checking_cgm_missing_values.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.signal import find_peaks, peak_prominences\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>ID</th>\n",
       "      <th>Gender</th>\n",
       "      <th>HbA1c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13</td>\n",
       "      <td>MALE</td>\n",
       "      <td>5.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>5.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>6.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>5.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2</td>\n",
       "      <td>MALE</td>\n",
       "      <td>5.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>5.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>5.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>5.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>9</td>\n",
       "      <td>MALE</td>\n",
       "      <td>6.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>11</td>\n",
       "      <td>MALE</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>12</td>\n",
       "      <td>MALE</td>\n",
       "      <td>5.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>14</td>\n",
       "      <td>MALE</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>16</td>\n",
       "      <td>MALE</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    ID  Gender  HbA1c\n",
       "0   13    MALE    5.7\n",
       "1    1  FEMALE    5.5\n",
       "2    3  FEMALE    5.9\n",
       "3    4  FEMALE    6.4\n",
       "4    5  FEMALE    5.7\n",
       "5    2    MALE    5.6\n",
       "6    6  FEMALE    5.8\n",
       "7    7  FEMALE    5.3\n",
       "8    8  FEMALE    5.6\n",
       "9   10  FEMALE    6.0\n",
       "10   9    MALE    6.1\n",
       "11  11    MALE    6.0\n",
       "12  12    MALE    5.6\n",
       "13  14    MALE    5.5\n",
       "14  15  FEMALE    5.5\n",
       "15  16    MALE    5.5"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dg_df = pd.read_csv('./big-ideas-lab-glycemic-variability-and-wearable-device-data-1.1.2/Demographics.csv')\n",
    "dg_df.head(16)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ID별 CGM 결측치 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ID: 1] Now processing ...\n",
      "[ID 1] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-02-15 09:08:30      NaN        NaN\n",
      "2020-02-15 09:13:30      NaN        NaN\n",
      "2020-02-19 23:43:26      NaN        NaN\n",
      "2020-02-19 23:48:26      NaN        NaN\n",
      "2020-02-19 23:53:26      NaN        NaN\n",
      "2020-02-19 23:58:26      NaN        NaN\n",
      "2020-02-20 00:03:26      NaN        NaN\n",
      "2020-02-20 00:08:26      NaN        NaN\n",
      "2020-02-20 00:13:26      NaN        NaN\n",
      "2020-02-20 00:18:26      NaN        NaN\n",
      "2020-02-20 00:23:26      NaN        NaN\n",
      "2020-02-20 00:28:26      NaN        NaN\n",
      "2020-02-20 00:33:26      NaN        NaN\n",
      "2020-02-20 00:38:26      NaN        NaN\n",
      "2020-02-20 00:43:26      NaN        NaN\n",
      "2020-02-20 00:48:26      NaN        NaN\n",
      "2020-02-20 00:53:26      NaN        NaN\n",
      "2020-02-20 00:58:26      NaN        NaN\n",
      "2020-02-20 01:03:26      NaN        NaN\n",
      "2020-02-20 01:08:26      NaN        NaN\n",
      "2020-02-20 01:13:26      NaN        NaN\n",
      "2020-02-20 01:18:26      NaN        NaN\n",
      "2020-02-20 01:23:26      NaN        NaN\n",
      "2020-02-20 01:28:26      NaN        NaN\n",
      "2020-02-20 01:33:26      NaN        NaN\n",
      "2020-02-20 01:38:26      NaN        NaN\n",
      "2020-02-20 01:43:26      NaN        NaN\n",
      "2020-02-20 01:48:26      NaN        NaN\n",
      "2020-02-20 01:53:26      NaN        NaN\n",
      "2020-02-20 01:58:26      NaN        NaN\n",
      "2020-02-20 02:03:26      NaN        NaN\n",
      "2020-02-20 02:08:26      NaN        NaN\n",
      "2020-02-20 02:13:26      NaN        NaN\n",
      "2020-02-20 02:18:26      NaN        NaN\n",
      "2020-02-20 02:23:26      NaN        NaN\n",
      "2020-02-20 02:28:26      NaN        NaN\n",
      "2020-02-20 02:33:26      NaN        NaN\n",
      "2020-02-20 02:38:26      NaN        NaN\n",
      "2020-02-20 02:43:26      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 2] Now processing ...\n",
      "[ID 2] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-02-24 22:43:31      NaN        NaN\n",
      "2020-02-24 22:48:31      NaN        NaN\n",
      "2020-02-24 22:53:31      NaN        NaN\n",
      "2020-02-24 22:58:31      NaN        NaN\n",
      "2020-02-24 23:03:31      NaN        NaN\n",
      "...                      ...        ...\n",
      "2020-02-25 12:18:31      NaN        NaN\n",
      "2020-02-25 12:23:31      NaN        NaN\n",
      "2020-02-25 12:28:31      NaN        NaN\n",
      "2020-02-25 12:33:31      NaN        NaN\n",
      "2020-02-25 12:38:31      NaN        NaN\n",
      "\n",
      "[168 rows x 2 columns]\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 3] Now processing ...\n",
      "[ID 3] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-02-25 10:21:40      NaN        NaN\n",
      "2020-02-25 10:26:40      NaN        NaN\n",
      "2020-02-25 10:31:40      NaN        NaN\n",
      "2020-02-25 10:36:40      NaN        NaN\n",
      "2020-02-25 10:41:40      NaN        NaN\n",
      "2020-02-25 10:46:40      NaN        NaN\n",
      "2020-02-25 10:51:40      NaN        NaN\n",
      "2020-02-25 10:56:40      NaN        NaN\n",
      "2020-02-25 11:01:40      NaN        NaN\n",
      "2020-02-25 11:06:40      NaN        NaN\n",
      "2020-02-25 11:11:40      NaN        NaN\n",
      "2020-02-25 11:16:40      NaN        NaN\n",
      "2020-02-25 11:21:40      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 4] Now processing ...\n",
      "[ID 4] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-02-29 03:31:18      NaN        NaN\n",
      "2020-02-29 03:36:18      NaN        NaN\n",
      "2020-03-02 21:36:12      NaN        NaN\n",
      "2020-03-02 21:41:12      NaN        NaN\n",
      "2020-03-02 21:46:12      NaN        NaN\n",
      "...                      ...        ...\n",
      "2020-03-03 04:46:12      NaN        NaN\n",
      "2020-03-03 04:51:12      NaN        NaN\n",
      "2020-03-03 04:56:12      NaN        NaN\n",
      "2020-03-03 05:01:12      NaN        NaN\n",
      "2020-03-03 05:06:12      NaN        NaN\n",
      "\n",
      "[93 rows x 2 columns]\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 5] Now processing ...\n",
      "[ID 5] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-03-03 23:54:52      NaN        NaN\n",
      "2020-03-03 23:59:52      NaN        NaN\n",
      "2020-03-04 00:04:52      NaN        NaN\n",
      "2020-03-04 00:09:52      NaN        NaN\n",
      "2020-03-04 00:14:52      NaN        NaN\n",
      "2020-03-04 00:19:52      NaN        NaN\n",
      "2020-03-04 00:24:52      NaN        NaN\n",
      "2020-03-04 00:29:52      NaN        NaN\n",
      "2020-03-04 00:34:52      NaN        NaN\n",
      "2020-03-04 00:39:52      NaN        NaN\n",
      "2020-03-04 00:44:52      NaN        NaN\n",
      "2020-03-04 00:49:52      NaN        NaN\n",
      "2020-03-04 00:54:52      NaN        NaN\n",
      "2020-03-04 00:59:52      NaN        NaN\n",
      "2020-03-04 01:04:52      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 6] Now processing ...\n",
      "[ID 6] 보간 전 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 7] Now processing ...\n",
      "[ID 7] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-03-18 03:42:09      NaN        NaN\n",
      "2020-03-18 03:47:09      NaN        NaN\n",
