income_model_final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "import warnings\n",
    "import random\n",
    "warnings.filterwarnings(action='ignore')\n",
    "from scipy.stats import randint\n",
    "from scipy.stats import uniform\n",
    "from tqdm import tqdm\n",
    "\n",
    "from pycaret.classification import *\n",
    "\n",
    "# 사이킷런\n",
    "from sklearn import preprocessing\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import GradientBoostingClassifier, RandomForestClassifier, AdaBoostClassifier, BaggingClassifier, VotingClassifier\n",
    "from sklearn.metrics import accuracy_score, f1_score, roc_auc_score,log_loss\n",
    "from sklearn.model_selection import StratifiedKFold, cross_val_score, GridSearchCV\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "\n",
    "# 모델\n",
    "from lightgbm import LGBMClassifier\n",
    "from xgboost import XGBClassifier\n",
    "from catboost import CatBoostClassifier\n",
    "\n",
    "# 선형회귀분석\n",
    "import statsmodels.api as sm\n",
    "\n",
    "# optuna 파라미터 튜닝\n",
    "import optuna\n",
    "from optuna.samplers import TPESampler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./data/train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>educational-num</th>\n",
       "      <th>marital-status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>gender</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "      <th>native-country</th>\n",
       "      <th>income_&gt;50K</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>67</td>\n",
       "      <td>Private</td>\n",
       "      <td>366425</td>\n",
       "      <td>Doctorate</td>\n",
       "      <td>16</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>99999</td>\n",
       "      <td>0</td>\n",
       "      <td>60</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17</td>\n",
       "      <td>Private</td>\n",
       "      <td>244602</td>\n",
       "      <td>12th</td>\n",
       "      <td>8</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Own-child</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>United-States</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>31</td>\n",
       "      <td>Private</td>\n",
       "      <td>174201</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>110199</td>\n",
       "      <td>7th-8th</td>\n",
       "      <td>4</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Transport-moving</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>149248</td>\n",
       "      <td>Some-college</td>\n",
       "      <td>10</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>Black</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  workclass  fnlwgt     education  educational-num      marital-status  \\\n",
       "0   67    Private  366425     Doctorate               16            Divorced   \n",
       "1   17    Private  244602          12th                8       Never-married   \n",
       "2   31    Private  174201     Bachelors               13  Married-civ-spouse   \n",
       "3   58  State-gov  110199       7th-8th                4  Married-civ-spouse   \n",
       "4   25  State-gov  149248  Some-college               10       Never-married   \n",
       "\n",
       "         occupation   relationship   race gender  capital-gain  capital-loss  \\\n",
       "0   Exec-managerial  Not-in-family  White   Male         99999             0   \n",
       "1     Other-service      Own-child  White   Male             0             0   \n",
       "2   Exec-managerial        Husband  White   Male             0             0   \n",
       "3  Transport-moving        Husband  White   Male             0             0   \n",
       "4     Other-service  Not-in-family  Black   Male             0             0   \n",
       "\n",
       "   hours-per-week native-country  income_>50K  \n",
       "0              60  United-States            1  \n",
       "1              15  United-States            0  \n",
       "2              40  United-States            1  \n",
       "3              40  United-States            0  \n",
       "4              40  United-States            0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 43957 entries, 0 to 43956\n",
      "Data columns (total 15 columns):\n",
      " #   Column           Non-Null Count  Dtype \n",
      "---  ------           --------------  ----- \n",
      " 0   age              43957 non-null  int64 \n",
      " 1   workclass        41459 non-null  object\n",
      " 2   fnlwgt           43957 non-null  int64 \n",
      " 3   education        43957 non-null  object\n",
      " 4   educational-num  43957 non-null  int64 \n",
      " 5   marital-status   43957 non-null  object\n",
      " 6   occupation       41451 non-null  object\n",
      " 7   relationship     43957 non-null  object\n",
      " 8   race             43957 non-null  object\n",
      " 9   gender           43957 non-null  object\n",
      " 10  capital-gain     43957 non-null  int64 \n",
      " 11  capital-loss     43957 non-null  int64 \n",
      " 12  hours-per-week   43957 non-null  int64 \n",
      " 13  native-country   43194 non-null  object\n",
      " 14  income_>50K      43957 non-null  int64 \n",
      "dtypes: int64(7), object(8)\n",
      "memory usage: 5.0+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(43957, 15)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 전처리"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>educational-num</th>\n",
       "      <th>marital-status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>gender</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "      <th>native-country</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>67</td>\n",
       "      <td>Private</td>\n",
       "      <td>366425</td>\n",
       "      <td>Doctorate</td>\n",
       "      <td>16</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>99999</td>\n",
       "      <td>0</td>\n",
       "      <td>60</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17</td>\n",
       "      <td>Private</td>\n",
       "      <td>244602</td>\n",
       "      <td>12th</td>\n",
       "      <td>8</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Own-child</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>United-States</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>31</td>\n",
       "      <td>Private</td>\n",
       "      <td>174201</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>110199</td>\n",
       "      <td>7th-8th</td>\n",
       "      <td>4</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Transport-moving</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>149248</td>\n",
       "      <td>Some-college</td>\n",
       "      <td>10</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>Black</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  workclass  fnlwgt     education  educational-num      marital-status  \\\n",
       "0   67    Private  366425     Doctorate               16            Divorced   \n",
       "1   17    Private  244602          12th                8       Never-married   \n",
       "2   31    Private  174201     Bachelors               13  Married-civ-spouse   \n",
       "3   58  State-gov  110199       7th-8th                4  Married-civ-spouse   \n",
       "4   25  State-gov  149248  Some-college               10       Never-married   \n",
       "\n",
       "         occupation   relationship   race gender  capital-gain  capital-loss  \\\n",
       "0   Exec-managerial  Not-in-family  White   Male         99999             0   \n",
       "1     Other-service      Own-child  White   Male             0             0   \n",
       "2   Exec-managerial        Husband  White   Male             0             0   \n",
       "3  Transport-moving        Husband  White   Male             0             0   \n",
       "4     Other-service  Not-in-family  Black   Male             0             0   \n",
       "\n",
       "   hours-per-week native-country  income  \n",
       "0              60  United-States       1  \n",
       "1              15  United-States       0  \n",
       "2              40  United-States       1  \n",
       "3              40  United-States       0  \n",
       "4              40  United-States       0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_col = {'income_>50K':'income'}\n",
    "df = df.rename(columns=new_col)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "age                0\n",
       "workclass          0\n",
       "fnlwgt             0\n",
       "education          0\n",
       "educational-num    0\n",
       "marital-status     0\n",
       "occupation         0\n",
       "relationship       0\n",
       "race               0\n",
       "gender             0\n",
       "capital-gain       0\n",
       "capital-loss       0\n",
       "hours-per-week     0\n",
       "native-country     0\n",
       "income             0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#NULL값 처리\n",
    "\n",
    "df['workclass'] = df['workclass'].fillna('None')\n",
    "df['occupation'] = df['occupation'].fillna('None')\n",
    "df['native-country'] = df['native-country'].fillna('Others')\n",
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[df['native-country'] == 'United-States']\n",
    "df.drop(columns = ['native-country'], inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 39429 entries, 0 to 43956\n",
      "Data columns (total 14 columns):\n",
      " #   Column           Non-Null Count  Dtype \n",
      "---  ------           --------------  ----- \n",
      " 0   age              39429 non-null  int64 \n",
      " 1   workclass        39429 non-null  object\n",
      " 2   fnlwgt           39429 non-null  int64 \n",
      " 3   education        39429 non-null  object\n",
      " 4   educational-num  39429 non-null  int64 \n",
      " 5   marital-status   39429 non-null  object\n",
      " 6   occupation       39429 non-null  object\n",
      " 7   relationship     39429 non-null  object\n",
      " 8   race             39429 non-null  object\n",
      " 9   gender           39429 non-null  object\n",
      " 10  capital-gain     39429 non-null  int64 \n",
      " 11  capital-loss     39429 non-null  int64 \n",
      " 12  hours-per-week   39429 non-null  int64 \n",
      " 13  income           39429 non-null  int64 \n",
      "dtypes: int64(7), object(7)\n",
      "memory usage: 4.5+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(39429, 14)\n",
      "(39407, 14)\n",
      "3     13099\n",
      "4      9079\n",
      "7      6468\n",
      "2      3121\n",
      "8      2123\n",
      "5      1702\n",
      "6      1326\n",
      "1      1201\n",
      "9       655\n",
      "10      432\n",
      "0       201\n",
      "Name: educational-num, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# educational-num 편집\n",
    "print(df.shape)\n",
    "df = df[df['educational-num'] != 1]\n",
    "print(df.shape)\n",
    "df.loc[df['educational-num'].isin([2, 3]), 'educational-num'] = 0\n",
    "df.loc[df['educational-num'].isin([4, 5]), 'educational-num'] = 1\n",
    "df.loc[df['educational-num'].isin([6, 7, 8]), 'educational-num'] = 2\n",
    "for i in range(9, 17):\n",
    "    df.loc[df['educational-num'].isin([i]), 'educational-num'] = i - 6\n",
    "print(df['educational-num'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>educational-num</th>\n",
       "      <th>marital-status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>67</td>\n",
       "      <td>Private</td>\n",
       "      <td>366425</td>\n",
       "      <td>10</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>99999</td>\n",
       "      <td>0</td>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17</td>\n",
       "      <td>Private</td>\n",
       "      <td>244602</td>\n",
       "      <td>2</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Own-child</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>31</td>\n",
       "      <td>Private</td>\n",
       "      <td>174201</td>\n",
       "      <td>7</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Husband</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>110199</td>\n",
       "      <td>1</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Transport-moving</td>\n",
       "      <td>Husband</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>149248</td>\n",
       "      <td>4</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  workclass  fnlwgt  educational-num      marital-status  \\\n",
       "0   67    Private  366425               10            Divorced   \n",
       "1   17    Private  244602                2       Never-married   \n",
       "2   31    Private  174201                7  Married-civ-spouse   \n",
       "3   58  State-gov  110199                1  Married-civ-spouse   \n",
       "4   25  State-gov  149248                4       Never-married   \n",
       "\n",
       "         occupation   relationship  capital-gain  capital-loss  \\\n",
       "0   Exec-managerial  Not-in-family         99999             0   \n",
       "1     Other-service      Own-child             0             0   \n",
       "2   Exec-managerial        Husband             0             0   \n",
       "3  Transport-moving        Husband             0             0   \n",
       "4     Other-service  Not-in-family             0             0   \n",
       "\n",
       "   hours-per-week  income  \n",
       "0              60       1  \n",
       "1              15       0  \n",
       "2              40       1  \n",
       "3              40       0  \n",
       "4              40       0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.drop(columns = ['education', 'race', 'gender'], inplace = True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 39407 entries, 0 to 43956\n",
      "Data columns (total 11 columns):\n",
      " #   Column           Non-Null Count  Dtype \n",
      "---  ------           --------------  ----- \n",
      " 0   age              39407 non-null  int64 \n",
      " 1   workclass        39407 non-null  object\n",
      " 2   fnlwgt           39407 non-null  int64 \n",
      " 3   educational-num  39407 non-null  int64 \n",
      " 4   marital-status   39407 non-null  object\n",
      " 5   occupation       39407 non-null  object\n",
      " 6   relationship     39407 non-null  object\n",
      " 7   capital-gain     39407 non-null  int64 \n",
      " 8   capital-loss     39407 non-null  int64 \n",
      " 9   hours-per-week   39407 non-null  int64 \n",
      " 10  income           39407 non-null  int64 \n",
      "dtypes: int64(7), object(4)\n",
      "memory usage: 3.6+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 범주형 변수 처리"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.get_dummies(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 39407 entries, 0 to 43956\n",
      "Data columns (total 44 columns):\n",
      " #   Column                                Non-Null Count  Dtype\n",
      "---  ------                                --------------  -----\n",
      " 0   age                                   39407 non-null  int64\n",
      " 1   fnlwgt                                39407 non-null  int64\n",
      " 2   educational-num                       39407 non-null  int64\n",
      " 3   capital-gain                          39407 non-null  int64\n",
      " 4   capital-loss                          39407 non-null  int64\n",
      " 5   hours-per-week                        39407 non-null  int64\n",
      " 6   income                                39407 non-null  int64\n",
      " 7   workclass_Federal-gov                 39407 non-null  uint8\n",
      " 8   workclass_Local-gov                   39407 non-null  uint8\n",
      " 9   workclass_Never-worked                39407 non-null  uint8\n",
      " 10  workclass_None                        39407 non-null  uint8\n",
      " 11  workclass_Private                     39407 non-null  uint8\n",
      " 12  workclass_Self-emp-inc                39407 non-null  uint8\n",
      " 13  workclass_Self-emp-not-inc            39407 non-null  uint8\n",
      " 14  workclass_State-gov                   39407 non-null  uint8\n",
      " 15  workclass_Without-pay                 39407 non-null  uint8\n",
      " 16  marital-status_Divorced               39407 non-null  uint8\n",
      " 17  marital-status_Married-AF-spouse      39407 non-null  uint8\n",
      " 18  marital-status_Married-civ-spouse     39407 non-null  uint8\n",
      " 19  marital-status_Married-spouse-absent  39407 non-null  uint8\n",
      " 20  marital-status_Never-married          39407 non-null  uint8\n",
      " 21  marital-status_Separated              39407 non-null  uint8\n",
      " 22  marital-status_Widowed                39407 non-null  uint8\n",
      " 23  occupation_Adm-clerical               39407 non-null  uint8\n",
      " 24  occupation_Armed-Forces               39407 non-null  uint8\n",
      " 25  occupation_Craft-repair               39407 non-null  uint8\n",
      " 26  occupation_Exec-managerial            39407 non-null  uint8\n",
      " 27  occupation_Farming-fishing            39407 non-null  uint8\n",
      " 28  occupation_Handlers-cleaners          39407 non-null  uint8\n",
      " 29  occupation_Machine-op-inspct          39407 non-null  uint8\n",
      " 30  occupation_None                       39407 non-null  uint8\n",
      " 31  occupation_Other-service              39407 non-null  uint8\n",
      " 32  occupation_Priv-house-serv            39407 non-null  uint8\n",
      " 33  occupation_Prof-specialty             39407 non-null  uint8\n",
      " 34  occupation_Protective-serv            39407 non-null  uint8\n",
      " 35  occupation_Sales                      39407 non-null  uint8\n",
      " 36  occupation_Tech-support               39407 non-null  uint8\n",
      " 37  occupation_Transport-moving           39407 non-null  uint8\n",
      " 38  relationship_Husband                  39407 non-null  uint8\n",
      " 39  relationship_Not-in-family            39407 non-null  uint8\n",
      " 40  relationship_Other-relative           39407 non-null  uint8\n",
      " 41  relationship_Own-child                39407 non-null  uint8\n",
      " 42  relationship_Unmarried                39407 non-null  uint8\n",
      " 43  relationship_Wife                     39407 non-null  uint8\n",
      "dtypes: int64(7), uint8(37)\n",
      "memory usage: 3.8 MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### OLS 회귀분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>educational-num</td> <th>  R-squared:         </th> <td>   0.405</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.404</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   705.4</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Mon, 13 Dec 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:51:42</td>     <th>  Log-Likelihood:    </th> <td> -73957.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td> 39407</td>      <th>  AIC:               </th> <td>1.480e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td> 39368</td>      <th>  BIC:               </th> <td>1.483e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    38</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                    <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>                                  <td>   -0.0028</td> <td>    0.001</td> <td>   -3.730</td> <td> 0.000</td> <td>   -0.004</td> <td>   -0.001</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fnlwgt</th>                               <td>-1.581e-07</td> <td>  7.6e-08</td> <td>   -2.081</td> <td> 0.037</td> <td>-3.07e-07</td> <td> -9.2e-09</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-gain</th>                         <td> 1.211e-05</td> <td> 1.09e-06</td> <td>   11.140</td> <td> 0.000</td> <td> 9.98e-06</td> <td> 1.42e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-loss</th>                         <td>    0.0001</td> <td> 1.98e-05</td> <td>    6.228</td> <td> 0.000</td> <td> 8.47e-05</td> <td>    0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>hours-per-week</th>                       <td>    0.0089</td> <td>    0.001</td> <td>   12.514</td> <td> 0.000</td> <td>    0.008</td> <td>    0.010</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>income</th>                               <td>    0.9331</td> <td>    0.023</td> <td>   40.952</td> <td> 0.000</td> <td>    0.888</td> <td>    0.978</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Federal-gov</th>                <td>    0.8889</td> <td>    0.060</td> <td>   14.734</td> <td> 0.000</td> <td>    0.771</td> <td>    1.007</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Local-gov</th>                  <td>    1.0743</td> <td>    0.053</td> <td>   20.272</td> <td> 0.000</td> <td>    0.970</td> <td>    1.178</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Never-worked</th>               <td>   -0.4670</td> <td>    0.366</td> <td>   -1.276</td> <td> 0.202</td> <td>   -1.184</td> <td>    0.250</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_None</th>                       <td>    0.8191</td> <td>    0.196</td> <td>    4.190</td> <td> 0.000</td> <td>    0.436</td> <td>    1.202</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Private</th>                    <td>    0.7553</td> <td>    0.045</td> <td>   16.729</td> <td> 0.000</td> <td>    0.667</td> <td>    0.844</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-inc</th>               <td>    0.9126</td> <td>    0.060</td> <td>   15.132</td> <td> 0.000</td> <td>    0.794</td> <td>    1.031</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-not-inc</th>           <td>    0.9265</td> <td>    0.052</td> <td>   17.807</td> <td> 0.000</td> <td>    0.824</td> <td>    1.028</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_State-gov</th>                  <td>    1.2157</td> <td>    0.057</td> <td>   21.431</td> <td> 0.000</td> <td>    1.105</td> <td>    1.327</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Without-pay</th>                <td>    0.7643</td> <td>    0.329</td> <td>    2.326</td> <td> 0.020</td> <td>    0.120</td> <td>    1.408</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Divorced</th>              <td>    1.0698</td> <td>    0.055</td> <td>   19.395</td> <td> 0.000</td> <td>    0.962</td> <td>    1.178</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-AF-spouse</th>     <td>    0.7920</td> <td>    0.270</td> <td>    2.937</td> <td> 0.003</td> <td>    0.263</td> <td>    1.321</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-civ-spouse</th>    <td>    0.9539</td> <td>    0.084</td> <td>   11.397</td> <td> 0.000</td> <td>    0.790</td> <td>    1.118</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-spouse-absent</th> <td>    1.2516</td> <td>    0.090</td> <td>   13.970</td> <td> 0.000</td> <td>    1.076</td> <td>    1.427</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Never-married</th>         <td>    1.3226</td> <td>    0.051</td> <td>   25.800</td> <td> 0.000</td> <td>    1.222</td> <td>    1.423</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Separated</th>             <td>    0.8279</td> <td>    0.066</td> <td>   12.614</td> <td> 0.000</td> <td>    0.699</td> <td>    0.957</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Widowed</th>               <td>    0.6720</td> <td>    0.068</td> <td>    