Credit Card Fraud Detection..ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Importing the required Python libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:23.942908Z",
     "start_time": "2019-09-19T18:31:19.696764Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import warnings\n",
    "\n",
    "%matplotlib inline\n",
    "sns.set()\n",
    "warnings.simplefilter('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's get the dataset into a pandas dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:36.105763Z",
     "start_time": "2019-09-19T18:31:23.942908Z"
    }
   },
   "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>378.66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>123.50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>69.99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9  ...         V21       V22       V23       V24  \\\n",
       "0  0.098698  0.363787  ...   -0.018307  0.277838 -0.110474  0.066928   \n",
       "1  0.085102 -0.255425  ...   -0.225775 -0.638672  0.101288 -0.339846   \n",
       "2  0.247676 -1.514654  ...    0.247998  0.771679  0.909412 -0.689281   \n",
       "3  0.377436 -1.387024  ...   -0.108300  0.005274 -0.190321 -1.175575   \n",
       "4 -0.270533  0.817739  ...   -0.009431  0.798278 -0.137458  0.141267   \n",
       "\n",
       "        V25       V26       V27       V28  Amount  Class  \n",
       "0  0.128539 -0.189115  0.133558 -0.021053  149.62      0  \n",
       "1  0.167170  0.125895 -0.008983  0.014724    2.69      0  \n",
       "2 -0.327642 -0.139097 -0.055353 -0.059752  378.66      0  \n",
       "3  0.647376 -0.221929  0.062723  0.061458  123.50      0  \n",
       "4 -0.206010  0.502292  0.219422  0.215153   69.99      0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(r\"C:\\Users\\hp\\Downloads\\creditcard.csv\\creditcard.csv\")\n",
    "df = data.copy() # To keep the data as backup\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:36.116220Z",
     "start_time": "2019-09-19T18:31:36.107243Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(284807, 31)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:36.296124Z",
     "start_time": "2019-09-19T18:31:36.118213Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Time      0\n",
       "V1        0\n",
       "V2        0\n",
       "V3        0\n",
       "V4        0\n",
       "V5        0\n",
       "V6        0\n",
       "V7        0\n",
       "V8        0\n",
       "V9        0\n",
       "V10       0\n",
       "V11       0\n",
       "V12       0\n",
       "V13       0\n",
       "V14       0\n",
       "V15       0\n",
       "V16       0\n",
       "V17       0\n",
       "V18       0\n",
       "V19       0\n",
       "V20       0\n",
       "V21       0\n",
       "V22       0\n",
       "V23       0\n",
       "V24       0\n",
       "V25       0\n",
       "V26       0\n",
       "V27       0\n",
       "V28       0\n",
       "Amount    0\n",
       "Class     0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:36.308240Z",
     "start_time": "2019-09-19T18:31:36.300262Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Time      float64\n",
       "V1        float64\n",
       "V2        float64\n",
       "V3        float64\n",
       "V4        float64\n",
       "V5        float64\n",
       "V6        float64\n",
       "V7        float64\n",
       "V8        float64\n",
       "V9        float64\n",
       "V10       float64\n",
       "V11       float64\n",
       "V12       float64\n",
       "V13       float64\n",
       "V14       float64\n",
       "V15       float64\n",
       "V16       float64\n",
       "V17       float64\n",
       "V18       float64\n",
       "V19       float64\n",
       "V20       float64\n",
       "V21       float64\n",
       "V22       float64\n",
       "V23       float64\n",
       "V24       float64\n",
       "V25       float64\n",
       "V26       float64\n",
       "V27       float64\n",
       "V28       float64\n",
       "Amount    float64\n",
       "Class       int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:36.326205Z",
     "start_time": "2019-09-19T18:31:36.315240Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "284792    172774.0\n",
       "284793    172775.0\n",
       "284794    172777.0\n",
       "284795    172778.0\n",
       "284796    172780.0\n",
       "284797    172782.0\n",
       "284798    172782.0\n",
       "284799    172783.0\n",
       "284800    172784.0\n",
       "284801    172785.0\n",
       "284802    172786.0\n",
       "284803    172787.0\n",
       "284804    172788.0\n",
       "284805    172788.0\n",
       "284806    172792.0\n",
       "Name: Time, dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Time.tail(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:37.284064Z",
     "start_time": "2019-09-19T18:31:36.334169Z"
    },
    "scrolled": true
   },
   "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>284807.000000</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>...</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>2.848070e+05</td>\n",
       "      <td>284807.000000</td>\n",
       "      <td>284807.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>94813.859575</td>\n",
       "      <td>3.919560e-15</td>\n",
       "      <td>5.688174e-16</td>\n",
       "      <td>-8.769071e-15</td>\n",
       "      <td>2.782312e-15</td>\n",
       "      <td>-1.552563e-15</td>\n",
       "      <td>2.010663e-15</td>\n",
       "      <td>-1.694249e-15</td>\n",
       "      <td>-1.927028e-16</td>\n",
       "      <td>-3.137024e-15</td>\n",
       "      <td>...</td>\n",
       "      <td>1.537294e-16</td>\n",
       "      <td>7.959909e-16</td>\n",
       "      <td>5.367590e-16</td>\n",
       "      <td>4.458112e-15</td>\n",
       "      <td>1.453003e-15</td>\n",
       "      <td>1.699104e-15</td>\n",
       "      <td>-3.660161e-16</td>\n",
       "      <td>-1.206049e-16</td>\n",
       "      <td>88.349619</td>\n",
       "      <td>0.001727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>47488.145955</td>\n",
       "      <td>1.958696e+00</td>\n",
       "      <td>1.651309e+00</td>\n",
       "      <td>1.516255e+00</td>\n",
       "      <td>1.415869e+00</td>\n",
       "      <td>1.380247e+00</td>\n",
       "      <td>1.332271e+00</td>\n",
       "      <td>1.237094e+00</td>\n",
       "      <td>1.194353e+00</td>\n",
       "      <td>1.098632e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>7.345240e-01</td>\n",
       "      <td>7.257016e-01</td>\n",
       "      <td>6.244603e-01</td>\n",
       "      <td>6.056471e-01</td>\n",
       "      <td>5.212781e-01</td>\n",
       "      <td>4.822270e-01</td>\n",
       "      <td>4.036325e-01</td>\n",
       "      <td>3.300833e-01</td>\n",
       "      <td>250.120109</td>\n",
       "      <td>0.041527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>-5.640751e+01</td>\n",
       "      <td>-7.271573e+01</td>\n",
       "      <td>-4.832559e+01</td>\n",
       "      <td>-5.683171e+00</td>\n",
       "      <td>-1.137433e+02</td>\n",
       "      <td>-2.616051e+01</td>\n",
       "      <td>-4.355724e+01</td>\n",
       "      <td>-7.321672e+01</td>\n",
       "      <td>-1.343407e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.483038e+01</td>\n",
       "      <td>-1.093314e+01</td>\n",
       "      <td>-4.480774e+01</td>\n",
       "      <td>-2.836627e+00</td>\n",
       "      <td>-1.029540e+01</td>\n",
       "      <td>-2.604551e+00</td>\n",
       "      <td>-2.256568e+01</td>\n",
       "      <td>-1.543008e+01</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>54201.500000</td>\n",
       "      <td>-9.203734e-01</td>\n",
       "      <td>-5.985499e-01</td>\n",
       "      <td>-8.903648e-01</td>\n",
       "      <td>-8.486401e-01</td>\n",
       "      <td>-6.915971e-01</td>\n",
       "      <td>-7.682956e-01</td>\n",
       "      <td>-5.540759e-01</td>\n",
       "      <td>-2.086297e-01</td>\n",
       "      <td>-6.430976e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.283949e-01</td>\n",
       "      <td>-5.423504e-01</td>\n",
       "      <td>-1.618463e-01</td>\n",
       "      <td>-3.545861e-01</td>\n",
       "      <td>-3.171451e-01</td>\n",
       "      <td>-3.269839e-01</td>\n",
       "      <td>-7.083953e-02</td>\n",
       "      <td>-5.295979e-02</td>\n",
       "      <td>5.600000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>84692.000000</td>\n",
       "      <td>1.810880e-02</td>\n",
       "      <td>6.548556e-02</td>\n",
       "      <td>1.798463e-01</td>\n",
       "      <td>-1.984653e-02</td>\n",
       "      <td>-5.433583e-02</td>\n",
       "      <td>-2.741871e-01</td>\n",
       "      <td>4.010308e-02</td>\n",
       "      <td>2.235804e-02</td>\n",
       "      <td>-5.142873e-02</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.945017e-02</td>\n",
       "      <td>6.781943e-03</td>\n",
       "      <td>-1.119293e-02</td>\n",
       "      <td>4.097606e-02</td>\n",
       "      <td>1.659350e-02</td>\n",
       "      <td>-5.213911e-02</td>\n",
       "      <td>1.342146e-03</td>\n",
       "      <td>1.124383e-02</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>139320.500000</td>\n",
       "      <td>1.315642e+00</td>\n",
       "      <td>8.037239e-01</td>\n",
       "      <td>1.027196e+00</td>\n",
       "      <td>7.433413e-01</td>\n",
       "      <td>6.119264e-01</td>\n",
       "      <td>3.985649e-01</td>\n",
       "      <td>5.704361e-01</td>\n",
       "      <td>3.273459e-01</td>\n",
       "      <td>5.971390e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>1.863772e-01</td>\n",
       "      <td>5.285536e-01</td>\n",
       "      <td>1.476421e-01</td>\n",
       "      <td>4.395266e-01</td>\n",
       "      <td>3.507156e-01</td>\n",
       "      <td>2.409522e-01</td>\n",
       "      <td>9.104512e-02</td>\n",
       "      <td>7.827995e-02</td>\n",
       "      <td>77.165000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>172792.000000</td>\n",
       "      <td>2.454930e+00</td>\n",
       "      <td>2.205773e+01</td>\n",
       "      <td>9.382558e+00</td>\n",
       "      <td>1.687534e+01</td>\n",
       "      <td>3.480167e+01</td>\n",
       "      <td>7.330163e+01</td>\n",
       "      <td>1.205895e+02</td>\n",
       "      <td>2.000721e+01</td>\n",
       "      <td>1.559499e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>2.720284e+01</td>\n",
       "      <td>1.050309e+01</td>\n",
       "      <td>2.252841e+01</td>\n",
       "      <td>4.584549e+00</td>\n",
       "      <td>7.519589e+00</td>\n",
       "      <td>3.517346e+00</td>\n",
       "      <td>3.161220e+01</td>\n",
       "      <td>3.384781e+01</td>\n",
       "      <td>25691.160000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                Time            V1            V2            V3            V4  \\\n",
       "count  284807.000000  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
       "mean    94813.859575  3.919560e-15  5.688174e-16 -8.769071e-15  2.782312e-15   \n",
       "std     47488.145955  1.958696e+00  1.651309e+00  1.516255e+00  1.415869e+00   \n",
       "min         0.000000 -5.640751e+01 -7.271573e+01 -4.832559e+01 -5.683171e+00   \n",
       "25%     54201.500000 -9.203734e-01 -5.985499e-01 -8.903648e-01 -8.486401e-01   \n",
       "50%     84692.000000  1.810880e-02  6.548556e-02  1.798463e-01 -1.984653e-02   \n",
       "75%    139320.500000  1.315642e+00  8.037239e-01  1.027196e+00  7.433413e-01   \n",
       "max    172792.000000  2.454930e+00  2.205773e+01  9.382558e+00  1.687534e+01   \n",
