{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>domisili</th>\n",
       "      <th>jenis_kelamin</th>\n",
       "      <th>date_of_birth</th>\n",
       "      <th>join_date</th>\n",
       "      <th>resign_date</th>\n",
       "      <th>marriage_stat</th>\n",
       "      <th>dependant</th>\n",
       "      <th>education</th>\n",
       "      <th>absent_90D</th>\n",
       "      <th>...</th>\n",
       "      <th>active_work_category</th>\n",
       "      <th>work_stability_score</th>\n",
       "      <th>married_dependent_ratio</th>\n",
       "      <th>position_score</th>\n",
       "      <th>job_income_position_score</th>\n",
       "      <th>education_score</th>\n",
       "      <th>education_income_ratio</th>\n",
       "      <th>weighted_satisfaction_performance</th>\n",
       "      <th>resign_risk_indicator</th>\n",
       "      <th>adjusted_work_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>EM10510</td>\n",
       "      <td>Kota Jakarta Utara</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1983-09-11</td>\n",
       "      <td>2021-02-09</td>\n",
       "      <td>2023-06-22</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>9.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1418218.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.418218e+06</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.246870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>EM4322</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1987-03-22</td>\n",
       "      <td>2022-02-28</td>\n",
       "      <td>2023-04-04</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1060575.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.060575e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.650000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>EM1637</td>\n",
       "      <td>Kota Jakarta Barat</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1970-04-27</td>\n",
       "      <td>2020-12-23</td>\n",
       "      <td>2023-03-25</td>\n",
       "      <td>Married</td>\n",
       "      <td>4</td>\n",
       "      <td>D2</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.400000</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>4885136.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.628379e+06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.813826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>EM14613</td>\n",
       "      <td>Kota Jakarta Pusat</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1988-06-10</td>\n",
       "      <td>2022-11-21</td>\n",
       "      <td>2024-03-23</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>D3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.333333</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4602479.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.150620e+06</td>\n",
       "      <td>2.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.756440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>EM1084</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1977-05-25</td>\n",
       "      <td>2021-06-07</td>\n",
       "      <td>2023-07-21</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1066966.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.066966e+06</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.080000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  employee_id            domisili jenis_kelamin date_of_birth   join_date  \\\n",
       "0     EM10510  Kota Jakarta Utara     Laki-laki    1983-09-11  2021-02-09   \n",
       "1      EM4322    Kabupaten Bekasi     Perempuan    1987-03-22  2022-02-28   \n",
       "2      EM1637  Kota Jakarta Barat     Laki-laki    1970-04-27  2020-12-23   \n",
       "3     EM14613  Kota Jakarta Pusat     Laki-laki    1988-06-10  2022-11-21   \n",
       "4      EM1084     Kabupaten Bogor     Perempuan    1977-05-25  2021-06-07   \n",
       "\n",
       "  resign_date marriage_stat  dependant education  absent_90D  ...  \\\n",
       "0  2023-06-22       Married          1      SLTA         9.0  ...   \n",
       "1  2023-04-04       Married          1      SLTA         0.0  ...   \n",
       "2  2023-03-25       Married          4        D2         4.0  ...   \n",
       "3  2024-03-23       Married          1        D3         2.0  ...   \n",
       "4  2023-07-21       Married          3      SLTA         0.0  ...   \n",
       "\n",
       "   active_work_category work_stability_score married_dependent_ratio  \\\n",
       "0              Mid-term             2.800000                       2   \n",
       "1              Mid-term            13.000000                       2   \n",
       "2              Mid-term             5.400000                       5   \n",
       "3              Mid-term             5.333333                       2   \n",
       "4              Mid-term            25.000000                       4   \n",
       "\n",
       "   position_score  job_income_position_score education_score  \\\n",
       "0               1                  1418218.0               1   \n",
       "1               1                  1060575.0               1   \n",
       "2               1                  4885136.0               3   \n",
       "3               1                  4602479.0               4   \n",
       "4               1                  1066966.0               1   \n",
       "\n",
       "  education_income_ratio  weighted_satisfaction_performance  \\\n",
       "0           1.418218e+06                                2.2   \n",
       "1           1.060575e+06                                1.6   \n",
       "2           1.628379e+06                                1.0   \n",
       "3           1.150620e+06                                2.4   \n",
       "4           1.066966e+06                                2.6   \n",
       "\n",
       "   resign_risk_indicator  adjusted_work_time  \n",
       "0                 Medium            9.246870  \n",
       "1                 Medium            9.650000  \n",
       "2                 Medium            9.813826  \n",
       "3                 Medium            9.756440  \n",
       "4                 Medium            9.080000  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.read_csv(\"D:/Tugas Akhir/Codingan/Notebook - Playground/preprocessed_data_train_3.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13770"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>churn_status</th>\n",
       "      <th>Count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>10696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3074</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   churn_status  Count\n",
       "0             0  10696\n",
       "1             1   3074"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "churn = df['churn_status']\n",
    "exploded_churn = churn.explode()\n",
    "\n",
    "churn_count = exploded_churn.value_counts().reset_index()\n",
    "churn_count.columns = ['churn_status', 'Count']\n",
    "churn_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>domisili</th>\n",
       "      <th>jenis_kelamin</th>\n",
       "      <th>date_of_birth</th>\n",
       "      <th>join_date</th>\n",
       "      <th>resign_date</th>\n",
       "      <th>marriage_stat</th>\n",
       "      <th>dependant</th>\n",
       "      <th>education</th>\n",
       "      <th>absent_90D</th>\n",
       "      <th>...</th>\n",
       "      <th>active_work_category</th>\n",
       "      <th>work_stability_score</th>\n",
       "      <th>married_dependent_ratio</th>\n",
       "      <th>position_score</th>\n",
       "      <th>job_income_position_score</th>\n",
       "      <th>education_score</th>\n",
       "      <th>education_income_ratio</th>\n",
       "      <th>weighted_satisfaction_performance</th>\n",
       "      <th>resign_risk_indicator</th>\n",
       "      <th>adjusted_work_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12683</th>\n",
       "      <td>EM12967</td>\n",
       "      <td>Kota Jakarta Utara</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1998-03-13</td>\n",
       "      <td>2020-05-03</td>\n",
       "      <td>2024-07-04</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Long-term</td>\n",
       "      <td>6.250000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1776619.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1776619.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Low</td>\n",
       "      <td>9.884603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11885</th>\n",
       "      <td>EM8515</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>2000-03-31</td>\n",
       "      <td>2022-08-23</td>\n",
       "      <td>2024-10-31</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1104966.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1104966.0</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.260000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13621</th>\n",
       "      <td>EM9679</td>\n",
       "      <td>Kota Bekasi</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1975-07-01</td>\n",
       "      <td>2021-12-06</td>\n",
       "      <td>2024-10-31</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>S1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.833333</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2519458.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1007783.2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.335163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8399</th>\n",
       "      <td>EM15269</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1988-01-19</td>\n",
       "      <td>2022-11-01</td>\n",
       "      <td>2024-10-31</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>S1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3015150.5</td>\n",
       "      <td>5</td>\n",
       "      <td>1206060.2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.740976</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5525</th>\n",
       "      <td>EM2598</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1996-03-11</td>\n",
       "      <td>2021-06-08</td>\n",
       "      <td>2024-10-31</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Long-term</td>\n",
       "      <td>8.200000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4488998.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1122249.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Low</td>\n",
       "      <td>9.120106</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      employee_id            domisili jenis_kelamin date_of_birth   join_date  \\\n",
       "12683     EM12967  Kota Jakarta Utara     Laki-laki    1998-03-13  2020-05-03   \n",
       "11885      EM8515           Tangerang     Laki-laki    2000-03-31  2022-08-23   \n",
       "13621      EM9679         Kota Bekasi     Perempuan    1975-07-01  2021-12-06   \n",
       "8399      EM15269    Kabupaten Bekasi     Laki-laki    1988-01-19  2022-11-01   \n",
       "5525       EM2598           Tangerang     Laki-laki    1996-03-11  2021-06-08   \n",
       "\n",
       "      resign_date marriage_stat  dependant education  absent_90D  ...  \\\n",
       "12683  2024-07-04        Single          0      SLTA         7.0  ...   \n",
       "11885  2024-10-31        Single          0      SLTA         0.0  ...   \n",
       "13621  2024-10-31       Married          2        S1         5.0  ...   \n",
       "8399   2024-10-31       Married          2        S1         2.0  ...   \n",
       "5525   2024-10-31        Single          0        D3         4.0  ...   \n",
       "\n",
       "       active_work_category work_stability_score married_dependent_ratio  \\\n",
       "12683             Long-term             6.250000                       1   \n",
       "11885              Mid-term            26.000000                       1   \n",
       "13621              Mid-term             5.833333                       3   \n",
       "8399               Mid-term             8.000000                       3   \n",
       "5525              Long-term             8.200000                       1   \n",
       "\n",
       "       position_score  job_income_position_score education_score  \\\n",
       "12683               1                  1776619.0               1   \n",
       "11885               1                  1104966.0               1   \n",
       "13621               2                  2519458.0               5   \n",
       "8399                2                  3015150.5               5   \n",
       "5525                1                  4488998.0               4   \n",
       "\n",
       "      education_income_ratio  weighted_satisfaction_performance  \\\n",
       "12683              1776619.0                                1.6   \n",
       "11885              1104966.0                                2.6   \n",
       "13621              1007783.2                                2.0   \n",
       "8399               1206060.2                                1.0   \n",
       "5525               1122249.5                                1.0   \n",
       "\n",
       "       resign_risk_indicator  adjusted_work_time  \n",
       "12683                    Low            9.884603  \n",
       "11885                 Medium            9.260000  \n",
       "13621                 Medium            9.335163  \n",
       "8399                  Medium            9.740976  \n",
       "5525                     Low            9.120106  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.DataFrame()\n",
    "for index, row in churn_count.iterrows():\n",
    "    churn = row['churn_status']  \n",
    "    count = row['Count']\n",
    "    if count > 3074:\n",
    "        filtered_data = df[df['churn_status'] == churn].sample(3074)\n",
    "        data = pd.concat([data, filtered_data])\n",
    "\n",
    "for index, row in churn_count.iterrows():\n",
    "    churn = row['churn_status']  \n",
    "    count = row['Count']\n",
    "    if count <= 3074:\n",
    "        filtered_data = df[df['churn_status'] == churn]\n",
    "        data = pd.concat([data, filtered_data])\n",
    "\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "cat_feature = ['departemen', 'position', 'domisili', 'marriage_stat', 'job_satisfaction', 'performance_rating',\n",
    "               'education', 'active_work_category', 'resign_risk_indicator', 'jenis_kelamin']\n",
    "\n",
    "X = data.drop(columns=['churn_status', 'employee_id', 'date_of_birth', 'join_date', 'resign_date', 'active_work_months'])\n",
    "y = data['churn_status']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>domisili</th>\n",
       "      <th>jenis_kelamin</th>\n",
       "      <th>marriage_stat</th>\n",
       "      <th>dependant</th>\n",
       "      <th>education</th>\n",
       "      <th>absent_90D</th>\n",
       "      <th>avg_time_work</th>\n",
       "      <th>departemen</th>\n",
       "      <th>position</th>\n",
       "      <th>income</th>\n",
       "      <th>...</th>\n",
       "      <th>active_work_category</th>\n",
       "      <th>work_stability_score</th>\n",
       "      <th>married_dependent_ratio</th>\n",
       "      <th>position_score</th>\n",
       "      <th>job_income_position_score</th>\n",
       "      <th>education_score</th>\n",
       "      <th>education_income_ratio</th>\n",
       "      <th>weighted_satisfaction_performance</th>\n",
       "      <th>resign_risk_indicator</th>\n",
       "      <th>adjusted_work_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12683</th>\n",
       "      <td>Kota Jakarta Utara</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>9.90</td>\n",
       "      <td>Creative &amp; Design</td>\n",
       "      <td>Staff</td>\n",
       "      <td>1776619</td>\n",
       "      <td>...</td>\n",
       "      <td>Long-term</td>\n",
       "      <td>6.250000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1776619.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1776619.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Low</td>\n",
       "      <td>9.884603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11885</th>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.26</td>\n",
       "      <td>Operations</td>\n",
       "      <td>Staff</td>\n",
       "      <td>1104966</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1104966.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1104966.0</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.260000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13621</th>\n",
       "      <td>Kota Bekasi</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>S1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>9.35</td>\n",
       "      <td>Engineering &amp; IT</td>\n",
       "      <td>Junior</td>\n",
       "      <td>5038916</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.833333</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2519458.0</td>\n",
       "      <td>5</td>\n",
       "      <td>1007783.2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.335163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8399</th>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>S1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>9.75</td>\n",
       "      <td>HR</td>\n",
       "      <td>Junior</td>\n",
       "      <td>6030301</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3015150.5</td>\n",
       "      <td>5</td>\n",
       "      <td>1206060.2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.740976</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5525</th>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>9.13</td>\n",
       "      <td>Service &amp; Support</td>\n",
       "      <td>Staff</td>\n",
       "      <td>4488998</td>\n",
       "      <td>...</td>\n",
       "      <td>Long-term</td>\n",
       "      <td>8.200000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4488998.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1122249.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Low</td>\n",
       "      <td>9.120106</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 domisili jenis_kelamin marriage_stat  dependant education  \\\n",
       "12683  Kota Jakarta Utara     Laki-laki        Single          0      SLTA   \n",
       "11885           Tangerang     Laki-laki        Single          0      SLTA   \n",
       "13621         Kota Bekasi     Perempuan       Married          2        S1   \n",
       "8399     Kabupaten Bekasi     Laki-laki       Married          2        S1   \n",
       "5525            Tangerang     Laki-laki        Single          0        D3   \n",
       "\n",
       "       absent_90D  avg_time_work         departemen position   income  ...  \\\n",
       "12683         7.0           9.90  Creative & Design    Staff  1776619  ...   \n",
       "11885         0.0           9.26         Operations    Staff  1104966  ...   \n",
       "13621         5.0           9.35   Engineering & IT   Junior  5038916  ...   \n",
       "8399          2.0           9.75                 HR   Junior  6030301  ...   \n",
       "5525          4.0           9.13  Service & Support    Staff  4488998  ...   \n",
       "\n",
       "       active_work_category work_stability_score married_dependent_ratio  \\\n",
       "12683             Long-term             6.250000                       1   \n",
       "11885              Mid-term            26.000000                       1   \n",
       "13621              Mid-term             5.833333                       3   \n",
       "8399               Mid-term             8.000000                       3   \n",
       "5525              Long-term             8.200000                       1   \n",
       "\n",
       "       position_score  job_income_position_score  education_score  \\\n",
       "12683               1                  1776619.0                1   \n",
       "11885               1                  1104966.0                1   \n",
       "13621               2                  2519458.0                5   \n",
       "8399                2                  3015150.5                5   \n",
       "5525                1                  4488998.0                4   \n",
       "\n",
       "       education_income_ratio  weighted_satisfaction_performance  \\\n",
       "12683               1776619.0                                1.6   \n",
       "11885               1104966.0                                2.6   \n",
       "13621               1007783.2                                2.0   \n",
       "8399                1206060.2                                1.0   \n",
       "5525                1122249.5                                1.0   \n",
       "\n",
       "       resign_risk_indicator  adjusted_work_time  \n",
       "12683                    Low            9.884603  \n",
       "11885                 Medium            9.260000  \n",
       "13621                 Medium            9.335163  \n",
       "8399                  Medium            9.740976  \n",
       "5525                     Low            9.120106  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "X.to_csv(r\"D:\\Tugas Akhir\\Codingan\\Development\\App\\X_train.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\ttest: 0.9491030\tbest: 0.9491030 (0)\ttotal: 266ms\tremaining: 4m 26s\n",
      "200:\ttest: 0.9793245\tbest: 0.9793245 (200)\ttotal: 14.5s\tremaining: 57.8s\n",
      "400:\ttest: 0.9820451\tbest: 0.9820530 (398)\ttotal: 27s\tremaining: 40.3s\n",
      "600:\ttest: 0.9827616\tbest: 0.9827616 (597)\ttotal: 40.5s\tremaining: 26.9s\n",
      "800:\ttest: 0.9831529\tbest: 0.9831820 (788)\ttotal: 54.9s\tremaining: 13.6s\n",
      "999:\ttest: 0.9837425\tbest: 0.9837425 (999)\ttotal: 1m 9s\tremaining: 0us\n",
      "\n",
      "bestTest = 0.9837424813\n",
      "bestIteration = 999\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x1da6d681090>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from catboost import CatBoostClassifier\n",
    "import pandas as pd\n",
    "\n",
    "model = CatBoostClassifier(\n",
    "    iterations=1000,\n",
    "    learning_rate=0.01,\n",
    "    depth=6,\n",
    "    cat_features= cat_feature,\n",
    "    loss_function='Logloss',\n",
    "    eval_metric='AUC',\n",
    "    scale_pos_weight=len(y_train[y_train == 0]) / len(y_train[y_train == 1]),\n",
    "    verbose=200\n",
    ")\n",
    "\n",
    "# Melatih model\n",
    "model.fit(X_train, y_train, eval_set=(X_test, y_test), use_best_model=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\Tugas Akhir\\Codingan\\Notebook - Playground\\.venv\\lib\\site-packages\\tqdm\\auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import optuna\n",
    "from catboost import CatBoostClassifier\n",
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "# Fungsi objective untuk Optuna\n",
    "def objective(trial):\n",
    "    # Definisikan parameter yang akan dioptimasi\n",
    "    params = {\n",
    "        'iterations': trial.suggest_int('iterations', 500, 1000),\n",
    "        'learning_rate': trial.suggest_float('learning_rate', 0.001, 0.1, log=True),\n",
    "        'depth': trial.suggest_int('depth', 4, 6),\n",
    "        'subsample': trial.suggest_float('subsample', 0.5, 0.8),\n",
    "        'colsample_bylevel': trial.suggest_float('colsample_bylevel', 0.5, 0.8),\n",
    "        'l2_leaf_reg': trial.suggest_float('l2_leaf_reg', 5, 20),\n",
    "        'random_strength': trial.suggest_float('random_strength', 5, 10),\n",
    "        'cat_features': cat_feature,\n",
    "        'loss_function': 'Logloss',\n",
    "        'random_state': 42,\n",
    "        'verbose': 0\n",
    "    }\n",
    "\n",
    "    # Inisialisasi model dengan parameter yang dioptimasi\n",
    "    model = CatBoostClassifier(**params)\n",
    "\n",
    "    # Melatih model dengan validasi\n",
    "    model.fit(X_train, y_train, eval_set=(X_test, y_test), use_best_model=True)\n",
    "\n",
    "    # Prediksi probabilitas untuk menghitung AUC\n",
    "    y_pred = model.predict_proba(X_test)[:, 1]\n",
    "    auc = roc_auc_score(y_test, y_pred)\n",
    "\n",
    "    return auc  # Mengembalikan AUC sebagai skor yang ingin dimaksimalkan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[I 2025-01-31 18:51:48,706] A new study created in memory with name: no-name-64301d1e-ebb0-4383-9c80-96d0c991cd74\n",
      "[I 2025-01-31 18:52:32,766] Trial 0 finished with value: 0.9773838323749091 and parameters: {'iterations': 898, 'learning_rate': 0.011053171335873363, 'depth': 5, 'subsample': 0.6792060501802821, 'colsample_bylevel': 0.677770309507521, 'l2_leaf_reg': 6.799321957299509, 'random_strength': 9.022785536036583}. Best is trial 0 with value: 0.9773838323749091.\n",
      "[I 2025-01-31 18:53:04,401] Trial 1 finished with value: 0.9764399497653513 and parameters: {'iterations': 751, 'learning_rate': 0.010856162412518163, 'depth': 4, 'subsample': 0.7382693336863855, 'colsample_bylevel': 0.7186692583510685, 'l2_leaf_reg': 8.362893889778585, 'random_strength': 8.702993428810624}. Best is trial 0 with value: 0.9773838323749091.\n",
      "[I 2025-01-31 18:54:04,581] Trial 2 finished with value: 0.9846493489325138 and parameters: {'iterations': 923, 'learning_rate': 0.04798765627548158, 'depth': 6, 'subsample': 0.6754895839217451, 'colsample_bylevel': 0.5237363809797543, 'l2_leaf_reg': 19.582992922585383, 'random_strength': 8.441609405121403}. Best is trial 2 with value: 0.9846493489325138.\n",
      "[I 2025-01-31 18:54:58,991] Trial 3 finished with value: 0.9753136360631899 and parameters: {'iterations': 955, 'learning_rate': 0.0015368102043846466, 'depth': 6, 'subsample': 0.7011449309209402, 'colsample_bylevel': 0.7103505124550098, 'l2_leaf_reg': 5.113987964193796, 'random_strength': 8.460971358767354}. Best is trial 2 with value: 0.9846493489325138.\n",
      "[I 2025-01-31 18:55:25,136] Trial 4 finished with value: 0.9841575781611475 and parameters: {'iterations': 578, 'learning_rate': 0.063678671058667, 'depth': 4, 'subsample': 0.7677253729534457, 'colsample_bylevel': 0.7531320019655664, 'l2_leaf_reg': 15.93900442556415, 'random_strength': 5.432113965360059}. Best is trial 2 with value: 0.9846493489325138.\n",
      "[I 2025-01-31 18:56:16,998] Trial 5 finished with value: 0.9776138541873224 and parameters: {'iterations': 826, 'learning_rate': 0.009797116924664805, 'depth': 6, 'subsample': 0.5890878538611122, 'colsample_bylevel': 0.7373385731110815, 'l2_leaf_reg': 15.73191814912146, 'random_strength': 6.050835373364055}. Best is trial 2 with value: 0.9846493489325138.\n",
      "[I 2025-01-31 18:57:15,482] Trial 6 finished with value: 0.9749619935223743 and parameters: {'iterations': 809, 'learning_rate': 0.001699677627520717, 'depth': 6, 'subsample': 0.578445828928962, 'colsample_bylevel': 0.7436210936117759, 'l2_leaf_reg': 11.117190568598893, 'random_strength': 7.04383990081959}. Best is trial 2 with value: 0.9846493489325138.\n",
      "[I 2025-01-31 18:57:54,127] Trial 7 finished with value: 0.9816352700112367 and parameters: {'iterations': 861, 'learning_rate': 0.01541861782108474, 'depth': 4, 'subsample': 0.7886446490601242, 'colsample_bylevel': 0.6735700671825722, 'l2_leaf_reg': 11.919032864668194, 'random_strength': 9.492143464663224}. Best is trial 2 with value: 0.9846493489325138.\n",
