Fraud_DetectionNotebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fraud-Detection Kaggle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, roc_auc_score\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('insuranceFraud.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>months_as_customer</th>\n",
       "      <th>age</th>\n",
       "      <th>policy_number</th>\n",
       "      <th>policy_bind_date</th>\n",
       "      <th>policy_state</th>\n",
       "      <th>policy_csl</th>\n",
       "      <th>policy_deductable</th>\n",
       "      <th>policy_annual_premium</th>\n",
       "      <th>umbrella_limit</th>\n",
       "      <th>insured_zip</th>\n",
       "      <th>...</th>\n",
       "      <th>witnesses</th>\n",
       "      <th>police_report_available</th>\n",
       "      <th>total_claim_amount</th>\n",
       "      <th>injury_claim</th>\n",
       "      <th>property_claim</th>\n",
       "      <th>vehicle_claim</th>\n",
       "      <th>auto_make</th>\n",
       "      <th>auto_model</th>\n",
       "      <th>auto_year</th>\n",
       "      <th>fraud_reported</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>328</td>\n",
       "      <td>48</td>\n",
       "      <td>521585</td>\n",
       "      <td>10/17/2014</td>\n",
       "      <td>OH</td>\n",
       "      <td>250/500</td>\n",
       "      <td>1000</td>\n",
       "      <td>1406.91</td>\n",
       "      <td>0</td>\n",
       "      <td>466132</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>YES</td>\n",
       "      <td>71610</td>\n",
       "      <td>6510</td>\n",
       "      <td>13020</td>\n",
       "      <td>52080</td>\n",
       "      <td>Saab</td>\n",
       "      <td>92x</td>\n",
       "      <td>2004</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>228</td>\n",
       "      <td>42</td>\n",
       "      <td>342868</td>\n",
       "      <td>6/27/2006</td>\n",
       "      <td>IN</td>\n",
       "      <td>250/500</td>\n",
       "      <td>2000</td>\n",
       "      <td>1197.22</td>\n",
       "      <td>5000000</td>\n",
       "      <td>468176</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>?</td>\n",
       "      <td>5070</td>\n",
       "      <td>780</td>\n",
       "      <td>780</td>\n",
       "      <td>3510</td>\n",
       "      <td>Mercedes</td>\n",
       "      <td>E400</td>\n",
       "      <td>2007</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>134</td>\n",
       "      <td>29</td>\n",
       "      <td>687698</td>\n",
       "      <td>9/6/2000</td>\n",
       "      <td>OH</td>\n",
       "      <td>100/300</td>\n",
       "      <td>2000</td>\n",
       "      <td>1413.14</td>\n",
       "      <td>5000000</td>\n",
       "      <td>430632</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>NO</td>\n",
       "      <td>34650</td>\n",
       "      <td>7700</td>\n",
       "      <td>3850</td>\n",
       "      <td>23100</td>\n",
       "      <td>Dodge</td>\n",
       "      <td>RAM</td>\n",
       "      <td>2007</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>256</td>\n",
       "      <td>41</td>\n",
       "      <td>227811</td>\n",
       "      <td>5/25/1990</td>\n",
       "      <td>IL</td>\n",
       "      <td>250/500</td>\n",
       "      <td>2000</td>\n",
       "      <td>1415.74</td>\n",
       "      <td>6000000</td>\n",
       "      <td>608117</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>NO</td>\n",
       "      <td>63400</td>\n",
       "      <td>6340</td>\n",
       "      <td>6340</td>\n",
       "      <td>50720</td>\n",
       "      <td>Chevrolet</td>\n",
       "      <td>Tahoe</td>\n",
       "      <td>2014</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>228</td>\n",
       "      <td>44</td>\n",
       "      <td>367455</td>\n",
       "      <td>6/6/2014</td>\n",
       "      <td>IL</td>\n",
       "      <td>500/1000</td>\n",
       "      <td>1000</td>\n",
       "      <td>1583.91</td>\n",
       "      <td>6000000</td>\n",
       "      <td>610706</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>NO</td>\n",
       "      <td>6500</td>\n",
       "      <td>1300</td>\n",
       "      <td>650</td>\n",
       "      <td>4550</td>\n",
       "      <td>Accura</td>\n",
       "      <td>RSX</td>\n",
       "      <td>2009</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   months_as_customer  age  policy_number policy_bind_date policy_state  \\\n",
       "0                 328   48         521585       10/17/2014           OH   \n",
       "1                 228   42         342868        6/27/2006           IN   \n",
       "2                 134   29         687698         9/6/2000           OH   \n",
       "3                 256   41         227811        5/25/1990           IL   \n",
       "4                 228   44         367455         6/6/2014           IL   \n",
       "\n",
       "  policy_csl  policy_deductable  policy_annual_premium  umbrella_limit  \\\n",
       "0    250/500               1000                1406.91               0   \n",
       "1    250/500               2000                1197.22         5000000   \n",
       "2    100/300               2000                1413.14         5000000   \n",
       "3    250/500               2000                1415.74         6000000   \n",
       "4   500/1000               1000                1583.91         6000000   \n",
       "\n",
       "   insured_zip  ... witnesses police_report_available total_claim_amount  \\\n",
       "0       466132  ...         2                     YES              71610   \n",
       "1       468176  ...         0                       ?               5070   \n",
       "2       430632  ...         3                      NO              34650   \n",
       "3       608117  ...         2                      NO              63400   \n",
       "4       610706  ...         1                      NO               6500   \n",
       "\n",
       "  injury_claim property_claim  vehicle_claim  auto_make auto_model auto_year  \\\n",
       "0         6510          13020          52080       Saab        92x      2004   \n",
       "1          780            780           3510   Mercedes       E400      2007   \n",
       "2         7700           3850          23100      Dodge        RAM      2007   \n",
       "3         6340           6340          50720  Chevrolet      Tahoe      2014   \n",
       "4         1300            650           4550     Accura        RSX      2009   \n",
       "\n",
       "  fraud_reported  \n",
       "0              Y  \n",
       "1              Y  \n",
       "2              N  \n",
       "3              Y  \n",
       "4              N  \n",
       "\n",
       "[5 rows x 39 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1000, 39)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Correct Data Format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replace \"?\" values with NAN for imputer\n",
    "data = data.replace('?', np.nan)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Drop Uselesss Columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['months_as_customer', 'age', 'policy_number', 'policy_bind_date',\n",
       "       'policy_state', 'policy_csl', 'policy_deductable',\n",
       "       'policy_annual_premium', 'umbrella_limit', 'insured_zip', 'insured_sex',\n",
       "       'insured_education_level', 'insured_occupation', 'insured_hobbies',\n",
       "       'insured_relationship', 'capital-gains', 'capital-loss',\n",
       "       'incident_date', 'incident_type', 'collision_type', 'incident_severity',\n",
       "       'authorities_contacted', 'incident_state', 'incident_city',\n",
       "       'incident_location', 'incident_hour_of_the_day',\n",
       "       'number_of_vehicles_involved', 'property_damage', 'bodily_injuries',\n",
