Day_009_HW-outlier.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 檢視與處理 Outliers\n",
    "### 為何會有 outliers, 常見的 outlier 原因\n",
    "* 未知值，隨意填補 (約定俗成的代入)，如年齡常見 0,999\n",
    "* 可能的錯誤紀錄/手誤/系統性錯誤，如某本書在某筆訂單的銷售量 = 1000 本"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [作業目標]\n",
    "- 依照下列提示與引導, 以幾種不同的方式, 檢視可能的離群值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [作業重點]\n",
    "- 從原始資料篩選可能的欄位, 看看那些欄位可能有離群值 (In[3], Out[3])\n",
    "- 繪製目標值累積密度函數(ECDF)的圖形, 和常態分布的累積密度函數對比, 以確認是否有離群值的情形 (In[6], Out[6], In[7], Out[7])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import 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",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "# 設定 data_path\n",
    "# dir_data = \".\\data'\"\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": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>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": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f_app = pd.read_csv(r\"C:\\Users\\user\\100 days\\data\\application_train.csv\")\n",
    "f_app.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 請參考 HomeCredit_columns_description.csv 的欄位說明，觀察並列出三個你覺得可能有 outlier 的欄位並解釋可能的原因"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 307511 entries, 0 to 307510\n",
      "Data columns (total 122 columns):\n",
      "SK_ID_CURR                      int64\n",
      "TARGET                          int64\n",
      "NAME_CONTRACT_TYPE              object\n",
      "CODE_GENDER                     object\n",
      "FLAG_OWN_CAR                    object\n",
      "FLAG_OWN_REALTY                 object\n",
      "CNT_CHILDREN                    int64\n",
      "AMT_INCOME_TOTAL                float64\n",
      "AMT_CREDIT                      float64\n",
      "AMT_ANNUITY                     float64\n",
      "AMT_GOODS_PRICE                 float64\n",
      "NAME_TYPE_SUITE                 object\n",
      "NAME_INCOME_TYPE                object\n",
      "NAME_EDUCATION_TYPE             object\n",
      "NAME_FAMILY_STATUS              object\n",
      "NAME_HOUSING_TYPE               object\n",
      "REGION_POPULATION_RELATIVE      float64\n",
      "DAYS_BIRTH                      int64\n",
      "DAYS_EMPLOYED                   int64\n",
      "DAYS_REGISTRATION               float64\n",
      "DAYS_ID_PUBLISH                 int64\n",
      "OWN_CAR_AGE                     float64\n",
      "FLAG_MOBIL                      int64\n",
      "FLAG_EMP_PHONE                  int64\n",
      "FLAG_WORK_PHONE                 int64\n",
      "FLAG_CONT_MOBILE                int64\n",
      "FLAG_PHONE                      int64\n",
      "FLAG_EMAIL                      int64\n",
      "OCCUPATION_TYPE                 object\n",
      "CNT_FAM_MEMBERS                 float64\n",
      "REGION_RATING_CLIENT            int64\n",
      "REGION_RATING_CLIENT_W_CITY     int64\n",
      "WEEKDAY_APPR_PROCESS_START      object\n",
      "HOUR_APPR_PROCESS_START         int64\n",
      "REG_REGION_NOT_LIVE_REGION      int64\n",
      "REG_REGION_NOT_WORK_REGION      int64\n",
      "LIVE_REGION_NOT_WORK_REGION     int64\n",
      "REG_CITY_NOT_LIVE_CITY          int64\n",
      "REG_CITY_NOT_WORK_CITY          int64\n",
      "LIVE_CITY_NOT_WORK_CITY         int64\n",
      "ORGANIZATION_TYPE               object\n",
      "EXT_SOURCE_1                    float64\n",
      "EXT_SOURCE_2                    float64\n",
      "EXT_SOURCE_3                    float64\n",
      "APARTMENTS_AVG                  float64\n",
      "BASEMENTAREA_AVG                float64\n",
      "YEARS_BEGINEXPLUATATION_AVG     float64\n",
      "YEARS_BUILD_AVG                 float64\n",
      "COMMONAREA_AVG                  float64\n",
      "ELEVATORS_AVG                   float64\n",
      "ENTRANCES_AVG                   float64\n",
      "FLOORSMAX_AVG                   float64\n",
      "FLOORSMIN_AVG                   float64\n",
      "LANDAREA_AVG                    float64\n",
      "LIVINGAPARTMENTS_AVG            float64\n",
      "LIVINGAREA_AVG                  float64\n",
      "NONLIVINGAPARTMENTS_AVG         float64\n",
      "NONLIVINGAREA_AVG               float64\n",
      "APARTMENTS_MODE                 float64\n",
      "BASEMENTAREA_MODE               float64\n",
      "YEARS_BEGINEXPLUATATION_MODE    float64\n",
      "YEARS_BUILD_MODE                float64\n",
