Employee Absenteeism.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Employee Absenteeism Project Work"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import the relevant libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix\n",
    "from sklearn import svm, tree\n",
    "import xgboost\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn import metrics\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "import pickle\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the preprocessed CSV data\n",
    "data = pd.read_excel('datasets_32809_42898_Absenteeism_at_work_Project.xls')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 740 entries, 0 to 739\n",
      "Data columns (total 21 columns):\n",
      " #   Column                           Non-Null Count  Dtype  \n",
      "---  ------                           --------------  -----  \n",
      " 0   ID                               740 non-null    int64  \n",
      " 1   Reason for absence               737 non-null    float64\n",
      " 2   Month of absence                 739 non-null    float64\n",
      " 3   Day of the week                  740 non-null    int64  \n",
      " 4   Seasons                          740 non-null    int64  \n",
      " 5   Transportation expense           733 non-null    float64\n",
      " 6   Distance from Residence to Work  737 non-null    float64\n",
      " 7   Service time                     737 non-null    float64\n",
      " 8   Age                              737 non-null    float64\n",
      " 9   Work load Average/day            730 non-null    float64\n",
      " 10  Hit target                       734 non-null    float64\n",
      " 11  Disciplinary failure             734 non-null    float64\n",
      " 12  Education                        730 non-null    float64\n",
      " 13  Son                              734 non-null    float64\n",
      " 14  Social drinker                   737 non-null    float64\n",
      " 15  Social smoker                    736 non-null    float64\n",
      " 16  Pet                              738 non-null    float64\n",
      " 17  Weight                           739 non-null    float64\n",
      " 18  Height                           726 non-null    float64\n",
      " 19  Body mass index                  709 non-null    float64\n",
      " 20  Absenteeism time in hours        718 non-null    float64\n",
      "dtypes: float64(18), int64(3)\n",
      "memory usage: 121.5 KB\n"
     ]
    }
   ],
   "source": [
    "#display some information about data\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#upon examination, there will be no need for the ID collumn, so it will be dropped\n",
    "data = data.drop(['ID'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 4.0,\n",
       " 5.0,\n",
       " 6.0,\n",
       " 7.0,\n",
       " 8.0,\n",
       " 9.0,\n",
       " 10.0,\n",
       " 11.0,\n",
       " 12.0,\n",
       " 13.0,\n",
       " 14.0,\n",
       " 15.0,\n",
       " 16.0,\n",
       " 17.0,\n",
       " 18.0,\n",
       " 19.0,\n",
       " 21.0,\n",
       " 22.0,\n",
       " 23.0,\n",
       " 24.0,\n",
       " 25.0,\n",
       " 26.0,\n",
       " 27.0,\n",
       " 28.0,\n",
       " nan]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quick check on the 'Reason for absence column'\n",
    "sorted(data['Reason for absence'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Given the below meaning for the categoy of CIDs I will grup for better repesentation ##\n",
    "'''\n",
    "I Certain infectious and parasitic diseases\n",
    "II Neoplasms\n",
    "III Diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism\n",
    "IV Endocrine, nutritional and metabolic diseases\n",
    "V Mental and behavioural disorders\n",
    "VI Diseases of the nervous system\n",
    "VII Diseases of the eye and adnexa\n",
    "VIII Diseases of the ear and mastoid process\n",
    "IX Diseases of the circulatory system\n",
    "X Diseases of the respiratory system\n",
    "XI Diseases of the digestive system\n",
    "XII Diseases of the skin and subcutaneous tissue\n",
    "XIII Diseases of the musculoskeletal system and connective tissue\n",
    "XIV Diseases of the genitourinary system\n",
    "XV Pregnancy, childbirth and the puerperium\n",
    "XVI Certain conditions originating in the perinatal period\n",
    "XVII Congenital malformations, deformations and chromosomal abnormalities\n",
    "XVIII Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified\n",
    "XIX Injury, poisoning and certain other consequences of external causes\n",
    "XX External causes of morbidity and mortality\n",
    "XXI Factors influencing health status and contact with health services.\n",
    "\n",
    "And 7 categories without (CID) patient follow-up (22), medical consultation (23), blood donation \n",
    "(24), laboratory examination (25), unjustified absence (26), physiotherapy (27), dental consultation (28).'''\n",
    "\n",
    "### the grouping is as follows\n",
    "'''\n",
    "1- 14 are various diseases\n",
    "15 -17 : pregnancy and given birth related\n",
    "18-21: poisons or diseases not elsewere categorise\n",
    "22 and above : light reason or less serious reasons\n",
    "\n",
    "'''\n",
    "### then apply the function below to the dataframe\n",
    "\n",
    "\n",
    "def Reason(data):\n",
    "    if data['Reason for absence'] < 15 :\n",
    "        d = 'R_Known'\n",
    "    elif data['Reason for absence'] >= 15 and data['Reason for absence'] <= 17  :\n",
    "        d = 'R_Preg_Birth'\n",
    "    elif data['Reason for absence'] >= 18 and data['Reason for absence'] <= 21  :\n",
    "        d = 'R_Pois_unclass'\n",
    "    elif data['Reason for absence'] == 22:\n",
    "        d = 'R_NotSerious'\n",
    "    else:\n",
    "        d = 'R_NotSerious'\n",
    "    return d\n",
    "\n",
    "data['ReasonGroups'] = data.apply(Reason, axis=1)\n",
    "\n",
    "# so lets drop the Reasons for absence column because is no more useful\n",
    "data = data.drop(['Reason for absence'], axis = 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>Month of absence</th>\n",
       "      <th>Day of the week</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Transportation expense</th>\n",
       "      <th>Distance from Residence to Work</th>\n",
       "      <th>Service time</th>\n",
       "      <th>Age</th>\n",
       "      <th>Work load Average/day</th>\n",
       "      <th>Hit target</th>\n",
       "      <th>Disciplinary failure</th>\n",
       "      <th>...</th>\n",
       "      <th>Social smoker</th>\n",
       "      <th>Pet</th>\n",
       "      <th>Weight</th>\n",
       "      <th>Height</th>\n",
       "      <th>Body mass index</th>\n",
       "      <th>Absenteeism time in hours</th>\n",
       "      <th>R_Known</th>\n",
       "      <th>R_NotSerious</th>\n",
       "      <th>R_Pois_unclass</th>\n",
       "      <th>R_Preg_Birth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>289.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>118.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>178.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>179.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>279.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>68.0</td>\n",
       "      <td>168.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>289.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>179.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>260.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>168.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>155.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>196.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>235.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Month of absence  Day of the week  Seasons  Transportation expense  \\\n",
       "0               7.0                3        1                   289.0   \n",
       "1               7.0                3        1                   118.0   \n",
       "2               7.0                4        1                   179.0   \n",
       "3               7.0                5        1                   279.0   \n",
       "4               7.0                5        1                   289.0   \n",
       "5               7.0                6        1                   179.0   \n",
       "6               7.0                6        1                     NaN   \n",
       "7               7.0                6        1                   260.0   \n",
       "8               7.0                2        1                   155.0   \n",
       "9               7.0                2        1                   235.0   \n",
       "\n",
       "   Distance from Residence to Work  Service time   Age  \\\n",
       "0                             36.0          13.0  33.0   \n",
       "1                             13.0          18.0  50.0   \n",
       "2                             51.0          18.0  38.0   \n",
       "3                              5.0          14.0  39.0   \n",
       "4                             36.0          13.0  33.0   \n",
       "5                             51.0          18.0  38.0   \n",
       "6                             52.0           3.0  28.0   \n",
       "7                             50.0          11.0  36.0   \n",
       "8                             12.0          14.0  34.0   \n",
       "9                             11.0          14.0  37.0   \n",
       "\n",
       "   Work load Average/day   Hit target  Disciplinary failure  ...  \\\n",
       "0                239554.0        97.0                   0.0  ...   \n",
       "1                239554.0        97.0                   1.0  ...   \n",
       "2                239554.0        97.0                   0.0  ...   \n",
       "3                239554.0        97.0                   0.0  ...   \n",
       "4                239554.0        97.0                   0.0  ...   \n",
       "5                239554.0        97.0                   0.0  ...   \n",
       "6                239554.0        97.0                   0.0  ...   \n",
       "7                239554.0        97.0                   0.0  ...   \n",
       "8                239554.0        97.0                   0.0  ...   \n",
       "9                239554.0        97.0                   0.0  ...   \n",
       "\n",
       "   Social smoker  Pet  Weight  Height  Body mass index  \\\n",