       "       'witnesses', 'police_report_available', 'total_claim_amount',\n",
       "       'injury_claim', 'property_claim', 'vehicle_claim', 'auto_make',\n",
       "       'auto_model', 'auto_year', 'fraud_reported'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "dropable_columns = ['policy_number','policy_bind_date','policy_state','insured_zip','incident_location','incident_date','incident_state','incident_city','auto_make','auto_model','auto_year']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(dropable_columns,1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1000, 28)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Check Null values and Replace by Mode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "months_as_customer               0\n",
       "age                              0\n",
       "policy_csl                       0\n",
       "policy_deductable                0\n",
       "policy_annual_premium            0\n",
       "umbrella_limit                   0\n",
       "insured_sex                      0\n",
       "insured_education_level          0\n",
       "insured_occupation               0\n",
       "insured_hobbies                  0\n",
       "insured_relationship             0\n",
       "capital-gains                    0\n",
       "capital-loss                     0\n",
       "incident_type                    0\n",
       "collision_type                 178\n",
       "incident_severity                0\n",
       "authorities_contacted            0\n",
       "incident_hour_of_the_day         0\n",
       "number_of_vehicles_involved      0\n",
       "property_damage                360\n",
       "bodily_injuries                  0\n",
       "witnesses                        0\n",
       "police_report_available        343\n",
       "total_claim_amount               0\n",
       "injury_claim                     0\n",
       "property_claim                   0\n",
       "vehicle_claim                    0\n",
       "fraud_reported                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_mode = dict(data.mode().iloc[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.fillna(data_mode)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Data Description"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>months_as_customer</th>\n",
       "      <th>age</th>\n",
       "      <th>policy_deductable</th>\n",
       "      <th>policy_annual_premium</th>\n",
       "      <th>umbrella_limit</th>\n",
       "      <th>capital-gains</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>incident_hour_of_the_day</th>\n",
       "      <th>number_of_vehicles_involved</th>\n",
       "      <th>bodily_injuries</th>\n",
       "      <th>witnesses</th>\n",
       "      <th>total_claim_amount</th>\n",
       "      <th>injury_claim</th>\n",
       "      <th>property_claim</th>\n",
       "      <th>vehicle_claim</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1.000000e+03</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.00000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.00000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>203.954000</td>\n",
       "      <td>38.948000</td>\n",
       "      <td>1136.000000</td>\n",
       "      <td>1256.406150</td>\n",
       "      <td>1.101000e+06</td>\n",
       "      <td>25126.100000</td>\n",
       "      <td>-26793.700000</td>\n",
       "      <td>11.644000</td>\n",
       "      <td>1.83900</td>\n",
       "      <td>0.992000</td>\n",
       "      <td>1.487000</td>\n",
       "      <td>52761.94000</td>\n",
       "      <td>7433.420000</td>\n",
       "      <td>7399.570000</td>\n",
       "      <td>37928.950000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>115.113174</td>\n",
       "      <td>9.140287</td>\n",
       "      <td>611.864673</td>\n",
       "      <td>244.167395</td>\n",
       "      <td>2.297407e+06</td>\n",
       "      <td>27872.187708</td>\n",
       "      <td>28104.096686</td>\n",
       "      <td>6.951373</td>\n",
       "      <td>1.01888</td>\n",
       "      <td>0.820127</td>\n",
       "      <td>1.111335</td>\n",
       "      <td>26401.53319</td>\n",
       "      <td>4880.951853</td>\n",
       "      <td>4824.726179</td>\n",
       "      <td>18886.252893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>500.000000</td>\n",
       "      <td>433.330000</td>\n",
       "      <td>-1.000000e+06</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-111100.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>70.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>115.750000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>500.000000</td>\n",
       "      <td>1089.607500</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-51500.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>41812.50000</td>\n",
       "      <td>4295.000000</td>\n",
       "      <td>4445.000000</td>\n",
       "      <td>30292.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>199.500000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1257.200000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-23250.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>58055.00000</td>\n",
       "      <td>6775.000000</td>\n",
       "      <td>6750.000000</td>\n",
       "      <td>42100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>276.250000</td>\n",
       "      <td>44.000000</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>1415.695000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>51025.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>3.00000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>70592.50000</td>\n",
       "      <td>11305.000000</td>\n",
       "      <td>10885.000000</td>\n",
       "      <td>50822.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>479.000000</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>2047.590000</td>\n",
       "      <td>1.000000e+07</td>\n",
       "      <td>100500.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>4.00000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>114920.00000</td>\n",
       "      <td>21450.000000</td>\n",
       "      <td>23670.000000</td>\n",
       "      <td>79560.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       months_as_customer          age  policy_deductable  \\\n",
       "count         1000.000000  1000.000000        1000.000000   \n",
       "mean           203.954000    38.948000        1136.000000   \n",
       "std            115.113174     9.140287         611.864673   \n",
       "min              0.000000    19.000000         500.000000   \n",
       "25%            115.750000    32.000000         500.000000   \n",
       "50%            199.500000    38.000000        1000.000000   \n",
       "75%            276.250000    44.000000        2000.000000   \n",
       "max            479.000000    64.000000        2000.000000   \n",
       "\n",
       "       policy_annual_premium  umbrella_limit  capital-gains   capital-loss  \\\n",
       "count            1000.000000    1.000000e+03    1000.000000    1000.000000   \n",
       "mean             1256.406150    1.101000e+06   25126.100000  -26793.700000   \n",
       "std               244.167395    2.297407e+06   27872.187708   28104.096686   \n",
       "min               433.330000   -1.000000e+06       0.000000 -111100.000000   \n",
       "25%              1089.607500    0.000000e+00       0.000000  -51500.000000   \n",
       "50%              1257.200000    0.000000e+00       0.000000  -23250.000000   \n",
       "75%              1415.695000    0.000000e+00   51025.000000       0.000000   \n",
       "max              2047.590000    1.000000e+07  100500.000000       0.000000   \n",
       "\n",
       "       incident_hour_of_the_day  number_of_vehicles_involved  bodily_injuries  \\\n",
       "count               1000.000000                   1000.00000      1000.000000   \n",