      "COMMONAREA_MODE                 float64\n",
      "ELEVATORS_MODE                  float64\n",
      "ENTRANCES_MODE                  float64\n",
      "FLOORSMAX_MODE                  float64\n",
      "FLOORSMIN_MODE                  float64\n",
      "LANDAREA_MODE                   float64\n",
      "LIVINGAPARTMENTS_MODE           float64\n",
      "LIVINGAREA_MODE                 float64\n",
      "NONLIVINGAPARTMENTS_MODE        float64\n",
      "NONLIVINGAREA_MODE              float64\n",
      "APARTMENTS_MEDI                 float64\n",
      "BASEMENTAREA_MEDI               float64\n",
      "YEARS_BEGINEXPLUATATION_MEDI    float64\n",
      "YEARS_BUILD_MEDI                float64\n",
      "COMMONAREA_MEDI                 float64\n",
      "ELEVATORS_MEDI                  float64\n",
      "ENTRANCES_MEDI                  float64\n",
      "FLOORSMAX_MEDI                  float64\n",
      "FLOORSMIN_MEDI                  float64\n",
      "LANDAREA_MEDI                   float64\n",
      "LIVINGAPARTMENTS_MEDI           float64\n",
      "LIVINGAREA_MEDI                 float64\n",
      "NONLIVINGAPARTMENTS_MEDI        float64\n",
      "NONLIVINGAREA_MEDI              float64\n",
      "FONDKAPREMONT_MODE              object\n",
      "HOUSETYPE_MODE                  object\n",
      "TOTALAREA_MODE                  float64\n",
      "WALLSMATERIAL_MODE              object\n",
      "EMERGENCYSTATE_MODE             object\n",
      "OBS_30_CNT_SOCIAL_CIRCLE        float64\n",
      "DEF_30_CNT_SOCIAL_CIRCLE        float64\n",
      "OBS_60_CNT_SOCIAL_CIRCLE        float64\n",
      "DEF_60_CNT_SOCIAL_CIRCLE        float64\n",
      "DAYS_LAST_PHONE_CHANGE          float64\n",
      "FLAG_DOCUMENT_2                 int64\n",
      "FLAG_DOCUMENT_3                 int64\n",
      "FLAG_DOCUMENT_4                 int64\n",
      "FLAG_DOCUMENT_5                 int64\n",
      "FLAG_DOCUMENT_6                 int64\n",
      "FLAG_DOCUMENT_7                 int64\n",
      "FLAG_DOCUMENT_8                 int64\n",
      "FLAG_DOCUMENT_9                 int64\n",
      "FLAG_DOCUMENT_10                int64\n",
      "FLAG_DOCUMENT_11                int64\n",
      "FLAG_DOCUMENT_12                int64\n",
      "FLAG_DOCUMENT_13                int64\n",
      "FLAG_DOCUMENT_14                int64\n",
      "FLAG_DOCUMENT_15                int64\n",
      "FLAG_DOCUMENT_16                int64\n",
      "FLAG_DOCUMENT_17                int64\n",
      "FLAG_DOCUMENT_18                int64\n",
      "FLAG_DOCUMENT_19                int64\n",
      "FLAG_DOCUMENT_20                int64\n",
      "FLAG_DOCUMENT_21                int64\n",
      "AMT_REQ_CREDIT_BUREAU_HOUR      float64\n",
      "AMT_REQ_CREDIT_BUREAU_DAY       float64\n",
      "AMT_REQ_CREDIT_BUREAU_WEEK      float64\n",
      "AMT_REQ_CREDIT_BUREAU_MON       float64\n",
      "AMT_REQ_CREDIT_BUREAU_QRT       float64\n",
      "AMT_REQ_CREDIT_BUREAU_YEAR      float64\n",
      "dtypes: float64(65), int64(41), object(16)\n",
      "memory usage: 286.2+ MB\n"
     ]
    }
   ],
   "source": [
    "f_app.info(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SK_ID_CURR</th>\n",
       "      <th>TARGET</th>\n",
       "      <th>CNT_CHILDREN</th>\n",
       "      <th>AMT_INCOME_TOTAL</th>\n",
       "      <th>AMT_CREDIT</th>\n",
       "      <th>AMT_ANNUITY</th>\n",
       "      <th>AMT_GOODS_PRICE</th>\n",
       "      <th>REGION_POPULATION_RELATIVE</th>\n",
       "      <th>DAYS_BIRTH</th>\n",
       "      <th>DAYS_EMPLOYED</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>count</th>\n",
       "      <td>307511.000000</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>3.075110e+05</td>\n",
       "      <td>3.075110e+05</td>\n",
       "      <td>307499.000000</td>\n",
       "      <td>3.072330e+05</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>307511.000000</td>\n",
       "      <td>265992.000000</td>\n",
       "      <td>265992.000000</td>\n",
       "      <td>265992.000000</td>\n",
       "      <td>265992.000000</td>\n",
       "      <td>265992.000000</td>\n",
       "      <td>265992.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>278180.518577</td>\n",
       "      <td>0.080729</td>\n",
       "      <td>0.417052</td>\n",
       "      <td>1.687979e+05</td>\n",
       "      <td>5.990260e+05</td>\n",
       "      <td>27108.573909</td>\n",