       "0            0.0  1.0    90.0   172.0             30.0   \n",
       "1            0.0  0.0    98.0   178.0             31.0   \n",
       "2            0.0  0.0    89.0   170.0             31.0   \n",
       "3            1.0  0.0    68.0   168.0             24.0   \n",
       "4            0.0  1.0    90.0   172.0             30.0   \n",
       "5            0.0  0.0    89.0   170.0             31.0   \n",
       "6            0.0  4.0    80.0   172.0             27.0   \n",
       "7            0.0  0.0    65.0   168.0             23.0   \n",
       "8            0.0  0.0    95.0   196.0             25.0   \n",
       "9            0.0  1.0    88.0   172.0             29.0   \n",
       "\n",
       "   Absenteeism time in hours  R_Known  R_NotSerious  R_Pois_unclass  \\\n",
       "0                        4.0        0             1               0   \n",
       "1                        0.0        1             0               0   \n",
       "2                        2.0        0             1               0   \n",
       "3                        4.0        1             0               0   \n",
       "4                        2.0        0             1               0   \n",
       "5                        NaN        0             1               0   \n",
       "6                        8.0        0             1               0   \n",
       "7                        4.0        0             1               0   \n",
       "8                       40.0        0             0               1   \n",
       "9                        8.0        0             1               0   \n",
       "\n",
       "   R_Preg_Birth  \n",
       "0             0  \n",
       "1             0  \n",
       "2             0  \n",
       "3             0  \n",
       "4             0  \n",
       "5             0  \n",
       "6             0  \n",
       "7             0  \n",
       "8             0  \n",
       "9             0  \n",
       "\n",
       "[10 rows x 23 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets get dummies\n",
    "R_dummies = pd.get_dummies(data['ReasonGroups'])\n",
    "\n",
    "#lets merge it\n",
    "data = pd.concat([data, R_dummies], axis = 1)\n",
    "\n",
    "data = data.drop(['ReasonGroups'], axis = 1)\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check for Missiing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Month of absence                    1\n",
       "Day of the week                     0\n",
       "Seasons                             0\n",
       "Transportation expense              7\n",
       "Distance from Residence to Work     3\n",
       "Service time                        3\n",
       "Age                                 3\n",
       "Work load Average/day              10\n",
       "Hit target                          6\n",
       "Disciplinary failure                6\n",
       "Education                          10\n",
       "Son                                 6\n",
       "Social drinker                      3\n",
       "Social smoker                       4\n",
       "Pet                                 2\n",
       "Weight                              1\n",
       "Height                             14\n",
       "Body mass index                    31\n",
       "Absenteeism time in hours          22\n",
       "R_Known                             0\n",
       "R_NotSerious                        0\n",
       "R_Pois_unclass                      0\n",
       "R_Preg_Birth                        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# data.isnull() # shows a df with the information whether a data point is null \n",
    "# Since True = the data point is missing, while False = the data point is not missing, we can sum them\n",
    "# This will give us the total number of missing values feature-wise\n",
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 25 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# As we can see there are quite a few missing data and for a datset with just 740 rows\n",
    "# it is not a good idea to drop this rows with missiing data so different method will be used to fill the columns\n",
    "# so let take a look at the spread of each column\n",
    "\n",
    "# Visulazing the distibution of the data for every feature\n",
    "data.hist(linewidth=1, histtype='stepfilled', facecolor='g', figsize=(20, 20));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>Month of absence</th>\n",
       "      <th>Day of the week</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Transportation expense</th>\n",
       "      <th>Distance from Residence to Work</th>\n",
       "      <th>Service time</th>\n",
       "      <th>Age</th>\n",
       "      <th>Work load Average/day</th>\n",
       "      <th>Hit target</th>\n",
       "      <th>Disciplinary failure</th>\n",
       "      <th>...</th>\n",
       "      <th>Social smoker</th>\n",
       "      <th>Pet</th>\n",
       "      <th>Weight</th>\n",
       "      <th>Height</th>\n",
       "      <th>Body mass index</th>\n",
       "      <th>Absenteeism time in hours</th>\n",
       "      <th>R_Known</th>\n",
       "      <th>R_NotSerious</th>\n",
       "      <th>R_Pois_unclass</th>\n",
       "      <th>R_Preg_Birth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>289.000000</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>178.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>179.000000</td>\n",
       "      <td>51.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>279.000000</td>\n",
       "      <td>5.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>68.0</td>\n",
       "      <td>168.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>289.000000</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>179.000000</td>\n",
       "      <td>51.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>221.035471</td>\n",
       "      <td>52.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>260.000000</td>\n",
       "      <td>50.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>168.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>155.000000</td>\n",
       "      <td>12.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>196.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>235.000000</td>\n",
       "      <td>11.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Month of absence  Day of the week  Seasons  Transportation expense  \\\n",
       "0               7.0                3        1              289.000000   \n",
       "1               7.0                3        1              118.000000   \n",
       "2               7.0                4        1              179.000000   \n",
       "3               7.0                5        1              279.000000   \n",
       "4               7.0                5        1              289.000000   \n",
       "5               7.0                6        1              179.000000   \n",
       "6               7.0                6        1              221.035471   \n",
       "7               7.0                6        1              260.000000   \n",
       "8               7.0                2        1              155.000000   \n",
       "9               7.0                2        1              235.000000   \n",
       "\n",
       "   Distance from Residence to Work  Service time   Age  \\\n",
       "0                             36.0          13.0  33.0   \n",
       "1                             13.0          18.0  50.0   \n",
       "2                             51.0          18.0  38.0   \n",
       "3                              5.0          14.0  39.0   \n",
       "4                             36.0          13.0  33.0   \n",
       "5                             51.0          18.0  38.0   \n",
       "6                             52.0           3.0  28.0   \n",
       "7                             50.0          11.0  36.0   \n",
       "8                             12.0          14.0  34.0   \n",
       "9                             11.0          14.0  37.0   \n",
       "\n",
       "   Work load Average/day   Hit target  Disciplinary failure  ...  \\\n",
       "0                239554.0        97.0                   0.0  ...   \n",
       "1                239554.0        97.0                   1.0  ...   \n",
       "2                239554.0        97.0                   0.0  ...   \n",
       "3                239554.0        97.0                   0.0  ...   \n",
       "4                239554.0        97.0                   0.0  ...   \n",
       "5                239554.0        97.0                   0.0  ...   \n",
       "6                239554.0        97.0                   0.0  ...   \n",
       "7                239554.0        97.0                   0.0  ...   \n",
       "8                239554.0        97.0                   0.0  ...   \n",
       "9                239554.0        97.0                   0.0  ...   \n",
       "\n",
       "   Social smoker  Pet  Weight  Height  Body mass index  \\\n",
       "0            0.0  1.0    90.0   172.0             30.0   \n",
       "1            0.0  0.0    98.0   178.0             31.0   \n",
       "2            0.0  0.0    89.0   170.0             31.0   \n",
       "3            1.0  0.0    68.0   168.0             24.0   \n",
       "4            0.0  1.0    90.0   172.0             30.0   \n",
       "5            0.0  0.0    89.0   170.0             31.0   \n",
       "6            0.0  4.0    80.0   172.0             27.0   \n",
       "7            0.0  0.0    65.0   168.0             23.0   \n",
       "8            0.0  0.0    95.0   196.0             25.0   \n",
       "9            0.0  1.0    88.0   172.0             29.0   \n",
       "\n",
       "   Absenteeism time in hours  R_Known  R_NotSerious  R_Pois_unclass  \\\n",
       "0                        4.0        0             1               0   \n",
       "1                        0.0        1             0               0   \n",
       "2                        2.0        0             1               0   \n",
       "3                        4.0        1             0               0   \n",
       "4                        2.0        0             1               0   \n",
       "5                        0.0        0             1               0   \n",
       "6                        8.0        0             1               0   \n",