      "[I 2025-01-31 18:58:51,816] Trial 8 finished with value: 0.9852865357921872 and parameters: {'iterations': 769, 'learning_rate': 0.06982641965149956, 'depth': 6, 'subsample': 0.6814518799518184, 'colsample_bylevel': 0.7775983370027374, 'l2_leaf_reg': 8.960252840750547, 'random_strength': 9.846274339599894}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 18:59:44,308] Trial 9 finished with value: 0.9754141053605658 and parameters: {'iterations': 746, 'learning_rate': 0.0016137224415552842, 'depth': 6, 'subsample': 0.5993549736218431, 'colsample_bylevel': 0.6989790297541257, 'l2_leaf_reg': 13.553876556590211, 'random_strength': 5.990982802533768}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:00:15,989] Trial 10 finished with value: 0.9838826095578029 and parameters: {'iterations': 597, 'learning_rate': 0.03144100490697419, 'depth': 5, 'subsample': 0.5363364194930027, 'colsample_bylevel': 0.7978652144122366, 'l2_leaf_reg': 8.249537524112156, 'random_strength': 9.984348987802527}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:01:10,331] Trial 11 finished with value: 0.9841760856632956 and parameters: {'iterations': 974, 'learning_rate': 0.09965915584325698, 'depth': 5, 'subsample': 0.6574603869223254, 'colsample_bylevel': 0.5501601205848138, 'l2_leaf_reg': 19.76070753486162, 'random_strength': 7.755479575146688}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:01:54,913] Trial 12 finished with value: 0.9840386013616234 and parameters: {'iterations': 691, 'learning_rate': 0.03424349599719855, 'depth': 6, 'subsample': 0.7224175773823622, 'colsample_bylevel': 0.5929425691182608, 'l2_leaf_reg': 19.25404560213221, 'random_strength': 7.759900759284147}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:02:25,479] Trial 13 finished with value: 0.9840068742150836 and parameters: {'iterations': 502, 'learning_rate': 0.038817996889561834, 'depth': 6, 'subsample': 0.6330719275681617, 'colsample_bylevel': 0.5095236622667528, 'l2_leaf_reg': 9.778730963469318, 'random_strength': 9.883633709104314}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:03:12,496] Trial 14 finished with value: 0.9760645118646308 and parameters: {'iterations': 913, 'learning_rate': 0.004711726870893525, 'depth': 5, 'subsample': 0.6341455516910446, 'colsample_bylevel': 0.6189316091763777, 'l2_leaf_reg': 14.590796785510822, 'random_strength': 8.280974259767127}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:03:54,780] Trial 15 finished with value: 0.983739837398374 and parameters: {'iterations': 684, 'learning_rate': 0.0770812480415403, 'depth': 6, 'subsample': 0.6841267228305123, 'colsample_bylevel': 0.6244873598756205, 'l2_leaf_reg': 17.527072672878234, 'random_strength': 6.997647156643534}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:04:35,490] Trial 16 finished with value: 0.9834939520126909 and parameters: {'iterations': 797, 'learning_rate': 0.02527170220411846, 'depth': 5, 'subsample': 0.5006836923455774, 'colsample_bylevel': 0.5022479394722035, 'l2_leaf_reg': 10.283699043522013, 'random_strength': 9.139463380029865}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:05:33,238] Trial 17 finished with value: 0.984297706391698 and parameters: {'iterations': 911, 'learning_rate': 0.051380928005540824, 'depth': 6, 'subsample': 0.7343100864781796, 'colsample_bylevel': 0.7831407839963198, 'l2_leaf_reg': 17.57103102855441, 'random_strength': 9.394666221253614}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:06:12,201] Trial 18 finished with value: 0.9830497719611342 and parameters: {'iterations': 668, 'learning_rate': 0.01944108063159131, 'depth': 6, 'subsample': 0.6583734530240242, 'colsample_bylevel': 0.5684416729202474, 'l2_leaf_reg': 12.834044527259307, 'random_strength': 8.062187605442572}. Best is trial 8 with value: 0.9852865357921872.\n",
      "[I 2025-01-31 19:06:52,557] Trial 19 finished with value: 0.9760327847180911 and parameters: {'iterations': 748, 'learning_rate': 0.004860150067425841, 'depth': 5, 'subsample': 0.6132382429039442, 'colsample_bylevel': 0.5408464943257619, 'l2_leaf_reg': 8.671812981324692, 'random_strength': 7.2908917904657535}. Best is trial 8 with value: 0.9852865357921872.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Trial:\n",
      "AUC: 0.9852865357921872\n",
      "Params:\n",
      "  iterations: 769\n",
      "  learning_rate: 0.06982641965149956\n",
      "  depth: 6\n",
      "  subsample: 0.6814518799518184\n",
      "  colsample_bylevel: 0.7775983370027374\n",
      "  l2_leaf_reg: 8.960252840750547\n",
      "  random_strength: 9.846274339599894\n"
     ]
    }
   ],
   "source": [
    "# Buat studi Optuna untuk memaksimalkan AUC\n",
    "study = optuna.create_study(direction='maximize')\n",
    "study.optimize(objective, n_trials=20)  # Lakukan 20 percobaan\n",
    "\n",
    "# Tampilkan hasil terbaik\n",
    "print(\"Best Trial:\")\n",
    "print(f\"AUC: {study.best_trial.value}\")\n",
    "print(\"Params:\")\n",
    "for key, value in study.best_trial.params.items():\n",
    "    print(f\"  {key}: {value}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 0.5922554\ttest: 0.5897562\tbest: 0.5897562 (0)\ttotal: 108ms\tremaining: 1m 22s\n",
      "200:\tlearn: 0.1113078\ttest: 0.1422183\tbest: 0.1422137 (199)\ttotal: 13.2s\tremaining: 37.4s\n",
      "Stopped by overfitting detector  (50 iterations wait)\n",
      "\n",
      "bestTest = 0.1357173793\n",
      "bestIteration = 347\n",
      "\n",
      "Shrink model to first 348 iterations.\n",
      "Final AUC: 0.9844008196179522\n"
     ]
    }
   ],
   "source": [
    "# Ambil parameter terbaik dari Optuna\n",
    "best_params = study.best_trial.params\n",
    "\n",
    "# Tambahkan parameter tetap (yang tidak dioptimasi)\n",
    "best_params.update({\n",
    "    'loss_function': 'Logloss',  # Gunakan Logloss sebagai loss function\n",
    "    'cat_features': cat_feature,\n",
    "    'random_state': 42,\n",
    "    'verbose': 200,  # Aktifkan output verbose\n",
    "    'od_type': 'Iter',\n",
    "    'od_wait': 50\n",
    "})\n",
    "\n",
    "# Latih model dengan parameter terbaik\n",
    "final_model = CatBoostClassifier(**best_params)\n",
    "final_model.fit(X_train, y_train, eval_set=(X_test, y_test), use_best_model=True)\n",
    "\n",
    "# Evaluasi model final\n",
    "y_pred = final_model.predict_proba(X_test)[:, 1]\n",
    "final_auc = roc_auc_score(y_test, y_pred)\n",
    "print(f\"Final AUC: {final_auc}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CatBoost Classification model saved to 'clasification_model.sav'\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "\n",
    "with open('clasification_model.sav', 'wb') as f:\n",
    "    pickle.dump(final_model, f)\n",
    "print(\"CatBoost Classification model saved to 'clasification_model.sav'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final Training Logloss: 0.07900948412964129\n",
      "Final Validation Logloss: 0.1360854668410178\n"
     ]
    }
   ],
   "source": [
    "evals_result = final_model.get_evals_result()\n",
    "\n",
    "# Menampilkan skor terakhir\n",
    "train_score = evals_result['learn']['Logloss'][-1]\n",
    "val_score = evals_result['validation']['Logloss'][-1]\n",
    "\n",
    "print(f\"Final Training Logloss: {train_score}\")\n",
    "print(f\"Final Validation Logloss: {val_score}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA04AAAIjCAYAAAA0vUuxAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAfFtJREFUeJzt3QecVOXZ9/Fr2vYKC0tbmoA06WqwF8ReYgpqEmvwiSXFGk3e2GMPGns0scRERY0lsaCIgqIICqJI7x2Wun13Zmfm/Vz32Vlml2V3gR3OmZnf9/mc50yfs/eOZP573fd1XOFwOCwAAAAAgD1y7/kuAAAAAIAiOAEAAABACwhOAAAAANACghMAAAAAtIDgBAAAAAAtIDgBAAAAQAsITgAAAADQAoITAAAAALSA4AQAAAAALSA4AQDiWs+ePeXiiy+2+zAAAAmO4AQAkOeff15cLpd8/fXXdh9K3KmurpaHHnpIDj/8cMnNzZW0tDTp16+fXH311bJkyRK7Dw8A0Ea8bfVCAADYYfHixeJ22/N3wK1bt8opp5wis2fPljPOOEMuuOACycrKMsf0yiuvyNNPPy1+v9+WYwMAtC2CEwDAMWprayUUCklKSkqrn5Oamip20SmC33zzjbz++uvyox/9qMF9d955p/zxj3+0bVwAAG2LqXoAgFZbv369XHrppVJYWGgCy6BBg+TZZ59t8BitsNxyyy0ycuRIM3UtMzNTjj76aPnkk08aPG7VqlVmeuCDDz4oDz/8sBx00EHmNRcsWCC33XabuW/ZsmUmnOTl5ZnXuuSSS6SysrLZNU6RaYeff/65XHvttdKhQwdzDD/84Q9ly5YtDZ6rYUTfq0uXLpKRkSHHH3+8ef/WrJuaOXOmvPvuu3LZZZftFpqU/iz6s0Ucd9xxZmtM30ffr6Vx0YDm9Xrl9ttv3+01tMKlz3nsscfqb9u5c6f87ne/k6KiIvP8Pn36yH333Wd+ZgDA3qPiBABolc2bN8sPfvAD8wVd1+9oIHn//fdNcCgtLTVf0pVe/vvf/y7nn3++jB8/XsrKyuQf//iHnHzyyTJr1iwZNmxYg9d97rnnzDqhyy+/3HzBb9euXf19P/3pT6VXr15yzz33yJw5c8zrduzY0QSAlvz617+W/Px8ufXWW00Y0RCixz1x4sT6x9x8881y//33y5lnnmmO79tvvzV7PZ6W/Pe//zX7X/ziFxILjcelc+fOcuyxx8qrr75qfqZo+jN5PB75yU9+Yq5ruNTHatD9v//7P+nevbt88cUX5ufduHGjGQsAwN4hOAEAWkWnnQWDQZk3b560b9/e3ParX/3KBCSt2ugX9PT0dBNWNKhETyvTANW/f3959NFHTYiKtm7dOlNZ0iDW2PDhwxs8ftu2beZ6a4KTHuOHH35ogp7SSssjjzwiJSUlpnqlQXDChAlyzjnnyJtvvln/PK3o6M/TkoULF5r9IYccIrHQ1LiMGzfOjPP3338vgwcPbhCcNChpJVDpz7V8+XJTperbt6+5TZ+nlbUHHnhArrvuOlOJAgC0HlP1AAAtCofD8p///MdUZvSyNkWIbFqh0TCiFSGllY9IaNKwsn37drNGZ9SoUfWPiabT3JoKTZFgFk2n/Gl40qpWS7RSEwlNkedq8Fu9erW5PmXKFHNcV1555W6VqtaIHEN2drbEQlPjcu6555rpetFVMw1ROr1QQ1XEa6+9Zn5eDbHRv6sxY8aYMfj0009jcswAkMioOAEAWqRrg3TNjHaJ060pxcXF9ZdfeOEF+ctf/iKLFi2SQCBQf7tOu2usqdsidIpZNA0CaseOHZKTk9PsMTf3XBUJULr2J5pOFYw8tjmR99epiLoGq601NS4FBQVy4oknmul62nxCaYjSMKWhKmLp0qXy3Xff7TGQRv+uAACtQ3ACALQo0lDg5z//uVx00UVNPmbIkCFm/69//cs0PNApcDfccINZk6RVKF2npNPHGtPpfXuiz2uKVr1asj/PbQ2deqh06qJWd1qi1a+m3lsrQE3Z07icd955pknG3LlzzXoxDVEapjRURf++TjrpJLnxxhubfA09zxQAYO8QnAAALdLKhU5J0y/5Ot2rOdqau3fv3vLGG280mCrXuKGB3Xr06GH2uo4ourqjUwEjVanm6LRFDYMaFFsTnLSKtWLFit1uj1S+WksDqa5XikzX05PsatOHaNqJr7y8vMXfFQCg9VjjBABoVfVG19zoOiddU9NYdJvvSKUnurqirbtnzJghTqJVGp3i9uSTTza4Pbqld3NGjx5tTn6rnf7eeuut3e7XtuzXX399gzCjUxejx0q7+Gnb9L2h0wJ1XZlWmvQku7qeTMNUNO1GqOP9wQcf7PZ8nXKpa7sAAHuHihMAoJ6ek2nSpEm73f7b3/5W7r33XnMupsMPP9x0yRs4cKBp/KANHz766CNzWZ1xxhmm2qTnTTr99NNl5cqV8tRTT5nHaxXEKbQDnf5cuhbrrLPOMiFIg4y2WNdpb9HVsj355z//KWPHjjXri7QCpWFMzxmla4w01Gjr78i5nPT8V9rtTkOPtnDXdUY6LnourNY0u4imjSB02uQTTzxhXq/xGiudIqnt0vV3odMm9ZxaFRUVZlqhVgS162H01D4AQMsITgCAeo2rLxH65btbt27mPEx33HGHCUb6pV1bfusX/+j24PrYTZs2yd/+9jdT8dDApNPZtNPb1KlTxUn0uPXEt88884wJf1pF0hbmRx11lKSlpbVqCqOeH0nHQqfOact2rTTpNEANYxrMIgYMGGCClp4cWE/Mq+Py4osvyksvvbTX46KvrWugtDFFdDe9CP2Zpk2bJnfffbcZd31fbWaha5u03bq2YwcA7B1XuK1WyQIAkAB0KpuuR7rrrrtMEAIAQLHGCQCQtKqqqna77eGHHzb74447zoYjAgA4FVP1AABJS6fXPf/883LaaadJVlaWTJ8+XV5++WWzbunII4+0+/AAAA5CcAIAJC0995R21rv//vtNg4ZIwwidpgcAQDTWOAEAAABAC1jjBAAAAAAtIDgBAAAAQAuSbo1TKBSSDRs2SHZ2dqtObggAAAAgMemqJT0nXpcuXcTtbr6mlHTBSUNTUVGR3YcBAAAAwCHWrl1rTvTenKQLTlppigyOnkXdboFAwJylXlvf+nw+uw8n4TC+sccYxx5jHFuMb+wxxrHHGMceY5yY46sdVbWoEskIzUm64BSZnqehySnBKSMjwxwL/xG2PcY39hjj2GOMY4vxjT3GOPYY49hjjBN7fFuzhIfmEAAAAADQAoITAAAAALSA4AQAAAAALUi6NU4AAADYP8Fg0KxJSSb683q9XqmurjY/P+JnfHXNlMfj2e/XITgBAACg1crLy2XdunXm/DfJRH/eTp06mc7MnAs0vsZXX09bjWdlZe3X6xCcAAAA0CpaCdDQpN3POnTokFQBIhQKmdCoX75bOlEqnDO+Gsi2bNliPrd9+/bdr8oTwQkAAACtnk6lX0Q1NKWnp0uyfbH3+/2SlpZGcIqz8dXP66pVq8znd3+Ck+2/9ccff1x69uxpBunwww+XWbNmNfv4nTt3ylVXXSWdO3eW1NRU6devn7z33nsH7HgBAACSXTJVmhD/XG30ebW14jRx4kS59tpr5amnnjKh6eGHH5aTTz5ZFi9eLB07dtzt8ZpCTzrpJHPf66+/Ll27dpXVq1dLXl6eLccPAAAAIDnYGpwmTJgg48ePl0suucRc1wD17rvvyrPPPis33XTTbo/X27dv3y5ffPFF/RmFtVoFAAAAAAkZnLR6NHv2bLn55pvrb9P5jGPGjJEZM2Y0+Zz//ve/Mnr0aDNV7+233zbzFS+44AL5/e9/v8f5ijU1NWaLKC0tNXud4+iENpqRY3DCsSQixjf2GOPYY4xji/GNPcY4ccY4ssZJ16PolkwiXQQjP3/v3r3lt7/9rdlaY+rUqXLiiSfKtm3b4n621O23326+i8+ZMydm49uW9PX0dZta47Q3/83YFpy2bt1qOrMUFhY2uF2vL1q0qMnnrFixQj7++GP52c9+ZtY1LVu2TK688krzA996661NPueee+4xv9zGPvzwQ9MRxikmT55s9yEkNMY39hjj2GOMY4vxjT3GOP7HWM+zoy2jtfuZ/hE8HuTn5zd7v/4BvqmZTntSVlZm9h999JH5Lhn5o3xLBg8ebL7j6nqb1j5nX0yfPl3OPPNM0wwhNzc3Ju9RU1NjvsfH4ueIjG9b0s9qVVWVfPrpp1JbW9vgvsrKysTsqqdpUdc3Pf300yYtjhw5UtavXy8PPPDAHoOTVrR0HVWE/oKLiopk7NixkpOTI3bT0Kf/yOnarcj0Q7Qdxjf2GOPYY4xji/GNPcY4ccZYT06q59nRltHa2Cse6HfFiFdffdV8Z1y4cGH9bfqzRM7vo1UJDQQaEBvT+/RLfXZ2tgk/+/I9sqCgQGItUhjQ44zVd93U1FTzXbwtX7/x+Lb151a7QB5zzDG7fW73JvzZFpz0g6MDvnnz5ga363X9S0ZTtJNe4zP/DhgwQDZt2mSSZEpKSpO/WN0a09dx0j/eTjueRMP4xh5jHHuMcWwxvrHHGMf/GGuo0C+1urxCN/2yWxUIih3SfZ5WfcHu0qVL/WWdIqfPidym0+eOP/54M5Pp//2//yfz5s0zs5L0j+z6h/cvv/xSKioqzPfNP//5z3LYYYfV//y6zv53v/ud2ZTe/swzz5j1+h988IFpYvaXv/xFzjrrrAbvtWPHDnMczz//vHmuNkvTvQbSo446Sp577jnznVdpdUSP45///Kf5/vvLX/7SfO8tKSmRt956q8mfN9LKO/I7akzfX6cX/u9//zOVo2OPPVYeeeQRc46jCP057rjjDjOtUBu3HX300ea6dreO/KzR76XFjbvuussUN/ScSTpe9957r5xyyinmfv2erj/Hf/7zH/P+OsPsV7/6lSlw6GdIZ4dpLwPNAe3bt5cf//jH5pjaih6nHnNT/33szX8vtgUnDTlaMZoyZYqcc8459YOu16+++uomn3PkkUfKSy+9ZB4X+UUtWbLEfLiaCk0AAACIHQ1NA2/5wJb3XnDHyZKR0jZfZXWq3oMPPmjWLenUPg0xp512mglL+gd4DS5nn322OW3OoEGD9vg6GgDuv/9+Mxvq0UcfNctLtAN0u3btmny8ThPT933xxRfNd9uf//zncv3118u///1vc/99991nLmuY0jDy17/+1QQmDWD76uKLL5alS5ea3gFaMdKpivqzLliwwISIzz//3IQafW8NfTol8U9/+lOzr6nHpSHxb3/7mwwfPtyEIH3u/PnzTSDTEKTvpxW/7t27m/HVTWmYeuihh8x3fL1Pg6oGWCeydaqeJs+LLrpIRo0aZRK8tiPXwYp02bvwwgtNWtd1SuqKK66Qxx57zKTkX//61+aXfvfdd8tvfvMbO38MAAAAxDGtpug0xwgNOkOHDq2/fuedd8qbb74p77//frPBSUPJ+eefby7rd1QNDBq2IpWXpqZYalfpgw46yFzX4oEeS4SGL63K/PCHPzTX9Xvw/py/NBKYNBwdccQR5jYNZlph00D2k5/8xLznqaeeagKc0nOmakfrd955Z4+vq+FPA9h5551nrmvo+uSTT8x3ez1n65o1a0yA0oqaVn569OhR/1y9T2ebaYM4XYekYe4HP/iBOJGtwWncuHGmnHfLLbeYsuOwYcNk0qRJ9Q0jdCCjS4z6S9XS5zXXXCNDhgwxoUpDlP6i4tHiTWWydFOJrK+w+0gAAAD2bbqcVn7seu+2on/Ej6bNL2677TYz7W7jxo1mypx+qV+3bl2zr6PfTyMyMzNNCCguLm52PVIkNCmdRRV5vE7H06lrWlyIiKzx39euc7q2S9dv6flTI3Rq3MEHH1y/7kvPpxoJahF6DO/sITjpGqENGzaYmWHR9Pq3335bHyg1mOr7aIg844wzTL8BpWFNA1afPn3khBNOMJUqre41tc7MbrYfkSbrPU3N07mgjWk7cp1vmghen71WnvlspZzQxS3j7T4YAACAvaTVg7aaLmcnDTnRtNqizTa0kqJf6LWxgK67aal1deP1Mjo+zYWcph4facudSEaMGCErV640FTud+vfTn/7UVJhef/11UxjRsKZry7SaprlAp/1NmzbNcWsid18xhgOmY2CdHOP+VjoENth9KAAAAKijU9m0SqKVl0MOOcRMJdP23geSthLXWVhfffVVg+Yc+3PuJF0npdWzmTNn1t+mDSA0uAwcONBc16pQ9HuqxtejaVVNm23omEXT65HXjDxOZ5tp4wltiKFrm7Zv327u02CqLdR1ip+eekjP6erEdU7x/yeCODZiy1syPuXf8k7N6SJyqd2HAwAAABGzHueNN94wX+a1CqTNEew44a+u6de1/lr16t+/v1l/pF3pWtNNUIOHtvaO0Ofoui2dBjd+/HjTyEHv18YYuvxFb4+8p7btnjBhgvn5NchopcjVzHvecMMNps27TjvUpTfazGLu3Ln1TS70tXQaojaO0GU4r732mgmjke6CGggPPfRQM8ZahdIgFb0OyikITnbyWJ0A3eGGJ+ICAACAffSL/qWXXmoaKOgpdHQ9fSxPWrsn+r7aB0Abpun6pssvv9y0B48+Nc+eaPiJps/RapOGGu0RoOuMtE24Pk6nyEWmxenaJG1YoR0CtUW7vp/2F3jsscf2+F7aqE3XZF133XVmjZZWmrQJRaTFuQY07TaozSn0ODQk6XtqiNLwpK3LtWmcBiit8GmrdF175TSucCJOpGyGfui19Km/XLtPgDv7hd/LyJVPyQfeE+WE30903DzORKBzkfU/TG2zyfjGBmMce4xxbDG+sccYJ84Y64lEda1Kr1694uYEuG1FqyH6PVK/PzZ1fqQDdQw63U7XCGmnvwNFK1SLFi2Szz77LC7Ht7nP7d5kAypODqg4eYSKEwAAABrSc0Bp0wQ9Sa2erFarPhoALrjggpi+rzbF0C542jRDp+m98MIL8sQTT0iyIzjZyOVlqh4AAACappUXXQOkXf50ktjgwYNNVzqtOsWSnntKp9aVlZWZkwLr+ah++ctfSrIjODkgOHnDzbe2BAAAQPLRVt2Nu9UdCK+++uoBf894QDtyG7m8qWbvCQftPhQAAAAAzSA42cgdqTgJFScAAADAyQhONnLXVZy8VJwAAAAARyM42cjtqwtOVJwAAAAARyM42cjti0zVo6seAAAA4GQEJxt5fNYJuLy0IwcAAAAcjeBkI0/dVD0fFScAAABHO+GEE+Tmm2+uv96zZ095+OGHm32Oy+WSt956a7/fu61ex24XX3yxnHPOORKvCE428tQ1hyA4AQAAxMaZZ54pp5xySpP3ffbZZyaUfPfdd3v9ul999ZVcfvnl0pZuu+02GTZs2G63b9y4UU499VSJJT3Rbl5eXkzfI94RnGzkTYkEJ5pDAAAAxMJll10mkydPlnXr1u1233PPPSejRo2SIUOG7PXrdujQQTIyMuRA6NSpk6SmWt8bYR+CkyOm6tGOHAAAxKFwWMRfYc+m790KZ5xxhgk5WlGJVl5eLq+99poJVtu2bZPzzz9funbtasLQIYccIi+//HKzr9t4qt7SpUvlmGOOkbS0NBk4cKAJa439/ve/l379+pn36N27t/zpT3+SQMD6A7oe3+233y7ffvutqYLpFjnmxlP15s2bZ6YOpqenS/v27U3lS3+exlPiHnzwQencubN5zFVXXVX/XvtizZo1cvbZZ0tWVpbk5OTIT3/6U9m8eXODx9x1113SsWNHyc7Oll/+8pdy0003NVlBi6ipqZHf/OY35jk6JloZ1EpexI4dO+RnP/uZ+f3pz9q3b18TdpXf75err77a/Hw65j169JB77rlHYskb01dHs3ypVnMIpuoBAIC4FKgUubuLPe/9hw0iKZktPszr9cqFF15oQsgf//hHE0KUhqZgMGgCk4aOkSNHmmCjoeDdd9+VX/ziF3LQQQfJYYcd1uJ7hEIhOffcc6WwsFBmzpwpJSUl8rvf/W63x2mg0OPo0qWLCT/jx483t914440ybtw4+f7772XSpEny0Ucfmcfn5ubu9hoVFRVy8skny+jRo03IKC4uNiFFQ0R0OPzkk09MqND9smXLzOtriNH33Fv6851dF5qmTZsmtbW1Jojpa06dOtU85t///rf8+c9/lieeeEKOPPJIeeWVV+Qvf/mL9OrVa4+vqz/3f/7zH3nhhRekqKhI7r77bjMlUY+3Xbt2JlguWLBA3n//fSkoKDC3V1VVmec+8sgj8t///ldeffVV6d69u6xdu9ZssURwspG3ruKUIgGpDYbE57P7iAAAABLPpZdeKg888ID50n/ccceZ27Ry8aMf/ciEE92uv/76+sf/+te/lg8++MB8KW9NcNKgs2jRIvMcDUUqEgKi/b//9/8aVKz0PTVgaIDQiooGEw16OjVvT1566SWprq6Wf/7zn5KZaQXHxx57zKzluu+++0x4U/n5+eZ2j8cj/fv3l9NPP12mTJmyT8FJnzdv3jxZuXKlCThK33/QoEEmvB166KHy6KOPmurdJZdcYu6/5ZZb5MMPP2xQCWscAJ988kkT9nScNJz99a9/NeHuH//4h9xwww2myjV8+HAznTIyZhF6n1agjjrqKBOGteIUawQnG/nq5qpqxakiGJZ0uw8IAABgb/gyrMqPXe/dShocjjjiCHn22WdNcNLKhTaGuOOOO8z9WnnSoKNBaf369WYamE4ja+0apoULF5pAEQlNSitCjU2cONFUSpYvX24ChVZutMK1N/S9hg4dWh+alFZ4NHgsXry4PjhpqNHQFKHVJw0/+yLy8xXVhSal0xG1mYTep8FJ3/vKK69s8DwNnR9//HGTr6ljoFMH9dgjfD6feS19TXXFFVeYcDtnzhwZO3asmX6ov8fIdMSTTjpJDj74YDPFT6dk6mNiiTVONvKlWFHJ4wqLP+C3+3AAAAD2jk570+lydmx1U+5aS6shOi2srKzMVJt0Gt6xxx5r7tNqlFY7dKqeTm2bO3eumQ6nAaqtzJgxw6zXOe200+Sdd96Rb775xkwdbMv3iKYhJJpWZTRcxZNTTz1VVq9eLddcc41s2LBBTjzxxPrK4IgRI0wF7M477zTT93TN1Y9//OOYHg/ByUZeX0r95YC/xtZjAQAASGT6xdrtdpupbjrNTKfvRdY7ff7552YNz89//nNTzdHGDUuWLGn1aw8YMMCsr9G24RFffvllg8d88cUXZjqZhiWdeqbTzDQUREtJSTHVr5beSxtI6FS3CD1+/dm0+hILkZ9vbdQaIl17tHPnTlN5Uvre0Y0dVOPr0TS46s+rxx6hFaivv/66/jWVNoa46KKL5F//+pdpxvH000/X36fVOl1n9cwzz5hqngbj7du3S6wwVc9GLq/VHEIF/NW2HgsAAEAi0/VD+iVbT2JbWlpqpnpFaIh5/fXXTbjRtUETJkwwHeOiv8A3Z8yYMaZbnn7B1+qVvr4GpGj6HrouR9c06XQ0bUDx5ptvNniMruHRKopWvLp162YaRzRuQ65Vq1tvvdW8l573acuWLWZNljaziEzT21ca2vS9o+n76893yCGHmPfW8KJTDHVanlbsIuuP9Bh0/ZRe1+l0GmT0/FgaQpuiUw11Kp6uZdJGEPrz6nTJyspKUx2MrJPSph067VCnTmqlTkOc0t+RTj/UNVAaGrXZh64Ni+W5qKg42cm9K7cGqTgBAADElH4h1xbXOg0vej2SNm3QqV96u66B0i/gup6mtfSLu4YgnTKm63q0y512mIt21llnmSln2v1OGyBoSNOucdF0PY+u1zn++ONNpaWplui67kqbUGhlRQOYTk/TKWzaCGJ/6borDSLRmzad0Mrc22+/bUKltlzXIKWBSMNRhIYqDaU6lS4yjU7DqbYK35N7773X/Mwa+jRwrVixwnTQ0/dRWpHS19TzbOn76potDZ5KQ+X9999vnqfjsGrVKnnvvffM7yJWXOFwK5vgJwj9C4B2TtE2kXu7GC8W/Le1lxSplSXnfyH9Dh5k9+EkHC356n9EOp+48VxftA3GOPYY49hifGOPMU6cMdZubvqFWFtMN/eFOBHp+iD9HqnfH2P55TyRnHTSSSaEvvjii7aOb3Of273JBkzVs1lAfCY41TJVDwAAAHGqsrJSnnrqKVO108qQVsu0TXtTJwKOVwQnmwXqfgXBAFP1AAAAEJ9cLpepeuoURa3waLMIbdag0/oSBcHJZrUun0iY4AQAAID4lZ6ebipMiYwJmjarjVScajmPEwAAAOBUBCcnVJx0QRwVJwAAECeSrLcY4ly4jT6vBCeHBCem6gEAAKfTRf/K72emDOJH5PMa+fzuK9Y42Szo9okERcJM1QMAAA7n9XrNeYT0pKva9jyZ2nJru2z9Aq6ND5Lp54738Q2FQubzqp9b/fzuD4KTzYL1FSeCEwAAcH7ntM6dO5tz4qxevVqSbbqXnuBWmyDoOCB+xleDWPfu3ff7dQlODglOUstUPQAA4HwpKSnSt2/fpJuupycZ/vTTT+WYY47hRM5xNr76mW2LKhbByWYhd4q1DxKcAABAfNAvoWlpaZJMdH1MbW2t+bkJTsk5vkzQtFnIbWVX1jgBAAAAzkVwsllIm0MQnAAAAABHIzg5ZKqeBAlOAAAAgFMRnGwW9hCcAAAAAKcjONksXDdVT4IBuw8FAAAAwB4QnBxScXLRVQ8AAABwLIKT3ag4AQAAAI5HcLKbN9XsXCHWOAEAAABORXCym8eqOLmpOAEAAACORXCyGxUnAAAAwPEITjZz1TWHcIeoOAEAAABORXByTHCi4gQAAAA4FcHJZi6vFZw8VJwAAAAAxyI4OSQ4ucMEJwAAAMCpCE42c/us5hBUnAAAAADnIjjZzFXXVc9LxQkAAABwLIKTzTyRNU7hWrsPBQAAAMAeEJwcMlWPihMAAADgXAQnm3kia5wITgAAAIBjEZxs5o1UnISpegAAAIBTEZxs5vFZa5x8VJwAAAAAxyI4OaTi5KPiBAAAADgWwclmnpQ0s/dJQMLhsN2HAwAAAKAJBCeHVJxSpFaCIYITAAAA4EQEJ5v56itOteIPhuw+HAAAAABNIDjZzJtS147cFZaAnwYRAAAAgBMRnBwyVU/5/dW2HgsAAACAphGcbOby7gpOAYITAAAA4EgEJ7u5vfUXa/01th4KAAAAgKYRnOzmcklN2Gcu1lJxAgAAAByJ4OQAAfGYPRUnAAAAwJkITg4QEKviFAwQnAAAAAAnIjg5QMBlrXNiqh4AAADgTAQnB6gVKzgFa/12HwoAAACAJhCcHCBQF5xCASpOAAAAgBMRnJxUcWKNEwAAAOBIjghOjz/+uPTs2VPS0tLk8MMPl1mzZu3xsc8//7y4XK4Gmz4vntVG1jgRnAAAAABHsj04TZw4Ua699lq59dZbZc6cOTJ06FA5+eSTpbi4eI/PycnJkY0bN9Zvq1evlngWrJ+qR3ACAAAAnMj24DRhwgQZP368XHLJJTJw4EB56qmnJCMjQ5599tk9PkerTJ06darfCgsLJZ4FXFY7ctY4AQAAAM5klTps4vf7Zfbs2XLzzTfX3+Z2u2XMmDEyY8aMPT6vvLxcevToIaFQSEaMGCF33323DBo0qMnH1tTUmC2itLTU7AOBgNnspsdQ60oxl4M1FY44pkQSGU/GNXYY49hjjGOL8Y09xjj2GOPYY4wTc3z35v1c4XA4LDbZsGGDdO3aVb744gsZPXp0/e033nijTJs2TWbOnLnbczRQLV26VIYMGSIlJSXy4IMPyqeffirz58+Xbt267fb42267TW6//fbdbn/ppZdMZcsJOs17VA6v/UpeybpI0vueaPfhAAAAAEmhsrJSLrjgApMrdDmQYytO+0IDVnTIOuKII2TAgAHyt7/9Te68887dHq/VLF1DFV1xKioqkrFjx7Y4OAcq5c75/m/mcmG7HDnqtNPsPqSEouM7efJkOemkk8Tns6ZEom0xxrHHGMcW4xt7jHHsMcaxxxgn5vhGZqO1hq3BqaCgQDwej2zevLnB7Xpd1y61hg7s8OHDZdmyZU3en5qaaramnueUD33QbR2HK+h3zDElGif9vhMVYxx7jHFsMb6xxxjHHmMce4xxYo3v3ryXrc0hUlJSZOTIkTJlypT623Tdkl6Prio1JxgMyrx586Rz584Sr4J1a5xctTSHAAAAAJzI9ql6Oo3uoosuklGjRslhhx0mDz/8sFRUVJgue+rCCy8066Duuecec/2OO+6QH/zgB9KnTx/ZuXOnPPDAA6Yd+S9/+UuJV8G68zi5amlHDgAAADiR7cFp3LhxsmXLFrnllltk06ZNMmzYMJk0aVJ9i/E1a9aYTnsRO3bsMO3L9bH5+fmmYqXNJbSVebwKuusqTkEqTgAAAIAT2R6c1NVXX222pkydOrXB9YceeshsiSRUt8bJHaTiBAAAADiR7SfAhQYnq+LkpuIEAAAAOBLByQHCdRUnT8hv96EAAAAAaALByUEVJ2+IihMAAADgRAQnB6DiBAAAADgbwckJ6oKTL0RzCAAAAMCJCE5O4KkLTmEqTgAAAIATEZwcIFy3ximF4AQAAAA4EsHJSRUnITgBAAAATkRwcgCXx6o4pVJxAgAAAByJ4OQArrqKU4oEJBQK2304AAAAABohODmAy2sFpzTxiz8YsvtwAAAAADRCcHIAd2SNkyso1TW0JAcAAACchuDkBHVrnJS/usrWQwEAAACwO4KTAwRdVsVJ1VRX2HosAAAAAHZHcHICl1v84jUXAzVUnAAAAACnITg5hF+s6XqB6kq7DwUAAABAIwQnh/C76oJTDcEJAAAAcBqCk8MqTrVM1QMAAAAch+DkEAG3FZyCfoITAAAA4DQEJ4cIuFLNPkjFCQAAAHAcgpND1EYqTgGCEwAAAOA0BCeHCLqtilOIqXoAAACA4xCcnBacqDgBAAAAjkNwcoigxwpOUltt96EAAAAAaITg5BChuuAUChCcAAAAAKchODlEyJNm9i4qTgAAAIDjEJycIjJVj4oTAAAA4DgEJ4cIeesqTkGCEwAAAOA0BCen8FoVJ1dtjd1HAgAAAKARgpNT+NLNzk3FCQAAAHAcgpNDuOoqTu6g3+5DAQAAANAIwckhXHUVJ0+IihMAAADgNAQnh3D7rOYQHipOAAAAgOMQnBzCnWJVnHxhKk4AAACA0xCcHMITqTiFqDgBAAAATkNwcghPXcUpJUw7cgAAAMBpCE4O4UnJMHtfOGD3oQAAAABohODkEN40Kk4AAACAUxGcHMJXP1WPNU4AAACA0xCcHMKbWhechKl6AAAAgNMQnBwiNc1a45QmfgkEQ3YfDgAAAIAoBCeH8NUFJ58rKDV+pusBAAAATkJwcojUuql6qrqqwtZjAQAAANAQwckhXHXtyJWf4AQAAAA4CsHJKVxu8YvXXPRXV9l9NAAAAACiEJwcpFpSzb6GihMAAADgKAQnB6lxWcGptobgBAAAADgJwclB/HXBKVBNcAIAAACchODkIAF3mtlTcQIAAACcheDkwOAUJDgBAAAAjkJwcpDa+uBEVz0AAADASQhODlLrsdY4hfyVdh8KAAAAgCgEJwcJedLNPkxwAgAAAByF4OQgQY81VS8cYKoeAAAA4CQEJwcJ+6yKkytAxQkAAABwEoKTg4S9VnCSWipOAAAAgJMQnBxYcXITnAAAAABHITg5iMuXYfYEJwAAAMBZCE4O4kqxgpMnWG33oQAAAACIQnByEE+KNVWP4AQAAAA4C8HJgRUnL8EJAAAAcBSCk4N4UjPN3heqsftQAAAAAEQhODmIL9WqOPnCVJwAAAAAJyE4OYg3zao4pYSpOAEAAABOQnBykJS64JRKcAIAAAAcheDkwIoTwQkAAABwFoKTg6SmZ5l9utRIKBS2+3AAAAAA1CE4OUhqulVxSnMFpCZQa/fhAAAAAKhDcHKQtAyr4qSqqipsPRYAAAAAuxCcHMRTdwJcVVNVbuuxAAAAANiF4OQkbo/UiM9c9FNxAgAAABzDEcHp8ccfl549e0paWpocfvjhMmvWrFY975VXXhGXyyXnnHOOJIpqSTV7Kk4AAACAc9genCZOnCjXXnut3HrrrTJnzhwZOnSonHzyyVJcXNzs81atWiXXX3+9HH300ZJI/C4rONVWU3ECAAAAnML24DRhwgQZP368XHLJJTJw4EB56qmnJCMjQ5599tk9PicYDMrPfvYzuf3226V3796SSGrqglOA4AQAAAA4htfON/f7/TJ79my5+eab629zu90yZswYmTFjxh6fd8cdd0jHjh3lsssuk88++6zZ96ipqTFbRGlpqdkHAgGz2S1yDPX7uuDkryxzxPHFu8bji7bHGMceYxxbjG/sMcaxxxjHHmOcmOO7N+9na3DaunWrqR4VFhY2uF2vL1q0qMnnTJ8+Xf7xj3/I3LlzW/Ue99xzj6lMNfbhhx+aypZTTJ482ez7hjxmv3TxfNlQaYUotN34InYY49hjjGOL8Y09xjj2GOPYY4wTa3wrKyvjIzjtrbKyMvnFL34hzzzzjBQUFLTqOVrN0jVU0RWnoqIiGTt2rOTk5IjdNOXqB+Skk04Sn88nSxdMEKkR6dGlkxx22ml2H17cazy+aHuMcewxxrHF+MYeYxx7jHHsMcaJOb6R2WiOD04afjwej2zevLnB7Xq9U6dOuz1++fLlpinEmWeeWX9bKBQye6/XK4sXL5aDDjqowXNSU1PN1pj+Qpz0oY8cT9CbboKT1FY56vjindN+34mIMY49xji2GN/YY4xjjzGOPcY4scZ3b97L1uYQKSkpMnLkSJkyZUqDIKTXR48evdvj+/fvL/PmzTPT9CLbWWedJccff7y5rJWkeBf0pJl9OFBl96EAAAAAcMpUPZ1Gd9FFF8moUaPksMMOk4cfflgqKipMlz114YUXSteuXc1aJT3P0+DBgxs8Py8vz+wb3x6vQlpx0uDkb/18SwAAAAAJHpzGjRsnW7ZskVtuuUU2bdokw4YNk0mTJtU3jFizZo3ptJcsQnUVJ1ctFScAAADAKWwPTurqq682W1OmTp3a7HOff/55SSRhn9XpzxWg4gQAAAA4RfKUcuKFz5qqR8UJAAAAcA6Ck8O46oPTrpP2AgAAALAXwclhXCnWVD1PkIoTAAAA4BQEJ4dxpVgVJ2+w2u5DAQAAAFCH4OQwnrqKkzdEcAIAAACcguDkMJ7UTLP3EZwAAAAAxyA4OYy3PjjRHAIAAABwCoKTw/jSrOCUSsUJAAAAcAyCk8OkZWabfWqY4AQAAAA4BcHJYdKy8s0+SyolFArbfTgAAAAACE7Ok5ljBacMV41UVFN1AgAAAJyA4OQwqZm59ZcrynbYeiwAAAAALAQnh3F5U6VafOZyVelOuw8HAAAAAMHJmSrE6qxXXUFwAgAAAJyA4ORAVe4Ms68pJzgBAAAATkBwcqAat1VxClSU2H0oAAAAAAhOzlTjzTL72ioqTgAAAIATEJwcKOC1Kk7BqlK7DwUAAAAAwcmZgr5s60INwQkAAABwAoKTA4VSrKl6Ul1m96EAAAAAIDg5Uzg1x+zdfoITAAAA4AQEJydKtabqeQIEJwAAAMAJCE4O5EnPNXtfbbndhwIAAACA4OT04FRh96EAAAAA2NfgNGnSJJk+fXr99ccff1yGDRsmF1xwgezYsaMtjy8p+TKs4JQWpOIEAAAAxG1wuuGGG6S01GqVPW/ePLnuuuvktNNOk5UrV8q1117b1seYdFIyreCUHqq0+1AAAAAAiIh3X56kAWngwIHm8n/+8x8544wz5O6775Y5c+aYAIX9k5qVb/YZYabqAQAAAHFbcUpJSZHKSqsa8tFHH8nYsWPN5Xbt2tVXorDvMrKt4JQpVRIKhe0+HAAAACDp7VPF6aijjjJT8o488kiZNWuWTJw40dy+ZMkS6datW1sfY9LJzLGCU6qrVsoqyyU7y2pPDgAAACCOKk6PPfaYeL1eef311+XJJ5+Url27mtvff/99OeWUU9r6GJNOaoZ1AlxVWUazDQAAACAuK07du3eXd955Z7fbH3roobY4pqTn8nilQtIkU6qlqnSHSOfudh8SAAAAkNT2qeKkTSC0m17E22+/Leecc4784Q9/EL/f35bHl7QqJNPsqypK7D4UAAAAIOntU3D6v//7P7OeSa1YsULOO+88ycjIkNdee01uvPHGtj7GpFTtzjB7f/lOuw8FAAAASHr7FJw0NOkJb5WGpWOOOUZeeuklef755017cuy/ao9VcQpUUnECAAAA4jI4hcNhCYVC9e3II+duKioqkq1bt7btESYpf11wClZRcQIAAADiMjiNGjVK7rrrLnnxxRdl2rRpcvrpp9efGLewsLCtjzEpBXxWC/JgFefFAgAAAOIyOD388MOmQcTVV18tf/zjH6VPnz7mdm1PfsQRR7T1MSalWl+WdaGa4AQAAADEZTvyIUOGNOiqF/HAAw+Ix+Npi+NKeqGUupPe1pTZfSgAAABA0tun4BQxe/ZsWbhwobk8cOBAGTFiRFsdF1Kt4OT2l9t9JAAAAEDS26fgVFxcLOPGjTPrm/Ly8sxtO3fulOOPP15eeeUV6dChQ1sfZ9JxpeWavTfAVD0AAAAgLtc4/frXv5by8nKZP3++bN++3Wzff/+9lJaWym9+85u2P8ok5MloZ/YpAdqRAwAAAHFZcZo0aZJpQz5gwID623Sq3uOPPy5jx45ty+NLWr7s9mafXktwAgAAAOKy4qTncPL5fLvdrrdFzu+E/ZOaU2D2WUGm6gEAAABxGZxOOOEE+e1vfysbNmyov239+vVyzTXXyIknntiWx5e0MnKt4JQdpjkEAAAAEJfB6bHHHjPrmXr27CkHHXSQ2Xr16mVue/TRR9v+KJNQVr51IuEsV5UE/NV2Hw4AAACQ1PZpjVNRUZE5Aa6uc1q0aJG5Tdc7jRkzpq2PL2ll57aXUNglbldYynYUS7vC7nYfEgAAAJC09vk8Ti6XS0466SSzoe15fT4pkQzJlQop37mV4AQAAADEQ3B65JFHWv2itCRvG6XuHMkNV0hVyRa7DwUAAABIaq0OTg899FCrK1EEp7ZR4c4RCW4Uf+lWuw8FAAAASGqtDk4rV66M7ZFgN9U+DU4iteXb7D4UAAAAIKntU1c9HBg1vlyzr60gOAEAAABx1xzi2muv3eM0vbS0NOnTp4+cffbZ0q5du/09vqRWm5ovUiYiVdvtPhQAAAAgqe1TcPrmm29MO/JgMCgHH3ywuW3JkiXi8Xikf//+8sQTT8h1110n06dPl4EDB7b1MSeNUFq+2Xuqd9h9KAAAAEBS26epelpN0nM2bdiwQWbPnm22devWmdbk559/vqxfv16OOeYYueaaa9r+iJNJuhWcvDUldh8JAAAAkNT2KTg98MADcuedd0pOTk79bbm5uXLbbbfJ/fffLxkZGXLLLbeYQIV958lsb/Yp/p12HwoAAACQ1PYpOJWUlEhxcfFut2/ZskVKS0vN5by8PPH7/ft/hEnMl20Fp/SgNaYAAAAA4myq3qWXXipvvvmmmaKnm16+7LLL5JxzzjGPmTVrlvTr16+tjzeppNQFp0yCEwAAABB/zSH+9re/mfVL5513ntTW1lov5PXKRRddVH+iXG0S8fe//71tjzbJZOR2NPvscJlIOKxtC+0+JAAAACAp7VNwysrKkmeeecaEpBUrVpjbevfubW6PGDZsWNsdZZLKzLeCU6oEJOyvEFfqrvEFAAAA4PDgFKFBKXKupujQhLaRm5MngbBHfK6gVJdtk3SCEwAAABA/a5xCoZDccccdppNejx49zKbNILTTnt6HtpGR6pWdYoWlsu27N+MAAAAA4OCK0x//+Ef5xz/+Iffee68ceeSR5jY92a22I6+urpY///nPbX2cScnlckmpK1s6SIlUlWyx+3AAAACApLVPwemFF14wjR/OOuus+tuGDBkiXbt2lSuvvJLg1IYqPDkiQTFT9QAAAADE0VS97du3m655jelteh/aTpUGJxGpLaPiBAAAAMRVcBo6dKg89thju92ut2nlCW2nOsVqvhEqJzgBAAAAcTVV7/7775fTTz9dPvroIxk9erS5bcaMGbJ27Vp577332voYk5o/rUCkXMRVQXMIAAAAIK4qTscee6wsWbJEfvjDH8rOnTvNdu6558r8+fPlxRdfbPujTGKhTOtcTp7KrXYfCgAAAJC09vk8Tl26dNmtCcS3335ruu09/fTTbXFs0MCUXXcS3BqCEwAAABBXFSccOCl5ncw+w09XPQAAAMAuBCeHS8vrYvY5wZ12HwoAAACQtAhODpdTYAWnDKkS8VfYfTgAAABAUtqrNU7aAKI52iRiXzz++OPywAMPyKZNm0yr80cffVQOO+ywJh/7xhtvyN133y3Lli2TQCAgffv2leuuu05+8YtfSCJq166dVId9kuYKSLCsWDzte9l9SAAAAEDS2avglJub2+L9F1544V4dwMSJE+Xaa6+Vp556Sg4//HB5+OGH5eSTT5bFixdLx45WY4TGQeKPf/yjOdluSkqKvPPOO3LJJZeYx+rzEk27zFTZJLnSTbZK6db1kk9wAgAAAJwdnJ577rk2P4AJEybI+PHjTfhRGqDeffddefbZZ+Wmm27a7fHHHXdcg+u//e1v5YUXXpDp06cnZHDyetyyw5VvglP5tg2Sb/cBAQAAAElon9uRtwW/3y+zZ8+Wm2++uf42t9stY8aMMSfUbUk4HJaPP/7YVKfuu+++Jh9TU1NjtojS0lKz12l+utktcgzNHUuZN1+kVqRq+3pHHHM8ac34Yv8wxrHHGMcW4xt7jHHsMcaxxxgn5vjuzfu5wpo+bLJhwwbp2rWrfPHFFzJ69Oj622+88UaZNm2azJw5s8nnlZSUmOdpIPJ4PPLEE0/IpZde2uRjb7vtNrn99tt3u/2ll16SjIwMiQfeb5+T00OfyMc5P5Syg35o9+EAAAAACaGyslIuuOACky9ycnKcW3HaV9nZ2TJ37lwpLy+XKVOmmDVSvXv33m0an9Jqlt4fXXEqKiqSsWPHtjg4ByrlTp48WU466STx+XxNPmby6ski20UKszxy9GmnHfBjjGetGV/sH8Y49hjj2GJ8Y48xjj3GOPYY48Qc38hstNawNTgVFBSYitHmzZsb3K7XO3WyTvzaFJ3O16dPH3N52LBhsnDhQrnnnnuaDE6pqalma0x/IU760Dd7PJkdTXDyVm111DHHE6f9vhMRYxx7jHFsMb6xxxjHHmMce4xxYo3v3ryXredx0q54I0eONFWjiFAoZK5HT91riT4neh1TonFnW90FU6q32n0oAAAAQFKyfaqeTqO76KKLZNSoUebcTdqOvKKior7LnrY31/VMWlFSutfHHnTQQSYsvffee/Liiy/Kk08+KYkqJbez2WcEttt9KAAAAEBSsj04jRs3TrZs2SK33HKLOQGuTr2bNGmSFBYWmvvXrFljpuZFaKi68sorZd26dZKenm7O5/Svf/3LvE6iSsu3pi1m1+6w+1AAAACApGR7cFJXX3212ZoyderUBtfvuususyWT7PZdzD5DqkT8FSIpmXYfEgAAAJBUbF3jhNZp1669VIethWvh8mK7DwcAAABIOgSnONA+K1W2hPPM5YptG+w+HAAAACDpEJziQJrPI9tc+eZy2dZ1dh8OAAAAkHQITnGi1Nve7Kt3UHECAAAADjSCU5yoTC0w+8DOjXYfCgAAAJB0CE5xIpBunQRXyjbZfSgAAABA0iE4xYlQlnVeK0/lZrsPBQAAAEg6BKc44cnpbPZp1VvtPhQAAAAg6RCc4kRavnUS3KwAwQkAAAA40AhOcSKzoJvZ54Z2igQDdh8OAAAAkFQITnEiv0MnCYQ91pXyYrsPBwAAAEgqBKc40SE7XbZKrrlcW0pLcgAAAOBAIjjFifyMFNkSzjOXS7ess/twAAAAgKRCcIoTbrdLdnrbm8uV29bbfTgAAABAUiE4xZGKlAKz9+/YYPehAAAAAEmF4BRH/GkdzD5UtsnuQwEAAACSCsEpjgQzC83eXbHZ7kMBAAAAkgrBKY64czqZfWoV7cgBAACAA4ngFEd8eV3MPsO/1e5DAQAAAJIKwSmOZBQUmX1OcIdIrd/uwwEAAACSBsEpjuQVdJXycJp4JCSyY5XdhwMAAAAkDYJTHOmYmy4rw9Y6p+DWpXYfDgAAAJA0CE5xpFNOmqwWa51T2fpFdh8OAAAAkDQITnHE43bJ9rTu5nLVpsV2Hw4AAACQNAhOcaYmt5fZu7Ytt/tQAAAAgKRBcIoznoI+Zp9ZTnMIAAAA4EAhOMWZrK79zT47sFWkutTuwwEAAACSAsEpznTr1Em2hHOtK9uZrgcAAAAcCASnONOzIFNWhDuby8EttCQHAAAADgSCUxy2JF8jVnAqpSU5AAAAcEAQnOKM2+2Skowe5nLN5iV2Hw4AAACQFAhOcSiQ19vsPTtY4wQAAAAcCASnOOTt2M/ss8tXi4TDdh8OAAAAkPAITnEor0tfCYZdkhaqECkvtvtwAAAAgIRHcIpD3Tvmy7pwB+vKtmV2Hw4AAACQ8AhOcahXQaasrGtJXktLcgAAACDmCE5xqEN2qqx1dzGXy2hJDgAAAMQcwSkOuVwuKcu0WpIHimlJDgAAAMQawSlO1eb3MXvfzhV2HwoAAACQ8AhOcSq1sK4ledVakWCt3YcDAAAAJDSCU5xq36WXVId94g3XipSssftwAAAAgIRGcIpTvTpkycpwJ+vKtuV2Hw4AAACQ0AhOcapn+10tyQPFi+0+HAAAACChEZziVLvMFFnt6W4uV66ea/fhAAAAAAmN4BTHLcmLcw4xl70bvrL7cAAAAICERnCKY1WFI8w+s3yVSMU2uw8HAAAASFgEpzhW1KWLLA11ta6so+oEAAAAxArBKY4NL8qTOaG+1pW1M+0+HAAAACBhEZzi2JCiPPkmbAUn/+ov7T4cAAAAIGERnOJYVqpXtuUPM5c9G74RCdbafUgAAABAQiI4xbmCnoOlJJwhnmCVyObv7T4cAAAAICERnOLc0O7t5JvIOicaRAAAAAAxQXCKc8O759c3iAitoUEEAAAAEAsEpzjXp2OWLPT2N5drVxOcAAAAgFggOMU5j9sloS4jJBR2SUrZGpGyzXYfEgAAAJBwCE4JoH/PrrI43M26sm6W3YcDAAAAJByCUwIYVpS/q0HEWoITAAAA0NYITglgWFGezA71M5draRABAAAAtDmCUwLokJ0qG3MGm8tuPRFuoNruQwIAAAASCsEpQbTvPkg2hNuJO+QXWT7F7sMBAAAAEgrBKYHO5/Ru8AfWle//Y/fhAAAAAAmF4JQghnXPk/8FR5vL4cXvi/gr7D4kAAAAIGEQnBLEoC45stjTR1aHOoorUCmyZJLdhwQAAAAkDIJTgkj1euSE/oXyv5BVdZJ5TNcDAAAA2grBKYGcMaRL/Tqn8PKPRQJVdh8SAAAAkBAITgnk+P4dZJW3l+mu56qtEln5md2HBAAAACQEglMCyUjxyokDCuWT4HDrhqUf2n1IAAAAQEIgOCXgdL2PQ8PM5fDSD0TCYbsPCQAAAIh7BKcEc0y/AvnadYjUhH3i2rlGZMtiuw8JAAAAiHsEpwScrndIry7yZWiAdQNtyQEAAID9RnBKQMf26yAfhkZZV+b8UyQUtPuQAAAAgLjmiOD0+OOPS8+ePSUtLU0OP/xwmTVr1h4f+8wzz8jRRx8t+fn5ZhszZkyzj09Gxx7cQd4MHiU7w5ki25eLLH7f7kMCAAAA4prtwWnixIly7bXXyq233ipz5syRoUOHysknnyzFxcVNPn7q1Kly/vnnyyeffCIzZsyQoqIiGTt2rKxfv/6AH7tT9e2YJXm5efKv4Bjrhi8esfuQAAAAgLhme3CaMGGCjB8/Xi655BIZOHCgPPXUU5KRkSHPPvtsk4//97//LVdeeaUMGzZM+vfvL3//+98lFArJlClTDvixO5XL5TJVpxdqx0qtyyeydqbIonftPiwAAAAgbnntfHO/3y+zZ8+Wm2++uf42t9ttpt9pNak1KisrJRAISLt27Zq8v6amxmwRpaWlZq/P0c1ukWNo62M5qX8HeXlWvrwWPkHOlw8k/OpFEjznbxIecJYkk1iNL3ZhjGOPMY4txjf2GOPYY4xjjzFOzPHdm/dzhcP2nehnw4YN0rVrV/niiy9k9OjR9bffeOONMm3aNJk5c2aLr6HVpw8++EDmz59v1kg1dtttt8ntt9++2+0vvfSSqWwlqlBY5K5vPFJaE5TX8x6XodUzJSRu+azfLbIzs7fdhwcAAADYToswF1xwgZSUlEhOTo5zK077695775VXXnnFrHtqKjQprWbpGqroilNkXVRLg3OgUu7kyZPlpJNOEp/P16avvTF3pdz/wVL5Y9rN8t8efxf34nfkmJ2vSu25H4l4UiQZxHJ8YWGMY48xji3GN/YY49hjjGOPMU7M8Y3MRmsNW4NTQUGBeDwe2bx5c4Pb9XqnTp2afe6DDz5ogtNHH30kQ4YM2ePjUlNTzdaY/kKc9KGPxfGcd1hPeXjKcvl+U6V8PeYWOWztDHEVLxDfzCdEjr1BkonTft+JiDGOPcY4thjf2GOMY48xjj3GOLHGd2/ey9bmECkpKTJy5MgGjR0ijR6ip+41dv/998udd94pkyZNklGj6s5XhN20y0yRH43oai7/6s01suXIuimLnz4gsnOtvQcHAAAAxBHbu+rpNDo9N9MLL7wgCxculCuuuEIqKipMlz114YUXNmgecd9998mf/vQn03VPz/20adMms5WXl9v4UzjX/zt9oBzSNVe2V/jlx593ldruR4oEa0Q+vtPuQwMAAADihu3Bady4cWba3S233GJajM+dO9dUkgoLC839a9askY0bN9Y//sknnzTd+H784x9L586d6zd9DewuM9Urz158qHTNS5fV26vkHxm/tO74bqLIhm/sPjwAAAAgLjiiOcTVV19ttqZo44doq1atOkBHlTg6ZKfKfT8aIj//x0y599tU+cmgc6XdsjdE3r9J5NJJeuInuw8RAAAAcDTbK044MI7qWyDnDOsi2nz+mq1nSdiXKbL2S5FvX7b70AAAAADHIzglkT+ePlCyUr0ybVOKLD74CuvGD/8kUrXT7kMDAAAAHI3glGRT9i49qpe5fM3q0RIuOFikcqvIezeIKUUBAAAAaBLBKclcdlQvyUnzysItNTJ9wJ9EXB6Rea+KzP233YcGAAAAOBbBKcnkpvvk8mN6m8s3f50pgWPrWr2/e73IovfsPTgAAADAoQhOSUin63XKSZN1O6rk6dDZIn3HitRWibxyvshHt4kEa+0+RAAAAMBRCE5JKCPFKzed2t9cfnzqCtl82j9EDv+Vdef0h0RePEekvNjegwQAAAAchOCUpM4e1kVGdM+TSn9QbnxzkYROvlfkx8+KaJvyVZ+JPHW0yOoZdh8mAAAA4AgEpyTlcrnknnOHSKrXLdOWbJGnP1shMvhHIpd/IqLd9so3iTx/ush/fy2yfo7dhwsAAADYiuCUxA7ulC23nTXIXH7gg8XyzncbRDocLDL+Y5HBPxYJB0Xm/FPkmeNF3vg/kYptdh8yAAAAYAuCU5I779Ai+dGIbhIMheXXL38jz3y6Qmq9GSI/+rvIxe+JHPITEZdb5LtXRP46VOTtq0RWTBMJBe0+dAAAAOCAITglOZ2yd/+Ph8jPf9DdnAP3z+8tlLEPfypfLN8m0vNIK0BdNlmk4yARf5nIN/8S+edZIg8NFnnnWpFF79KFDwAAAAmP4ATxuF1y59mD5Y6zB0l+hk9WbKmQS57/Sr5Zs8N6QLdRIr+aLnLJ+yIjLxZJyxUp2yDy9T9EXrlA5KmjRJZOFgmF7P5RAAAAgJggOKG+8nTh6J7y6Y3Hy/EHd5Ca2pCM/+fX8tWq7RLWUpTbLdLjCJEz/ypy/VKR818ROexykfR8kS0LRf79Y5GHBol88EeR9bOpQgEAACCheO0+ADhLdppPHrtghPzkqRmyYGOp2fcrzJI7zh4sP+jd3nqQN1Xk4FOt7fg/iEx7wJrCp1WoGY9Zm0rNFcnrLpLbVcSXLpKaLZJVKJKSJeLxWZezO4uk5VhVLL2stwMAAAAOQ3DCbjJTvfLCpYfJve8vknfnbZAlm8vl/Ge+lPFH95arju8juelR4UYrTqfcLTLmVpFlH4nMe11kySSRQKVITYnI5nnW1hrahCK7i0huN2vLK6q7rPu6yxqyAAAAgAOM4IQmdchOlb/8dKjccuZAufvdhTLx67Xy9Kcr5OVZa+Syo3rJpUf1kpy0qAClVaj+p1ubTtOrLhGp2CKyc41ViaqtsW4r2yRSW21tZZut+/wVIpXbRUIBkdJ11rZ2DwemVazCgSJFh4u0P8gKWjldrMpWataBGh4AAAAkGYITmqXVpft+PETGDiqU+yYtMtWnhz9aKs9OXynnjugmpx3SWQ7tmW/WSNXzeEUy21tbx/6teyNtLFFRLFKyTqRkrcjOtXWXdVtj7at2WFWsNTOsrXG1qnCQSEE/kbQ8kfQ8M/3P7fJJj62LxfXtTpGcziKdh4hkd2rbQQIAAEDCIzihVU4cUCjHH9xR3vt+o/z1o6WytLhcnv9ildlOGlgo9557iLTPSt33N9DmExpodNMufk2pKbcqWBvmiKz7ygpTpRtFSteLVO8U2TTP2qJ4RGSYXlj7/K4bc7uL9DzKmmao76uhy+Wx9lo582WIpGSI+DLr9no9s9G+7n4NiQAAAEh4fOtDq7ndLjljSBc5dXBnmbq4WN6dt1H+9+0Gmbxgs3y9ars5ke6PR3WT/p1itA5Jp+LpND3dhv+84X0aoNbNqqtM7bSmBVaXSKi2WjavXyuFBfni1imAW5dYFaxvX2qbY/KkNAxU3jSrwYXebvbRl1N2XXbv4fbIXkNcY6aq56oLenV7c103j4jbK+KO7L17uN6axzT1HBpwAgCA5EZwwj6d90krULrpeqdrJs41U/j+Pn2l2QZ3zZFRPdqZdVK6dTRbmnRvnyFZqTH6yOk0vIFn73ZzMBCQWe+9J6eddpq4fT6rarX2S5G1X1nrrMKhXVsoaN2mjS38lSKBirq9Xq9oeLs+3ryB39q04pXQXE2Hq7Rc8XQeKv22ibhnrRXJyLcaeKTmRO1zrb03xe4fAgAAYJ8RnLBfBnXJlXd/c7RMW7xFXp+9TqYs2izfry81W2MpHrccd3AHGdw1V1K8binKzzCtznsWZIrPc4AqGlq16jPG2vaVntdKm11EB6pIqNLgFardFaiCgajLe7hdH6+vp80xIrfre0h41/vp5XDj66FdlzX0hYPWXl+vwRbcu+tN/9DW8ekWrXKruLcvlwF6edMbzY+bBqicutb0DQKYr9F1z64gq8ejP2ODKps0vB7ZIs+vr75FvZ5eNrd76n6curGLTM/USqHuzfU0ka4jRDr0r6vyAQAAEJzQBjT0jBlYaLbtFX75cP4mWbO9UraU1UhxZCutlm1634LNZmv4fJf0KsiUvoXZ0q9jtglTGq665ac3bDrhFHpMvjRry2gnCUXDhAktzQWtqNtKN0hwzSxZO3+mdC/MFbe/XKS6VKSmtG66ZKlVoVN10yfjhq6BKxxstcHX4BWZYqnhSqdmavguOmxXGAMAAAmN4IQ21S4zRc47rHuT9y3aVCrvzdtkAlVNICgrtlbI0s1lUuEPmql+ur0rG+sf3z4zRYYW5cmQbrkyoHOODOyc49wwlSjq10tpGGhFs4+OAyTU4xj5tuw96RqZDtmYVto0SJVr+/mNIrX+uupVXQDTKlvjUBZdKdJpgtEVNjNNsq4CZ7ZgVNgLRlXfokNepHoVjPyg1s9q3r/GqviZqZthK9yt+9rq4rjqsz3/7J89aIUoj1asdG2aNhbRylW6tdfg1e4gK1xHVxxNZbHuslbftJV+Xg+R/B7WZT0xNJ9xAAAch+CEA0abRjRuHBEOh2VDSbUs2VQmSzaX1QWoMhOytEL18aJis0Vkp3rlpEGFcs2YflLULsOGnwJ7TTsPanjQraOZ1Od8GqSKF4hsXmCmI1pTKOumUSptQrLkA6s9fm2VtTVlxdS9f2+d0njczSKH/R9NOQAAcBCCE2yl1aOueelmO75/x/rbqwNBWbixVL5du1O+W18iCzeWybLiMimrqZU35qw33fx0el+n3HTplJMqB3fKkZ+M6tbwpLzAvtL1Tl2GW9ueaJCq0FBVY102J3bWtW9V1qYngN621FoHZ7ooRnVZjEz9i7TY37na2pvW+iUik24S+eZfIh0HirTrLa6CAVJQtkBcq7JF8otEsjo27LwY6eYIAABihuAER0rzeWR493yzRfhrQ/Ldup3y1ylL5bOlW+un90U8/NESc06pwpw0yUzxmNfonp8m26qt5zY1iwzYZxpUtJtjW9IANuefIh/+SWTz99ZW9w/1kXphWTPP1SmCqdkimR1E+p8uMuwCkXa92vb4AABIYgQnxA3txDeqZzt58bLDZeXWClm7vVI2lVbLppJq+e+3G2RZcbmpRu3OK3d885FpOvGL0T3l1MGdpGB/TtYLxDKMHXqZyMGniqz63KpAbV0ioc3zpXzHVsnOTBeXrhPTJhyNRaYMVhSLFM8XmT5B5OjrRY6+jlbwAAC0AYIT4pJO09Mt4urj+8gni4tl0aYy03xCp/rptL4lm0plxZZyCYZdpjr1p7e+N1tuuk/Sfbu6oaX53PXnneqQFTn/VJrZ52b4THuCyNTCwpxU6ZSTRpMKxE5OF5EhP2lwPrJP6s5H5vN6rSmB9cLW1MCaMmvTtVnfvCiy8lORafeKfPmkSMf+VgVq+IWsmwIAYB8RnJAQ3FEn5Y0WCATk3Xffk8OOOVEmLdgiL89aI8u2lEtJVcBs0VZtq2z1+2WkeCQv3SeZqV7JSvOaE/tmpnit66kes89O80n7rJT6MKa3ufU8si6XOYmweYzepjcCe9sOP5p254u0xu80WOSQn4jMf1PkvRus5hZrZ1rb3JdEznlSpP1Bthw6AADxjOCEpPieqeHl0qN6ma3KH5TV2yukNhjedR5Xf1C2llvnm9pSXlN/DirdRwesUCgsm8tqzON1awsannLSfaaVe3qKR7xuK1jpppe9brek+tyS5vVYe59HUr1us9Vfrr9N13Y13Ec/J7I/YCcchn0f+sHnivQ/w2pQsewjkWn3W+Hp6eNFTnvAWv+ka6Jyi6wTQwMAgGYRnJB0NJw0bou+N7TRxLodlVJWXSsVNbVmSmBF3VZeE6zb10ppdUC2lfvrQ5hOH9T266GwSDAUFn9Qz0ck5vm6rd+5h5bWMaChrHH40n12mldy01NMYPN53dI5N03yM1LqK2VaHau/7BIT+EYU7ftYIsZ0bVPhIGsb/COR1y+1wtOblzd83EEnWmGKShQAAHtEcAL2oUlF7w77/xf6mtqgCV+lddMGt1f4paY2JLWhsARDIVMRC4U1YIXNCYP1vsi+utG+4eWgVAei9lGPidDgtqtq1nDK4r7I9HrkgYWfSlaaz0xb7JxntYnXoKXTFE/o37FNxgz7IbebyEXviHxyl8jCd6yTAetJfrX9+fIpIk+Mtqb4HfIjkS4jRNLz7D5iAAAcheAE2MRMo8vyHLAOf6G6KldNXahqHLaq/CFTJdtZGZBg2AprG0uqTajTABcOi9lrxcy6HpZ1O6rk+/UlUlHrkoqd1XoGLuvNVu9o8N53vbtQenfIlBP7d5Queen1UwnzMnwyoHMOzTYOZAXqpDusLWLbcpH3rhdZ/rHI3H9Zm0qtqySmZIlkF4ocdILIiAtF8npYUwEBAEgyBCcgSWj1J81tnd9KpO1OarW9rFJefWeyjDj8CKkJipRW6bTDSjNFUUPWks1l8uWKbbJiS4Ws2LKy2emDWq3Kz/BJXkaKmQaY4fOYqZV6zCkeXfflFl2epXszndDjlvQU3bz1j9VNL+tzPHWPiawXM3vzOtbascxUnaa4q7tiUtLpeT9/Q2TNDJFvXxZZNsVqg15Tat2v+7INIhu+EfnsL9YJfDPaW5ueMyq7s0hKhnUS3r4nifQ8hs59AICERHACsF+0e2DnDJHhRXni28NZhrWS9emSLfL5sm1SVh2on0q4ubRalm+pMFMHdavvdrgXHQ73h8/jkqHd8qR7+wyzzivFo2HKKyN75MvhvdubIJcUtILU4whrU1U7RSq27gpOWpXSStSKqSJBv4ieS0q3xmY8JpJRIOLLEEnJtEKZXtbn6GvrOisNXFSsAABxKEm+FQCwU06aT84Y0sVsjWmAKq+uNWu7dL3Xjkrd/OZytU4n9AelKhCU2mBk/Ve4fq9TD7VLYqW/1qzXqqp7rO51GmLjx9eGQmYfqOuoqPuvV+8wW2NaoRrePU9GdM+XonYZpjrl0v9ziQlURxxUYKpbCUnXN0Wvceo6wjqvlJ4vqnKbFaq0zXn5FitA6XmltCo1/y3r9ogtC3ddXvCWyPs3WpddbhG3VyQcEgkFrfVXHQeIpOfvCl266WW3Z9fjlLk/w9pHHls4WCSz/YEaHQBAkiI4AXDEWi9VmNPo/EQxpAFq7fZKE5q2ldeYbokaxLQt/RfLt8nqbZXy1aodZtvTubw0WOm0wYKsFDmoQ5a0z0ypP7dXl9x06VeYlVhrt/R8URpydGvKKfeKbFlkzslrGk9sWyYSqrUqTgveFtk413qcBiG9LaJkrbXtM5dI15Eiw38uMmScFayi6QI93awrVnBLpN8LAOCAIDgBSEq6zqlnQabZmqKh6vNlW2XRpjLTBEMrY+b7t4RNqNLbdOphc/p0zJIf9G4nGSleSY+swYraa8OMQV1yJWFo9UcDTETfMbsuH32t1cEvGLCqR6GAFWA09GxfIbJ1iUhNmUigUsRfLuKvtC5ryDJBp27dlLm/7j7dNKDp89d/bW3vXGOFIhOUdp2rrSGXdQ6rbqNEjrpWpGCApARKRUo3iLj1fAEBEX+FNU1RH5fVScSbar1u5Fh22zys7QKABEdwAoAm6PS88w7r3uR92lFw7tqdpuGFdhjUtVortlaY6YXaYl7P47WsuLx+a87RfQtk3KFFZq1Vt/z0xKpQNZa2h5CY01mk55H7/roaeL5/Q2TW0yI7V0dVl/YkbIUi7SS4/GPTKuVUvfn7fT8EE8ayOorkdN1VlTOXu1oNNNyR9X9RxxZ9nFrN6ziQ8AUADkZwAoC9pOFmePd8s+2JNsH4cP5mWb290qy30nVY2vK9KqB7PVFyUOas2SGfLd1qNtW/U7YJUecf1r2u+yFaJaeLyBFXi/zgSpGK4robtTrkamKv8zQD1uO+fk7km3+JBGvMzWG3T1zaNdDjtdZPaUt2rZKVb26mehURth6n24Y5+/Zz5BZZbd+1BbzHZ3Uw1L2uB9NNL5vKVt1mKnd1UyEjlTxd63XwaSLZnfbtGAAAe0RwAoAYdRv80cg9rAWKmg747Ocr5etVO2TRplIzLfD2/y2Q579YJbefNUiO7dchsStQbU2rNa0NDHpuqjMmiJx6nwQCfnlv0mQ57fTTm+4MGVkjpdMG97RFGmSUrBMpWS9SGtmvt8KUPqaBRr9Xbbqh67zmvCD77Z1rRToPtaZNpmZZgUqDlZ70ODKtMLLOSwNY9HTDyGV33d6Tak1T1IqYtpzXy5F9ZPpkK7hqayWvcoU1DdMX9QcHc4K44K5jAQAHIzgBgI3TAW89c5C5XFIZkLe/XS+Pf7LMrKG6+LmvzBqoHw7rKsf2bS+Bxt+70Ta0iqNj21xANWub9H53y2Gsy/B9Ow7tWLj0Q5HN83dVkDSM6VqwYG3dvi78mEpT0AoakcqUTgXUsFO8SGTdLKsRR6QZh0O+bByrFx64TcSXaVXKdNOfR2kIy+kmkpHf8OeNrqhp+NTKm7a0b9fLqszVBz9PoyBYt49U7Mzr6Gv6626vq+bV7yOX667Xh8KoSmX0ZRMuo56n76H3aaUys8C6vXHg1qqk7tPbiWQV7tu0TBM068ZOwyyAA4rgBAAOkJvhkwtH95QfDu8qD01eKi/PWmPWUP1l8hL5y2R9hFf+35yPpG/HbBnQWbcc6ZqXbppc6OUueXyJimv6JXjg2da2v3TN15ovRTZ+a4UFnXrojoSBugrPnipnDe4LitT6RWqrrRBXW1W3rxYJVDfffKMRbapSs3OTpNWWigQqdn+Kvl/JGmtriba837pY4pqGLW2mYip6dcErMvZmXzcNs/62uuCn+8i461RSXT9nnu8Tj9srR+4oEc/WJ6Kqd5GgF/U7iQ6CkesaZnXKq9m6ivjS9tAEpa7xSn3wDe0KqSa01jVJiUwpjd7XB9vox0e9dvR0VCPSDTNqb26Ovq0VdJwz2u33rwx70OAPBMHW/14iIo134gTBCQAcNsXvljMHyrVj+8l7322UDxdskunLtpqTBus2b32J2aLp/+b8oFd7ObpfgRzas52M6pHPFL9kpl9+B59rbQ5RGwjIB++9J6cdd7j4QjW71m2ZzWOFsR2rRKpLd1WATEWnLvRp0NDPtH5Z16mP+lgNb/Vf1iKBI9zweqRip68Tqczpl25TxaqrZNXvI5f9TXw5b3w5KsxEnq+3azfGii110w8jbe8j3Rjr/pvUTpD6nKqolvz7Qhuc6FZHY3GBXmgilyY9PUecNqdp0G0zav2jCW111Ub9Im8+K3XnkIsKBp5QUI7esU08xY/s+hxEOn+az7J+bj0NPxv1oTfyGYtcjlRW/Vb3ztzu1ntHVyfNVveHhd1ujxxb3b7BfU1UOhv9LLtPH26sieAabiLI7q/rl1qNdeIEwQkAHEhPsvvTQ4vMVl3jlzffeV9GHnGsLN9aJQs3lsqCjaWyrcJvGk3o2qgZK7aZTQ3tlis3ntJfjuxjvkYBzqHT7JpaR6Zauz6t0JreGrf0i3PZJqudvqng6ZTM2l1NPxpUa+qCZXTQ1AqTftnXAKmvU/clvNZfLXPmfC0jhg8Xr8fTTPCL7uhYt9cAppVKXZena/X0S32TVclgw+PSwNCgSta4ahbZRz1/t8pmVAUrevrmnpq8mJsbNXxpjoZyDau67ScNp+1iEU61CU31vDZ+UcQCwQkAHE6n42V4RXoVZEq/znly6iGdd2sy8cH8TaZL39TFW+TbdSXys7/PlHOHd5WbTusvHbMP3ImFAbRAqxp5Rfv/Oul5Ih0Orr8aDgRk4wqXhAectudwGg80SLVlxVzP+6bnetOg2iCARVVeTAOV6ApkTZNVw9pQWGbP+UZGjholXg2wDaYv1k2lNCE4qsIZXTWtvxxdVfVZoW7nWuu5Dc4XF12xjLqt/tx2jR7b+Fxzu90ftW9V8NxTd1JXw33j190bulYxjhCcACABmkz88uje5vLW8hp57ONl8s8Zq+SNb9abrUN2qpwzrItcfUJfyU2P4y9UABJfW08zTskQ6TS4TV5Kw+mmFW4J9zu1bcOpVls7Dmi710PMcKY9AEggBVmpcttZg+T1K46QwV1zzG1bymrkmc9WynEPfCKvz15nTuALAAD2DhUnAEhAI7rnyzu/PloqamplxvJtcv8Hi2TJ5nK5/rVv5aWZq6VTbpqZwjeyR74ce3AHyUmjEgUAQHMITgCQwDJTvTJmYKEcd3AHefqzFfLw5KUyZ83O+vv1ZLv5GT654eT+Mu7QIrOeCgAA7I7gBABJwOtxy5XH9ZFTB3eWL5ZvldpgWFZurZCpi4tl1bZK+cOb8+TJacvk0iN7yU9HFZnABQAAduF/GQEgiWhnPt0iAsEB8uKM1fLIx0tl7fYquf1/C+ShyUtMeNLufcOL8sRNFQoAAIITACQzn8ctlx7VS84/rLu8PmedPDt9palE/X36SrN1y0+Xi0b3NEEqN4N1UACA5EVwAgBIeopHfvGDHvKzw7rLx4uK5X/fbZCPFxbLuh1V8uf3FsqEyUvknOFdZOzATqbxRHaal0oUACCpEJwAAPU0DGkzCd2qA0F5e+56ee7zVbJoU5m8PGut2SKnWumaly7Du+dLp5xUyU7zyQ+HdzXnlAIAIBERnAAATUrzeWTcod3NNL1ZK7fLW3PXy/RlW81aKD0VlFajdIt4cupyufK4g6R7+wwpzEkzlakUL6cLBAAkBoITAKBZLpdLDu/d3myqyh+U0uqALN1cLt+u2ymlVQGZvXqHfL16h/xl8pL652WnemVEj3wZ2CVHThvcWQ7plmvjTwEAwP4hOAEA9no9lG5aVTqqb4G5LRQKy6tfr5VJ8zdJTSAkSzaXybYKv0xbssVsWo0a1SNfxg4qlFE920m3vHQpyEplnRQAIG4QnAAA+00D0HmHdTdbJEjNW19iNp3m9968jaYipVtERopH+hZmy/EHdzAn3+2cm27jTwAAQPMITgCAmASpoUV5Zvv5D3rIH04bIO/O2yjTl26RhRvLpLisWir9Qfl27U6zPTJlqalEndC/owzsnCNd89MlxeOWVJ9bUj0es9frVKgAAHYhOAEAYq5TbppcdlQvs6lAMCSrt1XK3LU75bWv18rMldtNZUq3PfG4XdK9XYY5gW9vPZFvB91nybCiPDN1EACAWCI4AQBsOfFun45ZZvvxyG6ydnulTF6w2QSnJcVlsqWsRvy1IfEHQ6aDnwqGwubkvLp9HPVaWaleOe2QTvKjEd3k0J7tqEoBAGKC4AQAsJ2e/+nSo3qZLVo4HJZAMGwClHbvW7W1QlbUhacVW8rNtL9NpdXy6tfrzJaX4ZPMFK+pTnl187jE43aby3qbz+OS9BSvdMxOlZE98s25p6hVAQBag+AEAHB0K/QUr25uU1nqkpcuR/SxOvlFmlB8tWq7vDFnvVlDtbMyYLbWeH32Onngg8XSq32G7NzhkZc2fSVZqT5TBdOmFQcXZktRu3TJTfeZ4wAAJDeCEwAgbum0vMg5pm47a5CpRNWGQlIbCkttMGwuB+svhyUYCklZda05ce9/5qwz++0Vfo1osrzM6vg3ZVFxg/fQ0FaYkyqdctKkY06aFGanmevajl0rV6aBhddjApeeNBgAkJgITgCAhKANIvRku6316xP6mKYUO8qrZfacOTJs2HAp9Ydk6eYycx6qZcXlsrXcb9Zard1eZbbmaNe/voVZJlAN7por5x9Gi3UASCQEJwBAUvJ63HJknwIJBAISXB02DSZ8Pl+Dx1QHgqZRhbZP31xaI5tKqmVzWbUUl9bI5tJqKS6rMR0Cdf3VjsqAzN9QaraPFxXL458sk/6dsqUoP8NUpdwul+iMP61KtctIkVSv1V79oA6ZMrx7vglcAADnIjgBALAHGnK0cYVuzdEmFmu2V8rSzeWmWcX/vt1gqlmRINUaev6qH/Rub04MrNMDNVhpG3dtwa7HoVvn3DSmAwKATQhOAADsJ20e0aN9ptmUnvR3zbZKWVpcZtZR6foqDVehcFgqaoKyo9JvKlU1tSHTGXDxplJZsNHaWpKd6jWVqvZZKVao8nrE7baOoSAzRY47uKOMPqg9AQsA2hjBCQCAGOjePsNsrbGtvMY0pdB1VbqmSgOVThNcv6NK1u2olEBIA1etVPqDUlZTa55TUhWQFVsqdnutF2asNlWro/oUyJBuudI1P13SfV5TwdJKlnYJ1MpVVppXfG5ruiAAIA6C0+OPPy4PPPCAbNq0SYYOHSqPPvqoHHbYYU0+dv78+XLLLbfI7NmzZfXq1fLQQw/J7373uwN+zAAAtKX2Wany01FFzT5GK1baAbC0utZ0B9Q1VxqsrJMEa8fAsCwtLpcpC4vNdMEPF2w2W0si57zSkxLrea507Zenbj2WBrDsNJ9kp3mlQ3aqHNI1V/oVZpuugtpJUJ+r3QYJXwCSga3BaeLEiXLttdfKU089JYcffrg8/PDDcvLJJ8vixYulY8eOuz2+srJSevfuLT/5yU/kmmuuseWYAQCwg07F04Clm+rTMbvJx911Ttisq5q2ZIupSG0sqTIVLH/dtr3SbxpeRGjg0k0f0xI9X1ZjGq4O6pBl9qk+T31lS/carjJTPNItL1W2l1rv1bD9BgDED1uD04QJE2T8+PFyySWXmOsaoN5991159tln5aabbtrt8YceeqjZVFP3AwCQ7DRgaTt03fZE11fpVMCAnt8qGDJVKz3Xld6ut4Wi1mOVVgfqzn1VKd+tKzFNMLSjYKDuOTp9cN76klYcmVf+df9UObxXe9OmPT3FqmzlpPtME4yj+3Yw0wgBwKlsC05+v99Mubv55pvrb3O73TJmzBiZMWNGm71PTU2N2SJKS62Ft9p+Vje7RY7BCceSiBjf2GOMY48xjq1kHd80j246xU6bSOxbIwkNXSu3VZowFVmbFdki18uqrbVYM1dske0VAXn/+01NvpZOE+zbMUvS66pW2sK9R7sMc2JhbXqh7dr7dsw0VS3sLlk/xwcSY5yY47s372dbcNq6dasEg0EpLCxscLteX7RoUZu9zz333CO33377brd/+OGHkpHRukW7B8LkyZPtPoSExvjGHmMce4xxbDG++y+lbms8iXBIe5Ez80WWl7lkU6XIDr9LdGZgKCxSFRRZX+GSTVUiCzaWNfv6bglL10yRg3PDkpMSFo19uhZL6S7DK9IpPWz2uuwq3Svic0tS4XMce4xxYo2vLgWKm+YQsaYVLV1HFV1xKioqkrFjx0pOTuvPMB/LlKsfkJNOOmm3Ey9i/zG+sccYxx5jHFuM74Eb46t+PGaPY7xya4Ws2mZVrnTqoE4TXL6lXFZurTTt29fuqDInGV5bIbK2ovXNKNJ9VifBjBSvCVMFWSmmiqUNL1I8btPWvUNWqml+kZuu3Qc9puqV5rPWaenUx3jA5zj2GOPEHN/IbDRHB6eCggLxeDyyeXPDjj96vVOnTm32PqmpqWZrTH8hTvrQO+14Eg3jG3uMcewxxrHF+No7xv0655ltT7Rz4MaSavlyxTaZuWK7VPhrJVx3ezis1auwaXqhLd113ZWeO0tVBUJSFdAp+9a0/WVbKuTLlTtafcwanvScWJEwFTkZcV6GT/LSfaYboXYi1KmG3fLTzXm0dG91K3SbsHYgwxef49hjjBNrfPfmvWwLTikpKTJy5EiZMmWKnHPOOea2UChkrl999dV2HRYAAHAgDR9d8tLl3BHdzNaSUCgs5f5aKakMyM7KgAlaetv6nVWyfEuFaY6h29Zy7TJYLcVlNaYJRlUgaLr/RUTWbOl5s1rj7vd2X26gISrS9r1rXroc0i1XerbPlI7ZqXXBa9f9HrfbPKZ3h0xOYgw4jK1T9XQK3UUXXSSjRo0y527SduQVFRX1XfYuvPBC6dq1q1mnFGkosWDBgvrL69evl7lz50pWVpb06dPHzh8FAAA4iJ5bKifNZ7aidnv33EjXweqAta+pDUqVPyTVtVbYqvIHZWdVwISyQGhXR0LtLjhj+bbdWrtHWr77Rcy5tnRr8fhdIj0LMqVfx2zpV5hlncg4xWtOXqy35WZQ8QCSKjiNGzdOtmzZYk5qqyfAHTZsmEyaNKm+YcSaNWtMp72IDRs2yPDhw+uvP/jgg2Y79thjZerUqbb8DAAAILFYJwN2S3ba3j830t49EpZqo/aB2pBZt/X9+lLZsLNKtpbXSMDcb4UvnW6ooWvV1gpzomPtRqjbpPm7v49Wq3oVZJp27noOrTSvWzatc8u8D5ZIms9rjl9bvmvYyvB5zGPSU3TvjbrskQyfV7LSvKbiBcDhzSF0Wt6epuY1DkM9e/Y0c5kBAACcSNc76bYnWkU6cUDDjsKN6XcdnTq4ZHOZLN5UJks3l8uW8hqpqKmVtdsrZUOJNbVQt4bcMm3Tqr0+Zs1MHbPTpEf7DBnQOUeK2mVIp5w0s6ZLT2asTTRSok5snOLx1F+P3K9rvOKlkQYQt8EJAAAAu2gA0fNW6aYnBm5Mz42lTTD0/FnafbDSXyulVX5ZuHipdO/ZS4Jhl6lc6bRCvU+bZej0QrNvdJtWwnRJ16bSarPNXLl9n487OmBpM438TJ+0z0w1nQszU6yqltvlMiErLyPFNNjQkyDrtEoNb+6oy9rdsGf7DOmWn2ECHKEMTkBwAgAAiCPZaT4Z3j3fbNGtnN+rWiynnXLwXnUJ0/bvOyv9pmOhhrHFm8tMA40tpTVmbZc5mbFOP6w7qXH05UjnwvrXqpumWNfA0LxOW9AgpY1BDumaK51z0017eW0dXz9N0bdrCqLpgFg3ddHae0wYA9oCwQkAACBJaXWoY06a2YYW7bkdfFO0S6EGpZomQpVWtfT8W9vK/aZzoVa69PHBsLXWa3ulX0qras26Lt10HVikrbxe1i6IK7dUSIU/aCpi63ZUmW1v6dTCgzpkmfN0aeXq4E45MqpHvglf0VMNzRTEqOvNTbdE8iI4AQAAYK9pJSfNbVV5YkHXepXV1EplTVBWbC2XBRtKzVovDVylVQEprQ7UT0G0piVa0xCr68Kb0lC3YGOpyEbrNT+Yv7l1P5vLCpXW9EOPaRWvswVTgx6ZXP6dZKTq1EO3uV1PptwpN81MT9SmHAVZqea6dkDMTOWrdiLhtwkAAADH0XVNkZbyGkSOOKig1c/VqpVONSwutU6KrO3jdW3Y3LU75fv1JabVvFUpC9ZXyqJnHuplqx29BrDa6KOSVfM2tfo4slK9pgOiVrxMZS871QQrrXDpVEPtjNinQxbt5eMEwQkAAAAJRRtR6LqnngW6Ze5VK/nIdMPI+q6agIaqsFTV+OXdT2ZIp4MGSEhcpoW8PkdPjrw5ak3YlrIa2VRSbapl5XXbiq0Vzb53+0yrWYYes67N0uqVaRtft2Yrcj3NtJCvu00fG9VqPvo5qT5P3ZRD68TKNNdoGwQnAAAAJL1IK/mMlKbv1wYcGwrCctqRPVvVgEMD0+bSalP1Ki6rNoFKW8jrui8NWdsr/OY8XdrNcFuF32yxop0MdRqhhqjIecq026Gu9+rTMcs0HMnRhhtm7zNTDH1ulxmPyHN9HnfSn++L4AQAAAC0MZ2ml9UhyzSnaI6en2vVtgqzdiuyVkubaVT5a+vax0eu17WVb3S52uxrpcofMs/R2xqf9jQQDEsgGGxwmwa2+RtK9+pnapeZYqYb6rTDwuxU0zK/Y06qOQ9Yt/x0KcrPSOgTKhOcAAAAAJtodWdQl9w2baqhUwytqYR1l0MhCdSGJaD7oHVZzwP21artpiqmjTZM041qXQtmTS/UxzUOYNsr/GZbtKms2WPQ6YL6c7XL1PVp6dJZg1ZumuSkadt4r2SmWi3kj+pTYKYZxguCEwAAAJAgdD2TnkC4pYZ+h3TLldOHdG6xyYYGKD1nV3UgKFvLa8x6Lg1bOvVQ97ptKq2R9TsqTet5ZSpkdY9fsrl8j68/8w8nEpwAAAAAxDedcudxe+qnHhZkpUr/Tnt+vK7dqqjRzapa6XquDSVVplmGBiy9Xc/NFdnHW7v2+DpaAAAAAI6UaipdHrMWKhFxWmQAAAAAaAHBCQAAAABaQHACAAAAgBYQnAAAAACgBQQnAAAAAGgBwQkAAAAAWkBwAgAAAIAWEJwAAAAAoAUEJwAAAABoAcEJAAAAAFpAcAIAAACAFhCcAAAAAKAFBCcAAAAAaAHBCQAAAABaQHACAAAAgBYQnAAAAACgBQQnAAAAAGgBwQkAAAAAWuCVJBMOh82+tLRUnCAQCEhlZaU5Hp/PZ/fhJBzGN/YY49hjjGOL8Y09xjj2GOPYY4wTc3wjmSCSEZqTdMGprKzM7IuKiuw+FAAAAAAOyQi5ubnNPsYVbk28SiChUEg2bNgg2dnZ4nK57D4ck3I1xK1du1ZycnLsPpyEw/jGHmMce4xxbDG+sccYxx5jHHuMcWKOr0YhDU1dunQRt7v5VUxJV3HSAenWrZs4jX5A+I8wdhjf2GOMY48xji3GN/YY49hjjGOPMU688W2p0hRBcwgAAAAAaAHBCQAAAABaQHCyWWpqqtx6661mj7bH+MYeYxx7jHFsMb6xxxjHHmMce4xxbMXD+CZdcwgAAAAA2FtUnAAAAACgBQQnAAAAAGgBwQkAAAAAWkBwAgAAAIAWEJxs9Pjjj0vPnj0lLS1NDj/8cJk1a5bdhxS3brvtNnG5XA22/v37199fXV0tV111lbRv316ysrLkRz/6kWzevNnWY3ayTz/9VM4880xzFm0dy7feeqvB/dpT5pZbbpHOnTtLenq6jBkzRpYuXdrgMdu3b5ef/exn5iR2eXl5ctlll0l5efkB/knid4wvvvji3T7Tp5xySoPHMMZ7ds8998ihhx4q2dnZ0rFjRznnnHNk8eLFDR7Tmn8X1qxZI6effrpkZGSY17nhhhuktrb2AP808TvGxx133G6f41/96lcNHsMY79mTTz4pQ4YMqT8h6OjRo+X999+vv5/PcOzHmM9w27r33nvNGP7ud7+Ly88xwckmEydOlGuvvda0XZwzZ44MHTpUTj75ZCkuLrb70OLWoEGDZOPGjfXb9OnT6++75ppr5H//+5+89tprMm3aNNmwYYOce+65th6vk1VUVJjPpIb7ptx///3yyCOPyFNPPSUzZ86UzMxM8/nVf/wi9Av9/PnzZfLkyfLOO++YoHD55ZcfwJ8ivsdYaVCK/ky//PLLDe5njPdM/zvX/yH+8ssvzfgEAgEZO3asGffW/rsQDAbN/1D7/X754osv5IUXXpDnn3/e/NEArRtjNX78+AafY/33I4Ixbl63bt3MF83Zs2fL119/LSeccIKcffbZ5r97xWc49mOs+Ay3ja+++kr+9re/maAaLa4+x9qOHAfeYYcdFr7qqqvqrweDwXCXLl3C99xzj63HFa9uvfXW8NChQ5u8b+fOnWGfzxd+7bXX6m9buHChtuEPz5gx4wAeZXzScXrzzTfrr4dCoXCnTp3CDzzwQIMxTk1NDb/88svm+oIFC8zzvvrqq/rHvP/++2GXyxVev379Af4J4m+M1UUXXRQ+++yz9/gcxnjvFBcXm/GaNm1aq/9deO+998Jutzu8adOm+sc8+eST4ZycnHBNTY0NP0V8jbE69thjw7/97W/3+BzGeO/l5+eH//73v/MZPgBjrPgMt42ysrJw3759w5MnT24wpvH2OabiZANNzPqXDZ3eFOF2u831GTNm2Hps8Uynium0p969e5u/xGtZV+lY619Co8dbp/F1796d8d4HK1eulE2bNjUYz9zcXDPdNDKeutepY6NGjap/jD5eP+daoULrTJ061UxJOPjgg+WKK66Qbdu21d/HGO+dkpISs2/Xrl2r/13Q/SGHHCKFhYX1j9HKamlpaYO/RqPpMY7497//LQUFBTJ48GC5+eabpbKysv4+xrj19K/ur7zyiqno6XQyPsOxH+MIPsP776qrrjJVo+jPq4q3z7H3gL4bjK1bt5r/OKM/AEqvL1q0yLbjimf6pV3LtvoFU8vot99+uxx99NHy/fffmy/5KSkp5ktm4/HW+7B3ImPW1Oc3cp/u9Qt/NK/Xa75QMeato9P0dKpCr169ZPny5fKHP/xBTj31VPM/IB6PhzHeC6FQyMynP/LII80XH9Wafxd039TnPHIfmh9jdcEFF0iPHj3MH7W+++47+f3vf2/WQb3xxhvmfsa4ZfPmzTNf4nUqtK7/ePPNN2XgwIEyd+5cPsMxHmPFZ3j/vfLKK2ZZik7Vayze/i0mOCEh6BfKCJ07q0FK/6F79dVXTfMCIN6cd9559Zf1L236uT7ooINMFerEE0+09dji8S+d+keU6HWPODBjHL3mTj/H2lBGP7/6xwD9PKNl+gdBDUla0Xv99dfloosuMutAEPsx1vDEZ3j/rF27Vn7729+adZDaDC3eMVXPBlru1b8YN+4Yotc7depk23ElEv3LRb9+/WTZsmVmTHV65M6dOxs8hvHeN5Exa+7zq/vGjU60+412gWPM941OQdV/O/QzrRjj1rn66qtN44xPPvnELAKPaM2/C7pv6nMeuQ/Nj3FT9I9aKvpzzBg3T/8a36dPHxk5cqTpZKhNZf7617/yGT4AY9wUPsN7R6fi6f9WjRgxwsyK0E1DqTaY0staOYqnzzHByab/QPU/zilTpjSY5qDXo+fUYt9pS2b9a5D+ZUjH2ufzNRhvLbPrGijGe+/p1DH9hyp6PHWesa6riYyn7vUfQf0HM+Ljjz82n/PI/+hg76xbt86scdLPtGKMm6c9N/QLvU650XHRz2201vy7oHudwhMdUPWvptqyODKNJ5m1NMZN0b/qq+jPMWO8d/S/8ZqaGj7DB2CMm8JneO9odU7HR8ctsunaXF2LHrkcV5/jA9qKAvVeeeUV04Xs+eefN92xLr/88nBeXl6DjiFoveuuuy48derU8MqVK8Off/55eMyYMeGCggLT5Un96le/Cnfv3j388ccfh7/++uvw6NGjzYY9d7/55ptvzKb/TEyYMMFcXr16tbn/3nvvNZ/Xt99+O/zdd9+Z7m+9evUKV1VV1b/GKaecEh4+fHh45syZ4enTp5tuOueff76NP1X8jLHed/3115uOQvqZ/uijj8IjRowwY1hdXV3/Gozxnl1xxRXh3Nxc8+/Cxo0b67fKysr6x7T070JtbW148ODB4bFjx4bnzp0bnjRpUrhDhw7hm2++2aafKr7GeNmyZeE77rjDjK1+jvXfi969e4ePOeaY+tdgjJt30003mS6FOn76b61e186ZH374obmfz3Bsx5jPcGwc26hTYTx9jglONnr00UfNByUlJcW0J//yyy/tPqS4NW7cuHDnzp3NWHbt2tVc13/wIvQL/ZVXXmlajGZkZIR/+MMfmv+BR9M++eQT82W+8aYtsiMtyf/0pz+FCwsLzR8ATjzxxPDixYsbvMa2bdvMl/isrCzTMvSSSy4xgQAtj7F+8dT/gdD/YdA2rT169AiPHz9+tz+sMMZ71tTY6vbcc8/t1b8Lq1atCp966qnh9PR088cY/SNNIBCw4SeKvzFes2aN+YLZrl078+9Enz59wjfccEO4pKSkweswxnt26aWXmv/+9X/b9N8D/bc2EpoUn+HYjjGf4QMTnKri6HPs0v93YGtcAAAAABBfWOMEAAAAAC0gOAEAAABACwhOAAAAANACghMAAAAAtIDgBAAAAAAtIDgBAAAAQAsITgAAAADQAoITAAAAALSA4AQAQDN69uwpDz/8sN2HAQCwGcEJAOAYF198sZxzzjnm8nHHHSe/+93vDth7P//885KXl7fb7V999ZVcfvnlB+w4AADO5LX7AAAAiCW/3y8pKSn7/PwOHTq06fEAAOITFScAgCMrT9OmTZO//vWv4nK5zLZq1Spz3/fffy+nnnqqZGVlSWFhofziF7+QrVu31j9XK1VXX321qVYVFBTIySefbG6fMGGCHHLIIZKZmSlFRUVy5ZVXSnl5ublv6tSpcskll0hJSUn9+912221NTtVbs2aNnH322eb9c3Jy5Kc//als3ry5/n593rBhw+TFF180z83NzZXzzjtPysrKDtj4AQDaHsEJAOA4GphGjx4t48ePl40bN5pNw87OnTvlhBNOkOHDh8vXX38tkyZNMqFFw0u0F154wVSZPv/8c3nqqafMbW63Wx555BGZP3++uf/jjz+WG2+80dx3xBFHmHCkQSjyftdff/1uxxUKhUxo2r59uwl2kydPlhUrVsi4ceMaPG758uXy1ltvyTvvvGM2fey9994b0zEDAMQWU/UAAI6jVRoNPhkZGdKpU6f62x977DETmu6+++7625599lkTqpYsWSL9+vUzt/Xt21fuv//+Bq8ZvV5KK0F33XWX/OpXv5InnnjCvJe+p1aaot+vsSlTpsi8efNk5cqV5j3VP//5Txk0aJBZC3XooYfWByxdM5WdnW2ua1VMn/vnP/+5zcYIAHBgUXECAMSNb7/9Vj755BMzTS6y9e/fv77KEzFy5MjdnvvRRx/JiSeeKF27djWBRsPMtm3bpLKystXvv3DhQhOYIqFJDRw40DSV0Puig1kkNKnOnTtLcXHxPv3MAABnoOIEAIgbuibpzDPPlPvuu2+3+zScROg6pmi6PuqMM86QK664wlR92rVrJ9OnT5fLLrvMNI/QylZb8vl8Da5rJUurUACA+EVwAgA4kk6fCwaDDW4bMWKE/Oc//zEVHa+39f8TNnv2bBNc/vKXv5i1TurVV19t8f0aGzBggKxdu9ZskarTggULzNorrTwBABIXU/UAAI6k4WjmzJmmWqRd8zT4XHXVVaYxw/nnn2/WFOn0vA8++MB0xGsu9PTp00cCgYA8+uijppmDdryLNI2Ifj+taOlaJH2/pqbwjRkzxnTm+9nPfiZz5syRWbNmyYUXXijHHnusjBo1KibjAABwBoITAMCRtKudx+MxlRw9l5K2Ae/SpYvplKchaezYsSbEaNMHXWMUqSQ1ZejQoaYduU7xGzx4sPz73/+We+65p8FjtLOeNovQDnn6fo2bS0Sm3L399tuSn58vxxxzjAlSvXv3lokTJ8ZkDAAAzuEKh8Nhuw8CAAAAAJyMihMAAAAAtIDgBAAAAAAtIDgBAAAAQAsITgAAAADQAoITAAAAALSA4AQAAAAALSA4AQAAAEALCE4AAAAA0AKCEwAAAAC0gOAEAAAAAC0gOAEAAACANO//A9vkteryR9+gAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Ambil skor training dan validation dari evals_result\n",
    "train_logloss = evals_result['learn']['Logloss']\n",
    "val_logloss = evals_result['validation']['Logloss']\n",
    "\n",
    "# Plot learning curve\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(train_logloss, label='Training Logloss')\n",
    "plt.plot(val_logloss, label='Validation Logloss')\n",
    "plt.xlabel('Iteration')\n",
    "plt.ylabel('Logloss')\n",
    "plt.title('Learning Curve')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\ttotal: 53.9ms\tremaining: 53.8s\n",
      "200:\ttotal: 15.8s\tremaining: 1m 2s\n",
      "400:\ttotal: 32.7s\tremaining: 48.9s\n",
      "600:\ttotal: 48.3s\tremaining: 32.1s\n",
      "800:\ttotal: 1m 3s\tremaining: 15.8s\n",
      "999:\ttotal: 1m 18s\tremaining: 0us\n",
      "0:\ttotal: 67.1ms\tremaining: 1m 7s\n",
      "200:\ttotal: 16.2s\tremaining: 1m 4s\n",
      "400:\ttotal: 30.7s\tremaining: 45.8s\n",
      "600:\ttotal: 42.6s\tremaining: 28.2s\n",
      "800:\ttotal: 56.8s\tremaining: 14.1s\n",
      "999:\ttotal: 1m 10s\tremaining: 0us\n",
      "0:\ttotal: 115ms\tremaining: 1m 54s\n",
      "200:\ttotal: 13.6s\tremaining: 54.1s\n",
      "400:\ttotal: 26.6s\tremaining: 39.7s\n",
      "600:\ttotal: 39.4s\tremaining: 26.1s\n",
      "800:\ttotal: 52.8s\tremaining: 13.1s\n",
      "999:\ttotal: 1m 6s\tremaining: 0us\n",
      "0:\ttotal: 55.5ms\tremaining: 55.4s\n",
      "200:\ttotal: 13.7s\tremaining: 54.6s\n",
      "400:\ttotal: 26.3s\tremaining: 39.3s\n",
      "600:\ttotal: 38.2s\tremaining: 25.4s\n",
      "800:\ttotal: 50.6s\tremaining: 12.6s\n",
      "999:\ttotal: 1m 3s\tremaining: 0us\n",
      "0:\ttotal: 58.5ms\tremaining: 58.4s\n",
      "200:\ttotal: 12.5s\tremaining: 49.6s\n",
      "400:\ttotal: 25.3s\tremaining: 37.8s\n",
      "600:\ttotal: 37.7s\tremaining: 25.1s\n",
      "800:\ttotal: 51.5s\tremaining: 12.8s\n",
      "999:\ttotal: 1m 6s\tremaining: 0us\n",
      "0:\ttotal: 86.8ms\tremaining: 1m 26s\n",
      "200:\ttotal: 11.9s\tremaining: 47.4s\n",
      "400:\ttotal: 26.5s\tremaining: 39.6s\n",
      "600:\ttotal: 39.9s\tremaining: 26.5s\n",
      "800:\ttotal: 52.7s\tremaining: 13.1s\n",
      "999:\ttotal: 1m 7s\tremaining: 0us\n",
      "0:\ttotal: 108ms\tremaining: 1m 47s\n",
      "200:\ttotal: 13.3s\tremaining: 53s\n",
      "400:\ttotal: 27.2s\tremaining: 40.7s\n",
      "600:\ttotal: 39.7s\tremaining: 26.4s\n",
      "800:\ttotal: 53.8s\tremaining: 13.4s\n",
      "999:\ttotal: 1m 8s\tremaining: 0us\n",
      "0:\ttotal: 61.8ms\tremaining: 1m 1s\n",
      "200:\ttotal: 13.3s\tremaining: 52.8s\n",
      "400:\ttotal: 27.4s\tremaining: 40.9s\n",
      "600:\ttotal: 41.3s\tremaining: 27.4s\n",
      "800:\ttotal: 54.1s\tremaining: 13.4s\n",
      "999:\ttotal: 1m 8s\tremaining: 0us\n",
      "0:\ttotal: 61.2ms\tremaining: 1m 1s\n",
      "200:\ttotal: 14.1s\tremaining: 56.2s\n",
      "400:\ttotal: 28.4s\tremaining: 42.4s\n",
      "600:\ttotal: 42.1s\tremaining: 28s\n",
      "800:\ttotal: 56.9s\tremaining: 14.1s\n",
      "999:\ttotal: 1m 11s\tremaining: 0us\n",
      "0:\ttotal: 61.5ms\tremaining: 1m 1s\n",
      "200:\ttotal: 14.3s\tremaining: 56.9s\n",
      "400:\ttotal: 26.4s\tremaining: 39.5s\n",
      "600:\ttotal: 38.8s\tremaining: 25.7s\n",
      "800:\ttotal: 52.1s\tremaining: 12.9s\n",
      "999:\ttotal: 1m 5s\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Scores for each fold: [0.95772358 0.95121951 0.9495935  0.96585366 0.95772358 0.94796748\n",
      " 0.95284553 0.9398374  0.96416938 0.95602606]\n",
      "Mean Accuracy: 0.95\n",
      "Standard Deviation: 0.01\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score, StratifiedKFold\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "# Fungsi untuk menghitung skor cross-validation dan visualisasi\n",
    "def cross_validate_and_visualize_accuracy(model, X, y, cv=10):\n",
    "    # Stratified K-Fold untuk mempertahankan distribusi label\n",
    "    skf = StratifiedKFold(n_splits=cv, shuffle=True, random_state=42)\n",
    "\n",
    "    # Hitung skor cross-validation dengan metrik akurasi\n",
    "    scores = cross_val_score(model, X, y, scoring='accuracy', cv=skf)\n",
    "\n",
    "    # Rata-rata dan standar deviasi\n",
    "    mean_score = np.mean(scores)\n",
    "    std_score = np.std(scores)\n",
    "\n",
    "    # Visualisasi hasil cross-validation\n",
    "    plt.figure(figsize=(8, 5))\n",
    "    plt.plot(range(1, cv + 1), scores, marker='o', linestyle='-', color='b', label='Fold Score')\n",
    "    plt.axhline(y=mean_score, color='r', linestyle='--', label=f'Mean Accuracy: {mean_score:.2f}')\n",
    "    plt.fill_between(range(1, cv + 1), mean_score - std_score, mean_score + std_score, color='r', alpha=0.2, label='±1 Std Dev')\n",
    "    plt.title('Cross-Validation Scores (Accuracy)')\n",
    "    plt.xlabel('Fold')\n",
    "    plt.ylabel('Accuracy')\n",
    "    plt.legend()\n",
    "    plt.grid()\n",
    "    plt.show()\n",
    "\n",
    "    # Cetak hasil skor\n",
    "    print(f'Accuracy Scores for each fold: {scores}')\n",
    "    print(f'Mean Accuracy: {mean_score:.2f}')\n",
    "    print(f'Standard Deviation: {std_score:.2f}')\n",
    "\n",
    "# Contoh penggunaan\n",
    "# Ganti model dengan model Anda, misalnya `model`\n",
    "cross_validate_and_visualize_accuracy(model, X, y, cv=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 0.5888696\ttotal: 136ms\tremaining: 1m 44s\n",
      "200:\tlearn: 0.1175942\ttotal: 12s\tremaining: 34s\n",
      "400:\tlearn: 0.0930919\ttotal: 25.9s\tremaining: 23.8s\n",
      "600:\tlearn: 0.0725604\ttotal: 40.8s\tremaining: 11.4s\n",
      "768:\tlearn: 0.0585621\ttotal: 52.3s\tremaining: 0us\n",
      "0:\tlearn: 0.5886265\ttotal: 73.6ms\tremaining: 56.5s\n",
      "200:\tlearn: 0.1198413\ttotal: 13.5s\tremaining: 38.1s\n",
      "400:\tlearn: 0.0881541\ttotal: 27.7s\tremaining: 25.5s\n",
      "600:\tlearn: 0.0662039\ttotal: 41.6s\tremaining: 11.6s\n",
      "768:\tlearn: 0.0527943\ttotal: 53.4s\tremaining: 0us\n",
      "0:\tlearn: 0.5897432\ttotal: 60.9ms\tremaining: 46.7s\n",
      "200:\tlearn: 0.1123212\ttotal: 12.1s\tremaining: 34.3s\n",
      "400:\tlearn: 0.0833892\ttotal: 25.9s\tremaining: 23.8s\n",
      "600:\tlearn: 0.0613133\ttotal: 40.7s\tremaining: 11.4s\n",
      "768:\tlearn: 0.0478552\ttotal: 52.5s\tremaining: 0us\n",
      "0:\tlearn: 0.5899386\ttotal: 56.5ms\tremaining: 43.4s\n",
      "200:\tlearn: 0.1199977\ttotal: 12.3s\tremaining: 34.7s\n",
      "400:\tlearn: 0.0868800\ttotal: 26.2s\tremaining: 24.1s\n",
      "600:\tlearn: 0.0666446\ttotal: 40.4s\tremaining: 11.3s\n",
      "768:\tlearn: 0.0538801\ttotal: 52.7s\tremaining: 0us\n",
      "0:\tlearn: 0.5884433\ttotal: 59.2ms\tremaining: 45.5s\n",
      "200:\tlearn: 0.1198657\ttotal: 11.1s\tremaining: 31.5s\n",
      "400:\tlearn: 0.0896701\ttotal: 25.1s\tremaining: 23s\n",
      "600:\tlearn: 0.0667894\ttotal: 38.6s\tremaining: 10.8s\n",
      "768:\tlearn: 0.0541702\ttotal: 49.8s\tremaining: 0us\n",
      "0:\tlearn: 0.5885144\ttotal: 94.7ms\tremaining: 1m 12s\n",
      "200:\tlearn: 0.1134256\ttotal: 10.7s\tremaining: 30.2s\n",
      "400:\tlearn: 0.0799281\ttotal: 24.6s\tremaining: 22.5s\n",
      "600:\tlearn: 0.0591933\ttotal: 38.8s\tremaining: 10.8s\n",
      "768:\tlearn: 0.0468819\ttotal: 49.5s\tremaining: 0us\n",
      "0:\tlearn: 0.5893625\ttotal: 30.6ms\tremaining: 23.5s\n",
      "200:\tlearn: 0.1135640\ttotal: 9.38s\tremaining: 26.5s\n",
      "400:\tlearn: 0.0816812\ttotal: 25.6s\tremaining: 23.5s\n",
      "600:\tlearn: 0.0613164\ttotal: 38.7s\tremaining: 10.8s\n",
      "768:\tlearn: 0.0484291\ttotal: 49.8s\tremaining: 0us\n",
      "0:\tlearn: 0.5876099\ttotal: 68.4ms\tremaining: 52.5s\n",
      "200:\tlearn: 0.1100839\ttotal: 11.6s\tremaining: 32.8s\n",
      "400:\tlearn: 0.0779947\ttotal: 24.8s\tremaining: 22.8s\n",
      "600:\tlearn: 0.0553376\ttotal: 37.9s\tremaining: 10.6s\n",
      "768:\tlearn: 0.0420982\ttotal: 49.3s\tremaining: 0us\n",
      "0:\tlearn: 0.5793230\ttotal: 53.9ms\tremaining: 41.4s\n",
      "200:\tlearn: 0.1217541\ttotal: 11.3s\tremaining: 32.1s\n",
      "400:\tlearn: 0.0942607\ttotal: 25.7s\tremaining: 23.6s\n",
      "600:\tlearn: 0.0723347\ttotal: 40.6s\tremaining: 11.4s\n",