      "2020-03-18 03:52:09      NaN        NaN\n",
      "2020-03-18 03:57:09      NaN        NaN\n",
      "2020-03-18 04:02:09      NaN        NaN\n",
      "...                      ...        ...\n",
      "2020-03-18 09:52:09      NaN        NaN\n",
      "2020-03-18 09:57:09      NaN        NaN\n",
      "2020-03-18 10:02:09      NaN        NaN\n",
      "2020-03-18 10:07:09      NaN        NaN\n",
      "2020-03-18 10:12:09      NaN        NaN\n",
      "\n",
      "[79 rows x 2 columns]\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 8] Now processing ...\n",
      "[ID 8] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-03-15 00:49:51      NaN        NaN\n",
      "2020-03-15 00:54:51      NaN        NaN\n",
      "2020-03-15 00:59:51      NaN        NaN\n",
      "2020-03-15 01:04:51      NaN        NaN\n",
      "2020-03-15 01:09:51      NaN        NaN\n",
      "...                      ...        ...\n",
      "2020-03-20 18:34:37      NaN        NaN\n",
      "2020-03-20 18:39:37      NaN        NaN\n",
      "2020-03-20 18:44:37      NaN        NaN\n",
      "2020-03-20 18:49:37      NaN        NaN\n",
      "2020-03-20 18:54:37      NaN        NaN\n",
      "\n",
      "[64 rows x 2 columns]\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 9] Now processing ...\n",
      "[ID 9] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-03-22 02:46:50      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 10] Now processing ...\n",
      "[ID 10] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-03-23 22:29:14      NaN        NaN\n",
      "2020-03-23 22:34:14      NaN        NaN\n",
      "2020-03-23 22:39:14      NaN        NaN\n",
      "2020-03-23 22:44:14      NaN        NaN\n",
      "2020-03-23 22:49:14      NaN        NaN\n",
      "...                      ...        ...\n",
      "2020-03-29 01:14:06      NaN        NaN\n",
      "2020-03-29 01:19:06      NaN        NaN\n",
      "2020-03-29 01:24:06      NaN        NaN\n",
      "2020-03-29 01:29:06      NaN        NaN\n",
      "2020-03-29 01:34:06      NaN        NaN\n",
      "\n",
      "[133 rows x 2 columns]\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 11] Now processing ...\n",
      "[ID 11] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-04-09 15:36:01      NaN        NaN\n",
      "2020-04-09 15:41:01      NaN        NaN\n",
      "2020-04-09 15:46:01      NaN        NaN\n",
      "2020-04-09 15:51:01      NaN        NaN\n",
      "2020-04-09 15:56:01      NaN        NaN\n",
      "2020-04-09 16:01:01      NaN        NaN\n",
      "2020-04-09 16:06:01      NaN        NaN\n",
      "2020-04-09 16:11:01      NaN        NaN\n",
      "2020-04-09 16:16:01      NaN        NaN\n",
      "2020-04-10 12:56:00      NaN        NaN\n",
      "2020-04-10 13:01:00      NaN        NaN\n",
      "2020-04-10 13:06:00      NaN        NaN\n",
      "2020-04-10 13:11:00      NaN        NaN\n",
      "2020-04-10 13:16:00      NaN        NaN\n",
      "2020-04-10 13:21:00      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 12] Now processing ...\n",
      "[ID 12] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-05-04 12:37:12      NaN        NaN\n",
      "2020-05-04 12:42:12      NaN        NaN\n",
      "2020-05-04 12:47:12      NaN        NaN\n",
      "2020-05-07 04:52:07      NaN        NaN\n",
      "2020-05-07 04:57:07      NaN        NaN\n",
      "2020-05-07 05:02:07      NaN        NaN\n",
      "2020-05-07 05:07:07      NaN        NaN\n",
      "2020-05-07 05:12:07      NaN        NaN\n",
      "2020-05-07 05:17:07      NaN        NaN\n",
      "2020-05-07 05:22:07      NaN        NaN\n",
      "2020-05-07 05:27:07      NaN        NaN\n",
      "2020-05-07 05:32:07      NaN        NaN\n",
      "2020-05-07 05:37:07      NaN        NaN\n",
      "2020-05-07 05:42:07      NaN        NaN\n",
      "2020-05-07 05:47:07      NaN        NaN\n",
      "2020-05-10 11:57:00      NaN        NaN\n",
      "2020-05-10 12:02:00      NaN        NaN\n",
      "2020-05-10 12:07:00      NaN        NaN\n",
      "2020-05-10 12:12:00      NaN        NaN\n",
      "2020-05-10 12:17:00      NaN        NaN\n",
      "2020-05-10 12:22:00      NaN        NaN\n",
      "2020-05-10 12:27:00      NaN        NaN\n",
      "2020-05-10 12:32:00      NaN        NaN\n",
      "2020-05-10 12:37:00      NaN        NaN\n",
      "2020-05-10 12:42:00      NaN        NaN\n",
      "2020-05-10 12:47:00      NaN        NaN\n",
      "2020-05-10 12:52:00      NaN        NaN\n",
      "2020-05-10 12:57:00      NaN        NaN\n",
      "2020-05-10 13:02:00      NaN        NaN\n",
      "2020-05-10 13:07:00      NaN        NaN\n",
      "2020-05-10 13:12:00      NaN        NaN\n",
      "2020-05-10 13:17:00      NaN        NaN\n",
      "2020-05-10 13:22:00      NaN        NaN\n",
      "2020-05-10 13:27:00      NaN        NaN\n",
      "2020-05-10 13:32:00      NaN        NaN\n",
      "2020-05-10 13:37:00      NaN        NaN\n",
      "2020-05-10 13:42:00      NaN        NaN\n",
      "2020-05-10 13:47:00      NaN        NaN\n",
      "2020-05-10 13:52:00      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 13] Now processing ...\n",
      "[ID 13] 보간 전 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 14] Now processing ...\n",
      "[ID 14] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "datetime                               \n",
      "2020-06-06 08:05:00      NaN   1.883333\n",
      "2020-06-07 05:20:00      NaN   1.866667\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 15] Now processing ...\n",
      "[ID 15] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-07-19 20:08:52      NaN        NaN\n",
      "2020-07-19 20:13:52      NaN        NaN\n",
      "2020-07-20 00:08:53      NaN        NaN\n",
      "2020-07-20 00:13:53      NaN        NaN\n",
      "2020-07-20 00:18:53      NaN        NaN\n",
      "...                      ...        ...\n",
      "2020-07-26 09:03:48      NaN        NaN\n",
      "2020-07-26 09:08:48      NaN        NaN\n",
      "2020-07-26 09:13:48      NaN        NaN\n",
      "2020-07-26 09:18:48      NaN        NaN\n",
      "2020-07-26 09:23:48      NaN        NaN\n",
      "\n",
      "[552 rows x 2 columns]\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n",
      "[ID: 16] Now processing ...\n",
      "[ID 16] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-07-18 10:28:20      NaN        NaN\n",
      "2020-07-18 10:33:20      NaN        NaN\n",
      "2020-07-18 10:38:20      NaN        NaN\n",
      "2020-07-18 10:43:20      NaN        NaN\n",
      "2020-07-18 10:48:20      NaN        NaN\n",
      "2020-07-20 23:03:16      NaN        NaN\n",
      "2020-07-20 23:08:16      NaN        NaN\n",
      "2020-07-20 23:13:16      NaN        NaN\n",
      "2020-07-20 23:18:16      NaN        NaN\n",
      "2020-07-20 23:23:16      NaN        NaN\n",
      "2020-07-20 23:28:16      NaN        NaN\n",
      "2020-07-20 23:33:16      NaN        NaN\n",
      "2020-07-20 23:38:16      NaN        NaN\n",
      "2020-07-20 23:43:16      NaN        NaN\n",
      "2020-07-20 23:48:16      NaN        NaN\n",
      "2020-07-20 23:53:16      NaN        NaN\n",
      "2020-07-20 23:58:16      NaN        NaN\n",
      "2020-07-21 00:03:16      NaN        NaN\n",
      "2020-07-21 00:08:16      NaN        NaN\n",
      "2020-07-21 00:13:16      NaN        NaN\n",
      "2020-07-21 00:18:16      NaN        NaN\n",
      "2020-07-21 00:23:16      NaN        NaN\n",
      "2020-07-21 00:28:16      NaN        NaN\n",
      "2020-07-21 00:33:16      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "for id in range(1, len(dg_df) 1):\n",
    "    #if id == 15: # Skip ID 15\n",
    "    #    continue\n",
