boosting_with_reg.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "os.chdir('app/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from tqdm import tqdm\n",
    "\n",
    "from sklearn.metrics import roc_auc_score, f1_score, mean_squared_error, roc_curve\n",
    "from sklearn.model_selection import cross_val_score, train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.linear_model import LogisticRegression, LinearRegression\n",
    "from lightgbm import LGBMClassifier, LGBMRegressor\n",
    "from xgboost import XGBClassifier\n",
    "from catboost import CatBoostClassifier\n",
    "\n",
    "from sklearn.ensemble import StackingClassifier\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "from imblearn.over_sampling import RandomOverSampler, SMOTE"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "path_to_data = 'data'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = pd.read_parquet(path_to_data + '/train_dataset_hackaton2023_train.gzip')\n",
    "df_test = pd.read_parquet(path_to_data + '/hackaton2023_test.gzip')\n",
    "\n",
    "df_test['date_diff_post'] = 0\n",
    "df_test['buy_post'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def has_ws(x):\n",
    "    return x in ['Отдельно стоящий с внешней зоной', 'Отдельно стоящий без внешней зоны', 'Отдельно стоящий', 'Фудкорт с туалетом']\n",
    "\n",
    "def is_food_court(x):\n",
    "    return 'Фудкорт' in x\n",
    "\n",
    "\n",
    "def preprocess(df):\n",
    "    # df = df.drop_duplicates()\n",
    "\n",
    "    # some new features\n",
    "    df['discounts'] = df['revenue'] == 1\n",
    "    df['has_ws'] = df['format_name'].apply(has_ws)\n",
    "    df['is_food_court'] = df['format_name'].apply(is_food_court)\n",
    "\n",
    "    # aggregate each check\n",
    "    data_check_agg = df.groupby(['customer_id', 'startdatetime']).agg({\n",
    "        'revenue': ['sum', 'count'], \n",
    "        'discounts': 'sum',\n",
    "        'has_ws': 'first',\n",
    "        'is_food_court': 'first',\n",
    "        'buy_post': 'first',\n",
    "        'date_diff_post': 'first',\n",
    "    }).reset_index()\n",
    "\n",
    "    data_check_agg.columns = [f'{header}_{stat}' if header == 'revenue' else header for header, stat in data_check_agg.columns]\n",
    "    data_check_agg = data_check_agg.rename({'revenue_sum': 'revenue', 'revenue_count': 'positions_count'}, axis=1)\n",
    "\n",
    "    # calculate time delta between checks\n",
    "    data_check_agg['delta'] = data_check_agg.groupby('customer_id')['startdatetime'].diff() \n",
    "\n",
    "    data_check_agg['delta_hours'] = (data_check_agg['delta'].dt.total_seconds() // (60*60)).fillna(0)\n",
    "    data_check_agg['delta_days'] = (data_check_agg['delta'].dt.total_seconds() // (24*60*60)).fillna(0)\n",
    "\n",
    "    # add date features\n",
    "    data_check_agg = data_check_agg.drop('delta', axis=1)\n",
    "    data_check_agg['month'] = data_check_agg['startdatetime'].dt.month\n",
    "    data_check_agg['day_of_week'] = data_check_agg['startdatetime'].dt.day_of_week\n",
    "    data_check_agg['hour'] = data_check_agg['startdatetime'].dt.hour\n",
    "\n",
    "    # aggregate features for each user\n",
    "    data_cust_agg = data_check_agg.groupby('customer_id').agg({\n",
    "        'revenue': ['mean', 'median', 'std', 'max', 'min', 'count'],\n",
    "        'positions_count': ['mean', 'median', 'std', 'max', 'min'],\n",
    "        'discounts': ['mean', 'median', 'std', 'max', 'min'],\n",
    "        'delta_hours': ['mean', 'median', 'std', 'max'],\n",
    "        'delta_days': ['mean', 'median', 'std', 'max'],\n",
    "        'hour': ['mean', 'median', 'std', 'max', 'min'],\n",
    "        'startdatetime': ['min', 'max'],\n",
    "        'has_ws': 'mean',\n",
    "        'is_food_court': 'mean',\n",
    "        'buy_post': 'first',\n",
    "        'date_diff_post': 'first',\n",
    "    }).reset_index()\n",
    "\n",
    "    data_cust_agg.columns = [f'{header}_{stat}' for header, stat in data_cust_agg.columns]\n",
    "\n",
    "    # add some features\n",
    "    data_cust_agg['recency'] = (data_cust_agg['startdatetime_max'] - data_cust_agg['startdatetime_min']).dt.days\n",
    "    data_cust_agg['T'] = (pd.to_datetime('2023-08-02') - data_cust_agg['startdatetime_min']).dt.days\n",
    "    data_cust_agg['days_from_last_purchase'] = (pd.to_datetime('2023-08-02') - data_cust_agg['startdatetime_max']).dt.days\n",
    "    data_cust_agg['lambda'] = data_cust_agg['revenue_count'] / (data_cust_agg['recency'] + 1)\n",
    "    data_cust_agg = data_cust_agg.drop(['startdatetime_min', 'startdatetime_max'], axis=1)\n",
    "    \n",
    "    return data_cust_agg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_train_val = preprocess(df_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>customer_id_</th>\n",
       "      <th>revenue_mean</th>\n",
       "      <th>revenue_median</th>\n",
       "      <th>revenue_std</th>\n",
       "      <th>revenue_max</th>\n",
       "      <th>revenue_min</th>\n",
       "      <th>revenue_count</th>\n",
       "      <th>positions_count_mean</th>\n",
       "      <th>positions_count_median</th>\n",
       "      <th>positions_count_std</th>\n",
       "      <th>...</th>\n",
       "      <th>hour_max</th>\n",
       "      <th>hour_min</th>\n",
       "      <th>has_ws_mean</th>\n",
       "      <th>is_food_court_mean</th>\n",
       "      <th>buy_post_first</th>\n",
       "      <th>date_diff_post_first</th>\n",
       "      <th>recency</th>\n",
       "      <th>T</th>\n",
       "      <th>days_from_last_purchase</th>\n",
       "      <th>lambda</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>29891</td>\n",
       "      <td>203.494000</td>\n",
       "      <td>199.960</td>\n",
       "      <td>123.170275</td>\n",
       "      <td>439.98</td>\n",
       "      <td>1.00</td>\n",
       "      <td>25</td>\n",
       "      <td>1.360000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.860233</td>\n",
       "      <td>...</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>9.0</td>\n",
       "      <td>57</td>\n",
       "      <td>239</td>\n",
       "      <td>181</td>\n",
       "      <td>0.431034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>30477</td>\n",
       "      <td>227.024000</td>\n",
       "      <td>229.990</td>\n",
       "      <td>124.933425</td>\n",
       "      <td>499.95</td>\n",
       "      <td>44.99</td>\n",
       "      <td>25</td>\n",
       "      <td>2.440000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.916594</td>\n",
       "      <td>...</td>\n",
       "      <td>16</td>\n",
       "      <td>8</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>10.0</td>\n",
       "      <td>51</td>\n",
       "      <td>301</td>\n",
       "      <td>249</td>\n",
       "      <td>0.480769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>31426</td>\n",
       "      <td>391.399583</td>\n",
       "      <td>344.985</td>\n",
       "      <td>334.849322</td>\n",
       "      <td>1079.97</td>\n",
       "      <td>1.00</td>\n",
       "      <td>24</td>\n",
       "      <td>3.583333</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.909169</td>\n",
       "      <td>...</td>\n",
       "      <td>22</td>\n",
       "      <td>9</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>50</td>\n",
       "      <td>81</td>\n",
       "      <td>31</td>\n",
       "      <td>0.470588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>44491</td>\n",
       "      <td>128.725000</td>\n",
       "      <td>59.980</td>\n",
       "      <td>144.471912</td>\n",
       "      <td>344.97</td>\n",
       "      <td>49.97</td>\n",
       "      <td>4</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>2.5</td>\n",
       "      <td>0.577350</td>\n",
       "      <td>...</td>\n",
       "      <td>22</td>\n",
       "      <td>18</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>1</td>\n",
       "      <td>42.0</td>\n",
       "      <td>27</td>\n",
       "      <td>52</td>\n",
       "      <td>24</td>\n",
       "      <td>0.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>44939</td>\n",
       "      <td>554.943333</td>\n",
       "      <td>554.940</td>\n",
       "      <td>49.985000</td>\n",
       "      <td>604.93</td>\n",
       "      <td>504.96</td>\n",
       "      <td>3</td>\n",
       "      <td>8.333333</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>...</td>\n",
       "      <td>22</td>\n",
       "      <td>11</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1</td>\n",
       "      <td>9.0</td>\n",
       "      <td>15</td>\n",
       "      <td>234</td>\n",
       "      <td>219</td>\n",
       "      <td>0.187500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   customer_id_  revenue_mean  revenue_median  revenue_std  revenue_max  \\\n",
       "0         29891    203.494000         199.960   123.170275       439.98   \n",
       "1         30477    227.024000         229.990   124.933425       499.95   \n",
       "2         31426    391.399583         344.985   334.849322      1079.97   \n",
       "3         44491    128.725000          59.980   144.471912       344.97   \n",
       "4         44939    554.943333         554.940    49.985000       604.93   \n",
       "\n",
       "   revenue_min  revenue_count  positions_count_mean  positions_count_median  \\\n",
       "0         1.00             25              1.360000                     1.0   \n",
       "1        44.99             25              2.440000                     1.0   \n",
       "2         1.00             24              3.583333                     3.5   \n",
       "3        49.97              4              2.500000                     2.5   \n",
       "4       504.96              3              8.333333                     9.0   \n",
       "\n",
       "   positions_count_std  ...  hour_max  hour_min  has_ws_mean  \\\n",
       "0             0.860233  ...        19         0     1.000000   \n",
       "1             1.916594  ...        16         8     1.000000   \n",
       "2             1.909169  ...        22         9     0.000000   \n",
       "3             0.577350  ...        22        18     0.000000   \n",
       "4             1.154701  ...        22        11     0.333333   \n",
       "\n",
       "   is_food_court_mean  buy_post_first  date_diff_post_first  recency    T  \\\n",
       "0            0.000000               1                   9.0       57  239   \n",
       "1            0.000000               1                  10.0       51  301   \n",
       "2            1.000000               1                   4.0       50   81   \n",
       "3            0.250000               1                  42.0       27   52   \n",
       "4            0.666667               1                   9.0       15  234   \n",
       "\n",
       "   days_from_last_purchase    lambda  \n",
       "0                      181  0.431034  \n",
       "1                      249  0.480769  \n",
       "2                       31  0.470588  \n",
       "3                       24  0.142857  \n",
       "4                      219  0.187500  \n",
       "\n",
       "[5 rows x 38 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train_val.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_test = preprocess(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>customer_id_</th>\n",
       "      <th>revenue_mean</th>\n",
       "      <th>revenue_median</th>\n",
       "      <th>revenue_std</th>\n",
       "      <th>revenue_max</th>\n",
       "      <th>revenue_min</th>\n",
       "      <th>revenue_count</th>\n",
       "      <th>positions_count_mean</th>\n",
       "      <th>positions_count_median</th>\n",
       "      <th>positions_count_std</th>\n",
       "      <th>...</th>\n",
       "      <th>hour_max</th>\n",
       "      <th>hour_min</th>\n",
       "      <th>has_ws_mean</th>\n",
       "      <th>is_food_court_mean</th>\n",
       "      <th>buy_post_first</th>\n",
       "      <th>date_diff_post_first</th>\n",
       "      <th>recency</th>\n",
       "      <th>T</th>\n",
       "      <th>days_from_last_purchase</th>\n",
       "      <th>lambda</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>52341</td>\n",
       "      <td>563.942500</td>\n",
       "      <td>549.95</td>\n",
       "      <td>65.700330</td>\n",
       "      <td>655.93</td>\n",
       "      <td>499.94</td>\n",
       "      <td>4</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.816497</td>\n",
       "      <td>...</td>\n",
       "      <td>14</td>\n",
       "      <td>13</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>178</td>\n",
       "      <td>146</td>\n",
       "      <td>0.121212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>69175</td>\n",
       "      <td>430.306667</td>\n",
       "      <td>399.98</td>\n",
       "      <td>62.142941</td>\n",
       "      <td>554.96</td>\n",
       "      <td>374.98</td>\n",
       "      <td>9</td>\n",
       "      <td>5.444444</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.726483</td>\n",
       "      <td>...</td>\n",
       "      <td>21</td>\n",
       "      <td>8</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>40</td>\n",
       "      <td>5</td>\n",
       "      <td>0.257143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>73427</td>\n",
       "      <td>254.240323</td>\n",
       "      <td>169.98</td>\n",
       "      <td>218.360745</td>\n",
       "      <td>869.95</td>\n",
       "      <td>1.00</td>\n",
       "      <td>31</td>\n",
       "      <td>2.064516</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.112431</td>\n",
       "      <td>...</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>0.935484</td>\n",
       "      <td>0.129032</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>58</td>\n",
       "      <td>212</td>\n",
       "      <td>153</td>\n",
       "      <td>0.525424</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>134577</td>\n",
       "      <td>372.970000</td>\n",
       "      <td>290.98</td>\n",
       "      <td>310.195853</td>\n",
       "      <td>799.93</td>\n",
       "      <td>109.99</td>\n",
       "      <td>4</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.708013</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>16</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>316</td>\n",
       "      <td>310</td>\n",
       "      <td>0.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>156357</td>\n",
       "      <td>558.963333</td>\n",
       "      <td>644.96</td>\n",
       "      <td>245.551523</td>\n",
       "      <td>749.95</td>\n",
       "      <td>281.98</td>\n",
       "      <td>3</td>\n",
       "      <td>4.333333</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>...</td>\n",
       "      <td>17</td>\n",
       "      <td>10</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>43</td>\n",
       "      <td>227</td>\n",
       "      <td>184</td>\n",
       "      <td>0.068182</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   customer_id_  revenue_mean  revenue_median  revenue_std  revenue_max  \\\n",
       "0         52341    563.942500          549.95    65.700330       655.93   \n",
       "1         69175    430.306667          399.98    62.142941       554.96   \n",
       "2         73427    254.240323          169.98   218.360745       869.95   \n",
       "3        134577    372.970000          290.98   310.195853       799.93   \n",
       "4        156357    558.963333          644.96   245.551523       749.95   \n",
       "\n",
       "   revenue_min  revenue_count  positions_count_mean  positions_count_median  \\\n",
       "0       499.94              4              7.000000                     7.0   \n",
       "1       374.98              9              5.444444                     6.0   \n",
       "2         1.00             31              2.064516                     1.0   \n",