      "768:\tlearn: 0.0577402\ttotal: 53.2s\tremaining: 0us\n",
      "0:\tlearn: 0.5793558\ttotal: 78.5ms\tremaining: 1m\n",
      "200:\tlearn: 0.1156503\ttotal: 13s\tremaining: 36.8s\n",
      "400:\tlearn: 0.0851988\ttotal: 27.6s\tremaining: 25.3s\n",
      "600:\tlearn: 0.0650702\ttotal: 42.9s\tremaining: 12s\n",
      "768:\tlearn: 0.0549370\ttotal: 54.7s\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Scores for each fold: [0.95772358 0.95447154 0.94796748 0.96097561 0.95447154 0.95284553\n",
      " 0.9495935  0.94308943 0.96579805 0.95439739]\n",
      "Mean Accuracy: 0.95\n",
      "Standard Deviation: 0.01\n"
     ]
    }
   ],
   "source": [
    "cross_validate_and_visualize_accuracy(final_model, X, y, cv=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Metrics:\n",
      "Accuracy: 0.97\n",
      "Precision: 0.94\n",
      "Recall: 0.99\n",
      "F1 Score: 0.97\n",
      "------------------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing Metrics:\n",
      "Accuracy: 0.95\n",
      "Precision: 0.93\n",
      "Recall: 0.98\n",
      "F1 Score: 0.95\n",
      "------------------------------\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, accuracy_score, precision_score, recall_score, f1_score\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Fungsi untuk menampilkan confusion matrix dan metrik evaluasi\n",
    "def evaluate_model(y_true, y_pred, dataset_name):\n",
    "    # Confusion matrix\n",
    "    cm = confusion_matrix(y_true, y_pred)\n",
    "    \n",
    "    # Plot confusion matrix\n",
    "    plt.figure(figsize=(6, 4))\n",
    "    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', xticklabels=['Not Churn', 'Churn'], yticklabels=['Not Churn', 'Churn'])\n",
    "    plt.xlabel('Predicted')\n",
    "    plt.ylabel('Actual')\n",
    "    plt.title(f'Confusion Matrix ({dataset_name})')\n",
    "    plt.show()\n",
    "    \n",
    "    # Hitung metrik evaluasi\n",
    "    accuracy = accuracy_score(y_true, y_pred)\n",
    "    precision = precision_score(y_true, y_pred, zero_division=0)\n",
    "    recall = recall_score(y_true, y_pred, zero_division=0)\n",
    "    f1 = f1_score(y_true, y_pred, zero_division=0)\n",
    "    \n",
    "    print(f'{dataset_name} Metrics:')\n",
    "    print(f'Accuracy: {accuracy:.2f}')\n",
    "    print(f'Precision: {precision:.2f}')\n",
    "    print(f'Recall: {recall:.2f}')\n",
    "    print(f'F1 Score: {f1:.2f}')\n",
    "    print('-' * 30)\n",
    "\n",
    "# Prediksi untuk data training dan testing\n",
    "y_train_pred = final_model.predict(X_train)\n",
    "y_test_pred = final_model.predict(X_test)\n",
    "\n",
    "# Evaluasi untuk data training\n",
    "evaluate_model(y_train, y_train_pred, 'Training')\n",
    "\n",
    "# Evaluasi untuk data testing\n",
    "evaluate_model(y_test, y_test_pred, 'Testing')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Metrics:\n",
      "Accuracy: 0.96\n",
      "Precision: 0.94\n",
      "Recall: 0.99\n",
      "F1 Score: 0.96\n",
      "------------------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing Metrics:\n",
      "Accuracy: 0.95\n",
      "Precision: 0.92\n",
      "Recall: 0.98\n",
      "F1 Score: 0.95\n",
      "------------------------------\n"
     ]
    }
   ],
   "source": [
    "y_train_pred = model.predict(X_train)\n",
    "y_test_pred = model.predict(X_test)\n",
    "\n",
    "# Evaluasi untuk data training\n",
    "evaluate_model(y_train, y_train_pred, 'Training')\n",
    "\n",
    "# Evaluasi untuk data testing\n",
    "evaluate_model(y_test, y_test_pred, 'Testing')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final Training Logloss: 0.10264213391001191\n",
      "Final Validation Logloss: 0.14014741622356627\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA04AAAIjCAYAAAA0vUuxAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAeYZJREFUeJzt3Qd4VFX6x/F3Uia9QQg1dKQ3QRB7odjbrotlBdHFteDaV13/a1dsa0dZC3ZX7LqKKKJYWVAQRbrSe0nvyWT+z3tubpiEhGRCkjuT+X6e5+69c+fOzJlwFvPjnPNel9fr9QoAAAAAoFZhtT8FAAAAAFAEJwAAAACoA8EJAAAAAOpAcAIAAACAOhCcAAAAAKAOBCcAAAAAqAPBCQAAAADqQHACAAAAgDoQnAAAAACgDgQnAEBQ69q1q1x44YVONwMA0MIRnAAA8uKLL4rL5ZIff/zR6aYEnaKiInnkkUdk5MiRkpSUJNHR0XLQQQfJlClTZPXq1U43DwDQSCIa640AAHDCqlWrJCzMmX8H3L17t5xwwgmyaNEiOeWUU+S8886T+Ph406Y33nhDnnnmGSkpKXGkbQCAxkVwAgAEjLKyMikvLxe3213v10RFRYlTdIrgTz/9JG+//bb84Q9/qPLcXXfdJbfccotjPxcAQONiqh4AoN62bNkiF110kbRt29YElv79+8uMGTOqXKMjLLfeeqsMGzbMTF2Li4uTI488Ur788ssq161fv95MD3zooYfk0UcflR49epj3XL58udx+++3mud9++82Ek+TkZPNekyZNkoKCgv2ucbKnHX733Xdy7bXXSps2bUwbzjzzTNm1a1eV12oY0c/q0KGDxMbGyrHHHms+vz7rphYsWCAff/yxXHzxxfuEJqXfRb+b7ZhjjjFbdfo5+nl1/Vw0oEVERMgdd9yxz3voCJe+5sknn6w8l5WVJVdffbWkp6eb1/fs2VPuv/9+850BAP5jxAkAUC87duyQQw891PyCrut3NJB88sknJjjk5OSYX9KVHj/33HNy7rnnyuTJkyU3N1eef/55GTdunCxcuFCGDBlS5X1feOEFs07okksuMb/gt2rVqvK5P/3pT9KtWzeZOnWqLF682LxvWlqaCQB1ufLKKyUlJUVuu+02E0Y0hGi7Z86cWXnNzTffLA888ICceuqppn0///yz2Wt76vLhhx+a/QUXXCBNofrPpX379nL00UfLm2++ab6TL/1O4eHhcvbZZ5vHGi71Wg26f/3rX6Vz587y/fffm++7bds287MAAPiH4AQAqBeddubxeGTp0qXSunVrc+7SSy81AUlHbfQX9JiYGBNWNKj4TivTANWnTx954oknTIjytXnzZjOypEGsuqFDh1a5fs+ePeZxfYKTtvGzzz4zQU/pSMvjjz8u2dnZZvRKg+DDDz8sZ5xxhrz33nuVr9MRHf0+dVmxYoXZDxw4UJpCTT+X8ePHm5/zr7/+KgMGDKgSnDQo6Uig0u/1+++/m1GqXr16mXP6Oh1Ze/DBB+W6664zI1EAgPpjqh4AoE5er1feeecdMzKjx1oUwd50hEbDiI4IKR35sEOThpWMjAyzRmf48OGV1/jSaW41hSY7mPnSKX8annRUqy46UmOHJvu1Gvw2bNhgHs+dO9e06/LLL99npKo+7DYkJCRIU6jp53LWWWeZ6Xq+o2YaonR6oYYq21tvvWW+r4ZY3z+r0aNHm5/B119/3SRtBoCWjBEnAECddG2QrpnRKnG61WTnzp2Vxy+99JL861//kpUrV0ppaWnleZ12V11N52w6xcyXBgGVmZkpiYmJ+23z/l6r7ACla3986VRB+9r9sT9fpyLqGqzGVtPPJTU1VY4//ngzXU+LTygNURqmNFTZ1qxZI7/88kutgdT3zwoAUD8EJwBAneyCAn/+859l4sSJNV4zaNAgs3/11VdNwQOdAnfDDTeYNUk6CqXrlHT6WHU6va82+rqa6KhXXQ7ktfWhUw+VTl3U0Z266OhXTZ+tI0A1qe3ncs4555giGUuWLDHrxTREaZjSUOX75zVmzBj5+9//XuN76H2mAAD+ITgBAOqkIxc6JU1/ydfpXvujpbm7d+8u7777bpWpctULGjitS5cuZq/riHxHd3QqoD0qtT86bVHDoAbF+gQnHcVau3btPuftka/60kCq65Xs6Xp6k10t+uBLK/Hl5eXV+WcFAKg/1jgBAOo1eqNrbnSdk66pqc63zLc90uM7uqKlu+fPny+BREdpdIrb008/XeW8b0nv/Rk1apS5+a1W+nv//ff3eV7Lsl9//fVVwoxOXfT9WWkVPy2b7g+dFqjrynSkSW+yq+vJNEz50mqE+vP+9NNP93m9TrnUtV0AAP8w4gQAqKT3ZJo9e/Y+56+66iq57777zL2YRo4caark9evXzxR+0IIPn3/+uTlWp5xyihlt0vsmnXzyybJu3TqZPn26uV5HQQKFVqDT76VrsU477TQTgjTIaIl1nfbmO1pWm5dfflnGjh1r1hfpCJSGMb1nlK4x0lCjpb/teznp/a+02p2GHi3hruuM9Oei98KqT7ELX1oIQqdNPvXUU+b9qq+x0imSWi5d/yx02qTeUys/P99MK9QRQa166Du1DwBQN4ITAKBS9dEXm/7y3alTJ3MfpjvvvNMEI/2lXUt+6y/+vuXB9drt27fLv//9bzPioYFJp7Nppbd58+ZJINF2641vn332WRP+dBRJS5gfccQREh0dXa8pjHp/JP1Z6NQ5LdmuI006DVDDmAYzW9++fU3Q0psD64159efyyiuvyOuvv+73z0XfW9dAaWEK32p6Nv1OX331ldx7773m566fq8UsdG2TllvXcuwAAP+4vI21ShYAgBZAp7LpeqS7777bBCEAABRrnAAAIauwsHCfc48++qjZH3PMMQ60CAAQqJiqBwAIWTq97sUXX5STTjpJ4uPj5dtvv5X//Oc/Zt3S4Ycf7nTzAAABhOAEAAhZeu8praz3wAMPmAINdsEInaYHAIAv1jgBAAAAQB1Y4wQAAAAAdSA4AQAAAEAwrHGaNm2aPPjgg+a+H4MHD5YnnnhCRowYUeO1WuVI701RnS7s/fjjj+v8rPLyctm6daskJCTU6+aGAAAAAFomXbWk98Tr0KGDhIWFBXZw0opGeiNAvXu63o1ey8DqXdBXrVolaWlp+1yvN13Umwva9uzZY8LW2WefXa/P09CUnp7eqN8BAAAAQPDatGmTudF7QBeH0LB0yCGHyJNPPlk5IqTB5sorr5Sbbrqpztdr0NK7sG/btk3i4uLqvD47O1uSk5PND0fvou600tJSc5d6LX0bGRnpdHMQBOgz8Bd9Bv6iz8Bf9BkEa5/RiqqaPfTm50lJSYE74qQjR4sWLZKbb7658pwOkY0ePVrmz59fr/d4/vnn5Zxzzqk1NBUXF5vNpkNxKiYmxmxO0zK4sbGxpi38RYP6oM/AX/QZ+Is+A3/RZxCsfUYDnKrPEh5Hg9Pu3bvF4/GY+2b40scrV66s8/ULFy6UX3/91YSn2kydOlXuuOOOfc5rwtU/rEAxZ84cp5uAIEOfgb/oM/AXfQb+os8g2PpMQUFBva91fI3TgdDANHDgwFoLSSgdzdI1VNWH43RYMFCm6mmHGTNmDP9Cg3qhz8Bf9Bn4iz4Df9FnEKx9RrNBUASn1NRUCQ8Plx07dlQ5r4/btWu339fm5+fLG2+8IXfeeed+r4uKijJbdfoHFEj/xw609iDw0WfgL/oM/EWfgb/oMwi2PuPPZzsanNxutwwbNkzmzp0rZ5xxRmVxCH08ZcqU/b72rbfeMmuX/vznPzdTawEAAKB0qYW9NkTpsa5ZKSoqMs8BdWnOPqPhSAdrDpTjU/V0Gt3EiRNl+PDhZsqdVsnT0aRJkyaZ5ydMmCAdO3Y0a5WqT9PTsNW6dWuHWg4AABB68vLyZPPmzeb+NzY91tlCWrWY+2SiPpqzz+j7a6nx+Pj44A5O48ePl127dpmS4noD3CFDhsjs2bMrC0Zs3Lhxn5tR6T2evv32W1PgAQAAAM1DRwY0NGmBrTZt2lT+wqszhjRQ6S+mdd1EFGjOPqMBTbOG9ttevXod0MiT48FJ6bS82qbmzZs3b59zvXv3rvKvHAAAAGie6VX6O5iGJt/buugvwXqbmejoaIIT6qU5+4z21/Xr15v+eyDBiZ4NAAAAvzAdD6HYXwlOAAAAAFAHghMAAAAA1IHgBAAAAPipa9euphp0fem6fZ0ylpWVJcHu9ttvNwXdQg3BCQAAAC2WhpX9bRoCGuKHH36QSy65pN7XH3bYYbJt2zZJSkqSptSSAlqgCYiqegAAAEBT0LBimzlzprkFjt7axuZ7bx+tGKgl1/XGrPWp1OYPt9tt7luE4MWIEwAAABpEg0ZBSZnZCks8lcfNsdX31jQaVuxNR3t0NMZ+vHLlSklISJBPPvlEhg0bJlFRUeZeob///rucfvrp5r6iGqwOOeQQ+fzzz/c7VU/f97nnnpMzzzzT3OdK7xn04Ycf1joS9OKLL0pycrJ8+umn0rdvX/M5J5xwQpWgV1ZWJn/729/Mda1bt5Ybb7xRJk6cKGeccUaD/8wyMzNlwoQJkpKSYtp54oknypo1a6pc8+yzz0p6erp5Xr/Pww8/bNqwv9Lid955p7nJrP4M7fuy2rTsuN56qH379qb8eJcuXeS+++4zz+mfo476de7c2by2Q4cO5jsHIkacAAAA0CCFpR7pd+unjnz28jvHSay7cX6Vvemmm+Shhx6S7t27m0CxadMmOemkk+See+4xv8y//PLLcuqpp5qRKv0FvzZ33HGHPPDAA/Lggw/KE088Ieeff75s2LBBWrVqVeP1BQUF5nNfeeUVcy+jP//5z3L99dfLa6+9Zp6///77zfELL7xgwtVjjz0m77//vhx77LEN/q4XXnihCUoa6hITE00Y0++6fPlyiYyMlO+++04uvfRS89mnnXaaCYz//Oc/9/ue2q5//etf8u9//1uGDh0qM2bMMK9dtmyZCZCPP/64+bw333zT/Pz056s/F/XOO+/II488Im+88Yb0799ftm/fLj///LMEIoITAAAAQpqOlowZM6bysQadwYMHVz6+66675L333jO//OvIyf5CybnnnmuO7733XhMYFi5caEaSaqI3ZJ0+fbr06NHDPNb31rbYNHzdfPPNZtRHPfnkkzJr1qwGf087MGk40jVXSoOZji5pIDv77LPNZ+oolAY4ddBBB8n3338vH330Ua3vq+FPA9g555xjHmvo+vLLL82I3LRp02Tjxo0mQB1xxBFm1E1HnPTzc3JyTIjS0b/Ro0eb4KbBasSIERKICE4OWr96qWxf96sUZGQ63RQAAAC/xUSGm5EfnaqVm5MrCYkJZuSkuT67sQwfPrzK47y8PDN97OOPPzZT53TKXGFhoQkA+zNo0KDK47i4ODOis3Pnzlqv16lwdmhSOpXNvj47O1t27NhRJUSEh4ebKYX6826IFStWmPVbI0eOrDynUwB79+5tnlM6qmYHNZu24aNagpOGn61bt8rhhx9e5bw+tkeONFBqMNXP0RB5yimnmKCk/vjHP5oRKx3t0+d09EtH9+qzzqy5scbJQds/f0IOnX+pJO34zummAAAA+E1HD3S6nG4x7vDK4+bY9LMbi4YcXzraoiNMOmr0zTffyJIlS2TgwIFmrc7+6IhJ9Z/P/kJOTdfXd+1WMDn44INl3bp1ZuROA+if/vQnM7qldLRLw9pTTz0lMTExcvnll8tRRx1lRuMCDcHJQd5oqxylu7zQ6aYAAACggk5l01ESHXnRwKRTydavX9+sbdBCFlqcQsue27Ti3+LFixv8nrpOSkfPFixYUHluz549Jrj069fPPNZRId/PVNUf+9JRNS3ooD8zX/rYfk/7uvHjx5vCE1rd8N133zWFKpQGJh1l0qmNWkRj/vz5snTpUgk0gTcGFkJc0YlmH+XJd7opAAAAqKDrcfQXe/1lXkeBtDhCQ6fHHYgrr7xSpk6dKj179pQ+ffqY9UcaNuoz2qbBQysG2vQ1um5LqwVOnjzZFHLQ57UwRseOHc15+zN1xEcr6en3/+KLL0zVQdd+PvOGG26Q2267zUw71Ip6WsxCR+nsIhf6XjoNUQtH6FTOt956q7LKoVYX1FE2nT6oUxdfffVVE6R0HVSgITg5KCzGKusY7WXECQAAIFDoL/oXXXSRKWCQmppqCh/oWp7mpp+rVea0fLiub9Ib7o4bN84c10XDjy99jY42aai56qqrzDojnXqo12nBCXvaoK5N0oIVWiHw//7v/8znXXPNNaYwRW20fLiuybruuuvMGi0dadIiFBpAlQY0rTaoxSm0HVreXddMaYjSMuf63LXXXmtG1HSE77///a9ZexVoXN6WOJFyP7TTa7rVP1wdMnTST5+9IkO/nyK/Si/pfcv8fea5AjXROb/6F5wunqTPoD7oM/AXfQa1KSoqMmtVunXrZu7HY9PRGP0dS3+3aq7iEKFIf8463U7XCOl6oeaiI1QrV640670aS3P2mdr6rb/ZgBEnB0XGWmucYqXA6aYAAAAgwOi9jj777DM5+uijpbi42Iz6aAA477zzmvRztby4VsHTohk6Te+ll14yxRtCHcHJQdEJ1s3Q4rwEJwAAAFSlIzG6Bkir/OkksQEDBpgb0uqoU1PSe0/p9Lnc3FxTJlyLNvzlL3+RUEdwclBsRXCKZ8QJAAAA1Wip7urV6prDm2++2eyfGQyYhOqguKSKESdXsZSW7v++AAAAAACcQ3ByUFxCSuVxXnaGo20BAAAAUDuCk4Mi3FFS4I0yxwV51g3AAAAAAAQegpPD8lxxZl+cQ3ACAAAAAhXByWEFYVZwKsonOAEAAACBiuDksKKweLMvzc9yuikAAAAAakFwclhxhBWcygqynW4KAAAAanHMMcfI1VdfXfm4a9eu8uijj+73NS6XS95///0D/uzGeh+nXXjhhXLGGWdIsCI4Oay0IjiVFxKcAAAAGtupp54qJ5xwQo3PffPNNyaU/PLLL36/7w8//CCXXHKJNKbbb79dhgwZss/5bdu2yYknnihNSW+0m5yc3KSfEewITg7zuBOsgyKCEwAAQGO7+OKLZc6cObJ58+Z9nnvhhRdk+PDhMmjQIL/ft02bNhIbGyvNoV27dhIVZVVihnMITg4rdyeZvas41+mmAAAA+MfrFSnJt7bSgr3HzbHpZ9fDKaecYkKOjqj4ysvLk7feessEqz179si5554rHTt2NGFo4MCB8p///Ge/71t9qt6aNWvkqKOOkujoaOnXr58Ja9XdeOONctBBB5nP6N69u/zzn/+U0tJS85y274477pCff/7ZjILpZre5+lS9pUuXynHHHScxMTHSunVrM/Kl36f6lLiHHnpI2rdvb6654oorKj+rITZu3Cinn366xMfHS2JiovzpT3+SHTt2VLnm7rvvlrS0NElISJC//OUvctNNN9U4gmYrLi6Wq666yrxGf25HHHGEGcmzZWZmyvnnn2/+/PS79urVy4RdVVJSIlOmTDHfT1/bpUsXmTp1qjSliCZ9d9Qt2hpxCivJcbolAAAA/tGwdG8H8y/xzT7J6x9bRdxWdeL9iYiIkAkTJpgQcsstt5gQojQ0eTweE5g0dAwbNswEGw0FH3/8sVxwwQXSo0cPGTFiRJ2fUV5eLmeddZa0bdtWFixYINnZ2VXWQ9k0UGg7OnToYMLP5MmTzbm///3vMn78ePn1119l9uzZ8vnnn5vrk5Ksf2D3lZ+fL+PGjZNRo0aZkLFz504TUjRE+IbDL7/80oQK3f/222/m/TXE6Gf6S7/f6RWh6auvvpKysjITxPQ9582bZ6557bXX5J577pGnnnpKDj/8cHnjjTfkX//6l3Tr1q3W973tttvkv//9r7z00ksm+DzwwAPmu2l7W7VqZYLl8uXL5ZNPPpHU1FRzvrCw0Lz28ccflw8//FDefPNN6dy5s2zatMlsTYng5DBXtPV/iIhSRpwAAACawkUXXSQPPvig+aVfizwoHbn4wx/+YMKJbtdff33l9VdeeaV8+umn5pfy+gQnDTorV640r9FQpO6999591iX93//9X5URK/1MDRganHRERYOJBj2dmleb119/XYqKiuTll1+WuDgrOD755JNmLdf9999vwptKSUkx58PDw6VPnz5y8skny9y5cxsUnPR1S5culXXr1kl6ero5p5/fv39/E94OOeQQeeKJJ8zo3aRJk8zzt956q3z22WdVRsKqB8AZM2aYzf45Pfvss2ak7vnnn5cbbrjBjHINHTrUTKe0f2Y2fU5HoHSUSsOwBq+mRnByWFiMFZyiyghOAAAgyETGmpEfHZHIyc2VxIQECQsLa77PricNDocddpj5JV2Dk45caGGIO++80zyvI08adDQobdmyxUwD02lk9V3DtGLFChMo7NCkdESoupkzZ5qRkt9//90ECh250REuf+hnDR48uDI0KR3h0T+DVatWVQYnDTUammw6+qThpyHs75deEZqUTkfUYhL6nAYn/ezLL7+8yus0dH7xxRc1vqf+DHTqoLbdFhkZaV6j76kuu+wyE24XL14sY8eONdMP9c/Rno44ZswY6d27tyn+oVMy9ZqmxBonh0XGWQPbUZ58p5sCAADgH532ptPldNMgYx83x1Yx5a6+dDTknXfekdzcXDPapNPwjj76aPOcjkY99thjZqqeTm1bsmSJmTKmAaqxzJ8/36zXOemkk+Sjjz6Sn376yUwdbMzP8KUhxJeOymi4CiYnnniibNiwQa655hrZunWrHH/88ZUjgwcffLAZAbvrrrvM9D1dc/XHP/6xSdtDcHKYOy7F7GPKax7GBAAAwIHTX6x1NEynuuk0M52+Z693+u6778wanj//+c9mNEcLN6xevbre7923b1+zvkbLhtv+97//Vbnm+++/N9PJNCzp1DOdZqahwJfb7TajX3V9lhaQ0KluNm2/fjcdfWkK9vfb5LOGSNceZWVlmZEnpZ/tW9hBVX/sS4Orfl9tu01HoPQ19nsqLQwxceJEefXVV00xjmeeeabyOR2t03VWOsVPR/M0GGdkZEhTYaqew6LjreAU52XECQAAoKno+iH9Jfvmm2+WnJwcM9XLpiHm7bffNuFG1wY9/PDDpmKc7y/w+zN69GhTLU9/wdfRK31/DUi+9DN0XY6uadKpbVqA4r333qtyja7h0VEUHfHq1KmTKRxRvQy5jlppUQX9LL3v065du8yaLC1mYU/TaygNbfrZvvTz9fsNHDjQfLaGF51iqNPydMTOXn+kbdD1U/pYp9NpkNH7Y2kIrYlONdTwqqN8WvhBCzxocYiCggIzOmivk9KiHTrtUKdO6kidhjilf0Y6/VDXQGlo1GIfujasKe9FxYiTw2ISrT/ceG+BeINs+BQAACCY6C/kWuJap+H5rkfSog069UvP6xoo/QVc19PUl/7iriFIp4zpGh2tcqcV5nyddtppZsqZVr/T6nYa0rRqnC9dz6PrdY499lgz0lJTSXRdd6VFKHRkRQOYTk/TKWxaCOJA6borDSK+mxad0JG5Dz74wIRKLbmuQUoDkYYjm4YqDaU6lc6eRqfhVEuF10YDoFYj1NCnr9G1Z/rd9HOUjkjpe+p9tvRzdc2WBk+loVKDlgY1/TmsX79eZs2a1aRr7Fxebz2L4LcQ+i8AWjlFy0T6uxivKWRl7JLkx3ua44IbNklsnPNtQmDTYWz9i0HnSFefvwzUhD4Df9FnUBut5qa/EGuJad9fiE1xiJwc87tVsxWHQMAbM2aMCaGvvPLKPs81Z5+prd/6mw2YqucwDUoer0vCXV7Jz84kOAEAACDoFBQUyPTp082onY4M6WiZlmmv6UbAwYp/EnCYKyxM8sQqdZmfs8fp5gAAAAB+c7lcZqRap9TpuiS9sa0Wa9BpfS0FI04BIFdiJUnypSg30+mmAAAAAH6LiYkxI0wtGSNOAaDAZY04FecRnAAAAIBARHAKoOBUWpDldFMAAADqFGK1xRDkvI3UXwlOAaCoIjh58glOAAAgcOmif1VSUuJ0U4B6s/ur3X8bijVOAaAoLFakXKS8KNvppgAAANQqIiLC3EdIb7qqpertMtJaWlp/OdWyz5QjR300V5/Rz9H+qv1W+++BIDgFgJKwGOuA4AQAAAK8clr79u3NPXE2bNhQZSqU3vxVCwToNUBdmrPPaDDr3LnzAX8OwSkAlITHmb2rJMfppgAAAOyX2+2WXr16VZmupzdN/vrrr00pam6ajPpozj6jfbYxRrUITgGgLNwacYooznW6KQAAAHXSX0Kjo6MrH+vakbKyMnOO4IT6CMY+wyTUAOCJqAhOZQQnAAAAIBARnAKAJ8KaqhdVlud0UwAAAADUgOAUALwVI07RHoITAAAAEIgITgHAFWkFp5jyfKebAgAAAKAGBKdAEGndADfeS3ACAAAAAhHBKQCEuSuCk6tQykpLnW4OAAAAgGoITgEgvGLESeXnZjraFgAAAAD7IjgFgLCICCn0us1xfg7BCQAAAAg0BKcAke+yRp0KcvY43RQAAAAA1RCcAkRBmHUvp+K8LKebAgAAAKAaglOAKAyLN/vivAynmwIAAACgGoJTgCiJsIJTWUG2000BAAAAUA3BKUCURCSYvYfgBAAAAAQcglOAKIu0gpMUEZwAAACAQENwChDl7kTroJjgBAAAAAQaglOA8EZbI07hJTlONwUAAABANQSnABEWnWT24SW5TjcFAAAAQKAFp2nTpknXrl0lOjpaRo4cKQsXLtzv9VlZWXLFFVdI+/btJSoqSg466CCZNWuWBLuwGCs4ucsITgAAAECgiXDyw2fOnCnXXnutTJ8+3YSmRx99VMaNGyerVq2StLS0fa4vKSmRMWPGmOfefvtt6dixo2zYsEGSk5Ml2EXEWt8hypPvdFMAAAAABFJwevjhh2Xy5MkyadIk81gD1McffywzZsyQm266aZ/r9XxGRoZ8//33EhkZac7paFVL4I63glOMJ8/ppgAAAAAIlOCko0eLFi2Sm2++ufJcWFiYjB49WubPn1/jaz788EMZNWqUmar3wQcfSJs2beS8886TG2+8UcLDw2t8TXFxsdlsOTlW8YXS0lKzOc1uQ2RFcYhYb35AtAuBy+4f9BPUF30G/qLPwF/0GQRrn/Hn8x0LTrt37xaPxyNt27atcl4fr1y5ssbXrF27Vr744gs5//zzzbqm3377TS6//HLzhW+77bYaXzN16lS544479jn/2WefSWxsrASKX1askd4iEu8tkI8/niUul9MtQqCbM2eO001AkKHPwF/0GfiLPoNg6zMFBQXBMVXPX+Xl5WZ90zPPPGNGmIYNGyZbtmyRBx98sNbgpCNauo7Kd8QpPT1dxo4dK4mJFfdOcpCGPu0wxx4/TmSNSJSrVI4/7miJjolzumkIUHaf0fV+9pRVYH/oM/AXfQb+os8gWPuMPRstoINTamqqCT87duyocl4ft2vXrsbXaCU9/cH6Tsvr27evbN++3Uz9c7vd+7xGK+/pVp2+TyD9HzsxJVU8XpeEu7xSlJctCYnBX/ACTSvQ+jACH30G/qLPwF/0GQRbn/Hnsx0rR64hR0eM5s6dW2VESR/rOqaaHH744WZ6nl5nW716tQlUNYWmYOIKC5dclzXKlJ+92+nmAAAAAAiU+zjpFLpnn31WXnrpJVmxYoVcdtllkp+fX1llb8KECVWKR+jzWlXvqquuMoFJK/Dde++9plhES5DnsgpEFOTscbopAAAAAAJljdP48eNl165dcuutt5rpdkOGDJHZs2dXFozYuHGjqbRn07VJn376qVxzzTUyaNAgcx8nDVFaVa8lKAhPECnbJsW5BCcAAAAgkDheHGLKlClmq8m8efP2OafT+P73v/9JS1QUkShSJlJKcAIAAAACiqNT9VBVaaRV5c9TkOl0UwAAAAD4IDgFkLKoikp6hQQnAAAAIJAQnAJIebQVnFxFWU43BQAAAIAPglMAccVYwSmiONvppgAAAADwQXAKIGGxrczeXUpwAgAAAAIJwSmAuOOt4BRVluN0UwAAAAD4IDgFEHdCa7OP9eQ63RQAAAAAPghOASQ2KdXs470EJwAAACCQEJwCSFxFcEr05km5x+N0cwAAAABUIDgFkIRkKziFu7ySl0tJcgAAACBQEJwCSHRsvBR5I81xXtZup5sDAAAAoALBKcDkuBLMviB7l9NNAQAAAFCB4BRg8sOs4FSUs8fppgAAAACoQHAKMIXhVnAqzs1wuikAAAAAKhCcAkxxZKLZl+Uz4gQAAAAECoJTgCl1J5t9eQEjTgAAAECgIDgFmPKoJLN3FVKOHAAAAAgUBKdAE5NidmHFBCcAAAAgUBCcAkxYrDVVL6I4x+mmAAAAAKhAcAowEXGtzN5dlu10UwAAAABUIDgFmKiEVLOPKWPECQAAAAgUBKcAE53Y2uzjyvOcbgoAAACACgSnABOX3MbsE7y54vV6nW4OAAAAAIJT4ElItqbqxbmKpbCo0OnmAAAAACA4BZ7YhBQp97rMcU7GbqebAwAAAIDgFHhc4RGS64ozx7lZO51uDgAAAACCU2DKc8WbfVE2I04AAABAICA4BaD8COsmuMU5u5xuCgAAAACCU2AqjLSCU1keU/UAAACAQEBwCkAl7hSz9+bvcbopAAAAAAhOgaksupXZuwoynG4KAAAAAIJTYPLGtDb7yCKCEwAAABAICE4ByBVnBSd3SabTTQEAAABAcApMkQltzD6mLMvppgAAAAAgOAUmd6IVnOIITgAAAEBAIDgFoJjktmaf6M1xuikAAAAACE6BKaGVFZzipEi8pYVONwcAAAAIeQSnAJSckiql3nBznJfJTXABAAAApxGcAlC0O0KyJMEc52Zsd7o5AAAAQMgjOAWonLBEsy9gxAkAAABwHMEpQOWHJ5l9cQ7BCQAAAHAawSlAFUammH1p7m6nmwIAAACEPIJTgCqNsoJTeT7BCQAAAHAawSlAeaJbmb2rYI/TTQEAAABCHsEpQHljW5t9eFGG000BAAAAQh7BKUCFx6eavbs40+mmAAAAACGP4BSgIhKs4BRbluV0UwAAAICQR3AKUDFJbc0+zpPtdFMAAACAkEdwClCxyVZwSvLmiHi9TjcHAAAACGkEpwCV2NoKThHiEW8Ro04AAACAkwhOASo5MUHyvNHmOC9zh9PNAQAAAEIawSlARUeGS5YkmOO8DIITAAAA4CSCUwDLDUsy+4KsnU43BQAAAAhpBKcAlh9hBaeSHIITAAAA4CSCUwArikwx+9Lc3U43BQAAAAhpBKcAVhplBSdvwR6nmwIAAACENIJTAPPEtDZ7VwEjTgAAAICTCE6BLNYKThFFmU63BAAAAAhpBKcAFhFvBaeoEoITAAAA4CSCUwCLTEwz+9iyLKebAgAAAIQ0glMAi0myglO8J9vppgAAAAAhjeAUwOJS2pp9guSLeEqdbg4AAAAQsgIiOE2bNk26du0q0dHRMnLkSFm4cGGt17744ovicrmqbPq6ligppY14vC5zXJ5PSXIAAAAgZIPTzJkz5dprr5XbbrtNFi9eLIMHD5Zx48bJzp07a31NYmKibNu2rXLbsGGDtEQpCdGSJfHmODdjh9PNAQAAAEKW48Hp4YcflsmTJ8ukSZOkX79+Mn36dImNjZUZM2bU+hodZWrXrl3l1ratNaWtpYmKCJdsV6I5zsnY7nRzAAAAgJAV4eSHl5SUyKJFi+Tmm2+uPBcWFiajR4+W+fPn1/q6vLw86dKli5SXl8vBBx8s9957r/Tv37/Ga4uLi81my8nJMfvS0lKzOc1uQ21tyQtPEvFskZzdWwOivZCA7zNAdfQZ+Is+A3/RZxCsfcafz3c0OO3evVs8Hs8+I0b6eOXKlTW+pnfv3mY0atCgQZKdnS0PPfSQHHbYYbJs2TLp1KnTPtdPnTpV7rjjjn3Of/bZZ2ZkK1DMmTOnxvOty602/r78J/mtsFUztwqBrLY+A9SGPgN/0WfgL/oMgq3PFBQUBEdwaohRo0aZzaahqW/fvvLvf/9b7rrrrn2u19EsXUPlO+KUnp4uY8eONWulnKYpVzvMmDFjJDIycp/nF66dKZIt0i45RoaedJIjbURgqavPANXRZ+Av+gz8RZ9BsPYZezZawAen1NRUCQ8Plx07qhY+0Me6dqk+9Ac9dOhQ+e2332p8Pioqymw1vS6Q/o9dW3u8MakmOLkK9wRUe+G8QOvDCHz0GfiLPgN/0WcQbH3Gn892tDiE2+2WYcOGydy5cyvP6bolfew7qrQ/OtVv6dKl0r59e2mJwuJbW/vCDKebAgAAAIQsx6fq6TS6iRMnyvDhw2XEiBHy6KOPSn5+vqmypyZMmCAdO3Y0a5XUnXfeKYceeqj07NlTsrKy5MEHHzTlyP/yl79ISxSZkGb2UcUEJwAAACBkg9P48eNl165dcuutt8r27dtlyJAhMnv27MqCERs3bjSV9myZmZmmfLlem5KSYkasvv/+e1PKvCWKTrFG0uJKCU4AAABAyAYnNWXKFLPVZN68eVUeP/LII2YLFXGpHcw+uZzgBAAAAITsDXCxf8mpVon1FMmVkuIip5sDAAAAhCSCU4BLbt1WSr3h5jhz1xanmwMAAACEJIJTgAsLD5dMV5I5ziE4AQAAAI4gOAWBrHCrJHl+xlanmwIAAACEJIJTEChwW8GpJGub000BAAAAQhLBKQgUR6eafXnOdqebAgAAAIQkglMQKI+1boLryt/pdFMAAACAkERwCgKuBOtmwO7CXU43BQAAAAhJBKcg4E5uZ/YxJXucbgoAAAAQkghOQSAmpYPZJ5QRnAAAAAAnEJyCQEJqJ7NPKc8S8Xqdbg4AAAAQcghOQSAlraPZx7qKJT830+nmAAAAACGH4BQE4hKSJM8bY44zd2x2ujkAAABAyCE4BYnMsBSzz92zxemmAAAAACGH4BQkciNamX1h5janmwIAAACEHIJTkCiMam32pVkEJwAAAKC5EZyCRElMmtl783Y63RQAAAAg5BCcgoQ3zgpOEfk7nG4KAAAAEHIITkEiPKGt2UcV73a6KQAAAEDIITgFiaiU9mYfW7LH6aYAAAAAIYfgFCRiW3Uw+yRPhtNNAQAAAEIOwSlIJLXpZPYp3mwpLytzujkAAABASCE4BYmUNh2k3OuScJdXsvdQkhwAAABoTgSnIOF2uyXDlWiOs3dtcbo5AAAAQEghOAWR7LBWZp+3Z7PTTQEAAABCCsEpiORFWsGpKJOpegAAAEBzIjgFkeKoVLP35HATXAAAAKA5EZyCSFlsmnWQR3ACAAAAmhPBKZgkWMEpsnCX0y0BAAAAQgrBKYhEJLU3+5ji3U43BQAAAAgpBKcgEp3cwezjyzKcbgoAAAAQUghOQSQ+taPZJ5cTnAAAAIDmRHAKIilpncw+QQqluDDX6eYAAAAAIYPgFESSklpJkTfSHGft3OJ0cwAAAICQQXAKImHhYZLhSjHHObs2O90cAAAAIGQQnIJMTkQrsy/I2Op0UwAAAICQQXAKMvnuVLMvyd7udFMAAACAkEFwCjIl0VZwKs/Z4XRTAAAAgJBBcAoy5XFpZh9WsNPppgAAAAAhg+AUZMIS2pq9u3CX000BAAAAQgbBKci4k9qbfVzJbqebAgAAAIQMglOQiW1lBadET4bTTQEAAABCBsEpyCSkpZt9q/JM8XrKnG4OAAAAEBIITkGmTfsuUuYNkwhXuWTt2uJ0cwAAAICQQHAKMlFut+x0tTbHGVt/d7o5AAAAQEggOAWhzAirJHnezvVONwUAAAAICQSnIJQXbRWIKNmzwemmAAAAACGB4BSESuM7mL0re7PTTQEAAABCAsEpCLmSrcp67vytTjcFAAAACAkEpyAU1bqz2ccXb3e6KQAAAEBIIDgFoYS0bmbfumyn000BAAAAQgLBKQi17tjD7JMkT0rys51uDgAAANDiNSg4zZ49W7799tvKx9OmTZMhQ4bIeeedJ5mZmY3ZPtSgVavWkuONNcd7tq51ujkAAABAi9eg4HTDDTdITk6OOV66dKlcd911ctJJJ8m6devk2muvbew2opqwMJfsCmtjjrO3r3O6OQAAAECLF9GQF2lA6tevnzl+55135JRTTpF7771XFi9ebAIUml62u61I8QYp3M1NcAEAAICAHHFyu91SUFBgjj///HMZO3asOW7VqlXlSBSaVlGsdS8nTyb3cgIAAAACcsTpiCOOMFPyDj/8cFm4cKHMnDnTnF+9erV06tSpsduIGngSOopkioTlEpwAAACAgBxxevLJJyUiIkLefvttefrpp6Vjx47m/CeffCInnHBCY7cRNQhvZd3LKbZgm9NNAQAAAFq8Bo04de7cWT766KN9zj/yyCON0SbUQ2ybrmafVLLD6aYAAAAALV6DRpy0CIRW07N98MEHcsYZZ8g//vEPKSkpacz2oRbJ7ayb4KaW7xYp9zjdHAAAAKBFa1Bw+utf/2rWM6m1a9fKOeecI7GxsfLWW2/J3//+98ZuI2qQ1rGLlHnDJNLlkdw9W5xuDgAAANCiNSg4aWjSG94qDUtHHXWUvP766/Liiy+a8uRoerHR0bLL1coccxNcAAAAIACDk9frlfLy8spy5Pa9m9LT02X37t1+v9+0adOka9euEh0dLSNHjjSV+urjjTfeEJfLZaYJhqI9EWlmn7eDezkBAAAAARechg8fLnfffbe88sor8tVXX8nJJ59ceWPctm3b+vVeWspcS5vfdtttZu3U4MGDZdy4cbJz5879vm79+vVy/fXXy5FHHimhKi+qvdkX79ngdFMAAACAFq1BwenRRx81IWfKlClyyy23SM+ePc15LU9+2GGH+fVeDz/8sEyePFkmTZok/fr1k+nTp5v1UjNmzKj1NR6PR84//3y54447pHv37hKqSuOt4CTZ3MsJAAAACLhy5IMGDapSVc/24IMPSnh4eL3fRyvwLVq0SG6++ebKc2FhYTJ69GiZP39+ra+78847JS0tTS6++GL55ptv9vsZxcXFZrPl5OSYfWlpqdmcZrehIW0pT+woskMkMndLQHwXBH6fQWiiz8Bf9Bn4iz6DYO0z/nx+g4KTTUPPihUrzLGOFh188MF+vV7XQ+noUfXpffp45cqVNb7m22+/leeff16WLFlSr8+YOnWqGZmq7rPPPjMjW4Fizpw5fr8mN7PM7GPzN8msWbOaoFUIZA3pMwht9Bn4iz4Df9FnEGx9pqCgoGmDk64/Gj9+vFnflJycbM5lZWXJscceawo2tGnTRppCbm6uXHDBBfLss89KampqvV6jo1m6hsp3xEmLWIwdO1YSExPFaZpytcOMGTNGIiMj/Xrtql/binzwL2knu6XLiSeKuFxN1k4EjgPpMwhN9Bn4iz4Df9FnEKx9xp6N1mTB6corr5S8vDxZtmyZ9O3b15xbvny5TJw4Uf72t7/Jf/7zn3q9j4Yfndq3Y8eOKuf1cbt27fa5/vfffzdFIU499dTKc3Z1v4iICFm1apX06NGjymuioqLMVp3+AQXS/7Eb0p4O3ayffbwUSFFhtkQnNU1gRWAKtD6MwEefgb/oM/AXfQbB1mf8+ewGFYeYPXu2PPXUU5WhyZ6qp2XFP/nkk3q/j9vtlmHDhsncuXOrBCF9PGrUqH2u79Onj1lbpdP07O20004zI116rCNJoSQlKVF2eFPM8e5Nq5xuDgAAANBiNWjEScNNTelMz9kjQPWl0+h0pEpLnI8YMcJU7MvPzzdV9tSECROkY8eOZq2S3udpwIABVV5vTxWsfj4U6D2sdkW0l7aeTMneukY6DTjC6SYBAAAALVKDgtNxxx0nV111lZmS16FDB3Nuy5Ytcs0118jxxx/v13vpWqldu3bJrbfeKtu3b5chQ4aYES27YMTGjRtNpT3ULCemo0jecinetdbppgAAAAAtVoOC05NPPmmmyHXt2rVyetymTZvMqM+rr77q9/vp/aB0q8m8efP2+9oXX3xRQllpYheRPBFX5nqnmwIAAAC0WA0KThqW9Aa4n3/+eWXZcF3vpPdfQvMKb91VZKtITP4mp5sCAAAAtFgRB7K+RssH6gbnxLbtKbJUJLl4q9NNAQAAAFqsegenxx9/vN5vqiXJ0TxSOh5k9qmeXeItKxFXhNvpJgEAAAChG5weeeSReo9EEZyaT/uOXaTIGynRrlLJ2rFekiuCFAAAAAAHgtO6desa8WPRWKLdEbI2rJ10926S3ZtWE5wAAACAJkCd7xYg022VhM/f/pvTTQEAAABapAYVh9Cb1tY2TU9vUtuzZ085/fTTpVWrVgfaPtRDflwnkeIFUraHezkBAAAAAROcfvrpJ1OO3OPxSO/evc251atXS3h4uPTp00eeeuopue666+Tbb7+Vfv36NXabUY03qatIhkhkzkanmwIAAAC0SA2aqqejSXrPpq1bt8qiRYvMtnnzZlOa/Nxzz5UtW7bIUUcdJddcc03jtxj7cLfpavbxBZudbgoAAADQIjUoOD344INy1113SWJiYuW5pKQkuf322+WBBx6Q2NhYufXWW02gQtNL7GiN+rUt3Szi9TrdHAAAAKDFaVBwys7Olp07d+5zfteuXZKTk2OOk5OTpaSk5MBbiDq179ZfyrxhEieFUpjBqBMAAAAQMFP1LrroInnvvffMFD3d9Pjiiy+WM844w1yzcOFCOeggSmM3h1aJ8bLJ1d4cb/9tidPNAQAAAFqcBgWnf//733L88cfLOeecI126dDGbHuu56dOnm2u0SMRzzz3X2O1FLXZGWeuc8jYvc7opAAAAQIvToKp68fHx8uyzz8ojjzwia9daJbC7d+9uztuGDBnSeK1EnfKTeors/E7Kd650uikAAABAi9Og4GTToGTfq8k3NKH5udL6iOwUicvmJrgAAABAQEzVKy8vlzvvvNNU0rOn6mkxCK20p8+h+SWkDzD7tKJ1VNYDAAAAAmHE6ZZbbpHnn39e7rvvPjn88MPNOb3ZrZYjLyoqknvuuaex24k6tO8xQDxelyS68qQsZ4dEJLVzukkAAABAaAenl156yRR+OO200yrPDRo0SDp27CiXX345wckBHVqnyEZpK11ku+xY+7N0HEpwAgAAABydqpeRkWGq5lWn5/Q5ND+XyyXb3F3McfbGX51uDgAAANCiNCg4DR48WJ588sl9zus5HXmCM/ISepi9Z8cKp5sCAAAAtCgNmqr3wAMPyMknnyyff/65jBo1ypybP3++bNq0SWbNmtXYbUQ9edscJJIhEpO1xummAAAAAC1Kg0acjj76aFm9erWceeaZkpWVZbazzjpLli1bJq+88krjtxL1Etexv9m3LlzvdFMAAACAFqXB93Hq0KHDPkUgfv75Z1Nt75lnnmmMtsFPbXsMEvlCJMWbJd78PeKKa+10kwAAAIDQHXFCYOrcLk02e1PN8Z71S51uDgAAANBiEJxakMjwMNkSYVXWy9hAcAIAAAAaC8GphcmO7272pduWO90UAAAAIDTXOGkBiP3RIhFwlif1IJFsEXcmlfUAAAAAR4JTUlJSnc9PmDDhQNuEAxDTob/I7yKtCtY53RQAAAAgNIPTCy+80HQtQaNo022QyDcirct3ixRli0TvP+wCAAAAqBtrnFqYbp3ayzZvK3Ocs2mZ080BAAAAWgSCUwsT646QTeHp5njP+l+cbg4AAADQIhCcWqDM2G5mX7yVynoAAABAYyA4tUClrQ4y+4g9q51uCgAAANAiEJxaoKj2/cw+OX+t000BAAAAWgSCUwvUuusgs0/17BApznO6OQAAAEDQIzi1QF07p8tOb7I5Ltq61OnmAAAAAEGP4NQCtYpzy5qwruZ495ofnW4OAAAAEPQITi3Unvg+Zl+06SenmwIAAAAEPYJTC+VpO9Dso3dzE1wAAADgQBGcWqiErgebfVrh7yKeMqebAwAAAAQ1glML1bVXf8n1xohbSqV81yqnmwMAAAAENYJTC9WtTaKslC7meM9vFIgAAAAADgTBqYUKD3PJjtiDzHHu+kVONwcAAAAIagSnFqyodX+zj9jJvZwAAACAA0FwasGiOw81+9Z5q0W8XqebAwAAAAQtglML1qHnECnxhktceZ5I1kanmwMAAAAELYJTC9a7Y2tZ7U03x9nrWOcEAAAANBTBqQWLi4qQje6e5jhrLcEJAAAAaCiCUwuXl9LX7L3bfna6KQAAAEDQIji1cBEdB5t9cvYKp5sCAAAABC2CUwuX2nO4eLwuSS7bLZKzzenmAAAAAEGJ4NTC9enSXlZ5O5vjwvULnG4OAAAAEJQITi1cWkK0rI7sbY4zVn3ndHMAAACAoERwCgG5rYdYB5t/dLopAAAAQFAiOIWA6G4jzD41Z7mIp8zp5gAAAABBh+AUArr0Hio53hiJ8haJd+cyp5sDAAAABB2CUwgY2ClFfvZaN8LNXvM/p5sDAAAABB2CUwiIcYfL5th+5jjv9/lONwcAAAAIOgSnEFHSfpjZR+/4yemmAAAAAEGH4BQiknseavapRetFCrOcbg4AAAAQVAhOIaJfz+6yvrytOfZsXuR0cwAAAICgQnAKET3axMsyVy9zvIcb4QIAAADBF5ymTZsmXbt2lejoaBk5cqQsXLiw1mvfffddGT58uCQnJ0tcXJwMGTJEXnnllWZtbzAKC3PJ7pRB5rh0Q+0/XwAAAAABGJxmzpwp1157rdx2222yePFiGTx4sIwbN0527txZ4/WtWrWSW265RebPny+//PKLTJo0yWyffvpps7c92ISlH2L2yRm/iHi9TjcHAAAACBqOB6eHH35YJk+ebMJPv379ZPr06RIbGyszZsyo8fpjjjlGzjzzTOnbt6/06NFDrrrqKhk0aJB8++23zd72YNPuoOFS7I2UOE+2SMZap5sDAAAABI0IJz+8pKREFi1aJDfffHPlubCwMBk9erQZUaqL1+uVL774QlatWiX3339/jdcUFxebzZaTk2P2paWlZnOa3YbmaEv/jq3kV29XGeZaI3m/fS9RiZ2b/DMR3H0GLQN9Bv6iz8Bf9BkEa5/x5/MdDU67d+8Wj8cjbdta1d5s+njlypW1vi47O1s6duxoAlF4eLg89dRTMmbMmBqvnTp1qtxxxx37nP/ss8/MyFagmDNnTrN8To6rpwyTNfLbt+/Khu2JzfKZCO4+g5aDPgN/0WfgL/oMgq3PFBQUBEdwaqiEhARZsmSJ5OXlydy5c80aqe7du5tpfNXpaJY+7zvilJ6eLmPHjpXEROeDg6Zc7TAa/CIjI5v8857btl5k+yfSuWyd9D/ppCb/PAR/n0Hwo8/AX/QZ+Is+g2DtM/ZstIAPTqmpqWbEaMeOHVXO6+N27drV+jqdztezZ09zrFX1VqxYYUaWagpOUVFRZqtO/4AC6f/YzdWeuF5HimwXaVW4TqQ4SyS+TZN/JppGoPVhBD76DPxFn4G/6DMItj7jz2c7WhzC7XbLsGHDzKiRrby83DweNWpUvd9HX+O7jgm169ujq6wsTzfH3g3czwkAAAAIiql6Oo1u4sSJ5t5MI0aMkEcffVTy8/NNlT01YcIEs55JR5SU7vVarainYWnWrFnmPk5PP/20w98kOAzokCRveftKH9kkeWu+loT+ZzjdJAAAACDgOR6cxo8fL7t27ZJbb71Vtm/fbqbezZ49u7JgxMaNG83UPJuGqssvv1w2b94sMTEx0qdPH3n11VfN+6BuMe5w2Z58sEjuZ+JZ+43TzQEAAACCguPBSU2ZMsVsNZk3b16Vx3fffbfZ0HBRvY4SWXyfJOesFsnbKRKf5nSTAAAAgIDm+A1w0fwG9+4ly8q7WA/WVg2mAAAAAPZFcApBh3RtJd96B5nj/OWfOd0cAAAAIOARnEJQXFSEbG1tVS0MW/eliNfrdJMAAACAgEZwClFJvY+QQq9bYop3i+xc7nRzAAAAgIBGcApRR/RJlwXlfc1x+W9fON0cAAAAIKARnELUwZ2TZUH4EHOcxzonAAAAYL8ITiEqIjxMijsfbY5jty0QKcl3ukkAAABAwCI4hbA+Aw+RTeVtJKK8WOT3L51uDgAAABCwCE4h7JjeaTKnfJg5Lvr1v043BwAAAAhYBKcQlpYYLauTjzLHrjWfipR7nG4SAAAAEJAITiEudcDRkuWNk6iSTJFNC5xuDgAAABCQCE4h7ug+HeSL8qHmuHzFx043BwAAAAhIBKcQNzQ9Wb4LH2GOS5Z9JOL1Ot0kAAAAIOAQnEKcliX39jxeir0REp27XmTXKqebBAAAAAQcghNkVJ8u8n15f+vBKqbrAQAAANURnCBH924jc8qHm+PS5QQnAAAAoDqCEyQtIVo2p1llySO3LRLJ3e50kwAAAICAQnCCMahPX1lS3t16sGqW080BAAAAAgrBCcaxfdrIpx6rul750necbg4AAAAQUAhOMIakp8gXkdZ0vbAN34pkbXK6SQAAAEDAIDjBCA9zSa+D+sr/yvtaJ5a+5XSTAAAAgIBBcEKlkwa2l3c9R5hj7y8zuRkuAAAAUIHghErH9UmTbyMOk2JvpLh2rRTZvtTpJgEAAAABgeCEStGR4XLEwB7yeflQ64SOOgEAAAAgOKGqM4Z0lPc8R5pjr65zKvc43SQAAADAcQQnVDGye2tZGTdCMr3x4srbIbLuK6ebBAAAADiO4IR9quudNLSLfOQ51Drxy5tONwkAAABwHMEJtUzXq6iut/xDkaIcp5sEAAAAOIrghH30bZ8geW2Gym/lHcRVmk+RCAAAAIQ8ghP24XK55IyDO8nLnjHWiYXPcE8nAAAAhDSCE2p02uAO8q7nSMnzRovsXk2RCAAAAIQ0ghNq1CklVvp16yTvVJQml4XPOt0kAAAAwDEEJ9TqzKEd5WXPWHPsXTVLJGuj000CAAAAHEFwQq1OHtRetkSkyzeeAeLylov88JzTTQIAAAAcQXBCrRKjI+XkgR3kJc8468Til0VKCpxuFgAAANDsCE7Yr/GHpMsX5UNlkzdNpDCT0uQAAAAISQQn7NchXVOka2qCzCirGHX69hERT6nTzQIAAACaFcEJdd7TSUed/uM5TrJcySJZGxh1AgAAQMghOKFOfxjWScojYmRayUnWia8fEvGUOd0sAAAAoNkQnFCn1Pgo+cPBHeVVz2jJCUsWyVzHqBMAAABCCsEJ9XLxEd2lUKJlWsmJ1omvH2CtEwAAAEIGwQn10jMtXkb3TZOXy8ZIbniKSOZ6kSWvO90sAAAAoFkQnFBvlx/b04w6PVZ8inXiqwdESoucbhYAAADQ5AhOqLeDO6fI4T1byytlx0t2ZBuRnM0i/5vmdLMAAACAJkdwgl+mHNtLisUtdxaNt058/S+RnG1ONwsAAABoUgQn+OXQ7q1keJcUead0lGyKGyBSmi/y+e1ONwsAAABoUgQn+H1D3CuO66lHcm3OudbJX94QWf+d000DAAAAmgzBCX475qA2MqBjovxQ2k1+TjvdOvnB5SIl+U43DQAAAGgSBCc0aNRJ1zqpv2w/QzwJHa3y5HPvdLppAAAAQJMgOKFBxvVvK4M7Jcmukih5tc111skF05myBwAAgBaJ4IQGjzrddGJfc3zXyvaS3bdivdMHVzBlDwAAAC0OwQkNNqpHaxndN03Kyr1ydeYfxZuoU/bWMWUPAAAALQ7BCQfktlP7S3RkmHy5vli+63fr3il7v3/hdNMAAACARkNwwgFJbxUrVx5nFYq4+sdUKRl8gfXEmxNFMtY52zgAAACgkRCccMAmH9ldureJk915xfJA2EUi6SNFinNE3rtUpNzjdPMAAACAA0ZwwgFzR4TJHaf1N8cvLNguG455TMSdILLpfyLfPep08wAAAIADRnBCoziyVxtTKMJT7pXbv84V74n3W098ea/Itp+dbh4AAABwQAhOaDQ3ndhH3OFh8uWqXTKz5AiRvqeJlJeJvDNZpLTQ6eYBAAAADUZwQqPpmZYg1487yBzf+fEK2XT4PSLxbUV2rxKZeYFIUY7TTQQAAAAahOCERnXxEd1lRLdWUlDikWv+u1k8p08XCYsU+W2OyDt/oVgEAAAAghLBCY0qPMwl/zp7sMRHRciPGzLlua1dRCZ9IhIRLbLmU5HPb3e6iQAAAIDfCE5okns73XpKP3P8yOerZVNcf5EznrKe/P5xkR9fcLaBAAAAgJ8ITmgSZw/vJId2byVFpeXyf+//Kt7+Z4kcfaP15EfXiCz5j9NNBAAAAIIrOE2bNk26du0q0dHRMnLkSFm4cGGt1z777LNy5JFHSkpKitlGjx693+vhDJfLJfecOdBU2ftq9S555uu1IsfcLDLiEhHxinxwuciv7zjdTAAAACA4gtPMmTPl2muvldtuu00WL14sgwcPlnHjxsnOnTtrvH7evHly7rnnypdffinz58+X9PR0GTt2rGzZsqXZ247969EmvrLK3tRPVsrrCzeJnHC/yMETRbzlVpny5R863UwAAACgThHisIcfflgmT54skyZNMo+nT58uH3/8scyYMUNuuummfa5/7bXXqjx+7rnn5J133pG5c+fKhAkT9rm+uLjYbLacHKskdmlpqdmcZrchENrSFC48NF0y84vl6a/WyZ0fLZMhnRKk1wkPSnhpoYQtfVO8b10onpMfFe/gc51uatBo6X0GjY8+A3/RZ+Av+gyCtc/48/kur9frFYeUlJRIbGysvP3223LGGWdUnp84caJkZWXJBx98UOd75ObmSlpamrz11ltyyimn7PP87bffLnfcccc+519//XXz2Wh65V6Rf68Ik5XZYdI+1ivXDfSI2+WRoRuek/TM78w1y9v/Uda0PVXn+DndXAAAAISIgoICOe+88yQ7O1sSExMDd8Rp9+7d4vF4pG3btlXO6+OVK1fW