    "    dexcom_df = pd.read_csv(f'./big-ideas-lab-glycemic-variability-and-wearable-device-data-1.1.2/{id:03}/Dexcom_{id:03}.csv')\n",
    "\n",
    "    print(f'[ID: {id}] Now processing ...')\n",
    "\n",
    "    ## Dexcom Glucose\n",
    "    dexcom_df.drop(dexcom_df.loc[:11].index, inplace=True)\n",
    "    dexcom_df.rename(columns = {'Timestamp (YYYY-MM-DDThh:mm:ss)': 'datetime'}, inplace = True)\n",
    "    dexcom_df.rename(columns = {'Glucose Value (mg/dL)': 'glucose'}, inplace = True)\n",
    "    dexcom_df = dexcom_df[['datetime', 'glucose']]\n",
    "    dexcom_df['datetime']=pd.to_datetime(dexcom_df['datetime'])\n",
    "    dexcom_df.set_index('datetime', inplace=True)\n",
    "\n",
    "    # 시간 간격 계산\n",
    "    dexcom_df['time_diff'] = dexcom_df.index.to_series().diff().dt.total_seconds() / 60.0\n",
    "\n",
    "    # 결측치에 대해 NaN을 삽입\n",
    "    # 4분에서 6분 사이가 아닌 시간 간격을 찾기\n",
    "    new_rows = []\n",
    "    for i in range(1, len(dexcom_df)):\n",
    "        if dexcom_df['time_diff'].iloc[i] > 6 or dexcom_df['time_diff'].iloc[i] < 4:\n",
    "            num_missing = int(dexcom_df['time_diff'].iloc[i] // 5)  # 5분 간격으로 결측값 삽입\n",
    "            for j in range(1, num_missing   1):\n",
    "                missing_time = dexcom_df.index[i-1]   pd.Timedelta(minutes=5*j)\n",
    "                new_rows.append(pd.DataFrame({'glucose': [np.nan]}, index=[missing_time]))\n",
    "\n",
    "    # 새로 생성된 행을 기존 데이터프레임에 추가\n",
    "    if new_rows:\n",
    "        new_rows_df = pd.concat(new_rows)\n",
    "        dexcom_df = pd.concat([dexcom_df, new_rows_df]).sort_index()\n",
    "\n",
    "    # 시간순으로 정렬\n",
    "    dexcom_df.sort_index(inplace=True)\n",
    "\n",
    "    # 보간 전 결측값 구간 출력\n",
    "    print(f\"[ID {id}] 보간 전 결측값이 있는 구간:\")\n",
    "    print(dexcom_df[dexcom_df['glucose'].isnull()])\n",
    "    dexcom_df[dexcom_df['glucose'].isnull()].to_csv(rf'./missing_csv/id_{id}_missing.csv', sep='\\t', encoding='utf-8', header='true')\n",
    "\n",
    "    # 다양한 보간법 사용 (선형 보간을 예시로 사용)\n",
    "    dexcom_df['glucose'] = dexcom_df['glucose'].interpolate(method='linear')\n",
    "\n",
    "    # 보간 후 결측값 구간 출력\n",
    "    print(\"\\n[ID {id}] 보간 후 결측값이 있는 구간:\")\n",
    "    print(dexcom_df[dexcom_df['glucose'].isnull()])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 특정 ID, 001에 대해 데이터 보간 후 CGM 시각화해보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ID: 1] Now processing ...\n",
      "[ID 1] 보간 전 결측값이 있는 구간:\n",
      "                     glucose  time_diff\n",
      "2020-02-15 09:08:30      NaN        NaN\n",
      "2020-02-15 09:13:30      NaN        NaN\n",
      "2020-02-19 23:43:26      NaN        NaN\n",
      "2020-02-19 23:48:26      NaN        NaN\n",
      "2020-02-19 23:53:26      NaN        NaN\n",
      "2020-02-19 23:58:26      NaN        NaN\n",
      "2020-02-20 00:03:26      NaN        NaN\n",
      "2020-02-20 00:08:26      NaN        NaN\n",
      "2020-02-20 00:13:26      NaN        NaN\n",
      "2020-02-20 00:18:26      NaN        NaN\n",
      "2020-02-20 00:23:26      NaN        NaN\n",
      "2020-02-20 00:28:26      NaN        NaN\n",
      "2020-02-20 00:33:26      NaN        NaN\n",
      "2020-02-20 00:38:26      NaN        NaN\n",
      "2020-02-20 00:43:26      NaN        NaN\n",
      "2020-02-20 00:48:26      NaN        NaN\n",
      "2020-02-20 00:53:26      NaN        NaN\n",
      "2020-02-20 00:58:26      NaN        NaN\n",
      "2020-02-20 01:03:26      NaN        NaN\n",
      "2020-02-20 01:08:26      NaN        NaN\n",
      "2020-02-20 01:13:26      NaN        NaN\n",
      "2020-02-20 01:18:26      NaN        NaN\n",
      "2020-02-20 01:23:26      NaN        NaN\n",
      "2020-02-20 01:28:26      NaN        NaN\n",
      "2020-02-20 01:33:26      NaN        NaN\n",
      "2020-02-20 01:38:26      NaN        NaN\n",
      "2020-02-20 01:43:26      NaN        NaN\n",
      "2020-02-20 01:48:26      NaN        NaN\n",
      "2020-02-20 01:53:26      NaN        NaN\n",
      "2020-02-20 01:58:26      NaN        NaN\n",
      "2020-02-20 02:03:26      NaN        NaN\n",
      "2020-02-20 02:08:26      NaN        NaN\n",
      "2020-02-20 02:13:26      NaN        NaN\n",
      "2020-02-20 02:18:26      NaN        NaN\n",
      "2020-02-20 02:23:26      NaN        NaN\n",
      "2020-02-20 02:28:26      NaN        NaN\n",
      "2020-02-20 02:33:26      NaN        NaN\n",
      "2020-02-20 02:38:26      NaN        NaN\n",
      "2020-02-20 02:43:26      NaN        NaN\n",
      "\n",
      "[ID {id}] 보간 후 결측값이 있는 구간:\n",
      "Empty DataFrame\n",
      "Columns: [glucose, time_diff]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "id = 1\n",
    "\n",
    "dexcom_df = pd.read_csv(f'./big-ideas-lab-glycemic-variability-and-wearable-device-data-1.1.2/{id:03}/Dexcom_{id:03}.csv')\n",
    "\n",
    "print(f'[ID: {id}] Now processing ...')\n",
    "\n",
    "## Dexcom Glucose\n",
    "dexcom_df.drop(dexcom_df.loc[:11].index, inplace=True)\n",
    "dexcom_df.rename(columns = {'Timestamp (YYYY-MM-DDThh:mm:ss)': 'datetime'}, inplace = True)\n",
    "dexcom_df.rename(columns = {'Glucose Value (mg/dL)': 'glucose'}, inplace = True)\n",
    "dexcom_df = dexcom_df[['datetime', 'glucose']]\n",
    "dexcom_df['datetime']=pd.to_datetime(dexcom_df['datetime'])\n",
    "dexcom_df.set_index('datetime', inplace=True)\n",
    "\n",
    "# 시간 간격 계산\n",
    "dexcom_df['time_diff'] = dexcom_df.index.to_series().diff().dt.total_seconds() / 60.0\n",
    "\n",
    "# 결측치에 대해 NaN을 삽입\n",
    "# 4분에서 6분 사이가 아닌 시간 간격을 찾기\n",
    "new_rows = []\n",
    "for i in range(1, len(dexcom_df)):\n",
    "    if dexcom_df['time_diff'].iloc[i] > 6 or dexcom_df['time_diff'].iloc[i] < 4:\n",
    "        num_missing = int(dexcom_df['time_diff'].iloc[i] // 5)  # 5분 간격으로 결측값 삽입\n",
    "        for j in range(1, num_missing   1):\n",
    "            missing_time = dexcom_df.index[i-1]   pd.Timedelta(minutes=5*j)\n",
    "            new_rows.append(pd.DataFrame({'glucose': [np.nan]}, index=[missing_time]))\n",
    "\n",
    "# 새로 생성된 행을 기존 데이터프레임에 추가\n",
    "if new_rows:\n",
    "    new_rows_df = pd.concat(new_rows)\n",
    "    dexcom_df = pd.concat([dexcom_df, new_rows_df]).sort_index()\n",
    "\n",
    "# 시간순으로 정렬\n",
    "dexcom_df.sort_index(inplace=True)\n",
    "\n",
    "# 보간 전 결측값 구간 출력\n",
    "print(f\"[ID {id}] 보간 전 결측값이 있는 구간:\")\n",
    "print(dexcom_df[dexcom_df['glucose'].isnull()])\n",
    "\n",
    "# 다양한 보간법 사용 (선형 보간을 예시로 사용)\n",
    "dexcom_df['glucose'] = dexcom_df['glucose'].interpolate(method='linear')\n",
    "\n",
    "# 보간 후 결측값 구간 출력\n",
    "print(\"\\n[ID {id}] 보간 후 결측값이 있는 구간:\")\n",
    "print(dexcom_df[dexcom_df['glucose'].isnull()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "fl_df = pd.read_csv('./big-ideas-lab-glycemic-variability-and-wearable-device-data-1.1.2/001/Food_Log_001.csv')\n",
    "\n",
    "fl_df['date']=pd.to_datetime(fl_df['date'])\n",
    "fl_df['time_begin']=pd.to_datetime(fl_df['time_begin'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "glucose      119.0\n",
      "time_diff      5.0\n",
      "Name: 2020-02-19 22:43:26, dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1IAAAE CAYAAACHjyEPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAChMElEQVR4nOzdd3gUVffA8e/upncSUiEQCCX0FkCQpnRUQEVBAUFBsNef7VUBe3ktKApSVFBBwPIiIgIKKCA1NCGhpxNCeu 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 ejI2NDVOnTqVHjx6o1WpOnz7Nzz//zJIlS4iKiqJ169b13dQ607t3b0JCQvj   958803K9xmzZo1AEybNq0um1anpk fzpQpU7C3tzcuW7x4Mc2bN2fmzJn117BS fn52NjU/dfptm3brH6MhIQEXnvtNYKCgujZs6fVj2dpERERvPbaawwbNkwyPUIIi5NASghR7y5cuMCUKVNo3bo127dvx9/fv8z69957j8WLF6NWX39J9KlTp/Lqq6 yf/9 brjhhnLrv// e0JCQujdu3c9tK5uaDQaNBpNfTejUg4ODvVyXDs7u3o5roDc3FycnZ3ruxlCiHp2/V2VCCEanPfff5/c3Fy /vrrckEUgI2NDU888QSBgYGV7iM6OhqVSlVhd6 KxrBcvHiRWbNmERAQgL29PW3atOHhhx mqKjIuE1kZCR33XUXnp6eODk5ccMNN/Dbb7 V2/ iRYvo0qULTk5ONGvWjNDQUGOm6OrjPfDAA/j6 mJvb0 XLl346quvqjkz kAKKLc/gMOHD3PmzBnjNr/88gu33HKL8TkFBwfzxhtvoNVqqzzGX3/9hUql4q //iqzvLJzevr0aSZNmoSnpycODg6EhoaycePGap9L7969ueOOO8os69atGyqVqkzXzXXr1qFSqTh16hRQfnxQUFAQ4eHh/P3336hUKlQqVbmxQoWFhTzzzDN4e3vj7OzM7bffTnJycrVtrImKXl9//fUXoaGhODg4EBwczNKlS6scE7Rhwwa6du1qfG1s2bKl2uNWNEYqJiaG8ePH4 zsjI PD08//TRbt26t8O8L ozNTTfdhJOTEy1atOD9998v8xz69u0LwP33328811e/Hg4cOMCYMWNwd3fHycmJoUOH8s8//5Q7zp49e jbt2 Z81Ebhi6We/bsoV /fjg4ONC2bVu  eYb4zYrV67krrvuAuCmm24ytv/q8/D7778zePBgnJ2dcXV15ZZbbiE8PLzMsQxdCy9cuMC4ceNwdXVl6tSpPPbYY7i4uJCXl1eufffccw9 fn7G915N35tCiIZNMlJCiHq3adMm2rVrR// /evkeAkJCfTr14 MjAzmzJlDSEgIFy9e5McffyQvLw87OzsuX77MwIEDycvL44knnsDLy4tVq1Yxfvx4fvzxR26//XYAli9fzhNPPMGkSZN48sknKSgo4N9// XAgQPce  9AFy fJkbbrjBWFjA29ub33//nVmzZpGVlcVTTz1VaVvbtGnDwIEDWb9 PR9//HGZzIwhuDIcZ XKlbi4uPDMM8/g4uLCjh07mDdvHllZWfz3v/ 1yLkLDw/nxhtvpEWLFrz44os4Ozuzfv16Jk6cyE8//WQ8LxUZPHgw33//vfH3tLQ0wsPDUavV7N69m 7duwOwe/duvL296dSpU4X7WbhwIY8//jguLi68/PLLAPj6 pbZ5vHHH6dZs2bMnz f6OhoFi5cyGOPPca6detMep5paWkVLtfpdNU 9ujRo4wZMwZ/f39ee 01tFotr7/ Ot7e3hVuv2fPHn7  WceeeQRXF1d fTTT7nzzjuJjY3Fy8vLpPaCPkty8803c nSJZ588kn8/PxYs2YNO3furHD79PR0xowZwx133MHdd9/Njz/ yAsvvEC3bt0YO3YsnTp14vXXX2fevHnMmTOHwYMHAzBw4EAAduzYwdixY nTpw/z589HrVbz9ddfc/PNN7N792769esHwIkTJxg1ahTe3t4sWLCAkpIS5s fX 5vZq7z588zadIkZs2axYwZM/jqq6 YOXMmffr0oUuXLgwZMoQnnniCTz/9lP/85z/G15Ph32 //ZYZM2YwevRo3nvvPfLy8liyZAmDBg3i6NGjZboClpSUMHr0aAYNGsQHH3yAk5MTQUFBfP755/z222/GgA0gLy PX3/9lZkzZxrfr3Xx3hRC1ANFCCHqUWZmpgIoEydOLLcuPT1dSU5ONv7k5eUZ182fP1 5 iMsKipKAZSvv/663H4AZf78 cbf77vvPkWtViuHDh0qt61Op1MURVGeeuopBVB2795tXJedna20adNGCQoKUrRaraIoijJhwgSlS5cuVT7HWbNmKf7 /kpKSkqZ5VOmTFHc3d3LPK KfP755wqgbN261bhMq9UqLVq0UAYMGGBcVtF 5s6dqzg5OSkFBQXGZTNmzFBat25t/H3nzp0KoOzcubPMYys6p8OHD1e6detWZn86nU4ZOHCg0r59 yqfxw8//KAASkREhKIoirJx40bF3t5eGT9 vDJ58mTjdt27d1duv/124 9ff/21AihRUVHGZV26dFGGDh1a7hiGbUeMGGH8WyqKojz99NOKRqNRMjIyqmyj4XVV1c8tt9xS5jHXvr5uu 02xcnJSbl48aJx2blz5xQbGxvl2q9dQLGzs1POnz9vXHb8 HEFUBYtWlTlORg6dGiZc/Dhhx8qgLJhwwbjsvz8fCUkJKTc33fo0KEKoHzzzTfGZYWFhYqfn59y5513GpcdOnSowveVTqdT2rdvr4wePbrMec7Ly1PatGmjjBw50rhs4sSJioODgxITE2NcFhERoWg0mnLnoyIzZsxQnJ2dyyxr3bq1Aii7du0yLktKSlLs7e2VZ5991rjM8Jq79rWdnZ2teHh4KA8  GCZ5YmJiYq7u3uZ5TNmzFAA5cUXXyx3Dlq0aFHmfCmKoqxfv75c22r63hRCNGzStU8IUa ysrIAKqzKNWzYMLy9vY0/n3/ ea2Pp9Pp2LBhA7fddhuhoaHl1hu6Xm3evJl /foxaNAg4zoXFxfmzJlDdHQ0ERERAHh4eBAfH8 hQ4cqPJ6iKPz000/cdtttKIpCSkqK8Wf06NFkZmZy5MiRKts8efJkbG1ty3Tv /vvv7l48aKxWx Ao6Oj8f/Z2dmkpKQwePBg8vLyOH36tAlnp2ppaWns2LGDu   27j/lJQUUlNTGT16NOfOnePixYuVPt6Q0di1axegzzz17duXkSNHsnv3bgAyMjI4efKkcduamjNnTpludIMHD0ar1RITE2PS43/66Sf  OOPcj/VZVG0Wi1//vknEydOJCAgwLi8Xbt2jB07tsLHjBgxguDgYOPv3bt3x83NjcjISJPaarBlyxZatGjB PHjjcscHBx48MEHK9zexcWlTJESOzs7 vXrZ9Jxjx07xrlz57j33ntJTU01vhZyc3MZPnw4u3btQqfTodVq2bp1KxMnTqRVq1bGx3fq1InRo0eb9fyu1blz5zKvE29vbzp27GhS //44w8yMjK45557yrwnNRoN/fv3rzCL9/DDD5f5XaVScdddd7F582ZycnKMy9etW0eLFi3KfHZY 70phKgf0rVPCFGvXF1dAcpciBgsXbqU7OxsLl  bLGqdMnJyWRlZVU7h05MTEyFXQ0N3YJiYmLo2rUrL7zwAn/  Sf9 vWjXbt2jBo1invvvZcbb7zReLyMjAyWLVvGsmXLKjxWUlJSlW3x8vJi9OjR/O9// OLL77AwcGBNWvWYGNjw913323cLjw8nFdeeYUdO3YYA1SDzMzMKo9hivPnz6MoCq   iqvvvpqpc lRYsWFa7z9fWlffv27N69m7lz57J7925uuukmhgwZwuOPP05kZCSnTp1Cp9PVOpC6 qIdoFmzZoC O5sphgwZQvPmzcstr66wRFJSEvn5 bRr167cuoqWVdRW0LfX1LYaxMTEEBwcXG4cVmXHbdmyZbltmzVrZtJUA fOnQNgxowZlW6TmZlJYWEh fn5tG/fvtz6jh07snnz5mqPVZnanDdD 2   eYK17u5uZX53cbGhpYtW5bbbvLkySxcuJCNGzdy7733kpOTw bNm5k7d26Zc2vt96YQon5IICWEqFfu7u74 /tz8uTJcusMgYwpc8BUNojf2oO5O3XqxJkzZ9i0aRNbtmzhp59 YvHixcybN4/XXnvNOKZm2rRplV50GsYGVWXatGls2rSJTZs2MX78eH766SfjuBPQZ3KGDh2Km5sbr7/ OsHBwTg4OHDkyBFeeOGFKsf2mHruDPv4v//7v0qzCZVdtBsMGjSI7du3k5 fz HDh5k3bx5du3bFw8OD3bt3c rUKVxcXOjVq1eV 6lOZVX FEWp1X6tob7aWpvjGl4L//3vfysti 7i4kJhYWGN21cdS7T/22 /xc/Pr9z6a8vZ29vbV1g19IYbbiAoKIj169dz77338uuvv5Kfn8/kyZON29TmvSmEaNgkkBJC1LtbbrmFFStWcPDgQeMAdXMZMg4ZGRllll/blcvb2xs3N7cKA7ertW7dmjNnzpRbbuiGc/V8Vs7OzkyePJnJkydTVFTEHXfcwVtvvcVLL72Et7c3rq6uaLVaRowYUZOnBsD48eNxdXVlzZo12Nrakp6eXqZb319//UVqaio///wzQ4YMMS6Pioqqdt mnru2bdsCYGtrW PnMnjwYL7  mvWrl2LVqtl4MCBqNVqBg0aZAykBg4cWG2588qCv/rm4 ODg4MD58 fL7euomWW1Lp1ayIiIlAUpcz5qc1xKzvPhq6Ibm5uVb4WvL29cXR0NGaArlbR 8vSqmu/j49Prd6XAHfffTeffPIJWVlZrFu3jqCgoDJTFdTmvSmEaNhkjJQQot49//zzODk58cADD3D58uVy6025w zm5kbz5s2N428MFi9eXOZ3tVrNxIkT fXXXwkLC6v0WOPGjePgwYPs27fPuC43N5dly5YRFBRE586dAUhNTS3zeDs7Ozp37oyiKBQXF6PRaLjzzjv56aefKgzeTC3J7ejoyO23387mzZtZsmQJzs7OTJgwwbjeEHhcfa6KiorKPf KtG7dGo1GU 258/HxYdiwYSxdupRLly7V6LkYuuy99957dO/eHXd3d Py7du3ExYWZlK3Pmdn53KBX0Og0WgYMWIEGzZsICEhwbj8/Pnz/P7771Y99ujRo7l48WKZUvQFBQUsX768xvs0zJV07bnu06cPwcHBfPDBBxV2yzW8FjQaDaNHj2bDhg3ExsYa1586dYqtW7fWuF2mqqz9o0ePxs3Njbfffpvi4uJyjzOnVP7kyZMpLCxk1apVbNmypUx3W6jde1MI0bBJRkoIUe/at2/PmjVruOeee jYsSNTp06lR48eKIpCVFQUa9asQa1WVzhG4WqzZ8/m3XffZfbs2YSGhrJr1y7Onj1bbru3336bbdu2MXToUObMmUOnTp24dOkSP/zwA3v27MHDw4MXX3yR77//nrFjx/LEE0/g6enJqlWriIqK4qeffjJ28xk1ahR fn7ceOON Pr6curUKT777DNuueUW4/ivd999l507d9K/f38efPBBOnfuTFpaGkeOHOHPP/ stNT2taZNm8Y333zD1q1bmTp1apkJQQcOHEizZs2YMWMGTzzxBCqVim //dakINTd3Z277rqLRYsWoVKpCA4OZtOmTRWO3fr8888ZNGgQ3bp148EHH6Rt27ZcvnyZffv2ER8fz/Hjx6s8Vrt27fDz8 PMmTM8/vjjxuVDhgzhhRdeADApkOrTpw9LlizhzTffpF27dvj4 FQ63qWuLViwgG3btnHjjTfy8MMPo9Vq eyzz jatSvHjh2z2nHnzp3LZ599xj333MOTTz6Jv78/q1evNo7rqkkWLzg4GA8PD7744gtcXV1xdnamf// tGnThhUrVjB27Fi6dOnC/fffT4sWLbh48SI7d 7Ezc2NX3/9FYDXXnuNLVu2MHjwYB555BFKSkqMc6 ZMh6rNnr27IlGo G9994jMzMTe3t7br75Znx8fFiyZAnTp0 nd /eTJkyBW9vb2JjY/ntt9 48cYb eyzz0w6Ru/evWnXrh0vv/wyhYWFZbr1Qe3em0KIBq7uCwUKIUTFzp8/rzz88MNKu3btFAcHB8XR0VEJCQlRHnroIeXYsWNltr22/Lmi6EsMz5o1S3F3d1dcXV2Vu   W0lKSipXnlpRFCUmJka57777FG9vb8Xe3l5p27at8uijjyqFhYXGbS5cuKBMmjRJ8fDwUBwcHJR /fopmzZtKrOfpUuXKkOGDFG8vLwUe3t7JTg4WHnuueeUzMzMMttdvnxZefTRR5XAwEDF1tZW8fPzU4YPH64sW7bM5PNTUlKi Pv7K4CyefPmcuv/ ecf5YYbblAcHR2VgIAA5fnnn1e2bt1arvxzRSWWk5OTlTvvvFNxcnJSmjVrpsydO1c5efJkhaWvL1y4oNx3332Kn5 fYmtrq7Ro0UK59dZblR9//NGk53HXXXcpgLJu3TrjsqKiIsXJyUmxs7NT8vPzy2xfUenvxMRE5ZZbblFcXV0VwFgG3LDttaXtKyvxfi3D6yo5ObnC9a1bt662/LmiKMr27duVXr16KXZ2dkpwcLCyYsUK5dlnn1UcHBzKPfbRRx t8DgzZsyo8hxcW/5cURQlMjJSueWWWxRHR0fF29tbefbZZ5WffvpJAZT9 /eXeWxFZfsrem388ssvSufOnY3l269 PRw9elS54447jK//1q1bK3fffbeyffv2Mvv4   /lT59 ih2dnZK27ZtlS   KLC93BFKit/fu3fobJzsnz5cqVt27bGcutXvwZ27typjB49WnF3d1ccHByU4OBgZebMmUpYWFiVx7/Wyy /rABKu3btKlxfm/emEKLhUimK3BIRQgghrG3ixImEh4dXOF7ImhYuXMjTTz9NfHx8pRUVhRBCmE/GSAkhhBAWlp fX b3c fOsXnzZoYNG1anxy0oKGDp0qW0b99egighhLAwGSMlhBBCWFjbtm2ZOXMmbdu2JSYmhiVLlmBnZ8fzzz9v1ePecccdtGrVip49e5KZmcl3333H6dOnWb16tVWPK4QQ1yMJpIQQQggLGzNmDN9//z2JiYnY29szYMAA3n777QonprWk0aNHs2LFClavXo1Wq6Vz586sXbu2XAEEIYQQtSdjpIQQQgghhBDCTDJGSgghhBBCCCHMJIGUEEIIIYQQQphJxkgBOp2OhIQEXF1dazRhoRBCCCGEEKJpUBSF7OxsAgICUKsrzztJIAUkJCQQGBhY380QQgghhBBCNBBxcXG0bNmy0vUSSAGurq6A/mS5ubnVc2uEECIOmA84lf4A5JX vAbIjR9Rkdq8bjKBHwBHwB4oBPKBuwB3K7VXCFNc 9qEhvv6rK6tIO zxiErK4vAwEBjjFAZCaTA2J3Pzc1NAikhRAPgCtgCHlz5cs0EikvXyeeUqEhtXjcK ou7ZoAzkFu63K2axwlhbde NqHhvj6rayvI 6xxqW7IjxSbEEIIIYQQQggzSSAlhBBCCCGEEGaSQEoIIYQQQgghzCRjpEyk1WopLi6u72aIUhqNBhsbGylXL4QQQggh6oUEUibIyckhPj4eRVHquyniKk5OTvj7 2NnZ1ffTRFCCCGEqJH8Ii1apQQXe7ksb2zkL1YNrVZLfHw8Tk5OeHt7SwakAVAUhaKiIpKTk4mKiqJ9 /ZVTpYmhBBCCNEQFZbALYuOkV2gY tTQ/B0lpvDjYkEUtUoLi5GURS8vb1xdHSs7 aIUo6Ojtja2hITE0NRUREODg713SQhhBBCCLP8fMSZyOR8AFbtjebpkR3quUXCHBJImUgyUQ2PZKGEEEII0VhpdbD07yvzR63aF82cIW1xrsMufknZBSz7O5K8Yi0Ag9o1Z1w3/zo7fmMngZQQQgghhBB17PeTNkSn2uLhZIO7ox0xqXl8fzCW2YPb1lkbPttxnm/2xRh//yEsjiEdvGW8lonklr4QQgghhBB1SFFgyV/68VAzBvjz0NBgAFbsjqKoRFcnbdDpFLaFXwbgnn6taO5iT7FW4VhsRp0c/2rpuUXkFpbU XFrSwKp61hQUBALFy6s72YIIYQQQlxX9l2A8AQNjrY6Zg4M4I7eLfBxtScxq4CNxxPqpA0nLmaSmFWAs52GBeM7M6idFwCHotPq5PhXW747ku6vbePjP87W bFrQwIpIYQQQggh6tDh0t50Y7rm08zZFnsbDfcNaA3Ar3UUSG0NTwRgWIgP9jYaQoM8gfoJpMKi09HqFFp4NK7CbhJICSGEEEIIUYdi0/RFzNo2LzYuG9NVX