       "7                        4.0        0             1               0   \n",
       "8                       40.0        0             0               1   \n",
       "9                        8.0        0             1               0   \n",
       "\n",
       "   R_Preg_Birth  \n",
       "0             0  \n",
       "1             0  \n",
       "2             0  \n",
       "3             0  \n",
       "4             0  \n",
       "5             0  \n",
       "6             0  \n",
       "7             0  \n",
       "8             0  \n",
       "9             0  \n",
       "\n",
       "[10 rows x 23 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#from the chat above we can make some assumption on how best to fill the missing data\n",
    "# 1. Transportation expense ,Age, Distance from Residence to Work,Service time, Work load Average/day,Hit \n",
    "#    target,Weight, Height,Body mass index \n",
    "#    we will fill with there mean value\n",
    "#2.  Disciplinary failure,Education,Son,Social drinker ,Social smoker, Pet ,Absenteeism time in hours  \n",
    "#    we will fill with 0\n",
    "#3.  Month of absence also have values from 0 - 12 and a missing, since we know there are only 12 month in a year, \n",
    "#    we can infer that there is an issue with the data, will deal with this later\n",
    "\n",
    "#so lets start by filling the 2nd list of variables\n",
    "data[['Month of absence','Disciplinary failure','Education','Son','Social drinker' ,'Social smoker', 'Pet' ,'Absenteeism time in hours']] = data[['Month of absence','Disciplinary failure','Education','Son','Social drinker' ,'Social smoker', 'Pet' ,'Absenteeism time in hours']].fillna(0)\n",
    "\n",
    "#then we proceed with the rest\n",
    "data = data.fillna(data.mean())\n",
    "\n",
    "data.isnull().sum()\n",
    "###\n",
    "# lets drop rows where the month is not a recognised value e.g. btw 1 and 12\n",
    "data = data[data['Month of absence']>0]\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Month of absence                   0\n",
       "Day of the week                    0\n",
       "Seasons                            0\n",
       "Transportation expense             0\n",
       "Distance from Residence to Work    0\n",
       "Service time                       0\n",
       "Age                                0\n",
       "Work load Average/day              0\n",
       "Hit target                         0\n",
       "Disciplinary failure               0\n",
       "Education                          0\n",
       "Son                                0\n",
       "Social drinker                     0\n",
       "Social smoker                      0\n",
       "Pet                                0\n",
       "Weight                             0\n",
       "Height                             0\n",
       "Body mass index                    0\n",
       "Absenteeism time in hours          0\n",
       "R_Known                            0\n",
       "R_NotSerious                       0\n",
       "R_Pois_unclass                     0\n",
       "R_Preg_Birth                       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#then we proceed with the rest\n",
    "data.isnull().sum()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>Month of absence</th>\n",
       "      <th>Day of the week</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Transportation expense</th>\n",
       "      <th>Distance from Residence to Work</th>\n",
       "      <th>Service time</th>\n",
       "      <th>Age</th>\n",
       "      <th>Work load Average/day</th>\n",
       "      <th>Hit target</th>\n",
       "      <th>Disciplinary failure</th>\n",
       "      <th>...</th>\n",
       "      <th>Social smoker</th>\n",
       "      <th>Pet</th>\n",
       "      <th>Weight</th>\n",
       "      <th>Height</th>\n",
       "      <th>Body mass index</th>\n",
       "      <th>Absenteeism time in hours</th>\n",
       "      <th>R_Known</th>\n",
       "      <th>R_NotSerious</th>\n",
       "      <th>R_Pois_unclass</th>\n",
       "      <th>R_Preg_Birth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "      <td>736.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>6.345109</td>\n",
       "      <td>3.910326</td>\n",
       "      <td>2.544837</td>\n",
       "      <td>221.276153</td>\n",
       "      <td>29.631797</td>\n",
       "      <td>12.553937</td>\n",
       "      <td>36.416233</td>\n",
       "      <td>271212.818754</td>\n",
       "      <td>94.587667</td>\n",
       "      <td>0.052989</td>\n",
       "      <td>...</td>\n",
       "      <td>0.073370</td>\n",
       "      <td>0.733696</td>\n",
       "      <td>78.998728</td>\n",
       "      <td>172.157800</td>\n",
       "      <td>26.659247</td>\n",
       "      <td>6.802989</td>\n",
       "      <td>0.311141</td>\n",
       "      <td>0.589674</td>\n",
       "      <td>0.091033</td>\n",
       "      <td>0.008152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.419102</td>\n",
       "      <td>1.420971</td>\n",
       "      <td>1.111792</td>\n",
       "      <td>66.671730</td>\n",
       "      <td>14.813973</td>\n",
       "      <td>4.387540</td>\n",
       "      <td>6.453401</td>\n",
       "      <td>38816.932786</td>\n",
       "      <td>3.786997</td>\n",
       "      <td>0.224164</td>\n",
       "      <td>...</td>\n",
       "      <td>0.260919</td>\n",
       "      <td>1.293524</td>\n",
       "      <td>12.847139</td>\n",
       "      <td>6.037074</td>\n",
       "      <td>4.189822</td>\n",
       "      <td>13.356159</td>\n",
       "      <td>0.463275</td>\n",
       "      <td>0.492227</td>\n",
       "      <td>0.287851</td>\n",
       "      <td>0.089982</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>205917.000000</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>56.000000</td>\n",
       "      <td>163.000000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>179.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>244387.000000</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>69.000000</td>\n",
       "      <td>169.000000</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>225.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>264426.500000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>81.500000</td>\n",
       "      <td>171.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>9.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>260.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>284853.000000</td>\n",
       "      <td>97.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>388.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>378884.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>108.000000</td>\n",
       "      <td>196.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>120.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Month of absence  Day of the week     Seasons  Transportation expense  \\\n",
       "count        736.000000       736.000000  736.000000              736.000000   \n",
       "mean           6.345109         3.910326    2.544837              221.276153   \n",
       "std            3.419102         1.420971    1.111792               66.671730   \n",
       "min            1.000000         2.000000    1.000000              118.000000   \n",
       "25%            3.000000         3.000000    2.000000              179.000000   \n",
       "50%            6.000000         4.000000    3.000000              225.000000   \n",
       "75%            9.000000         5.000000    4.000000              260.000000   \n",
       "max           12.000000         6.000000    4.000000              388.000000   \n",
       "\n",
       "       Distance from Residence to Work  Service time         Age  \\\n",
       "count                       736.000000    736.000000  736.000000   \n",
       "mean                         29.631797     12.553937   36.416233   \n",
       "std                          14.813973      4.387540    6.453401   \n",
       "min                           5.000000      1.000000   27.000000   \n",
       "25%                          16.000000      9.000000   31.000000   \n",
       "50%                          26.000000     13.000000   37.000000   \n",
       "75%                          50.000000     16.000000   40.000000   \n",
       "max                          52.000000     29.000000   58.000000   \n",
       "\n",
       "       Work load Average/day   Hit target  Disciplinary failure  ...  \\\n",
       "count              736.000000  736.000000            736.000000  ...   \n",
       "mean            271212.818754   94.587667              0.052989  ...   \n",
       "std              38816.932786    3.786997              0.224164  ...   \n",
       "min             205917.000000   81.000000              0.000000  ...   \n",
       "25%             244387.000000   93.000000              0.000000  ...   \n",
       "50%             264426.500000   95.000000              0.000000  ...   \n",
       "75%             284853.000000   97.000000              0.000000  ...   \n",
       "max             378884.000000  100.000000              1.000000  ...   \n",
       "\n",
       "       Social smoker         Pet      Weight      Height  Body mass index  \\\n",
       "count     736.000000  736.000000  736.000000  736.000000       736.000000   \n",
       "mean        0.073370    0.733696   78.998728  172.157800        26.659247   \n",
       "std         0.260919    1.293524   12.847139    6.037074         4.189822   \n",
       "min         0.000000    0.000000   56.000000  163.000000        19.000000   \n",
       "25%         0.000000    0.000000   69.000000  169.000000        24.000000   \n",
       "50%         0.000000    0.000000   81.500000  171.000000        25.000000   \n",
       "75%         0.000000    1.000000   89.000000  172.000000        31.000000   \n",
       "max         1.000000    8.000000  108.000000  196.000000        38.000000   \n",
       "\n",
       "       Absenteeism time in hours     R_Known  R_NotSerious  R_Pois_unclass  \\\n",
       "count                 736.000000  736.000000    736.000000      736.000000   \n",
       "mean                    6.802989    0.311141      0.589674        0.091033   \n",
       "std                    13.356159    0.463275      0.492227        0.287851   \n",