       "mean                  11.644000                      1.83900         0.992000   \n",
       "std                    6.951373                      1.01888         0.820127   \n",
       "min                    0.000000                      1.00000         0.000000   \n",
       "25%                    6.000000                      1.00000         0.000000   \n",
       "50%                   12.000000                      1.00000         1.000000   \n",
       "75%                   17.000000                      3.00000         2.000000   \n",
       "max                   23.000000                      4.00000         2.000000   \n",
       "\n",
       "         witnesses  total_claim_amount  injury_claim  property_claim  \\\n",
       "count  1000.000000          1000.00000   1000.000000     1000.000000   \n",
       "mean      1.487000         52761.94000   7433.420000     7399.570000   \n",
       "std       1.111335         26401.53319   4880.951853     4824.726179   \n",
       "min       0.000000           100.00000      0.000000        0.000000   \n",
       "25%       1.000000         41812.50000   4295.000000     4445.000000   \n",
       "50%       1.000000         58055.00000   6775.000000     6750.000000   \n",
       "75%       2.000000         70592.50000  11305.000000    10885.000000   \n",
       "max       3.000000        114920.00000  21450.000000    23670.000000   \n",
       "\n",
       "       vehicle_claim  \n",
       "count    1000.000000  \n",
       "mean    37928.950000  \n",
       "std     18886.252893  \n",
       "min        70.000000  \n",
       "25%     30292.500000  \n",
       "50%     42100.000000  \n",
       "75%     50822.500000  \n",
       "max     79560.000000  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>months_as_customer</th>\n",
       "      <th>age</th>\n",
       "      <th>policy_deductable</th>\n",
       "      <th>policy_annual_premium</th>\n",
       "      <th>umbrella_limit</th>\n",
       "      <th>capital-gains</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>incident_hour_of_the_day</th>\n",
       "      <th>number_of_vehicles_involved</th>\n",
       "      <th>bodily_injuries</th>\n",
       "      <th>witnesses</th>\n",
       "      <th>total_claim_amount</th>\n",
       "      <th>injury_claim</th>\n",
       "      <th>property_claim</th>\n",
       "      <th>vehicle_claim</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>months_as_customer</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.922098</td>\n",
       "      <td>0.026807</td>\n",
       "      <td>0.005018</td>\n",
       "      <td>0.015498</td>\n",
       "      <td>0.006399</td>\n",
       "      <td>0.020209</td>\n",
       "      <td>0.070639</td>\n",
       "      <td>0.014736</td>\n",
       "      <td>-0.010162</td>\n",
       "      <td>0.058383</td>\n",
       "      <td>0.062108</td>\n",
       "      <td>0.065329</td>\n",
       "      <td>0.034940</td>\n",
       "      <td>0.061013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>age</th>\n",
       "      <td>0.922098</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.029188</td>\n",
       "      <td>0.014404</td>\n",
       "      <td>0.018126</td>\n",
       "      <td>-0.007075</td>\n",
       "      <td>0.007368</td>\n",
       "      <td>0.087161</td>\n",
       "      <td>0.022102</td>\n",
       "      <td>-0.015679</td>\n",
       "      <td>0.052359</td>\n",
       "      <td>0.069863</td>\n",
       "      <td>0.075522</td>\n",
       "      <td>0.060898</td>\n",
       "      <td>0.062588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>policy_deductable</th>\n",
       "      <td>0.026807</td>\n",
       "      <td>0.029188</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.003245</td>\n",
       "      <td>0.010870</td>\n",
       "      <td>0.035212</td>\n",
       "      <td>-0.023544</td>\n",
       "      <td>0.060935</td>\n",
       "      <td>0.051214</td>\n",
       "      <td>-0.022765</td>\n",
       "      <td>0.066639</td>\n",
       "      <td>0.022839</td>\n",
       "      <td>0.039107</td>\n",
       "      <td>0.064792</td>\n",
       "      <td>0.005269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>policy_annual_premium</th>\n",
       "      <td>0.005018</td>\n",
       "      <td>0.014404</td>\n",
       "      <td>-0.003245</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.006247</td>\n",
       "      <td>-0.013738</td>\n",
       "      <td>0.023547</td>\n",
       "      <td>-0.001578</td>\n",
       "      <td>-0.045991</td>\n",
       "      <td>0.026780</td>\n",
       "      <td>0.002332</td>\n",
       "      <td>0.009094</td>\n",
       "      <td>-0.017633</td>\n",
       "      <td>-0.011654</td>\n",
       "      <td>0.020246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>umbrella_limit</th>\n",
       "      <td>0.015498</td>\n",
       "      <td>0.018126</td>\n",
       "      <td>0.010870</td>\n",
       "      <td>-0.006247</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.047268</td>\n",
       "      <td>-0.024056</td>\n",
       "      <td>-0.023257</td>\n",
       "      <td>-0.021270</td>\n",
       "      <td>0.022743</td>\n",
       "      <td>-0.006738</td>\n",
       "      <td>-0.040344</td>\n",
       "      <td>-0.045412</td>\n",
       "      <td>-0.023790</td>\n",
       "      <td>-0.038584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>capital-gains</th>\n",
       "      <td>0.006399</td>\n",
       "      <td>-0.007075</td>\n",
       "      <td>0.035212</td>\n",
       "      <td>-0.013738</td>\n",
       "      <td>-0.047268</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.046904</td>\n",
       "      <td>-0.016406</td>\n",
       "      <td>0.061643</td>\n",
       "      <td>0.055829</td>\n",
       "      <td>-0.017651</td>\n",
       "      <td>0.015980</td>\n",
       "      <td>0.025934</td>\n",
       "      <td>-0.000779</td>\n",
       "      <td>0.015836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>capital-loss</th>\n",
       "      <td>0.020209</td>\n",
       "      <td>0.007368</td>\n",
       "      <td>-0.023544</td>\n",
       "      <td>0.023547</td>\n",
       "      <td>-0.024056</td>\n",
       "      <td>-0.046904</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.025054</td>\n",
       "      <td>-0.014895</td>\n",
       "      <td>-0.024418</td>\n",
       "      <td>-0.041330</td>\n",
       "      <td>-0.036060</td>\n",
       "      <td>-0.046060</td>\n",
       "      <td>-0.022863</td>\n",
       "      <td>-0.032665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>incident_hour_of_the_day</th>\n",
       "      <td>0.070639</td>\n",
       "      <td>0.087161</td>\n",
       "      <td>0.060935</td>\n",
       "      <td>-0.001578</td>\n",
       "      <td>-0.023257</td>\n",
       "      <td>-0.016406</td>\n",
       "      <td>-0.025054</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.120794</td>\n",
       "      <td>-0.034563</td>\n",
       "      <td>0.006527</td>\n",
       "      <td>0.217702</td>\n",
       "      <td>0.165768</td>\n",
       "      <td>0.179536</td>\n",
       "      <td>0.215626</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number_of_vehicles_involved</th>\n",
       "      <td>0.014736</td>\n",
       "      <td>0.022102</td>\n",
       "      <td>0.051214</td>\n",
       "      <td>-0.045991</td>\n",
       "      <td>-0.021270</td>\n",
       "      <td>0.061643</td>\n",
       "      <td>-0.014895</td>\n",
       "      <td>0.120794</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.014030</td>\n",
       "      <td>-0.014669</td>\n",
       "      <td>0.274278</td>\n",
       "      <td>0.224650</td>\n",
       "      <td>0.219084</td>\n",
