n",
       "REG_CITY_NOT_LIVE_CITY         0.044395\n",
       "                                 ...   \n",
       "OBS_60_CNT_SOCIAL_CIRCLE       0.009022\n",
       "DEF_60_CNT_SOCIAL_CIRCLE       0.031276\n",
       "DAYS_LAST_PHONE_CHANGE         0.055218\n",
       "FLAG_DOCUMENT_2                0.005417\n",
       "FLAG_DOCUMENT_3                0.044346\n",
       "FLAG_DOCUMENT_4               -0.002672\n",
       "FLAG_DOCUMENT_5               -0.000316\n",
       "FLAG_DOCUMENT_6               -0.028602\n",
       "FLAG_DOCUMENT_7               -0.001520\n",
       "FLAG_DOCUMENT_8               -0.008040\n",
       "FLAG_DOCUMENT_9               -0.004352\n",
       "FLAG_DOCUMENT_10              -0.001414\n",
       "FLAG_DOCUMENT_11              -0.004229\n",
       "FLAG_DOCUMENT_12              -0.000756\n",
       "FLAG_DOCUMENT_13              -0.011583\n",
       "FLAG_DOCUMENT_14              -0.009464\n",
       "FLAG_DOCUMENT_15              -0.006536\n",
       "FLAG_DOCUMENT_16              -0.011615\n",
       "FLAG_DOCUMENT_17              -0.003378\n",
       "FLAG_DOCUMENT_18              -0.007952\n",
       "FLAG_DOCUMENT_19              -0.001358\n",
       "FLAG_DOCUMENT_20               0.000215\n",
       "FLAG_DOCUMENT_21               0.003709\n",
       "AMT_REQ_CREDIT_BUREAU_HOUR     0.000930\n",
       "AMT_REQ_CREDIT_BUREAU_DAY      0.002704\n",
       "AMT_REQ_CREDIT_BUREAU_WEEK     0.000788\n",
       "AMT_REQ_CREDIT_BUREAU_MON     -0.012462\n",
       "AMT_REQ_CREDIT_BUREAU_QRT     -0.002022\n",
       "AMT_REQ_CREDIT_BUREAU_YEAR     0.019930\n",
       "DAYS_EMPLOYED_ANOM            -0.045987\n",
       "Name: TARGET, Length: 110, dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 觀察相關係數\n",
    "app_train.corr()['TARGET']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 練習時間\n",
    "列出目標 (TARGET) 與所有欄位之間相關係數，數值最大以及最小各 15 個\n",
    "\n",
    "通過相關係數的結果觀察有興趣的欄位與 TARGET 或其他欄位的相關係數，並嘗試找出有趣的訊息\n",