       "min                     0.000000    0.000000      0.000000        0.000000   \n",
       "25%                     2.000000    0.000000      0.000000        0.000000   \n",
       "50%                     3.000000    0.000000      1.000000        0.000000   \n",
       "75%                     8.000000    1.000000      1.000000        0.000000   \n",
       "max                   120.000000    1.000000      1.000000        1.000000   \n",
       "\n",
       "       R_Preg_Birth  \n",
       "count    736.000000  \n",
       "mean       0.008152  \n",
       "std        0.089982  \n",
       "min        0.000000  \n",
       "25%        0.000000  \n",
       "50%        0.000000  \n",
       "75%        0.000000  \n",
       "max        1.000000  \n",
       "\n",
       "[8 rows x 23 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets look at the basic description of the data\n",
    "data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create the targets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4470108695652174"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Data looks fairly OK to me. No obvious error\n",
    "# so let compute our Target which will be from 'Absenteeism time in hours'\n",
    "# This taret will be ctegorical and I have decided to use median rather than chosing an abitrary cut off, which might \n",
    "# make the data unbalanced. this reason for this is because we have very few rows of data.\n",
    "targets = np.where(data['Absenteeism time in hours'] > data['Absenteeism time in hours'].median(),1,0)\n",
    "#let do a quick check if targets is balanced in the data\n",
    "targets.sum() / targets.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# trgets looks good to me because it shows a ratio of 55/45\n",
    "\n",
    "# create a Series in the original data frame that will contain the targets for the regression\n",
    "data['Absenteeism'] = targets\n",
    "#drop the old absenteesim column\n",
    "adata = data.drop(['Absenteeism time in hours'], axis = 1)\n",
    "adata = adata.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>Month of absence</th>\n",
       "      <th>Day of the week</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Transportation expense</th>\n",
       "      <th>Distance from Residence to Work</th>\n",
       "      <th>Service time</th>\n",
       "      <th>Age</th>\n",
       "      <th>Work load Average/day</th>\n",
       "      <th>Hit target</th>\n",
       "      <th>Disciplinary failure</th>\n",
       "      <th>...</th>\n",
       "      <th>Social smoker</th>\n",
       "      <th>Pet</th>\n",
       "      <th>Weight</th>\n",
       "      <th>Height</th>\n",
       "      <th>Body mass index</th>\n",
       "      <th>R_Known</th>\n",
       "      <th>R_NotSerious</th>\n",
       "      <th>R_Pois_unclass</th>\n",
       "      <th>R_Preg_Birth</th>\n",
       "      <th>Absenteeism</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>289.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>118.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>178.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>179.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>279.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>68.0</td>\n",
       "      <td>168.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>289.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>239554.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>731</th>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>361.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>264604.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>732</th>\n",
       "      <td>7.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>225.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>264604.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>169.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>733</th>\n",
       "      <td>7.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>369.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>264604.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>70.0</td>\n",
       "      <td>169.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>734</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>289.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>264604.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>735</th>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>235.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>264604.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>172.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>736 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Month of absence  Day of the week  Seasons  Transportation expense  \\\n",
       "0                 7.0                3        1                   289.0   \n",
       "1                 7.0                3        1                   118.0   \n",
       "2                 7.0                4        1                   179.0   \n",
       "3                 7.0                5        1                   279.0   \n",
       "4                 7.0                5        1                   289.0   \n",
       "..                ...              ...      ...                     ...   \n",
       "731               7.0                4        1                   361.0   \n",
       "732               7.0                4        1                   225.0   \n",
       "733               7.0                2        1                   369.0   \n",
       "734               7.0                3        1                   289.0   \n",
       "735               7.0                3        1                   235.0   \n",
       "\n",
       "     Distance from Residence to Work  Service time   Age  \\\n",
       "0                               36.0          13.0  33.0   \n",
       "1                               13.0          18.0  50.0   \n",
       "2                               51.0          18.0  38.0   \n",
       "3                                5.0          14.0  39.0   \n",
       "4                               36.0          13.0  33.0   \n",
       "..                               ...           ...   ...   \n",
       "731                             52.0           3.0  28.0   \n",
       "732                             26.0           9.0  28.0   \n",
       "733                             17.0          12.0  31.0   \n",
       "734                             36.0          13.0  33.0   \n",
       "735                             11.0          14.0  37.0   \n",
       "\n",
       "     Work load Average/day   Hit target  Disciplinary failure  ...  \\\n",
       "0                  239554.0        97.0                   0.0  ...   \n",
       "1                  239554.0        97.0                   1.0  ...   \n",
       "2                  239554.0        97.0                   0.0  ...   \n",
       "3                  239554.0        97.0                   0.0  ...   \n",
       "4                  239554.0        97.0                   0.0  ...   \n",
       "..                      ...         ...                   ...  ...   \n",
       "731                264604.0        93.0                   0.0  ...   \n",
       "732                264604.0        93.0                   0.0  ...   \n",
       "733                264604.0        93.0                   0.0  ...   \n",
       "734                264604.0        93.0                   0.0  ...   \n",
       "735                264604.0        93.0                   0.0  ...   \n",
       "\n",
       "     Social smoker  Pet  Weight  Height  Body mass index  R_Known  \\\n",
       "0              0.0  1.0    90.0   172.0             30.0        0   \n",
       "1              0.0  0.0    98.0   178.0             31.0        1   \n",
       "2              0.0  0.0    89.0   170.0             31.0        0   \n",
       "3              1.0  0.0    68.0   168.0             24.0        1   \n",
       "4              0.0  1.0    90.0   172.0             30.0        0   \n",
       "..             ...  ...     ...     ...              ...      ...   \n",
       "731            0.0  4.0    80.0   172.0             27.0        0   \n",
       "732            0.0  2.0    69.0   169.0             24.0        0   \n",
       "733            0.0  0.0    70.0   169.0             25.0        1   \n",
       "734            0.0  1.0    90.0   172.0             30.0        1   \n",
       "735            0.0  1.0    88.0   172.0             29.0        1   \n",
       "\n",
       "     R_NotSerious  R_Pois_unclass  R_Preg_Birth  Absenteeism  \n",
       "0               1               0             0            1  \n",
       "1               0               0             0            0  \n",
       "2               1               0             0            0  \n",
       "3               0               0             0            1  \n",
       "4               1               0             0            0  \n",
       "..            ...             ...           ...          ...  \n",
       "731             1               0             0            1  \n",
       "732             1               0             0            1  \n",
       "733             0               0             0            1  \n",
       "734             0               0             0            1  \n",
       "735             0               0             0            1  \n",
       "\n",
       "[736 rows x 23 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check what happened\n",
    "# maybe manually see how the targets were created\n",
    "adata\n",
    "#targets.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a variable that will contain the inputs (everything without the targets)\n",
    "unscaled_inputs = adata.iloc[:,:-1]\n",
    "targets = adata['Absenteeism']\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##Standardise data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# define scaler as an object\n",
    "absenteeism_scaler = StandardScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create the Custom Scaler class\n",
    "\n",
    "class CustomScaler(BaseEstimator,TransformerMixin): \n",
    "    \n",
    "    # init or what information we need to declare a CustomScaler object\n",
    "    # and what is calculated/declared as we do\n",
    "    \n",
    "    def __init__(self,columns,copy=True,with_mean=True,with_std=True):\n",