9.881</td> <td> 0.000</td> <td>    0.539</td> <td>    0.805</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Adm-clerical</th>              <td>    0.6560</td> <td>    0.040</td> <td>   16.560</td> <td> 0.000</td> <td>    0.578</td> <td>    0.734</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Armed-Forces</th>              <td>    0.6857</td> <td>    0.431</td> <td>    1.589</td> <td> 0.112</td> <td>   -0.160</td> <td>    1.531</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Craft-repair</th>              <td>   -0.1683</td> <td>    0.039</td> <td>   -4.262</td> <td> 0.000</td> <td>   -0.246</td> <td>   -0.091</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Exec-managerial</th>           <td>    1.5597</td> <td>    0.040</td> <td>   39.277</td> <td> 0.000</td> <td>    1.482</td> <td>    1.638</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Farming-fishing</th>           <td>   -0.1564</td> <td>    0.056</td> <td>   -2.794</td> <td> 0.005</td> <td>   -0.266</td> <td>   -0.047</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Handlers-cleaners</th>         <td>   -0.2647</td> <td>    0.050</td> <td>   -5.343</td> <td> 0.000</td> <td>   -0.362</td> <td>   -0.168</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Machine-op-inspct</th>         <td>   -0.3136</td> <td>    0.046</td> <td>   -6.868</td> <td> 0.000</td> <td>   -0.403</td> <td>   -0.224</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_None</th>                      <td>    0.3521</td> <td>    0.174</td> <td>    2.020</td> <td> 0.043</td> <td>    0.011</td> <td>    0.694</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Other-service</th>             <td>   -0.0299</td> <td>    0.041</td> <td>   -0.731</td> <td> 0.465</td> <td>   -0.110</td> <td>    0.050</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Priv-house-serv</th>           <td>   -0.2904</td> <td>    0.130</td> <td>   -2.235</td> <td> 0.025</td> <td>   -0.545</td> <td>   -0.036</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Prof-specialty</th>            <td>    2.9261</td> <td>    0.040</td> <td>   73.587</td> <td> 0.000</td> <td>    2.848</td> <td>    3.004</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Protective-serv</th>           <td>    0.3072</td> <td>    0.063</td> <td>    4.859</td> <td> 0.000</td> <td>    0.183</td> <td>    0.431</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Sales</th>                     <td>    0.8379</td> <td>    0.040</td> <td>   21.003</td> <td> 0.000</td> <td>    0.760</td> <td>    0.916</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Tech-support</th>              <td>    1.2726</td> <td>    0.054</td> <td>   23.617</td> <td> 0.000</td> <td>    1.167</td> <td>    1.378</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Transport-moving</th>          <td>   -0.4842</td> <td>    0.048</td> <td>  -10.158</td> <td> 0.000</td> <td>   -0.578</td> <td>   -0.391</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Husband</th>                 <td>    1.3558</td> <td>    0.081</td> <td>   16.823</td> <td> 0.000</td> <td>    1.198</td> <td>    1.514</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Not-in-family</th>           <td>    1.4214</td> <td>    0.050</td> <td>   28.709</td> <td> 0.000</td> <td>    1.324</td> <td>    1.518</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Other-relative</th>          <td>    0.9153</td> <td>    0.059</td> <td>   15.595</td> <td> 0.000</td> <td>    0.800</td> <td>    1.030</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Own-child</th>               <td>    0.9707</td> <td>    0.048</td> <td>   20.237</td> <td> 0.000</td> <td>    0.877</td> <td>    1.065</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Unmarried</th>               <td>    1.0829</td> <td>    0.053</td> <td>   20.612</td> <td> 0.000</td> <td>    0.980</td> <td>    1.186</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Wife</th>                    <td>    1.1437</td> <td>    0.085</td> <td>   13.507</td> <td> 0.000</td> <td>    0.978</td> <td>    1.310</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>827.444</td> <th>  Durbin-Watson:     </th> <td>   2.013</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 908.139</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.335</td>  <th>  Prob(JB):          </th> <td>6.31e-198</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.323</td>  <th>  Cond. No.          </th> <td>1.24e+16</td> \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The smallest eigenvalue is 1.19e-17. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:        educational-num   R-squared:                       0.405\n",
       "Model:                            OLS   Adj. R-squared:                  0.404\n",
       "Method:                 Least Squares   F-statistic:                     705.4\n",
       "Date:                Mon, 13 Dec 2021   Prob (F-statistic):               0.00\n",
       "Time:                        11:51:42   Log-Likelihood:                -73957.\n",
       "No. Observations:               39407   AIC:                         1.480e+05\n",
       "Df Residuals:                   39368   BIC:                         1.483e+05\n",
       "Df Model:                          38                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "========================================================================================================\n",
       "                                           coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------------------\n",
       "age                                     -0.0028      0.001     -3.730      0.000      -0.004      -0.001\n",
       "fnlwgt                               -1.581e-07    7.6e-08     -2.081      0.037   -3.07e-07    -9.2e-09\n",
       "capital-gain                          1.211e-05   1.09e-06     11.140      0.000    9.98e-06    1.42e-05\n",
       "capital-loss                             0.0001   1.98e-05      6.228      0.000    8.47e-05       0.000\n",
       "hours-per-week                           0.0089      0.001     12.514      0.000       0.008       0.010\n",
       "income                                   0.9331      0.023     40.952      0.000       0.888       0.978\n",
       "workclass_Federal-gov                    0.8889      0.060     14.734      0.000       0.771       1.007\n",
       "workclass_Local-gov                      1.0743      0.053     20.272      0.000       0.970       1.178\n",
       "workclass_Never-worked                  -0.4670      0.366     -1.276      0.202      -1.184       0.250\n",
       "workclass_None                           0.8191      0.196      4.190      0.000       0.436       1.202\n",
       "workclass_Private                        0.7553      0.045     16.729      0.000       0.667       0.844\n",
       "workclass_Self-emp-inc                   0.9126      0.060     15.132      0.000       0.794       1.031\n",
       "workclass_Self-emp-not-inc               0.9265      0.052     17.807      0.000       0.824       1.028\n",
       "workclass_State-gov                      1.2157      0.057     21.431      0.000       1.105       1.327\n",
       "workclass_Without-pay                    0.7643      0.329      2.326      0.020       0.120       1.408\n",
       "marital-status_Divorced                  1.0698      0.055     19.395      0.000       0.962       1.178\n",
       "marital-status_Married-AF-spouse         0.7920      0.270      2.937      0.003       0.263       1.321\n",
       "marital-status_Married-civ-spouse        0.9539      0.084     11.397      0.000       0.790       1.118\n",
       "marital-status_Married-spouse-absent     1.2516      0.090     13.970      0.000       1.076       1.427\n",
       "marital-status_Never-married             1.3226      0.051     25.800      0.000       1.222       1.423\n",
       "marital-status_Separated                 0.8279      0.066     12.614      0.000       0.699       0.957\n",
       "marital-status_Widowed                   0.6720      0.068      9.881      0.000       0.539       0.805\n",
       "occupation_Adm-clerical                  0.6560      0.040     16.560      0.000       0.578       0.734\n",
       "occupation_Armed-Forces                  0.6857      0.431      1.589      0.112      -0.160       1.531\n",
       "occupation_Craft-repair                 -0.1683      0.039     -4.262      0.000      -0.246      -0.091\n",
       "occupation_Exec-managerial               1.5597      0.040     39.277      0.000       1.482       1.638\n",
       "occupation_Farming-fishing              -0.1564      0.056     -2.794      0.005      -0.266      -0.047\n",
       "occupation_Handlers-cleaners            -0.2647      0.050     -5.343      0.000      -0.362      -0.168\n",
       "occupation_Machine-op-inspct            -0.3136      0.046     -6.868      0.000      -0.403      -0.224\n",
       "occupation_None                          0.3521      0.174      2.020      0.043       0.011       0.694\n",
       "occupation_Other-service                -0.0299      0.041     -0.731      0.465      -0.110       0.050\n",
       "occupation_Priv-house-serv              -0.2904      0.130     -2.235      0.025      -0.545      -0.036\n",
       "occupation_Prof-specialty                2.9261      0.040     73.587      0.000       2.848       3.004\n",
       "occupation_Protective-serv               0.3072      0.063      4.859      0.000       0.183       0.431\n",
       "occupation_Sales                         0.8379      0.040     21.003      0.000       0.760       0.916\n",
       "occupation_Tech-support                  1.2726      0.054     23.617      0.000       1.167       1.378\n",
       "occupation_Transport-moving             -0.4842      0.048    -10.158      0.000      -0.578      -0.391\n",
       "relationship_Husband                     1.3558      0.081     16.823      0.000       1.198       1.514\n",
       "relationship_Not-in-family               1.4214      0.050     28.709      0.000       1.324       1.518\n",
       "relationship_Other-relative              0.9153      0.059     15.595      0.000       0.800       1.030\n",
       "relationship_Own-child                   0.9707      0.048     20.237      0.000       0.877       1.065\n",
       "relationship_Unmarried                   1.0829      0.053     20.612      0.000       0.980       1.186\n",
       "relationship_Wife                        1.1437      0.085     13.507      0.000       0.978       1.310\n",
       "==============================================================================\n",
       "Omnibus:                      827.444   Durbin-Watson:                   2.013\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              908.139\n",
       "Skew:                           0.335   Prob(JB):                    6.31e-198\n",
       "Kurtosis:                       3.323   Cond. No.                     1.24e+16\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The smallest eigenvalue is 1.19e-17. This might indicate that there are\n",
       "strong multicollinearity problems or that the design matrix is singular.\n",
       "\"\"\""
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns = ['educational-num'])\n",
    "y = df[['educational-num']]\n",
    "\n",
    "# for b0, 상수항 추가\n",
    "# x_data = sm.add_constant(x_data, has_constant = \"add\")\n",
    "\n",
    "# OLS 검정\n",
    "multi_model = sm.OLS(y.astype(float), X.astype(float))\n",
    "fitted_multi_model = multi_model.fit()\n",
    "fitted_multi_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>educational-num</td> <th>  R-squared:         </th> <td>   0.405</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.405</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   655.6</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Mon, 13 Dec 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:51:42</td>     <th>  Log-Likelihood:    </th> <td> -67317.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td> 35617</td>      <th>  AIC:               </th> <td>1.347e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td> 35579</td>      <th>  BIC:               </th> <td>1.350e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    37</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                    <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>                                  <td>   -0.0020</td> <td>    0.001</td> <td>   -2.450</td> <td> 0.014</td> <td>   -0.004</td> <td>   -0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fnlwgt</th>                               <td>-1.618e-07</td> <td> 8.15e-08</td> <td>   -1.986</td> <td> 0.047</td> <td>-3.22e-07</td> <td> -2.1e-09</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-gain</th>                         <td> 1.215e-05</td> <td> 1.11e-06</td> <td>   10.935</td> <td> 0.000</td> <td> 9.97e-06</td> <td> 1.43e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-loss</th>                         <td>    0.0001</td> <td> 2.06e-05</td> <td>    6.328</td> <td> 0.000</td> <td>    9e-05</td> <td>    0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>hours-per-week</th>                       <td>    0.0094</td> <td>    0.001</td> <td>   12.326</td> <td> 0.000</td> <td>    0.008</td> <td>    0.011</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>income</th>                               <td>    0.9321</td> <td>    0.023</td> <td>   39.689</td> <td> 0.000</td> <td>    0.886</td> <td>    0.978</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Federal-gov</th>                <td>    0.8748</td> <td>    0.063</td> <td>   13.945</td> <td> 0.000</td> <td>    0.752</td> <td>    0.998</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Local-gov</th>                  <td>    1.1065</td> <td>    0.056</td> <td>   19.839</td> <td> 0.000</td> <td>    0.997</td> <td>    1.216</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Never-worked</th>               <td>   -0.4690</td> <td>    0.371</td> <td>   -1.265</td> <td> 0.206</td> <td>   -1.196</td> <td>    0.258</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_None</th>                       <td>    0.8082</td> <td>    0.198</td> <td>    4.076</td> <td> 0.000</td> <td>    0.420</td> <td>    1.197</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Private</th>                    <td>    0.7298</td> <td>    0.047</td> <td>   15.415</td> <td> 0.000</td> <td>    0.637</td> <td>    0.823</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-inc</th>               <td>    0.8813</td> <td>    0.063</td> <td>   14.073</td> <td> 0.000</td> <td>    0.759</td> <td>    1.004</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-not-inc</th>           <td>    0.8946</td> <td>    0.055</td> <td>   16.388</td> <td> 0.000</td> <td>    0.788</td> <td>    1.002</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_State-gov</th>                  <td>    1.2433</td> <td>    0.060</td> <td>   20.781</td> <td> 0.000</td> <td>    1.126</td> <td>    1.361</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Without-pay</th>                <td>    0.7400</td> <td>    0.342</td> <td>    2.164</td> <td> 0.030</td> <td>    0.070</td> <td>    1.410</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Divorced</th>              <td>    1.0349</td> <td>    0.060</td> <td>   17.325</td> <td> 0.000</td> <td>    0.918</td> <td>    1.152</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-AF-spouse</th>     <td>    0.7898</td> <td>    0.294</td> <td>    2.687</td> <td> 0.007</td> <td>    0.214</td> <td>    1.366</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-civ-spouse</th>    <td>    0.9439</td> <td>    0.090</td> <td>   10.457</td> <td> 0.000</td> <td>    0.767</td> <td>    1.121</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-spouse-absent</th> <td>    1.2724</td> <td>    0.097</td> <td>   13.094</td> <td> 0.000</td> <td>    1.082</td> <td>    1.463</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Never-married</th>         <td>    1.3176</td> <td>    0.056</td> <td>   23.682</td> <td> 0.000</td> <td>    1.209</td> <td>    1.427</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Separated</th>             <td>    0.8086</td> <td>    0.072</td> <td>   11.239</td> <td> 0.000</td> <td>    0.668</td> <td>    0.950</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Widowed</th>               <td>    0.6423</td> <td>    0.074</td> <td>    8.673</td> <td> 0.000</td> <td>    0.497</td> <td>    0.787</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Adm-clerical</th>              <td>    0.6500</td> <td>    0.042</td> <td>   15.460</td> <td> 0.000</td> <td>    0.568</td> <td>    0.732</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Armed-Forces</th>              <td>    0.6712</td> <td>    0.436</td> <td>    1.541</td> <td> 0.123</td> <td>   -0.183</td> <td>    1.525</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Craft-repair</th>              <td>   -0.1698</td> <td>    0.042</td> <td>   -4.052</td> <td> 0.000</td> <td>   -0.252</td> <td>   -0.088</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Exec-managerial</th>           <td>    1.5513</td> <td>    0.042</td> <td>   36.798</td> <td> 0.000</td> <td>    1.469</td> <td>    1.634</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Farming-fishing</th>           <td>   -0.1621</td> <td>    0.058</td> <td>   -2.792</td> <td> 0.005</td> <td>   -0.276</td> <td>   -0.048</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Handlers-cleaners</th>         <td>   -0.2631</td> <td>    0.052</td> <td>   -5.088</td> <td> 0.000</td> <td>   -0.364</td> <td>   -0.162</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Machine-op-inspct</th>         <td>   -0.3122</td> <td>    0.048</td> <td>   -6.515</td> <td> 0.000</td> <td>   -0.406</td> <td>   -0.218</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_None</th>                      <td>    0.3392</td> <td>    0.177</td> <td>    1.920</td> <td> 0.055</td> <td>   -0.007</td> <td>    0.685</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Priv-house-serv</th>           <td>   -0.2901</td> <td>    0.132</td> <td>   -2.200</td> <td> 0.028</td> <td>   -0.549</td> <td>   -0.032</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Prof-specialty</th>            <td>    2.9084</td> <td>    0.042</td> <td>   68.908</td> <td> 0.000</td> <td>    2.826</td> <td>    2.991</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Protective-serv</th>           <td>    0.2703</td> <td>    0.065</td> <td>    4.140</td> <td> 0.000</td> <td>    0.142</td> <td>    0.398</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Sales</th>                     <td>    0.8391</td> <td>    0.042</td> <td>   19.826</td> <td> 0.000</td> <td>    0.756</td> <td>    0.922</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Tech-support</th>              <td>    1.2684</td> <td>    0.056</td> <td>   22.694</td> <td> 0.000</td> <td>    1.159</td> <td>    1.378</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Transport-moving</th>          <td>   -0.4913</td> <td>    0.050</td> <td>   -9.846</td> <td> 0.000</td> <td>   -0.589</td> <td>   -0.393</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Husband</th>                 <td>    1.3341</td> <td>    0.087</td> <td>   15.373</td> <td> 0.000</td> <td>    1.164</td> <td>    1.504</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Not-in-family</th>           <td>    1.4138</td> <td>    0.053</td> <td>   26.581</td> <td> 0.000</td> <td>    1.310</td> <td>    1.518</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Other-relative</th>          <td>    0.9133</td> <td>    0.064</td> <td>   14.194</td> <td> 0.000</td> <td>    0.787</td> <td>    1.039</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Own-child</th>               <td>    0.9570</td> <td>    0.052</td> <td>   18.506</td> <td> 0.000</td> <td>    0.856</td> <td>    1.058</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Unmarried</th>               <td>    1.0717</td> <td>    0.057</td> <td>   18.885</td> <td> 0.000</td> <td>    0.960</td> <td>    1.183</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Wife</th>                    <td>    1.1195</td> <td>    0.091</td> <td>   12.267</td> <td> 0.000</td> <td>    0.941</td> <td>    1.298</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>537.997</td> <th>  Durbin-Watson:     </th> <td>   2.016</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 570.000</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.290</td>  <th>  Prob(JB):          </th> <td>1.68e-124</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.217</td>  <th>  Cond. No.          </th> <td>1.23e+16</td> \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:        educational-num   R-squared:                       0.405\n",
       "Model:                            OLS   Adj. R-squared:                  0.405\n",
       "Method:                 Least Squares   F-statistic:                     655.6\n",
       "Date:                Mon, 13 Dec 2021   Prob (F-statistic):               0.00\n",
       "Time:                        11:51:42   Log-Likelihood:                -67317.\n",
       "No. Observations:               35617   AIC:                         1.347e+05\n",
       "Df Residuals:                   35579   BIC:                         1.350e+05\n",
       "Df Model:                          37                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "========================================================================================================\n",
       "                                           coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------------------\n",
       "age                                     -0.0020      0.001     -2.450      0.014      -0.004      -0.000\n",
       "fnlwgt                               -1.618e-07   8.15e-08     -1.986      0.047   -3.22e-07    -2.1e-09\n",
       "capital-gain                          1.215e-05   1.11e-06     10.935      0.000    9.97e-06    1.43e-05\n",
       "capital-loss                             0.0001   2.06e-05      6.328      0.000       9e-05       0.000\n",
       "hours-per-week                           0.0094      0.001     12.326      0.000       0.008       0.011\n",
       "income                                   0.9321      0.023     39.689      0.000       0.886       0.978\n",
       "workclass_Federal-gov                    0.8748      0.063     13.945      0.000       0.752       0.998\n",
       "workclass_Local-gov                      1.1065      0.056     19.839      0.000       0.997       1.216\n",
       "workclass_Never-worked                  -0.4690      0.371     -1.265      0.206      -1.196       0.258\n",
       "workclass_None                           0.8082      0.198      4.076      0.000       0.420       1.197\n",
       "workclass_Private                        0.7298      0.047     15.415      0.000       0.637       0.823\n",
       "workclass_Self-emp-inc                   0.8813      0.063     14.073      0.000       0.759       1.004\n",
       "workclass_Self-emp-not-inc               0.8946      0.055     16.388      0.000       0.788       1.002\n",
       "workclass_State-gov                      1.2433      0.060     20.781      0.000       1.126       1.361\n",