       "      <td>0.269393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bodily_injuries</th>\n",
       "      <td>-0.010162</td>\n",
       "      <td>-0.015679</td>\n",
       "      <td>-0.022765</td>\n",
       "      <td>0.026780</td>\n",
       "      <td>0.022743</td>\n",
       "      <td>0.055829</td>\n",
       "      <td>-0.024418</td>\n",
       "      <td>-0.034563</td>\n",
       "      <td>0.014030</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.005606</td>\n",
       "      <td>0.047093</td>\n",
       "      <td>0.047319</td>\n",
       "      <td>0.039749</td>\n",
       "      <td>0.043449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>witnesses</th>\n",
       "      <td>0.058383</td>\n",
       "      <td>0.052359</td>\n",
       "      <td>0.066639</td>\n",
       "      <td>0.002332</td>\n",
       "      <td>-0.006738</td>\n",
       "      <td>-0.017651</td>\n",
       "      <td>-0.041330</td>\n",
       "      <td>0.006527</td>\n",
       "      <td>-0.014669</td>\n",
       "      <td>-0.005606</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.011114</td>\n",
       "      <td>-0.024843</td>\n",
       "      <td>0.052640</td>\n",
       "      <td>-0.022564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_claim_amount</th>\n",
       "      <td>0.062108</td>\n",
       "      <td>0.069863</td>\n",
       "      <td>0.022839</td>\n",
       "      <td>0.009094</td>\n",
       "      <td>-0.040344</td>\n",
       "      <td>0.015980</td>\n",
       "      <td>-0.036060</td>\n",
       "      <td>0.217702</td>\n",
       "      <td>0.274278</td>\n",
       "      <td>0.047093</td>\n",
       "      <td>-0.011114</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.805025</td>\n",
       "      <td>0.810686</td>\n",
       "      <td>0.982773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>injury_claim</th>\n",
       "      <td>0.065329</td>\n",
       "      <td>0.075522</td>\n",
       "      <td>0.039107</td>\n",
       "      <td>-0.017633</td>\n",
       "      <td>-0.045412</td>\n",
       "      <td>0.025934</td>\n",
       "      <td>-0.046060</td>\n",
       "      <td>0.165768</td>\n",
       "      <td>0.224650</td>\n",
       "      <td>0.047319</td>\n",
       "      <td>-0.024843</td>\n",
       "      <td>0.805025</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.563866</td>\n",
       "      <td>0.722878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>property_claim</th>\n",
       "      <td>0.034940</td>\n",
       "      <td>0.060898</td>\n",
       "      <td>0.064792</td>\n",
       "      <td>-0.011654</td>\n",
       "      <td>-0.023790</td>\n",
       "      <td>-0.000779</td>\n",
       "      <td>-0.022863</td>\n",
       "      <td>0.179536</td>\n",
       "      <td>0.219084</td>\n",
       "      <td>0.039749</td>\n",
       "      <td>0.052640</td>\n",
       "      <td>0.810686</td>\n",
       "      <td>0.563866</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.732090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vehicle_claim</th>\n",
       "      <td>0.061013</td>\n",
       "      <td>0.062588</td>\n",
       "      <td>0.005269</td>\n",
       "      <td>0.020246</td>\n",
       "      <td>-0.038584</td>\n",
       "      <td>0.015836</td>\n",
       "      <td>-0.032665</td>\n",
       "      <td>0.215626</td>\n",
       "      <td>0.269393</td>\n",
       "      <td>0.043449</td>\n",
       "      <td>-0.022564</td>\n",
       "      <td>0.982773</td>\n",
       "      <td>0.722878</td>\n",
       "      <td>0.732090</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             months_as_customer       age  policy_deductable  \\\n",
       "months_as_customer                     1.000000  0.922098           0.026807   \n",
       "age                                    0.922098  1.000000           0.029188   \n",
       "policy_deductable                      0.026807  0.029188           1.000000   \n",
       "policy_annual_premium                  0.005018  0.014404          -0.003245   \n",
       "umbrella_limit                         0.015498  0.018126           0.010870   \n",
       "capital-gains                          0.006399 -0.007075           0.035212   \n",
       "capital-loss                           0.020209  0.007368          -0.023544   \n",
       "incident_hour_of_the_day               0.070639  0.087161           0.060935   \n",
       "number_of_vehicles_involved            0.014736  0.022102           0.051214   \n",
       "bodily_injuries                       -0.010162 -0.015679          -0.022765   \n",
       "witnesses                              0.058383  0.052359           0.066639   \n",
       "total_claim_amount                     0.062108  0.069863           0.022839   \n",
       "injury_claim                           0.065329  0.075522           0.039107   \n",
       "property_claim                         0.034940  0.060898           0.064792   \n",
       "vehicle_claim                          0.061013  0.062588           0.005269   \n",
       "\n",
       "                             policy_annual_premium  umbrella_limit  \\\n",
       "months_as_customer                        0.005018        0.015498   \n",
       "age                                       0.014404        0.018126   \n",
       "policy_deductable                        -0.003245        0.010870   \n",
       "policy_annual_premium                     1.000000       -0.006247   \n",
       "umbrella_limit                           -0.006247        1.000000   \n",
       "capital-gains                            -0.013738       -0.047268   \n",
       "capital-loss                              0.023547       -0.024056   \n",
       "incident_hour_of_the_day                 -0.001578       -0.023257   \n",
       "number_of_vehicles_involved              -0.045991       -0.021270   \n",
       "bodily_injuries                           0.026780        0.022743   \n",
       "witnesses                                 0.002332       -0.006738   \n",
       "total_claim_amount                        0.009094       -0.040344   \n",
       "injury_claim                             -0.017633       -0.045412   \n",
       "property_claim                           -0.011654       -0.023790   \n",
       "vehicle_claim                             0.020246       -0.038584   \n",
       "\n",
       "                             capital-gains  capital-loss  \\\n",
       "months_as_customer                0.006399      0.020209   \n",
       "age                              -0.007075      0.007368   \n",
       "policy_deductable                 0.035212     -0.023544   \n",
       "policy_annual_premium            -0.013738      0.023547   \n",
       "umbrella_limit                   -0.047268     -0.024056   \n",
       "capital-gains                     1.000000     -0.046904   \n",
       "capital-loss                     -0.046904      1.000000   \n",
       "incident_hour_of_the_day         -0.016406     -0.025054   \n",
       "number_of_vehicles_involved       0.061643     -0.014895   \n",
       "bodily_injuries                   0.055829     -0.024418   \n",
       "witnesses                        -0.017651     -0.041330   \n",
       "total_claim_amount                0.015980     -0.036060   \n",
       "injury_claim                      0.025934     -0.046060   \n",
       "property_claim                   -0.000779     -0.022863   \n",
       "vehicle_claim                     0.015836     -0.032665   \n",
       "\n",
       "                             incident_hour_of_the_day  \\\n",
       "months_as_customer                           0.070639   \n",
       "age                                          0.087161   \n",
       "policy_deductable                            0.060935   \n",
       "policy_annual_premium                       -0.001578   \n",