6z1uvPFG6dChg1nrVJObb77ZTAX0HXGyp/fV9cNprpQ7Z84cGTNmjERGRkpLNfKoYjl12nzZll8i3xWny/1nDRCXnCyeL++S8PlPSL9tb0vv9glSPm6qSJjjA6EBLVT6DBoPfQb+os/AX/QZBGufsWej1UdQ/4Z63333yRtvvGHWPWlhiZpERUWZrTr9Awqk/2MHWnsaW4dWkfLoOUPkwhd+kPeWbJOuqQly1eheIuPuFknqJDL7Jglf/IKE52wS+eMMkegkp5sc8Fp6n0Hjo8/AX/QZ+Is+g2DrM/58tqPFIVJTUyU8PFx27NhR5bw+bteu3X5f+9BDD5ng9Nlnn8mgQYOauKVoDEf2aiN3nT6g8v5O7y7ebD1x6KUi418RiYgR+e1zkWePF9m9xtnGAgAAAIESnNxutwwbNswUdrCVl5ebx6NGjar1dQ888IDcddddMnv2bBk+fHgztRaN4byRneWvR3c3xze+84vM/32P9UTfU0Uu+kQksZPInjUizx5n3evJuSV4AAAAQOCUI9f1R3pvppdeeklWrFghl112meTn51dW2dNKebpOyXb//ffLP//5T1N1T+/9tH37drPl5eU5+C3gjxvH9ZGTB7aXUo9XLnnlR/l5U5b1RIehIpfME+lyuEhxjsj7l4r85xyRnG1ONxkAAAAhzvHgNH78eDPt7tZbb5UhQ4bIkiVLzEiSXTBi48aNsm3b3l+cn376aVON749//KO0b9++ctP3QHAIC3PJv/40WEZ0bSW5RWXy5+cX7A1P8W1EJnwocvytIuFukdWzRZ4aKfLru043GwAAACEsIIpDTJkyxWw10cIPvtavX99MrUJTio4MlxcmHSIXvrBQflifacLTqxePlMHpySLhESJHXifS+ySR9y8T2fqTyNsXieRsERl5mfU8AAAAEEojTghdcVER8uKkEXJI1xQz8nTB8wtk+VafkpBpfUUu/lxk2IUi4hX57P9Enj1WZMtiJ5sNAACAEERwQkCEp2FdUiSnpvCko0snPyJy6mNWifLtv4g8d7zIx9eJ5Gx1sukAAAAIIQQnBER4mnHhITKgY6LsyS+R8c/Ml4XrMvZeEBZmjTpN+VFk4J9EvOUiPzwn8thgkf9eJZKxzsnmAwAAIAQQnBAQkmIi5bW/HFpZMEJHnj5dtr3qRfFpIn94VmTif63Ke54SkUUvijwxTOTdv4rsWuVU8wEAANDCEZwQUOHp5YtHyOi+aVJcVi6XvrpIpn/1u3jKq93LqdtRIpNmiUz6RKTH8SJej8gvb4hMGyky8wKRbb849RUAAADQQhGcEHDV9qb/eZi5Ua7e+/a+T1bKec/+T7ZnF+17cZfDRC54V2TylyJ9TrEKSKz4UOSZo0W+nCpSzL29AAAA0DgITgg4EeFhcs8ZA2TqWQMlzh0uC9ZlyFlPfSebMgpqfkHHg0XOeU3ksvkifU+11kB9dZ/IY4NEPrlJJGNtc38FAAAAtDAEJwQkl8sl547oLB//7UjpnhonW7OL5OTHv9l33ZOvtv1E/vSKyB9niKR0FSnYI7LgaZFph1proNZ9I2YYCwAAAPATwQkBrWtqnLw++VBzY1wtV/7XVxbJbR/8KkWlnppf4HKJDPiDVYHvvDet9VCeYmsN1EuniDw+VOTrByllDgAAAL8QnBDw2iVFy1t/HSWXHNXdPH5p/gY5Y9p38uN6n5Ll1YVHihw0TmTCh9ZNdA+eKOJOEMlcJ/LF3SKP9Bd59Y8iqz4RKStpvi8DAACAoBThdAOA+nBHhMk/Tuoro3q0luve/FlWbs+VP06fL2cO7Sg3ndhH2iZG1z4ClX6ItZ0wVWT5ByI/vSqy4TuR3+ZYmztepOuRIj2Os7bWPazXAQAAABUITggqx/ZOkznXHCUPfrpKZv64Sd77aYt8tmy7/O34XjLp8G4mYNXKHScy5Dxr2/O7yI8zRH5+Q6Rgt8jqT6xNJaWL9DjWClHdjhaJbdVs3w8AAACBial6CDqt46Pkvj8Mkg+vOEIO7pws+SUemfrJSjnhsa/lq9W76vkmPUTG3SNy/RqRv34tMvp2az1UuFske5PI4pdF3rpQ5IHuIs8eJzL3LpEti5v6qwEAACBAMeKEoDWwU5K8felh8s7izXL/7JWydle+TJyxUE7o304ePHuQJERH1v0mYWEi7Qdb2xHXiJTki2z4XuT3L0R+/1Jk1wqRLYus7ZuHRFr3FOl6hDW1T8ugp3RjWh8AAEAIIDghqIWFueTs4ekybkA7eezzNfLS9+tl9rLtsnpHrtxxen85omeqKW1ebzqdr9cYa1NafU8D1JrPRFZ+LLLnN2tb9KL1fHSSSL/TRTqPEkk9SKRNb5GohKb5sgAAAHAMwQktQmJ0pPzzlH5y+pAOcvFLP8ra3flywfMLZUDHRLnyuF4ytl9b/wJU5Rt3EBl6vrUVZolsnC+y/luruMSOZSJF2da0Pt1sukaq3UCR7seK9Bot0sqqBggAAIDgRXBCizKoU7J8fu3R8ujnq+X1BRvl1y055t5Ph/dsLVcc09NU5WtQgFIxySK9T7Q2pWXMN34vsuIjkd2rRHatFsnbbq2R0m3VLBGtN5HYUSR9pEhKF5HkLlao0i0iqlG/OwAAAJoOwQktTlJMpNx2an/523G95Llv18qzX6+T737bY7b+HRLlosO7ySmD20tURPiBfVCEW6T7MdZmK8gQ2bXKGpVa95XI+m9EcraILHu36mvDIq3w1Gm4SEfdhlEGHQAAIIARnNBipcS55YZxfeScQzrLc9+slTd/3CzLtubIdW/9LI98vlpuGNdbTh3UwayTajRaurzLKGs7+gaR4jxrep+GqawNIhlrRbb+JFKwR2TrYmuTZ6zX6shU50OtghOtuomkdLWOE9pbRSwAAADgGIITWrz0VrFyx+kD5OrRB8nrCzeaAhKbMwvlqjeWyFNf/i6Tj+oupw3usP97QDVUVHzVYhPK6xXJXL+3Wt/mH0W2/WyNTP36zr7vERFtTfGLTzPrpVz9zpIIT2HjtxUAAAC1IjghpEagrji2p0w6vKs8/806+ffXa2XVjly5/q2f5Z6Pl8tJA9vL6UM6yvAuKY07ClWdTsfTESXdBv7ROldaZE3t05GpzHUiGeusfdYmkbIiaw2Vbuu/kYjFL8nJmr9WXCuS0E4kro017a/PSSLph4q4Y5uu7QAAACGK4ISQE+uOkCuP7yUTDusqry3YIC9+t1525hbLaws2mq1DUrScPKi9nDm0k/TrkNg8jYqMFjlonLX58pRahSZ0hCp/t8i6r8W78iNxFWaKqzRfJON3a9v0P5EfnrVeE9NKJKmjSJs+Imn9rMqAOt3P7NtRLh0AAKABCE4I6SISlx/TUy45srv8b22GvL9ki8z+dbtszS6SZ79ZZzatxnf64I4ytn9bSY51N38jwyOtcuZ2SfNBf5Kykx6RT//7row7fLBEFu4Wyd1uApWsni2St0OkMMPati+t+T3dCSJt+1VM/eshEpdqBasuh1nhCgAAAPsgOCHkRYSHyRG9Us129xkD5MuVO+W/v2w1IcquxveP91xyTO828teje5ipfA0uad5IPOHRVuiJ7GOd0Cl/unZK7yuVvdkapdr+q3Wz3txtFdt2keIckZJckU0Lan5jHa3SmwDruiqd8qc39dVz0Yki7YdUPE4WcceLRMZQBRAAAIQMghPgIzoyXE4c2N5smzIK5IMlW+TjpdtlxbYc+XzFTrN1aR1r1kOdPLC9KW/udIiqpO3QUKNbuwF77zflS6v8abDS0Sgdldq10jq3e7XI9l/2jlbZtGhFbbSkuo6Ete4pEp1kBS0NXLrmSu9blZwuEpdmlW0HAAAIcgQnYD/V+KYc18tsv+3Mk+e/XSvvLt4iG/YUyNPzfjdb99Q4E6IO7d5ahnROlvioAP+/lFb5S9O1TxUjVb6KcqyRKi1UUVZojV5psYqSfJH8nVb1v+wtIsXZ1vXlpXuLVuyPBimtCqgjWXqvqnC3SFiENQ0xPMo6r5ue13LuSelW+HOFW2EsWo8DJJwCAICQFeC/5QGBoWdavEw9a5D838n95IuVO+XjX7bJl6t2ytrd+fLkl7+ZLTzMZabxHdEzVYZ1TZGR3Vqbc0FDp+NF9696ro/W76umvFyktMC6F5VOBdR7U2m4Ki20Apc+3rLYWm9VXiaSv8valFYO9JeGrMhYK1hp2NL7XaV0qXjss0VU7HUKYXJna0QsLFzEFWaFML0Xln2sQc0OZPZ76rUAAAC1IDgBfoiLipBTB3cwW15xmcxdscMEqR/XZ8qWrEJZsC7DbEqr8/1xWCc5pk+aDE1PDpwpfQdKA4iOXOmmAUaOr/k6DVhFWVaQytspUrDbGrHSkSqtFqihqqxYpCTPCl6eEqtyoI56FeeKlHusa/U6XZtl0/VaW35s3O8UEWMVydDwpMUz4tuIxLa2ph5qqAqPqAhfvkEsXCQiSiS+nRU6NdyZQJYkEqvvxU2LAQBoSQhOQAPptDy975NuStdEmRC1IVO+Xr3LVOd7/IvfzJYa75aBHZNkYKdkGdwpSYakJ0vr+Chp0TQ46NQ73RpKpw3qyJbey0o3DVy6Lqsw0wpaZSXW3nczBTK2iHjLRbwea68hzBx7rWMNa3qdPvYUWyNlGthsOw7wu+sUxPi21miWjpjpFlF9hCxq73RF3xGzGp+PrHjcwNd7W0hoBwDAQQQnoBHXRE08rKvZiko98umy7aYy37xVu2R3Xol8uWqX2Wy92ybIqB6t5aiDUuWwHqmmMAVquL+V3pPKV4chjfsZGqT0Plk6OubR0a3cirLumVbA0hEuE7y8VviyQ5judXqiXquv0emLJQXW6JiGseyNEigiReRUCRfXr9H7hjA91qmNOsXRhDwNWxVhz0x31PVo1Y+rPdb3iUkRiUrcG9gide1axRRL+z3NZ0YxGgcACEoEJ6AJaAiyR6M0RGlVvl82Z1dsWbJmZ56s2pFrthe/Xy8xkeFyZK9UU2RicHqS9GufJDFuglSz0Cl3WrSiseioWM4Wa3qiTjO0pxzqeTNKVlxxXLE3j31GzPb3fJXnahhx831eP9v3a4oGPV2LVkObdy6XZqPhSQOW7hPaWuXuK0vbV4yMVU5rdVU9rnyutusq1qxpeNNplpUjb/qZCdam0yz1WjPlsmKrnIJpb77Ph/tcU22tnH3Oft5sdrCMqPae1d9Xw6P92Pd7AgACFcEJaIYQNbRzitlse/KKzVqo737bbe4bpdP6Plu+w2xKi0r0SouXfh0STWGKnm3ipUdavHRpFevgN0G96C/uKV2tzUm6xqwiUJUW5csXc2bLcUcfIZEub9XgpWEqa6M12mavKbPXoNV4rJvH57yGuxJrSqVZq1a89311BE6f1+mSNv1cXe+mcrc69uMJDtUDYg3n7FBoBzU7RNY7rPk+rng+LELCXWEyKiNTwv/zkkiEjhi6rfV7dlVMs1ULkJVB0rXvOTucVg+b9nN2pU3zvuEVodcOv9EiCRVrCQHAQQQnwAG6vknLmOvm9Xpl+bYcM6Xvp41ZZkRqZ26xrNyeazZfCdERcsrAdhKV7ZJ+e/KlR1qShAVT5T40H/2lNizamu4YHiNF7lZWmIvUiXsOhDgz6qahKcOa1qjTHPWmzDpFUo/Fa12rUyJtlce+z1W/TqdQ2ufK9wa3Kvtiawql3rPMXK/t0TVw+9t8p2aWVzv2WTdnn/N9rIFSjw9YDT+Tak+Zz9KfZyPTCJWmB7nLJHDYgcw3sFUbNdRCL9rnq4wWVh8ltJ+r9n41jT7u83xN19bwOfs8X/FZOm3Vns66z4ikb9itPgq6n5HLBl1Tw/PaJi2KQ4VRoFYEJ8BhWm2vf4cks9m2ZxfJzzqlb0euuYfUb7vy5Ped+ZJbVCb/+WGz/jOzvLj6O4lzh0uf9onmRrxacOKIXqnSJj6q5VTwQwsKcTp9LsoaHQkFdiESM9rm3U84q3hu7wurvkdd57Voiq6xq/Ke5fv5XN9zFUG0cr83MJaVFsvPixfJ4EEDJUJHKfVz7HWA9gikCYi+xVd8A2T1sFlDyDTH+j723mczI6JaFKZk733ltI3VpqCikZnbNcTvDXpK+4Q+jqqoHlrTSKO4RP/3MB2lfO05n9BYQ0Azn1PtnD29Vqft2tNrda9rJesKmZW3rKioflrb9NuaRlf3mYZb8X30/cx3rRgB5b+pqEBwAgJQu6RoaZfUTsb1b1d5rrzcK/PX7pH//rxF5q/YJNuLwiW/xCOLNmSa7eX5G8x1CVER0jU1TgZ1SpJDuraS3u0SpEebeHFHsCAfaDbmHmHB+59Yb2mpbF4fJYMGneTMKGV1hVnWyGFtlTLtEKbXaNCqvMYnNNY4elgxslj9+Sqjkp79P1fvzyqvun6xSoj1fX1tQbu2ANzA5/dpc8VIqX2T8+p0Ou5+6H9h2uhB3gppcUzAsm9LUTFNtXIto88535HCmqau1nh+PyOotb6mhoBY43vV5/NrGCHV76KhVW87oqOkVcKnq+LvN7cVsitvau9bNKjiPootUPD+rQ6EGJ2Sd3jPVBnRJUlmRayXsePGyObsEjPNb9lWneq30xSdyC0uk6Vbss322gKrsltEmMtU/euUEiP92idK3/aJkpYYJZ1bxUrH5BhGqAAEtphkp1vQ8mmQ0qI2ulbRDnOG/vfBa4366bTaGkcUy6XMUyY/LV4kQ4cMlgj9xbuuEKfskKlhUm8/odNO7UI3uulje2Sztk1HQX2vqx5yq4zs+o6s+gZjn+9jr930ZT6npNn/SIJaeMX6S50+W71wTkXwDHeFi7vVJAkmBCcgSEWEh0mvtglm0+p9/zipr6ngtzGjwEzv+2F9hizdnG0q9+kUv3W78832zZqKhfkVdISqc+tYGd4lRbq0jpMOyTHSNjFK0hKjzbQ/RqoAIAToL7eJ7Q9olHLrukgZ0j9ARikPhI4ImjDmewsKu0qqHdB8H9uhq/q01RpGIPf3XJ2vqTbKWv01tU2d9fc15nYb+da6UFPgx2dar7fi2NzAvuLm9TWt6TRrWusepZSUCyWYEJyAFlbB76C2CWbTwhNKi09syy6SDXsKZP2efFm2NVtWb8+T3fnFsnFPgRmh0hEr3WrSKs4t7RKjZUDHRFMZUB+nxLqla+tYaZPAeioAQAtjbh6+d90x6uD1Vq28ao8C6vpLXatYJWDaaxk9UlZSJGVrarpHRuAiOAEtnAYbHUXSTW+460tHqDZlFJjqfVrNb2tWkWzJKpRducWyM7dISj1eycgvMZtOCXzzRy1MsZfef6pLa2sKYGJMpCRGR0qH5GgZ0a21pKfEmIBF1T8AAFowV8U99HSrVPX3jdpGKct/nyXBhOAEhPgIlT3d79TBHao8p8UosgpLTYDSkaklm7LMuqm84jITrLZmFUphqafGsum2yHCXGf3qlhonybGR0rttgglYWlY9JjJCYt3hZtN26J6gBQAAAhXBCUCNNMDotDzd+rRLlLE+Ff5USVm5bM4sMFMAdSpgTlGp5BSWyuodefLTxkzZk19iRqz2Nw2wOi2vPqBjknRvEy+J0RESHxVhQlZCtBW24qMjJD0l1oye6U2CAQAAmgvBCUCDaNEIDTi61aTUU27uR6WjVDtyisyxrrHSQhW66WhVYYlHCkrKpKDEI8Vl5aa8+oJ1GWarS1REWMWIVUTlaFXreLdZd9U9NU7ioiIkxh1upg/qOR3x0kpPWmFQQ5gW1wAAAKgvghOAJhEZHmZKoOtWH2WecnOj3182ZZsRrLzi0sqQpaNZ1nGpGeEqK/eaoKVbZoH/C0vd4WFm9EpHraIjw8x9rrQ0u4avlNhISU2Ikq6t48yIl16new1qFMIAACB0EZwABAQdAdIpgbrtT3GZR/KKrFEqHbXSfUFxmWQUlMievBJT3EILXtgjWtlmnVaxmUaogUuVeMpNwQvbpozCOtun67V0ymBcVLgJhTpypaNdPdPipX1StCTFRJrgpSNbegNjXdelzwMAgJaB/6oDCCpREeESFR9ej3o9NfOUa3n2QhOqdA2WFrvQe13tzCmSzIISycwvla3ZhbIls1Dyi8vM9EG1t8Jg1ffTohm10XVaGp50yqCOWOleKxFqyEqO1fVjkZVTDFvFRUnrOLeZYqjXasEMHQ3T6xnpAgDAeQQnACFFp+d1Sqk6fXBEt1a1Xq/VBfNLrCmDuukUwjKP1wQwrTq4Zkee7MkvNlMGszR4FZSYsu4asnLMNMOyA2qv1sDQ8KUjXrquTEe9NHhpKNO9Oa4oBa+hS9d76XUavnSv0xLDXV7Zki/y+658iYt2m+eiIsPNzY+pYggAQP0QnABgPzRYWFX9fO9P4WNgzac1RO3OKzH3ytLNnjqoe50+qMEqq6DUVB/MyC820wz1nF6ja7d0OqHS2YU6KmbbkVPcwG8SIQ/88t0+IVLXdKUlRJt1XTGRYZUjZDrS5VsqXrf4qL3VDRN9qh0yKgYACAUEJwBoAjoVT7eG0hEtDVz2dEEtnqGBStdqWaXfy0wA02Pd28UzikrLTal4XQtmAljFcW5+obk5oQYyPa83etfP0HCnm2xr+HfVAGaVjddCGlaY0kCmo2CRFaNfOrql69j0OCXObaYl6s/HlJ2vCGFajp4ABgAIVAQnAAhAGkZ06p1uB6q0tFRmzZolJ500TiIjI8Xr9ZoApSNeu/OKTal4nWpYWGKVibcLb5gRMp8iHHnVKhzqSJiOiJlpi2aqolY4rLvQRm101qCOcGnhDQ1ZujeFOMIr9mHWdEU91qmLutepiPY19rFeoz83DXA6UqYjYtY6M2tvHptzYVXO6WdzfzAAQG0ITgAQYnRURwND20TdoqV/h6QGvY8GMA1UdpDS9VwapnRUzB4NKy3zSlFFJURTRr7UUzE9sUSyCksq145p+NIQpu/nJA1d0T6FPKwiHXvDl31OQ1lUpBXm9P5gGuTCw7SoR5jE6b3FosLNXq/T630Dn1mDFr53LRrrzAAgOBCcAAANDmD2qJiWYG8oDWA6xdCealhaXm4KcOhNlDVQlZXrlENrr+f0WPfmcZk1elZauVn3+NKgpu9XWG2Nmb5/5eOKc3q9Tac26nagRT38YcKUTmOMtKov2oFKw+3eY/u87zU+z0eG7X2PilE3uziICX7ucFNKPzbSCnX2SBxTIwGg/ghOAABH6S/vZjTHHe7I52vlRA1PvmGqelGPorJyKfKdtlhcKsWlGt6sQGcFPX0f63ldm5ZXsel1JtzpejMT/PYGNaWvK/V4KkvfNydrpMwaNbODljU65pKi/HB5YfMCE9A0aGnJfHv6o13BMc5tBWd7hM1+Lq7iMdMfAbQkBCcAQEjTqXLNGdx0hE3DkinUUepbxMMq5GEf+xb5qNxKPRWv2/caPdawZ0blKgKa/V7mRtEl1o2jfadDWqFNbwxdLrJPwUaXbCnIPuDvq9MZ7fuSmRGzSJ0Oue/efl73VjXHCEmJs0ru24VHNJAlREWagiJ6rNcDQHMhOAEA0MwjbO4Ia61TfCMU/2jICJuuO8sv9lSOlunImR20SjweKSgqle/mL5DBBw8Tj7jMCJquTSso1tE2DWBWtceCiqqP5nGxNdJmRtxKykzlRqXvbY2+Nf530ZExe/2ZBi1rPdre+53ZYcysXau4obReo1MaYytK7duvtd8nMSZCkmPc5jUA4IvgBABAiI2waVDQbX+VGDNWemV03zRTibGh69bMlEcdEdNje2/OWaNn1fc6OlZUsdeiIXo/NK3WqEHMrFvToFYRzvaOmNk3mm7cZGbWjOm0xYqpi3aBD9/qjXZ1R98bTldff2YXFzEhriK4Vb63WZdmBzp7xG3va/V5pjoCgYPgBAAAgmrdmt7XTEe6NET53lzalNIvsYKUVnfcO83RJ7SZrVwKzOuskTI9Z09n1HCmo2VmuqPeiLoJRsr8ocFpb0CzioNo2Ko8No/3HpsRNZ+9PZIWXUvxEfuxXtchOaZRboEAtFT8vwMAAAQVvc9XUoxu/o+G1UUrOdr3KbNGtPauFdPAZq1P07Vo1nP28/a6Mjus2evPNKRpQDMjcOaxrkPbOzWy8hoz0mYdV2+PdUPscsmVpq/2qDenTjTTHPdWX7SmM1Yty6+P7bVoehwZ5pWVWS5J25ApCTFRVe6ZZo+iMXqGYEdwAgAAqKC/3CfHus3mBLt4iB3IdF/ss/ctJGIXCbGnN9a83zsi5/t63+Cme3MPtiJrLZtuDRMuT6/4odZndeTMLgCyd/TM2mv4sraIKntTSt9+XFlK3yoOUv06nQJJiX00JYITAABAABYPiYtq3s/Wm1Zvzig0a8o0bJmpj6V6rJUZy3ymNO6d3mhPlcwvLpUde7LEHRNX5b5pZrpjBXtNWlMUCrFDb6zvCFllCNs7SqbnrCAWLtFuqzKj71q0vQVDwk3xFh19M9dGhHOzahCcAAAAIJIYHSn9OjRs+qMWFJk1a5acdNIRVQqK6PTGfdaaVRsts++ZZocyrdaoa9DMXh/7Hlcrra8BzQ5nZpplsbVOrSlomLWnH2q4MmvDfKckVgSsGLddYn9veX1TGKTaGjTf56uX6NdrGD0LPAQnAAAANNl6NN2acvRMpzHaIcoOVb7FQnxDlhXE9o6oVV2LZr/Oel4rO/qGMHt6Y3ZhqTQ1zUzVi31U3u/MLqtfcV5HxtISoqRNgnWTan2smxb6MMd63zN3BOvMGgHBCQAAAEErsgmLhdj3PTNTD8uscGWPkJl9SdVj32vsCo6+Jfkrz+nNrO0RN5+Rt/KK+59pZUfruqrFQg6U5ia7nH6rOLekxkeZqYj2lEW7ymLreLe0S4w24cuE3zCXtVVUeNSfdYpZCxhpwlmojI45HpymTZsmDz74oGzfvl0GDx4sTzzxhIwYMaLGa5ctWya33nqrLFq0SDZs2CCPPPKIXH311c3eZgAAALR89bnvWWMXBrFHwOziHnrse27vc9Ze16btyCmW3XnF5n5n5r5netNpPa64SbVNg5k9dVJH1DbsKWiUtrt1TZ473IQpHeHSYKUjYLqvXkZfpzVqcY+ocC37L0HF0eA0c+ZMufbaa2X69OkycuRIefTRR2XcuHGyatUqSUtL2+f6goIC6d69u5x99tlyzTXXONJmAAAAoCkLgyREN+6oma4T0/Vmuh5Mw5kGrsz8EtmVW2xGzHyrLepzu/KKZXt2kXlO146VebxSVl4uZeVeM3qm0xWzCksqR8TsaYyZBf5NY7xzmAQVR4PTww8/LJMnT5ZJkyaZxxqgPv74Y5kxY4bcdNNN+1x/yCGHmE3V9DwAAACAqqNmOp1uH20O/L2LSq0bUdsjWBn5JeZxZkGJCV866lW9fL6GMb0mr7hUosIzJJg4FpxKSkrMlLubb7658lxYWJiMHj1a5s+f32ifU1xcbDZbTk5OZfUX3ZxmtyEQ2oLgQJ+Bv+gz8Bd9Bv6iz4SmcK3GGBUmEhUmbeIipHvr+g+VaV+ZM2eO433Gn893LDjt3r1bPB6PtG3btsp5fbxy5cpG+5ypU6fKHXfcsc/5zz77TGJjYyVQaMcB/EGfgb/oM/AXfQb+os8g2PqMLgUKmuIQTU1HtHQdle+IU3p6uowdO1YSExPFaXbaHjNmTJX7HgC1oc/AX/QZ+Is+A3/RZxCsfcaejRbQwSk1NVXCw8Nlx44dVc7r43bt2jXa50RFRZmtOv0DCqT/YwdaexD46DPwF30G/qLPwF/0GQRbn/Hns8PEIW63W4YNGyZz586tPFdeXm4ejxo1yqlmAQAAAEBgTdXTKXQTJ06U4cOHm3s3aTny/Pz8yip7EyZMkI4dO5p1SnZBieXLl1ceb9myRZYsWSLx8fHSs2dPJ78KAAAAgBbM0eA0fvx42bVrl7mprd4Ad8iQITJ79uzKghEbN240lfZsW7dulaFDh1Y+fuihh8x29NFHy7x58xz5DgAAAABaPseLQ0yZMsVsNakehrp27WruqgwAAAAAzcmxNU4AAAAAECwITgAAAABQB4ITAAAAANSB4AQAAAAAdSA4AQAAAEAdCE4AAAAAUAeCEwAAAADUgeAEAAAAAHUgOAEAAABAHQhOAAAAAFAHghMAAAAA1CFCQozX6zX7nJwcCQSlpaVSUFBg2hMZGel0cxAE6DPwF30G/qLPwF/0GQRrn7EzgZ0R9ifkglNubq7Zp6enO90UAAAAAAGSEZKSkvZ7jctbn3jVgpSXl8vWrVslISFBXC6X080xKVdD3KZNmyQxMdHp5iAI0GfgL/oM/EWfgb/oMwjWPqNRSENThw4dJCxs/6uYQm7ESX8gnTp1kkCjHYa/aOAP+gz8RZ+Bv+gz8Bd9BsHYZ+oaabJRHAIAAAAA6kBwAgAAAIA6EJwcFhUVJbfddpvZA/VBn4G/6DPwF30G/qLPIBT6TMgVhwAAAAAAfzHiBAAAAAB1IDgBAAAAQB0ITgAAAABQB4ITAAAAANSB4OSgadOmSdeuXSU6OlpGjhwpCxcudLpJcMDUqVPlkEMOkYSEBElLS5MzzjhDVq1aVeWaoqIiueKKK6R169YSHx8vf/jDH2THjh1Vrtm4caOcfPLJEhsba97nhhtukLKysmb+NnDCfffdJy6XS66++urKc/QZVLdlyxb585//bPpETEyMDBw4UH788cfK57VW1K233irt27c3z48ePVrWrFlT5T0yMjLk/PPPNzerTE5Olosvvljy8vIc+DZoDh6PR/75z39Kt27dTJ/o0aOH3HXXXaav2Og3oe3rr7+WU089VTp06GD+O/T+++9Xeb6x+scvv/wiRx55pPmdOT09XR544AFxhFbVQ/N74403vG632ztjxgzvsmXLvJMnT/YmJyd7d+zY4XTT0MzGjRvnfeGFF7y//vqrd8mSJd6TTjrJ27lzZ29eXl7lNZdeeqk3PT3dO3fuXO+PP/7oPfTQQ72HHXZY5fNlZWXeAQMGeEePHu396aefvLNmzfKmpqZ6b775Zoe+FZrLwoULvV27dvUOGjTIe9VVV1Wep8/AV0ZGhrdLly7eCy+80LtgwQLv2rVrvZ9++qn3t99+q7zmvvvu8yYlJXnff/99788//+w97bTTvN26dfMWFhZWXnPCCSd4Bw8e7P3f//7n/eabb7w9e/b0nnvuuQ59KzS1e+65x9u6dWvvRx995F23bp33rbfe8sbHx3sfe+yxymvoN6Ft1qxZ3ltuucX77rvvapr2vvfee1Web4z+kZ2d7W3btq33/PPPN78r/ec///HGxMR4//3vf3ubG8HJISNGjPBeccUVlY89Ho+3Q4cO3qlTpzraLjhv586d5i+fr776yjzOysryRkZGmv9g2VasWGGumT9/fuVfXGFhYd7t27dXXvP00097ExMTvcXFxQ58CzSH3Nxcb69evbxz5szxHn300ZXBiT6D6m688UbvEUccUevz5eXl3nbt2nkffPDBynPaj6KioswvKWr58uWmD/3www+V13zyySdel8vl3bJlSxN/Azjh5JNP9l500UVVzp111lnmF1hFv4Gv6sGpsfrHU0895U1JSany3yb9O613797e5sZUPQeUlJTIokWLzHClLSwszDyeP3++o22D87Kzs82+VatWZq99pbS0tEp/6dOnj3Tu3Lmyv+hep920bdu28ppx48ZJTk6OLFu2rNm/A5qHTsXTqXa+fUPRZ1Ddhx9+KMOHD5ezzz7bTMscOnSoPPvss5XPr1u3TrZv316lzyQlJZlp5L59RqfR6PvY9Hr979eCBQua+RuhORx22GEyd+5cWb16tXn8888/y7fffisnnniieUy/wf40Vv/Qa4466ihxu91V/nulyxoyMzOlOUU066fB2L17t5k37PsLi9LHK1eudKxdcF55eblZp3L44YfLgAEDzDn9S0f/stC/WKr3F33Ovqam/mQ/h5bnjTfekMWLF8sPP/ywz3P0GVS3du1aefrpp+Xaa6+Vf/zjH6bf/O1vfzP9ZOLEiZV/5jX1Cd8+o6HLV0REhPlHHvpMy3TTTTeZf0zRf3gJDw83v7vcc889Zj2Kot9gfxqrf+he19lVfw/7uZSUFGkuBCcgwEYQfv31V/MvekBtNm3aJFdddZXMmTPHLJQF6vOPMvovuvfee695rCNO+nfN9OnTTXACavLmm2/Ka6+9Jq+//rr0799flixZYv5xTwsB0G8Qipiq54DU1FTzLzfVK1zp43bt2jnWLjhrypQp8tFHH8mXX34pnTp1qjyvfUKnd2ZlZdXaX3RfU3+yn0PLolPxdu7cKQcffLD5lzndvvrqK3n88cfNsf5LHH0GvrSiVb9+/aqc69u3r6ms6Ptnvr//Lule+50vrcKoFbHoMy2TVtrUUadzzjnHTO294IIL5JprrjHVYBX9BvvTWP0jkP57RXBygE6NGDZsmJk37Puvgfp41KhRjrYNzU/XU2poeu+99+SLL77YZzha+0pkZGSV/qLzevUXHru/6H7p0qVV/vLR0Qgt7Vn9lyUEv+OPP978eeu//tqbjibo9Bn7mD4DXzr9t/ptDnTdSpcuXcyx/r2jv4D49hmdoqVrDHz7jIZxDe42/TtL//ulaxbQ8hQUFJi1Jr70H371z1zRb7A/jdU/9Bote65rd33/e9W7d+9mnaZnNHs5ClSWI9eqIi+++KKpKHLJJZeYcuS+Fa4QGi677DJTqnPevHnebdu2VW4FBQVVSktrifIvvvjClJYeNWqU2aqXlh47dqwpaT579mxvmzZtKC0dQnyr6in6DKqXrY+IiDDlpdesWeN97bXXvLGxsd5XX321Stlg/e/QBx984P3ll1+8p59+eo1lg4cOHWpKmn/77bemqiNlpVuuiRMnejt27FhZjlxLTuttC/7+979XXkO/CW25ubnmlha6aax4+OGHzfGGDRsarX9oJT4tR37BBReYcuT6O7T+/UU58hDzxBNPmF9s9H5OWp5c69cj9OhfNDVtem8nm/4Fc/nll5tynPqXxZlnnmnCla/169d7TzzxRHNvA/0P23XXXectLS114BshEIITfQbV/fe//zVhWf/Rrk+fPt5nnnmmyvNaOvif//yn+QVFrzn++OO9q1atqnLNnj17zC80ei8fLV0/adIk84sTWqacnBzz94r+rhIdHe3t3r27uWePb1lo+k1o+/LLL2v8HUZDd2P2D70HlN5SQd9Dw7wGMie49H+ad4wLAAAAAIILa5wAAAAAoA4EJwAAAACoA8EJAAAAAOpAcAIAAACAOhCcAAAAAKAOBCcAAAAAqAPBCQAAAADqQHACAAAAgDoQnAAA2I+uXbvKo48+6nQzAAAOIzgBAALGhRdeKGeccYY5PuaYY+Tqq69uts9+8cUXJTk5eZ/zP/zwg1xyySXN1g4AQGCKcLoBAAA0pZKSEnG73Q1+fZs2bRq1PQCA4MSIEwAgIEeevvrqK3nsscfE5XKZbf369ea5X3/9VU488USJj4+Xtm3bygUXXCC7d++ufK2OVE2ZMsWMVqWmpsq4cePM+YcfflgGDhwocXFxkp6eLpdffrnk5eWZ5+bNmyeTJk2S7Ozsys+7/fbba5yqt3HjRjn99NPN5ycmJsqf/vQn2bFjR+Xz+rohQ4bIK6+8Yl6blJQk55xzjuTm5jbbzw8A0PgITgCAgKOBadSoUTJ58mTZtm2b2TTsZGVlyXHHHSdDhw6VH3/8UWbPnm1Ci4YXXy+99JIZZfruu+9k+vTp5lxYWJg8/vjjsmzZMvP8F198IX//+9/Nc4cddpgJRxqE7M+7/vrr92lXeXm5CU0ZGRkm2M2ZM0fWrl0r48ePr3Ld77//Lu+//7589NFHZtNr77vvvib9mQEAmhZT9QAAAUdHaTT4xMbGSrt27SrPP/nkkyY03XvvvZXnZsyYYULV6tWr5aCDDjLnevXqJQ888ECV9/RdL6UjQXfffbdceuml8tRTT5nP0s/UkSbfz6tu7ty5snTpUlm3bp35TPXyyy9L//79zVqoQw45pDJg6ZqphIQE81hHxfS199xzT6P9jAAAzYsRJwBA0Pj555/lyy+/NNPk7K1Pnz6Vozy2YcOG7fPazz//XI4//njp2LGjCTQaZvbs2SMFBQX1/vwVK1aYwGSHJtWvXz9TVEKf8w1mdmhS7du3l507dzboOwMAAgMjTgCAoKFrkk499VS5//7793lOw4lN1zH50vVRp5xyilx22WVm1KdVq1by7bffysUXX2yKR+jIVmOKjIys8lhHsnQUCgAQvAhOAICApNPnPB5PlXMHH3ywvPPOO2ZEJyKi/v8JW7RokQku//rXv8xaJ/Xmm2/W+XnV9e3bVzZt2mQ2e9Rp+fLlZu2VjjwBAFoupuoBAAKShqMFCxaY0SKtmqfB54orrjCFGc4991yzpkin53366aemIt7+Qk/Pnj2ltLRUnnjiCVPMQSve2UUjfD9PR7R0LZJ+Xk1T+EaPHm0q851//vmyePFiWbhwoUyYMEGOPvpoGT58eJP8HAAAgYHgBAAISFrVLjw83Izk6L2UtAx4hw4dTKU8DUljx441IUaLPugaI3skqSaDBw825ch1it+AAQPktddek6lTp1a5RivrabEIrZCnn1e9uIQ95e6DDz6QlJQUOeqoo0yQ6t69u8ycObNJfgYAgMDh8nq9XqcbAQAAAACBjBEnAAAAAKgDwQkAAAAA6kBwAgAAAIA6EJwAAAAAoA4EJwAAAACoA8EJAAAAAOpAcAIAAACAOhCcAAAAAKAOBCcAAAAAqAPBCQAAAADqQHACAAAAANm//wdt+JO7bWDoZwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "evals_result = model.get_evals_result()\n",
    "\n",
    "# Menampilkan skor terakhir\n",
    "train_score = evals_result['learn']['Logloss'][-1]\n",
    "val_score = evals_result['validation']['Logloss'][-1]\n",
    "\n",
    "print(f\"Final Training Logloss: {train_score}\")\n",
    "print(f\"Final Validation Logloss: {val_score}\")\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Ambil skor training dan validation dari evals_result\n",
    "train_logloss = evals_result['learn']['Logloss']\n",
    "val_logloss = evals_result['validation']['Logloss']\n",
    "\n",
    "# Plot learning curve\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(train_logloss, label='Training Logloss')\n",
    "plt.plot(val_logloss, label='Validation Logloss')\n",
    "plt.xlabel('Iteration')\n",
    "plt.ylabel('Logloss')\n",
    "plt.title('Learning Curve')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "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>domisili</th>\n",
       "      <th>jenis_kelamin</th>\n",
       "      <th>date_of_birth</th>\n",
       "      <th>join_date</th>\n",
       "      <th>resign_date</th>\n",
       "      <th>marriage_stat</th>\n",
       "      <th>dependant</th>\n",
       "      <th>education</th>\n",
       "      <th>absent_90D</th>\n",
       "      <th>...</th>\n",
       "      <th>active_work_category</th>\n",
       "      <th>work_stability_score</th>\n",
       "      <th>married_dependent_ratio</th>\n",
       "      <th>position_score</th>\n",
       "      <th>job_income_position_score</th>\n",
       "      <th>education_score</th>\n",
       "      <th>education_income_ratio</th>\n",
       "      <th>weighted_satisfaction_performance</th>\n",
       "      <th>resign_risk_indicator</th>\n",
       "      <th>adjusted_work_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>EM0120</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1990-02-18</td>\n",
       "      <td>2023-01-11</td>\n",
       "      <td>2024-01-30</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>11.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1140915.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1140915.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.393432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>EM13985</td>\n",
       "      <td>Kepulauan Seribu</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1987-02-01</td>\n",
       "      <td>2022-09-26</td>\n",
       "      <td>2023-11-08</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2103348.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2103348.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>EM2560</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1999-11-01</td>\n",
       "      <td>2023-01-05</td>\n",
       "      <td>2024-05-04</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.454545</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2145814.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2145814.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.205670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>EM0343</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1990-10-12</td>\n",
       "      <td>2022-02-01</td>\n",
       "      <td>2023-07-17</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>8.500000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2331081.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2331081.0</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.154017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>EM14458</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1996-04-24</td>\n",
       "      <td>2022-10-23</td>\n",
       "      <td>2023-12-30</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>12.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.076923</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1798725.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1798725.0</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.706741</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  employee_id          domisili jenis_kelamin date_of_birth   join_date  \\\n",
       "0      EM0120         Tangerang     Laki-laki    1990-02-18  2023-01-11   \n",
       "1     EM13985  Kepulauan Seribu     Perempuan    1987-02-01  2022-09-26   \n",
       "2      EM2560         Tangerang     Perempuan    1999-11-01  2023-01-05   \n",
       "3      EM0343  Kabupaten Bekasi     Laki-laki    1990-10-12  2022-02-01   \n",
       "4     EM14458   Kabupaten Bogor     Perempuan    1996-04-24  2022-10-23   \n",
       "\n",
       "  resign_date marriage_stat  dependant education  absent_90D  ...  \\\n",
       "0  2024-01-30       Married          1      SLTA        11.0  ...   \n",
       "1  2023-11-08       Married          1      SLTA         0.0  ...   \n",
       "2  2024-05-04        Single          0      SLTA        10.0  ...   \n",
       "3  2023-07-17        Single          0      SLTA         1.0  ...   \n",
       "4  2023-12-30       Married          1      SLTA        12.0  ...   \n",
       "\n",
       "   active_work_category work_stability_score married_dependent_ratio  \\\n",
       "0              Mid-term             1.000000                       2   \n",
       "1              Mid-term            13.000000                       2   \n",
       "2              Mid-term             1.454545                       1   \n",
       "3              Mid-term             8.500000                       1   \n",
       "4              Mid-term             1.076923                       2   \n",
       "\n",
       "   position_score  job_income_position_score education_score  \\\n",
       "0               1                  1140915.0               1   \n",
       "1               1                  2103348.0               1   \n",
       "2               1                  2145814.0               1   \n",
       "3               1                  2331081.0               1   \n",
       "4               1                  1798725.0               1   \n",
       "\n",
       "  education_income_ratio  weighted_satisfaction_performance  \\\n",
       "0              1140915.0                                1.4   \n",
       "1              2103348.0                                1.8   \n",
       "2              2145814.0                                1.6   \n",
       "3              2331081.0                                2.6   \n",
       "4              1798725.0                                2.6   \n",
       "\n",
       "   resign_risk_indicator  adjusted_work_time  \n",
       "0                 Medium            9.393432  \n",
       "1                 Medium            9.300000  \n",
       "2                 Medium            9.205670  \n",
       "3                 Medium            9.154017  \n",
       "4                 Medium            9.706741  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = pd.read_csv('D:/Tugas Akhir/Codingan/Notebook - Playground/preprocessed_data_test_5.csv')\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "churn_status\n",
       "1    809\n",
       "0    798\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test['churn_status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1607"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9477286869943995\n",
      "Precision: 0.9142857142857143\n",
      "Recall: 0.9888751545117429\n",
      "F1 Score: 0.9501187648456056\n"
     ]
    }
   ],
   "source": [
    "X_test = df_test.drop(['churn_status', 'employee_id', 'date_of_birth', 'join_date', 'resign_date', 'active_work_months'], axis=1)\n",
    "\n",
    "cat_feature = ['departemen', 'position', 'domisili', 'marriage_stat', 'job_satisfaction', 'performance_rating',\n",
    "               'education', 'active_work_category', 'resign_risk_indicator', 'jenis_kelamin']\n",
    "\n",
    "y_pred = final_model.predict(X_test)\n",
    "\n",
    "X_test['predicted_churn'] = y_pred\n",
    "\n",
    "accuracy = accuracy_score(df_test['churn_status'], y_pred)\n",
    "precision = precision_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "recall = recall_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "f1 = f1_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "\n",
    "print(\"Accuracy:\", accuracy)\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1 Score:\", f1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>domisili</th>\n",
       "      <th>jenis_kelamin</th>\n",
       "      <th>date_of_birth</th>\n",
       "      <th>join_date</th>\n",
       "      <th>resign_date</th>\n",
       "      <th>marriage_stat</th>\n",
       "      <th>dependant</th>\n",
       "      <th>education</th>\n",
       "      <th>absent_90D</th>\n",
       "      <th>...</th>\n",
       "      <th>active_work_category</th>\n",
       "      <th>work_stability_score</th>\n",
       "      <th>married_dependent_ratio</th>\n",
       "      <th>position_score</th>\n",
       "      <th>job_income_position_score</th>\n",
       "      <th>education_score</th>\n",
       "      <th>education_income_ratio</th>\n",
       "      <th>weighted_satisfaction_performance</th>\n",
       "      <th>resign_risk_indicator</th>\n",
       "      <th>adjusted_work_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>EM0406</td>\n",
       "      <td>Kota Jakarta Utara</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1975-01-07</td>\n",
       "      <td>2021-09-25</td>\n",
       "      <td>2023-12-07</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>6.500000</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1578410.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1578410.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.797428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>EM1772</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1993-04-18</td>\n",
       "      <td>2021-02-23</td>\n",
       "      <td>2023-06-24</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>9.333333</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2003154.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2003154.00</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.342582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>EM7996</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1998-02-12</td>\n",
       "      <td>2023-05-04</td>\n",
       "      <td>2024-06-29</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1394384.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1394384.00</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.551975</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>EM13978</td>\n",
       "      <td>Kota Jakarta Barat</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1982-12-26</td>\n",
       "      <td>2021-09-11</td>\n",
       "      <td>2023-04-03</td>\n",
       "      <td>Married</td>\n",
       "      <td>0</td>\n",
       "      <td>D3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4151999.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1037999.75</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.180000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>EM9860</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1997-03-26</td>\n",
       "      <td>2023-06-20</td>\n",
       "      <td>2024-10-02</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>15.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>0.937500</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1560817.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1560817.00</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.414301</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  employee_id            domisili jenis_kelamin date_of_birth   join_date  \\\n",