       "      <td>5.383962e+05</td>\n",
       "      <td>0.020868</td>\n",
       "      <td>-16036.995067</td>\n",
       "      <td>63815.045904</td>\n",
       "      <td>...</td>\n",
       "      <td>0.008130</td>\n",
       "      <td>0.000595</td>\n",
       "      <td>0.000507</td>\n",
       "      <td>0.000335</td>\n",
       "      <td>0.006402</td>\n",
       "      <td>0.007000</td>\n",
       "      <td>0.034362</td>\n",
       "      <td>0.267395</td>\n",
       "      <td>0.265474</td>\n",
       "      <td>1.899974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>102790.175348</td>\n",
       "      <td>0.272419</td>\n",
       "      <td>0.722121</td>\n",
       "      <td>2.371231e+05</td>\n",
       "      <td>4.024908e+05</td>\n",
       "      <td>14493.737315</td>\n",
       "      <td>3.694465e+05</td>\n",
       "      <td>0.013831</td>\n",
       "      <td>4363.988632</td>\n",
       "      <td>141275.766519</td>\n",
       "      <td>...</td>\n",
       "      <td>0.089798</td>\n",
       "      <td>0.024387</td>\n",
       "      <td>0.022518</td>\n",
       "      <td>0.018299</td>\n",
       "      <td>0.083849</td>\n",
       "      <td>0.110757</td>\n",
       "      <td>0.204685</td>\n",
       "      <td>0.916002</td>\n",
       "      <td>0.794056</td>\n",
       "      <td>1.869295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>100002.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.565000e+04</td>\n",
       "      <td>4.500000e+04</td>\n",
       "      <td>1615.500000</td>\n",
       "      <td>4.050000e+04</td>\n",
       "      <td>0.000290</td>\n",
       "      <td>-25229.000000</td>\n",
       "      <td>-17912.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>189145.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.125000e+05</td>\n",
       "      <td>2.700000e+05</td>\n",
       "      <td>16524.000000</td>\n",
       "      <td>2.385000e+05</td>\n",
       "      <td>0.010006</td>\n",
       "      <td>-19682.000000</td>\n",
       "      <td>-2760.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>278202.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.471500e+05</td>\n",
       "      <td>5.135310e+05</td>\n",
       "      <td>24903.000000</td>\n",
       "      <td>4.500000e+05</td>\n",
       "      <td>0.018850</td>\n",
       "      <td>-15750.000000</td>\n",
       "      <td>-1213.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>367142.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.025000e+05</td>\n",
       "      <td>8.086500e+05</td>\n",
       "      <td>34596.000000</td>\n",
       "      <td>6.795000e+05</td>\n",
       "      <td>0.028663</td>\n",
       "      <td>-12413.000000</td>\n",
       "      <td>-289.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>456255.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>1.170000e+08</td>\n",
       "      <td>4.050000e+06</td>\n",
       "      <td>258025.500000</td>\n",
       "      <td>4.050000e+06</td>\n",
       "      <td>0.072508</td>\n",
       "      <td>-7489.000000</td>\n",
       "      <td>365243.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>261.000000</td>\n",
       "      <td>25.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 106 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          SK_ID_CURR         TARGET   CNT_CHILDREN  AMT_INCOME_TOTAL  \\\n",
       "count  307511.000000  307511.000000  307511.000000      3.075110e+05   \n",
       "mean   278180.518577       0.080729       0.417052      1.687979e+05   \n",
       "std    102790.175348       0.272419       0.722121      2.371231e+05   \n",
       "min    100002.000000       0.000000       0.000000      2.565000e+04   \n",
       "25%    189145.500000       0.000000       0.000000      1.125000e+05   \n",
       "50%    278202.000000       0.000000       0.000000      1.471500e+05   \n",
       "75%    367142.500000       0.000000       1.000000      2.025000e+05   \n",
       "max    456255.000000       1.000000      19.000000      1.170000e+08   \n",
       "\n",
       "         AMT_CREDIT    AMT_ANNUITY  AMT_GOODS_PRICE  \\\n",
       "count  3.075110e+05  307499.000000     3.072330e+05   \n",
       "mean   5.990260e+05   27108.573909     5.383962e+05   \n",
       "std    4.024908e+05   14493.737315     3.694465e+05   \n",
       "min    4.500000e+04    1615.500000     4.050000e+04   \n",