       "3       109.99              4              3.000000                     2.0   \n",
       "4       281.98              3              4.333333                     5.0   \n",
       "\n",
       "   positions_count_std  ...  hour_max  hour_min  has_ws_mean  \\\n",
       "0             0.816497  ...        14        13     1.000000   \n",
       "1             0.726483  ...        21         8     1.000000   \n",
       "2             2.112431  ...        23         0     0.935484   \n",
       "3             2.708013  ...        20        16     0.000000   \n",
       "4             1.154701  ...        17        10     1.000000   \n",
       "\n",
       "   is_food_court_mean  buy_post_first  date_diff_post_first  recency    T  \\\n",
       "0            0.000000               0                     0       32  178   \n",
       "1            0.000000               0                     0       34   40   \n",
       "2            0.129032               0                     0       58  212   \n",
       "3            1.000000               0                     0        5  316   \n",
       "4            1.000000               0                     0       43  227   \n",
       "\n",
       "   days_from_last_purchase    lambda  \n",
       "0                      146  0.121212  \n",
       "1                        5  0.257143  \n",
       "2                      153  0.525424  \n",
       "3                      310  0.666667  \n",
       "4                      184  0.068182  \n",
       "\n",
       "[5 rows x 38 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_val = data_train_val.drop(['date_diff_post_first', 'buy_post_first', 'customer_id_'], axis=1)\n",
    "y_train_val_class = data_train_val['buy_post_first']\n",
    "y_train_val_reg = data_train_val['date_diff_post_first']\n",
    "\n",
    "X_test = data_test.drop(['date_diff_post_first', 'buy_post_first', 'customer_id_'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_val, y_train_class, y_val_class, y_train_reg, y_val_reg = train_test_split(X_train_val, y_train_val_class, y_train_val_reg, test_size=.2, random_state=69)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LGBM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1: 0.694 ± 0.017\n"
     ]
    }
   ],
   "source": [
    "model_class = LGBMClassifier(\n",
    "    n_estimators=1000, \n",
    "    reg_alpha=20, \n",
    "    max_depth=3, \n",
    "    importance_type='gain', \n",
    "    reg_lambda=20, \n",
    "    class_weight={0: 0.75, 1:0.25},\n",
    "    verbose=-1,\n",
    ")\n",
    "\n",
    "res = cross_val_score(model_class, X_train_val, y_train_val_class, scoring='f1')\n",
    "print(f'f1: {res.mean():.3f} ± {res.std():.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, importance_type=&#x27;gain&#x27;,\n",
       "               max_depth=3, n_estimators=1000, reg_alpha=20, reg_lambda=20,\n",
       "               verbose=-1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMClassifier</label><div class=\"sk-toggleable__content\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, importance_type=&#x27;gain&#x27;,\n",
       "               max_depth=3, n_estimators=1000, reg_alpha=20, reg_lambda=20,\n",
       "               verbose=-1)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, importance_type='gain',\n",
       "               max_depth=3, n_estimators=1000, reg_alpha=20, reg_lambda=20,\n",
       "               verbose=-1)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_class.fit(X_train_val, y_train_val_class)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fi = pd.Series(model_class.feature_importances_, index=X_train.columns).sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<BarContainer object of 10 artists>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(9, 6))\n",
    "\n",
    "plt.barh(fi.index[-10:], fi.iloc[-10:], color='#502214') # "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 16.673 ± 0.207\n"
     ]
    }
   ],
   "source": [
    "model_reg = LGBMRegressor(\n",
    "    n_estimators=1000, \n",
    "    reg_alpha=20, \n",
    "    max_depth=3, \n",
    "    importance_type='gain', \n",
    "    reg_lambda=20,\n",
    "    verbose=-1,\n",
    ")\n",
    "\n",
    "res = cross_val_score(model_reg, X_train_val, y_train_val_reg.fillna(0), scoring='neg_root_mean_squared_error')\n",
    "print(f'RMSE: {-res.mean():.3f} ± {res.std():.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-7 {color: black;}#sk-container-id-7 pre{padding: 0;}#sk-container-id-7 div.sk-toggleable {background-color: white;}#sk-container-id-7 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-7 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-7 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-7 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-7 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-7 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-7 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-7 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-7 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-7 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-7 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-7 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-7 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-7 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-7 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-7 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-7 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-7 div.sk-item {position: relative;z-index: 1;}#sk-container-id-7 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-7 div.sk-item::before, #sk-container-id-7 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-7 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-7 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-7 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-7 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-7 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-7 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-7 div.sk-label-container {text-align: center;}#sk-container-id-7 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-7 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-7\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LGBMRegressor(importance_type=&#x27;gain&#x27;, max_depth=3, n_estimators=1000,\n",
       "              reg_alpha=20, reg_lambda=20, verbose=-1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" checked><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMRegressor</label><div class=\"sk-toggleable__content\"><pre>LGBMRegressor(importance_type=&#x27;gain&#x27;, max_depth=3, n_estimators=1000,\n",
       "              reg_alpha=20, reg_lambda=20, verbose=-1)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "LGBMRegressor(importance_type='gain', max_depth=3, n_estimators=1000,\n",
       "              reg_alpha=20, reg_lambda=20, verbose=-1)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_reg.fit(X_train_val, y_train_val_reg.fillna(0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_class = model_class.predict(X_test)\n",
    "pred_reg = model_reg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub = pd.DataFrame()\n",
    "\n",
    "sub['customer_id'] = data_test['customer_id_']\n",
    "sub['date_diff_post'] = pred_reg\n",
    "sub['buy_post'] = pred_class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub.to_csv('lgbm_3.csv', sep=';', index=False)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC-AUC: 0.736 ± 0.001\n"
     ]
    }
   ],
   "source": [
    "model_class = Pipeline([('scaling', StandardScaler()), ('logreg', LogisticRegression(max_iter=1000, class_weight={0: 0.7, 1:0.3}))])\n",
    "\n",
    "res = cross_val_score(model_class, X_train_val, y_train_val_class, scoring='roc_auc')\n",
    "print(f'ROC-AUC: {res.mean():.3f} ± {res.std():.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-10 {color: black;}#sk-container-id-10 pre{padding: 0;}#sk-container-id-10 div.sk-toggleable {background-color: white;}#sk-container-id-10 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-10 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-10 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-10 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-10 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-10 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-10 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-10 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-10 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-10 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-10 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-10 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-10 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-10 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-10 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-10 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-10 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-10 div.sk-item {position: relative;z-index: 1;}#sk-container-id-10 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-10 div.sk-item::before, #sk-container-id-10 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-10 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-10 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-10 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-10 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-10 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-10 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-10 div.sk-label-container {text-align: center;}#sk-container-id-10 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-10 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-10\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;scaling&#x27;, StandardScaler()),\n",
       "                (&#x27;logreg&#x27;,\n",
       "                 LogisticRegression(class_weight={0: 0.7, 1: 0.3},\n",
       "                                    max_iter=1000))])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;scaling&#x27;, StandardScaler()),\n",
       "                (&#x27;logreg&#x27;,\n",
       "                 LogisticRegression(class_weight={0: 0.7, 1: 0.3},\n",
       "                                    max_iter=1000))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-13\" type=\"checkbox\" ><label for=\"sk-estimator-id-13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-14\" type=\"checkbox\" ><label for=\"sk-estimator-id-14\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(class_weight={0: 0.7, 1: 0.3}, max_iter=1000)</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('scaling', StandardScaler()),\n",
       "                ('logreg',\n",
       "                 LogisticRegression(class_weight={0: 0.7, 1: 0.3},\n",
       "                                    max_iter=1000))])"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_class.fit(X_train_val, y_train_val_class)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArwAAAGdCAYAAADwoqBNAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAADdGklEQVR4nOzde1zP9///8du7Uqp3SSE10UjEklOOm0IfmdXChmGS4yxGH8cZUw6bYxibzRwKM+Yzx68RORRrLUI5pSWRbW3MITKKev3+cOn1663UO+fyuF4u78ul9+vwfD5eL/t+e35ePV/Pu0ZRFAUhhBBCCCHKKYPnXYAQQgghhBBPkwx4hRBCCCFEuSYDXiGEEEIIUa7JgFcIIYQQQpRrMuAVQgghhBDlmgx4hRBCCCFEuSYDXiGEEEIIUa7JgFcIIYQQQpRrRs+7ACFeBHl5efz5559YWFig0WiedzlCCCGE0IOiKNy8eRN7e3sMDB7+HFcGvEIAf/75Jw4ODs+7DCGEEEI8gosXL1KjRo2H7pcBryjzSnoiGxwcTEhISLHHWFhYAPf/D8bS0vJJlSaEEEKIp+jGjRs4ODiov8cfRga8oszLyMhQf/7hhx+YMmUKycnJ6jatVltiG/mDZktLSxnwCiF0OH780/MuQYgy7/yst55q+yU9/JIBryjzqlevrv5cqVIlNBqNzjYhhBBCvNxkwCteStnZ2WRnZ6vfb9y48RyrEUIIIcTTJMuSiZfSzJkzqVSpkvqRF9aEEEKI8ksGvOKlNHHiRDIzM9XPxYsXn3dJQgghhHhKZEqDeCmZmJhgYmLyvMsQQgghxDMgA14hhBCiGE/77XIhxNMnUxqEEEIIIUS5JgNeIYQQQghRrsmA9ynz9PQkKChIr2PDw8OxsrJ6qvW8SP0+DQEBAVy/fh1HR0cWLlz4vMsRQgghxAtABrwvsJCQEBo3bvy8y3hhlKeBuRBCCCGeHXlpTQghhCiGRAsLUXov2sue8oT3Cbp16xb+/v5otVrs7OwIDQ3V2Z+dnc3YsWN55ZVXMDc3p2XLlkRFRRXZVnh4OFOnTiUxMRGNRoNGoyE8PByA+fPn4+rqirm5OQ4ODgQGBpKVlaV3neHh4dSsWRMzMzO6devGlStXdPanpqbi5+eHra0tWq0Wd3d39uzZo+6fNm0ar732WqF2GzduzKeffgpAVFQULVq0wNzcHCsrK9q2bcuFCxdKrC0xMZH27dtjYWGBpaUlzZo1Iz4+nqioKAYMGEBmZqZ6P0JCQgC4dOkSvr6+mJqa8uqrr7J27Vq974UQQgghyj8Z8D5B48aNIzo6mq1bt7J7926ioqI4evSoun/EiBHExsayfv16jh8/To8ePejcuTMpKSmF2urVqxdjxoyhYcOGZGRkkJGRQa9evQAwMDBg0aJFnDp1ilWrVrFv3z7Gjx+vV41xcXEMGjSIESNGkJCQQPv27ZkxY4bOMVlZWXTp0oW9e/dy7NgxOnfujK+vL+np6QAMHDiQpKQkDh8+rJ5z7Ngxjh8/zoABA7h37x5du3bFw8OD48ePExsby9ChQ9FoNCXW17dvX2rUqMHhw4c5cuQIH3/8MRUqVKBNmzYsXLgQS0tL9X6MHTsWuD9v9+LFi+zfv58ff/yRJUuWcOnSpWL7yc7O5saNGzofIYQQQpRPMqXhCcnKymLFihV89913dOzYEYBVq1ZRo0YNANLT0wkLCyM9PR17e3sAxo4dS0REBGFhYXz++ec67ZmamqLVajEyMqJ69eo6+wq+BOfo6MiMGTMYNmwYS5YsKbHOL774gs6dO6sDZGdnZ3755RciIiLUY9zc3HBzc1O/T58+nc2bN7Nt2zZGjBhBjRo18Pb2JiwsDHd3dwDCwsLw8PCgdu3aXL16lczMTHx8fKhTpw4ALi4uet3H9PR0xo0bR/369QGoW7euuq9SpUpoNBqd+/Hbb7+xc+dODh06pNayYsWKEvubOXMmU6dO1asmIYQQQpRt8oT3CUlNTSUnJ4eWLVuq26ytralXrx4AJ06cIDc3F2dnZ7RarfqJjo4mNTW1VH3t2bOHjh078sorr2BhYUG/fv24cuUK//77b4nnJiUl6dQI0Lp1a53vWVlZjB07FhcXF6ysrNBqtSQlJalPeAGGDBnCunXruHPnDjk5OXz//fcMHDhQve6AgAC8vb3x9fXliy++ICMjQ69rGz16NIMHD8bLy4tZs2aVeG+SkpIwMjKiWbNm6rb69euX+HKbRAsLIYQQLw95wvuMZGVlYWhoyJEjRzA0NNTZp9Vq9W7n/Pnz+Pj48OGHH/LZZ59hbW3Nzz//zKBBg8jJycHMzOyxax07diyRkZHMmzcPJycnTE1Neffdd8nJyVGP8fX1xcTEhM2bN2NsbMzdu3d599131f1hYWGMHDmSiIgIfvjhByZPnkxkZCStWrUqtu+QkBD69OnDTz/9xM6dOwkODmb9+vV069btsa+rIIkWFkLo60V7+UYIUXoy4H1C6tSpQ4UKFYiLi6NmzZoAXLt2jd9++w0PDw+aNGlCbm4uly5d4o033tCrTWNjY3Jzc3W2HTlyhLy8PEJDQzEwuP+AfsOGDXrX6eLiQlxcnM62X3/9Ved7TEwMAQEB6iAzKyuL8+fP6xxjZGRE//79CQsLw9jYmPfeew9TU1OdY5o0aUKTJk2YOHEirVu35vvvvy9xwAv3p1k4Ozvz3//+l969exMWFka3bt2KvB/169fn3r17HDlyRJ3SkJyczPXr1/W5HUIIIYR4CciA9wnRarUMGjSIcePGYWNjQ7Vq1Zg0aZI6KHV2dqZv3774+/sTGhpKkyZNuHz5Mnv37qVRo0a89VbhJwiOjo6kpaWRkJBAjRo1sLCwwMnJibt377J48WJ8fX2JiYnhm2++0bvOkSNH0rZtW+bNm4efnx+7du3Smb8L9+fNbtq0CV9fXzQaDZ9++il5eXmF2ho8eLA6VzYmJkbdnpaWxrfffsvbb7+Nvb09ycnJpKSk4O/vX2xtt2/fZty4cbz77ru8+uqr/P777xw+fJh33nlHvR9ZWVns3bsXNzc3zMzMqFevHp07d+aDDz7g66+/xsjIiKCgoEKDbyGEEEK8xBTxxNy8eVN5//33FTMzM8XW1laZM2eO4uHhoYwaNUpRFEXJyclRpkyZojg6OioVKlRQ7OzslG7duinHjx9XFEVRwsLClEqVKqnt3blzR3nnnXcUKysrBVDCwsIURVGU+fPnK3Z2doqpqani7e2trF69WgGUa9eu6VXnihUrlBo1aiimpqaKr6+vMm/ePJ1+09LSlPbt2yumpqaKg4OD8uWXX+pcR0FvvPGG0rBhQ51tf/31l9K1a1fFzs5OMTY2VmrVqqVMmTJFyc3NLbau7Oxs5b333lMcHBwUY2Njxd7eXhkxYoRy+/Zt9Zhhw4YpNjY2CqAEBwcriqIoGRkZyltvvaWYmJgoNWvWVFavXq3UqlVLWbBggV73Q1EUJTMzUwGUzMxMvc8RQgghxPOl7+9vjaIoynMdcYsyS1EU6tatS2BgIKNHj37e5TyWGzduUKlSJTIzM7G0tHze5QghhBBCD/r+/pYpDaJYOTk5GBsbF9p++fJl1q9fz19//cWAAQOeQ2VCCPFsSNKaeFTywuOLQ5YlK2fefPNNnWXPCn4eXOu3KJ6enowYMYKgoCCqVKmCt7c3J0+eVNu1tbWlX79+VKtWjWnTpvHtt99SqVIl5syZg5OTEyYmJtSsWZPPPvtMbfPixYv07NkTKysrDA0NMTIywtzcXK2rQoUKGBkZ0adPH+zs7LCxsWH48OHcvXtXbSM7O5sJEybg4OCAiYkJTk5OrFixAkVRcHJyYt68eTrXkZCQgEaj4ezZs0/u5gohhBCiTJInvOXM8uXLuX37dpH7rK2t9Wpj1apVfPjhh8TExHD9+nU6dOjA4MGDWbBgAbdv32bChAm0b9+effv2ATBhwgSWLVvGggULeP3118nIyODMmTMA3L17F29vb1q3bs3Bgwf5+++/WbRoESdPnmT79u0YGxszfvx4IiMjMTU1Zf/+/Zw9e5ZevXrRuHFjhgwZAoC/vz+xsbEsWrQINzc30tLS+Oeff9BoNAwcOJCwsDA1eQ3uL4vWrl07nJycirzG7OxssrOz1e+StCaEEEKUXzKHV+jw9PTkxo0baiTyjBkzOHjwILt27VKP+f3333FwcCA5ORk7OzuqVq3Kl19+yeDBgwu199133zFjxgySkpLUaOGcnBysrKzYsmULnTp1IiAggKioKFJTU9U1inv27ImBgQHr16/nt99+o169ekRGRuLl5VWojz///JOaNWvyyy+/0KJFC+7evYu9vT3z5s2jf//+RV5nSEhIkUlrModXCPEgmdIgHpVMaXj69J3DK1MaRCEFU8sSExPZv3+/ztSI/Njf1NRUkpKSyM7OVuOUH5SYmMjZs2exsLBQz7e2tubOnTs6KWoNGzbUCeSws7Pj0qVLwP3pCYaGhnh4eBTZh729PW+99RYrV64E4P/+7//Izs6mR48eD71GSVoTQgghXh4ypUEUYm5urv6clZWFr68vs2fPLnScnZ0d586dK7atrKwsmjVrxtq1awvtq1q1qvpzhQoVdPZpNBp17V991tQdPHgw/fr1Y8GCBYSFhdGrV69iU+ckaU0IIYR4eciAVxSradOmbNy4EUdHR4yMCv/nUrduXUxNTdm7d2+RUxqaNm3KDz/8QLVq1R55qoCrqyt5eXlER0cXOaUBoEuXLpibm/P1118TERHBgQMHHqkvIYR4kPxZWoiyT6Y0iGINHz6cq1ev0rt3bw4fPkxqaiq7du1iwIAB5ObmUrFiRSZMmMD48eNZvXo1qamp/Prrr6xYsQKAvn37UqVKFfz8/Dh48CBpaWlERUUxcuRIfv/9d71qcHR0pH///gwcOJAtW7aobRSMVDY0NCQgIICJEydSt25dWrdu/VTuhxBCCCHKHhnwimLZ29