       "0      EM0406  Kota Jakarta Utara     Laki-laki    1975-01-07  2021-09-25   \n",
       "1      EM1772     Kabupaten Bogor     Perempuan    1993-04-18  2021-02-23   \n",
       "2      EM7996           Tangerang     Laki-laki    1998-02-12  2023-05-04   \n",
       "3     EM13978  Kota Jakarta Barat     Perempuan    1982-12-26  2021-09-11   \n",
       "4      EM9860           Tangerang     Perempuan    1997-03-26  2023-06-20   \n",
       "\n",
       "  resign_date marriage_stat  dependant education  absent_90D  ...  \\\n",
       "0  2023-12-07       Married          3      SLTA         3.0  ...   \n",
       "1  2023-06-24       Married          1      SLTA         2.0  ...   \n",
       "2  2024-06-29        Single          0      SLTA         5.0  ...   \n",
       "3  2023-04-03       Married          0        D3         0.0  ...   \n",
       "4  2024-10-02        Single          0      SLTA        15.0  ...   \n",
       "\n",
       "   active_work_category work_stability_score married_dependent_ratio  \\\n",
       "0              Mid-term             6.500000                       4   \n",
       "1              Mid-term             9.333333                       2   \n",
       "2              Mid-term             2.333333                       1   \n",
       "3              Mid-term            18.000000                       1   \n",
       "4              Mid-term             0.937500                       1   \n",
       "\n",
       "   position_score  job_income_position_score education_score  \\\n",
       "0               1                  1578410.0               1   \n",
       "1               1                  2003154.0               1   \n",
       "2               1                  1394384.0               1   \n",
       "3               1                  4151999.0               4   \n",
       "4               1                  1560817.0               1   \n",
       "\n",
       "  education_income_ratio  weighted_satisfaction_performance  \\\n",
       "0             1578410.00                                1.0   \n",
       "1             2003154.00                                2.2   \n",
       "2             1394384.00                                3.0   \n",
       "3             1037999.75                                2.6   \n",
       "4             1560817.00                                2.6   \n",
       "\n",
       "   resign_risk_indicator  adjusted_work_time  \n",
       "0                 Medium            9.797428  \n",
       "1                 Medium            9.342582  \n",
       "2                 Medium            9.551975  \n",
       "3                 Medium            9.180000  \n",
       "4                 Medium            9.414301  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = pd.read_csv('D:/Tugas Akhir/Codingan/Notebook - Playground/preprocessed_data_test_7.csv')\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "churn_status\n",
       "1    161\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test['churn_status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "161"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 1.0\n",
      "Precision: 1.0\n",
      "Recall: 1.0\n",
      "F1 Score: 1.0\n"
     ]
    }
   ],
   "source": [
    "X_test = df_test.drop(['churn_status', 'employee_id', 'date_of_birth', 'join_date', 'resign_date', 'active_work_months'], axis=1)\n",
    "\n",
    "cat_feature = ['departemen', 'position', 'domisili', 'marriage_stat', 'job_satisfaction', 'performance_rating',\n",
    "               'education', 'active_work_category', 'resign_risk_indicator', 'jenis_kelamin']\n",
    "\n",
    "y_pred = final_model.predict(X_test)\n",
    "\n",
    "X_test['predicted_churn'] = y_pred\n",
    "\n",
    "accuracy = accuracy_score(df_test['churn_status'], y_pred)\n",
    "precision = precision_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "recall = recall_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "f1 = f1_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "\n",
    "print(\"Accuracy:\", accuracy)\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1 Score:\", f1)"
   ]
  },
  {
   "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>domisili</th>\n",
       "      <th>jenis_kelamin</th>\n",
       "      <th>date_of_birth</th>\n",
       "      <th>join_date</th>\n",
       "      <th>resign_date</th>\n",
       "      <th>marriage_stat</th>\n",
       "      <th>dependant</th>\n",
       "      <th>education</th>\n",
       "      <th>absent_90D</th>\n",
       "      <th>...</th>\n",
       "      <th>active_work_category</th>\n",
       "      <th>work_stability_score</th>\n",
       "      <th>married_dependent_ratio</th>\n",
       "      <th>position_score</th>\n",
       "      <th>job_income_position_score</th>\n",
       "      <th>education_score</th>\n",
       "      <th>education_income_ratio</th>\n",
       "      <th>weighted_satisfaction_performance</th>\n",
       "      <th>resign_risk_indicator</th>\n",
       "      <th>adjusted_work_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>EM0012</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1970-12-21</td>\n",
       "      <td>2023-02-23</td>\n",
       "      <td>2024-08-07</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>D3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.666667</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>4708861.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.177215e+06</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.857106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>EM0026</td>\n",
       "      <td>Kota Depok</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1986-11-14</td>\n",
       "      <td>2022-04-17</td>\n",
       "      <td>2024-08-04</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.600000</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1430853.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.430853e+06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.694593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>EM0041</td>\n",
       "      <td>Kota Jakarta Barat</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1983-03-16</td>\n",
       "      <td>2023-06-15</td>\n",
       "      <td>2024-09-06</td>\n",
       "      <td>Divorce</td>\n",
       "      <td>3</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.750000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1379381.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.379381e+06</td>\n",
       "      <td>2.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.059429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>EM0053</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1979-07-13</td>\n",
       "      <td>2023-07-11</td>\n",
       "      <td>2024-09-21</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1911583.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.911583e+06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.842189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>EM0057</td>\n",
       "      <td>Kota Jakarta Barat</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>2000-03-13</td>\n",
       "      <td>2022-07-14</td>\n",
       "      <td>2024-08-29</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D2</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.777778</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3724157.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.241386e+06</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.047730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>EM0058</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1996-04-23</td>\n",
       "      <td>2023-07-18</td>\n",
       "      <td>2024-09-26</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>9.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.400000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2229928.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.229928e+06</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.114481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>EM0064</td>\n",
       "      <td>Kota Jakarta Utara</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1987-04-20</td>\n",
       "      <td>2022-07-25</td>\n",
       "      <td>2024-08-23</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1257855.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.257855e+06</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.320000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>EM0180</td>\n",
       "      <td>Kota Jakarta Utara</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>2000-06-25</td>\n",
       "      <td>2022-09-04</td>\n",
       "      <td>2024-10-07</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D2</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>3.125000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3034058.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.011353e+06</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.091639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>EM0259</td>\n",
       "      <td>Kepulauan Seribu</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1993-10-04</td>\n",
       "      <td>2023-06-08</td>\n",
       "      <td>2024-08-29</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4513378.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2.256689e+06</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.479833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>EM0263</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1995-02-15</td>\n",
       "      <td>2022-06-15</td>\n",
       "      <td>2024-07-27</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.785714</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1599099.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.599099e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.077272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>EM0268</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1984-09-25</td>\n",
       "      <td>2023-07-04</td>\n",
       "      <td>2024-09-21</td>\n",
       "      <td>Married</td>\n",
       "      <td>5</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>2869178.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.869178e+06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.599453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>EM0274</td>\n",
       "      <td>Kota Bogor</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1995-07-09</td>\n",
       "      <td>2023-07-15</td>\n",
       "      <td>2024-10-02</td>\n",
       "      <td>Married</td>\n",
       "      <td>0</td>\n",
       "      <td>D1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3040879.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.520440e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.781063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>EM0360</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1979-08-13</td>\n",
       "      <td>2022-04-17</td>\n",
       "      <td>2024-09-19</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>D3</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>3.625000</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>4658718.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.164680e+06</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.494477</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>EM0368</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1979-12-25</td>\n",
       "      <td>2022-05-25</td>\n",
       "      <td>2024-08-02</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>D1</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.888889</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3326206.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.663103e+06</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.786442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>EM0384</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1976-08-11</td>\n",
       "      <td>2022-07-05</td>\n",
       "      <td>2024-09-21</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>D3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.200000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3215076.0</td>\n",
       "      <td>4</td>\n",
       "      <td>8.037690e+05</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.773272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>EM0388</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1970-11-15</td>\n",
       "      <td>2023-07-10</td>\n",
       "      <td>2024-09-07</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>12.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.076923</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1178459.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.178459e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.072831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>EM0398</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1999-05-03</td>\n",
       "      <td>2023-09-01</td>\n",
       "      <td>2024-10-12</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1527441.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.527441e+06</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.390000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>EM0481</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1997-12-24</td>\n",
       "      <td>2022-06-08</td>\n",
       "      <td>2024-09-27</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>9.333333</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2890639.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.890639e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.562408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>EM0483</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1975-08-05</td>\n",
       "      <td>2023-06-14</td>\n",
       "      <td>2024-07-25</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>12.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1193560.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.193560e+06</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.313570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>EM0491</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1969-04-10</td>\n",
       "      <td>2023-06-28</td>\n",
       "      <td>2024-09-14</td>\n",
       "      <td>Married</td>\n",
       "      <td>5</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>2048458.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.048458e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.372688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>EM0493</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1996-08-05</td>\n",
       "      <td>2023-07-04</td>\n",
       "      <td>2024-07-25</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1267701.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.267701e+06</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.190102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>EM0499</td>\n",
       "      <td>Kota Jakarta Pusat</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1990-10-23</td>\n",
       "      <td>2022-07-20</td>\n",
       "      <td>2024-09-25</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.857143</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1544522.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.544522e+06</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.825135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>EM0504</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>2000-04-19</td>\n",
       "      <td>2023-08-01</td>\n",
       "      <td>2024-09-21</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>3.250000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1486463.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.486463e+06</td>\n",
       "      <td>1.8</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.705073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>EM0509</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1992-11-23</td>\n",
       "      <td>2023-08-15</td>\n",
       "      <td>2024-10-02</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>15.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>0.812500</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1214155.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.214155e+06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.733698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>EM0520</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>2000-09-11</td>\n",
       "      <td>2022-09-12</td>\n",
       "      <td>2024-10-13</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.785714</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1098601.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.098601e+06</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.087215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>EM0590</td>\n",
       "      <td>Kota Jakarta Pusat</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1980-06-14</td>\n",
       "      <td>2023-04-13</td>\n",
       "      <td>2024-08-29</td>\n",
       "      <td>Married</td>\n",
       "      <td>0</td>\n",
       "      <td>D3</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.666667</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4646268.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.161567e+06</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.108286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>EM0597</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1980-11-30</td>\n",
       "      <td>2023-05-15</td>\n",
       "      <td>2024-09-21</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>D1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3975285.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.987642e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.580014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>EM0602</td>\n",
       "      <td>Kota Bekasi</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1990-07-28</td>\n",
       "      <td>2023-05-29</td>\n",
       "      <td>2024-08-26</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D2</td>\n",
       "      <td>9.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3496995.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.165665e+06</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.188379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>EM0606</td>\n",
       "      <td>Kota Bogor</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1987-08-01</td>\n",
       "      <td>2023-06-12</td>\n",
       "      <td>2024-08-07</td>\n",
       "      <td>Divorce</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.750000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2928866.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.928866e+06</td>\n",
       "      <td>2.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.357764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>EM0621</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>2000-05-14</td>\n",
       "      <td>2022-07-25</td>\n",
       "      <td>2024-09-07</td>\n",
       "      <td>Married</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.785714</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1374872.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.374872e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.107099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>EM0626</td>\n",
       "      <td>Kota Depok</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1978-08-28</td>\n",
       "      <td>2023-08-14</td>\n",
       "      <td>2024-09-21</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.181818</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2436465.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.436465e+06</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.250325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>EM0638</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1991-06-05</td>\n",
       "      <td>2023-09-11</td>\n",
       "      <td>2024-10-02</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.090909</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1191009.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.191009e+06</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.134709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>EM0640</td>\n",
       "      <td>Tangerang</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1986-05-10</td>\n",
       "      <td>2023-09-25</td>\n",
       "      <td>2024-10-22</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>15.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>0.812500</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1106988.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.106988e+06</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.220359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>EM0722</td>\n",
       "      <td>Kota Jakarta Selatan</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1978-05-29</td>\n",
       "      <td>2022-07-10</td>\n",
       "      <td>2024-10-26</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>D2</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>3.375000</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3502617.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.167539e+06</td>\n",
       "      <td>1.8</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.133624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>EM0726</td>\n",
       "      <td>Kabupaten Bekasi</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1980-09-08</td>\n",
       "      <td>2023-07-20</td>\n",
       "      <td>2024-08-31</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>11.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.083333</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1592248.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.592248e+06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.668318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>EM0728</td>\n",
       "      <td>Kota Bekasi</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1983-04-20</td>\n",
       "      <td>2023-07-25</td>\n",
       "      <td>2024-09-11</td>\n",
       "      <td>Divorce</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1798264.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.798264e+06</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.720000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>EM0730</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1978-08-22</td>\n",
       "      <td>2023-08-03</td>\n",
       "      <td>2024-09-14</td>\n",
       "      <td>Married</td>\n",
       "      <td>3</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1658463.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.658463e+06</td>\n",
       "      <td>1.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.150000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>EM0732</td>\n",
       "      <td>Kota Jakarta Selatan</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1981-03-17</td>\n",
       "      <td>2022-08-04</td>\n",
       "      <td>2024-09-13</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.777778</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1461380.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.461380e+06</td>\n",
       "      <td>2.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.695420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>EM0733</td>\n",
       "      <td>Kota Jakarta Utara</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1975-05-11</td>\n",
       "      <td>2023-08-04</td>\n",
       "      <td>2024-09-18</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.857143</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2041027.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.041027e+06</td>\n",
       "      <td>1.8</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.550811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>EM0736</td>\n",
       "      <td>Kota Depok</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1999-12-24</td>\n",
       "      <td>2022-08-08</td>\n",
       "      <td>2024-09-30</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>5.200000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4568518.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.142130e+06</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.773272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>EM0741</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1997-01-24</td>\n",
       "      <td>2023-08-21</td>\n",
       "      <td>2024-09-23</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.857143</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3317052.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.658526e+06</td>\n",
       "      <td>2.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.371734</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>EM0819</td>\n",
       "      <td>Kota Depok</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1998-10-15</td>\n",
       "      <td>2022-05-09</td>\n",
       "      <td>2024-10-11</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>D1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>4.833333</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3966514.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.983257e+06</td>\n",
       "      <td>2.6</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.342076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>EM0837</td>\n",
       "      <td>Kota Jakarta Selatan</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1991-10-25</td>\n",
       "      <td>2023-07-03</td>\n",
       "      <td>2024-09-15</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>D1</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.750000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3765986.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.882993e+06</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.327930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>EM0845</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1998-10-17</td>\n",
       "      <td>2022-07-24</td>\n",
       "      <td>2024-08-23</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.666667</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1258904.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.258904e+06</td>\n",
       "      <td>2.2</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.798676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>EM0865</td>\n",
       "      <td>Kota Depok</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1996-10-04</td>\n",
       "      <td>2023-09-05</td>\n",
       "      <td>2024-09-27</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1126688.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.126688e+06</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>EM0868</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1977-06-10</td>\n",
       "      <td>2023-09-07</td>\n",
       "      <td>2024-10-26</td>\n",
       "      <td>Married</td>\n",
       "      <td>5</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>1144246.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.144246e+06</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>EM0930</td>\n",
       "      <td>Kota Depok</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1970-08-04</td>\n",
       "      <td>2023-03-08</td>\n",
       "      <td>2024-09-18</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>D3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3918148.0</td>\n",
       "      <td>4</td>\n",
       "      <td>9.795370e+05</td>\n",
       "      <td>2.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.478291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>EM0933</td>\n",
       "      <td>Kota Jakarta Timur</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1981-10-31</td>\n",
       "      <td>2022-03-20</td>\n",
       "      <td>2024-09-08</td>\n",
       "      <td>Married</td>\n",