       "workclass_Without-pay                    0.7400      0.342      2.164      0.030       0.070       1.410\n",
       "marital-status_Divorced                  1.0349      0.060     17.325      0.000       0.918       1.152\n",
       "marital-status_Married-AF-spouse         0.7898      0.294      2.687      0.007       0.214       1.366\n",
       "marital-status_Married-civ-spouse        0.9439      0.090     10.457      0.000       0.767       1.121\n",
       "marital-status_Married-spouse-absent     1.2724      0.097     13.094      0.000       1.082       1.463\n",
       "marital-status_Never-married             1.3176      0.056     23.682      0.000       1.209       1.427\n",
       "marital-status_Separated                 0.8086      0.072     11.239      0.000       0.668       0.950\n",
       "marital-status_Widowed                   0.6423      0.074      8.673      0.000       0.497       0.787\n",
       "occupation_Adm-clerical                  0.6500      0.042     15.460      0.000       0.568       0.732\n",
       "occupation_Armed-Forces                  0.6712      0.436      1.541      0.123      -0.183       1.525\n",
       "occupation_Craft-repair                 -0.1698      0.042     -4.052      0.000      -0.252      -0.088\n",
       "occupation_Exec-managerial               1.5513      0.042     36.798      0.000       1.469       1.634\n",
       "occupation_Farming-fishing              -0.1621      0.058     -2.792      0.005      -0.276      -0.048\n",
       "occupation_Handlers-cleaners            -0.2631      0.052     -5.088      0.000      -0.364      -0.162\n",
       "occupation_Machine-op-inspct            -0.3122      0.048     -6.515      0.000      -0.406      -0.218\n",
       "occupation_None                          0.3392      0.177      1.920      0.055      -0.007       0.685\n",
       "occupation_Priv-house-serv              -0.2901      0.132     -2.200      0.028      -0.549      -0.032\n",
       "occupation_Prof-specialty                2.9084      0.042     68.908      0.000       2.826       2.991\n",
       "occupation_Protective-serv               0.2703      0.065      4.140      0.000       0.142       0.398\n",
       "occupation_Sales                         0.8391      0.042     19.826      0.000       0.756       0.922\n",
       "occupation_Tech-support                  1.2684      0.056     22.694      0.000       1.159       1.378\n",
       "occupation_Transport-moving             -0.4913      0.050     -9.846      0.000      -0.589      -0.393\n",
       "relationship_Husband                     1.3341      0.087     15.373      0.000       1.164       1.504\n",
       "relationship_Not-in-family               1.4138      0.053     26.581      0.000       1.310       1.518\n",
       "relationship_Other-relative              0.9133      0.064     14.194      0.000       0.787       1.039\n",
       "relationship_Own-child                   0.9570      0.052     18.506      0.000       0.856       1.058\n",
       "relationship_Unmarried                   1.0717      0.057     18.885      0.000       0.960       1.183\n",
       "relationship_Wife                        1.1195      0.091     12.267      0.000       0.941       1.298\n",
       "==============================================================================\n",
       "Omnibus:                      537.997   Durbin-Watson:                   2.016\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              570.000\n",
       "Skew:                           0.290   Prob(JB):                    1.68e-124\n",
       "Kurtosis:                       3.217   Cond. No.                     1.23e+16\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are\n",
       "strong multicollinearity problems or that the design matrix is singular.\n",
       "\"\"\""
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drop_occupation = [f'occupation_{occupation}' for occupation in ['Other-service']]\n",
    "df = df[(df['occupation_Other-service'] == 0)]\n",
    "\n",
    "X = df.drop(columns = ['educational-num'] + drop_occupation)\n",
    "y = df[['educational-num']]\n",
    "\n",
    "\n",
    "# OLS 검정\n",
    "multi_model = sm.OLS(y.astype(float), X.astype(float))\n",
    "fitted_multi_model = multi_model.fit()\n",
    "fitted_multi_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>educational-num</td> <th>  R-squared:         </th> <td>   0.405</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.405</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   673.8</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Mon, 13 Dec 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:51:42</td>     <th>  Log-Likelihood:    </th> <td> -67292.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td> 35605</td>      <th>  AIC:               </th> <td>1.347e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td> 35568</td>      <th>  BIC:               </th> <td>1.350e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    36</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                    <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>                                  <td>   -0.0020</td> <td>    0.001</td> <td>   -2.461</td> <td> 0.014</td> <td>   -0.004</td> <td>   -0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fnlwgt</th>                               <td>-1.628e-07</td> <td> 8.15e-08</td> <td>   -1.998</td> <td> 0.046</td> <td>-3.23e-07</td> <td> -3.1e-09</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-gain</th>                         <td> 1.215e-05</td> <td> 1.11e-06</td> <td>   10.933</td> <td> 0.000</td> <td> 9.97e-06</td> <td> 1.43e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-loss</th>                         <td>    0.0001</td> <td> 2.06e-05</td> <td>    6.334</td> <td> 0.000</td> <td> 9.02e-05</td> <td>    0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>hours-per-week</th>                       <td>    0.0094</td> <td>    0.001</td> <td>   12.338</td> <td> 0.000</td> <td>    0.008</td> <td>    0.011</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>income</th>                               <td>    0.9312</td> <td>    0.023</td> <td>   39.649</td> <td> 0.000</td> <td>    0.885</td> <td>    0.977</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Federal-gov</th>                <td>    0.8584</td> <td>    0.063</td> <td>   13.713</td> <td> 0.000</td> <td>    0.736</td> <td>    0.981</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Local-gov</th>                  <td>    1.0900</td> <td>    0.055</td> <td>   19.936</td> <td> 0.000</td> <td>    0.983</td> <td>    1.197</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Never-worked</th>               <td>   -0.4593</td> <td>    0.371</td> <td>   -1.239</td> <td> 0.215</td> <td>   -1.186</td> <td>    0.267</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_None</th>                       <td>    0.8182</td> <td>    0.198</td> <td>    4.127</td> <td> 0.000</td> <td>    0.430</td> <td>    1.207</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Private</th>                    <td>    0.7133</td> <td>    0.046</td> <td>   15.492</td> <td> 0.000</td> <td>    0.623</td> <td>    0.804</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-inc</th>               <td>    0.8651</td> <td>    0.062</td> <td>   14.037</td> <td> 0.000</td> <td>    0.744</td> <td>    0.986</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-not-inc</th>           <td>    0.8781</td> <td>    0.053</td> <td>   16.431</td> <td> 0.000</td> <td>    0.773</td> <td>    0.983</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_State-gov</th>                  <td>    1.2268</td> <td>    0.059</td> <td>   20.861</td> <td> 0.000</td> <td>    1.112</td> <td>    1.342</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Without-pay</th>                <td>    0.7231</td> <td>    0.342</td> <td>    2.117</td> <td> 0.034</td> <td>    0.054</td> <td>    1.393</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Divorced</th>              <td>    1.0202</td> <td>    0.059</td> <td>   17.275</td> <td> 0.000</td> <td>    0.904</td> <td>    1.136</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-AF-spouse</th>     <td>    0.7784</td> <td>    0.294</td> <td>    2.651</td> <td> 0.008</td> <td>    0.203</td> <td>    1.354</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-civ-spouse</th>    <td>    0.9327</td> <td>    0.090</td> <td>   10.344</td> <td> 0.000</td> <td>    0.756</td> <td>    1.109</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-spouse-absent</th> <td>    1.2578</td> <td>    0.097</td> <td>   13.003</td> <td> 0.000</td> <td>    1.068</td> <td>    1.447</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Never-married</th>         <td>    1.3030</td> <td>    0.055</td> <td>   23.700</td> <td> 0.000</td> <td>    1.195</td> <td>    1.411</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Separated</th>             <td>    0.7939</td> <td>    0.071</td> <td>   11.123</td> <td> 0.000</td> <td>    0.654</td> <td>    0.934</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Widowed</th>               <td>    0.6279</td> <td>    0.073</td> <td>    8.542</td> <td> 0.000</td> <td>    0.484</td> <td>    0.772</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Adm-clerical</th>              <td>    0.6962</td> <td>    0.030</td> <td>   22.880</td> <td> 0.000</td> <td>    0.637</td> <td>    0.756</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Craft-repair</th>              <td>   -0.1234</td> <td>    0.030</td> <td>   -4.129</td> <td> 0.000</td> <td>   -0.182</td> <td>   -0.065</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Exec-managerial</th>           <td>    1.5979</td> <td>    0.030</td> <td>   52.571</td> <td> 0.000</td> <td>    1.538</td> <td>    1.657</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Farming-fishing</th>           <td>   -0.1159</td> <td>    0.050</td> <td>   -2.324</td> <td> 0.020</td> <td>   -0.214</td> <td>   -0.018</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Handlers-cleaners</th>         <td>   -0.2169</td> <td>    0.042</td> <td>   -5.106</td> <td> 0.000</td> <td>   -0.300</td> <td>   -0.134</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Machine-op-inspct</th>         <td>   -0.2659</td> <td>    0.038</td> <td>   -7.042</td> <td> 0.000</td> <td>   -0.340</td> <td>   -0.192</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_None</th>                      <td>    0.3589</td> <td>    0.176</td> <td>    2.040</td> <td> 0.041</td> <td>    0.014</td> <td>    0.704</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Priv-house-serv</th>           <td>   -0.2439</td> <td>    0.128</td> <td>   -1.907</td> <td> 0.057</td> <td>   -0.495</td> <td>    0.007</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Prof-specialty</th>            <td>    2.9550</td> <td>    0.030</td> <td>   97.082</td> <td> 0.000</td> <td>    2.895</td> <td>    3.015</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Protective-serv</th>           <td>    0.3167</td> <td>    0.058</td> <td>    5.438</td> <td> 0.000</td> <td>    0.203</td> <td>    0.431</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Sales</th>                     <td>    0.8855</td> <td>    0.030</td> <td>   29.113</td> <td> 0.000</td> <td>    0.826</td> <td>    0.945</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Tech-support</th>              <td>    1.3148</td> <td>    0.048</td> <td>   27.650</td> <td> 0.000</td> <td>    1.222</td> <td>    1.408</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Transport-moving</th>          <td>   -0.4449</td> <td>    0.040</td> <td>  -11.049</td> <td> 0.000</td> <td>   -0.524</td> <td>   -0.366</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Husband</th>                 <td>    1.3157</td> <td>    0.086</td> <td>   15.288</td> <td> 0.000</td> <td>    1.147</td> <td>    1.484</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Not-in-family</th>           <td>    1.3988</td> <td>    0.052</td> <td>   26.763</td> <td> 0.000</td> <td>    1.296</td> <td>    1.501</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Other-relative</th>          <td>    0.8990</td> <td>    0.064</td> <td>   14.106</td> <td> 0.000</td> <td>    0.774</td> <td>    1.024</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Own-child</th>               <td>    0.9423</td> <td>    0.051</td> <td>   18.565</td> <td> 0.000</td> <td>    0.843</td> <td>    1.042</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Unmarried</th>               <td>    1.0567</td> <td>    0.056</td> <td>   18.913</td> <td> 0.000</td> <td>    0.947</td> <td>    1.166</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Wife</th>                    <td>    1.1014</td> <td>    0.091</td> <td>   12.160</td> <td> 0.000</td> <td>    0.924</td> <td>    1.279</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>537.065</td> <th>  Durbin-Watson:     </th> <td>   2.016</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 569.036</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.290</td>  <th>  Prob(JB):          </th> <td>2.72e-124</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.217</td>  <th>  Cond. No.          </th> <td>1.23e+16</td> \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:        educational-num   R-squared:                       0.405\n",
       "Model:                            OLS   Adj. R-squared:                  0.405\n",
       "Method:                 Least Squares   F-statistic:                     673.8\n",
       "Date:                Mon, 13 Dec 2021   Prob (F-statistic):               0.00\n",
       "Time:                        11:51:42   Log-Likelihood:                -67292.\n",
       "No. Observations:               35605   AIC:                         1.347e+05\n",
       "Df Residuals:                   35568   BIC:                         1.350e+05\n",
       "Df Model:                          36                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "========================================================================================================\n",
       "                                           coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------------------\n",
       "age                                     -0.0020      0.001     -2.461      0.014      -0.004      -0.000\n",
       "fnlwgt                               -1.628e-07   8.15e-08     -1.998      0.046   -3.23e-07    -3.1e-09\n",
       "capital-gain                          1.215e-05   1.11e-06     10.933      0.000    9.97e-06    1.43e-05\n",
       "capital-loss                             0.0001   2.06e-05      6.334      0.000    9.02e-05       0.000\n",
       "hours-per-week                           0.0094      0.001     12.338      0.000       0.008       0.011\n",
       "income                                   0.9312      0.023     39.649      0.000       0.885       0.977\n",
       "workclass_Federal-gov                    0.8584      0.063     13.713      0.000       0.736       0.981\n",
       "workclass_Local-gov                      1.0900      0.055     19.936      0.000       0.983       1.197\n",
       "workclass_Never-worked                  -0.4593      0.371     -1.239      0.215      -1.186       0.267\n",
       "workclass_None                           0.8182      0.198      4.127      0.000       0.430       1.207\n",
       "workclass_Private                        0.7133      0.046     15.492      0.000       0.623       0.804\n",
       "workclass_Self-emp-inc                   0.8651      0.062     14.037      0.000       0.744       0.986\n",
       "workclass_Self-emp-not-inc               0.8781      0.053     16.431      0.000       0.773       0.983\n",
       "workclass_State-gov                      1.2268      0.059     20.861      0.000       1.112       1.342\n",
       "workclass_Without-pay                    0.7231      0.342      2.117      0.034       0.054       1.393\n",
       "marital-status_Divorced                  1.0202      0.059     17.275      0.000       0.904       1.136\n",
       "marital-status_Married-AF-spouse         0.7784      0.294      2.651      0.008       0.203       1.354\n",
       "marital-status_Married-civ-spouse        0.9327      0.090     10.344      0.000       0.756       1.109\n",
       "marital-status_Married-spouse-absent     1.2578      0.097     13.003      0.000       1.068       1.447\n",
       "marital-status_Never-married             1.3030      0.055     23.700      0.000       1.195       1.411\n",
       "marital-status_Separated                 0.7939      0.071     11.123      0.000       0.654       0.934\n",
       "marital-status_Widowed                   0.6279      0.073      8.542      0.000       0.484       0.772\n",
       "occupation_Adm-clerical                  0.6962      0.030     22.880      0.000       0.637       0.756\n",
       "occupation_Craft-repair                 -0.1234      0.030     -4.129      0.000      -0.182      -0.065\n",
       "occupation_Exec-managerial               1.5979      0.030     52.571      0.000       1.538       1.657\n",
       "occupation_Farming-fishing              -0.1159      0.050     -2.324      0.020      -0.214      -0.018\n",
       "occupation_Handlers-cleaners            -0.2169      0.042     -5.106      0.000      -0.300      -0.134\n",
       "occupation_Machine-op-inspct            -0.2659      0.038     -7.042      0.000      -0.340      -0.192\n",
       "occupation_None                          0.3589      0.176      2.040      0.041       0.014       0.704\n",
       "occupation_Priv-house-serv              -0.2439      0.128     -1.907      0.057      -0.495       0.007\n",
       "occupation_Prof-specialty                2.9550      0.030     97.082      0.000       2.895       3.015\n",
       "occupation_Protective-serv               0.3167      0.058      5.438      0.000       0.203       0.431\n",
       "occupation_Sales                         0.8855      0.030     29.113      0.000       0.826       0.945\n",
       "occupation_Tech-support                  1.3148      0.048     27.650      0.000       1.222       1.408\n",
       "occupation_Transport-moving             -0.4449      0.040    -11.049      0.000      -0.524      -0.366\n",
       "relationship_Husband                     1.3157      0.086     15.288      0.000       1.147       1.484\n",
       "relationship_Not-in-family               1.3988      0.052     26.763      0.000       1.296       1.501\n",
       "relationship_Other-relative              0.8990      0.064     14.106      0.000       0.774       1.024\n",
       "relationship_Own-child                   0.9423      0.051     18.565      0.000       0.843       1.042\n",
       "relationship_Unmarried                   1.0567      0.056     18.913      0.000       0.947       1.166\n",
       "relationship_Wife                        1.1014      0.091     12.160      0.000       0.924       1.279\n",
       "==============================================================================\n",
       "Omnibus:                      537.065   Durbin-Watson:                   2.016\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              569.036\n",
       "Skew:                           0.290   Prob(JB):                    2.72e-124\n",
       "Kurtosis:                       3.217   Cond. No.                     1.23e+16\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are\n",
       "strong multicollinearity problems or that the design matrix is singular.\n",
       "\"\"\""
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drop_occupation = [f'occupation_{occupation}' for occupation in ['Armed-Forces', 'Other-service']]\n",
    "df = df[(df['occupation_Armed-Forces'] == 0) & (df['occupation_Other-service'] == 0)]\n",
    "\n",
    "X = df.drop(columns = ['educational-num'] + drop_occupation)\n",
    "y = df[['educational-num']]\n",
    "\n",
    "\n",
    "# OLS 검정\n",
    "multi_model = sm.OLS(y.astype(float), X.astype(float))\n",
    "fitted_multi_model = multi_model.fit()\n",