       "25%    2.700000e+05   16524.000000     2.385000e+05   \n",
       "50%    5.135310e+05   24903.000000     4.500000e+05   \n",
       "75%    8.086500e+05   34596.000000     6.795000e+05   \n",
       "max    4.050000e+06  258025.500000     4.050000e+06   \n",
       "\n",
       "       REGION_POPULATION_RELATIVE     DAYS_BIRTH  DAYS_EMPLOYED  ...  \\\n",
       "count               307511.000000  307511.000000  307511.000000  ...   \n",
       "mean                     0.020868  -16036.995067   63815.045904  ...   \n",
       "std                      0.013831    4363.988632  141275.766519  ...   \n",
       "min                      0.000290  -25229.000000  -17912.000000  ...   \n",
       "25%                      0.010006  -19682.000000   -2760.000000  ...   \n",
       "50%                      0.018850  -15750.000000   -1213.000000  ...   \n",
       "75%                      0.028663  -12413.000000    -289.000000  ...   \n",
       "max                      0.072508   -7489.000000  365243.000000  ...   \n",
       "\n",
       "       FLAG_DOCUMENT_18  FLAG_DOCUMENT_19  FLAG_DOCUMENT_20  FLAG_DOCUMENT_21  \\\n",
       "count     307511.000000     307511.000000     307511.000000     307511.000000   \n",
       "mean           0.008130          0.000595          0.000507          0.000335   \n",
       "std            0.089798          0.024387          0.022518          0.018299   \n",
       "min            0.000000          0.000000          0.000000          0.000000   \n",
       "25%            0.000000          0.000000          0.000000          0.000000   \n",
       "50%            0.000000          0.000000          0.000000          0.000000   \n",
       "75%            0.000000          0.000000          0.000000          0.000000   \n",
       "max            1.000000          1.000000          1.000000          1.000000   \n",
       "\n",
       "       AMT_REQ_CREDIT_BUREAU_HOUR  AMT_REQ_CREDIT_BUREAU_DAY  \\\n",
       "count               265992.000000              265992.000000   \n",
       "mean                     0.006402                   0.007000   \n",
       "std                      0.083849                   0.110757   \n",
       "min                      0.000000                   0.000000   \n",
       "25%                      0.000000                   0.000000   \n",
       "50%                      0.000000                   0.000000   \n",
       "75%                      0.000000                   0.000000   \n",
       "max                      4.000000                   9.000000   \n",
       "\n",
       "       AMT_REQ_CREDIT_BUREAU_WEEK  AMT_REQ_CREDIT_BUREAU_MON  \\\n",
       "count               265992.000000              265992.000000   \n",
       "mean                     0.034362                   0.267395   \n",
       "std                      0.204685                   0.916002   \n",
       "min                      0.000000                   0.000000   \n",
       "25%                      0.000000                   0.000000   \n",
       "50%                      0.000000                   0.000000   \n",
       "75%                      0.000000                   0.000000   \n",
       "max                      8.000000                  27.000000   \n",
       "\n",
       "       AMT_REQ_CREDIT_BUREAU_QRT  AMT_REQ_CREDIT_BUREAU_YEAR  \n",
       "count              265992.000000               265992.000000  \n",
       "mean                    0.265474                    1.899974  \n",
       "std                     0.794056                    1.869295  \n",
       "min                     0.000000                    0.000000  \n",
       "25%                     0.000000                    0.000000  \n",
       "50%                     0.000000                    1.000000  \n",
       "75%                     0.000000                    3.000000  \n",
       "max                   261.000000                   25.000000  \n",
       "\n",
       "[8 rows x 106 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f_app.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "float64    65\n",
       "int64      41\n",
       "object     16\n",
       "dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 資料中有float64/int64/object等3種types\n",
    "f_app.dtypes.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 先篩選數值型的欄位\n",
    "\n",
    "f_app.dtype_select = ['float64','int64']\n",
    "\n",
