Updates.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sb\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import accuracy_score ,confusion_matrix , classification_report\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import cross_val_score , cross_validate\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn import metrics\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "import tensorflow as tf\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn import neighbors\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score, confusion_matrix, precision_score, recall_score, f1_score\n",
    "\n",
    "def print_scoreN(clf, X_train, y_train, X_test, y_test, train=True):\n",
    "    if train:\n",
    "        pred = clf.predict(X_train)\n",
    "        pred = (pred > 0.5)\n",
    "        fpr, tpr, _ = metrics.roc_curve(y_train, pred)\n",
    "        auc_score = metrics.auc(fpr, tpr)\n",
    "        print(\"Train Result:\\n================================================\")\n",
    "        print(f\"Accuracy Score: {accuracy_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tAUC score: {auc_score * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_train, pred)}\\n\")\n",
    "        print(classification_report(y_train,pred, digits = 3))\n",
    "        \n",
    "    elif train==False:\n",
    "        pred = clf.predict(X_test)\n",
    "        pred = (pred > 0.5)\n",
    "        fpr, tpr, _ = metrics.roc_curve(y_test, pred)\n",
    "        auc_score = metrics.auc(fpr, tpr)\n",
    "        print(\"Test Result:\\n================================================\")        \n",
    "        print(f\"Accuracy Score: {accuracy_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tAUC score: {auc_score * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_test, pred)}\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score, confusion_matrix, precision_score, recall_score, f1_score\n",
    "\n",
    "def print_score(clf, X_train, y_train, X_test, y_test, train=True):\n",
    "    if train:\n",
    "        pred = clf.predict(X_train)\n",
    "        fpr, tpr, _ = metrics.roc_curve(y_train, pred)\n",
    "        auc_score = metrics.auc(fpr, tpr)\n",
    "        print(\"Train Result:\\n================================================\")\n",
    "        print(f\"Accuracy Score: {accuracy_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tAUC score: {auc_score * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_train, pred)}\\n\")\n",
    "        print(classification_report(y_train,pred, digits = 3))\n",
    "        \n",
    "    elif train==False:\n",
    "        pred = clf.predict(X_test)\n",
    "        fpr, tpr, _ = metrics.roc_curve(y_test, pred)\n",
    "        auc_score = metrics.auc(fpr, tpr)\n",
    "        print(\"Test Result:\\n================================================\")        \n",
    "        print(f\"Accuracy Score: {accuracy_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tAUC score: {auc_score * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_test, pred)}\\n\")\n",
    "        print(classification_report(y_test,pred, digits = 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('train_LZdllcl (2).csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>employee_id</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>KPIs_met &gt;80%</th>\n",
       "      <th>awards_won?</th>\n",
       "      <th>avg_training_score</th>\n",
       "      <th>is_promoted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>54808.000000</td>\n",
       "      <td>54808.000000</td>\n",
       "      <td>54808.000000</td>\n",
       "      <td>50684.000000</td>\n",
       "      <td>54808.000000</td>\n",
       "      <td>54808.000000</td>\n",
       "      <td>54808.000000</td>\n",
       "      <td>54808.000000</td>\n",
       "      <td>54808.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>39195.830627</td>\n",
       "      <td>1.253011</td>\n",
       "      <td>34.803915</td>\n",
       "      <td>3.329256</td>\n",
       "      <td>5.865512</td>\n",
       "      <td>0.351974</td>\n",
       "      <td>0.023172</td>\n",
       "      <td>63.386750</td>\n",
       "      <td>0.085170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>22586.581449</td>\n",
       "      <td>0.609264</td>\n",
       "      <td>7.660169</td>\n",
       "      <td>1.259993</td>\n",
       "      <td>4.265094</td>\n",
       "      <td>0.477590</td>\n",
       "      <td>0.150450</td>\n",
       "      <td>13.371559</td>\n",
       "      <td>0.279137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>19669.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>39225.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>58730.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>78298.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        employee_id  no_of_trainings           age  previous_year_rating  \\\n",
       "count  54808.000000     54808.000000  54808.000000          50684.000000   \n",
       "mean   39195.830627         1.253011     34.803915              3.329256   \n",
       "std    22586.581449         0.609264      7.660169              1.259993   \n",
       "min        1.000000         1.000000     20.000000              1.000000   \n",
       "25%    19669.750000         1.000000     29.000000              3.000000   \n",
       "50%    39225.500000         1.000000     33.000000              3.000000   \n",
       "75%    58730.500000         1.000000     39.000000              4.000000   \n",
       "max    78298.000000        10.000000     60.000000              5.000000   \n",
       "\n",
       "       length_of_service  KPIs_met >80%   awards_won?  avg_training_score  \\\n",
       "count       54808.000000   54808.000000  54808.000000        54808.000000   \n",
       "mean            5.865512       0.351974      0.023172           63.386750   \n",
       "std             4.265094       0.477590      0.150450           13.371559   \n",
       "min             1.000000       0.000000      0.000000           39.000000   \n",
       "25%             3.000000       0.000000      0.000000           51.000000   \n",
       "50%             5.000000       0.000000      0.000000           60.000000   \n",
       "75%             7.000000       1.000000      0.000000           76.000000   \n",
       "max            37.000000       1.000000      1.000000           99.000000   \n",
       "\n",
       "        is_promoted  \n",
       "count  54808.000000  \n",
       "mean       0.085170  \n",
       "std        0.279137  \n",
       "min        0.000000  \n",
       "25%        0.000000  \n",
       "50%        0.000000  \n",
       "75%        0.000000  \n",
       "max        1.000000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>KPIs_met &gt;80%</th>\n",
       "      <th>awards_won?</th>\n",
       "      <th>avg_training_score</th>\n",
       "      <th>is_promoted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>Sales &amp; Marketing</td>\n",
       "      <td>region_7</td>\n",
       "      <td>Master's &amp; above</td>\n",
       "      <td>f</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>Operations</td>\n",
       "      <td>region_22</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>Sales &amp; Marketing</td>\n",
       "      <td>region_19</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>Sales &amp; Marketing</td>\n",
       "      <td>region_23</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>Technology</td>\n",
       "      <td>region_26</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>43923</td>\n",
       "      <td>Analytics</td>\n",
       "      <td>region_14</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>31</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>86</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>64109</td>\n",
       "      <td>Technology</td>\n",
       "      <td>region_26</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>f</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>77</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>41983</td>\n",
       "      <td>Operations</td>\n",
       "      <td>region_2</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>f</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>21440</td>\n",
       "      <td>Sales &amp; Marketing</td>\n",
       "      <td>region_31</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>41</td>\n",
       "      <td>3.0</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>51256</td>\n",
       "      <td>Finance</td>\n",
       "      <td>region_2</td>\n",
       "      <td>Master's &amp; above</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "      <td>5.0</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    employee_id         department     region         education gender  \\\n",
       "0         65438  Sales & Marketing   region_7  Master's & above      f   \n",
       "1         65141         Operations  region_22        Bachelor's      m   \n",
       "2          7513  Sales & Marketing  region_19        Bachelor's      m   \n",
       "3          2542  Sales & Marketing  region_23        Bachelor's      m   \n",
       "4         48945         Technology  region_26        Bachelor's      m   \n",
       "..          ...                ...        ...               ...    ...   \n",
       "95        43923          Analytics  region_14        Bachelor's      m   \n",
       "96        64109         Technology  region_26        Bachelor's      f   \n",
       "97        41983         Operations   region_2        Bachelor's      f   \n",
       "98        21440  Sales & Marketing  region_31        Bachelor's      m   \n",
       "99        51256            Finance   region_2  Master's & above      m   \n",
       "\n",
       "   recruitment_channel  no_of_trainings  age  previous_year_rating  \\\n",
       "0             sourcing                1   35                   5.0   \n",
       "1                other                1   30                   5.0   \n",
       "2             sourcing                1   34                   3.0   \n",
       "3                other                2   39                   1.0   \n",
       "4                other                1   45                   3.0   \n",
       "..                 ...              ...  ...                   ...   \n",
       "95            sourcing                1   31                   4.0   \n",
       "96            sourcing                1   23                   NaN   \n",
       "97            sourcing                1   33                   3.0   \n",
       "98               other                1   41                   3.0   \n",
       "99               other                1   60                   5.0   \n",
       "\n",
       "    length_of_service  KPIs_met >80%  awards_won?  avg_training_score  \\\n",
       "0                   8              1            0                  49   \n",
       "1                   4              0            0                  60   \n",
       "2                   7              0            0                  50   \n",
       "3                  10              0            0                  50   \n",
       "4                   2              0            0                  73   \n",
       "..                ...            ...          ...                 ...   \n",
       "95                  4              1            0                  86   \n",
       "96                  1              0            0                  77   \n",
       "97                  5              0            0                  57   \n",
       "98                 12              1            0                  52   \n",
       "99                 17              0            0                  59   \n",
       "\n",
       "    is_promoted  \n",
       "0             0  \n",
       "1             0  \n",
       "2             0  \n",
       "3             0  \n",
       "4             0  \n",
       "..          ...  \n",
       "95            0  \n",
       "96            0  \n",
       "97            0  \n",
       "98            0  \n",
       "99            0  \n",
       "\n",
       "[100 rows x 14 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head(100)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().values.any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "department : ['Sales & Marketing' 'Operations' 'Technology' 'Analytics' 'R&D'\n",
      " 'Procurement' 'Finance' 'HR' 'Legal']\n",
      "--------------------------------\n",
      "region : ['region_7' 'region_22' 'region_19' 'region_23' 'region_26' 'region_2'\n",
      " 'region_20' 'region_34' 'region_1' 'region_4' 'region_29' 'region_31'\n",
      " 'region_15' 'region_14' 'region_11' 'region_5' 'region_28' 'region_17'\n",
      " 'region_13' 'region_16' 'region_25' 'region_10' 'region_27' 'region_30'\n",
      " 'region_12' 'region_21' 'region_8' 'region_32' 'region_6' 'region_33'\n",
      " 'region_24' 'region_3' 'region_9' 'region_18']\n",
      "--------------------------------\n",
      "education : [\"Master's & above\" \"Bachelor's\" nan 'Below Secondary']\n",
      "--------------------------------\n",
      "gender : ['f' 'm']\n",
      "--------------------------------\n",
      "recruitment_channel : ['sourcing' 'other' 'referred']\n",
      "--------------------------------\n"
     ]
    }
   ],
   "source": [
    "categorical_col = []\n",
    "for column in data.columns:\n",
    "    if data[column].dtype == object and len(data[column].unique()) <= 50:\n",
    "        categorical_col.append(column)\n",
    "        print(f\"{column} : {data[column].unique()}\")\n",
    "        print(\"--------------------------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Bachelor's          36669\n",
       "Master's & above    14925\n",
       "Below Secondary       805\n",
       "Name: education, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.education.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# col=['mediumaquamarine', 'lightskyblue']\n",
    "# sb.set_palette(col)\n",
    "# sb.countplot('is_promoted', data=data, hatch=[\"//\",\"x\"],edgecolor='black',)\n",
    "# plt.show()\n",
    "label1=data[data[\"is_promoted\"]==0]\n",
    "label2=data[data[\"is_promoted\"]==1]\n",
    "\n",
    "n_groups = 1\n",
    "means_frank = label1['is_promoted'].value_counts()\n",
    "means_guido = label2['is_promoted'].value_counts()\n",
    "\n",
    "# create plot\n",
    "fig, ax = plt.subplots()\n",
    "index = np.arange(n_groups)\n",
    "bar_width = 0.35\n",
    "opacity = 0.8\n",
    "\n",
    "rects1 = plt.bar(index, means_frank, bar_width,\n",
    "alpha=opacity,\n",
    "color='mediumaquamarine',\n",
    "hatch=\"//\",\n",
    "edgecolor='black',\n",
    "label='not promoted')\n",
    "\n",
    "rects2 = plt.bar(index + bar_width , means_guido, bar_width, \n",
    "alpha=opacity,\n",
    "color='lightskyblue',\n",
    "hatch=\"x\",\n",
    "edgecolor='black',\n",
    "label='promoted')\n",
    "\n",
    "plt.xlabel('is_promoted')\n",
    "plt.ylabel('Count')\n",
    "#plt.title('no_of_trainings')\n",
    "plt.legend()\n",
    "\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>is_promoted</th>\n",
       "      <th>department_Analytics</th>\n",
       "      <th>department_Finance</th>\n",
       "      <th>department_HR</th>\n",
       "      <th>department_Legal</th>\n",
       "      <th>department_Operations</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>Master's &amp; above</td>\n",
       "      <td>f</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_19</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_23</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>Bachelor's</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   employee_id  department     region         education gender  \\\n",
       "0        65438         0.0   region_7  Master's & above      f   \n",
       "1        65141         0.0  region_22        Bachelor's      m   \n",
       "2         7513         0.0  region_19        Bachelor's      m   \n",
       "3         2542         0.0  region_23        Bachelor's      m   \n",
       "4        48945         0.0  region_26        Bachelor's      m   \n",
       "\n",
       "  recruitment_channel  no_of_trainings  age  previous_year_rating  \\\n",
       "0            sourcing                1   35                   5.0   \n",
       "1               other                1   30                   5.0   \n",
       "2            sourcing                1   34                   3.0   \n",
       "3               other                2   39                   1.0   \n",
       "4               other                1   45                   3.0   \n",
       "\n",
       "   length_of_service  ...  is_promoted  department_Analytics  \\\n",
       "0                  8  ...            0                     0   \n",
       "1                  4  ...            0                     0   \n",
       "2                  7  ...            0                     0   \n",
       "3                 10  ...            0                     0   \n",
       "4                  2  ...            0                     0   \n",
       "\n",
       "   department_Finance  department_HR  department_Legal  department_Operations  \\\n",
       "0                   0              0                 0                      0   \n",
       "1                   0              0                 0                      1   \n",
       "2                   0              0                 0                      0   \n",
       "3                   0              0                 0                      0   \n",
       "4                   0              0                 0                      0   \n",
       "\n",
       "   department_Procurement  department_R&D  department_Sales & Marketing  \\\n",
       "0                       0               0                             1   \n",
       "1                       0               0                             0   \n",
       "2                       0               0                             1   \n",
       "3                       0               0                             1   \n",
       "4                       0               0                             0   \n",
       "\n",
       "   department_Technology  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      1  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#using one hot encoder\n",
    "from sklearn.preprocessing import OneHotEncoder \n",
    "from sklearn.compose import ColumnTransformer \n",
    "  \n",
    "#encode department column\n",
    "enc = OneHotEncoder(sparse=False)\n",
    "cols_to_convert = ['department'] # specify the list of columns here\n",
    "department_col = pd.get_dummies(data['department'], prefix='department')\n",
    "data['department'] = enc.fit_transform(data.loc[:,cols_to_convert])\n",
    "data = data.join(department_col,how=\"right\") #Join the new columns\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 54808 entries, 0 to 54807\n",
      "Data columns (total 23 columns):\n",
      " #   Column                        Non-Null Count  Dtype  \n",
      "---  ------                        --------------  -----  \n",
      " 0   employee_id                   54808 non-null  int64  \n",
      " 1   department                    54808 non-null  float64\n",
      " 2   region                        54808 non-null  object \n",
      " 3   education                     52399 non-null  object \n",
      " 4   gender                        54808 non-null  object \n",
      " 5   recruitment_channel           54808 non-null  object \n",
      " 6   no_of_trainings               54808 non-null  int64  \n",
      " 7   age                           54808 non-null  int64  \n",
      " 8   previous_year_rating          50684 non-null  float64\n",
      " 9   length_of_service             54808 non-null  int64  \n",
      " 10  KPIs_met >80%                 54808 non-null  int64  \n",
      " 11  awards_won?                   54808 non-null  int64  \n",
      " 12  avg_training_score            54808 non-null  int64  \n",
      " 13  is_promoted                   54808 non-null  int64  \n",
      " 14  department_Analytics          54808 non-null  uint8  \n",
      " 15  department_Finance            54808 non-null  uint8  \n",
      " 16  department_HR                 54808 non-null  uint8  \n",
      " 17  department_Legal              54808 non-null  uint8  \n",
      " 18  department_Operations         54808 non-null  uint8  \n",
      " 19  department_Procurement        54808 non-null  uint8  \n",
      " 20  department_R&D                54808 non-null  uint8  \n",
      " 21  department_Sales & Marketing  54808 non-null  uint8  \n",
      " 22  department_Technology         54808 non-null  uint8  \n",
      "dtypes: float64(2), int64(8), object(4), uint8(9)\n",
      "memory usage: 6.3+ MB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 48660 entries, 0 to 54807\n",
      "Data columns (total 23 columns):\n",
      " #   Column                        Non-Null Count  Dtype  \n",
      "---  ------                        --------------  -----  \n",
      " 0   employee_id                   48660 non-null  int64  \n",
      " 1   department                    48660 non-null  float64\n",
      " 2   region                        48660 non-null  object \n",
      " 3   education                     48660 non-null  object \n",
      " 4   gender                        48660 non-null  object \n",
      " 5   recruitment_channel           48660 non-null  object \n",
      " 6   no_of_trainings               48660 non-null  int64  \n",
      " 7   age                           48660 non-null  int64  \n",
      " 8   previous_year_rating          48660 non-null  float64\n",
      " 9   length_of_service             48660 non-null  int64  \n",
      " 10  KPIs_met >80%                 48660 non-null  int64  \n",
      " 11  awards_won?                   48660 non-null  int64  \n",
      " 12  avg_training_score            48660 non-null  int64  \n",
      " 13  is_promoted                   48660 non-null  int64  \n",
      " 14  department_Analytics          48660 non-null  uint8  \n",
      " 15  department_Finance            48660 non-null  uint8  \n",
      " 16  department_HR                 48660 non-null  uint8  \n",
      " 17  department_Legal              48660 non-null  uint8  \n",
      " 18  department_Operations         48660 non-null  uint8  \n",
      " 19  department_Procurement        48660 non-null  uint8  \n",
      " 20  department_R&D                48660 non-null  uint8  \n",
      " 21  department_Sales & Marketing  48660 non-null  uint8  \n",
      " 22  department_Technology         48660 non-null  uint8  \n",
      "dtypes: float64(2), int64(8), object(4), uint8(9)\n",
      "memory usage: 6.0+ MB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    44428\n",
       "1     4232\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.is_promoted.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.08517160216760633"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "4668/54807"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>department_HR</th>\n",
       "      <th>department_Legal</th>\n",
       "      <th>department_Operations</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>sourcing</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   employee_id  department     region  education gender recruitment_channel  \\\n",
       "0        65438         0.0   region_7        0.0      f            sourcing   \n",
       "1        65141         0.0  region_22        1.0      m               other   \n",
       "2         7513         0.0  region_19        1.0      m            sourcing   \n",
       "3         2542         0.0  region_23        1.0      m               other   \n",
       "4        48945         0.0  region_26        1.0      m               other   \n",
       "\n",
       "   no_of_trainings  age  previous_year_rating  length_of_service  ...  \\\n",
       "0                1   35                   5.0                  8  ...   \n",
       "1                1   30                   5.0                  4  ...   \n",
       "2                1   34                   3.0                  7  ...   \n",
       "3                2   39                   1.0                 10  ...   \n",
       "4                1   45                   3.0                  2  ...   \n",
       "\n",
       "   department_HR  department_Legal  department_Operations  \\\n",
       "0              0                 0                      0   \n",
       "1              0                 0                      1   \n",
       "2              0                 0                      0   \n",
       "3              0                 0                      0   \n",
       "4              0                 0                      0   \n",
       "\n",
       "   department_Procurement  department_R&D  department_Sales & Marketing  \\\n",
       "0                       0               0                             1   \n",
       "1                       0               0                             0   \n",
       "2                       0               0                             1   \n",
       "3                       0               0                             1   \n",
       "4                       0               0                             0   \n",
       "\n",
       "   department_Technology  education_Bachelor's  education_Below Secondary  \\\n",
       "0                      0                     0                          0   \n",
       "1                      0                     1                          0   \n",
       "2                      0                     1                          0   \n",
       "3                      0                     1                          0   \n",
       "4                      1                     1                          0   \n",
       "\n",
       "   education_Master's & above  \n",
       "0                           1  \n",
       "1                           0  \n",
       "2                           0  \n",
       "3                           0  \n",
       "4                           0  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#encode education column\n",
    "enc = OneHotEncoder(sparse=False)\n",
    "cols_to_convert = ['education'] # specify the list of columns here\n",
    "education_col = pd.get_dummies(data['education'], prefix='education')\n",
    "data['education'] = enc.fit_transform(data.loc[:,cols_to_convert])\n",
    "data = data.join(education_col,how=\"right\") #Join the new columns\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   employee_id  department     region  education gender  recruitment_channel  \\\n",
       "0        65438         0.0   region_7        0.0      f                  0.0   \n",
       "1        65141         0.0  region_22        1.0      m                  1.0   \n",
       "2         7513         0.0  region_19        1.0      m                  0.0   \n",
       "3         2542         0.0  region_23        1.0      m                  1.0   \n",
       "4        48945         0.0  region_26        1.0      m                  1.0   \n",
       "\n",
       "   no_of_trainings  age  previous_year_rating  length_of_service  ...  \\\n",
       "0                1   35                   5.0                  8  ...   \n",
       "1                1   30                   5.0                  4  ...   \n",
       "2                1   34                   3.0                  7  ...   \n",
       "3                2   39                   1.0                 10  ...   \n",
       "4                1   45                   3.0                  2  ...   \n",
       "\n",
       "   department_Procurement  department_R&D  department_Sales & Marketing  \\\n",
       "0                       0               0                             1   \n",
       "1                       0               0                             0   \n",
       "2                       0               0                             1   \n",
       "3                       0               0                             1   \n",
       "4                       0               0                             0   \n",
       "\n",
       "   department_Technology  education_Bachelor's  education_Below Secondary  \\\n",
       "0                      0                     0                          0   \n",
       "1                      0                     1                          0   \n",
       "2                      0                     1                          0   \n",
       "3                      0                     1                          0   \n",
       "4                      1                     1                          0   \n",
       "\n",
       "   education_Master's & above  recruitment_channel_other  \\\n",
       "0                           1                          0   \n",
       "1                           0                          1   \n",
       "2                           0                          0   \n",
       "3                           0                          1   \n",
       "4                           0                          1   \n",
       "\n",
       "   recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "0                             0                             1  \n",
       "1                             0                             0  \n",
       "2                             0                             1  \n",
       "3                             0                             0  \n",
       "4                             0                             0  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#encode recruitment_channel column\n",
    "enc = OneHotEncoder(sparse=False)\n",
    "cols_to_convert = ['recruitment_channel'] # specify the list of columns here\n",
    "recruitment_channel_col = pd.get_dummies(data['recruitment_channel'], prefix='recruitment_channel')\n",
    "data['recruitment_channel'] = enc.fit_transform(data.loc[:,cols_to_convert])\n",
    "data = data.join(recruitment_channel_col,how=\"right\") #Join the new columns\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(48660, 29)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 48660 entries, 0 to 54807\n",
      "Data columns (total 29 columns):\n",
      " #   Column                        Non-Null Count  Dtype  \n",
      "---  ------                        --------------  -----  \n",
      " 0   employee_id                   48660 non-null  int64  \n",
      " 1   department                    48660 non-null  float64\n",
      " 2   region                        48660 non-null  object \n",
      " 3   education                     48660 non-null  float64\n",
      " 4   gender                        48660 non-null  object \n",
      " 5   recruitment_channel           48660 non-null  float64\n",
      " 6   no_of_trainings               48660 non-null  int64  \n",
      " 7   age                           48660 non-null  int64  \n",
      " 8   previous_year_rating          48660 non-null  float64\n",
      " 9   length_of_service             48660 non-null  int64  \n",
      " 10  KPIs_met >80%                 48660 non-null  int64  \n",
      " 11  awards_won?                   48660 non-null  int64  \n",
      " 12  avg_training_score            48660 non-null  int64  \n",
      " 13  is_promoted                   48660 non-null  int64  \n",
      " 14  department_Analytics          48660 non-null  uint8  \n",
      " 15  department_Finance            48660 non-null  uint8  \n",
      " 16  department_HR                 48660 non-null  uint8  \n",
      " 17  department_Legal              48660 non-null  uint8  \n",
      " 18  department_Operations         48660 non-null  uint8  \n",
      " 19  department_Procurement        48660 non-null  uint8  \n",
      " 20  department_R&D                48660 non-null  uint8  \n",
      " 21  department_Sales & Marketing  48660 non-null  uint8  \n",
      " 22  department_Technology         48660 non-null  uint8  \n",
      " 23  education_Bachelor's          48660 non-null  uint8  \n",
      " 24  education_Below Secondary     48660 non-null  uint8  \n",
      " 25  education_Master's & above    48660 non-null  uint8  \n",
      " 26  recruitment_channel_other     48660 non-null  uint8  \n",
      " 27  recruitment_channel_referred  48660 non-null  uint8  \n",
      " 28  recruitment_channel_sourcing  48660 non-null  uint8  \n",
      "dtypes: float64(4), int64(8), object(2), uint8(15)\n",
      "memory usage: 7.5+ MB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "employee_id                     0\n",
       "department                      0\n",
       "region                          0\n",
       "education                       0\n",
       "gender                          0\n",
       "recruitment_channel             0\n",
       "no_of_trainings                 0\n",
       "age                             0\n",
       "previous_year_rating            0\n",
       "length_of_service               0\n",
       "KPIs_met >80%                   0\n",
       "awards_won?                     0\n",
       "avg_training_score              0\n",
       "is_promoted                     0\n",
       "department_Analytics            0\n",
       "department_Finance              0\n",
       "department_HR                   0\n",
       "department_Legal                0\n",
       "department_Operations           0\n",
       "department_Procurement          0\n",
       "department_R&D                  0\n",
       "department_Sales & Marketing    0\n",
       "department_Technology           0\n",
       "education_Bachelor's            0\n",
       "education_Below Secondary       0\n",
       "education_Master's & above      0\n",
       "recruitment_channel_other       0\n",
       "recruitment_channel_referred    0\n",
       "recruitment_channel_sourcing    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.0    17867\n",
       "5.0    11362\n",
       "4.0     9509\n",
       "1.0     5887\n",
       "2.0     4035\n",
       "Name: previous_year_rating, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.previous_year_rating.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#data['previous_year_rating'] = data['previous_year_rating'].replace(np.nan,0) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "48660"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "region : ['region_7' 'region_22' 'region_19' 'region_23' 'region_26' 'region_2'\n",
      " 'region_20' 'region_34' 'region_1' 'region_4' 'region_29' 'region_31'\n",
