4. Credit Card Fraud Detection Capstone Project.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a067974a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "pd.set_option('display.max_rows',1000)\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "191105d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>cc_num</th>\n",
       "      <th>merchant</th>\n",
       "      <th>category</th>\n",
       "      <th>amt</th>\n",
       "      <th>first</th>\n",
       "      <th>last</th>\n",
       "      <th>gender</th>\n",
       "      <th>street</th>\n",
       "      <th>...</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>city_pop</th>\n",
       "      <th>job</th>\n",
       "      <th>dob</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>unix_time</th>\n",
       "      <th>merch_lat</th>\n",
       "      <th>merch_long</th>\n",
       "      <th>is_fraud</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>2703186189652095</td>\n",
       "      <td>fraud_Rippin, Kub and Mann</td>\n",
       "      <td>misc_net</td>\n",
       "      <td>4.97</td>\n",
       "      <td>Jennifer</td>\n",
       "      <td>Banks</td>\n",
       "      <td>F</td>\n",
       "      <td>561 Perry Cove</td>\n",
       "      <td>...</td>\n",
       "      <td>36.0788</td>\n",
       "      <td>-81.1781</td>\n",
       "      <td>3495</td>\n",
       "      <td>Psychologist, counselling</td>\n",
       "      <td>1988-03-09</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>1325376018</td>\n",
       "      <td>36.011293</td>\n",
       "      <td>-82.048315</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>630423337322</td>\n",
       "      <td>fraud_Heller, Gutmann and Zieme</td>\n",
       "      <td>grocery_pos</td>\n",
       "      <td>107.23</td>\n",
       "      <td>Stephanie</td>\n",
       "      <td>Gill</td>\n",
       "      <td>F</td>\n",
       "      <td>43039 Riley Greens Suite 393</td>\n",
       "      <td>...</td>\n",
       "      <td>48.8878</td>\n",
       "      <td>-118.2105</td>\n",
       "      <td>149</td>\n",
       "      <td>Special educational needs teacher</td>\n",
       "      <td>1978-06-21</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>1325376044</td>\n",
       "      <td>49.159047</td>\n",
       "      <td>-118.186462</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>38859492057661</td>\n",
       "      <td>fraud_Lind-Buckridge</td>\n",
       "      <td>entertainment</td>\n",
       "      <td>220.11</td>\n",
       "      <td>Edward</td>\n",
       "      <td>Sanchez</td>\n",
       "      <td>M</td>\n",
       "      <td>594 White Dale Suite 530</td>\n",
       "      <td>...</td>\n",
       "      <td>42.1808</td>\n",
       "      <td>-112.2620</td>\n",
       "      <td>4154</td>\n",
       "      <td>Nature conservation officer</td>\n",
       "      <td>1962-01-19</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>1325376051</td>\n",
       "      <td>43.150704</td>\n",
       "      <td>-112.154481</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>3534093764340240</td>\n",
       "      <td>fraud_Kutch, Hermiston and Farrell</td>\n",
       "      <td>gas_transport</td>\n",
       "      <td>45.00</td>\n",
       "      <td>Jeremy</td>\n",
       "      <td>White</td>\n",
       "      <td>M</td>\n",
       "      <td>9443 Cynthia Court Apt. 038</td>\n",
       "      <td>...</td>\n",
       "      <td>46.2306</td>\n",
       "      <td>-112.1138</td>\n",
       "      <td>1939</td>\n",
       "      <td>Patent attorney</td>\n",
       "      <td>1967-01-12</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>1325376076</td>\n",
       "      <td>47.034331</td>\n",
       "      <td>-112.561071</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>375534208663984</td>\n",
       "      <td>fraud_Keeling-Crist</td>\n",
       "      <td>misc_pos</td>\n",
       "      <td>41.96</td>\n",
       "      <td>Tyler</td>\n",
       "      <td>Garcia</td>\n",
       "      <td>M</td>\n",
       "      <td>408 Bradley Rest</td>\n",
       "      <td>...</td>\n",
       "      <td>38.4207</td>\n",
       "      <td>-79.4629</td>\n",
       "      <td>99</td>\n",
       "      <td>Dance movement psychotherapist</td>\n",
       "      <td>1986-03-28</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>1325376186</td>\n",
       "      <td>38.674999</td>\n",
       "      <td>-78.632459</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0 trans_date_trans_time            cc_num  \\\n",
       "0           0   2019-01-01 00:00:18  2703186189652095   \n",
       "1           1   2019-01-01 00:00:44      630423337322   \n",
       "2           2   2019-01-01 00:00:51    38859492057661   \n",
       "3           3   2019-01-01 00:01:16  3534093764340240   \n",
       "4           4   2019-01-01 00:03:06   375534208663984   \n",
       "\n",
       "                             merchant       category     amt      first  \\\n",
       "0          fraud_Rippin, Kub and Mann       misc_net    4.97   Jennifer   \n",
       "1     fraud_Heller, Gutmann and Zieme    grocery_pos  107.23  Stephanie   \n",
       "2                fraud_Lind-Buckridge  entertainment  220.11     Edward   \n",
       "3  fraud_Kutch, Hermiston and Farrell  gas_transport   45.00     Jeremy   \n",
       "4                 fraud_Keeling-Crist       misc_pos   41.96      Tyler   \n",
       "\n",
       "      last gender                        street  ...      lat      long  \\\n",
       "0    Banks      F                561 Perry Cove  ...  36.0788  -81.1781   \n",
       "1     Gill      F  43039 Riley Greens Suite 393  ...  48.8878 -118.2105   \n",
       "2  Sanchez      M      594 White Dale Suite 530  ...  42.1808 -112.2620   \n",
       "3    White      M   9443 Cynthia Court Apt. 038  ...  46.2306 -112.1138   \n",
       "4   Garcia      M              408 Bradley Rest  ...  38.4207  -79.4629   \n",
       "\n",
       "   city_pop                                job         dob  \\\n",
       "0      3495          Psychologist, counselling  1988-03-09   \n",
       "1       149  Special educational needs teacher  1978-06-21   \n",
       "2      4154        Nature conservation officer  1962-01-19   \n",
       "3      1939                    Patent attorney  1967-01-12   \n",
       "4        99     Dance movement psychotherapist  1986-03-28   \n",
       "\n",
       "                          trans_num   unix_time  merch_lat  merch_long  \\\n",
       "0  0b242abb623afc578575680df30655b9  1325376018  36.011293  -82.048315   \n",
       "1  1f76529f8574734946361c461b024d99  1325376044  49.159047 -118.186462   \n",
       "2  a1a22d70485983eac12b5b88dad1cf95  1325376051  43.150704 -112.154481   \n",
       "3  6b849c168bdad6f867558c3793159a81  1325376076  47.034331 -112.561071   \n",
       "4  a41d7549acf90789359a9aa5346dcb46  1325376186  38.674999  -78.632459   \n",
       "\n",
       "   is_fraud  \n",
       "0         0  \n",
       "1         0  \n",
       "2         0  \n",
       "3         0  \n",
       "4         0  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reading the train and test dataset\n",
    "\n",
    "df_train = pd.read_csv(\"fraudTrain.csv\")\n",
    "df_test = pd.read_csv('fraudTest.csv')\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "646f48db",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "72fbfd9d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1296675, 23)\n",
      "(555719, 23)\n"
     ]
    }
   ],
   "source": [
    "print(df_train.shape)\n",
    "print(df_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "46bb1f29",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 23 columns):\n",
      " #   Column                 Non-Null Count    Dtype  \n",
      "---  ------                 --------------    -----  \n",
      " 0   Unnamed: 0             1296675 non-null  int64  \n",
      " 1   trans_date_trans_time  1296675 non-null  object \n",
      " 2   cc_num                 1296675 non-null  int64  \n",
      " 3   merchant               1296675 non-null  object \n",
      " 4   category               1296675 non-null  object \n",
      " 5   amt                    1296675 non-null  float64\n",
      " 6   first                  1296675 non-null  object \n",
      " 7   last                   1296675 non-null  object \n",
      " 8   gender                 1296675 non-null  object \n",
      " 9   street                 1296675 non-null  object \n",
      " 10  city                   1296675 non-null  object \n",
      " 11  state                  1296675 non-null  object \n",
      " 12  zip                    1296675 non-null  int64  \n",
      " 13  lat                    1296675 non-null  float64\n",
      " 14  long                   1296675 non-null  float64\n",
      " 15  city_pop               1296675 non-null  int64  \n",
      " 16  job                    1296675 non-null  object \n",
      " 17  dob                    1296675 non-null  object \n",
      " 18  trans_num              1296675 non-null  object \n",
      " 19  unix_time              1296675 non-null  int64  \n",
      " 20  merch_lat              1296675 non-null  float64\n",
      " 21  merch_long             1296675 non-null  float64\n",
      " 22  is_fraud               1296675 non-null  int64  \n",
      "dtypes: float64(5), int64(6), object(12)\n",
      "memory usage: 227.5+ MB\n"
     ]
    }
   ],
   "source": [
    "df_train.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "eb60c2ab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 555719 entries, 0 to 555718\n",
      "Data columns (total 23 columns):\n",
      " #   Column                 Non-Null Count   Dtype  \n",
      "---  ------                 --------------   -----  \n",
      " 0   Unnamed: 0             555719 non-null  int64  \n",
      " 1   trans_date_trans_time  555719 non-null  object \n",
      " 2   cc_num                 555719 non-null  int64  \n",
      " 3   merchant               555719 non-null  object \n",
      " 4   category               555719 non-null  object \n",
      " 5   amt                    555719 non-null  float64\n",
      " 6   first                  555719 non-null  object \n",
      " 7   last                   555719 non-null  object \n",
      " 8   gender                 555719 non-null  object \n",
      " 9   street                 555719 non-null  object \n",
      " 10  city                   555719 non-null  object \n",
      " 11  state                  555719 non-null  object \n",
      " 12  zip                    555719 non-null  int64  \n",
      " 13  lat                    555719 non-null  float64\n",
      " 14  long                   555719 non-null  float64\n",
      " 15  city_pop               555719 non-null  int64  \n",
      " 16  job                    555719 non-null  object \n",
      " 17  dob                    555719 non-null  object \n",
      " 18  trans_num              555719 non-null  object \n",
      " 19  unix_time              555719 non-null  int64  \n",
      " 20  merch_lat              555719 non-null  float64\n",
      " 21  merch_long             555719 non-null  float64\n",
      " 22  is_fraud               555719 non-null  int64  \n",
      "dtypes: float64(5), int64(6), object(12)\n",
      "memory usage: 97.5+ MB\n"
     ]
    }
   ],
   "source": [
    "df_test.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "14ceb2cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>cc_num</th>\n",
       "      <th>merchant</th>\n",
       "      <th>category</th>\n",
       "      <th>amt</th>\n",
       "      <th>first</th>\n",
       "      <th>last</th>\n",
       "      <th>gender</th>\n",
       "      <th>street</th>\n",
       "      <th>city</th>\n",
       "      <th>...</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>city_pop</th>\n",
       "      <th>job</th>\n",
       "      <th>dob</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>unix_time</th>\n",
       "      <th>merch_lat</th>\n",
       "      <th>merch_long</th>\n",
       "      <th>is_fraud</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>2703186189652095</td>\n",
       "      <td>fraud_Rippin, Kub and Mann</td>\n",
       "      <td>misc_net</td>\n",
       "      <td>4.97</td>\n",
       "      <td>Jennifer</td>\n",
       "      <td>Banks</td>\n",
       "      <td>F</td>\n",
       "      <td>561 Perry Cove</td>\n",
       "      <td>Moravian Falls</td>\n",
       "      <td>...</td>\n",
       "      <td>36.0788</td>\n",
       "      <td>-81.1781</td>\n",
       "      <td>3495</td>\n",
       "      <td>Psychologist, counselling</td>\n",
       "      <td>1988-03-09</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>1325376018</td>\n",
       "      <td>36.011293</td>\n",
       "      <td>-82.048315</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>630423337322</td>\n",
       "      <td>fraud_Heller, Gutmann and Zieme</td>\n",
       "      <td>grocery_pos</td>\n",
       "      <td>107.23</td>\n",
       "      <td>Stephanie</td>\n",
       "      <td>Gill</td>\n",
       "      <td>F</td>\n",
       "      <td>43039 Riley Greens Suite 393</td>\n",
       "      <td>Orient</td>\n",
       "      <td>...</td>\n",
       "      <td>48.8878</td>\n",
       "      <td>-118.2105</td>\n",
       "      <td>149</td>\n",
       "      <td>Special educational needs teacher</td>\n",
       "      <td>1978-06-21</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>1325376044</td>\n",
       "      <td>49.159047</td>\n",
       "      <td>-118.186462</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>38859492057661</td>\n",
       "      <td>fraud_Lind-Buckridge</td>\n",
       "      <td>entertainment</td>\n",
       "      <td>220.11</td>\n",
       "      <td>Edward</td>\n",
       "      <td>Sanchez</td>\n",
       "      <td>M</td>\n",
       "      <td>594 White Dale Suite 530</td>\n",
       "      <td>Malad City</td>\n",
       "      <td>...</td>\n",
       "      <td>42.1808</td>\n",
       "      <td>-112.2620</td>\n",
       "      <td>4154</td>\n",
       "      <td>Nature conservation officer</td>\n",
       "      <td>1962-01-19</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>1325376051</td>\n",
       "      <td>43.150704</td>\n",
       "      <td>-112.154481</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>3534093764340240</td>\n",
       "      <td>fraud_Kutch, Hermiston and Farrell</td>\n",
       "      <td>gas_transport</td>\n",
       "      <td>45.00</td>\n",
       "      <td>Jeremy</td>\n",
       "      <td>White</td>\n",
       "      <td>M</td>\n",
       "      <td>9443 Cynthia Court Apt. 038</td>\n",
       "      <td>Boulder</td>\n",
       "      <td>...</td>\n",
       "      <td>46.2306</td>\n",
       "      <td>-112.1138</td>\n",
       "      <td>1939</td>\n",
       "      <td>Patent attorney</td>\n",
       "      <td>1967-01-12</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>1325376076</td>\n",
       "      <td>47.034331</td>\n",
       "      <td>-112.561071</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>375534208663984</td>\n",
       "      <td>fraud_Keeling-Crist</td>\n",
       "      <td>misc_pos</td>\n",
       "      <td>41.96</td>\n",
       "      <td>Tyler</td>\n",
       "      <td>Garcia</td>\n",
       "      <td>M</td>\n",
       "      <td>408 Bradley Rest</td>\n",
       "      <td>Doe Hill</td>\n",
       "      <td>...</td>\n",
       "      <td>38.4207</td>\n",
       "      <td>-79.4629</td>\n",
       "      <td>99</td>\n",
       "      <td>Dance movement psychotherapist</td>\n",
       "      <td>1986-03-28</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>1325376186</td>\n",
       "      <td>38.674999</td>\n",
       "      <td>-78.632459</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time            cc_num                            merchant  \\\n",
       "0   2019-01-01 00:00:18  2703186189652095          fraud_Rippin, Kub and Mann   \n",
       "1   2019-01-01 00:00:44      630423337322     fraud_Heller, Gutmann and Zieme   \n",
       "2   2019-01-01 00:00:51    38859492057661                fraud_Lind-Buckridge   \n",
       "3   2019-01-01 00:01:16  3534093764340240  fraud_Kutch, Hermiston and Farrell   \n",
       "4   2019-01-01 00:03:06   375534208663984                 fraud_Keeling-Crist   \n",
       "\n",
       "        category     amt      first     last gender  \\\n",
       "0       misc_net    4.97   Jennifer    Banks      F   \n",
       "1    grocery_pos  107.23  Stephanie     Gill      F   \n",
       "2  entertainment  220.11     Edward  Sanchez      M   \n",
       "3  gas_transport   45.00     Jeremy    White      M   \n",
       "4       misc_pos   41.96      Tyler   Garcia      M   \n",
       "\n",
       "                         street            city  ...      lat      long  \\\n",
       "0                561 Perry Cove  Moravian Falls  ...  36.0788  -81.1781   \n",
       "1  43039 Riley Greens Suite 393          Orient  ...  48.8878 -118.2105   \n",
       "2      594 White Dale Suite 530      Malad City  ...  42.1808 -112.2620   \n",
       "3   9443 Cynthia Court Apt. 038         Boulder  ...  46.2306 -112.1138   \n",
       "4              408 Bradley Rest        Doe Hill  ...  38.4207  -79.4629   \n",
       "\n",
       "   city_pop                                job         dob  \\\n",
       "0      3495          Psychologist, counselling  1988-03-09   \n",
       "1       149  Special educational needs teacher  1978-06-21   \n",
       "2      4154        Nature conservation officer  1962-01-19   \n",
       "3      1939                    Patent attorney  1967-01-12   \n",
       "4        99     Dance movement psychotherapist  1986-03-28   \n",
       "\n",
       "                          trans_num   unix_time  merch_lat  merch_long  \\\n",
       "0  0b242abb623afc578575680df30655b9  1325376018  36.011293  -82.048315   \n",
       "1  1f76529f8574734946361c461b024d99  1325376044  49.159047 -118.186462   \n",
       "2  a1a22d70485983eac12b5b88dad1cf95  1325376051  43.150704 -112.154481   \n",
       "3  6b849c168bdad6f867558c3793159a81  1325376076  47.034331 -112.561071   \n",
       "4  a41d7549acf90789359a9aa5346dcb46  1325376186  38.674999  -78.632459   \n",
       "\n",
       "   is_fraud  \n",
       "0         0  \n",
       "1         0  \n",
       "2         0  \n",
       "3         0  \n",
       "4         0  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dropping Unnamed column\n",
    "\n",
    "df_train.drop(\"Unnamed: 0\",axis=1,inplace=True)\n",
    "df_test.drop(\"Unnamed: 0\",axis=1,inplace=True)\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "db289fce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>cc_num</th>\n",
       "      <th>merchant</th>\n",
       "      <th>category</th>\n",
       "      <th>amt</th>\n",
       "      <th>first</th>\n",
       "      <th>last</th>\n",
       "      <th>gender</th>\n",
       "      <th>street</th>\n",
       "      <th>city</th>\n",
       "      <th>...</th>\n",
       "      <th>long</th>\n",
       "      <th>city_pop</th>\n",
       "      <th>job</th>\n",
       "      <th>dob</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>unix_time</th>\n",
       "      <th>merch_lat</th>\n",
       "      <th>merch_long</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>trans_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>2703186189652095</td>\n",
       "      <td>fraud_Rippin, Kub and Mann</td>\n",
       "      <td>misc_net</td>\n",
       "      <td>4.97</td>\n",
       "      <td>Jennifer</td>\n",
       "      <td>Banks</td>\n",
       "      <td>F</td>\n",
       "      <td>561 Perry Cove</td>\n",
       "      <td>Moravian Falls</td>\n",
       "      <td>...</td>\n",
       "      <td>-81.1781</td>\n",
       "      <td>3495</td>\n",
       "      <td>Psychologist, counselling</td>\n",
       "      <td>1988-03-09</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>1325376018</td>\n",
       "      <td>36.011293</td>\n",
       "      <td>-82.048315</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>630423337322</td>\n",
       "      <td>fraud_Heller, Gutmann and Zieme</td>\n",
       "      <td>grocery_pos</td>\n",
       "      <td>107.23</td>\n",
       "      <td>Stephanie</td>\n",
       "      <td>Gill</td>\n",
       "      <td>F</td>\n",
       "      <td>43039 Riley Greens Suite 393</td>\n",
       "      <td>Orient</td>\n",
       "      <td>...</td>\n",
       "      <td>-118.2105</td>\n",
       "      <td>149</td>\n",
       "      <td>Special educational needs teacher</td>\n",
       "      <td>1978-06-21</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>1325376044</td>\n",
       "      <td>49.159047</td>\n",
       "      <td>-118.186462</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>38859492057661</td>\n",
       "      <td>fraud_Lind-Buckridge</td>\n",
       "      <td>entertainment</td>\n",
       "      <td>220.11</td>\n",
       "      <td>Edward</td>\n",
       "      <td>Sanchez</td>\n",
       "      <td>M</td>\n",
       "      <td>594 White Dale Suite 530</td>\n",
       "      <td>Malad City</td>\n",
       "      <td>...</td>\n",
       "      <td>-112.2620</td>\n",
       "      <td>4154</td>\n",
       "      <td>Nature conservation officer</td>\n",
       "      <td>1962-01-19</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>1325376051</td>\n",
       "      <td>43.150704</td>\n",
       "      <td>-112.154481</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>3534093764340240</td>\n",
       "      <td>fraud_Kutch, Hermiston and Farrell</td>\n",
       "      <td>gas_transport</td>\n",
       "      <td>45.00</td>\n",
       "      <td>Jeremy</td>\n",
       "      <td>White</td>\n",
       "      <td>M</td>\n",
       "      <td>9443 Cynthia Court Apt. 038</td>\n",
       "      <td>Boulder</td>\n",
       "      <td>...</td>\n",
       "      <td>-112.1138</td>\n",
       "      <td>1939</td>\n",
       "      <td>Patent attorney</td>\n",
       "      <td>1967-01-12</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>1325376076</td>\n",
       "      <td>47.034331</td>\n",
       "      <td>-112.561071</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>375534208663984</td>\n",
       "      <td>fraud_Keeling-Crist</td>\n",
       "      <td>misc_pos</td>\n",
       "      <td>41.96</td>\n",
       "      <td>Tyler</td>\n",
       "      <td>Garcia</td>\n",
       "      <td>M</td>\n",
       "      <td>408 Bradley Rest</td>\n",
       "      <td>Doe Hill</td>\n",
       "      <td>...</td>\n",
       "      <td>-79.4629</td>\n",
       "      <td>99</td>\n",
       "      <td>Dance movement psychotherapist</td>\n",
       "      <td>1986-03-28</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>1325376186</td>\n",
       "      <td>38.674999</td>\n",
       "      <td>-78.632459</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time            cc_num                            merchant  \\\n",