Rh74UUsguKK3ycJW2L0HfrG9XZF4B bfSB1NHYDIq1ddO9EKCwRMux AwAQoOa1dlxLUECKTMpikJeUUm9/Jg7IXB2djZTp07F2dkZf39/Pv74Y4YNG8ZTTz1Vbtvo6GhUKhXHjh0zLsvIyEClUvHXX38Zl4WHh3Prrbfi5uaGq6srgwcP5sKFCwDodDpef/11WrZsib29PT179mTLli3GxxYVFfHYY4/h7  Pg4MDrVu35p133ilzvNmzZ Pt7Y2bmxs333wzx48fN s5CyGEEEI0dDGlSZ9WXlfGBbXzcaGttzPFWoW/ziRb9fgXknM4n5SDrUbFTSE  uN7u DuaEt sZaIhCyrHv9qJ IzKSrR0dzFjjbNnevsuJYgJTnMlF spfO8rfVy7IjXR NkZ/qf7JlnnuGff/5h48aN Pr6Mm/ePI4cOULPnj1rdPyLFy8yZMgQhg0bxo4dO3Bzc Off/6hpET/IfDJJ5/w4YcfsnTpUnr16sVXX33F PHjCQ8Pp3379nz66ads3LiR9evX06pVK Li4oiLizPu/6677sLR0ZHff/8dd3d3li5dyvDhwzl79iyenp41arMQQgghREMTZwikPMsWWBjV2Y8v/r7AtojL3NYjwGrH/6M0G3VDWy/cHGwBUKtVhLZuxvbTSRyKTqNHoIfVjn 1Q9HpAIS29mx00w1JINVEZWdns2rVKtasWcPw4cMB PrrrwkIqPmb8vPPP8fd3Z21a9dia6t/03XocGXiuA8  IAXXniBKVOmAPDee  xc dOFi5cyOeff05sbCzt27dn0KBBqFQqWrdubXzsnj17OHjwIElJSdjb2xv3t2HDBn788UfmzJlT43YLIYQQQjQUBcWQWJrwuTaQGt3Fly/ vsDO00kUlmixt9FYpQ2G8VGjuviVWR4a5Mn200mERacze7BVDl2OYUxW3zaN76a5BFJmcrTVEPH66Ho7tqkiIyMpLi6mX79 xmXu7u507Nixxsc/duwYgwcPNgZRV8vKyiIhIYEbb7yxzPIbb7zR2D1v5syZjBw5ko4dOzJmzBhuvfVWRo0aBcDx48fJycnBy8urzOPz8/ONXQeFEEIIIRq7 HQ1iqLC2U7B07nsWKQeLT3wcbUnKbuQ/ZFpDO3gbfHjZxUUczwuA4ARnXzKrOtbOkYpLCYNRVGsniHS6RTCDIFUIxsfBRJImU2lUpnVva6xUKv1w WuHodVXFx2oKOjY 0qqfTu3ZuoqCh //13/vzzT 6  25GjBjBjz/ SE5ODv7 /mXGYxl4eHjU6rhCXE KSnScupRFtxbuqNWNq4uEEEJcD JKC0208tJxbZyiVqsY2dmX1Qdi2RaeaJVA6mhsBjoFWnk64e9e9tquW0t37GzUpOQUEZ2aZ/UxS2eTsskqKMHJTkNnfzerHssapNhEE9W2bVtsbW05dOiQcVlmZiZnz1Zcn9/bW/9GvXTpknHZ1YUnALp3787u3bvLBVgAbm5uBAQE8M8//5RZ/s8//9C5c cy202ePJnly5ezbt06fvrpJ9LS0ujduzeJiYnY2NjQrl27Mj/Nmzc3 /kLcb1atOMcEz7/h6fWHUOrM69AjRBCCOuLSdVffrfyrLgynqG73R8Rl9FZ4XPckAGqqEKevY2GHi3dy2xnTYbxUb1bNcNG0/jCksbXYmESV1dXZsyYwXPPPcfOnTsJDw9n1qxZqNXqCtO0jo6O3HDDDbz77rucOnWKv// m1deeaXMNo899hhZWVlMmTKFsLAwzp07x7fffsuZM2cAeO6553jvvfdYt24dZ86c4cUXX TYsWM8 eSTAHz00Ud8//33nD59mrNnz/LDDz/g5 eHh4cHI0aMYMCAAUycOJFt27YRHR3N3r17efnllwkLC7P CROiCVAUhQ3HLgKw8XgCr/5y0uxqn0IIIawrNs0QSFX8 TygrReu9jYkZRcay4Jb0sEoQ1e6isckdQnQB1LnknIsfuxrVRXUNQYSSDVhH330EQMGDODWW29lxIgR3HjjjXTq1AkHB4cKt//qq68oKSmhT58 PPXUU7z55ptl1nt5ebFjxw5ycnIYOnQoffr0Yfny5cYxU0888QTPPPMMzz77LN26dWPLli1s3LiR9u3bA/rg7v333yc0NJS ffsSHR3N5s2bjcHd5s2bGTJkCPfffz8dOnRgypQpxMTE4Ovra90TJUQTcToxm7i0fGw1KlQqWHMgttHNEi EEE1d7FVd ypiZ6NmWGlJ8m3hly167KISHcdKx0dVNiYp2McFgAt1EEgdKg3q lUS1DV0TW wjzBydXVl9erVxt9zc3N57bXXjBXwoqOjy2zfqVMn9u7dW2bZtXezu3fvztatFZd/V6vVzJ8/n/nz51e4/sEHH TBBx ssr2ffvopn376aaXbCCEqZ/jCHdbRh EhPrz48wmW/H2BB4e0xdWhfJEYIYQQde9KRqrySW9Hdfbl1 MJbItI5MWxIRY79smETApLdDRzsiXY26XCbYK99eOiLiRbN5BKyiogIbMAtQp6tvKw6rGsRTJSTdjRo0f5/vvvuXDhAkeOHGHq1KkATJgwoZ5bJoSwBmM5286 TOnXqs4mdhRCCGEaRTEtkBrW0RtbjYrI5FzOWzAzdKUrXeVzNrUrDbBi0/IoLNFa7NjXCr krwEf7O3SaAu51SiQio2NZffu3WzdupUjR45QWFhYo4Pv2rWL2267jYCAAFQqFRs2bKh024ceegiVSsXChQvLLE9LS2Pq1Km4ubnh4eHBrFmzyMmxfiqysfjggw/o0aMHI0aMIDc3l927d0vxBiGaoLi0PCIuZaFWwfBO u6wo0sHLG LsGzXECGEEDWTnK2moFiFWqXQwqPyMayuDrYMDNZfr22LSLTY8Q3FHaoqNe7tao rvQ06BWJS8yx27GtFJOgDqc4Bja9an4HJgVR0dDQvvPACrVu3pk2bNgwdOpSxY8cSGhqKu7s7I0eO5IcffkCnqzy6vlZubi49evTg888/r3K7//3vf zfv7/CyWSnTp1KeHg4f/zxB5s2bWLXrl0yeWupXr16cfjwYXJyckhLS OPP/6gW7du9d0sIYQVGGap7xvkiaezHaDPTAHGiR2FEELUr9g0febF3x2qS8KM6qL/DLfUOKmr52wKrWJMkkqlom0djJOKKM1INcay5wYmBVJPPPEEPXr0ICoqijfffJOIiAgyMzMpKioiMTGRzZs3M2jQIObNm0f37t3LlNyuytixY3nzzTe5/fbbK93m4sWLPP7446xevbrcRLCnTp1iy5YtrFixgv79 zNo0CAWLVrE2rVrSUhIqHSfhYWFZGVllfkRQojGzHDH8upZ6g0TO YUlrA/0vplbIUQQlTNEEi1MqG2wsjS3gXH4jK4nFVQ62NHpuSQnleMg62arqWV SpTF OkrpuMlLOzM5GRkaxfv57p06fTsWNHXF1dsbGxwcfHh5tvvpn58 dz6tQpPvjgA Li4izSOJ1Ox/Tp03nuuefo0qVLufX79u3Dw8OD0NBQ47IRI0agVqs5cOBApft95513cHd3N/4EBgZW2xYpIdzwyN9ECL3CEq2xu8bVs9QbJnaEK OnhBBC1J YVH1SoLVX9dv6uDnQM9ADgL/OJNX62IbviZ6BHtjZVB0CGApRXEjOrfVxK5JTWEJ0qn7fnZp6Ruqdd97By8uEvzgwZswY7rjjjlo1yuC9997DxsaGJ554osL1iYmJ Pj4lFlmY2ODp6cniYmVXzS89NJLZGZmGn qCvw0Gg0ARUVFNXgGwpry8vT9dq/NVApxvYlPz0erU3Cy09DK06nMOmtP7CiEEMJ0caUZqcBK5pC61sBg/fW3IQiqjUPRVc8fdbUrgZR1MlJnErNQFPB1s6e5i71VjlEXLFYi499//yU0NNRiAcfhw4f55JNPOHLkSKVVRWrK3t4ee3vT/mg2NjY4OTmRnJyMra0tarUUOqxviqKQl5dHUlISHh4exmBXiOtVbJr pkIrT6dyn5eGiR2TSyd27N2qcU56KIQQTUFsmv6axZSMFEDfNp7w1wXj2KbaOGTC CiDdj6lXfuSclAUxeLX4sZufY04GwUWDKQURUGrtdxg5t27d5OUlESrVq2My7RaLc8  ywLFy4kOjoaPz8/kpLKpjpLSkpIS0vDz8/v2l3WiEqlwt/fn6ioKGJiYiyyT2EZHh4eFvs7C9GYxZUGUoHXZKNAP7Fj/7Ze/HnqMv/GSSAlhBD1KTHrSrEJU/Ru1QyVCqJT80jKLsDH1aFmx80sIC4tH7UKepswZ1MrT2c0ahW5RVouZxXi516z41bGWGiiEY PggY8Ie/06dMZMWJEmWWjR49m vTp3H///QAMGDCAjIwMDh8 TJ8 fQDYsWMHOp2O/v37W6wtdnZ2tG/fXrr3NSC2traSiRKilKE8besKAimA9r4u/HnqstX6ugshhDBNeq6 Z5Ons2nbuzvaEuLnxqlLWYRFpzOum3 NjhsWo89GdfJ3M2mCdjsbNa09nYhMyeVCco7lAyljRsrEiLKBqtdAKicnh/Pnzxt/j4qK4tixY3h6etKqVaty47JsbW3x8/OjY8eOAHTq1IkxY8bw4IMP8sUXX1BcXMxjjz3GlClTKiyVXhtqtRoHB8u iIQQwhKMXfu8Kg6krN3XXQghRPUKinXkFpUGUhV/XFeob1AzTl3K4lB0Ws0DKeP8USaUCyzV1tvFGEjd2M5yc5CWaHWcTswGGn9GyuQBP9eWC7/2Jzs72 yDh4WF0atXL3r16gXAM888Q69evZg3b57J 1i9ejUhISEMHz6ccePGMWjQIJYtW2Z2W4QQorGKTb0yRqoidVHGVgghRNUy8ooB0KgVzOmhZxjTFFaLghNXxkeZ3r07 KpxUpYUlZJLYYkOJztNpT0pGguTM1IeHh5VDjSryUC0YcOGmVXCOjo6utwyT09P1qxZY9ZxhRCiqVAUpUyxiYoEl06seDmrkOyCYpO6dQghhLCstFx9INXMScGc2mV9S4Of8IRMcgpLcLE3r0NZdkExp0rHJJmTkbJWCXTD KhO/m6o1ZYtYlHXTP5L7Ny505rtEEIIUQMpOUXkF2tRqaBFM8cKt3FzsMXH1Z6k7EIik3PpUToviRBCiLqTnlcCgKezApgeQPi7O9KymSPx6fkcjU1ncHtvs457JDYDnaK/2ebrZnoqzFrdwptKxT4wI5AaOnSoNdshhBCiBmLT9HcKA9wdsbepvABLsLcLSdmFnE/KkUBKCCHqwdUZKXMCKYB QZ7Ep1/kULT5gVRYDbr1wZVu4ZcyC2qUCatMU6nYByaOkapufNTVP0IIIepOrLH0ecXZKANjX3cZJyWEEPUivXSMlD4jZR7DOKldZ5PNfuyucymAed36ADyc7GjuYgdApIW OxRFuf4yUtWNj7qaJeeSEkIIUbXY1HwAWldTS1cq9wkhRP1Ky9V37WtWg0BqeCcf7DaqORaXQVh0mkmT6oK yMTxuAzsNGqGh/iYfdy23i6k5KRxITmH7i09zH78tZKyC0nNLUKtgo5 rrXeX30zKZC6enxUdHQ0L774IjNnzmTAgAEA7Nu3j1WrVvHOO 9Yp5VCCCEqFFPata y0ucG1ho0LIQQwjTppV37PJ3MD6R83Ry4o3cL1h6K44u/L7DCxEDqi78uAHBnnxb4mDE yqCdjwsHo9K4kGSZ7w5DNirY2wUH28Y/H6hJgdTV46Nef/11PvroI 655x7jsvHjx9OtWzeWLVvGjBkzLN9KIYQQFYozdu2rJpAqrdwXk5pLsVaHrcaMklFCCCFqLa20a19NMlIAc4a0ZV1YHH eSuJMYna1GZ3TiVlsP52ESgVzhgTX6JiW7s3QlMZHgRnzSBns27eP0NDQcstDQ0M5ePCgRRolhBDCNDGlc0hVNxeHv5sDjrYairWKMfgSQghRd9Jzr67aZ7623i6M66qfkPeLvy9Uu/3SvyMBGNfVnzbNq 7 XRlLz0PYlMZHQQ0CqcDAQJYvX15u YoVKwgMDLRIo4QQQlQvv0hLUnYhUPkcUgZqteqqghPSvU8IIepa2ap9NfPQUH1maePxBGM1voocjklj4/GEMo pCUNGKjoljxKtrsb7MWhqGSmz6xh /PHH3Hnnnfz   /0798fgIMHD3Lu3Dl  uknizdQCCFExeLT9ZklVwcbPJyqn2Q32NuFkxezuJCcw0h8rd08IYQQV6lN1T6Dbi3dGdfNj80nErl/5SG f/AGurZwL7NNeEImM78 hFanMLarH91auleyt q18HDE3kZNYYmO PR8gmqY2QLIKSwhKkV/I69TE8lImR1IjRs3jrNnz7JkyRJOnz4NwG233cZDDz0kGSkhhKhDhm59rTydTKqsauzrniSV 4QQoi4pikKqBTJSAB/e1ZOU7IMcjE5jxlcHmdSn5ZVpqRT48XA82QUl9A1qxkd396zVsdRqFW29XTh1SX8TrjaB1OnSbJSvmz3NXexr1a6GwuRA6quvvmL8 PE0b96cwMBA3n77bWu2SwghRDUMGanAZlV36zNoW9rX3XBHUAghRN3IK9JRVKIPoGqTkQJwtNOwYmYo9y7fz8mLWSzdFVlum64t3PhyZl8c7WpfGS/Y29kYSA3vVPPeDMZufU0kGwVmBFLfffcdjzzyCL1792bChAlMmDCBkJAQa7ZNCCFEFdJyiwBo7mpn0vZ paVvDeOqhBBC1A3D Cg7GwUn0z6yq TmYMt3s/qz kAsGXlFZdZ5ONkxtX8r3Byq7/Jtiiu9GWp3E85YaKKJjI8CMwKpHTt2kJ6ezm //cbGjRt566238PX1Zfz48UyYMIFBgwahVks5XSGEqCtppV eniZ K3u76rtSJGcXoiiKyROtCyGEqB3j CgnLZb66PVwsuPRm9pZZmdVMEyfUdvKfeHGin01H7PV0JgV TRr1oxp06axfv16UlJSWLRoEfn5 UydOhUfHx/uu 8 fvzxR3JzpduIEEJYm2Fyx2bO5gVS cVacou0VmuXEEKIstJKS583c6595bu6ZokS6Gm5RYQnZALQI/A6DaSuZmdnx5gxY1i8eDFxcXFs2bKFoKAg3njjDT766CNLtlEIIUQFDF37PE0MpJzsbHCx13dESJbufUIIUWeuVOxrfIFU2 b6jFR6XrHxe8dc209dRqfox0e1NHFcb2NgdtW yoSGhhIaGsrrr79OcXGxpXYrhBCiEumlXfuamdHh3tvVnpzCEpKzC2s8QaMQQgjzXJlDqvH1BnC009DCw5GLGfmcScxmQLCX2fvYGn4ZgFFdmtbUG2YHUs8880yFy1UqFQ4ODrRv357x48fj6elZ68YJIYSonLkZKQBvF3uiUnIlIyWEEHUovbRrX2PMSIG O97FjHyOxKabHUjlFZWw 1wyAKO7 FmjefXG7EDq6NGjHDlyBK1WS8eOHQE4e/YsGo2GkJAQFi9ezDPPPMOePXvo3LmzxRsshBBCPyeJISNlViBlLDhRYJV2CSGEKC/N2LVPC9S JHldC23tyeYTiRyKTjP7sbvOplBYoiPQ05EQP1crtK7 mD1GasKECYwYMYKEhAQOHz7M4cOHiY PZ TIkdxzzz1cvHiRIUOG8PTTT1ujvUIIIdDPEF s1c9FYm7XPpAS6EIIUZcMxYEaa0aqXxt9T7PDMelodebNg7UtIhGAUZ39mly1WLMDqf/ 97 