      " 'region_15' 'region_14' 'region_11' 'region_5' 'region_28' 'region_17'\n",
      " 'region_13' 'region_16' 'region_25' 'region_10' 'region_27' 'region_30'\n",
      " 'region_12' 'region_21' 'region_32' 'region_6' 'region_33' 'region_8'\n",
      " 'region_24' 'region_3' 'region_9' 'region_18']\n",
      "--------------------------------\n",
      "gender : ['f' 'm']\n",
      "--------------------------------\n"
     ]
    }
   ],
   "source": [
    "categorical_col = []\n",
    "for column in data.columns:\n",
    "    if data[column].dtype == object and len(data[column].unique()) <= 50:\n",
    "        categorical_col.append(column)\n",
    "        print(f\"{column} : {data[column].unique()}\")\n",
    "        print(\"--------------------------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>58896</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>20379</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_20</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>31</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>16290</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_34</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>73202</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_20</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>28911</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>32</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>49017</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>60051</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>49</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>38401</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_29</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>3.0</td>\n",
       "      <td>16</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>77040</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>43931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>7152</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>38</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>9403</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>17436</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>54461</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_15</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>3.0</td>\n",
       "      <td>9</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>12067</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>58789</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_11</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>52057</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>26585</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>37749</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>31</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>2370</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>8219</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>29</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>27395</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_15</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>5.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>58536</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>51756</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>57</td>\n",
       "      <td>4.0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>57011</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>27</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>77064</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>40</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>64062</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_28</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>47123</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>40</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>56275</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>58304</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_28</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>25014</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_15</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>40097</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>23</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>23560</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>59</td>\n",
       "      <td>4.0</td>\n",
       "      <td>26</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>61758</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>44</td>\n",
       "      <td>5.0</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>75009</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>50</td>\n",
       "      <td>5.0</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>17876</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_13</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>36924</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_16</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>40</td>\n",
       "      <td>3.0</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>50408</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_20</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>18013</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>56</td>\n",
       "      <td>3.0</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>24104</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>32083</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_25</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>31</td>\n",
       "      <td>3.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>38330</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_15</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>40</td>\n",
       "      <td>3.0</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>33690</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>73161</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>39</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>13596</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>26</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>50 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    employee_id  department     region  education gender  recruitment_channel  \\\n",
       "0         65438         0.0   region_7        0.0      f                  0.0   \n",
       "1         65141         0.0  region_22        1.0      m                  1.0   \n",
       "2          7513         0.0  region_19        1.0      m                  0.0   \n",
       "3          2542         0.0  region_23        1.0      m                  1.0   \n",
       "4         48945         0.0  region_26        1.0      m                  1.0   \n",
       "5         58896         1.0   region_2        1.0      m                  0.0   \n",
       "6         20379         0.0  region_20        1.0      f                  1.0   \n",
       "7         16290         0.0  region_34        0.0      m                  0.0   \n",
       "8         73202         1.0  region_20        1.0      m                  1.0   \n",
       "9         28911         0.0   region_1        0.0      m                  0.0   \n",
       "11        49017         0.0   region_7        1.0      f                  0.0   \n",
       "12        60051         0.0   region_4        1.0      m                  0.0   \n",
       "13        38401         0.0  region_29        0.0      m                  1.0   \n",
       "14        77040         0.0   region_2        0.0      m                  0.0   \n",
       "15        43931         0.0   region_7        1.0      m                  1.0   \n",
       "16         7152         0.0   region_2        1.0      m                  1.0   \n",
       "17         9403         0.0  region_31        1.0      m                  1.0   \n",
       "18        17436         0.0  region_31        1.0      m                  1.0   \n",
       "19        54461         0.0  region_15        1.0      m                  1.0   \n",
       "20        12067         0.0  region_14        1.0      m                  1.0   \n",
       "22        58789         0.0  region_11        1.0      f                  1.0   \n",
       "24        52057         0.0  region_22        0.0      m                  0.0   \n",
       "25        26585         0.0  region_22        1.0      m                  1.0   \n",
       "26        37749         0.0  region_26        1.0      m                  1.0   \n",
       "27         2370         0.0  region_26        1.0      m                  1.0   \n",
       "28         8219         1.0   region_5        1.0      m                  1.0   \n",
       "30        27395         0.0  region_15        1.0      m                  0.0   \n",
       "31        58536         0.0  region_31        1.0      m                  1.0   \n",
       "33        51756         0.0   region_2        1.0      m                  0.0   \n",
       "34        57011         1.0  region_22        1.0      m                  1.0   \n",
       "35        77064         0.0  region_22        1.0      m                  1.0   \n",
       "36        64062         0.0  region_28        1.0      m                  0.0   \n",
       "37        47123         0.0   region_7        1.0      f                  0.0   \n",
       "38        56275         0.0   region_1        0.0      m                  1.0   \n",
       "39        58304         0.0  region_28        1.0      m                  0.0   \n",
       "40        25014         0.0  region_15        1.0      m                  1.0   \n",
       "41        40097         0.0   region_2        1.0      f                  1.0   \n",
       "42        23560         0.0   region_2        1.0      m                  0.0   \n",
       "44        61758         1.0   region_2        1.0      m                  1.0   \n",
       "45        75009         0.0  region_17        1.0      f                  0.0   \n",
       "46        17876         0.0  region_13        0.0      f                  1.0   \n",
       "47        36924         0.0  region_16        1.0      m                  0.0   \n",
       "48        50408         0.0  region_20        1.0      m                  0.0   \n",
       "49        18013         0.0   region_2        0.0      f                  0.0   \n",
       "50        24104         0.0   region_2        1.0      m                  0.0   \n",
       "51        32083         0.0  region_25        1.0      f                  0.0   \n",
       "52        38330         0.0  region_15        1.0      m                  0.0   \n",
       "53        33690         0.0  region_22        1.0      m                  1.0   \n",
       "54        73161         0.0  region_22        1.0      m                  1.0   \n",
       "55        13596         1.0  region_31        1.0      m                  1.0   \n",
       "\n",
       "    no_of_trainings  age  previous_year_rating  length_of_service  ...  \\\n",
       "0                 1   35                   5.0                  8  ...   \n",
       "1                 1   30                   5.0                  4  ...   \n",
       "2                 1   34                   3.0                  7  ...   \n",
       "3                 2   39                   1.0                 10  ...   \n",
       "4                 1   45                   3.0                  2  ...   \n",
       "5                 2   31                   3.0                  7  ...   \n",
       "6                 1   31                   3.0                  5  ...   \n",
       "7                 1   33                   3.0                  6  ...   \n",
       "8                 1   28                   4.0                  5  ...   \n",
       "9                 1   32                   5.0                  5  ...   \n",
       "11                1   35                   5.0                  3  ...   \n",
       "12                1   49                   5.0                  5  ...   \n",
       "13                2   39                   3.0                 16  ...   \n",
       "14                1   37                   3.0                  7  ...   \n",
       "15                1   37                   1.0                 10  ...   \n",
       "16                1   38                   3.0                  5  ...   \n",
       "17                1   34                   1.0                  4  ...   \n",
       "18                1   34                   5.0                  8  ...   \n",
       "19                1   37                   3.0                  9  ...   \n",
       "20                1   35                   3.0                  7  ...   \n",
       "22                1   28                   3.0                  4  ...   \n",
       "24                2   39                   5.0                  7  ...   \n",
       "25                1   27                   5.0                  3  ...   \n",
       "26                1   31                   5.0                  4  ...   \n",
       "27                1   33                   4.0                  4  ...   \n",
       "28                1   29                   5.0                  3  ...   \n",
       "30                1   37                   5.0                 10  ...   \n",
       "31                2   26                   1.0                  2  ...   \n",
       "33                2   57                   4.0                  6  ...   \n",
       "34                2   27                   5.0                  3  ...   \n",
       "35                1   40                   2.0                  4  ...   \n",
       "36                1   33                   4.0                  4  ...   \n",
       "37                1   40                   5.0                  3  ...   \n",
       "38                1   42                   5.0                  5  ...   \n",
       "39                1   33                   5.0                  6  ...   \n",
       "40                1   34                   3.0                 10  ...   \n",
       "41                2   23                   4.0                  2  ...   \n",
       "42                1   59                   4.0                 26  ...   \n",
       "44                1   44                   5.0                 12  ...   \n",
       "45                1   50                   5.0                 11  ...   \n",
       "46                1   37                   4.0                  4  ...   \n",
       "47                1   40                   3.0                 11  ...   \n",
       "48                1   30                   3.0                  3  ...   \n",
       "49                1   56                   3.0                 12  ...   \n",
       "50                1   20                   4.0                  2  ...   \n",
       "51                1   31                   3.0                  8  ...   \n",
       "52                1   40                   3.0                 12  ...   \n",
       "53                1   30                   1.0                  6  ...   \n",
       "54                1   39                   3.0                  5  ...   \n",
       "55                2   26                   2.0                  2  ...   \n",
       "\n",
       "    department_Procurement  department_R&D  department_Sales & Marketing  \\\n",
       "0                        0               0                             1   \n",
       "1                        0               0                             0   \n",
       "2                        0               0                             1   \n",
       "3                        0               0                             1   \n",
       "4                        0               0                             0   \n",
       "5                        0               0                             0   \n",
       "6                        0               0                             0   \n",
       "7                        0               0                             0   \n",
       "8                        0               0                             0   \n",
       "9                        0               0                             1   \n",
       "11                       0               0                             1   \n",
       "12                       0               0                             1   \n",
       "13                       0               0                             0   \n",
       "14                       0               1                             0   \n",
       "15                       0               0                             0   \n",
       "16                       0               0                             0   \n",
       "17                       0               0                             1   \n",
       "18                       0               0                             1   \n",
       "19                       0               0                             0   \n",
       "20                       1               0                             0   \n",
       "22                       0               0                             0   \n",
       "24                       0               0                             0   \n",
       "25                       0               0                             0   \n",
       "26                       1               0                             0   \n",
       "27                       0               0                             0   \n",
       "28                       0               0                             0   \n",
       "30                       1               0                             0   \n",
       "31                       0               0                             0   \n",
       "33                       0               0                             0   \n",
       "34                       0               0                             0   \n",
       "35                       0               0                             1   \n",
       "36                       0               0                             0   \n",
       "37                       0               0                             1   \n",
       "38                       0               0                             1   \n",
       "39                       0               0                             1   \n",
       "40                       0               0                             0   \n",
       "41                       0               0                             0   \n",
       "42                       0               0                             0   \n",
       "44                       0               0                             0   \n",
       "45                       1               0                             0   \n",
       "46                       1               0                             0   \n",
       "47                       0               0                             0   \n",
       "48                       0               0                             1   \n",
       "49                       1               0                             0   \n",
       "50                       0               0                             0   \n",
       "51                       0               0                             1   \n",
       "52                       1               0                             0   \n",
       "53                       0               0                             0   \n",
       "54                       0               0                             0   \n",
       "55                       0               0                             0   \n",
       "\n",
       "    department_Technology  education_Bachelor's  education_Below Secondary  \\\n",
       "0                       0                     0                          0   \n",
       "1                       0                     1                          0   \n",
       "2                       0                     1                          0   \n",
       "3                       0                     1                          0   \n",
       "4                       1                     1                          0   \n",
       "5                       0                     1                          0   \n",
       "6                       0                     1                          0   \n",
       "7                       0                     0                          0   \n",
       "8                       0                     1                          0   \n",
       "9                       0                     0                          0   \n",
       "11                      0                     1                          0   \n",
       "12                      0                     1                          0   \n",
       "13                      1                     0                          0   \n",
       "14                      0                     0                          0   \n",
       "15                      0                     1                          0   \n",
       "16                      1                     1                          0   \n",
       "17                      0                     1                          0   \n",
       "18                      0                     1                          0   \n",
       "19                      0                     1                          0   \n",
       "20                      0                     1                          0   \n",
       "22                      0                     1                          0   \n",
       "24                      0                     0                          0   \n",
       "25                      1                     1                          0   \n",
       "26                      0                     1                          0   \n",
       "27                      1                     1                          0   \n",
       "28                      0                     1                          0   \n",
       "30                      0                     1                          0   \n",
       "31                      0                     1                          0   \n",
       "33                      0                     1                          0   \n",
       "34                      0                     1                          0   \n",
       "35                      0                     1                          0   \n",
       "36                      1                     1                          0   \n",
       "37                      0                     1                          0   \n",
       "38                      0                     0                          0   \n",
       "39                      0                     1                          0   \n",
       "40                      1                     1                          0   \n",
       "41                      0                     1                          0   \n",
       "42                      0                     1                          0   \n",
       "44                      0                     1                          0   \n",
       "45                      0                     1                          0   \n",
       "46                      0                     0                          0   \n",
       "47                      1                     1                          0   \n",
       "48                      0                     1                          0   \n",
       "49                      0                     0                          0   \n",
       "50                      0                     1                          0   \n",
       "51                      0                     1                          0   \n",
       "52                      0                     1                          0   \n",
       "53                      0                     1                          0   \n",
       "54                      0                     1                          0   \n",
       "55                      0                     1                          0   \n",
       "\n",
       "    education_Master's & above  recruitment_channel_other  \\\n",
       "0                            1                          0   \n",
       "1                            0                          1   \n",
       "2                            0                          0   \n",
       "3                            0                          1   \n",
       "4                            0                          1   \n",
       "5                            0                          0   \n",
       "6                            0                          1   \n",
       "7                            1                          0   \n",
       "8                            0                          1   \n",
       "9                            1                          0   \n",
       "11                           0                          0   \n",
       "12                           0                          0   \n",
       "13                           1                          1   \n",
       "14                           1                          0   \n",
       "15                           0                          1   \n",
       "16                           0                          1   \n",
       "17                           0                          1   \n",
       "18                           0                          1   \n",
       "19                           0                          1   \n",
       "20                           0                          1   \n",
       "22                           0                          1   \n",
       "24                           1                          0   \n",
       "25                           0                          1   \n",
       "26                           0                          1   \n",
       "27                           0                          1   \n",
       "28                           0                          1   \n",
       "30                           0                          0   \n",
       "31                           0                          1   \n",
       "33                           0                          0   \n",
       "34                           0                          1   \n",
       "35                           0                          1   \n",
       "36                           0                          0   \n",
       "37                           0                          0   \n",
       "38                           1                          1   \n",
       "39                           0                          0   \n",
       "40                           0                          1   \n",
       "41                           0                          1   \n",
       "42                           0                          0   \n",
       "44                           0                          1   \n",
       "45                           0                          0   \n",
       "46                           1                          1   \n",
       "47                           0                          0   \n",
       "48                           0                          0   \n",
       "49                           1                          0   \n",
       "50                           0                          0   \n",
       "51                           0                          0   \n",
       "52                           0                          0   \n",
       "53                           0                          1   \n",
       "54                           0                          1   \n",
       "55                           0                          1   \n",
       "\n",
       "    recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "0                              0                             1  \n",
       "1                              0                             0  \n",
       "2                              0                             1  \n",
       "3                              0                             0  \n",
       "4                              0                             0  \n",
       "5                              0                             1  \n",
       "6                              0                             0  \n",
       "7                              0                             1  \n",
       "8                              0                             0  \n",
       "9                              0                             1  \n",
       "11                             0                             1  \n",
       "12                             0                             1  \n",
       "13                             0                             0  \n",
       "14                             0                             1  \n",
       "15                             0                             0  \n",
       "16                             0                             0  \n",
       "17                             0                             0  \n",
       "18                             0                             0  \n",
       "19                             0                             0  \n",
       "20                             0                             0  \n",
       "22                             0                             0  \n",
       "24                             0                             1  \n",
       "25                             0                             0  \n",
       "26                             0                             0  \n",
       "27                             0                             0  \n",
       "28                             0                             0  \n",
       "30                             0                             1  \n",
       "31                             0                             0  \n",
       "33                             0                             1  \n",
       "34                             0                             0  \n",
       "35                             0                             0  \n",
       "36                             0                             1  \n",
       "37                             0                             1  \n",
       "38                             0                             0  \n",
       "39                             0                             1  \n",
       "40                             0                             0  \n",
       "41                             0                             0  \n",
       "42                             0                             1  \n",
       "44                             0                             0  \n",
       "45                             0                             1  \n",
       "46                             0                             0  \n",
       "47                             0                             1  \n",
       "48                             0                             1  \n",
       "49                             0                             1  \n",
       "50                             0                             1  \n",
       "51                             0                             1  \n",
       "52                             0                             1  \n",
       "53                             0                             0  \n",
       "54                             0                             0  \n",
       "55                             0                             0  \n",
       "\n",
       "[50 rows x 29 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head(50)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# split The data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54802</th>\n",
       "      <td>6915</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54803</th>\n",
       "      <td>3030</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>3.0</td>\n",
       "      <td>17</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54804</th>\n",
       "      <td>74592</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_27</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54805</th>\n",
       "      <td>13918</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54807</th>\n",
       "      <td>51526</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>48660 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       employee_id  department     region  education gender  \\\n",
       "0            65438         0.0   region_7        0.0      f   \n",
       "1            65141         0.0  region_22        1.0      m   \n",
       "2             7513         0.0  region_19        1.0      m   \n",
       "3             2542         0.0  region_23        1.0      m   \n",
       "4            48945         0.0  region_26        1.0      m   \n",
       "...            ...         ...        ...        ...    ...   \n",
       "54802         6915         0.0  region_14        1.0      m   \n",
       "54803         3030         0.0  region_14        1.0      m   \n",
       "54804        74592         0.0  region_27        0.0      f   \n",
       "54805        13918         1.0   region_1        1.0      m   \n",
       "54807        51526         0.0  region_22        1.0      m   \n",
       "\n",
       "       recruitment_channel  no_of_trainings  age  previous_year_rating  \\\n",
       "0                      0.0                1   35                   5.0   \n",
       "1                      1.0                1   30                   5.0   \n",
       "2                      0.0                1   34                   3.0   \n",
       "3                      1.0                2   39                   1.0   \n",
       "4                      1.0                1   45                   3.0   \n",
       "...                    ...              ...  ...                   ...   \n",
       "54802                  1.0                2   31                   1.0   \n",
       "54803                  0.0                1   48                   3.0   \n",
       "54804                  1.0                1   37                   2.0   \n",
       "54805                  1.0                1   27                   5.0   \n",
       "54807                  1.0                1   27                   1.0   \n",
       "\n",
       "       length_of_service  ...  department_Procurement  department_R&D  \\\n",
       "0                      8  ...                       0               0   \n",