       "umbrella_limit                              -0.023257   \n",
       "capital-gains                               -0.016406   \n",
       "capital-loss                                -0.025054   \n",
       "incident_hour_of_the_day                     1.000000   \n",
       "number_of_vehicles_involved                  0.120794   \n",
       "bodily_injuries                             -0.034563   \n",
       "witnesses                                    0.006527   \n",
       "total_claim_amount                           0.217702   \n",
       "injury_claim                                 0.165768   \n",
       "property_claim                               0.179536   \n",
       "vehicle_claim                                0.215626   \n",
       "\n",
       "                             number_of_vehicles_involved  bodily_injuries  \\\n",
       "months_as_customer                              0.014736        -0.010162   \n",
       "age                                             0.022102        -0.015679   \n",
       "policy_deductable                               0.051214        -0.022765   \n",
       "policy_annual_premium                          -0.045991         0.026780   \n",
       "umbrella_limit                                 -0.021270         0.022743   \n",
       "capital-gains                                   0.061643         0.055829   \n",
       "capital-loss                                   -0.014895        -0.024418   \n",
       "incident_hour_of_the_day                        0.120794        -0.034563   \n",
       "number_of_vehicles_involved                     1.000000         0.014030   \n",
       "bodily_injuries                                 0.014030         1.000000   \n",
       "witnesses                                      -0.014669        -0.005606   \n",
       "total_claim_amount                              0.274278         0.047093   \n",
       "injury_claim                                    0.224650         0.047319   \n",
       "property_claim                                  0.219084         0.039749   \n",
       "vehicle_claim                                   0.269393         0.043449   \n",
       "\n",
       "                             witnesses  total_claim_amount  injury_claim  \\\n",
       "months_as_customer            0.058383            0.062108      0.065329   \n",
       "age                           0.052359            0.069863      0.075522   \n",
       "policy_deductable             0.066639            0.022839      0.039107   \n",
       "policy_annual_premium         0.002332            0.009094     -0.017633   \n",
       "umbrella_limit               -0.006738           -0.040344     -0.045412   \n",
       "capital-gains                -0.017651            0.015980      0.025934   \n",
       "capital-loss                 -0.041330           -0.036060     -0.046060   \n",
       "incident_hour_of_the_day      0.006527            0.217702      0.165768   \n",
       "number_of_vehicles_involved  -0.014669            0.274278      0.224650   \n",
       "bodily_injuries              -0.005606            0.047093      0.047319   \n",
       "witnesses                     1.000000           -0.011114     -0.024843   \n",
       "total_claim_amount           -0.011114            1.000000      0.805025   \n",
       "injury_claim                 -0.024843            0.805025      1.000000   \n",
       "property_claim                0.052640            0.810686      0.563866   \n",
       "vehicle_claim                -0.022564            0.982773      0.722878   \n",
       "\n",
       "                             property_claim  vehicle_claim  \n",
       "months_as_customer                 0.034940       0.061013  \n",
       "age                                0.060898       0.062588  \n",
       "policy_deductable                  0.064792       0.005269  \n",
       "policy_annual_premium             -0.011654       0.020246  \n",
       "umbrella_limit                    -0.023790      -0.038584  \n",
       "capital-gains                     -0.000779       0.015836  \n",
       "capital-loss                      -0.022863      -0.032665  \n",
       "incident_hour_of_the_day           0.179536       0.215626  \n",
       "number_of_vehicles_involved        0.219084       0.269393  \n",
       "bodily_injuries                    0.039749       0.043449  \n",
       "witnesses                          0.052640      -0.022564  \n",
       "total_claim_amount                 0.810686       0.982773  \n",
       "injury_claim                       0.563866       0.722878  \n",
       "property_claim                     1.000000       0.732090  \n",
       "vehicle_claim                      0.732090       1.000000  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Graph Plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,15))\n",
    "ax = sns.heatmap(data.corr(),linewidths=.5,square=True,center=0,annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['months_as_customer', 'age', 'policy_csl', 'policy_deductable',\n",
       "       'policy_annual_premium', 'umbrella_limit', 'insured_sex',\n",
       "       'insured_education_level', 'insured_occupation', 'insured_hobbies',\n",
       "       'insured_relationship', 'capital-gains', 'capital-loss',\n",
       "       'incident_type', 'collision_type', 'incident_severity',\n",
       "       'authorities_contacted', 'incident_hour_of_the_day',\n",
       "       'number_of_vehicles_involved', 'property_damage', 'bodily_injuries',\n",
       "       'witnesses', 'police_report_available', 'total_claim_amount',\n",
       "       'injury_claim', 'property_claim', 'vehicle_claim', 'fraud_reported'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f47eeed8c50>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, ax = plt.subplots(figsize=(18, 8))\n",
    "sns.countplot(x='insured_hobbies',hue='fraud_reported',data=data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###### Check Data is Balanced Or Not\n",
    "> Data has Imbalanced Classses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f47ef54dfd0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x='fraud_reported',data=data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Label Encoder and OneHotEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_df = data.select_dtypes(include=['object']).copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_df = data.select_dtypes(include=[\"int64\"]).copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>policy_csl</th>\n",
       "      <th>insured_sex</th>\n",
       "      <th>insured_education_level</th>\n",
       "      <th>insured_occupation</th>\n",
       "      <th>insured_hobbies</th>\n",
       "      <th>insured_relationship</th>\n",
       "      <th>incident_type</th>\n",
       "      <th>collision_type</th>\n",
       "      <th>incident_severity</th>\n",
       "      <th>authorities_contacted</th>\n",
       "      <th>property_damage</th>\n",
       "      <th>police_report_available</th>\n",
       "      <th>fraud_reported</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>250/500</td>\n",
       "      <td>MALE</td>\n",
       "      <td>MD</td>\n",
       "      <td>craft-repair</td>\n",
       "      <td>sleeping</td>\n",
       "      <td>husband</td>\n",
       "      <td>Single Vehicle Collision</td>\n",
       "      <td>Side Collision</td>\n",
       "      <td>Major Damage</td>\n",
       "      <td>Police</td>\n",
       "      <td>YES</td>\n",
       "      <td>YES</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>250/500</td>\n",
       "      <td>MALE</td>\n",
       "      <td>MD</td>\n",
       "      <td>machine-op-inspct</td>\n",
       "      <td>reading</td>\n",
       "      <td>other-relative</td>\n",
       "      <td>Vehicle Theft</td>\n",
       "      <td>Rear Collision</td>\n",