       "\n",
       "                 V5            V6            V7            V8            V9  \\\n",
       "count  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
       "mean  -1.552563e-15  2.010663e-15 -1.694249e-15 -1.927028e-16 -3.137024e-15   \n",
       "std    1.380247e+00  1.332271e+00  1.237094e+00  1.194353e+00  1.098632e+00   \n",
       "min   -1.137433e+02 -2.616051e+01 -4.355724e+01 -7.321672e+01 -1.343407e+01   \n",
       "25%   -6.915971e-01 -7.682956e-01 -5.540759e-01 -2.086297e-01 -6.430976e-01   \n",
       "50%   -5.433583e-02 -2.741871e-01  4.010308e-02  2.235804e-02 -5.142873e-02   \n",
       "75%    6.119264e-01  3.985649e-01  5.704361e-01  3.273459e-01  5.971390e-01   \n",
       "max    3.480167e+01  7.330163e+01  1.205895e+02  2.000721e+01  1.559499e+01   \n",
       "\n",
       "           ...                 V21           V22           V23           V24  \\\n",
       "count      ...        2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
       "mean       ...        1.537294e-16  7.959909e-16  5.367590e-16  4.458112e-15   \n",
       "std        ...        7.345240e-01  7.257016e-01  6.244603e-01  6.056471e-01   \n",
       "min        ...       -3.483038e+01 -1.093314e+01 -4.480774e+01 -2.836627e+00   \n",
       "25%        ...       -2.283949e-01 -5.423504e-01 -1.618463e-01 -3.545861e-01   \n",
       "50%        ...       -2.945017e-02  6.781943e-03 -1.119293e-02  4.097606e-02   \n",
       "75%        ...        1.863772e-01  5.285536e-01  1.476421e-01  4.395266e-01   \n",
       "max        ...        2.720284e+01  1.050309e+01  2.252841e+01  4.584549e+00   \n",
       "\n",
       "                V25           V26           V27           V28         Amount  \\\n",
       "count  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05  284807.000000   \n",
       "mean   1.453003e-15  1.699104e-15 -3.660161e-16 -1.206049e-16      88.349619   \n",
       "std    5.212781e-01  4.822270e-01  4.036325e-01  3.300833e-01     250.120109   \n",
       "min   -1.029540e+01 -2.604551e+00 -2.256568e+01 -1.543008e+01       0.000000   \n",
       "25%   -3.171451e-01 -3.269839e-01 -7.083953e-02 -5.295979e-02       5.600000   \n",
       "50%    1.659350e-02 -5.213911e-02  1.342146e-03  1.124383e-02      22.000000   \n",
       "75%    3.507156e-01  2.409522e-01  9.104512e-02  7.827995e-02      77.165000   \n",
       "max    7.519589e+00  3.517346e+00  3.161220e+01  3.384781e+01   25691.160000   \n",
       "\n",
       "               Class  \n",
       "count  284807.000000  \n",
       "mean        0.001727  \n",
       "std         0.041527  \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 31 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Checking the frequency of frauds before moving forward"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:37.299683Z",
     "start_time": "2019-09-19T18:31:37.284064Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    284315\n",
       "1       492\n",
       "Name: Class, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Class.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:37.612109Z",
     "start_time": "2019-09-19T18:31:37.299683Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1ff9d53ebe0>"
      ]
     },
     "execution_count": 9,
     "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=df.Class, hue=df.Class)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By looking at the above statistics, we can see that the data is highly imbalanced. Only 492 out of 284807 are fraud."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Checking the distribution of amount"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.096369Z",
     "start_time": "2019-09-19T18:31:37.612109Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1ff9d536160>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 5))\n",
    "sns.distplot(df.Amount)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since, it is a little difficult to see. Let's engineer a new feature of bins."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.111991Z",
     "start_time": "2019-09-19T18:31:38.096369Z"
    }
   },
   "outputs": [],
   "source": [
    "df['Amount-Bins'] = ''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's set the bins and their labels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.127613Z",
     "start_time": "2019-09-19T18:31:38.111991Z"
    },
    "run_control": {
     "marked": false
    }
   },
   "outputs": [],
   "source": [
    "def make_bins(predictor, size=50):\n",
    "    '''\n",
    "    Takes the predictor (a series or a dataframe of single predictor) and size of bins\n",
    "    Returns bins and bin labels\n",
    "    '''\n",
    "    bins = np.linspace(predictor.min(), predictor.max(), num=size)\n",
    "\n",
    "    bin_labels = []\n",
    "\n",
    "    # Index of the final element in bins list\n",
    "    bins_last_index = bins.shape[0] - 1\n",
    "\n",
    "    for id, val in enumerate(bins):\n",
    "        if id == bins_last_index:\n",
    "            continue\n",
    "        val_to_put = str(int(bins[id])) + ' to ' + str(int(bins[id + 1]))\n",
    "        bin_labels.append(val_to_put)\n",
    "    \n",
    "    return bins, bin_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.158855Z",
     "start_time": "2019-09-19T18:31:38.127613Z"
    }
   },
   "outputs": [],
   "source": [
    "bins, bin_labels = make_bins(df.Amount, size=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, adding bins in the column Amount-Bins."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.190100Z",
     "start_time": "2019-09-19T18:31:38.158855Z"
    }
   },
   "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>Amount-Bins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0 to 2854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0 to 2854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0 to 2854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0 to 2854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0 to 2854</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Amount-Bins\n",
       "0   0 to 2854\n",
       "1   0 to 2854\n",
       "2   0 to 2854\n",
       "3   0 to 2854\n",
       "4   0 to 2854"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Amount-Bins'] = pd.cut(df.Amount, bins=bins,\n",
    "                           labels=bin_labels, include_lowest=True)\n",
    "df['Amount-Bins'].head().to_frame()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's plot the bins."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.205720Z",
     "start_time": "2019-09-19T18:31:38.190100Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0 to 2854         284484\n",
       "2854 to 5709         285\n",
       "5709 to 8563          28\n",
       "8563 to 11418          4\n",
       "11418 to 14272         3\n",
       "17127 to 19982         2\n",
       "22836 to 25691         1\n",
       "19982 to 22836         0\n",
       "14272 to 17127         0\n",
       "Name: Amount-Bins, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Amount-Bins'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.484871Z",
     "start_time": "2019-09-19T18:31:38.205720Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1, 2, 3, 4, 5, 6, 7, 8]), <a list of 9 Text xticklabel objects>)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 10))\n",
    "sns.countplot(x='Amount-Bins', data=df)\n",
    "plt.xticks(rotation=45)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since, count of values of Bins other than '0 to 2854' are difficult to view. Let's not insert the first one."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.672541Z",
     "start_time": "2019-09-19T18:31:38.486867Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1, 2, 3, 4, 5, 6, 7, 8]), <a list of 9 Text xticklabel objects>)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 10))\n",
    "sns.countplot(x='Amount-Bins', data=df[~(df['Amount-Bins'] == '0 to 2854')])\n",
    "plt.xticks(rotation=45)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that mostly the amount is between 0 and 2854 euros. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predictive Modelling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's predict whether a credit card is fraud or not using machine learning."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## One-hot encoding the Amount-Bins"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since, for classification, we need to pass the data in numerical form. That's why we need to One-Hot encode the Amount-Bins column.<br>\n",
    "```Note: We can also label encode values.```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.813698Z",
     "start_time": "2019-09-19T18:31:38.674535Z"
    }
   },
   "outputs": [],
   "source": [
    "df_encoded = pd.get_dummies(data=df, columns=['Amount-Bins'])\n",
    "df = df_encoded.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.839162Z",
     "start_time": "2019-09-19T18:31:38.813698Z"
    }
   },
   "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>Class</th>\n",
       "      <th>Amount-Bins_0 to 2854</th>\n",
       "      <th>Amount-Bins_2854 to 5709</th>\n",
       "      <th>Amount-Bins_5709 to 8563</th>\n",
       "      <th>Amount-Bins_8563 to 11418</th>\n",
       "      <th>Amount-Bins_11418 to 14272</th>\n",
       "      <th>Amount-Bins_14272 to 17127</th>\n",
       "      <th>Amount-Bins_17127 to 19982</th>\n",
       "      <th>Amount-Bins_19982 to 22836</th>\n",
       "      <th>Amount-Bins_22836 to 25691</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 40 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9             ...              Class  \\\n",
       "0  0.098698  0.363787             ...                  0   \n",
       "1  0.085102 -0.255425             ...                  0   \n",
       "2  0.247676 -1.514654             ...                  0   \n",
       "3  0.377436 -1.387024             ...                  0   \n",
       "4 -0.270533  0.817739             ...                  0   \n",
       "\n",
       "   Amount-Bins_0 to 2854  Amount-Bins_2854 to 5709  Amount-Bins_5709 to 8563  \\\n",
       "0                      1                         0                         0   \n",
       "1                      1                         0                         0   \n",
       "2                      1                         0                         0   \n",
       "3                      1                         0                         0   \n",
       "4                      1                         0                         0   \n",
       "\n",
       "   Amount-Bins_8563 to 11418  Amount-Bins_11418 to 14272  \\\n",
       "0                          0                           0   \n",
       "1                          0                           0   \n",
       "2                          0                           0   \n",
       "3                          0                           0   \n",
       "4                          0                           0   \n",
       "\n",
       "   Amount-Bins_14272 to 17127  Amount-Bins_17127 to 19982  \\\n",
       "0                           0                           0   \n",
       "1                           0                           0   \n",
       "2                           0                           0   \n",
       "3                           0                           0   \n",
       "4                           0                           0   \n",
       "\n",
       "   Amount-Bins_19982 to 22836  Amount-Bins_22836 to 25691  \n",
       "0                           0                           0  \n",