       "1                      4  ...                       0               0   \n",
       "2                      7  ...                       0               0   \n",
       "3                     10  ...                       0               0   \n",
       "4                      2  ...                       0               0   \n",
       "...                  ...  ...                     ...             ...   \n",
       "54802                  2  ...                       0               0   \n",
       "54803                 17  ...                       0               0   \n",
       "54804                  6  ...                       0               0   \n",
       "54805                  3  ...                       0               0   \n",
       "54807                  5  ...                       0               0   \n",
       "\n",
       "       department_Sales & Marketing  department_Technology  \\\n",
       "0                                 1                      0   \n",
       "1                                 0                      0   \n",
       "2                                 1                      0   \n",
       "3                                 1                      0   \n",
       "4                                 0                      1   \n",
       "...                             ...                    ...   \n",
       "54802                             1                      0   \n",
       "54803                             0                      1   \n",
       "54804                             0                      0   \n",
       "54805                             0                      0   \n",
       "54807                             0                      0   \n",
       "\n",
       "       education_Bachelor's  education_Below Secondary  \\\n",
       "0                         0                          0   \n",
       "1                         1                          0   \n",
       "2                         1                          0   \n",
       "3                         1                          0   \n",
       "4                         1                          0   \n",
       "...                     ...                        ...   \n",
       "54802                     1                          0   \n",
       "54803                     1                          0   \n",
       "54804                     0                          0   \n",
       "54805                     1                          0   \n",
       "54807                     1                          0   \n",
       "\n",
       "       education_Master's & above  recruitment_channel_other  \\\n",
       "0                               1                          0   \n",
       "1                               0                          1   \n",
       "2                               0                          0   \n",
       "3                               0                          1   \n",
       "4                               0                          1   \n",
       "...                           ...                        ...   \n",
       "54802                           0                          1   \n",
       "54803                           0                          0   \n",
       "54804                           1                          1   \n",
       "54805                           0                          1   \n",
       "54807                           0                          1   \n",
       "\n",
       "       recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "0                                 0                             1  \n",
       "1                                 0                             0  \n",
       "2                                 0                             1  \n",
       "3                                 0                             0  \n",
       "4                                 0                             0  \n",
       "...                             ...                           ...  \n",
       "54802                             0                             0  \n",
       "54803                             0                             1  \n",
       "54804                             0                             0  \n",
       "54805                             0                             0  \n",
       "54807                             0                             0  \n",
       "\n",
       "[48660 rows x 29 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {},
   "outputs": [],
   "source": [
    "s0 = df.is_promoted[df.is_promoted.eq(0)].sample(200).index\n",
    "s1 = df.is_promoted[df.is_promoted.eq(1)].sample(200).index "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [],
   "source": [
    "test = df.loc[s0.union(s1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>71</th>\n",
       "      <td>24040</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>351</th>\n",
       "      <td>8284</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>502</th>\n",
       "      <td>44381</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>557</th>\n",
       "      <td>28850</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_14</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>36</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>618</th>\n",
       "      <td>9072</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_14</td>\n",
       "      <td>0.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54259</th>\n",
       "      <td>7652</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>3.0</td>\n",
       "      <td>16</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54310</th>\n",
       "      <td>31961</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_29</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>29</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54404</th>\n",
       "      <td>40891</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54649</th>\n",
       "      <td>26846</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54758</th>\n",
       "      <td>15084</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>43</td>\n",
       "      <td>4.0</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>400 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       employee_id  department     region  education gender  \\\n",
       "71           24040         0.0  region_22        0.0      m   \n",
       "351           8284         0.0   region_2        1.0      f   \n",
       "502          44381         0.0   region_7        1.0      m   \n",
       "557          28850         0.0  region_14        0.0      f   \n",
       "618           9072         0.0  region_14        0.0      m   \n",
       "...            ...         ...        ...        ...    ...   \n",
       "54259         7652         0.0   region_2        1.0      f   \n",
       "54310        31961         1.0  region_29        1.0      m   \n",
       "54404        40891         0.0  region_27        1.0      m   \n",
       "54649        26846         0.0  region_22        1.0      m   \n",
       "54758        15084         0.0   region_2        1.0      f   \n",
       "\n",
       "       recruitment_channel  no_of_trainings  age  previous_year_rating  \\\n",
       "71                     1.0                1   34                   5.0   \n",
       "351                    1.0                1   34                   5.0   \n",
       "502                    1.0                1   33                   5.0   \n",
       "557                    0.0                1   36                   3.0   \n",
       "618                    1.0                1   33                   4.0   \n",
       "...                    ...              ...  ...                   ...   \n",
       "54259                  1.0                1   48                   3.0   \n",
       "54310                  0.0                1   29                   3.0   \n",
       "54404                  0.0                1   30                   5.0   \n",
       "54649                  1.0                1   27                   3.0   \n",
       "54758                  0.0                1   43                   4.0   \n",
       "\n",
       "       length_of_service  ...  department_Procurement  department_R&D  \\\n",
       "71                     7  ...                       0               0   \n",
       "351                    5  ...                       0               0   \n",
       "502                    8  ...                       0               0   \n",
       "557                    5  ...                       0               0   \n",
       "618                    3  ...                       0               0   \n",
       "...                  ...  ...                     ...             ...   \n",
       "54259                 16  ...                       1               0   \n",
       "54310                  2  ...                       0               0   \n",
       "54404                  5  ...                       0               0   \n",
       "54649                  2  ...                       0               0   \n",
       "54758                 13  ...                       0               0   \n",
       "\n",
       "       department_Sales & Marketing  department_Technology  \\\n",
       "71                                0                      1   \n",
       "351                               1                      0   \n",
       "502                               0                      0   \n",
       "557                               0                      1   \n",
       "618                               1                      0   \n",
       "...                             ...                    ...   \n",
       "54259                             0                      0   \n",
       "54310                             0                      0   \n",
       "54404                             1                      0   \n",
       "54649                             0                      1   \n",
       "54758                             0                      0   \n",
       "\n",
       "       education_Bachelor's  education_Below Secondary  \\\n",
       "71                        0                          0   \n",
       "351                       1                          0   \n",
       "502                       1                          0   \n",
       "557                       0                          0   \n",
       "618                       0                          0   \n",
       "...                     ...                        ...   \n",
       "54259                     1                          0   \n",
       "54310                     1                          0   \n",
       "54404                     1                          0   \n",
       "54649                     1                          0   \n",
       "54758                     1                          0   \n",
       "\n",
       "       education_Master's & above  recruitment_channel_other  \\\n",
       "71                              1                          1   \n",
       "351                             0                          1   \n",
       "502                             0                          1   \n",
       "557                             1                          0   \n",
       "618                             1                          1   \n",
       "...                           ...                        ...   \n",
       "54259                           0                          1   \n",
       "54310                           0                          0   \n",
       "54404                           0                          0   \n",
       "54649                           0                          1   \n",
       "54758                           0                          0   \n",
       "\n",
       "       recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "71                                0                             0  \n",
       "351                               0                             0  \n",
       "502                               0                             0  \n",
       "557                               0                             1  \n",
       "618                               0                             0  \n",
       "...                             ...                           ...  \n",
       "54259                             0                             0  \n",
       "54310                             0                             1  \n",
       "54404                             0                             1  \n",
       "54649                             0                             0  \n",
       "54758                             0                             1  \n",
       "\n",
       "[400 rows x 29 columns]"
      ]
     },
     "execution_count": 268,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.concat([df, test, test]).drop_duplicates(keep=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54802</th>\n",
       "      <td>6915</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54803</th>\n",
       "      <td>3030</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>3.0</td>\n",
       "      <td>17</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54804</th>\n",
       "      <td>74592</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_27</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54805</th>\n",
       "      <td>13918</td>\n",
       "      <td>1.0</td>\n",
       "      <td>region_1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54807</th>\n",
       "      <td>51526</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>48260 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       employee_id  department     region  education gender  \\\n",
       "0            65438         0.0   region_7        0.0      f   \n",
       "1            65141         0.0  region_22        1.0      m   \n",
       "2             7513         0.0  region_19        1.0      m   \n",
       "3             2542         0.0  region_23        1.0      m   \n",
       "4            48945         0.0  region_26        1.0      m   \n",
       "...            ...         ...        ...        ...    ...   \n",
       "54802         6915         0.0  region_14        1.0      m   \n",
       "54803         3030         0.0  region_14        1.0      m   \n",
       "54804        74592         0.0  region_27        0.0      f   \n",
       "54805        13918         1.0   region_1        1.0      m   \n",
       "54807        51526         0.0  region_22        1.0      m   \n",
       "\n",
       "       recruitment_channel  no_of_trainings  age  previous_year_rating  \\\n",
       "0                      0.0                1   35                   5.0   \n",
       "1                      1.0                1   30                   5.0   \n",
       "2                      0.0                1   34                   3.0   \n",
       "3                      1.0                2   39                   1.0   \n",
       "4                      1.0                1   45                   3.0   \n",
       "...                    ...              ...  ...                   ...   \n",
       "54802                  1.0                2   31                   1.0   \n",
       "54803                  0.0                1   48                   3.0   \n",
       "54804                  1.0                1   37                   2.0   \n",
       "54805                  1.0                1   27                   5.0   \n",
       "54807                  1.0                1   27                   1.0   \n",
       "\n",
       "       length_of_service  ...  department_Procurement  department_R&D  \\\n",
       "0                      8  ...                       0               0   \n",
       "1                      4  ...                       0               0   \n",
       "2                      7  ...                       0               0   \n",
       "3                     10  ...                       0               0   \n",
       "4                      2  ...                       0               0   \n",
       "...                  ...  ...                     ...             ...   \n",
       "54802                  2  ...                       0               0   \n",
       "54803                 17  ...                       0               0   \n",
       "54804                  6  ...                       0               0   \n",
       "54805                  3  ...                       0               0   \n",
       "54807                  5  ...                       0               0   \n",
       "\n",
       "       department_Sales & Marketing  department_Technology  \\\n",
       "0                                 1                      0   \n",
       "1                                 0                      0   \n",
       "2                                 1                      0   \n",
       "3                                 1                      0   \n",
       "4                                 0                      1   \n",
       "...                             ...                    ...   \n",
       "54802                             1                      0   \n",
       "54803                             0                      1   \n",
       "54804                             0                      0   \n",
       "54805                             0                      0   \n",
       "54807                             0                      0   \n",
       "\n",
       "       education_Bachelor's  education_Below Secondary  \\\n",
       "0                         0                          0   \n",
       "1                         1                          0   \n",
       "2                         1                          0   \n",
       "3                         1                          0   \n",
       "4                         1                          0   \n",
       "...                     ...                        ...   \n",
       "54802                     1                          0   \n",
       "54803                     1                          0   \n",
       "54804                     0                          0   \n",
       "54805                     1                          0   \n",
       "54807                     1                          0   \n",
       "\n",
       "       education_Master's & above  recruitment_channel_other  \\\n",
       "0                               1                          0   \n",
       "1                               0                          1   \n",
       "2                               0                          0   \n",
       "3                               0                          1   \n",
       "4                               0                          1   \n",
       "...                           ...                        ...   \n",
       "54802                           0                          1   \n",
       "54803                           0                          0   \n",
       "54804                           1                          1   \n",
       "54805                           0                          1   \n",
       "54807                           0                          1   \n",
       "\n",
       "       recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "0                                 0                             1  \n",
       "1                                 0                             0  \n",
       "2                                 0                             1  \n",
       "3                                 0                             0  \n",
       "4                                 0                             0  \n",
       "...                             ...                           ...  \n",
       "54802                             0                             0  \n",
       "54803                             0                             1  \n",
       "54804                             0                             0  \n",
       "54805                             0                             0  \n",
       "54807                             0                             0  \n",
       "\n",
       "[48260 rows x 29 columns]"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    44228\n",
       "1     4032\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.is_promoted.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    200\n",
       "0    200\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.is_promoted.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9732.0"
      ]
     },
     "execution_count": 203,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "48660 * 20/100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3892.8"
      ]
     },
     "execution_count": 204,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "9732 * 40/100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5839.2"
      ]
     },
     "execution_count": 205,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "9732 * 60/100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    44428\n",
       "1     4232\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.is_promoted.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# select data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(48660, 24)\n",
      "(48660,)\n"
     ]
    }
   ],
   "source": [
    "X = data.iloc[:,[2,4,6,7,8,9,10,11,12,14,15,16,17,18,19,20,21,22,23, 24, 25, 26,27,28]]\n",
    "Y = data.iloc[:,13]\n",
    "print(X.shape) \n",
    "print(Y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(48260, 24)\n",
      "(48260,)\n"
     ]
    }
   ],
   "source": [
    "X_train = train.iloc[:,[2,4,6,7,8,9,10,11,12,14,15,16,17,18,19,20,21,22,23, 24, 25, 26,27,28]]\n",
    "Y_train = train.iloc[:,13]\n",
    "print(X_train.shape) \n",
    "print(Y_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(400, 24)\n",
      "(400,)\n"
     ]
    }
   ],
   "source": [
    "X_test = test.iloc[:,[2,4,6,7,8,9,10,11,12,14,15,16,17,18,19,20,21,22,23, 24, 25, 26,27,28]]\n",
    "Y_test = test.iloc[:,13]\n",
    "print(X_test.shape) \n",
    "print(Y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [],
   "source": [
    "#X = data.iloc[:,[1,2,3,4,5,6,7,8,9,10,11,12]]\n",
    "#Y = data.iloc[:,-1]\n",
    "#print(X.shape)\n",
    "#print(Y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['employee_id', 'department', 'region', 'education', 'gender',\n",
       "       'recruitment_channel', 'no_of_trainings', 'age', 'previous_year_rating',\n",
       "       'length_of_service', 'KPIs_met >80%', 'awards_won?',\n",
       "       'avg_training_score', 'is_promoted', 'department_Analytics',\n",
       "       'department_Finance', 'department_HR', 'department_Legal',\n",
       "       'department_Operations', 'department_Procurement', 'department_R&D',\n",
       "       'department_Sales & Marketing', 'department_Technology',\n",
       "       'education_Bachelor's', 'education_Below Secondary',\n",
       "       'education_Master's & above', 'recruitment_channel_other',\n",
       "       'recruitment_channel_referred', 'recruitment_channel_sourcing'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 211,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "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>employee_id</th>\n",
       "      <th>department</th>\n",
       "      <th>region</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>recruitment_channel</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65438</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>65141</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_19</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2542</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_23</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>48945</td>\n",
       "      <td>0.0</td>\n",
       "      <td>region_26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>m</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   employee_id  department     region  education gender  recruitment_channel  \\\n",
       "0        65438         0.0   region_7        0.0      f                  0.0   \n",
       "1        65141         0.0  region_22        1.0      m                  1.0   \n",
       "2         7513         0.0  region_19        1.0      m                  0.0   \n",
       "3         2542         0.0  region_23        1.0      m                  1.0   \n",
       "4        48945         0.0  region_26        1.0      m                  1.0   \n",
       "\n",
       "   no_of_trainings  age  previous_year_rating  length_of_service  ...  \\\n",
       "0                1   35                   5.0                  8  ...   \n",
       "1                1   30                   5.0                  4  ...   \n",
       "2                1   34                   3.0                  7  ...   \n",
       "3                2   39                   1.0                 10  ...   \n",
       "4                1   45                   3.0                  2  ...   \n",
       "\n",
       "   department_Procurement  department_R&D  department_Sales & Marketing  \\\n",
       "0                       0               0                             1   \n",
       "1                       0               0                             0   \n",
       "2                       0               0                             1   \n",
       "3                       0               0                             1   \n",
       "4                       0               0                             0   \n",
       "\n",
       "   department_Technology  education_Bachelor's  education_Below Secondary  \\\n",
       "0                      0                     0                          0   \n",
       "1                      0                     1                          0   \n",
       "2                      0                     1                          0   \n",
       "3                      0                     1                          0   \n",
       "4                      1                     1                          0   \n",
       "\n",
       "   education_Master's & above  recruitment_channel_other  \\\n",
       "0                           1                          0   \n",
       "1                           0                          1   \n",
       "2                           0                          0   \n",
       "3                           0                          1   \n",
       "4                           0                          1   \n",
       "\n",
       "   recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "0                             0                             1  \n",
       "1                             0                             0  \n",
       "2                             0                             1  \n",
       "3                             0                             0  \n",
       "4                             0                             0  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 212,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {},
   "outputs": [],
   "source": [
    "#label encoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "# encoder lables \n",
    "enc= LabelEncoder()\n",
    "X.iloc[:,[0,1]]= \\\n",
    "X.iloc[:,[0,1]].apply(enc.fit_transform)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "# encoder lables \n",
    "enc= LabelEncoder()\n",
    "X_train.iloc[:,[0,1]]= \\\n",
    "X_train.iloc[:,[0,1]].apply(enc.fit_transform)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "# encoder lables \n",
    "enc= LabelEncoder()\n",
    "X_test.iloc[:,[0,1]]= \\\n",
    "X_test.iloc[:,[0,1]].apply(enc.fit_transform)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "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>region</th>\n",
       "      <th>gender</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>KPIs_met &gt;80%</th>\n",
       "      <th>awards_won?</th>\n",
       "      <th>avg_training_score</th>\n",
       "      <th>department_Analytics</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>85</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>31</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>27</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>63</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>83</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>32</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>28</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>49</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>39</td>\n",
       "      <td>3.0</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>77</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>24</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>51</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>24</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>3.0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>75</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    region  gender  no_of_trainings  age  previous_year_rating  \\\n",
       "0       31       0                1   35                   5.0   \n",
       "1       14       1                1   30                   5.0   \n",
       "2       10       1                1   34                   3.0   \n",
       "3       15       1                2   39                   1.0   \n",
       "4       18       1                1   45                   3.0   \n",
       "5       11       1                2   31                   3.0   \n",
       "6       12       0                1   31                   3.0   \n",
       "7       27       1                1   33                   3.0   \n",
       "8       12       1                1   28                   4.0   \n",
       "9        0       1                1   32                   5.0   \n",
       "11      31       0                1   35                   5.0   \n",
       "12      28       1                1   49                   5.0   \n",
       "13      21       1                2   39                   3.0   \n",
       "14      11       1                1   37                   3.0   \n",
       "15      31       1                1   37                   1.0   \n",
       "16      11       1                1   38                   3.0   \n",
       "17      24       1                1   34                   1.0   \n",
       "18      24       1                1   34                   5.0   \n",
       "19       6       1                1   37                   3.0   \n",
       "20       5       1                1   35                   3.0   \n",
       "\n",
       "    length_of_service  KPIs_met >80%  awards_won?  avg_training_score  \\\n",
       "0                   8              1            0                  49   \n",
       "1                   4              0            0                  60   \n",
       "2                   7              0            0                  50   \n",
       "3                  10              0            0                  50   \n",
       "4                   2              0            0                  73   \n",
       "5                   7              0            0                  85   \n",
       "6                   5              0            0                  59   \n",
       "7                   6              0            0                  63   \n",
       "8                   5              0            0                  83   \n",
       "9                   5              1            0                  54   \n",
       "11                  3              1            0                  50   \n",
       "12                  5              1            0                  49   \n",
       "13                 16              0            0                  80   \n",
       "14                  7              0            0                  84   \n",
       "15                 10              0            0                  60   \n",
       "16                  5              1            0                  77   \n",
       "17                  4              0            0                  51   \n",
       "18                  8              1            0                  46   \n",
       "19                  9              0            0                  59   \n",
       "20                  7              0            0                  75   \n",
       "\n",
       "    department_Analytics  ...  department_Procurement  department_R&D  \\\n",
       "0                      0  ...                       0               0   \n",
       "1                      0  ...                       0               0   \n",
       "2                      0  ...                       0               0   \n",
       "3                      0  ...                       0               0   \n",
       "4                      0  ...                       0               0   \n",
       "5                      1  ...                       0               0   \n",
       "6                      0  ...                       0               0   \n",
       "7                      0  ...                       0               0   \n",
       "8                      1  ...                       0               0   \n",
       "9                      0  ...                       0               0   \n",
       "11                     0  ...                       0               0   \n",
       "12                     0  ...                       0               0   \n",
       "13                     0  ...                       0               0   \n",
       "14                     0  ...                       0               1   \n",
       "15                     0  ...                       0               0   \n",
       "16                     0  ...                       0               0   \n",
       "17                     0  ...                       0               0   \n",
       "18                     0  ...                       0               0   \n",
       "19                     0  ...                       0               0   \n",
       "20                     0  ...                       1               0   \n",
       "\n",
       "    department_Sales & Marketing  department_Technology  education_Bachelor's  \\\n",
       "0                              1                      0                     0   \n",
       "1                              0                      0                     1   \n",
       "2                              1                      0                     1   \n",