       "0   2019-01-01 00:00:18  2703186189652095          fraud_Rippin, Kub and Mann   \n",
       "1   2019-01-01 00:00:44      630423337322     fraud_Heller, Gutmann and Zieme   \n",
       "2   2019-01-01 00:00:51    38859492057661                fraud_Lind-Buckridge   \n",
       "3   2019-01-01 00:01:16  3534093764340240  fraud_Kutch, Hermiston and Farrell   \n",
       "4   2019-01-01 00:03:06   375534208663984                 fraud_Keeling-Crist   \n",
       "\n",
       "        category     amt      first     last gender  \\\n",
       "0       misc_net    4.97   Jennifer    Banks      F   \n",
       "1    grocery_pos  107.23  Stephanie     Gill      F   \n",
       "2  entertainment  220.11     Edward  Sanchez      M   \n",
       "3  gas_transport   45.00     Jeremy    White      M   \n",
       "4       misc_pos   41.96      Tyler   Garcia      M   \n",
       "\n",
       "                         street            city  ...      long  city_pop  \\\n",
       "0                561 Perry Cove  Moravian Falls  ...  -81.1781      3495   \n",
       "1  43039 Riley Greens Suite 393          Orient  ... -118.2105       149   \n",
       "2      594 White Dale Suite 530      Malad City  ... -112.2620      4154   \n",
       "3   9443 Cynthia Court Apt. 038         Boulder  ... -112.1138      1939   \n",
       "4              408 Bradley Rest        Doe Hill  ...  -79.4629        99   \n",
       "\n",
       "                                 job        dob  \\\n",
       "0          Psychologist, counselling 1988-03-09   \n",
       "1  Special educational needs teacher 1978-06-21   \n",
       "2        Nature conservation officer 1962-01-19   \n",
       "3                    Patent attorney 1967-01-12   \n",
       "4     Dance movement psychotherapist 1986-03-28   \n",
       "\n",
       "                          trans_num   unix_time  merch_lat  merch_long  \\\n",
       "0  0b242abb623afc578575680df30655b9  1325376018  36.011293  -82.048315   \n",
       "1  1f76529f8574734946361c461b024d99  1325376044  49.159047 -118.186462   \n",
       "2  a1a22d70485983eac12b5b88dad1cf95  1325376051  43.150704 -112.154481   \n",
       "3  6b849c168bdad6f867558c3793159a81  1325376076  47.034331 -112.561071   \n",
       "4  a41d7549acf90789359a9aa5346dcb46  1325376186  38.674999  -78.632459   \n",
       "\n",
       "   is_fraud  trans_date  \n",
       "0         0  2019-01-01  \n",
       "1         0  2019-01-01  \n",
       "2         0  2019-01-01  \n",
       "3         0  2019-01-01  \n",
       "4         0  2019-01-01  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Converting date columns to datetime format\n",
    "\n",
    "df_train['trans_date_trans_time']=pd.to_datetime(df_train['trans_date_trans_time'])\n",
    "df_train['trans_date']=df_train['trans_date_trans_time'].dt.strftime('%Y-%m-%d')\n",
    "df_train['trans_date']=pd.to_datetime(df_train['trans_date'])\n",
    "df_train['dob']=pd.to_datetime(df_train['dob'])\n",
    "\n",
    "df_test['trans_date_trans_time']=pd.to_datetime(df_test['trans_date_trans_time'])\n",
    "df_test['trans_date']=df_test['trans_date_trans_time'].dt.strftime('%Y-%m-%d')\n",
    "df_test['trans_date']=pd.to_datetime(df_test['trans_date'])\n",
    "df_test['dob']=pd.to_datetime(df_test['dob'])\n",
    "\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "47684ead",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 23 columns):\n",
      " #   Column                 Non-Null Count    Dtype         \n",
      "---  ------                 --------------    -----         \n",
      " 0   trans_date_trans_time  1296675 non-null  datetime64[ns]\n",
      " 1   cc_num                 1296675 non-null  int64         \n",
      " 2   merchant               1296675 non-null  object        \n",
      " 3   category               1296675 non-null  object        \n",
      " 4   amt                    1296675 non-null  float64       \n",
      " 5   first                  1296675 non-null  object        \n",
      " 6   last                   1296675 non-null  object        \n",
      " 7   gender                 1296675 non-null  object        \n",
      " 8   street                 1296675 non-null  object        \n",
      " 9   city                   1296675 non-null  object        \n",
      " 10  state                  1296675 non-null  object        \n",
      " 11  zip                    1296675 non-null  int64         \n",
      " 12  lat                    1296675 non-null  float64       \n",
      " 13  long                   1296675 non-null  float64       \n",
      " 14  city_pop               1296675 non-null  int64         \n",
      " 15  job                    1296675 non-null  object        \n",
      " 16  dob                    1296675 non-null  datetime64[ns]\n",
      " 17  trans_num              1296675 non-null  object        \n",
      " 18  unix_time              1296675 non-null  int64         \n",
      " 19  merch_lat              1296675 non-null  float64       \n",
      " 20  merch_long             1296675 non-null  float64       \n",
      " 21  is_fraud               1296675 non-null  int64         \n",
      " 22  trans_date             1296675 non-null  datetime64[ns]\n",
      "dtypes: datetime64[ns](3), float64(5), int64(5), object(10)\n",
      "memory usage: 227.5+ MB\n"
     ]
    }
   ],
   "source": [
    "df_train.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7f989382",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 555719 entries, 0 to 555718\n",
      "Data columns (total 23 columns):\n",
      " #   Column                 Non-Null Count   Dtype         \n",
      "---  ------                 --------------   -----         \n",
      " 0   trans_date_trans_time  555719 non-null  datetime64[ns]\n",
      " 1   cc_num                 555719 non-null  int64         \n",
      " 2   merchant               555719 non-null  object        \n",
      " 3   category               555719 non-null  object        \n",
      " 4   amt                    555719 non-null  float64       \n",
      " 5   first                  555719 non-null  object        \n",
      " 6   last                   555719 non-null  object        \n",
      " 7   gender                 555719 non-null  object        \n",
      " 8   street                 555719 non-null  object        \n",
      " 9   city                   555719 non-null  object        \n",
      " 10  state                  555719 non-null  object        \n",
      " 11  zip                    555719 non-null  int64         \n",
      " 12  lat                    555719 non-null  float64       \n",
      " 13  long                   555719 non-null  float64       \n",
      " 14  city_pop               555719 non-null  int64         \n",
      " 15  job                    555719 non-null  object        \n",
      " 16  dob                    555719 non-null  datetime64[ns]\n",
      " 17  trans_num              555719 non-null  object        \n",
      " 18  unix_time              555719 non-null  int64         \n",
      " 19  merch_lat              555719 non-null  float64       \n",
      " 20  merch_long             555719 non-null  float64       \n",
      " 21  is_fraud               555719 non-null  int64         \n",
      " 22  trans_date             555719 non-null  datetime64[ns]\n",
      "dtypes: datetime64[ns](3), float64(5), int64(5), object(10)\n",
      "memory usage: 97.5+ MB\n"
     ]
    }
   ],
   "source": [
    "df_test.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "56bb3c3c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 15 columns):\n",
      " #   Column                 Non-Null Count    Dtype         \n",
      "---  ------                 --------------    -----         \n",
      " 0   trans_date_trans_time  1296675 non-null  datetime64[ns]\n",
      " 1   category               1296675 non-null  object        \n",
      " 2   amt                    1296675 non-null  float64       \n",
      " 3   gender                 1296675 non-null  object        \n",
      " 4   city                   1296675 non-null  object        \n",
      " 5   state                  1296675 non-null  object        \n",
      " 6   lat                    1296675 non-null  float64       \n",
      " 7   long                   1296675 non-null  float64       \n",
      " 8   city_pop               1296675 non-null  int64         \n",
      " 9   job                    1296675 non-null  object        \n",
      " 10  dob                    1296675 non-null  datetime64[ns]\n",
      " 11  merch_lat              1296675 non-null  float64       \n",
      " 12  merch_long             1296675 non-null  float64       \n",
      " 13  is_fraud               1296675 non-null  int64         \n",
      " 14  trans_date             1296675 non-null  datetime64[ns]\n",
      "dtypes: datetime64[ns](3), float64(5), int64(2), object(5)\n",
      "memory usage: 148.4+ MB\n"
     ]
    }
   ],
   "source": [
    "# Dropping fields not necessary for model building\n",
    "\n",
    "drop_cols = ['cc_num','merchant','first','last','street','zip','trans_num','unix_time']\n",
    "\n",
    "df_train1 = df_train.drop(drop_cols,axis=1)\n",
    "df_test1 = df_test.drop(drop_cols,axis=1)\n",
    "\n",
    "df_train1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "51dde319",
   "metadata": {},
   "outputs": [
    {
     "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": [
    "# Checking Class Imbalance\n",
    "\n",
    "classes_train=df_train1['is_fraud'].value_counts()\n",
    "normal_share_train=classes_train[0]/df_train1['is_fraud'].count()*100\n",
    "fraud_share_train=classes_train[1]/df_train1['is_fraud'].count()*100\n",
    "\n",
    "classes_test=df_test1['is_fraud'].value_counts()\n",
    "normal_share_test=classes_test[0]/df_test1['is_fraud'].count()*100\n",
    "fraud_share_test=classes_test[1]/df_test1['is_fraud'].count()*100\n",
    "\n",
    "plt.subplot(2,1,1)\n",
    "plt.bar(['Non-Fraud','Fraud'], classes_train, color=['y','b'])\n",
    "plt.title('Train')\n",
    "plt.ylabel('Number of transactions')\n",
    "plt.annotate(\"{0:.4}%\".format(normal_share_train),(0.2, 0.5), xycoords='axes fraction')\n",
    "plt.annotate(\"{0:.4}%\".format(fraud_share_train),(0.7, 0.5), xycoords='axes fraction')\n",
    "\n",
    "plt.subplot(2,1,2)\n",
    "plt.bar(['Non-Fraud','Fraud'], classes_test, color=['y','b'])\n",
    "plt.title('Test')\n",
    "plt.ylabel('Number of transactions')\n",
    "plt.annotate(\"{0:.4}%\".format(normal_share_test),(0.2, 0.5), xycoords='axes fraction')\n",
    "plt.annotate(\"{0:.4}%\".format(fraud_share_test),(0.7, 0.5), xycoords='axes fraction')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "eb9de121",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>category</th>\n",
       "      <th>amt</th>\n",
       "      <th>gender</th>\n",
       "      <th>city</th>\n",
       "      <th>state</th>\n",
       "      <th>lat</th>\n",
       "      <th>long</th>\n",
       "      <th>city_pop</th>\n",
       "      <th>job</th>\n",
       "      <th>dob</th>\n",
       "      <th>merch_lat</th>\n",
       "      <th>merch_long</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>trans_date</th>\n",
       "      <th>age_at_trans</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>misc_net</td>\n",
       "      <td>4.97</td>\n",
       "      <td>F</td>\n",
       "      <td>Moravian Falls</td>\n",
       "      <td>NC</td>\n",
       "      <td>36.0788</td>\n",
       "      <td>-81.1781</td>\n",
       "      <td>3495</td>\n",
       "      <td>Psychologist, counselling</td>\n",
       "      <td>1988-03-09</td>\n",
       "      <td>36.011293</td>\n",
       "      <td>-82.048315</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>grocery_pos</td>\n",
       "      <td>107.23</td>\n",
       "      <td>F</td>\n",
       "      <td>Orient</td>\n",
       "      <td>WA</td>\n",
       "      <td>48.8878</td>\n",
       "      <td>-118.2105</td>\n",
       "      <td>149</td>\n",
       "      <td>Special educational needs teacher</td>\n",
       "      <td>1978-06-21</td>\n",
       "      <td>49.159047</td>\n",
       "      <td>-118.186462</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>40.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>entertainment</td>\n",
       "      <td>220.11</td>\n",
       "      <td>M</td>\n",
       "      <td>Malad City</td>\n",
       "      <td>ID</td>\n",
       "      <td>42.1808</td>\n",
       "      <td>-112.2620</td>\n",
       "      <td>4154</td>\n",
       "      <td>Nature conservation officer</td>\n",
       "      <td>1962-01-19</td>\n",
       "      <td>43.150704</td>\n",
       "      <td>-112.154481</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>56.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>gas_transport</td>\n",
       "      <td>45.00</td>\n",
       "      <td>M</td>\n",
       "      <td>Boulder</td>\n",
       "      <td>MT</td>\n",
       "      <td>46.2306</td>\n",
       "      <td>-112.1138</td>\n",
       "      <td>1939</td>\n",
       "      <td>Patent attorney</td>\n",
       "      <td>1967-01-12</td>\n",
       "      <td>47.034331</td>\n",
       "      <td>-112.561071</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>51.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>misc_pos</td>\n",
       "      <td>41.96</td>\n",
       "      <td>M</td>\n",
       "      <td>Doe Hill</td>\n",
       "      <td>VA</td>\n",
       "      <td>38.4207</td>\n",
       "      <td>-79.4629</td>\n",
       "      <td>99</td>\n",
       "      <td>Dance movement psychotherapist</td>\n",
       "      <td>1986-03-28</td>\n",
       "      <td>38.674999</td>\n",
       "      <td>-78.632459</td>\n",
       "      <td>0</td>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>32.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time       category     amt gender            city state  \\\n",
       "0   2019-01-01 00:00:18       misc_net    4.97      F  Moravian Falls    NC   \n",
       "1   2019-01-01 00:00:44    grocery_pos  107.23      F          Orient    WA   \n",
       "2   2019-01-01 00:00:51  entertainment  220.11      M      Malad City    ID   \n",
       "3   2019-01-01 00:01:16  gas_transport   45.00      M         Boulder    MT   \n",
       "4   2019-01-01 00:03:06       misc_pos   41.96      M        Doe Hill    VA   \n",
       "\n",
       "       lat      long  city_pop                                job        dob  \\\n",
       "0  36.0788  -81.1781      3495          Psychologist, counselling 1988-03-09   \n",
       "1  48.8878 -118.2105       149  Special educational needs teacher 1978-06-21   \n",
       "2  42.1808 -112.2620      4154        Nature conservation officer 1962-01-19   \n",
       "3  46.2306 -112.1138      1939                    Patent attorney 1967-01-12   \n",
       "4  38.4207  -79.4629        99     Dance movement psychotherapist 1986-03-28   \n",
       "\n",
       "   merch_lat  merch_long  is_fraud trans_date  age_at_trans  \n",
       "0  36.011293  -82.048315         0 2019-01-01          30.0  \n",
       "1  49.159047 -118.186462         0 2019-01-01          40.0  \n",
       "2  43.150704 -112.154481         0 2019-01-01          56.0  \n",
       "3  47.034331 -112.561071         0 2019-01-01          51.0  \n",
       "4  38.674999  -78.632459         0 2019-01-01          32.0  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating field for age on transaction date\n",
    "\n",
    "df_train1['age_at_trans'] = df_train1['trans_date'] - df_train1['dob']\n",
    "df_train1['age_at_trans'] = df_train1['age_at_trans'].astype('timedelta64[Y]')\n",
    "\n",
    "df_test1['age_at_trans'] = df_test1['trans_date'] - df_test1['dob']\n",
    "df_test1['age_at_trans'] = df_test1['age_at_trans'].astype('timedelta64[Y]')\n",
    "\n",
    "df_train1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c78097b1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 16 columns):\n",
      " #   Column                 Non-Null Count    Dtype         \n",
      "---  ------                 --------------    -----         \n",
      " 0   trans_date_trans_time  1296675 non-null  datetime64[ns]\n",
      " 1   category               1296675 non-null  object        \n",
      " 2   amt                    1296675 non-null  float64       \n",
      " 3   gender                 1296675 non-null  object        \n",
      " 4   city                   1296675 non-null  object        \n",
      " 5   state                  1296675 non-null  object        \n",
      " 6   lat                    1296675 non-null  float64       \n",
      " 7   long                   1296675 non-null  float64       \n",
      " 8   city_pop               1296675 non-null  int64         \n",
      " 9   job                    1296675 non-null  object        \n",
      " 10  dob                    1296675 non-null  datetime64[ns]\n",
      " 11  merch_lat              1296675 non-null  float64       \n",
      " 12  merch_long             1296675 non-null  float64       \n",
      " 13  is_fraud               1296675 non-null  int64         \n",
      " 14  trans_date             1296675 non-null  datetime64[ns]\n",
      " 15  age_at_trans           1296675 non-null  float64       \n",
      "dtypes: datetime64[ns](3), float64(6), int64(2), object(5)\n",
      "memory usage: 158.3+ MB\n"
     ]
    }
   ],
   "source": [
    "df_train1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "79c70be9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 555719 entries, 0 to 555718\n",
      "Data columns (total 16 columns):\n",
      " #   Column                 Non-Null Count   Dtype         \n",
      "---  ------                 --------------   -----         \n",
      " 0   trans_date_trans_time  555719 non-null  datetime64[ns]\n",
      " 1   category               555719 non-null  object        \n",
      " 2   amt                    555719 non-null  float64       \n",
      " 3   gender                 555719 non-null  object        \n",
      " 4   city                   555719 non-null  object        \n",
      " 5   state                  555719 non-null  object        \n",
      " 6   lat                    555719 non-null  float64       \n",
      " 7   long                   555719 non-null  float64       \n",
      " 8   city_pop               555719 non-null  int64         \n",
      " 9   job                    555719 non-null  object        \n",
      " 10  dob                    555719 non-null  datetime64[ns]\n",
      " 11  merch_lat              555719 non-null  float64       \n",
      " 12  merch_long             555719 non-null  float64       \n",
      " 13  is_fraud               555719 non-null  int64         \n",
      " 14  trans_date             555719 non-null  datetime64[ns]\n",
      " 15  age_at_trans           555719 non-null  float64       \n",
      "dtypes: datetime64[ns](3), float64(6), int64(2), object(5)\n",
      "memory usage: 67.8+ MB\n"
     ]
    }
   ],
   "source": [
    "df_test1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b9c98957",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Achille', 'Acworth', 'Afton', 'Albany', 'Albuquerque', 'Alder',\n",
       "       'Aledo', 'Alexandria', 'Allenhurst', 'Allentown',\n",
       "       ...\n",
       "       'Wilmington', 'Winfield', 'Winger', 'Winnsboro', 'Winslow', 'Winter',\n",
       "       'Winthrop', 'Wittenberg', 'Yellowstone National Park', 'Zaleski'],\n",
       "      dtype='object', name='city', length=702)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fraud_city = pd.DataFrame(df_train1.groupby('city')['is_fraud'].sum())\n",
    "fraud_city = fraud_city[fraud_city['is_fraud']>0]\n",
    "fraud_city.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a234e4a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['AK', 'AL', 'AR', 'AZ', 'CA', 'CO', 'CT', 'DC', 'DE', 'FL', 'GA', 'HI',\n",
       "       'IA', 'ID', 'IL', 'IN', 'KS', 'KY', 'LA', 'MA', 'MD', 'ME', 'MI', 'MN',\n",
       "       'MO', 'MS', 'MT', 'NC', 'ND', 'NE', 'NH', 'NJ', 'NM', 'NV', 'NY', 'OH',\n",
       "       'OK', 'OR', 'PA', 'RI', 'SC', 'SD', 'TN', 'TX', 'UT', 'VA', 'VT', 'WA',\n",
       "       'WI', 'WV', 'WY'],\n",
       "      dtype='object', name='state')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fraud_state = pd.DataFrame(df_train1.groupby('state')['is_fraud'].sum())\n",
    "fraud_state = fraud_state[fraud_state['is_fraud']>0]\n",
    "fraud_state.index\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e35d3650",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Academic librarian', 'Accountant, chartered',\n",
       "       'Accountant, chartered certified',\n",
       "       'Accountant, chartered public finance', 'Accounting technician',\n",
       "       'Acupuncturist', 'Administrator', 'Administrator, arts',\n",
       "       'Administrator, charities/voluntary organisations',\n",
       "       'Administrator, education',\n",
       "       ...\n",
       "       'Veterinary surgeon', 'Video editor', 'Volunteer coordinator',\n",
       "       'Warden/ranger', 'Warehouse manager', 'Waste management officer',\n",
       "       'Water quality scientist', 'Web designer', 'Wellsite geologist',\n",
       "       'Writer'],\n",
       "      dtype='object', name='job', length=443)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fraud_job = pd.DataFrame(df_train1.groupby('job')['is_fraud'].sum())\n",
    "fraud_job = fraud_job[fraud_job['is_fraud']>0]\n",
    "fraud_job.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "9061d462",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>merch_lat</th>\n",
       "      <th>lat</th>\n",
       "      <th>lat_dist</th>\n",
       "      <th>merch_long</th>\n",
       "      <th>long</th>\n",
       "      <th>long_dist</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>36.011293</td>\n",
       "      <td>36.0788</td>\n",
       "      <td>0.068</td>\n",
       "      <td>-82.048315</td>\n",
       "      <td>-81.1781</td>\n",
       "      <td>0.870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>49.159047</td>\n",
       "      <td>48.8878</td>\n",
       "      <td>0.271</td>\n",
       "      <td>-118.186462</td>\n",
       "      <td>-118.2105</td>\n",
       "      <td>0.024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>43.150704</td>\n",
       "      <td>42.1808</td>\n",
       "      <td>0.970</td>\n",
       "      <td>-112.154481</td>\n",
       "      <td>-112.2620</td>\n",
       "      <td>0.108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>47.034331</td>\n",
       "      <td>46.2306</td>\n",
       "      <td>0.804</td>\n",
       "      <td>-112.561071</td>\n",
       "      <td>-112.1138</td>\n",
       "      <td>0.447</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>38.674999</td>\n",
       "      <td>38.4207</td>\n",
       "      <td>0.254</td>\n",