       "1                           0                           0  \n",
       "2                           0                           0  \n",
       "3                           0                           0  \n",
       "4                           0                           0  \n",
       "\n",
       "[5 rows x 40 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Breaking the dataset into training and testing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, separating the response variable from the explanatory variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:38.908684Z",
     "start_time": "2019-09-19T18:31:38.841155Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((284807, 39), (284807,))"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(labels='Class', axis=1)\n",
    "Y = df['Class']\n",
    "\n",
    "X.shape, Y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:41.110543Z",
     "start_time": "2019-09-19T18:31:38.915665Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:41.221404Z",
     "start_time": "2019-09-19T18:31:41.110543Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(199364, 39) (199364,)\n",
      "(85443, 39) (85443,)\n"
     ]
    }
   ],
   "source": [
    "xtrain, xtest, ytrain, ytest = train_test_split(\n",
    "    X, Y, random_state=42, test_size=0.3, shuffle=True)\n",
    "\n",
    "print(xtrain.shape, ytrain.shape)\n",
    "print(xtest.shape, ytest.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Applying Machine Learning Algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's apply different Machine Learning Algorithms then compare their metrics to select the most suitable ML algorithm. Algorithms to be used are:\n",
    "* Logistic Regression\n",
    "* Support Vector Machine\n",
    "* Naive Bayes\n",
    "* K-Nearest Neighbors\n",
    "* Random Forest\n",
    "* Ada Boost\n",
    "* XGBoost\n",
    "\n",
    "The metrics we'll use initially are:\n",
    "* Accuracy\n",
    "* Precision\n",
    "* F1-Score\n",
    "\n",
    "The main metrics we'll look at are (Reason is mentioned later):\n",
    "* Recall\n",
    "* AUC/RUC Curve"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:41.514920Z",
     "start_time": "2019-09-19T18:31:41.223399Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "lr_model = LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.289958Z",
     "start_time": "2019-09-19T18:31:41.514920Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='warn',\n",
       "          n_jobs=None, penalty='l2', random_state=None, solver='warn',\n",
       "          tol=0.0001, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Training the algorithm\n",
    "lr_model.fit(xtrain, ytrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.385169Z",
     "start_time": "2019-09-19T18:31:47.291955Z"
    }
   },
   "outputs": [],
   "source": [
    "# Predictions on training and testing data\n",
    "lr_pred_train = lr_model.predict(xtrain)\n",
    "lr_pred_test = lr_model.predict(xtest)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before going further into metrics, let's first decide either Type-I or Type-II error is more important to consider.<br><br>\n",
    "```Type-I  Error or False Positives:``` False Positives are the ones which are actually not fraud but the prediction said that they are fraud.<br>\n",
    "```Type-II Error or False Negatives:``` False Negatives are the ones which are actually fraud but the system said that they aren't.\n",
    "\n",
    "Well, we can say that Type-II Error is more significant because we don't want system to have a fraudulent credit card because that can be more dangerous.\n",
    "\n",
    "So, for Type-II Error, We can say that **recall** is the important metric."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.393150Z",
     "start_time": "2019-09-19T18:31:47.387167Z"
    }
   },
   "outputs": [],
   "source": [
    "# Importing the required metrics\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, confusion_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's first look at the **confusion matrix**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.469889Z",
     "start_time": "2019-09-19T18:31:47.395144Z"
    }
   },
   "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>Predicted Fraud</th>\n",
       "      <th>Predicted Not Fraud</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Fraud</th>\n",
       "      <td>82</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Not Fraud</th>\n",
       "      <td>13</td>\n",
       "      <td>85294</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Predicted Fraud  Predicted Not Fraud\n",
       "Fraud                   82                   54\n",
       "Not Fraud               13                85294"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tn, fp, fn, tp = confusion_matrix(ytest, lr_pred_test).ravel()\n",
    "conf_matrix = pd.DataFrame(\n",
    "    {\n",
    "        'Predicted Fraud': [tp, fp],\n",
    "        'Predicted Not Fraud': [fn, tn]\n",
    "    }, index=['Fraud', 'Not Fraud'])\n",
    "conf_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's draw a heatmap for the above confusion matrix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.629488Z",
     "start_time": "2019-09-19T18:31:47.469889Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1ffa0a89320>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(conf_matrix, annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Heatmap also suggests that the data is highly imbalanced."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the accuracy score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.643450Z",
     "start_time": "2019-09-19T18:31:47.631483Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9992158515033414"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_accuracy = accuracy_score(ytest, lr_pred_test)\n",
    "lr_accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see here that accuracy is great. Around 99%.<br>\n",
    "```BUT WAIT```<br>\n",
    "We know that the dataset is highly unbalanced and accuracy takes into account the whole confusion matrix. So we can say that this measure is not suitable."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at precision and recall."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.670737Z",
     "start_time": "2019-09-19T18:31:47.645446Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8631578947368421"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_precision = precision_score(ytest, lr_pred_test)\n",
    "lr_precision"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.693702Z",
     "start_time": "2019-09-19T18:31:47.672183Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6029411764705882"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_recall = recall_score(ytest, lr_pred_test)\n",
    "lr_recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall is very low in case of logistic regression. However, we may try to increase it by increasing the complexity of the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although, let's check the recall for training dataset to get the idea of any overfitting we may be having."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.739917Z",
     "start_time": "2019-09-19T18:31:47.694697Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6376404494382022"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_recall_train = recall_score(ytrain, lr_pred_train)\n",
    "lr_recall_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Well, we can see that the delta is small, only around 0.03. So, we can say that the model is not overfitting."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the F1-Score. F1-Score may tell us that one of the precision or recall is very low."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.748903Z",
     "start_time": "2019-09-19T18:31:47.741918Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.783545Z",
     "start_time": "2019-09-19T18:31:47.750888Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7099567099567099"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_f1 = f1_score(ytest, lr_pred_test)\n",
    "lr_f1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the classification report."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.788531Z",
     "start_time": "2019-09-19T18:31:47.784542Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.863424Z",
     "start_time": "2019-09-19T18:31:47.790526Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     85307\n",
      "           1       0.86      0.60      0.71       136\n",
      "\n",
      "   micro avg       1.00      1.00      1.00     85443\n",
      "   macro avg       0.93      0.80      0.85     85443\n",
      "weighted avg       1.00      1.00      1.00     85443\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(ytest, lr_pred_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the ROC curve."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, for the ROC Curve, we need the probabilites of Fraud happening (which is the probability of occurance of 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.898520Z",
     "start_time": "2019-09-19T18:31:47.864421Z"
    }
   },
   "outputs": [],
   "source": [
    "lr_pred_test_prob = lr_model.predict_proba(xtest)[:, 1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, to draw the ROC Curve, we need to have ```True Positive Rate``` and ```False Positive Rate```."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.914142Z",
     "start_time": "2019-09-19T18:31:47.898520Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve, roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.945384Z",
     "start_time": "2019-09-19T18:31:47.914142Z"
    }
   },
   "outputs": [],
   "source": [
    "fpr, tpr, threshold = roc_curve(ytest, lr_pred_test_prob)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also, let's get the auc score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.981106Z",
     "start_time": "2019-09-19T18:31:47.945384Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9646767143445232"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_auc = roc_auc_score(ytest, lr_pred_test_prob)\n",
    "lr_auc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's define a function to plot the roc curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:47.988087Z",
     "start_time": "2019-09-19T18:31:47.983100Z"
    }
   },
   "outputs": [],
   "source": [
    "def plot_roc_curve(fpr, tpr, label=None):\n",
    "    plt.figure(figsize=(8, 6))\n",
    "    plt.title('ROC Curve', fontsize=15)\n",
    "    plt.plot([0, 1], [0, 1], 'k--')\n",
    "    plt.plot(fpr, tpr, linewidth=2, label=label)\n",
    "    plt.xticks(np.arange(0, 1, 0.05), rotation=90)\n",
    "    plt.xlabel('False Positive Rates', fontsize=15)\n",
    "    plt.ylabel('True Positive Rates', fontsize=15)\n",
    "    plt.legend(loc='best')\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting ROC Curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:48.310330Z",
     "start_time": "2019-09-19T18:31:47.992077Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_roc_curve(fpr=fpr, tpr=tpr, label=\"AUC = %.3f\" % lr_auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "AUC is quite good. i.e. 0.965. Based on the data being highly imbalanced, we'll only check the AUC metric in later algorithms."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model Complexity"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's try to train the Logistic Regression models on the 2nd degree of polynomials. Not going further 2nd degree because features are already too much. Otherwise, computer gives the MemoryError."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:31:48.325959Z",