    "numeric_columns = list(f_app.columns[list(f_app.dtypes.isin(dtype_select))])\n",
    "\n",
    "# https://www.cupoy.com/qa/kwassist/ai_tw/0000016A3B608C23000003B66375706F795F72656C656173655155455354"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Numbers of remain columns 0\n"
     ]
    }
   ],
   "source": [
    "# 再把只有 2 值 (通常是 0,1) 的欄位去掉\n",
    "numeric_columns = list(f_app[numeric_columns].columns[list(f_app[numeric_columns].apply(lambda x:len(x.unique())!=2 ))])\n",
    "print(\"Numbers of remain columns %s\" % len(numeric_columns))\n",
    "\n",
    "### 補充一下這種用法叫做格式化字串（formatting string），其用法是「在字串中篩入變數」，其用法如下：\n",
    "# w = 'world'\n",
    "# p = 'python'\n",
    "# s = 'hello %s, I love %s' % (w, p)\n",
    "# 執行 s = .. 這行的時候，會將 % 後面的變數，依序取代掉左邊字串中的 %s，這就是格式化字串。\n",
    "# print(\"Numbers of remain columns %s\" % len(nume... 即可."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 檢視這些欄位的數值範圍\n",
    "for col in numeric_columns:\n",
    "    f_app.boxplot(col)    \n",
    "    f_app.plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([5.3501e+04, 3.4636e+04, 1.0248e+04, 2.3360e+03, 4.8600e+02,\n",
       "        3.0000e+01, 8.0000e+00, 3.3350e+03, 0.0000e+00, 2.0000e+00]),\n",
       " array([ 0. ,  9.1, 18.2, 27.3, 36.4, 45.5, 54.6, 63.7, 72.8, 81.9, 91. ]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 36,
     "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": [
    "a = f_app['OWN_CAR_AGE'].dropna()\n",
    "plt.hist(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 從上面的圖檢查的結果，至少這三個欄位好像有點可疑\n",
    "\n",
    "# AMT_INCOME_TOTAL\n",
    "# REGION_POPULATION_RELATIVE\n",
    "# OBS_60_CNT_SOCIAL_CIRCLE\n",
    "# ?? 看不懂"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hints: Emprical Cumulative Density Plot, [ECDF](https://zh.wikipedia.org/wiki/%E7%BB%8F%E9%AA%8C%E5%88%86%E5%B8%83%E5%87%BD%E6%95%B0), [ECDF with Python](https://stackoverflow.com/questions/14006520/ecdf-in-python-without-step-function)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    3.075110e+05\n",
      "mean     1.687979e+05\n",
      "std      2.371231e+05\n",
      "min      2.565000e+04\n",
      "25%      1.125000e+05\n",
      "50%      1.471500e+05\n",
      "75%      2.025000e+05\n",
      "max      1.170000e+08\n",
      "Name: AMT_INCOME_TOTAL, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# 最大值離平均與中位數很遠\n",
    "print(f_app['AMT_INCOME_TOTAL'].describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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": [
    "# 繪製 Empirical Cumulative Density Plot (ECDF)\n",
    "# 經驗分布函數\n",
    "\n",
    "\"\"\"\n",
    "YOUR CODE HERE\n",
    "\"\"\"\n",
    "cdf = f_app.AMT_INCOME_TOTAL.value_counts().sort_index().cumsum()\n",
    "\n",
    "\n",
    "plt.plot(list(cdf.index), cdf/cdf.max())\n",
    "plt.xlabel('Value')\n",
    "plt.ylabel('ECDF')\n",
    "plt.xlim([cdf.index.min(), cdf.index.max() * 1.05]) # 限制顯示圖片的範圍\n",
    "plt.ylim([-0.05,1.05]) # 限制顯示圖片的範圍\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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": [
    "# 改變 y 軸的 Scale, 讓我們可以正常檢視 ECDF\n",
    "plt.plot(np.log(list(cdf.index)), cdf/cdf.max())\n",
    "plt.xlabel('Value (log-scale)')\n",
    "plt.ylabel('ECDF')\n",
    "\n",
    "plt.ylim([-0.05,1.05]) # 限制顯示圖片的範圍\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 補充：Normal dist 的 ECDF\n",
    "![ecdf_normal](https://au.mathworks.com/help/examples/stats/win64/PlotEmpiricalCdfAndCompareWithSamplingDistributionExample_01.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    307511.000000\n",
      "mean          0.020868\n",
      "std           0.013831\n",
      "min           0.000290\n",
      "25%           0.010006\n",
      "50%           0.018850\n",
      "75%           0.028663\n",
      "max           0.072508\n",
      "Name: REGION_POPULATION_RELATIVE, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# 最大值落在分布之外\n",
    "print(f_app['REGION_POPULATION_RELATIVE'].describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.035792    16408\n",
       "0.046220    13442\n",
       "0.030755    12163\n",
       "0.025164    11950\n",
       "0.026392    11601\n",
       "0.031329    11321\n",
       "0.028663    11157\n",
       "0.019101     8694\n",
       "0.072508     8412\n",
       "0.020713     8066\n",
       "0.018850     7668\n",
       "0.020246     7178\n",
       "0.018634     7038\n",