       "3                              1                      0                     1   \n",
       "4                              0                      1                     1   \n",
       "5                              0                      0                     1   \n",
       "6                              0                      0                     1   \n",
       "7                              0                      0                     0   \n",
       "8                              0                      0                     1   \n",
       "9                              1                      0                     0   \n",
       "11                             1                      0                     1   \n",
       "12                             1                      0                     1   \n",
       "13                             0                      1                     0   \n",
       "14                             0                      0                     0   \n",
       "15                             0                      0                     1   \n",
       "16                             0                      1                     1   \n",
       "17                             1                      0                     1   \n",
       "18                             1                      0                     1   \n",
       "19                             0                      0                     1   \n",
       "20                             0                      0                     1   \n",
       "\n",
       "    education_Below Secondary  education_Master's & above  \\\n",
       "0                           0                           1   \n",
       "1                           0                           0   \n",
       "2                           0                           0   \n",
       "3                           0                           0   \n",
       "4                           0                           0   \n",
       "5                           0                           0   \n",
       "6                           0                           0   \n",
       "7                           0                           1   \n",
       "8                           0                           0   \n",
       "9                           0                           1   \n",
       "11                          0                           0   \n",
       "12                          0                           0   \n",
       "13                          0                           1   \n",
       "14                          0                           1   \n",
       "15                          0                           0   \n",
       "16                          0                           0   \n",
       "17                          0                           0   \n",
       "18                          0                           0   \n",
       "19                          0                           0   \n",
       "20                          0                           0   \n",
       "\n",
       "    recruitment_channel_other  recruitment_channel_referred  \\\n",
       "0                           0                             0   \n",
       "1                           1                             0   \n",
       "2                           0                             0   \n",
       "3                           1                             0   \n",
       "4                           1                             0   \n",
       "5                           0                             0   \n",
       "6                           1                             0   \n",
       "7                           0                             0   \n",
       "8                           1                             0   \n",
       "9                           0                             0   \n",
       "11                          0                             0   \n",
       "12                          0                             0   \n",
       "13                          1                             0   \n",
       "14                          0                             0   \n",
       "15                          1                             0   \n",
       "16                          1                             0   \n",
       "17                          1                             0   \n",
       "18                          1                             0   \n",
       "19                          1                             0   \n",
       "20                          1                             0   \n",
       "\n",
       "    recruitment_channel_sourcing  \n",
       "0                              1  \n",
       "1                              0  \n",
       "2                              1  \n",
       "3                              0  \n",
       "4                              0  \n",
       "5                              1  \n",
       "6                              0  \n",
       "7                              1  \n",
       "8                              0  \n",
       "9                              1  \n",
       "11                             1  \n",
       "12                             1  \n",
       "13                             0  \n",
       "14                             1  \n",
       "15                             0  \n",
       "16                             0  \n",
       "17                             0  \n",
       "18                             0  \n",
       "19                             0  \n",
       "20                             0  \n",
       "\n",
       "[20 rows x 24 columns]"
      ]
     },
     "execution_count": 279,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['region', 'gender', 'no_of_trainings', 'age', 'previous_year_rating',\n",
       "       'length_of_service', 'KPIs_met >80%', 'awards_won?',\n",
       "       'avg_training_score', 'department_Analytics', 'department_Finance',\n",
       "       'department_HR', 'department_Legal', 'department_Operations',\n",
       "       'department_Procurement', 'department_R&D',\n",
       "       'department_Sales & Marketing', 'department_Technology',\n",
       "       'education_Bachelor's', 'education_Below Secondary',\n",
       "       'education_Master's & above', 'recruitment_channel_other',\n",
       "       'recruitment_channel_referred', 'recruitment_channel_sourcing'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 280,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "metadata": {},
   "outputs": [],
   "source": [
    "Col_name = ['no_of_trainings','age' , 'previous_year_rating' , 'length_of_service' , 'avg_training_score' ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "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>region</th>\n",
       "      <th>gender</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>KPIs_met &gt;80%</th>\n",
       "      <th>awards_won?</th>\n",
       "      <th>avg_training_score</th>\n",
       "      <th>department_Analytics</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>23</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>26</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>61</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>51</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>51</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>3.0</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>51</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>266</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>41</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>48</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>335</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>32</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>476</th>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>3.0</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>87</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>605</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>4.0</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>48</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>762</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>49</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>834</th>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>69</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1075</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>49</td>\n",
       "      <td>4.0</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1080</th>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>5.0</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1217</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>36</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>67</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1767</th>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>37</td>\n",
       "      <td>4.0</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>92</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1773</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>29</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1774</th>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>43</td>\n",
       "      <td>3.0</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>55</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2336</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>35</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>93</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2371</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>84</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2589</th>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>96</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2669</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>44</td>\n",
       "      <td>5.0</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>49</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      region  gender  no_of_trainings  age  previous_year_rating  \\\n",
       "31        23       1                2   26                   1.0   \n",
       "39        19       1                1   33                   5.0   \n",
       "121       16       1                1   34                   5.0   \n",
       "146       13       1                1   46                   3.0   \n",
       "266       10       0                1   41                   4.0   \n",
       "335        3       0                1   32                   5.0   \n",
       "476        8       1                1   45                   3.0   \n",
       "605       10       1                1   47                   4.0   \n",
       "762       10       0                1   49                   5.0   \n",
       "834       11       0                1   33                   2.0   \n",
       "1075       4       0                1   49                   4.0   \n",
       "1080      18       1                2   31                   5.0   \n",
       "1217       4       0                1   36                   3.0   \n",
       "1767      21       1                1   37                   4.0   \n",
       "1773       4       1                1   29                   4.0   \n",
       "1774      19       1                1   43                   3.0   \n",
       "2336       4       1                2   35                   3.0   \n",
       "2371       4       1                2   29                   4.0   \n",
       "2589      13       1                1   27                   3.0   \n",
       "2669      10       0                1   44                   5.0   \n",
       "\n",
       "      length_of_service  KPIs_met >80%  awards_won?  avg_training_score  \\\n",
       "31                    2              0            0                  61   \n",
       "39                    6              1            0                  51   \n",
       "121                   7              1            0                  51   \n",
       "146                  11              1            0                  51   \n",
       "266                   5              1            0                  48   \n",
       "335                   5              0            1                  84   \n",
       "476                  16              0            0                  87   \n",
       "605                  16              1            0                  48   \n",
       "762                   7              1            0                  73   \n",
       "834                   3              1            0                  69   \n",
       "1075                 18              0            0                  60   \n",
       "1080                  9              1            0                  59   \n",
       "1217                  6              1            0                  67   \n",
       "1767                 10              1            0                  92   \n",
       "1773                  2              1            0                  50   \n",
       "1774                 11              1            0                  55   \n",
       "2336                  3              0            0                  93   \n",
       "2371                  2              0            0                  84   \n",
       "2589                  2              1            1                  96   \n",
       "2669                 12              1            0                  49   \n",
       "\n",
       "      department_Analytics  ...  department_Procurement  department_R&D  \\\n",
       "31                       0  ...                       0               0   \n",
       "39                       0  ...                       0               0   \n",
       "121                      0  ...                       0               0   \n",
       "146                      0  ...                       0               0   \n",
       "266                      0  ...                       0               0   \n",
       "335                      0  ...                       1               0   \n",
       "476                      0  ...                       1               0   \n",
       "605                      0  ...                       0               0   \n",
       "762                      0  ...                       1               0   \n",
       "834                      0  ...                       1               0   \n",
       "1075                     0  ...                       0               0   \n",
       "1080                     0  ...                       0               0   \n",
       "1217                     0  ...                       1               0   \n",
       "1767                     0  ...                       0               0   \n",
       "1773                     0  ...                       0               0   \n",
       "1774                     0  ...                       0               0   \n",
       "2336                     0  ...                       1               0   \n",
       "2371                     1  ...                       0               0   \n",
       "2589                     0  ...                       0               0   \n",
       "2669                     0  ...                       0               0   \n",
       "\n",
       "      department_Sales & Marketing  department_Technology  \\\n",
       "31                               0                      0   \n",
       "39                               1                      0   \n",
       "121                              1                      0   \n",
       "146                              1                      0   \n",
       "266                              1                      0   \n",
       "335                              0                      0   \n",
       "476                              0                      0   \n",
       "605                              1                      0   \n",
       "762                              0                      0   \n",
       "834                              0                      0   \n",
       "1075                             0                      0   \n",
       "1080                             0                      0   \n",
       "1217                             0                      0   \n",
       "1767                             1                      0   \n",
       "1773                             0                      0   \n",
       "1774                             1                      0   \n",
       "2336                             0                      0   \n",
       "2371                             0                      0   \n",
       "2589                             1                      0   \n",
       "2669                             0                      0   \n",
       "\n",
       "      education_Bachelor's  education_Below Secondary  \\\n",
       "31                       1                          0   \n",
       "39                       1                          0   \n",
       "121                      0                          0   \n",
       "146                      0                          0   \n",
       "266                      0                          0   \n",
       "335                      1                          0   \n",
       "476                      0                          0   \n",
       "605                      1                          0   \n",
       "762                      0                          0   \n",
       "834                      1                          0   \n",
       "1075                     0                          0   \n",
       "1080                     1                          0   \n",
       "1217                     0                          0   \n",
       "1767                     0                          0   \n",
       "1773                     1                          0   \n",
       "1774                     0                          0   \n",
       "2336                     1                          0   \n",
       "2371                     1                          0   \n",
       "2589                     1                          0   \n",
       "2669                     1                          0   \n",
       "\n",
       "      education_Master's & above  recruitment_channel_other  \\\n",
       "31                             0                          1   \n",
       "39                             0                          0   \n",
       "121                            1                          1   \n",
       "146                            1                          0   \n",
       "266                            1                          1   \n",
       "335                            0                          1   \n",
       "476                            1                          1   \n",
       "605                            0                          1   \n",
       "762                            1                          1   \n",
       "834                            0                          1   \n",
       "1075                           1                          1   \n",
       "1080                           0                          0   \n",
       "1217                           1                          1   \n",
       "1767                           1                          0   \n",
       "1773                           0                          1   \n",
       "1774                           1                          1   \n",
       "2336                           0                          1   \n",
       "2371                           0                          1   \n",
       "2589                           0                          1   \n",
       "2669                           0                          0   \n",
       "\n",
       "      recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "31                               0                             0  \n",
       "39                               0                             1  \n",
       "121                              0                             0  \n",
       "146                              0                             1  \n",
       "266                              0                             0  \n",
       "335                              0                             0  \n",
       "476                              0                             0  \n",
       "605                              0                             0  \n",
       "762                              0                             0  \n",
       "834                              0                             0  \n",
       "1075                             0                             0  \n",
       "1080                             0                             1  \n",
       "1217                             0                             0  \n",
       "1767                             0                             1  \n",
       "1773                             0                             0  \n",
       "1774                             0                             0  \n",
       "2336                             0                             0  \n",
       "2371                             0                             0  \n",
       "2589                             0                             0  \n",
       "2669                             1                             0  \n",
       "\n",
       "[20 rows x 24 columns]"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exp1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### split data train , test "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from sklearn.model_selection import train_test_split\n",
    "#X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.3 , random_state=0 , stratify = Y , shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "metadata": {},
   "outputs": [],
   "source": [
    "#y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [],
   "source": [
    "#y_test.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Balancing Data "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Over sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.over_sampling import SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Implementation over sampling SMOTAT\n",
    "smo = SMOTE()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [],
   "source": [
    "# X and Y after balance\n",
    "x_SM,y_SM = smo.fit_sample(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    44228\n",
       "1     4032\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 227,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    44228\n",
       "0    44228\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 228,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_SM.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Under sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Implementation under sampling Tomek"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.under_sampling import TomekLinks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [],
   "source": [
    "tl = TomekLinks()\n",
    "X_tl, y_tl = tl.fit_resample(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    44228\n",
       "1     4032\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 232,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    42971\n",
       "1     4032\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 233,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_tl.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Over and Under sampling "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.combine import SMOTETomek "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {},
   "outputs": [],
   "source": [
    "smt = SMOTETomek()\n",
    "x_SMtl,y_SMtl = smt.fit_sample(X_tl,y_tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    42911\n",
       "0    42911\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 236,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_SMtl.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Standard Scaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {},
   "outputs": [],
   "source": [
    "#sca = StandardScaler()\n",
    "#x_SM.iloc[:,[2,3,4,5,8]] = x_SM.iloc[:,[2,3,4,5,8]].apply(sca.fit_transform)\n",
    "\n",
    "\n",
    "#X_tl.iloc[:,[2,3,4,5,8]] = X_tl.iloc[:,[2,3,4,5,8]].apply(sca.fit_transform)\n",
    "\n",
    "#x_SMtl.iloc[:,[2,3,4,5,8]] = x_SMtl.iloc[:,[2,3,4,5,8]].apply(sca.fit_transform)\n",
    "\n",
    "#X_test.iloc[:,[2,3,4,5,8]] = X_test.iloc[:,[2,3,4,5,8]].apply(sca.fit_transform)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['no_of_trainings',\n",
       " 'age',\n",
       " 'previous_year_rating',\n",
       " 'length_of_service',\n",
       " 'avg_training_score']"
      ]
     },
     "execution_count": 239,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Col_name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {},
   "outputs": [],
   "source": [
    "sca = StandardScaler()\n",
    "\n",
    "scale_SM = x_SM[Col_name]\n",
    "x_SM[Col_name] = sca.fit_transform(scale_SM.values)\n",
    "\n",
    "scale_tl = X_tl[Col_name]\n",
    "X_tl[Col_name]= sca.transform(scale_tl.values)\n",
    "#X = sca.transform(X)\n",
    "\n",
    "scale_SMtl = x_SMtl[Col_name]\n",
    "x_SMtl[Col_name] = sca.fit_transform(scale_SMtl.values)\n",
    "\n",
    "\n",
    "\n",
    "scale_test = X_test[Col_name]\n",
    "X_test[Col_name] = sca.fit_transform(scale_test.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "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>region</th>\n",
       "      <th>gender</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>KPIs_met &gt;80%</th>\n",
       "      <th>awards_won?</th>\n",
       "      <th>avg_training_score</th>\n",
       "      <th>department_Analytics</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>23</td>\n",
       "      <td>1</td>\n",
       "      <td>1.416426</td>\n",
       "      <td>-1.334237</td>\n",
       "      <td>-2.251463</td>\n",
       "      <td>-1.039088</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.431369</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>-0.326179</td>\n",
       "      <td>1.099552</td>\n",
       "      <td>-0.044745</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.116625</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>-0.182170</td>\n",
       "      <td>1.099552</td>\n",
       "      <td>0.203840</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.116625</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>1.545929</td>\n",
       "      <td>-0.575956</td>\n",
       "      <td>1.198182</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.116625</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>266</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>0.825887</td>\n",
       "      <td>0.261798</td>\n",
       "      <td>-0.293331</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.322202</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54305</th>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>-0.902212</td>\n",
       "      <td>-2.251463</td>\n",
       "      <td>-0.044745</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.007669</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54315</th>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>0.249854</td>\n",
       "      <td>-0.575956</td>\n",
       "      <td>0.203840</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.979574</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54452</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>-1.478245</td>\n",
       "      <td>-2.251463</td>\n",
       "      <td>-1.039088</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.911048</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54529</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>0.969896</td>\n",
       "      <td>0.261798</td>\n",
       "      <td>-0.541916</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.390939</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54722</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.411220</td>\n",
       "      <td>-0.182170</td>\n",
       "      <td>1.099552</td>\n",
       "      <td>-1.039088</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.322413</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>400 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       region  gender  no_of_trainings       age  previous_year_rating  \\\n",
       "31         23       1         1.416426 -1.334237             -2.251463   \n",
       "39         19       1        -0.411220 -0.326179              1.099552   \n",
       "121        16       1        -0.411220 -0.182170              1.099552   \n",
       "146        13       1        -0.411220  1.545929             -0.575956   \n",
       "266        10       0        -0.411220  0.825887              0.261798   \n",
       "...       ...     ...              ...       ...                   ...   \n",
       "54305      13       1        -0.411220 -0.902212             -2.251463   \n",
       "54315      22       1        -0.411220  0.249854             -0.575956   \n",
       "54452       1       1        -0.411220 -1.478245             -2.251463   \n",
       "54529      10       1        -0.411220  0.969896              0.261798   \n",
       "54722       1       1        -0.411220 -0.182170              1.099552   \n",
       "\n",
       "       length_of_service  KPIs_met >80%  awards_won?  avg_training_score  \\\n",
       "31             -1.039088              0            0           -0.431369   \n",
       "39             -0.044745              1            0           -1.116625   \n",
       "121             0.203840              1            0           -1.116625   \n",
       "146             1.198182              1            0           -1.116625   \n",
       "266            -0.293331              1            0           -1.322202   \n",
       "...                  ...            ...          ...                 ...   \n",
       "54305          -0.044745              0            0            1.007669   \n",
       "54315           0.203840              1            0           -0.979574   \n",
       "54452          -1.039088              0            0           -0.911048   \n",
       "54529          -0.541916              0            0            0.390939   \n",
       "54722          -1.039088              1            0            0.322413   \n",
       "\n",
       "       department_Analytics  ...  department_Procurement  department_R&D  \\\n",
       "31                        0  ...                       0               0   \n",
       "39                        0  ...                       0               0   \n",
       "121                       0  ...                       0               0   \n",
       "146                       0  ...                       0               0   \n",
       "266                       0  ...                       0               0   \n",
       "...                     ...  ...                     ...             ...   \n",
       "54305                     1  ...                       0               0   \n",
       "54315                     0  ...                       0               0   \n",
       "54452                     0  ...                       0               0   \n",
       "54529                     0  ...                       1               0   \n",
       "54722                     0  ...                       1               0   \n",
       "\n",
       "       department_Sales & Marketing  department_Technology  \\\n",
       "31                                0                      0   \n",
       "39                                1                      0   \n",
       "121                               1                      0   \n",
       "146                               1                      0   \n",
       "266                               1                      0   \n",
       "...                             ...                    ...   \n",
       "54305                             0                      0   \n",
       "54315                             1                      0   \n",
       "54452                             1                      0   \n",
       "54529                             0                      0   \n",
       "54722                             0                      0   \n",
       "\n",
       "       education_Bachelor's  education_Below Secondary  \\\n",
       "31                        1                          0   \n",
       "39                        1                          0   \n",
       "121                       0                          0   \n",
       "146                       0                          0   \n",
       "266                       0                          0   \n",
       "...                     ...                        ...   \n",
       "54305                     1                          0   \n",
       "54315                     0                          0   \n",
       "54452                     1                          0   \n",
       "54529                     0                          0   \n",
       "54722                     0                          0   \n",
       "\n",
       "       education_Master's & above  recruitment_channel_other  \\\n",
       "31                              0                          1   \n",
       "39                              0                          0   \n",
       "121                             1                          1   \n",
       "146                             1                          0   \n",
       "266                             1                          1   \n",
       "...                           ...                        ...   \n",
       "54305                           0                          0   \n",
       "54315                           1                          0   \n",
       "54452                           0                          0   \n",
       "54529                           1                          1   \n",
       "54722                           1                          0   \n",
       "\n",
       "       recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "31                                0                             0  \n",
       "39                                0                             1  \n",
       "121                               0                             0  \n",
       "146                               0                             1  \n",