     "start_time": "2019-09-19T18:31:48.310330Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import PolynomialFeatures"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:52.095604Z",
     "start_time": "2019-09-19T18:31:48.325959Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.9089129413350895, 0.9033804549519763)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Getting the polynomial features\n",
    "poly = PolynomialFeatures(degree=2)\n",
    "xtrain_poly = poly.fit_transform(xtrain)\n",
    "xtest_poly = poly.fit_transform(xtest)\n",
    "\n",
    "# Training the model\n",
    "model = LogisticRegression()\n",
    "model.fit(xtrain_poly, ytrain)\n",
    "\n",
    "# Getting the probabilities\n",
    "train_prob = model.predict_proba(xtrain_poly)[:, 1]\n",
    "test_prob = model.predict_proba(xtest_poly)[:, 1]\n",
    "\n",
    "# Computing the ROC Score\n",
    "roc_auc_score(ytrain, train_prob), roc_auc_score(ytest, test_prob)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting ROC Curve for the Test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:52.136492Z",
     "start_time": "2019-09-19T18:32:52.103581Z"
    }
   },
   "outputs": [],
   "source": [
    "fpr_poly, tpr_poly, threshold_poly = roc_curve(ytest, test_prob)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:52.609442Z",
     "start_time": "2019-09-19T18:32:52.138488Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_roc_curve(fpr=fpr_poly, tpr=tpr_poly, label='AUC = %.3f' %  roc_auc_score(ytest, test_prob))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First degree is better in Logistic Regression case which gives 0.965 AUC Score."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's also check the Recall in case of model complexity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:52.779543Z",
     "start_time": "2019-09-19T18:32:52.611435Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7426470588235294"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(ytest, model.predict(xtest_poly))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall has increased when the model is made complex."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Support Vector Machine"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's try the Support Vector Machine algorithm."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, for support vector machines, we need to train the model after scaling the features. Let's first do that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:52.785526Z",
     "start_time": "2019-09-19T18:32:52.781536Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:52.799488Z",
     "start_time": "2019-09-19T18:32:52.787521Z"
    }
   },
   "outputs": [],
   "source": [
    "mms = MinMaxScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:52.855478Z",
     "start_time": "2019-09-19T18:32:52.801484Z"
    }
   },
   "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Amount-Bins_0 to 2854</th>\n",
       "      <th>Amount-Bins_2854 to 5709</th>\n",
       "      <th>Amount-Bins_5709 to 8563</th>\n",
       "      <th>Amount-Bins_8563 to 11418</th>\n",
       "      <th>Amount-Bins_11418 to 14272</th>\n",
       "      <th>Amount-Bins_14272 to 17127</th>\n",
       "      <th>Amount-Bins_17127 to 19982</th>\n",
       "      <th>Amount-Bins_19982 to 22836</th>\n",
       "      <th>Amount-Bins_22836 to 25691</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>149.62</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>2.69</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>378.66</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>123.50</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>69.99</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9             ...              Amount  \\\n",
       "0  0.098698  0.363787             ...              149.62   \n",
       "1  0.085102 -0.255425             ...                2.69   \n",
       "2  0.247676 -1.514654             ...              378.66   \n",
       "3  0.377436 -1.387024             ...              123.50   \n",
       "4 -0.270533  0.817739             ...               69.99   \n",
       "\n",
       "   Amount-Bins_0 to 2854  Amount-Bins_2854 to 5709  Amount-Bins_5709 to 8563  \\\n",
       "0                      1                         0                         0   \n",
       "1                      1                         0                         0   \n",
       "2                      1                         0                         0   \n",
       "3                      1                         0                         0   \n",
       "4                      1                         0                         0   \n",
       "\n",
       "   Amount-Bins_8563 to 11418  Amount-Bins_11418 to 14272  \\\n",
       "0                          0                           0   \n",
       "1                          0                           0   \n",
       "2                          0                           0   \n",
       "3                          0                           0   \n",
       "4                          0                           0   \n",
       "\n",
       "   Amount-Bins_14272 to 17127  Amount-Bins_17127 to 19982  \\\n",
       "0                           0                           0   \n",
       "1                           0                           0   \n",
       "2                           0                           0   \n",
       "3                           0                           0   \n",
       "4                           0                           0   \n",
       "\n",
       "   Amount-Bins_19982 to 22836  Amount-Bins_22836 to 25691  \n",
       "0                           0                           0  \n",
       "1                           0                           0  \n",
       "2                           0                           0  \n",
       "3                           0                           0  \n",
       "4                           0                           0  \n",
       "\n",
       "[5 rows x 39 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's first check the head of the explanatory variables which are to be scaled.\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:53.252047Z",
     "start_time": "2019-09-19T18:32:52.857473Z"
    }
   },
   "outputs": [],
   "source": [
    "X_scaled = mms.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:53.298912Z",
     "start_time": "2019-09-19T18:32:53.252047Z"
    }
   },
   "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Amount-Bins_0 to 2854</th>\n",
       "      <th>Amount-Bins_2854 to 5709</th>\n",
       "      <th>Amount-Bins_5709 to 8563</th>\n",
       "      <th>Amount-Bins_8563 to 11418</th>\n",
       "      <th>Amount-Bins_11418 to 14272</th>\n",
       "      <th>Amount-Bins_14272 to 17127</th>\n",
       "      <th>Amount-Bins_17127 to 19982</th>\n",
       "      <th>Amount-Bins_19982 to 22836</th>\n",
       "      <th>Amount-Bins_22836 to 25691</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.935192</td>\n",
       "      <td>0.766490</td>\n",
       "      <td>0.881365</td>\n",
       "      <td>0.313023</td>\n",
       "      <td>0.763439</td>\n",
       "      <td>0.267669</td>\n",
       "      <td>0.266815</td>\n",
       "      <td>0.786444</td>\n",
       "      <td>0.475312</td>\n",
       "      <td>...</td>\n",
       "      <td>0.005824</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.978542</td>\n",
       "      <td>0.770067</td>\n",
       "      <td>0.840298</td>\n",
       "      <td>0.271796</td>\n",
       "      <td>0.766120</td>\n",
       "      <td>0.262192</td>\n",
       "      <td>0.264875</td>\n",
       "      <td>0.786298</td>\n",
       "      <td>0.453981</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.935217</td>\n",
       "      <td>0.753118</td>\n",
       "      <td>0.868141</td>\n",
       "      <td>0.268766</td>\n",
       "      <td>0.762329</td>\n",
       "      <td>0.281122</td>\n",
       "      <td>0.270177</td>\n",
       "      <td>0.788042</td>\n",
       "      <td>0.410603</td>\n",
       "      <td>...</td>\n",
       "      <td>0.014739</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.941878</td>\n",
       "      <td>0.765304</td>\n",
       "      <td>0.868484</td>\n",
       "      <td>0.213661</td>\n",
       "      <td>0.765647</td>\n",
       "      <td>0.275559</td>\n",
       "      <td>0.266803</td>\n",
       "      <td>0.789434</td>\n",
       "      <td>0.414999</td>\n",
       "      <td>...</td>\n",
       "      <td>0.004807</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.938617</td>\n",
       "      <td>0.776520</td>\n",
       "      <td>0.864251</td>\n",
       "      <td>0.269796</td>\n",
       "      <td>0.762975</td>\n",
       "      <td>0.263984</td>\n",
       "      <td>0.268968</td>\n",
       "      <td>0.782484</td>\n",
       "      <td>0.490950</td>\n",
       "      <td>...</td>\n",
       "      <td>0.002724</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Time        V1        V2        V3        V4        V5        V6  \\\n",
       "0  0.000000  0.935192  0.766490  0.881365  0.313023  0.763439  0.267669   \n",
       "1  0.000000  0.978542  0.770067  0.840298  0.271796  0.766120  0.262192   \n",
       "2  0.000006  0.935217  0.753118  0.868141  0.268766  0.762329  0.281122   \n",
       "3  0.000006  0.941878  0.765304  0.868484  0.213661  0.765647  0.275559   \n",
       "4  0.000012  0.938617  0.776520  0.864251  0.269796  0.762975  0.263984   \n",
       "\n",
       "         V7        V8        V9             ...                Amount  \\\n",
       "0  0.266815  0.786444  0.475312             ...              0.005824   \n",
       "1  0.264875  0.786298  0.453981             ...              0.000105   \n",
       "2  0.270177  0.788042  0.410603             ...              0.014739   \n",
       "3  0.266803  0.789434  0.414999             ...              0.004807   \n",
       "4  0.268968  0.782484  0.490950             ...              0.002724   \n",
       "\n",
       "   Amount-Bins_0 to 2854  Amount-Bins_2854 to 5709  Amount-Bins_5709 to 8563  \\\n",
       "0                    1.0                       0.0                       0.0   \n",
       "1                    1.0                       0.0                       0.0   \n",
       "2                    1.0                       0.0                       0.0   \n",
       "3                    1.0                       0.0                       0.0   \n",
       "4                    1.0                       0.0                       0.0   \n",
       "\n",
       "   Amount-Bins_8563 to 11418  Amount-Bins_11418 to 14272  \\\n",
       "0                        0.0                         0.0   \n",
       "1                        0.0                         0.0   \n",
       "2                        0.0                         0.0   \n",
       "3                        0.0                         0.0   \n",
       "4                        0.0                         0.0   \n",
       "\n",
       "   Amount-Bins_14272 to 17127  Amount-Bins_17127 to 19982  \\\n",
       "0                         0.0                         0.0   \n",
       "1                         0.0                         0.0   \n",
       "2                         0.0                         0.0   \n",
       "3                         0.0                         0.0   \n",
       "4                         0.0                         0.0   \n",