    "        \n",
    "        # scaler is nothing but a Standard Scaler object\n",
    "        self.scaler = StandardScaler(copy,with_mean,with_std)\n",
    "        # with some columns 'twist'\n",
    "        self.columns = columns\n",
    "        self.mean_ = None\n",
    "        self.var_ = None\n",
    "        \n",
    "    \n",
    "    # the fit method, which, again based on StandardScale\n",
    "    \n",
    "    def fit(self, X, y=None):\n",
    "        self.scaler.fit(X[self.columns], y)\n",
    "        self.mean_ = np.mean(X[self.columns])\n",
    "        self.var_ = np.var(X[self.columns])\n",
    "        return self\n",
    "    \n",
    "    # the transform method which does the actual scaling\n",
    "\n",
    "    def transform(self, X, y=None, copy=None):\n",
    "        \n",
    "        # record the initial order of the columns\n",
    "        init_col_order = X.columns\n",
    "        \n",
    "        # scale all features that you chose when creating the instance of the class\n",
    "        X_scaled = pd.DataFrame(self.scaler.transform(X[self.columns]), columns=self.columns)\n",
    "        \n",
    "        # declare a variable containing all information that was not scaled\n",
    "        X_not_scaled = X.loc[:,~X.columns.isin(self.columns)]\n",
    "        \n",
    "        # return a data frame which contains all scaled features and all 'not scaled' features\n",
    "        # use the original order (that you recorded in the beginning)\n",
    "        return pd.concat([X_not_scaled, X_scaled], axis=1)[init_col_order]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# choose the columns to scale\n",
    "# we later augmented this code and put it in comments\n",
    "# select the columns to omit\n",
    "columns_to_omit = ['R_Known', 'R_NotSerious', 'R_Pois_unclass', 'R_Preg_Birth','Education']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create the columns to scale, based on the columns to omit\n",
    "# use list comprehension to iterate over the list\n",
    "columns_to_scale = [x for x in unscaled_inputs.columns.values if x not in columns_to_omit]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\stanl\\anaconda3\\envs\\py3-TF2.0\\lib\\site-packages\\sklearn\\base.py:197: FutureWarning: From version 0.24, get_params will raise an AttributeError if a parameter cannot be retrieved as an instance attribute. Previously it would return None.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "CustomScaler(columns=['Month of absence', 'Day of the week', 'Seasons',\n",
       "                      'Transportation expense',\n",
       "                      'Distance from Residence to Work', 'Service time', 'Age',\n",
       "                      'Work load Average/day ', 'Hit target',\n",
       "                      'Disciplinary failure', 'Son', 'Social drinker',\n",
       "                      'Social smoker', 'Pet', 'Weight', 'Height',\n",
       "                      'Body mass index'],\n",
       "             copy=None, with_mean=None, with_std=None)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# declare a scaler object, specifying the columns you want to scale\n",
    "absenteeism_scaler = CustomScaler(columns_to_scale)\n",
    "# fit the data (calculate mean and standard deviation); they are automatically stored inside the object \n",
    "absenteeism_scaler.fit(unscaled_inputs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# standardizes the data, using the transform method \n",
    "# in the last line, we fitted the data - in other words\n",
    "# we found the internal parameters of a model that will be used to transform data. \n",
    "# transforming applies these parameters to our data\n",
    "# note that when you get new data, you can just call 'scaler' again and transform it in the same way as now\n",
    "scaled_inputs = absenteeism_scaler.transform(unscaled_inputs)\n",
    "#scaled_inputs= scaled_inputs.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>Month of absence</th>\n",
       "      <th>Day of the week</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Transportation expense</th>\n",
       "      <th>Distance from Residence to Work</th>\n",
       "      <th>Service time</th>\n",
       "      <th>Age</th>\n",
       "      <th>Work load Average/day</th>\n",
       "      <th>Hit target</th>\n",
       "      <th>Disciplinary failure</th>\n",
       "      <th>...</th>\n",
       "      <th>Social drinker</th>\n",
       "      <th>Social smoker</th>\n",
       "      <th>Pet</th>\n",
       "      <th>Weight</th>\n",
       "      <th>Height</th>\n",
       "      <th>Body mass index</th>\n",
       "      <th>R_Known</th>\n",
       "      <th>R_NotSerious</th>\n",
       "      <th>R_Pois_unclass</th>\n",
       "      <th>R_Preg_Birth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>-0.641072</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>1.016471</td>\n",
       "      <td>0.430170</td>\n",
       "      <td>0.101735</td>\n",
       "      <td>-0.529729</td>\n",
       "      <td>-0.816148</td>\n",
       "      <td>0.637437</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>0.206015</td>\n",
       "      <td>0.856903</td>\n",
       "      <td>-0.026156</td>\n",
       "      <td>0.797892</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>-0.641072</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>-1.550078</td>\n",
       "      <td>-1.123474</td>\n",
       "      <td>1.242101</td>\n",
       "      <td>2.106332</td>\n",
       "      <td>-0.816148</td>\n",
       "      <td>0.637437</td>\n",
       "      <td>4.227505</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>-0.567593</td>\n",
       "      <td>1.480033</td>\n",
       "      <td>0.968379</td>\n",
       "      <td>1.036728</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>0.063150</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>-0.634525</td>\n",
       "      <td>1.443417</td>\n",
       "      <td>1.242101</td>\n",
       "      <td>0.245583</td>\n",
       "      <td>-0.816148</td>\n",
       "      <td>0.637437</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>-0.567593</td>\n",
       "      <td>0.779012</td>\n",
       "      <td>-0.357668</td>\n",
       "      <td>1.036728</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>0.767373</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>0.866381</td>\n",
       "      <td>-1.663871</td>\n",
       "      <td>0.329808</td>\n",
       "      <td>0.400645</td>\n",
       "      <td>-0.816148</td>\n",
       "      <td>0.637437</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>3.553819</td>\n",
       "      <td>-0.567593</td>\n",
       "      <td>-0.856705</td>\n",
       "      <td>-0.689179</td>\n",
       "      <td>-0.635124</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>0.767373</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>1.016471</td>\n",
       "      <td>0.430170</td>\n",
       "      <td>0.101735</td>\n",
       "      <td>-0.529729</td>\n",
       "      <td>-0.816148</td>\n",
       "      <td>0.637437</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>0.206015</td>\n",
       "      <td>0.856903</td>\n",
       "      <td>-0.026156</td>\n",
       "      <td>0.797892</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>731</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>0.063150</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>2.097124</td>\n",
       "      <td>1.510966</td>\n",
       "      <td>-2.178996</td>\n",
       "      <td>-1.305041</td>\n",
       "      <td>-0.170372</td>\n",
       "      <td>-0.419527</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>2.526839</td>\n",
       "      <td>0.077990</td>\n",
       "      <td>-0.026156</td>\n",
       "      <td>0.081384</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>732</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>0.063150</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>0.055891</td>\n",
       "      <td>-0.245327</td>\n",
       "      <td>-0.810558</td>\n",
       "      <td>-1.305041</td>\n",
       "      <td>-0.170372</td>\n",
       "      <td>-0.419527</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.137029</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>0.979623</td>\n",
       "      <td>-0.778814</td>\n",
       "      <td>-0.523424</td>\n",
       "      <td>-0.635124</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>733</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>-1.345295</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>2.217196</td>\n",
       "      <td>-0.853275</td>\n",
       "      <td>-0.126338</td>\n",
       "      <td>-0.839854</td>\n",
       "      <td>-0.170372</td>\n",
       "      <td>-0.419527</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>-0.567593</td>\n",
       "      <td>-0.700922</td>\n",
       "      <td>-0.523424</td>\n",
       "      <td>-0.396288</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>734</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>-0.641072</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>1.016471</td>\n",
       "      <td>0.430170</td>\n",
       "      <td>0.101735</td>\n",
       "      <td>-0.529729</td>\n",
       "      <td>-0.170372</td>\n",
       "      <td>-0.419527</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>0.879485</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>0.206015</td>\n",
       "      <td>0.856903</td>\n",
       "      <td>-0.026156</td>\n",
       "      <td>0.797892</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>735</th>\n",
       "      <td>0.191669</td>\n",
       "      <td>-0.641072</td>\n",
       "      <td>-1.390446</td>\n",
       "      <td>0.205982</td>\n",
       "      <td>-1.258573</td>\n",
       "      <td>0.329808</td>\n",
       "      <td>0.090520</td>\n",
       "      <td>-0.170372</td>\n",
       "      <td>-0.419527</td>\n",
       "      <td>-0.236546</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.137029</td>\n",
       "      <td>-0.281387</td>\n",
       "      <td>0.206015</td>\n",
       "      <td>0.701121</td>\n",
       "      <td>-0.026156</td>\n",
       "      <td>0.559056</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>736 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Month of absence  Day of the week   Seasons  Transportation expense  \\\n",