       "0.022625     6943\n",
       "0.015221     6824\n",
       "0.032561     6636\n",
       "0.024610     6404\n",
       "0.019689     6172\n",
       "0.018029     6167\n",
       "0.018801     6108\n",
       "0.018209     6052\n",
       "0.014520     4785\n",
       "0.016612     4408\n",
       "0.007020     4105\n",
       "0.022800     3807\n",
       "0.010032     3570\n",
       "0.010006     3563\n",
       "0.014464     3422\n",
       "0.004960     3300\n",
       "0.011703     3252\n",
       "            ...  \n",
       "0.006008     1796\n",
       "0.003069     1783\n",
       "0.008575     1763\n",
       "0.008474     1740\n",
       "0.007274     1724\n",
       "0.007114     1684\n",
       "0.006305     1672\n",
       "0.008068     1668\n",
       "0.008866     1654\n",
       "0.006671     1653\n",
       "0.002042     1642\n",
       "0.003813     1616\n",
       "0.005084     1482\n",
       "0.006296     1225\n",
       "0.005313     1218\n",
       "0.003818     1212\n",
       "0.003122     1147\n",
       "0.006233     1087\n",
       "0.005002     1068\n",
       "0.002134     1051\n",
       "0.004849     1038\n",
       "0.005144      967\n",
       "0.002506      944\n",
       "0.003541      627\n",
       "0.001276      558\n",
       "0.001417      467\n",
       "0.001333      235\n",
       "0.000533       39\n",
       "0.000938       28\n",
       "0.000290        2\n",
       "Name: REGION_POPULATION_RELATIVE, Length: 81, dtype: int64"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 繪製 Empirical Cumulative Density Plot (ECDF)\n",
    "\"\"\"\n",
    "Your Code Here\n",
    "\"\"\"\n",
    "cdf = f_app.REGION_POPULATION_RELATIVE.value_counts().sort_index().cumsum()\n",
    "\n",
    "\n",
    "plt.plot(list(cdf.index), cdf/cdf.max())\n",
    "plt.xlabel('Value')\n",
    "plt.ylabel('ECDF')\n",
    "plt.ylim([-0.05,1.05]) # 限制顯示圖片的範圍\n",
    "plt.show()\n",
    "\n",
    "f_app['REGION_POPULATION_RELATIVE'].hist()\n",
    "plt.show()\n",
    "\n",
    "f_app['REGION_POPULATION_RELATIVE'].value_counts()\n",
    "\n",
    "# 就以這個欄位來說，雖然有資料掉在分布以外，也不算異常，僅代表這間公司在稍微熱鬧的地區有的據點較少，\n",
    "# 導致 region population relative 在少的部分較為密集，但在大的部分較為疏漏"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    306490.000000\n",
      "mean          1.405292\n",
      "std           2.379803\n",
      "min           0.000000\n",
      "25%           0.000000\n",
      "50%           0.000000\n",
      "75%           2.000000\n",
      "max         344.000000\n",
      "Name: OBS_60_CNT_SOCIAL_CIRCLE, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# 最大值落在分布之外\n",
    "print(f_app['OBS_60_CNT_SOCIAL_CIRCLE'].describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAD8CAYAAACLrvgBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAFw1JREFUeJzt3X+MXfWZ3/H3szYQi/wwP8LIsq3iNJYaJ3QdmCWWUq2msAJD/zCRQDVCwc1a8ioL2kRipcCuVNIQpFCJoIIIK6e4mKwbh5JEtrZOvRb4Klop/EwcbMfLehasMLGFRWwIkyikZp/+cb+T3Ax3Zr6+M+Nzb3m/pKt77nO+55znHI35zPkxl8hMJEmq8QdNNyBJGhyGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmqZmhIkqoZGpKkagubbmCuXXjhhXnxxRf3tOwvf/lLzj333LltaB4NWr8weD3b7/watH5h8Hqu7ff5559/LTM/OOPAzPz/6nXZZZdlr/bu3dvzsk0YtH4zB69n+51fg9Zv5uD1XNsv8FxW/DfWy1OSpGqGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmqZmhIkqoZGpKkajN+jUhEvAf4PnBOGf94Zt4ZESuA7cD5wA+BT2fmbyLiHOBR4DLg58B/zMwjZV13ABuBt4G/yMzdpb4W+G/AAuC/Z+ZXSr3rNuZo399h/8/e4D/d/r/na/XTOvKV/9DIdiXpdNScabwFXJGZfwisBtZGxBrgHuC+zFwJnKQdBpT3k5n5YeC+Mo6IWAWsBz4KrAW+FhELImIB8CBwDbAKuLGMZZptSJIaMGNolK8lGS8fzyqvBK4AHi/1rcB1ZXpd+UyZf2VERKlvz8y3MvNlYBS4vLxGM/OlchaxHVhXlplqG5KkBlTd0yhnBPuA48Ae4J+B1zPzVBkyBiwt00uBVwDK/DeACzrrk5aZqn7BNNuQJDWg6qvRM/NtYHVELAa+C3yk27DyHlPMm6reLbimG/8OEbEJ2AQwNDREq9XqNmxGQ4vgtktOzTxwHvTS8/j4eM/72pRB69l+59eg9QuD1/Nc93ta/z+NzHw9IlrAGmBxRCwsZwLLgKNl2BiwHBiLiIXAB4ATHfUJnct0q782zTYm97UZ2AwwPDycIyMjp7Nbv/XAth3cu7+Z/8XIkZtGTnuZVqtFr/valEHr2X7n16D1C4PX81z3O+PlqYj4YDnDICIWAX8CHAL2AteXYRuAHWV6Z/lMmf9k+a72ncD6iDinPBW1EngGeBZYGRErIuJs2jfLd5ZlptqGJKkBNb9WLwG2lqec/gB4LDP/LiJ+AmyPiC8DPwIeLuMfBr4REaO0zzDWA2TmwYh4DPgJcAq4pVz2IiJuBXbTfuR2S2YeLOv6whTbkCQ1YMbQyMwXgI93qb9E+8mnyfVfAzdMsa67gbu71HcBu2q3IUlqhn8RLkmqZmhIkqoZGpKkaoaGJKmaoSFJqmZoSJKqGRqSpGqGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmqZmhIkqoZGpKkaoaGJKmaoSFJqmZoSJKqGRqSpGqGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmqNmNoRMTyiNgbEYci4mBEfK7UvxgRP4uIfeV1bccyd0TEaES8GBFXd9TXltpoRNzeUV8REU9HxOGI+FZEnF3q55TPo2X