       "\n",
       "   Amount-Bins_19982 to 22836  Amount-Bins_22836 to 25691  \n",
       "0                         0.0                         0.0  \n",
       "1                         0.0                         0.0  \n",
       "2                         0.0                         0.0  \n",
       "3                         0.0                         0.0  \n",
       "4                         0.0                         0.0  \n",
       "\n",
       "[5 rows x 39 columns]"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_scaled = pd.DataFrame(data=X_scaled, columns=X.columns)\n",
    "X_scaled.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's train test split on the scaled data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:53.533230Z",
     "start_time": "2019-09-19T18:32:53.298912Z"
    }
   },
   "outputs": [],
   "source": [
    "xtrainS, xtestS, ytrainS, ytestS = train_test_split(\n",
    "    X_scaled, Y, random_state=42, test_size=0.30, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:53.548858Z",
     "start_time": "2019-09-19T18:32:53.533230Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(199364, 39) (199364,)\n",
      "(85443, 39) (85443,)\n"
     ]
    }
   ],
   "source": [
    "print(xtrainS.shape, ytrainS.shape)\n",
    "print(xtestS.shape, ytestS.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:53.564473Z",
     "start_time": "2019-09-19T18:32:53.548858Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:32:53.580095Z",
     "start_time": "2019-09-19T18:32:53.564473Z"
    },
    "run_control": {
     "marked": false
    }
   },
   "outputs": [],
   "source": [
    "svc_model = SVC(kernel='linear', probability=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:02.474056Z",
     "start_time": "2019-09-19T18:32:53.580095Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma='auto_deprecated',\n",
       "  kernel='linear', max_iter=-1, probability=True, random_state=None,\n",
       "  shrinking=True, tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_model.fit(xtrainS, ytrainS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:03.728275Z",
     "start_time": "2019-09-19T18:34:02.476201Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_pred = svc_model.predict(xtestS)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's first check the recall score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:03.759176Z",
     "start_time": "2019-09-19T18:34:03.728275Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_recall = recall_score(ytestS, svc_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:03.781311Z",
     "start_time": "2019-09-19T18:34:03.759176Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8014705882352942"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall quite increased in case of SVC."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:05.015960Z",
     "start_time": "2019-09-19T18:34:03.781311Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_pred_prob = svc_model.predict_proba(xtestS)[:, 1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's draw the ROC Curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:05.186078Z",
     "start_time": "2019-09-19T18:34:05.015960Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# First, getting the auc score\n",
    "svc_auc = roc_auc_score(ytestS, svc_pred_prob)\n",
    "\n",
    "# Now, let's get the fpr and tpr\n",
    "fpr, tpr, threshold = roc_curve(ytestS, svc_pred_prob)\n",
    "\n",
    "# Now, let's draw the curve\n",
    "plot_roc_curve(fpr, tpr, 'AUC: %.3f' % svc_auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The score AUC Score SVC gave is also pretty great. But it's still less than Logistic Regression Model. But the Recall increased significantly."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Tuning the Hyper-parameters"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's tune some of the hyper-parameters of SVM and then compare the scores."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:05.194015Z",
     "start_time": "2019-09-19T18:34:05.188033Z"
    }
   },
   "outputs": [],
   "source": [
    "# For Kernel = rbf\n",
    "tuned_rbf = {'kernel': ['rbf'], 'gamma': [\n",
    "    1e-2, 1e-3, 1e-4, 1e-5], 'C': [0.001, 0.10, 0.1, 10, 25, 50, 100, 1000]}\n",
    "\n",
    "# For kernel = sigmoid\n",
    "tuned_sigmoid = {'kernel': ['sigmoid'], 'gamma': [\n",
    "    1e-2, 1e-3, 1e-4, 1e-5], 'C': [0.001, 0.10, 0.1, 10, 25, 50, 100, 1000]}\n",
    "\n",
    "# For kernel = linear\n",
    "tuned_linear = {'kernel': ['linear'], 'C': [\n",
    "    0.001, 0.10, 0.1, 10, 25, 50, 100, 1000]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:05.210972Z",
     "start_time": "2019-09-19T18:34:05.196011Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T18:34:05.223937Z",
     "start_time": "2019-09-19T18:34:05.212965Z"
    }
   },
   "outputs": [],
   "source": [
    "rs_rbf = RandomizedSearchCV(estimator=SVC(probability=True), \n",
    "        param_distributions=tuned_rbf, n_iter=500, n_jobs=4, scoring='roc_auc')\n",
    "\n",
    "rs_sigmoid = RandomizedSearchCV(estimator=SVC(probability=True), \n",
    "        param_distributions=tuned_sigmoid, n_iter=500, n_jobs=4, scoring='roc_auc')\n",
    "\n",
    "rs_linear = RandomizedSearchCV(estimator=SVC(probability=True), \n",
    "        param_distributions=tuned_linear, n_iter=500, n_jobs=4, scoring='roc_auc')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**For kernel rbf:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:20:42.469945Z",
     "start_time": "2019-09-19T18:34:05.225930Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv='warn', error_score='raise-deprecating',\n",
       "          estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma='auto_deprecated',\n",
       "  kernel='rbf', max_iter=-1, probability=True, random_state=None,\n",
       "  shrinking=True, tol=0.001, verbose=False),\n",
       "          fit_params=None, iid='warn', n_iter=500, n_jobs=4,\n",
       "          param_distributions={'kernel': ['rbf'], 'gamma': [0.01, 0.001, 0.0001, 1e-05], 'C': [0.001, 0.1, 0.1, 10, 25, 50, 100, 1000]},\n",
       "          pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "          return_train_score='warn', scoring='roc_auc', verbose=0)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rs_rbf.fit(xtrainS, ytrainS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:20:42.489678Z",
     "start_time": "2019-09-19T19:20:42.479581Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1000, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.01, kernel='rbf',\n",
       "  max_iter=-1, probability=True, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rs_rbf.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:20:42.529758Z",
     "start_time": "2019-09-19T19:20:42.489678Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_rbf_best_est = rs_rbf.best_estimator_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's fit the model on the best rbf estimator."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:22:42.327793Z",
     "start_time": "2019-09-19T19:20:42.529758Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1000, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.01, kernel='rbf',\n",
       "  max_iter=-1, probability=True, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_rbf_best_est.fit(xtrainS, ytrainS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:22:44.304763Z",
     "start_time": "2019-09-19T19:22:42.329787Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_rbf_best_est_pred = svc_rbf_best_est.predict(xtestS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:22:46.175032Z",
     "start_time": "2019-09-19T19:22:44.304763Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_rbf_best_est_pred_proba = svc_rbf_best_est.predict_proba(xtestS)[:, 1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Getting the AUC Score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:22:46.225197Z",
     "start_time": "2019-09-19T19:22:46.175032Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_rbf_auc = roc_auc_score(ytestS, svc_rbf_best_est_pred_proba)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Getting the Recall too."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:22:46.260106Z",
     "start_time": "2019-09-19T19:22:46.230182Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8308823529411765"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_rbf_recall = recall_score(ytestS, svc_rbf_best_est_pred)\n",
    "svc_rbf_recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that in this model, both recall and ROC Score are great. Let's draw the ROC Curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T19:22:46.559964Z",
     "start_time": "2019-09-19T19:22:46.266088Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, threshold = roc_curve(ytestS, svc_rbf_best_est_pred_proba)\n",
    "plot_roc_curve(fpr, tpr, 'AUC = %.3f' % svc_rbf_auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, for kernel sigmoid."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:08:53.178201Z",
     "start_time": "2019-09-19T19:22:46.562956Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv='warn', error_score='raise-deprecating',\n",
       "          estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma='auto_deprecated',\n",
       "  kernel='rbf', max_iter=-1, probability=True, random_state=None,\n",
       "  shrinking=True, tol=0.001, verbose=False),\n",
       "          fit_params=None, iid='warn', n_iter=500, n_jobs=4,\n",
       "          param_distributions={'kernel': ['sigmoid'], 'gamma': [0.01, 0.001, 0.0001, 1e-05], 'C': [0.001, 0.1, 0.1, 10, 25, 50, 100, 1000]},\n",
       "          pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "          return_train_score='warn', scoring='roc_auc', verbose=0)"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rs_sigmoid.fit(xtrainS, ytrainS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:08:53.198537Z",
     "start_time": "2019-09-19T20:08:53.183208Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_sigmoid = rs_sigmoid.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:10:14.025945Z",
     "start_time": "2019-09-19T20:08:53.203556Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1000, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.001, kernel='sigmoid',\n",
       "  max_iter=-1, probability=True, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_sigmoid.fit(xtrainS, ytrainS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:10:17.653100Z",
     "start_time": "2019-09-19T20:10:14.030931Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_sigmoid_pred = svc_sigmoid.predict(xtestS)\n",
    "svc_sigmoid_pred_proba = svc_sigmoid.predict_proba(xtestS)[:, 1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "AUC:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:10:17.695173Z",
     "start_time": "2019-09-19T20:10:17.653100Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9618103369215271"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_sigmoid_auc = roc_auc_score(ytestS, svc_sigmoid_pred_proba)\n",
    "svc_sigmoid_auc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:10:17.732910Z",
     "start_time": "2019-09-19T20:10:17.698165Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8014705882352942"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_sigmoid_recall = recall_score(ytestS, svc_sigmoid_pred)\n",
    "svc_sigmoid_recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:10:17.932864Z",
     "start_time": "2019-09-19T20:10:17.732910Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, threshold = roc_curve(ytestS, svc_sigmoid_pred_proba)\n",