       "0            0.191669        -0.641072 -1.390446                1.016471   \n",
       "1            0.191669        -0.641072 -1.390446               -1.550078   \n",
       "2            0.191669         0.063150 -1.390446               -0.634525   \n",
       "3            0.191669         0.767373 -1.390446                0.866381   \n",
       "4            0.191669         0.767373 -1.390446                1.016471   \n",
       "..                ...              ...       ...                     ...   \n",
       "731          0.191669         0.063150 -1.390446                2.097124   \n",
       "732          0.191669         0.063150 -1.390446                0.055891   \n",
       "733          0.191669        -1.345295 -1.390446                2.217196   \n",
       "734          0.191669        -0.641072 -1.390446                1.016471   \n",
       "735          0.191669        -0.641072 -1.390446                0.205982   \n",
       "\n",
       "     Distance from Residence to Work  Service time       Age  \\\n",
       "0                           0.430170      0.101735 -0.529729   \n",
       "1                          -1.123474      1.242101  2.106332   \n",
       "2                           1.443417      1.242101  0.245583   \n",
       "3                          -1.663871      0.329808  0.400645   \n",
       "4                           0.430170      0.101735 -0.529729   \n",
       "..                               ...           ...       ...   \n",
       "731                         1.510966     -2.178996 -1.305041   \n",
       "732                        -0.245327     -0.810558 -1.305041   \n",
       "733                        -0.853275     -0.126338 -0.839854   \n",
       "734                         0.430170      0.101735 -0.529729   \n",
       "735                        -1.258573      0.329808  0.090520   \n",
       "\n",
       "     Work load Average/day   Hit target  Disciplinary failure  ...  \\\n",
       "0                 -0.816148    0.637437             -0.236546  ...   \n",
       "1                 -0.816148    0.637437              4.227505  ...   \n",
       "2                 -0.816148    0.637437             -0.236546  ...   \n",
       "3                 -0.816148    0.637437             -0.236546  ...   \n",
       "4                 -0.816148    0.637437             -0.236546  ...   \n",
       "..                      ...         ...                   ...  ...   \n",
       "731               -0.170372   -0.419527             -0.236546  ...   \n",
       "732               -0.170372   -0.419527             -0.236546  ...   \n",
       "733               -0.170372   -0.419527             -0.236546  ...   \n",
       "734               -0.170372   -0.419527             -0.236546  ...   \n",
       "735               -0.170372   -0.419527             -0.236546  ...   \n",
       "\n",
       "     Social drinker  Social smoker       Pet    Weight    Height  \\\n",
       "0          0.879485      -0.281387  0.206015  0.856903 -0.026156   \n",
       "1          0.879485      -0.281387 -0.567593  1.480033  0.968379   \n",
       "2          0.879485      -0.281387 -0.567593  0.779012 -0.357668   \n",
       "3          0.879485       3.553819 -0.567593 -0.856705 -0.689179   \n",
       "4          0.879485      -0.281387  0.206015  0.856903 -0.026156   \n",
       "..              ...            ...       ...       ...       ...   \n",
       "731        0.879485      -0.281387  2.526839  0.077990 -0.026156   \n",
       "732       -1.137029      -0.281387  0.979623 -0.778814 -0.523424   \n",
       "733        0.879485      -0.281387 -0.567593 -0.700922 -0.523424   \n",
       "734        0.879485      -0.281387  0.206015  0.856903 -0.026156   \n",
       "735       -1.137029      -0.281387  0.206015  0.701121 -0.026156   \n",
       "\n",
       "     Body mass index  R_Known  R_NotSerious  R_Pois_unclass  R_Preg_Birth  \n",
       "0           0.797892        0             1               0             0  \n",
       "1           1.036728        1             0               0             0  \n",
       "2           1.036728        0             1               0             0  \n",
       "3          -0.635124        1             0               0             0  \n",
       "4           0.797892        0             1               0             0  \n",
       "..               ...      ...           ...             ...           ...  \n",
       "731         0.081384        0             1               0             0  \n",
       "732        -0.635124        0             1               0             0  \n",
       "733        -0.396288        1             0               0             0  \n",
       "734         0.797892        1             0               0             0  \n",
       "735         0.559056        1             0               0             0  \n",
       "\n",
       "[736 rows x 22 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the scaled_inputs are now an ndarray, because sklearn works with ndarrays\n",
    "scaled_inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Split the data into train & test and shuffle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[     Month of absence  Day of the week   Seasons  Transportation expense  \\\n",
       " 353          1.655036         0.063150 -0.490386                0.581209   \n",
       " 331          1.362362        -1.345295  1.309734                0.371082   \n",
       " 414         -0.686351        -1.345295  0.409674                2.217196   \n",
       " 542          1.362362         0.767373  1.309734                0.055891   \n",
       " 401         -0.979024         1.471596  0.409674               -1.550078   \n",
       " ..                ...              ...       ...                     ...   \n",
       " 4            0.191669         0.767373 -1.390446                1.016471   \n",
       " 699         -0.393677         1.471596  0.409674               -0.634525   \n",
       " 64           1.069689         0.063150  1.309734               -1.550078   \n",
       " 138         -1.271697         0.063150 -0.490386               -0.634525   \n",
       " 603         -1.271697         0.063150 -0.490386               -0.634525   \n",
       " \n",
       "      Distance from Residence to Work  Service time       Age  \\\n",
       " 353                         1.375867     -0.354411 -0.064542   \n",
       " 331                        -0.312877      0.785955  0.710770   \n",
       " 414                        -0.853275     -0.126338 -0.839854   \n",
       " 542                        -0.245327     -0.810558 -1.305041   \n",
       " 401                        -1.123474      1.242101  2.106332   \n",
       " ..                               ...           ...       ...   \n",
       " 4                           0.430170      0.101735 -0.529729   \n",
       " 699                        -0.245327     -0.810558 -0.994917   \n",
       " 64                         -1.123474      1.242101  2.106332   \n",
       " 138                         1.443417      1.242101  0.245583   \n",
       " 603                        -0.245327     -0.810558 -0.994917   \n",
       " \n",
       "      Work load Average/day   Hit target  Disciplinary failure  ...  \\\n",
       " 353               -0.891553   -0.419527             -0.236546  ...   \n",
       " 331                0.330446    0.637437             -0.236546  ...   \n",
       " 414               -0.819886    0.901678             -0.236546  ...   \n",
       " 542               -0.069445   -0.419527             -0.236546  ...   \n",
       " 401               -0.691555    0.901678             -0.236546  ...   \n",
       " ..                      ...         ...                   ...  ...   \n",
       " 4                 -0.816148    0.637437             -0.236546  ...   \n",
       " 699               -0.865077    1.165919             -0.236546  ...   \n",
       " 64                -0.457529   -0.419527              4.227505  ...   \n",
       " 138                0.808758    1.165919             -0.236546  ...   \n",
       " 603               -0.179524    0.637437             -0.236546  ...   \n",
       " \n",
       "      Social drinker  Social smoker       Pet    Weight    Height  \\\n",
       " 353        0.879485      -0.281387 -0.567593 -1.090379 -0.689179   \n",
       " 331        0.879485      -0.281387 -0.567593 -0.934596 -0.357668   \n",
       " 414        0.879485      -0.281387 -0.567593 -0.700922 -0.523424   \n",
       " 542       -1.137029      -0.281387  0.979623 -0.778814 -0.523424   \n",
       " 401        0.879485      -0.281387 -0.567593  1.480033  0.968379   \n",
       " ..              ...            ...       ...       ...       ...   \n",
       " 4          0.879485      -0.281387  0.206015  0.856903 -0.026156   \n",
       " 699       -1.137029      -0.281387 -0.567593 -1.791400 -0.191912   \n",
       " 64         0.879485      -0.281387 -0.567593  1.480033  0.968379   \n",
       " 138        0.879485      -0.281387 -0.567593  0.779012 -0.357668   \n",
       " 603       -1.137029      -0.281387 -0.567593 -1.791400 -0.191912   \n",
       " \n",
       "      Body mass index  R_Known  R_NotSerious  R_Pois_unclass  R_Preg_Birth  \n",
       " 353        -0.873960        0             1               0             0  \n",
       " 331        -0.873960        1             0               0             0  \n",
       " 414        -0.396288        0             1               0             0  \n",
       " 542        -0.635124        1             0               0             0  \n",
       " 401         1.036728        0             1               0             0  \n",
       " ..               ...      ...           ...             ...           ...  \n",
       " 4           0.797892        0             1               0             0  \n",
       " 699        -1.829304        0             1               0             0  \n",
       " 64          0.005927        1             0               0             0  \n",
       " 138         1.036728        1             0               0             0  \n",
       " 603        -1.829304        0             1               0             0  \n",
       " \n",
       " [552 rows x 22 columns],\n",
       "      Month of absence  Day of the week   Seasons  Transportation expense  \\\n",