    "fitted_multi_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>educational-num</td> <th>  R-squared:         </th> <td>   0.405</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.404</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   688.0</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Mon, 13 Dec 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:51:43</td>     <th>  Log-Likelihood:    </th> <td> -67037.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td> 35464</td>      <th>  AIC:               </th> <td>1.341e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td> 35428</td>      <th>  BIC:               </th> <td>1.345e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    35</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                    <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>                                  <td>   -0.0019</td> <td>    0.001</td> <td>   -2.369</td> <td> 0.018</td> <td>   -0.003</td> <td>   -0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fnlwgt</th>                               <td>-1.527e-07</td> <td> 8.17e-08</td> <td>   -1.870</td> <td> 0.061</td> <td>-3.13e-07</td> <td> 7.31e-09</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-gain</th>                         <td> 1.212e-05</td> <td> 1.11e-06</td> <td>   10.908</td> <td> 0.000</td> <td> 9.95e-06</td> <td> 1.43e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-loss</th>                         <td>    0.0001</td> <td> 2.06e-05</td> <td>    6.365</td> <td> 0.000</td> <td> 9.09e-05</td> <td>    0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>hours-per-week</th>                       <td>    0.0094</td> <td>    0.001</td> <td>   12.188</td> <td> 0.000</td> <td>    0.008</td> <td>    0.011</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>income</th>                               <td>    0.9299</td> <td>    0.024</td> <td>   39.564</td> <td> 0.000</td> <td>    0.884</td> <td>    0.976</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Federal-gov</th>                <td>    0.8649</td> <td>    0.063</td> <td>   13.821</td> <td> 0.000</td> <td>    0.742</td> <td>    0.988</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Local-gov</th>                  <td>    1.0963</td> <td>    0.055</td> <td>   20.069</td> <td> 0.000</td> <td>    0.989</td> <td>    1.203</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Never-worked</th>               <td>   -0.4634</td> <td>    0.371</td> <td>   -1.250</td> <td> 0.211</td> <td>   -1.190</td> <td>    0.263</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_None</th>                       <td>    0.8137</td> <td>    0.198</td> <td>    4.099</td> <td> 0.000</td> <td>    0.425</td> <td>    1.203</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Private</th>                    <td>    0.7197</td> <td>    0.046</td> <td>   15.615</td> <td> 0.000</td> <td>    0.629</td> <td>    0.810</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-inc</th>               <td>    0.8717</td> <td>    0.062</td> <td>   14.146</td> <td> 0.000</td> <td>    0.751</td> <td>    0.992</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-not-inc</th>           <td>    0.8843</td> <td>    0.053</td> <td>   16.556</td> <td> 0.000</td> <td>    0.780</td> <td>    0.989</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_State-gov</th>                  <td>    1.2331</td> <td>    0.059</td> <td>   20.981</td> <td> 0.000</td> <td>    1.118</td> <td>    1.348</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Without-pay</th>                <td>    0.7297</td> <td>    0.341</td> <td>    2.138</td> <td> 0.033</td> <td>    0.061</td> <td>    1.399</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Divorced</th>              <td>    1.0295</td> <td>    0.059</td> <td>   17.369</td> <td> 0.000</td> <td>    0.913</td> <td>    1.146</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-AF-spouse</th>     <td>    0.7694</td> <td>    0.294</td> <td>    2.618</td> <td> 0.009</td> <td>    0.193</td> <td>    1.345</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-civ-spouse</th>    <td>    0.9228</td> <td>    0.091</td> <td>   10.136</td> <td> 0.000</td> <td>    0.744</td> <td>    1.101</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-spouse-absent</th> <td>    1.2689</td> <td>    0.097</td> <td>   13.093</td> <td> 0.000</td> <td>    1.079</td> <td>    1.459</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Never-married</th>         <td>    1.3121</td> <td>    0.055</td> <td>   23.779</td> <td> 0.000</td> <td>    1.204</td> <td>    1.420</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Separated</th>             <td>    0.8043</td> <td>    0.072</td> <td>   11.222</td> <td> 0.000</td> <td>    0.664</td> <td>    0.945</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Widowed</th>               <td>    0.6429</td> <td>    0.074</td> <td>    8.672</td> <td> 0.000</td> <td>    0.498</td> <td>    0.788</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Adm-clerical</th>              <td>    0.6780</td> <td>    0.030</td> <td>   22.690</td> <td> 0.000</td> <td>    0.619</td> <td>    0.737</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Craft-repair</th>              <td>   -0.1415</td> <td>    0.029</td> <td>   -4.851</td> <td> 0.000</td> <td>   -0.199</td> <td>   -0.084</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Exec-managerial</th>           <td>    1.5800</td> <td>    0.030</td> <td>   53.159</td> <td> 0.000</td> <td>    1.522</td> <td>    1.638</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Farming-fishing</th>           <td>   -0.1339</td> <td>    0.049</td> <td>   -2.719</td> <td> 0.007</td> <td>   -0.230</td> <td>   -0.037</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Handlers-cleaners</th>         <td>   -0.2353</td> <td>    0.042</td> <td>   -5.612</td> <td> 0.000</td> <td>   -0.317</td> <td>   -0.153</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Machine-op-inspct</th>         <td>   -0.2842</td> <td>    0.037</td> <td>   -7.635</td> <td> 0.000</td> <td>   -0.357</td> <td>   -0.211</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_None</th>                      <td>    0.3503</td> <td>    0.176</td> <td>    1.994</td> <td> 0.046</td> <td>    0.006</td> <td>    0.695</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Prof-specialty</th>            <td>    2.9370</td> <td>    0.030</td> <td>   98.860</td> <td> 0.000</td> <td>    2.879</td> <td>    2.995</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Protective-serv</th>           <td>    0.2988</td> <td>    0.058</td> <td>    5.193</td> <td> 0.000</td> <td>    0.186</td> <td>    0.412</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Sales</th>                     <td>    0.8673</td> <td>    0.030</td> <td>   29.113</td> <td> 0.000</td> <td>    0.809</td> <td>    0.926</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Tech-support</th>              <td>    1.2967</td> <td>    0.047</td> <td>   27.600</td> <td> 0.000</td> <td>    1.205</td> <td>    1.389</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Transport-moving</th>          <td>   -0.4631</td> <td>    0.040</td> <td>  -11.670</td> <td> 0.000</td> <td>   -0.541</td> <td>   -0.385</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Husband</th>                 <td>    1.3357</td> <td>    0.087</td> <td>   15.408</td> <td> 0.000</td> <td>    1.166</td> <td>    1.506</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Not-in-family</th>           <td>    1.4003</td> <td>    0.052</td> <td>   26.710</td> <td> 0.000</td> <td>    1.298</td> <td>    1.503</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Other-relative</th>          <td>    0.8934</td> <td>    0.064</td> <td>   13.932</td> <td> 0.000</td> <td>    0.768</td> <td>    1.019</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Own-child</th>               <td>    0.9453</td> <td>    0.051</td> <td>   18.571</td> <td> 0.000</td> <td>    0.845</td> <td>    1.045</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Unmarried</th>               <td>    1.0525</td> <td>    0.056</td> <td>   18.767</td> <td> 0.000</td> <td>    0.943</td> <td>    1.162</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Wife</th>                    <td>    1.1228</td> <td>    0.091</td> <td>   12.304</td> <td> 0.000</td> <td>    0.944</td> <td>    1.302</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>529.927</td> <th>  Durbin-Watson:     </th> <td>   2.016</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 561.059</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.289</td>  <th>  Prob(JB):          </th> <td>1.47e-122</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.215</td>  <th>  Cond. No.          </th> <td>1.23e+16</td> \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:        educational-num   R-squared:                       0.405\n",
       "Model:                            OLS   Adj. R-squared:                  0.404\n",
       "Method:                 Least Squares   F-statistic:                     688.0\n",
       "Date:                Mon, 13 Dec 2021   Prob (F-statistic):               0.00\n",
       "Time:                        11:51:43   Log-Likelihood:                -67037.\n",
       "No. Observations:               35464   AIC:                         1.341e+05\n",
       "Df Residuals:                   35428   BIC:                         1.345e+05\n",
       "Df Model:                          35                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "========================================================================================================\n",
       "                                           coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------------------\n",
       "age                                     -0.0019      0.001     -2.369      0.018      -0.003      -0.000\n",
       "fnlwgt                               -1.527e-07   8.17e-08     -1.870      0.061   -3.13e-07    7.31e-09\n",
       "capital-gain                          1.212e-05   1.11e-06     10.908      0.000    9.95e-06    1.43e-05\n",
       "capital-loss                             0.0001   2.06e-05      6.365      0.000    9.09e-05       0.000\n",
       "hours-per-week                           0.0094      0.001     12.188      0.000       0.008       0.011\n",
       "income                                   0.9299      0.024     39.564      0.000       0.884       0.976\n",
       "workclass_Federal-gov                    0.8649      0.063     13.821      0.000       0.742       0.988\n",
       "workclass_Local-gov                      1.0963      0.055     20.069      0.000       0.989       1.203\n",
       "workclass_Never-worked                  -0.4634      0.371     -1.250      0.211      -1.190       0.263\n",
       "workclass_None                           0.8137      0.198      4.099      0.000       0.425       1.203\n",
       "workclass_Private                        0.7197      0.046     15.615      0.000       0.629       0.810\n",
       "workclass_Self-emp-inc                   0.8717      0.062     14.146      0.000       0.751       0.992\n",
       "workclass_Self-emp-not-inc               0.8843      0.053     16.556      0.000       0.780       0.989\n",
       "workclass_State-gov                      1.2331      0.059     20.981      0.000       1.118       1.348\n",
       "workclass_Without-pay                    0.7297      0.341      2.138      0.033       0.061       1.399\n",
       "marital-status_Divorced                  1.0295      0.059     17.369      0.000       0.913       1.146\n",
       "marital-status_Married-AF-spouse         0.7694      0.294      2.618      0.009       0.193       1.345\n",
       "marital-status_Married-civ-spouse        0.9228      0.091     10.136      0.000       0.744       1.101\n",
       "marital-status_Married-spouse-absent     1.2689      0.097     13.093      0.000       1.079       1.459\n",
       "marital-status_Never-married             1.3121      0.055     23.779      0.000       1.204       1.420\n",
       "marital-status_Separated                 0.8043      0.072     11.222      0.000       0.664       0.945\n",
       "marital-status_Widowed                   0.6429      0.074      8.672      0.000       0.498       0.788\n",
       "occupation_Adm-clerical                  0.6780      0.030     22.690      0.000       0.619       0.737\n",
       "occupation_Craft-repair                 -0.1415      0.029     -4.851      0.000      -0.199      -0.084\n",
       "occupation_Exec-managerial               1.5800      0.030     53.159      0.000       1.522       1.638\n",
       "occupation_Farming-fishing              -0.1339      0.049     -2.719      0.007      -0.230      -0.037\n",
       "occupation_Handlers-cleaners            -0.2353      0.042     -5.612      0.000      -0.317      -0.153\n",
       "occupation_Machine-op-inspct            -0.2842      0.037     -7.635      0.000      -0.357      -0.211\n",
       "occupation_None                          0.3503      0.176      1.994      0.046       0.006       0.695\n",
       "occupation_Prof-specialty                2.9370      0.030     98.860      0.000       2.879       2.995\n",
       "occupation_Protective-serv               0.2988      0.058      5.193      0.000       0.186       0.412\n",
       "occupation_Sales                         0.8673      0.030     29.113      0.000       0.809       0.926\n",
       "occupation_Tech-support                  1.2967      0.047     27.600      0.000       1.205       1.389\n",
       "occupation_Transport-moving             -0.4631      0.040    -11.670      0.000      -0.541      -0.385\n",
       "relationship_Husband                     1.3357      0.087     15.408      0.000       1.166       1.506\n",
       "relationship_Not-in-family               1.4003      0.052     26.710      0.000       1.298       1.503\n",
       "relationship_Other-relative              0.8934      0.064     13.932      0.000       0.768       1.019\n",
       "relationship_Own-child                   0.9453      0.051     18.571      0.000       0.845       1.045\n",
       "relationship_Unmarried                   1.0525      0.056     18.767      0.000       0.943       1.162\n",
       "relationship_Wife                        1.1228      0.091     12.304      0.000       0.944       1.302\n",
       "==============================================================================\n",
       "Omnibus:                      529.927   Durbin-Watson:                   2.016\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              561.059\n",
       "Skew:                           0.289   Prob(JB):                    1.47e-122\n",
       "Kurtosis:                       3.215   Cond. No.                     1.23e+16\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are\n",
       "strong multicollinearity problems or that the design matrix is singular.\n",
       "\"\"\""
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drop_occupation = [f'occupation_{occupation}' for occupation in ['Armed-Forces', 'Other-service', 'Priv-house-serv']]\n",
    "df = df[(df['occupation_Armed-Forces'] == 0) & (df['occupation_Other-service'] == 0) & (df['occupation_Priv-house-serv'] == 0)]\n",
    "\n",
    "X = df.drop(columns = ['educational-num'] + drop_occupation)\n",
    "y = df[['educational-num']]\n",
    "\n",
    "# OLS 검정\n",
    "multi_model = sm.OLS(y.astype(float), X.astype(float))\n",
    "fitted_multi_model = multi_model.fit()\n",
    "fitted_multi_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>educational-num</td> <th>  R-squared:         </th> <td>   0.405</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.404</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   707.7</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Mon, 13 Dec 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:51:43</td>     <th>  Log-Likelihood:    </th> <td> -67025.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td> 35456</td>      <th>  AIC:               </th> <td>1.341e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td> 35421</td>      <th>  BIC:               </th> <td>1.344e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    34</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                    <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>                                  <td>   -0.0019</td> <td>    0.001</td> <td>   -2.375</td> <td> 0.018</td> <td>   -0.004</td> <td>   -0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>fnlwgt</th>                               <td> -1.54e-07</td> <td> 8.17e-08</td> <td>   -1.886</td> <td> 0.059</td> <td>-3.14e-07</td> <td> 6.05e-09</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-gain</th>                         <td> 1.212e-05</td> <td> 1.11e-06</td> <td>   10.907</td> <td> 0.000</td> <td> 9.95e-06</td> <td> 1.43e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-loss</th>                         <td>    0.0001</td> <td> 2.06e-05</td> <td>    6.365</td> <td> 0.000</td> <td> 9.09e-05</td> <td>    0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>hours-per-week</th>                       <td>    0.0093</td> <td>    0.001</td> <td>   12.179</td> <td> 0.000</td> <td>    0.008</td> <td>    0.011</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>income</th>                               <td>    0.9300</td> <td>    0.024</td> <td>   39.564</td> <td> 0.000</td> <td>    0.884</td> <td>    0.976</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Federal-gov</th>                <td>    0.8472</td> <td>    0.061</td> <td>   13.840</td> <td> 0.000</td> <td>    0.727</td> <td>    0.967</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Local-gov</th>                  <td>    1.0786</td> <td>    0.053</td> <td>   20.315</td> <td> 0.000</td> <td>    0.975</td> <td>    1.183</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_None</th>                       <td>    0.5676</td> <td>    0.025</td> <td>   22.286</td> <td> 0.000</td> <td>    0.518</td> <td>    0.617</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Private</th>                    <td>    0.7021</td> <td>    0.044</td> <td>   15.850</td> <td> 0.000</td> <td>    0.615</td> <td>    0.789</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-inc</th>               <td>    0.8540</td> <td>    0.060</td> <td>   14.175</td> <td> 0.000</td> <td>    0.736</td> <td>    0.972</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-not-inc</th>           <td>    0.8666</td> <td>    0.052</td> <td>   16.718</td> <td> 0.000</td> <td>    0.765</td> <td>    0.968</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_State-gov</th>                  <td>    1.2154</td> <td>    0.057</td> <td>   21.199</td> <td> 0.000</td> <td>    1.103</td> <td>    1.328</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Without-pay</th>                <td>    0.7122</td> <td>    0.341</td> <td>    2.090</td> <td> 0.037</td> <td>    0.044</td> <td>    1.380</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Divorced</th>              <td>    1.0433</td> <td>    0.058</td> <td>   17.840</td> <td> 0.000</td> <td>    0.929</td> <td>    1.158</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-AF-spouse</th>     <td>    0.7835</td> <td>    0.294</td> <td>    2.669</td> <td> 0.008</td> <td>    0.208</td> <td>    1.359</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-civ-spouse</th>    <td>    0.9368</td> <td>    0.090</td> <td>   10.352</td> <td> 0.000</td> <td>    0.759</td> <td>    1.114</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-spouse-absent</th> <td>    1.2802</td> <td>    0.097</td> <td>   13.242</td> <td> 0.000</td> <td>    1.091</td> <td>    1.470</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Never-married</th>         <td>    1.3256</td> <td>    0.054</td> <td>   24.407</td> <td> 0.000</td> <td>    1.219</td> <td>    1.432</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Separated</th>             <td>    0.8179</td> <td>    0.071</td> <td>   11.524</td> <td> 0.000</td> <td>    0.679</td> <td>    0.957</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Widowed</th>               <td>    0.6564</td> <td>    0.073</td> <td>    8.934</td> <td> 0.000</td> <td>    0.512</td> <td>    0.800</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Adm-clerical</th>              <td>    0.6668</td> <td>    0.028</td> <td>   23.415</td> <td> 0.000</td> <td>    0.611</td> <td>    0.723</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Craft-repair</th>              <td>   -0.1528</td> <td>    0.028</td> <td>   -5.511</td> <td> 0.000</td> <td>   -0.207</td> <td>   -0.098</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Exec-managerial</th>           <td>    1.5688</td> <td>    0.028</td> <td>   55.438</td> <td> 0.000</td> <td>    1.513</td> <td>    1.624</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Farming-fishing</th>           <td>   -0.1451</td> <td>    0.048</td> <td>   -2.997</td> <td> 0.003</td> <td>   -0.240</td> <td>   -0.050</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Handlers-cleaners</th>         <td>   -0.2465</td> <td>    0.041</td> <td>   -6.021</td> <td> 0.000</td> <td>   -0.327</td> <td>   -0.166</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Machine-op-inspct</th>         <td>   -0.2954</td> <td>    0.036</td> <td>   -8.186</td> <td> 0.000</td> <td>   -0.366</td> <td>   -0.225</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_None</th>                      <td>    0.5676</td> <td>    0.025</td> <td>   22.286</td> <td> 0.000</td> <td>    0.518</td> <td>    0.617</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Prof-specialty</th>            <td>    2.9257</td> <td>    0.028</td> <td>  103.443</td> <td> 0.000</td> <td>    2.870</td> <td>    2.981</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Protective-serv</th>           <td>    0.2875</td> <td>    0.057</td> <td>    5.062</td> <td> 0.000</td> <td>    0.176</td> <td>    0.399</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Sales</th>                     <td>    0.8561</td> <td>    0.028</td> <td>   30.170</td> <td> 0.000</td> <td>    0.800</td> <td>    0.912</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Tech-support</th>              <td>    1.2855</td> <td>    0.046</td> <td>   27.895</td> <td> 0.000</td> <td>    1.195</td> <td>    1.376</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Transport-moving</th>          <td>   -0.4744</td> <td>    0.039</td> <td>  -12.281</td> <td> 0.000</td> <td>   -0.550</td> <td>   -0.399</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Husband</th>                 <td>    1.3513</td> <td>    0.086</td> <td>   15.744</td> <td> 0.000</td> <td>    1.183</td> <td>    1.520</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Not-in-family</th>           <td>    1.4165</td> <td>    0.051</td> <td>   27.831</td> <td> 0.000</td> <td>    1.317</td> <td>    1.516</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Other-relative</th>          <td>    0.9084</td> <td>    0.063</td> <td>   14.418</td> <td> 0.000</td> <td>    0.785</td> <td>    1.032</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Own-child</th>               <td>    0.9612</td> <td>    0.049</td> <td>   19.446</td> <td> 0.000</td> <td>    0.864</td> <td>    1.058</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Unmarried</th>               <td>    1.0684</td> <td>    0.055</td> <td>   19.557</td> <td> 0.000</td> <td>    0.961</td> <td>    1.176</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Wife</th>                    <td>    1.1379</td> <td>    0.090</td> <td>   12.575</td> <td> 0.000</td> <td>    0.961</td> <td>    1.315</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>529.643</td> <th>  Durbin-Watson:     </th> <td>   2.016</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 560.715</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.289</td>  <th>  Prob(JB):          </th> <td>1.75e-122</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.214</td>  <th>  Cond. No.          </th> <td>1.23e+16</td> \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:        educational-num   R-squared:                       0.405\n",
       "Model:                            OLS   Adj. R-squared:                  0.404\n",
       "Method:                 Least Squares   F-statistic:                     707.7\n",
       "Date:                Mon, 13 Dec 2021   Prob (F-statistic):               0.00\n",
       "Time:                        11:51:43   Log-Likelihood:                -67025.\n",
       "No. Observations:               35456   AIC:                         1.341e+05\n",
       "Df Residuals:                   35421   BIC:                         1.344e+05\n",
       "Df Model:                          34                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "========================================================================================================\n",
       "                                           coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------------------\n",
       "age                                     -0.0019      0.001     -2.375      0.018      -0.004      -0.000\n",
       "fnlwgt                                -1.54e-07   8.17e-08     -1.886      0.059   -3.14e-07    6.05e-09\n",
       "capital-gain                          1.212e-05   1.11e-06     10.907      0.000    9.95e-06    1.43e-05\n",
       "capital-loss                             0.0001   2.06e-05      6.365      0.000    9.09e-05       0.000\n",
       "hours-per-week                           0.0093      0.001     12.179      0.000       0.008       0.011\n",
       "income                                   0.9300      0.024     39.564      0.000       0.884       0.976\n",
       "workclass_Federal-gov                    0.8472      0.061     13.840      0.000       0.727       0.967\n",
       "workclass_Local-gov                      1.0786      0.053     20.315      0.000       0.975       1.183\n",
       "workclass_None                           0.5676      0.025     22.286      0.000       0.518       0.617\n",
       "workclass_Private                        0.7021      0.044     15.850      0.000       0.615       0.789\n",
       "workclass_Self-emp-inc                   0.8540      0.060     14.175      0.000       0.736       0.972\n",
       "workclass_Self-emp-not-inc               0.8666      0.052     16.718      0.000       0.765       0.968\n",
