Classification using MLP and kernel trick.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploring the Data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "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>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Customer Type</th>\n",
       "      <th>Age</th>\n",
       "      <th>Type of Travel</th>\n",
       "      <th>Class</th>\n",
       "      <th>Flight Distance</th>\n",
       "      <th>Inflight wifi service</th>\n",
       "      <th>Departure/Arrival time convenient</th>\n",
       "      <th>...</th>\n",
       "      <th>Inflight entertainment</th>\n",
       "      <th>On-board service</th>\n",
       "      <th>Leg room service</th>\n",
       "      <th>Baggage handling</th>\n",
       "      <th>Checkin service</th>\n",
       "      <th>Inflight service</th>\n",
       "      <th>Cleanliness</th>\n",
       "      <th>Departure Delay in Minutes</th>\n",
       "      <th>Arrival Delay in Minutes</th>\n",
       "      <th>satisfaction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>70172</td>\n",
       "      <td>Male</td>\n",
       "      <td>Loyal Customer</td>\n",
       "      <td>13</td>\n",
       "      <td>Personal Travel</td>\n",
       "      <td>Eco Plus</td>\n",
       "      <td>460</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>25</td>\n",
       "      <td>18.0</td>\n",
       "      <td>neutral or dissatisfied</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>5047</td>\n",
       "      <td>Male</td>\n",
       "      <td>disloyal Customer</td>\n",
       "      <td>25</td>\n",
       "      <td>Business travel</td>\n",
       "      <td>Business</td>\n",
       "      <td>235</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>neutral or dissatisfied</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>110028</td>\n",
       "      <td>Female</td>\n",
       "      <td>Loyal Customer</td>\n",
       "      <td>26</td>\n",
       "      <td>Business travel</td>\n",
       "      <td>Business</td>\n",
       "      <td>1142</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>satisfied</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>24026</td>\n",
       "      <td>Female</td>\n",
       "      <td>Loyal Customer</td>\n",
       "      <td>25</td>\n",
       "      <td>Business travel</td>\n",
       "      <td>Business</td>\n",
       "      <td>562</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>9.0</td>\n",
       "      <td>neutral or dissatisfied</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>119299</td>\n",
       "      <td>Male</td>\n",
       "      <td>Loyal Customer</td>\n",
       "      <td>61</td>\n",
       "      <td>Business travel</td>\n",
       "      <td>Business</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>satisfied</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0      id  Gender      Customer Type  Age   Type of Travel  \\\n",
       "0           0   70172    Male     Loyal Customer   13  Personal Travel   \n",
       "1           1    5047    Male  disloyal Customer   25  Business travel   \n",
       "2           2  110028  Female     Loyal Customer   26  Business travel   \n",
       "3           3   24026  Female     Loyal Customer   25  Business travel   \n",
       "4           4  119299    Male     Loyal Customer   61  Business travel   \n",
       "\n",
       "      Class  Flight Distance  Inflight wifi service  \\\n",
       "0  Eco Plus              460                      3   \n",
       "1  Business              235                      3   \n",
       "2  Business             1142                      2   \n",
       "3  Business              562                      2   \n",
       "4  Business              214                      3   \n",
       "\n",
       "   Departure/Arrival time convenient  ...  Inflight entertainment  \\\n",
       "0                                  4  ...                       5   \n",
       "1                                  2  ...                       1   \n",
       "2                                  2  ...                       5   \n",
       "3                                  5  ...                       2   \n",
       "4                                  3  ...                       3   \n",
       "\n",
       "   On-board service  Leg room service  Baggage handling  Checkin service  \\\n",
       "0                 4                 3                 4                4   \n",
       "1                 1                 5                 3                1   \n",
       "2                 4                 3                 4                4   \n",
       "3                 2                 5                 3                1   \n",
       "4                 3                 4                 4                3   \n",
       "\n",
       "   Inflight service  Cleanliness  Departure Delay in Minutes  \\\n",
       "0                 5            5                          25   \n",
       "1                 4            1                           1   \n",
       "2                 4            5                           0   \n",
       "3                 4            2                          11   \n",
       "4                 3            3                           0   \n",
       "\n",