       "      <td>-78.632459</td>\n",
       "      <td>-79.4629</td>\n",
       "      <td>0.830</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   merch_lat      lat  lat_dist  merch_long      long  long_dist\n",
       "0  36.011293  36.0788     0.068  -82.048315  -81.1781      0.870\n",
       "1  49.159047  48.8878     0.271 -118.186462 -118.2105      0.024\n",
       "2  43.150704  42.1808     0.970 -112.154481 -112.2620      0.108\n",
       "3  47.034331  46.2306     0.804 -112.561071 -112.1138      0.447\n",
       "4  38.674999  38.4207     0.254  -78.632459  -79.4629      0.830"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Finding distance from customer location to merchant location in degrees latitude and degrees longitude\n",
    "\n",
    "df_train1['lat_dist'] = abs(round(df_train1['merch_lat']-df_train1['lat'],3))\n",
    "df_train1['long_dist'] = abs(round(df_train1['merch_long']-df_train1['long'],3))\n",
    "\n",
    "df_test1['lat_dist'] = abs(round(df_test1['merch_lat']-df_test1['lat'],3))\n",
    "df_test1['long_dist'] = abs(round(df_test1['merch_long']-df_test1['long'],3))\n",
    "\n",
    "df_train1[['merch_lat','lat','lat_dist','merch_long','long','long_dist']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e060adaa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 18 columns):\n",
      " #   Column                 Non-Null Count    Dtype         \n",
      "---  ------                 --------------    -----         \n",
      " 0   trans_date_trans_time  1296675 non-null  datetime64[ns]\n",
      " 1   category               1296675 non-null  object        \n",
      " 2   amt                    1296675 non-null  float64       \n",
      " 3   gender                 1296675 non-null  object        \n",
      " 4   city                   1296675 non-null  object        \n",
      " 5   state                  1296675 non-null  object        \n",
      " 6   lat                    1296675 non-null  float64       \n",
      " 7   long                   1296675 non-null  float64       \n",
      " 8   city_pop               1296675 non-null  int64         \n",
      " 9   job                    1296675 non-null  object        \n",
      " 10  dob                    1296675 non-null  datetime64[ns]\n",
      " 11  merch_lat              1296675 non-null  float64       \n",
      " 12  merch_long             1296675 non-null  float64       \n",
      " 13  is_fraud               1296675 non-null  int64         \n",
      " 14  trans_date             1296675 non-null  datetime64[ns]\n",
      " 15  age_at_trans           1296675 non-null  float64       \n",
      " 16  lat_dist               1296675 non-null  float64       \n",
      " 17  long_dist              1296675 non-null  float64       \n",
      "dtypes: datetime64[ns](3), float64(8), int64(2), object(5)\n",
      "memory usage: 178.1+ MB\n"
     ]
    }
   ],
   "source": [
    "df_train1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "008a0c2b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>trans_month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time  trans_month\n",
       "0   2019-01-01 00:00:18            1\n",
       "1   2019-01-01 00:00:44            1\n",
       "2   2019-01-01 00:00:51            1\n",
       "3   2019-01-01 00:01:16            1\n",
       "4   2019-01-01 00:03:06            1"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating month column\n",
    "\n",
    "df_train1['trans_month'] = pd.DatetimeIndex(df_train1['trans_date']).month\n",
    "df_test1['trans_month'] = pd.DatetimeIndex(df_test1['trans_date']).month\n",
    "\n",
    "df_train1[['trans_date_trans_time','trans_month']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "0750208c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 19 columns):\n",
      " #   Column                 Non-Null Count    Dtype         \n",
      "---  ------                 --------------    -----         \n",
      " 0   trans_date_trans_time  1296675 non-null  datetime64[ns]\n",
      " 1   category               1296675 non-null  object        \n",
      " 2   amt                    1296675 non-null  float64       \n",
      " 3   gender                 1296675 non-null  object        \n",
      " 4   city                   1296675 non-null  object        \n",
      " 5   state                  1296675 non-null  object        \n",
      " 6   lat                    1296675 non-null  float64       \n",
      " 7   long                   1296675 non-null  float64       \n",
      " 8   city_pop               1296675 non-null  int64         \n",
      " 9   job                    1296675 non-null  object        \n",
      " 10  dob                    1296675 non-null  datetime64[ns]\n",
      " 11  merch_lat              1296675 non-null  float64       \n",
      " 12  merch_long             1296675 non-null  float64       \n",
      " 13  is_fraud               1296675 non-null  int64         \n",
      " 14  trans_date             1296675 non-null  datetime64[ns]\n",
      " 15  age_at_trans           1296675 non-null  float64       \n",
      " 16  lat_dist               1296675 non-null  float64       \n",
      " 17  long_dist              1296675 non-null  float64       \n",
      " 18  trans_month            1296675 non-null  int64         \n",
      "dtypes: datetime64[ns](3), float64(8), int64(3), object(5)\n",
      "memory usage: 188.0+ MB\n"
     ]
    }
   ],
   "source": [
    "df_train1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "3f43f3b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "trans_date_trans_time    1274791\n",
       "category                      14\n",
       "amt                        52928\n",
       "gender                         2\n",
       "city                         894\n",
       "state                         51\n",
       "lat                          968\n",
       "long                         969\n",
       "city_pop                     879\n",
       "job                          494\n",
       "dob                          968\n",
       "merch_lat                1247805\n",
       "merch_long               1275745\n",
       "is_fraud                       2\n",
       "trans_date                   537\n",
       "age_at_trans                  83\n",
       "lat_dist                    1001\n",
       "long_dist                   1001\n",
       "trans_month                   12\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train1.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "771cc93d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "trans_date_trans_time    544760\n",
       "category                     14\n",
       "amt                       37256\n",
       "gender                        2\n",
       "city                        849\n",
       "state                        50\n",
       "lat                         910\n",
       "long                        910\n",
       "city_pop                    835\n",
       "job                         478\n",
       "dob                         910\n",
       "merch_lat                546490\n",
       "merch_long               551770\n",
       "is_fraud                      2\n",
       "trans_date                  194\n",
       "age_at_trans                 82\n",
       "lat_dist                   1001\n",
       "long_dist                  1001\n",
       "trans_month                   7\n",
       "dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test1.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b882fcdf",
   "metadata": {},
   "source": [
    "Checking if there are any highly skewed categorical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "1771b202",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gas_transport     10.15\n",
       "grocery_pos        9.54\n",
       "home               9.49\n",
       "shopping_pos       9.00\n",
       "kids_pets          8.72\n",
       "shopping_net       7.52\n",
       "entertainment      7.25\n",
       "food_dining        7.05\n",
       "personal_care      7.00\n",
       "health_fitness     6.62\n",
       "misc_pos           6.14\n",
       "misc_net           4.88\n",
       "grocery_net        3.51\n",
       "travel             3.12\n",
       "Name: category, dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(df_train1['category'].value_counts()/len(df_train1.index)*100,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "6e7318f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "F    54.74\n",
       "M    45.26\n",
       "Name: gender, dtype: float64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(df_train1['gender'].value_counts()/len(df_train1.index)*100,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "fb561507",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Birmingham                   0.43\n",
       "San Antonio                  0.40\n",
       "Utica                        0.39\n",
       "Phoenix                      0.39\n",
       "Meridian                     0.39\n",
       "Thomas                       0.36\n",
       "Conway                       0.36\n",
       "Cleveland                    0.36\n",
       "Warren                       0.35\n",
       "Houston                      0.32\n",
       "Burbank                      0.32\n",
       "Naples                       0.32\n",
       "Arcadia                      0.32\n",
       "Indianapolis                 0.31\n",
       "Brandon                      0.31\n",
       "Fulton                       0.31\n",
       "Lahoma                       0.28\n",
       "Allentown                    0.28\n",
       "Lakeland                     0.28\n",
       "Dallas                       0.28\n",
       "Washington                   0.28\n",
       "Andrews                      0.28\n",
       "Detroit                      0.28\n",
       "Huntsville                   0.27\n",
       "Hudson                       0.27\n",
       "Orient                       0.27\n",
       "Fort Washakie                0.27\n",
       "Philadelphia                 0.27\n",
       "Uledi                        0.24\n",
       "Elberta                      0.24\n",
       "Superior                     0.24\n",
       "Bradley                      0.24\n",
       "Moorhead                     0.24\n",
       "Hinesburg                    0.24\n",
       "Wilmington                   0.24\n",
       "Hinckley                     0.24\n",
       "Bowdoin                      0.24\n",
       "De Witt                      0.24\n",
       "Morrisdale                   0.24\n",
       "Cottekill                    0.24\n",
       "East Canaan                  0.24\n",
       "Littleton                    0.24\n",
       "Centerview                   0.24\n",
       "Tulsa                        0.24\n",
       "Jones                        0.24\n",
       "De Soto                      0.24\n",
       "Bauxite                      0.24\n",
       "Plainfield                   0.24\n",
       "Whaleyville                  0.24\n",
       "Wetmore                      0.24\n",
       "Heart Butte                  0.24\n",
       "Walnut Ridge                 0.24\n",
       "Norwalk                      0.24\n",
       "Thrall                       0.24\n",
       "Baton Rouge                  0.24\n",
       "Sutherland                   0.24\n",
       "Gadsden                      0.24\n",
       "Shields                      0.24\n",
       "Tupper Lake                  0.24\n",
       "Goodrich                     0.24\n",
       "Florence                     0.23\n",
       "Rocky Mount                  0.23\n",
       "Thompson                     0.23\n",
       "Grandview                    0.23\n",
       "Norman                       0.23\n",
       "Roma                         0.23\n",
       "Newhall                      0.23\n",
       "Topeka                       0.23\n",
       "Sebring                      0.23\n",
       "Manistique                   0.23\n",
       "Sun City                     0.23\n",
       "Reno                         0.23\n",
       "Lomax                        0.23\n",
       "Ranier                       0.23\n",
       "Westport                     0.23\n",
       "Thida                        0.23\n",
       "Kingsford Heights            0.23\n",
       "Clarks Mills                 0.23\n",
       "Edisto Island                0.23\n",
       "Pembroke Township            0.22\n",
       "Ruckersville                 0.20\n",
       "Oakland                      0.20\n",
       "Glendale                     0.20\n",
       "Espanola                     0.20\n",
       "Issaquah                     0.20\n",
       "Ronceverte                   0.20\n",
       "Lawrence                     0.20\n",
       "San Diego                    0.20\n",
       "Sachse                       0.20\n",
       "Sherman                      0.20\n",
       "Knoxville                    0.20\n",
       "Johns Island                 0.20\n",
       "Grover                       0.20\n",
       "Bagley                       0.20\n",
       "Key West                     0.20\n",
       "West Sayville                0.20\n",
       "Scotland                     0.20\n",
       "Red River                    0.20\n",
       "Bellmore                     0.20\n",
       "Murfreesboro                 0.20\n",
       "Esbon                        0.20\n",
       "Stanchfield                  0.20\n",
       "Lanark Village               0.20\n",
       "North Wilkesboro             0.20\n",
       "New Holstein                 0.20\n",
       "Elizabeth                    0.20\n",
       "Kensington                   0.20\n",
       "Moundsville                  0.20\n",
       "Smiths Grove                 0.20\n",
       "Leonard                      0.20\n",
       "New York City                0.20\n",
       "Rock Springs                 0.20\n",
       "Alpharetta                   0.20\n",
       "Hannawa Falls                0.20\n",
       "Edmond                       0.20\n",
       "Muskegon                     0.20\n",
       "Wichita                      0.20\n",
       "Lake Jackson                 0.20\n",
       "Meadville                    0.20\n",
       "Blooming Grove               0.20\n",
       "Haw River                    0.20\n",
       "Moriarty                     0.20\n",
       "Bessemer                     0.20\n",
       "Spencer                      0.20\n",
       "Cochranton                   0.20\n",
       "Winfield                     0.20\n",
       "Fenelton                     0.20\n",
       "Romulus                      0.20\n",
       "Springfield                  0.20\n",
       "Drakes Branch                0.20\n",
       "Diamond                      0.20\n",
       "Oolitic                      0.20\n",
       "Denham Springs               0.20\n",
       "Armagh                       0.20\n",
       "Mobile                       0.20\n",
       "Camden                       0.20\n",
       "Mulberry Grove               0.20\n",
       "Redford                      0.20\n",
       "Milford                      0.19\n",
       "Fayetteville                 0.19\n",
       "Saint Amant                  0.19\n",
       "Fort Myers                   0.19\n",
       "Hancock                      0.19\n",
       "Brunson                      0.19\n",
       "Tekoa                        0.19\n",
       "Murrayville                  0.19\n",
       "Woodville                    0.16\n",
       "Harrodsburg                  0.16\n",
       "Tyler                        0.16\n",
       "Barneveld                    0.16\n",
       "Acworth                      0.16\n",
       "Saxon                        0.16\n",
       "Wilmette                     0.16\n",
       "Napa                         0.16\n",
       "Barnstable                   0.16\n",
       "Basye                        0.16\n",
       "Altonah                      0.16\n",
       "Grassflat                    0.16\n",
       "Coffeeville                  0.16\n",
       "Mount Hope                   0.16\n",
       "Sontag                       0.16\n",
       "Mount Saint Joseph           0.16\n",
       "Bridger                      0.16\n",
       "Holstein                     0.16\n",
       "Steuben                      0.16\n",
       "Falconer                     0.16\n",
       "Catawba                      0.16\n",
       "Cisco                        0.16\n",
       "Comfrey                      0.16\n",
       "Summerfield                  0.16\n",
       "Cokeburg                     0.16\n",
       "Pointe Aux Pins              0.16\n",
       "Heiskell                     0.16\n",
       "Creedmoor                    0.16\n",
       "High Rolls Mountain Park     0.16\n",
       "Lawn                         0.16\n",
       "Lohrville                    0.16\n",
       "Corriganville                0.16\n",
       "Atlantic                     0.16\n",
       "Brownville                   0.16\n",
       "Noonan                       0.16\n",
       "New Boston                   0.16\n",
       "Dunlevy                      0.16\n",
       "Holliday                     0.16\n",
       "Louisville                   0.16\n",
       "North Prairie                0.16\n",
       "Saint Paul                   0.16\n",
       "Keisterville                 0.16\n",
       "Oconto Falls                 0.16\n",
       "Mc Veytown                   0.16\n",
       "Mc Nabb                      0.16\n",
       "Winger                       0.16\n",
       "Methuen                      0.16\n",
       "Scotia                       0.16\n",
       "Pittsburgh                   0.16\n",
       "Holcomb                      0.16\n",
       "Georgetown                   0.16\n",
       "Bronx                        0.16\n",
       "Plantersville                0.16\n",
       "Shippingport                 0.16\n",
       "Matawan                      0.16\n",
       "Freedom                      0.16\n",
       "Tallmansville                0.16\n",
       "Rock Tavern                  0.16\n",
       "O Brien                      0.16\n",
       "Egan                         0.16\n",
       "Altair                       0.16\n",
       "West Hartford                0.16\n",
       "Sacramento                   0.16\n",
       "Grant                        0.16\n",
       "Nobleboro                    0.16\n",
       "Westerville                  0.16\n",
       "Southfield                   0.16\n",
       "Notrees                      0.16\n",
       "Fiddletown                   0.16\n",
       "Saint Bonaventure            0.16\n",
       "Bolton                       0.16\n",
       "Breesport                    0.16\n",
       "Marathon                     0.16\n",
       "Hurricane                    0.16\n",
       "Hooper                       0.16\n",
       "Mesa                         0.16\n",
       "Cascade Locks                0.16\n",
       "Kansas City                  0.16\n",
       "Graniteville                 0.16\n",
       "Belgrade                     0.16\n",
       "Logan                        0.16\n",
       "Wheaton                      0.16\n",
       "Nazareth                     0.16\n",
       "Ogdensburg                   0.16\n",
       "Tomales                      0.16\n",
       "Wilton                       0.16\n",
       "Clayton                      0.16\n",
       "Burlington                   0.16\n",
       "Fordoche                     0.16\n",
       "Meredith                     0.16\n",
       "Parsonsfield                 0.16\n",
       "Mound City                   0.16\n",
       "North Washington             0.16\n",
       "Coleharbor                   0.16\n",
       "Carlotta                     0.16\n",
       "Altona                       0.16\n",
       "Mayersville                  0.16\n",
       "Monmouth Beach               0.16\n",
       "Afton                        0.16\n",
       "Leo                          0.16\n",
       "Falmouth                     0.16\n",
       "Ratcliff                     0.16\n",
       "Brooklyn                     0.16\n",
       "Schaefferstown               0.16\n",
       "Cromona                      0.16\n",
       "Enola                        0.16\n",
       "Mountain Park                0.16\n",
       "North Loup                   0.16\n",
       "Saint Petersburg             0.16\n",
       "San Jose                     0.16\n",
       "Avoca                        0.16\n",
       "Lonsdale                     0.16\n",
       "Moravian Falls               0.16\n",
       "Mifflin                      0.16\n",
       "Roosevelt                    0.16\n",
       "Plymouth                     0.16\n",
       "Harper                       0.16\n",
       "Manderson                    0.16\n",
       "Greenwich                    0.16\n",
       "Shenandoah Junction          0.16\n",
       "Hewitt                       0.16\n",
       "Sheffield                    0.16\n",
       "May                          0.16\n",
       "Laramie                      0.16\n",
       "Clarinda                     0.16\n",
       "Prairie Hill                 0.16\n",
       "Doe Hill                     0.16\n",
       "Surrency                     0.16\n",
       "Mallie                       0.16\n",
       "Tyaskin                      0.16\n",
       "Cord                         0.16\n",
       "Arnold                       0.16\n",
       "Port Costa                   0.16\n",
       "Helm                         0.15\n",
       "Manville                     0.15\n",
       "Riverton                     0.15\n",
       "Phenix City                  0.15\n",
       "Garrattsville                0.15\n",
       "Kirby                        0.15\n",
       "Lakeport                     0.15\n",
       "Battle Creek                 0.15\n",
       "Delhi                        0.12\n",
       "Blackville                   0.12\n",
       "Beaver Falls                 0.12\n",
       "Sturgis                      0.12\n",
       "Mount Morris                 0.12\n",
       "Kissee Mills                 0.12\n",
       "Lake Oswego                  0.12\n",
       "Milwaukee                    0.12\n",
       "Clarksville                  0.12\n",
       "West Palm Beach              0.12\n",
       "Fairview                     0.12\n",
       "Deadwood                     0.12\n",
       "Wales                        0.12\n",
       "Luray                        0.12\n",
       "Oaks                         0.12\n",
       "Great Mills                  0.12\n",
       "Cherokee Village             0.12\n",
       "Grenada                      0.12\n",
       "Moulton                      0.12\n",
       "Harborcreek                  0.12\n",
       "Cass                         0.12\n",
       "Columbia                     0.12\n",
       "Bay Minette                  0.12\n",
       "Coleman                      0.12\n",