       "266                               0                             0  \n",
       "...                             ...                           ...  \n",
       "54305                             1                             0  \n",
       "54315                             0                             1  \n",
       "54452                             0                             1  \n",
       "54529                             0                             0  \n",
       "54722                             0                             1  \n",
       "\n",
       "[400 rows x 24 columns]"
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "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>region</th>\n",
       "      <th>gender</th>\n",
       "      <th>no_of_trainings</th>\n",
       "      <th>age</th>\n",
       "      <th>previous_year_rating</th>\n",
       "      <th>length_of_service</th>\n",
       "      <th>KPIs_met &gt;80%</th>\n",
       "      <th>awards_won?</th>\n",
       "      <th>avg_training_score</th>\n",
       "      <th>department_Analytics</th>\n",
       "      <th>...</th>\n",
       "      <th>department_Procurement</th>\n",
       "      <th>department_R&amp;D</th>\n",
       "      <th>department_Sales &amp; Marketing</th>\n",
       "      <th>department_Technology</th>\n",
       "      <th>education_Bachelor's</th>\n",
       "      <th>education_Below Secondary</th>\n",
       "      <th>education_Master's &amp; above</th>\n",
       "      <th>recruitment_channel_other</th>\n",
       "      <th>recruitment_channel_referred</th>\n",
       "      <th>recruitment_channel_sourcing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>-0.022167</td>\n",
       "      <td>1.171059</td>\n",
       "      <td>0.485781</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.242033</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>-0.713965</td>\n",
       "      <td>1.171059</td>\n",
       "      <td>-0.512487</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.475667</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>-0.160527</td>\n",
       "      <td>-0.545872</td>\n",
       "      <td>0.236214</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.172363</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>1.682995</td>\n",
       "      <td>0.531271</td>\n",
       "      <td>-2.262802</td>\n",
       "      <td>0.984915</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1.172363</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>1.361428</td>\n",
       "      <td>-0.545872</td>\n",
       "      <td>-1.011622</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.430039</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85817</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>2.883383</td>\n",
       "      <td>0.251801</td>\n",
       "      <td>2.981451</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.893685</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85818</th>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>-0.575606</td>\n",
       "      <td>-0.545872</td>\n",
       "      <td>-1.011622</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.639048</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85819</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>0.392911</td>\n",
       "      <td>0.595372</td>\n",
       "      <td>0.984915</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.475667</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85820</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>0.392911</td>\n",
       "      <td>1.171059</td>\n",
       "      <td>-0.762055</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.824015</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85821</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.338842</td>\n",
       "      <td>0.669630</td>\n",
       "      <td>-0.545872</td>\n",
       "      <td>-0.262920</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.754345</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>85822 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       region  gender  no_of_trainings       age  previous_year_rating  \\\n",
       "0          31       0        -0.338842 -0.022167              1.171059   \n",
       "1          14       1        -0.338842 -0.713965              1.171059   \n",
       "2          10       1        -0.338842 -0.160527             -0.545872   \n",
       "3          15       1         1.682995  0.531271             -2.262802   \n",
       "4          18       1        -0.338842  1.361428             -0.545872   \n",
       "...       ...     ...              ...       ...                   ...   \n",
       "85817      11       1        -0.338842  2.883383              0.251801   \n",
       "85818      31       1        -0.338842 -0.575606             -0.545872   \n",
       "85819       4       0        -0.338842  0.392911              0.595372   \n",
       "85820      11       1        -0.338842  0.392911              1.171059   \n",
       "85821      10       0        -0.338842  0.669630             -0.545872   \n",
       "\n",
       "       length_of_service  KPIs_met >80%  awards_won?  avg_training_score  \\\n",
       "0               0.485781              1            0           -1.242033   \n",
       "1              -0.512487              0            0           -0.475667   \n",
       "2               0.236214              0            0           -1.172363   \n",
       "3               0.984915              0            0           -1.172363   \n",
       "4              -1.011622              0            0            0.430039   \n",
       "...                  ...            ...          ...                 ...   \n",
       "85817           2.981451              1            0           -0.893685   \n",
       "85818          -1.011622              0            0            0.639048   \n",
       "85819           0.984915              1            0           -0.475667   \n",
       "85820          -0.762055              1            0           -0.824015   \n",
       "85821          -0.262920              1            0           -0.754345   \n",
       "\n",
       "       department_Analytics  ...  department_Procurement  department_R&D  \\\n",
       "0                         0  ...                       0               0   \n",
       "1                         0  ...                       0               0   \n",
       "2                         0  ...                       0               0   \n",
       "3                         0  ...                       0               0   \n",
       "4                         0  ...                       0               0   \n",
       "...                     ...  ...                     ...             ...   \n",
       "85817                     0  ...                       0               0   \n",
       "85818                     0  ...                       0               0   \n",
       "85819                     0  ...                       0               0   \n",
       "85820                     0  ...                       0               0   \n",
       "85821                     0  ...                       0               0   \n",
       "\n",
       "       department_Sales & Marketing  department_Technology  \\\n",
       "0                                 1                      0   \n",
       "1                                 0                      0   \n",
       "2                                 1                      0   \n",
       "3                                 1                      0   \n",
       "4                                 0                      1   \n",
       "...                             ...                    ...   \n",
       "85817                             1                      0   \n",
       "85818                             0                      0   \n",
       "85819                             0                      0   \n",
       "85820                             0                      0   \n",
       "85821                             0                      0   \n",
       "\n",
       "       education_Bachelor's  education_Below Secondary  \\\n",
       "0                         0                          0   \n",
       "1                         1                          0   \n",
       "2                         1                          0   \n",
       "3                         1                          0   \n",
       "4                         1                          0   \n",
       "...                     ...                        ...   \n",
       "85817                     1                          0   \n",
       "85818                     1                          0   \n",
       "85819                     1                          0   \n",
       "85820                     0                          0   \n",
       "85821                     1                          0   \n",
       "\n",
       "       education_Master's & above  recruitment_channel_other  \\\n",
       "0                               1                          0   \n",
       "1                               0                          1   \n",
       "2                               0                          0   \n",
       "3                               0                          1   \n",
       "4                               0                          1   \n",
       "...                           ...                        ...   \n",
       "85817                           0                          0   \n",
       "85818                           0                          0   \n",
       "85819                           0                          1   \n",
       "85820                           0                          0   \n",
       "85821                           0                          1   \n",
       "\n",
       "       recruitment_channel_referred  recruitment_channel_sourcing  \n",
       "0                                 0                             1  \n",
       "1                                 0                             0  \n",
       "2                                 0                             1  \n",
       "3                                 0                             0  \n",
       "4                                 0                             0  \n",
       "...                             ...                           ...  \n",
       "85817                             0                             0  \n",
       "85818                             0                             0  \n",
       "85819                             0                             0  \n",
       "85820                             0                             1  \n",
       "85821                             0                             0  \n",
       "\n",
       "[85822 rows x 24 columns]"
      ]
     },
     "execution_count": 242,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_SMtl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       region  gender  no_of_trainings       age  previous_year_rating  \\\n",
      "0          31       0        -0.339715 -0.021611              1.172350   \n",
      "1          14       1        -0.339715 -0.711912              1.172350   \n",
      "2          10       1        -0.339715 -0.159671             -0.553907   \n",
      "3          15       1         1.689033  0.530631             -2.280164   \n",
      "4          18       1        -0.339715  1.358993             -0.553907   \n",
      "...       ...     ...              ...       ...                   ...   \n",
      "88451      31       1        -0.339715 -0.435792              1.172350   \n",
      "88452      28       1        -0.339715  0.392570             -0.553907   \n",
      "88453       6       1        -0.339715  0.944812              0.373667   \n",
      "88454       6       0        -0.339715 -0.159671             -0.553907   \n",
      "88455       5       0        -0.339715 -0.573852             -0.083514   \n",
      "\n",
      "       length_of_service  KPIs_met >80%  awards_won?  avg_training_score  \\\n",
      "0               0.486129              1            0           -1.246675   \n",
      "1              -0.512873              0            0           -0.479649   \n",
      "2               0.236379              0            0           -1.176945   \n",
      "3               0.985630              0            0           -1.176945   \n",
      "4              -1.012373              0            0            0.426836   \n",
      "...                  ...            ...          ...                 ...   \n",
      "88451          -0.762623              0            0            1.054403   \n",
      "88452           0.236379              0            0           -0.200730   \n",
      "88453           1.235380              1            0            0.984673   \n",
      "88454           0.486129              1            1            1.681970   \n",
      "88455          -0.512873              0            0            1.542510   \n",
      "\n",
      "       department_Analytics  ...  department_Procurement  department_R&D  \\\n",
      "0                         0  ...                       0               0   \n",
      "1                         0  ...                       0               0   \n",
      "2                         0  ...                       0               0   \n",
      "3                         0  ...                       0               0   \n",
      "4                         0  ...                       0               0   \n",
      "...                     ...  ...                     ...             ...   \n",
      "88451                     0  ...                       0               0   \n",
      "88452                     0  ...                       0               0   \n",
      "88453                     0  ...                       0               0   \n",
      "88454                     0  ...                       0               0   \n",
      "88455                     0  ...                       0               0   \n",
      "\n",
      "       department_Sales & Marketing  department_Technology  \\\n",
      "0                                 1                      0   \n",
      "1                                 0                      0   \n",
      "2                                 1                      0   \n",
      "3                                 1                      0   \n",
      "4                                 0                      1   \n",
      "...                             ...                    ...   \n",
      "88451                             0                      0   \n",
      "88452                             0                      0   \n",
      "88453                             0                      1   \n",
      "88454                             0                      0   \n",
      "88455                             0                      1   \n",
      "\n",
      "       education_Bachelor's  education_Below Secondary  \\\n",
      "0                         0                          0   \n",
      "1                         1                          0   \n",
      "2                         1                          0   \n",
      "3                         1                          0   \n",
      "4                         1                          0   \n",
      "...                     ...                        ...   \n",
      "88451                     1                          0   \n",
      "88452                     0                          0   \n",
      "88453                     0                          0   \n",
      "88454                     1                          0   \n",
      "88455                     0                          0   \n",
      "\n",
      "       education_Master's & above  recruitment_channel_other  \\\n",
      "0                               1                          0   \n",
      "1                               0                          1   \n",
      "2                               0                          0   \n",
      "3                               0                          1   \n",
      "4                               0                          1   \n",
      "...                           ...                        ...   \n",
      "88451                           0                          0   \n",
      "88452                           1                          0   \n",
      "88453                           0                          1   \n",
      "88454                           0                          1   \n",
      "88455                           0                          0   \n",
      "\n",
      "       recruitment_channel_referred  recruitment_channel_sourcing  \n",
      "0                                 0                             1  \n",
      "1                                 0                             0  \n",
      "2                                 0                             1  \n",
      "3                                 0                             0  \n",
      "4                                 0                             0  \n",
      "...                             ...                           ...  \n",
      "88451                             0                             0  \n",
      "88452                             0                             1  \n",
      "88453                             0                             0  \n",
      "88454                             0                             0  \n",
      "88455                             0                             0  \n",
      "\n",
      "[88456 rows x 24 columns]\n"
     ]
    }
   ],
   "source": [
    "print(x_SM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       region  gender  no_of_trainings       age  previous_year_rating  \\\n",
      "0          31       0        -0.339715 -0.021611              1.172350   \n",
      "1          14       1        -0.339715 -0.711912              1.172350   \n",
      "2          10       1        -0.339715 -0.159671             -0.553907   \n",
      "3          15       1         1.689033  0.530631             -2.280164   \n",
      "4          18       1        -0.339715  1.358993             -0.553907   \n",
      "...       ...     ...              ...       ...                   ...   \n",
      "46998       5       1         1.689033 -0.573852             -2.280164   \n",
      "46999       5       1        -0.339715  1.773174             -0.553907   \n",
      "47000      19       0        -0.339715  0.254510             -1.417035   \n",
      "47001       0       1        -0.339715 -1.126093              1.172350   \n",
      "47002      14       1        -0.339715 -1.126093             -2.280164   \n",
      "\n",
      "       length_of_service  KPIs_met >80%  awards_won?  avg_training_score  \\\n",
      "0               0.486129              1            0           -1.246675   \n",
      "1              -0.512873              0            0           -0.479649   \n",
      "2               0.236379              0            0           -1.176945   \n",
      "3               0.985630              0            0           -1.176945   \n",
      "4              -1.012373              0            0            0.426836   \n",
      "...                  ...            ...          ...                 ...   \n",
      "46998          -1.012373              0            0           -1.246675   \n",
      "46999           2.733882              0            0            0.775484   \n",
      "47000          -0.013372              0            0           -0.758568   \n",
      "47001          -0.762623              1            0            0.845214   \n",
      "47002          -0.263122              0            0           -1.246675   \n",
      "\n",
      "       department_Analytics  ...  department_Procurement  department_R&D  \\\n",
      "0                         0  ...                       0               0   \n",
      "1                         0  ...                       0               0   \n",
      "2                         0  ...                       0               0   \n",
      "3                         0  ...                       0               0   \n",
      "4                         0  ...                       0               0   \n",
      "...                     ...  ...                     ...             ...   \n",
      "46998                     0  ...                       0               0   \n",
      "46999                     0  ...                       0               0   \n",
      "47000                     0  ...                       0               0   \n",
      "47001                     1  ...                       0               0   \n",
      "47002                     0  ...                       0               0   \n",
      "\n",
      "       department_Sales & Marketing  department_Technology  \\\n",
      "0                                 1                      0   \n",
      "1                                 0                      0   \n",
      "2                                 1                      0   \n",
      "3                                 1                      0   \n",
      "4                                 0                      1   \n",
      "...                             ...                    ...   \n",
      "46998                             1                      0   \n",
      "46999                             0                      1   \n",
      "47000                             0                      0   \n",
      "47001                             0                      0   \n",
      "47002                             0                      0   \n",
      "\n",
      "       education_Bachelor's  education_Below Secondary  \\\n",
      "0                         0                          0   \n",
      "1                         1                          0   \n",
      "2                         1                          0   \n",
      "3                         1                          0   \n",
      "4                         1                          0   \n",
      "...                     ...                        ...   \n",
      "46998                     1                          0   \n",
      "46999                     1                          0   \n",
      "47000                     0                          0   \n",
      "47001                     1                          0   \n",
      "47002                     1                          0   \n",
      "\n",
      "       education_Master's & above  recruitment_channel_other  \\\n",
      "0                               1                          0   \n",
      "1                               0                          1   \n",
      "2                               0                          0   \n",
      "3                               0                          1   \n",
      "4                               0                          1   \n",
      "...                           ...                        ...   \n",
      "46998                           0                          1   \n",
      "46999                           0                          0   \n",
      "47000                           1                          1   \n",
      "47001                           0                          1   \n",
      "47002                           0                          1   \n",
      "\n",
      "       recruitment_channel_referred  recruitment_channel_sourcing  \n",
      "0                                 0                             1  \n",
      "1                                 0                             0  \n",
      "2                                 0                             1  \n",
      "3                                 0                             0  \n",
      "4                                 0                             0  \n",
      "...                             ...                           ...  \n",
      "46998                             0                             0  \n",
      "46999                             0                             1  \n",
      "47000                             0                             0  \n",
      "47001                             0                             0  \n",
      "47002                             0                             0  \n",
      "\n",
      "[47003 rows x 24 columns]\n"
     ]
    }
   ],
   "source": [
    "print(X_tl)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Split the train to train and validation set"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split for Over sampling "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "Xc_train, X_cv, yc_train, y_cv = train_test_split(x_SM, y_SM, test_size=0.2 , random_state=1 , stratify = y_SM )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    35382\n",
       "0    35382\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 246,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yc_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    8846\n",
       "0    8846\n",
       "Name: is_promoted, dtype: int64"
      ]
     },
     "execution_count": 247,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_cv.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(criterion='entropy', n_estimators=150, random_state=0)"
      ]
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "classifierOvRF = RandomForestClassifier(n_estimators = 150 ,criterion = 'entropy' , random_state = 0)\n",
    "classifierOvRF.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9408205344396432\n",
      "0.9410595758797792\n",
      "0.9408214459186542\n",
      "0.9408129509503936\n",
      "0.985896829286153\n"
     ]
    }
   ],
   "source": [
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOvRF ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.685"
      ]
     },
     "execution_count": 254,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = classifierOvRF.predict(X_test)\n",
    "classifierOvRF.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 88.89%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 89.06%\n",
      "\t\t\tRecall Score: 88.68%\n",
      "\t\t\tF1 score: 88.87%\n",
      "\t\t\tAUC score: 88.89%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[31527  3855]\n",
      " [ 4007 31375]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.887     0.891     0.889     35382\n",
      "           1      0.891     0.887     0.889     35382\n",
      "\n",
      "    accuracy                          0.889     70764\n",
      "   macro avg      0.889     0.889     0.889     70764\n",
      "weighted avg      0.889     0.889     0.889     70764\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 68.00%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 68.56%\n",
      "\t\t\tRecall Score: 66.50%\n",
      "\t\t\tF1 score: 67.51%\n",
      "\t\t\tAUC score: 68.00%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[139  61]\n",
      " [ 67 133]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.675     0.695     0.685       200\n",
      "           1      0.686     0.665     0.675       200\n",
      "\n",
      "    accuracy                          0.680       400\n",
      "   macro avg      0.680     0.680     0.680       400\n",
      "weighted avg      0.680     0.680     0.680       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.454     0.490     0.471       200\n",
      "           1      0.446     0.410     0.427       200\n",
      "\n",
      "    accuracy                          0.450       400\n",
      "   macro avg      0.450     0.450     0.449       400\n",
      "weighted avg      0.450     0.450     0.449       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierOvRF, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOvRF, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Desicon Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(criterion='entropy', random_state=0)"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifierOVD = DecisionTreeClassifier(criterion = 'entropy' , random_state = 0)\n",
    "classifierOVD.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9144239938935783\n",
      "0.9145818555940839\n",
      "0.9144233325664033\n",
      "0.9144159862767175\n",
      "0.9144233325664033\n"
     ]
    }
   ],
   "source": [
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOVD ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.575"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = classifierOVD.predict(X_test)\n",
    "classifierOVD.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 99.98%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 99.99%\n",
      "\t\t\tRecall Score: 99.98%\n",
      "\t\t\tF1 score: 99.98%\n",
      "\t\t\tAUC score: 99.98%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[35378     4]\n",
      " [    7 35375]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      1.000     1.000     1.000     35382\n",
      "           1      1.000     1.000     1.000     35382\n",
      "\n",
      "    accuracy                          1.000     70764\n",
      "   macro avg      1.000     1.000     1.000     70764\n",
      "weighted avg      1.000     1.000     1.000     70764\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 57.50%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 54.10%\n",
      "\t\t\tRecall Score: 99.00%\n",
      "\t\t\tF1 score: 69.96%\n",
      "\t\t\tAUC score: 57.50%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[ 32 168]\n",
      " [  2 198]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.941     0.160     0.274       200\n",
      "           1      0.541     0.990     0.700       200\n",
      "\n",
      "    accuracy                          0.575       400\n",
      "   macro avg      0.741     0.575     0.487       400\n",
      "weighted avg      0.741     0.575     0.487       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.941     0.160     0.274       200\n",
      "           1      0.541     0.990     0.700       200\n",
      "\n",
      "    accuracy                          0.575       400\n",
      "   macro avg      0.741     0.575     0.487       400\n",