       "      <td>1</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>3.750000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2490863.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.490863e+06</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.106338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>EM0957</td>\n",
       "      <td>Kota Jakarta Selatan</td>\n",
       "      <td>Perempuan</td>\n",
       "      <td>1998-11-24</td>\n",
       "      <td>2022-07-05</td>\n",
       "      <td>2024-10-31</td>\n",
       "      <td>Married</td>\n",
       "      <td>2</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>10.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>2.545455</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2615137.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.615137e+06</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.342793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>EM0967</td>\n",
       "      <td>Kabupaten Bogor</td>\n",
       "      <td>Laki-laki</td>\n",
       "      <td>1996-02-01</td>\n",
       "      <td>2023-08-07</td>\n",
       "      <td>2024-10-03</td>\n",
       "      <td>Single</td>\n",
       "      <td>0</td>\n",
       "      <td>SLTA</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Mid-term</td>\n",
       "      <td>1.750000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1745824.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.745824e+06</td>\n",
       "      <td>1.4</td>\n",
       "      <td>Medium</td>\n",
       "      <td>9.208596</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>50 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   employee_id              domisili jenis_kelamin date_of_birth   join_date  \\\n",
       "0       EM0012             Tangerang     Laki-laki    1970-12-21  2023-02-23   \n",
       "1       EM0026            Kota Depok     Laki-laki    1986-11-14  2022-04-17   \n",
       "2       EM0041    Kota Jakarta Barat     Laki-laki    1983-03-16  2023-06-15   \n",
       "3       EM0053    Kota Jakarta Timur     Laki-laki    1979-07-13  2023-07-11   \n",
       "4       EM0057    Kota Jakarta Barat     Perempuan    2000-03-13  2022-07-14   \n",
       "5       EM0058             Tangerang     Perempuan    1996-04-23  2023-07-18   \n",
       "6       EM0064    Kota Jakarta Utara     Perempuan    1987-04-20  2022-07-25   \n",
       "7       EM0180    Kota Jakarta Utara     Perempuan    2000-06-25  2022-09-04   \n",
       "8       EM0259      Kepulauan Seribu     Laki-laki    1993-10-04  2023-06-08   \n",
       "9       EM0263       Kabupaten Bogor     Laki-laki    1995-02-15  2022-06-15   \n",
       "10      EM0268    Kota Jakarta Timur     Laki-laki    1984-09-25  2023-07-04   \n",
       "11      EM0274            Kota Bogor     Laki-laki    1995-07-09  2023-07-15   \n",
       "12      EM0360             Tangerang     Perempuan    1979-08-13  2022-04-17   \n",
       "13      EM0368    Kota Jakarta Timur     Laki-laki    1979-12-25  2022-05-25   \n",
       "14      EM0384    Kota Jakarta Timur     Laki-laki    1976-08-11  2022-07-05   \n",
       "15      EM0388    Kota Jakarta Timur     Laki-laki    1970-11-15  2023-07-10   \n",
       "16      EM0398    Kota Jakarta Timur     Laki-laki    1999-05-03  2023-09-01   \n",
       "17      EM0481      Kabupaten Bekasi     Perempuan    1997-12-24  2022-06-08   \n",
       "18      EM0483       Kabupaten Bogor     Perempuan    1975-08-05  2023-06-14   \n",
       "19      EM0491    Kota Jakarta Timur     Perempuan    1969-04-10  2023-06-28   \n",
       "20      EM0493             Tangerang     Perempuan    1996-08-05  2023-07-04   \n",
       "21      EM0499    Kota Jakarta Pusat     Laki-laki    1990-10-23  2022-07-20   \n",
       "22      EM0504      Kabupaten Bekasi     Laki-laki    2000-04-19  2023-08-01   \n",
       "23      EM0509    Kota Jakarta Timur     Laki-laki    1992-11-23  2023-08-15   \n",
       "24      EM0520    Kota Jakarta Timur     Perempuan    2000-09-11  2022-09-12   \n",
       "25      EM0590    Kota Jakarta Pusat     Perempuan    1980-06-14  2023-04-13   \n",
       "26      EM0597      Kabupaten Bekasi     Perempuan    1980-11-30  2023-05-15   \n",
       "27      EM0602           Kota Bekasi     Perempuan    1990-07-28  2023-05-29   \n",
       "28      EM0606            Kota Bogor     Laki-laki    1987-08-01  2023-06-12   \n",
       "29      EM0621      Kabupaten Bekasi     Laki-laki    2000-05-14  2022-07-25   \n",
       "30      EM0626            Kota Depok     Laki-laki    1978-08-28  2023-08-14   \n",
       "31      EM0638       Kabupaten Bogor     Laki-laki    1991-06-05  2023-09-11   \n",
       "32      EM0640             Tangerang     Laki-laki    1986-05-10  2023-09-25   \n",
       "33      EM0722  Kota Jakarta Selatan     Perempuan    1978-05-29  2022-07-10   \n",
       "34      EM0726      Kabupaten Bekasi     Laki-laki    1980-09-08  2023-07-20   \n",
       "35      EM0728           Kota Bekasi     Laki-laki    1983-04-20  2023-07-25   \n",
       "36      EM0730    Kota Jakarta Timur     Perempuan    1978-08-22  2023-08-03   \n",
       "37      EM0732  Kota Jakarta Selatan     Laki-laki    1981-03-17  2022-08-04   \n",
       "38      EM0733    Kota Jakarta Utara     Laki-laki    1975-05-11  2023-08-04   \n",
       "39      EM0736            Kota Depok     Laki-laki    1999-12-24  2022-08-08   \n",
       "40      EM0741    Kota Jakarta Timur     Laki-laki    1997-01-24  2023-08-21   \n",
       "41      EM0819            Kota Depok     Laki-laki    1998-10-15  2022-05-09   \n",
       "42      EM0837  Kota Jakarta Selatan     Perempuan    1991-10-25  2023-07-03   \n",
       "43      EM0845    Kota Jakarta Timur     Laki-laki    1998-10-17  2022-07-24   \n",
       "44      EM0865            Kota Depok     Laki-laki    1996-10-04  2023-09-05   \n",
       "45      EM0868       Kabupaten Bogor     Laki-laki    1977-06-10  2023-09-07   \n",
       "46      EM0930            Kota Depok     Laki-laki    1970-08-04  2023-03-08   \n",
       "47      EM0933    Kota Jakarta Timur     Laki-laki    1981-10-31  2022-03-20   \n",
       "48      EM0957  Kota Jakarta Selatan     Perempuan    1998-11-24  2022-07-05   \n",
       "49      EM0967       Kabupaten Bogor     Laki-laki    1996-02-01  2023-08-07   \n",
       "\n",
       "   resign_date marriage_stat  dependant education  absent_90D  ...  \\\n",
       "0   2024-08-07       Married          3        D3         2.0  ...   \n",
       "1   2024-08-04       Married          2      SLTA         4.0  ...   \n",
       "2   2024-09-06       Divorce          3      SLTA         7.0  ...   \n",
       "3   2024-09-21        Single          0      SLTA         1.0  ...   \n",
       "4   2024-08-29        Single          0        D2         8.0  ...   \n",
       "5   2024-09-26        Single          0      SLTA         9.0  ...   \n",
       "6   2024-08-23       Married          3      SLTA         0.0  ...   \n",
       "7   2024-10-07        Single          0        D2         7.0  ...   \n",
       "8   2024-08-29        Single          0        D1         4.0  ...   \n",
       "9   2024-07-27       Married          1      SLTA        13.0  ...   \n",
       "10  2024-09-21       Married          5      SLTA         4.0  ...   \n",
       "11  2024-10-02       Married          0        D1         5.0  ...   \n",
       "12  2024-09-19       Married          3        D3         7.0  ...   \n",
       "13  2024-08-02       Married          1        D1         8.0  ...   \n",
       "14  2024-09-21       Married          1        D3         4.0  ...   \n",
       "15  2024-09-07       Married          2      SLTA        12.0  ...   \n",
       "16  2024-10-12        Single          0      SLTA         0.0  ...   \n",
       "17  2024-09-27        Single          0      SLTA         2.0  ...   \n",
       "18  2024-07-25       Married          1      SLTA        12.0  ...   \n",
       "19  2024-09-14       Married          5      SLTA         5.0  ...   \n",
       "20  2024-07-25       Married          1      SLTA         7.0  ...   \n",
       "21  2024-09-25       Married          1      SLTA        13.0  ...   \n",
       "22  2024-09-21        Single          0      SLTA         3.0  ...   \n",
       "23  2024-10-02       Married          1      SLTA        15.0  ...   \n",
       "24  2024-10-13        Single          0      SLTA        13.0  ...   \n",
       "25  2024-08-29       Married          0        D3         5.0  ...   \n",
       "26  2024-09-21       Married          3        D1         3.0  ...   \n",
       "27  2024-08-26        Single          0        D2         9.0  ...   \n",
       "28  2024-08-07       Divorce          0      SLTA         7.0  ...   \n",
       "29  2024-09-07       Married          0      SLTA        13.0  ...   \n",
       "30  2024-09-21       Married          2      SLTA        10.0  ...   \n",
       "31  2024-10-02        Single          0      SLTA        10.0  ...   \n",
       "32  2024-10-22       Married          1      SLTA        15.0  ...   \n",
       "33  2024-10-26       Married          3        D2         7.0  ...   \n",
       "34  2024-08-31       Married          2      SLTA        11.0  ...   \n",
       "35  2024-09-11       Divorce          2      SLTA         0.0  ...   \n",
       "36  2024-09-14       Married          3      SLTA         0.0  ...   \n",
       "37  2024-09-13       Married          2      SLTA         8.0  ...   \n",
       "38  2024-09-18       Married          2      SLTA         6.0  ...   \n",
       "39  2024-09-30        Single          0        D3         4.0  ...   \n",
       "40  2024-09-23        Single          0        D1         6.0  ...   \n",
       "41  2024-10-11        Single          0        D1         5.0  ...   \n",
       "42  2024-09-15       Married          1        D1         7.0  ...   \n",
       "43  2024-08-23        Single          0      SLTA        14.0  ...   \n",
       "44  2024-09-27        Single          0      SLTA         0.0  ...   \n",
       "45  2024-10-26       Married          5      SLTA         0.0  ...   \n",
       "46  2024-09-18       Married          2        D3         2.0  ...   \n",
       "47  2024-09-08       Married          1      SLTA         7.0  ...   \n",
       "48  2024-10-31       Married          2      SLTA        10.0  ...   \n",
       "49  2024-10-03        Single          0      SLTA         7.0  ...   \n",
       "\n",
       "    active_work_category work_stability_score married_dependent_ratio  \\\n",
       "0               Mid-term             5.666667                       4   \n",
       "1               Mid-term             5.600000                       3   \n",
       "2               Mid-term             1.750000                       1   \n",
       "3               Mid-term             7.000000                       1   \n",
       "4               Mid-term             2.777778                       1   \n",
       "5               Mid-term             1.400000                       1   \n",
       "6               Mid-term            25.000000                       4   \n",
       "7               Mid-term             3.125000                       1   \n",
       "8               Mid-term             2.800000                       1   \n",
       "9               Mid-term             1.785714                       2   \n",
       "10              Mid-term             2.800000                       6   \n",
       "11              Mid-term             2.333333                       1   \n",
       "12              Mid-term             3.625000                       4   \n",
       "13              Mid-term             2.888889                       2   \n",
       "14              Mid-term             5.200000                       2   \n",
       "15              Mid-term             1.076923                       3   \n",
       "16              Mid-term            13.000000                       1   \n",
       "17              Mid-term             9.333333                       1   \n",
       "18              Mid-term             1.000000                       2   \n",
       "19              Mid-term             2.333333                       6   \n",
       "20              Mid-term             1.500000                       2   \n",
       "21              Mid-term             1.857143                       2   \n",
       "22              Mid-term             3.250000                       1   \n",
       "23              Mid-term             0.812500                       2   \n",
       "24              Mid-term             1.785714                       1   \n",
       "25              Mid-term             2.666667                       1   \n",
       "26              Mid-term             4.000000                       4   \n",
       "27              Mid-term             1.500000                       1   \n",
       "28              Mid-term             1.750000                       1   \n",
       "29              Mid-term             1.785714                       1   \n",
       "30              Mid-term             1.181818                       3   \n",
       "31              Mid-term             1.090909                       1   \n",
       "32              Mid-term             0.812500                       2   \n",
       "33              Mid-term             3.375000                       4   \n",
       "34              Mid-term             1.083333                       3   \n",
       "35              Mid-term            13.000000                       1   \n",
       "36              Mid-term            13.000000                       4   \n",
       "37              Mid-term             2.777778                       3   \n",
       "38              Mid-term             1.857143                       3   \n",
       "39              Mid-term             5.200000                       1   \n",
       "40              Mid-term             1.857143                       1   \n",
       "41              Mid-term             4.833333                       1   \n",
       "42              Mid-term             1.750000                       2   \n",
       "43              Mid-term             1.666667                       1   \n",
       "44              Mid-term            12.000000                       1   \n",
       "45              Mid-term            13.000000                       6   \n",
       "46              Mid-term             6.000000                       3   \n",
       "47              Mid-term             3.750000                       2   \n",
       "48              Mid-term             2.545455                       3   \n",
       "49              Mid-term             1.750000                       1   \n",
       "\n",
       "    position_score  job_income_position_score education_score  \\\n",
       "0                1                  4708861.0               4   \n",
       "1                1                  1430853.0               1   \n",
       "2                1                  1379381.0               1   \n",
       "3                1                  1911583.0               1   \n",
       "4                1                  3724157.0               3   \n",
       "5                1                  2229928.0               1   \n",
       "6                1                  1257855.0               1   \n",
       "7                1                  3034058.0               3   \n",
       "8                1                  4513378.0               2   \n",
       "9                1                  1599099.0               1   \n",
       "10               1                  2869178.0               1   \n",
       "11               1                  3040879.0               2   \n",
       "12               1                  4658718.0               4   \n",
       "13               1                  3326206.0               2   \n",
       "14               1                  3215076.0               4   \n",
       "15               1                  1178459.0               1   \n",
       "16               1                  1527441.0               1   \n",
       "17               1                  2890639.0               1   \n",
       "18               1                  1193560.0               1   \n",
       "19               1                  2048458.0               1   \n",
       "20               1                  1267701.0               1   \n",
       "21               1                  1544522.0               1   \n",
       "22               1                  1486463.0               1   \n",
       "23               1                  1214155.0               1   \n",
       "24               1                  1098601.0               1   \n",
       "25               1                  4646268.0               4   \n",
       "26               1                  3975285.0               2   \n",
       "27               1                  3496995.0               3   \n",
       "28               1                  2928866.0               1   \n",
       "29               1                  1374872.0               1   \n",
       "30               1                  2436465.0               1   \n",
       "31               1                  1191009.0               1   \n",
       "32               1                  1106988.0               1   \n",
       "33               1                  3502617.0               3   \n",
       "34               1                  1592248.0               1   \n",
       "35               1                  1798264.0               1   \n",
       "36               1                  1658463.0               1   \n",
       "37               1                  1461380.0               1   \n",
       "38               1                  2041027.0               1   \n",
       "39               1                  4568518.0               4   \n",
       "40               1                  3317052.0               2   \n",
       "41               1                  3966514.0               2   \n",
       "42               1                  3765986.0               2   \n",
       "43               1                  1258904.0               1   \n",
       "44               1                  1126688.0               1   \n",
       "45               1                  1144246.0               1   \n",
       "46               1                  3918148.0               4   \n",
       "47               1                  2490863.0               1   \n",
       "48               1                  2615137.0               1   \n",
       "49               1                  1745824.0               1   \n",
       "\n",
       "   education_income_ratio  weighted_satisfaction_performance  \\\n",
       "0            1.177215e+06                                1.4   \n",
       "1            1.430853e+06                                1.0   \n",
       "2            1.379381e+06                                2.4   \n",
       "3            1.911583e+06                                1.0   \n",
       "4            1.241386e+06                                2.0   \n",
       "5            2.229928e+06                                1.4   \n",
       "6            1.257855e+06                                2.2   \n",
       "7            1.011353e+06                                2.2   \n",
       "8            2.256689e+06                                2.6   \n",
       "9            1.599099e+06                                1.6   \n",
       "10           2.869178e+06                                1.0   \n",
       "11           1.520440e+06                                1.6   \n",
       "12           1.164680e+06                                3.0   \n",
       "13           1.663103e+06                                2.6   \n",
       "14           8.037690e+05                                2.0   \n",
       "15           1.178459e+06                                1.6   \n",
       "16           1.527441e+06                                3.0   \n",
       "17           2.890639e+06                                1.6   \n",
       "18           1.193560e+06                                2.6   \n",
       "19           2.048458e+06                                1.6   \n",
       "20           1.267701e+06                                1.4   \n",
       "21           1.544522e+06                                2.6   \n",
       "22           1.486463e+06                                1.8   \n",
       "23           1.214155e+06                                1.0   \n",
       "24           1.098601e+06                                2.6   \n",
       "25           1.161567e+06                                1.4   \n",
       "26           1.987642e+06                                1.6   \n",
       "27           1.165665e+06                                2.2   \n",
       "28           2.928866e+06                                2.4   \n",
       "29           1.374872e+06                                1.6   \n",
       "30           2.436465e+06                                3.0   \n",
       "31           1.191009e+06                                2.2   \n",
       "32           1.106988e+06                                2.2   \n",
       "33           1.167539e+06                                1.8   \n",
       "34           1.592248e+06                                1.0   \n",
       "35           1.798264e+06                                2.0   \n",
       "36           1.658463e+06                                1.6   \n",
       "37           1.461380e+06                                2.4   \n",
       "38           2.041027e+06                                1.8   \n",
       "39           1.142130e+06                                1.0   \n",
       "40           1.658526e+06                                2.4   \n",
       "41           1.983257e+06                                2.6   \n",
       "42           1.882993e+06                                3.0   \n",
       "43           1.258904e+06                                2.2   \n",
       "44           1.126688e+06                                1.4   \n",
       "45           1.144246e+06                                1.4   \n",
       "46           9.795370e+05                                2.4   \n",
       "47           2.490863e+06                                2.0   \n",
       "48           2.615137e+06                                2.0   \n",
       "49           1.745824e+06                                1.4   \n",
       "\n",
       "    resign_risk_indicator  adjusted_work_time  \n",
       "0                  Medium            9.857106  \n",
       "1                  Medium            9.694593  \n",
       "2                  Medium            9.059429  \n",
       "3                  Medium            9.842189  \n",
       "4                  Medium            9.047730  \n",
       "5                  Medium            9.114481  \n",
       "6                  Medium            9.320000  \n",
       "7                  Medium            9.091639  \n",
       "8                  Medium            9.479833  \n",
       "9                  Medium            9.077272  \n",
       "10                 Medium            9.599453  \n",
       "11                 Medium            9.781063  \n",
       "12                 Medium            9.494477  \n",
       "13                 Medium            9.786442  \n",
       "14                 Medium            9.773272  \n",
       "15                 Medium            9.072831  \n",
       "16                 Medium            9.390000  \n",
       "17                 Medium            9.562408  \n",
       "18                 Medium            9.313570  \n",
       "19                 Medium            9.372688  \n",
       "20                 Medium            9.190102  \n",
       "21                 Medium            9.825135  \n",
       "22                 Medium            9.705073  \n",
       "23                 Medium            9.733698  \n",
       "24                 Medium            9.087215  \n",
       "25                 Medium            9.108286  \n",
       "26                 Medium            9.580014  \n",
       "27                 Medium            9.188379  \n",
       "28                 Medium            9.357764  \n",
       "29                 Medium            9.107099  \n",
       "30                 Medium            9.250325  \n",
       "31                 Medium            9.134709  \n",
       "32                 Medium            9.220359  \n",
       "33                 Medium            9.133624  \n",
       "34                 Medium            9.668318  \n",
       "35                 Medium            9.720000  \n",
       "36                 Medium            9.150000  \n",
       "37                 Medium            9.695420  \n",
       "38                 Medium            9.550811  \n",
       "39                 Medium            9.773272  \n",
       "40                 Medium            9.371734  \n",
       "41                 Medium            9.342076  \n",
       "42                 Medium            9.327930  \n",
       "43                 Medium            9.798676  \n",
       "44                 Medium            9.400000  \n",
       "45                 Medium            9.140000  \n",
       "46                 Medium            9.478291  \n",
       "47                 Medium            9.106338  \n",
       "48                 Medium            9.342793  \n",
       "49                 Medium            9.208596  \n",
       "\n",
       "[50 rows x 37 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df_test = pd.read_csv('D:\\Tugas Akhir\\Codingan\\Development\\Data\\data_testing_resign_6.csv')\n",
    "df_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "employee_id                          0\n",
       "domisili                             0\n",
       "jenis_kelamin                        0\n",
       "date_of_birth                        0\n",
       "join_date                            0\n",
       "resign_date                          0\n",
       "marriage_stat                        0\n",
       "dependant                            0\n",
       "education                            0\n",
       "absent_90D                           0\n",
       "avg_time_work                        0\n",
       "departemen                           0\n",
       "position                             0\n",
       "income                               0\n",
       "total_komp                           0\n",
       "job_satisfaction                     0\n",
       "performance_rating                   0\n",
       "churn_status                         0\n",
       "age_years                            0\n",
       "active_work                          0\n",
       "active_work_months                   0\n",
       "income_3_months                      0\n",
       "income_6_months                      0\n",
       "total_income_work                    0\n",
       "absence_ratio                        0\n",
       "income_dependant_ratio               0\n",
       "work_efficiency                      0\n",
       "active_work_category                 0\n",
       "work_stability_score                 0\n",
       "married_dependent_ratio              0\n",
       "position_score                       0\n",
       "job_income_position_score            0\n",
       "education_score                      0\n",
       "education_income_ratio               0\n",
       "weighted_satisfaction_performance    0\n",
       "resign_risk_indicator                0\n",
       "adjusted_work_time                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "final_model = pickle.load(open('clasification_model.sav', 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['departemen', 'position', 'income', 'domisili', 'marriage_stat', 'dependant', 'education', 'absent_90D', 'avg_time_work', 'total_komp', 'job_satisfaction', 'performance_rating', 'age_years', 'active_work', 'active_work_months', 'income_3_months', 'income_6_months', 'total_income_work', 'absence_ratio', 'income_dependant_ratio', 'work_efficiency', 'active_work_category', 'work_stability_score', 'married_dependent_ratio', 'position_score', 'job_income_position_score', 'education_score', 'education_income_ratio', 'weighted_satisfaction_performance', 'resign_risk_indicator', 'adjusted_work_time']\n"
     ]
    }
   ],