       "Mc Cracken                   0.12\n",
       "Dieterich                    0.12\n",
       "Greenview                    0.12\n",
       "Independence                 0.12\n",
       "Seneca                       0.12\n",
       "Eugene                       0.12\n",
       "East Andover                 0.12\n",
       "New Ellenton                 0.12\n",
       "Elizabethtown                0.12\n",
       "Zaleski                      0.12\n",
       "Oxford                       0.12\n",
       "West Columbia                0.12\n",
       "Lowell                       0.12\n",
       "Loxahatchee                  0.12\n",
       "Big Indian                   0.12\n",
       "Indian Wells                 0.12\n",
       "Juliette                     0.12\n",
       "Des Moines                   0.12\n",
       "Montandon                    0.12\n",
       "North Judson                 0.12\n",
       "Fullerton                    0.12\n",
       "Vero Beach                   0.12\n",
       "Viola                        0.12\n",
       "Quanah                       0.12\n",
       "Smackover                    0.12\n",
       "Athena                       0.12\n",
       "Jermyn                       0.12\n",
       "Paulding                     0.12\n",
       "Shelter Island               0.12\n",
       "Saint James City             0.12\n",
       "Cuyahoga Falls               0.12\n",
       "Greendale                    0.12\n",
       "Washington Court House       0.12\n",
       "Oak Hill                     0.12\n",
       "Rule                         0.12\n",
       "Stirling                     0.12\n",
       "Irwinton                     0.12\n",
       "Fields Landing               0.12\n",
       "Lima                         0.12\n",
       "Ballwin                      0.12\n",
       "Colorado Springs             0.12\n",
       "Ruth                         0.12\n",
       "Bowersville                  0.12\n",
       "Tomahawk                     0.12\n",
       "Manchester                   0.12\n",
       "Cazenovia                    0.12\n",
       "New Waverly                  0.12\n",
       "Ozawkie                      0.12\n",
       "Howells                      0.12\n",
       "Laguna Hills                 0.12\n",
       "Browning                     0.12\n",
       "Clay Center                  0.12\n",
       "University                   0.12\n",
       "Grand Junction               0.12\n",
       "Margaretville                0.12\n",
       "Randolph                     0.12\n",
       "Belmond                      0.12\n",
       "Mansfield                    0.12\n",
       "Irvine                       0.12\n",
       "Cedar                        0.12\n",
       "Comfort                      0.12\n",
       "Eldridge                     0.12\n",
       "Newark Valley                0.12\n",
       "Ruidoso                      0.12\n",
       "Moro                         0.12\n",
       "Santa Monica                 0.12\n",
       "Hawthorne                    0.12\n",
       "Baroda                       0.12\n",
       "Stephensport                 0.12\n",
       "Henderson                    0.12\n",
       "Zavalla                      0.12\n",
       "Manley                       0.12\n",
       "Palermo                      0.12\n",
       "Port Gibson                  0.12\n",
       "Arlington                    0.12\n",
       "Jefferson                    0.12\n",
       "Lamberton                    0.12\n",
       "Mount Perry                  0.12\n",
       "Turner                       0.12\n",
       "Kilgore                      0.12\n",
       "Darien                       0.12\n",
       "Chester Heights              0.12\n",
       "Hahira                       0.12\n",
       "North Augusta                0.12\n",
       "Daly City                    0.12\n",
       "Laredo                       0.12\n",
       "Paint Rock                   0.12\n",
       "Hopkins                      0.12\n",
       "Bonfield                     0.12\n",
       "Amanda                       0.12\n",
       "Tower Hill                   0.12\n",
       "Cecilton                     0.12\n",
       "Kenner                       0.12\n",
       "Center Point                 0.12\n",
       "Lowville                     0.12\n",
       "Port Ewen                    0.12\n",
       "June Lake                    0.12\n",
       "Princeton                    0.12\n",
       "Jelm                         0.12\n",
       "Paauilo                      0.12\n",
       "Scarborough                  0.12\n",
       "Titusville                   0.12\n",
       "Highland                     0.12\n",
       "Gardiner                     0.12\n",
       "Joliet                       0.12\n",
       "Purmela                      0.12\n",
       "Karnack                      0.12\n",
       "Holloway                     0.12\n",
       "West Harrison                0.12\n",
       "Whigham                      0.12\n",
       "Hawley                       0.12\n",
       "Marienville                  0.12\n",
       "Ollie                        0.12\n",
       "Mounds                       0.12\n",
       "Keller                       0.12\n",
       "Valdosta                     0.12\n",
       "Cadiz                        0.12\n",
       "Gretna                       0.12\n",
       "Munith                       0.12\n",
       "Mill Creek                   0.12\n",
       "River                        0.12\n",
       "Coyle                        0.12\n",
       "Norman Park                  0.12\n",
       "Minneapolis                  0.12\n",
       "Matthews                     0.12\n",
       "Eureka                       0.12\n",
       "Orangeburg                   0.12\n",
       "Cranks                       0.12\n",
       "Harrington Park              0.12\n",
       "Mooresville                  0.12\n",
       "Big Creek                    0.12\n",
       "Corsica                      0.12\n",
       "Bridgeport                   0.12\n",
       "Powell                       0.12\n",
       "Collettsville                0.12\n",
       "Hatch                        0.12\n",
       "Ash Flat                     0.12\n",
       "Blairstown                   0.12\n",
       "Bristol                      0.12\n",
       "Sula                         0.12\n",
       "Lolita                       0.12\n",
       "Timberville                  0.12\n",
       "Campbell                     0.12\n",
       "Bailey                       0.12\n",
       "Dumont                       0.12\n",
       "Cowlesville                  0.12\n",
       "Ford                         0.12\n",
       "Payson                       0.12\n",
       "Elkhart                      0.12\n",
       "Waupaca                      0.12\n",
       "Varnell                      0.12\n",
       "Belfast                      0.12\n",
       "Creola                       0.12\n",
       "Moab                         0.12\n",
       "Farmington                   0.12\n",
       "Spring Church                0.12\n",
       "Rhame                        0.11\n",
       "Burke                        0.11\n",
       "West Henrietta               0.11\n",
       "Deane                        0.11\n",
       "Hampton                      0.08\n",
       "Christine                    0.08\n",
       "Greenville                   0.08\n",
       "Dubre                        0.08\n",
       "North Tonawanda              0.08\n",
       "Texarkana                    0.08\n",
       "Albany                       0.08\n",
       "Oran                         0.08\n",
       "Winter                       0.08\n",
       "Lonetree                     0.08\n",
       "Ridgeland                    0.08\n",
       "Bristow                      0.08\n",
       "Parks                        0.08\n",
       "Alva                         0.08\n",
       "Maysville                    0.08\n",
       "Pomona                       0.08\n",
       "Pelham                       0.08\n",
       "Smock                        0.08\n",
       "Alexandria                   0.08\n",
       "Beasley                      0.08\n",
       "Pueblo                       0.08\n",
       "Colton                       0.08\n",
       "Halma                        0.08\n",
       "Boonton                      0.08\n",
       "Amsterdam                    0.08\n",
       "Heislerville                 0.08\n",
       "Clifton                      0.08\n",
       "Brinson                      0.08\n",
       "Stayton                      0.08\n",
       "Hurley                       0.08\n",
       "Kent                         0.08\n",
       "American Fork                0.08\n",
       "Kirk                         0.08\n",
       "Umatilla                     0.08\n",
       "Gibsonville                  0.08\n",
       "Edinburg                     0.08\n",
       "Sixes                        0.08\n",
       "Carroll                      0.08\n",
       "Carlisle                     0.08\n",
       "Sauk Rapids                  0.08\n",
       "Iliff                        0.08\n",
       "Pewee Valley                 0.08\n",
       "Orr                          0.08\n",
       "Ashford                      0.08\n",
       "Winslow                      0.08\n",
       "Queenstown                   0.08\n",
       "Duncan                       0.08\n",
       "Mount Clemens                0.08\n",
       "Downsville                   0.08\n",
       "Pikesville                   0.08\n",
       "West Green                   0.08\n",
       "East Troy                    0.08\n",
       "Slayden                      0.08\n",
       "Belle Fourche                0.08\n",
       "Port Richey                  0.08\n",
       "Weeping Water                0.08\n",
       "Spearsville                  0.08\n",
       "Grantham                     0.08\n",
       "Jordanville                  0.08\n",
       "Auburn                       0.08\n",
       "Pembroke                     0.08\n",
       "Easton                       0.08\n",
       "Paradise Valley              0.08\n",
       "Eagarville                   0.08\n",
       "Aledo                        0.08\n",
       "Montrose                     0.08\n",
       "Olmsted                      0.08\n",
       "Higganum                     0.08\n",
       "Milner                       0.08\n",
       "Westhampton Beach            0.08\n",
       "Rumely                       0.08\n",
       "Grand Bay                    0.08\n",
       "Honokaa                      0.08\n",
       "Marion                       0.08\n",
       "Smith River                  0.08\n",
       "Ashfield                     0.08\n",
       "Oklahoma City                0.08\n",
       "Halstad                      0.08\n",
       "Bethel Springs               0.08\n",
       "Stillwater                   0.08\n",
       "Saint Louis                  0.08\n",
       "Phil Campbell                0.08\n",
       "Scotts Mills                 0.08\n",
       "Central                      0.08\n",
       "Lakeview                     0.08\n",
       "Grand Ridge                  0.08\n",
       "Lubbock                      0.08\n",
       "Vinton                       0.08\n",
       "Cape Coral                   0.08\n",
       "Cuthbert                     0.08\n",
       "Glen Rock                    0.08\n",
       "Metairie                     0.08\n",
       "Unionville                   0.08\n",
       "Port Charlotte               0.08\n",
       "Corona                       0.08\n",
       "Tampa                        0.08\n",
       "West Decatur                 0.08\n",
       "Beaverdam                    0.08\n",
       "Nelson                       0.08\n",
       "Avera                        0.08\n",
       "Cressona                     0.08\n",
       "Deltona                      0.08\n",
       "Cross Plains                 0.08\n",
       "Center Tuftonboro            0.08\n",
       "Melbourne                    0.08\n",
       "Albuquerque                  0.08\n",
       "Barnard                      0.08\n",
       "Shrewsbury                   0.08\n",
       "Minnesota Lake               0.08\n",
       "Llano                        0.08\n",
       "Goreville                    0.08\n",
       "Bolivar                      0.08\n",
       "Portland                     0.08\n",
       "Hedley                       0.08\n",
       "Clune                        0.08\n",
       "Blairsden-Graeagle           0.08\n",
       "Watertown                    0.08\n",
       "Dongola                      0.08\n",
       "South Hero                   0.08\n",
       "Bay City                     0.08\n",
       "Union                        0.08\n",
       "Falls Church                 0.08\n",
       "West Eaton                   0.08\n",
       "Heidelberg                   0.08\n",
       "Ringwood                     0.08\n",
       "Brantley                     0.08\n",
       "Greenwood                    0.08\n",
       "Alder                        0.08\n",
       "Sterling City                0.08\n",
       "Armonk                       0.08\n",
       "West Monroe                  0.08\n",
       "Chester                      0.08\n",
       "Roseland                     0.08\n",
       "Emporium                     0.08\n",
       "Canton                       0.08\n",
       "Newberg                      0.08\n",
       "Mountain Center              0.08\n",
       "Brooklin                     0.08\n",
       "Tamaroa                      0.08\n",
       "Churubusco                   0.08\n",
       "Gregory                      0.08\n",
       "Montgomery                   0.08\n",
       "Schaumburg                   0.08\n",
       "Allenhurst                   0.08\n",
       "Kirtland Afb                 0.08\n",
       "Veedersburg                  0.08\n",
       "Haynes                       0.08\n",
       "Oriskany Falls               0.08\n",
       "Atglen                       0.08\n",
       "Tickfaw                      0.08\n",
       "Lebanon                      0.08\n",
       "Manor                        0.08\n",
       "Parkers Lake                 0.08\n",
       "South Londonderry            0.08\n",
       "Haines City                  0.08\n",
       "Harwood                      0.08\n",
       "Tiptonville                  0.08\n",
       "Glade Spring                 0.08\n",
       "Prosperity                   0.08\n",
       "Wauchula                     0.08\n",
       "Vanderbilt                   0.08\n",
       "Humble                       0.08\n",
       "Crownpoint                   0.08\n",
       "Norwich                      0.08\n",
       "Leetsdale                    0.08\n",
       "Miamisburg                   0.08\n",
       "Rockwood                     0.08\n",
       "Hedrick                      0.08\n",
       "Grimesland                   0.08\n",
       "Mendon                       0.08\n",
       "Belmont                      0.08\n",
       "Luzerne                      0.08\n",
       "Waukesha                     0.08\n",
       "Reynolds                     0.08\n",
       "Kingsport                    0.08\n",
       "Lagrange                     0.08\n",
       "Alton                        0.08\n",
       "Tryon                        0.08\n",
       "Hovland                      0.08\n",
       "Falls City                   0.08\n",
       "Burrton                      0.07\n",
       "Clearwater                   0.04\n",
       "Dublin                       0.04\n",
       "Providence                   0.04\n",
       "Huslia                       0.04\n",
       "Sprague                      0.04\n",
       "Curlew                       0.04\n",
       "El Paso                      0.04\n",
       "Vienna                       0.04\n",
       "Pecos                        0.04\n",
       "Old Hickory                  0.04\n",
       "Winthrop                     0.04\n",
       "Linthicum Heights            0.04\n",
       "East Rochester               0.04\n",
       "Gainesville                  0.04\n",
       "Smithfield                   0.04\n",
       "Maria Stein                  0.04\n",
       "Jaffrey                      0.04\n",
       "New Memphis                  0.04\n",
       "Mc Intosh                    0.04\n",
       "North Haverhill              0.04\n",
       "Achille                      0.04\n",
       "Prairie Creek                0.04\n",
       "Tuscarora                    0.04\n",
       "Brainard                     0.04\n",
       "Benton                       0.04\n",
       "Paris                        0.04\n",
       "Dadeville                    0.04\n",
       "Hazel                        0.04\n",
       "Daniels                      0.04\n",
       "Claremont                    0.04\n",
       "Elk Rapids                   0.04\n",
       "White Sulphur Springs        0.04\n",
       "Pearlington                  0.04\n",
       "Burns Flat                   0.04\n",
       "Newton                       0.04\n",
       "Fairhope                     0.04\n",
       "Livonia                      0.04\n",
       "Annapolis                    0.04\n",
       "Saint Francis                0.04\n",
       "Spring                       0.04\n",
       "Yellowstone National Park    0.04\n",
       "Galatia                      0.04\n",
       "Clutier                      0.04\n",
       "Dalton                       0.04\n",
       "Williamsburg                 0.04\n",
       "Monitor                      0.04\n",
       "Oakford                      0.04\n",
       "North Las Vegas              0.04\n",
       "Skytop                       0.04\n",
       "Powell Butte                 0.04\n",
       "Odessa                       0.04\n",
       "Knowlesville                 0.04\n",
       "Preston                      0.04\n",
       "Ironton                      0.04\n",
       "Remer                        0.04\n",
       "Damascus                     0.04\n",
       "Emmons                       0.04\n",
       "Sardis                       0.04\n",
       "Bigelow                      0.04\n",
       "Red Cliff                    0.04\n",
       "Los Angeles                  0.04\n",
       "Du Pont                      0.04\n",
       "Jordan Valley                0.04\n",
       "Cassatt                      0.04\n",
       "De Queen                     0.04\n",
       "New Goshen                   0.04\n",
       "Cross                        0.04\n",
       "Ravenna                      0.04\n",
       "Kittery Point                0.04\n",
       "Stoneham                     0.04\n",
       "Paxton                       0.04\n",
       "Newport                      0.04\n",
       "Apison                       0.04\n",
       "Washoe Valley                0.04\n",
       "Boyd                         0.04\n",
       "Springfield Gardens          0.04\n",
       "West Long Branch             0.04\n",
       "Ehrhardt                     0.04\n",
       "Waynesfield                  0.04\n",
       "Bonita Springs               0.04\n",
       "Westfir                      0.04\n",
       "West Chazy                   0.04\n",
       "Ferney                       0.04\n",
       "Dayton                       0.04\n",
       "Omaha                        0.04\n",
       "Greenbush                    0.04\n",
       "Republic                     0.04\n",
       "Hills                        0.04\n",
       "Brashear                     0.04\n",
       "Richland                     0.04\n",
       "Williams                     0.04\n",
       "Moscow                       0.04\n",
       "Adams                        0.04\n",
       "Arvada                       0.04\n",
       "North Brookfield             0.04\n",
       "Dexter                       0.04\n",
       "Harmony                      0.04\n",
       "Girard                       0.04\n",
       "Wendel                       0.04\n",
       "Woods Cross                  0.04\n",
       "Etlan                        0.04\n",
       "Rockwell                     0.04\n",
       "Iselin                       0.04\n",
       "Gaithersburg                 0.04\n",
       "Port Saint Lucie             0.04\n",
       "Kingsville                   0.04\n",
       "Sunflower                    0.04\n",
       "Liberty Mills                0.04\n",
       "Putnam                       0.04\n",
       "Aurora                       0.04\n",
       "Akron                        0.04\n",
       "Wittenberg                   0.04\n",
       "Cardwell                     0.04\n",
       "West Bethel                  0.04\n",
       "Manquin                      0.04\n",
       "Amorita                      0.04\n",
       "Trenton                      0.04\n",
       "Huntington Beach             0.04\n",
       "Rosewood                     0.04\n",
       "Marietta                     0.04\n",
       "Mc Clellandtown              0.04\n",
       "Parker                       0.04\n",
       "Oakdale                      0.04\n",
       "Bynum                        0.04\n",
       "Premier                      0.04\n",
       "Syracuse                     0.04\n",
       "Valentine                    0.04\n",
       "Louisiana                    0.04\n",
       "Bethel                       0.04\n",
       "Parker Dam                   0.04\n",
       "Lepanto                      0.04\n",
       "Hartford                     0.04\n",
       "Palmdale                     0.04\n",
       "San Angelo                   0.04\n",
       "Spirit Lake                  0.04\n",
       "Lorenzo                      0.04\n",
       "Broomfield                   0.04\n",
       "Hopewell                     0.04\n",
       "Rock Glen                    0.04\n",
       "Monetta                      0.04\n",
       "Collegeville                 0.04\n",
       "Riverview                    0.04\n",
       "Dresden                      0.04\n",
       "Howes Cave                   0.04\n",
       "Whittemore                   0.04\n",
       "Nokomis                      0.04\n",
       "Moores Hill                  0.04\n",
       "Bryant                       0.04\n",
       "Malad City                   0.04\n",
       "Desdemona                    0.04\n",
       "Ragland                      0.04\n",
       "Shedd                        0.04\n",
       "Lamy                         0.04\n",
       "Moriches                     0.04\n",
       "Owensville                   0.04\n",
       "Jay                          0.04\n",
       "Azusa                        0.04\n",
       "Chatham                      0.04\n",
       "Vancouver                    0.04\n",
       "Lane                         0.04\n",
       "South Richmond Hill          0.04\n",
       "Northport                    0.04\n",
       "Garfield                     0.04\n",