sTExNDbm4unTp1wtXVlaCgIKysrNTY5E8//ZQxY8YwZcoUXFxc6NWrlzr/1szMjAMHDlCzZk26d++Oi4sLgwYN4s6dO6V64vv111/z7rvvEhgYSP369RkyZAi3bt3SOWbQoEHk5OTIusBCCCGE0PHSr9Lg6elJ48aNWbhwYYnHhoeHExQUxPXr159aPc+ij/Lq4MGDdOzYkYsXL2Jra1uqcyVpTQghhCh7ZJWGZyAkJITGjRs/7zJeetnZ2fz++++EhITQo0ePUg92hRBCCFG+yUtrAoDc3Fw0Go06TaEsWbduHYMGDaJx48asXr36eZcjhChnZB3e8kFePny5lb3RzWO4desW/v7+aLVa7OzsCA0N1dmfnZ3N2LFjeeWVVzA3N6dly5ZERUUV2VZ4eDhTp04lMTERjUaDRqMhPDwcgPnz5+Pq6oq5uTkODg4EBgaSlZVVqlp37dqFi4sLWq2Wzp07k5GRoe7Ly8tj2rRp1KhRAxMTExo3bkxERIS6PyoqCo1GozMtIj9q9/z582r9VlZWbNu2jQYNGmBiYkJ6ejpRUVG0aNECc3NzrKysaNu2LRcuXCix3vyn3StXrqRmzZpotVoCAwPJzc1lzpw5VK9enWrVqulEDgNcv36dwYMHU7VqVSwtLenQoQOJiYnq/tTUVPz8/LC1tUWr1eLu7s6ePXsK9T19+nTc3NyoX78+NWvW5Ntvvy3N7RZCCCFEOfZSDXjHjRtHdHQ0W7duZffu3URFRamJYgAjRowgNjaW9evXc/z4cXr06EHnzp1JSUkp1FavXr0YM2YMDRs2JCMjg4yMDHr16gWAgYEBixYt4tSpU6xatYp9+/Yxfvx4vev8999/mTdvHmvWrOHAgQOkp6frxOZ+8cUXhIaGMm/ePI4fP463tzdvv/12kXWW1M/s2bNZvnw5p06dwtramq5du+Lh4cHx48eJjY1l6NChakJaSVJTU9m5cycRERGsW7eOFStW8NZbb/H7778THR3N7NmzmTx5MnFxceo5PXr04NKlS+zcuZMjR47QtGlTOnbsyNWrV4H76/h26dKFvXv3cuzYMTp37oyvry/p6ek6fYeGhtK8eXOOHTtGYGAgH374IcnJyQ+tNTs7mxs3buh8hBBCCFE+vTRTGrKyslixYgXfffedmgq2atUqatSoAUB6ejphYWGkp6djb28PwNixY4mIiCAsLIzPP/9cpz1TU1O0Wi1GRkZUr15dZ19QUJD6s6OjIzNmzGDYsGEsWbJEr1rv3r3LN998Q506dYD7A/Fp06ap++fNm8eECRN47733AJg9ezb79+9n4cKFfPXVV3rfk7t377JkyRLc3NwAuHr1KpmZmfj4+Kh9u7i46N1eXl4eK1euxMLCggYNGtC+fXuSk5PZsWMHBgYG1KtXT621ZcuW/Pzzzxw6dIhLly6pIRDz5s1jy5Yt/PjjjwwdOhQ3Nze1PoDp06ezefNmtm3bxogRI9TtXbp0ITAwEIAJEyawYMEC9u/fT7169YqsdebMmUVGCwshhBCi/HlpBrypqank5OTQsmVLdZu1tbU6IDpx4gS5ubk4OzvrnJednY2NjU2p+tqzZw8zZ87kzJkz3Lhxg3v37nHnzh3+/fffYtO/8pmZmakDTtCN2b1x4wZ//vknbdu21Tmnbdu2OlMB9GFsbEyjRo3U79bW1gQEBODt7c1//vMfvLy86NmzJ3Z2dnq15+joiIWFhfrd1tYWQ0NDnXnBtra26rUkJiaSlZVV6P7evn1bjR3OysoiJCSEn376iYyMDO7du8ft27cLPeEteB0ajYbq1aur/RRl4sSJjB49Wv1+48YNHBwc9LpOIYQQQpQtL82AtyRZWVkYGhpy5MgRDA0NdfZptVq92zl//jw+Pj58+OGHfPbZZ1hbW/Pzzz+ra8TqM+AtKma3NKvH5Q8wC55z9+7dQseZmpoWmq4QFhbGyJEjiYiI4IcffmDy5MlERkbSqlWrR6q7uMjgrKws7OzsipwnbWVlBdx/yh4ZGcm8efNwcnLC1NSUd999l5ycnBL7zu+nKBItLITQl7zsJETZ99IMeOvUqUOFChWIi4ujZs2aAFy7do3ffvsNDw8PmjRpQm5uLpcuXeKNN97Qq01jY2Nyc3N1th05coS8vDxCQ0PVgWfBRLDHZWlpqYZBeHh4qNtjYmJo0aIFAFWrVgUgIyODypUrA/dfWtNXkyZNaNKkCRMnTqR169Z8//33eg14S6tp06b89ddfGBkZ4ejoWOQxMTExBAQE0K1bN+D+IDn/xTshhBBCCH28NC+tabVaBg0axLhx49i3bx8nT54kICBAHZQ6OzvTt29f/P392bRpE2lpaRw6dIiZM2fy009FL0nj6OhIWloaCQkJ/PPPP2RnZ+Pk5MTdu3dZvHgx586dY82aNXzzzTdP9FrGjRvH7Nmz+eGHH0hOTubjjz8mISGBUaNGAeDk5ISDgwMhISGkpKTw008/FVqRoihpaWlMnDiR2NhYLly4wO7du0lJSSnVPN7S8PLyonXr1nTt2pXdu3dz/vx5fvnlFyZNmkR8fDwAdevWZdOmTSQkJJCYmEifPn2KfXIrhBBCCPGgl+YJL8DcuXPJysrC19cXCwsLxowZQ2Zmpro/LCyMGTNmMGbMGP744w+qVKlCq1at2LhxI5GRkYVCJt555x02bdpE+/btuX79OmFhYQBUrFiR2bNnM3HiRNq1a8fMmTPx9/d/YtdhaWlJdnY2Y8aM4dKlSzRo0IBt27ZRt25d4P6f99etW8eHH35Io0aNcHd3Z8aMGfTo0aPYds3MzDhz5gyrVq3iypUr2NnZMXz4cD744IMnVntBGo2GHTt2MGnSJAYMGMDly5epXr067dq1U8Mj5s+fz8CBA2nTpg1VqlRhwoQJsqKCEEIIIUrlpY8W1sfjxA+HhISwZcuWUk0pKG0f4vFJtLAQQghR9ki0sHjicnNzZTqBEEIIIcqcl2pKgz5u3brFhx9+yKZNm7CwsNAJfID7y5RNmjSJdevWcf36dV577TVmz56Np6dnobby09gAdTUEV1dXzp07x927d7l37x55eXloNBoMDQ355JNPSrU27K5duwgKCuLixYu8/vrrhIWFqUuI5eXlMWPGDL799lsuX76Mi4sLs2bNonPnzsD9NLb27dtz7do1dUWEhIQEmjRpQlpaGo6OjuqT5NWrV9OjRw91lYm8vDxycnLUwW/dunWJjIykVq1axdab/7R75MiRhISEcPXqVfz9/Vm8eDGhoaHMnz+fvLw8Ro0axaRJk9Tz5s+fT1hYGOfOncPa2hpfX1/mzJmjrp4xcOBA4uPjOXz4MCYmJuryc66urhI1LIR4bBIt/HCygoUoK2TA+4CCaWzVqlXjk08+4ejRo+r83REjRnD69GnWr1+Pvb09mzdvpnPnzpw4cUKdQ5uvV69enDx5koiICDUONz9iOCwsDBcXF2rUqMHFixcJDg7m4sWLetdZMI3NwMCA999/n7Fjx7J27Vrg/6exLV26lCZNmrBy5UrefvttTp06VajOkvqZPXs269atQ6vVUrVqVdq1a0ffvn3p3bs3d+/e5fz584+Uxpaamsq7777LuXPncHZ2Jjo6ml9++YWBAwfi5eWlrpmcn1z36quvcu7cOQIDAxk/frwa5LFo0SLc3Nz4+OOPWbBgAZMmTeL69et8+eWXD60jOzub7Oxs9bvMCxZCCCHKLxnwFvAs09g+++yzQscOGzZM71qfZxpbVlYW/fr101kWTV+lTWODkpPrtFot3333HR4eHlhYWLBw4UL2799f7FweSVoTQgghXh4y4C1A0tgKe95pbKDfvWrdujVjx45l+vTpTJgwgddff73YOiRpTQghhHh5yEtrpVAwjS0hIUH9JCUl8cUXX+jdTn4aW6NGjdi4cSNHjhxRn7o+mCD2MM87jS02NpY2bdrwww8/4OzszK+//vrIdReXkqbvvcrLyyMmJgZDQ0POnj1bYh0mJiZYWlrqfIQQQghRPskT3gIkjS1B7z6eVRqbvvdq7ty5nDlzhujoaLy9vQkLC2PAgAFPvB4hxMtHXswSouyTJ7wFSBrbi5fGps+9OnbsGFOmTGH58uW0bduW+fPnM2rUKM6dO/dUahJCCCFE2SID3gfMnTuXN954A19fX7y8vHj99ddp1qyZuj8sLAx/f3/GjBlDvXr16Nq1K4cPH1afCD/onXfeoXPnzrRv356qVauybt063NzcmD9/PrNnz+a1115j7dq1zJw584lex8iRIxk9ejRjxozB1dWViIiIItPYzpw5Q6NGjZg9ezYzZswosd38NLZ33nkHZ2dnhg4d+lTT2Eq6V3fu3OH9998nICAAX19fAIYOHUr79u3p169foafrQgghhHj5lOuktaLWmi2Ko6MjQUFBOqsBiJeLJK0JIYQQZY8krQFt2rQhIyODSpUqAfeDIIoa+B4+fJihQ4c+4+pebBqNhi1btjzvMoQQQgghHlu5fmnN2Ni40Pq3Rcl/getF8eabb3Lw4MEi933yySd88sknz7gi/TRs2JALFy4UuW/p0qX07dv3GVckhBCP72klrcnLcEI8O8/9Ca+npycjRoxgxIgRVKpUiSpVqvDpp5+qy2Vdu3YNf39/KleujJmZGW+++SYpKSnq+RcuXMDX15fKlStjbm5Ow4YN2bFjB3B/SoNGo+H69etERUUxYMAAMjMz0Wg0aDQaQkJCgPtTGhYuXKi2mZ6ejp+fH1qtFktLS3r27Mnff/+t7g8JCaFx48asWbMGR0dHKlWqxHvvvcfNmzfVY3788UdcXV0xNTXFxsYGLy8vbt26pdc98fLyws7OjpycHExNTenatau6BJqPj0+xtQUEBNC1a1ed9oKCgnSijz09PRk5ciTjx4/H2tqa6tWrq/ci/34AdOvWDY1Go34vTkhICIqiMGnSJCwsLMjLy8PPz48jR44QGBjIf//7X6pVq1YocOP69esMHjyYqlWrYmlpSYcOHXTWCk5NTcXPzw9bW1u0Wi3u7u5qal3Bej///HMGDhyIhYUFNWvW5Ntvvy2xZiGEEEK8HJ77gBfup5kZGRlx6NAhvvjiC+bPn8/y5cuB+wO4+Ph4tm3bRmxsLIqi0KVLF3XN2OHDh5Odnc2BAwc4ceIEs2fPRqvVFuqjTZs2LFy4EEtLSzIyMsjIyGDs2LGFjssfqF29epXo6GgiIyM5d+4cvXr10jkuNTWVLVu2sH37drZv3050dDSzZs0C7i/11bt3bwYOHEhSUhJRUVF0795dr3Vyv/76ayZPnsyIESM4efIkO3bsoHnz5jg5OVG7dm369+9fYm363nNzc3Pi4uKYM2cO06ZNIzIyErg/xQPuv6CXkZGhfi/JxYsXOXr0KHv37uWHH37gxx9/ZNSoUdy5c4eDBw8ye/ZsJk+eTFxcnHpOjx49uHTpEjt37uTIkSM0bdqUjh07cvXqVeD+2sddunRh7969HDt2jM6dO+Pr60t6erpO36GhoTRv3pxjx44RGBjIhx9+SHJy8kNrzc7O5saNGzofIYQQQpRPL8SUBgcHBxYsWIBGo6FevXqcOHGCBQsW4OnpybZt24iJiaFNmzYArF27FgcHB7Zs2UKPHj1IT0/nnXfewdXVFYDatWsX2YexsTGVKlVCo9EUO81h7969nDhxgrS0NDV5a/Xq1TRs2JDDhw/j7u4O3B8Yh4eHq6lh/fr1Y+/evXz22WdkZGRw7949unfvTq1atQDU+koyY8YMxowZoy4fBqh96lubPho1akRwcDAAdevW5csvv2Tv3r385z//Uad4WFlZ6TUlJF9pY4N//vlnDh06xKVLlzAxMQHuRyJv2bKFH3/8kaFDh+Lm5qbGGgNMnz6dzZs3s23bNkaMGKFu79KlC4GBgQBMmDCBBQsWsH//fjUl70ESLSyEEEK8PF6IJ7ytWrXSSfNq3bo1KSkpnD59GiMjI52oXxsbG+rVq0dSUhJwf/mtGTNm0LZtW4KDgzl+/Phj1ZKUlISDg4NOzGyDBg2wsrJS+4TCEbkFo33d3Nzo2LEjrq6u9OjRg2XLlnHt2rUS+7506RJ//vknHTt2fKza9FEwLvjB+h9VUbHBDRo0eGhscGJiIllZWdjY2KDVatVPWloaqampwP0nvGPHjsXFxQUrKyu0Wi1JSUmFnvAWvJ78/1FT3PVMnDiRzMxM9XPx4sXHunYhhBBCvLheiCe8j2Pw4MF4e3vz008/sXv3bmbOnEloaCgfffTRU+23uDhcQ0NDIiMj+eWXX9i9ezeLFy9m0qRJxMXF8eqrrz60TVNT08euy8DAoNDUiaIig4ur/1GVNjY4KysLOzs7oqKiCrWVv5rG2LFjiYyMZN68eTg5OWFqasq7775bKIK5tNdjYmKiPlUWQojiyMtlQpR9L8QT3oJzOgF+/fVX6tatS4MGDbh3757O/itXrpCcnEyDBg3UbQ4ODgwbNoxNmzYxZswYli1bVmQ/RcX8PsjFxYWLFy/qPPE7ffo0169f1+mzJBqNhrZt2zJ16lSOHTuGsbExmzdvLvYcCwsLHB0d2bt37yPXVrVqVTIyMnTOK01kcL4KFSo89dCGpk2b8tdff2FkZISTk5POp0qVKsD9OOSAgAC6deuGq6sr1atX5/z580+1LiGEEEKULy/EgDc9PZ3Ro0eTnJzMunXrWLx4MaNGjaJu3br4+fkxZMgQfv75ZxITE3n//fd55ZVX8PPzA+6vQLBr1y7S0tI4evQo+/fvf2jMraOjI1lZWezdu5d//vmHf//9t9AxXl5euLq60rdvX44ePcqhQ4fw9/fHw8OD5s2b63U9cXFxfP7558THx5Oens6mTZu4fPmyXvG7ISEhhIaGsmjRIlJSUjh69CiLFy/Wu7YOHToQHx/P6tWrSUlJITg4mJMnT+pVd0H5A++//vpLr+kYj8LLy4vWrVvTtWtXdu/ezfnz5/nll1+YNGkS8fHxwP35xZs2bSIhIYHExET69Onz2E+ihRBCCPFyeSEGvP7+/ty+fZsWLVowfPhwRo0apQZBhIWF0axZM3x8fGjdujWKorBjxw71T9i5ubkMHz4cFxcXOnfujLOzM0uWLCmynzZt2jBs2DB69epF1apVmTNnTqFjNBoNW7dupXLlyrRr1w4vLy9q167NDz/8oPf1WFpacuDAAbp06YKzszOTJ08mNDSUN998s8Rz+/fvz8KFC1myZAkNGzbEx8dHXYZNn9q8vb359NNPGT9+PO7u7ty8eRN/f3+9a88XGhpKZGQkDg4ONGnSpNTn60Oj0bBjxw7atWvHgAEDcHZ25r333uPChQvY2toCMH/+fCpXrkybNm3w9fXF29ubpk2bPpV6hBBCCFE+PfdoYU9PTxo3bqyzDu7LpuA9kJjjJ0Oj0bB58+ZCaxI/jEQLCyGEEGWPRAuXUWUt5rhguMfTlh/4IYQQQghRGmV+lYaypqhQjNu3b/PLL7/wzjvv8MYbbzyHqkr2sNjgp/1imxBCPG9PIlpYVnoQ4vl67k94o6KiXqrpDL/88gteXl7k5eVhamrKRx99RPPmzenTpw/NmzfXiTlWFIWQkBBq1qyJiYkJ9vb2jBw5Um0rOzubCRMm4ODggImJCU5OTqxYsULdHx0dTYsWLTAxMcHOzo6PP/6Ye/fuqfsfjFQGaNy4sU7MsEajYfny5dSoUYN79+5hY2NDaGgoCQkJbN++nTt37gBQuXJlNBoNAQEBwKNHK0dFRdGiRQvMzc2xsrKibdu2XLhwgfDwcKZOnUpiYqIaDR0eHg5ASkoK7dq1o2LFijRo0EBNjCuOJK0JIYQQLw95wvuMffPNNxw7doz/+7//o1q1anzyySckJSXRunXrQuvwbty4kQULFrB+/XoaNmzIX3/9RWJiorrf39+f2NhYFi1ahJubG2lpafzzzz8A/PHHH3Tp0oWAgABWr17NmTNnGDJkCBUrVtQZ0Opj6tSpzJkzh6+++orFixczduxYLly4wKuvvsrGjRt55513SE5OxtLSElNTUzVaec6cOXTr1o2bN29y8ODBEqOV7927R9euXRkyZAjr1q0jJyeHQ4cOodFo6NWrFydPniQiIoI9e/YAUKlSJfLy8ujevTu2trbExcWRmZmp1/xnSVoTQgghXh4y4H2GsrKyWLFiBd99952aprZq1Spq1KhR5PHp6elUr14dLy8vKlSoQM2aNWnRogUAv/32Gxs2bCAyMhIvLy9AN1Z5yZIlODg48OWXX6LRaKhfvz5//vknEyZMYMqUKTrpZyUJCAigd+/eAHz++ecsWrSIQ4cO0blzZ6ytrQGoVq2aGhaRmpr6SNHKN27cIDMzEx8fH+rUqQOgs5SbVqvFyMhIJ+549+7dnDlzhl27dmFvb6/WWNKKGBMnTmT06NE6fRdMsBNCCCFE+fHcpzS8TFJTU8nJydGJSra2tqZevXpFHt+jRw9u375N7dq1GTJkCJs3b1anJCQkJGBoaIiHh0eR5+Y/NS4Y2dy2bVuysrL4/fffS1V3wdhec3NzLC0ti43tfdRoZWtrawICAvD29sbX15cvvviiUIjGg/LjlvMHu3A/mrokJiYmWFpa6nyEEEIIUT7JE94XmIODA8nJyezZs4fIyEgCAwOZO3cu0dHRL3QM8aNGK8P9dZdHjhxJREQEP/zwA5MnTyYyMpJWrVqV4sqEEOLJkRfOhCj75AnvM1SnTh0qVKigE5V87do1fvvtt4eeY2pqiq+vL4sWLSIqKorY2FhOnDiBq6sreXl5REdHF3mei4sLsbGxOgPamJgYLCws1CkUD8YQ37hxg7S0tFJdk7GxMVB4tYZHiVbO16RJEyZOnMgvv/zCa6+9xvfff6/29WA/+XHLBa/j119/LdU1CCGEEKJ8kwHvM6TVahk0aBDjxo1j3759nDx5koCAgIfOpw0PD2fFihWcPHmSc+fO8d1332FqakqtWrVwdHSkf//+DBw4kC1btpCWlkZUVBQbNmwAIDAwkIsXL/LRRx9x5swZtm7dSnBwMKNHj1b769ChA2vWrOHgwYOcOHGC/v37Y2hoWKprqlWrFhqNhu3bt3P58mWysrIeOVo5LS2NiRMnEhsby4ULF9i9ezcpKSnqeY6OjqSlpZGQkMA///xDdnY2Xl5eODs7079/fxITEzl48CCTJk0q1TUIIYQQopxTxDN18+ZN5f3331fMzMwUW1tbZc6cOYqHh4cyatQoRVEUpVatWsqCBQsURVGUzZs3Ky1btlQsLS0Vc3NzpVWrVsqePXvUtm7fvq3897//Vezs7BRjY2PFyclJWblypbo/KipKcXd3V4yNjZXq1asrEyZMUO7evavuz8zMVHr16qVYWloqDg4OSnh4uOLm5qYEBwerxwDK5s2bda6hUqVKSlhYmPp92rRpSvXq1RWNRqP0799fOX36tOLt7a1UrVpVMTExUZydnZXFixeXeG/++usvpWvXrur11KpVS5kyZYqSm5urKIqi3LlzR3nnnXcUKysrBVBrSE5OVl5//XXF2NhYcXZ2ViIiIoqsuziZmZkKoGRmZup9jhBCCCGeL31/fz/3aGEhnqTw8HCCgoJKnfwm0cJCCCFE2aPv7295ae0py8nJUee5CiGEKHtKk7QmL7gJ8WKSObxPmKenJyNGjCAoKIgqVarg7e3NyZMnefPNN9Fqtdja2tKvXz81IOLbb7/F3t6+0KoHfn5+DBw4UP2+detWmjZtSsWKFalduzZTp07VSU3LT0Tr1q0bZmZm1K1bl23btqn7w8PD1XVy823ZskVn2TJ9+inO9evX+eCDD7C1taVixYq89tprbN++Xd2/ceNGDAwM0Gg0GBgYYGJiglarVT8ajYYtW7botGllZaUmqp0/fx6NRsOmTZto3749ZmZmuLm5ERsbC9xPaRswYACZmZlqGltpQzaEEEIIUf7IgPcpWLVqFcbGxsTExDBr1iw6dOhAkyZNiI+PJyIigr///puePXsC99favXLlCvv371fPv3r1KhEREfTt2xeAgwcP4u/vz6hRozh9+jRLly4lPDyczz77TKffqVOn0rNnT44fP06XLl3o27cvV69e1btuffspSl5eHm+++SYxMTF89913nD59mlmzZqkvwR05coSePXsycuRIdu/ezcyZM9FoNHz66ackJCSQkJCgd52TJk1i7NixJCQk4OzsTO/evbl37x5t2rRh4cKFWFpakpGRQUZGBmPHji2yDYkWFkIIIV4iz2RG8UvEw8NDadKkifp9+vTpSqdOnXSOuXjxogIoycnJiqIoip+fnzJw4EB1/9KlSxV7e3v1Za2OHTsqn3/+uU4ba9asUezs7NTvgDJ58mT1e1ZWlgIoO3fuVBRFUcLCwpRKlSrptLF582al4H8C+vTzMLt27VIMDAzUa3pQnz59lP/85z8628aNG6c0aNBA5xqKe0EuLS1NAZTly5er+0+dOqUASlJS0kOvsyjBwcEKUOgjL60JIR5Ua8J2vT9CiGdL35fW5AnvU9CsWTP158TERPbv36/zp/v69esD95PXAPr27cvGjRvJzs4GYO3atbz33nvq8mGJiYlMmzZNp40hQ4aQkZHBv//+q/ZV2kS0B+nbT1ESEhKoUaMGzs7ORe5PSkqibdu2Otvatm1LSkpKobV1S1LwOu3s7ABKdZ1wP1o4MzNT/Vy8eLFU5wshhBCi7JCX1p4Cc3Nz9eesrCx8fX2ZPXt2oePyB2u+vr4oisJPP/2Eu7s7Bw8eZMGCBTptTJ06le7duxdqo2LFiurPxSWi6ZOqpm8/RXkSyW8ajabUyW/5c5CLS34riomJCSYmJo9QpRBCCCHKGhnwPmVNmzZl48aNODo6YmRU9O2uWLEi3bt3Z+3atZw9e5Z69erRtGlTnTaSk5NxcnJ65DqqVq3KzZs3uXXrljogf3De7OP006hRI37//Xd+++23Ip/yuri4EBMTo7MtJiYGZ2dndZ7vg8lvKSkpJT5ZflBRaWxCCPE4ZOUFIco+GfA+ZcOHD2fZsmX07t2b8ePHY21tzdmzZ1m/fj3Lly9XB3t9+/bFx8eHU6dO8f777+u0MWXKFHx8fKhZsybvvvsuBgYGJCYmcvLkSWbMmKFXHS1btsTMzIxPPvmEkSNHEhcXp65+8CT68fDwoF27drzzzjvMnz8fJycnzpw5g0ajoXPnzowZMwZ3d3emT59Or169iI2N5csvv2TJkiVqGx06dODLL7+kdevW5ObmMmHChEJPrUvi6OhIVlYWe/fuxc3NDTMzM8zMzErVhhBCCCHKF5nD+5TZ29sTExNDbm4unTp1wtXVlaCgIKysrHQihTt06IC1tTXJycn06dNHpw1vb2+2b9/O7t27cXd3p1WrVixYsIBatWrpXYe1tTXfffcdO3bswNXVlXXr1hVasutx+9m4cSPu7u707t2bBg0aMH78ePVpa9OmTdmwYQPr16/ntddeY8qUKUybNo2AgAD1/NDQUBwcHHjjjTfo06cPY8eOLfVgtU2bNgwbNoxevXpRtWpV5syZU6rzhRBCCFH+SNJaGePp6Unjxo1ZuHDh8y6lXJGkNSGEEKLs0ff3tzzhFUIIIYQQ5ZrM4RUlysnJ4X//+x8ffPBBkftr1arFqVOnnnFVQgjxbBQXLSwvtAlRNsgT3jIoLy9PfQGuevXqOnNx09PT8fPzQ6vVYmlpSc+ePfn777/V/QEBAXTt2lWnvaCgIDw9PdXvRcUjv/3222oi2oOfHTt2oNFoWLp0KT4+PpiZmeHi4kJsbCxnz57F09MTc3Nz2rRpo649nK+kKOP58+fj6uqKubk5Dg4OBAYGkpWVpe7Pj0zetWsXLi4uaLVaOnfurLPaQ1EkaU0IIYR4eciAtwxatWoV5ubmxMXFMWfOHKZNm0ZkZCR5eXn4+flx9epVoqOjiYyM5Ny5c/Tq1euR+siPR/7mm2+wsLDAycmpyE/+S23Tp0/H39+fhIQE6tevT58+ffjggw+YOHEi8fHxKIrCiBEj1D70iTI2MDBg0aJFnDp1ilWrVrFv3z7Gjx+vU+u///7LvHnzWLNmDQcOHCA9Pf2hkcL5Zs6cSaVKldSPg4NDqe+REEIIIcoGeWmtjPH09CQ3N5eDBw+q21q0aEGHDh3o2LEjb775JmlpaeoA7vTp0zRs2JBDhw7h7u5OQEAA169fZ8uWLer5QUFBJCQkEBUVpfZx48YNjh49qnddGo2GyZMnM336dAB+/fVXWrduzYoVKxg4cCAA69evZ8CAAdy+fRsALy8vOnbsyMSJE9V2vvvuO8aPH8+ff/5ZZD8//vgjw4YN459//gHuP+EdMGAAZ8+epU6dOgAsWbKEadOm8ddffz203uzsbDXZDu5PendwcJCX1oQQhciUBiFeXPq+tCZzeMuggtG6cD+x7dKlSyQlJeHg4KDztLJBgwZYWVmRlJSEu7u73n0UjEd+lLpsbW0BcHV11dl2584dbty4gaWlJYmJicTExOg80c3NzeXOnTv8+++/mJmZsWfPHmbOnMmZM2e4ceMG9+7d09kPYGZmpg52C96P4kjSmhBCCPHykAFvGVRchHBJ9IkYBt145EepKz/yt7gY4JKijM+fP4+Pjw8ffvghn332GdbW1vz8888MGjSInJwcdcBb1P2QP1wIIZ4UeYorRNknA95yxMXFhYsXL3Lx4kWdKQ3Xr1+nQYMGwP343pMnT+qcl5CQUOpEsyehpCjjI0eOkJeXR2hoqBrSsWHDhmdZohBCCCHKAXlprRzx8vLC1dWVvn37cvToUQ4dOoS/vz8eHh40b94cuJ/oFh8fz+rVq0lJSSE4OLjQAPhZmTJlCqtXr2bq1KmcOnWKpKQk1q9fz+TJkwFwcnLi7t27LF68mHPnzrFmzRq++eab51KrEEIIIcouGfCWIxqNhq1bt1K5cmXatWuHl5cXtWvX5ocfflCP8fb25tNPP2X8+PG4u7tz8+ZN/P39n0u9JUUZu7m5MX/+fGbPns1rr73G2rVrmTlz5nOpVQghhBBll6zSIAQSLSyEEEKURbJKQxmRk5ODsbHx8y5DCCHEQzy4LJm8xCZE2SNTGp6xolLMTp48yZtvvolWq8XW1pZ+/fqp68x+++232NvbF1qFwc/PT13fFkpOLNNoNCxfvpxu3bphZmZG3bp12bZtm7o/P7GsoC1btqDRaFi7di1arRatVoupqSmGhoZoNBoMDAyoVq2aTj/FeZQ0ttTUVPz8/LC1tUWr1eLu7s6ePXvU/WfOnMHMzIzvv/9e3bZhwwZMTU05ffq0XnUJIYQQonyTAe9zUDDFbNasWXTo0IEmTZoQHx9PREQEf//9Nz179gSgR48eXLlyhf3796vnX716lYiICPr27Qvol1gGMHXqVHr27Mnx48fp0qULffv25erVqyXWmx8rvGzZMoyMjJg5cyZ79+5l5cqVmJiYFOqnOKVNY8vKyqJLly7s3buXY8eO0blzZ3x9fUlPTwegfv36zJs3j8DAQNLT0/n9998ZNmwYs2fPVlemKIpECwshhBAvEUU8Ux4eHkqTJk3U79OnT1c6deqkc8zFixcVQElOTlYURVH8/PyUgQMHqvuXLl2q2NvbK7m5uYqiKErHjh2Vzz//XKeNNWvWKHZ2dup3QJk8ebL6PSsrSwGUnTt3KoqiKGFhYUqlSpV02ti8ebNS8D8RffopzoM1xMbGKoCyYsUKddu6deuUihUrFttOw4YNlcWLF+tse+utt5Q33nhD6dixo9KpUyclLy+v2DaCg4MVoNAnMzNTr2sRQrw8ak3YrvMRQrw4MjMz9fr9LXN4n4OCKWaJiYns378frVZb6LjU1FScnZ3p27cvQ4YMYcmSJZiYmLB27Vree+89dW1afRLLQDcJzdzcHEtLyxITyQrSt5/ilDaNLSsri5CQEH766ScyMjK4d+8et2/fVp/w5lu5ciXOzs4YGBhw6tQpNeTiYSZOnMjo0aPV7/nRwkIIIYQof2TA+xwUTDHLysrC19eX2bNnFzrOzs4OAF9fXxRF4aeffsLd3Z2DBw+yYMECnTaKSyzLV1xCmz4JbPr2U5zSprGNHTuWyMhI5s2bh5OTE6amprz77rvk5OTotJuYmMitW7cwMDAgIyNDvXcPI9HCQgghxMtDBrzPWdOmTdm4cSOOjo4YGRX9z1GxYkW6d+/O2rVrOXv2LPXq1aNp06Y6bRSXWKaPqlWrcvPmTW7duqUOyBMSEgrV+rj9lFZMTAwBAQF069YNuD/oPn/+vM4xV69eJSAggEmTJpGRkaEGb5iamj6zOoUQ5ZesyiBE2ScvrT1nw4cP5+rVq/Tu3ZvDhw+TmprKrl27GDBgALm5uepxffv25aeffmLlypXqy2r5Skos00fLli0xMzPjk08+ITU1le+//57w8PAn3k9p1a1bl02bNpGQkEBiYiJ9+vQptGLFsGHDcHBwYPLkycyfP5/c3FzGjh371GoSQgghRNkiA97nzN7enpiYGHJzc+nUqROurq4EBQVhZWWlztGF+5HA1tbWJCcn06dPH502Skos04e1tTXfffcdO3bswNXVlXXr1hESEvLE+ymt+fPnU7lyZdq0aYOvry/e3t46T7dXr17Njh07WLNmDUZGRpibm/Pdd9+xbNkydu7c+dTqEkIIIUTZ8dInrUVFRdG+fXuuXbtWaB3aghwdHQkKCiIoKOiZ1SaeHUlaE0IIIcoefX9/v/RPeNu0aUNGRgaVKlUCig5gADh8+DBDhw59xtW92DQaDVu2bHneZQghhBBCFOulf2nN2NiY6tWrl3hc1apVn0E1ZdfatWv54IMPitxXq1YtTp069YwrEkKIJ0OihYUo+8rEE978ON4RI0ZQqVIlqlSpwqeffqouo3Xt2jX8/f2pXLkyZmZmvPnmm6SkpKjnX7hwAV9fXypXroy5uTkNGzZkx44dwP0pDRqNhuvXrxMVFcWAAQPIzMxEo9Gg0WjUeayOjo4sXLhQbTM9PR0/Pz+0Wi2Wlpb07NmTv//+W90fEhJC48aNWbNmDY6OjlSqVIn33nuPmzdvqsf8+OOPuLq6Ympqio2NDV5eXty6dUuve7Jy5UoaNmyIiYkJdnZ2OulkJdUWEBBA165dddoLCgrC09NT556PHDmS8ePHY21tTfXq1XXm9Do6OgLQrVs3NBoNEydOJCEhochP/r3OvycrV66kZs2aaLVaAgMDyc3NZc6cOVSvXp1q1aoVSm6bP38+rq6umJub4+DgQGBgIFlZWer+gQMH0qhRI7KzswHIycmhSZMm+Pv763UvhRBCCFG+lYkBL9yP4zUyMuLQoUN88cUXzJ8/n+XLlwP3B3Dx8fFs27aN2NhYFEWhS5cu6jqyw4cPJzs7mwMHDnDixAlmz55dZNBDmzZtWLhwIZaWlmRkZJCRkVHk2/55eXn4+flx9epVoqOjiYyM5Ny5c/Tq1UvnuNTUVLZs2cL27dvZvn070dHRzJo1C4CMjAx69+7NwIEDSUpKIioqiu7duxdaC7coX3/9NcOHD2fo0KGcOHGCbdu2qUuF6VubPlatWoW5uTlxcXHMmTOHadOmERkZCdyf4gEQFhZGRkYGR44cwcnJqchPwZfaUlNT2blzJxEREaxbt44VK1bw1ltv8fvvvxMdHc3s2bOZPHkycXFx6jkGBgYsWrSIU6dOsWrVKvbt28f48ePV/YsWLeLWrVt8/PHHAEyaNInr16/z5ZdfPvTaJFpYCCGEeIk87ci3J8HDw0NxcXHRiYudMGGC4uLiovz2228KoMTExKj7/vnnH8XU1FTZsGGDoiiK4urqqoSEhBTZ9v79+xVAuXbtmqIoRUfsKoqi1KpVS1mwYIGiKIqye/duxdDQUElPT1f3nzp1SgGUQ4cOKYpyP7rWzMxMuXHjhnrMuHHjlJYtWyqKoihHjhxRAOX8+fOlvh/29vbKpEmTitynT239+/dX/Pz8dM4bNWqU4uHhoX738PBQXn/9dZ1j3N3dlQkTJqjfAWXz5s16113UPfH29lYcHR3VmGRFUZR69eopM2fOfGg7//vf/xQbGxudbb/88otSoUIF5dNPP1WMjIyUgwcPllgLEi0shNCDRAsL8eLSN