88cYbuLldqQHv7u7OggULeP/993FycmLevHkcPnzYog0VQghxheFL2dFWY9aEi1eXQBdCCFE3royRapyBVIifK852GrILSjiTmG3y40q0OrafSgJgVOemNT4KahBIZWZmkpSUVG55cnIyWVn6 vAeHh4UFdWsqocQQojqpdWgWx9cFUjlSCAlhBB1JT3PMEaq8RWbALDRqOnduhkAYTGmd /bcz6FzPxiPJ3tCA1qevUTatS174EHHuB///sf8fHxxMfH87///Y9Zs2YxceJEAA4ePEiHDh0s3VYhhBCl0ksLTTRzNm/meslICSFE3WvsGSmAvqWB0KHodJMfs3x3JADjewSgUTetbn1Qg2ITS5cu5emnn2bKlCmUlOijaxsbG2bMmMHHH38MQEhICCtWrLBsS4UQQhgZKvaZMz4K9FX7QAIpIYSoK4rSuOeRMggN0mekDkWloShKteOdjsdl8M/5VGzUKh4c0rYumljnzA6kXFxcWL58OR9//DGRkfoos23btri4uBi36dmzp8UaKIQQoryaVOwD8CnNSKXmFqHVKU3yDqEQQjQkWQUqtKXxk0cjnEfKoFdgM2zUKhKzCohPzyfQs qJdZf8dQGACT1b0MLDsS6aWOdqPCGvi4sL3bt3t2RbhBBCmKimGSlPZztUKtDq9OXTm5dmqIQQQlhHeq6 IJCznQYH83pjNyiOdhq6tnDnWFwGYTFpFQZSxVodWfnFxKfns7W0Wt9DQ5tmNgpqEEgVFBSwaNEidu7cSVJSEjpd2RTlkSNHLNY4IYQQFatpRspGo8bL2Y6UnCKSsgolkBJCCCtLy9OXJGjmXOP8RYPRN6gZx Iy2Hs ldt7tSyz7kxiNjO/PsilzCvzFI7s7Et736Y1d9TVzP6Lzpo1i23btjFp0iT69evX5OrBCyFEY2DMSJkZSAF4uzqQklMklfuEEKIOpOXoM1KeZhYHaohuCvFh e4ofv03gRfHhuBVejMuOiWXaV8eKDP 1t3RlqdHNO3ic2YHUps2bWLz5s3ceOON1miPEEIIExgndzSzax/oK/eduiQFJ4QQoi4YM1JOjT QGtDWi24t3DlxMZNVe6N5ZlRHLmXmM3WFPogK8XNl7ZwbcHfUP9emnnAxu/x5ixYtcHVtuik6IYRoDAzzSJlb/hykcp8QQtSl9Fz95bZnE jap1KpeHhYMAAr90YTm5rHtBUHuJiRT5CXE9/O6o Hkx0qlarJB1FQg0Dqww8/5IUXXiAmJsYa7RFCCGECwzxSXs7mj3GSuaSEEKLupOXpu/Y1hYwUwOgufrRt7kxWQQnjPt3NheRcAtwd G52f P3y/XC7EAqNDSUgoIC2rZti6urK56enmV hBBCWJeutOIe1DAjZQikZIyUEEJY3ZWMVNMIpDRqFXNLK/HlFJbg5WzHt7P707JZ1eXQmyKzc4z33HMPFy9e5O2338bX1/e6SNsJIURDklVQjE7R/9/c8udwdUaqoJothRBC1FZabtOp2mdwe6 WLN0VSVpuEd/M6kewt0v1D2qCzP6L7t27l3379tGjRw9rtEcIIUQ1Uku79bk62GCrMbtjgXFS3iTp2ieEaMBOXcrC3iaftt713ZLaSS/t2ufZRLr2AdjZqNn8xGC0OgVn 6YTIJrL7GceEhJCfn6 NdoihBDCBIbxUebOIWUgY6SEEA1dbGoeEz77B7UavnnAnn5t6rtFNZduzEg1nUAKwMFWU99NqHdm38p89913efbZZ/nrr79ITU0lKyurzI8QQgjrMs4hVYNufXAlkMouKKGgWGuxdgkhhKUs3XWBIq2OgmIdD6z05kS8 dn3hsJQ/tyriQVSogYZqTFjxgAwfPjwMssVRUGlUqHVypeyEEJYk6HQRE0zUq72NtjbqCks0ZGcXUig5/U3QFgI0XAlZRfww F4AIK9HbmQnM99Xzmy8TEIbGR1zUq0kJnf9MZICT2z/6I7d 60RjuEEEKYKK10Mt6aZqRUKhXNXey5mJFPam6RBFJCiAbl63 iKSrR0auVB988EMKUZbsIT7Bj9QGFF8cq9d08s2Tmq1AUfWE2D0fJSDU1ZgdSQ4cOtUY7hBBCmOhKRqrmX8quDvqP/ yCYou0SQghLCGroJjv9unnKn1kWDtcHWx4aGgWj3/fnG3h8MIYaEwFo9Pz9I11d7TBRtOIGi5MYlKH09jYWLN2evHixRo1RgghRPWMY6Rq2LUPrg6kSizSJiGEsIQfwuLJLiyhvY8Lw0N8ABjWMR87jUJkiooLyfXcQDOl5eqDJ0/p1tckmRRI9e3bl7lz53Lo0KFKt8nMzGT58uV07dqVn376yWINFEIIUZaxal8Nu/YBuDros1mSkRJCNCT/nE8BYHLfQNRqfRDi6qAwsJ1 DP7W8HprWo0YAqlmTaj0ubjCpEAqIiICZ2dnRo4ciZ fH7fccgsPPvggjz/ ONOmTaN37974 Pjw1Vdf8f777/PEE0 YdPBdu3Zx2223ERAQgEqlYsOGDcZ1xcXFvPDCC3Tr1g1nZ2cCAgK47777SEhIKLOPtLQ0pk6dipubGx4eHsyaNYucnBzTz4AQQjQyaXmSkRJCND06nUJYdBoA/dqUrSoxqrP s2pbROPqHmfo2lebrtii4TIpkPLy8uKjjz7i0qVLfPbZZ7Rv356UlBTOnTsHwNSpUzl8 DD79u1j3LhxJh88NzeXHj168Pnnn5dbl5eXx5EjR3j11Vc5cuQIP//8M2fOnGH8 PFltps6dSrh4eH88ccfbNq0iV27djFnzhyT2yCEEI1NWi3nkYIrgVSWBFJCiAbiXFIOWQUlONlp6OzvVmbdiM4lqFQKx NUJGY2nmDqSkZKuvY1RWb9VR0dHZk0aRKTJk2yyMHHjh3L2LFjK1zn7u7OH3/8UWbZZ599Rr9 /YiNjaVVq1acOnWKLVu2cOjQIUJDQwFYtGgR48aN44MPPiAgIMAi7RRCiIYkLUcfSHnVKpCSrn1CiIblUGk2qlcrD2w0Ze/1 7gq9AqEI7HwR4QN0wfURwvNl54rGammrFHNbpaZmYlKpcLDwwOAffv24eHhYQyiAEaMGIFarebAgQOV7qewsFAmEhZCNEoFxTqyC/VZJC8X xrvR7r2CSEaGkO3vtDWFU8WNaqLvvT5tojGk90xZqQkkGqSGk0gVVBQwAsvvMA999yDm5s 3ZuYmIiPj0 Z7WxsbPD09CQxMbHSfb3zzju4u7sbfwIDA63adiGEsJS0XH3gY6tR4eZQ84sJyUgJIRqaQ9HpAPQNqiSQ6qz/d98FDZn5jaN7X5phjJR07WuSGkUgVVxczN13342iKCxZsqTW 3vppZfIzMw0/sTFxVmglbVXUKzlu/0xfLUnqr6bIoRooFJz9JWrvJztUdViMhU3yUgJM QXaVm1N5rPd57n853n2Xshpb6bJJqYhIx8Lmbko1Gr6NnKo8Jt2npDex FEp2Kv8441m0DayhdMlJNWoMPjw1BVExMDDt27DBmowD8/PxISkoqs31JSQlpaWn4 flVuk97e3vs7WveJcZa/jmfwisbTuJqb8Ok0Ja4OcibTghRVkquoVtfzcdHgXTtE ZZvjuSj/44a/zdTqNm/3 G16rgiRBXC4vRZ6M6 7vhYl/55emoLnAuCbaFOzKhZx01rhbSpGpfk9agM1KGIOrcuXP8 eefeHl5lVk/YMAAMjIyOHz4sHHZjh070Ol09O/fv66bW2s3dfShg68L2YUlrN5v3iTIQojrQ1pO7cdHgXTtE b5/aS u/yQDt74uTlQpNWx/dTlem6VaEoORenHR1XWrc9gVGf9OKm/zjhSUKyzertqK13mkWrSahRIffvtt9x4440EBAQQExMDwMKFC/nll1/M2k9OTg7Hjh3j2LFjAERFRXHs2DFiY2MpLi5m0qRJhIWFsXr1arRaLYmJiSQmJlJUpK9Y1alTJ8aMGcODDz7IwYMH eeff3jssceYMmVKo6zYp1areGhoMABf7omioFhbzy0SQjQ0qYaMVC0zAZKREqaKS8vj1KUsNGoVn0zuyZR  nHFW8MlkBKWY6jY1zeoWZXbdWsBfm46covU7L2QUQctq7nCEsgpNGSkGnwnMFEDZgdSS5Ys4ZlnnmHcuHFkZGSg1eov9j08PFi4cKFZ woLC6NXr1706tULgGeeeYZevXoxb948Ll68yMaNG4mPj6dnz574 /sbf/bu3Wvcx rVqwkJCWH48OGMGzeOQYMGsWzZMnOfVoNxW48AWng4kpJTyI H4 u7OUKIBiY1x1KBlCEjJYGUqNrWcH02ql QJ82c7RjVWd91fve5ZPKK5PUjai /SMvZy9kA9G5ddSClVsNIw S84alWb1ttZOTp/1WrlFoVBxINl9mB1KJFi1i fDkvv/wyGo3GuDw0NJQTJ06Yta9hw4ahKEq5n5UrVxIUFFThOkVRGDZsmHEfnp6erFmzhuzsbDIzM/nqq69wcXEx92k1GLYaNQ8ObgPAsl2RlGgbftpaCFF3UizWtU//pV6k1Un2W1RpW4Q 8zSqiy8AnfxdadnMkcISHbvONpyiE4qicDQ2nYy8ovpuijDTmcvZ6BRo7mKHj2v1n22ju g/B/88lYZWp1i7eTWWlqv/t5mTDrW6cVQZFOYxO5CKiooyZpCuZm9vT25urkUadb2b3LcVns52xKbl8eeppOofIIS4bqRaqNiEi50NhqJ/kpUSlUnNKTbO7TOysz6QUqlUjO6iz0pti6h8qpG69sn2c9y eC8PfhOGojTci2tRXkSCfj7PTv5uJlUj7d9Wi6uDjpSc4gZdQTK99LLY01luVjVVZgdSbdq0MY5putqWLVvo1KmTJdp03XO003BHrxbAlS4VQggBV8qfN69lIKVWq3CxM4yTkoITomLbT6WiU6BLgBstmzkZl48qDaq2n0qiuAH0nPhyTxQL/zwH6Oci2h ZVs8tEuaIuJQJQJcAd5O2t9XAHb30UcqyXZFWa1dtpZV27WvmXP/vEWEdZgdSzzzzDI8  ijr1q1DURQOHjzIW2 9xUsvvcTzzz9vjTZel0Z31d/t237qcoP4khJCNAxXxkjVfgoHN0cZJyWqti1CH5AYMlAGoUGeeDrbkZlfbKy2Vl82Hk/gjU0RALT20gd7S/6 UJ9NEmYKL81IdQ5wq2bLK2YPzkKjht3nUjgRn1nhNvlFWhZsDGf5rsh6yVIaM1JOch3XVJkdSM2ePZv33nuPV155hby8PO69916WLFnCJ598wpQpU6zRxutS71bN8HK2I6ughIP1/CUlhGgYFMVy80iBVO4TVdPpYH k/gL15hCfMus0apVx2d/nkuu8bVf7crc I3H/jUF8N6s/GrWKXWeTOXmx4otr0bBodQqnL kLTXT2Nz2QCvTUMr6HNwBL/j5fbn1RiY6HVx9m5d5o3tp8ig 3nS23jbUZx0hJ174mq0blz6dOncq5c fIyckhMTGR Ph4Zs2aZem2Xdc0ahUjOum7TmyT7n1CCCC3yIaiEv1dVUtkpK4EUtK1T5QXl25DTqEWOxs1Hf1cy63v30Y/309YdHpdN82oRKvjdKL InzGgCACPZ24tbs/IFmpxiI6NZf8Yi0OtmraNHc267EPDW0J6Oc5i0zOMS7X6hSeXneMv84kY6fRX p tvM8S v4NZFunIxXMlJNVa0m5HVycsLHx6f6DUWNjO5aGkhFXJaBs0IIUnMcAHCy0 Bop6lm6 pJCXRRlYgE/eujo68rtprylwv9SgOpf Mz6q3yY1RKLoUlOpztNLTy1HfrM8zHuPnEJS5m5NdLu4TpDIUmQvzc0JhZ2a6jnzMjOvmgKPDarxEUa3UoisJ/fj7BbycuYatRsXxGKC MCQHgnd9Ps ts3WVQU6 q2ieaJrOL2rdp06bKiiqRkQ130F9jMzC4OU52Gi5lFnDyYhbdWpo2CFMI0TSllAZSlujWB1cyUlmSkRIViLikf511qWTcSitPJ7xd7UnOLuTf ExjYFWXIi5dqfZmKC/dyd NXq08OBqbwb4LqUzq07LO2yVMZ/gbmjM 6mpPjejA7nMp/H02mf/74TjNXexZFxaHWgWfTunF0A7eDO3gTXx6HqsPxLJoxzmGdPC25FOoVEKG/l8/d na11SZHUg99dRTZX4vLi7m6NGjbNmyheeee85S7RKAg62GYR292Xwika3hiRJICXGdM2SkLNGtD2SMlKhaRII kKrsAlelUtE3qBmbTyRyKDqtfgKpSooU9GvjydHYDMKi0ySQauAMf8PKAvbqdG3hzpJpvZnzzWF OZZgXP7 pB6M7eZv/P2J4e35ISyeQ9HpHIpOo2 Q9V vsaVD3Ft7ys2qpsrsQOrJJ5 scPnnn39OWFhYrRskyhrZ2ZfNJxLZfS6Z/xvdsb6bI4SoR6m5 kCqtqXPDaRrn6hKxCX966OqAgChrT3ZfCLRONdUXTNWe7umjX1be7KUSA7VU7uE6Sr7G5rj5hBfPry7B0 tO4aiwILbOpcLoH3dHLizTwu PxjHkr8u0HemdQOpvCJIztZnSVt5ymdsU1WrMVJXGzt2LD/99JOldidK9QpsBsDpxGxKpAy6aAS0OoVirc7YV11YjvUyUnK3VJSVlqviUqb 9RFSxQWu4a5 WEw6Ol3dvt8VRam0W1if1vrvzgvJuaTmFFrkWMLykrILSMkpRK3Sj5GqjQk9W7BuzgBW3t XmTe2qXCbuUOCUatgx kkTpW dqwlrjSGd3NQcHeS109TZbFA6scff8TTs 7T k1dK08nnO00FJboiEzJre/mCFGl3/69RK/Xt9H 5d9p//Lv3LpoD5cyZbC3pVh jJRkpETFIhL0lwdBXg642FfeeaWTvytOdhqyC0o4czm7rpoHwOWsQtJyi9CoVXTwLVtVsJmzHe19XAA4HFO7qoJbTibSfcE2VuyWMeCWZujW16a5s0UK6PRr48mwjpUXQQtq7mzs7mftCn7Gbn1echO8KTM7kOrVqxe9e/c2/vTq1Qt/f3/ 85//8J///McabbyuqdUqOpXeDTR84AjREO04fZkn1x4l66qL8vCELKatOGCRO8IC0nL1mShPZ8sEUm6GjFShZKREWRGX9JcHnQOqLkdto1HTu5U  1PX3fsiLunniQr2dsbBtvxFeOhV2bKa0uoU3t9ymuzCEt787RTf7Y p8b5EeVcyinU3Bvzh0qqOv/57ibi0PKsdJyZV/28rTwmkmjKzx0hNnDixzO9qtRpvb2 GDRtGSEiIpdolrtI5wI2wmHQiLmUxsVcLkx8Xl5bHHxGX0UmXhAbB2d6GW7v7G7MApjoel2FSP3 NWsXoLn4EeDjWtIk1tj8ylYe/O0KJTmF8jwDemNCV5JxCpn95gAvJucz4 iBrHrwBNzOfuyjryhgpKTYhrCsiQR YdPZ3qXbbvkGe7DmfwqHodKYPCLJyy66IqGZsTd gZnx/MLZWk9pvC08kMiUXtQp0Crz6y0lcHWyY0NP072JRuer htbQtYU7Qzp4s tsMst2RfLGxK5WOU5cmn58VKAEUk2a2YHU/PnzrdEOUYXONchInUnMZvKyfWTkyZ3mhuT7g7Gsnt3f5GBqy8lLPLL6CKYOPfhsx3l eGgAbb2rv/ixlMISLU tPUZhiY7hIT58eHcPbDVq3J1s W52f 7 Yh8nL2Yxa Uhvnmgv0W6b1yvUqVrn6gjpmakQB wQH1kpKoum20Yv3XyYib5RVqzP3sURWHxX/ruX4/d1I6M/GK 2RfDSz fYGRnX5zszL6EEteobenzmnp4aDC7ziazPiyOJ4a3x9vVMjenrnala5/czG7KTPoUyMoy/QLeza1u3wzXgy6lKe