       " 652         -0.979024         0.063150  0.409674                2.217196   \n",
       " 503          0.777016        -1.345295 -1.390446               -0.634525   \n",
       " 256          0.484343         0.767373 -1.390446                1.181571   \n",
       " 176         -0.979024        -0.641072  0.409674                0.055891   \n",
       " 177         -0.979024         0.063150  0.409674                0.055891   \n",
       " ..                ...              ...       ...                     ...   \n",
       " 375         -1.271697         0.063150 -0.490386               -0.634525   \n",
       " 363         -1.564371         0.063150 -0.490386                1.046489   \n",
       " 668         -0.686351         1.471596  0.409674               -0.634525   \n",
       " 339          1.655036        -1.345295  1.309734               -0.634525   \n",
       " 378         -1.271697         1.471596 -0.490386               -0.634525   \n",
       " \n",
       "      Distance from Residence to Work  Service time       Age  \\\n",
       " 652                        -0.853275     -0.126338 -0.839854   \n",
       " 503                        -0.245327     -0.810558 -0.994917   \n",
       " 256                        -0.245327      0.101735  1.020895   \n",
       " 176                        -0.245327     -0.810558 -1.305041   \n",
       " 177                        -0.245327     -0.810558 -1.305041   \n",
       " ..                               ...           ...       ...   \n",
       " 375                         1.443417      1.242101  0.245583   \n",
       " 363                         0.092422     -0.126338  0.555708   \n",
       " 668                        -0.245327     -0.810558 -0.994917   \n",
       " 339                         1.443417      1.242101  0.245583   \n",
       " 378                         1.443417      1.242101  0.245583   \n",
       " \n",
       "      Work load Average/day   Hit target  Disciplinary failure  ...  \\\n",
       " 652               -1.263628    1.165919             -0.236546  ...   \n",
       " 503               -0.243792   -2.004973             -0.236546  ...   \n",
       " 256               -0.144309   -0.155286             -0.236546  ...   \n",
       " 176                1.857158    0.108955             -0.236546  ...   \n",
       " 177                1.857158    0.108955             -0.236546  ...   \n",
       " ..                      ...         ...                   ...  ...   \n",
       " 375               -0.499988    0.373196             -0.236546  ...   \n",
       " 363                1.517075    1.430160             -0.236546  ...   \n",
       " 668               -0.642549   -0.948009             -0.236546  ...   \n",
       " 339               -0.891553   -0.419527             -0.236546  ...   \n",
       " 378               -0.499988    0.373196             -0.236546  ...   \n",
       " \n",
       "      Social drinker  Social smoker       Pet    Weight    Height  \\\n",
       " 652        0.879485      -0.281387 -0.567593 -0.700922 -0.523424   \n",
       " 503       -1.137029      -0.281387 -0.567593 -1.791400 -0.000813   \n",
       " 256        0.879485       3.553819  0.206015 -0.155683  0.471111   \n",
       " 176       -1.137029      -0.281387  0.979623 -0.778814 -0.523424   \n",
       " 177       -1.137029      -0.281387  0.979623 -0.778814 -0.523424   \n",
       " ..              ...            ...       ...       ...       ...   \n",
       " 375        0.879485      -0.281387 -0.567593  0.779012 -0.357668   \n",
       " 363        0.879485      -0.281387  0.206015 -0.467249 -0.191912   \n",
       " 668       -1.137029      -0.281387 -0.567593 -1.791400 -0.191912   \n",
       " 339        0.879485      -0.281387 -0.567593  0.779012 -0.357668   \n",
       " 378        0.879485      -0.281387 -0.567593  0.779012 -0.357668   \n",
       " \n",
       "      Body mass index  R_Known  R_NotSerious  R_Pois_unclass  R_Preg_Birth  \n",
       " 652        -0.396288        0             1               0             0  \n",
       " 503        -1.829304        0             1               0             0  \n",
       " 256        -0.396288        1             0               0             0  \n",
       " 176        -0.635124        1             0               0             0  \n",
       " 177        -0.635124        1             0               0             0  \n",
       " ..               ...      ...           ...             ...           ...  \n",
       " 375         1.036728        0             1               0             0  \n",
       " 363        -0.396288        1             0               0             0  \n",
       " 668        -1.829304        0             1               0             0  \n",
       " 339         1.036728        0             1               0             0  \n",
       " 378         1.036728        0             1               0             0  \n",
       " \n",
       " [184 rows x 22 columns],\n",
       " 353    1\n",
       " 331    1\n",
       " 414    1\n",
       " 542    0\n",
       " 401    0\n",
       "       ..\n",
       " 4      0\n",
       " 699    0\n",
       " 64     0\n",
       " 138    1\n",
       " 603    0\n",
       " Name: Absenteeism, Length: 552, dtype: int32,\n",
       " 652    1\n",
       " 503    0\n",
       " 256    0\n",
       " 176    1\n",
       " 177    1\n",
       "       ..\n",
       " 375    0\n",
       " 363    1\n",
       " 668    0\n",
       " 339    0\n",
       " 378    0\n",
       " Name: Absenteeism, Length: 184, dtype: int32]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check how this method works\n",
    "train_test_split(scaled_inputs, targets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# declare 4 variables for the split\n",
    "x_train, x_test, y_train, y_test = train_test_split(scaled_inputs, targets, #train_size = 0.75, \n",
    "                                                                            test_size = 0.25, random_state = 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(552, 22) (552,)\n"
     ]
    }
   ],
   "source": [
    "# check the shape of the train inputs and targets\n",
    "print (x_train.shape, y_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(184, 22) (184,)\n"
     ]
    }
   ],
   "source": [
    "# check the shape of the test inputs and targets\n",
    "print (x_test.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ML Applications"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Now, we will create an array of Classifiers and append different classification models to our array\n",
    "classifiers = [] \n",
    "\n",
    "mod1 = xgboost.XGBClassifier()\n",
    "classifiers.append(mod1)\n",
    "mod2 = svm.SVC()\n",
    "classifiers.append(mod2)\n",
    "mod3 = RandomForestClassifier()\n",
    "classifiers.append(mod3)\n",
    "mod4 = LogisticRegression()\n",
    "classifiers.append(mod4)\n",
    "mod5 = KNeighborsClassifier(3)\n",
    "classifiers.append(mod5)\n",
    "mod6 = AdaBoostClassifier()\n",
    "classifiers.append(mod6)\n",
    "mod7= GaussianNB()\n",
    "classifiers.append(mod7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
      "              colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
      "              importance_type='gain', interaction_constraints='',\n",
      "              learning_rate=0.300000012, max_delta_step=0, max_depth=6,\n",
      "              min_child_weight=1, missing=nan, monotone_constraints='()',\n",
      "              n_estimators=100, n_jobs=0, num_parallel_tree=1,\n",
      "              objective='binary:logistic', random_state=0, reg_alpha=0,\n",
      "              reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
      "              tree_method='exact', validate_parameters=1, verbosity=None)\n",
      "Accuracy of trainset XGBClassifier is 0.9836956521739131\n",
      "Accuracy of testset XGBClassifier is 0.7880434782608695\n",
      "Confusion Matrix of testset XGBClassifier is [[87 22]\n",
      " [17 58]]\n",
      "SVC(C=1.0, break_ties=False, cache_size=200, class_weight=None, coef0=0.0,\n",
      "    decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',\n",
      "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "    tol=0.001, verbose=False)\n",
      "Accuracy of trainset SVC is 0.7934782608695652\n",
      "Accuracy of testset SVC is 0.8260869565217391\n",
      "Confusion Matrix of testset SVC is [[91 18]\n",
      " [14 61]]\n",
      "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,\n",
      "                       criterion='gini', max_depth=None, max_features='auto',\n",
      "                       max_leaf_nodes=None, max_samples=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
      "                       n_jobs=None, oob_score=False, random_state=None,\n",
      "                       verbose=0, warm_start=False)\n",
      "Accuracy of trainset RandomForestClassifier is 0.9836956521739131\n",
      "Accuracy of testset RandomForestClassifier is 0.7989130434782609\n",
      "Confusion Matrix of testset RandomForestClassifier is [[91 18]\n",
      " [19 56]]\n",
      "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
      "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
      "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
      "                   random_state=None, solver='lbfgs', tol=0.0001, verbose=0,\n",
      "                   warm_start=False)\n",
      "Accuracy of trainset LogisticRegression is 0.7355072463768116\n",
      "Accuracy of testset LogisticRegression is 0.7989130434782609\n",
      "Confusion Matrix of testset LogisticRegression is [[89 20]\n",
      " [17 58]]\n",
      "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "                     metric_params=None, n_jobs=None, n_neighbors=3, p=2,\n",
      "                     weights='uniform')\n",
      "Accuracy of trainset KNeighborsClassifier is 0.8170289855072463\n",
      "Accuracy of testset KNeighborsClassifier is 0.7663043478260869\n",
      "Confusion Matrix of testset KNeighborsClassifier is [[92 17]\n",
      " [26 49]]\n",
      "AdaBoostClassifier(algorithm='SAMME.R', base_estimator=None, learning_rate=1.0,\n",
      "                   n_estimators=50, random_state=None)\n",
      "Accuracy of trainset AdaBoostClassifier is 0.75\n",
      "Accuracy of testset AdaBoostClassifier is 0.8206521739130435\n",