       "workclass_State-gov                      1.2154      0.057     21.199      0.000       1.103       1.328\n",
       "workclass_Without-pay                    0.7122      0.341      2.090      0.037       0.044       1.380\n",
       "marital-status_Divorced                  1.0433      0.058     17.840      0.000       0.929       1.158\n",
       "marital-status_Married-AF-spouse         0.7835      0.294      2.669      0.008       0.208       1.359\n",
       "marital-status_Married-civ-spouse        0.9368      0.090     10.352      0.000       0.759       1.114\n",
       "marital-status_Married-spouse-absent     1.2802      0.097     13.242      0.000       1.091       1.470\n",
       "marital-status_Never-married             1.3256      0.054     24.407      0.000       1.219       1.432\n",
       "marital-status_Separated                 0.8179      0.071     11.524      0.000       0.679       0.957\n",
       "marital-status_Widowed                   0.6564      0.073      8.934      0.000       0.512       0.800\n",
       "occupation_Adm-clerical                  0.6668      0.028     23.415      0.000       0.611       0.723\n",
       "occupation_Craft-repair                 -0.1528      0.028     -5.511      0.000      -0.207      -0.098\n",
       "occupation_Exec-managerial               1.5688      0.028     55.438      0.000       1.513       1.624\n",
       "occupation_Farming-fishing              -0.1451      0.048     -2.997      0.003      -0.240      -0.050\n",
       "occupation_Handlers-cleaners            -0.2465      0.041     -6.021      0.000      -0.327      -0.166\n",
       "occupation_Machine-op-inspct            -0.2954      0.036     -8.186      0.000      -0.366      -0.225\n",
       "occupation_None                          0.5676      0.025     22.286      0.000       0.518       0.617\n",
       "occupation_Prof-specialty                2.9257      0.028    103.443      0.000       2.870       2.981\n",
       "occupation_Protective-serv               0.2875      0.057      5.062      0.000       0.176       0.399\n",
       "occupation_Sales                         0.8561      0.028     30.170      0.000       0.800       0.912\n",
       "occupation_Tech-support                  1.2855      0.046     27.895      0.000       1.195       1.376\n",
       "occupation_Transport-moving             -0.4744      0.039    -12.281      0.000      -0.550      -0.399\n",
       "relationship_Husband                     1.3513      0.086     15.744      0.000       1.183       1.520\n",
       "relationship_Not-in-family               1.4165      0.051     27.831      0.000       1.317       1.516\n",
       "relationship_Other-relative              0.9084      0.063     14.418      0.000       0.785       1.032\n",
       "relationship_Own-child                   0.9612      0.049     19.446      0.000       0.864       1.058\n",
       "relationship_Unmarried                   1.0684      0.055     19.557      0.000       0.961       1.176\n",
       "relationship_Wife                        1.1379      0.090     12.575      0.000       0.961       1.315\n",
       "==============================================================================\n",
       "Omnibus:                      529.643   Durbin-Watson:                   2.016\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              560.715\n",
       "Skew:                           0.289   Prob(JB):                    1.75e-122\n",
       "Kurtosis:                       3.214   Cond. No.                     1.23e+16\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The smallest eigenvalue is 1.08e-17. This might indicate that there are\n",
       "strong multicollinearity problems or that the design matrix is singular.\n",
       "\"\"\""
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "drop_workclass = [f'workclass_{status}' for status in ['Never-worked']]\n",
    "df = df[(df['workclass_Never-worked'] == 0)]\n",
    "\n",
    "X = df.drop(columns = ['educational-num'] + drop_occupation + drop_workclass)\n",
    "y = df[['educational-num']]\n",
    "\n",
    "\n",
    "# OLS 검정\n",
    "multi_model = sm.OLS(y.astype(float), X.astype(float))\n",
    "fitted_multi_model = multi_model.fit()\n",
    "fitted_multi_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>educational-num</td> <th>  R-squared:         </th> <td>   0.404</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.404</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   729.0</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Mon, 13 Dec 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:51:43</td>     <th>  Log-Likelihood:    </th> <td> -67027.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td> 35456</td>      <th>  AIC:               </th> <td>1.341e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td> 35422</td>      <th>  BIC:               </th> <td>1.344e+05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>    33</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                    <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age</th>                                  <td>   -0.0018</td> <td>    0.001</td> <td>   -2.286</td> <td> 0.022</td> <td>   -0.003</td> <td>   -0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-gain</th>                         <td> 1.212e-05</td> <td> 1.11e-06</td> <td>   10.899</td> <td> 0.000</td> <td> 9.94e-06</td> <td> 1.43e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>capital-loss</th>                         <td>    0.0001</td> <td> 2.06e-05</td> <td>    6.369</td> <td> 0.000</td> <td>  9.1e-05</td> <td>    0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>hours-per-week</th>                       <td>    0.0094</td> <td>    0.001</td> <td>   12.204</td> <td> 0.000</td> <td>    0.008</td> <td>    0.011</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>income</th>                               <td>    0.9288</td> <td>    0.023</td> <td>   39.527</td> <td> 0.000</td> <td>    0.883</td> <td>    0.975</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Federal-gov</th>                <td>    0.8396</td> <td>    0.061</td> <td>   13.745</td> <td> 0.000</td> <td>    0.720</td> <td>    0.959</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Local-gov</th>                  <td>    1.0703</td> <td>    0.053</td> <td>   20.227</td> <td> 0.000</td> <td>    0.967</td> <td>    1.174</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_None</th>                       <td>    0.5611</td> <td>    0.025</td> <td>   22.234</td> <td> 0.000</td> <td>    0.512</td> <td>    0.611</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Private</th>                    <td>    0.6936</td> <td>    0.044</td> <td>   15.739</td> <td> 0.000</td> <td>    0.607</td> <td>    0.780</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-inc</th>               <td>    0.8467</td> <td>    0.060</td> <td>   14.082</td> <td> 0.000</td> <td>    0.729</td> <td>    0.965</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Self-emp-not-inc</th>           <td>    0.8595</td> <td>    0.052</td> <td>   16.625</td> <td> 0.000</td> <td>    0.758</td> <td>    0.961</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_State-gov</th>                  <td>    1.2082</td> <td>    0.057</td> <td>   21.119</td> <td> 0.000</td> <td>    1.096</td> <td>    1.320</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>workclass_Without-pay</th>                <td>    0.7032</td> <td>    0.341</td> <td>    2.064</td> <td> 0.039</td> <td>    0.035</td> <td>    1.371</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Divorced</th>              <td>    1.0351</td> <td>    0.058</td> <td>   17.748</td> <td> 0.000</td> <td>    0.921</td> <td>    1.149</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-AF-spouse</th>     <td>    0.7745</td> <td>    0.294</td> <td>    2.639</td> <td> 0.008</td> <td>    0.199</td> <td>    1.350</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-civ-spouse</th>    <td>    0.9279</td> <td>    0.090</td> <td>   10.267</td> <td> 0.000</td> <td>    0.751</td> <td>    1.105</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Married-spouse-absent</th> <td>    1.2721</td> <td>    0.097</td> <td>   13.171</td> <td> 0.000</td> <td>    1.083</td> <td>    1.461</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Never-married</th>         <td>    1.3168</td> <td>    0.054</td> <td>   24.334</td> <td> 0.000</td> <td>    1.211</td> <td>    1.423</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Separated</th>             <td>    0.8068</td> <td>    0.071</td> <td>   11.406</td> <td> 0.000</td> <td>    0.668</td> <td>    0.945</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>marital-status_Widowed</th>               <td>    0.6490</td> <td>    0.073</td> <td>    8.845</td> <td> 0.000</td> <td>    0.505</td> <td>    0.793</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Adm-clerical</th>              <td>    0.6613</td> <td>    0.028</td> <td>   23.343</td> <td> 0.000</td> <td>    0.606</td> <td>    0.717</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Craft-repair</th>              <td>   -0.1582</td> <td>    0.028</td> <td>   -5.737</td> <td> 0.000</td> <td>   -0.212</td> <td>   -0.104</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Exec-managerial</th>           <td>    1.5639</td> <td>    0.028</td> <td>   55.494</td> <td> 0.000</td> <td>    1.509</td> <td>    1.619</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Farming-fishing</th>           <td>   -0.1470</td> <td>    0.048</td> <td>   -3.037</td> <td> 0.002</td> <td>   -0.242</td> <td>   -0.052</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Handlers-cleaners</th>         <td>   -0.2518</td> <td>    0.041</td> <td>   -6.167</td> <td> 0.000</td> <td>   -0.332</td> <td>   -0.172</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Machine-op-inspct</th>         <td>   -0.3006</td> <td>    0.036</td> <td>   -8.354</td> <td> 0.000</td> <td>   -0.371</td> <td>   -0.230</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_None</th>                      <td>    0.5611</td> <td>    0.025</td> <td>   22.234</td> <td> 0.000</td> <td>    0.512</td> <td>    0.611</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Prof-specialty</th>            <td>    2.9212</td> <td>    0.028</td> <td>  103.660</td> <td> 0.000</td> <td>    2.866</td> <td>    2.976</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Protective-serv</th>           <td>    0.2806</td> <td>    0.057</td> <td>    4.951</td> <td> 0.000</td> <td>    0.170</td> <td>    0.392</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Sales</th>                     <td>    0.8510</td> <td>    0.028</td> <td>   30.126</td> <td> 0.000</td> <td>    0.796</td> <td>    0.906</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Tech-support</th>              <td>    1.2806</td> <td>    0.046</td> <td>   27.832</td> <td> 0.000</td> <td>    1.190</td> <td>    1.371</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>occupation_Transport-moving</th>          <td>   -0.4798</td> <td>    0.039</td> <td>  -12.458</td> <td> 0.000</td> <td>   -0.555</td> <td>   -0.404</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Husband</th>                 <td>    1.3415</td> <td>    0.086</td> <td>   15.658</td> <td> 0.000</td> <td>    1.174</td> <td>    1.509</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Not-in-family</th>           <td>    1.4057</td> <td>    0.051</td> <td>   27.795</td> <td> 0.000</td> <td>    1.307</td> <td>    1.505</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Other-relative</th>          <td>    0.8977</td> <td>    0.063</td> <td>   14.305</td> <td> 0.000</td> <td>    0.775</td> <td>    1.021</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Own-child</th>               <td>    0.9510</td> <td>    0.049</td> <td>   19.355</td> <td> 0.000</td> <td>    0.855</td> <td>    1.047</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Unmarried</th>               <td>    1.0578</td> <td>    0.054</td> <td>   19.465</td> <td> 0.000</td> <td>    0.951</td> <td>    1.164</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>relationship_Wife</th>                    <td>    1.1285</td> <td>    0.090</td> <td>   12.490</td> <td> 0.000</td> <td>    0.951</td> <td>    1.306</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>529.426</td> <th>  Durbin-Watson:     </th> <td>   2.015</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 560.492</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.289</td>  <th>  Prob(JB):          </th> <td>1.95e-122</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.215</td>  <th>  Cond. No.          </th> <td>1.02e+16</td> \n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The smallest eigenvalue is 2.2e-20. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:        educational-num   R-squared:                       0.404\n",
       "Model:                            OLS   Adj. R-squared:                  0.404\n",
       "Method:                 Least Squares   F-statistic:                     729.0\n",
       "Date:                Mon, 13 Dec 2021   Prob (F-statistic):               0.00\n",
       "Time:                        11:51:43   Log-Likelihood:                -67027.\n",
       "No. Observations:               35456   AIC:                         1.341e+05\n",
       "Df Residuals:                   35422   BIC:                         1.344e+05\n",
       "Df Model:                          33                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "========================================================================================================\n",
       "                                           coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------------------\n",
       "age                                     -0.0018      0.001     -2.286      0.022      -0.003      -0.000\n",
       "capital-gain                          1.212e-05   1.11e-06     10.899      0.000    9.94e-06    1.43e-05\n",
       "capital-loss                             0.0001   2.06e-05      6.369      0.000     9.1e-05       0.000\n",
       "hours-per-week                           0.0094      0.001     12.204      0.000       0.008       0.011\n",
       "income                                   0.9288      0.023     39.527      0.000       0.883       0.975\n",
       "workclass_Federal-gov                    0.8396      0.061     13.745      0.000       0.720       0.959\n",
       "workclass_Local-gov                      1.0703      0.053     20.227      0.000       0.967       1.174\n",
       "workclass_None                           0.5611      0.025     22.234      0.000       0.512       0.611\n",
       "workclass_Private                        0.6936      0.044     15.739      0.000       0.607       0.780\n",
       "workclass_Self-emp-inc                   0.8467      0.060     14.082      0.000       0.729       0.965\n",
       "workclass_Self-emp-not-inc               0.8595      0.052     16.625      0.000       0.758       0.961\n",
       "workclass_State-gov                      1.2082      0.057     21.119      0.000       1.096       1.320\n",
       "workclass_Without-pay                    0.7032      0.341      2.064      0.039       0.035       1.371\n",
       "marital-status_Divorced                  1.0351      0.058     17.748      0.000       0.921       1.149\n",
       "marital-status_Married-AF-spouse         0.7745      0.294      2.639      0.008       0.199       1.350\n",
       "marital-status_Married-civ-spouse        0.9279      0.090     10.267      0.000       0.751       1.105\n",
       "marital-status_Married-spouse-absent     1.2721      0.097     13.171      0.000       1.083       1.461\n",
       "marital-status_Never-married             1.3168      0.054     24.334      0.000       1.211       1.423\n",
       "marital-status_Separated                 0.8068      0.071     11.406      0.000       0.668       0.945\n",
       "marital-status_Widowed                   0.6490      0.073      8.845      0.000       0.505       0.793\n",
       "occupation_Adm-clerical                  0.6613      0.028     23.343      0.000       0.606       0.717\n",
       "occupation_Craft-repair                 -0.1582      0.028     -5.737      0.000      -0.212      -0.104\n",
       "occupation_Exec-managerial               1.5639      0.028     55.494      0.000       1.509       1.619\n",
       "occupation_Farming-fishing              -0.1470      0.048     -3.037      0.002      -0.242      -0.052\n",
       "occupation_Handlers-cleaners            -0.2518      0.041     -6.167      0.000      -0.332      -0.172\n",
       "occupation_Machine-op-inspct            -0.3006      0.036     -8.354      0.000      -0.371      -0.230\n",
       "occupation_None                          0.5611      0.025     22.234      0.000       0.512       0.611\n",
       "occupation_Prof-specialty                2.9212      0.028    103.660      0.000       2.866       2.976\n",
       "occupation_Protective-serv               0.2806      0.057      4.951      0.000       0.170       0.392\n",
       "occupation_Sales                         0.8510      0.028     30.126      0.000       0.796       0.906\n",
       "occupation_Tech-support                  1.2806      0.046     27.832      0.000       1.190       1.371\n",
       "occupation_Transport-moving             -0.4798      0.039    -12.458      0.000      -0.555      -0.404\n",
       "relationship_Husband                     1.3415      0.086     15.658      0.000       1.174       1.509\n",
       "relationship_Not-in-family               1.4057      0.051     27.795      0.000       1.307       1.505\n",
       "relationship_Other-relative              0.8977      0.063     14.305      0.000       0.775       1.021\n",
       "relationship_Own-child                   0.9510      0.049     19.355      0.000       0.855       1.047\n",
       "relationship_Unmarried                   1.0578      0.054     19.465      0.000       0.951       1.164\n",
       "relationship_Wife                        1.1285      0.090     12.490      0.000       0.951       1.306\n",
       "==============================================================================\n",
       "Omnibus:                      529.426   Durbin-Watson:                   2.015\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              560.492\n",
       "Skew:                           0.289   Prob(JB):                    1.95e-122\n",
       "Kurtosis:                       3.215   Cond. No.                     1.02e+16\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The smallest eigenvalue is 2.2e-20. This might indicate that there are\n",
       "strong multicollinearity problems or that the design matrix is singular.\n",
       "\"\"\""
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns = ['educational-num', 'fnlwgt'] + drop_occupation + drop_workclass)\n",
    "y = df[['educational-num']]\n",
    "\n",
    "\n",
    "# OLS 검정\n",
    "multi_model = sm.OLS(y.astype(float), X.astype(float))\n",
    "fitted_multi_model = multi_model.fit()\n",
    "fitted_multi_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "X.reset_index(drop=True, inplace=True)\n",
    "y.reset_index(drop=True, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### pycaret"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>age</th>\n",
       "      <th>capital-gain</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "      <th>income</th>\n",
       "      <th>workclass_Federal-gov</th>\n",
       "      <th>workclass_Local-gov</th>\n",
       "      <th>workclass_None</th>\n",
       "      <th>workclass_Private</th>\n",
       "      <th>workclass_Self-emp-inc</th>\n",
       "      <th>workclass_Self-emp-not-inc</th>\n",
       "      <th>workclass_State-gov</th>\n",
       "      <th>workclass_Without-pay</th>\n",
       "      <th>marital-status_Divorced</th>\n",
       "      <th>marital-status_Married-AF-spouse</th>\n",
       "      <th>marital-status_Married-civ-spouse</th>\n",
       "      <th>marital-status_Married-spouse-absent</th>\n",
       "      <th>marital-status_Never-married</th>\n",
       "      <th>marital-status_Separated</th>\n",
       "      <th>marital-status_Widowed</th>\n",
       "      <th>occupation_Adm-clerical</th>\n",
       "      <th>occupation_Craft-repair</th>\n",
       "      <th>occupation_Exec-managerial</th>\n",
       "      <th>occupation_Farming-fishing</th>\n",
       "      <th>occupation_Handlers-cleaners</th>\n",
       "      <th>occupation_Machine-op-inspct</th>\n",
       "      <th>occupation_None</th>\n",
       "      <th>occupation_Prof-specialty</th>\n",
       "      <th>occupation_Protective-serv</th>\n",
       "      <th>occupation_Sales</th>\n",
       "      <th>occupation_Tech-support</th>\n",
       "      <th>occupation_Transport-moving</th>\n",
       "      <th>relationship_Husband</th>\n",
       "      <th>relationship_Not-in-family</th>\n",
       "      <th>relationship_Other-relative</th>\n",
       "      <th>relationship_Own-child</th>\n",
       "      <th>relationship_Unmarried</th>\n",
       "      <th>relationship_Wife</th>\n",
       "      <th>educational-num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2330</th>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42464</th>\n",
       "      <td>44</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10669</th>\n",
       "      <td>44</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>48</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9194</th>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8084</th>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "      <td>1579</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7333</th>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43644</th>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9906</th>\n",
       "      <td>78</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10331</th>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38402</th>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>55</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>35101 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       age  capital-gain  capital-loss  hours-per-week  income  \\\n",
       "2330    37             0             0              40       1   \n",
       "42464   44             0             0              40       1   \n",
       "10669   44             0             0              48       0   \n",
       "9194    45             0             0              25       0   \n",
       "8084    59             0          1579              60       0   \n",
       "...    ...           ...           ...             ...     ...   \n",
       "7333    29             0             0              40       0   \n",
       "43644   33             0             0              35       0   \n",
       "9906    78             0             0              36       0   \n",
       "10331   20             0             0              10       0   \n",
       "38402   37             0             0              55       1   \n",
       "\n",
       "       workclass_Federal-gov  workclass_Local-gov  workclass_None  \\\n",
       "2330                       0                    0               0   \n",