1q4zDzhbdWqlU5cbOvWrUlJSeH06dMYGRnRsmVLdZ+NjQ316tUjKSkJgJEjRzJjxgzatm1LcHAwx48ff6xakpKScHBw0ImibdCgAVZWVmqfcP/P/hYWFup3Ozs7NcrXzc2Njh074urqSo8ePVi2bBnXrl0rse9Lly7x559/0rFjx8eqTR8FY4AfrP9RPXhPbG1tadCggc4SbLa2tjr97Nmzh44dO/LKK69gYWFBv379uHLlCv/++696TOvWrRk7dizTp09nzJgxvP7668XWMXHiRDIzM9XPxYsXH+u6hBBCCPHiKjMD3scxePBgzp07R79+/Thx4gTNmzdn8eLFT73f4qJ8DQ0NiYyMZOfOnTRo0IDFixdTr1490tLSim3zSaSH6RMjDMXX/6iKarO4fs6fP4+Pjw+NGjVi48aNHDlyhK+++gpAJ144Ly+PmJgYDA0NOXv2bIl1mJiYYGlpqfMRQoiinJ/1ls5HCFH2lJkBb8E5nQC//vordevWpUGDBty7d09n/5UrV0hOTqZBgwbqNgcHB4YNG8amTZsYM2YMy5YtK7IfY2NjnYSzori4uHDx4kWdp4KnT5/m+vXrOn2WRKPR0LZtW6ZOncqxY8cwNjZm8+bNxZ5jYWGBo6Mje/fufeTaqlatSkZGhs55D8YI66NChQol3qvHdeTIEfLy8ggNDaVVq1Y4Ozvz559/Fjpu7ty5nDlzhujoaCIiIggLC3uqdQkhhBCi7CgzA9709HRGjx5NcnIy69atY/HixYwaNYq6devi5+fHkCFD+Pnnn0lMTOT999/nlVdewc/PD7i/AsGuXbtIS0vj6NGj7N+/HxcXlyL7cXR0JCsri7179/LPP//o/Nk8n5eXF66urvTt25ejR49y6NAh/P398fDwoHnz5npdT1xcHJ9//jnx8fGkp6ezadMmLl++/NC6CgoJCSE0NJRFixaRkpLC0aNH1SfW+tTWoUMH4uPjWb16NSkpKQQHB3Py5Em96i4of+D9119/6TUd41E4OTlx9+5dFi9ezLlz51izZg3ffPONzjHHjh1jypQpLF++nLZt2zJ//nxGjRrFuXPnnkpNQgghhChbysyA19/fn9u3b9OiRQuGDx/OqFGj1CCIsLAwmjVrho+PD61bt0ZRFHbs2KH+qTw3N5fhw4fj4uJC586dcXZ2ZsmSJUX206ZNG4YNG0avXr2oWrUqc+bMKXSMRqNh69atVK5cmXbt2uHl5UXt2rX54Ycf9L4eS0tLDhw4QJcuXXB2dmby5MmEhoby5ptvlnhu//79WbhwIUuWLKFhw4b4+Pioy7DpU5u3tzeffvop48ePx93dnZs3bz7SEl6hoaFERkbi4OBAkyZNSn1+vq1bt6oBFufPnyc6OprLly8D9+c6z58/n9mzZ/Paa6+xdu1aZs6cqZ57584d3n//fQICAvD19QVg6NChtG/fnn79+j31J9BCCCGEePGViWhhT09PGjdurLMOrig/NBoNmzdvpmvXruTm5nL58mWqVKmCkdGzy0WRaGEhhBCi7NH39/dLn7RWGjk5ORgbGz/vMso1Q0NDvZLvhBDiWZGkNSHKvjIzpeF5yE94CwoKokqVKnh7e3Py5EnefPNNtFottra29OvXj3/++QeAb7/9Fnt7+0IrGfj5+TFw4ED1+9atW2natCkVK1akdu3aTJ06lXv37qn7NRoNFStWxMjICI1Gg4GBAaampmi1Wg4ePEh4eDhWVlY6fWzZskVn2TZ9+imORqNh6dKl+Pj4YGZmhouLC7GxsZw9exZPT0/Mzc1p06YNqamp6jkNGzbE1NQUQ0NDtW5jY2O0Wi1r164FICUlhXbt2lGxYkUaNGigBlnkO3/+PBqNRn2JLjc3l0GDBvHqq69iampKvXr1+OKLL3TOyU+OmzdvHnZ2dtjY2DB8+PAiV54QQgghxMunTAx4o6Kintt0hlWrVmFsbExMTAyzZs2iQ4cONGnShPj4eCIiIvj777/p2bMnAD169ODKlSvs379fPf/q1atERETQt29fAA4ePIi/vz+jRo3i9OnTLF26lPDw8EJxupUrV2bu3Lns2bOHfv36YWBgQFRUlN4vxenbT3GmT5+Ov78/CQkJ1K9fnz59+vDBBx8wceJE4uPjURRFJ9I4JCQEIyMjZs6cyd69e1m5ciXVqlVj4MCBvP322+Tl5dG9e3eMjY2Ji4vjm2++YcKECcXWkJeXR40aNfjf//7H6dOnmTJlCp988gkbNmzQOW7//v2kpqayf/9+Vq1aRXh4OOHh4Q9tV5LWhBBCiJfI08/AKLs8PDyUJk2aqN+nT5+udOrUSeeYixcvKoCSnJysKIqi+Pn5KQMHDlT3L126VLG3t1eTxDp27Kh8/vnnOm2sWbNGsbOzU78DyuTJk9XvWVlZCqDs3LlTUZSi0+A2b96sFPzn1Kef4jxYQ2xsrAIoK1asULetW7dOqVixot597tq1SzEyMlL++OMPdf/OnTt1EtvS0tIUQDl27NhDaxs+fLjyzjvvqN/79++v1KpVS7l37566rUePHkqvXr0e2oYkrQkh9CVJa0K8uPRNWpM5vCVo1qyZ+nNiYiL79+9Hq9UWOi41NRVnZ2f69u3LkCFDWLJkCSYmJqxdu5b33ntPTRJLTEwkJiZG50lrbm4ud+7c4d9//8XMzAzQTTkzNzfH0tKyVCln+vZTnII12NraAuDq6qqz7c6dO9y4cQNLS8sS+8xPgbO3t1f3t27dusQ6vvrqK1auXEl6ejq3b98mJyeHxo0b6xzTsGFDDA0N1e92dnacOHHioW1OnDiR0aNHq99v3Lihk04nhBBCiPJDBrwlMDc3V3/OysrC19eX2bNnFzrOzs4OAF9fXxRF4aeffsLd3Z2DBw+yYMECnTamTp1K9+7dC7VRsWJF9efi0sf0SUrTt5/iFKwhf35wUdvy63oSfT5o/fr1jB07ltDQUFq3bo2FhQVz584tFERS2lQ4ExMTTExMHqkmIYQQQpQtMuAthaZNm7Jx40YcHR0fumRWxYoV6d69O2vXruXs2bPUq1ePpk2b6rSRnJyMk5PTI9dRtWpVbt68ya1bt9QB+YNJaU+in9Iqqc/8FLiMjAz1fyD8+uuvxbYZExNDmzZtCAwMVLcVfFFOCCGeNlmVQYiyTwa8pTB8+HCWLVtG7969GT9+PNbW1pw9e5b169ezfPly9U/qffv2xcfHh1OnTvH+++/rtDFlyhR8fHyoWbMm7777LgYGBiQmJnLy5ElmzJihVx0tW7bEzMyMTz75hJEjRxIXF1foBa0n0U9pldSnl5cXzs7O9O/fn7lz53Ljxg0mTZpUbJt169Zl9erV7Nq1i1dffZU1a9Zw+PBhXn311adyDUIIIYQof8rEKg0vCnt7e2JiYsjNzaVTp064uroSFBSElZWVOkcX7kf3Wltbk5ycTJ8+fXTa8Pb2Zvv27ezevRt3d3datWrFggULqFWrlt51WFtb891337Fjxw5cXV1Zt24dISEhT7yf0iqpTwMDAzZv3qwm5g0ePLjEVSM++OADunfvTq9evWjZsiVXrlzRedorhBBCCFGSMpG09qwUTHRzdHQkKCiIoKCg511WuRcQEMD169fVeOHnkawnSWtCCCFE2SNJa4/p8OHDOi+sveiioqJo3749165dKxRKUdZs2rSp0EtoQgghhBCPSga8D1G1atXnXcJTs3btWj744IMi99WqVYtTp04944p0WVtbP9f+hRCiIIkWFqLse2nn8N66dQt/f3+0Wi12dnaEhobq7Hd0dFT/pK4oCiEhIdSsWRMTExPs7e0ZOXKkemx2djYTJkzAwcEBExMTnJycWLFihbo/OjqaFi1aYGJigp2dHR9//LFOxG/BvvI1btxYZ16uRqNh+fLldOvWDTMzM+rWrcu2bduA+3G87du3B+4ntGk0GgICAgD48ccfcXV1xdTUFBsbG7y8vOjYsSMJCQlFfnbs2AH8/7jezz//HFtbW6ysrJg2bRr37t1j3LhxWFtbU6NGDcLCwnTqvnjxIj179sTKygpra2v8/Pw4f/68uj83N5fRo0djZWWFjY0N48ePL7TEmqenp85UkjVr1tC8eXMsLCyoXr06ffr00VmTOCoqCo1Gw969e2nevDlmZma0adOG5OTkB//ZhRBCCPESemkHvOPGjSM6OpqtW7eye/duoqKiOHr0aJHHbty4kQULFrB06VJSUlLYsmWLTgCDv78/69atY9GiRSQlJbF06VI1nOKPP/6gS5cuuLu7k5iYyNdff82KFSseaaWEqVOn0rNnT44fP06XLl3o27cvV69excHBgY0bNwKQnJxMRkYGX3zxBRkZGfTu3ZuBAweSlJREVFQU3bt3R6vV4uTkVOSn4Ett+/bt488//+TAgQPMnz+f4OBgfHx8qFy5MnFxcQwbNowPPviA33//Hbi/FrC3tzcWFhYcPHiQmJgYtFotnTt3JicnB4DQ0FDCw8NZuXIlP//8M1evXmXz5s3FXvfdu3eZPn06iYmJbNmyhfPnz6sD+oImTZpEaGgo8fHxGBkZMXDgwIe2KdHCQgghxEvkqWe+vYBu3rypGBsbKxs2bFC3XblyRTE1NVVGjRqlKIqi1KpVS1mwYIGiKIoSGhqqODs7Kzk5OYXaSk5OVgAlMjKyyL4++eQTpV69ekpeXp667auvvlK0Wq0aN1ywr3xubm5KcHCw+p0S4ob379+vAMq1a9fUY44cOaIAyvnz50u8Jw/Kj+vNr1FRFKVevXrKG2+8oX6/d++eYm5urqxbt05RlPsxwg9ea3Z2tmJqaqrs2rVLURRFsbOzU+bMmaPuv3v3rlKjRg3Fz89P3ebh4aH+OxTl8OHDCqDcvHlT59r37NmjHvPTTz8pgHL79u0i25BoYSGEviRaWIgXl77Rwi/lE97U1FRycnJo2bKlus3a2pp69eoVeXyPHj24ffs2tWvXZsiQIWzevFmdkpCQkIChoSEeHh5FnpuUlETr1q3VVDKAtm3bkpWVpT4Z1Vdp44bd3Nzo2LEjrq6u9OjRg2XLlnHt2jW9+2vYsKHOcmu2trY6T7YNDQ2xsbFRa0hMTOTs2bNYWFig1WrRarVYW1tz584dUlNTyczMJCMjQ+e+GxkZ0bx582LrOHLkCL6+vtSsWRMLCwv1Xqenp+scV/D+5AdbPOz+TJw4kczMTPVz8eJFfW6JEEIIIcogeWlNDw4ODiQnJ7Nnzx4iIyMJDAxk7ty5REdHY2pq+tjt6xMVDKWPzzU0NCQyMpJffvmF3bt3s3jxYiZNmkRcXJxewQ1F9VdcDVlZWTRr1oy1a9cWautRXwK8desW3t7eeHt7s3btWqpWrUp6ejre3t7qNImi6n0w9vhBEi0shNCXvKQmRNn3Uj7hrVOnDhUqVCAuLk7ddu3aNX777beHnmNqaoqvry+LFi0iKiqK2NhYTpw4gaurK3l5eURHRxd5nouLC7GxsToD2piYGCwsLKhRowZwfzCYkZGh7r9x4wZpaWmluiZjY2Pg/kthBWk0Gtq2bcvUqVM5duwYxsbGJc6ZfVRNmzYlJSWFatWqFZobXKlSJSpVqoSdnZ3Ofb937x5Hjhx5aJtnzpzhypUrzJo1izfeeIP69esX+1RbCCGEEOJBL+WAV6vVMmjQIMaNG8e+ffs4efIkAQEBOn++Lyg8PJwVK1Zw8uRJzp07x3fffYepqSm1atXC0dGR/v37M3DgQLZs2UJaWhpRUVFs2LABgMDAQC5evMhHH33EmTNn2Lp1K8HBwYwePVrtr0OHDqxZs4aDBw9y4sQJ+vfvr8YU66tWrVpoNBq2b9/O5cuXycrKIi4ujs8//5z4+HjS09PZtGkTly9fxsXF5fFu4EP07duXKlWq4Ofnx8GDB9V7MXLkSHX6xqhRo5g1axZbtmzhzJkzBAYGcv369Ye2WbNmTYyNjVm8eDHnzp1j27ZtTJ8+/anUL4QQQojy6aUc8ALMnTuXN954A19fX7y8vHj99ddp1qxZkcdaWVmxbNky2rZtS6NGjdizZw//93//h42NDQBff/017777LoGBgdSvX58hQ4Zw69YtAF555RV27NjBoUOHcHNzY9iwYZiamnL16lXg/pJklSpVwsPDAx8fH9566y26du1KnTp1SnU9r7zyClOnTuXjjz/G1taWESNGYGlpyYEDB+jSpQvOzs5MnjyZ0NBQ3nzzzce4cw9nZmbGgQMHqFmzJt27d8fFxYVBgwZx584dNf1kzJgx9OvXj/79+9O6dWssLCzo1q3bQ9usWrUq4eHh/O9//6NBgwbMmjWLcePGAciyY0IIIYTQi0QLPwcFo3MvX76Mubk5ZmZmz7ssvbwIiW65ublcvnyZKlWqYGT0ZKahS7SwEEIIUfZItHAZUZ4T3Z4WQ0NDqlev/rzLEEIIIUQZ8dJOaXhWXtREt/xlwwwNDTE2Nla/azQaxo8f/0QS3fKndRTnURLdzp8/j0ajISEhAZCkNSHE0+X48U86HyFE2SMD3qfsRU10y48SdnZ2ZtiwYep3gO+///6JJLrpO1umtIluDyNJa0IIIYQo0lOPwHiJSaJbyR4l0S0tLU0BlGPHjunUJElrQoinQZLWhHhxSdLaC0AS3fRT2kQ3feqWpDUhhBBC5JMB7wskP9FtyZIlmJqaEhgYSLt27bh7926ZSHTbuXMnDRo0YPHixdSrV0/v8IzSJrrp044+SWuWlpY6HyGEEEKUTzLgfYok0e3pJLoJIcSzdH7WWzofIUTZI8uSPUUFE91sbGyoVq0akyZNKjbRLTc3l5YtW2JmZqaT6GZjY6Mmui1atAg3NzcuXLjApUuX6NmzJ4GBgSxcuJCPPvqIESNGkJycXGSiW3h4OL6+vlhZWTFlypTHSnTr0qULpqamnDp1ir1799KpUyeqVatGXFzcU010E0IIIYQoDRnwPmVz584lKysLX19fLCwsGDNmDJmZmUUea2VlxaxZsxg9ejS5ubm4uroWSnT75JNPCAwM5MqVK9SsWZNPPvkE+P+JbuPGjcPNzQ1ra2sGDRrE5MmT1fYnTpxIWloaPj4+VKpUienTp5f6CW/BRLcBAwbg7+/PhAkTOHDgAAsXLuTGjRvUqlXrqSa6CSGEEEKUhiStiTInJCSELVu2qMuoPQmStCaEEEKUPfr+/pYBryhzsrKyyM7OVp98Pwky4BVCCCHKHokWLsNycnLUl8PKuvxgjKLs3LmTN95445HaLK5dIYR4kgqmq8lLa0KUTbJKwwvA09OTESNGEBQURJUqVfD29ubkyZO8+eabaLVabG1t6devH//88w8A3377Lfb29oWW3PLz89NJF9u6dStNmzalYsWK1K5dm6lTp+pEDWs0GpYvX15kjDDcf4nOyspKp48tW7borPVbUj/5CW5FfZo3b45Go2Hp0qX4+PhgZmamrjZx9uxZPD09MTc3p02bNqSmpqr9hYSE0LhxY/V7fjzxvHnzsLOzw8bGhuHDhxe55JoQQgghXj4y4H1BrFq1CmNjY2JiYpg1axYdOnSgSZMmxMfHExERwd9//03Pnj2B+wEVV65cYf/+/er5V69eJSIigr59+wJw8OBB/P39GTVqFKdPn2bp0qWEh4fz2Wef6fQ7derUImOE9VVSP05OTg/95K8tPH36dPz9/UlISKB+/fr06dOHDz74gIkTJxIfH4+iKIwYMaLYOvbv309qair79+9n1apVhIeHEx4e/tDjJVpYCCGEeIk87cg3UTIPDw+lSZMm6vfp06crnTp10jnm4sWLCqAkJycriqIofn5+ysCBA9X9S5cuVezt7dWI3o4dOyqff/65Thtr1qxR7Ozs1O+UECMcFhamVKpUSaeNzZs3KwX/s9Gnn+I8WENsbKwCKCtWrFC3rVu3TqlYsaL6PTg4WHFzc1O/58cT37t3T93Wo0cPpVevXg/tV6KFhRD6klhhIV5cEi1cxjRr1kz9OTExkf3796tzVbVaLfXr1wdQ/7Tft29fNm7cSHZ2NgBr167lvffeU9fcTUxMZNq0aTptDBkyhIyMDP7991+1r9LGCD9I336KU7AGW1tbAJ1oYVtbW+7cuVPsU9iGDRvqrClsZ2dX7HVItLAQQgjx8pCX1l4Q5ubm6s/56/bOnj270HF2dnYA+Pr6oigKP/30E+7u7hw8eJAFCxbotDF16lS6d+9eqI2KFSuqPxcX4atPFLG+/RSnqEjg0sQEP3h8/jnFHW9iYoKJiYle9QkhXm7yopoQZZ8MeF9ATZs2ZePGjTg6OmJkVPQ/UcWKFenevTtr167l7Nmz1KtXj6ZNm+q0kZycjJOT0yPXUbVqVW7evMmtW7fUAfmDa98+iX6EEEIIIZ4mGfC+gIYPH86yZcvo3bs348ePx9ramrNnz7J+/XqWL1+u/um+b9+++Pj4cOrUKd5//32dNqZMmYKPjw81a9bk3XffxcDAgMTERE6ePMmMGTP0qiM/4viTTz5h5MiRxMXFFXoR7En0I4QQQgjxNMkc3heQvb09MTEx5Obm0qlTJ1xdXQkKCsLKykqdowvQoUMHrK2tSU5Opk+fPjpteHt7s337dnbv3o27uzutWrViwYIF1KpVS+86rK2t+e6779ixYweurq6sW7eOkJCQJ96PEEIIIcTTJElrLwlPT08aN27MwoULn3cpj02ihYUQQggB+v/+lie8oswZO3Yse/fufd5lCCGEEKKMkAGveGpWrVqls1xZwU/Dhg0fuV2tVouNjc0TrFQIIR6uYLSwEKJskgHvSyQvL099Ca569eo683HT09Px8/NDq9ViaWlJz549+fvvv9X9+fG9BQUFBeHp6al+fzAiefny5Q+NFd6xYwfAc4sWlqQ1IYQQ4uUhA96XyKpVqzA3NycuLo45c+Ywbdo0IiMjycvLw8/Pj6tXrxIdHU1kZCTnzp2jV69ej9RHfkTy8uXLHxorXPCltucRLTxz5kwqVaqkfhwcHEp9rUIIIYQoG2RZspdIo0aNCA4OBqBu3bp8+eWX6lzYEydOkJaWpg78Vq9eTcOGDTl8+DDu7u5691G3bl3mzJlTqroGDBhAz549AZgwYQKtW7fm008/xdvbG4BRo0YxYMCAYtuoXLkyX375JYaGhtSvX5+33nqLvXv3MmTIkCKPnzhxIqNHj1a/37hxQwa9QgghRDklT3hfIgUjfOH/x+8mJSXh4OCgM+Br0KABVlZWJCUllaqPghHJj1LXs4oWNjExwdLSUucjhBBCiPJJBrwvkdLG7xakT8ww6EYkP0pdzypaWAgh9CXRwkKUfTLgFbi4uHDx4kUuXryobjt9+jTXr1+nQYMGwP2Y4YyMDJ3znuQ6uEIIIYQQT4sMeAVeXl64urrSt29fjh49yqFDh/D398fDw4PmzZsD91Pd4uPjWb16NSkpKQQHB3Py5MnnXLkQQgghRMlkwPuMeHp6EhQU9LzLKJJGo2Hr1q1UrlyZdu3a4eXlRe3atfnhhx/UY7y9vfn0008ZP3487u7u3Lx5E39//+dYtRBCCCGEfiRa+BkpT9G+5ZFECwshhBBlj0QLCyGEEE+AJK0JUfbJgPcZKi7pbP78+bi6umJubo6DgwOBgYFkZWWp+y9cuICvry+VK1fG3Nychg0bqmllxWnevDnz5s1Tv3ft2pUKFSqobf/+++9oNBrOnj0LwJIlS6hbty4VK1bE1taWd999V69r8/T05KOPPiIoKIjKlStja2vL4MGD0Wq1VKhQAY1Gg4GBAaampjrRwidPnuTNN99Eq9Via2tLv379+Oeff9R2IyIieP3117GyssLGxgYfHx+d1LXz58+j0WjYtGkT7du3x8zMDDc3N2JjY/WqWwghhBDlnwx4n6GHJZ3B/WW/Fi1axKlTp1i1ahX79u1j/Pjx6rnDhw8nOzubAwcOcOLECWbPno1Wqy2xTw8PD6KiogBQFIWDBw9iZWXFzz//DEB0dDSvvPIKTk5OxMfHM3LkSKZNm0ZycjIRERG0a9euVNdXpUoVDh06xEcffURYWBjNmjUjJCSEyMhIevfuTcWKFYmNjWXHjh1cv36dDh060KRJE+Lj44mIiODvv/9WQygAbt26xejRo4mPj2fv3r0YGBjQrVu3QkuOTZo0ibFjx5KQkICzszO9e/fm3r17D61VooWFEEKIl4gingkPDw/l9ddf19nm7u6uTJgwocjj//e//yk2Njbqd1dXVyUkJKTU/W7btk2pVKmScu/ePSUhIUGpXr26MmrUKLXfwYMHK3369FEURVE2btyoWFpaKjdu3Ch1Pw9e37179xRzc3OlX79+6raMjAwFUGJjYxVFUZTp06crnTp10mnn4sWLCqAkJycX2c/ly5cVQDlx4oSiKIqSlpamAMry5cvVY06dOqUASlJS0kPrDQ4OVoBCn8zMzFJfuxCifKs1YfvzLkEI8RCZmZl6/f6WJ7zP0MOSzgD27NlDx44deeWVV7CwsKBfv35cuXKFf//9F4CRI0cyY8YM2rZtS3BwMMePH9erzzfeeIObN29y7NgxoqOj8fDwwNPTU33qGx0djaenJwD/+c9/qFWrFrVr16Zfv36sXbtW7b+012doaIiNjU2hxDRAvebExET279+PVqtVP/Xr1wdQpy2kpKTQu3dvateujaWlJY6OjgCkp6c/tG87OzudfooyceJEMjMz1U/BNYiFEEIIUb7IgPcZelga2Pnz5/Hx8aFRo0Zs3LiRI0eO8NVXXwGQk5MDwODBgzl37hz9+vXjxIkTNG/enMWLF5fYp5WVFW5ubkRFRamD23bt2nHs2DF+++03UlJS8PDwAMDCwoKjR4+ybt067OzsmDJlCm5ubly/fv2Rr6+4xLSsrCx8fX1JSEjQ+aSkpKhTKXx9fbl69SrLli0jLi6OuLg4nftSVN/6JLNJtLAQQgjx8pAB7wvgyJEj5OXlERoaSqtWrXB2dubPP/8sdJyDgwPDhg1j06ZNjBkzhmXLlunVvoeHB/v37+fAgQN4enpibW2Ni4sLn332GXZ2djg7O6vHGhkZ4eXlxZw5czh+/Djnz59n3759T+xaC2ratCmnTp3C0dERJycnnY+5uTlXrlwhOTmZyZMn07FjR1xcXLh27dpTqUUIIR5GooWFKPtkwPsCcHJy4u7duyxevJhz586xZs0avvnmG51jgoKC2LVrF2lpaRw9epT9+/fj4uKiV/uenp7s2rULIyMjdcqAp6cna9euVZ/uAmzfvp1FixaRkJDAhQsXWL16NXl5edSrV+/JXWwBw4cP5+rVq/Tu3ZvDhw+TmprKrl27GDBgALm5uVSuXBkbGxu+/fZbzp49y759+xg9evRTqUUIIYQQ5ZcMeF8Abm5uzJ8/n9mzZ/Paa6+xdu1aZs6cqXNMbm4uw4cPx8XFhc6dO+Ps7MySJUv0av+NN94gLy9PZ3Dr6elJbm6uOn8X7k9/2LRpEx06dMDFxYVvvvmGdevWqUuIPWn29vbExMSQm5tLp06dcHV1JSgoCCsrKwwMDDAwMGD9+vUcOXKE1157jf/+97/MnTv3qdQihBBCiPJLktZeUuU5+S08PJygoCC95x6DJK0JIYQQZZEkrYlyx9HRsVwO0IUQQgjxdMmAt4wbNmyYzrJeBT/Dhg17In2kp6c/tA+tVltoibCHeXBlBSGEEEKIZ0EGvGXctGnTCi3rlf+ZNm1asecWF3Wcnp6On5+fGgPcrl079uzZo7bdqVMnWrVqRUJCAvb29sD9F+sKzgn29PRkxIgRBAUFUaVKFby9vYutR1EUQkJCqFmzJiYmJtjb2zNy5Ei1rQsXLvDf//4XjUajLj0G96cw1KxZEzMzM7p168aVK1dKvG+StCaEEEK8PIyedwHi8VSrVo1q1ao90rmrVq1i9OjRxMXFERsbS0BAAG3btqVjx47qYDc6Opp79+4xfPhwPv74YzWwwtLSkry8PJycnErs48MPPyQmJqbEejZu3MiCBQtYv349DRs25K+//iIxMRGATZs24ebmxtChQxkyZIh6TlxcHIMGDWLmzJl07dqViIgIgoODS+xr5syZTJ06tcTjhBBCCFH2yYD3JdaoUSN1cFi3bl2+/PJL9u7dC8CJEydIS0vDwcEBgNWrV9OwYUMOHz6Mu7u73n3UrVuXOXPm6HVseno61atXx8vLiwoVKlCzZk1atGgBgLW1NYaGhlhYWFC9enX1nC+++ILOnTszfvx4AJydnfnll1+IiIgotq+JEyfqLHF248YN9VqFEEIIUb7IlIaX2MOijpOSknBwcNAZADZo0AArKyuSkpJK1UezZs30PrZHjx7cvn2b2rVrM2TIEDZv3sy9e/eKPScpKYmWLVvqbGvdunWJfUnSmhBCCPHykAHvS+xhUcf6MDAw4MEV7e7evVvoOHNzc73rcXBwIDk5mSVLlmBqakpgYCDt2rUrsl0hhBBCCH3JgFcU4uLiwsWLF7l48aK67fTp01y/fp0GDRoAULVqVTIyMnTOS0hIeOy+TU1N8fX1ZdGiRURFRREbG8uJEycAMDY2Jjc3t1CtcXFxOtt+/fXXx65DCCGEEOWHDHhFIV5eXri6utK3b1+OHj3KoUOH8Pf3x8PDg+bNmwPQoUMH4uPjWb16NSkpKQQHB3Py5MnH6jc8PJwVK1Zw8uRJzp07x3fffYepqSm1atUC7q/De+DAAf744w/++ecfAEaOHElERATz5s0jJSWFL7/8ssT5u0IIIYR4uZT5Aa+npydBQUGP1YaiKAwdOhRra2s0Gs0TeVL5ME+i3qdNo9GwdetWKleuTLt27fDy8qJ27dr88MMP6jHe3t58+umnjB8/Hnd3d27evIm/v/9j9WtlZcWyZcto27YtjRo1Ys+ePfzf//0fNjY2wP0l2M6fP0+dOnWoWrUqAK1atWLZsmV88cUXuLm5sXv3biZPnvxYdQghhBCifCnz0cJXr16lQoUKWFhYPHIbO3fuxM/Pj6ioKGrXrk2VKlUwMno6C1iUhUjfqKgo2rdvz7Vr17Cysnre5TwTEi0shBBClD0vTbSwtbX1Yw12AVJTU7Gzs6NNmzZUr179qQ12ywJ5QUwIIf4/x49/et4lCCGegDI/4C04RWDJkiXUrVuXihUrYmtry7vvvlvi+QEBAXz00Uekp6ej0WhwdHQE7idxjRw5kmrVqlGxYkVef/11Dh8+rHNudHQ0LVq0wMTEBDs7Oz7++GOdZbRu3bqFv78/Wq0WOzs7QkNDS3Vt2dnZTJgwAQcHB0xMTHBycmLFihV69+/o6FjoSXLjxo11EtU0Gg1ff/01b7/9Nubm5gwZMoT27dsDULlyZTQaDQEBASXW6unpyUcffURQUBCVK1fG1taWZcuWcevWLQYMGICFhQW2traYmprqxBKbmZlhZGSEgYEBtra29OvXT52fCxAREcHrr7+OlZUVNjY2+Pj4kJqaqu4/f/48Go2GTZs20b59e8zMzHBzcyM2NrZU91oIIYQQ5VeZH/Dmi4+PZ+TIkUybNo3k5GQiIiJo165died98cUXTJs2jRo1apCRkaEOasePH8/GjRtZtWoVR48excnJCW9vb65evQrAH3/8QZcuXXB3dycxMZGvv/6aFStWMGPGDLXtcePGER0dzdatW9m9ezdRUVEcPXpU72vy9/dn3bp1LFq0iKSkJJYuXYpWq9W7f32FhITQrVs3Tpw4wdSpU9m4cSMAycnJZGRk8MUXX+jVzqpVq6hSpQqHDh3io48+4sMPP6RHjx60adOGo0eP4uvrS8WKFYmNjSUhIYEDBw5gamrK4MGDiYyMJCIigr///puePXuqbd66dYvRo0cTHx/P3r17MTAwoFu3boWWT5s0aRJjx44lISEBZ2dnevfuXewavhItLIQQQrxElDLOw8NDGTVqlLJx40bF0tJSuXHjRqnbWLBggVKrVi31e1ZWllKhQgVl7dq16racnBzF3t5emTNnjqIoivLJJ58o9erVU/Ly8tRjvvrqK0Wr1Sq5ubnKzZs3FWNjY2XDhg3q/itXriimpqbKqFGjSqwpOTlZAZTIyMgi95fUv6IoSq1atZQFCxbonOfm5qYEBwer3wElKChI55j9+/crgHLt2rUS68zn4eGhvP766+r3e/fuKebm5kq/fv3UbRkZGQqgxMbGKoqiKNOnT1c6deqk087FixcVQElOTi6yn8uXLyuAcuLECUVRFCUtLU0BlOXLl6vHnDp1SgGUpKSkh9YbHBysAIU+mZmZel+zEKL8qzVh+/MuQQhRjMzMTL1+f5ebJ7z/+c9/qFWrFrVr16Zfv36sXbuWf//995HaSk1N5e7du7Rt21bdVqFCBVq0aKEmjSUlJdG6dWs0Go16TNu2bcnKyuL3338nNTWVnJwcnRQwa2tr6tWrp1cNCQkJGBoa4uHhUeT+kvovjfylxh5XweQ2Q0NDbGxscHV1VbfZ2toCcOnSJQASExPZv3+/zhSH+vXrA6jTFlJSUujduze1a9fG0tJSnXKSnp7+0L7t7Ox0+inKxIkTyczMVD8F1xwWQgghRPlSbt7OsrCw4OjRo0RFRbF7926mTJlCSEgIhw8fLpMrDZiamj52G08jDa04RSW3FdyWPzjPn46QlZWFr68vs2fPLtRW/qDV19eXWrVqsWzZMuzt7cnLy+O1114jJyfnoX0/2E9RTExMMDExKc3lCSGEEKKMKjdPeAGMjIzw8vJizpw5HD9+nPPnz7Nv375St1OnTh2MjY2JiYlRt929e5fDhw+rSWMuLi7ExsbqDChjYmKwsLCgRo0a1KlThwoVKuikgF27do3ffvtNrxpcXV3Jy8sjOjq6yP0l9Q+F09Bu3LhBWlpaiX0bGxsDFEo1e9KaNm3KqVOncHR0xMnJSedjbm7OlStXSE5OZvLkyXTs2BEXFxeuXbv2VGsSQoiCzs9663mXIIR4AsrNgHf79u0sWrSIhIQELly4wOrVq8nLy9N7CkFB5ubmfPjhh4wbN46IiAhOnz7NkCFD+Pfffxk0aBAAgYGBXLx4kY8++ogzZ86wdetWgoODGT16NAYGBmi1WgYNGsS4cePYt28fJ0+eJCAgAAMD/W65o6Mj/fv3Z+DAgWzZsoW0tDSioqLYsGGDXv3D/TS0NWvWcPDgQU6cOEH//v0xNDQsse9atWqh0WjYvn07ly9fJisrq9T3UB/Dhw/n6tWr9O7dm8OHD5OamsquXbsYMGAAubm5VK5cGRsbG7799lvOnj3Lvn37GD169FOpRQghhBDlV7mZ0mBlZcWmTZsICQnhzp071K1bl3Xr1tGwYcNHam/WrFnk5eXRr18/bt68SfPmzdm1axeVK1cG4JVXXmHHjh2MGzcONzc3rK2tGTRokE7K19y5c9U/21tYWDBmzBgyMzP1ruHrr7/mk08+ITAwkCtXrlCzZk0++eQTvfufOHEiaWlp+Pj4UKlSJaZPn67XE95XXnmFqVOn8vHHHzNgwAD8/f0JDw/Xu2592dvbExMTw4QJE+jUqRPZ2dnUqlWLzp07Y2BggEajYf369YwcOZLXXnuNevXqsWjRIjw9PZ94LUIIIYQov8p80lpZVxaS1542jUbD5s2b6dq1K+fPn+fVV1/l2LFjNG7c+JnVIElrQgghRNmj7+/vcvOEV5QPDg4OZGRkUKVKleddihBCCCHKiXIzh/dh0tPTdZa9evDz4PJWz8rBgwfRarUcPHiQJUuWFKrraXpwhYOSPMt7aGho+NLHOwshXgyOH/8k0cJClBPlfsBrb29PQkLCQz/29vbPpa7mzZuTkJBA8+bNee+99+jduzeGhoZUrFiRgQMHqselp6fj5+eHVqvF0tKSnj178vfff6v7AwIC6Nq1q07bQUFBOvNcPT09GTFiBEFBQVSpUgVvb+8S69NoNCxduhQfHx/q1auHra0tYWFhbN26lYYNG5Kbm0vdunXZunWrzj3cunUrTZs2pWLFitSuXZupU6fqJJ6lpKTQrl07KlasSIMGDYiMjNTpNz8qOCEhAbi/UsSgQYN49dVXMTU1pV69eoWS3/Lvwbx587Czs8PGxobhw4cXuQSbEEIIIV4+5f4xmpGREU5OTs+7jEJMTU1xcnLC1NSUrVu3qvG5sbGxBAQE4OvrS8eOHdXBbnR0NPfu3WP48OH06tWLqKioUvW3atUqPvzwQ52l1koyffp05s+fz/z585kwYQLjx4+ndu3aTJs2jZo1azJw4EDmzZtHx44dgftPrf39/Vm0aBFvvPEGqampDB06FIDg4GDy8vLo3r07tra2xMXFkZmZSVBQULE15OXlUaNGDf73v/9hY2PDL7/8wtChQ7Gzs9OJIN6/fz92dnbs37+fs2fP0qtXLxo3bsyQIUOKbDc7O5vs7Gz1u0QLCyGEEOXY0w99E8V5MJJXURTF3d1dmTBhgrJ7927F0NBQSU9PV/flx+YeOnRIURRF6d+/v+Ln56dz/qhRoxQPDw+dPpo0aVKqugBl8uTJ6vfY2FgFUFasWKFuW7dunVKxYkX1e8eOHZXPP/9cp501a9YodnZ2iqIoyq5duxQjIyPljz/+UPfv3LlTAZTNmzcrivL/o4KPHTv20NqGDx+uvPPOO+r3/v37K7Vq1VLu3bunbuvRo4fSq1evh7Yh0cJCiJLUmrBdooWFeMG9dNHCZVnBWFy4nzJ26dIlkpKScHBwwMHBQd3XoEEDrKys1IhjfTVr1uyx6sqPBX4wKvjOnTvq09HExESmTZumM793yJAhZGRk8O+//6rXU3AKROvWrUus46uvvqJZs2ZUrVoVrVbLt99+W2jecMOGDXXWGM6/hw8j0cJCCCHEy6PcT2koC4qK5C0uFregpxkfXFRcb0lRwVOnTqV79+6F2qpYsWKp+wdYv349Y8eOJTQ0lNatW2NhYcHcuXN1EuwerCu/NokWFkI8DklZE6L8kAHvC8zFxYWLFy9y8eJF9Snv6dOnuX79uhpxXLVqVU6ePKlzXkJCQqEB4LPQtGlTkpOTHzpnOv96MjIysLOzA+DXX38tts2YmBjatGlDYGCgui01NfXJFS2EEEKIck+mNLzAvLy8cHV1pW/fvhw9epRDhw7h7++Ph4cHzZs3B+7HB8fHx7N69WpSUlIIDg4uNAB+VqZMmcLq1auZOnUqp06dIikpifXr16vpb15eXjg7O9O/f38SExM5ePAgkyZNKrbNunXrEh8fz65du/jtt9/49NNPOXz48LO4HCGEEEKUE2VmwBsVFYVGo+H69evFHufo6FhuUss0Gg1bt26lcuXKtGvXDi8vL2rXrs0PP/ygHuPt7c2nn37K+PHjcXd35+bNm/j7+z+Xer29vdm+fTu7d+/G3d2dVq1asWDBAmrVqgXcn36xefNmbt++TYsWLRg8eDD169fXaWPs2LE63z/44AO6d+9Or169aNmyJVeuXNF52iuEEEIIUZIyEy2ck5PD1atXsbW1RaPREB4eTlBQUKEB8OXLlzE3N8fMzOz5FPoCKhjd+6IJCQlhy5Yt6rq7mZmZKIqClZXVM61DooWFEEKIsqfcRQsbGxtTvXr1Eo+rWrXqM6hGPC2VKlV63iUIIYQQopx5olMa8hO9RowYQaVKlahSpQqffvqpuorAtWvX8Pf3p3LlypiZmfHmm2+SkpKinn/hwgV8fX2pXLky5ubmNGzYkB07dgC6UxqioqIYMGAAmZmZaDQaNBoNISEhQOEpDSUllYWEhNC4cWPWrFmDo6MjlSpV4r333uPmzZvqMT/++COurq6YmppiY2ODl5cXt27d0uuerFy5koYNG2JiYoKdnR0jRozQuzZ9U9RGjhzJ+PHjsba2pnr16uq9yL8fAN26dUOj0eDo6MjatWsfGhPcsGFD9Z6sXLmSmjVrotVqCQwMJDc3lzlz5lC9enWqVavGZ599plPb9evXGTx4MFWrVsXS0pIOHTqQmJioc8ysWbOwtbXFwsKCQYMGcefOHZ39D15zREQEr7/+OlZWVtjY2ODj46Pz0lp+MtumTZto3749ZmZmuLm5ERsbq88/jxBCFCk/VliihYUoH574HN5Vq1ZhZGTEoUOH+OKLL5g/fz7Lly8H7g9m4uPj2bZtG7GxsSiKQpcuXdRltIYPH052djYHDhzgxIkTzJ49G61WW6iPNm3asHDhQiwtLcnIyCAjI6PQ3E+4v1yWn58fV69eJTo6msjISM6dO0evXr10jktNTWXLli1s376d7du3Ex0dzaxZswDIyMigd+/eDBw4kKSkJKKioujevXuhpcCK8vXXXzN8+HCGDh3KiRMn2LZtm7qCgb616WPVqlWYm5sTFxfHnDlzmDZtmhrZm/+CV1hYGBkZGRw+fJi33377oVHL+f8DIzU1lZ07dxIREcG6detYsWIFb731Fr///jvR0dHMnj2byZMn6ywP1qNHDy5dusTOnTs5cuQITZs2pWPHjly9ehWADRs2EBISwueff058fDx2dnYsWbKk2Gu7deuWmkK3d+9eDAwM6NatW6ElxyZNmsTYsWNJSEjA2dmZ3r1760QaPyg7O5sbN27ofIQQQghRTj3JtAsPDw/FxcVFycvLU7dNmDBBcXFxUX777TcFUGJiYtR9//zzj2Jqaqps2LBBURRFcXV1VUJCQopse//+/QqgXLt2TVEURQkLC1MqVapU6LhatWopCxYsUBRF0SupLDg4WDEzM1Nu3LihHjNu3DilZcuWiqIoypEjRxRAOX/+fKnvh729vTJp0qQi9z3JFLWHJbXlo0CSmT6Kuife3t6Ko6Ojkpubq26rV6+eMnPmTEVRFOXgwYOKpaWlcufOHZ226tSpoyxdulRRFEVp3bq1EhgYqLO/ZcuWipubm/q9qGsu6PLlywqgnDhxQlGU/5/Mtnz5cvWY/PuYlJRU7DUiSWtCiIfIT1mTpDUhXmzPLWmtVatWaiAB3E/SSklJ4fTp0xgZGdGyZUt1n42NDfXq1VNTw0aOHMmMGTNo27YtwcHBHD9+/LFq0TepzNHREQsLC/V7wZQuNzc3OnbsiKurKz169GDZsmVcu3atxL4vXbrEn3/+SceOHR+rNn08LKntcTx4T2xtbWnQoAEGBgY62/L7SUxMJCsrCxsbG50pEmlpaeoUhKSkJJ1/fyg5aS0lJYXevXtTu3ZtLC0t1SkaDyatFbwH+Wv8StKaEEIIIeAFW5Zs8ODBnDt3jn79+nHixAmaN2/O4sWLn3q/xaV0GRoaEhkZyc6dO2nQoAGLFy+mXr16pKWlFdumqanpY9elb4ra4yS1PUxRbRbXT1ZWFnZ2doWmSCQnJzNu3LhHrsPX15erV6+ybNky4uLi1CkUOTk5D633wQS4opiYmGBpaanzEUIIIUT59MQHvA9Gvv7666/UrVuXBg0acO/ePZ39V65cITk5WU0NA3BwcGDYsGFs2rSJMWPGsGzZsiL7MTY2Jjc3t9haCiaV5XswqUwfGo2Gtm3bMnXqVI4dO4axsTGbN28u9hwLCwscHR3Zu3fvI9dWtWpVMjIydM7LX76rNCpUqFDivXpcTZs25a+//sLIyAgnJyedT5UqVYD711zUfx8Pk//fx+TJk+nYsSMuLi56PV0XQojHdX7WW+pHCFH2PfEBb3p6OqNHjyY5OZl169axePFiRo0aRd26dfHz82PIkCH8/PPPJCYm8v777/PKK6/g5+cH3F+BYNeuXaSlpXH06FH279+Pi4tLkf04OjqSlZXF3r17+eeff/j3338LHaNPUllJ4uLi1Jes0tPT2bRpE5cvX35oXQWFhIQQGhrKokWLSElJ4ejRo+oT62eZopY/8P7rr7+e2oDRy8uL1q1b07VrV3bv3s358+f55ZdfmDRpEvHx8QCMGjWKlStXEhYWxm+//UZwcDCnTp16aJuVK1fGxsaGb7/9lrNnz7Jv3z5Gjx79VOoXQgghRPn1xAe8/v7+apLW8OHDGTVqFEOHDgXurxTQrFkzfHx8aN26NYqisGPHDvXP0bm5uQwfPhwXFxc6d+6Ms7PzQ9/ib9OmDcOGDaNXr15UrVqVOXPmFDpGn6SyklhaWnLgwAG6dOmCs7MzkydPJjQ0lDfffLPEc/v378/ChQtZsmQJDRs2xMfHR12G7VmmqIWGhhIZGYmDgwNNmjQp9fn60Gg07Nixg3bt2jFgwACcnZ157733uHDhAra2tgD06tVLvZ5mzZpx4cIFPvzww4e2aWBgwPr16zly5AivvfYa//3vf5k7d+5TqV8IIYQQ5dcTTVrz9PSkcePG5Sbat6ySf4fSk6Q1IYQQouzR9/f3C/XSmhBCCCGEEE9amYkWfhEVFYqRb+fOnbzxxhvPsBr9NWzYkAsXLhS5b+nSpfTt27fUbebk5GBsbPy4pQkhxAuhYMKavLgmRNn3RJ/wRkVFvVR/Rn9YWllCQoLeL8U9LXl5eQ+NG162bBktW7ZUl+1q164de/bsISEhgbffflvvSOMRI0YQFBRElSpV8Pb2LrEmjUbD0qVL8fHxwczMDBcXF2JjYzl79iyenp6Ym5vTpk0bnejg1NRU/Pz8sLW1RavV4u7uzp49e9T9Z86cwczMjO+//17dtmHDBkxNTTl9+nQp75oQQgghyiOZ0vAYHlx+q+DnSazD+zgeFjecl5fH8OHDycnJ4eDBg+zdu5dLly7x8ccf4+TkpBM2oU8fxsbGxMTE8M033+h1zvTp0/H39ychIYH69evTp08fPvjgAyZOnEh8fDyKojBixAj1+KysLLp06cLevXs5duwYnTt3xtfXVw2eqF+/PvPmzSMwMJD09HR+//13hg0bxuzZs4tdek6ihYUQQoiXh0xpKKcaNWpEcHAwAHXr1uXLL79U1wQ+ceIEaWlpasrb6tWradiwIYcPH8bd3V3vPurWrVvk6hjFGTBgAD179gRgwoQJtG7dmk8//VR9Qjxq1CgGDBigHu/m5oabm5v6ffr06WzevJlt27apA+PAwEB27NjB+++/j7GxMe7u7nz00UfF1jFz5kymTp1aqtqFEEIIUTbJE95y6mFxw08y0rhZs2aPVVf+cmWurq462+7cuaM+cc3KymLs2LG4uLhgZWWFVqslKSmpULTwypUrOX78OEePHiU8PFwn3rooEi0shBBCvDzkCW859Thxw/pGGpubmz9WXfmD0uJigceOHUtkZCTz5s1Tp4q8++67haKFExMTuXXrFgYGBmRkZGBnZ1dsHSYmJpiYmJS6fiHEy0FeVBOifJEB70umYKRx/lPeoiKNH0x0S0hIKDSIfhZiYmIICAigW7duwP0nvufPn9c55urVqwQEBDBp0iQyMjLU9LrnPY9aCCGEEC8GmdLwknmWkcZPQt26ddm0aRMJCQkkJibSp0+fQk+qhw0bhoODA5MnT2b+/Pnk5uYyduzY51KvEEIIIV485W7A6+npSVBQkF7HhoeHY2Vl9VTreRZ9lMazjDTW1zfffEPjxo11tuWvBTx//nwqV65MmzZt8PX1xdvbm6ZNm6rHrV69mh07drBmzRqMjIwwNzfnu+++Y9myZezcufOp1SyEEEKIsuOJRgu/CEoTqxseHk5QUBDXr18HICQkhC1btpCQkPDE6nmwD1HYg/c9MzMTRVGe6f9QkGhhIYQQouzR9/e3zOEtp3Jzc9FoNBgYlL2H+JUqVXreJQghhBCiHCl7o6ECbt26hb+/P1qtFjs7O0JDQ3X2Z2dnM3bsWF555RXMzc1p2bIlUVFRRbYVHh7O1KlTSUxMRKPRoNFoCA8PB+7/Wd3V1RVzc3McHBwIDAwkKyurVLXu2rULFxcXtFotnTt3JiMjQ92Xl5fHtGnTqFGjBiYmJjRu3JiIiAh1f1RUFBqNRucpcUJCAhqNRn2BK3/qxLZt22jQoAEmJiakp6cTFRVFixYtMDc3x8rKirZt2z40VrigkJAQGjduzMqVK6lZsyZarZbAwEByc3OZM2cO1atXp1q1anz22WcArF27Fq1Wi1arpUKFCuo9NDIyok6dOjptz5o1C1tbWywsLBg0aBB37tzR2f9g0ltERASvv/46VlZW2NjY4OPjo5PGdv78eTQaDZs2baJ9+/aYmZnh5uZGbGxsidcphBAPcvz4J52PEKLsK9MD3nHjxhEdHc3WrVvZvXs3UVFRHD16VN0/YsQIYmNjWb9+PcePH6dHjx507tyZlJSUQm316tWLMWPG0LBhQzIyMsjIyKBXr17A/WW6Fi1axKlTp1i1ahX79u1j/Pjxetf577//Mm/ePNasWcOBAwdIT0/Xeanqiy++IDQ0lHnz5nH8+HG8vb15++23i6yzpH5mz57N8uXLOXXqFNbW1nTt2hUPDw+OHz9ObGwsQ4cOLXGN2nypqans3LmTiIgI1q1bx4oVK3jrrbf4/fffiY6OZvbs2UyePJm4uDjefvttEhIScHNzo127dmzatInIyEj69+/P1atXuXr1KnA/9jckJITPP/+c+Ph47OzsWLJkSbF13Lp1i9GjRxMfH8/evXsxMDCgW7duhV5emzRpEmPHjiUhIQFnZ2d69+7NvXv3imxTktaEEEKIl4hSRt28eVMxNjZWNmzYoG67cuWKYmpqqowaNUq5cOGCYmhoqPzxxx8653Xs2FGZOHGioiiKEhYWplSqVEndFxwcrLi5uZXY9//+9z/FxsZGrzrDwsIUQDl79qy67auvvlJsbW3V7/b29spnn32mc567u7sSGBioKIqi7N+/XwGUa9euqfuPHTumAEpaWppOPwkJCeoxV65cUQAlKipKr1oLCg4OVszMzJQbN26o27y9vRVHR0clNzdX3VavXj1l5syZiqIoysGDBxVLS0vlzp07Om3VqVNHWbp0qaIoitK6dWv1uvK1bNlS5773799f8fPze2htly9fVgDlxIkTiqIoSlpamgIoy5cvV485deqUAihJSUkPvT6g0CczM7OYuyKEeBnUmrBd5yOEeHFlZmbq9fu7zD7hTU1NJScnh5YtW6rbrK2tqVevHnA/Pjc3NxdnZ2f1T+1arZbo6GidP4frY8+ePXTs2JFXXnkFCwsL+vXrx5UrV/j333/1Ot/MzEznz/r5qWdwf7L1n3/+Sdu2bXXOadu2bamTz4yNjXWSzKytrQkICMDb2xtfX1+++OILnakUJXF0dMTCwkL9bmtrS4MGDXTmBdva2qrXkpiYSFZWFjY2Njr3PC0tTb3nSUlJOv9mAK1bty62jpSUFHr37k3t2rWxtLTE0dERoFDaWsFrzw+eyK/tQZK0JoQQQrw8yu1La1lZWRgaGnLkyBEMDQ119mm1Wr3bOX/+PD4+Pnz44Yd89tlnWFtb8/PPPzNo0CBycnIwMzMrsY2iUs+UUiyOkT/ALHhOUclnpqamhaYrhIWFMXLkSCIiIvjhhx+YPHkykZGRtGrV6pHqLi7BLSsrCzs7uyLnST/Oigu+vr7UqlWLZcuWYW9vT15eHq+99lqhtLXiEtseJElrQgghxMujzA5469SpQ4UKFYiLi6NmzZoAXLt2jd9++w0PDw+aNGlCbm4uly5d4o033tCrTWNjY3Jzc3W2HTlyhLy8PEJDQ9WB54YNG57YdVhaWmJvb09MTAweHh7q9piYGFq0aAHcTz4DyMjIoHLlygClWjqtSZMmNGnShIkTJ9K6dWu+//57vQa8pdW0aVP++usvjIyM1KewD3JxcSEuLk5nXd9ff/31oW1euXKF5ORkli1bpv47/vzzz0+0biGEKEhihYUof8rsgFer1TJo0CDGjRuHjY0N1apVY9KkSeqg1NnZmb59++Lv709oaChNmjTh8uXL7N27l0aNGvHWW4X/H5qjoyNpaWkkJCRQo0YNLCwscHJy4u7duyxevBhfX19iYmL45ptvnui1jBs3juDgYOrUqUPjxo0JCwsjISGBtWvXAuDk5ISDgwMhISF89tln/Pbbb4VWpChKWloa3377LW+//Tb29vYkJyeTkpLy1EIkvLy8aN26NV27dmXOnDk4Ozvz559/8tNPP9GtWzeaN2/OqFGjCAgIoHnz5rRt25a1a9dy6tQpateuXWSblStXxsbGhm+//RY7OzvS09P5+OOPn0r9QgghhCifyuwcXoC5c+fyxhtv4Ovri5eXF6+//jrNmjVT94eFheHv78+YMWOoV68eXbt25fDhw+oT4Qe98847dO7cmfbt21O1alXWrVuHm5sb8+fPZ/bs2bz22musXbuWmTNnPtHrGDlyJKNHj2bMmDG4uroSERHBtm3bqFu3LnD/T/Xr1q3jzJkzNGrUiNmzZzNjxowS2zUzM+PMmTO88847ODs7M3ToUIYPH84HH3zwROvPp9Fo2LFjB+3atWPAgAE4Ozvz3nvvceHCBWxtbYH7q2Hkp7g1a9aMCxcu8OGHHz60TQMDA9avX8+RI0d47bXX+O9//8vcuXOfSv1CCCGEKJ/KXdLai6Zg8pujoyNBQf+vvTuPqzn7/wD+um23W7dFaaW6kxZJUoQ0ZGmmpDvFyNaoLKWhb4x9r+xbDGYYQhlDZqHGoEhkaZokldDcSSIzMowlsqU6vz969Pl1lbpRqN7Px+PzeMy9n+W8P0cPznw6n/OaKnP0MXl3KGmNEEIIaX4oae0DlJ6eDlVV1fddhsySk5PRv39/PHjw4J3G/BJCCCGENCYa8L6lQYMG4cyZM7XumzdvntTnqpfPPgTW1tavTVzbunUrfHx83nFFhBDy/tWWrEYvsRHS/DXrObwfgu3btyMrKwtZWVn4/fff4eLigoqKCggEghorPohEInz99dcAKpcYCwsLg7GxMfh8PgwNDRESEsId++LFC8yePRtGRkbg8/kwMzPDjh07uP2nTp1Cjx49wOfzYWBggDlz5kililVvq0rXrl0RFhYGADhy5AiePHmCefPmwdHREWVlZdDW1kZERAQ+++wzXL9+Hf379wdQ+eIYj8eDv78/AOCXX36BjY0NBAIBtLW14eLigidPntTbV1WRwcuXL4eenh40NTWxePFilJWVYebMmdDS0kL79u0RFRUldd7s2bNhYWEBFRUVmJqaYuHChdyybIwxuLi4wNXVlVu27f79+2jfvj0WLVpUb02EEEIIaflowPuW2rVrBzMzM5iZmeG7775DZmYmfvvtN5w4cQLnzp2Tijqubv/+/Vi/fj22bt2KvLw8xMXFwcbGhtvv6+uLmJgYbNy4Ebm5udi6dSu3fvA///wDd3d3ODg4IDs7G1u2bMGOHTtkepGtiomJCQBgy5YtGDduHC5dugQvLy/MmDEDL1++hJGREfbv3w8AkEgkKCoq4oIrRo0ahXHjxiE3NxfJyckYOnSozOsKnzhxArdu3cLp06exbt06hIaGwsPDA23atEFaWhqCgoIwceJE/P3339w5ampqiI6OxpUrV7BhwwZERkZi/fr1ACpflNu1axfS09OxceNGAEBQUBDatWtX54CXooUJIYSQVqSJE99ajfqijhljzMTEhK1fv54xxlhERASzsLBgpaWlNa4lkUgYAJaYmFhrW/PmzWOWlpasoqKC++7bb79lQqGQi/2t3lYVW1tbFhoayn0GwBYsWMB9LikpYQBYfHw8Y6z2SOOMjAwGgF2/fr3ePnmVn58fMzExqRFN3KdPH+5zWVkZU1VVZTExMa+9zpo1a1i3bt2kvvvpp5+YsrIymzNnDlNVVWV//fVXnbVQtDAhpDavxgpTtDAhH7YWHy38oakv6vhV3t7eePbsGUxNTREQEIDY2FhuSkJWVhbk5eWlgiiqy83NhaOjo1SqmpOTE0pKSqSejMqiehyvqqoq1NXVXxvHCwC2trYYOHAgbGxs4O3tjcjISDx48EDm9qytrWtEE1d/si0vLw9tbW2pGn788Uc4OTlBX18fQqEQCxYsqBEr7O3tjSFDhmDlypVYu3Ytt6Tb61C0MCGEENJ60ID3PTEyMoJEIsHmzZshEAgwadIk9O3bFy9fvoRAIHjr68vJydWYZlBbHHFdUcG1kZeXR2JiIuLj49GpUyds2rQJlpaWKCgokKmuhsYVp6amwsfHB+7u7jh06BAyMzMxf/78GrHCT58+5WKk8/Ly6q2Dz+dDXV1daiOEkOsrB9fYCCHNHw14G0n1qOMqVVHHryMQCCAWi7Fx40YkJycjNTUVOTk5sLGxQUVFBU6dOlXreVZWVkhNTZUa0KakpEBNTQ3t27cHULkiRFFREbf/0aNHMg9KqygpKQFAjZfveDwenJycEB4ejszMTCgpKSE2NrZB15bV77//DhMTE8yfPx/du3eHubl5ratLTJ8+HXJycoiPj8fGjRtx4sSJJqmHEEIIIc0PLUvWSOqLOn5VdHQ0ysvL0bNnT6ioqOCHH36AQCCAiYkJtLW14efnh3HjxmHjxo2wtbXFjRs3cOfOHQwfPhyTJk3C119/jf/9738IDg6GRCJBaGgopk2bxrU3YMAAREdHQywWQ1NTE4sWLYK8vHyD7snExAQ8Hg+HDh2Cu7s7BAIBLl++jKSkJHz66afQ1dVFWloa7t69Cysrq7fuw9qYm5ujsLAQ+/btg4ODAw4fPlxjcH348GHs3LkTqampsLe3x8yZM+Hn54eLFy+iTZs2TVIXIYQQQpoPesLbiOqLOq5OU1MTkZGRcHJyQpcuXXD8+HH89ttv0NbWBlC5esKwYcMwadIkdOzYEQEBAdzSX+3atcORI0dw7tw52NraIigoCOPHj8eCBQu468+dOxfOzs7w8PDA4MGD4eXlhQ4dOjToftq1a4fw8HDMmTMHenp6CA4Ohrq6Ok6fPg13d3dYWFhgwYIFiIiIwKBBg96w1+r22Wef4auvvkJwcDC6du2K33//HQsXLuT23717F+PHj0dYWBjs7e0BAOHh4dDT00NQUFCT1EQIIYSQ5oWihV8ha7oYxQS/PzweD7GxsfDy8mq0a1K0MCGEENL8yPrvNz3hfUXv3r1RVFQEDQ0NAJVTD2ob+KanpyMwMPAdV/dh4/F4iIuLa/J2ioqKmuyJMiGEEEJaHprD+wolJSXo6+vXe9yHFBP8IRk1alStc4Xj4+PRp0+fRmlDlj8fQgh5ExQtTEjL1Cyf8Pbr1w/BwcEIDg6GhoYG2rZti4ULF3KrFjx48AC+vr5o06YNVFRUMGjQIKmlqm7cuAGxWIw2bdpAVVUV1tbWOHLkCIDKKQ08Hg8PHz5EcnIyxo4di+LiYvB4PPB4PC6a99Xo3sLCQnh6ekIoFEJdXR3Dhw/Hv//+y+0PCwtD165dsXv3bohEImhoaGDkyJF4/Pgxd8ybRvYCwM6dO2Ftbc1FDQcHB8tcW1Xkb3VTp05Fv379pPo8JCQEs2bNgpaWFvT19bm+qOoPAHj+/DmePHkCDQ0NLnI5KysL3bt3r1FzVZ/s3LkTxsbGEAqFmDRpEsrLy7F69Wro6+tDV1cXy5Ytkzqv+pPk69evg8fj4cCBA+jfvz9UVFRga2uL1NTUOvuLktYIIYSQ1qNZDngBYNeuXVBQUMC5c+ewYcMGrFu3Dtu3bwdQOYA7f/48Dh48yC3f5e7uzq1DO3nyZLx48QKnT59GTk4OVq1axcX2Vte7d298/fXXUFdXR1FREYqKijBjxowax1VUVMDT0xP379/HqVOnkJiYiGvXrmHEiBFSx+Xn5yMuLg6HDh3CoUOHcOrUKaxcuRIA3iqyd8uWLZg8eTICAwORk5ODgwcPwszMrEG1yWLXrl1QVVVFWloaVq9ejcWLFyMxMRFA5RQPAIiKikJRURGysrK4yGUzM7PXri2cn5+P+Ph4JCQkICYmBjt27MDgwYPx999/49SpU1i1ahUWLFggtdxbbebPn48ZM2YgKysLFhYWGDVqFBfkUZsVK1ZAQ0OD24yMjBrcH4QQQghpJpo68q0pODs7MysrK6lo3dmzZzMrKyv2119/MQAsJSWF2/fff/8xgUDAxf7a2NiwsLCwWq/9apxuVFQU09DQqHFc9ejeY8eOMXl5eVZYWMjtv3z5MgPAzp07xxirjLJVUVFhjx494o6ZOXMm69mzJ2Ps7SJ7DQ0N2fz582vdJ0ttfn5+zNPTU+q8KVOmMGdnZ+6zs7Mz+/jjj6WOcXBwYLNnz+Y+A2CxsbEy111bn7i6ujKRSFQjfnjFihW1tlNQUMAAsO3bt9e4v9zc3Ne2/fz5c1ZcXMxtN2/epGhhQghFCxPSzLT4aOFevXpJRes6OjoiLy8PV65cgYKCglTEr7a2NiwtLZGbmwsACAkJwdKlS+Hk5ITQ0FBcvHjxrWrJzc2FkZGR1FPCTp06QVNTk2sTqPy1v5qaGvfZwMCAi9B908jeO3fu4NatWxg4cOBb1SaL6jHEr9b/pl7tEz09PXTq1KlG/HB97VSvzcDAAADqPIeS1gghhJDWo9kOeN/GhAkTcO3aNYwZMwY5OTno3r07Nm3a1OTt1hWh+6aRvR9yDLEsGho1LMt1qv5H6G1rI4S0PhQtTEjL1GwHvK/O6fzjjz9gbm6OTp06oaysTGr/vXv3IJFI0KlTJ+47IyMjBAUF4cCBA5g+fToiIyNrbUdJSalGtO6rrKyscPPmTdy8eZP77sqVK3j48KFUm/V5k8heNTU1iEQiJCUlvXFtr8YQA0BWVpbMdVdRVFSst68IIYQQQt61ZjvgLSwsxLRp0yCRSBATE4NNmzZhypQpMDc3h6enJwICAnD27FlkZ2fjiy++QLt27eDp6QmgcgWCo0ePoqCgABcuXMDJkydfG40rEolQUlKCpKQk/Pfff3j69GmNY1xcXGBjYwMfHx9cuHAB586dg6+vL5ydnWtdnaA2aWlpWL58Oc6fP4/CwkIcOHBA5sjesLAwREREYOPGjcjLy8OFCxe4J9ay1DZgwACcP38e33//PfLy8hAaGopLly7JVHd1VQPv27dvyzQdgxBCCCHkXWi2A15fX188e/YMPXr0wOTJkzFlyhQuCCIqKgrdunWDh4cHHB0dwRjDkSNHuF97l5eXY/LkybCysoKbmxssLCywefPmWtvp3bs3goKCMGLECOjo6GD16tU1juHxePj111/Rpk0b9O3bFy4uLjA1NcWPP/4o8/28TWSvn58fvv76a2zevBnW1tbw8PDglmGTpTZXV1csXLgQs2bNgoODAx4/fgxfX1+Za68SERGBxMREGBkZwc7OrsHnN5a9e/e+t7YJIYQQ8uFpltHC/fr1Q9euXaXWwSUtn6xxztHR0Zg6dSoePnwo87UpWpgQQghpfmT995uS1pqB0tJSKCkpve8yCCGkRastZQ2gpDVCWoJmO6WhJatKkps6dSratm0LgUAAFRUVKCgogMfjQU5ODoqKilBVVcWZM2ewbds2GBoa1liVwNPTE+PGjeM+//rrr7C3t4eysjJMTU0RHh4uFc7A4/Gwfft2DBkyBCoqKjA3N8fBgwe5/dHR0dDU1JRqIy4uTmp5uOrtyMnJQU5ODkpKShAKhVLbnj17atw3YwxhYWEwNjYGn8+HoaEhQkJCuD65ceMGvvrqKy71rnpdxsbGUFFRwZAhQ3Dv3r2GdzohhBBCWqxmOeBNTk5u8dMZdu3aBSUlJaSkpODHH3+EQCDAhAkTkJCQgNjYWPTs2RM2Njbo3r07vL29ce/ePZw8eZI7//79+0hISICPjw8A4MyZM/D19cWUKVNw5coVbN26FdHR0TVie8PDwzF8+HBcvHgR7u7u8PHxwf3792Wuu3o7p06dws6dO6Grq4tx48ZJRQ1/9tlnNc7dv38/1q9fj61btyIvLw9xcXGwsbEBABw4cADt27fH4sWLudQ7oPJlv/HjxyM4OBhZWVno378/li5dWm+dFC1MCCGEtCJNn4FBGsrZ2ZnZ2dlxn5csWcI+/fRTqWOqksEkEgljjDFPT082btw4bv/WrVuZoaEhl1g2cOBAtnz5cqlr7N69mxkYGHCfAbAFCxZwn0tKShgAFh8fzxirPXUuNjaWVf8xkqWd14mIiGAWFhastLS01v3V0+2qjBo1irm7u0t9N2LEiFrT8aoLDQ1lAGpslLRGSOtVW8oaJa0R8mFr8UlrLV23bt24/87OzsbJkyelpgR07NgRAJCfnw8A8PHxwf79+/HixQsAwJ49ezBy5EgusSw7OxuLFy+WukZAQACKioqkllqrnlimqqoKdXX1BqWpydpObby9vfHs2TOYmpoiICAAsbGxUlMuapObmyuVqgdUpu7VZ+7cuSguLua26usUE0IIIaRloZfWPlCqqqrcf5eUlEAsFmPVqlU1jquK0RWLxWCM4fDhw3BwcMCZM2ewfv16qWuEh4dj6NChNa6hrKzM/XddKWeyJLLJ2k5tjIyMIJFIcPz4cSQmJmLSpElYs2YNTp06VaOut8Xn88Hn8xv1moQQQgj5MNGAtxmwt7fH/v37IRKJoKBQ+x+ZsrIyhg4dij179uDq1auwtLSEvb291DUkEgnMzMzeuA4dHR08fvwYT5484QbkryayvW07AoEAYrEYYrEYkydPRseOHZGTkwN7e/taU++srKxqTd0jhJCGotUYCGm5aMDbDEyePBmRkZEYNWoUZs2aBS0tLVy9ehX79u3D9u3bIS8vD6ByWoOHhwcuX76ML774QuoaixYtgoeHB4yNjTFs2DDIyckhOzsbly5dkuklLwDo2bMnVFRUMG/ePISEhCAtLQ3R0dGN1k50dDTKy8u5dn744QcIBAKYmJgAqFyH9/Tp0xg5ciT4fD7atm2LkJAQODk5Ye3atfD09MTRo0eRkJAgY88SQgghpDWgObzNgKGhIVJSUlBeXo5PP/0UNjY2mDp1KjQ1Nbk5ukBlRLCWlhYkEglGjx4tdQ1XV1ccOnQIx44dg4ODA3r16oX169dzg0lZaGlp4YcffsCRI0dgY2ODmJgYhIWFNVo7mpqaiIyMhJOTE7p06YLjx4/jt99+g7a2NgBg8eLFuH79Ojp06AAdHR0AQK9evRAZGYkNGzbA1tYWx44dw4IFC2S+J0IIIYS0fM0yae1dS05ORv/+/fHgwYMa69BWJ2sSGGl6Df2zoKQ1QgghpPmR9d9vesIrg969e6OoqAgaGhoAag9gAID09HQEBga+4+o+bDweD3FxcY1yrdf1OyGEEEJIXWgOrwyUlJSgr69f73FVv2YntduzZw8mTpxY6z4TExNcvnz5HVdECCGvjxSuQi+zEdL8tZgnvFVxvMHBwdDQ0EDbtm2xcOFCbhmtBw8ewNfXF23atIGKigoGDRqEvLw87vwbN25ALBajTZs2UFVVhbW1NY4cOQKgckoDj8fDw4cPkZycjLFjx6K4uJiLuK2axyoSiaQS4AoLC+Hp6QmhUAh1dXUMHz4c//77L7c/LCwMXbt2xe7duyESiaChoYGRI0fi8ePH3DG//PILbGxsIBAIoK2tDRcXFzx58kSmPtm5cyesra3B5/NhYGCA4OBgmWvz9/eHl5eX1PWmTp2Kfv36SfV5SEgI9yKdvr6+1JxekUgEABgyZAh4PB7mzp0rlbZWfavq6+zsbPTv3x9qampQV1dHt27dcP78+Tr7/c6dOxCLxRAIBPjoo49qjS0mhBBCSOvVop7w7tq1C+PHj8e5c+dw/vx5BAYGwtjYGAEBAfD390deXh4OHjwIdXV1zJ49G+7u7rhy5QoUFRUxefJklJaW4vTp01BVVcWVK1cgFAprtNG7d298/fXXWLRoESQSCQDUelxFRQU3oDx16hTKysowefJkjBgxAsnJydxx+fn5iIuLw6FDh/DgwQMMHz4cK1euxLJly1BUVIRRo0Zh9erVGDJkCB4/fowzZ87UWAu3Nlu2bMG0adOwcuVKDBo0CMXFxUhJSWlQbbL2+bRp05CWlobU1FT4+/vDyckJn3zyCdLT06Grq4uoqCi4ublBXl6+3qfgPj4+sLOzw5YtWyAvL4+srCwoKirW2e/+/v64desWTp48CUVFRYSEhNQblvHixQsupAMARQsTQgghLViLGvAaGRlh/fr14PF4sLS0RE5ODtavX49+/frh4MGDSElJQe/evQFU/nrdyMgIcXFx8Pb2RmFhIT7//HPY2NgAAExNTWttQ0lJCRoaGuDxeHVOc0hKSkJOTg4KCgpgZGQEAPj+++9hbW2N9PR0ODg4AKgcfEZHR0NNTQ0AMGbMGCQlJXED3rKyMgwdOpRb5aCqvvosXboU06dPx5QpU7jvqtqUtTZZdOnSBaGhoQAAc3NzfPPNN0hKSsInn3zCDW41NTVlmhICVD55njlzJpckZ25uzu2rrd//+usvxMfH49y