       "Loami                        0.04\n",
       "Rossville                    0.04\n",
       "Jackson                      0.04\n",
       "Pea Ridge                    0.04\n",
       "Palmyra                      0.04\n",
       "Webster City                 0.04\n",
       "Michigan                     0.04\n",
       "Boulder                      0.04\n",
       "Kings Bay                    0.04\n",
       "Sea Island                   0.04\n",
       "Thornville                   0.04\n",
       "Loving                       0.04\n",
       "Rochester                    0.04\n",
       "New Franken                  0.04\n",
       "Dell City                    0.04\n",
       "Early                        0.04\n",
       "Lithopolis                   0.04\n",
       "De Lancey                    0.04\n",
       "Humboldt                     0.04\n",
       "Stittville                   0.04\n",
       "West Finley                  0.04\n",
       "Hubbell                      0.00\n",
       "Melville                     0.00\n",
       "Coulee Dam                   0.00\n",
       "Queen Anne                   0.00\n",
       "West Frankfort               0.00\n",
       "Mineral                      0.00\n",
       "Byesville                    0.00\n",
       "Granbury                     0.00\n",
       "Kaktovik                     0.00\n",
       "Clinton                      0.00\n",
       "La Grande                    0.00\n",
       "Phelps                       0.00\n",
       "Mount Vernon                 0.00\n",
       "Beacon                       0.00\n",
       "Walkertown                   0.00\n",
       "Winnsboro                    0.00\n",
       "Roland                       0.00\n",
       "Wartburg                     0.00\n",
       "Madisonville                 0.00\n",
       "Angwin                       0.00\n",
       "Marshall                     0.00\n",
       "Isanti                       0.00\n",
       "Seattle                      0.00\n",
       "Nanuet                       0.00\n",
       "Orange Park                  0.00\n",
       "Moss Point                   0.00\n",
       "Ridge Spring                 0.00\n",
       "Ashland                      0.00\n",
       "Waukau                       0.00\n",
       "Downey                       0.00\n",
       "Greenport                    0.00\n",
       "Mountain City                0.00\n",
       "Medford                      0.00\n",
       "Lockhart                     0.00\n",
       "Nicholson                    0.00\n",
       "Brookfield                   0.00\n",
       "East China                   0.00\n",
       "Grenola                      0.00\n",
       "Freeport                     0.00\n",
       "North East                   0.00\n",
       "Oakton                       0.00\n",
       "Clarion                      0.00\n",
       "Gaines                       0.00\n",
       "Morven                       0.00\n",
       "Noblesville                  0.00\n",
       "Las Vegas                    0.00\n",
       "Buellton                     0.00\n",
       "Crouse                       0.00\n",
       "Irvington                    0.00\n",
       "Pleasant Hill                0.00\n",
       "Wappapello                   0.00\n",
       "Norfolk                      0.00\n",
       "Streator                     0.00\n",
       "Vacaville                    0.00\n",
       "Claypool                     0.00\n",
       "Karns City                   0.00\n",
       "Bruce                        0.00\n",
       "Chattanooga                  0.00\n",
       "Name: city, dtype: float64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(df_train1['city'].value_counts()/len(df_train1.index)*100,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "c6ae4c18",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TX    7.32\n",
       "NY    6.44\n",
       "PA    6.16\n",
       "CA    4.35\n",
       "OH    3.58\n",
       "MI    3.56\n",
       "IL    3.34\n",
       "FL    3.29\n",
       "AL    3.16\n",
       "MO    2.96\n",
       "MN    2.45\n",
       "AR    2.40\n",
       "NC    2.33\n",
       "WI    2.26\n",
       "VA    2.26\n",
       "SC    2.25\n",
       "KY    2.20\n",
       "IN    2.13\n",
       "IA    2.08\n",
       "OK    2.06\n",
       "MD    2.02\n",
       "GA    2.01\n",
       "WV    1.98\n",
       "NJ    1.90\n",
       "NE    1.86\n",
       "KS    1.77\n",
       "MS    1.63\n",
       "LA    1.62\n",
       "WY    1.49\n",
       "WA    1.46\n",
       "OR    1.43\n",
       "TN    1.35\n",
       "ME    1.27\n",
       "NM    1.27\n",
       "ND    1.14\n",
       "CO    1.07\n",
       "MA    0.95\n",
       "SD    0.95\n",
       "VT    0.91\n",
       "MT    0.91\n",
       "AZ    0.83\n",
       "UT    0.83\n",
       "NH    0.64\n",
       "CT    0.59\n",
       "NV    0.43\n",
       "ID    0.43\n",
       "DC    0.28\n",
       "HI    0.20\n",
       "AK    0.16\n",
       "RI    0.04\n",
       "DE    0.00\n",
       "Name: state, dtype: float64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(df_train1['state'].value_counts()/len(df_train1.index)*100,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "3808a0bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Film/video editor                                              0.75\n",
       "Exhibition designer                                            0.71\n",
       "Naval architect                                                0.67\n",
       "Surveyor, land/geomatics                                       0.67\n",
       "Materials engineer                                             0.64\n",
       "Designer, ceramics/pottery                                     0.63\n",
       "Systems developer                                              0.59\n",
       "IT trainer                                                     0.59\n",
       "Financial adviser                                              0.59\n",
       "Environmental consultant                                       0.58\n",
       "Chartered public finance accountant                            0.56\n",
       "Scientist, audiological                                        0.55\n",
       "Chief Executive Officer                                        0.55\n",
       "Copywriter, advertising                                        0.55\n",
       "Comptroller                                                    0.52\n",
       "Sub                                                            0.52\n",
       "Podiatrist                                                     0.51\n",
       "Agricultural consultant                                        0.51\n",
       "Paramedic                                                      0.51\n",
       "Magazine features editor                                       0.51\n",
       "Historic buildings inspector/conservation officer              0.48\n",
       "Librarian, public                                              0.48\n",
       "Building surveyor                                              0.48\n",
       "Musician                                                       0.48\n",
       "Water engineer                                                 0.48\n",
       "Scientist, research (maths)                                    0.47\n",
       "Mining engineer                                                0.47\n",
       "Barrister                                                      0.47\n",
       "Audiological scientist                                         0.47\n",
       "Immunologist                                                   0.47\n",
       "Clothing/textile technologist                                  0.47\n",
       "Prison officer                                                 0.44\n",
       "Financial trader                                               0.44\n",
       "Claims inspector/assessor                                      0.44\n",
       "Energy engineer                                                0.44\n",
       "Petroleum engineer                                             0.44\n",
       "Quantity surveyor                                              0.44\n",
       "Make                                                           0.44\n",
       "Engineer, automotive                                           0.43\n",
       "Mechanical engineer                                            0.43\n",
       "Geoscientist                                                   0.43\n",
       "Counsellor                                                     0.43\n",
       "Secondary school teacher                                       0.43\n",
       "Land/geomatics surveyor                                        0.43\n",
       "Firefighter                                                    0.43\n",
       "Early years teacher                                            0.43\n",
       "Sales professional, IT                                         0.43\n",
       "Pensions consultant                                            0.42\n",
       "Psychologist, forensic                                         0.42\n",
       "Psychotherapist, child                                         0.42\n",
       "Librarian, academic                                            0.40\n",
       "Research scientist (physical sciences)                         0.40\n",
       "Electrical engineer                                            0.40\n",
       "Insurance underwriter                                          0.40\n",
       "Science writer                                                 0.40\n",
       "Medical sales representative                                   0.40\n",
       "Chartered loss adjuster                                        0.40\n",
       "Drilling engineer                                              0.40\n",
       "Engineer, biomedical                                           0.39\n",
       "Trade mark attorney                                            0.39\n",
       "Television production assistant                                0.39\n",
       "Special educational needs teacher                              0.39\n",
       "Scientist, marine                                              0.39\n",
       "Cytogeneticist                                                 0.39\n",
       "Chemical engineer                                              0.39\n",
       "Multimedia programmer                                          0.36\n",
       "Accounting technician                                          0.36\n",
       "Dealer                                                         0.36\n",
       "Counselling psychologist                                       0.36\n",
       "Technical brewer                                               0.36\n",
       "Trading standards officer                                      0.36\n",
       "Toxicologist                                                   0.36\n",
       "Engineer, production                                           0.36\n",
       "Social researcher                                              0.36\n",
       "Radio producer                                                 0.36\n",
       "Chief Strategy Officer                                         0.35\n",
       "Physiotherapist                                                0.35\n",
       "Fine artist                                                    0.35\n",
       "Teacher, special educational needs                             0.35\n",
       "Race relations officer                                         0.35\n",
       "Engineer, control and instrumentation                          0.35\n",
       "Radio broadcast assistant                                      0.35\n",
       "Therapist, occupational                                        0.35\n",
       "Surveyor, minerals                                             0.35\n",
       "Educational psychologist                                       0.35\n",
       "Ceramics designer                                              0.35\n",
       "Advertising account planner                                    0.32\n",
       "Town planner                                                   0.32\n",
       "Programme researcher, broadcasting/film/video                  0.32\n",
       "Exhibitions officer, museum/gallery                            0.32\n",
       "Surgeon                                                        0.32\n",
       "Futures trader                                                 0.32\n",
       "Television/film/video producer                                 0.32\n",
       "Further education lecturer                                     0.32\n",
       "Scientist, biomedical                                          0.32\n",
       "Designer, industrial/product                                   0.32\n",
       "English as a second language teacher                           0.32\n",
       "Applications developer                                         0.32\n",
       "Probation officer                                              0.32\n",
       "Senior tax professional/tax inspector                          0.32\n",
       "Buyer, industrial                                              0.32\n",
       "Curator                                                        0.32\n",
       "Child psychotherapist                                          0.32\n",
       "Lexicographer                                                  0.32\n",
       "Engineering geologist                                          0.32\n",
       "Occupational psychologist                                      0.31\n",
       "Herbalist                                                      0.31\n",
       "Development worker, community                                  0.31\n",
       "Pilot, airline                                                 0.31\n",
       "Radiographer, therapeutic                                      0.31\n",
       "Press photographer                                             0.31\n",
       "Archaeologist                                                  0.31\n",
       "Public relations account executive                             0.31\n",
       "Facilities manager                                             0.31\n",
       "Editor, commissioning                                          0.28\n",
       "Forensic psychologist                                          0.28\n",
       "Fisheries officer                                              0.28\n",
       "Research officer, political party                              0.28\n",
       "Engineer, electronics                                          0.28\n",
       "Dispensing optician                                            0.28\n",
       "Administrator                                                  0.28\n",
       "Broadcast presenter                                            0.28\n",
       "Insurance risk surveyor                                        0.28\n",
       "Exercise physiologist                                          0.28\n",
       "Engineer, mining                                               0.28\n",
       "Systems analyst                                                0.28\n",
       "Retail merchandiser                                            0.28\n",
       "Chiropodist                                                    0.28\n",
       "Operations geologist                                           0.28\n",
       "Lecturer, further education                                    0.28\n",
       "Call centre manager                                            0.28\n",
       "Physicist, medical                                             0.28\n",
       "Insurance broker                                               0.28\n",
       "Learning disability nurse                                      0.28\n",
       "Hospital doctor                                                0.28\n",
       "Heritage manager                                               0.28\n",
       "Research officer, trade union                                  0.27\n",
       "Private music teacher                                          0.27\n",
       "Structural engineer                                            0.27\n",
       "Travel agency manager                                          0.27\n",
       "Amenity horticulturist                                         0.27\n",
       "Psychologist, counselling                                      0.27\n",
       "Colour technologist                                            0.27\n",
       "Tree surgeon                                                   0.24\n",
       "Television floor manager                                       0.24\n",
       "Metallurgist                                                   0.24\n",
       "Geologist, engineering                                         0.24\n",
       "Producer, radio                                                0.24\n",
       "Arboriculturist                                                0.24\n",
       "Phytotherapist                                                 0.24\n",
       "Education officer, museum                                      0.24\n",
       "Journalist, newspaper                                          0.24\n",
       "Occupational hygienist                                         0.24\n",
       "Site engineer                                                  0.24\n",
       "Electronics engineer                                           0.24\n",
       "Tourist information centre manager                             0.24\n",
       "Regulatory affairs officer                                     0.24\n",
       "Video editor                                                   0.24\n",
       "Designer, jewellery                                            0.24\n",
       "Biomedical scientist                                           0.24\n",
       "Leisure centre manager                                         0.24\n",
       "Licensed conveyancer                                           0.24\n",
       "Press sub                                                      0.24\n",
       "Production manager                                             0.24\n",
       "Designer, furniture                                            0.24\n",
       "Manufacturing engineer                                         0.24\n",
       "Development worker, international aid                          0.24\n",
       "Barrister's clerk                                              0.24\n",
       "Camera operator                                                0.24\n",
       "Commercial horticulturist                                      0.24\n",
       "Warden/ranger                                                  0.24\n",
       "Tax inspector                                                  0.23\n",
       "Chief Operating Officer                                        0.23\n",
       "Radiographer, diagnostic                                       0.23\n",
       "Music tutor                                                    0.23\n",
       "Health physicist                                               0.23\n",
       "Product/process development scientist                          0.23\n",
       "Designer, exhibition/display                                   0.23\n",
       "Engineer, land                                                 0.23\n",
       "Contractor                                                     0.23\n",
       "Gaffer                                                         0.23\n",
       "Freight forwarder                                              0.23\n",
       "Lecturer, higher education                                     0.23\n",
       "Commercial/residential surveyor                                0.23\n",
       "Careers information officer                                    0.23\n",
       "Energy manager                                                 0.23\n",
       "Administrator, local government                                0.20\n",
       "Osteopath                                                      0.20\n",
       "Wellsite geologist                                             0.20\n",
       "Statistician                                                   0.20\n",
       "Market researcher                                              0.20\n",
       "Risk analyst                                                   0.20\n",
       "Nurse, children's                                              0.20\n",
       "Doctor, general practice                                       0.20\n",
       "Accountant, chartered public finance                           0.20\n",
       "Chief Technology Officer                                       0.20\n",
       "Health and safety adviser                                      0.20\n",
       "Advice worker                                                  0.20\n",
       "Programmer, multimedia                                         0.20\n",
       "Soil scientist                                                 0.20\n",
       "Tourism officer                                                0.20\n",
       "Equality and diversity officer                                 0.20\n",
       "Administrator, education                                       0.20\n",
       "Human resources officer                                        0.20\n",
       "Community arts worker                                          0.20\n",
       "Operational researcher                                         0.20\n",
       "Therapist, sports                                              0.20\n",
       "Police officer                                                 0.20\n",
       "Music therapist                                                0.20\n",
       "Building control surveyor                                      0.20\n",
       "Chief Financial Officer                                        0.20\n",
       "Media buyer                                                    0.20\n",
       "Bookseller                                                     0.20\n",
       "Database administrator                                         0.20\n",
       "Optician, dispensing                                           0.20\n",
       "Fitness centre manager                                         0.20\n",
       "Health service manager                                         0.20\n",
       "Land                                                           0.20\n",
       "Mental health nurse                                            0.20\n",
       "Web designer                                                   0.20\n",
       "Maintenance engineer                                           0.20\n",
       "Therapist, art                                                 0.20\n",
       "Doctor, hospital                                               0.20\n",