       "      <td>Minor Damage</td>\n",
       "      <td>Police</td>\n",
       "      <td>NO</td>\n",
       "      <td>NO</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>100/300</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>PhD</td>\n",
       "      <td>sales</td>\n",
       "      <td>board-games</td>\n",
       "      <td>own-child</td>\n",
       "      <td>Multi-vehicle Collision</td>\n",
       "      <td>Rear Collision</td>\n",
       "      <td>Minor Damage</td>\n",
       "      <td>Police</td>\n",
       "      <td>NO</td>\n",
       "      <td>NO</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>250/500</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>PhD</td>\n",
       "      <td>armed-forces</td>\n",
       "      <td>board-games</td>\n",
       "      <td>unmarried</td>\n",
       "      <td>Single Vehicle Collision</td>\n",
       "      <td>Front Collision</td>\n",
       "      <td>Major Damage</td>\n",
       "      <td>Police</td>\n",
       "      <td>NO</td>\n",
       "      <td>NO</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500/1000</td>\n",
       "      <td>MALE</td>\n",
       "      <td>Associate</td>\n",
       "      <td>sales</td>\n",
       "      <td>board-games</td>\n",
       "      <td>unmarried</td>\n",
       "      <td>Vehicle Theft</td>\n",
       "      <td>Rear Collision</td>\n",
       "      <td>Minor Damage</td>\n",
       "      <td>None</td>\n",
       "      <td>NO</td>\n",
       "      <td>NO</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  policy_csl insured_sex insured_education_level insured_occupation  \\\n",
       "0    250/500        MALE                      MD       craft-repair   \n",
       "1    250/500        MALE                      MD  machine-op-inspct   \n",
       "2    100/300      FEMALE                     PhD              sales   \n",
       "3    250/500      FEMALE                     PhD       armed-forces   \n",
       "4   500/1000        MALE               Associate              sales   \n",
       "\n",
       "  insured_hobbies insured_relationship             incident_type  \\\n",
       "0        sleeping              husband  Single Vehicle Collision   \n",
       "1         reading       other-relative             Vehicle Theft   \n",
       "2     board-games            own-child   Multi-vehicle Collision   \n",
       "3     board-games            unmarried  Single Vehicle Collision   \n",
       "4     board-games            unmarried             Vehicle Theft   \n",
       "\n",
       "    collision_type incident_severity authorities_contacted property_damage  \\\n",
       "0   Side Collision      Major Damage                Police             YES   \n",
       "1   Rear Collision      Minor Damage                Police              NO   \n",
       "2   Rear Collision      Minor Damage                Police              NO   \n",
       "3  Front Collision      Major Damage                Police              NO   \n",
       "4   Rear Collision      Minor Damage                  None              NO   \n",
       "\n",
       "  police_report_available fraud_reported  \n",
       "0                     YES              Y  \n",
       "1                      NO              Y  \n",
       "2                      NO              N  \n",
       "3                      NO              Y  \n",
       "4                      NO              N  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['policy_csl', 'insured_sex', 'insured_education_level',\n",
       "       'insured_occupation', 'insured_hobbies', 'insured_relationship',\n",
       "       'incident_type', 'collision_type', 'incident_severity',\n",
       "       'authorities_contacted', 'property_damage', 'police_report_available',\n",
       "       'fraud_reported'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "for column in cat_df.columns:\n",
    "    le = LabelEncoder()\n",
    "    cat_df[column] = le.fit_transform(cat_df[column])"
   ]
  },
  {
   "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>policy_csl</th>\n",
       "      <th>insured_sex</th>\n",
       "      <th>insured_education_level</th>\n",
       "      <th>insured_occupation</th>\n",
       "      <th>insured_hobbies</th>\n",
       "      <th>insured_relationship</th>\n",
       "      <th>incident_type</th>\n",
       "      <th>collision_type</th>\n",
       "      <th>incident_severity</th>\n",
       "      <th>authorities_contacted</th>\n",
       "      <th>property_damage</th>\n",
       "      <th>police_report_available</th>\n",
       "      <th>fraud_reported</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   policy_csl  insured_sex  insured_education_level  insured_occupation  \\\n",
       "0           1            1                        4                   2   \n",
       "1           1            1                        4                   6   \n",
       "2           0            0                        6                  11   \n",
       "3           1            0                        6                   1   \n",
       "4           2            1                        0                  11   \n",
       "\n",
       "   insured_hobbies  insured_relationship  incident_type  collision_type  \\\n",
       "0               17                     0              2               2   \n",
       "1               15                     2              3               1   \n",
       "2                2                     3              0               1   \n",
       "3                2                     4              2               0   \n",
       "4                2                     4              3               1   \n",
       "\n",
       "   incident_severity  authorities_contacted  property_damage  \\\n",
       "0                  0                      4                1   \n",
       "1                  1                      4                0   \n",
       "2                  1                      4                0   \n",
       "3                  0                      4                0   \n",
       "4                  1                      2                0   \n",
       "\n",
       "   police_report_available  fraud_reported  \n",
       "0                        1               1  \n",
       "1                        0               1  \n",
       "2                        0               0  \n",
       "3                        0               1  \n",
       "4                        0               0  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>months_as_customer</th>\n",
       "      <th>age</th>\n",
       "      <th>policy_deductable</th>\n",
       "      <th>umbrella_limit</th>\n",
       "      <th>capital-gains</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>incident_hour_of_the_day</th>\n",
       "      <th>number_of_vehicles_involved</th>\n",
       "      <th>bodily_injuries</th>\n",
       "      <th>witnesses</th>\n",
       "      <th>total_claim_amount</th>\n",
       "      <th>injury_claim</th>\n",
       "      <th>property_claim</th>\n",
       "      <th>vehicle_claim</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>328</td>\n",
       "      <td>48</td>\n",
       "      <td>1000</td>\n",
       "      <td>0</td>\n",
       "      <td>53300</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>71610</td>\n",
       "      <td>6510</td>\n",
       "      <td>13020</td>\n",
       "      <td>52080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>228</td>\n",
       "      <td>42</td>\n",
       "      <td>2000</td>\n",
       "      <td>5000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5070</td>\n",
       "      <td>780</td>\n",
       "      <td>780</td>\n",
       "      <td>3510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>134</td>\n",
       "      <td>29</td>\n",