       "42464                      0                    0               0   \n",
       "10669                      1                    0               0   \n",
       "9194                       0                    0               0   \n",
       "8084                       0                    0               0   \n",
       "...                      ...                  ...             ...   \n",
       "7333                       0                    0               1   \n",
       "43644                      0                    0               0   \n",
       "9906                       0                    0               0   \n",
       "10331                      0                    0               1   \n",
       "38402                      0                    0               0   \n",
       "\n",
       "       workclass_Private  workclass_Self-emp-inc  workclass_Self-emp-not-inc  \\\n",
       "2330                   1                       0                           0   \n",
       "42464                  1                       0                           0   \n",
       "10669                  0                       0                           0   \n",
       "9194                   1                       0                           0   \n",
       "8084                   0                       0                           1   \n",
       "...                  ...                     ...                         ...   \n",
       "7333                   0                       0                           0   \n",
       "43644                  0                       0                           1   \n",
       "9906                   1                       0                           0   \n",
       "10331                  0                       0                           0   \n",
       "38402                  0                       0                           1   \n",
       "\n",
       "       workclass_State-gov  workclass_Without-pay  marital-status_Divorced  \\\n",
       "2330                     0                      0                        0   \n",
       "42464                    0                      0                        0   \n",
       "10669                    0                      0                        0   \n",
       "9194                     0                      0                        0   \n",
       "8084                     0                      0                        0   \n",
       "...                    ...                    ...                      ...   \n",
       "7333                     0                      0                        0   \n",
       "43644                    0                      0                        0   \n",
       "9906                     0                      0                        0   \n",
       "10331                    0                      0                        0   \n",
       "38402                    0                      0                        0   \n",
       "\n",
       "       marital-status_Married-AF-spouse  marital-status_Married-civ-spouse  \\\n",
       "2330                                  0                                  1   \n",
       "42464                                 0                                  1   \n",
       "10669                                 0                                  1   \n",
       "9194                                  0                                  0   \n",
       "8084                                  0                                  1   \n",
       "...                                 ...                                ...   \n",
       "7333                                  0                                  0   \n",
       "43644                                 0                                  1   \n",
       "9906                                  0                                  1   \n",
       "10331                                 0                                  0   \n",
       "38402                                 0                                  0   \n",
       "\n",
       "       marital-status_Married-spouse-absent  marital-status_Never-married  \\\n",
       "2330                                      0                             0   \n",
       "42464                                     0                             0   \n",
       "10669                                     0                             0   \n",
       "9194                                      0                             0   \n",
       "8084                                      0                             0   \n",
       "...                                     ...                           ...   \n",
       "7333                                      0                             1   \n",
       "43644                                     0                             0   \n",
       "9906                                      0                             0   \n",
       "10331                                     0                             1   \n",
       "38402                                     0                             1   \n",
       "\n",
       "       marital-status_Separated  marital-status_Widowed  \\\n",
       "2330                          0                       0   \n",
       "42464                         0                       0   \n",
       "10669                         0                       0   \n",
       "9194                          1                       0   \n",
       "8084                          0                       0   \n",
       "...                         ...                     ...   \n",
       "7333                          0                       0   \n",
       "43644                         0                       0   \n",
       "9906                          0                       0   \n",
       "10331                         0                       0   \n",
       "38402                         0                       0   \n",
       "\n",
       "       occupation_Adm-clerical  occupation_Craft-repair  \\\n",
       "2330                         0                        0   \n",
       "42464                        0                        0   \n",
       "10669                        0                        0   \n",
       "9194                         1                        0   \n",
       "8084                         0                        0   \n",
       "...                        ...                      ...   \n",
       "7333                         0                        0   \n",
       "43644                        0                        1   \n",
       "9906                         0                        0   \n",
       "10331                        0                        0   \n",
       "38402                        0                        0   \n",
       "\n",
       "       occupation_Exec-managerial  occupation_Farming-fishing  \\\n",
       "2330                            1                           0   \n",
       "42464                           0                           0   \n",
       "10669                           1                           0   \n",
       "9194                            0                           0   \n",
       "8084                            0                           1   \n",
       "...                           ...                         ...   \n",
       "7333                            0                           0   \n",
       "43644                           0                           0   \n",
       "9906                            0                           0   \n",
       "10331                           0                           0   \n",
       "38402                           0                           0   \n",
       "\n",
       "       occupation_Handlers-cleaners  occupation_Machine-op-inspct  \\\n",
       "2330                              0                             0   \n",
       "42464                             0                             0   \n",
       "10669                             0                             0   \n",
       "9194                              0                             0   \n",
       "8084                              0                             0   \n",
       "...                             ...                           ...   \n",
       "7333                              0                             0   \n",
       "43644                             0                             0   \n",
       "9906                              0                             0   \n",
       "10331                             0                             0   \n",
       "38402                             0                             0   \n",
       "\n",
       "       occupation_None  occupation_Prof-specialty  occupation_Protective-serv  \\\n",
       "2330                 0                          0                           0   \n",
       "42464                0                          1                           0   \n",
       "10669                0                          0                           0   \n",
       "9194                 0                          0                           0   \n",
       "8084                 0                          0                           0   \n",
       "...                ...                        ...                         ...   \n",
       "7333                 1                          0                           0   \n",
       "43644                0                          0                           0   \n",
       "9906                 0                          0                           0   \n",
       "10331                1                          0                           0   \n",
       "38402                0                          0                           1   \n",
       "\n",
       "       occupation_Sales  occupation_Tech-support  occupation_Transport-moving  \\\n",
       "2330                  0                        0                            0   \n",
       "42464                 0                        0                            0   \n",
       "10669                 0                        0                            0   \n",
       "9194                  0                        0                            0   \n",
       "8084                  0                        0                            0   \n",
       "...                 ...                      ...                          ...   \n",
       "7333                  0                        0                            0   \n",
       "43644                 0                        0                            0   \n",
       "9906                  0                        0                            1   \n",
       "10331                 0                        0                            0   \n",
       "38402                 0                        0                            0   \n",
       "\n",
       "       relationship_Husband  relationship_Not-in-family  \\\n",
       "2330                      1                           0   \n",
       "42464                     1                           0   \n",
       "10669                     1                           0   \n",
       "9194                      0                           0   \n",
       "8084                      1                           0   \n",
       "...                     ...                         ...   \n",
       "7333                      0                           0   \n",
       "43644                     1                           0   \n",
       "9906                      1                           0   \n",
       "10331                     0                           0   \n",
       "38402                     0                           1   \n",
       "\n",
       "       relationship_Other-relative  relationship_Own-child  \\\n",
       "2330                             0                       0   \n",
       "42464                            0                       0   \n",
       "10669                            0                       0   \n",
       "9194                             0                       0   \n",
       "8084                             0                       0   \n",
       "...                            ...                     ...   \n",
       "7333                             0                       1   \n",
       "43644                            0                       0   \n",
       "9906                             0                       0   \n",
       "10331                            0                       1   \n",
       "38402                            0                       0   \n",
       "\n",
       "       relationship_Unmarried  relationship_Wife  educational-num  \n",
       "2330                        0                  0                3  \n",
       "42464                       0                  0               10  \n",
       "10669                       0                  0                7  \n",
       "9194                        1                  0                2  \n",
       "8084                        0                  0                4  \n",
       "...                       ...                ...              ...  \n",
       "7333                        0                  0                2  \n",
       "43644                       0                  0                7  \n",
       "9906                        0                  0                0  \n",
       "10331                       0                  0                4  \n",
       "38402                       0                  0                7  \n",
       "\n",
       "[35101 rows x 39 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = pd.concat([X_train, y_train], axis = 1)\n",
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_fc1c2_row18_col1 {\n",
       "  background-color: lightgreen;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_fc1c2_\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th class=\"col_heading level0 col0\" >Description</th>\n",
       "      <th class=\"col_heading level0 col1\" >Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_fc1c2_row0_col0\" class=\"data row0 col0\" >session_id</td>\n",
       "      <td id=\"T_fc1c2_row0_col1\" class=\"data row0 col1\" >123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_fc1c2_row1_col0\" class=\"data row1 col0\" >Target</td>\n",
       "      <td id=\"T_fc1c2_row1_col1\" class=\"data row1 col1\" >educational-num</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_fc1c2_row2_col0\" class=\"data row2 col0\" >Target Type</td>\n",
       "      <td id=\"T_fc1c2_row2_col1\" class=\"data row2 col1\" >Multiclass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_fc1c2_row3_col0\" class=\"data row3 col0\" >Label Encoded</td>\n",
       "      <td id=\"T_fc1c2_row3_col1\" class=\"data row3 col1\" >0: 0, 1: 1, 10: 2, 2: 3, 3: 4, 4: 5, 5: 6, 6: 7, 7: 8, 8: 9, 9: 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_fc1c2_row4_col0\" class=\"data row4 col0\" >Original Data</td>\n",
       "      <td id=\"T_fc1c2_row4_col1\" class=\"data row4 col1\" >(35101, 39)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_fc1c2_row5_col0\" class=\"data row5 col0\" >Missing Values</td>\n",
       "      <td id=\"T_fc1c2_row5_col1\" class=\"data row5 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_fc1c2_row6_col0\" class=\"data row6 col0\" >Numeric Features</td>\n",
       "      <td id=\"T_fc1c2_row6_col1\" class=\"data row6 col1\" >37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_fc1c2_row7_col0\" class=\"data row7 col0\" >Categorical Features</td>\n",
       "      <td id=\"T_fc1c2_row7_col1\" class=\"data row7 col1\" >1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_fc1c2_row8_col0\" class=\"data row8 col0\" >Ordinal Features</td>\n",
       "      <td id=\"T_fc1c2_row8_col1\" class=\"data row8 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_fc1c2_row9_col0\" class=\"data row9 col0\" >High Cardinality Features</td>\n",
       "      <td id=\"T_fc1c2_row9_col1\" class=\"data row9 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_fc1c2_row10_col0\" class=\"data row10 col0\" >High Cardinality Method</td>\n",
       "      <td id=\"T_fc1c2_row10_col1\" class=\"data row10 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_fc1c2_row11_col0\" class=\"data row11 col0\" >Transformed Train Set</td>\n",
       "      <td id=\"T_fc1c2_row11_col1\" class=\"data row11 col1\" >(24570, 37)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_fc1c2_row12_col0\" class=\"data row12 col0\" >Transformed Test Set</td>\n",
       "      <td id=\"T_fc1c2_row12_col1\" class=\"data row12 col1\" >(10531, 37)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_fc1c2_row13_col0\" class=\"data row13 col0\" >Shuffle Train-Test</td>\n",
       "      <td id=\"T_fc1c2_row13_col1\" class=\"data row13 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_fc1c2_row14_col0\" class=\"data row14 col0\" >Stratify Train-Test</td>\n",
       "      <td id=\"T_fc1c2_row14_col1\" class=\"data row14 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_fc1c2_row15_col0\" class=\"data row15 col0\" >Fold Generator</td>\n",
       "      <td id=\"T_fc1c2_row15_col1\" class=\"data row15 col1\" >StratifiedKFold</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_fc1c2_row16_col0\" class=\"data row16 col0\" >Fold Number</td>\n",
       "      <td id=\"T_fc1c2_row16_col1\" class=\"data row16 col1\" >10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_fc1c2_row17_col0\" class=\"data row17 col0\" >CPU Jobs</td>\n",
       "      <td id=\"T_fc1c2_row17_col1\" class=\"data row17 col1\" >-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "      <td id=\"T_fc1c2_row18_col0\" class=\"data row18 col0\" >Use GPU</td>\n",
       "      <td id=\"T_fc1c2_row18_col1\" class=\"data row18 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "      <td id=\"T_fc1c2_row19_col0\" class=\"data row19 col0\" >Log Experiment</td>\n",
       "      <td id=\"T_fc1c2_row19_col1\" class=\"data row19 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "      <td id=\"T_fc1c2_row20_col0\" class=\"data row20 col0\" >Experiment Name</td>\n",
       "      <td id=\"T_fc1c2_row20_col1\" class=\"data row20 col1\" >clf-default-name</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "      <td id=\"T_fc1c2_row21_col0\" class=\"data row21 col0\" >USI</td>\n",
       "      <td id=\"T_fc1c2_row21_col1\" class=\"data row21 col1\" >cf93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "      <td id=\"T_fc1c2_row22_col0\" class=\"data row22 col0\" >Imputation Type</td>\n",
       "      <td id=\"T_fc1c2_row22_col1\" class=\"data row22 col1\" >simple</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row23\" class=\"row_heading level0 row23\" >23</th>\n",
       "      <td id=\"T_fc1c2_row23_col0\" class=\"data row23 col0\" >Iterative Imputation Iteration</td>\n",
       "      <td id=\"T_fc1c2_row23_col1\" class=\"data row23 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row24\" class=\"row_heading level0 row24\" >24</th>\n",
       "      <td id=\"T_fc1c2_row24_col0\" class=\"data row24 col0\" >Numeric Imputer</td>\n",
       "      <td id=\"T_fc1c2_row24_col1\" class=\"data row24 col1\" >mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row25\" class=\"row_heading level0 row25\" >25</th>\n",
       "      <td id=\"T_fc1c2_row25_col0\" class=\"data row25 col0\" >Iterative Imputation Numeric Model</td>\n",
       "      <td id=\"T_fc1c2_row25_col1\" class=\"data row25 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row26\" class=\"row_heading level0 row26\" >26</th>\n",
       "      <td id=\"T_fc1c2_row26_col0\" class=\"data row26 col0\" >Categorical Imputer</td>\n",
       "      <td id=\"T_fc1c2_row26_col1\" class=\"data row26 col1\" >constant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row27\" class=\"row_heading level0 row27\" >27</th>\n",
       "      <td id=\"T_fc1c2_row27_col0\" class=\"data row27 col0\" >Iterative Imputation Categorical Model</td>\n",
       "      <td id=\"T_fc1c2_row27_col1\" class=\"data row27 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row28\" class=\"row_heading level0 row28\" >28</th>\n",
       "      <td id=\"T_fc1c2_row28_col0\" class=\"data row28 col0\" >Unknown Categoricals Handling</td>\n",
       "      <td id=\"T_fc1c2_row28_col1\" class=\"data row28 col1\" >least_frequent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row29\" class=\"row_heading level0 row29\" >29</th>\n",
       "      <td id=\"T_fc1c2_row29_col0\" class=\"data row29 col0\" >Normalize</td>\n",
       "      <td id=\"T_fc1c2_row29_col1\" class=\"data row29 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row30\" class=\"row_heading level0 row30\" >30</th>\n",
       "      <td id=\"T_fc1c2_row30_col0\" class=\"data row30 col0\" >Normalize Method</td>\n",
       "      <td id=\"T_fc1c2_row30_col1\" class=\"data row30 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row31\" class=\"row_heading level0 row31\" >31</th>\n",
       "      <td id=\"T_fc1c2_row31_col0\" class=\"data row31 col0\" >Transformation</td>\n",
       "      <td id=\"T_fc1c2_row31_col1\" class=\"data row31 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row32\" class=\"row_heading level0 row32\" >32</th>\n",
       "      <td id=\"T_fc1c2_row32_col0\" class=\"data row32 col0\" >Transformation Method</td>\n",
       "      <td id=\"T_fc1c2_row32_col1\" class=\"data row32 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row33\" class=\"row_heading level0 row33\" >33</th>\n",
       "      <td id=\"T_fc1c2_row33_col0\" class=\"data row33 col0\" >PCA</td>\n",
       "      <td id=\"T_fc1c2_row33_col1\" class=\"data row33 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row34\" class=\"row_heading level0 row34\" >34</th>\n",
       "      <td id=\"T_fc1c2_row34_col0\" class=\"data row34 col0\" >PCA Method</td>\n",
       "      <td id=\"T_fc1c2_row34_col1\" class=\"data row34 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row35\" class=\"row_heading level0 row35\" >35</th>\n",
       "      <td id=\"T_fc1c2_row35_col0\" class=\"data row35 col0\" >PCA Components</td>\n",
       "      <td id=\"T_fc1c2_row35_col1\" class=\"data row35 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row36\" class=\"row_heading level0 row36\" >36</th>\n",
       "      <td id=\"T_fc1c2_row36_col0\" class=\"data row36 col0\" >Ignore Low Variance</td>\n",
       "      <td id=\"T_fc1c2_row36_col1\" class=\"data row36 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row37\" class=\"row_heading level0 row37\" >37</th>\n",
       "      <td id=\"T_fc1c2_row37_col0\" class=\"data row37 col0\" >Combine Rare Levels</td>\n",
       "      <td id=\"T_fc1c2_row37_col1\" class=\"data row37 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row38\" class=\"row_heading level0 row38\" >38</th>\n",
       "      <td id=\"T_fc1c2_row38_col0\" class=\"data row38 col0\" >Rare Level Threshold</td>\n",
       "      <td id=\"T_fc1c2_row38_col1\" class=\"data row38 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row39\" class=\"row_heading level0 row39\" >39</th>\n",
       "      <td id=\"T_fc1c2_row39_col0\" class=\"data row39 col0\" >Numeric Binning</td>\n",