      "weighted avg      0.741     0.575     0.487       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbkAAAGDCAYAAAC2tW7jAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXwV5d3+8c9XVhGoFLAqoCwiiBQsplL6YEFUVhW0WKAuPEWL+1JrXVurVq36uFXBBWml8vwA8UFEZBFk06IoqIgCYgERg2AjgiiIIcn398d9jCEkIYGczDlzrvfrlZeZOZNzrowhV2bOzH2buyMiIhJHB0QdQEREJFlUciIiElsqORERiS2VnIiIxJZKTkREYkslJyIisaWSExGR2FLJiewHM1tnZt+Y2ddmtsnMxphZ3WLb/NzM5prZV2b2pZlNNbN2xbapb2YPmdn6xHOtTiw3KuV1zcyuNLP3zWy7mWWb2bNm9uNkfr8i6UYlJ7L/Tnf3usBxwE+AG797wMy6ALOAKcDhQAvgXWChmbVMbFMTmAMcC/QG6gM/BzYDJ5Tymn8DrgKuBH4IHA08D/SraHgzq17RrxFJF6YRT0T2nZmtAy5095cTy/cCx7p7v8Tyq8B77n5psa+bAeS4+/lmdiFwJ9DK3b8ux2u2Bj4Aurj7m6VsMx/4X3cfnVj+70TOrollBy4HrgaqAy8BX7v7tUWeYwqwwN0fMLPDgUeAXwBfAw+6+8Pl2EUikdKRnEglMbOmQB9gdWK5DuGI7NkSNp8InJr4/BRgZnkKLuFkILu0gquAAUBnoB0wDhhkZgZgZg2AnsAEMzsAmEo4Am2SeP2rzazXfr6+SNKp5ET23/Nm9hXwCfAf4M+J9T8k/BvbWMLXbAS+e7+tYSnblKai25fmr+7+hbt/A7wKOHBi4rGBwOvu/inwU6Cxu9/u7rnuvhZ4EhhcCRlEkkolJ7L/Brh7PaA70Jbvy2sLUAAcVsLXHAZ8nvh8cynblKai25fmk+8+8fC+xQRgSGLVr4H/l/j8SOBwM9v63QdwE/CjSsggklQqOZFK4u4LgDHAfYnl7cDrwNklbP4rwsUmAC8DvczsoHK+1BygqZlllbHNdqBOkeVDS4pcbHk8MNDMjiScxpyUWP8J8JG7H1zko5679y1nXpHIqOREKtdDwKlmdlxi+QZgaOJy/3pm1sDM7gC6ALclthlLKJJJZtbWzA4ws4ZmdpOZ7VEk7v5v4FFgvJl1N7OaZlbbzAab2Q2JzZYCZ5lZHTM7Crhgb8Hd/R0gBxgNvOTuWxMPvQlsM7PrzexAM6tmZu3N7Kf7soNEqpJKTqQSuXsO8DTwp8Tyv4BewFmE99E+Jtxm0DVRVrj7t4SLTz4AZgPbCMXSCHijlJe6EhgBjAS2AmuAMwkXiAA8COQCnwH/5PtTj3szPpFlXJHvKR84nXCLxEeE06yjgR+U8zlFIqNbCEREJLZ0JCciIrGlkhMRkdhSyYmISGyp5EREJLZUciIiEltpN/p4o0aNvHnz5lHHEBGRFPLWW2997u6Ni69Pu5Jr3rw5S5YsiTqGiIikEDP7uKT1Ol0pIiKxpZITEZHYUsmJiEhsqeRERCS2VHIiIhJbKjkREYktlZyIiMSWSk5ERGJLJSciIrGVtJIzs3+Y2X/M7P1SHjcze9jMVpvZMjPrlKwsIiKSmZJ5JDcG6F3G432A1omP4cBjScwiIiIZKGljV7r7K2bWvIxN+gNPu7sDi8zsYDM7zN03JiuTiIhEo6AAvv4avvoqfGzbFv5bqxZ07Zq8141ygOYmwCdFlrMT6/YoOTMbTjja44gjjqiScCIimS4/f89SKs/nJa37+uuSX+P44yGZY+5HWXJWwjovaUN3HwWMAsjKyipxGxERgV27Kl5ApX2+Y0f5XrNGDahfH+rVCx/160OjRtCy5e7rSvq88R6T41SuKEsuG2hWZLkp8GlEWUREUsqXX8KsWeG/FTly2rmzfM9fu/aepXPooXD00WWXUkmf16qV3H2xP6IsuReAy81sAtAZ+FLvx4mIwGefwcknw/Llu6+vU2fPomnWrOwCKq2gatSI5nurakkrOTMbD3QHGplZNvBnoAaAuz8OTAf6AquBHcBvkpVFRCRdbNwIPXrA+vUweXJ4z6pePahbF6qn3TTX0Uvm1ZVD9vK4A5cl6/VFRNLNhg2h4DZsgOnToVu3qBOlP/1dICKSArKz4aSTYNMmmDkzuZfVZxKVnIhIxD7+OBzB5eSEi026dIk6UXyo5EREIvTRR+EIbutWmD0bOneOOlG8qORERCKyZk04gvvqK3j5ZcjKijpR/KjkREQi8O9/h4LbsQPmzIGf/CTqRPGkkhMRqWKrVoWCy82FuXOhY8eoE8WXSk5EpAqtXBkKLj8f5s2D9u2jThRvmjRVRKSKLF8O3buDO8yfr4KrCio5EZEqsGxZKLhq1ULBtWsXdaLMoJITEUmypUvDKcpatWDBAmjbNupEmUMlJyKSRG+9FQquTp1QcK1bR50os6jkRESS5M03w2wC9euHgmvVKupEmUclJyKSBIsWwamnwg9/GAquRYuoE2UmlZyISCVbuBB69gyzXi9YAEceGXWizKWSExGpRK+8Ar16hVm2FywIk5pKdFRyIiKVZP586NMnFNuCBdCkSdSJRCUnIlIJ5syBvn2hefMwkslhh0WdSEAlJyKy3156CU47LVw9OW9eOFUpqUElJyKyH6ZPh/79oU2bUHCHHBJ1IilKJSciso+mToUzz4Rjjw2zCTRqFHUiKU4lJyKyD55/Hn75S+jQIUx4+sMfRp1ISqKSExGpoEmT4OyzoVMnmD0bGjSIOpGURiUnIlIBEyfCoEFwwgkwaxYcfHDUiaQsKjkRkXIaNw6GDIEuXWDmzDAmpaQ2lZyISDmMHQvnnQcnnggzZkC9elEnkvJQyYmI7MVTT8HQoWHS02nToG7dqBNJeankRETK8OSTMGwYnHJKuGXgoIOiTiQVoZITESnF44/D8OHQuze88EKY+FTSi0pORKQEI0bAJZdAv37hnrjataNOJPtCJSciUsxDD8EVV4ThuiZNglq1ok4k+0olJyJSxP33w+9+B2edFe6JU8GlN5WciEjCPffAtdeG0UwmTICaNaNOJPtLJSciAtx5J9xwAwweHG76rlEj6kRSGVRyIpLR3OG22+CPf4Rzzw03fVevHnUqqSz6XykiGcsdbrkF7rgD/vu/YfRoqFYt6lRSmVRyIpKR3OGmm+Duu+HCC+GJJ+AAnduKHZWciGQcd7juOrjvPrjoInj0URVcXKnkRCSjuMM114R74S67DB55BMyiTiXJor9dRCRjuMOVV4aCu+oqFVwmUMmJSEYoKAhHbiNGhCO5Bx9UwWUClZyIxF5BAVx8MTz22PfvxangMoNKTkRiLT8/XD355JPfX02pgsscuvBERGIrPx9+85twg/ef/xw+VHCZRSUnIrGUlxdm8x43Dm6/Hf70p6gTSRRUciISO7t2hSG6Jk6Eu+6CG2+MOpFERSUnIrGyaxcMGRLmgbv3XvjDH6JOJFFSyYlIbOTmwqBBYSbvBx4I88JJZlPJiUgsfPttmAdu6lR4+OEws7eISk5E0t7OnfDLX8L06TByJFx6adSJJFWo5EQkrX3zDQwYALNmhZkEhg+POpGkkqTeDG5mvc1slZmtNrMbSnj8B2Y21czeNbPlZvabZOYRkXjZsQNOPx1mz4a//10FJ3tKWsmZWTVgJNAHaAcMMbN2xTa7DFjh7h2B7sD9ZlYzWZlEJD62b4d+/WDuXBgzBoYNizqRpKJkHsmdAKx297XungtMAPoX28aBemZmQF3gCyAviZlEJAa++gr69IFXXgmjmZx/ftSJJFUl8z25JsAnRZazgc7FthkBvAB8CtQDBrl7QRIziUia27YN+vaFRYvCaCaDBkWdSFJZMo/kShohzost9wKWAocDxwEjzKz+Hk9kNtzMlpjZkpycnMpPKiJp4csvoVcveOMNmDBBBSd7l8ySywaaFVluSjhiK+o3wHMerAY+AtoWfyJ3H+XuWe6e1bhx46QFFpHUtXUrnHoqLFkShusaODDqRJIOkllyi4HWZtYicTHJYMKpyaLWAycDmNmPgDbA2iRmEpE09MUXcMopsHRpGK7rzDOjTiTpImnvybl7npldDrwEVAP+4e7LzezixOOPA38BxpjZe4TTm9e7++fJyiQi6efzz8MR3IoVMHlyuKJSpLySejO4u08Hphdb93iRzz8FeiYzg4ikr5wcOPlk+PBDmDIFeveOOpGkG414IiIp6bPPQsGtWQMvvhhOV4pUlEpORFLOxo3QowesXw/TpoXPRfaFSk5EUsqGDaHUNmwIAy536xZ1IklnKjkRSRnZ2XDSSbBpE8ycCV27Rp1I0p1KTkRSwvr1oeBycsKMAl26RJ1I4kAlJyKRW7cuFNyWLWFGgc7FBwAU2UcqORGJ1Jo14T24r76Cl1+GrKyoE0mcqOREJDL//ncouB07YM4c+MlPok4kcaOSE5FIrFoVCi43N8wJ17Fj1IkkjlRyIlLlVq4MBZefD/PmQfv2USeSuErmAM0iIntYvhy6dwd3mD9fBSfJpZITkSqzbFkouGrVQsG1axd1Iok7lZyIVImlS8Mpylq1YMECaLvHzJEilU8lJyJJ9/bboeDq1AkF17p11IkkU6jkRCSpFi8OswnUrx8KrlWrqBNJJlHJiUjSLFoUpshp0CAUXIsWUSeSTKOSE5GkWLgQevaExo1DwR15ZNSJJBOp5ESk0r3yCvTqBYceGgquWbOoE0mmUsmJSKWaPx/69AnFtmABNGkSdSLJZCo5Eak0c+ZA377QvHkYyeSww6JOJJlOJScilWLWLDjttHD15Lx54VSlSNRUciKy32bMgDPOgDZtQsEdckjUiUQClZyI7JepU2HAgDBE15w50KhR1IlEvqeSE5F99vzz8MtfQocOoeAaNow6kcjuVHIisk8mTYKzz4ZOnWD27HDDt0iqUcmJSIVNnAiDBsEJJ4QLTg4+OOpEIiVTyYlIhYwbB0OGQJcuMHNmGJNSJFWp5ESk3MaOhfPOgxNPDFdU1qsXdSKRsqnkRKRcxoyBoUPDpKfTpkHdulEnEtk7lZyI7NXo0TBsWJhRYOpUOOigqBOJlI9KTkTK9Pjj8NvfhgGXX3ghTHwqki5UciJSqhEj4JJLoF+/cE9c7dpRJxKpGJWciJTooYfgiiugf/9wT1ytWlEnEqk4lZyI7OH+++F3v4Ozzgr3xKngJF2p5ERkN/fcA9deG0YzmTABataMOpHIvlPJiUihO++EG26AwYPDTd81akSdSGT/qOREBIDbboM//hHOPTfc9F29etSJRPafSk4kw7nDLbfArbeGm73HjFHBSXzoR1kkg7nDTTfB3XfDBRfAqFFwgP70lRjRj7NIhnKH664LBXfRRSo4iScdyYlkIHe45ppwL9xll8Ejj4BZ1KlEKp/+bhPJMO5w5ZWh4K66SgUn8aaSE8kgBQXhyG3EiHAk9+CDKjiJN5WcSIYoKICLL4bHHgvvxd13nwpO4k8lJ5IB8vPhwgvhySe/v5pSBSeZQCUnEnP5+WEuuKeeCvfD3XGHCk4yh66uFImxvLxwg/e4cXD77fCnP0WdSKRqqeREYmrXrjBE18SJcNddcOONUScSqXoqOZEY2rULhgwJ88Ddey/84Q9RJxKJhkpOJGZyc2HQoDCT9wMPhHnhRDKVSk4kRr79NswDN3UqPPxwmNlbJJMl9epKM+ttZqvMbLWZ3VDKNt3NbKmZLTezBcnMIxJnO3eGmbynToWRI1VwIpDEIzkzqwaMBE4FsoHFZvaCu68oss3BwKNAb3dfb2aHJCuPSJx98w2ceSa89BI88QQMHx51IpHUkMwjuROA1e6+1t1zgQlA/2Lb/Bp4zt3XA7j7f5KYRySWduyAM86AWbPg739XwYkUlcySawJ8UmQ5O7GuqKOBBmY238zeMrPzS3oiMxtuZkvMbElOTk6S4oqkn+3b4bTTYM6cMNnpsGFRJxJJLcm88KSkMRW8hNc/HjgZOBB43cwWufuHu32R+yhgFEBWVlbx5xDJSF99Bf36wcKFMHYsnHNO1IlEUk8ySy4baFZkuSnwaQnbfO7u24HtZvYK0BH4EBEp1bZt0LcvLFoURjMZNCjqRCKpKZmnKxcDrc2shZnVBAYDLxTbZgpwoplVN7M6QGdgZRIziaS9L7+EXr3gjTdgwgQVnEhZknYk5+55ZnY58BJQDfiHuy83s4sTjz/u7ivNbCawDCgARrv7+8nKJJLutm6Fnj3hnXfCcF1nnhl1IpHUZu7p9RZXVlaWL1myJOoYIlXuiy9CwS1bBv/3f+GKShEJzOwtd88qvl4jnoikgc2b4ZRTYMUKmDw5XHAiInunkhNJcTk5oeBWrYIpU6B376gTiaQPlZxICvvsMzj5ZFizBl58MZSdiJSfSk4kRW3cCD16wPr1MG1a+FxEKkYlJ5KCNmwIpbZhA0yfDt26RZ1IJD2p5ERSTHY2nHQSbNoEM2dC165RJxJJXyo5kRSyfn0ouJycMOByly5RJxJJbyo5kRSxbl0ouC1bYPZs6Nw56kQi6U8lJ5IC1q4NBbdtG7z8MmTtcUuriOwLlZxIxFavDgW3YwfMnQs/+UnUiUTiQyUnEqFVq8JVlLm5oeA6dow6kUi8qOREIrJyZSi4/HyYNw/at486kUj8JHOqHREpxfLl0L07uMP8+So4kWRRyYlUsWXLQsFVqxYKrl27qBOJxJdKTqQKLV0aTlHWqgULFkDbtlEnEok3lZxIFXn77VBwdeqEgmvdOupEIvFX4ZIzs2pmdk4ywojE1eLFYTaB+vVDwbVqFXUikcxQasmZWX0zu9HMRphZTwuuANYCv6q6iCLpbdGiMEVOgwah4Fq0iDqRSOYo6xaCscAW4HXgQuAPQE2gv7svrYJsImlv4ULo0wcOOSTcJtCsWdSJRDJLWSXX0t1/DGBmo4HPgSPc/asqSSaS5l55Bfr2hcMPDwXXpEnUiUQyT1nvye367hN3zwc+UsGJlM/8+eEIrlmzcIpSBScSjbKO5Dqa2TbAEssHFll2d6+f9HQiaWjOHDj99PDe25w5cOihUScSyVyllpy7V6vKICJxMGsW9O8PRx0VCu6QQ6JOJJLZSi05M6sNXAwcBSwD/uHueVUVTCTdzJgBZ54ZbvB++WVo1CjqRCJS1nty/wSygPeAvsD9VZJIJA29+CIMGBCG6JozRwUnkirKek+uXZGrK/8OvFk1kUTSy5QpcPbZYZqcWbPC/XAikhrKe3WlTlOKlGDSJBg4EDp1gtmzVXAiqaasI7njEldTQriiUldXihQxcSL8+tfQuXN4P66+/kWIpJyySu5dd/9JlSURSSPjxsF558HPfw7Tp0O9elEnEpGSlHW60qsshUgaGTs2FNyJJ4YjOBWcSOoq60juEDO7prQH3f2BJOQRSWljxsCwYXDSSfDCC3DQQVEnEpGylFVy1YC6fD/iiUhGGz0ahg8PMwo8/3yYF05EUltZJbfR3W+vsiQiKeyJJ+Dii6F3b3juOTjwwKgTiUh5lPWenI7gRICRI0PB9esHkyer4ETSSVkld3KVpRBJUX/7G1x+eRiPctIkqF076kQiUhGllpy7f1GVQURSzf33w9VXw1lnhXviatWKOpGIVFRZR3IiGeuee+Daa8NwXRMmQM2aUScSkX2hkhMp5s474YYbYPDgcNN3jRpRJxKRfaWSEynittvgj3+Ec88NN31XL+v6YxFJeSo5EcAdbrkFbr0Vhg4NN32r4ETSn/4ZS8Zzh5tvhr/+FS64AEaNggP0559ILOifsmQ0d7j++lBwF12kghOJGx3JScZyh2uugYcegssug0ceAdMQCCKxor9ZJSO5w5VXhoK76ioVnEhcqeQk4xQUhCO3ESPCkdyDD6rgROJKJScZpaAgjEP52GNw3XVw330qOJE4U8lJxsjPhwsvhCefhJtugrvvVsGJxJ1KTjJCfn6Y7PSpp8L9cHfcoYITyQS6ulJiLy8v3OA9bhzcfjv86U9RJxKRqqKSk1jLywtDdD3zDNx1F9x4Y9SJRKQqJfV0pZn1NrNVZrbazG4oY7ufmlm+mQ1MZh7JLLt2hUGWn3kG7r1XBSeSiZJWcmZWDRgJ9AHaAUPMrF0p290DvJSsLJJ5cnPhV78KE50+8AD84Q9RJxKRKCTzSO4EYLW7r3X3XGAC0L+E7a4AJgH/SWIWySDffgsDB8Lzz8PDD8Pvfhd1IhGJSjJLrgnwSZHl7MS6QmbWBDgTeLysJzKz4Wa2xMyW5OTkVHpQiY+dO8NM3lOnwsiRcMUVUScSkSgls+RKukDbiy0/BFzv7vllPZG7j3L3LHfPaty4caUFlHj55hsYMACmT4cnnoBLL406kYhELZlXV2YDzYosNwU+LbZNFjDBwg1LjYC+Zpbn7s8nMZfE0I4d0L8/zJkDf/97uCdORCSZJbcYaG1mLYANwGDg10U3cPcW331uZmOAF1VwUlHbt8Ppp8P8+eFm76FDo04kIqkiaSXn7nlmdjnhqslqwD/cfbmZXZx4vMz34UTK4+uvoV8/+Ne/YOxYOOecqBOJSCpJ6s3g7j4dmF5sXYnl5u7/ncwsEj/btkHfvrBoURjNZNCgqBOJSKrRiCeSlr78Enr3hiVLYMKEcMuAiEhxKjlJO1u3Qs+e8M47MHEinHlm1IlEJFWp5CStfPFFKLhly8JoJmecEXUiEUllKjlJG5s3wymnwIoVMHlyuOBERKQsKjlJCzk5oeBWrYIpU8L7cSIie6OSk5T32Wdw8smwZk0YruvUU6NOJCLpQiUnKW3TJujRAz7+GKZNC5+LiJSXSk5S1qefhlLLzg7jUXbrFnUiEUk3KjlJSdnZcNJJ4Uhu5kzo2jXqRCKSjlRyknLWrw8Fl5MDs2ZBly5RJxKRdKWSk5Sybl0ouC1bYPZs6Nw56kQiks5UcpIy1q4NBbdtG7z8MmRlRZ1IRNKdSk5SwurVoeB27AhzwnXqFHUiEYkDlZxE7sMPQ8Hl5sLcudCxY9SJRCQuVHISqQ8+CAWXnw/z5kH79lEnEpE4OSDqAJK5li8P9765h1m9VXAiUtlUchKJZcuge3eoVi0UXLt2UScSkThSyUmVW7o0jGRSqxYsWABt20adSETiSiUnVertt0PB1akTCq5166gTiUicqeSkyixeHGYTqF8/FFyrVlEnEpG4U8lJlVi0KMwH16BBeA+uRYuoE4lIJlDJSdK99hr07AmNG4cjuObNo04kIplCJSdJ9eqr0KsXHHpoKLhmzaJOJCKZRCUnSTN/PvTuDU2bhoJr0iTqRCKSaVRykhRz5kDfvuHU5Lx5cNhhUScSkUykkpNKN2sWnHZauHpy3rxwqlJEJAoqOalUM2bAGWdAmzah4A45JOpEIpLJVHJSaV58EQYMCEN0zZkDjRpFnUhEMp1KTirFlClw1lnQoUMouIYNo04kIqKSk0rw3HMwcGCY6HT27HDDt4hIKlDJyX559ln41a/ghBPCBScHHxx1IhGR76nkZJ+NHw9DhkCXLjBzZhiTUkQklajkZJ+MHQvnngtdu4YrKuvVizqRiMieVHJSYWPGwNChYdLTadOgbt2oE4mIlEwlJxUyejQMGxZmFJg6FQ46KOpEIiKlU8lJuT3xBPz2t2HA5SlTwsSnIiKpTCUn5TJyJFx8MfTrB5Mnw4EHRp1IRGTvVHKyV3/7G1x+OfTvD5MmQe3aUScSESkflZyU6YEH4Oqrw2gmEydCrVpRJxIRKT+VnJTqnnvg97+Hs8+GCROgZs2oE4mIVIxKTkp0551www0weDCMGwc1akSdSESk4lRysofbboM//jHc7D12LFSvHnUiEZF9o5KTQu5wyy1w663hZu8xY1RwIpLe9CtMgFBwN98Mf/0rXHABjBoFB+hPIBFJc/o1JrjD9deHgrvoIhWciMSHjuQynHu4gvLBB+Gyy+CRR8As6lQiIpVDf69nMHe46qpQcFddpYITkfhRyWWogoLvj9yuuSYUnQpOROJGJZeBCgrCOJSPPQbXXQf33aeCE5F4UsllmPx8uPBCePJJuOkmuPtuFZyIxFdSS87MepvZKjNbbWY3lPD4OWa2LPHxmpl1TGaeTJefH+aCe+qpcD/cHXeo4EQk3pJ2daWZVQNGAqcC2cBiM3vB3VcU2ewjoJu7bzGzPsAooHOyMmWyvLxwg/e4cWFEk1tuiTqRiEjyJfNI7gRgtbuvdfdcYALQv+gG7v6au29JLC4CmiYxT8bKywtDdI0bB3fdpYITkcyRzJJrAnxSZDk7sa40FwAzkpgnI+3aBUOGwDPPwL33wo03Rp1IRKTqJPNm8JLe7fESNzQ7iVByXUt5fDgwHOCII46orHyxl5sbZhGYPDnMC/e730WdSESkaiXzSC4baFZkuSnwafGNzKwDMBro7+6bS3oidx/l7lnuntW4ceOkhI2bb7+FgQNDwT38sApORDJTMktuMdDazFqYWU1gMPBC0Q3M7AjgOeA8d/8wiVkyys6dYSbvqVNh5Ei44oqoE4mIRCNppyvdPc/MLgdeAqoB/3D35WZ2ceLxx4FbgIbAoxauZc9z96xkZcoE33wDZ54JL70ETzwBw4dHnUhEJDrmXuLbZCkrKyvLlyxZEnWMlLRjB/TvD3PmwOjR4Z44EZFMYGZvlXSQpFkIYmL7djj9dJg/P9zsPXRo1IlERKKnkouBr7+Gfv3gX/+CsWPhnHOiTiQikhpUcmnuq6+gTx9YtCjc7D1oUNSJRERSh0oujX35ZSi4xYthwoRwy4CIiHxPJZemtm6FXr3g7bdh4sRwRaWIiOxOJZeGvvgCevaEZctg0iQ444yoE4mIpCaVXJrZvBlOOQVWrAijmfTrF3UiEZHUpZJLIzk5oeBWrYIpU6B376gTiYikNpVcmvjsMzj5ZFizJgzXdeqpUScSEUl9Krk0sGkT9OgBH38M06aFz0VEZO9Ucinu009DqWVnw/Tp0K1b1IlERNKHSi6FZWeHgtu4EWbOhK4lzrYnIiKlUcmlqPXr4aSTwsUms2ZBly5RJ7svEp4AABEcSURBVBIRST8quRS0bl0ouC1bYPZs6Nw56kQiIulJJZdi1q4NBbdtG7z8MmRpdj0RkX2mkkshq1eHgtuxI8wJ16lT1IlERNKbSi5FfPhhKLjcXJg7Fzp2jDqRiEj6U8mlgA8+CAWXnw/z5kH79lEnEhGJhwOiDpDpli+H7t3BPczqrYITEak8KrkIvfdeOII74IBQcO3aRZ1IRCReVHIRWbo0FFzNmrBgAbRtG3UiEZH4UclF4O23w0gmdeqEgmvdOupEIiLxpJKrYosXh9kE6tcPBdeqVdSJRETiSyVXhRYtCvPBNWgQ3oNr0SLqRCIi8aaSqyKvvQY9e0LjxuEIrnnzqBOJiMSfSq4KvPoq9OoFhx4aCq5Zs6gTiYhkBpVcks2fD717Q9OmoeCaNIk6kYhI5lDJJdGcOdC3bzg1OW8eHHZY1IlERDKLSi5JZs2C004LV0/OmxdOVYqISNVSySXBjBlwxhnQpk0ouEMOiTqRiEhmUslVshdfhAEDwhBdc+ZAo0ZRJxIRyVwquUo0ZQqcdRZ06BAKrmHDqBOJiGQ2lVwlee45GDgwTHQ6e3a44VtERKKlkqsEzz4Lv/oVnHBCuODk4IOjTiQiIqCS22/jx8OQIdClC8ycGcakFBGR1KCS2w//+79w7rnQtWu4orJevagTiYhIUSq5fTRmDJx/fpjVe9o0qFs36kQiIlKcSm4fjB4Nw4aFGQWmToWDDoo6kYiIlEQlV0FPPAG//W0YcHnKlDDxqYiIpCaVXAWMHAkXXwz9+sHkyXDggVEnEhGRsqjkyulvf4PLL4f+/WHSJKhdO+pEIiKyNyq5cnjgAbj66jCaycSJUKtW1IlERKQ8VHJ7ce+98Pvfw9lnw4QJULNm1IlERKS8VHJluOsuuP56GDwYxo2DGjWiTiQiIhWhkivF7bfDzTeHm73HjoXq1aNOJCIiFaWSK8YdbrkF/vxnGDo03PStghMRSU/69V2Eezh6++tf4YILYNQoOEB/BohUiV27dpGdnc3OnTujjiIprHbt2jRt2pQa5Xz/SCWX4B7ef/uf/4Hhw+Gxx1RwIlUpOzubevXq0bx5c8ws6jiSgtydzZs3k52dTYsWLcr1Nfo1Tii43/8+FNyll6rgRKKwc+dOGjZsqIKTUpkZDRs2rNDRfsb/KneHq66CBx8M/x0xQgUnEhUVnOxNRX9GMvrXeUEBXHYZPPIIXHNNKDr9GxMRiY+klpyZ9TazVWa22sxuKOFxM7OHE48vM7NOycxTVEEBXHJJODV53XVw330qOJFMV61aNY477jjat2/P6aefztatWwsfW758OT169ODoo4+mdevW/OUvf8HdCx+fMWMGWVlZHHPMMbRt25Zrr722xNd4/vnnuf3223db17FjR4YMGbLbuu7du7NkyZLC5XXr1tG+ffvC5TfffJNf/OIXtGnThrZt23LhhReyY8eO/fr+P/roIzp37kzr1q0ZNGgQubm5JW733X467rjjOOOMMwrXn3jiiYXrDz/8cAYMGADA/Pnz+cEPflD4WNHvf+bMmbRp04ajjjqKu+++u3D9tddey9y5c/fr+wHCG3nJ+ACqAWuAlkBN4F2gXbFt+gIzAAN+Bryxt+c9/vjjfX/l5bkPG+YO7jfd5F5QsN9PKSL7acWKFVFH8IMOOqjw8/PPP9/vuOMOd3ffsWOHt2zZ0l966SV3d9++fbv37t3bR4wY4e7u7733nrds2dJXrlzp7u67du3ykSNHlvgaXbp08ZycnMLlFStWePv27f3www/3r7/+unB9t27dfPHixYXLH330kR977LHu7r5p0yY/4ogj/LXXXnN394KCAn/22Wd906ZN+/X9n3322T5+/Hh3d7/ooov80UcfLXG7ovupNGeddZb/85//dHf3efPmeb9+/fbYJi8vz1u2bOlr1qzxb7/91jt06ODLly93d/d169b5qaeeWuJzl/SzAizxEjojmVdXngCsdve1AGY2AegPrCiyTX/g6UTARWZ2sJkd5u4bkxUqPz/MBff00+F+uFtv1RGcSKq5+mpYurRyn/O44+Chh8q/fZcuXVi2bBkA48aN47/+67/o2bMnAHXq1GHEiBF0796dyy67jHvvvZebb76Ztm3bAlC9enUuvfTSPZ7zww8/pFatWjRq1Khw3bhx4zjvvPNYuXIlL7zwwh5HdCUZOXIkQ4cOpUuXLkB4n2rgwIHl/+ZK4O7MnTuXcePGATB06FBuvfVWLrnkkgo/11dffcXcuXN56qmnytzuzTff5KijjqJly5YADB48mClTptCuXTuOPPJINm/ezKZNmzj00EMr/g0lJPN0ZRPgkyLL2Yl1Fd0GMxtuZkvMbElOTs5+hTrggDCL9223hQ8VnIgUl5+fz5w5cwpPxS1fvpzjjz9+t21atWrF119/zbZt23j//ff3eLwkCxcupFOn3d+VeeaZZxg0aBBDhgxh/Pjx5cpX3tdbtWpV4SnC4h9FT8UCbN68mYMPPpjqidEvmjZtyoYNG0p83p07d5KVlcXPfvYznn/++T0enzx5MieffDL169cvXPf666/TsWNH+vTpw/LlywHYsGEDzZo1K9ym+Gt26tSJhQsX7vX7LEsyj+RKqg/fh21w91HAKICsrKw9Hq9QKAtXUKrcRFJXRY64KtM333zDcccdx7p16zj++OM59dRTgXCUU9pVfRW52m/jxo00bty4cHnx4sU0btyYI488kqZNmzJs2DC2bNlCgwYNSnzeil5Z2KZNG5aW85DYfc9fraW93vr16zn88MNZu3YtPXr04Mc//jGtWrUqfHz8+PFceOGFhcudOnXi448/pm7dukyfPp0BAwbw73//e6+vecghh/Dpp5+WK39pknkklw00K7LcFCietjzbVDoVnIiU5MADD2Tp0qV8/PHH5ObmMnLkSACOPfbY3S4CAVi7di1169alXr16HHvssbz11lvlev6i93iNHz+eDz74gObNm9OqVSu2bdvGpEmTAGjYsCFbtmwp3PaLL74oPM1Z3teryJFco0aN2Lp1K3l5eUC4Of/www8v8Xm/W9+yZUu6d+/OO++8U/jY5s2befPNN+nXr1/huvr161O3bl0A+vbty65du/j8889p2rQpn3zy/cm84q+5c+dODtzf2alLeqOuMj4IR4lrgRZ8f+HJscW26cfuF568ubfnrYwLT0Qk9aTahSdvv/22N2vWzHNzc33Hjh3eokULnz17truHC1H69evnDz/8sLu7v/vuu96qVStftWqVu7vn5+f7/fffv8fzz5gxw88555zCbZo2berZ2dmFj8+dO9d79Ojh7u6PPPKIn3/++V6QuDLuyiuv9Ntuu83dv7/wZNGiRYVfO3bsWN+4ceN+ff8DBw7c7cKTki6e+eKLL3znzp3u7p6Tk+NHHXVU4cUi7u6PPfaYn3/++bt9zcaNGwu/jzfeeMObNWvmBQUFvmvXLm/RooWvXbu28MKT999/v/DrTjvtNH/99df3yFCRC0+SVnL+/dWTHxKusrw5se5i4OLE5waMTDz+HpC1t+dUyYnEU6qVnHv4Jfv000+7u/uyZcu8W7dufvTRR3urVq381ltvLfzF7e4+depU79Spk7dt29aPOeYYv/baa/d4/u3bt3u7du28oKDA582b5507d97t8by8PD/00EP9008/9W+//dYvu+wy//GPf+wdOnTwYcOG+fbt2wu3fe2117xr165+9NFHe9u2bX348OG7Pb4v1qxZ4z/96U+9VatWPnDgwMIyW7x4sV9wwQXu7r5w4UJv3769d+jQwdu3b++jR4/e7Tm6devmM2bM2G3dI4884u3atfMOHTp4586dfeHChYWPTZs2zVu3bu0tW7YsvJrV3T03N9fbtm3ru3bt2iNnRUrOvIRzoqksKyvLi582EJH0t3LlSo455pioYyTdVVddxemnn84pp5wSdZSUNnnyZN5++23+8pe/7PFYST8rZvaWu2cV3zajRzwREalqN910037ftJ0J8vLy+P3vf7/fz6NZCEREqtCPfvSj3UYJkZKdffbZlfI8OpITkZSRbm+fSNWr6M+ISk5EUkLt2rXZvHmzik5K5R7mk6tdu3a5v0anK0UkJTRt2pTs7Gz2d1QjibfvZgYvL5WciKSEGjVqlHu2Z5Hy0ulKERGJLZWciIjElkpORERiK+1GPDGzHODjSniqRsDnlfA8caR9Uzrtm9Jp35RO+6Z0lbVvjnT3xsVXpl3JVRYzW1LSEDCifVMW7ZvSad+UTvumdMneNzpdKSIisaWSExGR2MrkkhsVdYAUpn1TOu2b0mnflE77pnRJ3TcZ+56ciIjEXyYfyYmISMzFvuTMrLeZrTKz1WZ2QwmPm5k9nHh8mZl1iiJnFMqxb85J7JNlZvaamXWMImcU9rZvimz3UzPLN7OBVZkvSuXZN2bW3cyWmtlyM1tQ1RmjUo5/Uz8ws6lm9m5i3/wmipxVzcz+YWb/MbP3S3k8eb+HS5ouPC4fQDVgDdASqAm8C7Qrtk1fYAZgwM+AN6LOnUL75udAg8TnfbRvStxuLjAdGBh17lTZN8DBwArgiMTyIVHnTqF9cxNwT+LzxsAXQM2os1fBvvkF0Al4v5THk/Z7OO5HcicAq919rbvnAhOA/sW26Q887cEi4GAzO6yqg0Zgr/vG3V9z9y2JxUVA+Yf+Tm/l+bkBuAKYBPynKsNFrDz75tfAc+6+HsDdM2X/lGffOFDPzAyoSyi5vKqNWfXc/RXC91qapP0ejnvJNQE+KbKcnVhX0W3iqKLf9wWEv7QywV73jZk1Ac4EHq/CXKmgPD83RwMNzGy+mb1lZudXWbpolWffjACOAT4F3gOucveCqomX0pL2ezjuU+1YCeuKX05anm3iqNzft5mdRCi5rklNlDrKs28eAq539/zwR3nGKM++qQ4cD5wMHAi8bmaL3P3DZIeLWHn2TS9gKdADaAXMNrNX3X1bssOluKT9Ho57yWUDzYosNyX8BVXRbeKoXN+3mXUARgN93H1zFWWLWnn2TRYwIVFwjYC+Zpbn7s9XTcTIlPff1Ofuvh3YbmavAB2BuJdcefbNb4C7PbwRtdrMPgLaAm9WTcSUlbTfw3E/XbkYaG1mLcysJjAYeKHYNi8A5yeu7vkZ8KW7b6zqoBHY674xsyOA54DzMuCv8KL2um/cvYW7N3f35sD/AZdmQMFB+f5NTQFONLPqZlYH6AysrOKcUSjPvllPOMLFzH4EtAHWVmnK1JS038OxPpJz9zwzuxx4iXDl0z/cfbmZXZx4/HHClXF9gdXADsJfWrFXzn1zC9AQeDRxxJLnGTDIbDn3TUYqz75x95VmNhNYBhQAo929xEvH46ScPzd/AcaY2XuEU3TXu3vsZycws/FAd6CRmWUDfwZqQPJ/D2vEExERia24n64UEZEMppITEZHYUsmJiEhsqeRERCS2VHIiIhJbKjmRFJWY3WBpkY/midH9vzSzd8xspZn9ObFt0fUfmNl9UecXSQWxvk9OJM194+7HFV1hZs2BV939NDM7CFhqZi8mHv5u/YHAO2Y22d0XVm1kkdSiIzmRNJUYNustwhiIRdd/QxgfMRMGGhcpk0pOJHUdWORU5eTiD5pZQ8LcW8uLrW8AtAZeqZqYIqlLpytFUtcepysTTjSzdwhDZt2dGDqqe2L9MsJ4iHe7+6YqzCqSklRyIunnVXc/rbT1ZnY08K/Ee3JLqzqcSCrR6UqRmEnMGPFX4Pqos4hETSUnEk+PA78wsxZRBxGJkmYhEBGR2NKRnIiIxJZKTkREYkslJyIisaWSExGR2FLJiYhIbKnkREQktlRyIiISWyo5ERGJrf8PehiZ38P8WRgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierOVD, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOVD, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(random_state=0)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifierOVS = SVC(kernel='rbf', random_state = 0)\n",
    "classifierOVS.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8382317565862802\n",
      "0.8443801868612434\n",
      "0.838232673773551\n",
      "0.8375064067268909\n",
      "0.9307270149568676\n"
     ]
    }
   ],
   "source": [
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOVS ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5675"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_pred = classifierOVS.predict(X_test)\n",
    "classifierOVS.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 86.17%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 99.11%\n",
      "\t\t\tRecall Score: 72.99%\n",
      "\t\t\tF1 score: 84.07%\n",
      "\t\t\tAUC score: 86.17%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[35150   232]\n",
      " [ 9557 25825]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.786     0.993     0.878     35382\n",
      "           1      0.991     0.730     0.841     35382\n",
      "\n",
      "    accuracy                          0.862     70764\n",
      "   macro avg      0.889     0.862     0.859     70764\n",
      "weighted avg      0.889     0.862     0.859     70764\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 56.75%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 96.55%\n",
      "\t\t\tRecall Score: 14.00%\n",
      "\t\t\tF1 score: 24.45%\n",
      "\t\t\tAUC score: 56.75%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[199   1]\n",
      " [172  28]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.536     0.995     0.697       200\n",
      "           1      0.966     0.140     0.245       200\n",
      "\n",
      "    accuracy                          0.568       400\n",
      "   macro avg      0.751     0.568     0.471       400\n",
      "weighted avg      0.751     0.568     0.471       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.536     0.995     0.697       200\n",
      "           1      0.966     0.140     0.245       200\n",
      "\n",
      "    accuracy                          0.568       400\n",
      "   macro avg      0.751     0.568     0.471       400\n",
      "weighted avg      0.751     0.568     0.471       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierOVS, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOVS, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,Y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, Y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='brute', n_neighbors=1)"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifierOVN = neighbors.KNeighborsClassifier(n_neighbors=1, weights='uniform', algorithm = 'brute')\n",
    "classifierOVN.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "#CrossV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8919286327939115\n",
      "0.8923228971128108\n",
      "0.8919277424137844\n",
      "0.8919015301412356\n",
      "0.8919277424137843\n"
     ]
    }
   ],
   "source": [
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOVN ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='brute', n_neighbors=1)"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_pred = classifierOVN.predict(X_test)\n",
    "classifierOVN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 99.99%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 99.99%\n",
      "\t\t\tRecall Score: 99.98%\n",
      "\t\t\tF1 score: 99.99%\n",
      "\t\t\tAUC score: 99.99%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[35380     2]\n",
      " [    8 35374]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      1.000     1.000     1.000     35382\n",
      "           1      1.000     1.000     1.000     35382\n",
      "\n",
      "    accuracy                          1.000     70764\n",
      "   macro avg      1.000     1.000     1.000     70764\n",
      "weighted avg      1.000     1.000     1.000     70764\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 66.00%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 86.36%\n",
      "\t\t\tRecall Score: 38.00%\n",
      "\t\t\tF1 score: 52.78%\n",
      "\t\t\tAUC score: 66.00%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[188  12]\n",
      " [124  76]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.603     0.940     0.734       200\n",
      "           1      0.864     0.380     0.528       200\n",
      "\n",
      "    accuracy                          0.660       400\n",
      "   macro avg      0.733     0.660     0.631       400\n",
      "weighted avg      0.733     0.660     0.631       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.603     0.940     0.734       200\n",
      "           1      0.864     0.380     0.528       200\n",
      "\n",
      "    accuracy                          0.660       400\n",
      "   macro avg      0.733     0.660     0.631       400\n",
      "weighted avg      0.733     0.660     0.631       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierOVN, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOVN, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,Y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, Y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ANN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "2212/2212 [==============================] - 5s 2ms/step - loss: 0.4865 - accuracy: 0.7330\n",
      "Epoch 2/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.2336 - accuracy: 0.8775\n",
      "Epoch 3/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.2163 - accuracy: 0.8871\n",
      "Epoch 4/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.2117 - accuracy: 0.8893\n",
      "Epoch 5/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.2098 - accuracy: 0.8897\n",
      "Epoch 6/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.2055 - accuracy: 0.8928\n",
      "Epoch 7/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.2022 - accuracy: 0.8943\n",
      "Epoch 8/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.2021 - accuracy: 0.8913\n",
      "Epoch 9/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1973 - accuracy: 0.8964\n",
      "Epoch 10/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1935 - accuracy: 0.8988\n",
      "Epoch 11/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1963 - accuracy: 0.8956\n",
      "Epoch 12/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1940 - accuracy: 0.9004\n",
      "Epoch 13/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1951 - accuracy: 0.8968\n",
      "Epoch 14/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1946 - accuracy: 0.8978\n",
      "Epoch 15/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1929 - accuracy: 0.8984\n",
      "Epoch 16/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1936 - accuracy: 0.8988\n",
      "Epoch 17/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1931 - accuracy: 0.8996\n",
      "Epoch 18/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1939 - accuracy: 0.8996\n",
      "Epoch 19/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1928 - accuracy: 0.8999\n",
      "Epoch 20/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1925 - accuracy: 0.8995\n",
      "Epoch 21/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1930 - accuracy: 0.9006\n",
      "Epoch 22/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1921 - accuracy: 0.9003\n",
      "Epoch 23/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1920 - accuracy: 0.8994\n",
      "Epoch 24/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1917 - accuracy: 0.9015\n",
      "Epoch 25/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1895 - accuracy: 0.9020\n",
      "Epoch 26/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1913 - accuracy: 0.9010\n",
      "Epoch 27/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1928 - accuracy: 0.8994\n",
      "Epoch 28/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1919 - accuracy: 0.9000\n",
      "Epoch 29/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1906 - accuracy: 0.9014\n",
      "Epoch 30/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1893 - accuracy: 0.9036\n",
      "Epoch 31/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1911 - accuracy: 0.9006\n",
      "Epoch 32/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1899 - accuracy: 0.9017\n",
      "Epoch 33/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1901 - accuracy: 0.9028\n",
      "Epoch 34/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1872 - accuracy: 0.9043\n",
      "Epoch 35/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1892 - accuracy: 0.9017\n",
      "Epoch 36/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1887 - accuracy: 0.9023\n",
      "Epoch 37/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1890 - accuracy: 0.9038\n",
      "Epoch 38/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1898 - accuracy: 0.9019\n",
      "Epoch 39/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1885 - accuracy: 0.9023\n",
      "Epoch 40/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1856 - accuracy: 0.9049\n",
      "Epoch 41/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1877 - accuracy: 0.9041\n",
      "Epoch 42/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1857 - accuracy: 0.9052\n",
      "Epoch 43/100\n",
      "2212/2212 [==============================] - ETA: 0s - loss: 0.1870 - accuracy: 0.90 - 3s 2ms/step - loss: 0.1870 - accuracy: 0.9028\n",
      "Epoch 44/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1872 - accuracy: 0.9037\n",
      "Epoch 45/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1882 - accuracy: 0.9043\n",
      "Epoch 46/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1894 - accuracy: 0.9027\n",
      "Epoch 47/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1886 - accuracy: 0.9035\n",
      "Epoch 48/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1863 - accuracy: 0.9048\n",
      "Epoch 49/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1878 - accuracy: 0.9034\n",
      "Epoch 50/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1872 - accuracy: 0.9040\n",
      "Epoch 51/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1875 - accuracy: 0.9034\n",
      "Epoch 52/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1901 - accuracy: 0.9018\n",
      "Epoch 53/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1902 - accuracy: 0.9015\n",
      "Epoch 54/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1873 - accuracy: 0.9052\n",
      "Epoch 55/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1873 - accuracy: 0.9054\n",
      "Epoch 56/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1874 - accuracy: 0.9041\n",
      "Epoch 57/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1878 - accuracy: 0.9042\n",
      "Epoch 58/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1872 - accuracy: 0.9053\n",
      "Epoch 59/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1876 - accuracy: 0.9042\n",
      "Epoch 60/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1860 - accuracy: 0.9046\n",
      "Epoch 61/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1865 - accuracy: 0.9051\n",
      "Epoch 62/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1890 - accuracy: 0.9026\n",
      "Epoch 63/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1858 - accuracy: 0.9048\n",
      "Epoch 64/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1879 - accuracy: 0.9037\n",
      "Epoch 65/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1861 - accuracy: 0.9051\n",
      "Epoch 66/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1860 - accuracy: 0.9048\n",
      "Epoch 67/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1863 - accuracy: 0.9039\n",
      "Epoch 68/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1840 - accuracy: 0.9057\n",
      "Epoch 69/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1870 - accuracy: 0.9035\n",
      "Epoch 70/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1863 - accuracy: 0.9056\n",
      "Epoch 71/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1880 - accuracy: 0.9049\n",
      "Epoch 72/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1863 - accuracy: 0.9034\n",
      "Epoch 73/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1850 - accuracy: 0.9046\n",
      "Epoch 74/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1860 - accuracy: 0.9041\n",
      "Epoch 75/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1875 - accuracy: 0.9022\n",
      "Epoch 76/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1865 - accuracy: 0.9050\n",
      "Epoch 77/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1877 - accuracy: 0.9044\n",
      "Epoch 78/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1845 - accuracy: 0.9062\n",
      "Epoch 79/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1842 - accuracy: 0.9053\n",
      "Epoch 80/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1835 - accuracy: 0.9078\n",
      "Epoch 81/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1853 - accuracy: 0.9040\n",
      "Epoch 82/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1847 - accuracy: 0.9054\n",
      "Epoch 83/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1881 - accuracy: 0.9030\n",
      "Epoch 84/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1872 - accuracy: 0.9050\n",
      "Epoch 85/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1861 - accuracy: 0.9056\n",
      "Epoch 86/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1844 - accuracy: 0.9055\n",
      "Epoch 87/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1836 - accuracy: 0.9075\n",
      "Epoch 88/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1864 - accuracy: 0.9037\n",
      "Epoch 89/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1876 - accuracy: 0.9039\n",
      "Epoch 90/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1849 - accuracy: 0.9061\n",
      "Epoch 91/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1851 - accuracy: 0.9053\n",
      "Epoch 92/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1860 - accuracy: 0.9047\n",
      "Epoch 93/100\n",
      "2212/2212 [==============================] - 3s 2ms/step - loss: 0.1869 - accuracy: 0.9030\n",
      "Epoch 94/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1854 - accuracy: 0.9048\n",
      "Epoch 95/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1840 - accuracy: 0.9055\n",
      "Epoch 96/100\n",
      "2212/2212 [==============================] - 3s 1ms/step - loss: 0.1858 - accuracy: 0.9058\n",
      "Epoch 97/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1852 - accuracy: 0.9053\n",
      "Epoch 98/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1841 - accuracy: 0.9053\n",
      "Epoch 99/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1847 - accuracy: 0.9038\n",
      "Epoch 100/100\n",
      "2212/2212 [==============================] - 4s 2ms/step - loss: 0.1854 - accuracy: 0.9039\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x21c014109d0>"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train\n",
    "annO = tf.keras.models.Sequential()\n",
    "annO.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "annO.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "annO.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "annO.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "annO.fit(Xc_train,yc_train , batch_size = 32, epochs = 30 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "#CrossV\n",
    "#CrossV\n",
    "from keras.wrappers.scikit_learn import KerasClassifier\n",
    "def create_network():\n",
    "    model = tf.keras.models.Sequential()\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "    model.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "    model.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "    return model\n",
    "\n",
    "\n",
    "\n",
    "network = KerasClassifier(build_fn=create_network, \n",
    "                                 epochs=30, \n",
    "                                 batch_size=32\n",
    "                                 )\n",
    "    \n",
    "    \n",
    "    \n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(network ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "Cannot clone object '<tensorflow.python.keras.engine.sequential.Sequential object at 0x0000021C0046F430>' (type <class 'tensorflow.python.keras.engine.sequential.Sequential'>): it does not seem to be a scikit-learn estimator as it does not implement a 'get_params' method.",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mEmpty\u001b[0m                                     Traceback (most recent call last)",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[1;34m(self, iterator)\u001b[0m\n\u001b[0;32m    807\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 808\u001b[1;33m                 \u001b[0mtasks\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_ready_batches\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mblock\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    809\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mqueue\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mEmpty\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\queue.py\u001b[0m in \u001b[0;36mget\u001b[1;34m(self, block, timeout)\u001b[0m\n\u001b[0;32m    166\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_qsize\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 167\u001b[1;33m                     \u001b[1;32mraise\u001b[0m \u001b[0mEmpty\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    168\u001b[0m             \u001b[1;32melif\u001b[0m \u001b[0mtimeout\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mEmpty\u001b[0m: ",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-109-eac606359759>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      4\u001b[0m            \u001b[1;34m'f1_macro'\u001b[0m\u001b[1;33m:\u001b[0m \u001b[1;34m'f1_macro'\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m           ,'roc_auc' : 'roc_auc'}\n\u001b[1;32m----> 6\u001b[1;33m scores = cross_validate(annO ,X_cv, y_cv ,scoring=scoring,\n\u001b[0m\u001b[0;32m      7\u001b[0m                          cv=10, return_train_score=False)\n\u001b[0;32m      8\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py\u001b[0m in \u001b[0;36minner_f\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     71\u001b[0m                           FutureWarning)\n\u001b[0;32m     72\u001b[0m         \u001b[0mkwargs\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m{\u001b[0m\u001b[0mk\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mk\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msig\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mparameters\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 73\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     74\u001b[0m     \u001b[1;32mreturn\u001b[0m \u001b[0minner_f\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\u001b[0m in \u001b[0;36mcross_validate\u001b[1;34m(estimator, X, y, groups, scoring, cv, n_jobs, verbose, fit_params, pre_dispatch, return_train_score, return_estimator, error_score)\u001b[0m\n\u001b[0;32m    240\u001b[0m     parallel = Parallel(n_jobs=n_jobs, verbose=verbose,\n\u001b[0;32m    241\u001b[0m                         pre_dispatch=pre_dispatch)\n\u001b[1;32m--> 242\u001b[1;33m     scores = parallel(\n\u001b[0m\u001b[0;32m    243\u001b[0m         delayed(_fit_and_score)(\n\u001b[0;32m    244\u001b[0m             \u001b[0mclone\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscorers\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtrain\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, iterable)\u001b[0m\n\u001b[0;32m   1027\u001b[0m             \u001b[1;31m# remaining jobs.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1028\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1029\u001b[1;33m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdispatch_one_batch\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1030\u001b[0m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_original_iterator\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1031\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[1;34m(self, iterator)\u001b[0m\n\u001b[0;32m    817\u001b[0m                 \u001b[0mbig_batch_size\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    818\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 819\u001b[1;33m                 \u001b[0mislice\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mitertools\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mislice\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mbig_batch_size\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    820\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mislice\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m==\u001b[0m \u001b[1;36m0\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    821\u001b[0m                     \u001b[1;32mreturn\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\u001b[0m in \u001b[0;36m<genexpr>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m    242\u001b[0m     scores = parallel(\n\u001b[0;32m    243\u001b[0m         delayed(_fit_and_score)(\n\u001b[1;32m--> 244\u001b[1;33m             \u001b[0mclone\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscorers\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtrain\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    245\u001b[0m             \u001b[0mfit_params\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreturn_train_score\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mreturn_train_score\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    246\u001b[0m             \u001b[0mreturn_times\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreturn_estimator\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mreturn_estimator\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py\u001b[0m in \u001b[0;36minner_f\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     71\u001b[0m                           FutureWarning)\n\u001b[0;32m     72\u001b[0m         \u001b[0mkwargs\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m{\u001b[0m\u001b[0mk\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mk\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msig\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mparameters\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 73\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     74\u001b[0m     \u001b[1;32mreturn\u001b[0m \u001b[0minner_f\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\base.py\u001b[0m in \u001b[0;36mclone\u001b[1;34m(estimator, safe)\u001b[0m\n\u001b[0;32m     76\u001b[0m                                 \"scikit-learn estimator instead of a class.\")\n\u001b[0;32m     77\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 78\u001b[1;33m                 raise TypeError(\"Cannot clone object '%s' (type %s): \"\n\u001b[0m\u001b[0;32m     79\u001b[0m                                 \u001b[1;34m\"it does not seem to be a scikit-learn \"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     80\u001b[0m                                 \u001b[1;34m\"estimator as it does not implement a \"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mTypeError\u001b[0m: Cannot clone object '<tensorflow.python.keras.engine.sequential.Sequential object at 0x0000021C0046F430>' (type <class 'tensorflow.python.keras.engine.sequential.Sequential'>): it does not seem to be a scikit-learn estimator as it does not implement a 'get_params' method."