    "- 最好的方式當然是畫圖，舉例來說，我們知道  EXT_SOURCE_3 這個欄位和 TARGET 之間的相關係數是 -0.178919 (在已經這個資料集已經是最負的了！)，那我們可以 EXT_SOURCE_3  為 x 軸， TARGET 為 y 軸，把資料給畫出來"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "hw_solution_1 = app_train.corr()['TARGET'].sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "EXT_SOURCE_3                 -0.178919\n",
       "EXT_SOURCE_2                 -0.160472\n",
       "EXT_SOURCE_1                 -0.155317\n",
       "DAYS_BIRTH                   -0.078239\n",
       "DAYS_EMPLOYED_ANOM           -0.045987\n",
       "FLOORSMAX_AVG                -0.044003\n",
       "FLOORSMAX_MEDI               -0.043768\n",
       "FLOORSMAX_MODE               -0.043226\n",
       "AMT_GOODS_PRICE              -0.039645\n",
       "REGION_POPULATION_RELATIVE   -0.037227\n",
       "ELEVATORS_AVG                -0.034199\n",
       "ELEVATORS_MEDI               -0.033863\n",
       "FLOORSMIN_AVG                -0.033614\n",
       "FLOORSMIN_MEDI               -0.033394\n",
       "LIVINGAREA_AVG               -0.032997\n",
       "Name: TARGET, dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hw_solution_1.head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DEF_60_CNT_SOCIAL_CIRCLE       0.031276\n",
       "DEF_30_CNT_SOCIAL_CIRCLE       0.032248\n",
       "LIVE_CITY_NOT_WORK_CITY        0.032518\n",
       "OWN_CAR_AGE                    0.037612\n",
       "DAYS_REGISTRATION              0.041975\n",
       "FLAG_DOCUMENT_3                0.044346\n",
       "REG_CITY_NOT_LIVE_CITY         0.044395\n",
       "FLAG_EMP_PHONE                 0.045982\n",
       "REG_CITY_NOT_WORK_CITY         0.050994\n",
       "DAYS_ID_PUBLISH                0.051457\n",
       "DAYS_LAST_PHONE_CHANGE         0.055218\n",
       "REGION_RATING_CLIENT           0.058899\n",
       "REGION_RATING_CLIENT_W_CITY    0.060893\n",
       "DAYS_EMPLOYED                  0.074958\n",
       "TARGET                         1.000000\n",
       "Name: TARGET, dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hw_solution_1.tail(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(app_train['EXT_SOURCE_3'] , app_train['TARGET'], '.')\n",