       "   Arrival Delay in Minutes             satisfaction  \n",
       "0                      18.0  neutral or dissatisfied  \n",
       "1                       6.0  neutral or dissatisfied  \n",
       "2                       0.0                satisfied  \n",
       "3                       9.0  neutral or dissatisfied  \n",
       "4                       0.0                satisfied  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_data = pd.read_csv('data/train.csv')\n",
    "raw_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PreProcessing: \n",
    "- drop unnamed column\n",
    "- drop id column\n",
    "- drop rows that have Nan "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = raw_data.drop(['id', 'Unnamed: 0'], axis=1)\n",
    "data = data.dropna(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Flight Distance</th>\n",
       "      <th>Inflight wifi service</th>\n",
       "      <th>Departure/Arrival time convenient</th>\n",
       "      <th>Ease of Online booking</th>\n",
       "      <th>Gate location</th>\n",
       "      <th>Food and drink</th>\n",
       "      <th>Online boarding</th>\n",
       "      <th>Seat comfort</th>\n",
       "      <th>Inflight entertainment</th>\n",
       "      <th>On-board service</th>\n",
       "      <th>Leg room service</th>\n",
       "      <th>Baggage handling</th>\n",
       "      <th>Checkin service</th>\n",
       "      <th>Inflight service</th>\n",
       "      <th>Cleanliness</th>\n",
       "      <th>Departure Delay in Minutes</th>\n",
       "      <th>Arrival Delay in Minutes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>39.380466</td>\n",
       "      <td>1189.325202</td>\n",
       "      <td>2.729753</td>\n",
       "      <td>3.060081</td>\n",
       "      <td>2.756984</td>\n",
       "      <td>2.977026</td>\n",
       "      <td>3.202126</td>\n",
       "      <td>3.250497</td>\n",
       "      <td>3.439765</td>\n",
       "      <td>3.358341</td>\n",
       "      <td>3.382609</td>\n",
       "      <td>3.351401</td>\n",
       "      <td>3.631687</td>\n",
       "      <td>3.304323</td>\n",
       "      <td>3.640761</td>\n",
       "      <td>3.286397</td>\n",
       "      <td>14.747939</td>\n",
       "      <td>15.178678</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>15.113125</td>\n",
       "      <td>997.297235</td>\n",
       "      <td>1.327866</td>\n",
       "      <td>1.525233</td>\n",
       "      <td>1.398934</td>\n",
       "      <td>1.277723</td>\n",
       "      <td>1.329401</td>\n",
       "      <td>1.349433</td>\n",
       "      <td>1.318896</td>\n",
       "      <td>1.333030</td>\n",
       "      <td>1.288284</td>\n",
       "      <td>1.315409</td>\n",
       "      <td>1.181051</td>\n",
       "      <td>1.265396</td>\n",
       "      <td>1.175603</td>\n",
       "      <td>1.312194</td>\n",
       "      <td>38.116737</td>\n",
       "      <td>38.698682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>7.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>27.000000</td>\n",
       "      <td>414.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>40.000000</td>\n",
       "      <td>842.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>1743.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>85.000000</td>\n",
       "      <td>4983.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1592.000000</td>\n",
       "      <td>1584.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Age  Flight Distance  Inflight wifi service  \\\n",
       "count  103594.000000    103594.000000          103594.000000   \n",
       "mean       39.380466      1189.325202               2.729753   \n",
       "std        15.113125       997.297235               1.327866   \n",
       "min         7.000000        31.000000               0.000000   \n",
       "25%        27.000000       414.000000               2.000000   \n",
       "50%        40.000000       842.000000               3.000000   \n",
       "75%        51.000000      1743.000000               4.000000   \n",
       "max        85.000000      4983.000000               5.000000   \n",
       "\n",
       "       Departure/Arrival time convenient  Ease of Online booking  \\\n",
       "count                      103594.000000           103594.000000   \n",
       "mean                            3.060081                2.756984   \n",
       "std                             1.525233                1.398934   \n",
       "min                             0.000000                0.000000   \n",
       "25%                             2.000000                2.000000   \n",
       "50%                             3.000000                3.000000   \n",
       "75%                             4.000000                4.000000   \n",
       "max                             5.000000                5.000000   \n",
       "\n",
       "       Gate location  Food and drink  Online boarding   Seat comfort  \\\n",
       "count  103594.000000   103594.000000    103594.000000  103594.000000   \n",
       "mean        2.977026        3.202126         3.250497       3.439765   \n",
       "std         1.277723        1.329401         1.349433       1.318896   \n",
       "min         0.000000        0.000000         0.000000       0.000000   \n",