    "plot_roc_curve(fpr, tpr, 'AUC = %.3f' % svc_sigmoid_auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check for Linear kernel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:59:56.290644Z",
     "start_time": "2019-09-19T20:10:17.937871Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv='warn', error_score='raise-deprecating',\n",
       "          estimator=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma='auto_deprecated',\n",
       "  kernel='rbf', max_iter=-1, probability=True, random_state=None,\n",
       "  shrinking=True, tol=0.001, verbose=False),\n",
       "          fit_params=None, iid='warn', n_iter=500, n_jobs=4,\n",
       "          param_distributions={'kernel': ['linear'], 'C': [0.001, 0.1, 0.1, 10, 25, 50, 100, 1000]},\n",
       "          pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "          return_train_score='warn', scoring='roc_auc', verbose=0)"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rs_linear.fit(xtrainS, ytrainS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T20:59:56.304607Z",
     "start_time": "2019-09-19T20:59:56.298622Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_linear = rs_linear.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:11.896073Z",
     "start_time": "2019-09-19T20:59:56.308595Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=0.1, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma='auto_deprecated',\n",
       "  kernel='linear', max_iter=-1, probability=True, random_state=None,\n",
       "  shrinking=True, tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_linear.fit(xtrainS, ytrainS)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Getting the predictions and probabilities."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:15.311062Z",
     "start_time": "2019-09-19T21:01:11.899066Z"
    }
   },
   "outputs": [],
   "source": [
    "svc_linear_pred = svc_linear.predict(xtestS)\n",
    "svc_linear_pred_proba = svc_linear.predict_proba(xtestS)[:, 1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "AUC and ROC Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:15.564776Z",
     "start_time": "2019-09-19T21:01:15.317012Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "svc_linear_auc = roc_auc_score(ytestS, svc_linear_pred_proba)\n",
    "\n",
    "fpr, tpr, threshold = roc_curve(ytestS, svc_linear_pred_proba)\n",
    "plot_roc_curve(fpr, tpr, 'AUC = %.3f' % svc_linear_auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check the recall too."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:15.599684Z",
     "start_time": "2019-09-19T21:01:15.569764Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6102941176470589"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_linear_recall = recall_score(ytestS, svc_linear_pred)\n",
    "svc_linear_recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "AUC is great in case of a linear kernel however it's less than that of rbf kernel. And its recall decreased quite a bit."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Naive Bayes Algorithm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's try the famous Naive Bayes Machine Learning Algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:15.636584Z",
     "start_time": "2019-09-19T21:01:15.602675Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import GaussianNB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:15.643566Z",
     "start_time": "2019-09-19T21:01:15.639577Z"
    },
    "run_control": {
     "marked": false
    }
   },
   "outputs": [],
   "source": [
    "nb = GaussianNB()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's first train the algorithm on the default settings."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:16.362727Z",
     "start_time": "2019-09-19T21:01:15.646558Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianNB(priors=None, var_smoothing=1e-09)"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nb.fit(xtrain, ytrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:16.810476Z",
     "start_time": "2019-09-19T21:01:16.364723Z"
    }
   },
   "outputs": [],
   "source": [
    "nb_pred = nb.predict(xtest)\n",
    "nb_pred_proba = nb.predict_proba(xtest)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:16.835351Z",
     "start_time": "2019-09-19T21:01:16.812411Z"
    }
   },
   "outputs": [],
   "source": [
    "nb_auc = roc_auc_score(ytest, nb_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:17.012463Z",
     "start_time": "2019-09-19T21:01:16.841408Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAGWCAYAAABy/1oRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzs3Xd8U2X7x/FP08EoVFYpG1T0RlDBhcoPFRkOeBzIRhkiouIEVBBBwIE8MgQVHh8FFJQhuBXBjfo4ceG+BREQZLdQVmfy+yNpKRhKWpKcNPm+Xy9eJCfnXLlaaK/c97lHnMfjQURERKKfy+kEREREJDxU9EVERGKEir6IiEiMUNEXERGJESr6IiIiMUJFX0REJEYkOJ2AiATOGLMWaHjI4f3AOuApa+2jh5xfGbgL6A40AnYCnwETrbWf+4l/hu/884EqwFpgLvCotTb7CLmV+loRCQ+19EXKnn8DtYv8OQt4D5hijOlRcJIxphrwOdATGAM0AS4DdgAfG2OuLRrUGNPTd/4u4ErgZOB+4DbgFWPMYX9fHM21IhI+aumLlD17rLWbizzfDNxqjLkU6AG84Dv+OFAZOM1am+47thZYYYzZDMwwxvzPWrvKGFMHmAlMtdbeXST2GmPMn8AXQLcisQsdzbUiEl4q+iLRIwfIBTDG1MD7AeDOIgW/qAfxtsKvB+4GrgbKA+MPPdFa+6Ux5kLgu8O8b8DXGmOeBepZa9sXnFP0mDGmDbDMl99QYCXe2xJzrLVji1xzN3CL7zWAEcANQA3gF2CMtfatw+QrErPU5SZSxhljKhpj7gJOAub5Dp8FxOO9f/8PvnvsnwOtfIfOAH6z1u48zPnLrbW7DpPC0VzrTzngQqAl3g8mzwG9DjmnN/CctdYNPAxcCwwCmgNzgJd9HyBEpAi19EXKntHGmBG+x3F4W9k/AL2sta/7jlf1/b2jmDg7gGOLnO+3aAfgaK49nEestasBjDH78H7Np1trvzXGNMNb3HsYYyoBtwNdrLVv+659whjTHLgHWB7kvETKNLX0Rcqe6UALvC3sccA+4Blr7cIi5xQU+5Ri4lQBtvkebweqlTKfo7n2cNYUPLDW/gF8yoHW/tXAl9Zai7d3oxyw2Bizp+AP0Nf3mogUoaIvUvakW2tXW69H8Bb+acaYol3gX+G9x9/aXwBjTBJwNt4ufnx/G2NM1cOcP8cYM/gw+RzNteC/x3H/Ic/nAD19swB64Z0KCN6vEeAqvB+ECv40Ay4o5j1FYpKKvkjZNxn4H97R+LUArLUZeAvlXcaY6n6uuQuoBDzte74Y2Iu3S/wgxpj/w9tyzjzM+5fk2hz+2ftwwuG+sCIWAdWBm4E6QEGvxiq8gxfr+T4IrfbdFrga731+ESlC9/RFyjhrrccYcz3wPfAY3oV4AO4ETgU+N8aMBFbgvf8+ALgJuMFa+7svxhZjzK3AbGNMCt4PA5lAG7yj8l8DFhzm/Uty7efAAN+8/i/xfiA4hcMMOCzyHruMMa8BDwFvFsxIsNbuM8ZMAR42xmQCXwP/Au4Drjvyd08ktqilLxIFrLW/4S2w3Ywxl/uOFRTemXgX5/kVeBtvS/k8a+3sQ2LMAToADYC38E6Xuw2YAPSw1uYX8/6BXvs8MMP3ZyVQH5ga4Jc5B++6A3MPOT4K+A8wyfc1FnygeTbAuCIxI87j8Tidg4iIiISBWvoiIiIxQkVfREQkRqjoi4iIxAgVfRERkRgRLVP2yuFda3wTcNgRxiIiIlEkHu/22iuA7EAuiJaifxbwidNJiIiIOOA8vAt0HVG0FP1NABkZe3G7gzMFsXr1SuzYsScosUIVM9LjhSJmpMcLRcxYixeKmJEeLxQxYy1eKGKGIsdgcrniqFo1GXw1MBDRUvTzAdxuT9CKfkG8YAt2zEiPF4qYkR4vFDFjLV4oYkZ6vFDEjLV4oYgZihxDIODb2hrIJyIiEiNU9EVERGJEtHTvH1Z+fh4ZGdvIy8s58slFbN3qwu12BzWXYMeM9HjFxUxISKJq1VTi46P+v6CISMSI+t+4GRnbKF++IsnJtYiLiwv4uoQEF3l5wS2AwY4Z6fEOF9Pj8bB3byYZGduoUaN2UN9PREQOL+q79/PyckhOTilRwZfQiouLIzk5pcS9LyIicnSivugDKvgRSP8mIiLhFxNFX0RERFT0HbNmzWpatz6T5cvfLzx2yy2D+Pbbrw8676GHxvLWW28A3nvhCxc+T//+venfvzd9+/bmvffeLnUOubm5PPDAaK6+uisDBlzN2rV/+j1v/vy5XH11V665phsLFz5fePy1116mT5/u9O3bg/Hjx5GbmwuAtb8xcGBf+vXrxbBht7F79+5S5ygiIsET9qJvjEkxxvxkjGnk57UWxpivjTG/G2NmGmOidqDhkiWvc+GF7XnttZcDvuapp2awYsVXPPHEUzz77HwmTXrUd+zLUuWwePFCypevwLx5L3LbbcN44IEx/zhnw4a/eOWVF5k9+3mefnouixcvZMOGv1i/fh0LFjzHk0/OZs6chbjdbl5+eREA06ZN4rrrbmDOnAU0aNCIBQueK1V+IiISXGEtqsaYs4GngRMPc8rzwEBr7RfGmFnA9cB/wpVfuOTl5fHOO8uYPv1pbrppABs3bqBu3XrFXrNv3z4WLZrPnDkLqVSpEgA1a6Yxbtx4ypUrf9C5W7ZsZvjwof+IMWPG01SsmFz4/PPP/8fAgTcC0KLF6ezcuZPNmzdTq1atwnPcbje5ublkZ+cAHjweDwkJ3v82w4YNJznZm8txxzVmy5bNhdfs27cXgKysLCpVqlySb4+IiIRIuFvS1wM3A/9o+hljGgIVrLVf+A49C4wjiEV/6uKV/PDHjmCFO8ipx1fnjm7NAzr3s8/+R61atWjQoCHnndeG1157mcGDbyv2mvXr15KQkEC9evUPOn7SSc3+cW5aWi2efXb+EfPYvn0b1avXKHxevXoNtm3bclDRb9CgIe3bX0y3bpfhdnu47LIrqFXLO82u4O+MjAxefnkRI0d6ewpuuWUIQ4fewmOPTaZChQr897/PHjEXkWDKy3fz4vI/yNjt3XisXLkEsrPzghY/2PFCETPW4oUiZjDj5ebmUKNKJbq2OZ6kxPigxCyNsBZ9a+1AAGOMv5frcPCmAZuA4pu/h6hevdI/jm3d6iIhwXsXI5QDxuPiKHyf4iQkuFi69A0uuugSEhJcdOhwMWPH3stNN91MfLyL+HjXQXG8ceNJSIgnKSnpH+/h7z23bNnMnXfe8Y/jTz45i+TkAy19b6v9wPt5PB4SE+MPivn555/y+++/8cYbb+PxuBky5DaWL3+P9u0vAmDr1q0MGXILl19+JS1btiQrK4t///sBHn/8PzRrdjLz5z/PQw+NZcqUx/6Rj8vlIjW15L0ApbkmnPFCETPW4h1tzJ/+2M47K/4KYjYiwbCTTucfT90Q/LwEKpLumbuAojsbxAElWilmx449/9gcwe12Fy4Oc3vXwFriULqFao50fkKCi23btvP555/y22+/8sILC/B4PGRmZvL+++9RqVJldu3KPCjOjh3pVKxYifr1G5KVlcWGDX8XtsQTElwsW7aU9PR0unfvVXhN9eo1eeYZ/y39orFTU2uyZcs2atWqC0B6+g6qVKlx0Dkff/wxF1xwIUlJ3lsI7dpdxDfffEObNu1Zt24tQ4feQteuPenV6xry8tysWrWKpKRyGNOUvDw3nTt34amnZvj93rjdbrZtK9kgv9TUyiW+JpzxQhEz0Hi5eW4+/XETu/fnFntecnISe/cGb42EYMcLRsytGfsAqFsjmcv+rxEplSuQuXt/sNILerxQxIy1eKGIGax4mZmZLFw4j97du5CS5Ara7weXK85vY7c4kVT0NwBFl2erBfztUC4hs2zZW5xxRksmTz7Q8p0167+8+upLtG3bgWXLltCqVWsSEhJYv34t1v7KySefSrly5bnqqu5MnvwwY8c+RHJyJf7++2/++98Z3HXXPaXK5dxz/49ly5bQvHkLVq78nqSkpIO69gEaNz6Bjz76kCuv7IrH4+HLLz+jTZt27Nu3lyFDbmbQoMFcckmnwvPr1q3P1q1bWL9+LQ0aNOLjj5fTpEnT0n2zpER++GM7c9+2TqcRUWpWrUDLk9L0YTEG4oUi5tHE++uv9SxY8DzDhg0nPj6N9mePDVpeRyNiir61dp0xJssY83/W2k+BPsBSp/MKtqVL32DQoJsPOnbVVd2ZP38uQ4fezcaNG+jfvzcuVxzlypVjzJgHqFKlCgCDBg3m2WdncsMN1xIfn0B8fDw33ngLLVueU6pcunTpwcSJ47nmmu4kJSUyZswDAPz22y/MnPkkkyY9xmWXXcn69Wvp06c78fHxnHtuay699F8sWjSfjIx0Fi6cx8KF8wBo3fp8Bg68kZEjxzB69D2Ah6pVqxXe65fg27pzP39v8w6atOt3AlAvNZnmjWsc9pqKFZPYty94LfNgxwtWTFdcHOc0SwtSRiKBW7r0TR58cAwuVzxdu/bguOOOdzqlQnEeT/j3CjbGrAXaWGvXGmPeAu6z1n5tjGmOd3R/CvAtcK21NjuAkI2AP/1172/evI5atRqWOEetvR/6mKX5t4mkT/Lhinm4eLl5bu54/BP2Zx+8lfZ5p9bm2o4nOZ5fJMWM9HihiBlr8UIRs6Tx9u/fx4QJD/Laay/TvPlpPPzwJOrUqRu0fA5VpHv/WGBtINc40tK31jYq8rhjkccrgZZO5CSxwePxkLE7m4LPup6EeNJ3ZQX3PYIc83Dx9mXnsT87n3hXHCcfWw3wfshqe3qJxr+KSJAMG3Ybn3/+KQMH3siNN95SOL05kkReRiIhNHvJr3z602an0wiqconx3B7gdFERCS6Px0N+fj4JCQncdNOt9O8/sNS3XMNBRV9iyrotewBISU4iIT4Ol8uF2x3cWxrBjnmkeGefpPvWIk7IyMhg7NiR1K/fgDvvvIdTTon8D98xUfQ9Ho92dYswwR5L4vZ4eHTRStb8nVnseVm+hTaGdm9Og7TKUXlfUURC7+uvv2LkyLvIyEjnjjvucjqdgEV90U9ISGLv3kySk1NU+COEx+Nh795MEhKSghZzz/5cfv4zPaBzj6mURGqVCkF7bxGJHXl5eTz11AxmznySevXqM3fuQr8ro0aqqC/6VaumkpGxjT17dpboumjo9nU6XnExExKSqFo1tVQxt+7cz4J3f2d/keUx83yzNuJdcUy7rXWx1yclxpMQrw0mRaTkNm7cwJw5s+jU6XJGjBhVuP9IWRH1RT8+PoEaNWof+cRDxGK3b1n4mgG+tdtYeZg9FOrVrETF8olBfT8RkZ9//pFmzU6hYcNGvPjiG9Sv38DplEol6ou+RIYvf9nCqg0l623xp0KFJH5b6y34LU+qyYWnHTwHtkGadvQTkeDJyspiypRHWLRoPo8+Op0LL2xXZgs+qOhLGOTlu5n55i/ku4M7eK9+zUqYBlWDGlNEpMCqVau4/vpBrFr1O337Xkvr1uc5ndJRU9GXkPllbTr7V+9g56795Ls9xLvi6NnuhKOKWalSOfbsySYp0cWZpmaQMhUROdibb77Ggw+OoUKFijzxxFO0bn2+0ykFhYq+hMTWnfuZtPD7g46VS4yn3RlHt1qcpq+JSDiUL1+es846i/vue4jU1OhpYKjoR6msnDw2+jZiCdSOfbns9G1JerT+3uF978oVkzjtBO/mL6ccVy0osUVEQuHHH1fy559ruPzyzrRvfzE9e3Zh+/Y9TqcVVCr6UWrCvG9Zv8X5/6y1qlek/6VNnE5DROSw3G43c+bMYvr0adStW49LL+1EYmJSVK7toqLvoLx891ENbsvKySM7N9/va9t2ejdoaVirMvGuwP7jJibEk5vnP15pxMXBFedHzpaSIiKH2r59G/feO5wvv/yMDh0uYfTo+0lMDN7CYZFGRd8hv/+1kykvfE9OkLeyPdSdPVuQHOC89bIyT19EJBj27NlDz55XsWfPbkaPvp+rruoWla37olT0HfLnpszCgp+UUMrV4eLioJg17BvXO4aK5fRPLCJSVMF+LJUqVWLQoMGcfvqZNG58dDOLygpVBIcs+nA1ABedVb/U09jUihYRKZn169cxcuRd3HbbUFq2PIfu3Xs5nVJYaQFyB+Tluwsb6HVrJDubjIhIjHjrrTfo1esq1q1bS1bWfqfTcYRa+iH0wbcb+HVtRuHzpHIJ5GTn4S7SJX9e8zpOpCYiEjP27dvLhAkP8vrrr9CixemMHz+ROnXqHvnCKKSiHyJut4f57646qMAfqmrlcmHMSEQkNi1b9hZvvPEqgwYNZtCgwSQkxG7pi92vPMQ+/3lzYcEffOXJAKQcU4HMXQe6lBrV1uYwIiKh4PF4+Ouv9TRo0JDOnbvStGkzmjRp6nRajlPRD4Et6fuYteRXACqUS+DMJt4lHDXwTkQk9DIyMhg7diTffvsNL7/8JqmpNVXwfVT0gyTf7ea3dTvJysln284DrflbOp/sYFYiIrFlxYovGTnyLnbuzGDIkLuoUSPV6ZQiiop+kCz/7m/mvfv7QceOrZ3CSY203ryISKi53W4mTpzItGnTaNCgIY8//qRa936o6AfJrr3ZANSuXpFa1SriiovTyHwRkTBxuVxs2LCByy67khEjRlGxoqZD+6OiHyRf/7YNgJYnpXFF62MdzkZEJDZ88MG7NGjQiMaNT2Dy5MlkZMTm/PtAaXGeIEmtUgGAxNIuqSsiIgHLysrioYfGMXTorTzzzNMAMT0VL1D6Dh2Fj1f+zQsfrCLf7SHXt45+vVR1KYmIhNIff6xmxIihrFr1O337Xsuttw5xOqUyQ0X/KHy/ajv7sw9sRZtcPoF6qZUczEhEJLqtXPkdN9xwLRUqVOSJJ56idevznU6pTFHRL6VPVv7N96u3A3DjFc1ofnwNEhLiiHepe19EJFROOqkZnTt3Y8CA60lNrel0OmWOKlQpLftqfeHjWtUqUi4pXgVfRCQEfvjhewYO7EtmZiZJSUkMH36vCn4pqUqV0qYd+wAY0r05DdK0nK6ISLC53W5mz36Ka6+9mo0bN7B16xanUyrz1L1fCkU30VHBFxEJvu3bt3HvvcP58svP6NDhEkaPvp+UlBSn0yrzVPRLoWCkPsAxyUkOZiIiEp0eeWQ8K1d+y+jR93PVVd2