       "      <td>2000</td>\n",
       "      <td>5000000</td>\n",
       "      <td>35100</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>34650</td>\n",
       "      <td>7700</td>\n",
       "      <td>3850</td>\n",
       "      <td>23100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>256</td>\n",
       "      <td>41</td>\n",
       "      <td>2000</td>\n",
       "      <td>6000000</td>\n",
       "      <td>48900</td>\n",
       "      <td>-62400</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>63400</td>\n",
       "      <td>6340</td>\n",
       "      <td>6340</td>\n",
       "      <td>50720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>228</td>\n",
       "      <td>44</td>\n",
       "      <td>1000</td>\n",
       "      <td>6000000</td>\n",
       "      <td>66000</td>\n",
       "      <td>-46000</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>6500</td>\n",
       "      <td>1300</td>\n",
       "      <td>650</td>\n",
       "      <td>4550</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   months_as_customer  age  policy_deductable  umbrella_limit  capital-gains  \\\n",
       "0                 328   48               1000               0          53300   \n",
       "1                 228   42               2000         5000000              0   \n",
       "2                 134   29               2000         5000000          35100   \n",
       "3                 256   41               2000         6000000          48900   \n",
       "4                 228   44               1000         6000000          66000   \n",
       "\n",
       "   capital-loss  incident_hour_of_the_day  number_of_vehicles_involved  \\\n",
       "0             0                         5                            1   \n",
       "1             0                         8                            1   \n",
       "2             0                         7                            3   \n",
       "3        -62400                         5                            1   \n",
       "4        -46000                        20                            1   \n",
       "\n",
       "   bodily_injuries  witnesses  total_claim_amount  injury_claim  \\\n",
       "0                1          2               71610          6510   \n",
       "1                0          0                5070           780   \n",
       "2                2          3               34650          7700   \n",
       "3                1          2               63400          6340   \n",
       "4                0          1                6500          1300   \n",
       "\n",
       "   property_claim  vehicle_claim  \n",
       "0           13020          52080  \n",
       "1             780           3510  \n",
       "2            3850          23100  \n",
       "3            6340          50720  \n",
       "4             650           4550  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_df = pd.concat([num_df,cat_df], axis=1)"
   ]
  },
  {
   "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>months_as_customer</th>\n",
       "      <th>age</th>\n",
       "      <th>policy_deductable</th>\n",
       "      <th>umbrella_limit</th>\n",
       "      <th>capital-gains</th>\n",
       "      <th>capital-loss</th>\n",
       "      <th>incident_hour_of_the_day</th>\n",
       "      <th>number_of_vehicles_involved</th>\n",
       "      <th>bodily_injuries</th>\n",
       "      <th>witnesses</th>\n",
       "      <th>...</th>\n",
       "      <th>insured_occupation</th>\n",
       "      <th>insured_hobbies</th>\n",
       "      <th>insured_relationship</th>\n",
       "      <th>incident_type</th>\n",
       "      <th>collision_type</th>\n",
       "      <th>incident_severity</th>\n",
       "      <th>authorities_contacted</th>\n",
       "      <th>property_damage</th>\n",
       "      <th>police_report_available</th>\n",
       "      <th>fraud_reported</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>328</td>\n",
       "      <td>48</td>\n",
       "      <td>1000</td>\n",
       "      <td>0</td>\n",
       "      <td>53300</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>228</td>\n",
       "      <td>42</td>\n",
       "      <td>2000</td>\n",
       "      <td>5000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>134</td>\n",
       "      <td>29</td>\n",
       "      <td>2000</td>\n",
       "      <td>5000000</td>\n",
       "      <td>35100</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>256</td>\n",
       "      <td>41</td>\n",
       "      <td>2000</td>\n",
       "      <td>6000000</td>\n",
       "      <td>48900</td>\n",
       "      <td>-62400</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>228</td>\n",
       "      <td>44</td>\n",
       "      <td>1000</td>\n",
       "      <td>6000000</td>\n",
       "      <td>66000</td>\n",
       "      <td>-46000</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   months_as_customer  age  policy_deductable  umbrella_limit  capital-gains  \\\n",
       "0                 328   48               1000               0          53300   \n",
       "1                 228   42               2000         5000000              0   \n",
       "2                 134   29               2000         5000000          35100   \n",
       "3                 256   41               2000         6000000          48900   \n",
       "4                 228   44               1000         6000000          66000   \n",
       "\n",
       "   capital-loss  incident_hour_of_the_day  number_of_vehicles_involved  \\\n",
       "0             0                         5                            1   \n",
       "1             0                         8                            1   \n",
       "2             0                         7                            3   \n",
       "3        -62400                         5                            1   \n",
       "4        -46000                        20                            1   \n",
       "\n",
       "   bodily_injuries  witnesses  ...  insured_occupation  insured_hobbies  \\\n",
       "0                1          2  ...                   2               17   \n",
       "1                0          0  ...                   6               15   \n",
       "2                2          3  ...                  11                2   \n",
       "3                1          2  ...                   1                2   \n",
       "4                0          1  ...                  11                2   \n",
       "\n",
       "   insured_relationship  incident_type  collision_type  incident_severity  \\\n",
       "0                     0              2               2                  0   \n",
       "1                     2              3               1                  1   \n",
       "2                     3              0               1                  1   \n",
       "3                     4              2               0                  0   \n",
       "4                     4              3               1                  1   \n",
       "\n",
       "   authorities_contacted  property_damage  police_report_available  \\\n",
       "0                      4                1                        1   \n",
       "1                      4                0                        0   \n",
       "2                      4                0                        0   \n",
       "3                      4                0                        0   \n",
       "4                      2                0                        0   \n",
       "\n",
       "   fraud_reported  \n",
       "0               1  \n",
       "1               1  \n",
       "2               0  \n",
       "3               1  \n",
       "4               0  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_value = 'fraud_reported'\n",