       "      <td id=\"T_fc1c2_row39_col1\" class=\"data row39 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row40\" class=\"row_heading level0 row40\" >40</th>\n",
       "      <td id=\"T_fc1c2_row40_col0\" class=\"data row40 col0\" >Remove Outliers</td>\n",
       "      <td id=\"T_fc1c2_row40_col1\" class=\"data row40 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row41\" class=\"row_heading level0 row41\" >41</th>\n",
       "      <td id=\"T_fc1c2_row41_col0\" class=\"data row41 col0\" >Outliers Threshold</td>\n",
       "      <td id=\"T_fc1c2_row41_col1\" class=\"data row41 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row42\" class=\"row_heading level0 row42\" >42</th>\n",
       "      <td id=\"T_fc1c2_row42_col0\" class=\"data row42 col0\" >Remove Multicollinearity</td>\n",
       "      <td id=\"T_fc1c2_row42_col1\" class=\"data row42 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row43\" class=\"row_heading level0 row43\" >43</th>\n",
       "      <td id=\"T_fc1c2_row43_col0\" class=\"data row43 col0\" >Multicollinearity Threshold</td>\n",
       "      <td id=\"T_fc1c2_row43_col1\" class=\"data row43 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row44\" class=\"row_heading level0 row44\" >44</th>\n",
       "      <td id=\"T_fc1c2_row44_col0\" class=\"data row44 col0\" >Clustering</td>\n",
       "      <td id=\"T_fc1c2_row44_col1\" class=\"data row44 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row45\" class=\"row_heading level0 row45\" >45</th>\n",
       "      <td id=\"T_fc1c2_row45_col0\" class=\"data row45 col0\" >Clustering Iteration</td>\n",
       "      <td id=\"T_fc1c2_row45_col1\" class=\"data row45 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row46\" class=\"row_heading level0 row46\" >46</th>\n",
       "      <td id=\"T_fc1c2_row46_col0\" class=\"data row46 col0\" >Polynomial Features</td>\n",
       "      <td id=\"T_fc1c2_row46_col1\" class=\"data row46 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row47\" class=\"row_heading level0 row47\" >47</th>\n",
       "      <td id=\"T_fc1c2_row47_col0\" class=\"data row47 col0\" >Polynomial Degree</td>\n",
       "      <td id=\"T_fc1c2_row47_col1\" class=\"data row47 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row48\" class=\"row_heading level0 row48\" >48</th>\n",
       "      <td id=\"T_fc1c2_row48_col0\" class=\"data row48 col0\" >Trignometry Features</td>\n",
       "      <td id=\"T_fc1c2_row48_col1\" class=\"data row48 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row49\" class=\"row_heading level0 row49\" >49</th>\n",
       "      <td id=\"T_fc1c2_row49_col0\" class=\"data row49 col0\" >Polynomial Threshold</td>\n",
       "      <td id=\"T_fc1c2_row49_col1\" class=\"data row49 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row50\" class=\"row_heading level0 row50\" >50</th>\n",
       "      <td id=\"T_fc1c2_row50_col0\" class=\"data row50 col0\" >Group Features</td>\n",
       "      <td id=\"T_fc1c2_row50_col1\" class=\"data row50 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row51\" class=\"row_heading level0 row51\" >51</th>\n",
       "      <td id=\"T_fc1c2_row51_col0\" class=\"data row51 col0\" >Feature Selection</td>\n",
       "      <td id=\"T_fc1c2_row51_col1\" class=\"data row51 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row52\" class=\"row_heading level0 row52\" >52</th>\n",
       "      <td id=\"T_fc1c2_row52_col0\" class=\"data row52 col0\" >Features Selection Threshold</td>\n",
       "      <td id=\"T_fc1c2_row52_col1\" class=\"data row52 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row53\" class=\"row_heading level0 row53\" >53</th>\n",
       "      <td id=\"T_fc1c2_row53_col0\" class=\"data row53 col0\" >Feature Interaction</td>\n",
       "      <td id=\"T_fc1c2_row53_col1\" class=\"data row53 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row54\" class=\"row_heading level0 row54\" >54</th>\n",
       "      <td id=\"T_fc1c2_row54_col0\" class=\"data row54 col0\" >Feature Ratio</td>\n",
       "      <td id=\"T_fc1c2_row54_col1\" class=\"data row54 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row55\" class=\"row_heading level0 row55\" >55</th>\n",
       "      <td id=\"T_fc1c2_row55_col0\" class=\"data row55 col0\" >Interaction Threshold</td>\n",
       "      <td id=\"T_fc1c2_row55_col1\" class=\"data row55 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row56\" class=\"row_heading level0 row56\" >56</th>\n",
       "      <td id=\"T_fc1c2_row56_col0\" class=\"data row56 col0\" >Fix Imbalance</td>\n",
       "      <td id=\"T_fc1c2_row56_col1\" class=\"data row56 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_fc1c2_level0_row57\" class=\"row_heading level0 row57\" >57</th>\n",
       "      <td id=\"T_fc1c2_row57_col0\" class=\"data row57 col0\" >Fix Imbalance Method</td>\n",
       "      <td id=\"T_fc1c2_row57_col1\" class=\"data row57 col1\" >SMOTE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x22d4b2c25b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 환경 설정\n",
    "clf = setup(data = train, target = 'educational-num', session_id = 123, silent = True, use_gpu = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_56707_ th {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_56707_row0_col0, #T_56707_row0_col1, #T_56707_row0_col2, #T_56707_row0_col3, #T_56707_row0_col4, #T_56707_row0_col7, #T_56707_row1_col0, #T_56707_row1_col1, #T_56707_row1_col2, #T_56707_row1_col3, #T_56707_row1_col4, #T_56707_row1_col5, #T_56707_row1_col6, #T_56707_row1_col7, #T_56707_row2_col0, #T_56707_row2_col3, #T_56707_row2_col5, #T_56707_row2_col6, #T_56707_row3_col0, #T_56707_row3_col1, #T_56707_row3_col2, #T_56707_row3_col3, #T_56707_row3_col4, #T_56707_row3_col5, #T_56707_row3_col6, #T_56707_row3_col7, #T_56707_row4_col0, #T_56707_row4_col1, #T_56707_row4_col2, #T_56707_row4_col3, #T_56707_row4_col4, #T_56707_row4_col5, #T_56707_row4_col6, #T_56707_row4_col7, #T_56707_row5_col0, #T_56707_row5_col1, #T_56707_row5_col2, #T_56707_row5_col3, #T_56707_row5_col4, #T_56707_row5_col5, #T_56707_row5_col6, #T_56707_row5_col7, #T_56707_row6_col0, #T_56707_row6_col1, #T_56707_row6_col2, #T_56707_row6_col3, #T_56707_row6_col4, #T_56707_row6_col5, #T_56707_row6_col6, #T_56707_row6_col7, #T_56707_row7_col0, #T_56707_row7_col1, #T_56707_row7_col2, #T_56707_row7_col3, #T_56707_row7_col4, #T_56707_row7_col5, #T_56707_row7_col6, #T_56707_row7_col7, #T_56707_row8_col0, #T_56707_row8_col1, #T_56707_row8_col2, #T_56707_row8_col3, #T_56707_row8_col4, #T_56707_row8_col5, #T_56707_row8_col6, #T_56707_row8_col7, #T_56707_row9_col0, #T_56707_row9_col1, #T_56707_row9_col2, #T_56707_row9_col3, #T_56707_row9_col4, #T_56707_row9_col5, #T_56707_row9_col6, #T_56707_row9_col7, #T_56707_row10_col0, #T_56707_row10_col1, #T_56707_row10_col2, #T_56707_row10_col3, #T_56707_row10_col4, #T_56707_row10_col5, #T_56707_row10_col6, #T_56707_row10_col7, #T_56707_row11_col0, #T_56707_row11_col1, #T_56707_row11_col2, #T_56707_row11_col4, #T_56707_row11_col5, #T_56707_row11_col6, #T_56707_row11_col7, #T_56707_row12_col0, #T_56707_row12_col1, #T_56707_row12_col2, #T_56707_row12_col3, #T_56707_row12_col4, #T_56707_row12_col5, #T_56707_row12_col6, #T_56707_row12_col7, #T_56707_row13_col0, #T_56707_row13_col1, #T_56707_row13_col2, #T_56707_row13_col3, #T_56707_row13_col4, #T_56707_row13_col5, #T_56707_row13_col6, #T_56707_row13_col7, #T_56707_row14_col0, #T_56707_row14_col1, #T_56707_row14_col2, #T_56707_row14_col3, #T_56707_row14_col4, #T_56707_row14_col5, #T_56707_row14_col6, #T_56707_row14_col7 {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_56707_row0_col5, #T_56707_row0_col6, #T_56707_row2_col1, #T_56707_row2_col2, #T_56707_row2_col4, #T_56707_row2_col7, #T_56707_row11_col3 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "}\n",
       "#T_56707_row0_col8, #T_56707_row1_col8, #T_56707_row2_col8, #T_56707_row3_col8, #T_56707_row4_col8, #T_56707_row5_col8, #T_56707_row6_col8, #T_56707_row8_col8, #T_56707_row9_col8, #T_56707_row10_col8, #T_56707_row11_col8, #T_56707_row12_col8, #T_56707_row13_col8, #T_56707_row14_col8 {\n",
       "  text-align: left;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "#T_56707_row7_col8 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_56707_\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th class=\"col_heading level0 col0\" >Model</th>\n",
       "      <th class=\"col_heading level0 col1\" >Accuracy</th>\n",
       "      <th class=\"col_heading level0 col2\" >AUC</th>\n",
       "      <th class=\"col_heading level0 col3\" >Recall</th>\n",
       "      <th class=\"col_heading level0 col4\" >Prec.</th>\n",
       "      <th class=\"col_heading level0 col5\" >F1</th>\n",
       "      <th class=\"col_heading level0 col6\" >Kappa</th>\n",
       "      <th class=\"col_heading level0 col7\" >MCC</th>\n",
       "      <th class=\"col_heading level0 col8\" >TT (Sec)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row0\" class=\"row_heading level0 row0\" >catboost</th>\n",
       "      <td id=\"T_56707_row0_col0\" class=\"data row0 col0\" >CatBoost Classifier</td>\n",
       "      <td id=\"T_56707_row0_col1\" class=\"data row0 col1\" >0.4402</td>\n",
       "      <td id=\"T_56707_row0_col2\" class=\"data row0 col2\" >0.7395</td>\n",
       "      <td id=\"T_56707_row0_col3\" class=\"data row0 col3\" >0.2251</td>\n",
       "      <td id=\"T_56707_row0_col4\" class=\"data row0 col4\" >0.4089</td>\n",
       "      <td id=\"T_56707_row0_col5\" class=\"data row0 col5\" >0.3929</td>\n",
       "      <td id=\"T_56707_row0_col6\" class=\"data row0 col6\" >0.2515</td>\n",
       "      <td id=\"T_56707_row0_col7\" class=\"data row0 col7\" >0.2632</td>\n",
       "      <td id=\"T_56707_row0_col8\" class=\"data row0 col8\" >16.8020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row1\" class=\"row_heading level0 row1\" >lightgbm</th>\n",
       "      <td id=\"T_56707_row1_col0\" class=\"data row1 col0\" >Light Gradient Boosting Machine</td>\n",
       "      <td id=\"T_56707_row1_col1\" class=\"data row1 col1\" >0.4381</td>\n",
       "      <td id=\"T_56707_row1_col2\" class=\"data row1 col2\" >0.7368</td>\n",
       "      <td id=\"T_56707_row1_col3\" class=\"data row1 col3\" >0.2254</td>\n",
       "      <td id=\"T_56707_row1_col4\" class=\"data row1 col4\" >0.4130</td>\n",
       "      <td id=\"T_56707_row1_col5\" class=\"data row1 col5\" >0.3877</td>\n",
       "      <td id=\"T_56707_row1_col6\" class=\"data row1 col6\" >0.2469</td>\n",
       "      <td id=\"T_56707_row1_col7\" class=\"data row1 col7\" >0.2614</td>\n",
       "      <td id=\"T_56707_row1_col8\" class=\"data row1 col8\" >1.6420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row2\" class=\"row_heading level0 row2\" >gbc</th>\n",
       "      <td id=\"T_56707_row2_col0\" class=\"data row2 col0\" >Gradient Boosting Classifier</td>\n",
       "      <td id=\"T_56707_row2_col1\" class=\"data row2 col1\" >0.4439</td>\n",
       "      <td id=\"T_56707_row2_col2\" class=\"data row2 col2\" >0.7428</td>\n",
       "      <td id=\"T_56707_row2_col3\" class=\"data row2 col3\" >0.2253</td>\n",
       "      <td id=\"T_56707_row2_col4\" class=\"data row2 col4\" >0.4213</td>\n",
       "      <td id=\"T_56707_row2_col5\" class=\"data row2 col5\" >0.3875</td>\n",
       "      <td id=\"T_56707_row2_col6\" class=\"data row2 col6\" >0.2505</td>\n",
       "      <td id=\"T_56707_row2_col7\" class=\"data row2 col7\" >0.2693</td>\n",
       "      <td id=\"T_56707_row2_col8\" class=\"data row2 col8\" >24.7120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row3\" class=\"row_heading level0 row3\" >xgboost</th>\n",
       "      <td id=\"T_56707_row3_col0\" class=\"data row3 col0\" >Extreme Gradient Boosting</td>\n",
       "      <td id=\"T_56707_row3_col1\" class=\"data row3 col1\" >0.4331</td>\n",
       "      <td id=\"T_56707_row3_col2\" class=\"data row3 col2\" >0.7337</td>\n",
       "      <td id=\"T_56707_row3_col3\" class=\"data row3 col3\" >0.2216</td>\n",
       "      <td id=\"T_56707_row3_col4\" class=\"data row3 col4\" >0.3996</td>\n",
       "      <td id=\"T_56707_row3_col5\" class=\"data row3 col5\" >0.3846</td>\n",
       "      <td id=\"T_56707_row3_col6\" class=\"data row3 col6\" >0.2407</td>\n",
       "      <td id=\"T_56707_row3_col7\" class=\"data row3 col7\" >0.2533</td>\n",
       "      <td id=\"T_56707_row3_col8\" class=\"data row3 col8\" >9.8720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row4\" class=\"row_heading level0 row4\" >lda</th>\n",
       "      <td id=\"T_56707_row4_col0\" class=\"data row4 col0\" >Linear Discriminant Analysis</td>\n",
       "      <td id=\"T_56707_row4_col1\" class=\"data row4 col1\" >0.4033</td>\n",
       "      <td id=\"T_56707_row4_col2\" class=\"data row4 col2\" >0.7183</td>\n",
       "      <td id=\"T_56707_row4_col3\" class=\"data row4 col3\" >0.2376</td>\n",
       "      <td id=\"T_56707_row4_col4\" class=\"data row4 col4\" >0.3675</td>\n",
       "      <td id=\"T_56707_row4_col5\" class=\"data row4 col5\" >0.3666</td>\n",
       "      <td id=\"T_56707_row4_col6\" class=\"data row4 col6\" >0.2175</td>\n",
       "      <td id=\"T_56707_row4_col7\" class=\"data row4 col7\" >0.2226</td>\n",
       "      <td id=\"T_56707_row4_col8\" class=\"data row4 col8\" >0.1330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row5\" class=\"row_heading level0 row5\" >ada</th>\n",
       "      <td id=\"T_56707_row5_col0\" class=\"data row5 col0\" >Ada Boost Classifier</td>\n",
       "      <td id=\"T_56707_row5_col1\" class=\"data row5 col1\" >0.4108</td>\n",
       "      <td id=\"T_56707_row5_col2\" class=\"data row5 col2\" >0.6209</td>\n",
       "      <td id=\"T_56707_row5_col3\" class=\"data row5 col3\" >0.2020</td>\n",
       "      <td id=\"T_56707_row5_col4\" class=\"data row5 col4\" >0.3791</td>\n",
       "      <td id=\"T_56707_row5_col5\" class=\"data row5 col5\" >0.3644</td>\n",
       "      <td id=\"T_56707_row5_col6\" class=\"data row5 col6\" >0.2174</td>\n",
       "      <td id=\"T_56707_row5_col7\" class=\"data row5 col7\" >0.2268</td>\n",
       "      <td id=\"T_56707_row5_col8\" class=\"data row5 col8\" >0.8480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row6\" class=\"row_heading level0 row6\" >rf</th>\n",
       "      <td id=\"T_56707_row6_col0\" class=\"data row6 col0\" >Random Forest Classifier</td>\n",
       "      <td id=\"T_56707_row6_col1\" class=\"data row6 col1\" >0.3582</td>\n",
       "      <td id=\"T_56707_row6_col2\" class=\"data row6 col2\" >0.6741</td>\n",
       "      <td id=\"T_56707_row6_col3\" class=\"data row6 col3\" >0.2078</td>\n",
       "      <td id=\"T_56707_row6_col4\" class=\"data row6 col4\" >0.3369</td>\n",
       "      <td id=\"T_56707_row6_col5\" class=\"data row6 col5\" >0.3448</td>\n",
       "      <td id=\"T_56707_row6_col6\" class=\"data row6 col6\" >0.1754</td>\n",
       "      <td id=\"T_56707_row6_col7\" class=\"data row6 col7\" >0.1762</td>\n",
       "      <td id=\"T_56707_row6_col8\" class=\"data row6 col8\" >4.5650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row7\" class=\"row_heading level0 row7\" >ridge</th>\n",
       "      <td id=\"T_56707_row7_col0\" class=\"data row7 col0\" >Ridge Classifier</td>\n",
       "      <td id=\"T_56707_row7_col1\" class=\"data row7 col1\" >0.4170</td>\n",
       "      <td id=\"T_56707_row7_col2\" class=\"data row7 col2\" >0.0000</td>\n",
       "      <td id=\"T_56707_row7_col3\" class=\"data row7 col3\" >0.1544</td>\n",
       "      <td id=\"T_56707_row7_col4\" class=\"data row7 col4\" >0.3345</td>\n",
       "      <td id=\"T_56707_row7_col5\" class=\"data row7 col5\" >0.3373</td>\n",
       "      <td id=\"T_56707_row7_col6\" class=\"data row7 col6\" >0.2061</td>\n",
       "      <td id=\"T_56707_row7_col7\" class=\"data row7 col7\" >0.2217</td>\n",
       "      <td id=\"T_56707_row7_col8\" class=\"data row7 col8\" >0.0370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row8\" class=\"row_heading level0 row8\" >et</th>\n",
       "      <td id=\"T_56707_row8_col0\" class=\"data row8 col0\" >Extra Trees Classifier</td>\n",
       "      <td id=\"T_56707_row8_col1\" class=\"data row8 col1\" >0.3434</td>\n",
       "      <td id=\"T_56707_row8_col2\" class=\"data row8 col2\" >0.6402</td>\n",
       "      <td id=\"T_56707_row8_col3\" class=\"data row8 col3\" >0.2041</td>\n",
       "      <td id=\"T_56707_row8_col4\" class=\"data row8 col4\" >0.3270</td>\n",
       "      <td id=\"T_56707_row8_col5\" class=\"data row8 col5\" >0.3334</td>\n",
       "      <td id=\"T_56707_row8_col6\" class=\"data row8 col6\" >0.1614</td>\n",
       "      <td id=\"T_56707_row8_col7\" class=\"data row8 col7\" >0.1619</td>\n",
       "      <td id=\"T_56707_row8_col8\" class=\"data row8 col8\" >6.0540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row9\" class=\"row_heading level0 row9\" >knn</th>\n",
       "      <td id=\"T_56707_row9_col0\" class=\"data row9 col0\" >K Neighbors Classifier</td>\n",
       "      <td id=\"T_56707_row9_col1\" class=\"data row9 col1\" >0.3401</td>\n",
       "      <td id=\"T_56707_row9_col2\" class=\"data row9 col2\" >0.6186</td>\n",
       "      <td id=\"T_56707_row9_col3\" class=\"data row9 col3\" >0.1609</td>\n",
       "      <td id=\"T_56707_row9_col4\" class=\"data row9 col4\" >0.3060</td>\n",
       "      <td id=\"T_56707_row9_col5\" class=\"data row9 col5\" >0.3155</td>\n",
       "      <td id=\"T_56707_row9_col6\" class=\"data row9 col6\" >0.1370</td>\n",
       "      <td id=\"T_56707_row9_col7\" class=\"data row9 col7\" >0.1393</td>\n",
       "      <td id=\"T_56707_row9_col8\" class=\"data row9 col8\" >1.7230</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row10\" class=\"row_heading level0 row10\" >dt</th>\n",
       "      <td id=\"T_56707_row10_col0\" class=\"data row10 col0\" >Decision Tree Classifier</td>\n",
       "      <td id=\"T_56707_row10_col1\" class=\"data row10 col1\" >0.3199</td>\n",
       "      <td id=\"T_56707_row10_col2\" class=\"data row10 col2\" >0.5856</td>\n",
       "      <td id=\"T_56707_row10_col3\" class=\"data row10 col3\" >0.1973</td>\n",
       "      <td id=\"T_56707_row10_col4\" class=\"data row10 col4\" >0.3129</td>\n",
       "      <td id=\"T_56707_row10_col5\" class=\"data row10 col5\" >0.3154</td>\n",
       "      <td id=\"T_56707_row10_col6\" class=\"data row10 col6\" >0.1400</td>\n",
       "      <td id=\"T_56707_row10_col7\" class=\"data row10 col7\" >0.1402</td>\n",
       "      <td id=\"T_56707_row10_col8\" class=\"data row10 col8\" >0.1340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row11\" class=\"row_heading level0 row11\" >nb</th>\n",
       "      <td id=\"T_56707_row11_col0\" class=\"data row11 col0\" >Naive Bayes</td>\n",
       "      <td id=\"T_56707_row11_col1\" class=\"data row11 col1\" >0.3010</td>\n",
       "      <td id=\"T_56707_row11_col2\" class=\"data row11 col2\" >0.6955</td>\n",
       "      <td id=\"T_56707_row11_col3\" class=\"data row11 col3\" >0.2768</td>\n",
       "      <td id=\"T_56707_row11_col4\" class=\"data row11 col4\" >0.3430</td>\n",
       "      <td id=\"T_56707_row11_col5\" class=\"data row11 col5\" >0.3062</td>\n",
       "      <td id=\"T_56707_row11_col6\" class=\"data row11 col6\" >0.1653</td>\n",
       "      <td id=\"T_56707_row11_col7\" class=\"data row11 col7\" >0.1686</td>\n",
       "      <td id=\"T_56707_row11_col8\" class=\"data row11 col8\" >0.0390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row12\" class=\"row_heading level0 row12\" >svm</th>\n",
       "      <td id=\"T_56707_row12_col0\" class=\"data row12 col0\" >SVM - Linear Kernel</td>\n",
       "      <td id=\"T_56707_row12_col1\" class=\"data row12 col1\" >0.2954</td>\n",
       "      <td id=\"T_56707_row12_col2\" class=\"data row12 col2\" >0.0000</td>\n",
       "      <td id=\"T_56707_row12_col3\" class=\"data row12 col3\" >0.1442</td>\n",
       "      <td id=\"T_56707_row12_col4\" class=\"data row12 col4\" >0.3175</td>\n",
       "      <td id=\"T_56707_row12_col5\" class=\"data row12 col5\" >0.2268</td>\n",
       "      <td id=\"T_56707_row12_col6\" class=\"data row12 col6\" >0.0958</td>\n",
       "      <td id=\"T_56707_row12_col7\" class=\"data row12 col7\" >0.1102</td>\n",
       "      <td id=\"T_56707_row12_col8\" class=\"data row12 col8\" >0.3140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row13\" class=\"row_heading level0 row13\" >lr</th>\n",
       "      <td id=\"T_56707_row13_col0\" class=\"data row13 col0\" >Logistic Regression</td>\n",
       "      <td id=\"T_56707_row13_col1\" class=\"data row13 col1\" >0.3263</td>\n",
       "      <td id=\"T_56707_row13_col2\" class=\"data row13 col2\" >0.5200</td>\n",
       "      <td id=\"T_56707_row13_col3\" class=\"data row13 col3\" >0.0965</td>\n",
       "      <td id=\"T_56707_row13_col4\" class=\"data row13 col4\" >0.1654</td>\n",
       "      <td id=\"T_56707_row13_col5\" class=\"data row13 col5\" >0.1706</td>\n",
       "      <td id=\"T_56707_row13_col6\" class=\"data row13 col6\" >0.0075</td>\n",
       "      <td id=\"T_56707_row13_col7\" class=\"data row13 col7\" >0.0273</td>\n",
       "      <td id=\"T_56707_row13_col8\" class=\"data row13 col8\" >15.8900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_56707_level0_row14\" class=\"row_heading level0 row14\" >qda</th>\n",
       "      <td id=\"T_56707_row14_col0\" class=\"data row14 col0\" >Quadratic Discriminant Analysis</td>\n",
       "      <td id=\"T_56707_row14_col1\" class=\"data row14 col1\" >0.0464</td>\n",
       "      <td id=\"T_56707_row14_col2\" class=\"data row14 col2\" >0.4993</td>\n",
       "      <td id=\"T_56707_row14_col3\" class=\"data row14 col3\" >0.0922</td>\n",
       "      <td id=\"T_56707_row14_col4\" class=\"data row14 col4\" >0.2192</td>\n",
       "      <td id=\"T_56707_row14_col5\" class=\"data row14 col5\" >0.0435</td>\n",
       "      <td id=\"T_56707_row14_col6\" class=\"data row14 col6\" >-0.0017</td>\n",
       "      <td id=\"T_56707_row14_col7\" class=\"data row14 col7\" >-0.0022</td>\n",
       "      <td id=\"T_56707_row14_col8\" class=\"data row14 col8\" >0.1100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x22d4a8acfd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x22d4b34ce20>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 모델들의 평가 지표 점수 출력(F1 기준 정렬)\n",
    "compare_models(sort = 'F1', n_select = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train test 데이터 분리"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                  y, \n",
    "                                                  test_size=0.01, \n",
    "                                                  random_state=42,\n",
    "                                                  stratify = y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 모델 성능 비교"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBoostClassifier 정확도: 0.4817\n",
      "LightGBMBoostClassifier 정확도: 0.4732\n",
      "CatBoostClassifier 정확도: 0.4676\n"
     ]
    }
   ],
   "source": [
    "xgb = XGBClassifier(objective = 'multi:softprob')\n",
    "lgb = LGBMClassifier()\n",
    "cat = CatBoostClassifier(silent = True)\n",
    "\n",
    "\n",
    "# XGBoostClassifier 학습,예측,정확도 출력\n",
    "xgb.fit(X_train, y_train, eval_metric = 'mlogloss')\n",
    "xgb_pred = xgb.predict(X_test)\n",