+xXO585Kk01NzpnEKuC0zPwKsAW6JiFVl3n2Zubq8dgGUeeuBjwJrga9FxIKIWAA8CFwDrAJu7FjPPWVdK4GTwMZS3wiczMwPA/eVcZKkhswYGpl5LDN/WKbfBA4BS6dZZB2wPTPfysyXgVHg8vIazcyXMvM3wHZgXUQEcAXweFl+K3Bdx7q2lunHgSvLeElSA07rnka5PPRx4OlSujUiXoiILRFxXqktBV7pWGys1KaqXwC8npmnJtV/b11l/htlvCSpAQtrB0bEe4FvA5/PzF9ExEPAXUCW93uBPwW6nQkk3QMqpxnPDPM6e9sEbAIYGhqi1WpNuy9TGVoEt11yauaB86CXnsfHx3ve16YMWs/2O78GrV8YvJ7nut+q0IiIs2gHxrbM/A5AZr7aMf/rwN+Vj2PA8o7FlwFHy3S3+mvA4ohYWM4mOsdPrGssIhYCHwBOTO4vMzcDmwGGh4dzZGSkZrfe4YFtO7h3f3WOzqkjN42c9jKtVote97Upg9az/c6vQesXBq/nue635umpAB4GDmXmVzvqSzqGfQo4UKZ3AuvLk08rgJXAM8CzwMrypNTZtG+W78zMBPYC15flNwA7Ota1oUxfDzxZxkuSGlDza/UngU8D+yNiX6n9Fe2nn1bTvlx0BPgzgMw8GBGPAT+h/eTVLZn5NkBE3ArsBhYAWzLzYFnfF4DtEfFl4Ee0Q4ry/o2IGKV9hrF+FvsqSZqlGUMjM/+B7vcWdk2zzN3A3V3qu7otl5kv0X66anL918ANM/UoSToz/ItwSVI1Q0OSVM3QkCRVMzQkSdUMDUlSNUNDklTN0JAkVTM0JEnVDA1JUjVDQ5JUzdCQJFUzNCRJ1QwNSVI1Q0OSVM3QkCRVMzQkSdUMDUlSNUNDklTN0JAkVTM0JEnVDA1JUjVDQ5JUzdCQJFUzNCRJ1QwNSVK1GUMjIpZHxN6IOBQRByPic6V+fkTsiYjD5f28Uo+IuD8iRiPihYi4tGNdG8r4wxGxoaN+WUTsL8vcHxEx3TYkSc2oOdM4BdyWmR8B1gC3RMQq4HbgicxcCTxRPgNcA6wsr03AQ9AOAOBO4BPA5cCdHSHwUBk7sdzaUp9qG5KkBswYGpl5LDN/WKbfBA4BS4F1wNYybCtwXZleBzyabU8BiyNiCXA1sCczT2TmSWAPsLbMe39m/iAzE3h00rq6bUOS1IDTuqcRERcDHweeBoYy8xi0gwW4qAxbCrzSsdhYqU1XH+tSZ5ptSJIasLB2YES8F/g28PnM/EW57dB1aJda9lCvFhGbaF/eYmhoiFardTqL/9bQIrjtklM9LTtbvfQ8Pj7e8742ZdB6tt/5NWj9wuD1PNf9VoVGRJxFOzC2ZeZ3SvnViFiSmcfKJabjpT4GLO9YfBlwtNRHJtVbpb6sy/jptvF7MnMzsBlgeHg4R0ZGug2b0QPbdnDv/uocnVNHbho57WVarRa97mtTBq1n+51fg9YvDF7Pc91vzdNTATwMHMrMr3bM2glMPAG1AdjRUb+5PEW1BnijXFraDVwVEeeVG+BXAbvLvDcjYk3Z1s2T1tVtG5KkBtT8Wv1J4NPA/ojYV2p/BXwFeCwiNgI/BW4o83YB1wKjwK+AzwBk5omIuAt4toz7UmaeKNOfBR4BFgHfKy+m2YYkqQEzhkZm/gPd7zsAXNllfAK3TLGuLcCWLvXngI91qf+82zYkSc3wL8IlSdUMDUlSNUNDklTN0JAkVTM0JEnVDA1JUjVDQ5JUzdCQJFUzNCRJ1QwNSVI1Q0OSVM3QkCRVMzQkSdUMDUlSNUNDklTN0JAkVTM0JEnVDA1JUjVDQ5JUzdCQJFUzNCRJ1QwNSVI1Q0OSVM3QkCRVMzQkSdVmDI2I2BIRxyPiQEftixHxs4jYV17Xdsy7IyJGI+LFiLi6o7621EYj4vaO+oqIeDoiDkfEtyLi7FI/p3weLfMvnqudliT1puZM4xFgbZf6fZm5urx2AUTEKmA98NGyzNciYkFELAAeBK4BVgE3lrEA95R1rQROAhtLfSNwMjM/DNxXxkmSGjRjaGTm94ETletbB2zPzLcy82VgFLi8vEYz86XM/A2wHVgXEQFcATxelt8KXNexrq1l+nHgyjJektSQhbNY9taIuBl4DrgtM08CS4GnOsaMlRrAK5PqnwAuAF7PzFNdxi+dWCYzT0XEG2X8a5MbiYhNwCaAoaEhWq1WTzs0tAhuu+TUzAPnQS89j4+P97yvTRm0nu13fg1avzB4Pc91v72GxkPAXUCW93uBPwW6nQkk3c9ocprxzDDv94uZm4HNAMPDwzkyMjJN61N7YNsO7t0/mxzt3ZGbRk57mVarRa/72pRB69l+59eg9QuD1/Nc99vT01OZ+Wpmvp2Z/wJ8nfblJ2ifKSzvGLoMODpN/TVgcUQsnFT/vXWV+R+g/jKZJGke9BQaEbGk4+OngIknq3YC68uTTyuAlcAzwLPAyvKk1Nm0b5bvzMwE9gLXl+U3ADs61rWhTF8PPFnGS5IaMuO1mIj4JjACXBgRY8CdwEhErKZ9uegI8GcAmXkwIh4DfgKcAm7JzLfLem4FdgMLgC2ZebBs4gvA9oj4MvAj4OFSfxj4RkSM0j7DWD/rvZUkzcqMoZGZN3YpP9ylNjH+buDuLvVdwK4u9Zf43eWtzvqvgRtm6k+SdOb4F+GSpGqGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmqZmhIkqoZGpKkaoaGJKmaoSFJqmZoSJKqGRqSpGqGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmqZmhIkqoZGpKkaoaGJKmaoSFJqmZoSJKqGRqSpGozhkZEbImI4xFxoKN2fkTsiYjD5f28Uo+IuD8iRiPihYi4tGOZDWX84YjY0FG/LCL2l2Xuj4iYbhuSpObUnGk8AqydVLsdeCIzVwJPlM8A1wAry2sT8BC0AwC4E/gEcDlwZ0cIPFTGTiy3doZtSJIaMmNoZOb3gROTyuuArWV6K3BdR/3RbHsKWBwRS4CrgT2ZeSIzTwJ7gLVl3vsz8weZmcCjk9bVbRuSpIYs7HG5ocw8BpCZxyLiolJfCrzSMW6s1Karj3WpT7eNd4iITbTPVhgaGqLVavW2U4vgtktO9bTsbPXS8/j4eM/72pRB69l+59eg9QuD1/Nc99traEwlutSyh/ppyczNwGaA4eHhHBkZOd1VAPDAth3cu3+uD0mdIzeNnPYyrVaLXve1KYPWs/3Or0HrFwav57nut9enp14tl5Yo78dLfQxY3jFuGXB0hvqyLvXptiFJakivobETmHgCagOwo6N+c3mKag3wRrnEtBu4KiLOKzfArwJ2l3lvRsSa8tTUzZPW1W0bkqSGzHgtJiK+CYwAF0bEGO2noL4CPBYRG4GfAjeU4buAa4FR4FfAZwAy80RE3AU8W8Z9KTMnbq5/lvYTWouA75UX02xDktSQGUMjM2+cYtaVXcYmcMsU69kCbOlSfw74WJf6z7ttQ5LUHP8iXJJUzdCQJFUzNCRJ1QwNSVI1Q0OSVM3QkCRVMzQkSdUMDUlSNUNDklTN0JAkVTM0JEnVDA1JUjVDQ5JUzdCQJFUzNCRJ1QwNSVI1Q0OSVM3QkCRVMzQkSdUMDUlSNUNDklTN0JAkVTM0JEnVDA1JUrVZhUZEHImI/RGxLyKeK7XzI2JPRBwu7+eVekTE/RExGhEvRMSlHevZUMYfjogNHfXLyvpHy7Ixm34lSbMzF2ca/z4zV2fmcPl8O/BEZq4EniifAa4BVpbXJuAhaIcMcCfwCeBy4M6JoCljNnUst3YO+pUk9Wg+Lk+tA7aW6a3AdR31R7PtKWBxRCwBrgb2ZOaJzDwJ7AHWlnnvz8wfZGYCj3asS5LUgNmGRgJ/HxHPR8SmUhvKzGMA5f2iUl8KvNKx7FipTVcf61KXJDVk4SyX/2RmHo2Ii4A9EfGP04ztdj8ie6i/c8XtwNoEMDQ0RKvVmrbpqQwtgtsuOdXTsrPVS8/j4+M972tTBq1n+51fg9YvDF7Pc93vrEIjM4+W9+MR8V3a9yRejYglmXmsXGI6XoaPAcs7Fl8GHC31kUn1Vqkv6zK+Wx+bgc0Aw8PDOTIy0m3YjB7YtoN79882R3tz5KaR016m1WrR6742ZdB6tt/5NWj9wuD1PNf99nx5KiLOjYj3TUwDVwEHgJ3AxBNQG4AdZXoncHN5imoN8Ea5fLUbuCoizis3wK8Cdpd5b0bEmvLU1M0d65IkNWA2v1YPAd8tT8EuBP5nZv6fiHgWeCwiNgI/BW4o43cB1wKjwK+AzwBk5omIuAt4toz7UmaeKNOfBR4BFgHfKy9JUkN6Do3MfAn4wy71nwNXdqkncMsU69oCbOlSfw74WK89SpLmln8RLkmqZmhIkqoZGpKkaoaGJKmaoSFJqmZoSJKqGRqSpGqGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmqZmhIkqoZGpKkaoaGJKmaoSFJqmZoSJKqGRqSpGqGhiSpmqEhSapmaEiSqhkakqRqhoYkqZqhIUmq1vehERFrI+LFiBiNiNub7keS3s36OjQiYgHwIHANsAq4MSJWNduVJL179XVoAJcDo5n5Umb+BtgOrGu4J0l61+r30FgKvNLxeazUJEkNWNh0AzOILrV8x6CITcCm8nE8Il7scXsXAq/1uOysxD09LdZYv7MwaD3b7/watH5h8Hqu7fdf1ays30NjDFje8XkZcHTyoMzcDGye7cYi4rnMHJ7tes6UQesXBq9n+51fg9YvDF7Pc91vv1+eehZYGRErIuJsYD2ws+GeJOldq6/PNDLzVETcCuwGFgBbMvNgw21J0rtWX4cGQGbuAnadoc3N+hLXGTZo/cLg9Wy/82vQ+oXB63lO+43Md9xXliSpq36/pyFJ6iOGRjEIX1cSEUciYn9E7IuI50rt/IjYExGHy/t5Dfa3JSKOR8SBjlrX/qLt/nK8X4iIS/uo5y9GxM/Kcd4XEdd2zLuj9PxiRFx9hntdHhF7I+JQRByMiM+Vet8e42l67tdj/J6IeCYiflz6/S+lviIini7H+FvlwRwi4pzyebTMv7hP+n0kIl7uOL6rS332PxOZ+a5/0b7J/s/Ah4CzgR8Dq5ruq0ufR4ALJ9X+K3B7mb4duKfB/v4YuBQ4MFN/wLXA92j/Lc4a4Ok+6vmLwF92Gbuq/GycA6woPzMLzmCvS4BLy/T7gH8qPfXtMZ6m5349xgG8t0yfBTxdjt1jwPpS/xvgs2X6z4G/KdPrgW+d4eM7Vb+PANd3GT/rnwnPNNoG+etK1gFby/RW4LqmGsnM7wMnJpWn6m8d8Gi2PQUsjoglZ6bT35mi56msA7Zn5luZ+TIwSvtn54zIzGOZ+cMy/SZwiPY3JPTtMZ6m56k0fYwzM8fLx7PKK4ErgMdLffIxnjj2jwNXRkS3P0qeF9P0O5VZ/0wYGm2D8nUlCfx9RDxf/goeYCgzj0H7HyhwUWPddTdVf/1+zG8tp+9bOi759U3P5TLIx2n/ZjkQx3hSz9CnxzgiFkTEPuA4sIf22c7rmXmqS0+/7bfMfwO4oMl+M3Pi+N5dju99EXHO5H6L0z6+hkZb1deV9IFPZualtL/195aI+OOmG5qFfj7mDwH/GlgNHAPuLfW+6Dki3gt8G/h8Zv5iuqFdao0c4y499+0xzsy3M3M17W+guBz4yDQ99V2/EfEx4A7g3wB/BJwPfKEMn3W/hkZb1deVNC0zj5b348B3af9AvzpxelnejzfXYVdT9de3xzwzXy3/EP8F+Dq/uzzSeM8RcRbt//huy8zvlHJfH+NuPffzMZ6Qma8DLdrX/hdHxMTftXX29Nt+y/wPUH+5c0519Lu2XBbMzHwL+B/M4fE1NNr6/utKIuLciHjfxDRwFXCAdp8byrANwI5mOpzSVP3tBG4uT3OsAd6YuMTStEnXeD9F+zhDu+f15YmZFcBK4Jkz2FcADwOHMvOrHbP69hhP1XMfH+MPRsTiMr0I+BPa92H2AteXYZOP8cSxvx54Mssd5wb7/ceOXyKC9v2XzuM7u5+JM3mnv59ftJ8q+Cfa1y//uul+uvT3IdpPlfwYODjRI+3rp08Ah8v7+Q32+E3alxr+L+3faDZO1R/t0+QHy/HeDwz3Uc/fKD29UP6RLekY/9el5xeBa85wr/+O9qWEF4B95XVtPx/jaXru12P8b4Eflb4OAP+51D9EO7xGgf8FnFPq7ymfR8v8D/VJv0+W43sA+Ft+94TVrH8m/ItwSVI1L09JkqoZGpKkaoaGJKmaoSFJqmZoSJKqGRqSpGqGhiSpmqEhSar2/wAHVdTHORx/MgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "344.0         1\n",
      "47.0          1\n",
      "30.0          1\n",
      "29.0          2\n",
      "28.0          1\n",
      "27.0          5\n",
      "26.0          1\n",
      "25.0         11\n",
      "24.0         12\n",
      "23.0         13\n",
      "22.0         20\n",
      "21.0         28\n",
      "20.0         30\n",
      "19.0         44\n",
      "18.0         49\n",
      "17.0         73\n",
      "16.0        137\n",
      "15.0        166\n",
      "14.0        238\n",
      "13.0        401\n",
      "12.0        620\n",
      "11.0        837\n",
      "10.0       1335\n",
      "9.0        1959\n",
      "8.0        2886\n",
      "7.0        4344\n",
      "6.0        6349\n",
      "5.0        9463\n",
      "4.0       13946\n",
      "3.0       20215\n",
      "2.0       29766\n",
      "1.0       48870\n",
      "0.0      164666\n",
      "Name: OBS_60_CNT_SOCIAL_CIRCLE, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# 繪製 Empirical Cumulative Density Plot (ECDF)\n",
    "\"\"\"\n",
    "Your Code Here\n",
    "\n",
    "cdf = f_app.OBS_60_CNT_SOCIAL_CIRCLE.value_counts().sort_index().cumsum()\n",
    "\n",
    "\n",
    "plt.plot(list(cdf.index), cdf/cdf.max())\n",
    "plt.xlabel('Value')\n",
    "plt.ylabel('ECDF')\n",
    "plt.xlim([cdf.index.min() * 0.95, cdf.index.max() * 1.05])\n",
    "plt.ylim([-0.05,1.05]) # 限制顯示圖片的範圍\n",
    "plt.show()\n",
    "\"\"\"\n",
    "f_app['OBS_60_CNT_SOCIAL_CIRCLE'].hist()\n",
    "plt.show()\n",
    "print(f_app['OBS_60_CNT_SOCIAL_CIRCLE'].value_counts().sort_index(ascending = False))\n",
    "\n",
    "# "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 注意：當 histogram 畫出上面這種圖 (只出現一條，但是 x 軸延伸很長導致右邊有一大片空白時，代表右邊有值但是數量稀少。這時可以考慮用 value_counts 去找到這些數值"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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": [
    "# 把一些極端值暫時去掉，在繪製一次 Histogram\n",
    "# 選擇 OBS_60_CNT_SOCIAL_CIRCLE 小於 20 的資料點繪製\n",
    "\"\"\"\n",
    "Your Code Here\n",
    "\"\"\"\n",
    "loc_a = f_app['OBS_60_CNT_SOCIAL_CIRCLE'] < 20\n",
    "loc_b = 'OBS_60_CNT_SOCIAL_CIRCLE'\n",
    "\n",
    "f_app.loc[loc_a, loc_b].hist()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 是選擇 OBS_60_CNT_SOCIAL_CIRCLE 小於 20 的資料點繪製，不是value_counts()小於 20。\n",
    "# 所以這題的 loc_a = app_train['OBS_60_CNT_SOCIAL_CIRCLE'] < 20 就好！\n",
    "# 2019/09/19"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}