IS1koilLlPF4AMam5TP/yABl5xbT3calxSVFhWX fTebf ExmrQrjmwf6VdgV5Wq7zibz PdH0Sn6ft8B7g5Vbv/vxUwik3OZtuIAPzw8kBZ1lJn65WgCiVkF LrZ8/nU3mUm7gz2dmHVA/24Z/l DkWn89B3h1l Xyh2NhYbnnldSZFiE6IOFBTDhWT9e7SLCYFUz1YeaNQqEjILuJiRX2efPVfKZlfcLaxlM0f83BxIzCrgWFwGA4K9zNr/P dTOXExEwdbNTNvbIOHoy07zyQRl5bPrrMpjOnqV vncD3LKyohqnTsd11mpABuaOtJz0APjsVlsHJvFM NtnyPqpjSQEq69jVtJgVSHh4e1V68Gy7wtVqZdMzS2vu6oFGrSMst4nJWIX5VXFDnFJYw7csDJGUXEuLnyto5N DhZJm716J2Tl7M5J5l zkYlcag93ZUe3f0cmYhxVqFW7r58 k9vaqd8T0lp5C7l 4zBlPr5w6wyl22q2l1Cl UDtidPahthcFh1xbufD2zL9O/PMjfZ5N5et0xk57P1TYcvcjnO89TUFLx54tGpeLRm9pxV2hgzZ5II6DTXRkjZeny51mSkRJXOZMIWp0KL2ctPq7Vv9ac7GzoGuDG8fhMwqLTaFEH3d5yCkuITtWPb lUyUW4SqUiNKgZm/69RFh0mtmBlOGzbUrfVsZxiaM6 /Hlnii2RSRKIFVLpy5loyjg42pv9e qa6lUKh4ZFsycbw/zzb4YHhoabHa3 6rodBAvgdR1waRAaufOndZuh6iCg62Gdt4unLmcTcSlzCoDqa0nE4lLyyfA3YFvZvWTIKoB6drCna/u78uMrw6SklNk0mNuDvHh48k9TQo6mrvYs3p2fyYt2UdUij4ruW7OANydrDcuyTB wM3Bhnv6t6p0u9AgT5ZO78OsVYf47cQlnO01vHdn92pv0ABsPJ7A0 v1c4NU5c3fTjGumz/OVVQYa8wy8m3RKfosQTMLFZswZKSKSnQUlmixt5FulwIiLun/7RxQZNJ7FPTv8ePxmRyKTquT8UP/xmUAEODuUGXxlb5Bnmz69xKHzCw4cTmrgD3nUwCYNehKKe1RnX35ck8U208lUaLVYaOR7HpN1Ve3PoMRnXxp5 PC aQcVh I5aHSIhSWcDlbRZFWhY1awd9duvY1ZSZdcQwdOtTa7RDV6BzgxpnL2YRfzOLmEN9Kt9sWkQjApNBAfFyr7gom6l7fIE/2vnizsTtDVextNHTydzX5QgbA392R72b3564v9nE6MZv7Vx7k21n9rRJcKIrCktI7tjMGBlVZIhlgSAdvPp3Si0fXHGF9WDyuDra8ckuncs9PURSOxKaTmV/MxYwCXtsYjqLAPf1acXdoywr3/cz640Sl5PL9wVhmD25rmSfYwKTm6C8WPZw0ZbpP1oaLnQ0qFSiKfpyUvYsEUgIiEvTvyc7 pnf57BvUjC/3RHEoqnalxk11KFp/HENlvsqElo7fOhKTjlanmJwJ3xZxGYBerTwI9HS6an eeDrbkZZbxMGoNAa2a16T5gvqp9DE1dRqFQ8NDeb/fjjOl3uimDkwqNou96aKSdV/RrdoBnJ/qmmr8dVVXl4esbGxFBWVvbPevXv3WjdKlNfZ343/Hb1ovINTkfwiLX fTQZgdJfKgy1Rvzyc7OjVynqZwjbNnfludj8mL93PkdgMPv7jLK/c2tnix9l7IZV/40vHDwwMMukxY7v5896d3Xnux3/5ck8Ubg62PDmivXG9Tqfwfz8e5 cjF8s87vZeLXhrYlfUlVwEzR3Slhd/PsHy3ZFMH9C6SWZWknMMpc8tFxSr1Spc7GzILiwhu6DEYtUAReN2dUbKVIaA5szlbDLziq2aCQeMlUwNhS4qE Lnhou9DTmFJZxOzKp0PNW1toXrb0qO7lK2 55GrWJEJx/Wh8WzLeKyBFK1UN8ZKYDxPQL4aNsZEjIL OlIPFP7t7bIfmNLK/a1kulVmzyzb2smJydz66234urqSpcuXejVq1eZH2Edhg aqgKp3eeSKSjW0cLDsd7u8IiGIcTPjXfv6AbAbycuoVihBP7iv84D vEDXmZcgN8VGsj82/SB3cd/nuWrPVGAPhP12q/h/HzkIhq1im4t3One0p0HB7fh/UndKw2iAG7v3QJfN3suZxXyy9GEWjyrhutiuj7D3MLDsheoUnBCXE2ng1OGQMrf9ECquYs9bZvrC1McjrVu9b4SrY4jsaZlpDRqFb1bG6oKmpYty8wvZt8F/SRA1wZSoB8nBfpgyxqfrdeDEq2O05fqNyMFYGej5sEh l4MS/ OpERrmfFMcWn6y2sJpJo swOpp556ioyMDA4cOICjoyNbtmxh1apVtG/fno0bN1qjjYIrg2ljUvMqveAxdEUY1cXXrO5gomm6KcQHJzsNlzILOHnRspM5H4/L4J/zqdioVcwe3Kb6B1zj/hvb8MzIDgC8vimCqSv2c9cX 1i1Tz/Z5Qd3defXxwex8bFBvHxL52q7stnbaJg9SP9l N6W08z46iCPrD7MifhMs9vWUMWm6bsXtfK0bDZTSqCLq8WkqcgrUmFvo9CmuXmvCUM3ukMmBiw1depSNnlFWlwdbOjo61rt9v1K23XQxPLsf51JokSn0N7HhTbNy1ctHNS OY62GhIyCwg3Y1oScUV0ai6FJTqc7DQEeVVfGdKaJvcNpJmTLbFpefx MtEi zR07ZPS502f2YHUjh07 OijjwgNDUWtVtO6dWumTZvG   /zzvvvGONNgrA09nOWFK2oslZS7Q6tp8qDaQ6SyUhoS9SMqyjNwBbwy3z5WBgqGY1vmcALZs5VbN1xR6/uR0PlgZh/5xPJax0MPgbE7pwe6 Kx0JV5Z7 rWjmZEtqbhF/n01m84lEpq7Yz6kqsriNSWya/v1v UBKMlLiCsNEvCF OrPHdhiyQ9aeT8rwHRjaulmVmeqK2mVKBmlrJd36DBxsNQztoP9s3Wbhz9brhSEA7eTvZtLf0Jqc7GyYOVD/XbT4rwsWyTLGSkbqumF2IJWbm4uPjw8AzZo1IzlZPyanW7duHDlyxLKtE2XcHKI/73 UZp6uFhaTTnpeMR5OttX2GRfXD2MXlAjLfdmfT8phS nFQ22qHKlUKv4zrhNrZvfnw7t68OFdPdjw6I1MHxBUo/252Nvw08MDjfvq3cqDrIISpn95kMMx6USn5Bp/YlPzGl2XnJhUa2Wk9IGUlEAXAOEJhol4zZ/KxDAB7vG4TAqKrTcVSljpBD3Vdesz6NHSA1uNistZhcSn51e5bUGxlr/O6K9rRlUx1nh0V/26reHlv49F9eq70MS1ZgxsjZOdhlOXsoxjzWsjrnSMVKAEUk2e2aOWO3bsyJkzZwgKCqJHjx4sXbqUoKAgvvjiC/z9/a3RRlFqVBdfvt0fwx8Rl3lzYtnqQ5tP6Du1Dw/xlXKswuimjj7YqFWcvZxDVEpuhd1UzLVs1wUURV86toMJ3WqqolKpLDpYu623C229XQAY0dmXKcv0Gak7l wtt22/IE  ur9vtdUGG4o4Q0bKS7r2CeuJuKRPQ3UOMH sSJCXE81d7EjJKeLExUxjYGVJiqIYuw6aun9HOw1dW7hzNDaDQ9FpZarwXWvfhVTyirT4uzvQrUXlhSlu7uiLRq3izOVsolNyCbLAZ v15Fhp fr6LDRxNQ8nO 7t14oVe6JY8tcFhnX0qfG MvNVpOZKRup6YfYV95NPPsmlS/qL9vnz5/P777/TqlUrPv30U95  22LN1Bc0b NF64ONqTkFHEs7kof9PTcIn4IiwdgQs A mqeaIDcnWy5oa1 Eso/LJCVupSZz/ O6ivqPTzMcnNuWIO7oy3fPNCPG9p64mpvU bHRq3iYHQac74Js qdc0vJKVSTWjoZr3TtE9YUYchI Zv/vlCpVPQM1PeIOHnROuMTY9PySM4uxE6jpntL0yrwwZWgq7rxW/sj9UUmhnX0rnKssf6zVb/PinqJiMqdupTFgag0VCoYaOYkydY0a3AbbDUqDkSlcdjMeceutuecvjBQm YKMgtN02dyIDVp0iS2bNnC1KlTmTlzJgB9 vQhJiaGQ4cOERcXx TJk63VToG uoyhe9 2q7oTrNoXTX6xli4BbgxuL6VYRVmG7imW6IKyYncUxVqF/m086dO64Xch9Xa1Z 2cAZx4bXSZn58eHoiznYa9F1J5bM1Rik2s1BSRkMWbmyJYsDGcNzdFWO1i8VqxqfpvY0/nAlwdLFvaXTJSwiA5W01SthqVSiHEr2bVy4wVZq1UhOFglL5bX7eW7mbN RNqrNxX9fitK2XVq08lWKPr9PXAMMZ2XFd/WtdzoYmr bs7ckfp Nwlf12o8X62husznqO6WKRZooEzOZBKT0/nlltuoVWrVsybN4/IyEgAnJyc6N27N82bywV8XTAMft1aWnY1t7CElXujAX2GQKr1iWuN7KwPpA7HpHM KafG 0nPLeL7g7FAw89GVadHoAcrZvTF3kbNn6cu89wPx9Hpqh4zdSwug7u 2MuKPVGs3BvNij1R3PXFvlrduTRVbJo GxXoWfO/X2WMY6TyJSN1vTt1SZ/tbOMFzjWcUsww5qWqqTpqI8zMbn0GhvFU55JyuJxVUOE2BcVaTpTeHDFl/4bP1rCYdFJyCs1qz/UqNjWPX4/rp6hoiN8jc4a2RaWCP09d5uzlbLMfX1SiY dpfTfsUZ0b1zhcUTMmB1Lbt28nMjKSWbNm8d1339G fXtuvvlm1qxZQ2GhfIDUlSEdvLGzUROdmse5pBzWHoojI6 Y1l5OjO0qY9REef7ujozopP/CX7ar5nfZvtkXQ16Rls7 bsaKVY3ZgGAvlkzrjY1axYZjCczbeLLSAhRnErOZ fVBcou09G7lwaM3BdM3qBn5xVru//qg1e6 G8Sl6TNSrawQSLkZu/ZJRup6F3FJn53sVIse4l1KM1LnLudQVGKZOXmudijGtIl4r XpbEevVh5A5V3xjsdlUKxV8HWzp2Uzx2r3GeDhSPeW7igK/Cnd 0yybPcFdAoMbt crlWMQasvwd4ujCm9YW3InJljX2Qm2YVqvF119Aq0dOtEQ2TWGKnWrVuzYMECIiMj eOPPwgICODBBx/E39 fRx99lMOHD1urnaKUi70Ng0oH549euIs3NkUAMHdIcJniE0Jc7ZGb9Hf /nf0Ipcyq65adbV1h2LpvmArbV/6jY//PAs0rcznzSG fDy5JyoVfLc/lve3nim3TUxqLtO PEBGXjE9Az34ZlZ/nhsdwqoH hHauhlZBSXc99UBIpMtH QY21CakWrtaf4d0uoYu/YVSkbqeheeoM9Idfav Z30ls0ccXWwoUir44KF3xOpOYVEJucC1Khr8ZWueBUHPYYpGEKDPE3 jBtVmpWqbJ/iiuTsQtaXjuduiNkoA0PbNh5LID49z6zHbgvXj7Eb2bkEtdT9ui7U M988803891335GYmMg777zD2rVr6d /vyXbJipxT79WqFVguHne1tuZO3q3qN9GiQatd6tm9G/jSbFWYcXuKJMes HoRV78 QRZBSUYer11CXBjbNemNU/ZbT0CePv2boC X/ziv84b1yVmFjB1xQGSswsJ8XNl5VVV/pzsbPhyZl86 7uRklPEtBUHuJhhepBqjlgrZqRcJSMlSh2L0wfs3WrxdaJSqa5077NwptYQ6HTwdcHDyfyiK4bxovsupJBVQXGVq enMn2f s/DPedTyCmU91BVvv4niqISHT0DPRjQtuEUmbhW95YeDGrXnBKd6d XADqdwh8R tfQ6C7yWrhe1Krub1RUFCtXrmTlypVkZmYyYsQIS7VLVGFkZ1 Ozx9Ffmm1MU8nOyl5Lqr18LBgDkSl8f3BWFp5OlFVAjMjr5iF28 hKHDfgNY8dnM7ALyc7Ztk5vOefq3ILijm7c2neX/LGTLyigls5siqfTHEp cT5OXEN7P6lbt4c3e05ZtZ/bj7i31EpuQybcUB1s8dgLdrDQeYVCI2tbRin5cVM1ISSF3XEjMLiUuzQa1S6NWqdvvqHODGgag0whOyuLOPZdoHVwpFmDp/1LWCvV0I9nbmQnIuO08nMaHnlYhRq1OM4x3NGX/V3seFNs2diUrJZdfZZMZ1ky72FckqKObbfTFA4 jV8PCwYPacT2HtoVgev7kdXi7Vf6Yfi88gKbsIV3sdA9o2/GqwwjLMDqQKCgr48ccf eqrr9i1axeBgYHMmjWL   /n8BA6RBaV1wdbI0XQEKYYmgHbzr7uxFxKYv5G8NNeswdvVuw4LYu9T7zfF2YMySY7IISFu04z7Jdkcbl/u4OfDe7Pz6V1LFt7mLPd7P7c9cX 4hKyeX1TREsuqeXxdql1UF8uqH0uTUzUtK173p2KLp0gtQAHa4OtXu/Xyk4YdmqlgeNhSZqXjF0VBc/lvx1gW0Rl8sEUmcvZ5NdUIKznYYQP9Pnx1OpVIzq7MvSXZFsC0 UQKoSq/fHkl1YQjsfF0Z2qnyi44ZiYLAXPVq6czw k2d/OM6y6aHY2VR wzq/SMu7m08DMCwkH7vGMT2hsACT/9QHDx7kq6  Yt26dRQUFHD77bezZcsWhg8f3uDvLAgh9F/470/qzvLdkSaV  7k58bDw4KviyDK4JmRHfBxc2DfhRQA3BxseWhoMC2bVT6BJ gHnS d3odbF 3ht38TeHZkB4tN0Hkp044SnRo7jQ5fN8t3HTRW7ZOM1HUtrDSQCm2tpZadVcqUQFcUxSLXCHlFJYSXVtQLbV3zWU5HlwZSf51OorBEi72NvoS6IdvVu3Uzs3t4jOqiD6S2n06iWKvDVnqIlFFQrOXLPfoucg8NbRzfKSqVigXju3DP8v38dSaZp9cf49MpvSrskVFUouPh1Yc5GJ2Gq4OGx27KAlzqvtGiXpj8aXnDDTfQo0cP3njjDaZOnUqzZg1/DhkhRFldW7jzyRTLZUuaGpVKxfQbWjP9htZmP7ZrC3eGdfTmrzPJLNsdaRx3VVuGOaRaeuahUVu nK4hs11UoitzYSmuL4aMVN g2gdS7X1csdWoyCoo4WJGfrU3IkxxLC6DEp2Cv7uDSRX1KtO9hTu bvZczipkx6kkxpZmkHaf0988qUmQ1iuwGc1d7EnJKWR/ZCqD2zf qqaW9N3 GFJyCglwd2B8j1qUhKxjvVo1Y n0UGavOsRv/17iWGwGjnYa/N0deHNiV1p7OaPVKTy9/hh/nUnGwVbN1zO70NHP9HFVovEz bZJWFgYR48e5bHHHpMgSgghKvDwUH21px/D4kmqZK4acxnmkGrlaZ1CFobiGSDjpK5XWQXFnE7UV8MLDar92A47GzXtfPTd4yxVcMIwf5Q5FfUqolaruLW7/mL  Z/ 5eTFTJbtumCsujekg/lzYqrVKuOcUtssMPF5U7I1PJF3ftd3eZs7NLjK7nEN0dAO3sZM1MWMfM4n5bD7XArTvjzApcx8Xv7fCX779xK2GhVLp4cSGuRW300WdczkV3Tv3r2t2Q4hhGj0 rXxpE/rZhRpdXyw7Qz/nE8hPCGz0vmpTGGo2Nfa07wyvKbSqFXGYEoCqevT0dgMdAq08izG180yWU/DfFKWmpjXUFGvNuOjDP5vVEf6BjUju6CEe5bt5 3SsS3Pje5Ir1Y127 hIuAfEZerndy7MdHpFMITMmv0nP45n8Lja46i1Snc2btljTL9DcHYbv7seeEm1s65ge9m9ae1lxNxafmM