Ii4tzOqWooKJfCl/9oi4mEZFgy83NYe/evVSpUpU77xzBDTfczPHHN3Y6raiie/qlsHrjLgBqHFPe4UxERKLD+vXr6NevN3ffPQSPx0PNmmkq+CGgol9CP/yxg09+2ATAGUa7N4mIHK0lS96gZ8/ObNjwFz17Xq2u/BBS934Jbd91YF3nc5vVcjATEZGybd++vTz88AO88carnHbaGYwfP5HatbVRWSip6JeAx+Ph+Xe82+e2aVFHI/dFRI5CXl4+3333DYMGDWbQoMFaOz8M9B0uAU+Rx6ccX92xPEREyiqPx8OSJa9z9dXdSUlJYfHi16lQoYLTacUMFf0AuT0eHn/xBwDigNNO0P18EZGSyMjIYMyYe/j44+WUK+eiQ4fLVPDDTAP5ApS+K4uVf+wAoH5NbaojIlISK1Z8SffuV/D5559y990j6dWrl9MpxSQV/QDluQ907o/qd6aDmYiIlC2LFy9k0KD+VKxYkeeee4HevftqhL5DVPQDtH3ngVH7CfH6tomIBOq0086gc+euLFjwEk2aNHU6nZim6hWggk+lDdS1LyJyRO+//w7//vdDADRufAL33fcAFSsmO5yVaCBfgDy+sfuVKyY6nImISOTKyspi8uQJLF68kKZNT2bfvr0q9hFERT9Aubm+TXZ0H0pExK/Vq1cxYsRQVq9eRd++13LrrUNITNSmZJFERT9A367aBhy8w56IiHjl5ORw003XkZeXxxNPPEXr1uc7nZL4oaIfILdv9L620hUROWDv3j1UrJhMUlISDz88iYYNG5GaWtPptOQwNJAvAFsz9vH5z97tdBvXPcbhbEREIsPKld/RtevlzJs3F4Azz2ypgh/hVPQDsDn9wHQ9Lb8rIrHO7XYza9Z/GTDgGuLi4mjevIXTKUmAwtq9b4zpDYwCEoGp1trph7x+OvBfIAn4C7jGWrsznDn64/Z4u/ZPPb46tapVdDgbERHnbNu2lVGjhvPll59z8cUdGTVqHJUra/OxsiJsLX1jTF3gIaA10AIYZIw5dJWGacB91trmgAXuDFd+xfH47ue7NHJfRGLc2rVr+PHHlYwZ8yATJkxWwS9jwtm93x74wFqbbq3dC7wIdD3knHggxfe4IrCfCFDQ0lfNF5FYlJubw4cffgjAWWedw9KlH9C5c1ctpVsGhbPo1wE2FXm+Cah3yDlDgaeNMZuADsCTYcqtWAXL7rtc+g8uIrFl/fp19OvXi759+7Ju3VoAjjmmirNJSamF856+i4O3pI8DCie9G2MqALOA9tbar4wxQ4G5QKdA36B69eAukZua6u22qrRhFwAVyicWHjvamMES6fFCETPS44UiZqzFC0XMSI8XiphHG++ll17innvuITExkaeffpozzzwlSJl5xcL3MNKEs+hvAM4r8rwW8HeR5ycD+621X/me/xd4oCRvsGPHnsL59EcrNbUy27btBmDnLu9dhtzc/MJjRxszGCI9XihiRnq8UMSMtXihiBnp8UIR82jj3X//aF5+eTGnn34m48dP5JRTToyo/MIRMxQ5BpPLFVfixm44u/ffA9oZY1KNMRWBLsCyIq+vBuobY4zv+RXAijDmd1i79uQAoN59EYkVjRufwKBBg3nqqWepVau20+lIkIStpW+t3WiMuRf4EO+UvJm+bvy38I7Y/9oY0x9YZIyJA7YC14Yrv+L8uGYHANm5WoJXRKKTx+Nh3ry51KlTh7ZtO9C7d1+nU5IQCOs8fWvtfGD+Icc6Fnm8FFgazpwCkVze+206rk7KEc4UESl70tPTGTPmHj755CP+9a8raNu2g9MpSYho7f0AFIwSqFmlgqN5iIgE24oVXzBy5F3s3LmT4cPvpWfPa5xOSUJIRT8AnuCMDRQRiSjW/sagQdfSsGEjHn/8KZo0OcnplCTEVPQD8J1vW12tQyEi0SAnJ4ekpCSMacLo0fdzySUdqVgx2em0JAy04U4A6tTw/jCUT9JnJBEp29577206dWrP6tWrALjqqm4q+DFEVSwQvu79Y5KTnM1DRKSUsrKymDTpYV588QWaNTuF8uXLO52SOEBFPwCFt/TVvS8iZdDq1asYMWIoq1evol+/67jllttJTFQjJhap6Afg7+17AdV8ESmbXn/9FdLT05kx42latTrvyBdI1NI9/SPYl5Vb+LiyuvdFpIzYtWtX4X37W265g8WLX1PBFxX9I8nLPzBfL6Wiir6IRL6VK7/joosuYsiQW8jLyyMpKYnq1Ws4nZZEABX9I3D7JumnqJUvIhEuPz+fmTOfZMCAa3C5XIwf/wgJCbqLKwfof8MRFOzap812RCSSZWZmcuedt/HVV19w8cUdmTp1MtnZ+sUlB1PRP4J9WXnAwd38IiKRJjk5mfLlyzN27ENcccVVpKSkRPS2sOIMFf0jyPe19Pfszz3CmSIi4ZWTk8NTT82gR4/epKbWZNq0/xCnpUOlGLqnfwQF9/Qb1qrscCYiIgesW7eWfv16MnPmk3z44fsAKvhyRGrpH8GBe/r6YRKRyPDmm68xfvw4EhISmTLlCdq2be90SlJGqKV/BAUtfZe+UyISAV54YT6jRg2nSZOmLFr0qgq+lIha+kdQsK2uus1ExEn5+fnEx8dz6aWd2L9/P9dc00/T8aTE1H49AnXvi4iTPB4Pzz33LNdeezW5uTmkpBxD//7XqeBLqajoH0Fh975qvoiEWXp6OrfeeiOTJ0+gWrVqZGVlO52SlHH6qHgEBfP0Xar6IhJGX331Bffeexe7du1ixIhR9OhxtW4zylErdUvfGBMTCznv3pcDwM49OQ5nIiKxIj8/n4kTx1OpUmWee24RPXteo4IvQRFQS98YUxWYCEwDfgZeAzoaY1YDnay1q0OXorPi472fi9KqVnA4ExGJdn//vZEqVapQsWIyU6fOoFq1alSoUNHptCSKBNrSnwa0AnKA7kA7oCfwE/BoaFKLDAUD+bThjoiE0pIlS+jRozOPPjoRgLp166ngS9AFWvQ7Av2stRa4DHjbWrsYGAWcH6rkIkG+Ru+LSAhlZWXx4INjGDRoEA0bNqJfv+ucTkmiWKAD+coDm32POwBjfI/dvj9Ra3P6PsA7bUZEJJjWrl3DsGG388cfqxg8eDDXXnsjiYnqVZTQCbTofw8MMMZsAWoAS4wxScDdvtei1jG+bv09vlH8IiLBkpiYRH5+HjNmzOSKKy7VrngScoF2798JDAamAxOtteuBqcCVeAt/1Cq4p1+rmu6ticjRy8zcxTPPzMTj8VC3bj1eeulNWrVq7XRaEiMCKvrW2i+A2kANa+1w3+FJwLHW2hWhSi4SaHEeEQmW7777lh49OjN9+lR+/fVnAOLj4x3OSmJJSRbncQEXGWNOAh4D6gOZIckqghzYcEdVX0RKJz8/n9mzn+bJJx+nVq3aPPPMPJo2PdnptCQGBdTSN8bUBn4EnsY7Yr8KMBT4yRjTNHTpOc/tG6ao0fsiUlqjRg1n+vSpdOhwMQsXvsIppzR3OiWJUYG29KfgXZSnBbDdd+waYAHebv6OwU8tMqzeuAuAeLX0RaSEPB4PcXFxdO7clbPPPpcrrrhKK+uJowIt+hcCF1lrs40xAFhrdxtjRgCfhCq5SJCY4O0M2Zet0fsiEpicnBymTZtMcnIygwffRsuW5zidkggQ+Oj9CkCun+PlgKj+2Pr39r0AHF/nGIczEZGyYN26P+nXryfz5s1hz549WuNDIkqgLf13geHGmGt9zz3GmGOAh4EPQ5JZhCgYyOfWD66IHMEbb7zK+PH3k5SUyKOPTufCC9s5nZLIQQIt+kPwFveNeFv9rwDHAjuA9qFJLTLs2eft4KipDXdEpBgbN25g3LjRNG/enPHjJ5GWVsvplET+IaCib639yxhzKtAb72C+HLwD++ZZa7NCmJ/jCtbeL7i9g3yuAAAgAElEQVS3LyJS1ObNm6hVqzZ169Zj9uznaNbsFM29l4gV6JS92UC8tXamtfYWa+1Qa+0soKIx5qXQphgZUo9RS19EDvB4PDz33LNcdtlFLF/+AQCnntpCBV8i2mFb+saYJkBN39N+wGvGmIxDTjsVuCREuTluf9ER+1E9XFFESmLHjh3ceuut/O9/H9GmTTtatDjN6ZREAlJc9/5xwJu+xx689/H9eTyoGUWQzH05hY+1OI+IAKxY8QWjRg0nIyODESNG06NHb829lzLjsEXfWvuWMaYe3lsA64HTgW1FTvEAu6210bstlG/Afs0q6toXEa/NmzdTuXJlHnvsv3g7REXKjmIH8llr//Y9POy9f2NM+WgfzKeufZHYtnHjBlat+p02bdpy2WVXcvXV3cnMzDnyhSIRJqDR+8aY6sC9wClAwSiVOLyL8zTFuxa/iEjUeeedZdx//2jKly/POee0onz58pQrVw7vJCaRsiXQeWj/xTtdbyNwPt7u/iTgHOCh0KTmPC3HIxK79u/fzwMP3Mfdd99Bo0bH8uyz8ylfvrzTaYkclUCLfjugn7W2P/ArMNVa+3/ADLzz9qNSdk4+oN59kVizf/9++vTpzksvLaJ//4E888w86tWr73RaIkct0BX5KgK/+B7/BpwGfA/8B3g7BHlFhIJ197fviu4hCyJysAoVKtChwyUMHTqcVq1aO52OSNAE2tJfBxQMU7UcaN3nAVWDnVSkWLvZOzHh+DopDmciIqGWmbmL4cOHsnLldwDccMPNKvgSdQIt+nOB540xl+Cduz/AGHMHMBX4IVTJOe3dr/8CoLpW4xOJat999y09enTm/fff4Y8/VjudjkjIBNq9/xCwH+9SvF8YYyYAI/HO2+8bquSclpTgIifPTcdzGjidioiEQH5+PrNnP8WTTz5B7dp1eOaZ+ZxyyqlOpyUSMoFuuOMBJhd5/hC+UfvGmIqhSc15Bdvpaoc9kei0ZMnrTJ8+jUsu6cSoUeOoVKmS0ymJhFSxRd9X0NsCucAn1tp9h7x+Cd7BfMeGLEMH+Wq+ltgUiTK7du3kmGOq0KnT5VSpUoXzzmujn3OJCcWttHc6sAZ4DVgK/GKMaeR7rZoxZh7wFrA3DHk6wu3bVlfr7otEh5ycHCZOfJjOnTuxbdtW4uPjOf/8C1XwJWYU19J/BNgMdMW79NQEYJIx5j680/RSgfspweI8xpjewCggEe9c/+mHvG7wLgRU1ffePa21h+7sFxYej6dwcR79PhAp+9at+5Phw4fx22+/0KvXNaSkHON0SiJhV1zRPwO40lr7PwBjzADgR7zL7m4BLrHW/hzoGxlj6uL9gHAGkA18Zoz50Fr7i+/1OOB14HZr7TLfYMERwPCSf1lHz13Yta/ufZGybvHixdxzz0iSkhKZOnUGbdq0dTolEUcUN2UvBfi94Im1di3eFvoaoFVJCr5Pe+ADa226tXYv8CLeXoQCpwN7rbXLfM/HA9NxiLr2RaKDx+PhnXfeoWnTZixa9JoKvsS04lr6cUD+IcdygbHW2tLsNFEH2FTk+SagZZHnjYHNxphZeFf8+xW4tSRvUL168Ebe5uR6v3SXK47U1MpBixvMWGUhXihiRnq8UMSMtXjBiPnDDz+QkpJCo0aNmDp1KuXLlyc+Pv7IFwYoEr/mWI8XipihyNFJgc7TLyq9lO/l4uA9bOIA9yG5tAHOt9Z+bYx5AJgC9A/0DXbs2FPYQj9alVMqFCa5bdvuoMRMTa0ctFhlIV4oYkZ6vFDEjLV4RxvT7XYzb94cpk2bwnnnXcCjjz4R9V+z4oUmZihyDCaXK67Ejd0jFf1axphDz0kzxhy0GL219u8A3msDcF7R2EDR6zYDq6y1X/ueL8B7C8ARBXP041zq3hcpK9LTd3DffSP53/8+4sIL2zN27INOpyQSUY5U9L875Hkc8L9DnnuAQPrM3gPGGmNS8U7z6wIMKvL6Z0CqMaa5tXYlcBnwTQBxQ2L3vlzgwE57IhLZVq/+nRtvvI7MzF2MGDGaHj16axCuyCGKK/oXBvONrLUbjTH3Ah8CScBMa+1Xxpi3gPt8XfqdgaeNMcl4ewb6BDOHksjPdx/5JBGJGHXr1qd58xYMGnQzxjQ58gUiMeiwRd9a+1Gw38xaOx+Yf8ixjkUef8nBg/scUzAyIE1L8IpErI0bNzB9+jRGjRpLxYrJTJ78uNMpiUS0QHfZizkFAwLVPSgSmd55Zxk9enTm448/ZNWq3498gYiUavR+TPAUDORTzReJKPv372fixPG8/PJiTjmlORMmTKZu3XpOpyVSJqjoH0bBZjtanEcksowfP44333yNAQOu56abbiMxMdHplETKDBX9w3CrpS8SMTweD9nZ2ZQvX54bb7yFjh0v49xz/8/ptETKnICLvjGmMTAMaAJcA1wB/Gqt/TBEuTnKU7jGj6q+iJMyM3cxbtwo8vLymDp1BnXr1lN3vkgpBTSQzxhzNrASOB5oBZQDmgHvGGP+Fbr0nFNwT19r84g457vvvqF79yv56KMPOf30Mwt/LkWkdAIdvT8BmGCtvQjvNrtYa2/Gu/3u2NCk5ixP4S57qvoi4Zafn8/TT/+H667rQ0JCAs8+u4B+/a7D5dKEI5GjEehP0Ol4l8U91CzgpOClEzny3N7FeVTzRcJv165dvPDCfC6+uCMLF77CySef4nRKIlEh0Hv6e4GawOpDjp8I7ApqRhEiN9db9NMzs45wpogEyzffrKB589NITa3GggUvUaNGqnrbRIIo0Jb+AuBRY8xJeBerK2+MaQs8gYOb4oRSwe+ZyslJziYiEgNycnJ45JHxXHddH158cSEAqak1VfBFgizQlv49wHPAz77nP/r+XgSMCHZSkSDftyJf5QqaAywSSuvW/cnw4cP47bdf6NWrD507d3M6JZGoFVDRt9bmAD18G+a0wDuY72dr7R+hTM5JBUU/XsP3RULmgw/e5d57h1OuXBLTps3gggvaOp2SSFQLqOgbYz4GngUWW2ujsjv/UAW77Gm0sEjopKXVonnzFowbN560tFpOpyMS9QKtaD/hnba32Rgz3xhzsTEmqpvAaumLhMbPP//I008/CUCzZqfw5JOzVfBFwiSgom+tHQzUBrrjHcj3ErDBGPOIMSYq59Ks2eidlKBxRCLB4Xa7mTt3Nv369ebllxeRmRmVE39EIlrAy/Baa/OBJcASY0xFYCDwAN6leeNDk55zyid5v6TMfTkOZyJS9qWn72D06BF8+ukntG3bgTFjHiAl5Rin0xKJOSXacMcYUwPoBvTCuxzvV8CcEOTlOF/vPifWq+JsIiJlXF5eHv369WbLlk2MHDmGbt16aiqeiEMCHcjXF2+hbwdswjt97zpr7aoQ5uYoT+Eue/rlJFIaeXl5xMfHk5CQwJAhd1GvXn1OPNE4nZZITAt0IN90YCtwCdDIWjsqmgs+aGtdkaOxceMGBgy4mlde8U72adu2vQq+SAQItHu/lrV2b0gziTDacEekdN5+eykPPDAagMqVKzucjYgUddiib4x5Chhqrd2Ddwnewwax1g4KQW6O+nvbHkBb64oEav/+/UycOJ6XX17MKac0Z8KEydr3XiTCFNfSP6HI6yfinaoXM7Jy8gHYsz/X4UxEyobvv/+WV199iQEDruemm24jMVFLWItEmsMWfWvthUUetzncecaYmkHOKSJUKOf91tStkexwJiKRy+Px8NNPP5GW1pBzz/0/XnllCQ0bHut0WiJyGAEN5DPG5BtjUv0cbwCsCXpWESQpMeqWIBAJil27dnLnnbfRqVMnVq/2jutVwReJbMXd0+8BXOx7Ggc8ZozZf8hpxwK7Q5SbiESo7777hnvuuZPt27dzzz33cNxxxzudkogEoLh7+p8C1+Et+AB18O6uV8ADpANXhyY1EYlEs2b9l+nTp1G3bj3mzJlPmzat2LZNn/1FyoLi7ulvAC4CMMY8A9xurc0MV2IiEpny8/O55JJOjBw5hkqVKjmdjoiUQHHd+3WstX/7nt4LVDLG+P0JL3Je1PhpzQ7gwHx9kVj20UcfkJiYRKtWrRk48Ebi4uK0hoVIGVRc9/5fxpja1tqtwAb8T9mL8x2PutFu9VIrsTV9H3lut9OpiDgmOzubqVMnsWDBc7Rq1ZpWrVrjcgW6kKeIRJriin5bvPfsAS4s5ryoVLD2fvWU8g5nIuKMtWvXMHz4MKz9ld69+3LHHXc6nZKIHKXi7ul/5O9xAWNMDWvt9lAl5rSCbg31YEosWrt2Db16daVcuSQee+w/nH9+zH3uF4lKge6yVxWYCEwDfgZeAzoaY1YDnay1q0OXojMKd9lDVV9ih8fjIS4ujoYNj2XAgOu5/PKrSEtLczotEQmSQG/OTQNa4Z2y1x3vFrs9gZ+AR0OTmrMObLjjbB4i4fLTTz/Sq1cX/vprPXFxcVx//U0q+CJRJtCi3xHoZ621wGXA29baxcAo4PxQJeekwqLvbBoiIed2u5kzZxb9+/di584Mdu3a5XRKIhIigW6tWx7Y7HvcARjje+z2/Yk6HtTUl+i3Y8d27rvvHj799BPatu3AmDEPcMwxVZxOS0RCJNCi/z0wwBizBagBLDHGJAF3+16LOgUtfW2tK9FszpxZrFjxJSNHjqFbt56aey8S5QIt+nfiHbxXA5horV1vjJkBXAlcEqrknOTRqjwSpXJzc9m8eRO1atXmpptu4/LLr6Jx4xOcTktEwiCge/rW2i+A2kANa+1w3+FJwLHW2hWhSs5JBwbyqeUj0WPjxg1cddVV3HjjAHJycqhQoYIKvkgMCbSlD1APuNkYczKQi3fq3lNAVK7HXzhlTzVfosTbby/lgQdG43K5uPfecSQlJTmdkoiEWUAtfWNMC+BHvNP09gL5QB/gB2PMKaFLzzkHFudR1ZeyLSsri3HjRjN8+BCOO+543nnnHS6++FKn0xIRBwQ6ZW8SsBRobK3tbq3tAhwPLMG7aE/UObA4j0jZFh8fz5o1qxkwYBCzZj1P/fr1nU5JRBwSaPf+uUBLa21uwQFrba4xZjzwWUgyc1jhOD5VfSmDPB4Pr776Em3atKNq1arMnDmXxMREp9MSEYcFWvQzgMp+jqcAecFLJ3IU1HyXuveljNm1ayfjxo3mgw/eZfv27Vx//Y0q+CICBN69vwR4whhzfMEBY8wJeJfnXRKKxJymKXtSFn377dd0734lH3+8nGHDhnPddYOcTklEIkigLf17gPeA340xBTvr1QC+AIaGIjGnHVicRy19KRuWLVvCyJF3UbdufebOXUDTpic7nZKIRJiAir61Nt0YcybehXiaAVnAz9baD0KZnJPU0peypmXLc+jevRe33jqE5ORKTqcjIhGo2KJvjEkG2gLZwGfW2reAt8KRmNO0y56UBcuXf8Brr73MxIlTqVatOiNGjHY6JRGJYIe9p2+MORVYjXf53WXAb8aYs8KVmNMOLM6jqi+RJzs7m3//+0HuuGMwmzb9za5dO51OSUTKgOIG8k0A/gBaAWcDFpgRjqQigWbsSaRau3YNffr0YMGC5+nduy9z5y6kevUaTqclImVAcd375wIXWmu/BzDGXI93IF+ytXZvWLJzkLr3JRJ5PB7uvnsoW7du5rHH/sP551/odEoiUoYUV/QrA5sLnlhr1xhj8oDqeJfiLTFjTG9gFJAITLXWTj/MeZ2AJ6y1x5bmfYLBo6ovEWTPnj0kJiZSrlw5xo9/hMqVjyEtLc3ptESkjCmue98FuA85lkvJNukpZIypCzwEtAZaAIOMMU39nJeGd9lfR6vtgSl7TmYhAt9//z29el3Fo48+AkDjxieq4ItIqQS6OE8wtAc+sNam+24PvAh09XPeTGBcGPPyS1P2xGlut5s5c2ZxxRVXkJeXx8UXd3I6JREp447Uar/dGFO0Kz8BGGyMSS96krV2fADvVQfYVOT5JqBl0ROMMbcB3+Jd9KfEqlcP3tzkgpJfvXolUlODFzc11d9qxtEbLxQxIz1eMGJu376doUNvZ/ny5XTs2JGJEydSpUqVIGUXG9/DshYvFDFjLV4oYoYiRycVV/TXA70PObYZ6HLIMQ8QSNF3caCWgrf7vvD2gTHmZF/sdkC9AOL9w44de3C7g9RC94XJyNhLEsGJmZpamW3bdgclVlmIF4qYkR4vWDHXrt3IypUruffesdx000C2b98TtDxj5XtYluKFImasxQtFzFDkGEwuV1yJG7uHLfrW2kZHm9AhNgDnFXleC/i7yPNuQG3gayAJqGOM+cRaW/SasHFra10Js9zcXN55ZxkdO/6LRo2OY8mS96hYMVlrRYhI0JRqUF4pvQeMNcak4h393wUo3A3EWjsGGANgjGkELHeq4EORefr6hSthsGHDX4wYMYyffvqBOnXqcNppZ1CxYrLTaYlIlAnbQD5r7UbgXuBD4HtgvrX2K2PMW751/SOKRy19CZO3336Lnj07s3btn0ycOJXTTjvD6ZREJEqFs6WPtXY+MP+QYx39nLcWaBSerPw7ME1fZV9C59FHJzJnzixOPbUFDz88ibp1SzWcRUQkIGEt+mXJgbX3HU5Eotrpp59BQkICN954C4mJiU6nIyJRrsRF33e/fQMQZ63NDXpGEULT9CUUPB4PixYtICcnhz59+nPBBW254IK2TqclIjEioHv6xpg4Y8xo35z91UAD4DljzCxjTJQ2T7TLngTXrl07GTr0Vh5++H6++eYrLQAlImEX6EC+ocBA4Hog23fsBeBfwIMhyMtxbi29L0H07bdf0737lXzyyUcMGzaCKVOe0AdKEQm7QIv+dcDNvoF4bgBr7SvAAKBXiHJzlgbySZBs2bKFG264lqSkcsydu4A+ffrjcoVzBWwREa9A7+kfC/zs57gFUoOXTuTQ4jxytPbt20vFismkpaXxyCNTadnybJKTg7eks4hISQXa3LDA+X6Od/G9FnUKb7eq6kspLF/+AZ06teezzz4B4MIL26ngi4jjAm3pjwWe922FmwBcbYw5AW/Xfp8Q5eYwb9V3qXtfSiA7O5vRox9h9uzZNGnSVPPuRSSiBNTSt9a+CnQHWgH5wBC8Xf7/stYuDF16zgnWvj0SO/78cw19+vRg9uzZXH11P+bOXUjDhsc6nZaISKGA5+lba98C3gphLpHFU9DSdzgPKTNWrPiSbdu28Oyzz9KixTlOpyMi8g8BFX1jzMjiXrfWBrK1bplyoKWvqi+Ht3v3blav/p3TTjuDbt160qHDJZx4YoOI3o5TRGJXoC396/1clwbkAp8CUVf0DyzO43AaErF+/PEHRowYyu7du1m69H2SkytRtWpVp9MSETmsgIq+tfYfNyaNMSnAM8D/gp1UJNDiPHI4brebOXNmM336VFJTa/LYY//RyHwRKRNKvUKItTYTuA8YFrx0IkfhLnvq3pcicnNzuPnm65k2bRJt2rTlhRdeoUWL051OS0QkIEe7y15loEowEok42mVP/EhMTKJRo2Np1+4iunTprhUbRaRMOZqBfCl45+l/ENSMIoS696VAbm4u//nPY1x0UUeaNDmJ4cNHOZ2SiEiplHYgH0AO8CFQ7Mj+sku77Als2PAXI0YM46effqBChWSaNDnJ6ZREREot0KJ/FfC9tTZmlqzR4jyybNkSHnxwDHFxLiZOnEqHDpc4nZKIyFEJdCDfMuDMUCYScXRPP6a9++4yRowYxvHHn8ALL7yigi8iUSHQln4GUC6UiUSawv12VPVjSm5uDomJSbRp04577rmPLl26k5BwtONdRUQiQ6C/zd4AlhpjXgfWAPuLvhiNK/IdmLInscDj8fDCC/OYN