       "25%         2.000000        2.000000         2.000000       2.000000   \n",
       "50%         3.000000        3.000000         3.000000       4.000000   \n",
       "75%         4.000000        4.000000         4.000000       5.000000   \n",
       "max         5.000000        5.000000         5.000000       5.000000   \n",
       "\n",
       "       Inflight entertainment  On-board service  Leg room service  \\\n",
       "count           103594.000000     103594.000000     103594.000000   \n",
       "mean                 3.358341          3.382609          3.351401   \n",
       "std                  1.333030          1.288284          1.315409   \n",
       "min                  0.000000          0.000000          0.000000   \n",
       "25%                  2.000000          2.000000          2.000000   \n",
       "50%                  4.000000          4.000000          4.000000   \n",
       "75%                  4.000000          4.000000          4.000000   \n",
       "max                  5.000000          5.000000          5.000000   \n",
       "\n",
       "       Baggage handling  Checkin service  Inflight service    Cleanliness  \\\n",
       "count     103594.000000    103594.000000     103594.000000  103594.000000   \n",
       "mean           3.631687         3.304323          3.640761       3.286397   \n",
       "std            1.181051         1.265396          1.175603       1.312194   \n",
       "min            1.000000         0.000000          0.000000       0.000000   \n",
       "25%            3.000000         3.000000          3.000000       2.000000   \n",
       "50%            4.000000         3.000000          4.000000       3.000000   \n",
       "75%            5.000000         4.000000          5.000000       4.000000   \n",
       "max            5.000000         5.000000          5.000000       5.000000   \n",
       "\n",
       "       Departure Delay in Minutes  Arrival Delay in Minutes  \n",
       "count               103594.000000             103594.000000  \n",
       "mean                    14.747939                 15.178678  \n",
       "std                     38.116737                 38.698682  \n",
       "min                      0.000000                  0.000000  \n",
       "25%                      0.000000                  0.000000  \n",
       "50%                      0.000000                  0.000000  \n",
       "75%                     12.000000                 13.000000  \n",
       "max                   1592.000000               1584.000000  "
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Change categorical data to numeric(using one hot encoding): "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAFuCAYAAAChovKPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAWZ0lEQVR4nO3df5Bd5X3f8fcHyQacGGKwIFgiEQ1yU8AxDjKlcdPalhvUX4F4IFFaG7lVq5QhHjvTpgOdaeKmowlM0jgxDvZQYyNoE9BACJQEbEaEurUZQNjEAmyKbBxQoUgYiknG0Ej59o/7rH21Wi0L6OzdZ3m/Zu7cc773PGefo6P96NFzzz03VYUkqR+HTLoDkqSXxuCWpM4Y3JLUGYNbkjpjcEtSZ5ZOugPzbe3atXXrrbdOuhuSNBeZqfiqG3E/9dRTk+6CJL0ir7rglqTeGdyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHVm0OBO8s0k25Pcl2Rbqx2V5LYkD7fnN4xtf1GSHUkeSnLmWP20tp8dST6WJK1+aJJrW/2uJCuHPB5JWgjmY8T9rqo6tapWt/ULga1VtQrY2tZJchKwDjgZWAtclmRJa/MJYCOwqj3WtvoG4JmqOhH4KHDJPByPJE3UJKZKzgI2t+XNwNlj9Wuq6oWqegTYAZye5DjgiKq6s0Zf13PVtDZT+7oOWDM1GpekxWro4C7gc0nuTbKx1Y6tqicA2vMxrb4ceGys7c5WW96Wp9f3aVNVe4BngaOndyLJxiTbkmzbvXv3QTkwSZqUoe8O+I6qejzJMcBtSb42y7YzjZRrlvpsbfYtVF0OXA6wevVqv2RTUtcGHXFX1ePteRdwA3A68GSb/qA972qb7wSOH2u+Ani81VfMUN+nTZKlwJHA00MciyQtFIONuJN8H3BIVT3Xln8K+DXgJmA9cHF7vrE1uQn4vSS/BbyJ0ZuQd1fV3iTPJTkDuAs4D7h0rM164E7gHOD28mvrF6VHf+0tk+5C137oV7ZPugs6iIacKjkWuKG9V7gU+L2qujXJPcCWJBuAR4FzAarqgSRbgAeBPcAFVbW37et84ErgcOCW9gC4Arg6yQ5GI+11Ax6PJC0IgwV3VX0DeOsM9W8Baw7QZhOwaYb6NuCUGerP04Jfkl4t/OSkJHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnRn6OyclLULvuPQdk+5C177wwS+8ovaOuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUmcGDO8mSJF9OcnNbPyrJbUkebs9vGNv2oiQ7kjyU5Myx+mlJtrfXPpYkrX5okmtb/a4kK4c+HkmatPkYcX8I+OrY+oXA1qpaBWxt6yQ5CVgHnAysBS5LsqS1+QSwEVjVHmtbfQPwTFWdCHwUuGTYQ5GkyRs0uJOsAP4h8Kmx8lnA5ra8GTh7rH5NVb1QVY8AO4DTkxwHHFFVd1ZVAVdNazO1r+uANVOjcUlarIYecf828G+BvxqrHVtVTwC052NafTnw2Nh2O1tteVueXt+nTVXtAZ4Fjj6oRyBJC8xgwZ3kHwG7qureuTaZoVaz1GdrM70vG5NsS7Jt9+7dc+yOJC1MQ4643wH8dJJvAtcA707yX4An2/QH7XlX234ncPxY+xXA462+Yob6Pm2SLAWOBJ6e3pGquryqVlfV6mXLlh2co5OkCRksuKvqoqpaUVUrGb3peHtVvQ+4CVjfNlsP3NiWbwLWtStFTmD0JuTdbTrluSRntPnr86a1mdrXOe1n7DfilqTFZOkEfubFwJYkG4BHgXMBquqBJFuAB4E9wAVVtbe1OR+4EjgcuKU9AK4Ark6yg9FIe918HYQkTcq8BHdV3QHc0Za/Baw5wHabgE0z1LcBp8xQf54W/JL0auEnJyWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOjNYcCc5LMndSf40yQNJ/kOrH5XktiQPt+c3jLW5KMmOJA8lOXOsflqS7e21jyVJqx+a5NpWvyvJyqGOR5IWiiFH3C8A766qtwKnAmuTnAFcCGytqlXA1rZOkpOAdcDJwFrgsiRL2r4+AWwEVrXH2lbfADxTVScCHwUuGfB4JGlBGCy4a+TP2+pr2qOAs4DNrb4ZOLstnwVcU1UvVNUjwA7g9CTHAUdU1Z1VVcBV09pM7es6YM3UaFySFqtB57iTLElyH7ALuK2q7gKOraonANrzMW3z5cBjY813ttrytjy9vk+bqtoDPAscPUM/NibZlmTb7t27D9LRSdJkDBrcVbW3qk4FVjAaPZ8yy+YzjZRrlvpsbab34/KqWl1Vq5ctW/YivZakhW1eriqpqv8L3MFobvrJNv1Be97VNtsJHD/WbAXweKuvmKG+T5skS4EjgaeHOAZJWiiGvKpkWZIfaMuHA+8BvgbcBKxvm60HbmzLNwHr2pUiJzB6E/LuNp3yXJIz2vz1edPaTO3rHOD2Ng8uSYvW0gH3fRywuV0ZcgiwpapuTnInsCXJBuBR4FyAqnogyRbgQWAPcEFV7W37Oh+4EjgcuKU9AK4Ark6yg9FIe92AxyNJC8JgwV1VXwHeNkP9W8CaA7TZBGyaob4N2G9+vKqepwW/JL1a+MlJSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmfmFNxJts6lJkka3qyfnExyGPA64I3tm2qm7sZ3BPCmgfsmSZrBi33k/ReADzMK6Xv5XnB/G/jd4bolSTqQWYO7qn4H+J0kH6yqS+epT5KkWczpJlNVdWmSnwBWjrepqqsG6pck6QDmFNxJrgZ+BLgPmLrV6tT3P0qS5tFcb+u6GjjJLymQpMmb63Xc9wM/OGRHJElzM9cR9xuBB5PcDbwwVayqnx6kV5KkA5prcH9kyE5IkuZurleV/PehOyJJmpu5XlXyHKOrSABeC7wG+IuqOmKojkmSZjbXEffrx9eTnA2cPkSHJEmze1l3B6yqPwTefXC7Ikmai7lOlbx3bPUQRtd1e023JE3AXK8q+cdjy3uAbwJnHfTeSJJe1FznuP/Z0B2RJM3NXL9IYUWSG5LsSvJkkuuTrBi6c5Kk/c31zcnPADcxui/3cuC/tZokaZ7NNbiXVdVnqmpPe1wJLBuwX5KkA5hrcD+V5H1JlrTH+4BvDdkxSdLM5hrc/xz4WeD/AE8A5wC+YSlJEzDXywH/I7C+qp4BSHIU8JuMAl2SNI/mOuL+sanQBqiqp4G3DdMlSdJs5hrchyR5w9RKG3HPdbQuSTqI5hq+/wn4YpLrGH3U/WeBTYP1SpJ0QHP95ORVSbYxurFUgPdW1YOD9kySNKM5T3e0oDasJWnCXtZtXSVJk2NwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6M1hwJzk+yZ8k+WqSB5J8qNWPSnJbkofb8/iXEF+UZEeSh5KcOVY/Lcn29trHkqTVD01ybavflWTlUMcjSQvFkCPuPcC/rqq/AZwBXJDkJOBCYGtVrQK2tnXaa+uAk4G1wGVJlrR9fQLYCKxqj7WtvgF4pqpOBD4KXDLg8UjSgjBYcFfVE1X1pbb8HPBVYDlwFrC5bbYZOLstnwVcU1UvVNUjwA7g9CTHAUdU1Z1VVcBV09pM7es6YM3UaFySFqt5meNuUxhvA+4Cjq2qJ2AU7sAxbbPlwGNjzXa22vK2PL2+T5uq2gM8Cxw9w8/fmGRbkm27d+8+SEclSZMxeHAn+X7geuDDVfXt2TadoVaz1Gdrs2+h6vKqWl1Vq5ctW/ZiXZakBW3Q4E7yGkah/V+r6g9a+ck2/UF73tXqO4Hjx5qvAB5v9RUz1Pdpk2QpcCTw9ME/EklaOIa8qiTAFcBXq+q3xl66CVjfltcDN47V17UrRU5g9Cbk3W065bkkZ7R9njetzdS+zgFub/PgkrRoLR1w3+8A3g9sT3Jfq/074GJgS5INwKPAuQBV9UCSLcCDjK5IuaCq9rZ25wNXAocDt7QHjP5huDrJDkYj7XUDHo8kLQiDBXdV/U9mnoMGWHOANpuATTPUtwGnzFB/nhb8kvRq4ScnJakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6M1hwJ/l0kl1J7h+rHZXktiQPt+c3jL12UZIdSR5KcuZY/bQk29trH0uSVj80ybWtfleSlUMdiyQtJEOOuK8E1k6rXQhsrapVwNa2TpKTgHXAya3NZUmWtDafADYCq9pjap8bgGeq6kTgo8Algx2JJC0ggwV3VX0eeHpa+Sxgc1veDJw9Vr+mql6oqkeAHcDpSY4DjqiqO6uqgKumtZna13XAmqnRuCQtZvM9x31sVT0B0J6PafXlwGNj2+1steVteXp9nzZVtQd4Fjh6ph+aZGOSbUm27d69+yAdiiRNxkJ5c3KmkXLNUp+tzf7FqsuranVVrV62bNnL7KIkLQzzHdxPtukP2vOuVt8JHD+23Qrg8VZfMUN9nzZJlgJHsv/UjCQtOvMd3DcB69vyeuDGsfq6dqXICYzehLy7Tac8l+SMNn993rQ2U/s6B7i9zYNL0qK2dKgdJ/l94J3AG5PsBH4VuBjYkmQD8ChwLkBVPZBkC/AgsAe4oKr2tl2dz+gKlcOBW9oD4Arg6iQ7GI201w11LJK0kAwW3FX18wd4ac0Btt8EbJqhvg04ZYb687Tgl6RXk4Xy5qQkaY4MbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6sxgH3nv3Wm/fNWku9C1e3/jvEl3QVq0HHFLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSeqMwS1JnTG4JakzBrckdcbglqTOGNyS1BmDW5I6Y3BLUmcMbknqjMEtSZ0xuCWpMwa3JHXG4JakzhjcktQZg1uSOmNwS1JnDG5J6ozBLUmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZwxuSepM98GdZG2Sh5LsSHLhpPsjSUPrOriTLAF+F/j7wEnAzyc5abK9kqRhdR3cwOnAjqr6RlX9P+Aa4KwJ90mSBpWqmnQfXrYk5wBrq+pftPX3A3+zqn5x2nYbgY1t9a8DD81rR4fxRuCpSXdC3+X5WFgWy/l4qqrWTi8unURPDqLMUNvvX6Kquhy4fPjuzJ8k26pq9aT7oRHPx8Ky2M9H71MlO4Hjx9ZXAI9PqC+SNC96D+57gFVJTkjyWmAdcNOE+yRJg+p6qqSq9iT5ReCzwBLg01X1wIS7NV8W1dTPIuD5WFgW9fno+s1JSXo16n2qRJJedQxuSeqMwT1PkuxNct/Y4xV9PD/JyiTfaft6MMknkxzS6vcfrH4vRmPn4k+TfCnJT7zM/fyrJOcd7P7pxR3s36fedP3mZGe+U1WnHuR9fr2qTk2yFLgdOBv40kH+GYvRd89FkjOBXwf+7kvdSVV98iD3S3M3xO9TNxxxT1iStyf5Yhv93Z3k9UkOS/KZJNuTfDnJu2bbR1XtAb4InDht3x9I8vGx9ZuTvDPJkiRXJrm//YxfGubounAE8AxA+7O5eeqFJB9P8oG2fHH7n81Xkvxmq30kyb9py3ckuaSdw/+V5CdbfUmS30hyT2v7C61+XJLPt9Hi/Ul+0vPyyh2M36ceOOKeP4cnuW9s/deBG4BrgZ+rqnuSHAF8B/gQQFW9JcmPAp9L8uaqen6mHSd5HbAG+JU59uVUYHlVndLa/8BLP5yuTZ2Lw4DjgHfPtnGSo4CfAX60qmqWP6+lVXV6kn8A/CrwHmAD8GxVvT3JocAXknwOeC/w2ara1G6W9jo8Ly/FYL9PPTC4589+/7VL8hbgiaq6B6Cqvt3qfxu4tNW+luTPgDcDX5m2zx9pf3kLuLGqbkmycg59+Qbw15JcCvwR8LmXe1CdGp8q+VvAVUlOmWX7bwPPA59K8kfAzQfY7g/a873Ayrb8U8CPtfvqABwJrGL04bFPJ3kN8IdVdV+SV/t5eSmG+H3qhlMlkxVmuLcKM9+DZSZfr6pTq+ptVfWRGV7fw77n+DCAqnoGeCtwB3AB8Km5dnixqao7Gd2QaBkH/vPaw+hOlNczeh/h1gPs7oX2vJfvDYoCfLCdp1Or6oSq+lxVfR74O8D/Bq5Ocp7n5RV7pb9P3TC4J+trwJuSvB2gzcctBT4P/NNWezPwQ7y8Oxp+Ezi1XW1yPKPwIckbgUOq6nrg3wM//koPpFftv85LgG8BfwaclOTQJEcymn4iyfcDR1bVHwMfZjSlMVefBc5vI2uSvDnJ9yX5YWBXVf1n4Argxz0vr9jQv08LhlMl82f6nNytVXVhkp8DLk1yOKP5uPcAlwGfTLKd0SjwA1X1wn57fHFfAB4BtgP3870rTpYDn0ky9Q/3RS9j3z0bPxcB1lfVXuCxJFsY/Rf6YeDLbZvXAzcmOaxt/1LeNPwUo2mTLyUJsJvRqP2dwC8n+Uvgz4Hz8Ly8FJP4fVow/Mi7JHXGqRJJ6ozBLUmdMbglqTMGtyR1xuCWpM4Y3NIMkvxgkmuSfL3do+SP2zXY3nlRE+d13NI07XrrG4DNVbWu1U4Fjp1kv6Qpjril/b0L+Mvx27ZW1X3AY1PrGd33/H9kdD/v797T27v+aT444pb2dwqjG0XNZhfw96rq+SSrgN8HVgP/BO/6p4EZ3NLL8xrg420KZS+ju82Bd/3TPHCqRNrfA8BpL7LNLwFPMrqb32rgtQDe9U/zweCW9nc7cGiSfzlVaHec++GxbY5kdO/nvwLez+gOg3jXP80Hp0qkadq33PwM8NsZfQnt84xukfvhsc0uA65Pci7wJ8BftPo78a5/Gph3B5SkzjhVIkmdMbglqTMGtyR1xuCWpM4Y3JLUGYNbkjpjcEtSZ/4/f6UQp+ETtQcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cat_cols = list(set(data.columns) - set(data._get_numeric_data().columns))\n",
    "for s in cat_cols:\n",
    "    sns.catplot(x=s, kind=\"count\", data=data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['Age', 'Flight Distance', 'Inflight wifi service',\n",
      "       'Departure/Arrival time convenient', 'Ease of Online booking',\n",
      "       'Gate location', 'Food and drink', 'Online boarding', 'Seat comfort',\n",
      "       'Inflight entertainment', 'On-board service', 'Leg room service',\n",
      "       'Baggage handling', 'Checkin service', 'Inflight service',\n",
      "       'Cleanliness', 'Departure Delay in Minutes', 'Arrival Delay in Minutes',\n",
      "       'satisfaction', 'Gender_Female', 'Gender_Male',\n",
      "       'Type of Travel_Business travel', 'Type of Travel_Personal Travel',\n",
      "       'Class_Business', 'Class_Eco', 'Class_Eco Plus',\n",