   "source": [
    "expected_columns = final_model.feature_names_\n",
    "print(expected_columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 1.0\n",
      "Precision: 1.0\n",
      "Recall: 1.0\n",
      "F1 Score: 1.0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
    "from catboost import Pool\n",
    "\n",
    "# Drop kolom yang tidak relevan\n",
    "X_test = df_test.drop(['churn_status', 'employee_id', 'date_of_birth', 'join_date', 'resign_date', 'active_work_months'], axis=1)\n",
    "\n",
    "# Konversi semua kolom kategori ke string\n",
    "cat_feature = ['departemen', 'position', 'domisili', 'marriage_stat', 'job_satisfaction', \n",
    "               'performance_rating', 'education', 'active_work_category', 'resign_risk_indicator', 'jenis_kelamin']\n",
    "\n",
    "# Pastikan semua fitur kategori adalah string\n",
    "for col in cat_feature:\n",
    "    if col in X_test.columns:\n",
    "        X_test[col] = X_test[col].astype(str)\n",
    "\n",
    "# Buat Pool untuk data uji\n",
    "test_pool = Pool(data=X_test, cat_features=cat_feature)\n",
    "\n",
    "# Prediksi dengan model menggunakan Pool\n",
    "y_pred = final_model.predict(test_pool)\n",
    "\n",
    "# Evaluasi\n",
    "accuracy = accuracy_score(df_test['churn_status'], y_pred)\n",
    "precision = precision_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "recall = recall_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "f1 = f1_score(df_test['churn_status'], y_pred, zero_division=0)\n",
    "\n",
    "print(\"Accuracy:\", accuracy)\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1 Score:\", f1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>income</th>\n",
       "      <th>dependant</th>\n",
       "      <th>absent_90D</th>\n",
       "      <th>avg_time_work</th>\n",
       "      <th>total_komp</th>\n",
       "      <th>churn_status</th>\n",
       "      <th>age_years</th>\n",
       "      <th>active_work</th>\n",
       "      <th>active_work_months</th>\n",
       "      <th>income_3_months</th>\n",
       "      <th>...</th>\n",
       "      <th>income_dependant_ratio</th>\n",
       "      <th>work_efficiency</th>\n",
       "      <th>work_stability_score</th>\n",
       "      <th>married_dependent_ratio</th>\n",
       "      <th>position_score</th>\n",
       "      <th>job_income_position_score</th>\n",
       "      <th>education_score</th>\n",
       "      <th>education_income_ratio</th>\n",
       "      <th>weighted_satisfaction_performance</th>\n",
       "      <th>adjusted_work_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>8.120000e+02</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.0</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.0</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>8.120000e+02</td>\n",
       "      <td>...</td>\n",
       "      <td>8.120000e+02</td>\n",
       "      <td>812.000</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>8.120000e+02</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>8.120000e+02</td>\n",
       "      <td>812.000000</td>\n",
       "      <td>812.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2.704077e+06</td>\n",
       "      <td>1.443350</td>\n",
       "      <td>5.703202</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.116995</td>\n",
       "      <td>1.0</td>\n",
       "      <td>37.488916</td>\n",
       "      <td>601.076355</td>\n",
       "      <td>19.556650</td>\n",
       "      <td>8.112230e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>1.494665e+06</td>\n",
       "      <td>1.125</td>\n",
       "      <td>5.940580</td>\n",
       "      <td>2.387931</td>\n",
       "      <td>1.008621</td>\n",
       "      <td>2.676485e+06</td>\n",
       "      <td>1.838670</td>\n",
       "      <td>1.650454e+06</td>\n",
       "      <td>2.018966</td>\n",
       "      <td>8.967059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.217016e+06</td>\n",
       "      <td>1.272101</td>\n",
       "      <td>4.420202</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.543866</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.914092</td>\n",
       "      <td>195.811635</td>\n",
       "      <td>6.545148</td>\n",
       "      <td>3.651049e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>1.103701e+06</td>\n",
       "      <td>0.000</td>\n",
       "      <td>6.556615</td>\n",
       "      <td>1.283876</td>\n",
       "      <td>0.092504</td>\n",
       "      <td>1.165877e+06</td>\n",
       "      <td>1.136691</td>\n",
       "      <td>5.549310e+05</td>\n",
       "      <td>0.580476</td>\n",
       "      <td>0.029627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.015570e+06</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>365.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>3.046710e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>1.907077e+05</td>\n",
       "      <td>1.125</td>\n",
       "      <td>0.705882</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.015570e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.524518e+05</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>8.866790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.570747e+06</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>428.000000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>4.712242e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>6.466449e+05</td>\n",
       "      <td>1.125</td>\n",
       "      <td>1.854396</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.570747e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.200502e+06</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>8.950040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2.561418e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>496.500000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>7.684256e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>1.168337e+06</td>\n",
       "      <td>1.125</td>\n",
       "      <td>3.200000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.561418e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.548386e+06</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>8.973353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3.692924e+06</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>44.000000</td>\n",
       "      <td>798.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>1.107877e+07</td>\n",
       "      <td>...</td>\n",
       "      <td>1.879129e+06</td>\n",
       "      <td>1.125</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.679853e+06</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.048790e+06</td>\n",
       "      <td>2.400000</td>\n",
       "      <td>8.991674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.855813e+06</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>57.000000</td>\n",
       "      <td>1095.000000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>2.356744e+07</td>\n",
       "      <td>...</td>\n",
       "      <td>7.855813e+06</td>\n",
       "      <td>1.125</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>4.982560e+06</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>2.996378e+06</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             income   dependant  absent_90D  avg_time_work  total_komp  \\\n",
       "count  8.120000e+02  812.000000  812.000000          812.0  812.000000   \n",
       "mean   2.704077e+06    1.443350    5.703202            9.0    0.116995   \n",
       "std    1.217016e+06    1.272101    4.420202            0.0    0.543866   \n",
       "min    1.015570e+06    0.000000    0.000000            9.0    0.000000   \n",
       "25%    1.570747e+06    0.000000    2.000000            9.0    0.000000   \n",
       "50%    2.561418e+06    1.000000    5.000000            9.0    0.000000   \n",
       "75%    3.692924e+06    2.000000    9.000000            9.0    0.000000   \n",
       "max    7.855813e+06    6.000000   16.000000            9.0   12.000000   \n",
       "\n",
       "       churn_status   age_years  active_work  active_work_months  \\\n",
       "count         812.0  812.000000   812.000000          812.000000   \n",
       "mean            1.0   37.488916   601.076355           19.556650   \n",
       "std             0.0    8.914092   195.811635            6.545148   \n",
       "min             1.0    1.000000   365.000000           12.000000   \n",
       "25%             1.0   30.000000   428.000000           14.000000   \n",
       "50%             1.0   37.000000   496.500000           16.000000   \n",
       "75%             1.0   44.000000   798.000000           26.000000   \n",
       "max             1.0   57.000000  1095.000000           36.000000   \n",
       "\n",
       "       income_3_months  ...  income_dependant_ratio  work_efficiency  \\\n",
       "count     8.120000e+02  ...            8.120000e+02          812.000   \n",
       "mean      8.112230e+06  ...            1.494665e+06            1.125   \n",
       "std       3.651049e+06  ...            1.103701e+06            0.000   \n",
       "min       3.046710e+06  ...            1.907077e+05            1.125   \n",
       "25%       4.712242e+06  ...            6.466449e+05            1.125   \n",
       "50%       7.684256e+06  ...            1.168337e+06            1.125   \n",
       "75%       1.107877e+07  ...            1.879129e+06            1.125   \n",
       "max       2.356744e+07  ...            7.855813e+06            1.125   \n",
       "\n",
       "       work_stability_score  married_dependent_ratio  position_score  \\\n",
       "count            812.000000               812.000000      812.000000   \n",
       "mean               5.940580                 2.387931        1.008621   \n",
       "std                6.556615                 1.283876        0.092504   \n",
       "min                0.705882                 1.000000        1.000000   \n",
       "25%                1.854396                 1.000000        1.000000   \n",
       "50%                3.200000                 2.000000        1.000000   \n",
       "75%                7.000000                 3.000000        1.000000   \n",
       "max               33.000000                 7.000000        2.000000   \n",
       "\n",
       "       job_income_position_score  education_score  education_income_ratio  \\\n",
       "count               8.120000e+02       812.000000            8.120000e+02   \n",
       "mean                2.676485e+06         1.838670            1.650454e+06   \n",
       "std                 1.165877e+06         1.136691            5.549310e+05   \n",
       "min                 1.015570e+06         1.000000            7.524518e+05   \n",
       "25%                 1.570747e+06         1.000000            1.200502e+06   \n",
       "50%                 2.561418e+06         1.000000            1.548386e+06   \n",
       "75%                 3.679853e+06         3.000000            2.048790e+06   \n",
       "max                 4.982560e+06         5.000000            2.996378e+06   \n",
       "\n",
       "       weighted_satisfaction_performance  adjusted_work_time  \n",
       "count                         812.000000          812.000000  \n",
       "mean                            2.018966            8.967059  \n",
       "std                             0.580476            0.029627  \n",
       "min                             1.000000            8.866790  \n",
       "25%                             1.600000            8.950040  \n",
       "50%                             2.000000            8.973353  \n",
       "75%                             2.400000            8.991674  \n",
       "max                             3.000000            9.000000  \n",
       "\n",
       "[8 rows x 23 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>Feature</th>\n",
       "      <th>Importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>active_work_category</td>\n",
       "      <td>54.255294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>active_work</td>\n",
       "      <td>7.777737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>position</td>\n",
       "      <td>7.155448</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>active_work_months</td>\n",
       "      <td>4.230350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>position_score</td>\n",
       "      <td>3.667581</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>education_score</td>\n",
       "      <td>3.106486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>income_6_months</td>\n",
       "      <td>3.089712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>education</td>\n",
       "      <td>2.826131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>income</td>\n",
       "      <td>2.385703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>total_income_work</td>\n",
       "      <td>2.048091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>income_3_months</td>\n",
       "      <td>1.777421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>resign_risk_indicator</td>\n",
       "      <td>1.190863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>job_income_position_score</td>\n",
       "      <td>1.018409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>age_years</td>\n",
       "      <td>0.762253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>income_dependant_ratio</td>\n",
       "      <td>0.684176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>marriage_stat</td>\n",
       "      <td>0.508788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>domisili</td>\n",
       "      <td>0.435528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>married_dependent_ratio</td>\n",
       "      <td>0.357841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>education_income_ratio</td>\n",
       "      <td>0.335609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>work_stability_score</td>\n",
       "      <td>0.309915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>adjusted_work_time</td>\n",
       "      <td>0.308920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>weighted_satisfaction_performance</td>\n",
       "      <td>0.283635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>absence_ratio</td>\n",
       "      <td>0.282780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>departemen</td>\n",
       "      <td>0.259452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>dependant</td>\n",
       "      <td>0.241856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>absent_90D</td>\n",
       "      <td>0.223434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>performance_rating</td>\n",
       "      <td>0.206559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>job_satisfaction</td>\n",
       "      <td>0.148087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>total_komp</td>\n",
       "      <td>0.119943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>avg_time_work</td>\n",
       "      <td>0.001582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>work_efficiency</td>\n",
       "      <td>0.000416</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              Feature  Importance\n",
       "21               active_work_category   54.255294\n",
       "13                        active_work    7.777737\n",
       "1                            position    7.155448\n",
       "14                 active_work_months    4.230350\n",
       "24                     position_score    3.667581\n",
       "26                    education_score    3.106486\n",
       "16                    income_6_months    3.089712\n",
       "6                           education    2.826131\n",
       "2                              income    2.385703\n",
       "17                  total_income_work    2.048091\n",
       "15                    income_3_months    1.777421\n",
       "29              resign_risk_indicator    1.190863\n",
       "25          job_income_position_score    1.018409\n",
       "12                          age_years    0.762253\n",
       "19             income_dependant_ratio    0.684176\n",
       "4                       marriage_stat    0.508788\n",
       "3                            domisili    0.435528\n",
       "23            married_dependent_ratio    0.357841\n",
       "27             education_income_ratio    0.335609\n",
       "22               work_stability_score    0.309915\n",
       "30                 adjusted_work_time    0.308920\n",
       "28  weighted_satisfaction_performance    0.283635\n",
       "18                      absence_ratio    0.282780\n",
       "0                          departemen    0.259452\n",
       "5                           dependant    0.241856\n",
       "7                          absent_90D    0.223434\n",
       "11                 performance_rating    0.206559\n",
       "10                   job_satisfaction    0.148087\n",
       "9                          total_komp    0.119943\n",
       "8                       avg_time_work    0.001582\n",
       "20                    work_efficiency    0.000416"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_names = X_train.columns.tolist()\n",
    "feature_importance = model.get_feature_importance()\n",
    "\n",
    "feature_importance_df = pd.DataFrame({\n",
    "    'Feature': feature_names,\n",
    "    'Importance': feature_importance\n",
    "}).sort_values(by='Importance', ascending=False)\n",
    "feature_importance_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CatBoost Classification model saved to 'clasification_model.sav'\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "\n",
    "with open('clasification_model.sav', 'wb') as f:\n",
    "    pickle.dump(final_model, f)\n",
    "print(\"CatBoost Classification model saved to 'clasification_model.sav'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting streamlit-option-menu\n",
      "  Using cached streamlit_option_menu-0.4.0-py3-none-any.whl.metadata (2.5 kB)\n",
      "Collecting streamlit>=1.36 (from streamlit-option-menu)\n",
      "  Downloading streamlit-1.41.1-py2.py3-none-any.whl.metadata (8.5 kB)\n",
      "Requirement already satisfied: altair<6,>=4.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (5.2.0)\n",
      "Requirement already satisfied: blinker<2,>=1.0.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (1.7.0)\n",
      "Requirement already satisfied: cachetools<6,>=4.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (4.2.2)\n",
      "Requirement already satisfied: click<9,>=7.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (7.1.2)\n",
      "Collecting numpy<3,>=1.23 (from streamlit>=1.36->streamlit-option-menu)\n",
      "  Downloading numpy-2.0.2-cp39-cp39-win_amd64.whl.metadata (59 kB)\n",
      "Requirement already satisfied: packaging<25,>=20 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (23.1)\n",
      "Requirement already satisfied: pandas<3,>=1.4.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (1.4.2)\n",
      "Requirement already satisfied: pillow<12,>=7.1.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (10.0.1)\n",
      "Requirement already satisfied: protobuf<6,>=3.20 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (3.20.3)\n",
      "Requirement already satisfied: pyarrow>=7.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (14.0.2)\n",
      "Requirement already satisfied: requests<3,>=2.27 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (2.31.0)\n",
      "Requirement already satisfied: rich<14,>=10.14.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (13.7.0)\n",
      "Requirement already satisfied: tenacity<10,>=8.1.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (8.2.2)\n",
      "Requirement already satisfied: toml<2,>=0.10.1 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (0.10.2)\n",
      "Requirement already satisfied: typing-extensions<5,>=4.3.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (4.7.1)\n",
      "Requirement already satisfied: watchdog<7,>=2.1.5 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (2.1.6)\n",
      "Requirement already satisfied: gitpython!=3.1.19,<4,>=3.0.7 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (3.1.41)\n",
      "Requirement already satisfied: pydeck<1,>=0.8.0b4 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (0.8.1b0)\n",
      "Requirement already satisfied: tornado<7,>=6.0.3 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from streamlit>=1.36->streamlit-option-menu) (6.3.3)\n",
      "Requirement already satisfied: jinja2 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (3.1.2)\n",
      "Requirement already satisfied: jsonschema>=3.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (3.2.0)\n",
      "Requirement already satisfied: toolz in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (0.12.0)\n",
      "Requirement already satisfied: gitdb<5,>=4.0.1 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from gitpython!=3.1.19,<4,>=3.0.7->streamlit>=1.36->streamlit-option-menu) (4.0.11)\n",
      "Requirement already satisfied: python-dateutil>=2.8.1 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from pandas<3,>=1.4.0->streamlit>=1.36->streamlit-option-menu) (2.8.2)\n",
      "Requirement already satisfied: pytz>=2020.1 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from pandas<3,>=1.4.0->streamlit>=1.36->streamlit-option-menu) (2023.3.post1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from requests<3,>=2.27->streamlit>=1.36->streamlit-option-menu) (2.0.4)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from requests<3,>=2.27->streamlit>=1.36->streamlit-option-menu) (2.10)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from requests<3,>=2.27->streamlit>=1.36->streamlit-option-menu) (1.26.16)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from requests<3,>=2.27->streamlit>=1.36->streamlit-option-menu) (2024.6.2)\n",
      "Requirement already satisfied: markdown-it-py>=2.2.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from rich<14,>=10.14.0->streamlit>=1.36->streamlit-option-menu) (3.0.0)\n",
      "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from rich<14,>=10.14.0->streamlit>=1.36->streamlit-option-menu) (2.15.1)\n",
      "Requirement already satisfied: smmap<6,>=3.0.1 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from gitdb<5,>=4.0.1->gitpython!=3.1.19,<4,>=3.0.7->streamlit>=1.36->streamlit-option-menu) (5.0.1)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from jinja2->altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (2.1.1)\n",
      "Requirement already satisfied: attrs>=17.4.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from jsonschema>=3.0->altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (23.1.0)\n",
      "Requirement already satisfied: pyrsistent>=0.14.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from jsonschema>=3.0->altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (0.18.0)\n",
      "Requirement already satisfied: setuptools in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from jsonschema>=3.0->altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (68.0.0)\n",
      "Requirement already satisfied: six>=1.11.0 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from jsonschema>=3.0->altair<6,>=4.0->streamlit>=1.36->streamlit-option-menu) (1.16.0)\n",
      "Requirement already satisfied: mdurl~=0.1 in c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages (from markdown-it-py>=2.2.0->rich<14,>=10.14.0->streamlit>=1.36->streamlit-option-menu) (0.1.2)\n",
      "Downloading streamlit_option_menu-0.4.0-py3-none-any.whl (829 kB)\n",
      "   ---------------------------------------- 0.0/829.3 kB ? eta -:--:--\n",
      "   ------------------------- -------------- 524.3/829.3 kB 4.2 MB/s eta 0:00:01\n",
      "   ---------------------------------------- 829.3/829.3 kB 3.3 MB/s eta 0:00:00\n",
      "Downloading streamlit-1.41.1-py2.py3-none-any.whl (9.1 MB)\n",
      "   ---------------------------------------- 0.0/9.1 MB ? eta -:--:--\n",
      "   --- ------------------------------------ 0.8/9.1 MB 4.8 MB/s eta 0:00:02\n",
      "   ----- ---------------------------------- 1.3/9.1 MB 3.5 MB/s eta 0:00:03\n",
      "   -------- ------------------------------- 1.8/9.1 MB 4.0 MB/s eta 0:00:02\n",
      "   ------------- -------------------------- 3.1/9.1 MB 3.9 MB/s eta 0:00:02\n",
      "   ----------------- ---------------------- 3.9/9.1 MB 4.1 MB/s eta 0:00:02\n",
      "   -------------------- ------------------- 4.7/9.1 MB 3.8 MB/s eta 0:00:02\n",
      "   ----------------------- ---------------- 5.2/9.1 MB 3.7 MB/s eta 0:00:02\n",
      "   -------------------------- ------------- 6.0/9.1 MB 3.7 MB/s eta 0:00:01\n",
      "   ---------------------------- ----------- 6.6/9.1 MB 3.5 MB/s eta 0:00:01\n",
      "   --------------------------------- ------ 7.6/9.1 MB 3.6 MB/s eta 0:00:01\n",
      "   -------------------------------------- - 8.7/9.1 MB 3.7 MB/s eta 0:00:01\n",
      "   ---------------------------------------- 9.1/9.1 MB 3.7 MB/s eta 0:00:00\n",
      "Downloading numpy-2.0.2-cp39-cp39-win_amd64.whl (15.9 MB)\n",
      "   ---------------------------------------- 0.0/15.9 MB ? eta -:--:--\n",
      "   - -------------------------------------- 0.5/15.9 MB 3.3 MB/s eta 0:00:05\n",
      "   ---- ----------------------------------- 1.8/15.9 MB 5.0 MB/s eta 0:00:03\n",
      "   ------- -------------------------------- 3.1/15.9 MB 5.4 MB/s eta 0:00:03\n",
      "   ---------- ----------------------------- 4.2/15.9 MB 5.5 MB/s eta 0:00:03\n",
      "   ------------- -------------------------- 5.5/15.9 MB 5.6 MB/s eta 0:00:02\n",
      "   ---------------- ----------------------- 6.6/15.9 MB 5.6 MB/s eta 0:00:02\n",
      "   ------------------- -------------------- 7.9/15.9 MB 5.5 MB/s eta 0:00:02\n",
      "   --------------------- ------------------ 8.7/15.9 MB 5.3 MB/s eta 0:00:02\n",
      "   ------------------------ --------------- 9.7/15.9 MB 5.2 MB/s eta 0:00:02\n",
      "   --------------------------- ------------ 11.0/15.9 MB 5.3 MB/s eta 0:00:01\n",
      "   ----------------------------- ---------- 11.8/15.9 MB 5.1 MB/s eta 0:00:01\n",
      "   ------------------------------ --------- 12.3/15.9 MB 4.9 MB/s eta 0:00:01\n",
      "   --------------------------------- ------ 13.4/15.9 MB 4.9 MB/s eta 0:00:01\n",
      "   ---------------------------------- ----- 13.9/15.9 MB 4.8 MB/s eta 0:00:01\n",
      "   -------------------------------------- - 15.2/15.9 MB 4.8 MB/s eta 0:00:01\n",
      "   ---------------------------------------- 15.9/15.9 MB 4.7 MB/s eta 0:00:00\n",
      "Installing collected packages: numpy, streamlit, streamlit-option-menu\n",
      "  Attempting uninstall: numpy\n",
      "    Found existing installation: numpy 1.22.4\n",
      "    Uninstalling numpy-1.22.4:\n",
      "      Successfully uninstalled numpy-1.22.4\n",
      "  Attempting uninstall: streamlit\n",
      "    Found existing installation: streamlit 1.31.0\n",
      "    Uninstalling streamlit-1.31.0:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Ignoring invalid distribution -pencv-python (c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages)\n",
      "WARNING: Ignoring invalid distribution -treamlit (c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages)\n",
      "WARNING: Ignoring invalid distribution -pencv-python (c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages)\n",
      "WARNING: Ignoring invalid distribution -treamlit (c:\\users\\jesselyn mu\\anaconda3\\lib\\site-packages)\n",
      "  WARNING: Failed to remove contents in a temporary directory 'C:\\Users\\Jesselyn Mu\\anaconda3\\Lib\\site-packages\\~umpy'.\n",
      "  You can safely remove it manually.\n",
      "ERROR: Could not install packages due to an OSError: [WinError 32] The process cannot access the file because it is being used by another process: 'c:\\\\users\\\\jesselyn mu\\\\anaconda3\\\\scripts\\\\streamlit.exe'\n",
      "Consider using the `--user` option or check the permissions.\n",
      "\n",
      "\n",
      "[notice] A new release of pip is available: 23.2.1 -> 24.3.1\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    }
   ],
   "source": [
    "%pip install streamlit-option-menu"
   ]
  }
 ],
 "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": "3.10.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}