    "plt.xlabel('EXT_SOURCE_3')\n",
    "plt.ylabel('TARGET')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "app_train.boxplot('EXT_SOURCE_3',by='TARGET')\n",
    "plt.title('EXT_SOURCE_3 corr')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2bf9cc18eb8>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "app_train.boxplot('DAYS_EMPLOYED',by='TARGET')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2bf3123a630>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_data = app_train\n",
    "plot_data['YEAR_EMPLOYED'] = plot_data['DAYS_EMPLOYED']/365\n",
    "plot_data.boxplot(column='YEAR_EMPLOYED', by = 'TARGET')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2bf31437cf8>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEcCAYAAADdtCNzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3XuYHFW97vHvuyGAXA634AAhELfgBVRA5wQRkEEBAeXECwqBE8GDxgu4URFFjoKiyFGQvVVEyJEQUC5y5GKOcotu2ohAJGGHS4hghCAxYCDhNoCyg7/9x1odKp01051JT3omeT/PU0+6Vq1atbqm0r+qtVZVKSIwMzNr9E+droCZmQ1NDhBmZlbkAGFmZkUOEGZmVuQAYWZmRQ4QZmZW5ABhq42kKZK+0el6dFp/+0HSMZJuWd11MitxgFgLSZov6QVJvZKelPRLSaM7Xa8qSSFpx07XY7iSNCf/fXslvSTpb5X5Uyr5evK+/kLD+mNyen2d+ZJOLmznCEkzJD0naVH+/ClJysunSHqxUk6vpLsk7VOZf65hW72Sth/8vWTNOECsvQ6NiI2BbYC/At/vcH0GjZK16liPiF0iYuP8N/4tcHx9PiK+Wcl6NLAk/1uyWS7jMOArkg6oL5B0IvBd4Cxga6AL+ASwF7BepYxvV7a9cUTsGhG/rdRvl+q28vTnVd8LtqrWqv80tqKI+BvwM2DnepqkTSVdIulxSQ9L+nL9B1bSDyX9rJL3W5J+nX+EeyQtkHSKpCfyWedRfW1b0sckzZO0RNJUSdvm9Ok5y135bPLwwrrrSPpO3s5Dko7PZ6Hr5uU1SWdI+h3wPPDPkrbN21mSt/uxSnnLNfvUv0tlfr6kL0m6L191XSRpg8ry90iaLekpSbdKelNl2e6S7pT0rKSfAsvW63vX6PuSnpb0B0nvzIkflDSrIeOJkq5tUl5fG9mQ9MN/HLCTpO6+8kbETGAOsFted1PgdOBTEfGziHg2kv+IiKMi4u8DqZMNLQ4Qa7n8I3E4cHsl+fvApsA/A/sCHwY+kpedCLxJqa18H+BY4Oh4+ZktWwMjgVGks9JJkl5b2O47gDOBD5GuYh4GrgCIiLfnbLvms8mfFqr+MeBg0g/Wm4H3FvJMACYCm+TyLwcWANuSfhi/Wf/xbdFRwLuAVwOvAb6cv8ubgcnAx4EtgQuAqZLWl7QecC3wY2AL4P8BH2iynT2AB0n78TTgaklbAFOBV0l6fSXv/8xlD8QHgN5cpxtJf+ciSW8F3gDMy0l7AusDPx/gtm04iAhPa9kEzCf9MDwFLAUWAm/My9YB/g7sXMn/caBWmR9LapZ4GBhfSe/J5W1USbsS+Er+PAX4Rv58IanpoZ5vY+A/gTF5PoAd+/kO/w58vDK/f15n3TxfA06vLB8NvARsUkk7E5jSWLfKd1nQsM8+UZk/BPhT/vxD4OsN9bufFFzfnvevKsturW6rYb1jCvl/D0yobOuM/HkX4Elg/SZ/7xrw0UL6r4B/y5/HA48DI/L8mLw/nwJeyJ/PrteLFJgeayjv1kr+t1f2699yen26uGG9+rbW7fT/DU/LT76CWHu9NyI2I50FHg/8RlL97H890o9/3cOkKwIAIuL3pDNckQJA1ZMR8VzDutsWtr9tdRsR0Qssrm6niW2BRyrzjxTyVNO2BZZExLMNdWt1e43lVb/XDsCJuXnpKUlPkQLStnn6S+Rfwsq6/Snlr2/rYuDI3Ak8AbgyBtCcozQoYT/g0pz0c1LT17sbso4kBe/Pk4LmiJy+GBhZb9IDiIi35WNqMcu3TpwdEZtVpr76O2yIcYBYy0XESxFxNensem/gCdKZ/A6VbNsDf6nPSDqOFFgWAsuNfgE2l7RRw7oLC5teWN1GXmfL6naaeBTYrjJfGoVV/ZFdCGwhaZOGutW39xywYWXZ1oXyqtuofq9HSGf11R/BDSPi8lzPUfVRPZV1+1PKvxAgIm4HXgT2AY5k4M1LE0j///+/pMdIAX8DCs1M+Rj5DulK4FM5+TbSlea4AW7fhgEHiLVc7lweB2wOzI2Il0hXBWdI2kTSDsDngJ/k/K8BvkFqYpgAfEHSbg3Ffk3SermP4j2kNu5GlwEfkbSbpPWBbwIzImJ+Xv5XUh9IX64ETpA0StJmwBf7+54R8QipCeRMSRvkTuRjefkMejZwiKQt8pXUZwrFHCdpu9wfcApQ7xv5v8AnJO2R9+dGkt6dg9FtpGa3f5G0rqT3k5ro+vPKnH+EpA8Crweuqyy/BDgXWBoRA71n4sPA10h9OPXpA8C7JW3Zxzr/h/T33iAinsrrnyfpMEkbS/qnfCxs1Mf6Ntx0uo3L06pNpLbxF0h9Co+R2nw3zsumkM42eyvTXQ3r/D1PfyOdTV8PvJ8UEB4n/bhNIZ1MrAv8R54Wk8665+dpfXK7PenMfSnwZ15uO/8GqYOz2s7/CeBPpP6MXwBvzev25voszfX8UOF7/ybnWQw8BHwW+Efle/4jr1+fPx/4YC5vSd7uA7mM0aSz55/m7/RiLm9B3tYReR+9mMtcSurU37BSn4OAO0ht7I+SguImpGDwu7xO5PrOyPujh0o/Ry7nmJx/bq7bA6RAXN8vvbmOAfys8ncOYGylnB2ByJ+fy2X1kq4U/14p75SG7c8hNTm+OufZs7LsqJx2diXtB3mfPk86XmaQBgas13AMPlfZB/8A7iadeKxDunIN0jF0feV7/ifLH78Xka5wexrqfBFweaf/L66JU8cr4GkV/4Dpx3n//HlrUgCod2JOoY/O0Lx8KjCLNGpmvTwdBHy3j/K3yPMX5W29gtS5+QxpVFAPLweIxcCRlXK+Qe4Q7qc+Y1i+o7k7/7AcUMj3EumH/oM57WDg4UqeFb47K3Y813I9J1XS9gfmV+ZPJF3NLAIOJfW77E668mjWObxn/mH7IqktX8BbSP0GK9SnYd1l9S/sl1cAzwI7VfIuBm6qrL8sQDSUW6PQYd1HHaYBJ1XmLyAFrsa0SU3KeTWpM/0cYJuc9lrSVeRmjd+vyd9wAvBH4BV5/p35b7NVp/8vromTm5jWIBHxGGm4YmOTzwok7Q8cAIyLiBkR8WKeboiIE/pY7bOkH7xjI+KxiHghUjv7GcB3GvJ+m9TUtG5jISvxfZYbe1/xYdLInhpwjKRRpOGg1wxgM98Dxqtw13Z1rD/pDPmFSFod638WacTOtyLiibzurIj40ADqWfVJ4I6I+GMl7WLS8ON9V7HsqumkUVh1+wDfKqRNp39fA26NiM9FxKMAEXF/RBwZqamqZRHxY9IIsdMlvYIUoP4lIh5fmXKsNQ4QaxBJ25HOpOc1y0s6U54REQua5nzZAcBVEfGPhvQrSR2p1U7cq0lXFsesRPnLKYy9r/sw6Z6J15KGm84mndmeOoDN/IXUh/DVwrIBj/XP95fsSboJsW0kzQdOIF3ZVD1P6sc5o42bmw7slfsWRpL6Fq4ExlbSXkfzALE/7d0PnwD+F+kYuDcirmhj2VbhALFmuFbSs6TRNItIZ9N1n68Ov5R0cU4fSeqzACB3zj6V7979Wx/bGUlqX29UT3swIuojiwL4CnBq7oReGU9IeoHUwXse6Uazej33Jo1+ujQidiGdTX4zIj4SEc+s5HbqzgQOlbRLQ/pI4ImIWBoRYyLiV/ku6aeUnmX19kJZdZuT/n+V9tdAPUFqktkUuLnhhjlIZ9PbSzq4TdubQRrZ9UbSlcItEfE8qc+nnvZwNH8sxpa0cT/kk5pTSYHnk+0q11bkALFmeG9EbEJq034d6Yetrq8x6ItJdzADEBFLIo1hfwvprLnkieo6FdtUli8TEdeROqonruT36WvsPaS7s2+KiPq2LqPv5wi1JDdPnEtqTqpambH+jZ4kdcaW9tdAjWz4W86tLsxNXl/Pk4olrIRIj2H5PalJ6e2kZzoB3FJJa3b1AA3HWpvMId1z084AbA0cINYgEfEbUsfe2S1k/zXw33OzVKt+BXxAKz747kOkq5cHCut8GfjfLH+PQVNRGHuf25w/BOwr6bE8fv+zwK6Sdl2Z8gvOIt049pZK2oDH+ucz7dto/liNdruIdIXxvjaVV++H2IeXA8RvK2mtBIhfsfr3g7WBA8Sa59+AAwr3JiwnIm4CbiY1T+2R71sYQRpq2pd/Bf4bcKGkrfP9BONJAeCkiIjGFSKiBtzDwM/yl429Jz1v6SXSgwXrY/dfT/rB6vM5Qq3InaXfoXLjX6z6WP8vkDrRT6rfWyBpV0nLtZnn/VidBnz2HxFLSf0p/d4XshKmkwLnaOC+nHYL6cpuN1oLEKcBb5N0Vr7HBEk7SvpJvofFhigHiDVMbi65hNT+D+nHtfqc/Woz0PtJ9x/8hDR+/yHSWPeD+ih7MWnM+gakH4vFpLHsE6L8QL26L5OGyA7EL0nNNR8jBZmLIuLPeRTVY3nk1rnAUasyYir7LikALRMR3yZ9xy+Q+nf+Smrr/yLpxrs+RcStwDvy9KCkJcAklr/pbRTpPoLq9Oo+inyq4W/5uT7y1e/gbodbSVckM+onAPk4eBxY1DCSqigi/kTqsB8DzJH0NHAVMJM0XNeGKBVO+szMzHwFYWZmZQ4QtlpJOqqhmaQ+zel03QZCy7/aszr1+aKkoUTS+X3U//yVLOf6Pso5pfnaNlS5icnMzIp8BWFmZkWrOupjUIwcOTLGjBnT6WqscZ577jk22shPYrbhw8fs4Jg1a9YTEbFVs3xDMkCMGTOGmTNndroaa5xarUZPT0+nq2HWMh+zg0NSs7caAm5iMjOzPjhAmJlZkQOEmZkVOUCYmVlR0wCRHx72e0l35ZuCvlbIs76kn0qaJ2mGpDGVZV/K6fdLeld7q29mZoOllVFMfwfeERG9+Wmft0i6PiJur+Q5lvRs9h0lHUF6LeHhknYmvfB9F2Bb4FeSXhMRLzVuxAZP6eGgvkHShjIfs0ND0yuI/B7d3jw7Ik+Nf6lxpHfiQnq14DvzI4vHAVdExN8j4iHSqyPHtqXm1pK+nhy9Ck+UNhtUPmaHjpb6ICStI2k26XHH0yJiRkOWUaQXxtSfR/806TWDy9KzBTnNVrOI4Oabb/ZZmA0bPmY7r6Ub5XKT0G755R7XSHpDRNxbyVIK7dFP+gokTSS/mrKrq4tardZK1axFtVqN3t7e5far97ENZT5mO2+lH9Yn6TTguYg4u5J2I/DViLgtv7TlMWAr4GSAiDizMV9/2+ju7g7fSd0e9cvyiFh2V2o1zWyo8TE7+CTNiojuZvmaXkFI2gr4z4h4Kr8TeH9SJ3TVVNLbvm4DDgP+PSJC0lTgMknnkDqpdyK9BN1WM7ff2nDjY7bzWmli2ga4WNI6pD6LKyPiF5JOB2ZGxFTgQuDHkuYBS0gjl4iIOZKuJL2ecilwnEcwrV4R4REhNqz4mB06huT7INzENDj84DMbbnzMDo5Wm5h8J7WZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZUSuvHB0NXAJsDfwDmBQR323IcxJwVKXM1wNbRcQSSfOBZ4GXgKWtvKTCzMw6r5VXji4FToyIOyVtAsySNC0i7qtniIizgLMAJB0KfDYillTK2C8inmhnxc3MbHA1bWKKiEcj4s78+VlgLjCqn1XGA5e3p3pmZtYprVxBLCNpDLA7MKOP5RsCBwHHV5IDuElSABdExKQ+1p0ITATo6uqiVqutTNWsBb29vd6vNqz4mO2slgOEpI2Bq4DPRMQzfWQ7FPhdQ/PSXhGxUNIrgWmS/hAR0xtXzIFjEkB3d3f4ReXt5xfA23DjY7azWhrFJGkEKThcGhFX95P1CBqalyJiYf53EXANMHZgVTUzs9WpaYCQJOBCYG5EnNNPvk2BfYGfV9I2yh3bSNoIOBC4d1UrbWZmg6+VJqa9gAnAPZJm57RTgO0BIuL8nPY+4KaIeK6ybhdwTYoxrAtcFhE3tKPiZmY2uJoGiIi4BVAL+aYAUxrSHgR2HWDdzMysg3wntZmZFTlAmJlZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVtfJGudGSbpY0V9IcSScU8vRIelrS7DydWll2kKT7Jc2TdHK7v4CZmQ2OVt4otxQ4MSLuzK8PnSVpWkTc15DvtxHxnmqCpHWAHwAHAAuAOyRNLaxrZmZDTNMriIh4NCLuzJ+fBeYCo1osfywwLyIejIgXgSuAcQOtrJmZrT4r1QchaQywOzCjsHhPSXdJul7SLjltFPBIJc8CWg8uZmbWQa00MQEgaWPgKuAzEfFMw+I7gR0iolfSIcC1wE6U32UdfZQ/EZgI0NXVRa1Wa7Vq1qLe3l7vVxtWfMx2VksBQtIIUnC4NCKublxeDRgRcZ2k8ySNJF0xjK5k3Q5YWNpGREwCJgF0d3dHT09Pq9/BWlSr1fB+teHEx2xntTKKScCFwNyIOKePPFvnfEgam8tdDNwB7CTpVZLWA44Aprar8mZmNnhauYLYC5gA3CNpdk47BdgeICLOBw4DPilpKfACcEREBLBU0vHAjcA6wOSImNPm72BmZoOgaYCIiFso9yVU85wLnNvHsuuA6wZUOzMz6xjfSW1mZkUOEGZmVuQAYWZmRQ4QZmZW5ABhZmZFDhBmZlbkAGFmZkUOEGZmVuQAYWZmRQ4QZmZW5ABhZmZFDhBmZlbkAGFmZkUOEGZmVuQAYWZmRQ4QZmZW1MorR0dLulnSXElzJJ1QyHOUpLvzdKukXSvL5ku6R9JsSTPb/QXMzGxwtPLK0aXAiRFxp6RNgFmSpkXEfZU8DwH7RsSTkg4GJgF7VJbvFxFPtK/aZmY22Fp55eijwKP587OS5gKjgPsqeW6trHI7sF2b62lmZqtZK1cQy0gaA+wOzOgn27HA9ZX5AG6SFMAFETGpj7InAhMBurq6qNVqK1M1a0Fvb6/3qw0rPmY7SxHRWkZpY+A3wBkRcXUfefYDzgP2jojFOW3biFgo6ZXANODTETG9v211d3fHzJnurmi3Wq1GT09Pp6th1jIfs4ND0qyI6G6Wr6VRTJJGAFcBl/YTHN4E/AgYVw8OABGxMP+7CLgGGNvKNs3MrLNaGcUk4EJgbkSc00ee7YGrgQkR8UAlfaPcsY2kjYADgXvbUXEzMxtcrfRB7AVMAO6RNDunnQJsDxAR5wOnAlsC56V4wtJ8+dIFXJPT1gUui4gb2voNzMxsULQyiukWQE3yfBT4aCH9QWDXFdcwM7OhzndSm5lZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVlRK2+UGy3pZklzJc2RdEIhjyR9T9I8SXdLenNl2dGS/pino9v9BczMbHC08ka5pcCJEXFnfn3oLEnTIuK+Sp6DgZ3