+85nntuIVWqVKVHj95OpyUiElSBFv2uwHa8G+60OuQ1D1G2Ip/Hc+CGvlr60W/nzgzGjh3F8uXv07r1BXg0nkNEolSpV+SLZmrlx44vvviCwYNvZseOHQwbNoKrr+6Ly1XqNatERCLaYYu+MWYNcJa1dkcY84kIHk3XixkzZswgKakcc+cuoGnTk51OR0QkpIpr6TcC4sOUR0TxaGGeqLZ58ybi4uJIS6vF1KlTyczM1tr5IhIT1I/ph+7pRq8PP3yfHj2u5P777wOgWrVqKvgiEjOOdE//LGNMxpGCWGs/C1I+EULd+9EmOzubKVP+zQsvzOekk5py991RupCkiEgxjlT0X+fI49k8RNltAK27H102btzAkCE38/vvlj59+nPrrUNJSkpyOi0RkbA7UtE/G9gWjkQiiop+VDnmmCqUL1+Bxx//L+edd4HT6YiIOKa4ou8B1ltrt4YrmUjhLliCV5P2yqzdu3fzzDNPccMNt1CpUiXmzFmg2zUiEvOKG8in35D6DpRJP/74Az17dmbOnNl8881XgMZniIhA8UV/DocstxsrCkbvu1QnyhS3280zz8zk2mt743a7mTXrOVq1Os/ptEREIsZhu/ettdeGM5FI4jmw3Y6jeUjJTJo0gfnz59K+/cXcd98DpKSkOJ2SiEhE0fZhfqilX7a43W5cLhfdu/fkuOOOp0uX7urOFxHxQ0XfD49W5ykTcnNzeOKJaWzduoXx4yfSqNFxNGp0nNNpiYhELK3I50dh575aixFrw4a/6N//aubMmUVycjL5+flOpyQiEvHU0vdDa+9HtqVL3+TBB8fgcsUzadI02re/2OmURETKBBV9fwrn6Uuk2blzJxMmPEDjxify8MOTqFOnrtMpiYiUGSr6fqh7P/KsW/cn9es3pEqVKsya9TyNGh1LQoL++4qIlITu6fvh0Yy9iOHxeFiw4Dm6dr2cBQueB6Bx4xNU8EVESkG/Of0oGL3vUkvfUTt3ZjB27L0sX/4B5513AZde+i+nUxIRKdNU9P3QjD3nfffdNwwfPpSMjHTuumskvXv30e0WEZGjpKLvR8GKfKoxzvF4PFSqVIlp02Zw0knNnE5HRCQq6J6+PwVT9nRTP6w2b97ESy8tAuD0089k8eLXVfBFRIJILX0/3L6/1dIPnw8/fJ8xY0bidudz4YXtqFatOvHx8U6nJSISVdTS98ej7v1wyc7O5uGH72fIkJupV68e8+a9SLVq1Z1OS0QkKqml74dH3fth4Xa7GTiwLz/+uJI+ffpz221DSUxMcjotEZGopaLvx4HFeRxNI+q5XC66du3BDTfcTOvW5zudjohI1FP3vh8eLb4fMrt372bEiKEsXfomAFdccZUKvohImKjo+3Gge1+C6ccfV9KzZ2feffdtduzY4XQ6IiIxJ6zd+8aY3sAoIBGYaq2dfsjrVwDj8NbbP4FrrbUZ4cwR1L0fbG63mzlzZjF9+jRSU2sya9ZztGhxutNpiYjEnLC19I0xdYGHgNZAC2CQMaZpkddTgP8Anay1zYEfgLHhyq8oT+HofVX9YFix4kumTZtMmzbteOGFV1XwRUQcEs7u/fbAB9badGvtXuBFoGuR1xOBm621G33PfwAahDG/A3RLPyi2bNkCwNlnn8vMmXOZOHEqKSkpDmclIhK7wln06wCbijzfBNQreGKt3WGtfQXAGFMBGAG8Gsb8CrkLWvpOvHkUyM3NYcqUR7j88ov49ddfATjzzJbqORERcVg47+m7OHC7HLw11X3oScaYY4BXgJXW2jkleYPq1SsdVYIFdud400pIiCc1tXJQYhaI9nhr167l5ptv5vvvv6dPnz40atSIChUqBCk7r0j7msMRM9bihSJmpMcLRcxYixeKmKHI0UnhLPobgPOKPK8F/F30BGNMbeBt4ANgSEnfYMeOPbjdR79FXnr6XgDc+W62bdt91PEKpKZWjup4S5e+yYMPjsHlimfSpGm0b38xFSpUiKgcQx0vFDFjLV4oYkZ6vFDEjLV4oYgZihyDyeWKK3FjN5xF/z1grDEmFdgLdAEGFbxojIkH3gAWWWsfDGNe/+BBc/ZK4/ffLSecYBg/fiJ16tR1Oh0RETlE2Iq+tXajMeZe4EMgCZhprf3KGPMWcB9QHzgdSDDGFAzw+9paOzBcORY4sDaPqv6R/Pbbr2RnZ9G8+WkMHnwbcXFxJCRooUcRkUgU1t/O1tr5wPxDjnX0PfyaCFksSIvzHJnH42HBgud49NGJNGnSlLlzF5KYmOh0WiIiUgw1yfwo6N5XS9+/jIwMxo4dyUcffcj557dh3LiH9b0SESkDVPT90NL7h7dp09/069eLjIx07r57JL169VHBFxEpI1T0/VH3/mGlpdXiggva0qVLN5o0aXrkC0REJGJExD30SKPu/YNt3ryJO+4YzObNm3C5XNx77xgVfBGRMkhF3w9P4Y47jqYRET744D26d7+Sr776gjVrVjudjoiIHAV17/tRsOFOLH8iysrKYsqUR1i0aD5NmzZjwoQpNGjQ0Om0RETkKKjo++HR3ro89dR0Fi2aT9++13LrrUNITExyOiURETlKKvp+xGrvvsfjITMzk5SUFAYMuIGzzjqHc8/9P6fTEhGRIInlHuzDKujej6WGfmZmJjfddBMDB/YhOzubSpUqqeCLiEQZFX0/DvTux0bV/+GH7+nZszNvvfUWl1zSSSvriYhEKXXv+xMji/O43W6efXYm06dPIy2tFq+88goNGpzodFoiIhIiaun7Udi973AeoZaXl8eyZW/Rrt1FLFz4CmeccYbTKYmISAippe9HtHfvf/HFZzRrdgqVK1dm5sy5VK5cOWq/VhEROUAtfT88UbrNXm5uDlOmPMKNNw5g9uynAEhJSVHBFxGJEWrp+3Gg5kdPMVy/fh0jRgzjl19+olu3Xtxww81OpyQiImGmou9HtK3N8/nnn3LnnbfhcsUzefJjtGt3kdMpiYiIA1T0/Yi2gXzHHXc8Z511NsOHj6J27TpOpyMiIg7RPX1/Cqfsld2y/9tvv/LQQ+Nwu92kpdVi6tQZKvgiIjFORd8Pdxmep+/xeJg/fy59+nRn+fL32bx5k9MpiYhIhFD3vjgWJJMAABbhSURBVF8Fy/CWraqfkZHB2LEj+eijDzn//DaMG/cwVatWdTotERGJECr6fpTFGXsej4fbb7+RX3/9hbvvHkmvXn3K3IcWEREJLRV9PwpG75eFqp+Xl4fH4yExMZFhw0ZQrlw5mjRp6nRaIiISgXRP348Du+xFdtXfuHEjAwf2Zfr0aQA0b36aCr6IiByWir4fZaF7//3336FDhw6sWmU58UTjdDoiIlIGqHvfD0/hQD6HE/EjKyuLyZMnsHjxQpo3b86DD06kfv0GTqclIiJlgFr6fngieJ7+hg3ref31V+jbdwCvvvqqCr6IiARMLX1/Iqx73+Px8M03KzjzzJY0bnwir7/+DmlpaSQlJQHZTqcnIiJlhFr6fkRS935mZiZ33z2EgQP78vnnnwKQlpbmcFYiIlIWqaXvhydC5uytXPkd99xzJ1u3buH224dx9tnnOpqPiIiUbSr6fhQUfZeDNX/evLlMmfJv0tJqMXv285x6agvnkhERkaig7n0/PBEwZ6969eq0a3cRCxe+ooIvIiJBoZa+HwW9++Eevf/pp5+wffs2rrjiKi65pBMXX9wxImcQiIhI2aSWvh+FK/KF6f1yc3OYMuXf3Hzz9SxatID8/Hzv+6vgi4hIEKml78eBln7o32v9+nWMGDGMX375ie7dezN06N3Ex8eH/o1FRCTmqOj7E6bFedLT0+nduwsuVzxTpjxO27YdQvp+IiIS21T0/Qh1935+fj7x8fFUq1aNIUPuplWr1tSuXSdE7yYiIuKle/p+hHIg32+//UK3bpfzzTcrAOjSpbsKvoiIhIWKvh+Fi/MEseZ7PB6efvpp+vTpwZ49e3C59K0XEZHwUve+H8Hu3k9PT2fMmHv45JOPuOCCCxk7djxVq1YNUnQREZHAqOj7Eezu/WXL3uSLLz7j/vvv57LLumkqnoiIOEJF349gLMiXl5fHunVrOf74xvTseQ2tWrXmrLOas23b/7d37nFWVWUf/85MiiKIN0rzTjpPkqSYmt3Uyqys11JTLBQEQcu8FSZomJqRoaipqGkXLyAhWuBrKpWXNLxGmhrqE9SrJN4QdAAd5TbvH2sd5zCeYTh79p6zz5zf9/M5n5lZa5/ffta+zLPXs56119JUbBRCCCHKRQPLpSiE9xP2yF98cQEjRgxh+PDBNDW9QX19PTvs0C9NC4UQQoiykdMvwep35+mX/9277/4TgwYdwty5zujRY+nTZ5N0jRNCCCESovD+WijH6a9cuZLx43/CzTdPpX//XRk//mK23Xa77IwTQgghykROvwSt2fvr7vUbGhpYtmwZQ4YM56STTmW99dbPyjwhhBAiEXL6JShk73fk81taWpg+/RYGDvwYO+7Yj3HjLtD8eyGEELlFHqoE7/b01+L0lyxZwumnf48f//gspk2bAiCHL4QQIteop1+C1il7pb3+E088zpgxo1i48FVOOWUUQ4ce24XWCSGEEMmQ0y/B2pbWfeihBzjxxOP4wAe25Nprb2TAgN261DYhhBAiKXL6JSgV3m9paaGuro499tiToUOHM2zYcfTu3btCFgohhBDlo0HoUrTJ5Js1636GDj2SZcuW0aNHD04+eZQcvhBCiKqjS3v6ZvYtYCywHvBzd7+iTf3uwK+AjYH7gW+7+8qutBFafX7L6lVMmHA+kydfT2Oj0dT0Br169epqc4QQQohU6LKevpltDYwDPg3sDhxnZv3bbDYZONHdGwnd7JFdZV8xhfD+jBm/Y/Lk6xk06FtMmjSNrbfephLmCCGEEKnQleH9A4B73H2xu78J3AJ8o1BpZtsDG7r7w7HoOuDwLrTvPSxZ0sTFF0/kjDN+RI8ePSppihBCCNFpujK8/0HgpaK/XwL27qC+rK715punE3r/7N7b8+z8RRxx9Cl8fPedU9Es0LdvurkAedfLQjPvello1ppeFpp518tCs9b0stDMwsZK0pVOv56iFDlC+H51GfUdsmjRMlavbul4ww7o06OBi079HAsXLk11Kdy+fXvXlF4WmnnXy0Kz1vSy0My7XhaataaXhWYWNqZJfX1d2Z3drgzvvwBsVfT3lsCLZdQLIYQQohN0pdO/C/i8mfU1s57AYcDMQqW7Pw+8bWafikVHA3d2oX1CCCFEt6bLnL67LwB+CNwL/AOY4u6PmtkdZrZn3GwwcImZPQv0Ai7rKvuEEEKI7k6XztN39ynAlDZlBxX9/gRrJvcJIYQQIiX0Rj4hhBCiRpDTF0IIIWoEOX0hhBCiRpDTF0IIIWoEOX0hhBCiRpDTF0IIIWoEOX0hhBCiRpDTF0IIIWqELn05T4Y0QFh8IE3S1stCM+96WWjmXS8LzVrTy0Iz73pZaNaaXhaaWdiYFkW2Nazrd+paWjq/Kl0O+DTw10obIYQQQlSAzwCz1mXD7uL0ewB7AS8BqypsixBCCNEVNBBWp/0b8M66fKG7OH0hhBBCdIAS+YQQQogaQU5fCCGEqBHk9IUQQogaQU5fCCGEqBHk9IUQQogaQU5fCCGEqBHk9IUQQogaQU5fCCGEqBG6y7v3heg2mFkD0BdYDSxy99TeMmlmvYGV7t6clqbovmR1LaZ1HepeKR+9kQ8ws57Aj4DDga0JF9CLwJ3AWHdvKlPvPHc/y8w2ASYCXwZWANOB0919aXezMe/2VYONZvZ+4LKo00SIxPUirCvxXXefX45e1Py1ux9rZtsAU4GPAC3AI8AId19Qpl49MBI4gvcew8vcfUWZet9x96vMrAcwFjiI1mN4kbuvLFMv1XMcNWvxXkn1WszgOsz9vZJXFN4P3AgsA/YDNgJ6A/sT3uX/2wR6X4k/LwfmAx8C+ke9G7qpjXm3rxpsnAb8AdjM3bdz922AzQn/gG5MoAcwMP6cCExy903dfTNCe5PY+AvCOhdnE9p/MHAO0Ahcm0BvZPw5AdgBOBb4DrBz3Fe5pH2OoTbvlbSvxbSvw2q4V/JJS0tLzX8aGxufXkvdPxPoPRZ/Plmibk53tDHv9lWDjWnb18bGJ0rUPZVA79kk9q+LfY2NjfVF5XVr21eFjmEuz3OV3StpXYe5v1fy+tGYfmChmR0O/M7dVwOYWR0wCFiYQG9LMxsEvGBm+7j7w1FzL+Dtbmpj3u2rBhv/Y2anE3oqLxX2AQwB/p1AD2A7MxsNLDKzr7v7jNjmw4AlCfSWmtle7v634kIz+wRQdhgZ2MzMPg48R+ihzi3YTQhRl0va5xhq815J+1pM+zqshnsllyi8HziKcLG8bmbzzWw+8AYwGBiaQG8M8ClgU+AHAGZ2KnArcHJKNj4PvJ6BjTMS2ph3+0rZmLfzPBjYljAu+Vb8/JUw7ntMAj2AQwnO8xXgwFh2BjAaGJFAbyTwGzOba2b3m9l9ZjYXuDqh3rXADwlDBpcAmNkwYDZh3Ltc0j7HkP55TvteOaMd+zpzr7S9FpsJ67UnvRbTvg7Tti8LG3OJEvmKMLP3AVsQHoZeLTeJqAPt3sCbhSf7TuhkaePGwLLO2Jh3+6JO7s9z3jGz7YAPAnXAC+7+3xQ0N3L3N2Mi1dvu/lontDI7x1G/0+e5Gu4V0f2Q02eNjOTDgW1ozaa9A7g8QUZyqnrVgJn1ISR0bQdMd/fJRXXXuPtxldSL33sfoRewGPgzIVFsAKG3MCZBFnaqeu3s4z5336+zOmlpFmXbr0/oiR8ELAd+D1ycINu+WO8sWrP3k+r1IfTMryAky11CiCL8HRiV5EFiLZqzgdPK1TSzJ4Hj3f2hcm3pCr2ouTFwGqEXfTUwiZB4+DdgmLs/n1CvmXCfdFavJ3A+8DVCWH85Iaw/FZiQZOpeFpp5RGP6gV8QnrbPIYwP1RFO+tGE8ONRFdbDzIasrd7dy8ouTVuP0K6ngCnAGDPbt8gx71mmVhZ6ANcAGwLvJzis24FxhIezXxBChhXTM7P/lCjeulDu7v3KtC8LzZHAVcBFwCbAcML1/V1Cm8sNg7bVO7aTelOBxwnTuK4E/g+4gDDLYBJhile5pK25GXC1md0HnNuZiEZGegCTgX8RIhGPEM7REcDXgV/SGv6ulN7VwJOEYz8IWEA4R98HLgVOLFOvlOYLwD86qZk75PQD+7r7h9uUzQNmmdnTOdAD+DwhoWQa4Z9iMS2UP6Ukbb0d3f1QADO7A/iDmV3k7qNK6FdCD2Avdx8Qn+jnu/sPY/m5ZvZ4DvROAi4kPCw+Qmjn7YTeb1Ky0ATYFxhYlIg2EngmB3pbufuXo0Z/dz86lj9jZmU/bGek+SphDP5M4J9mNp3wYPGQuy/PgR7Atu5+MICZLXD3S2L5lJhAV2m9AUXn4Rwze8TdP25mg0l+HWahmTuUyBdYGjNd16ATGclp6+HuQ4F7gFnuPqzNZ3il9QDMbMuo3UxIijnAzM4kPERUXA9YbWZbuPtbFEVb4hhyknshVT13v50wX/soQkLXfOAdd3++3PBnhppts+0LJM22T1vvdTP7Qvz9cTMbCGBmAwih6iSkrunuze5+FuH9Bk8RkhlfMrMX86AHrDCzAy3MMuhlZh8DMLOdc6JXZ2YWNQbQ+j+hEJbPi2buUE8/MBKYZGYbEMLxLYQkpWYShOIz0CtwPOWHoLtK7xzg72Z2grvf6u5NZvZFwgs0PpoDvYLm42a2g7vPBIj/zCfT+pKYSurh7q8CB5vZSYSHsp5JdDLU/A3BoexJGNv+qoVs+/GE66nSet8GbjWzZYQs7L9amF3QFzgkgV4Wmu9Gqtx9CWHI4EoAM+ubAz0Ix/5SwsPrgcD1ZvYWIWP+mBzojQHuN7PnosYQM9sF+BNQdr5Phpq5Q4l8RVhrRnI9ISO57Fc5ZqmXd2JG83ruvriorB442N1nVFovfr9n7JkX/t4UqCveRyX12mjvCnzD3c/prFYWmtaabb8t0NzZseQ09cxsD2AnQsfmZeBBd086Zz1VTTPbzd2f6IwtWeq1s48NgF2Bee7+Rh70LCRY7gzMjZ2CeqDF3RM7tSw0c0el3w6U909jY+MeedarBhvzrlcNNqrN+dOrBhvV5vxqVuqjMf2OOS/nello1ppeFpp518tCs9b0stDMu14WmnnXy0qzIii8HzGzDwPfYM159TPdfXYe9KrBxrzrVYONanP+9KrBRrU5n23OI+rpA2Z2AmGKC4SXRTwWf/+lmY2qtF412Jh3vWqwUW3On1412Kg257PNuaXS4wt5+DQ2NnpjY2PPEuU9E670lapeNdiYd71qsFFtzp9eNdioNuezzXn9qKcfWAmsV6J8Q5LNFU5bLwvNWtPLQjPvello1ppeFpp518tCM+96WWnmDs3TD4wjzLe+mzXn1X+OMIe40nrVYGPe9arBRrU5f3rVYKPanM825xIl8kXM7IPAARTNqwfucvdEb7RKW68abMy7XjXYqDbnT68abFSb89nmPCKnL4QQQtQIGtMXQgghagQ5fSGEEKJGUCKfEAkws78A+7VTfZK7T1wHjeuAbdz9gBRNK2jvQFj3vZhVhGVYZwKnpbQ+wDnAUe6+U/z7E0C9uz8Q/24Bjnb3yZ3dVzv7PrtN8SrgNeAu4PtxsaF10XofcKK7/zxVI4XIGerpC5GcKcBWJT6/qaRRbfgarXZtT1jR7CDgxpT0JwD7FP19P2HBkgJbAbektK9SPMeax34n4FRCG68vQ2cQYZU/Ibo16ukLkZxmd3+50kZ0wOI2Ni4ws0uBn5rZJp1dMc3dlwHLiorq2tRnfXxWldjHcxbWaf+xmfVx96Z10KnreBMhqh85fSEyIobYLwQ+C/QBFgBXuPuFJbZtIKwh/01gC8CB89z95lhfT1jv+/hY/zRwtrvfkcC0lYT3ii+P2rsCFwCfIMxN/gMhNP5arD8GGA30Iywpex1wrruvLg7vW1iHvAG41syOcff9C+H92PZ7gH7u/u6wg5nNAaa7+1gLS+peQlhvvRm4N9qRZLrUO7EtK+N+NiNEJb5MOH4LCdGO0cC+wKS4XQswzN2vM7PPEM7JQMK87Ztiu9/u6LgksFeILkHhfSGy4zZgA4LT3wW4AbjAzHYvse0JwCHAYYABNwO/NbMdY/35wDDgOGA3Quj692a2/7oaY2YNccz9FOBOd38rPpg8ACwGPkMYDtgN+HPc/qPA1YSXk+xMCJ3/ADiqxC72Ioypnwoc2qbuL8DzwJFF9uwO9AduMLON4jbNwCeBLwLrA/eY2fpltLHOzPaJbZzh7m/Gqhvivv4HaAR+AowCDgYeBE6M220F3BRt+yPwe2AAMCJ+96q4n3KOixC5QT19IZIz1MyObFM21d1HmNmGhJ7fVHdfAGBm5xKcxADgH22+txPwFvCcu79sZj8BHgUWm1kvghM7zN3/GLefaGa7AWcQnGV7/MnMCj3PDQg9/JnAyFh2AvAGoXe7Itp5JCGS8CWC420Bnnf3+cB8MzuA8NKSNXD3hWYG0NQ2SdDdW8xsEiGScX4sHgw87O7/MrMRwEbAMe6+KtrxTUJS3mHAb9tpXz8zKx5e6AEsAqYBZxaVzwTudfc58e8rzWw0MMDdZ5hZU7Tz5bjv04Db3X1C3H6emR0PzDKzM4EPretxESJPyOkLkZzprOlYAJYCuHuzmU0EBpnZ3oTe4O6E6FpDCa0rCb3jBWY2m+CkJrl7k5ntRXBmNxc5cAjvCX+lAxuHAX+Pvy8HXnH3d4rqdwUeLTj8aPszZvZarLsMeASYbWbzCL3fm6KjK5frgbFm9hHgGUKvf1ysGwj0BZrig0OBnoQoSXv8F/h8/H0n4HJgLjA25hsUuAr4Wny4aAQ+Slg+tdS5KNizc5sHisK4/y6E85PWcRGiy5DTFyI5S9x9XqmKGK6eRbjHbiGMTz9CCHG/B3d3M+tHcGAHEnrEp5nZV4DX42aHAm33t6oDGxe0Z2Pk7XbKG4AV7t4M7GdmexLGw78EnGBmY9z9gg72vQbuPs/MHiS07S6Ck78pVi8H5vDeYQEIkYj2WFHUvnlm9iXCkqg3mdlBMcJQB9xBGDa5kTB+/yhw91p0lxMeUsaXqHspzeMiRFcipy9ENuxP6NlvXgh1W+jC1lMiU9zCWt6L3X0qMDOGl58EjgBOI6zytY27zyz6ztkE5/yjTtg5BxhiZusVhff7A5sCT5vZF4B93P08YDZwnpldSZj6V8q5dfRe7+uB7wGbAbe5e+GBZg5h3HxRoczMNgYmE5L77l2Xxrj7v+Oxu4aQ/3A1YSz/QOBj7v5YkfZWtJ6LtnbPAXYpfmCK+RCjgW+b2Scp77gIkQvk9IXIhoXx52Az+19C6PniWNajxPZbAOfGcPJTwB7AjsCFMeHuYuB8M1tCcDJfJTj7Yztp50TgJELG/fkEZ3858AShJ/xJ4Ow45n0bsCUhMfHhdvSWAv3N7P3tvBjnJuBSQsLb4KLyGwn5DtPM7AxCBOJnwN4EB1wOv4ra483sNkKkZCVwhJktJjj7cYTzUDgXSwFiz/1ZQg//sXjcrwE+EHUXxJwLo7zjIkQuUPa+EBng7o8SsrlHE5zIFYRe672ELPe2/BT4NWFs/1+EqX5nu3vhBTNjCePSEwjj4d8Bjnf36zpp5yvAFwjj27OBGcDjwAHuvsLd7wOGE3rNT8f6+4CT25H8GSE58I+lKuOc+VsJTv3OovLmaMdbhKl9DxA6JZ9b17fqFWm1RHt7ABPjlL9hwOGEczGZMNQymdZzcW9s14PAce7+FPAV4FOEpMtpsf6QuI9yj4sQuUCr7AkhhBA1gnr6QgghRI0gpy+EEELUCHL6QgghRI0gpy+EEELUCHL6QgghRI0gpy+EEELUCHL6QgghRI0gpy+EEELUCHL6QgghRI3w/6ClSl7uDz8BAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, threshold = roc_curve(ytestS, nb_pred_proba)\n",
    "plot_roc_curve(fpr, tpr, 'AUC = %.3f' % nb_auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-19T21:01:17.041346Z",
     "start_time": "2019-09-19T21:01:17.015415Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6617647058823529"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nb_recall = recall_score(ytest, nb_pred)\n",
    "nb_recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Conclusion: Naive Bayes didn't perform well as compared to the other ones."
   ]
  }
 ],
 "metadata": {
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.13"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "279.273px"
   },
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}