5c1zdO3bsgJWVVZ3trFixAuHh4TLfJyGEEEKarxYzpQGoXKKKx+Nxnx0dHZGXl4crV65AQUFBKoJWW1sblpaWyM3NBQCEhIRg6dKlcHJyQmhoKC5evPhWteTm5sLIyIgbUAJAp06doKmpybUJVP7av2qwC1Qmp1U9nbS1tcXAgQNhY2MDb29vREZG4sGDB/W2fefOHdy6dQsDBw58q9pkUT2K+NX638S0adMwYcIEuLi4YOXKlVx08uvk5uZCQUFBKoq5Y8eO9b7cRtHChBBCSOvRoga8b2PChAm4du0axowZg5ycHHTv3h2bNm1q8nbrivKVl5dHYmIi4uPj0alTJ2zatAmWlpYoKCio85oCgeCt65IlRhiou/43ERYWhsuXL2Pw4ME4ceIEOnXqhNjY2De+3uvw+Xyoq6tLbYSQ1un6ysF1boSQ5q9FDXhri5g1NzdHp06dUFZWJrX/3r17kEgk6NSpE/edkZERgoKCcODAAUyfPh2RkZG1tlNbxO2rrKyscPPmTaknh1euXMHDhw+l2qwPj8eDk5MTwsPDkZmZCSUlpXoHgGpqahCJREhKSnrj2nR0dFBUVCR13qsxwrJQVFSst69eZWFhga+++grHjh3D0KFDERUVBaD2fu/YsSPKysqQkZHBfSeRSPDw4cMG10oIIYSQlqlFDXgLCwsxbdo0SCQSxMTEYNOmTZgyZQrMzc3h6emJgIAAnD17FtnZ2fjiiy/Qrl07eHp6AqhcgeDo0aMoKCjAhQsXcPLkydfOAxWJRCgpKUFSUhL+++8/PH36tMYxLi4usLGxgY+PDy5cuIBz587B19cXzs7O6N69u0z3k5aWhuXLl+P8+fMoLCzEgQMHcPfu3XrnpwKVT0ojIiKwceNG5OXl4cKFC9wTa1lqGzBgAM6fP4/vv/8eeXl5CA0NxaVLl2Squ7qqgfft27frnY7x7NkzBAcHIzk5GTdu3EBKSgrS09O5+62t3y0tLeHm5oaJEyciLS0NGRkZmDBhQqM85SaEEEJIy9CiBry+vr549uwZevTogcmTJ2PKlClcEERUVBS6desGDw8PODo6gjGGI0eOcL+SLy8vx+TJk2FlZQU3NzdYWFhg8+bNtbbTu3dvBAUFYcSIEdDR0cHq1atrHMPj8fDrr7+iTZs26Nu3L1xcXGBqaooff/xR5vtRV1fH6dOn4e7uDgsLCyxYsAAREREYNGhQvef6+fnh66+/xubNm2FtbQ0PDw9uGTZZanN1dcXChQsxa9YsODg44PHjx/D19ZW59ioRERFITEyEkZER7Ozs6jxWXl4e9+7dg6+vLywsLDB8+HAMGjSIe7nsdf0eFRUFQ0NDODs7Y+jQoQgMDISurm6DayWEEEJIy9RiooX79euHrl27Sq2DS4isKFqYEEIIaX5k/fe7RS1L9qEpLS2FkpLS+y6DEELIK+pLV6uOXlwjpPlrUVMa3reqtLepU6eibdu2cHV1xaVLlzBo0CAIhULo6elhzJgx+O+//wAA27Ztg6GhYY1VDTw9PTFu3Dju86+//gp7e3soKyvD1NQU4eHhEAqF3Mbj8aCsrAwFBQXweDy0b98eBw8e5M6Pjo6usUxXXFyc1BJur2unrKxMpnvn8XjYunUrPDw8oKKiAisrK6SmpuLq1avo168fVFVV0bt3b5ibm0vVLhAIIC8vDx6PB11d3Rptrlu3DjY2NlBVVYWRkREmTZqEkpKSGvd29OhRWFlZQSgUws3NrcYLd4QQQghpvVrMgDc5OfmDmM6wa9cuKCkpISUlBStXrsSAAQNgZ2eH8+fPIyEhAf/++y+GDx8OAPD29sa9e/dw8uRJ7vz79+8jISEBPj4+AIAzZ87A19cXU6ZMwZUrV7B161ZER0dj3LhxXCwvALRp0wZr1qzB8ePH4enpCR8fH9y/f1/mul/XzrJly2S+xpIlS+Dr64usrCx07NgRo0ePxsSJEzF37lycP38ejDG0a9eOqzsyMhIKCgpYsWIFkpKSEBkZWaNNOTk5bNy4EZcvX8auXbtw4sQJzJo1S6rdp0+fYu3atdi9ezdOnz6NwsJCzJgxo85aX7x4gUePHklthBBCCGmhGGk0zs7OzM7Ojvu8ZMkS9umnn0odc/PmTQaASSQSxhhjnp6ebNy4cdz+rVu3MkNDQ1ZeXs4YY2zgwIFs+fLlUtfYvXs3MzAw4D4DYAsWLOA+l5SUMAAsPj6eMcZYVFQU09DQkLpGbGwsq/7HL0s7dXm1htTUVAaA7dixg/suJiaGKSsrv1WbP//8M9PW1uY+R0VFMQDs6tWr3Hfffvst09PTq7Pe0NBQBqDGVlxcXP/NEkKaPZPZh2TeCCEfruLiYpn+/aY5vI2seuJXdnY2Tp48CaFQWOO4/Px8WFhYwMfHBwEBAdi8eTP4fD727NmDkSNHQk5OjrtGSkqK1FPP8vJyPH/+HE+fPoWKigoA6cQzVVVVqKurNyjxTNZ26lK9Bj09PQDSUch6enp4/vw5Hj16BHV1dZnaPH78OFasWIE///wTjx49QllZWY2aVFRU0KFDB+4asqS9zZ07F9OmTeM+P3r0SCp5jhBCCCEtBw14G5mqqir33yUlJRCLxVi1alWN4wwMDAAAYrEYjDEcPnwYDg4OOHPmDNavXy91jfDwcAwdOrTGNZSVlbn/rivxTJbUNFnbqUv1GqrmB9f2XVVd9bV5/fp1eHh44Msvv8SyZcugpaWFs2fPYvz48SgtLeUGvLXd+6v3+yo+nw8+ny/TfRFCCCGkeaMBbxOyt7fH/v37IRKJoKBQe1crKytj6NCh2LNnD65evQpLS0vY29tLXUMikcDMzOyN69DR0cHjx4/x5MkTbkD+ampaY7TTUPW1mZGRgYqKCkRERHBPvH/66ad3Vh8hpOWilRcIaV1owNuEJk+ejMjISIwaNQqzZs2ClpYWrl69in379mH79u2Ql5cHAPj4+MDDwwOXL1/GF198IXWNRYsWwcPDA8bGxhg2bBjk5OSQnZ2NS5cuYenSpTLV0bNnT6ioqGDevHkICQlBWloaoqOjG72dhqqvTTMzM7x8+RKbNm2CWCxGSkoKvvvuuyaphRBCCCEtV4tZpeFDZGhoiJSUFJSXl+PTTz+FjY0Npk6dCk1NTe6JJVAZ46ulpQWJRILRo0dLXcPV1RWHDh3CsWPH4ODggF69emH9+vUwMTGRuQ4tLS388MMPOHLkCGxsbBATE4OwsLBGb6eh6mvT1tYW69atw6pVq9C5c2fs2bMHK1asaLJ6CCGEENIytZiktfelesKbSCTC1KlTMXXq1PddFmkgSlojhBBCmh9KWnsP0tPTpV5a+9AlJyejf//+ePDgQY1gCkIIIYSQloIGvI1IR0fnfZfQJPbs2YOJEyfWus/ExASXL19+xxURQsiba0isMEAvuBHSEtAc3gZ48uQJfH19IRQKYWBggIiICKn9IpGIS3tjjCEsLAzGxsbg8/kwNDRESEgId+yLFy8we/ZsGBkZgc/nw8zMDDt27OD2nzp1Cj169ACfz4eBgQHmzJkjFblbva0qXbt2lZqby+PxsH37dgwZMgQqKiowNzfnIoevX7+O/v37A6hMaePxePD39wcA/PLLL7CxsYFAIIC2tjYiIyPx+++/cwlp1bcjR45w7fn7+8PLywvLly+Hnp4eNDU1sXjxYpSVlWHmzJnQ0tJC+/btERUVJVX3zZs3MXz4cGhqakJLSwuenp64fv06tz89PR2ffPIJ2rZtCw0NDTg7O+PChQtS16jrXgkhhBDSutGAtwFmzpyJU6dO4ddff8WxY8eQnJxcY+BVZf/+/Vi/fj22bt2KvLw8xMXFSYUw+Pr6IiYmBhs3bkRubi62bt3KBVT8888/cHd3h4ODA7Kzs7Flyxbs2LHjjVZLCA8Px/Dhw3Hx4kW4u7tzkcNGRkbYv38/AEAikaCoqAgbNmxAUVERRo0ahXHjxiE3NxfJyckYPnw4TE1NYWZmVmN79aW2EydO4NatWzh9+jTWrVuH0NBQeHh4oE2bNkhLS0NQUBAmTpyIv//+G0DlesCurq5QU1PDmTNnkJKSAqFQCDc3N5SWlgIAHj9+DD8/P5w9exZ//PEHzM3N4e7ujsePH8t0r7WhaGFCCCGkFWnyzLcW4vHjx0xJSYn99NNP3Hf37t1jAoGATZkyhTHGmImJCVu/fj1jjLGIiAhmYWHBSktLa1xLIpEwACwxMbHWtubNm8csLS1ZRUUF9923337LhEIhFzlcva0qtra2LDQ0lPuMeiKHT548yQCwBw8ecMdkZGQwAOz69ev19smr/Pz8mImJCVcjY4xZWlqyPn36cJ/LysqYqqoqi4mJYYxVRgm/eq8vXrxgAoGAHT16tNZ2ysvLmZqaGvvtt99kvtdXUbQwIa1XQ2KFKVqYkA+brNHC9IRXRvn5+SgtLUXPnj2577S0tGBpaVnr8d7e3nj27BlMTU0REBCA2NhYbkpCVlYW5OXl4ezsXOu5ubm5cHR05JLJAMDJyQklJSXck1FZNTRy2NbWFgMHDoSNjQ28vb0RGRmJBw8eyNyetbW11JJrenp6Uk+25eXloa2tzdWQnZ2Nq1evQk1NDUKhEEKhEFpaWnj+/Dny8/MBAP/++y8CAgJgbm4ODQ0NqKuro6SkBIWFhW98r3PnzkVxcTG33bx5U+Z7JIQQQkjzQi+tNREjIyNIJBIcP34ciYmJmDRpEtasWYNTp05BIBC89fVliQsG6o4cro28vDwSExPx+++/49ixY9i0aRPmz5+PtLQ0fPTRR/XWVVt7ddVQUlKCbt26Yc+ePTWuVfUSoJ+fH+7du4cNGzbAxMQEfD4fjo6O3JSHN7lXihYmpPWil9AIaX3oCa+MOnToAEVFRaSlpXHfPXjwAH/99ddrzxEIBBCLxdi4cSOSk5ORmpqKnJwc2NjYoKKiAqdOnar1PCsrK6SmpkoNaFNSUqCmpob27dsDqBwMFhUVcfsfPXqEgoKCBt2TkpISAKC8vFzqex6PBycnJ4SHhyMzMxNKSkqIjY1t0LVlZW9vj7y8POjq6taYH6yhoQGg8t5DQkLg7u4Oa2tr8Pl8/Pfff01SDyGEEEJaHhrwykgoFGL8+PGYOXMmTpw4gUuXLsHf31/q1/fVRUdHY8eOHbh06RKuXbuGH374AQKBACYmJhCJRPDz88O4ceMQFxeHgoICJCcn46effgIATJo0CTdv3sT//vc//Pnnn/j1118RGhqKadOmce0NGDAAu3fvxpkzZ5CTkwM/Pz8uqlhWJiYm4PF4OHToEO7evYuSkhKkpaVh+fLlOH/+PAoLC3HgwAHcvXsXVlZWb9eBr+Hj44O2bdvC09MTZ86c4foiJCSEm75hbm6O3bt3Izc3F2lpafDx8WmUp+SEEEIIaR1azYC3X79+MiegRUdH1xrEsGbNGvTp0wdisRguLi74+OOP0a1bt1qvoampicjISDg5OaFLly44fvw4fvvtN2hrawMAtmzZgmHDhmHSpEno2LEjAgICcPToUWhqaqJdu3Y4cuQIzp07B1tbWwQFBWH8+PFYsGABd/25c+fC2dkZHh4eGDx4MLy8vNChQ4cG9Um7du0QHh6OOXPmQE9PD8HBwVBXV8fp06fh7u4OCwsLLFiwABERERg0aFCDri0rFRUVfPbZZ8jOzsbQoUNhZWUFsViMgwcPcokpO3bswIMHD2Bvb48xY8YgJCQEurq6TVIPIYQQQlqeVhMtXD0CuD7R0dGYOnUqHj58CAAICwtDXFwcsrKymrTGV9ttLV7t3+LiYjDG3mn6G0ULE0IIIc0PRQuTZqtq7i4hhBBCSGNokVMa6ktEe/HiBWbMmIF27dpBVVUVPXv2RHJycq3Xio6ORnh4OLKzs8Hj8cDj8RAdHQ0AWLduHWxsbKCqqgojIyNMmjQJJSUlMtcZHR0NY2NjqKioYMiQIbh3757U/vz8fHh6ekJPTw9CoRAODg44fvw4t3/x4sXo3Llzjet27doVCxcuBAAkJyejR48eUFVVhaamJpycnHDjxo16awsLC0PXrl2xc+dOGBsbQygUQlFREUKhEHw+H3JycuDxeODz+RAKhThz5gwA4OHDh5gwYQJ0dHSgrq6OAQMGIDs7W+raK1euhJ6eHtTU1DB+/Hg8f/5can9VYluVhIQEfPzxx9DU1IS2tjY8PDy4JcuAytQ4Ho+HAwcOoH///lBRUYGtrS1SU1PrvU9CSOshmnP4jTZCSPPXIge89SWiBQcHIzU1Ffv27cPFixfh7e0NNzc35OXl1bjWiBEjMH36dFhbW6OoqAhFRUUYMWIEgMqlwTZu3IjLly9j165dOHHiBGbNmiVTjWlpaRg/fjyCg4ORlZWF/v3710hSKykpgbu7O5KSkpCZmQk3NzeIxWJu/dmqNLT09HTunMzMTFy8eBFjx45FWVkZvLy84OzsjIsXLyI1NRWBgYFS6/vWJT8/H/Hx8UhISEBMTAx4PB7s7OwwYsQIJCQkYMWKFSgtLUV0dDS6d+8OoHL94Tt37iA+Ph4ZGRmwt7fHwIEDucSzn376CWFhYdyLcQYGBti8eXOddTx58gTTpk3D+fPnkZSUBDk5OQwZMqTGkmPz58/HjBkzkJWVBQsLC4waNUoqjrk6SlojhBBCWpGmz8B4t+pLRLtx4waTl5dn//zzj9R5AwcOZHPnzmWMMRYVFcU0NDS4faGhoczW1rbetn/++Wemra0tU52jRo1i7u7uUt+NGDFCqt3aWFtbs02bNnGfBw0axL788kvu8//+9z/Wr18/xljlfQNgycnJMtVUXWhoKFNRUWGPHj3ivnN1dWUikahGktqKFSsYY4ydOXOGqaurs+fPn0tdq0OHDmzr1q2MMcYcHR3ZpEmTpPb37NlTqn/9/PyYp6fna2u7e/cuA8BycnIYY4wVFBQwAGz79u3cMZcvX2YAWG5u7mvvD5S0Rkir0tCENUpaI+TD12qT1upLRMvJyUF5eTksLCy4ZC+hUIhTp05J/ZpcFsePH8fAgQPRrl07qKmpYcyYMbh37x6ePn1a77m5ublSNQKAo6Oj1OeSkhLMmDEDVlZW0NTUhFAoRG5urlTCWEBAAGJiYvD8+XOUlpZi7969GDduHHff/v7+cHV1hVgsxoYNG6TW7q2PSCSCmpoa91lPTw+dOnWqkaRWPTWtpKQE2traUn1bUFDA9a0s9/2qvLw8jBo1CqamplBXV4dIJAKAOpPWDAwMAICS1gghhBDS+l5aKykpgby8PDIyMmqsWysUCmW+zvXr1+Hh4YEvv/wSy5Ytg5aWFs6ePYvx48ejtLQUKioqb13rjBkzkJiYiLVr18LMzAwCgQDDhg2TShgTi8Xg8/mIjY2FkpISXr58iWHDhnH7o6KiEBISgoSEBPz4449YsGABEhMT0atXr3rbf5PUNAMDg1rnQ7/NigtisRgmJiaIjIyEoaEhKioq0Llz5zqT1qqmbVDSGiGEEEJa3IC3eiKasbExgP9PRHN2doadnR3Ky8tx584d9OnTR6ZrKikp1Ugjy8jIQEVFBSIiIrgnnlXBEbKwsrKSSm0DgD/++EPqc0pKCvz9/TFkyBAAlQPK69evSx2joKAAPz8/REVFQUlJCSNHjqwRymBnZwc7OzvMnTsXjo6O2Lt3r0wD3oayt7fH7du3oaCgwD2FfVXVffv6+nLfvXrf1d27dw8SiQSRkZHcn9fZs2cbtW5CSOtAkcKEtF4tbsBbPRFNW1sburq6mD9/PjcotbCwgI+PD3x9fREREQE7OzvcvXsXSUlJ6NKlCwYPrvkXokgkQkFBAbKystC+fXuoqanBzMwML1++xKZNmyAWi5GSkoLvvvtO5jpDQkLg5OSEtWvXwtPTE0ePHkVCQoLUMebm5jhw4ADEYjF4PB4WLlxY6xPLCRMmcEloKSkp3PcFBQXYtm0bPvvsMxgaGkIikSAvL09qsNmYXFxc4OjoCC8vL6xevRoWFha4desWDh8+jCFDhqB79+6YMmUK/P390b17dzg5OWHPnj24fPkyTE1Na71mmzZtoK2tjW3btsHAwACFhYWYM2dOk9RPCCGEkJapxc3hBepPRIuKioKvry+mT58OS0tLeHl5IT09nXsi/KrPP/8cbm5u6N+/P3R0dBATEwNbW1usW7cOq1atQufOnbFnzx6sWLFC5hp79eqFyMhIbNiwAba2tjh27JhUkhpQuexZmzZt0Lt3b4jFYri6usLe3r7GtczNzdG7d2907NhRan6siooK/vzzT3z++eewsLBAYGAgJk+ejIkTJ8pcZ0PweDwcOXIEffv2xdixY2FhYYGRI0fixo0b0NPTA1C56sXChQsxa9YsdOvWDTdu3MCXX3752mvKyclh3759yMjIQOfOnfHVV19hzZo1TVI/IYQQQlqmVpO01pIxxmBubo5JkyZh2rRpjXLNhiTTves2RCIRpk6dKnNUtCwoaY0QQghpfihprZW4e/cu9u3bh9u3b2Ps2LHvuxxCCCGEkA8ODXibyKBBg7j0sVfNmzcP8+bNa5R2dHV10bZtW2zbtg1t2rSR+Txra+vXJq5t3bq1UWojhJD3qbFS0uhlN0KavxY5h/dDsH37dmRlZdW6BQUFNVo7jDHcvXsXo0ePbtB5R44ceW19n332mdSxu3fvRvfu3aGmpgZ9fX2MHj1aan3b5ORk8Hg8HD16FHZ2dhAIBBgwYACXuGZlZQV1dXWMHj26xhrFZWVlCA4OhoaGBtq2bYuFCxei+iybO3fuQCwWQyAQ4KOPPsKePXtq3MvbRjwTQgghpGWjJ7xNpF27du+7hDqZmJjIfOzLly+xZMkSWFpa4s6dO5g2bRr8/f1x5MgRqePCwsLwzTffQEVFBcOHD8fw4cPB5/Oxd+9elJSUYMiQIdi0aRNmz57NnbNr1y6MHz8e586dw/nz5xEYGAhjY2MEBAQAAPz9/XHr1i2cPHkSioqKCAkJqREmURXx/NFHH+HatWuYNGkSZs2aVWdk8YsXL/DixQvuM0ULE0IIIS0XDXhJvaqS2wDA1NQUGzduhIODA0pKSqTCOpYuXQonJycAwPjx4zF37lzk5+dzS44NGzYMJ0+elBrwGhkZYf369eDxeLC0tEROTg7Wr1+PgIAA/PXXX4iPj8e5c+fg4OAAANixYwe3BFuV6i+viUQiLF26FEFBQXUOeFesWIHw8PA37xRCCCGENBs0pYHUKyMjA2KxGMbGxlBTU4OzszOAuqN99fT0oKKiIrW+bvUY4iq9evXiUtGAypjhvLw8lJeXIzc3FwoKClJLynXs2LFGatubRDxTtDAhhBDSetATXlKnJ0+ewNXVFa6urtizZw90dHRQWFgIV1fXeqN964ohbixvGvFM0cKEtHz0shkhpAoNeEmd/vzzT9y7dw8rV66EkZERAOD8+fONdv3a4pXNzc0hLy+Pjh07oqysDBkZGdyUBolEgocPH3LHv23EMyGEEEJaPprSQOpkbGwMJSUlbNq0CdeuXcPBgwexZMmSRrt+YWEhpk2bBolEgpiYGGzatAlTpkwBAFhaWsLNzQ0TJ05EWloaMjIyMGHCBAgEAu786hHP165dw+7duxsU8UwIIYSQlo8GvKROOjo6iI6Oxs8//4xOnTph5cqVWLt2baNd39fXF8+ePUOPHj0wefJkTJkyBYGBgdz+qKgoGBoawtnZGUOHDkVgYCB0dXW5/W8b8UwIIYSQlo+ihd9CQ6Jxo6OjMXXqVKlfx78L76vd5oaihQkhhJDmR9Z/v+kJ73sSFhaGrl27vu8yCCGEEEJaPHppjRBC3pHGirol7xat9kBI80dPeGX05MkT+Pr6QigUwsDAABEREVL7X7x4gRkzZqBdu3ZQVVVFz549kZycXOu1oqOjER4ejuzsbPB4PPB4PERHRwN4+5jc6OhoGBsbQ0VFBUOGDMG9e/ek9ufn58PT0xN6enoQCoVwcHDA8ePHuf2LFy9G586da1y3a9euWLhwIYDKKOEePXpAVVUVmpqacHJywo0bN+qtreqp9s6dO2FsbAyhUIhJkyahvLwcq1evhr6+PnR1dbFs2TKp8+rrk3HjxqFLly5cclppaSns7Ozg6+v72lpevHiBR48eSW2EEEIIaZlowCujmTNn4tSpU/j1119x7NgxJCcn48KFC9z+4OBgpKamYt++fbh48SK8vb3h5uaGvLy8GtcaMWIEpk+fDmtraxQVFaGoqAgjRowA8P8xuZcvX8auXbtw4sQJzJo1S6Ya09LSMH78eAQHByMrKwv9+/fH0qVLpY4pKSmBu7s7kpKSkJmZCTc3N4jFYi5EYty4ccjNzUV6ejp3TmZmJi5evIixY8eirKwMXl5ecHZ2xsWLF5GamorAwECp8Ii65OfnIz4+HgkJCYiJicGOHTswePBg/P333zh16hRWrVqFBQsWSC1XVl+fbNy4EU+ePMGcOXMAAPPnz8fDhw/xzTffvLaOFStWQENDg9uqllwjhBBCSMtDL63JoKSkBNra2vjhhx/g7e0NALh//z7at2+PwMBATJs2DaampigsLIShoSF3nouLC3r06IHly5fXeHksLCwMcXFxyMrKqrPtX375BUFBQfjvv//qrXP06NEoLi7G4cP//2vTkSNHIiEhoc6X1jp37oygoCAEBwcDANzd3SESibho3pCQEOTk5ODkyZO4f/8+tLW1kZyczCWuySosLAxr1qzB7du3oaamBgBwc3ODRCJBfn4+t45ux44d4e/vzw1gX1Vbn6SmpsLZ2Rlz5szBihUrcPLkSXz88cevreXFixfcE2GgctK7kZERvbRGmhRNaWieaEoDIR8uemmtEeXn56O0tBQ9e/bkvtPS0oKlpSUAICcnB+Xl5bCwsIBQKOS2U6dOIT8/v0FtvUlMbpXc3FypGoHKqN7qSkpKMGPGDFhZWUFTUxNCoRC5ublSMcEBAQGIiYnB8+fPUVpair1792LcuHHcffv7+8PV1RVisRgbNmxAUVGRzPcnEom4wS5QGTfcqVMnbrBb9V31CGJZ+sTR0REzZszAkiVLMH369DoHu0Bl0pq6urrURgghhJCWiV5aawQlJSWQl5dHRkYG5OXlpfYJhUKZr/OmMbkNMWPGDCQmJmLt2rUwMzODQCDAsGHDpGKCxWIx+Hw+YmNjoaSkhJcvX2LYsGHc/qioKISEhCAhIQE//vgjFixYgMTERPTq1ave9muLG64rgljWPqmoqEBKSgrk5eVx9erVN+4fQpoSPSkkhJD3gwa8MujQoQMUFRWRlpYGY2NjAMCDBw/w119/wdnZGXZ2digvL8edO3fQp08fma6ppKSE8vJyqe/eNibXysqq1qje6lJSUuDv748hQ4YAqBysX79+XeoYBQUF+Pn5ISoqCkpKShg5cqRUuhkA2NnZwc7ODnPnzoWjoyP27t0r04C3oWTtkzVr1uDPP//EqVOn4OrqiqioKIwdO7bR6yGEEEJI80MDXhkIhUKMHz8eM2fOhLa2NnR1dTF//nxuAGZhYQEfHx/4+voiIiICdnZ2uHv3LpKSktClSxcMHlzzqY5IJEJBQQGysrLQvn17qKmpScXkisVipKSkNCgmNyQkBE5OTli7di08PT1x9OhRJCQkSB1jbm6OAwcOQCwWg8fjYeHChdzT1OomTJgAKysrAJWD5CoFBQXYtm0bPvvsMxgaGkIikSAvL6/OFRHehix9kpmZiUWLFuGXX36Bk5MT1q1bhylTpsDZ2RmmpqZNUhchhBBCmg+aw/sa/fr1w9SpU7nPa9asQZ8+fSAWi+Hi4oKPP/4Y3bp1A1C5FNhvv/0GX19fTJ8+HZaWlvDy8kJ6ejr3RPhVn3/+Odzc3NC/f3/o6OggJiamwTG50dHR0NTU5D736tULkZGR2LBhA2xtbXHs2DEsWLBA6px169ahTZs26N27N8RiMVxdXWFvb1/j2ubm5ujduzc6duwoNS9YRUUFf/75Jz7//HNYWFggMDAQkydPxsSJE2Xp1garr0+eP3+OL774Av7+/hCLxQCAwMBA9O/fH2PGjKnxFJ0QQgghrQ+t0vAabxMbLOsKDG+rKWODGWMwNzfHpEmTMG3atEa//oeGooUJIYSQ5kfWf79pSgOp4e7du9i3bx9u375N82BJi0bLhBFZ0MuGhDR/NKUBzSdFrUuXLpCTkwOPx4OCggImTpyI4uJiCIVCLF++vNFS1HR1dRESEoKKigp4eHjInKKmrKwMZWVlrkZFRUUIhULw+XxoaGjUmqL28OFDTJgwATo6OlBXV8eAAQOQnZ3N7a/vnoDK+dDLly/HuHHjoKamBmNjY2zbtk3mfiWEEEJIy0YDXjSfFLVLly5hxowZOHbsGObOnQs+nw+hUIisrCwEBQU1WorajBkzcPXqVWRkZDQoRU1eXh79+/fH4cOH8d1334HH48HOzg4jRozAiRMnak1R8/b2xp07dxAfH4+MjAzY29tj4MCBuH//PoD6k+GqREREoHv37sjMzMSkSZPw5ZdfQiKRvLZWihYmhBBCWo9WP4eXUtTeX4ra2bNnMXjwYNy5cwd8Pp+7lpmZGWbNmoXAwECZ7kkkEqFPnz7YvXs3gMr5x/r6+ggPD0dQUNBr6w0PD6/xPc3hbV1oSgORBU1pIOTDRUlrMqIUtfeXopadnc39D0f1vi0oKOD6VpZ7Aiqne1Th8XjQ19eXSmt71dy5c1FcXMxtN2/elPk+CSGEENK80Etr9aAUtaZLUSspKYGBgUGt86GrlluT5Z5e13Zt6wtX4fP5Uk+VCSGEENJytfoBL6Wovb8UNXt7e9y+fRsKCgoQiUS1HiPLPRHypuhX1YQQ0jq0+gEvpai9vxQ1FxcXODo6wsvLC6tXr4aFhQVu3bqFw4cPY8iQIejevbvM90QIIYQQ8jqtfg4vUHeKGlD5a/53naL2qpaYosbj8XDkyBH07dsXY8eOhYWFBUaOHIkbN25AT0+vQfdECCGEEPI6rX6VhsbyNslsTaGuNlpDippIJMLUqVOl4qHrQklrhBBCSPNDqzQ0I2FhYejates7aevu3bv45ptvmmWKWnR0NPcyGyGEEEKIrFr9HN4PxbVr11676sO8efMwb968RmlHV1cXbdu2xbZt29CmTRuZz7O2tn5t4trWrVvh4+MDACgvLwePx5NaiozQeq+ENGf0ciMhzR+NSt5AU0QRP378GE+ePMGTJ0+wcOFCZGVlISQkBIaGhli6dOkbRREDwNGjR2FlZQWhUAg3NzfcunULd+/exejRo1FRUYHFixejffv24PP56Nq1q9SLcMnJyeDxeHj48CGOHDmCrKws7N27F0+ePMGhQ4eQlZWFhQsXYtKkSTh48CA6deoEPp+PwsJCJCcno0ePHlBVVYWmpiacnJxkiijOzs5G//79oaamBnV1dXTr1g3nz59HcnIyxo4di+LiYi6yOSwsDABw584diMViCAQCfPTRR9izZ0+97VDSGiGEENJ60BPeN1A9ilhXVxfz5s3DhQsXuGkJwcHBuHLlCvbt2wdDQ0PExsbCzc0NOTk5MDc3l7rWiBEjcOnSJSQkJOD48eMAAA0NDQgEAujq6mLr1q346KOPcO3aNUyaNAmzZs3iUtLq8/TpU6xduxa7d++GnJwcvvjiC8yYMYMbEG7YsAERERHYunUr7OzssHPnTnz22We4fPlyjTpNTEwAgBtwi0QiiEQi6Onp4dmzZ1i1ahW2b98ObW1taGlpoWvXrlzIRWlpKc6dOydTRLGPjw/s7OywZcsWyMvLIysrC4qKiujduze+/vprLFq0iIsMrnoi7u/vj1u3buHkyZNQVFRESEhInaETALBixYpak9YIIYQQ0vLQgLeBSkpKsGPHDvzwww8YOHAgAGDXrl1o3749AKCwsBBRUVFSUcQzZsxAQkICoqKisHz5cqnrCQQCCIVCKCgoQF9fX2pf9ReuRCIRli5diqCgIJkHvC9fvsR3332HDh06AKgciC9evJjbv3btWsyePRsjR44EAKxatQonT57E119/jW+//VbmPnn58iU2b94MW1tbAJXRzMXFxfDw8ODarloGrT6FhYWYOXMmOnbsCABSA28NDQ0uRa3KX3/9hfj4eJw7dw4ODg4AgB07dtTb3ty5c6Ve2Hv06BGMjIxkqpEQQgghzQsNeBuoIVHE1b148QLa2toNauv48eNYsWIF/vzzTzx69AhlZWV4/vw5nj59KlMym4qKCjfgBAADAwPuyeejR49w69YtODk5SZ3j5OSE7OzsBtWppKQkFe1bPaL4k08+gYuLC4YPHw4DA4N6rzVt2jRMmDABu3fvhouLC7y9vaXu4VW5ublQUFCQWkauY8eO9b7cRklrhBBCSOtBA95G9iFFEdcWt9uQVeiqXjyrfs7Lly9rHCcQCGpMV3jTiOKwsDCMHj0ahw8fRnx8PEJDQ7Fv3z4uaa25opdeCCGEkPeHXlproOpRxFWqoogBSEURm5mZSW2vTlmoUl8Uca9evbgUssairq4OQ0NDqaQ1oDJ5rVOnTgAAHR0dAEBRURG3PysrS+Y2quKJf//9d3Tu3Bl79+6V6TwLCwt89dVXOHbsGIYOHYqoqCgAtfdTx44dUVZWhoyMDO47iUTSpGscE0IIIaR5oQFvA1WPIj5x4gQuXboEf3//WqOIDxw4gIKCApw7dw4rVqzA4cO1L01VPYr4v//+w4sXL6SiiK9du4bdu3c3KIpYFjNnzsSqVavw448/QiKRYM6cOcjKysKUKVMAAGZmZjAyMkJYWBjy8vJw+PDhGitS1KagoABz585Famoqbty4gWPHjiEvL6/eebXPnj1DcHAwkpOTcePGDaSkpCA9PZ07TyQSoaSkBElJSfjvv//w9OlTWFpaws3NDRMnTkRaWhoyMjIwYcIECASCt+8gQgghhLQMrBE4OzuzKVOmNMal3sqTJ0/Y0KFDmZqaGgPAHjx40CTtPH78mH3xxRdMRUWF6enpsdWrV0v1QWlpKVu0aBETiURMUVGRGRgYMHt7eyYUChljjEVFRTENDQ3ues+fP2eff/4509TUZABYVFQUY4yxdevWMQMDAyYQCJirqyv7/vvvZb6vV9tgjLHY2FhW/Y+8vLychYWFsXbt2jFFRUVma2vL4uPjpc45e/Yss7GxYcrKyqxPnz7s559/ZgBYQUHBa9u5ffs28/LyYgYGBkxJSYlpaGgwPT09Vl5eXmfNL168YCNHjmRGRkZMSUmJGRoasuDgYPbs2TPumKCgIKatrc0AsNDQUMYYY0VFRWzw4MGMz+czY2Nj9v333zMTExO2fv36evupSnFxMQPAiouLZT6HEEIIIe+XrP9+N0q0cENidZvSli1bEBoaihMnTqBt27bQ09OTaSmsd6Gx44Q/lD6XRVhYGOLi4ho0HeJdo2hhQgghpPmR9d/vFvXSWn5+PqysrNC5c+fXHlNaWgolJaV3WFXLRv0pG0paI6T5opdOCWn+GjyHt76Usd27d6N79+5QU1ODvr4+Ro8ezS2FxRiDmZkZ1q5dK3VOVlYWeDwerl69CsYYwsLCYGxsDD6fD0NDQ4SEhNRbV79+/RAREYHTp0+Dx+OhX79+ACrnfS5ZsgS+vr5QV1dHYGAgAGD//v2wtrYGn8+HSCSqcR9V695W3auJiQkOHjyIu3fvwtPTE0KhEF26dMH58+cb2oUAKgfnnp6e0NPTg1AohIODAxc8UWXz5s0wNzeHsrIy9PT0MGzYMACVQQunTp3Chg0buNQxVVVVCIVCCIXCGmv9Av+fmnb48GF06dIFysrK6NWrFy5dusQdExYWxoVnVPn6668hEom4z/7+/vDy8sKyZctgaGjILcf2999/Y9SoUdDS0oKqqiq6d+8u9WIfUPmzoaSkBB6PB0VFRa5eoVAIgUAACwsLaGpqQltbGx4eHsjPz+fOLS0tRXBwMAwMDKCsrAwTExOsWLGC2//w4UNMmDABOjo6UFdXx4ABAxq8vBohhBBCWqYGD3irp4wdO3YMycnJuHDhArf/5cuXWLJkCbKzsxEXF4fr16/D398fQOWyWOPGjePeuq8SFRWFvn37wszMDPv378f69euxdetW5OXlIS4uDjY2NvXWdeDAAQQEBMDR0RFFRUU4cOAAt2/t2rWwtbVFZmYmFi5ciIyMDAwfPhwjR45ETk4OwsLCsHDhQkRHR0tdc/369XByckJmZiYGDx6MMWPGwNfXF1988QUuXLiADh06wNfXt0FLfVUpKSmBu7s7kpKSkJmZCTc3N4jFYhQWFgIAzp8/j5CQECxevBgSiQQJCQno27cvgMqENHt7ewwfPhy///47fv/9d1y4cAFZWVnIyspCUFDQa9udOXMmIiIikJ6eDh0dHYjF4lqXGqtLUlISJBIJEhMTcejQIZSUlMDZ2Rn//PMPDh48iOzsbMyaNQsVFRXcOfn5+YiLi8OhQ4ewd+9eaGpqYsyYMVzNa9asQWhoKM6fP4+kpCTIyclhyJAh3DU2btyIgwcP4qeffoJEIsGePXukBuLe3t64c+cO4uPjkZGRAXt7ewwcOBD379+v9R4oWpgQQghpPRo0paG+lDEAGDduHPffpqam2LhxIxwcHFBSUgKhUAh/f38sWrQI586dQ48ePfDy5Uvs3buXe+pbWFgIfX19uLi4QFFREcbGxujRo0e9tWlpaUFFRQVKSko1lv8aMGAApk+fzn328fHBwIEDsXDhQgCVKytcuXIFa9as4QbnAODu7o6JEycCABYtWoQtW7bAwcEB3t7eAIDZs2fD0dER//7772uXHHsdW1tbLpkMAJYsWYLY2FgcPHgQwcHBKCwshKqqKjw8PKCmpgYTExPY2dkBqEwcU1NTg4GBARwdHRvUbmhoKD755BMA//9nFxsbi+HDh8t8DVVVVWzfvp2byrBt2zbcvXsX6enp0NLSAlC5wkN1FRUViI6OhpqaGgAgMzMTp0+f5o4LDg6WOn7nzp3Q0dHBlStX0LlzZxQWFsLc3Bwff/wxeDweF3UMAGfPnsW5c+dw584dLkxi7dq1iIuLwy+//MI91a+OooUJIYSQ1qNBT3jrSxkDKtePFYvFMDY2hpqaGpydnQGAe3JpaGiIwYMHY+fOnQCA3377DS9evOAGkd7e3nj27BlMTU0REBCA2NhYlJWVvdVNdu/eXepzbm5urQljeXl5Uuu8Vk8P09PTAwCpp81V31VN2WiIkpISzJgxA1ZWVtDU1IRQKERubi7XT5988glMTExgamqKMWPGYM+ePXj69GmD23lV9QFy1Z9dbm5ug65hY2MjNW83KysLdnZ23GC3NiKRiBvsAtKpbwCQl5eHUaNGwdTUFOrq6tzT26r+8Pf3R1ZWFiwtLRESEoJjx45x52ZnZ6OkpATa2tpS0yQKCgqkpkVUN3fuXBQXF3PbzZs3G9QHhBBCCGk+GnUd3idPnsDV1RXq6urYs2cP0tPTERsbC6ByDmaVCRMmYN++fXj27BmioqIwYsQILjnMyMgIEokEmzdvhkAgwKRJk9C3b98G/9q9OlVV1Tc6r3pSWdVqD7V9V/1X97KaMWMGYmNjsXz5cpw5cwZZWVmwsbHh+klNTQ0XLlxATEwMDAwMsGjRItja2jZpoIKcnFyN6Rm19fur/SnLmre1pb5V7zexWIz79+8jMjISaWlp3Pzfqv6wt7dHQUEBlixZgmfPnmH48OHcnOaSkhIYGBhw0yOqNolEgpkzZ9ZaD5/Ph7q6utRGCCGEkJapQVMaqqeMGRsbA/j/lDFnZ2f8+eefuHfvHlauXAkjIyMAqPWlLnd3d6iqqmLLli1ISEjA6dOnpfYLBAKIxWKIxWJMnjwZHTt2RE5ODuzt7d/0PqVYWVnVmjBmYWFRIw64qaSkpMDf35+LzC0pKcH169eljlFQUICLiwtcXFwQGhoKTU1NnDhxAkOHDq01dUwWf/zxR40/u6pgBx0dHdy+fRuMMW4wL8tSYl26dMH27dtx//79Op/yvs69e/cgkUgQGRmJPn36AKicpvAqdXV1jBgxAiNGjMCwYcPg5uaG+/fvw97eHrdv34aCgoLUvN4PCb3lTQghhLw/DRrwVk8Z09bWhq6uLubPn8+ljBkbG0NJSQmbNm1CUFAQLl26hCVLltS4jry8PPz9/TF37lyYm5tL/Zo9Ojoa5eXl6NmzJ1RUVPDDDz9AIBBIzdl8W9OnT4eDgwOWLFmCESNGIDU1Fd988w02b97caG3Ux9zcHAcOHIBYLAaPx8PChQulnngeOnQI165dQ9++fdGmTRscOXIEFRUV3PQRkUiEtLQ0XL9+HUKhEFpaWtyfQ10WL14MbW1t6OnpYf78+Wjbti28vLwAVK50cffuXaxevRrDhg1DQkIC4uPj6336OWrUKCxfvhxeXl5YsWIFDAwMkJmZCUNDQ5nmGLdp0wba2trYtm0bDAwMUFhYiDlz5kgds27dOhgYGMDOzg5ycnL4+eefoa+vD01NTbi4uMDR0RFeXl5YvXo1F8N8+PBhDBkypMaUltpUPdmml9cIIYSQ5qPq3+16FxBoaKJFfSlje/fuZSKRiPH5fObo6MgOHjzIALDMzEyp6+Tn5zMAbPXq1VLfx8bGsp49ezJ1dXWmqqrKevXqxY4fPy5TbVOmTGHOzs5S370uceuXX35hnTp1YoqKiszY2JitWbOm3vMAsNjYWO5zQUFBrfdWm1cTyQoKClj//v2ZQCBgRkZG7JtvvpHqxzNnzjBnZ2fWpk0bJhAIWJcuXdiPP/7InS+RSFivXr2YQCCQSj57nZMnTzIA7LfffmPW1tZMSUmJ9ejRg2VnZ0sdt2XLFmZkZMRUVVWZr68vW7ZsGTMxMeH2+/n5MU9PzxrXv379Ovv888+Zuro6U1FRYd27d2dpaWmMMcZCQ0OZra2t1PHr16+Xum5iYiKzsrJifD6fdenShSUnJ0v197Zt21jXrl2ZqqoqU1dXZwMHDmQXLlzgzn/06BH73//+xwwNDZmioiIzMjJiPj4+rLCwsM5+qXLz5k0GgDbaaKONNtpoa4bbzZs36/x3vlGS1t7EmTNnMHDgQNy8eZN7+Ys0neTkZPTv3x8PHjyApqbm+y7ng1NRUYFbt25BTU3tjdL5Hj16BCMjI9y8eZPmA78F6sfGQ33ZOKgfGwf1Y+OgfqyJMYbHjx/D0NCwzt90v/OktRcvXuDu3bsICwuDt7c3DXbJB0FOTk5qeb03RS/ANQ7qx8ZDfdk4qB8bB/Vj46B+lKahoVHvMY26SoMsYmJiYGJigocPH2L16tUyn3fmzBmpJade3d63QYMGvba22pLPGltQUNBr268riIIQQgghpKV7b1MaGurZs2f4559/Xrv/1aCDd+2ff/7Bs2fPat2npaX1RqsXNMSdO3de+8KVuro6dHV1m7T91u7Ro0fQ0NBAcXEx/V/3W6B+bDzUl42D+rFxUD82DurHN/fOpzS8KYFA8N4HtXVp167de21fV1eXBrXvEZ/PR2hoKJf0Rt4M9WPjob5sHNSPjYP6sXFQP765ZvOElxBCCCGEkDfxzufwEkIIIYQQ8i7RgJcQQgghhLRoNOAlhBBCCCEtGg14CSGEEEJIi0YDXkIIIYQQ0qLRgJeQN7Rs2TL07t0bKioqMsc1+/v7g8fjSW1ubm5NW+gH7k36kTGGRYsWwcDAAAKBAC4uLsjLy2vaQj9w9+/fh4+PD9TV1aGpqYnx48ejpKSkznP69etX4+exNQbVfPvttxCJRFBWVkbPnj1x7ty5Oo//+eef0bFjRygrK8PGxgZHjhx5R5V+2BrSj9HR0TV+9pSVld9htR+m06dPQywWw9DQEDweD3FxcfWek5ycDHt7e/D5fJiZmSE6OrrJ62yOaMBLyBsqLS2Ft7c3vvzyywad5+bmhqKiIm6LiYlpogqbhzfpx9WrV2Pjxo347rvvkJaWBlVVVbi6uuL58+dNWOmHzcfHB5cvX0ZiYiIOHTqE06dPIzAwsN7zAgICpH4eG5KA2RL8+OOPmDZtGkJDQ3HhwgXY2trC1dUVd+7cqfX433//HaNGjcL48eORmZkJLy8veHl54dKlS++48g9LQ/sRqAxFqv6zd+PGjXdY8YfpyZMnsLW1xbfffivT8QUFBRg8eDD69++PrKwsTJ06FRMmTMDRo0ebuNJmiBFC3kpUVBTT0NCQ6Vg/Pz/m6enZpPU0V7L2Y0VFBdPX12dr1qzhvnv48CHj8/ksJiamCSv8cF25coUBYOnp6dx38fHxjMfjsX/++ee15zk7O7MpU6a8gwo/XD169GCTJ0/mPpeXlzNDQ0O2YsWKWo8fPnw4Gzx4sNR3PXv2ZBMnTmzSOj90De3Hhvy92VoBYLGxsXUeM2vWLGZtbS313YgRI5irq2sTVtY80RNeQt6x5ORk6OrqwtLSEl9++SXu3bv3vktqVgoKCnD79m24uLhw32loaKBnz55ITU19j5W9P6mpqdDU1ET37t2571xcXCAnJ4e0tLQ6z92zZw/atm2Lzp07Y+7cuXj69GlTl/vBKC0tRUZGhtTPkpycHFxcXF77s5Samip1PAC4urq22p894M36EQBKSkpgYmICIyMjeHp64vLly++i3BaFfh5l12yihQlpCdzc3DB06FB89NFHyM/Px7x58zBo0CCkpqZCXl7+fZfXLNy+fRsAoKenJ/W9np4et6+1uX37do1ocQUFBWhpadXZJ6NHj4aJiQkMDQ1x8eJFzJ49GxKJBAcOHGjqkj8I//33H8rLy2v9Wfrzzz9rPef27dv0s/eKN+lHS0tL7Ny5E126dEFxcTHWrl2L3r174/Lly2jfvv27KLtFeN3P46NHj/Ds2TMIBIL3VNmHh57wElLNnDlzarxI8er2ur/AZTFy5Eh89tlnsLGxgZeXFw4dOoT09HQkJyc33k18AJq6H1uLpu7HwMBAuLq6wsbGBj4+Pvj+++8RGxuL/Pz8RrwLQmpydHSEr68vunbtCmdnZxw4cAA6OjrYunXr+y6NtFD0hJeQaqZPnw5/f/86jzE1NW209kxNTdG2bVtcvXoVAwcObLTrvm9N2Y/6+voAgH///RcGBgbc9//++y+6du36Rtf8UMnaj/r6+jVeDiorK8P9+/e5/pJFz549AQBXr15Fhw4dGlxvc9O2bVvIy8vj33//lfr+33//fW2/6evrN+j41uBN+vFVioqKsLOzw9WrV5uixBbrdT+P6urq9HT3FTTgJaQaHR0d6OjovLP2/v77b9y7d09q4NYSNGU/fvTRR9DX10dSUhI3wH306BHS0tIavGLGh07WfnR0dMTDhw+RkZGBbt26AQBOnDiBiooKbhAri6ysLABocT+Pr6OkpIRu3bohKSkJXl5eAICKigokJSUhODi41nMcHR2RlJSEqVOnct8lJibC0dHxHVT8YXqTfnxVeXk5cnJy4O7u3oSVtjyOjo41lsVr7T+Pr/W+35ojpLm6ceMGy8zMZOHh4UwoFLLMzEyWmZnJHj9+zB1jaWnJDhw4wBhj7PHjx2zGjBksNTWVFRQUsOPHjzN7e3tmbm7Onj9//r5u471raD8yxtjKlSuZpqYm+/XXX9nFixeZp6cn++ijj9izZ8/exy18ENzc3JidnR1LS0tjZ8+eZebm5mzUqFHc/r///ptZWlqytLQ0xhhjV69eZYsXL2bnz59nBQUF7Ndff2Wmpqasb9++7+sW3ot9+/YxPp/PoqOj2ZUrV1hgYCDT1NRkt2/fZowxNmbMGDZnzhzu+JSUFKagoMDWrl3LcnNzWWhoKFNUVGQ5OTnv6xY+CA3tx/DwcHb06FGWn5/PMjIy2MiRI5mysjK7fPny+7qFD8Ljx4+5vwMBsHXr1rHMzEx248YNxhhjc+bMYWPGjOGOv3btGlNRUWEzZ85kubm57Ntvv2Xy8vIsISHhfd3CB4sGvIS8IT8/Pwagxnby5EnuGAAsKiqKMcbY06dP2aeffsp0dHSYoqIiMzExYQEBAdw/CK1VQ/uRscqlyRYuXMj09PQYn89nAwcOZBKJ5N0X/wG5d+8eGzVqFBMKhUxdXZ2NHTtW6n8aCgoKpPq1sLCQ9e3bl2lpaTE+n8/MzMzYzJkzWXFx8Xu6g/dn06ZNzNjYmCkpKbEePXqwP/74g9vn7OzM/Pz8pI7/6aefmIWFBVNSUmLW1tbs8OHD77jiD1ND+nHq1KncsXp6eszd3Z1duHDhPVT9YTl58mStfx9W9Z2fnx9zdnaucU7Xrl2ZkpISMzU1lfq7kvw/HmOMveOHyoQQQgghhLwztEoDIYQQQghp0WjASwghhBBCWjQa8BJCCCGEkBaNBryEEEIIIaRFowEvIYQQQghp0WjASwghhBBCWjQa8BJCCCGEkBaNBryEEEIIIaRFowEvIYQQQghp0WjASwghhBBCWjQa8BJCCCGEkBbt/wBRGwFVgr5/HAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(model_class.named_steps['logreg'].coef_[0], index=X_train_val.columns).sort_values().plot(kind='barh')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_class = model_class.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub = pd.DataFrame()\n",
    "\n",
    "sub['customer_id'] = data_test['customer_id_']\n",
    "sub['date_diff_post'] = pred_reg\n",
    "sub['buy_post'] = pred_class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub.to_csv('logistic.csv', sep=';', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "customer_id       0\n",
       "date_diff_post    0\n",
       "buy_post          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sub.isna().sum()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ensemble"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "estimators = [\n",
    "    (\n",
    "        'lgbm1', \n",
    "        LGBMClassifier(\n",
    "            n_estimators=1000, \n",
    "            reg_alpha=20, \n",
    "            max_depth=3, \n",
    "            importance_type='gain', \n",
    "            reg_lambda=20, \n",
    "            class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1,\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm2', \n",
    "        LGBMClassifier(\n",
    "            n_estimators=500, \n",
    "            reg_alpha=20, \n",
    "            max_depth=3, \n",
    "            importance_type='gain', \n",
    "            reg_lambda=10, \n",
    "            class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1,\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm3', \n",
    "        LGBMClassifier(\n",
    "            n_estimators=1000, \n",
    "            reg_alpha=20, \n",
    "            max_depth=4, \n",
    "            importance_type='gain', \n",
    "            reg_lambda=10, \n",
    "            class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1,\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm4',\n",
    "        LGBMClassifier(\n",
    "            n_estimators=1000,\n",
    "            max_depth=5,\n",
    "            reg_lambda=20, \n",
    "            class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm5',\n",
    "        LGBMClassifier(\n",
    "            n_estimators=500,\n",
    "            max_depth=5,\n",
    "            reg_lambda=20, \n",
    "            class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1\n",
    "        ),\n",
    "    )\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-4 {color: black;background-color: white;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-4 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>StackingClassifier(estimators=[(&#x27;lgbm1&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               importance_type=&#x27;gain&#x27;,\n",
       "                                               max_depth=3, n_estimators=1000,\n",
       "                                               reg_alpha=20, reg_lambda=20,\n",
       "                                               verbose=-1)),\n",
       "                               (&#x27;lgbm2&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               importance_type=&#x27;gain&#x27;,\n",
       "                                               max_depth=3, n_estimators=500,\n",
       "                                               reg_alpha=20, reg_lambda=10,\n",
       "                                               verbose=-1)),\n",
       "                               (&#x27;lgbm3&#x27;,\n",
       "                                LGBMClassifier(class_we...\n",
       "                                               reg_alpha=20, reg_lambda=10,\n",
       "                                               verbose=-1)),\n",
       "                               (&#x27;lgbm4&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               max_depth=5, n_estimators=1000,\n",
       "                                               reg_lambda=20, verbose=-1)),\n",
       "                               (&#x27;lgbm5&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               max_depth=5, n_estimators=500,\n",
       "                                               reg_lambda=20, verbose=-1))],\n",
       "                   final_estimator=LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                                  max_depth=3, reg_lambda=20,\n",
       "                                                  verbose=-1))</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-22\" type=\"checkbox\" ><label for=\"sk-estimator-id-22\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StackingClassifier</label><div class=\"sk-toggleable__content\"><pre>StackingClassifier(estimators=[(&#x27;lgbm1&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               importance_type=&#x27;gain&#x27;,\n",
       "                                               max_depth=3, n_estimators=1000,\n",
       "                                               reg_alpha=20, reg_lambda=20,\n",
       "                                               verbose=-1)),\n",
       "                               (&#x27;lgbm2&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               importance_type=&#x27;gain&#x27;,\n",
       "                                               max_depth=3, n_estimators=500,\n",
       "                                               reg_alpha=20, reg_lambda=10,\n",
       "                                               verbose=-1)),\n",
       "                               (&#x27;lgbm3&#x27;,\n",
       "                                LGBMClassifier(class_we...\n",
       "                                               reg_alpha=20, reg_lambda=10,\n",
       "                                               verbose=-1)),\n",
       "                               (&#x27;lgbm4&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               max_depth=5, n_estimators=1000,\n",
       "                                               reg_lambda=20, verbose=-1)),\n",
       "                               (&#x27;lgbm5&#x27;,\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               max_depth=5, n_estimators=500,\n",
       "                                               reg_lambda=20, verbose=-1))],\n",
       "                   final_estimator=LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                                  max_depth=3, reg_lambda=20,\n",
       "                                                  verbose=-1))</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm1</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-23\" type=\"checkbox\" ><label for=\"sk-estimator-id-23\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMClassifier</label><div class=\"sk-toggleable__content\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, importance_type=&#x27;gain&#x27;,\n",
       "               max_depth=3, n_estimators=1000, reg_alpha=20, reg_lambda=20,\n",
       "               verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm2</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-24\" type=\"checkbox\" ><label for=\"sk-estimator-id-24\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMClassifier</label><div class=\"sk-toggleable__content\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, importance_type=&#x27;gain&#x27;,\n",
       "               max_depth=3, n_estimators=500, reg_alpha=20, reg_lambda=10,\n",
       "               verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm3</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-25\" type=\"checkbox\" ><label for=\"sk-estimator-id-25\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMClassifier</label><div class=\"sk-toggleable__content\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, importance_type=&#x27;gain&#x27;,\n",
       "               max_depth=4, n_estimators=1000, reg_alpha=20, reg_lambda=10,\n",
       "               verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm4</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-26\" type=\"checkbox\" ><label for=\"sk-estimator-id-26\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMClassifier</label><div class=\"sk-toggleable__content\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, max_depth=5, n_estimators=1000,\n",
       "               reg_lambda=20, verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm5</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-27\" type=\"checkbox\" ><label for=\"sk-estimator-id-27\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMClassifier</label><div class=\"sk-toggleable__content\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, max_depth=5, n_estimators=500,\n",
       "               reg_lambda=20, verbose=-1)</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>final_estimator</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-28\" type=\"checkbox\" ><label for=\"sk-estimator-id-28\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMClassifier</label><div class=\"sk-toggleable__content\"><pre>LGBMClassifier(class_weight={0: 0.7, 1: 0.3}, max_depth=3, reg_lambda=20,\n",
       "               verbose=-1)</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "StackingClassifier(estimators=[('lgbm1',\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               importance_type='gain',\n",
       "                                               max_depth=3, n_estimators=1000,\n",
       "                                               reg_alpha=20, reg_lambda=20,\n",
       "                                               verbose=-1)),\n",