      "       'Customer Type_Loyal Customer', 'Customer Type_disloyal Customer'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "encd_data = pd.get_dummies(data, columns = list(set(cat_cols)-set(['satisfaction'])))\n",
    "encd_data['satisfaction'].replace(['neutral or dissatisfied', 'satisfied'], [0, 1], inplace=True)\n",
    "\n",
    "print(encd_data.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Flight Distance</th>\n",
       "      <th>Inflight wifi service</th>\n",
       "      <th>Departure/Arrival time convenient</th>\n",
       "      <th>Ease of Online booking</th>\n",
       "      <th>Gate location</th>\n",
       "      <th>Food and drink</th>\n",
       "      <th>Online boarding</th>\n",
       "      <th>Seat comfort</th>\n",
       "      <th>Inflight entertainment</th>\n",
       "      <th>...</th>\n",
       "      <th>satisfaction</th>\n",
       "      <th>Gender_Female</th>\n",
       "      <th>Gender_Male</th>\n",
       "      <th>Type of Travel_Business travel</th>\n",
       "      <th>Type of Travel_Personal Travel</th>\n",
       "      <th>Class_Business</th>\n",
       "      <th>Class_Eco</th>\n",
       "      <th>Class_Eco Plus</th>\n",
       "      <th>Customer Type_Loyal Customer</th>\n",
       "      <th>Customer Type_disloyal Customer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>39.380466</td>\n",
       "      <td>1189.325202</td>\n",
       "      <td>2.729753</td>\n",
       "      <td>3.060081</td>\n",
       "      <td>2.756984</td>\n",
       "      <td>2.977026</td>\n",
       "      <td>3.202126</td>\n",
       "      <td>3.250497</td>\n",
       "      <td>3.439765</td>\n",
       "      <td>3.358341</td>\n",
       "      <td>...</td>\n",
       "      <td>0.433394</td>\n",
       "      <td>0.507520</td>\n",
       "      <td>0.492480</td>\n",
       "      <td>0.689857</td>\n",
       "      <td>0.310143</td>\n",
       "      <td>0.478145</td>\n",
       "      <td>0.449765</td>\n",
       "      <td>0.072089</td>\n",
       "      <td>0.817248</td>\n",
       "      <td>0.182752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>15.113125</td>\n",
       "      <td>997.297235</td>\n",
       "      <td>1.327866</td>\n",
       "      <td>1.525233</td>\n",
       "      <td>1.398934</td>\n",
       "      <td>1.277723</td>\n",
       "      <td>1.329401</td>\n",
       "      <td>1.349433</td>\n",
       "      <td>1.318896</td>\n",
       "      <td>1.333030</td>\n",
       "      <td>...</td>\n",
       "      <td>0.495546</td>\n",
       "      <td>0.499946</td>\n",
       "      <td>0.499946</td>\n",
       "      <td>0.462554</td>\n",
       "      <td>0.462554</td>\n",
       "      <td>0.499525</td>\n",
       "      <td>0.497472</td>\n",
       "      <td>0.258637</td>\n",
       "      <td>0.386465</td>\n",
       "      <td>0.386465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>7.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>27.000000</td>\n",
       "      <td>414.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>40.000000</td>\n",
       "      <td>842.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>1743.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>85.000000</td>\n",
       "      <td>4983.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Age  Flight Distance  Inflight wifi service  \\\n",
       "count  103594.000000    103594.000000          103594.000000   \n",
       "mean       39.380466      1189.325202               2.729753   \n",
       "std        15.113125       997.297235               1.327866   \n",
       "min         7.000000        31.000000               0.000000   \n",
       "25%        27.000000       414.000000               2.000000   \n",
       "50%        40.000000       842.000000               3.000000   \n",
       "75%        51.000000      1743.000000               4.000000   \n",
       "max        85.000000      4983.000000               5.000000   \n",
       "\n",
       "       Departure/Arrival time convenient  Ease of Online booking  \\\n",
       "count                      103594.000000           103594.000000   \n",
       "mean                            3.060081                2.756984   \n",
       "std                             1.525233                1.398934   \n",
       "min                             0.000000                0.000000   \n",
       "25%                             2.000000                2.000000   \n",
       "50%                             3.000000                3.000000   \n",
       "75%                             4.000000                4.000000   \n",
       "max                             5.000000                5.000000   \n",
       "\n",
       "       Gate location  Food and drink  Online boarding   Seat comfort  \\\n",
       "count  103594.000000   103594.000000    103594.000000  103594.000000   \n",
       "mean        2.977026        3.202126         3.250497       3.439765   \n",
       "std         1.277723        1.329401         1.349433       1.318896   \n",
       "min         0.000000        0.000000         0.000000       0.000000   \n",
       "25%         2.000000        2.000000         2.000000       2.000000   \n",
       "50%         3.000000        3.000000         3.000000       4.000000   \n",
       "75%         4.000000        4.000000         4.000000       5.000000   \n",