     ]
    }
   ],
   "source": [
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 90.62%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 93.50%\n",
      "\t\t\tRecall Score: 87.30%\n",
      "\t\t\tF1 score: 90.29%\n",
      "\t\t\tAUC score: 90.62%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[33236  2146]\n",
      " [ 4494 30888]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.881     0.939     0.909     35382\n",
      "           1      0.935     0.873     0.903     35382\n",
      "\n",
      "    accuracy                          0.906     70764\n",
      "   macro avg      0.908     0.906     0.906     70764\n",
      "weighted avg      0.908     0.906     0.906     70764\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 68.25%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 61.81%\n",
      "\t\t\tRecall Score: 95.50%\n",
      "\t\t\tF1 score: 75.05%\n",
      "\t\t\tAUC score: 68.25%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[ 82 118]\n",
      " [  9 191]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.901     0.410     0.564       200\n",
      "           1      0.618     0.955     0.750       200\n",
      "\n",
      "    accuracy                          0.682       400\n",
      "   macro avg      0.760     0.682     0.657       400\n",
      "weighted avg      0.760     0.682     0.657       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_scoreN(annO, Xc_train,yc_train, X_test,Y_test, train=True)\n",
    "print_scoreN(annO, Xc_train,yc_train, X_test,Y_test, train=False)\n",
    "\n",
    "y_pred = annO.predict(X_test)\n",
    "y_pred = (y_pred > 0.5)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# XGboot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "              colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "              importance_type='gain', interaction_constraints='',\n",
       "              learning_rate=0.300000012, max_delta_step=0, max_depth=6,\n",
       "              min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "              n_estimators=100, n_jobs=0, num_parallel_tree=1, random_state=0,\n",
       "              reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "              tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train \n",
    "classifierOXG = XGBClassifier()\n",
    "classifierOXG.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9355643170357026\n",
      "0.9372471526860593\n",
      "0.9355647800189175\n",
      "0.9355021912568148\n",
      "0.9848142129456914\n"
     ]
    }
   ],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOXG ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7025"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = classifierOXG.predict(X_test)\n",
    "classifierOXG.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 95.84%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 98.45%\n",
      "\t\t\tRecall Score: 93.15%\n",
      "\t\t\tF1 score: 95.73%\n",
      "\t\t\tAUC score: 95.84%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[34864   518]\n",
      " [ 2424 32958]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.935     0.985     0.960     35382\n",
      "           1      0.985     0.931     0.957     35382\n",
      "\n",
      "    accuracy                          0.958     70764\n",
      "   macro avg      0.960     0.958     0.958     70764\n",
      "weighted avg      0.960     0.958     0.958     70764\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 70.25%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 62.70%\n",
      "\t\t\tRecall Score: 100.00%\n",
      "\t\t\tF1 score: 77.07%\n",
      "\t\t\tAUC score: 70.25%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[ 81 119]\n",
      " [  0 200]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      1.000     0.405     0.577       200\n",
      "           1      0.627     1.000     0.771       200\n",
      "\n",
      "    accuracy                          0.703       400\n",
      "   macro avg      0.813     0.703     0.674       400\n",
      "weighted avg      0.813     0.703     0.674       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      1.000     0.405     0.577       200\n",
      "           1      0.627     1.000     0.771       200\n",
      "\n",
      "    accuracy                          0.703       400\n",
      "   macro avg      0.813     0.703     0.674       400\n",
      "weighted avg      0.813     0.703     0.674       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierOXG, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOXG, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split for imbalace data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "Xc_train, X_cv, yc_train, y_cv = train_test_split(X_train, Y_train, test_size=0.2 , random_state=3, stratify = Y_train )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(criterion='entropy', n_estimators=80, random_state=0)"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train\n",
    "classifierURF = RandomForestClassifier(n_estimators = 80 ,criterion = 'entropy' , random_state = 0)\n",
    "classifierURF.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9323459809695448\n",
      "0.8937272994955545\n",
      "0.6123141786653334\n",
      "0.6612865671036048\n",
      "0.8791528584553399\n"
     ]
    }
   ],
   "source": [
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierURF ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = classifierURF.predict(X_test)\n",
    "classifierURF.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 99.97%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 99.97%\n",
      "\t\t\tRecall Score: 99.66%\n",
      "\t\t\tF1 score: 99.81%\n",
      "\t\t\tAUC score: 99.83%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[35381     1]\n",
      " [   11  3215]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      1.000     1.000     1.000     35382\n",
      "           1      1.000     0.997     0.998      3226\n",
      "\n",
      "    accuracy                          1.000     38608\n",
      "   macro avg      1.000     0.998     0.999     38608\n",
      "weighted avg      1.000     1.000     1.000     38608\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 50.00%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 0.00%\n",
      "\t\t\tRecall Score: 0.00%\n",
      "\t\t\tF1 score: 0.00%\n",
      "\t\t\tAUC score: 50.00%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[200   0]\n",
      " [200   0]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.500     1.000     0.667       200\n",
      "           1      0.000     0.000     0.000       200\n",
      "\n",
      "    accuracy                          0.500       400\n",
      "   macro avg      0.250     0.500     0.333       400\n",
      "weighted avg      0.250     0.500     0.333       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.500     1.000     0.667       200\n",
      "           1      0.000     0.000     0.000       200\n",
      "\n",
      "    accuracy                          0.500       400\n",
      "   macro avg      0.250     0.500     0.333       400\n",
      "weighted avg      0.250     0.500     0.333       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierURF, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierURF, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "#AUC\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(criterion='entropy', random_state=0)"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train\n",
    "classifierUD = DecisionTreeClassifier(criterion = 'entropy' , random_state = 0)\n",
    "classifierUD.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8991917956639742\n",
      "0.6750656767299908\n",
      "0.6855792729835186\n",
      "0.6798959799324831\n",
      "0.6855792729835186\n"
     ]
    }
   ],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUD ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.57"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#test\n",
    "y_pred = classifierUD.predict(X_test)\n",
    "classifierUD.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 99.98%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 100.00%\n",
      "\t\t\tRecall Score: 99.81%\n",
      "\t\t\tF1 score: 99.91%\n",
      "\t\t\tAUC score: 99.91%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[35382     0]\n",
      " [    6  3220]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      1.000     1.000     1.000     35382\n",
      "           1      1.000     0.998     0.999      3226\n",
      "\n",
      "    accuracy                          1.000     38608\n",
      "   macro avg      1.000     0.999     0.999     38608\n",
      "weighted avg      1.000     1.000     1.000     38608\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 57.00%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 68.92%\n",
      "\t\t\tRecall Score: 25.50%\n",
      "\t\t\tF1 score: 37.23%\n",
      "\t\t\tAUC score: 57.00%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[177  23]\n",
      " [149  51]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.543     0.885     0.673       200\n",
      "           1      0.689     0.255     0.372       200\n",
      "\n",
      "    accuracy                          0.570       400\n",
      "   macro avg      0.616     0.570     0.523       400\n",
      "weighted avg      0.616     0.570     0.523       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.543     0.885     0.673       200\n",
      "           1      0.689     0.255     0.372       200\n",
      "\n",
      "    accuracy                          0.570       400\n",
      "   macro avg      0.616     0.570     0.523       400\n",
      "weighted avg      0.616     0.570     0.523       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierUD, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUD, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(random_state=0)"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train\n",
    "classifierUS = SVC(kernel='rbf', random_state = 0)\n",
    "classifierUS.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9164940623692596\n",
      "0.4582470311846298\n",
      "0.5\n",
      "0.478213845686631\n",
      "0.6958644440215284\n"
     ]
    }
   ],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUS ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#test\n",
    "y_pred = classifierUS.predict(X_test)\n",
    "classifierUS.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 91.64%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 0.00%\n",
      "\t\t\tRecall Score: 0.00%\n",
      "\t\t\tF1 score: 0.00%\n",
      "\t\t\tAUC score: 50.00%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[35382     0]\n",
      " [ 3226     0]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.916     1.000     0.956     35382\n",
      "           1      0.000     0.000     0.000      3226\n",
      "\n",
      "    accuracy                          0.916     38608\n",
      "   macro avg      0.458     0.500     0.478     38608\n",
      "weighted avg      0.840     0.916     0.876     38608\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 50.00%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 0.00%\n",
      "\t\t\tRecall Score: 0.00%\n",
      "\t\t\tF1 score: 0.00%\n",
      "\t\t\tAUC score: 50.00%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[200   0]\n",
      " [200   0]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.500     1.000     0.667       200\n",
      "           1      0.000     0.000     0.000       200\n",
      "\n",
      "    accuracy                          0.500       400\n",
      "   macro avg      0.250     0.500     0.333       400\n",
      "weighted avg      0.250     0.500     0.333       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.500     1.000     0.667       200\n",
      "           1      0.000     0.000     0.000       200\n",
      "\n",
      "    accuracy                          0.500       400\n",
      "   macro avg      0.250     0.500     0.333       400\n",
      "weighted avg      0.250     0.500     0.333       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbkAAAGDCAYAAAC2tW7jAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd5xU1fnH8c8jiCB2RSOiUgRpItFFRFAQRbFiwSAqKEU0Yu8/TaKxJJrEEgULwR4BGyIWrFjoVUGKKALKUiKCiHR29/n9cUZclt1ll907d3bm+369eGVn5u7Msze4X845zz3X3B0REZF0tEPcBYiIiERFISciImlLISciImlLISciImlLISciImlLISciImlLISciImlLISdSBma2wMzWmdlqM1tqZs+a2S4FjjnGzEaa2S9m9rOZvWlmjQscs5uZPWxm3yfea27i8T5FfK6Z2dVmNsPM1phZtpm9YmaHRfnzilQ0CjmRsjvD3XcBmgO/B/7v1xfMrBXwPvAGUBOoA0wDxphZ3cQxVYCPgCZAR2A34BhgOXBUEZ/5b+Aa4GpgL6ABMAw4rbTFm1nl0n6PSEVh2vFEZPuZ2QKgt7t/mHj8D6CJu5+WeDwK+NLdryjwfSOAZe7e3cx6A/cC9dx9dQk+sz7wFdDK3ScWccwnwH/dfWDi8SWJOtskHjtwJXAtUBl4D1jt7jfme483gE/d/UEzqwk8ChwHrAYecvdHSnCKRGKlkZxIOTGzWsApwNzE450JI7JXCjn8ZaBD4usTgXdLEnAJJwDZRQVcKZwFtAQaA4OALmZmAGa2J3ASMMTMdgDeJIxAD0h8/rVmdnIZP18kcgo5kbIbZma/AAuBH4A7Es/vRfhvbEkh37ME+HW9be8ijilKaY8vyt/dfYW7rwNGAQ4cm3itMzDO3RcDLYAa7n6Xu29093nAf4Dzy6EGkUgp5ETK7ix33xVoBzTkt/D6CcgD9i/ke/YHfkx8vbyIY4pS2uOLsvDXLzysWwwBuiaeugB4MfH1wUBNM1v56x/gNmC/cqhBJFIKOZFy4u6fAs8C/0o8XgOMA84r5PA/EJpNAD4ETjaz6iX8qI+AWmaWVcwxa4Cd8z3+XWElF3g8GOhsZgcTpjFfSzy/EJjv7nvk+7Oru59awnpFYqOQEylfDwMdzKx54vGtwMWJdv9dzWxPM7sHaAX8NXHMC4Qgec3MGprZDma2t5ndZmZbBYm7fwM8Bgw2s3ZmVsXMqprZ+WZ2a+KwL4BzzGxnMzsE6LWtwt39c2AZMBB4z91XJl6aCKwys1vMrJqZVTKzpmbWYntOkEgyKeREypG7LwOeB/6ceDwaOBk4h7CO9h3hMoM2ibDC3TcQmk++Aj4AVhGCZR9gQhEfdTXQD+gPrAS+Bc4mNIgAPARsBP4HPMdvU4/bMjhRy6B8P1MucAbhEon5hGnWgcDuJXxPkdjoEgIREUlbGsmJiEjaUsiJiEjaUsiJiEjaUsiJiEjaUsiJiEjaqnC7j++zzz5eu3btuMsQEZEUMmXKlB/dvUbB5ytcyNWuXZvJkyfHXYaIiKQQM/uusOc1XSkiImlLISciImlLISciImlLISciImlLISciImlLISciImlLISciImlLISciImlLISciImkrspAzs6fN7Aczm1HE62Zmj5jZXDObbmZHRFWLiIhkpihHcs8CHYt5/RSgfuJPH+DxCGsREZEMFFnIuftnwIpiDukEPO/BeGAPM9s/qnpERCS1fPstjBsX7WfEuSZ3ALAw3+PsxHNbMbM+ZjbZzCYvW7YsKcWJiEh0Pv0UjjoKevSA3NzoPifOkLNCnvPCDnT3Ae6e5e5ZNWpsdScFERGpQJ56Ck48EfbdF95+GypViu6z4gy5bODAfI9rAYtjqkVERCKWmwvXXw+9e8MJJ8D48VCvXrSfGWfIDQe6J7osjwZ+dvclMdYjIiIRWbUKzjwTHnoIrrkG3noLdt89+s+N7KapZjYYaAfsY2bZwB3AjgDu/gTwDnAqMBdYC/SIqhYREYnPvHlwxhnw9dfwxBNw2WXJ++zIQs7du27jdQf6RvX5IiISv1Gj4JxzwlTl++/D8ccn9/O144mIiETimWfC2tvee8OECckPOFDIiYhIOcvNhZtugp49oV270GBSv348tUQ2XSkiIpln1Sq48MLQWHLllaHRpHKMSaOQExGRcrFgQWgwmT0bHnsM/vjHuCtSyImISDkYPTo0mGzaBO++Gy72TgVakxMRkTJ57rnQYLLHHqHBJFUCDhRyIiKynXJz4ZZb4JJL4NhjQ8A1aBB3VVvSdKWIiJTa6tWhwWT48LD29u9/w447xl3V1hRyIiJSKt99F7bomjkT+vWDvim8rYdCTkRESmzcODjrLNiwAd55B046Ke6Kiqc1ORERKZEXXggXd++2W7jAO9UDDhRyIiKyDXl5cNtt0L07tG4dGkwaNoy7qpLRdKWIiBRp9Wro1g2GDYM+fcIaXCo2mBRFISciIoVauDA0mEyfHronr7oKzOKuqnQUciIispXx40ODybp18Pbb0LFj3BVtH63JiYjIFgYNCg0m1auHsKuoAQcKORERScjLgz/9KVzkffTRMHEiNGoUd1Vlo+lKERFhzZrQPTl0KPTuDf37Q5UqcVdVdgo5EZEMl50dGkymTQv3f7vmmorXYFIUhZyISAabOBE6dQojuTffhFNPjbui8qU1ORGRDDVkCLRtC9Wqhe260i3gQCEnIpJx8vLgjjuga1do0SKM5po0ibuqaGi6UkQkg6xdG+7/9sor0KMHPPFEejSYFEUhJyKSIRYtCutvU6fCv/4F11+fPg0mRVHIiYhkgMmTQwflL7+EG52efnrcFSWH1uRERNLcyy/DccfBTjvB2LGZE3CgkBMRSVvucNdd0KULHHFEuEXOYYfFXVVyabpSRCQNrVsHPXuGywQuvhiefDKM5DKNQk5EJM0sXhzuIDB5Mtx/P9x0U/o3mBRFIScikkamTg0NJitXhhudnnlm3BXFS2tyIiJp4rXXoE0bqFQpNJhkesCBQk5EpMJzh3vugc6doXnzsINJs2ZxV5UaNF0pIlKBrVsXbo0zaBB06wYDBkDVqnFXlToUciIiFdTSpaHBZMIE+Pvf4ZZbMrfBpCgKORGRCujzz8Oa24oV4UanZ58dd0WpSWtyIiIVzOuvhwYTgDFjFHDFUciJiFQQ7mFa8pxzws4lkyaFRhMpmqYrRUQqgPXr4dJL4b//hQsugKeeUoNJSWgkJyKS4v73P2jfPgTcPfeE/1XAlYxGciIiKWzatNBg8uOP8OqrcO65cVdUsWgkJyKSot54A1q3htxcGDVKAbc9FHIiIinGPWysfPbZ0LhxaDA54oi4q6qYNF0pIpJCNmyAPn3g+efDfeCeeQaqVYu7qopLIzkRkRTxww9wwgkh4P76Vxg8WAFXVhrJiYikgC+/hDPOCEH38stw3nlxV5QeNJITEYnZW2/BMcfApk3w2WcKuPKkkBMRiYk7/Otf4RKBhg3DLXKysuKuKr0o5EREYrBhA/TqBTfdFO4D9+mncMABcVeVfhRyIiJJtmwZdOgQOifvuAOGDIGdd467qvSkxhMRkSSaMSM0mCxdGsKtS5e4K0pvkY7kzKyjmc0xs7lmdmshr+9uZm+a2TQzm2lmPaKsR0QkTu+8ExpMNmwIDSYKuOhFFnJmVgnoD5wCNAa6mlnjAof1BWa5++FAO+ABM6sSVU0iInFwh4ceCiO4+vVDg0mLFnFXlRmiHMkdBcx193nuvhEYAnQqcIwDu5qZAbsAK4CcCGsSEUmqjRvDLXKuvz5s0/XZZ1CrVtxVZY4oQ+4AYGG+x9mJ5/LrBzQCFgNfAte4e16ENYmIJM2PP4YGk6eegj/9KVzkXb163FVlligbT6yQ57zA45OBL4D2QD3gAzMb5e6rtngjsz5AH4CDDjooglJFRMrXrFlhenLRInjxxXCjU0m+KEdy2cCB+R7XIozY8usBDPVgLjAfaFjwjdx9gLtnuXtWjRo1IitYRKQ8vPsutGoFa9aE698UcPGJMuQmAfXNrE6imeR8YHiBY74HTgAws/2AQ4F5EdYkIhIZd/j3v+G006Bu3XCLnJYt464qs0U2XenuOWZ2JfAeUAl42t1nmtnlidefAO4GnjWzLwnTm7e4+49R1SQiEpVNm+DKK2HAgNBg8sILWn9LBZFeDO7u7wDvFHjuiXxfLwZOirIGEZGoLV8eNlX++GO47Ta4+27YQftJpQTteCIiUgZffQWnnw4LF4bR20UXxV2R5KeQExHZTu+/D3/4A+y0E3zySWg2kdSiAbWISCm5Q79+cOqpcPDBYQcTBVxqUsiJiJTCpk3Qty9cdVXoohwzJgSdpCaFnIhICa1YAaecAo8/DrfcAq+/DrvsEndVUhytyYmIlMCcOWEHk+++g+eeg+7d465ISkIhJyKyDR9+GC4R2HFHGDkSWreOuyIpKU1XiogU47HHoGPHcOeAiRMVcBWNQk5EpBA5OWEHk759wzrc2LFQu3bcVUlpKeRERAr46adweUD//nDTTTBsGOy6a9xVyfbQmpyISD7ffBMaTObNg6efhh494q5IykIhJyKSMHIkdO4MlSrBRx/BscfGXZGUlaYrRUSAJ5+Ek0+GmjVDg4kCLj0o5EQko+XkwDXXwOWXw0knhQaTOnXirkrKi0JORDLWypXhDgKPPALXXw/Dh8Nuu8VdlZQnrcmJSEaaOzc0mMydCwMHQq9ecVckUVDIiUjG+eQTOPfc8PWHH0LbtrGWIxHSdKWIZJSBA6FDB9hvv9BgooBLbwo5EckIublw3XVw6aVw4okwbhzUqxd3VRI1hZyIpL2ffw7rbw8/DNdeC2++CbvvHndVkgxakxORtDZvXgi4r78O18L16RN3RZJMCjkRSVuffQbnnAN5efD++3D88XFXJMmm6UoRSUtPPx3W3vbZByZMUMBlKoWciKSV3Fy48cZw3dvxx8P48VC/ftxVSVw0XSkiaWPVKrjgAnj7bbjqKnjwQais33IZTf/3i0hamD8/NJh89RU8/njYi1JEISciFd7o0XD22WGz5ffegxNOiLsiSRVakxORCu3ZZ6F9e9hrr9BgooCT/BRyIlIh5ebCzTeHO3cfd1xoMGnQIO6qJNVoulJEKpxffoGLLgq3xrniirCTyY47xl2VpCKFnIhUKN99FxpMZs2Cfv2gb9+4K5JUppATkQpj7NjQYLJhA4wYEe4mIFIcrcmJSIXwwgvh4u7ddgvrbwo4KQmFnIiktLw8+L//g+7doXXr0EHZsGHcVUlFoelKEUlZq1dDt24wbBhcdhk8+qgaTKR0FHIikpK+/x7OPBO+/BIeeQSuvBLM4q5KKhqFnIiknHHjQoPJunVhH8qOHeOuSCoqrcmJSEp58cXQYLLLLqHBRAEnZaGQE5GUkJcHt98eLvI++ujQYNKoUdxVSUWn6UoRid2aNaF7cuhQuPTScJF3lSpxVyXpQCEnIrFauBA6dYJp0+Chh+Caa9RgIuVHIScisZk4MQTcmjXw1ltwyilxVyTpRmtyIhKLIUOgbVuoVi10UyrgJAoKORFJqrw8+MtfoGtXaNEijOaaNIm7KklXmq4UkaRZuxYuvhhefRV69oTHH1eDiURLISciSbFoUVh/mzoVHngArrtODSYSPYWciERu8uSwRdcvv4QbnZ5+etwVSabQmpyIROrll+HYY2GnnUKDiQJOkkkhJyKRcIe//hW6dIEjjww7mDRtGndVkmk0XSki5W7dOujRA156KTSaPPlkGMmJJJtCTkTK1eLFocFkyhT4xz/gxhvVYCLxiXS60sw6mtkcM5trZrcWcUw7M/vCzGaa2adR1iMi0ZoyJVz79tVX4UanN92kgJN4RTaSM7NKQH+gA5ANTDKz4e4+K98xewCPAR3d/Xsz2zeqekQkWq++GjZZrlEDxoyBZs3irkgk2pHcUcBcd5/n7huBIUCnAsdcAAx19+8B3P2HCOsRkQi4wz33wHnnwe9/H3YwUcBJqogy5A4AFuZ7nJ14Lr8GwJ5m9omZTTGz7oW9kZn1MbPJZjZ52bJlEZUrIqW1bh1ceCH8+c/QrRt89BHst1/cVYn8JsqQK2wm3gs8rgwcCZwGnAz82cwabPVN7gPcPcvds2rUqFH+lYpIqS1ZAu3ahY2W//53eO45qFo17qpEthRld2U2cGC+x7WAxYUc86O7rwHWmNlnwOHA1xHWJSJl9PnnYQeTn34KNzo966y4KxIpXJQjuUlAfTOrY2ZVgPOB4QWOeQM41swqm9nOQEtgdoQ1iUgZDR0KbdqErsnRoxVwktoiCzl3zwGuBN4jBNfL7j7TzC43s8sTx8wG3gWmAxOBge4+I6qaRGT7ucPf/gbnnguHHRYaTJo3j7sqkeKZe8FlstSWlZXlkydPjrsMkYyyfj307g0vvhgaTQYO1PqbpBYzm+LuWQWf196VIlKs//0Pjj8+BNy998ILLyjgpOLQtl4iUqRp0+CMM2D5cnjtNTjnnLgrEikdjeREpFDDhkHr1pCXB6NGKeCkYlLIicgW3OH++0OoNWkCkybBEUfEXZXI9tF0pYhstmED9OkDzz8P558PTz8N1arFXZXI9tNITkQA+OEHaN8+BNxdd8GgQQo4qfg0khMRpk8PO5j88AO88gp07hx3RSLlQyM5kQz35puhwWTTptBgooCTdKKQE8lQ7vDPf4a7eDdsGBpMjjwy7qpEypdCTiQDbdgAPXvCzTeH+8B9+inUrBl3VSLlTyEnkmGWLYMTT4Rnn4U77wy3ytl557irEomGGk9EMsiMGWEHk6VLQ7h16RJ3RSLR0khOJEO8/Ta0ahWmKj/7TAEnmUEhJ5Lm3OHBB8MIrkGD0GDSokXcVYkkh0JOJI1t3BhukXPDDeE+cJ99BgccEHdVIsmjkBNJUz/+CB06hK25/vxneOklqF497qpEkqvUIWdmlczswiiKEZHyMWsWtGwJEyaE+8DddRfsoH/SSgYq8q+9me1mZv9nZv3M7CQLrgLmAX9IXokiUhojRoQGk7Vrw/VvF1wQd0Ui8Snu33YvAIcCXwK9gfeBzkAnd++UhNpEpBTc4eGH4fTToW5dmDgxjOZEMllx18nVdffDAMxsIPAjcJC7/5KUykSkxDZuhCuvhP/8B84+G154QetvIlD8SG7Tr1+4ey4wXwEnknqWL4eTTw4Bd9tt8OqrCjiRXxU3kjvczFYBlnhcLd9jd/fdIq9ORIo1e3a4/i07O4zeLroo7opEUkuRIefulZJZiIiUznvvwR/+AFWrwscfh2YTEdlScd2VVc3s2kR3ZR8z0z6XIinAHR59FE49FerUCTuYKOBEClfcmtxzQBahu/JU4IGkVCQiRdq0Ca64Aq6+OkxTjh4NBx0Ud1Uiqau40VnjfN2VTwETk1OSiBRmxYpw77eRI+HWW+Hee3WBt8i2FBdy+bsrc8ysmENFJEpz5oTr377/Hp57Drp3j7sikYqhuJBrnuimhNBRqe5KkRh88EEYwVWpEkZxrVvHXZFIxVHcZMc0d98t8WdXd6+c72sFnEgS9O8Pp5wS1t0mTlTAiZRWcSHnSatCRLaQkwN9+4ZdTE49FcaMgdq1465KpOIpbrpyXzO7vqgX3f3BCOoRyXg//RSuf/vwQ7jpJvj736GSrloV2S7FhVwlYBd+2/FERCL2zTehwWT+fHjmGbjkkrgrEqnYigu5Je5+V9IqEclwI0dC585h1DZyJLRpE3dFIhVfcWtyGsGJJMkTT8BJJ0HNmqHBRAEnUj6KC7kTklaFSIbKyQm7l/zxj+FOAmPHhq26RKR8FBly7r4imYWIZJqVK+G008I+lNdfD8OHw266OEekXGnTZZEYzJ0b9p789lsYOBB69Yq7IpH0pJATSbKPPw4NJmZhN5O2beOuSCR9aXtXkST6z39Cg8nvfhcaTBRwItFSyIkkQU4OXHcd9OkDJ54YGkzq1o27KpH0p5ATidjPP4f1t4cfhmuvhTffhN13j7sqkcygNTmRCH37bQi4b76BJ58MIzkRSR6FnEhEPv0Uzj0X3OH99+H44+OuSCTzaLpSJAJPPQUdOkCNGjBhggJOJC4KOZFylJsLN9wAvXuHYBs3Dg45JO6qRDKXpitFysmqVdC1K7zzDlx1FTz4IFTWf2EisdJ/giLlYP780GDy1Vfw+ONw+eVxVyQioJATKbNRo+Ccc8JU5XvvwQna2lwkZWhNTqQMnn02hNpee8H48Qo4kVSjkBPZDrm5cPPN0KNH2Jpr/Hho0CDuqkSkoEhDzsw6mtkcM5trZrcWc1wLM8s1s85R1iNSHn75Bc4+G/75T+jbNzSa7Lln3FWJSGEiW5Mzs0pAf6ADkA1MMrPh7j6rkOPuB96LqhaR8rJgAZx5JsyaBf36hZATkdQVZePJUcBcd58HYGZDgE7ArALHXQW8BrSIsBaRMhszJozgNm2CESPCxd4iktqinK48AFiY73F24rnNzOwA4GzgieLeyMz6mNlkM5u8bNmyci9UZFuefx7at4c99gjrbwo4kYohypCzQp7zAo8fBm5x99zi3sjdB7h7lrtn1ahRo9wKFNmWvDy49Va4+GJo0yYE3KGHxl2ViJRUlNOV2cCB+R7XAhYXOCYLGGJmAPsAp5pZjrsPi7AukRJZvRouugjeeCNc3P3II7DjjnFXJSKlEWXITQLqm1kdYBFwPnBB/gPcvc6vX5vZs8BbCjhJBd9/HxpMvvwSHn00NJhYYXMTIpLSIgs5d88xsysJXZOVgKfdfaaZXZ54vdh1OJG4jBsHZ50FGzaEywNOPjnuikRke0W6rZe7vwO8U+C5QsPN3S+JshaRknjxRejVC2rVgk8+gUaN4q5IRMpCO56IEBpMbr89rMG1ahXuAaeAE6n4tEGzZLw1a6BbN3j9dbj00nCRd5UqcVclIuVBIScZbeHC0GAyfTo89BBcc40aTETSiUJOMtaECdCpE6xbB2+9BaecEndFIlLetCYnGWnw4HD3gOrVQzelAk4kPSnkJKPk5cGf/wwXXAAtW4bRXOPGcVclIlHRdKVkjDVrwvZcr70GPXvC44+rwUQk3SnkJCNkZ4f1t88/hwcegOuuU4OJSCZQyEnamzQpBNwvv8Cbb8Jpp8VdkYgki9bkJK299BIcdxzstFNoMFHAiWQWhZykJXe48044/3zIyoKJE6Fp07irEpFk03SlpJ21a6FHD3j5ZbjkEnjiiTCSE5HMo5CTtLJoUbiDwJQp8I9/wI03qsFEJJMp5CRtTJkStuhatSrc6PSMM+KuSETipjU5SQuvvALHHguVK8OYMQo4EQkUclKhucPdd8Mf/gC//324XKBZs7irEpFUoelKqbDWrQs7lwwZEm6VM2AAVK0ad1UikkoUclIhLVkSLvCePBnuuw9uvlkNJiKyNYWcVDiffx7W3FauhKFDQzeliEhhtCYnFcrQodCmDeywQ2gwUcCJSHEUclIhuMO998K554bGkokT4fDD465KRFKdpisl5a1fD717w4svwoUXwsCBajARkZLRSE5S2tKl0K5dCLh774UXXlDAiUjJaSQnKeuLL8IOJsuXhxudnnNO3BWJSEWjkZykpGHDoHXrsBY3erQCTkS2j0JOUop7uO7t7LPDrXEmTgw7mYiIbA9NV0rKWL8e+vQJ627nnw9PPw3VqsVdlYhUZBrJSUr43/+gffsQcHffDYMGKeBEpOw0kpPYTZ8edjBZtizcTaBz57grEpF0oZGcxGr4cDjmGMjNhVGjFHAiUr4UchIL93Dn7rPOgkaNQoPJkUfGXZWIpBuFnCTdhg3Qowfccgucdx58+inUrBl3VSKSjhRyklTLlsEJJ8Bzz8Gdd4Z7we28c9xViUi6UuOJJM2MGaHBZOlSeOmlcDdvEZEoaSQnSfHWW9CqVZiq/OwzBZyIJIdCTiLlDg88EPagbNAAJk2CFi3irkpEMoVCTiKzcSP06gU33hjuAzdqFBxwQNxViUgmUchJJH78EU48EZ55Bv7yl7AGpwYTEUk2NZ5IuZs5MzSYLF4ctufq2jXuikQkUynkpFyNGAFdukD16uH6t5Yt465IRDKZpiulXLjDQw/B6adDvXphBxMFnIjETSEnZbZxI1x2GVx/fdima/RoOPDAuKsSEVHISRktXw4nnQT/+Q/cfnu4i0D16nFXJSISaE1Ottvs2aHBJDsb/vtfuPDCuCsSEdmSQk62y7vvhgaTatXgk0/g6KPjrkhEZGuarpRScYdHHoHTToM6dUKDiQJORFKVQk5KbNMm+OMf4ZprwjTl6NFw0EFxVyUiUjSFnJTIihXQsSM8+STceisMHQq77BJ3VSIixdOanGzTnDnh+rfvv4fnn4du3eKuSESkZBRyUqwPPgh3765SBT7+GI45Ju6KRERKLtLpSjPraGZzzGyumd1ayOsXmtn0xJ+xZnZ4lPVI6fTvD6ecEtbdJk1SwIlIxRNZyJlZJaA/cArQGOhqZo0LHDYfaOvuzYC7gQFR1SMlt2kT9O0LV14Jp54KY8bAwQfHXZWISOlFOZI7Cpjr7vPcfSMwBOiU/wB3H+vuPyUejgdqRViPlMBPP4XR22OPwc03w+uvw667xl2ViMj2iXJN7gBgYb7H2UBxW/b2AkZEWI9sw9dfh0sD5s8P94G75JK4KxIRKZsoQ84Kec4LPdDseELItSni9T5AH4CDdGFWJD76CDp3hsqVYeRIaFPo/xMiIhVLlNOV2UD+vehrAYsLHmRmzYCBQCd3X17YG7n7AHfPcvesGjVqRFJsJnv8cTj5ZKhVK+xgooATkXQRZchNAuqbWR0zqwKcDwzPf4CZHQQMBbq5+9cR1iKFyMmBq66CK64IF3qPGRO26hIRSReRTVe6e46ZXQm8B1QCnnb3mWZ2eeL1J4C/AHsDj5kZQI67Z0VVk/xm5cqwwfL778MNN8D990OlSnFXJSJSvsy90GWylJWVleWTJ0+Ou4wKbe7csIPJvHnwxBPQs2fcFYmIlI2ZTSlskKQdTzLMxx/DuefCDjvAhx/CccfFXZGISHS0QXMGGTAg3MV7//1Dg4kCTkTSnUIuA+TkwLXXwnxp0gQAABHBSURBVGWXQYcOMHYs1K0bd1UiItHTdGWa+/nn0GDy3ntw3XXwz3+qwUREModCLo19+23YweSbb8JU5aWXxl2RiEhyKeTS1KefwjnnhK8/+ADatYu1HBGRWGhNLg099RSceCLsuy9MmKCAE5HMpZBLI7m5cP310Ls3nHACjBsHhxwSd1UiIvHRdGWaWLUKunaFd96Bq6+GBx4Imy2LiGQy/RpMA/PmhQaTOXPCZsuXXx53RSIiqUEhV8GNGhUaTHJzwz6U7dvHXZGISOrQmlwF9swzYe1tr71Cg4kCTkRkSwq5Cig3F266KWys3LYtjB8P9evHXZWISOrRdGUF88svcMEF8NZb0LcvPPywGkxERIqiX48VyIIFocFk9mzo3z/c7FRERIqmkKsgRo8ODSabNsG774aLvUVEpHhak6sAnnsuNJjssUdYf1PAiYiUjEIuheXmwi23wCWXwLHHhoA79NC4qxIRqTg0XZmiVq+GCy+E4cPDxd2PPAI77hh3VSIiFYtCLgV99x2ceSbMmAGPPhq6KM3irkpEpOJRyKWYcePgrLNgwwYYMQJOOinuikREKi6tyaWQ//433BZn113D+psCTkSkbBRyKSAvD267Dbp1g2OOCVt0NWwYd1UiIhWfpitjtnp1CLdhw+DSS6FfP6hSJe6qRETSg0IuRgsXhgaT6dPD9lxXX60GExGR8qSQi8n48aHBZN06ePtt6Ngx7opERNKP1uRiMGhQaDCpXj10UyrgRESioZBLorw8+NOfwkXeLVuGBpPGjeOuSkQkfWm6MknWrIHu3WHoUOjVCx57TA0mIiJRU8glQXZ2aDCZNg0efBCuvVYNJiIiyaCQi9jEidCpUxjJDR8Op50Wd0UiIplDa3IRGjIE2raFatVCg4kCTkQkuRRyEcjLgzvugK5dISsrNJg0aRJ3VSIimUfTleVs7dpw/7dXXoEePeDxx2GnneKuSkQkMynkytGiRWH9bepU+Oc/4YYb1GAiIhInhVw5mTw5dFD+8gu88QaccUbcFYmIiNbkysHLL8Nxx4Xr3saOVcCJiKQKhVwZuMNdd0GXLvD734fLBQ47LO6qRETkV5qu3E7r1kHPnuEyge7dYcAANZiIiKQahdx2WLw43EFg8mS4/3646SY1mIiIpCKFXClNnRoaTFauhNdfD92UIiKSmrQmVwqvvQZt2sAOO8CYMQo4EZFUp5ArAXe45x7o3BkOPzw0mBx+eNxViYjItmi6chvWrYPevcONTi+6CP7zH6haNe6qRESkJBRyxVi6NDSYTJgAf/sb3HqrGkxERCoShVwRvvgiNJgsXx5udHr22XFXJCIipaWQK8Trr4epyb32gtGjw4XeIhKtTZs2kZ2dzfr16+MuRVJY1apVqVWrFjvuuGOJjlfI5eMO990Ht90GLVuGsNt//7irEskM2dnZ7LrrrtSuXRvTuoAUwt1Zvnw52dnZ1KlTp0Tfo+7KhPXrw84lt90W7gP38ccKOJFkWr9+PXvvvbcCTopkZuy9996lGu0r5ID//Q/at4f//hfuvhtefDHczVtEkksBJ9tS2r8jGT9dOW1aaDBZtgxefRXOPTfuikREpLxEOpIzs45mNsfM5prZrYW8bmb2SOL16WZ2RJT1FDR8OLRuDbm5ocFEASeS2SpVqkTz5s1p2rQpZ5xxBitXrtz82syZM2nfvj0NGjSgfv363H333bj75tdHjBhBVlYWjRo1omHDhtx4442FfsawYcO46667tnju8MMPp2vXrls8165dOyZPnrz58YIFC2jatOnmxxMnTuS4447j0EMPpWHDhvTu3Zu1a9eW6eefP38+LVu2pH79+nTp0oWNGzcWetyv56l58+aceeaZ2/x+d+fqq6/mkEMOoVmzZkydOnXz97z77rsceuihHHLIIdx3332bn7/xxhsZOXJkmX6ezR8exR+gEvAtUBeoAkwDGhc45lRgBGDA0cCEbb3vkUce6WWVl+d+//3uZu4tWrgvWlTmtxSRMpo1a1bcJXj16tU3f929e3e/55573N197dq1XrduXX/vvffc3X3NmjXesWNH79evn7u7f/nll163bl2fPXu2u7tv2rTJ+/fvX+hntGrVypctW7b58axZs7xp06Zes2ZNX7169ebn27Zt65MmTdr8eP78+d6kSRN3d1+6dKkfdNBBPnbsWHd3z8vL81deecWXLl1app//vPPO88GDB7u7+2WXXeaPPfZYocflP08l+f63337bO3bs6Hl5eT5u3Dg/6qij3N09JyfH69at699++61v2LDBmzVr5jNnznR39wULFniHDh0K/ZzC/q4Ak72QzIhyuvIoYK67zwMwsyFAJ2BWvmM6Ac8nChxvZnuY2f7uviSqonJzoVcveO65cB+4Z57R+ptIqrn22nCtanlq3hwefrjkx7dq1Yrp06cDMGjQIFq3bs1JJ50EwM4770y/fv1o164dffv25R//+Ae33347DRs2BKBy5cpcccUVW73n119/zU477cQ+++yz+blBgwbRrVs3Zs+ezfDhw7ca0RWmf//+XHzxxbRq1QoI61SdO3cu+Q9XCHdn5MiRDBo0CICLL76YO++8kz/+8Y9l/v433niD7t27Y2YcffTRrFy5kiVLlrBgwQIOOeQQ6tatC8D555/PG2+8QePGjTn44INZvnw5S5cu5Xe/+912/1xRTlceACzM9zg78Vxpj8HM+pjZZDObvGzZsjIVtcMOUL063HknDB6sgBORreXm5vLRRx9tnoqbOXMmRx555BbH1KtXj9WrV7Nq1SpmzJix1euFGTNmDEccseWqzEsvvUSXLl3o2rUrgwcPLlF9Jf28OXPmbJ5WLPgn/1QswPLly9ljjz2oXDmMfWrVqsWiRYsKfd/169eTlZXF0UcfzbBhw7b5/YsWLeLAAw/c/P2/vlbU87864ogjGDNmTElOSZGiHMkV1gLj23EM7j4AGACQlZW11eulKsqgXz9tzyWSykoz4ipP69ato3nz5ixYsIAjjzySDh06AGGUUlRXX2m6/ZYsWUKNGjU2P540aRI1atTg4IMPplatWvTs2ZOffvqJPffcs9D3LW1n4aGHHsoXJRwSu2/9q7Woz/v++++pWbMm8+bNo3379hx22GHstttuRX5/Ue+9rc/cd999Wbx4cYnqL0qUI7ls4MB8j2sBBastyTHlTgEnIoWpVq0aX3zxBd999x0bN26kf//+ADRp0mSLJhCAefPmscsuu7DrrrvSpEkTpkyZUqL3z3+N1+DBg/nqq6+oXbs29erVY9WqVbz22msA7L333vz000+bj12xYsXmac6Sfl5pRnL77LMPK1euJCcnBwgX59esWbPQ9/31+bp169KuXTs+//zzYr+/Vq1aLFz426Tdr68V9fyv1q9fT7WyTrcVtlBXHn8Io8R5QB1+azxpUuCY09iy8WTitt63PBpPRCT1pFrjydSpU/3AAw/0jRs3+tq1a71OnTr+wQcfuHtoRDnttNP8kUcecXf3adOmeb169XzOnDnu7p6bm+sPPPDAVu8/YsQIv/DCCzcfU6tWLc/Ozt78+siRI719+/bu7v7oo4969+7dPS8vz93dr776av/rX//q7r81nowfP37z977wwgu+ZMmSMv38nTt33qJxpLDmmRUrVvj69evd3X3ZsmV+yCGHbG4WKer733rrrS0aT1q0aOHuoUGnTp06Pm/evM2NJzNmzNj8WaeffrqPGzduqxpK03gSWcj5b92TXxO6LG9PPHc5cHniawP6J17/Esja1nsq5ETSU6qFnHv4Jfv888+7u/v06dO9bdu23qBBA69Xr57feeedmwPI3f3NN9/0I444whs2bOiNGjXyG2+8cav3X7NmjTdu3Njz8vL8448/9pYtW27xek5Ojv/ud7/zxYsX+4YNG7xv375+2GGHebNmzbxnz56+Zs2azceOHTvW27Rp4w0aNPCGDRt6nz59tnh9e3z77bfeokULr1evnnfu3HlzmE2aNMl79erl7u5jxozxpk2berNmzbxp06Y+cODAbX5/Xl6eX3HFFV63bl1v2rTpFl2jb7/9ttevX9/r1q27uZvV3X3jxo3esGFD37Rp01Z1libkzAuZE01lWVlZXnDaQEQqvtmzZ9OoUaO4y4jcNddcwxlnnMGJJ54Ydykp7fXXX2fq1KncfffdW71W2N8VM5vi7lkFj9W2XiIiSXTbbbeV+aLtTJCTk8MNN9xQ5vfJ+G29RESSab/99ttilxAp3HnnnVcu76ORnIikjIq2fCLJV9q/Iwo5EUkJVatWZfny5Qo6KZJ7uJ9c1apVS/w9mq4UkZRQq1YtsrOzKeuuRpLefr0zeEkp5EQkJey4444lvtuzSElpulJERNKWQk5ERNKWQk5ERNJWhdvxxMyWAd+Vw1vtA/xYDu+TjnRuiqZzUzSdm6Lp3BStvM7Nwe5eo+CTFS7kyouZTS5sCxjRuSmOzk3RdG6KpnNTtKjPjaYrRUQkbSnkREQkbWVyyA2Iu4AUpnNTNJ2bouncFE3npmiRnpuMXZMTEZH0l8kjORERSXNpH3Jm1tHM5pjZXDO7tZDXzcweSbw+3cyOiKPOOJTg3FyYOCfTzWysmR0eR51x2Na5yXdcCzPLNbPOyawvTiU5N2bWzsy+MLOZZvZpsmuMSwn+m9rdzN40s2mJc9MjjjqTzcyeNrMfzGxGEa9H93u4sNuFp8sfoBLwLVAXqAJMAxoXOOZUYARgwNHAhLjrTqFzcwywZ+LrU3RuCj1uJPAO0DnuulPl3AB7ALOAgxKP94277hQ6N7cB9ye+rgGsAKrEXXsSzs1xwBHAjCJej+z3cLqP5I4C5rr7PHffCAwBOhU4phPwvAfjgT3MbP9kFxqDbZ4bdx/r7j8lHo4HSr71d8VWkr83AFcBrwE/JLO4mJXk3FwADHX37wHcPVPOT0nOjQO7mpkBuxBCLie5ZSafu39G+FmLEtnv4XQPuQOAhfkeZyeeK+0x6ai0P3cvwr+0MsE2z42ZHQCcDTyRxLpSQUn+3jQA9jSzT8xsipl1T1p18SrJuekHNAIWA18C17h7XnLKS2mR/R5O91vtWCHPFWwnLckx6ajEP7eZHU8IuTaRVpQ6SnJuHgZucffc8I/yjFGSc1MZOBI4AagGjDOz8e7+ddTFxawk5+Zk4AugPVAP+MDMRrn7qqiLS3GR/R5O95DLBg7M97gW4V9QpT0mHZXo5zazZsBA4BR3X56k2uJWknOTBQxJBNw+wKlmluPuw5JTYmxK+t/Uj+6+BlhjZp8BhwPpHnIlOTc9gPs8LETNNbP5QENgYnJKTFmR/R5O9+nKSUB9M6tjZlWA84HhBY4ZDnRPdPccDfzs7kuSXWgMtnluzOwgYCjQLQP+FZ7fNs+Nu9dx99ruXht4FbgiAwIOSvbf1BvAsWZW2cx2BloCs5NcZxxKcm6+J4xwMbP9gEOBeUmtMjVF9ns4rUdy7p5jZlcC7xE6n55295lmdnni9ScInXGnAnOBtYR/aaW9Ep6bvwB7A48lRiw5ngGbzJbw3GSkkpwbd59tZu8C04E8YKC7F9o6nk5K+PfmbuBZM/uSMEV3i7un/d0JzGww0A7Yx8yygTuAHSH638Pa8URERNJWuk9XiohIBlPIiYhI2lLIiYhI2lLIiYhI2lLIiYhI2lLIiaSoxN0Nvsj3p3Zid/+fzexzM5ttZnckjs3//Fdm9q+46xdJBWl9nZxIBbfO3Zvnf8LMagOj3P10M6sOfGFmbyVe/vX5asDnZva6u49JbskiqUUjOZEKKrFt1hTCHoj5n19H2B8xEzYaFymWQk4kdVXLN1X5esEXzWxvwr23ZhZ4fk+gPvBZcsoUSV2arhRJXVtNVyYca2afE7bMui+xdVS7xPPTCfsh3ufuS5NYq0hKUsiJVDyj3P30op43swbA6MSa3BfJLk4klWi6UiTNJO4Y8XfglrhrEYmbQk4kPT0BHGdmdeIuRCROuguBiIikLY3kREQkbSnkREQkbSnkREQkbSnkREQkbSnkREQkbSnkREQkbSnkREQkbSnkREQkbf0/JQnZKiUecyIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierUS, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUS, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='brute', n_neighbors=1)"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train\n",
    "classifierUN = neighbors.KNeighborsClassifier(n_neighbors=1, weights='uniform', algorithm = 'brute')\n",
    "classifierUN.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8760853474077173\n",
      "0.5717778131234297\n",
      "0.5602807514067484\n",
      "0.5646973252443279\n",
      "0.5602807514067484\n"
     ]
    }
   ],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUN ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#test\n",
    "y_pred = classifierUN.predict(X_test)\n",
    "classifierUN.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Result:\n",
      "================================================\n",
      "Accuracy Score: 99.98%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 99.88%\n",
      "\t\t\tRecall Score: 99.94%\n",
      "\t\t\tF1 score: 99.91%\n",
      "\t\t\tAUC score: 99.96%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[35378     4]\n",
      " [    2  3224]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      1.000     1.000     1.000     35382\n",
      "           1      0.999     0.999     0.999      3226\n",
      "\n",
      "    accuracy                          1.000     38608\n",
      "   macro avg      0.999     1.000     0.999     38608\n",
      "weighted avg      1.000     1.000     1.000     38608\n",
      "\n",
      "Test Result:\n",
      "================================================\n",
      "Accuracy Score: 50.00%\n",
      "_______________________________________________\n",
      "Classification Report:\tPrecision Score: 0.00%\n",
      "\t\t\tRecall Score: 0.00%\n",
      "\t\t\tF1 score: 0.00%\n",
      "\t\t\tAUC score: 50.00%\n",
      "_______________________________________________\n",
      "Confusion Matrix: \n",
      " [[200   0]\n",
      " [200   0]]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.500     1.000     0.667       200\n",
      "           1      0.000     0.000     0.000       200\n",
      "\n",
      "    accuracy                          0.500       400\n",
      "   macro avg      0.250     0.500     0.333       400\n",
      "weighted avg      0.250     0.500     0.333       400\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.500     1.000     0.667       200\n",
      "           1      0.000     0.000     0.000       200\n",
      "\n",
      "    accuracy                          0.500       400\n",
      "   macro avg      0.250     0.500     0.333       400\n",
      "weighted avg      0.250     0.500     0.333       400\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_score(classifierUN, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUN, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ANN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "1207/1207 [==============================] - 4s 2ms/step - loss: 0.3662 - accuracy: 0.9158\n",
      "Epoch 2/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2693 - accuracy: 0.9160\n",
      "Epoch 3/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2590 - accuracy: 0.9164\n",
      "Epoch 4/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2502 - accuracy: 0.9174\n",
      "Epoch 5/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2463 - accuracy: 0.9192\n",
      "Epoch 6/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.2414 - accuracy: 0.9194\n",
      "Epoch 7/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2425 - accuracy: 0.9172\n",
      "Epoch 8/100\n",
      "1207/1207 [==============================] - 2s 1ms/step - loss: 0.2310 - accuracy: 0.9231\n",
      "Epoch 9/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2316 - accuracy: 0.9225\n",
      "Epoch 10/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2301 - accuracy: 0.9212\n",
      "Epoch 11/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2304 - accuracy: 0.9218\n",
      "Epoch 12/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2200 - accuracy: 0.9250\n",