    "print('XGBoostClassifier 정확도: {0:.4f}'.format(accuracy_score(y_test, xgb_pred)))\n",
    "\n",
    "# LightGBMBoostClassifier 학습,예측,정확도 출력\n",
    "lgb.fit(X_train, y_train)\n",
    "lgb_pred = lgb.predict(X_test)\n",
    "print('LightGBMBoostClassifier 정확도: {0:.4f}'.format(accuracy_score(y_test, lgb_pred)))\n",
    "\n",
    "# CatBoostClassifier 학습,예측,정확도 출력\n",
    "cat.fit(X_train, y_train)\n",
    "cat_pred = cat.predict(X_test)\n",
    "print('CatBoostClassifier 정확도: {0:.4f}'.format(accuracy_score(y_test, cat_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBoostClassifier 정확도: 0.4817\n",
      "LightGBMBoostClassifier 정확도: 0.4732\n",
      "CatBoostClassifier 정확도: 0.4676\n"
     ]
    }
   ],
   "source": [
    "xgb = XGBClassifier(objective = 'multi:softprob')\n",
    "lgb = LGBMClassifier()\n",
    "cat = CatBoostClassifier(silent = True)\n",
    "\n",
    "\n",
    "# XGBoostClassifier 학습,예측,정확도 출력\n",
    "xgb.fit(X_train, y_train, eval_metric = 'mlogloss')\n",
    "xgb_pred = xgb.predict(X_test)\n",
    "print('XGBoostClassifier 정확도: {0:.4f}'.format(f1_score(y_test, xgb_pred, average='micro')))\n",
    "\n",
    "# LightGBMBoostClassifier 학습,예측,정확도 출력\n",
    "lgb.fit(X_train, y_train)\n",
    "lgb_pred = lgb.predict(X_test)\n",
    "print('LightGBMBoostClassifier 정확도: {0:.4f}'.format(f1_score(y_test, lgb_pred, average='micro')))\n",
    "\n",
    "# CatBoostClassifier 학습,예측,정확도 출력\n",
    "cat.fit(X_train, y_train)\n",
    "cat_pred = cat.predict(X_test)\n",
    "print('CatBoostClassifier 정확도: {0:.4f}'.format(f1_score(y_test, cat_pred, average='micro')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 파라미터 튜닝"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Catboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def objective(trial):\n",
    "#     param = {\n",
    "#       \"random_state\": 42,\n",
    "#       'learning_rate' : trial.suggest_loguniform('learning_rate', 0.01, 0.3),\n",
    "#       'bagging_temperature' :trial.suggest_loguniform('bagging_temperature', 0.01, 100.00),\n",
    "#       \"n_estimators\":trial.suggest_int(\"n_estimators\", 1000, 10000),\n",
    "#       \"max_depth\":trial.suggest_int(\"max_depth\", 4, 16),\n",
    "#       'random_strength' :trial.suggest_int('random_strength', 0, 100),\n",
    "#       \"colsample_bylevel\":trial.suggest_float(\"colsample_bylevel\", 0.4, 1.0),\n",
    "#       \"l2_leaf_reg\":trial.suggest_float(\"l2_leaf_reg\",1e-8,3e-5),\n",
    "#       \"min_child_samples\": trial.suggest_int(\"min_child_samples\", 5, 100),\n",
    "#       \"max_bin\": trial.suggest_int(\"max_bin\", 200, 500),\n",
    "#       'od_type': trial.suggest_categorical('od_type', ['IncToDec', 'Iter'])\n",
    "#     }\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                  y, \n",
    "                                                  test_size=0.01, \n",
    "                                                  random_state=42,\n",
    "                                                  stratify = y)\n",
    "    param = {\n",
    "      \"random_state\": 42,\n",
    "      'learning_rate' : trial.suggest_loguniform('learning_rate', 0.01, 0.3),\n",
    "      \"n_estimators\":trial.suggest_int(\"n_estimators\", 1000, 10000),\n",
    "      \"max_depth\":trial.suggest_int(\"max_depth\", 4, 16),\n",
    "      \"l2_leaf_reg\":trial.suggest_float(\"l2_leaf_reg\",1e-8,3e-5),\n",
    "      'random_strength' :trial.suggest_int('random_strength', 0, 100),\n",
    "    }\n",
    "    \n",
    "    cat = CatBoostClassifier(**param)\n",
    "    cat.fit(X_train, y_train,\n",
    "          eval_set=[(X_train, y_train), (X_test, y_test)],\n",
    "          early_stopping_rounds=35,\n",
    "          verbose=100)\n",
    "    cat_pred = cat.predict(X_test)\n",
    "    accuracy = accuracy_score(y_test, cat_pred)\n",
    "\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sampler = TPESampler(seed=42)\n",
    "study = optuna.create_study(\n",
    "    study_name = 'cat_parameter_opt',\n",
    "    direction = 'maximize',\n",
    "    sampler = sampler,\n",
    ")\n",
    "study.optimize(objective, n_trials = 10)\n",
    "print(\"Best Score:\",study.best_value)\n",
    "print(\"Best trial\", study.best_trial.params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### best 파라미터 활용 학습"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 2.3692126\ttest: 2.3692126\ttest1: 2.3693904\tbest: 2.3693904 (0)\ttotal: 230ms\tremaining: 14m 20s\n",
      "100:\tlearn: 1.6677635\ttest: 1.6677635\ttest1: 1.6997109\tbest: 1.6997109 (100)\ttotal: 8.69s\tremaining: 5m 12s\n",
      "200:\tlearn: 1.5644543\ttest: 1.5644543\ttest1: 1.6158797\tbest: 1.6158787 (199)\ttotal: 16.5s\tremaining: 4m 50s\n",
      "300:\tlearn: 1.5121873\ttest: 1.5121873\ttest1: 1.5856547\tbest: 1.5856547 (300)\ttotal: 25.2s\tremaining: 4m 47s\n",
      "400:\tlearn: 1.4751490\ttest: 1.4751490\ttest1: 1.5657094\tbest: 1.5657094 (400)\ttotal: 33.8s\tremaining: 4m 41s\n",
      "500:\tlearn: 1.4488069\ttest: 1.4488069\ttest1: 1.5534276\tbest: 1.5534276 (500)\ttotal: 41.6s\tremaining: 4m 28s\n",
      "600:\tlearn: 1.4236337\ttest: 1.4236337\ttest1: 1.5434614\tbest: 1.5434614 (600)\ttotal: 49.5s\tremaining: 4m 18s\n",
      "700:\tlearn: 1.3936462\ttest: 1.3936462\ttest1: 1.5341553\tbest: 1.5340142 (695)\ttotal: 59s\tremaining: 4m 15s\n",
      "800:\tlearn: 1.3498054\ttest: 1.3498054\ttest1: 1.5232376\tbest: 1.5231589 (799)\ttotal: 1m 11s\tremaining: 4m 21s\n",
      "Stopped by overfitting detector  (35 iterations wait)\n",
      "\n",
      "bestTest = 1.522803863\n",
      "bestIteration = 811\n",
      "\n",
      "Shrink model to first 812 iterations.\n",
      "CatBoostClassifier 정확도 :  0.4732394366197183\n"
     ]
    }
   ],
   "source": [
    "param = {'random_state': 42, 'learning_rate': 0.018659959624904916, 'n_estimators': 3738, \n",
    "         'max_depth': 10, 'l2_leaf_reg': 1.2964031109077052e-05, 'random_strength': 29}\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                  y, \n",
    "                                                  test_size=0.01, \n",
    "                                                  random_state=42,\n",
    "                                                  stratify = y)\n",
    "\n",
    "cat = CatBoostClassifier(**param)\n",
    "cat.fit(X_train, y_train,\n",
    "        eval_set=[(X_train, y_train), (X_test, y_test)],\n",
    "        early_stopping_rounds=35,\n",
    "        verbose=100)\n",
    "\n",
    "cat_pred = cat.predict(X_test) \n",
    "print(f'CatBoostClassifier 정확도 : ', accuracy_score(y_test, cat_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### feature importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_feature_importance(importance, names, model_type):\n",
    "    \n",
    "    #Create arrays from feature importance and feature names\n",
    "    feature_importance = np.array(importance)\n",
    "    feature_names = np.array(names)\n",
    "    \n",
    "    #Create a DataFrame using a Dictionary\n",
    "    data={'feature_names':feature_names,'feature_importance':feature_importance}\n",
    "    fi_df = pd.DataFrame(data)\n",
    "    \n",
    "    #Sort the DataFrame in order decreasing feature importance\n",
    "    fi_df.sort_values(by=['feature_importance'], ascending= False, inplace=True)\n",
    "    \n",
    "    #Define size of bar plot\n",
    "    plt.figure(figsize=(10,8))\n",
    "    #Plot Searborn bar chart\n",
    "    sns.barplot(x=fi_df['feature_importance'], y=fi_df['feature_names'])\n",
    "    #Add chart labels\n",
    "    plt.title(model_type + ' feature importance')\n",
    "    plt.xlabel('feature importance')\n",
    "    plt.ylabel('feature names')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_feature_importance(cat.get_feature_importance(), X_train.columns, 'catboost')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['age', 'capital-gain', 'capital-loss', 'hours-per-week', 'income',\n",
       "       'workclass_Federal-gov', 'workclass_Local-gov', 'workclass_None',\n",
       "       'workclass_Private', 'workclass_Self-emp-inc',\n",
       "       'workclass_Self-emp-not-inc', 'workclass_State-gov',\n",
       "       'workclass_Without-pay', 'marital-status_Divorced',\n",
       "       'marital-status_Married-AF-spouse', 'marital-status_Married-civ-spouse',\n",
       "       'marital-status_Married-spouse-absent', 'marital-status_Never-married',\n",
       "       'marital-status_Separated', 'marital-status_Widowed',\n",
       "       'occupation_Adm-clerical', 'occupation_Craft-repair',\n",
       "       'occupation_Exec-managerial', 'occupation_Farming-fishing',\n",
       "       'occupation_Handlers-cleaners', 'occupation_Machine-op-inspct',\n",
       "       'occupation_None', 'occupation_Prof-specialty',\n",
       "       'occupation_Protective-serv', 'occupation_Sales',\n",
       "       'occupation_Tech-support', 'occupation_Transport-moving',\n",
       "       'relationship_Husband', 'relationship_Not-in-family',\n",
       "       'relationship_Other-relative', 'relationship_Own-child',\n",
       "       'relationship_Unmarried', 'relationship_Wife'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " X_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([17.95913807,  3.64528461,  2.01757625, 11.5606266 , 10.67851688,\n",
       "        0.79411714,  1.40273643,  1.01344591,  3.4266205 ,  0.91146699,\n",
       "        1.6869141 ,  1.08161394,  0.02419708,  1.68328508,  0.02472063,\n",
       "        1.39333892,  0.33591652,  2.58578811,  0.69927403,  0.49703791,\n",
       "        1.88026772,  3.27013795,  3.12655914,  0.87694323,  1.21499929,\n",
       "        1.79271146,  0.53973268,  8.32507285,  0.33814478,  2.16106688,\n",
       "        1.14581073,  1.82377669,  1.99200125,  2.68486492,  0.77630305,\n",
       "        2.22399972,  1.5394866 ,  0.86650534])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat.get_feature_importance()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Lightgbm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "lgb = LGBMClassifier()\n",
    "\n",
    "params = {'boosting_type' : ['gbdt', 'dart', 'goss', 'rf'],\n",
    "          'num_leaves': randint(6, 127), \n",
    "          'min_child_samples': randint(10, 100), \n",
    "          'min_child_weight': [1e-5, 1e-3, 1e-2, 1e-1, 1, 1e1, 1e2, 1e3, 1e4],\n",
    "          'subsample': uniform(loc=0.2, scale=0.8), \n",
    "          'colsample_bytree': uniform(loc=0.4, scale=0.6),\n",
    "          'reg_alpha': [0, 1e-1, 1, 2, 5, 7, 10, 50, 100],\n",
    "          'reg_lambda': [0, 1e-1, 1, 5, 10, 20, 50, 100],\n",
    "          'learning_rate': uniform(loc = 0.001, scale = 0.5),\n",
    "          'objective': ['multiclass'],\n",
    "          'random_state' : randint(0, 100),\n",
    "          'n_estimators' : randint(100, 1000)\n",
    "          }\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                  y, \n",
    "                                                  test_size=0.01, \n",
    "                                                  random_state=42,\n",
    "                                                  stratify = y)\n",
    "\n",
    "param_dict = {} # 파라미터 저장\n",
    "acc_dict = {} # randomstate 별 정확도 저장\n",
    "# random search\n",
    "for i in tqdm(range(5)):\n",
    "    clf = RandomizedSearchCV(estimator = lgb, param_distributions = params, scoring='accuracy', random_state = i, n_jobs = -1, verbose=True, cv=5)\n",
    "    clf.fit(X_train, y_train, early_stopping_rounds = 30, eval_metric=\"logloss\", eval_set=[(X_train, y_train), (X_test, y_test)])\n",
    "    preds = clf.predict(X_test)\n",
    "    param_dict[i] = clf.best_params_\n",
    "    acc_dict[i] = accuracy_score(y_test, preds)\n",
    "    print('LightGBMBoostClassifier 정확도: {0:.4f}'.format(accuracy_score(y_test, preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[2]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# acc 가장 큰  random_state index\n",
    "\n",
    "[k for k,v in acc_dict.items() if max(acc_dict.values()) == v]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'boosting_type': 'dart',\n",
       " 'colsample_bytree': 0.4719703255578339,\n",
       " 'learning_rate': 0.2435879547999325,\n",
       " 'min_child_samples': 30,\n",
       " 'min_child_weight': 1,\n",
       " 'n_estimators': 905,\n",
       " 'num_leaves': 45,\n",
       " 'objective': 'multiclass',\n",
       " 'random_state': 67,\n",
       " 'reg_alpha': 10,\n",
       " 'reg_lambda': 10,\n",
       " 'subsample': 0.6694375824000827}"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 최대 값일 때 파라미터\n",
    "\n",
    "param_dict[2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Best 파라미터 활용 학습 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = {'boosting_type': 'dart',\n",
    " 'colsample_bytree': 0.4719703255578339,\n",
    " 'learning_rate': 0.2435879547999325,\n",
    " 'min_child_samples': 30,\n",
    " 'min_child_weight': 1,\n",
    " 'n_estimators': 905,\n",
    " 'num_leaves': 45,\n",
    " 'objective': 'multiclass',\n",
    " 'random_state': 67,\n",
    " 'reg_alpha': 10,\n",
    " 'reg_lambda': 10,\n",
    " 'subsample': 0.6694375824000827}\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                  y, \n",
    "                                                  test_size=0.01, \n",
    "                                                  random_state=42,\n",
    "                                                  stratify = y)\n",
    "\n",
    "lgb = LGBMClassifier(**params)\n",
    "lgb.fit(X_train, y_train,\n",
    "        eval_set=[(X_train, y_train), (X_test, y_test)],\n",
    "        early_stopping_rounds=30,\n",
    "        eval_metric = 'logloss',\n",
    "        verbose=True)\n",
    "\n",
    "lgb_pred = lgb.predict(X_test) \n",
    "print(f'LightGBMBoostClassifier 정확도 : ', accuracy_score(y_test, lgb_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LightGBMBoostClassifier 정확도 :  0.4788732394366197\n"
     ]
    }
   ],
   "source": [
    "# LightGBM 정확도\n",
    "print(f'LightGBMBoostClassifier 정확도 : ', accuracy_score(y_test, lgb_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### feature importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_feature_importance(lgb.feature_importances_, X_train.columns, 'lightgbm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['age', 'capital-gain', 'capital-loss', 'hours-per-week', 'income',\n",
       "       'workclass_Federal-gov', 'workclass_Local-gov', 'workclass_None',\n",
       "       'workclass_Private', 'workclass_Self-emp-inc',\n",
       "       'workclass_Self-emp-not-inc', 'workclass_State-gov',\n",
       "       'workclass_Without-pay', 'marital-status_Divorced',\n",
       "       'marital-status_Married-AF-spouse', 'marital-status_Married-civ-spouse',\n",
       "       'marital-status_Married-spouse-absent', 'marital-status_Never-married',\n",
       "       'marital-status_Separated', 'marital-status_Widowed',\n",
       "       'occupation_Adm-clerical', 'occupation_Craft-repair',\n",
       "       'occupation_Exec-managerial', 'occupation_Farming-fishing',\n",
       "       'occupation_Handlers-cleaners', 'occupation_Machine-op-inspct',\n",
       "       'occupation_None', 'occupation_Prof-specialty',\n",
       "       'occupation_Protective-serv', 'occupation_Sales',\n",
       "       'occupation_Tech-support', 'occupation_Transport-moving',\n",
       "       'relationship_Husband', 'relationship_Not-in-family',\n",
       "       'relationship_Other-relative', 'relationship_Own-child',\n",
       "       'relationship_Unmarried', 'relationship_Wife'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " X_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([33204,  6932,  6386, 26424,  2793,   423,  1066,  1302,  1340,\n",
       "         933,  1277,  1280,     0,   717,     0,   337,   596,  1313,\n",
       "        1062,  1350,  1277,  1964,  1853,  2041,  2059,  2082,   701,\n",
       "        3041,  1113,  1009,  1876,  2112,   738,  1024,  1166,   663,\n",
       "         815,   571])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lgb.feature_importances_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### XGBOOST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb = XGBClassifier(silent = True, nthread = 1)\n",
    "\n",
    "# A parameter grid for XGBoost\n",
    "params = {\n",
    "        'learning_rate': uniform(loc = 0.001, scale = 0.5),\n",
    "        'n_estimators' : randint(100, 1000),\n",
    "        'random_state' : randint(0, 100),\n",
    "        'min_child_weight': [1, 5, 10],\n",
    "        'gamma': [0.5, 1, 1.5, 2, 5],\n",
    "        'subsample': [0.6, 0.8, 1.0],\n",
    "        'colsample_bytree': [0.6, 0.8, 1.0],\n",
    "        'max_depth': [3, 4, 5],\n",
    "        'objective' : 'multi:softprob'\n",
    "        }\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                  y, \n",
    "                                                  test_size=0.01, \n",
    "                                                  random_state=42,\n",
    "                                                  stratify = y)\n",
    "\n",
    "param_dict = {} # 파라미터 저장\n",
    "acc_dict = {} # randomstate 별 정확도 저장\n",
    "# random search\n",
    "for i in tqdm(range(5)):\n",
    "    clf = RandomizedSearchCV(estimator = xgb, param_distributions = params, scoring='accuracy', random_state = i, n_jobs = -1, verbose=True, cv=5)\n",
    "    clf.fit(X_train, y_train, early_stopping_rounds = 30, eval_metric=\"mlogloss\", eval_set=[(X_train, y_train), (X_test, y_test)])\n",
    "    preds = clf.predict(X_test)\n",
    "    param_dict[i] = clf.best_params_\n",
    "    acc_dict[i] = accuracy_score(y_test, preds)\n",
    "    print('XGBoostClassifier 정확도: {0:.4f}'.format(accuracy_score(y_test, preds)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[4]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# acc 가장 큰  random_state index\n",
    "\n",
    "[k for k,v in acc_dict.items() if max(acc_dict.values()) == v]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'colsample_bytree': 1.0,\n",
       " 'gamma': 2,\n",
       " 'learning_rate': 0.19945000145086633,\n",
       " 'max_depth': 5,\n",
       " 'min_child_weight': 10,\n",
       " 'n_estimators': 843,\n",
       " 'objective': 'b',\n",
       " 'random_state': 71,\n",
       " 'subsample': 0.8}"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 최대 값일 때 파라미터\n",
    "\n",
    "param_dict[4]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Best 파라미터 활용 학습"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = {'colsample_bytree': 1.0,\n",
    " 'gamma': 2,\n",
    " 'learning_rate': 0.19945000145086633,\n",
    " 'max_depth': 5,\n",
    " 'min_child_weight': 10,\n",
    " 'n_estimators': 843,\n",
    " 'objective': 'b',\n",
    " 'random_state': 71,\n",
    " 'subsample': 0.8}\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                  y, \n",
    "                                                  test_size=0.01, \n",
    "                                                  random_state=42,\n",
    "                                                  stratify = y)\n",
    "\n",
    "xgb = XGBClassifier(silent = True, nthread = 1, **params)\n",
    "xgb.fit(X_train, y_train,\n",
    "        eval_set=[(X_train, y_train), (X_test, y_test)],\n",
    "        early_stopping_rounds=30,\n",
    "        eval_metric = 'mlogloss',\n",
    "        verbose=True)\n",
    "\n",
    "xgb_pred = xgb.predict(X_test) \n",
    "print(f'XGBoostClassifier 정확도 : ', accuracy_score(y_test, xgb_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBoostClassifier 정확도 :  0.48732394366197185\n"
     ]
    }
   ],
   "source": [
    "# XGBoostClassifier 정확도\n",
    "print(f'XGBoostClassifier 정확도 : ', accuracy_score(y_test, xgb_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### feature importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_feature_importance(xgb.feature_importances_, X_train.columns, 'XGboost')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['age', 'capital-gain', 'capital-loss', 'hours-per-week', 'income',\n",
       "       'workclass_Federal-gov', 'workclass_Local-gov', 'workclass_None',\n",
       "       'workclass_Private', 'workclass_Self-emp-inc',\n",
       "       'workclass_Self-emp-not-inc', 'workclass_State-gov',\n",
       "       'workclass_Without-pay', 'marital-status_Divorced',\n",
       "       'marital-status_Married-AF-spouse', 'marital-status_Married-civ-spouse',\n",
       "       'marital-status_Married-spouse-absent', 'marital-status_Never-married',\n",
       "       'marital-status_Separated', 'marital-status_Widowed',\n",
       "       'occupation_Adm-clerical', 'occupation_Craft-repair',\n",
       "       'occupation_Exec-managerial', 'occupation_Farming-fishing',\n",
       "       'occupation_Handlers-cleaners', 'occupation_Machine-op-inspct',\n",
       "       'occupation_None', 'occupation_Prof-specialty',\n",
       "       'occupation_Protective-serv', 'occupation_Sales',\n",
       "       'occupation_Tech-support', 'occupation_Transport-moving',\n",
       "       'relationship_Husband', 'relationship_Not-in-family',\n",
       "       'relationship_Other-relative', 'relationship_Own-child',\n",
       "       'relationship_Unmarried', 'relationship_Wife'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.02911009, 0.01293533, 0.01001622, 0.01346364, 0.05449927,\n",
       "       0.01003804, 0.02894605, 0.02558032, 0.01512738, 0.02390403,\n",
       "       0.02082296, 0.01689692, 0.        , 0.01398401, 0.        ,\n",
       "       0.01334029, 0.00923465, 0.02813907, 0.01088617, 0.01054839,\n",
       "       0.03250549, 0.03696651, 0.11006492, 0.01473634, 0.02378887,\n",
       "       0.03218851, 0.        , 0.17896049, 0.01333411, 0.05050922,\n",
       "       0.05064295, 0.02868803, 0.01401572, 0.01234785, 0.01043693,\n",
       "       0.01569035, 0.01839398, 0.00925688], dtype=float32)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb.feature_importances_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### parameter tuning 모델 성능 비교"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBoostClassifier 정확도 :  0.48732394366197185\n",
      "LightGBMBoostClassifier 정확도 :  0.4788732394366197\n",
      "CatBoostClassifier 정확도 :  0.4732394366197183\n"
     ]
    }
   ],
   "source": [
    "print(f'XGBoostClassifier 정확도 : ', accuracy_score(y_test, xgb_pred))\n",
    "print(f'LightGBMBoostClassifier 정확도 : ', accuracy_score(y_test, lgb_pred))\n",
    "print(f'CatBoostClassifier 정확도 : ', accuracy_score(y_test, cat_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### VotingClassifier 앙상블 수행"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# Catboost, LightGBM, XGBoost 보팅(소프트)\n",
    "vo_soft_clf = VotingClassifier( estimators=[('CAT', cat), ('LGB', lgb), ('XGB', xgb)], voting='soft')\n",
    "\n",
    "# VotingClassifier 학습, 예측, 정확도 계산\n",
    "vo_soft_clf.fit(X_train, y_train)\n",
    "soft_pred = vo_soft_clf.predict(X_test)\n",
    "print('VotingClassifier Soft 정확도: {0:.4f}'.format(accuracy_score(y_test, soft_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VotingClassifier Soft 정확도: 0.4704\n"
     ]
    }
   ],
   "source": [
    "print('VotingClassifier Soft 정확도: {0:.4f}'.format(accuracy_score(y_test, soft_pred)))"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}