       "Teacher, secondary school                                      0.20\n",
       "Engineer, petroleum                                            0.20\n",
       "Product designer                                               0.20\n",
       "Paediatric nurse                                               0.20\n",
       "Armed forces training and education officer                    0.20\n",
       "Waste management officer                                       0.20\n",
       "Tax adviser                                                    0.20\n",
       "Geochemist                                                     0.20\n",
       "Investment analyst                                             0.20\n",
       "Patent attorney                                                0.20\n",
       "Therapist, horticultural                                       0.19\n",
       "Surveyor, rural practice                                       0.19\n",
       "Administrator, charities/voluntary organisations               0.19\n",
       "Intelligence analyst                                           0.19\n",
       "Teacher, early years/pre                                       0.19\n",
       "Sports development officer                                     0.16\n",
       "Neurosurgeon                                                   0.16\n",
       "Transport planner                                              0.16\n",
       "Horticultural therapist                                        0.16\n",
       "Chief Marketing Officer                                        0.16\n",
       "Airline pilot                                                  0.16\n",
       "Retail buyer                                                   0.16\n",
       "Charity fundraiser                                             0.16\n",
       "Pension scheme manager                                         0.16\n",
       "Scientist, physiological                                       0.16\n",
       "Psychologist, sport and exercise                               0.16\n",
       "Landscape architect                                            0.16\n",
       "Telecommunications researcher                                  0.16\n",
       "Industrial/product designer                                    0.16\n",
       "Animal nutritionist                                            0.16\n",
       "Seismic interpreter                                            0.16\n",
       "Psychiatrist                                                   0.16\n",
       "Geologist, wellsite                                            0.16\n",
       "Museum/gallery conservator                                     0.16\n",
       "Glass blower/designer                                          0.16\n",
       "Politician's assistant                                         0.16\n",
       "Clinical research associate                                    0.16\n",
       "Sport and exercise psychologist                                0.16\n",
       "Editor, film/video                                             0.16\n",
       "Clinical biochemist                                            0.16\n",
       "Nutritional therapist                                          0.16\n",
       "Museum/gallery exhibitions officer                             0.16\n",
       "Therapist, drama                                               0.16\n",
       "Art therapist                                                  0.16\n",
       "Secretary/administrator                                        0.16\n",
       "Immigration officer                                            0.16\n",
       "Architect                                                      0.16\n",
       "Psychotherapist                                                0.16\n",
       "Retail banker                                                  0.16\n",
       "Communications engineer                                        0.16\n",
       "Animal technologist                                            0.16\n",
       "Engineer, drilling                                             0.16\n",
       "Scientist, research (physical sciences)                        0.16\n",
       "Furniture designer                                             0.16\n",
       "Editor, magazine features                                      0.16\n",
       "Community education officer                                    0.16\n",
       "Public house manager                                           0.16\n",
       "Cartographer                                                   0.16\n",
       "Health promotion specialist                                    0.16\n",
       "Horticulturist, commercial                                     0.16\n",
       "Biomedical engineer                                            0.16\n",
       "Farm manager                                                   0.16\n",
       "Scientific laboratory technician                               0.16\n",
       "Research scientist (life sciences)                             0.16\n",
       "Investment banker, corporate                                   0.16\n",
       "Herpetologist                                                  0.16\n",
       "Public affairs consultant                                      0.16\n",
       "Volunteer coordinator                                          0.16\n",
       "Pharmacist, community                                          0.16\n",
       "Arts development officer                                       0.16\n",
       "Medical secretary                                              0.16\n",
       "Corporate investment banker                                    0.16\n",
       "Information systems manager                                    0.16\n",
       "Civil engineer, contracting                                    0.16\n",
       "Dance movement psychotherapist                                 0.16\n",
       "Engineer, technical sales                                      0.16\n",
       "Aid worker                                                     0.16\n",
       "General practice doctor                                        0.15\n",
       "Theme park manager                                             0.15\n",
       "Health visitor                                                 0.15\n",
       "Animator                                                       0.12\n",
       "Medical physicist                                              0.12\n",
       "Acupuncturist                                                  0.12\n",
       "Engineer, agricultural                                         0.12\n",
       "Advertising account executive                                  0.12\n",
       "Planning and development surveyor                              0.12\n",
       "Embryologist, clinical                                         0.12\n",
       "Barista                                                        0.12\n",
       "Oceanographer                                                  0.12\n",
       "Pharmacologist                                                 0.12\n",
       "Event organiser                                                0.12\n",
       "Special effects artist                                         0.12\n",
       "Designer, interior/spatial                                     0.12\n",
       "Psychologist, clinical                                         0.12\n",
       "Learning mentor                                                0.12\n",
       "Theatre manager                                                0.12\n",
       "Orthoptist                                                     0.12\n",
       "Sales executive                                                0.12\n",
       "Surveyor, mining                                               0.12\n",
       "Network engineer                                               0.12\n",
       "Theatre director                                               0.12\n",
       "Psychiatric nurse                                              0.12\n",
       "Training and development officer                               0.12\n",
       "Community pharmacist                                           0.12\n",
       "Logistics and distribution manager                             0.12\n",
       "Equities trader                                                0.12\n",
       "Field seismologist                                             0.12\n",
       "Engineer, communications                                       0.12\n",
       "Set designer                                                   0.12\n",
       "Interpreter                                                    0.12\n",
       "Research scientist (maths)                                     0.12\n",
       "Therapist, music                                               0.12\n",
       "Environmental health practitioner                              0.12\n",
       "Emergency planning/management officer                          0.12\n",
       "Social research officer, government                            0.12\n",
       "Physiological scientist                                        0.12\n",
       "Building services engineer                                     0.12\n",
       "Optometrist                                                    0.12\n",
       "Diagnostic radiographer                                        0.12\n",
       "Production engineer                                            0.12\n",
       "Mudlogger                                                      0.12\n",
       "Garment/textile technologist                                   0.12\n",
       "Management consultant                                          0.12\n",
       "Armed forces logistics/support/administrative officer          0.12\n",
       "Higher education careers adviser                               0.12\n",
       "Public librarian                                               0.12\n",
       "Engineer, maintenance                                          0.12\n",
       "Aeronautical engineer                                          0.12\n",
       "Commissioning editor                                           0.12\n",
       "Hydrologist                                                    0.12\n",
       "Geophysicist/field seismologist                                0.12\n",
       "Archivist                                                      0.12\n",
       "Rural practice surveyor                                        0.12\n",
       "Quarry manager                                                 0.12\n",
       "Control and instrumentation engineer                           0.12\n",
       "Records manager                                                0.12\n",
       "Teaching laboratory technician                                 0.12\n",
       "Engineer, civil (contracting)                                  0.12\n",
       "Chartered accountant                                           0.12\n",
       "Location manager                                               0.12\n",
       "Illustrator                                                    0.11\n",
       "Teacher, primary school                                        0.11\n",
       "Conservation officer, historic buildings                       0.11\n",
       "Media planner                                                  0.08\n",
       "Medical technical officer                                      0.08\n",
       "Plant breeder/geneticist                                       0.08\n",
       "Pharmacist, hospital                                           0.08\n",
       "Ambulance person                                               0.08\n",
       "Sports administrator                                           0.08\n",
       "Scientist, clinical (histocompatibility and immunogenetics)    0.08\n",
       "Designer, multimedia                                           0.08\n",
       "Furniture conservator/restorer                                 0.08\n",
       "Engineer, broadcasting (operations)                            0.08\n",
       "Designer, textile                                              0.08\n",
       "Museum education officer                                       0.08\n",
       "Pathologist                                                    0.08\n",
       "Economist                                                      0.08\n",
       "Academic librarian                                             0.08\n",
       "Stage manager                                                  0.08\n",
       "Hydrographic surveyor                                          0.08\n",
       "Hospital pharmacist                                            0.08\n",
       "Presenter, broadcasting                                        0.08\n",
       "Company secretary                                              0.08\n",
       "Biochemist, clinical                                           0.08\n",
       "Environmental education officer                                0.08\n",
       "Air broker                                                     0.08\n",
       "Retail manager                                                 0.08\n",
       "Engineer, building services                                    0.08\n",
       "Production assistant, radio                                    0.08\n",
       "Advertising copywriter                                         0.08\n",
       "Cabin crew                                                     0.08\n",
       "Chief of Staff                                                 0.08\n",
       "Engineer, aeronautical                                         0.08\n",
       "Art gallery manager                                            0.08\n",
       "Education administrator                                        0.08\n",
       "Insurance claims handler                                       0.08\n",
       "Ecologist                                                      0.08\n",
       "Civil Service fast streamer                                    0.08\n",
       "Architectural technologist                                     0.08\n",
       "Chartered legal executive (England and Wales)                  0.08\n",
       "Outdoor activities/education manager                           0.08\n",
       "Nurse, mental health                                           0.08\n",
       "Loss adjuster, chartered                                       0.08\n",
       "Chemist, analytical                                            0.08\n",
       "Jewellery designer                                             0.08\n",
       "Research scientist (medical)                                   0.08\n",
       "Programmer, applications                                       0.08\n",
       "Teacher, English as a foreign language                         0.08\n",
       "Product manager                                                0.08\n",
       "Clinical psychologist                                          0.08\n",
       "IT consultant                                                  0.08\n",
       "Manufacturing systems engineer                                 0.08\n",
       "Scientist, research (medical)                                  0.08\n",
       "Visual merchandiser                                            0.08\n",
       "Geneticist, molecular                                          0.04\n",
       "Engineer, civil (consulting)                                   0.04\n",
       "Lawyer                                                         0.04\n",
       "Estate manager/land agent                                      0.04\n",
       "Merchandiser, retail                                           0.04\n",
       "Engineer, manufacturing                                        0.04\n",
       "Conservator, furniture                                         0.04\n",
       "Community development worker                                   0.04\n",
       "Administrator, arts                                            0.04\n",
       "Designer, television/film set                                  0.04\n",
       "Accountant, chartered certified                                0.04\n",
       "TEFL teacher                                                   0.04\n",
       "Magazine journalist                                            0.04\n",
       "Horticultural consultant                                       0.04\n",
       "Surveyor, hydrographic                                         0.04\n",
       "Production assistant, television                               0.04\n",
       "Minerals surveyor                                              0.04\n",
       "Associate Professor                                            0.04\n",
       "Oncologist                                                     0.04\n",
       "Restaurant manager, fast food                                  0.04\n",
       "Catering manager                                               0.04\n",
       "Local government officer                                       0.04\n",
       "Solicitor, Scotland                                            0.04\n",
       "Broadcast engineer                                             0.04\n",
       "Occupational therapist                                         0.04\n",
       "Textile designer                                               0.04\n",
       "Charity officer                                                0.04\n",
       "Artist                                                         0.04\n",
       "Field trials officer                                           0.04\n",
       "Television camera operator                                     0.04\n",
       "Air cabin crew                                                 0.04\n",
       "Engineer, materials                                            0.04\n",
       "Public relations officer                                       0.04\n",
       "Environmental manager                                          0.04\n",
       "Data scientist                                                 0.04\n",
       "Hotel manager                                                  0.04\n",
       "Conservator, museum/gallery                                    0.04\n",
       "Teacher, adult education                                       0.04\n",
       "Primary school teacher                                         0.04\n",
       "Education officer, community                                   0.04\n",
       "Nature conservation officer                                    0.04\n",
       "Professor Emeritus                                             0.04\n",
       "Investment banker, operational                                 0.04\n",
       "Water quality scientist                                        0.04\n",
       "Interior and spatial designer                                  0.04\n",
       "Clinical cytogeneticist                                        0.04\n",
       "Marketing executive                                            0.04\n",
       "Civil Service administrator                                    0.04\n",
       "Purchasing manager                                             0.04\n",
       "Copy                                                           0.04\n",
       "Writer                                                         0.04\n",
       "Analytical chemist                                             0.04\n",
       "Tour manager                                                   0.04\n",
       "Buyer, retail                                                  0.04\n",
       "English as a foreign language teacher                          0.04\n",
       "Producer, television/film/video                                0.04\n",
       "Engineer, structural                                           0.04\n",
       "Hydrogeologist                                                 0.04\n",
       "Dancer                                                         0.00\n",
       "Air traffic controller                                         0.00\n",
       "Careers adviser                                                0.00\n",
       "Sales promotion account executive                              0.00\n",
       "Legal secretary                                                0.00\n",
       "Personnel officer                                              0.00\n",
       "Engineer, site                                                 0.00\n",
       "Solicitor                                                      0.00\n",
       "Homeopath                                                      0.00\n",
       "Accountant, chartered                                          0.00\n",
       "Industrial buyer                                               0.00\n",
       "Broadcast journalist                                           0.00\n",
       "Forest/woodland manager                                        0.00\n",
       "Armed forces technical officer                                 0.00\n",
       "Information officer                                            0.00\n",
       "Veterinary surgeon                                             0.00\n",
       "Ship broker                                                    0.00\n",
       "Contracting civil engineer                                     0.00\n",
       "Warehouse manager                                              0.00\n",
       "Name: job, dtype: float64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round(df_train1['job'].value_counts()/len(df_train1.index)*100,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "1c91cd43",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 19 columns):\n",
      " #   Column                 Non-Null Count    Dtype         \n",
      "---  ------                 --------------    -----         \n",
      " 0   trans_date_trans_time  1296675 non-null  datetime64[ns]\n",
      " 1   category               1296675 non-null  object        \n",
      " 2   amt                    1296675 non-null  float64       \n",
      " 3   gender                 1296675 non-null  object        \n",
      " 4   city                   1296675 non-null  object        \n",
      " 5   state                  1296675 non-null  object        \n",
      " 6   lat                    1296675 non-null  float64       \n",
      " 7   long                   1296675 non-null  float64       \n",
      " 8   city_pop               1296675 non-null  int64         \n",
      " 9   job                    1296675 non-null  object        \n",
      " 10  dob                    1296675 non-null  datetime64[ns]\n",
      " 11  merch_lat              1296675 non-null  float64       \n",
      " 12  merch_long             1296675 non-null  float64       \n",
      " 13  is_fraud               1296675 non-null  int64         \n",
      " 14  trans_date             1296675 non-null  datetime64[ns]\n",