      "Epoch 13/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2199 - accuracy: 0.9250\n",
      "Epoch 14/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2186 - accuracy: 0.9246\n",
      "Epoch 15/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2094 - accuracy: 0.9285\n",
      "Epoch 16/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2111 - accuracy: 0.9280\n",
      "Epoch 17/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2197 - accuracy: 0.9235\n",
      "Epoch 18/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2058 - accuracy: 0.9291\n",
      "Epoch 19/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2130 - accuracy: 0.9267\n",
      "Epoch 20/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2099 - accuracy: 0.9282\n",
      "Epoch 21/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2051 - accuracy: 0.9305\n",
      "Epoch 22/100\n",
      "1207/1207 [==============================] - 3s 3ms/step - loss: 0.2112 - accuracy: 0.9271\n",
      "Epoch 23/100\n",
      "1207/1207 [==============================] - 3s 3ms/step - loss: 0.2118 - accuracy: 0.9282\n",
      "Epoch 24/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2094 - accuracy: 0.9277\n",
      "Epoch 25/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2042 - accuracy: 0.9295\n",
      "Epoch 26/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2074 - accuracy: 0.9277\n",
      "Epoch 27/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2086 - accuracy: 0.9291\n",
      "Epoch 28/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2044 - accuracy: 0.9313\n",
      "Epoch 29/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2112 - accuracy: 0.9292: 0s - los\n",
      "Epoch 30/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2055 - accuracy: 0.9296\n",
      "Epoch 31/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2047 - accuracy: 0.9297\n",
      "Epoch 32/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2060 - accuracy: 0.9289\n",
      "Epoch 33/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2081 - accuracy: 0.9281\n",
      "Epoch 34/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2001 - accuracy: 0.9320\n",
      "Epoch 35/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2080 - accuracy: 0.9282\n",
      "Epoch 36/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2090 - accuracy: 0.9282\n",
      "Epoch 37/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2073 - accuracy: 0.9292\n",
      "Epoch 38/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1989 - accuracy: 0.9327\n",
      "Epoch 39/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2014 - accuracy: 0.9309\n",
      "Epoch 40/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2081 - accuracy: 0.9306\n",
      "Epoch 41/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2083 - accuracy: 0.9287\n",
      "Epoch 42/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.2007 - accuracy: 0.9327\n",
      "Epoch 43/100\n",
      "1207/1207 [==============================] - 3s 3ms/step - loss: 0.2069 - accuracy: 0.9285\n",
      "Epoch 44/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.2044 - accuracy: 0.9297\n",
      "Epoch 45/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2060 - accuracy: 0.9298\n",
      "Epoch 46/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2010 - accuracy: 0.9306\n",
      "Epoch 47/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2008 - accuracy: 0.9314\n",
      "Epoch 48/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1995 - accuracy: 0.9313\n",
      "Epoch 49/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2035 - accuracy: 0.9306\n",
      "Epoch 50/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2007 - accuracy: 0.9303\n",
      "Epoch 51/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.2052 - accuracy: 0.9288\n",
      "Epoch 52/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2035 - accuracy: 0.9298\n",
      "Epoch 53/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2048 - accuracy: 0.9300\n",
      "Epoch 54/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1985 - accuracy: 0.9310\n",
      "Epoch 55/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2004 - accuracy: 0.9320\n",
      "Epoch 56/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1992 - accuracy: 0.9325\n",
      "Epoch 57/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.2015 - accuracy: 0.9308\n",
      "Epoch 58/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1996 - accuracy: 0.9330\n",
      "Epoch 59/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1896 - accuracy: 0.9347\n",
      "Epoch 60/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1906 - accuracy: 0.9361\n",
      "Epoch 61/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1905 - accuracy: 0.9353\n",
      "Epoch 62/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1881 - accuracy: 0.9382\n",
      "Epoch 63/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1799 - accuracy: 0.9402\n",
      "Epoch 64/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1818 - accuracy: 0.9400\n",
      "Epoch 65/100\n",
      "1207/1207 [==============================] - 2s 1ms/step - loss: 0.1813 - accuracy: 0.9391\n",
      "Epoch 66/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1809 - accuracy: 0.9399\n",
      "Epoch 67/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1788 - accuracy: 0.9402\n",
      "Epoch 68/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1761 - accuracy: 0.9426\n",
      "Epoch 69/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1759 - accuracy: 0.9411\n",
      "Epoch 70/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1767 - accuracy: 0.9409\n",
      "Epoch 71/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1751 - accuracy: 0.9420\n",
      "Epoch 72/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1706 - accuracy: 0.9431\n",
      "Epoch 73/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1698 - accuracy: 0.9435\n",
      "Epoch 74/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1702 - accuracy: 0.9430\n",
      "Epoch 75/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1726 - accuracy: 0.9415\n",
      "Epoch 76/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1759 - accuracy: 0.9398\n",
      "Epoch 77/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.1756 - accuracy: 0.9403\n",
      "Epoch 78/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1714 - accuracy: 0.9417\n",
      "Epoch 79/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1733 - accuracy: 0.9413\n",
      "Epoch 80/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1691 - accuracy: 0.9433\n",
      "Epoch 81/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1743 - accuracy: 0.9405\n",
      "Epoch 82/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1746 - accuracy: 0.9407\n",
      "Epoch 83/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.1723 - accuracy: 0.9416\n",
      "Epoch 84/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.1711 - accuracy: 0.9426\n",
      "Epoch 85/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.1718 - accuracy: 0.9431\n",
      "Epoch 86/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1705 - accuracy: 0.9420\n",
      "Epoch 87/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1767 - accuracy: 0.9397\n",
      "Epoch 88/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.1695 - accuracy: 0.9422\n",
      "Epoch 89/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1701 - accuracy: 0.9419\n",
      "Epoch 90/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.1739 - accuracy: 0.9409\n",
      "Epoch 91/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1681 - accuracy: 0.9438\n",
      "Epoch 92/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1696 - accuracy: 0.9423\n",
      "Epoch 93/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1707 - accuracy: 0.9421\n",
      "Epoch 94/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1703 - accuracy: 0.9419\n",
      "Epoch 95/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1743 - accuracy: 0.9403\n",
      "Epoch 96/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1635 - accuracy: 0.9450\n",
      "Epoch 97/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1714 - accuracy: 0.9419\n",
      "Epoch 98/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1734 - accuracy: 0.9409\n",
      "Epoch 99/100\n",
      "1207/1207 [==============================] - 3s 2ms/step - loss: 0.1761 - accuracy: 0.9399\n",
      "Epoch 100/100\n",
      "1207/1207 [==============================] - 2s 2ms/step - loss: 0.1699 - accuracy: 0.9416\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x21c1e414c10>"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train\n",
    "annU = tf.keras.models.Sequential()\n",
    "annU.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "annU.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "annU.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "annU.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "annU.fit(Xc_train,yc_train , batch_size = 32, epochs = 30 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "#CrossV\n",
    "from keras.wrappers.scikit_learn import KerasClassifier\n",
    "def create_network():\n",
    "    model = tf.keras.models.Sequential()\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "    model.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "    model.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "    return model\n",
    "\n",
    "\n",
    "\n",
    "network = KerasClassifier(build_fn=create_network, \n",
    "                                 epochs=30, \n",
    "                                 batch_size=32\n",
    "                                 )\n",
    "    \n",
    "    \n",
    "    \n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(network ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "Cannot clone object '<tensorflow.python.keras.engine.sequential.Sequential object at 0x0000021C1F3DA880>' (type <class 'tensorflow.python.keras.engine.sequential.Sequential'>): it does not seem to be a scikit-learn estimator as it does not implement a 'get_params' method.",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mEmpty\u001b[0m                                     Traceback (most recent call last)",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[1;34m(self, iterator)\u001b[0m\n\u001b[0;32m    807\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 808\u001b[1;33m                 \u001b[0mtasks\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_ready_batches\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mblock\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    809\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mqueue\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mEmpty\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\queue.py\u001b[0m in \u001b[0;36mget\u001b[1;34m(self, block, timeout)\u001b[0m\n\u001b[0;32m    166\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_qsize\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 167\u001b[1;33m                     \u001b[1;32mraise\u001b[0m \u001b[0mEmpty\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    168\u001b[0m             \u001b[1;32melif\u001b[0m \u001b[0mtimeout\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mEmpty\u001b[0m: ",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-138-3ae2e16bb932>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      4\u001b[0m            \u001b[1;34m'f1_macro'\u001b[0m\u001b[1;33m:\u001b[0m \u001b[1;34m'f1_macro'\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m           ,'roc_auc' : 'roc_auc'}\n\u001b[1;32m----> 6\u001b[1;33m scores = cross_validate(annU ,X_cv, y_cv ,scoring=scoring,\n\u001b[0m\u001b[0;32m      7\u001b[0m                          cv=10, return_train_score=False)\n\u001b[0;32m      8\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py\u001b[0m in \u001b[0;36minner_f\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     71\u001b[0m                           FutureWarning)\n\u001b[0;32m     72\u001b[0m         \u001b[0mkwargs\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m{\u001b[0m\u001b[0mk\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mk\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msig\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mparameters\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 73\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     74\u001b[0m     \u001b[1;32mreturn\u001b[0m \u001b[0minner_f\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\u001b[0m in \u001b[0;36mcross_validate\u001b[1;34m(estimator, X, y, groups, scoring, cv, n_jobs, verbose, fit_params, pre_dispatch, return_train_score, return_estimator, error_score)\u001b[0m\n\u001b[0;32m    240\u001b[0m     parallel = Parallel(n_jobs=n_jobs, verbose=verbose,\n\u001b[0;32m    241\u001b[0m                         pre_dispatch=pre_dispatch)\n\u001b[1;32m--> 242\u001b[1;33m     scores = parallel(\n\u001b[0m\u001b[0;32m    243\u001b[0m         delayed(_fit_and_score)(\n\u001b[0;32m    244\u001b[0m             \u001b[0mclone\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscorers\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtrain\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, iterable)\u001b[0m\n\u001b[0;32m   1027\u001b[0m             \u001b[1;31m# remaining jobs.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1028\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1029\u001b[1;33m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdispatch_one_batch\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1030\u001b[0m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_original_iterator\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1031\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[1;34m(self, iterator)\u001b[0m\n\u001b[0;32m    817\u001b[0m                 \u001b[0mbig_batch_size\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    818\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 819\u001b[1;33m                 \u001b[0mislice\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mitertools\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mislice\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mbig_batch_size\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    820\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mislice\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m==\u001b[0m \u001b[1;36m0\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    821\u001b[0m                     \u001b[1;32mreturn\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_validation.py\u001b[0m in \u001b[0;36m<genexpr>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m    242\u001b[0m     scores = parallel(\n\u001b[0;32m    243\u001b[0m         delayed(_fit_and_score)(\n\u001b[1;32m--> 244\u001b[1;33m             \u001b[0mclone\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscorers\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtrain\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    245\u001b[0m             \u001b[0mfit_params\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreturn_train_score\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mreturn_train_score\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    246\u001b[0m             \u001b[0mreturn_times\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreturn_estimator\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mreturn_estimator\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py\u001b[0m in \u001b[0;36minner_f\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     71\u001b[0m                           FutureWarning)\n\u001b[0;32m     72\u001b[0m         \u001b[0mkwargs\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m{\u001b[0m\u001b[0mk\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mk\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0marg\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msig\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mparameters\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 73\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     74\u001b[0m     \u001b[1;32mreturn\u001b[0m \u001b[0minner_f\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\sklearn\\base.py\u001b[0m in \u001b[0;36mclone\u001b[1;34m(estimator, safe)\u001b[0m\n\u001b[0;32m     76\u001b[0m                                 \"scikit-learn estimator instead of a class.\")\n\u001b[0;32m     77\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 78\u001b[1;33m                 raise TypeError(\"Cannot clone object '%s' (type %s): \"\n\u001b[0m\u001b[0;32m     79\u001b[0m                                 \u001b[1;34m\"it does not seem to be a scikit-learn \"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     80\u001b[0m                                 \u001b[1;34m\"estimator as it does not implement a \"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mTypeError\u001b[0m: Cannot clone object '<tensorflow.python.keras.engine.sequential.Sequential object at 0x0000021C1F3DA880>' (type <class 'tensorflow.python.keras.engine.sequential.Sequential'>): it does not seem to be a scikit-learn estimator as it does not implement a 'get_params' method."
     ]
    }
   ],
   "source": [
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "print_scoreN(annU, Xc_train,yc_train, X_test,Y_test, train=True)\n",
    "print_scoreN(annU, Xc_train,yc_train, X_test,Y_test, train=False)\n",
    "\n",
    "y_pred = annU.predict(X_test)\n",
    "y_pred = (y_pred > 0.5)\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# XGboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train \n",
    "classifierUXG = XGBClassifier()\n",
    "classifierUXG.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUXG ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierUXG.predict(X_test)\n",
    "classifierUXG.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierUXG, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUXG, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split for under-sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Xc_train, X_cv, yc_train, y_cv = train_test_split(X_tl, y_tl, test_size=0.2 , random_state=3 , stratify = y_tl)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifierUNRF = RandomForestClassifier(n_estimators = 80 ,criterion = 'entropy' , random_state = 0)\n",
    "classifierUNRF.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "\n",
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUNRF ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = classifierUNRF.predict(X_test)\n",
    "classifierUNRF.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierUNRF, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUNRF, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "#AUC\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "classifierUND = DecisionTreeClassifier(criterion = 'entropy' , random_state = 0)\n",
    "classifierUND.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUND ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierUND.predict(X_test)\n",
    "classifierUND.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierUND, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUND, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "#AUC\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "classifierUNS = SVC(kernel='rbf', random_state = 0)\n",
    "classifierUNS.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUNS ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierUNS.predict(X_test)\n",
    "classifierUNS.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierUNS, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUNS, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "classifierUNN = neighbors.KNeighborsClassifier(n_neighbors=1, weights='uniform', algorithm = 'brute')\n",
    "classifierUNN.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUNN ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierUNN.predict(X_test)\n",
    "classifierUNN.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierUNN, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUNN, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ANN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "annUN = tf.keras.models.Sequential()\n",
    "annUN.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "annUN.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "annUN.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "annUN.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "annUN.fit(Xc_train,yc_train , batch_size = 32, epochs = 30 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#CrossV\n",
    "from keras.wrappers.scikit_learn import KerasClassifier\n",
    "def create_network():\n",
    "    model = tf.keras.models.Sequential()\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "    model.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "    model.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "    return model\n",
    "\n",
    "\n",
    "\n",
    "network = KerasClassifier(build_fn=create_network, \n",
    "                                 epochs=30, \n",
    "                                 batch_size=32\n",
    "                                 )\n",
    "    \n",
    "    \n",
    "    \n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(network ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "print_scoreN(annUN, Xc_train,yc_train, X_test,Y_test, train=True)\n",
    "print_scoreN(annUN, Xc_train,yc_train, X_test,Y_test, train=False)\n",
    "\n",
    "y_pred = annUN.predict(X_test)\n",
    "y_pred = (y_pred > 0.5)\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# XGboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train \n",
    "classifierUNXG = XGBClassifier()\n",
    "classifierUNXG.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUNXG ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierUNXG.predict(X_test)\n",
    "classifierUNXG.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierUNXG, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUNXG, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### split for under and over sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Xc_train, X_cv, yc_train, y_cv = train_test_split(x_SMtl, y_SMtl, test_size=0.2 , random_state=3 , stratify = y_SMtl)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "classifierOURF = RandomForestClassifier(n_estimators = 90 ,criterion = 'entropy' , random_state = 0)\n",
    "classifierOURF .fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOURF ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierOURF .predict(X_test)\n",
    "classifierOURF .score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierOURF, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOURF, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "classifierOUD = DecisionTreeClassifier(criterion = 'entropy' , random_state = 0)\n",
    "classifierOUD.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOUD ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierOUD.predict(X_test)\n",
    "classifierOUD.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierOUD, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOUD, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "classifierOUS = SVC(kernel='rbf', random_state = 0)\n",
    "classifierOUS.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOUS ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierOUS.predict(X_test)\n",
    "classifierOUS.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierOUS, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOUS, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "classifierOUN = neighbors.KNeighborsClassifier(n_neighbors=1, weights='uniform', algorithm = 'brute')\n",
    "classifierOUN.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierOUN ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierOUN.predict(X_test)\n",
    "classifierOUN.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierOUN, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierOUN, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ANN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train\n",
    "annO = tf.keras.models.Sequential()\n",
    "annO.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "annO.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "annO.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "annO.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "annO.fit(Xc_train,yc_train , batch_size = 32, epochs = 30 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#CrossV\n",
    "from keras.wrappers.scikit_learn import KerasClassifier\n",
    "def create_network():\n",
    "    model = tf.keras.models.Sequential()\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu' , input_dim = 24))\n",
    "    model.add(tf.keras.layers.Dense(units = 16 , activation = 'relu'))\n",
    "    model.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))\n",
    "    model.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )\n",
    "    return model\n",
    "\n",
    "\n",
    "\n",
    "network = KerasClassifier(build_fn=create_network, \n",
    "                                 epochs=30, \n",
    "                                 batch_size=32\n",
    "                                 )\n",
    "    \n",
    "    \n",
    "    \n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(network ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "print_scoreN(annO, Xc_train,yc_train, X_test,Y_test, train=True)\n",
    "print_scoreN(annO, Xc_train,yc_train, X_test,Y_test, train=False)\n",
    "\n",
    "y_pred = annO.predict(X_test)\n",
    "y_pred = (y_pred > 0.5)\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# XGboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#train \n",
    "classifierUXG = XGBClassifier()\n",
    "classifierUXG.fit(Xc_train,yc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#crossV\n",
    "scoring = {'acc': 'accuracy',\n",
    "           'prec_macro': 'precision_macro',\n",
    "           'rec_micro': 'recall_macro',\n",
    "           'f1_macro': 'f1_macro' \n",
    "          ,'roc_auc' : 'roc_auc'}\n",
    "scores = cross_validate(classifierUXG ,X_cv, y_cv ,scoring=scoring,\n",
    "                         cv=10, return_train_score=False)\n",
    "\n",
    "\n",
    "print(scores['test_acc'].mean())\n",
    "print(scores['test_prec_macro'].mean())\n",
    "print(scores['test_rec_micro'].mean())\n",
    "print(scores['test_f1_macro'].mean())\n",
    "print(scores['test_roc_auc'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#test\n",
    "y_pred = classifierUXG.predict(X_test)\n",
    "classifierUXG.score(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierUXG, Xc_train, yc_train, X_test, Y_test, train=True)\n",
    "print_score(classifierUXG, Xc_train, yc_train, X_test, Y_test, train=False)\n",
    "print(classification_report(Y_test,y_pred, digits = 3))\n",
    "\n",
    "fpr, tpr, _ = metrics.roc_curve(Y_test, y_pred)\n",
    "auc_score = metrics.auc(fpr, tpr)\n",
    "plt.figure(figsize=(7, 6))\n",
    "plt.plot(fpr, tpr, color='blue',\n",
    "label='ROC (AUC = %0.4f)' % auc_score)\n",
    "plt.legend(loc='lower right')\n",
    "plt.title(\"ROC Curve\")\n",
    "plt.xlabel(\"FPR\")\n",
    "plt.ylabel(\"TPR\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The END EXP1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# solve it to visulation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib.colors import ListedColormap\n",
    "X_set, y_set = sca.inverse_transform(X_train), y_train\n",
    "X1, X2 = np.meshgrid(np.arange(start = X_set[:, 0].min() - 10, stop = X_set[:, 0].max() + 10, step = 0.25),\n",
    "                     np.arange(start = X_set[:, 1].min() - 1000, stop = X_set[:, 1].max() + 1000, step = 0.25))\n",
    "plt.contourf(X1, X2, classifier.predict(sca.transform(np.array([X1.ravel(), X2.ravel()]).T)).reshape(X1.shape),\n",
    "             alpha = 0.75, cmap = ListedColormap(('red', 'green')))\n",
    "plt.xlim(X1.min(), X1.max())\n",
    "plt.ylim(X2.min(), X2.max())\n",
    "for i, j in enumerate(np.unique(y_set)):\n",
    "    plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1], c = ListedColormap(('red', 'green'))(i), label = j)\n",
    "plt.title('Random Forest Classification (Training set)')\n",
    "plt.xlabel('Age')\n",
    "plt.ylabel('Estimated Salary')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#showing the RF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import tree\n",
    "fig, axes = plt.subplots(nrows = 1,ncols = 1,figsize = (10,6), dpi=800)\n",
    "tree.plot_tree(classifier.estimators_[0],\n",
    "               max_depth = 3 ,\n",
    "               feature_names = X.columns.values, \n",
    "               class_names= ['No', 'Yes'],\n",
    "               filled = True);\n",
    "fig.savefig('rf_individualtree.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(t_SM, u_SM, test_size=0.3 , random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "classifier = RandomForestClassifier(n_estimators = 30 , criterion = 'entropy' , random_state = 0)\n",
    "classifier.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = classifier.predict(X_test)\n",
    "classifier.score(X_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm = confusion_matrix(y_test,y_pred)\n",
    "print(cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "8427/13349"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "11095 /13308"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC\n",
    "classifier = SVC(kernel='rbf', random_state = 0)\n",
    "classifier.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_pred = classifier.predict(X_test)\n",
    "from sklearn.metrics import confusion_matrix\n",
    "cm = confusion_matrix(y_test,Y_pred)\n",
    "accuracy = float(cm.diagonal().sum())/len(y_test)\n",
    "print(\"\\nAccuracy Of SVM For The Given Dataset : \", accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "print(confusion_matrix(y_test,Y_pred))\n",
    "print(classification_report(y_test,Y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifier, X_train, y_train, X_test, y_test, train=True)\n",
    "print_score(classifier, X_train, y_train, X_test, y_test, train=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(x_SM, y_SM, test_size=0.2 , random_state=1)\n",
    "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.25 , random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifierD = DecisionTreeClassifier(criterion = 'entropy' , random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifierD.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = classifierD.predict(X_val)\n",
    "classifierD.score(X_val,y_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierD, X_train, y_train, X_val, y_val, train=True)\n",
    "print_score(classifierD, X_train, y_train, X_val, y_val, train=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = classifierD.predict(X_test)\n",
    "classifierD.score(X_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_score(classifierD, X_train, y_train, X_test, y_test, train=True)\n",
    "print_score(classifierD, X_train, y_train, X_test, y_test, train=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# over and under"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# artifichal neural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tf.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ann = tf.keras.models.Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ann.add(tf.keras.layers.Dense(units = 8 , activation = 'relu' , input_dim = 24))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ann.add(tf.keras.layers.Dense(units = 8 , activation = 'relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ann.add(tf.keras.layers.Dense(units = 1 , activation = 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ann.compile(optimizer ='adam' , loss = 'binary_crossentropy', metrics = ['accuracy'] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ann.fit(X_train, y_train , batch_size = 32 , epochs = 100  )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pre = "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print_scoreN(ann, X_train, y_train, X_test, y_test, train=True)\n",
    "print_scoreN(ann, X_train, y_train, X_test, y_test, train=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score, confusion_matrix, precision_score, recall_score, f1_score\n",
    "\n",
    "def print_score(clf, X_train, y_train, X_test, y_test, train=True):\n",
    "    if train:\n",
    "        pred = clf.predict(X_train)\n",
    "        print(\"Train Result:\\n================================================\")\n",
    "        print(f\"Accuracy Score: {accuracy_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_train, pred)}\\n\")\n",
    "        \n",
    "    elif train==False:\n",
    "        pred = clf.predict(X_test)\n",
    "        print(\"Test Result:\\n================================================\")        \n",
    "        print(f\"Accuracy Score: {accuracy_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_test, pred)}\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score, confusion_matrix, precision_score, recall_score, f1_score\n",
    "\n",
    "def print_scoreN(clf, X_train, y_train, X_test, y_test, train=True):\n",
    "    if train:\n",
    "        pred = clf.predict(X_train)\n",
    "        pred = (pred > 0.5)\n",
    "        print(\"Train Result:\\n================================================\")\n",
    "        print(f\"Accuracy Score: {accuracy_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_train, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_train, pred)}\\n\")\n",
    "        \n",
    "    elif train==False:\n",
    "        pred = clf.predict(X_test)\n",
    "        pred = (pred > 0.5)\n",
    "        print(\"Test Result:\\n================================================\")        \n",
    "        print(f\"Accuracy Score: {accuracy_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(\"Classification Report:\", end='')\n",
    "        print(f\"\\tPrecision Score: {precision_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tRecall Score: {recall_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(f\"\\t\\t\\tF1 score: {f1_score(y_test, pred) * 100:.2f}%\")\n",
    "        print(\"_______________________________________________\")\n",
    "        print(f\"Confusion Matrix: \\n {confusion_matrix(y_test, pred)}\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exp 2 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    " = df.sample(1000).index"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}