       "                               ('lgbm2',\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               importance_type='gain',\n",
       "                                               max_depth=3, n_estimators=500,\n",
       "                                               reg_alpha=20, reg_lambda=10,\n",
       "                                               verbose=-1)),\n",
       "                               ('lgbm3',\n",
       "                                LGBMClassifier(class_we...\n",
       "                                               reg_alpha=20, reg_lambda=10,\n",
       "                                               verbose=-1)),\n",
       "                               ('lgbm4',\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               max_depth=5, n_estimators=1000,\n",
       "                                               reg_lambda=20, verbose=-1)),\n",
       "                               ('lgbm5',\n",
       "                                LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                               max_depth=5, n_estimators=500,\n",
       "                                               reg_lambda=20, verbose=-1))],\n",
       "                   final_estimator=LGBMClassifier(class_weight={0: 0.7, 1: 0.3},\n",
       "                                                  max_depth=3, reg_lambda=20,\n",
       "                                                  verbose=-1))"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stack_model = StackingClassifier(\n",
    "    estimators,\n",
    "    LGBMClassifier(\n",
    "        n_estimators=100,\n",
    "        max_depth=3,\n",
    "        reg_lambda=20, \n",
    "        class_weight={0: 0.7, 1:0.3},\n",
    "        verbose=-1\n",
    "    ),\n",
    ")\n",
    "\n",
    "stack_model.fit(X_train, y_train_class)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7518152703806135"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_val_class, stack_model.predict_proba(X_val)[:, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7506196938712276"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(y_val_class, stack_model.predict(X_val))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7554322412906606"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(y_train_class, stack_model.predict(X_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn import metrics\n",
    "fpr, tpr, thresholds = metrics.roc_curve(y_val_class, stack_model.predict_proba(X_val)[:, 1])\n",
    "auc = roc_auc_score(y_val_class, stack_model.predict_proba(X_val)[:, 1])\n",
    "\n",
    "#create ROC curve\n",
    "plt.plot(fpr,tpr,label=\"AUC=\"+str(round(auc, 3))+' for ensemble', c='brown')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.legend(loc=4)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_class = stack_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "estimators = [\n",
    "    (\n",
    "        'lgbm1', \n",
    "        LGBMRegressor(\n",
    "            n_estimators=1000, \n",
    "            reg_alpha=20, \n",
    "            max_depth=3, \n",
    "            importance_type='gain', \n",
    "            reg_lambda=20, \n",
    "            # class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1,\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm2', \n",
    "        LGBMRegressor(\n",
    "            n_estimators=500, \n",
    "            reg_alpha=20, \n",
    "            max_depth=3, \n",
    "            importance_type='gain', \n",
    "            reg_lambda=10, \n",
    "            # class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1,\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm3', \n",
    "        LGBMRegressor(\n",
    "            n_estimators=1000, \n",
    "            reg_alpha=20, \n",
    "            max_depth=4, \n",
    "            importance_type='gain', \n",
    "            reg_lambda=10, \n",
    "            # class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1,\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm4',\n",
    "        LGBMRegressor(\n",
    "            n_estimators=1000,\n",
    "            max_depth=5,\n",
    "            reg_lambda=20, \n",
    "            # class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1\n",
    "        ),\n",
    "    ),\n",
    "    (\n",
    "        'lgbm5',\n",
    "        LGBMRegressor(\n",
    "            n_estimators=500,\n",
    "            max_depth=5,\n",
    "            reg_lambda=20, \n",
    "            # class_weight={0: 0.7, 1:0.3},\n",
    "            verbose=-1\n",
    "        ),\n",
    "    )\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-5 {color: black;background-color: white;}#sk-container-id-5 pre{padding: 0;}#sk-container-id-5 div.sk-toggleable {background-color: white;}#sk-container-id-5 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-5 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-5 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-5 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-5 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-5 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-5 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-5 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-5 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-5 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-5 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-5 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-5 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-5 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-5 div.sk-item {position: relative;z-index: 1;}#sk-container-id-5 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-5 div.sk-item::before, #sk-container-id-5 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-5 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-5 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-5 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-5 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-5 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-5 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-5 div.sk-label-container {text-align: center;}#sk-container-id-5 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-5 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-5\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>StackingRegressor(estimators=[(&#x27;lgbm1&#x27;,\n",
       "                               LGBMRegressor(importance_type=&#x27;gain&#x27;,\n",
       "                                             max_depth=3, n_estimators=1000,\n",
       "                                             reg_alpha=20, reg_lambda=20,\n",
       "                                             verbose=-1)),\n",
       "                              (&#x27;lgbm2&#x27;,\n",
       "                               LGBMRegressor(importance_type=&#x27;gain&#x27;,\n",
       "                                             max_depth=3, n_estimators=500,\n",
       "                                             reg_alpha=20, reg_lambda=10,\n",
       "                                             verbose=-1)),\n",
       "                              (&#x27;lgbm3&#x27;,\n",
       "                               LGBMRegressor(importance_type=&#x27;gain&#x27;,\n",
       "                                             max_depth=4, n_estimators=1000,\n",
       "                                             reg_alpha=20, reg_lambda=10,\n",
       "                                             verbose=-1)),\n",
       "                              (&#x27;lgbm4&#x27;,\n",
       "                               LGBMRegressor(max_depth=5, n_estimators=1000,\n",
       "                                             reg_lambda=20, verbose=-1)),\n",
       "                              (&#x27;lgbm5&#x27;,\n",
       "                               LGBMRegressor(max_depth=5, n_estimators=500,\n",
       "                                             reg_lambda=20, verbose=-1))],\n",
       "                  final_estimator=LGBMRegressor(max_depth=3, reg_lambda=20,\n",
       "                                                verbose=-1))</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-29\" type=\"checkbox\" ><label for=\"sk-estimator-id-29\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StackingRegressor</label><div class=\"sk-toggleable__content\"><pre>StackingRegressor(estimators=[(&#x27;lgbm1&#x27;,\n",
       "                               LGBMRegressor(importance_type=&#x27;gain&#x27;,\n",
       "                                             max_depth=3, n_estimators=1000,\n",
       "                                             reg_alpha=20, reg_lambda=20,\n",
       "                                             verbose=-1)),\n",
       "                              (&#x27;lgbm2&#x27;,\n",
       "                               LGBMRegressor(importance_type=&#x27;gain&#x27;,\n",
       "                                             max_depth=3, n_estimators=500,\n",
       "                                             reg_alpha=20, reg_lambda=10,\n",
       "                                             verbose=-1)),\n",
       "                              (&#x27;lgbm3&#x27;,\n",
       "                               LGBMRegressor(importance_type=&#x27;gain&#x27;,\n",
       "                                             max_depth=4, n_estimators=1000,\n",
       "                                             reg_alpha=20, reg_lambda=10,\n",
       "                                             verbose=-1)),\n",
       "                              (&#x27;lgbm4&#x27;,\n",
       "                               LGBMRegressor(max_depth=5, n_estimators=1000,\n",
       "                                             reg_lambda=20, verbose=-1)),\n",
       "                              (&#x27;lgbm5&#x27;,\n",
       "                               LGBMRegressor(max_depth=5, n_estimators=500,\n",
       "                                             reg_lambda=20, verbose=-1))],\n",
       "                  final_estimator=LGBMRegressor(max_depth=3, reg_lambda=20,\n",
       "                                                verbose=-1))</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm1</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-30\" type=\"checkbox\" ><label for=\"sk-estimator-id-30\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMRegressor</label><div class=\"sk-toggleable__content\"><pre>LGBMRegressor(importance_type=&#x27;gain&#x27;, max_depth=3, n_estimators=1000,\n",
       "              reg_alpha=20, reg_lambda=20, verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm2</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-31\" type=\"checkbox\" ><label for=\"sk-estimator-id-31\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMRegressor</label><div class=\"sk-toggleable__content\"><pre>LGBMRegressor(importance_type=&#x27;gain&#x27;, max_depth=3, n_estimators=500,\n",
       "              reg_alpha=20, reg_lambda=10, verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm3</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-32\" type=\"checkbox\" ><label for=\"sk-estimator-id-32\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMRegressor</label><div class=\"sk-toggleable__content\"><pre>LGBMRegressor(importance_type=&#x27;gain&#x27;, max_depth=4, n_estimators=1000,\n",
       "              reg_alpha=20, reg_lambda=10, verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm4</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-33\" type=\"checkbox\" ><label for=\"sk-estimator-id-33\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMRegressor</label><div class=\"sk-toggleable__content\"><pre>LGBMRegressor(max_depth=5, n_estimators=1000, reg_lambda=20, verbose=-1)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lgbm5</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-34\" type=\"checkbox\" ><label for=\"sk-estimator-id-34\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMRegressor</label><div class=\"sk-toggleable__content\"><pre>LGBMRegressor(max_depth=5, n_estimators=500, reg_lambda=20, verbose=-1)</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>final_estimator</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-35\" type=\"checkbox\" ><label for=\"sk-estimator-id-35\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LGBMRegressor</label><div class=\"sk-toggleable__content\"><pre>LGBMRegressor(max_depth=3, reg_lambda=20, verbose=-1)</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "StackingRegressor(estimators=[('lgbm1',\n",
       "                               LGBMRegressor(importance_type='gain',\n",
       "                                             max_depth=3, n_estimators=1000,\n",
       "                                             reg_alpha=20, reg_lambda=20,\n",
       "                                             verbose=-1)),\n",
       "                              ('lgbm2',\n",
       "                               LGBMRegressor(importance_type='gain',\n",
       "                                             max_depth=3, n_estimators=500,\n",
       "                                             reg_alpha=20, reg_lambda=10,\n",
       "                                             verbose=-1)),\n",
       "                              ('lgbm3',\n",
       "                               LGBMRegressor(importance_type='gain',\n",
       "                                             max_depth=4, n_estimators=1000,\n",
       "                                             reg_alpha=20, reg_lambda=10,\n",
       "                                             verbose=-1)),\n",
       "                              ('lgbm4',\n",
       "                               LGBMRegressor(max_depth=5, n_estimators=1000,\n",
       "                                             reg_lambda=20, verbose=-1)),\n",
       "                              ('lgbm5',\n",
       "                               LGBMRegressor(max_depth=5, n_estimators=500,\n",
       "                                             reg_lambda=20, verbose=-1))],\n",
       "                  final_estimator=LGBMRegressor(max_depth=3, reg_lambda=20,\n",
       "                                                verbose=-1))"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import StackingRegressor\n",
    "stack_model = StackingRegressor(\n",
    "    estimators,\n",
    "    LGBMRegressor(\n",
    "        n_estimators=100,\n",
    "        max_depth=3,\n",
    "        reg_lambda=20, \n",
    "        # class_weight={0: 0.7, 1:0.3},\n",
    "        verbose=-1\n",
    "    ),\n",
    ")\n",
    "\n",
    "stack_model.fit(X_train, y_train_reg.fillna(0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16.642235368154747"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_squared_error(y_val_reg.fillna(0), stack_model.predict(X_val), squared=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16.48287847968275"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_squared_error(y_train_reg.fillna(0), stack_model.predict(X_train), squared= False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_reg = stack_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub = pd.DataFrame()\n",
    "\n",
    "sub['customer_id'] = data_test['customer_id_']\n",
    "sub['date_diff_post'] = pred_reg\n",
    "sub['buy_post'] = pred_class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub.to_csv('stacking6.csv', sep=';', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "customer_id       2.074861e+07\n",
       "date_diff_post    1.755663e+01\n",
       "buy_post          4.913205e-01\n",
       "dtype: float64"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sub.mean()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "env",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "1d85f8d091281e8cdbbabc30d35abdaebd4eb091dd84d434d9fa6db6a3942f76"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}