      " 15  age_at_trans           1296675 non-null  float64       \n",
      " 16  lat_dist               1296675 non-null  float64       \n",
      " 17  long_dist              1296675 non-null  float64       \n",
      " 18  trans_month            1296675 non-null  int64         \n",
      "dtypes: datetime64[ns](3), float64(8), int64(3), object(5)\n",
      "memory usage: 188.0+ MB\n"
     ]
    }
   ],
   "source": [
    "df_train1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "4387ec0b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "trans_date_trans_time    1274791\n",
       "category                      14\n",
       "amt                        52928\n",
       "gender                         2\n",
       "city                         894\n",
       "state                         51\n",
       "lat                          968\n",
       "long                         969\n",
       "city_pop                     879\n",
       "job                          494\n",
       "dob                          968\n",
       "merch_lat                1247805\n",
       "merch_long               1275745\n",
       "is_fraud                       2\n",
       "trans_date                   537\n",
       "age_at_trans                  83\n",
       "lat_dist                    1001\n",
       "long_dist                   1001\n",
       "trans_month                   12\n",
       "dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train1.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "f192cc93",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>category</th>\n",
       "      <th>amt</th>\n",
       "      <th>gender</th>\n",
       "      <th>state</th>\n",
       "      <th>city_pop</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>age_at_trans</th>\n",
       "      <th>lat_dist</th>\n",
       "      <th>long_dist</th>\n",
       "      <th>trans_month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>misc_net</td>\n",
       "      <td>4.97</td>\n",
       "      <td>F</td>\n",
       "      <td>NC</td>\n",
       "      <td>3495</td>\n",
       "      <td>0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.068</td>\n",
       "      <td>0.870</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>grocery_pos</td>\n",
       "      <td>107.23</td>\n",
       "      <td>F</td>\n",
       "      <td>WA</td>\n",
       "      <td>149</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0.271</td>\n",
       "      <td>0.024</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>entertainment</td>\n",
       "      <td>220.11</td>\n",
       "      <td>M</td>\n",
       "      <td>ID</td>\n",
       "      <td>4154</td>\n",
       "      <td>0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.970</td>\n",
       "      <td>0.108</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>gas_transport</td>\n",
       "      <td>45.00</td>\n",
       "      <td>M</td>\n",
       "      <td>MT</td>\n",
       "      <td>1939</td>\n",
       "      <td>0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>0.804</td>\n",
       "      <td>0.447</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>misc_pos</td>\n",
       "      <td>41.96</td>\n",
       "      <td>M</td>\n",
       "      <td>VA</td>\n",
       "      <td>99</td>\n",
       "      <td>0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0.254</td>\n",
       "      <td>0.830</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        category     amt gender state  city_pop  is_fraud  age_at_trans  \\\n",
       "0       misc_net    4.97      F    NC      3495         0          30.0   \n",
       "1    grocery_pos  107.23      F    WA       149         0          40.0   \n",
       "2  entertainment  220.11      M    ID      4154         0          56.0   \n",
       "3  gas_transport   45.00      M    MT      1939         0          51.0   \n",
       "4       misc_pos   41.96      M    VA        99         0          32.0   \n",
       "\n",
       "   lat_dist  long_dist  trans_month  \n",
       "0     0.068      0.870            1  \n",
       "1     0.271      0.024            1  \n",
       "2     0.970      0.108            1  \n",
       "3     0.804      0.447            1  \n",
       "4     0.254      0.830            1  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dropping variables not useful for model building\n",
    "\n",
    "drop_cols = ['trans_date_trans_time','city','lat','long','job','dob','merch_lat','merch_long','trans_date']\n",
    "\n",
    "df_train2 = df_train1.drop(drop_cols,axis=1)\n",
    "df_test2 = df_test1.drop(drop_cols,axis=1)\n",
    "\n",
    "df_train2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "19d3a3ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>category</th>\n",
       "      <th>amt</th>\n",
       "      <th>gender</th>\n",
       "      <th>state</th>\n",
       "      <th>city_pop</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>age_at_trans</th>\n",
       "      <th>lat_dist</th>\n",
       "      <th>long_dist</th>\n",
       "      <th>trans_month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>misc_net</td>\n",
       "      <td>4.97</td>\n",
       "      <td>0</td>\n",
       "      <td>NC</td>\n",
       "      <td>3495</td>\n",
       "      <td>0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.068</td>\n",
       "      <td>0.870</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>grocery_pos</td>\n",
       "      <td>107.23</td>\n",
       "      <td>0</td>\n",
       "      <td>WA</td>\n",
       "      <td>149</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0.271</td>\n",
       "      <td>0.024</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>entertainment</td>\n",
       "      <td>220.11</td>\n",
       "      <td>1</td>\n",
       "      <td>ID</td>\n",
       "      <td>4154</td>\n",
       "      <td>0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.970</td>\n",
       "      <td>0.108</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>gas_transport</td>\n",
       "      <td>45.00</td>\n",
       "      <td>1</td>\n",
       "      <td>MT</td>\n",
       "      <td>1939</td>\n",
       "      <td>0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>0.804</td>\n",
       "      <td>0.447</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>misc_pos</td>\n",
       "      <td>41.96</td>\n",
       "      <td>1</td>\n",
       "      <td>VA</td>\n",
       "      <td>99</td>\n",
       "      <td>0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0.254</td>\n",
       "      <td>0.830</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        category     amt  gender state  city_pop  is_fraud  age_at_trans  \\\n",
       "0       misc_net    4.97       0    NC      3495         0          30.0   \n",
       "1    grocery_pos  107.23       0    WA       149         0          40.0   \n",
       "2  entertainment  220.11       1    ID      4154         0          56.0   \n",
       "3  gas_transport   45.00       1    MT      1939         0          51.0   \n",
       "4       misc_pos   41.96       1    VA        99         0          32.0   \n",
       "\n",
       "   lat_dist  long_dist  trans_month  \n",
       "0     0.068      0.870            1  \n",
       "1     0.271      0.024            1  \n",
       "2     0.970      0.108            1  \n",
       "3     0.804      0.447            1  \n",
       "4     0.254      0.830            1  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Binary mapping on Gender\n",
    "  \n",
    "df_train2['gender'] = df_train2['gender'].map({'M': 1, 'F': 0})\n",
    "df_test2['gender'] = df_test2['gender'].map({'M': 1, 'F': 0})\n",
    "\n",
    "df_train2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "38a00d8e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>amt</th>\n",
       "      <th>city_pop</th>\n",
       "      <th>age_at_trans</th>\n",
       "      <th>lat_dist</th>\n",
       "      <th>long_dist</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.296675e+06</td>\n",
       "      <td>1.296675e+06</td>\n",
       "      <td>1.296675e+06</td>\n",
       "      <td>1.296675e+06</td>\n",
       "      <td>1.296675e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.035104e+01</td>\n",
       "      <td>8.882444e+04</td>\n",
       "      <td>4.549592e+01</td>\n",
       "      <td>5.002626e-01</td>\n",
       "      <td>5.003366e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.603160e+02</td>\n",
       "      <td>3.019564e+05</td>\n",
       "      <td>1.739739e+01</td>\n",
       "      <td>2.885700e-01</td>\n",
       "      <td>2.888704e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>2.300000e+01</td>\n",
       "      <td>1.300000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>9.650000e+00</td>\n",
       "      <td>7.430000e+02</td>\n",
       "      <td>3.200000e+01</td>\n",
       "      <td>2.500000e-01</td>\n",
       "      <td>2.500000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.752000e+01</td>\n",
       "      <td>2.456000e+03</td>\n",
       "      <td>4.300000e+01</td>\n",
       "      <td>5.010000e-01</td>\n",
       "      <td>5.000000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>8.314000e+01</td>\n",
       "      <td>2.032800e+04</td>\n",
       "      <td>5.700000e+01</td>\n",
       "      <td>7.500000e-01</td>\n",
       "      <td>7.510000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90%</th>\n",
       "      <td>1.366700e+02</td>\n",
       "      <td>1.861400e+05</td>\n",
       "      <td>7.000000e+01</td>\n",
       "      <td>9.000000e-01</td>\n",
       "      <td>9.000000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95%</th>\n",
       "      <td>1.963100e+02</td>\n",
       "      <td>5.257130e+05</td>\n",
       "      <td>7.900000e+01</td>\n",
       "      <td>9.500000e-01</td>\n",
       "      <td>9.500000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99%</th>\n",
       "      <td>5.459926e+02</td>\n",
       "      <td>1.577385e+06</td>\n",
       "      <td>9.100000e+01</td>\n",
       "      <td>9.900000e-01</td>\n",
       "      <td>9.900000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.894890e+04</td>\n",
       "      <td>2.906700e+06</td>\n",
       "      <td>9.500000e+01</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                amt      city_pop  age_at_trans      lat_dist     long_dist\n",
       "count  1.296675e+06  1.296675e+06  1.296675e+06  1.296675e+06  1.296675e+06\n",
       "mean   7.035104e+01  8.882444e+04  4.549592e+01  5.002626e-01  5.003366e-01\n",
       "std    1.603160e+02  3.019564e+05  1.739739e+01  2.885700e-01  2.888704e-01\n",
       "min    1.000000e+00  2.300000e+01  1.300000e+01  0.000000e+00  0.000000e+00\n",
       "25%    9.650000e+00  7.430000e+02  3.200000e+01  2.500000e-01  2.500000e-01\n",
       "50%    4.752000e+01  2.456000e+03  4.300000e+01  5.010000e-01  5.000000e-01\n",
       "75%    8.314000e+01  2.032800e+04  5.700000e+01  7.500000e-01  7.510000e-01\n",
       "90%    1.366700e+02  1.861400e+05  7.000000e+01  9.000000e-01  9.000000e-01\n",
       "95%    1.963100e+02  5.257130e+05  7.900000e+01  9.500000e-01  9.500000e-01\n",
       "99%    5.459926e+02  1.577385e+06  9.100000e+01  9.900000e-01  9.900000e-01\n",
       "max    2.894890e+04  2.906700e+06  9.500000e+01  1.000000e+00  1.000000e+00"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train2[['amt','city_pop','age_at_trans','lat_dist','long_dist']].describe(percentiles=[.25, .5, .75, .90, .95, .99])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "c880dc66",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 22 columns):\n",
      " #   Column                   Non-Null Count    Dtype  \n",
      "---  ------                   --------------    -----  \n",
      " 0   amt                      1296675 non-null  float64\n",
      " 1   gender                   1296675 non-null  int64  \n",
      " 2   state                    1296675 non-null  object \n",
      " 3   city_pop                 1296675 non-null  int64  \n",
      " 4   is_fraud                 1296675 non-null  int64  \n",
      " 5   age_at_trans             1296675 non-null  float64\n",
      " 6   lat_dist                 1296675 non-null  float64\n",
      " 7   long_dist                1296675 non-null  float64\n",
      " 8   trans_month              1296675 non-null  int64  \n",
      " 9   category_food_dining     1296675 non-null  uint8  \n",
      " 10  category_gas_transport   1296675 non-null  uint8  \n",
      " 11  category_grocery_net     1296675 non-null  uint8  \n",
      " 12  category_grocery_pos     1296675 non-null  uint8  \n",
      " 13  category_health_fitness  1296675 non-null  uint8  \n",
      " 14  category_home            1296675 non-null  uint8  \n",
      " 15  category_kids_pets       1296675 non-null  uint8  \n",
      " 16  category_misc_net        1296675 non-null  uint8  \n",
      " 17  category_misc_pos        1296675 non-null  uint8  \n",
      " 18  category_personal_care   1296675 non-null  uint8  \n",
      " 19  category_shopping_net    1296675 non-null  uint8  \n",
      " 20  category_shopping_pos    1296675 non-null  uint8  \n",
      " 21  category_travel          1296675 non-null  uint8  \n",
      "dtypes: float64(4), int64(4), object(1), uint8(13)\n",
      "memory usage: 105.1+ MB\n"
     ]
    }
   ],
   "source": [
    "# Creating dummy variables for Category\n",
    "\n",
    "df_train3 = pd.get_dummies(df_train2,columns=['category'],drop_first=True)\n",
    "df_test3 = pd.get_dummies(df_test2,columns=['category'],drop_first=True)\n",
    "\n",
    "df_train3.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "354773de",
   "metadata": {},
   "source": [
    "Building Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "a00eea4d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1296675 entries, 0 to 1296674\n",
      "Data columns (total 21 columns):\n",
      " #   Column                   Non-Null Count    Dtype  \n",
      "---  ------                   --------------    -----  \n",
      " 0   amt                      1296675 non-null  float64\n",
      " 1   gender                   1296675 non-null  int64  \n",
      " 2   city_pop                 1296675 non-null  int64  \n",
      " 3   is_fraud                 1296675 non-null  int64  \n",
      " 4   age_at_trans             1296675 non-null  float64\n",
      " 5   lat_dist                 1296675 non-null  float64\n",
      " 6   long_dist                1296675 non-null  float64\n",
      " 7   trans_month              1296675 non-null  int64  \n",
      " 8   category_food_dining     1296675 non-null  uint8  \n",
      " 9   category_gas_transport   1296675 non-null  uint8  \n",
      " 10  category_grocery_net     1296675 non-null  uint8  \n",
      " 11  category_grocery_pos     1296675 non-null  uint8  \n",
      " 12  category_health_fitness  1296675 non-null  uint8  \n",
      " 13  category_home            1296675 non-null  uint8  \n",
      " 14  category_kids_pets       1296675 non-null  uint8  \n",
      " 15  category_misc_net        1296675 non-null  uint8  \n",
      " 16  category_misc_pos        1296675 non-null  uint8  \n",
      " 17  category_personal_care   1296675 non-null  uint8  \n",
      " 18  category_shopping_net    1296675 non-null  uint8  \n",
      " 19  category_shopping_pos    1296675 non-null  uint8  \n",
      " 20  category_travel          1296675 non-null  uint8  \n",
      "dtypes: float64(4), int64(4), uint8(13)\n",
      "memory usage: 95.2 MB\n"
     ]
    }
   ],
   "source": [
    "# Dropping State\n",
    "\n",
    "df_train4 = df_train3.drop('state',axis=1)\n",
    "df_test4 = df_test3.drop('state',axis=1)\n",
    "df_train4.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "a80bc968",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Splitting train and test dataset into X and y\n",
    "\n",
    "X_train = df_train4.drop(['is_fraud'],axis=1)\n",
    "y_train = df_train4['is_fraud']\n",
    "\n",
    "X_test = df_test4.drop(['is_fraud'],axis=1)\n",
    "y_test = df_test4['is_fraud']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fce12e7d",
   "metadata": {},
   "source": [
    "Oversampling data to compensate for the high class imbalance in the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "032dfa54",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting imbalanced-learn\n",
      "  Downloading imbalanced_learn-0.11.0-py3-none-any.whl (235 kB)\n",
      "Requirement already satisfied: numpy>=1.17.3 in c:\\users\\rahul\\anaconda3\\lib\\site-packages (from imbalanced-learn) (1.21.5)\n",
      "Requirement already satisfied: scipy>=1.5.0 in c:\\users\\rahul\\anaconda3\\lib\\site-packages (from imbalanced-learn) (1.7.3)\n",
      "Collecting joblib>=1.1.1\n",
      "  Downloading joblib-1.3.1-py3-none-any.whl (301 kB)\n",
      "Requirement already satisfied: scikit-learn>=1.0.2 in c:\\users\\rahul\\anaconda3\\lib\\site-packages (from imbalanced-learn) (1.0.2)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in c:\\users\\rahul\\anaconda3\\lib\\site-packages (from imbalanced-learn) (2.2.0)\n",
      "Installing collected packages: joblib, imbalanced-learn\n",
      "  Attempting uninstall: joblib\n",
      "    Found existing installation: joblib 1.1.0\n",
      "    Uninstalling joblib-1.1.0:\n",
      "      Successfully uninstalled joblib-1.1.0\n",
      "Successfully installed imbalanced-learn-0.11.0 joblib-1.3.1\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install -U imbalanced-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "12e46101",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting package metadata (current_repodata.json): ...working... done\n",
      "Solving environment: ...working... done\n",
      "\n",
      "# All requested packages already installed.\n",
      "\n",
      "Retrieving notices: ...working... done\n",
      "\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "==> WARNING: A newer version of conda exists. <==\n",
      "  current version: 4.14.0\n",
      "  latest version: 23.7.2\n",
      "\n",
      "Please update conda by running\n",
      "\n",
      "    $ conda update -n base -c conda-forge conda\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "conda install -c conda-forge imbalanced-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "10c00dc4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(0, 1289169), (1, 1288749)]\n"
     ]
    }
   ],
   "source": [
    "# Using ADASYN to oversample\n",
    "\n",
    "from imblearn.over_sampling import ADASYN\n",
    "X_train_res, y_train_res = ADASYN().fit_resample(X_train, y_train)\n",
    "\n",
    "from collections import Counter\n",
    "print(sorted(Counter(y_train_res).items()))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9774946",
   "metadata": {},
   "source": [
    "Building a Random Forest Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "ec30843b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing required library\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import plot_roc_curve\n",
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "57aabdba",
   "metadata": {},
   "outputs": [],
   "source": [
    "rf = RandomForestClassifier(n_estimators=10, max_depth=5, max_features=12, min_samples_leaf=100, random_state=100, oob_score=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "fad7817e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Rahul\\anaconda3\\lib\\site-packages\\sklearn\\ensemble\\_forest.py:560: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable OOB estimates.\n",
      "  warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(max_depth=5, max_features=12, min_samples_leaf=100,\n",
       "                       n_estimators=10, oob_score=True, random_state=100)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.fit(X_train_res, y_train_res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "b5aaf2fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8838337759385675"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.oob_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "275a4867",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Rahul\\anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function plot_roc_curve is deprecated; Function :func:`plot_roc_curve` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: :meth:`sklearn.metric.RocCurveDisplay.from_predictions` or :meth:`sklearn.metric.RocCurveDisplay.from_estimator`.\n",
      "  warnings.warn(msg, category=FutureWarning)\n"
     ]
    },
    {
     "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": [
    "plot_roc_curve(rf, X_train_res, y_train_res)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "951f7ff2",
   "metadata": {},
   "outputs": [],
   "source": [
    "imp_df = pd.DataFrame({\n",
    "    \"Varname\": X_train.columns,\n",
    "    \"Imp\": rf.feature_importances_\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "dbbd6021",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Varname</th>\n",
       "      <th>Imp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>amt</td>\n",
       "      <td>0.876627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>category_kids_pets</td>\n",
       "      <td>0.028953</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>category_gas_transport</td>\n",
       "      <td>0.023618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>category_home</td>\n",
       "      <td>0.013925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>category_shopping_pos</td>\n",
       "      <td>0.011293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>category_travel</td>\n",
       "      <td>0.010896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>category_grocery_pos</td>\n",
       "      <td>0.010110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>category_misc_pos</td>\n",
       "      <td>0.008930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>category_food_dining</td>\n",
       "      <td>0.004284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>category_shopping_net</td>\n",
       "      <td>0.003880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gender</td>\n",
       "      <td>0.003284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age_at_trans</td>\n",
       "      <td>0.001899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>city_pop</td>\n",
       "      <td>0.001509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>category_health_fitness</td>\n",
       "      <td>0.000416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>category_grocery_net</td>\n",
       "      <td>0.000194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>lat_dist</td>\n",
       "      <td>0.000141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>trans_month</td>\n",