       "max         5.000000        5.000000         5.000000       5.000000   \n",
       "\n",
       "       Inflight entertainment  ...   satisfaction  Gender_Female  \\\n",
       "count           103594.000000  ...  103594.000000  103594.000000   \n",
       "mean                 3.358341  ...       0.433394       0.507520   \n",
       "std                  1.333030  ...       0.495546       0.499946   \n",
       "min                  0.000000  ...       0.000000       0.000000   \n",
       "25%                  2.000000  ...       0.000000       0.000000   \n",
       "50%                  4.000000  ...       0.000000       1.000000   \n",
       "75%                  4.000000  ...       1.000000       1.000000   \n",
       "max                  5.000000  ...       1.000000       1.000000   \n",
       "\n",
       "         Gender_Male  Type of Travel_Business travel  \\\n",
       "count  103594.000000                   103594.000000   \n",
       "mean        0.492480                        0.689857   \n",
       "std         0.499946                        0.462554   \n",
       "min         0.000000                        0.000000   \n",
       "25%         0.000000                        0.000000   \n",
       "50%         0.000000                        1.000000   \n",
       "75%         1.000000                        1.000000   \n",
       "max         1.000000                        1.000000   \n",
       "\n",
       "       Type of Travel_Personal Travel  Class_Business      Class_Eco  \\\n",
       "count                   103594.000000   103594.000000  103594.000000   \n",
       "mean                         0.310143        0.478145       0.449765   \n",
       "std                          0.462554        0.499525       0.497472   \n",
       "min                          0.000000        0.000000       0.000000   \n",
       "25%                          0.000000        0.000000       0.000000   \n",
       "50%                          0.000000        0.000000       0.000000   \n",
       "75%                          1.000000        1.000000       1.000000   \n",
       "max                          1.000000        1.000000       1.000000   \n",
       "\n",
       "       Class_Eco Plus  Customer Type_Loyal Customer  \\\n",
       "count   103594.000000                 103594.000000   \n",
       "mean         0.072089                      0.817248   \n",
       "std          0.258637                      0.386465   \n",
       "min          0.000000                      0.000000   \n",
       "25%          0.000000                      1.000000   \n",
       "50%          0.000000                      1.000000   \n",
       "75%          0.000000                      1.000000   \n",
       "max          1.000000                      1.000000   \n",
       "\n",
       "       Customer Type_disloyal Customer  \n",
       "count                    103594.000000  \n",
       "mean                          0.182752  \n",
       "std                           0.386465  \n",
       "min                           0.000000  \n",
       "25%                           0.000000  \n",
       "50%                           0.000000  \n",
       "75%                           0.000000  \n",
       "max                           1.000000  \n",
       "\n",
       "[8 rows x 28 columns]"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encd_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Normalize, using min max scaling: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Flight Distance</th>\n",
       "      <th>Inflight wifi service</th>\n",
       "      <th>Departure/Arrival time convenient</th>\n",
       "      <th>Ease of Online booking</th>\n",
       "      <th>Gate location</th>\n",
       "      <th>Food and drink</th>\n",
       "      <th>Online boarding</th>\n",
       "      <th>Seat comfort</th>\n",
       "      <th>Inflight entertainment</th>\n",
       "      <th>...</th>\n",
       "      <th>satisfaction</th>\n",
       "      <th>Gender_Female</th>\n",
       "      <th>Gender_Male</th>\n",
       "      <th>Type of Travel_Business travel</th>\n",
       "      <th>Type of Travel_Personal Travel</th>\n",
       "      <th>Class_Business</th>\n",
       "      <th>Class_Eco</th>\n",
       "      <th>Class_Eco Plus</th>\n",
       "      <th>Customer Type_Loyal Customer</th>\n",
       "      <th>Customer Type_disloyal Customer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "      <td>103594.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.415134</td>\n",
       "      <td>0.233911</td>\n",
       "      <td>0.545951</td>\n",
       "      <td>0.612016</td>\n",
       "      <td>0.551397</td>\n",
       "      <td>0.595405</td>\n",
       "      <td>0.640425</td>\n",
       "      <td>0.650099</td>\n",
       "      <td>0.687953</td>\n",
       "      <td>0.671668</td>\n",
       "      <td>...</td>\n",
       "      <td>0.433394</td>\n",
       "      <td>0.507520</td>\n",
       "      <td>0.492480</td>\n",
       "      <td>0.689857</td>\n",
       "      <td>0.310143</td>\n",
       "      <td>0.478145</td>\n",
       "      <td>0.449765</td>\n",
       "      <td>0.072089</td>\n",
       "      <td>0.817248</td>\n",
       "      <td>0.182752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.193758</td>\n",
       "      <td>0.201393</td>\n",