ytAfwQ2APSVsApwHdQOR1p0bEk239FmZm1nZNryAi4tGIuDN/fhaYC4xqyDYOuCSS24HNJG0DvAuYFhFLclCYBhzU1m9gZmaDYqX6ICSNAXYHZjQsGgU8UplfkNP6SjczsyGulSYmACRtDFwFfCYinmlcXFgl+kkvlT8RmAjQ1dVFrVZrtWprpU8//OmBrXjxyq/y/R2+P7BtmVX4mB1+WgoQkkaQgsOlEXF1IcsCYHRlfjtgYU7vaUivlbYREZOASQDd3d3R09NTymbZPdyz0uvUajW8X61TfMwOP62MYhJwITA3Is7pI9tU4MN5NNNbgacj4lHgRuBASZtL2hw4MKeZmdkQ18oVxF7ABOAeSbNz2inA9gARcT5wHXAIMA94HvhIXrZE0teBO/J6p0fEkvZV38zMBkvTABERt1DuS6jmCeC4PpZNBiYPqHZmZtYxvpPazMyKHCDMzKzIAcLMzIocIMzMrMgBwszMihwgzMysyAHCzMyKHCDMzKzIAcLMzIocIMzMrMgBwszMihwgzMysyAHCzMyKHCDMzKzIAcLMzIocIMzMrKjpC4MkTQbeAyyKiDcUlp8EHFUp7/XAVvltcvOBZ4GXgKUR0d2uipuZ2eBq5QpiCnBQXwsj4qyI2C0idgO+BPym4bWi++XlDg5mZsNI0wAREdOBVt8jPR64fJVqZGZmQ0LTJqZWSdqQdKVxfCU5gJskBXBBREzqZ/2JwESArq4uarVau6pmWW9vr/erDSs+ZjurbQECOBT4XUPz0l4RsVDSK4Fpkv6Qr0hWkIPHJIDu7u7o6elpY9UMoFar4f1qw4mP2c5q5yimI2hoXoqIhfnfRcA1wNg2bs/MzAZRWwKEpE2BfYGfV9I2krRJ/TNwIHBvO7ZnZmaDr5VhrpcDPcBISQuA04ARABFxfs72PuCmiHiusmoXcI2k+nYui4gb2ld1MzMbTE0DRESMbyHPFNJw2Grag8CuA62YmZl1lu+kNjOzIgcIMzMrcoAwM7MiBwgzMytygDAzsyIHCDMzK3KAMDOzIgcIMzMrcoAwM7MiBwgzMytygDAzsyIHCDMzK3KAMDOzIgcIMzMrcoAwM7OipgFC0mRJiyQV3wYnqUfS05Jm5+nUyrKDJN0vaZ6kk9tZcTMzG1ytXEFMAQ5qkue3EbFbnk4HkLQO8APgYGBnYLyknVelsmZmtvo0DRARMR1YMoCyxwLzIuLBiHgRuAIYN4ByzMysA9rVB7GnpLskXS9pl5w2CnikkmdBTjMzs2Gg6TupW3AnsENE9Eo6BLgW2AlQIW/0VYikicBEgK6uLmq1WhuqZlW9vb3erzas+JjtrFUOEBHxTOXzdZLOkzSSdMUwupJ1O2BhP+VMAiYBdHd3R09Pz6pWzRrUajW8X2048THbWavcxCRpa0nKn8fmMhcDdwA7SXqVpPWAI4Cpq7o9MzNbPZpeQUi6HOgBRkpaAJwGjACIiPOBw4BPSloKvAAcEREBLJV0PHAjsA4wOSLmDMq3MDOztmsaICJifJPl5wLn9rHsOuC6gVXNzMw6yXdSm5lZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVmRA4SZmRU5QJiZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVmRA4SZmRU1DRCSJktaJOnePpYfJenuPN0qadfKsvmS7pE0W9LMdlbczMwGVytXEFOAg/pZ/hCwb0S8Cfg6MKlh+X4RsVtEdA+simZm1gmtvHJ0uqQx/Sy/tTJ7O7DdqlfLzMw6rWmAWEnHAtdX5gO4SVIAF0RE49XFMpImAhMBurq6qNVqba6a9fb2er/asOJjtrPaFiAk7UcKEHtXkveKiIWSXglMk/SHiJheWj8Hj0kA3d3d0dPT066qWVar1fB+teHEx2xntWUUk6Q3AT8CxkXE4np6RCzM/y4CrgHGtmN7ZmY2+FY5QEjaHrgamBARD1TSN5K0Sf0zcCBQHAllZmZDT9MmJkmXAz3ASEkLgNOAEQARcT5wKrAlcJ4kgKV5xFIXcE1OWxe4LCJuGITvYGZmg6CVUUzjmyz/KPDRQvqDwK4rrmFmZsOB76Q2M7MiBwgzMytygDAzsyIHCDMzK3KAMDOzIgcIMzMrcoAwM7MiBwgzMytygDAzsyIHCDMzK3KAMDOzIgcIMzMrcoAwM7MiBwgzMytygDAzs6KWAoSkyZIWSSq+EU7J9yTNk3S3pDdXlh0t6Y95OrpdFTczs8HV9IVB2RTgXOCSPpYfDOyUpz2AHwJ7SNqC9Aa6biCAWZKmRsSTq1JpWzn5rX7LiYgO1MSsNT5mh4aWriAiYjqwpJ8s44BLIrkd2EzSNsC7gGkRsSQHhWnAQataaWtd6T9af+lmneZjduhoVx/EKOCRyvyCnNZXuq1mEcHNN9/sszAbNnzMdl6rTUzNlEJ79JO+YgHSRGAiQFdXF7VarU1VM4BarUZvb+9y+9X72IYyH7Odp1ajs6QxwC8i4g2FZRcAtYi4PM/fD/TUp4j4eClfX7q7u2PmzJktfwnrW/2yPCKo1Wr09PQsl2Y21PiYHXySZkVEd7N87bqCmAocL+kKUif10xHxqKQbgW9K2jznOxD4Upu2aSvB7bc23PiY7byWAoSky0lXAyMlLSCNTBoBEBHnA9cBhwDzgOeBj+RlSyR9HbgjF3V6RPTX2W1tFhEeEWLDio/ZoaOlABER45ssD+C4PpZNBiavfNWsXer/seqX62ZDnY/ZocF3UpuZWZEDhJmZFTlAmJlZkQOEmZkVOUCYmVlRyzfKrU6SHgce7nQ91kAjgSc6XQmzleBjdnDsEBFbNcs0JAOEDQ5JM1u5e9JsqPAx21luYjIzsyIHCDMzK3KAWLtM6nQFzFaSj9kOch+EmZkV+QrCzMyKHCDWApIOknS/pHmSTu50fcyakTRZ0iJJ93a6LmszB4g1nKR1gB8ABwM7A+Ml7dzZWpk1NQW/v77jHCDWfGOBeRHxYES8CFwBjOtwncz6FRHTAb87psMcINZ8o4BHKvMLcpqZWb8cINZ8pfc2euiamTXlALHmWwCMrsxvByzsUF3MbBhxgFjz3QHsJOlVktYDjgCmdrhOZjYMOECs4SJiKXA8cCMwF7gyIuZ0tlZm/ZN0OXAb8FpJCyQd2+k6rY18J7WZmRX5CsLMzIocIMzMrMgBwszMihwgzMysyAHCzMyKHCBsrSNpS0mz8/SYpL9U5teT9D5JIel1lXXGSHoh57lP0iWSRlSWj5VUk/RHSXdK+qWkN+ZlX23YxmxJh1c+9+an7c6WdEkn9olZiYe52lpN0leB3og4u5J2JbAN8OuI+GpOGwP8IiLekJ+QOw24MCIuldQFzACOjIhbc/69gZERcW1pGw11qAGfj4iZg/IlzQbIVxBmFZI2BvYCjiXddb6CiHgJ+D0vP/TweODienDIeW6JiGsHubpmg8oBwmx57wVuiIgHgCWS3tyYQdIGwB7ADTlpF+DOJuV+ttKkdHNba2w2SBwgzJY3nvTODPK/4yvLXi1JLgUnAAAA50lEQVRpNrAY+HNE3F0qQNIMSXMlfbeS/K8RsVue9huUmpu1mQOEWSZpS+AdwI8kzQdOAg6XVH9k+p8iYjdgR+Ctkv5HTp8DLLvSiIg9gK8Am66uupsNBgcIs5cdBlwSETtExJiIGA08BOxdzRQRjwInA1/KST8AjpH0tkq2DVdHhc0GkwOE2cvGA9c0pF0FHFnIey2woaR9IuIx4HDgTEnzJN1KCjbnVvJX+yBm51FRZkOah7mamVmRryDMzKzIAcLMzIocIMzMrMgBwszMihwgzMysyAHCzMyKHCDMzKzIAcLMzIr+C7odq2cd0+1iAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_data.boxplot(column=['REGION_RATING_CLIENT_W_CITY'],by = 'TARGET')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}