       "      <td>0.000043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>long_dist</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>category_misc_net</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>category_personal_care</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Varname       Imp\n",
       "0                       amt  0.876627\n",
       "13       category_kids_pets  0.028953\n",
       "8    category_gas_transport  0.023618\n",
       "12            category_home  0.013925\n",
       "18    category_shopping_pos  0.011293\n",
       "19          category_travel  0.010896\n",
       "10     category_grocery_pos  0.010110\n",
       "15        category_misc_pos  0.008930\n",
       "7      category_food_dining  0.004284\n",
       "17    category_shopping_net  0.003880\n",
       "1                    gender  0.003284\n",
       "3              age_at_trans  0.001899\n",
       "2                  city_pop  0.001509\n",
       "11  category_health_fitness  0.000416\n",
       "9      category_grocery_net  0.000194\n",
       "4                  lat_dist  0.000141\n",
       "6               trans_month  0.000043\n",
       "5                 long_dist  0.000000\n",
       "14        category_misc_net  0.000000\n",
       "16   category_personal_care  0.000000"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imp_df.sort_values(by=\"Imp\", ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "cd6074df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_pred = rf.predict(X_train_res)\n",
    "y_train_pred[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "7247114e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.90      0.89      0.90   1289169\n",
      "           1       0.89      0.90      0.90   1288749\n",
      "\n",
      "    accuracy                           0.90   2577918\n",
      "   macro avg       0.90      0.90      0.90   2577918\n",
      "weighted avg       0.90      0.90      0.90   2577918\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_train_res, y_train_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9cbf914",
   "metadata": {},
   "source": [
    "Predicting on test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "16f2a6b8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 1, 0, 1, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test_pred = rf.predict(X_test)\n",
    "y_test_pred[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "37225f60",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.89      0.94    553574\n",
      "           1       0.03      0.92      0.06      2145\n",
      "\n",
      "    accuracy                           0.89    555719\n",
      "   macro avg       0.52      0.91      0.50    555719\n",
      "weighted avg       1.00      0.89      0.94    555719\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, y_test_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dbf0c61b",
   "metadata": {},
   "source": [
    "Testing model on original data without oversampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "cb2b3fd8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.89      0.94   1289169\n",
      "           1       0.05      0.93      0.09      7506\n",
      "\n",
      "    accuracy                           0.89   1296675\n",
      "   macro avg       0.52      0.91      0.52   1296675\n",
      "weighted avg       0.99      0.89      0.94   1296675\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.89      0.94    553574\n",
      "           1       0.03      0.92      0.06      2145\n",
      "\n",
      "    accuracy                           0.89    555719\n",
      "   macro avg       0.52      0.91      0.50    555719\n",
      "weighted avg       1.00      0.89      0.94    555719\n",
      "\n"
     ]
    }
   ],
   "source": [
    "y_train_pred = rf.predict(X_train)\n",
    "y_test_pred = rf.predict(X_test)\n",
    "\n",
    "print(classification_report(y_train, y_train_pred))\n",
    "print(classification_report(y_test, y_test_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "c5dd8ee1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "trans_date_trans_time    1274791\n",
       "cc_num                       983\n",
       "merchant                     693\n",
       "category                      14\n",
       "amt                        52928\n",
       "first                        352\n",
       "last                         481\n",
       "gender                         2\n",
       "street                       983\n",
       "city                         894\n",
       "state                         51\n",
       "zip                          970\n",
       "lat                          968\n",
       "long                         969\n",
       "city_pop                     879\n",
       "job                          494\n",
       "dob                          968\n",
       "trans_num                1296675\n",
       "unix_time                1274823\n",
       "merch_lat                1247805\n",
       "merch_long               1275745\n",
       "is_fraud                       2\n",
       "trans_date                   537\n",
       "dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "45d0034d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1296675, 23)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "641b597c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>4.97</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>107.23</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>220.11</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>45.00</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>41.96</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time     amt                         trans_num  is_fraud\n",
       "0   2019-01-01 00:00:18    4.97  0b242abb623afc578575680df30655b9         0\n",
       "1   2019-01-01 00:00:44  107.23  1f76529f8574734946361c461b024d99         0\n",
       "2   2019-01-01 00:00:51  220.11  a1a22d70485983eac12b5b88dad1cf95         0\n",
       "3   2019-01-01 00:01:16   45.00  6b849c168bdad6f867558c3793159a81         0\n",
       "4   2019-01-01 00:03:06   41.96  a41d7549acf90789359a9aa5346dcb46         0"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Subset of df_train\n",
    "\n",
    "df_train_merge = df_train[['trans_date_trans_time','amt','trans_num','is_fraud']]\n",
    "df_train_merge.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "890eeed8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   0\n",
       "0  0\n",
       "1  0\n",
       "2  0\n",
       "3  0\n",
       "4  0"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating dataframe of y_train_pred\n",
    "\n",
    "df_train_pred = pd.DataFrame(y_train_pred)\n",
    "df_train_pred.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "f6ad22f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>4.97</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>107.23</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>220.11</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>45.00</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>41.96</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time     amt                         trans_num  is_fraud  0\n",
       "0   2019-01-01 00:00:18    4.97  0b242abb623afc578575680df30655b9         0  0\n",
       "1   2019-01-01 00:00:44  107.23  1f76529f8574734946361c461b024d99         0  0\n",
       "2   2019-01-01 00:00:51  220.11  a1a22d70485983eac12b5b88dad1cf95         0  0\n",
       "3   2019-01-01 00:01:16   45.00  6b849c168bdad6f867558c3793159a81         0  0\n",
       "4   2019-01-01 00:03:06   41.96  a41d7549acf90789359a9aa5346dcb46         0  0"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merging the two dataframes\n",
    "\n",
    "df_train_final = pd.concat([df_train_merge, df_train_pred],axis=1)\n",
    "df_train_final.head()\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "c98570a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>is_fraud_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>4.97</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>107.23</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>220.11</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>45.00</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>41.96</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time     amt                         trans_num  is_fraud  \\\n",
       "0   2019-01-01 00:00:18    4.97  0b242abb623afc578575680df30655b9         0   \n",
       "1   2019-01-01 00:00:44  107.23  1f76529f8574734946361c461b024d99         0   \n",
       "2   2019-01-01 00:00:51  220.11  a1a22d70485983eac12b5b88dad1cf95         0   \n",
       "3   2019-01-01 00:01:16   45.00  6b849c168bdad6f867558c3793159a81         0   \n",
       "4   2019-01-01 00:03:06   41.96  a41d7549acf90789359a9aa5346dcb46         0   \n",
       "\n",
       "   is_fraud_pred  \n",
       "0              0  \n",
       "1              0  \n",
       "2              0  \n",
       "3              0  \n",
       "4              0  "
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Renaming column\n",
    "\n",
    "df_train_final= df_train_final.rename(columns={ 0 : 'is_fraud_pred'})\n",
    "df_train_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "3cec3bfe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_fraud\n",
       "0    1289169\n",
       "1       7506\n",
       "Name: is_fraud, dtype: int64"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train_final.groupby('is_fraud')['is_fraud'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "d1f138b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_fraud_pred\n",
       "0    1151366\n",
       "1     145309\n",
       "Name: is_fraud_pred, dtype: int64"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train_final.groupby('is_fraud_pred')['is_fraud_pred'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "72bc2b11",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-06-21 12:14:25</td>\n",
       "      <td>2.86</td>\n",
       "      <td>2da90c7d74bd46a0caf3777415b3ebd3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-06-21 12:14:33</td>\n",
       "      <td>29.84</td>\n",
       "      <td>324cc204407e99f51b0d6ca0055005e7</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-06-21 12:14:53</td>\n",
       "      <td>41.28</td>\n",
       "      <td>c81755dbbbea9d5c77f094348a7579be</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-06-21 12:15:15</td>\n",
       "      <td>60.05</td>\n",
       "      <td>2159175b9efe66dc301f149d3d5abf8c</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020-06-21 12:15:17</td>\n",
       "      <td>3.19</td>\n",
       "      <td>57ff021bd3f328f8738bb535c302a31b</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time    amt                         trans_num  is_fraud\n",
       "0   2020-06-21 12:14:25   2.86  2da90c7d74bd46a0caf3777415b3ebd3         0\n",
       "1   2020-06-21 12:14:33  29.84  324cc204407e99f51b0d6ca0055005e7         0\n",
       "2   2020-06-21 12:14:53  41.28  c81755dbbbea9d5c77f094348a7579be         0\n",
       "3   2020-06-21 12:15:15  60.05  2159175b9efe66dc301f149d3d5abf8c         0\n",
       "4   2020-06-21 12:15:17   3.19  57ff021bd3f328f8738bb535c302a31b         0"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Subset of df_test\n",
    "\n",
    "df_test_merge = df_test[['trans_date_trans_time','amt','trans_num','is_fraud']]\n",
    "df_test_merge.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "27224425",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   0\n",
       "0  0\n",
       "1  0\n",
       "2  0\n",
       "3  0\n",
       "4  0"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating dataframe of y_test_pred\n",
    "\n",
    "df_test_pred = pd.DataFrame(y_test_pred)\n",
    "df_test_pred.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "b9362e52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-06-21 12:14:25</td>\n",
       "      <td>2.86</td>\n",
       "      <td>2da90c7d74bd46a0caf3777415b3ebd3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-06-21 12:14:33</td>\n",
       "      <td>29.84</td>\n",
       "      <td>324cc204407e99f51b0d6ca0055005e7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-06-21 12:14:53</td>\n",
       "      <td>41.28</td>\n",
       "      <td>c81755dbbbea9d5c77f094348a7579be</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-06-21 12:15:15</td>\n",
       "      <td>60.05</td>\n",
       "      <td>2159175b9efe66dc301f149d3d5abf8c</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020-06-21 12:15:17</td>\n",
       "      <td>3.19</td>\n",
       "      <td>57ff021bd3f328f8738bb535c302a31b</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time    amt                         trans_num  is_fraud  0\n",
       "0   2020-06-21 12:14:25   2.86  2da90c7d74bd46a0caf3777415b3ebd3         0  0\n",
       "1   2020-06-21 12:14:33  29.84  324cc204407e99f51b0d6ca0055005e7         0  0\n",
       "2   2020-06-21 12:14:53  41.28  c81755dbbbea9d5c77f094348a7579be         0  0\n",
       "3   2020-06-21 12:15:15  60.05  2159175b9efe66dc301f149d3d5abf8c         0  0\n",
       "4   2020-06-21 12:15:17   3.19  57ff021bd3f328f8738bb535c302a31b         0  0"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merging the two dataframes\n",
    "\n",
    "df_test_final = pd.concat([df_test_merge, df_test_pred],axis=1)\n",
    "df_test_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "ca4cf82f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>is_fraud_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-06-21 12:14:25</td>\n",
       "      <td>2.86</td>\n",
       "      <td>2da90c7d74bd46a0caf3777415b3ebd3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-06-21 12:14:33</td>\n",
       "      <td>29.84</td>\n",
       "      <td>324cc204407e99f51b0d6ca0055005e7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-06-21 12:14:53</td>\n",
       "      <td>41.28</td>\n",
       "      <td>c81755dbbbea9d5c77f094348a7579be</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-06-21 12:15:15</td>\n",
       "      <td>60.05</td>\n",
       "      <td>2159175b9efe66dc301f149d3d5abf8c</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020-06-21 12:15:17</td>\n",
       "      <td>3.19</td>\n",
       "      <td>57ff021bd3f328f8738bb535c302a31b</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time    amt                         trans_num  is_fraud  \\\n",
       "0   2020-06-21 12:14:25   2.86  2da90c7d74bd46a0caf3777415b3ebd3         0   \n",
       "1   2020-06-21 12:14:33  29.84  324cc204407e99f51b0d6ca0055005e7         0   \n",
       "2   2020-06-21 12:14:53  41.28  c81755dbbbea9d5c77f094348a7579be         0   \n",
       "3   2020-06-21 12:15:15  60.05  2159175b9efe66dc301f149d3d5abf8c         0   \n",
       "4   2020-06-21 12:15:17   3.19  57ff021bd3f328f8738bb535c302a31b         0   \n",
       "\n",
       "   is_fraud_pred  \n",
       "0              0  \n",
       "1              0  \n",
       "2              0  \n",
       "3              0  \n",
       "4              0  "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Renaming column\n",
    "\n",
    "df_test_final= df_test_final.rename(columns={ 0 : 'is_fraud_pred'})\n",
    "df_test_final.head()\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "d9c2587c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_fraud\n",
       "0    553574\n",
       "1      2145\n",
       "Name: is_fraud, dtype: int64"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test_final.groupby('is_fraud')['is_fraud'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "cb2493f5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_fraud_pred\n",
       "0    494455\n",
       "1     61264\n",
       "Name: is_fraud_pred, dtype: int64"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test_final.groupby('is_fraud_pred')['is_fraud_pred'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "ffbd2dc5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>is_fraud_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>4.97</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>107.23</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>220.11</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>45.00</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>41.96</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time     amt                         trans_num  is_fraud  \\\n",
       "0   2019-01-01 00:00:18    4.97  0b242abb623afc578575680df30655b9         0   \n",
       "1   2019-01-01 00:00:44  107.23  1f76529f8574734946361c461b024d99         0   \n",
       "2   2019-01-01 00:00:51  220.11  a1a22d70485983eac12b5b88dad1cf95         0   \n",
       "3   2019-01-01 00:01:16   45.00  6b849c168bdad6f867558c3793159a81         0   \n",
       "4   2019-01-01 00:03:06   41.96  a41d7549acf90789359a9aa5346dcb46         0   \n",
       "\n",
       "   is_fraud_pred  \n",
       "0              0  \n",
       "1              0  \n",
       "2              0  \n",
       "3              0  \n",
       "4              0  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merging train and test dataset for cost benefit analysis\n",
    "\n",
    "df_merge_final = pd.concat([df_train_final, df_test_final],axis=0)\n",
    "df_merge_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "8e78d076",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1852394, 5)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_merge_final.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea80244c",
   "metadata": {},
   "source": [
    "Cost Benefit Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "d8d5e8d5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trans_date_trans_time</th>\n",
       "      <th>amt</th>\n",
       "      <th>trans_num</th>\n",
       "      <th>is_fraud</th>\n",
       "      <th>is_fraud_pred</th>\n",
       "      <th>month</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01 00:00:18</td>\n",
       "      <td>4.97</td>\n",
       "      <td>0b242abb623afc578575680df30655b9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-01 00:00:44</td>\n",
       "      <td>107.23</td>\n",
       "      <td>1f76529f8574734946361c461b024d99</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-01 00:00:51</td>\n",
       "      <td>220.11</td>\n",
       "      <td>a1a22d70485983eac12b5b88dad1cf95</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-01 00:01:16</td>\n",
       "      <td>45.00</td>\n",
       "      <td>6b849c168bdad6f867558c3793159a81</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-01 00:03:06</td>\n",
       "      <td>41.96</td>\n",
       "      <td>a41d7549acf90789359a9aa5346dcb46</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trans_date_trans_time     amt                         trans_num  is_fraud  \\\n",
       "0   2019-01-01 00:00:18    4.97  0b242abb623afc578575680df30655b9         0   \n",
       "1   2019-01-01 00:00:44  107.23  1f76529f8574734946361c461b024d99         0   \n",
       "2   2019-01-01 00:00:51  220.11  a1a22d70485983eac12b5b88dad1cf95         0   \n",
       "3   2019-01-01 00:01:16   45.00  6b849c168bdad6f867558c3793159a81         0   \n",
       "4   2019-01-01 00:03:06   41.96  a41d7549acf90789359a9aa5346dcb46         0   \n",
       "\n",
       "   is_fraud_pred  month  year  \n",
       "0              0      1  2019  \n",
       "1              0      1  2019  \n",
       "2              0      1  2019  \n",
       "3              0      1  2019  \n",
       "4              0      1  2019  "
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating month and year columns\n",
    "\n",
    "df_merge_final['month'] = pd.DatetimeIndex(df_merge_final['trans_date_trans_time']).month\n",
    "df_merge_final['year'] = pd.DatetimeIndex(df_merge_final['trans_date_trans_time']).year\n",
    "df_merge_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "9b6c75ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Group by variable creation\n",
    "\n",
    "g = df_merge_final.groupby([\"year\",\"month\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "f250f184",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "year  month\n",
       "2019  1         52525\n",
       "      2         49866\n",
       "      3         70939\n",
       "      4         68078\n",
       "      5         72532\n",
       "      6         86064\n",
       "      7         86596\n",
       "      8         87359\n",
       "      9         70652\n",
       "      10        68758\n",
       "      11        70421\n",
       "      12       141060\n",
       "2020  1         52202\n",
       "      2         47791\n",
       "      3         72850\n",
       "      4         66892\n",
       "      5         74343\n",
       "      6         87805\n",
       "      7         85848\n",
       "      8         88759\n",
       "      9         69533\n",
       "      10        69348\n",
       "      11        72635\n",
       "      12       139538\n",
       "Name: trans_num, dtype: int64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of transactions per month\n",
    "\n",
    "g.trans_num.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "0841ada1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "77183.08333333333"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Average number of transactions per month\n",
    "\n",
    "df_merge_final.trans_num.count()/24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "fad19f7f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "402.125"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Average Number of fraudulent transactions per month\n",
    "\n",
    "df_merge_final[df_merge_final['is_fraud']==1].trans_num.count()/24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "2c25e043",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "530.6614122888789"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Average amount per fraud transaction\n",
    "\n",
    "df_merge_final[df_merge_final['is_fraud']==1].amt.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "ad7e653b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8607.208333333334"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Average number of transactions per month detected as fraud by the model\n",
    "\n",
    "df_merge_final[df_merge_final['is_fraud_pred']==1].trans_num.count()/24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "69fa69c0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "27.125"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Average number of transactions per month that are fraudulent but are not detected by the model\n",
    "\n",
    "Undetected_frauds = df_merge_final.loc[(df_merge_final.is_fraud_pred==0) & (df_merge_final.is_fraud==1)]\n",
    "Undetected_frauds.trans_num.count()/24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9f01a04e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "550b75a1",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.9.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}