    "X = final_df.drop([target_value],1)\n",
    "y = final_df[target_value]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split data into train and test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_X, test_X , train_y, test_y = train_test_split(X, y, test_size =0.3, random_state = 2020)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train The Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = RandomForestClassifier(n_estimators=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_y = clf.fit(train_X, train_y).predict(test_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.74"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(test_y, pred_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[199,  29],\n",
       "       [ 49,  23]])"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(test_y, pred_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.596125730994152"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(test_y, pred_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#####  Roc Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, threshold = metrics.roc_curve(test_y, pred_y)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)\n",
    "plt.legend(loc = 'lower right')\n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.xlim([0, 1])\n",
    "plt.ylim([0, 1])\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Support Vector Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf2 = SVC()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred2_y = clf2.fit(train_X, train_y).predict(test_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.76"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(test_y, pred2_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[228,   0],\n",
       "       [ 72,   0]])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(test_y, pred2_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(test_y, pred2_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### GridSearchCv For Better Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = {\"kernel\": ['rbf','sigmoid'],\n",
    "             \"C\":[0.1,0.5,1.0],\n",
    "             \"random_state\":[0,10,20,30]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid = GridSearchCV(estimator=clf2, param_grid=param_grid, cv=5,  verbose=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 24 candidates, totalling 120 fits\n",
      "[CV] C=0.1, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.1, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.1, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.1, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=0 ...............................\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] ... C=0.1, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.1, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=20, score=0.750, total=   0.1s\n",
      "[CV] C=0.1, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=20, score=0.750, total=   0.1s\n",
      "[CV] C=0.1, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.1, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=10, score=0.750, total=   0.1s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.1, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.1, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.5, kernel=rbf, random_state=0, score=0.750, total=   0.1s\n",
      "[CV] C=0.5, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.5, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.5, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.5, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=0.5, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=20, score=0.750, total=   0.1s\n",
      "[CV] C=0.5, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=0.5, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=0, score=0.743, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=10 ..........................\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  C=0.5, kernel=sigmoid, random_state=10, score=0.743, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=20, score=0.743, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=20, score=0.750, total=   0.1s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=30, score=0.743, total=   0.1s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=30, score=0.750, total=   0.1s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=30, score=0.750, total=   0.1s\n",
      "[CV] C=0.5, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=0.5, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=1.0, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=1.0, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=1.0, kernel=rbf, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=1.0, kernel=rbf, random_state=0, score=0.750, total=   0.1s\n",
      "[CV] C=1.0, kernel=rbf, random_state=0 ...............................\n",
      "[CV] ... C=1.0, kernel=rbf, random_state=0, score=0.750, total=   0.1s\n",
      "[CV] C=1.0, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=10, score=0.750, total=   0.1s\n",
      "[CV] C=1.0, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=10 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=10, score=0.750, total=   0.1s\n",
      "[CV] C=1.0, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=20, score=0.750, total=   0.1s\n",
      "[CV] C=1.0, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=20 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=rbf, random_state=30 ..............................\n",
      "[CV] .. C=1.0, kernel=rbf, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=0, score=0.743, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=0, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=0 ...........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=0, score=0.757, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=10, score=0.743, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=10, score=0.750, total=   0.1s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=10, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=10 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=10, score=0.757, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=20, score=0.743, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=20, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=20 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=20, score=0.757, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=30, score=0.743, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=30, score=0.750, total=   0.0s\n",
      "[CV] C=1.0, kernel=sigmoid, random_state=30 ..........................\n",
      "[CV]  C=1.0, kernel=sigmoid, random_state=30, score=0.757, total=   0.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done 120 out of 120 | elapsed:    5.8s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score=nan,\n",
       "             estimator=SVC(C=1.0, break_ties=False, cache_size=200,\n",
       "                           class_weight=None, coef0=0.0,\n",
       "                           decision_function_shape='ovr', degree=3,\n",
       "                           gamma='scale', kernel='rbf', max_iter=-1,\n",
       "                           probability=False, random_state=None, shrinking=True,\n",
       "                           tol=0.001, verbose=False),\n",
       "             iid='deprecated', n_jobs=None,\n",
       "             param_grid={'C': [0.1, 0.5, 1.0], 'kernel': ['rbf', 'sigmoid'],\n",
       "                         'random_state': [0, 10, 20, 30]},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring=None, verbose=3)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=0.1, 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=0, shrinking=True, tol=0.001,\n",
       "    verbose=False)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}