      "Confusion Matrix of testset AdaBoostClassifier is [[95 14]\n",
      " [19 56]]\n",
      "GaussianNB(priors=None, var_smoothing=1e-09)\n",
      "Accuracy of trainset GaussianNB is 0.7083333333333334\n",
      "Accuracy of testset GaussianNB is 0.7717391304347826\n",
      "Confusion Matrix of testset GaussianNB is [[82 27]\n",
      " [15 60]]\n"
     ]
    }
   ],
   "source": [
    "#Lets fit the models into anarray\n",
    "\n",
    "for clf in classifiers:\n",
    "    clf.fit(x_train,y_train)\n",
    "    y_pred= clf.predict(x_test)\n",
    "    y_tr = clf.predict(x_train)\n",
    "    acc_tr = accuracy_score(y_train, y_tr)\n",
    "    acc = accuracy_score(y_test, y_pred)\n",
    "    mn = type(clf).__name__\n",
    "    \n",
    "    print(clf)\n",
    "    print(\"Accuracy of trainset %s is %s\"%(mn, acc_tr))\n",
    "    print(\"Accuracy of testset %s is %s\"%(mn, acc))\n",
    "    cm = confusion_matrix(y_test, y_pred)\n",
    "    print(\"Confusion Matrix of testset %s is %s\"%(mn, cm))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7934782608695652 0.8260869565217391\n"
     ]
    }
   ],
   "source": [
    "# So we stick with SVC as the best model in this case\n",
    "\n",
    "SVCclf = svm.SVC(C=1.0, break_ties=False, cache_size=200, class_weight=None, coef0=0.0,\n",
    "    decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',\n",
    "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
    "    tol=0.001, verbose=False)\n",
    "SVCclf.fit(x_train,y_train)\n",
    "ypredtrain = SVCclf.predict(x_train)\n",
    "y_pred = SVCclf.predict(x_test)\n",
    "acc_tr = accuracy_score(y_train, ypredtrain)\n",
    "acc = accuracy_score(y_test, y_pred)\n",
    "print(acc_tr, acc)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pickle the model file\n",
    "with open('model', 'wb') as file:\n",
    "    pickle.dump(SVCclf, file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Thats the end of the project, we can do abit more by tunning parameters for SVC and Adaboost by following the below process,\n",
    "## we can also look at logistic regression model by examing the odds and removing some features and retraining the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SVC parameter search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score for data1: 0.7174283374283374\n",
      "Best C: 10\n",
      "Best Kernel: linear\n",
      "Best Gamma: scale\n"
     ]
    }
   ],
   "source": [
    "param = [{'C': [1, 10, 20], 'gamma': ['scale', 'auto'], 'kernel': ['rbf','linear']}]\n",
    "\n",
    "# Create a classifier object with the classifier and parameter candidates\n",
    "svcclf = GridSearchCV(estimator=svm.SVC(), param_grid=param)\n",
    "\n",
    "# Train the classifier on data1's feature and target data\n",
    "svcclf.fit(x_train, y_train)  \n",
    "\n",
    "# View the accuracy score\n",
    "print('Best score for data1:', svcclf.best_score_) \n",
    "\n",
    "\n",
    "# View the best parameters for the model found using grid search\n",
    "print('Best C:',svcclf.best_estimator_.C) \n",
    "print('Best Kernel:',svcclf.best_estimator_.kernel)\n",
    "print('Best Gamma:',svcclf.best_estimator_.gamma)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Adaboost parameter search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7084029484029484\n",
      "{'learning_rate': 0.1, 'n_estimators': 150}\n"
     ]
    }
   ],
   "source": [
    "# inteesting result but SVC & Adaboost model seems to be doing very well, \n",
    "#So let see if we can improve on accuracy by tunning some parameters, otherwise we stick to the default parameters\n",
    "#lets start with Adaboost\n",
    "\n",
    "#Creating a grid of hyperparameters\n",
    "boost = AdaBoostClassifier(base_estimator=None)\n",
    "parameters = {'n_estimators': (50,100,150,200),\n",
    "              'learning_rate': (0.1,0.5,1, 2)}\n",
    "adab = GridSearchCV(boost, parameters)\n",
    "adab.fit(x_train, y_train)\n",
    "score = adab.best_score_\n",
    "param = adab.best_params_\n",
    "print(score)\n",
    "print(param)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logistic regression with sklearn to determine important features by examing Odds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>Feature name</th>\n",
       "      <th>Coefficient</th>\n",
       "      <th>Odds_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>R_Pois_unclass</td>\n",
       "      <td>0.769401</td>\n",
       "      <td>2.158473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>R_Known</td>\n",
       "      <td>0.500629</td>\n",
       "      <td>1.649759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Transportation expense</td>\n",
       "      <td>0.469942</td>\n",
       "      <td>1.599901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Weight</td>\n",
       "      <td>0.390210</td>\n",
       "      <td>1.477291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Son</td>\n",
       "      <td>0.378041</td>\n",
       "      <td>1.459423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Social drinker</td>\n",
       "      <td>0.360709</td>\n",
       "      <td>1.434346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Month of absence</td>\n",
       "      <td>0.199739</td>\n",
       "      <td>1.221084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Social smoker</td>\n",
       "      <td>0.144979</td>\n",
       "      <td>1.156016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Intercept</td>\n",
       "      <td>0.133547</td>\n",
       "      <td>1.142875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Work load Average/day</td>\n",
       "      <td>0.127955</td>\n",
       "      <td>1.136502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Education</td>\n",
       "      <td>0.028051</td>\n",
       "      <td>1.028448</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Hit target</td>\n",
       "      <td>0.021704</td>\n",
       "      <td>1.021941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Body mass index</td>\n",
       "      <td>-0.054167</td>\n",
       "      <td>0.947274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>R_Preg_Birth</td>\n",
       "      <td>-0.159500</td>\n",
       "      <td>0.852570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Service time</td>\n",
       "      <td>-0.172338</td>\n",
       "      <td>0.841694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Age</td>\n",
       "      <td>-0.181862</td>\n",
       "      <td>0.833716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Seasons</td>\n",
       "      <td>-0.213232</td>\n",
       "      <td>0.807969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Day of the week</td>\n",
       "      <td>-0.218986</td>\n",
       "      <td>0.803333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Pet</td>\n",
       "      <td>-0.220894</td>\n",
       "      <td>0.801802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Height</td>\n",
       "      <td>-0.227472</td>\n",
       "      <td>0.796545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Distance from Residence to Work</td>\n",
       "      <td>-0.229180</td>\n",
       "      <td>0.795185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Disciplinary failure</td>\n",
       "      <td>-1.102468</td>\n",
       "      <td>0.332051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>R_NotSerious</td>\n",
       "      <td>-1.108819</td>\n",
       "      <td>0.329948</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Feature name  Coefficient  Odds_ratio\n",
       "21                   R_Pois_unclass     0.769401    2.158473\n",
       "19                          R_Known     0.500629    1.649759\n",
       "4            Transportation expense     0.469942    1.599901\n",
       "16                           Weight     0.390210    1.477291\n",
       "12                              Son     0.378041    1.459423\n",
       "13                   Social drinker     0.360709    1.434346\n",
       "1                  Month of absence     0.199739    1.221084\n",
       "14                    Social smoker     0.144979    1.156016\n",
       "0                         Intercept     0.133547    1.142875\n",
       "8            Work load Average/day      0.127955    1.136502\n",
       "11                        Education     0.028051    1.028448\n",
       "9                        Hit target     0.021704    1.021941\n",
       "18                  Body mass index    -0.054167    0.947274\n",
       "22                     R_Preg_Birth    -0.159500    0.852570\n",
       "6                      Service time    -0.172338    0.841694\n",
       "7                               Age    -0.181862    0.833716\n",
       "3                           Seasons    -0.213232    0.807969\n",
       "2                   Day of the week    -0.218986    0.803333\n",
       "15                              Pet    -0.220894    0.801802\n",
       "17                           Height    -0.227472    0.796545\n",
       "5   Distance from Residence to Work    -0.229180    0.795185\n",
       "10             Disciplinary failure    -1.102468    0.332051\n",
       "20                     R_NotSerious    -1.108819    0.329948"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create a logistic regression object\n",
    "rr = LogisticRegression()\n",
    "# fit our train inputs\n",
    "# that is basically the whole training part of the machine learning\n",
    "rr.fit(x_train,y_train)\n",
    "rr.score(x_train,y_train)\n",
    "# save the names of the columns in an ad-hoc variable\n",
    "feature_name = unscaled_inputs.columns.values\n",
    "df = pd.DataFrame (columns=['Feature name'], data = feature_name)\n",
    "# add the coefficient values to the df\n",
    "df['Coefficient'] = np.transpose(reg.coef_)\n",
    "# move all indices by 1\n",
    "df.index = df.index + 1\n",
    "# add the intercept at index 0\n",
    "df.loc[0] = ['Intercept', reg.intercept_[0]]\n",
    "# sort the df by index\n",
    "df = df.sort_index()\n",
    "# create a new Series called: 'Odds ratio' which will show the.. odds ratio of each feature\n",
    "df['Odds_ratio'] = np.exp(df.Coefficient)\n",
    "# sort the table according to odds ratio\n",
    "df.sort_values('Odds_ratio', ascending=False)"
   ]
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}