       "      <td>0.265573</td>\n",
       "      <td>0.305047</td>\n",
       "      <td>0.279787</td>\n",
       "      <td>0.255545</td>\n",
       "      <td>0.265880</td>\n",
       "      <td>0.269887</td>\n",
       "      <td>0.263779</td>\n",
       "      <td>0.266606</td>\n",
       "      <td>...</td>\n",
       "      <td>0.495546</td>\n",
       "      <td>0.499946</td>\n",
       "      <td>0.499946</td>\n",
       "      <td>0.462554</td>\n",
       "      <td>0.462554</td>\n",
       "      <td>0.499525</td>\n",
       "      <td>0.497472</td>\n",
       "      <td>0.258637</td>\n",
       "      <td>0.386465</td>\n",
       "      <td>0.386465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.256410</td>\n",
       "      <td>0.077342</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.423077</td>\n",
       "      <td>0.163772</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.564103</td>\n",
       "      <td>0.345719</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Age  Flight Distance  Inflight wifi service  \\\n",
       "count  103594.000000    103594.000000          103594.000000   \n",
       "mean        0.415134         0.233911               0.545951   \n",
       "std         0.193758         0.201393               0.265573   \n",
       "min         0.000000         0.000000               0.000000   \n",
       "25%         0.256410         0.077342               0.400000   \n",
       "50%         0.423077         0.163772               0.600000   \n",
       "75%         0.564103         0.345719               0.800000   \n",
       "max         1.000000         1.000000               1.000000   \n",
       "\n",
       "       Departure/Arrival time convenient  Ease of Online booking  \\\n",
       "count                      103594.000000           103594.000000   \n",
       "mean                            0.612016                0.551397   \n",
       "std                             0.305047                0.279787   \n",
       "min                             0.000000                0.000000   \n",
       "25%                             0.400000                0.400000   \n",
       "50%                             0.600000                0.600000   \n",
       "75%                             0.800000                0.800000   \n",
       "max                             1.000000                1.000000   \n",
       "\n",
       "       Gate location  Food and drink  Online boarding   Seat comfort  \\\n",
       "count  103594.000000   103594.000000    103594.000000  103594.000000   \n",
       "mean        0.595405        0.640425         0.650099       0.687953   \n",
       "std         0.255545        0.265880         0.269887       0.263779   \n",
       "min         0.000000        0.000000         0.000000       0.000000   \n",
       "25%         0.400000        0.400000         0.400000       0.400000   \n",
       "50%         0.600000        0.600000         0.600000       0.800000   \n",
       "75%         0.800000        0.800000         0.800000       1.000000   \n",
       "max         1.000000        1.000000         1.000000       1.000000   \n",
       "\n",
       "       Inflight entertainment  ...   satisfaction  Gender_Female  \\\n",
       "count           103594.000000  ...  103594.000000  103594.000000   \n",
       "mean                 0.671668  ...       0.433394       0.507520   \n",
       "std                  0.266606  ...       0.495546       0.499946   \n",
       "min                  0.000000  ...       0.000000       0.000000   \n",
       "25%                  0.400000  ...       0.000000       0.000000   \n",
       "50%                  0.800000  ...       0.000000       1.000000   \n",
       "75%                  0.800000  ...       1.000000       1.000000   \n",
       "max                  1.000000  ...       1.000000       1.000000   \n",
       "\n",
       "         Gender_Male  Type of Travel_Business travel  \\\n",
       "count  103594.000000                   103594.000000   \n",
       "mean        0.492480                        0.689857   \n",
       "std         0.499946                        0.462554   \n",
       "min         0.000000                        0.000000   \n",
       "25%         0.000000                        0.000000   \n",
       "50%         0.000000                        1.000000   \n",
       "75%         1.000000                        1.000000   \n",
       "max         1.000000                        1.000000   \n",
       "\n",
       "       Type of Travel_Personal Travel  Class_Business      Class_Eco  \\\n",
       "count                   103594.000000   103594.000000  103594.000000   \n",
       "mean                         0.310143        0.478145       0.449765   \n",
       "std                          0.462554        0.499525       0.497472   \n",
       "min                          0.000000        0.000000       0.000000   \n",
       "25%                          0.000000        0.000000       0.000000   \n",
       "50%                          0.000000        0.000000       0.000000   \n",
       "75%                          1.000000        1.000000       1.000000   \n",
       "max                          1.000000        1.000000       1.000000   \n",
       "\n",
       "       Class_Eco Plus  Customer Type_Loyal Customer  \\\n",
       "count   103594.000000                 103594.000000   \n",
       "mean         0.072089                      0.817248   \n",
       "std          0.258637                      0.386465   \n",
       "min          0.000000                      0.000000   \n",
       "25%          0.000000                      1.000000   \n",
       "50%          0.000000                      1.000000   \n",
       "75%          0.000000                      1.000000   \n",
       "max          1.000000                      1.000000   \n",
       "\n",
       "       Customer Type_disloyal Customer  \n",
       "count                    103594.000000  \n",
       "mean                          0.182752  \n",
       "std                           0.386465  \n",
       "min                           0.000000  \n",
       "25%                           0.000000  \n",
       "50%                           0.000000  \n",
       "75%                           0.000000  \n",
       "max                           1.000000  \n",
       "\n",
       "[8 rows x 28 columns]"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def min_max_scaling(series):\n",
    "    return (series - series.min()) / (series.max() - series.min())  \n",
    "\n",
    "for col in encd_data.drop(['satisfaction'], axis=1).columns:\n",
    "    encd_data[col] = min_max_scaling(encd_data[col])\n",
    "    \n",
    "encd_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "# a function to do preprocessing\n",
    "def preprocess(df):\n",
    "    df = df.drop(['id', 'Unnamed: 0'], axis=1)\n",
    "    df = df.dropna(axis=0)\n",
    "    cat_cols = list(set(df.columns) - set(df._get_numeric_data().columns))\n",
    "    encd_data = pd.get_dummies(df, columns = list(set(cat_cols)-set(['satisfaction'])))\n",
    "    encd_data['satisfaction'].replace(['neutral or dissatisfied', 'satisfied'],\n",
    "                            [0, 1], inplace=True)\n",
    "    \n",
    "    for col in encd_data.drop(['satisfaction'], axis=1).columns:\n",
    "        encd_data[col] = min_max_scaling(encd_data[col])\n",
    "    \n",
    "    return encd_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Perceptron"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import Perceptron"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Perceptron(eta0=0.4, max_iter=1500, tol=None)"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = encd_data.drop('satisfaction', axis=1)\n",
    "y = encd_data['satisfaction']\n",
    "\n",
    "perceptron = Perceptron(eta0=0.4, tol=None, max_iter= 1500)\n",
    "# eta0 -> learing rate\n",
    "# tol  -> The stopping criterion. If it is not None, the iterations will stop when (loss > previous_loss - tol).\n",
    "# max_iterint -> The maximum number of passes over the training data (aka epochs)\n",
    "perceptron.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8305886441299689"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data_accuracy = perceptron.score(X, y)\n",
    "training_data_accuracy\n",
    "# Returns the mean accuracy on the given test data and labels."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8335843664310818"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data = pd.read_csv('data/test.csv')\n",
    "\n",
    "encd_test_data = preprocess(test_data)\n",
    "\n",
    "X_test = encd_test_data.drop('satisfaction', axis=1)\n",
    "y_test = encd_test_data['satisfaction']\n",
    "\n",
    "testing_data_accuracy = perceptron.score(X_test, y_test)\n",
    "testing_data_accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Kernel trick using feature maps:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RBF feature mapping:\n",
    "\n",
    "- [Feature mapping explained](https://xavierbourretsicotte.github.io/Kernel_feature_map.html#Gram-Matrix-vs-Feature-Map)\n",
    "- [rbf feature mapping](https://scikit-learn.org/stable/modules/kernel_approximation.html#rbf-kernel-approx)\n",
    "- [rbf sampler](https://scikit-learn.org/stable/modules/generated/sklearn.kernel_approximation.RBFSampler.html#sklearn.kernel_approximation.RBFSampler)\n",
    "- [Kernel perceptron](https://en.wikipedia.org/wiki/Kernel_perceptron)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.kernel_approximation import RBFSampler\n",
    "\n",
    "rbf_feature = RBFSampler(n_components=3200, gamma=1/40, random_state=1)\n",
    "X_rbf_mapping = rbf_feature.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(103594, 3200)"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_rbf_mapping.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9166457516844605"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf = Perceptron(eta0=0.1, tol=1e-3)\n",
    "clf.fit(X_rbf_mapping, y)\n",
    "clf.score(X_rbf_mapping, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9197852701502337"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data = pd.read_csv('data/test.csv')\n",
    "\n",
    "encd_test_data = preprocess(test_data)\n",
    "\n",
    "X_test = encd_test_data.drop('satisfaction', axis=1)\n",
    "y_test = encd_test_data['satisfaction']\n",
    "\n",
    "X_test_rbf = rbf_feature.transform(X_test)\n",
    "clf.score(X_test_rbf, y_test)"
   ]
  }
 ],
 "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
}