mgXaw/FoVbBJ1N6NYn5FYX5GtetASGEaMBUKpUxK7U LJ6pKw5wy6d7WLAxvMbBVGSyPpAKtFJGCqTgxPXuSjW8Qovt01Bw4kR87QtOFJXoOGKY46kW46MMHO00fDmzL11buJFdWrL8oaHBPHpTuxrvc2CwF852GhKzClgfFlfrNjYEJVr92J9bPt3D0qsK8JjicEw6D34TRpFWx5gufrx3Z7dGMTaqMv7ujtzQ1otB7Zvz3az  Ls7GF87797RXYqMXMckkBJCCAu6OcSHe/oFEuLnSkdfV1QqWLUvhg 3nTV7X0UlKv457w5Ar1YZFm7pFTKX1PXtSrbHcoFUvzb6gOefCynkF9Wuu D yFRyi7Q0d7GjoxkV9ari5mDLqvv7cUs3f54d2YEXxnSs1f7sbTTMHtwWgP/87wSbT1yyRDPrjU6n8PyP/7K1tKvisl0XyCsy7fPh1KUs7v/6IHlFWga3b84n9/RsUlO0BHo68d3s/gwP8eH9Sd25u69UrL6eSYFGIYSwILVaxTt3dDf /t3 GF7ZcJLPdp7H1cGGuaUZK1Psi/Qku9AGb9ciegVatpT01a7MJdW4M1JHY9P56p9oCoq1qIBbuvuXKXEtyisq0XE0NgMwBFLmT3RbkS4BbrTwcORiRj67zyUbJ66tiW0RiYB DkVLzmPn5WLP51MtN2zhqRHtuZxVwNpDcTy59ijO9jblunsVlehYtOMcLZs5MrlvLSfsshJFUVjwazg/H72IRq3Cw9GW1Nwi1h2K4/4b21T52Li0PKZ/eZCsghL6tG7G0ul9mmQBm2BvF76c2be myEagBrdIigpKeHPP/9k6dKlZGfrJ4hMSEggJ8fyc5wIIURjNu2G1rwwJgSAd34/zeoDMSY/dlu4DwAjO6ejtuIN3aZQAv1YXAbTVhzg1 MJ/BFxmW0Rl3n x39JybFclqUpOhyTTmGJDk9nG4K9Lff3V6lUxnFDhkIONaHTKfxR vjaBGN1QaVS8dbt3bi1uz/FWoW534YZs32gn/T3qXVHWbTjPC/8dIIv/r5Qj62t3IfbzvLNvhhUKvjo7h48PbIDAMt3VT91xofbzpCSU0gnfze mtkXJ5lQSTRxZn81x8TE0K1bNyZMmMCjjz5KcnIyAO 99x7/93//Z/EGCiFEY/fwsGAeGabPRL2y4SS/HLtY7WN0OvgjQh9IjeqcVs3WtXMlI9U4A6kzidnM OoguUVa rfx5O3bu9HJ343CEh1f/yOliKuyNVyf7bk5xBNLTwk5qrM 8Nl 6jIlJsxdV5F/L2ZyOasQZzsNA4O9LNk8q9CoVXx0d09u6uhNQbGOB74 xIHIVC5l5vPSz/ y UQihqTau2beWKkLS/  wGc7zwPwxoSuTOjZgkl9WtLcxZ6EzAJ OZZQ6WPj0vL49V99l8b/TuqOu6NtnbRZiPpkdiD15JNPEhoaSnp6Oo6OV ZyuP3229m fbtZ 9q1axe33XYbAQEBqFQqNmzYUGa9oijMmzcPf39/HB0dGTFiBOfOnSuzTVpaGlOnTsXNzQ0PDw9mzZolmTEhRIPz3OiO3DegNYoCz6w/zsmLVXfVOxbfnKRsB1zsSxgQbJnKZ5VpzMUm8opKmPHVQTLzi kZ6MFXM/tyb/9WPDWiPQDf7ItplM rLijKVdmezpYPUvoGNcPDyZb0vGLCSotFmMsQ6A0L8Wk0XcTsbNQsntqHfm08yS4sYfKy/Qx4Zwfrw JRq2Dx1D5m31ipCztOXzaWKn9hTAjTSqvsOdhqmDVI36Xvk 1nSa0ky7tsVyRancKQDt50beFeN40Wop6ZHUjt3r2bV155BTu7sv2og4KCuHjRvA D3NxcevToweeff17h vfff59PP/2UL774ggMHDuDs7Mzo0aMpKLgyP8vUqVMJDw/njz/ YNOmTezatYs5c aY 7SEEMKqVCoVC27rwqjOvmh1Cp/tOF/l9tvCWwIwrGMG9jbWLafcmItNrDsUR2JWAS2bObLy/r44l5ZyH9nJl2BvZ7ILSlhzILaeW9kwhSdkcTEjH0dbDUM6eFh8/zYaNcNDaje/0rbSQGpU6TxNjYWjnYYvZ4QyrKM3dho1dho1Xs52fHR3T8Z09eO50R2ZfoP xsqz64 z/VT9zj lKAqf/Km/UX3fgNY8PKzsWM5pN7QiwN2BuLR87vvqIFnX3JxIzi40Vit82IxxoEI0dmZ3XtXpdGi15SvwxMfH4 pqXjWdsWPHMnbs2ArXKYrCwoULeeWVV5gwYQIA33zzDb6 vmzYsIEpU6Zw6tQptmzZwqFDhwgNDQVg0aJFjBs3jg8  ICAgMYzg7YQoulTq1U8N7oj2yIuszUikQvJOQR7u1S47bYIfSWo0V1qdiffHG6NtNhEUYmO5aVlmR8eFoyH05UbfGq1ioeGBvPcj/ yYk8UMwYG4WDbODIadeXqSWitdW5Gd/HlpyPxbA1P5NVbO5k1me75pBwuJOdiq1FxU4iPVdpnTa4Otqy8v1 F61QqFa N70JOYQn/O3qRh1cf4ZsH nFD2/rpvrj3QirH4zNxsFXz5PD25da7Otjy7ez 3P3FPsITspi24gADrupqGZGQRWGJjl6tPLihbe1L1AvRWJidkRo1ahQLFy40/q5SqcjJyWH /PmMGzfOYg2LiooiMTGRESNGGJe5u7vTv39/9u3bB8C ffvw8PAwBlEAI0aMQK1Wc DAgUr3XVhYSFZWVpkfIYSoC 19XRnRyRdFgWV/Vzw3y5HYPCKT3bHV6BjWMcPqbWqsGamNxxNIyCzA29WeO3u3LLd Qs8WBLg7kJxdyM9HGkb3qYbkSrbHekUcBrf3xsFWzcWMfLMn5/3pSDwAA4KbG4P9pkStVvH pO6M7OxLUYmO5348XuOxZLW15C994YvJoYF4udhXuE2wtwvfzOqHq4MN/8ZnsvTvSOPP7nMpgD4bZU6wLERjZ3Yg9eGHH/LPP//QuXNnCgoKuPfee43d t577z2LNSwxUf8B7 tbNp3v6 trXJeYmIiPT9m7VDY2Nnh6ehq3qcg777yDu7u78ScwUOYAEELUHUO3mZ PxpOYWVBu/ZK/9EV8JvZMwNWhdnPwmKIxBlI6nWKsejZrUJsKMyp2Nmrj3D5Ld11Aq7NuF8nGJCY1l9OJ2WjUKoZ3sl62x9FOw5D2 hLg5nTvyyoo5rt9 kIM0/o3zDLhlmCrUfPplF54OdsRl5bPb/Uw/9S/8RnsOZ CRq0yvl8q0yXAnZ8fHsjDw4J5cHCbMj9vTOzKyEbWBVOI2jK7a1/Lli05fvw469at4/jx4 Tk5DBr1iymTp1apvhEQ/bSSy/xzDPPGH/PysqSYEoIUWf6tG5GvzaeHIxK4 M/zvLund2Md3HPXc7mj4gsVCqFuUOjqYvp/lzt66drX3ZBMU uPUZeUQmf39u70jvhFfnz1GXOJ Xg6mDD1CoutKf0C2TRjnPEpOax cQlbushXb7hShGH/m08S7tE5lvtWKO6 LGttCS9oZR2dVbvjyW7sIR2Pi6M6NS0L84d7TTcf2MQH2w7y5K/LjC R4DVsjqf7zzPsl2RFJVcyXyV6PT/n9AjgEBPp2r30d7X1TilgxDXuxrNTGJjY8PUqVN5//33Wbx4MbNnz7Z4EOXnp 9qcPly2TtYly9fNq7z8/MjKSmpzPqSkhLS0tKM21TE3t4eNze3Mj9CCFGXDOMQ1oXF8en2K4Unvijt7jeqcxztfHLrpC31kZEqKNYye1UYO04nsT8yjftKK  ZQlEUFpd2RZp Q2tj faKONnZMHOgvuLYkr8uoCiSlSrR6vimNNszrpu/1Y83PMQHjVrFqUtZxKXlVbt9QbGWL/foy9Y/NDQYtQUn4W2opt8QhLOdhtOJ2fx1Jtkqx/ji7wv8d sZMvOLyS/WGn KtQoOtmoeuUmKRAhhLrMDqVWrVvHbb78Zf3/  efx8PBg4MCBxMRYbj6ENm3a4OfnV6akelZWFgcOHGDAgAEADBgwgIyMDA4fPmzcZseOHeh0Ovr372 xtgghhKXd2K4582/rDMDHf57l7c2n HZ/jLEU8sPDwuusLYZApK4m5C3W6nhk9REORKXhYm Dl7Md4QlZzFp5iPyi6rsy7o9M41hcBvY2au6/sU212983oDVOdhoiLmWxq3Qsx/Vs07 XiE/Px8vZrsKxZZbWzNmOfkH6AgSmTM774 F4UnIKCXB3YELP6yOD6O5ky9TScuOL/6q6omdNfLc/hndLS5v/36gO7H7 pjI/h14eQTsf8wqGCSFqEEi9/fbbxuzTvn37 Oyzz3j//fdp3rw5Tz/9tFn7ysnJ4dixYxw7dgzQF5g4duwYsbGxqFQqnnrqKd588002btzIiRMnuO   wgICGDixIkAdOrUiTFjxvDggw9y8OBB/vnnHx577DGmTJkiFfuEEA3e/Te24ZnSrk7LdkXy6oaTlOgUBgY70zMwtc7aUdfzSH39TxQ7TifhYKvmq5l9jQPYw2LSmfvdYQpLqg6mlpSOjbo7NBBv1 q7AzZztuOefvruf4t3Wv4itTFRFMVYWOD G4NwtKubSoajuui75xm6FFamRKtjWWklxgeHtMVWU6OOM43SrEFtsNOoORSdzqFoy03C/cuxi7z6y0kAHr0pmMdubk gp1OZn6qyukKIypn9CRUXF0e7du0A2LBhA5MmTWLOnDm888477N6926x9hYWF0atXL3r16gXAM888Q69evZg3bx6gz3Y9/vjjzJkzh759 5KTk8OWLVtwcHAw7mP16tWEhIQwfPhwxo0bx6BBg1i2bJm5T0sIIerF4ze3442JXRke4sPwEB/GdfNjwfi6vRFkqIhWWKIrM3bCGgqKtSzfre 2teC2LvRr40mXAHdW3t8XR1sNu84m89TaY5VWLzt5MZNdZ5PRqFXMGVL1wPirzR7cBluNigNRaRyu4eSwTcHOM0mcuZyNs52G6TcE1dlxDUUIwqLTKp3QFWDzyURi0/Jo5mTL5L7X19hlXzcH7uzTAoAvSoPd2voj4jLPrD OosCMAa35v1EdLbJfIYSe2aOYXVxcSE1NpVWrVmzbts1YtMHBwYH8fPMGqw4bNqzK/uoqlYrXX3 d119/vdJtPD09WbNmjVnHFUKIhkKlUjH9htZML 3Wo1e3E8i6OFz5KsguKDar6IO5fj5ykeRsfbetO67qVtantSfL7uvDrJVh/H4ykRd/PsH7d3YvMz5GURQ 2a6fNPS27v4mDYw38Hd35PZeLVgfFs8Xf19g X2h1T oluLS8njt13ASMvSVGYN9XHhjQpcy813VpRKtzjgeb9oNrXF3qrssRMtmTnQJcCM8IYvNJxOveb3rlc2WtcHJzvqFVhqaOUOCWXsoju2nkzidmEWIn3ljuHU6hY//PMv2U/rx4 eTctDqFO7o3YL5t3WR0uRCWJjZGamRI0cye/ZsZs ezdmzZ41zR4WHhxMUFGTp9gkhhLAyjVqFc2kXL2sWnNDqFJbu0l8ozx7cFjubsl9Bg9t78 k9vdCoVfx4OJ7XN0WUudn23pYz/BFxGbUKHhpm/sD4OUOCUan0d nPXs6u3ZOpRlJWAdO PMCfp5KIuJRFxKUsfj2ewIyvD5FTWPdl5nU6hed//JdjcRk42Kp5YFD1Y8sszVAx8a3fIjgYVb7r2l9nkzl1KQsnOw33DSgfaF0P2jR3ZlxXfQEQc7NSiqKw4NdwFu04b3zNFWl1jO7iW 6mhBDCMswOpD7//HMGDBhAcnIyP/30E15e pmtDx8 zD333GPxBgohhLA wxgJawZSv5 8REyqvtvWlH4Vd9sa09WP9 /sDsDKvdG8u U0YdFpfLjtjHHeqLdv72b2nXqAdj4ujC6dfNawL2tIzy1i2pcHiEnNI9DTkS9nhLJ4am88nGw5HpfBnG/CKCg2f36w3MISwqLTCItOIyIhy QKhIoCC35N4OejF9GoVSy6pze bg7VP9DCZg1qw80hPhQU63hg5SFOxGeWWW/IRt3br1W9Ze0aAsM8c7/ e6naKofFWh1HYtMJi07j7c2n GZfDCoVvHJLJ1Y90I fHh7Akql9sLmOxpoJUZfMzpt7eHjw2WeflVv 2muvWaRBQggh6p6rgw2JWdYrOJGcXcj7W84AMHNg1d227uzTktyiEub9Es7SvyNZWloSHvQXiFP61XyC1oeHBbMlPJFfjiVwR6 WDGrfvMb7qkhOYQkzVx7i7OUcfFztWT3rBlp56bsgtvBw5N7l 9l7IZXH1hxlybTeJhdTiErJ5e6l 0jOvjK aELPAD66uyeaajINH27rwTf70lCp4KO7e9TbpKm2GjWLp/ZmxlcHORCVxn1fHWD93AG093Xlx8PxHIxKw1ZT/aSwTV3XFu4Mbt c3edSmPfLSZbdF1rh6yQzr5gpy/dz6lJWmeVvTOjKtAq6TgohLK9GtygyMjL48MMPjV38Pv74YzIzM6t/oBBCiAbJULnPGiXQM/OKmf7lAWLT8mjZzJGZA4Oqfcx9A4J4Y0IX2vu40Ka5M 18XJh3a daX2T3CPRgfI8AtDqFB78Js2jhCf3cWIc4HpdBMydbvpvd3xhEGY69YkZf7G3U/HnqMs/9cBydrvqsUkJGPtNWHCA5u5BmTra0ae6MjVrFL8cSeGXDySozU0v/DuKznd0A/QX2hJ4tav9Ea8HBVsOKGaH0aOlOel4x0748wMp/onj x MAPDi4LX7udZ8ta2ieG90Rexs1O88k8 z642iveZ3kFJYw4 uDnLqUhbOdhjbNnWnv48Jbt0sQJURdMjsjFRYWxujRo3F0dKRfv34AfPTRR7z11lts27aN3r17W7yRQgghrOtK1z7LZqRyC0uYufIgpxOz8Xa157tZ/U0ucjB9QBDTBwRZtD0A/72rO l5Rew l8L9Xx9kSAfvMuubu9jzyE3B LiafkFfrNXx2Joj7I/Uz4216oF dPAtPy/PgGAvlkzrzZxvDrPhWAIJmQX4VFPC/Xh8Bhcz8mnb3Jn1Dw2guYs9vx5P4Im1R/n YCwxqbl4OtsBecAgwB6wpaBYzZ n9PM3vTDGr8FcYLs62LLy/n5MXraPs5dzWPBrBACT rSUqnKlurf04ItpfXjwmzA2Hk8gMbMAH7crr5PzSTmcTszGw8mWdXMG0NFP5oASoj6YHUg9/fTTjB8/nuXLl2Njo394SUkJs2fP5qmnnmLXrl0Wb6QQQgjrujKXlOUyUgXFWuZ8G8bR2AzcHW35blZ/gpo7W2z/NWVvo2Hp9D7c9 VBwmLS2fTvpXLb7L2Qwro5A2jmXP1YHa1O4dn1x/nzVBL2NmpWzAile0uPSre/OcSXjyf35Im1RyssulCRAHcHvp3dn alFRVv6xFAbmEJL/58gr0Xrp5zLKjcYx8ZdpKHh3Uz6Th1pZmzHd/N6s kL/YRm5bH2K5 vHtHNymIcJWbQnyuvE4qmFfK2U7Dqvv7SRAlRD2qUUbq6iAKwMbGhueff57QUOuXkxVCCGF5li42UaLV8fj3R/nnfKr gu BhnXB52Rnwzez rH5RCI5V2XhdAos3XWBs5dzmPn1QZZM64O9TdW94D/84ywbjydgo1bxxbQ 3NDWq9rj39YjgFaeThyNrb5roY1GzegufuUmH57SrxXtfFw4edHQtT4d2Ai4APpsWmuvNIZ1PAU0vGJQPm4ObHj0RsKi07gpxEcKIlSgsteJSqViaAfvBnFjQojrmdmBlJubG7GxsYSEhJRZHhcXh6trw/mSFEIIYTo3Y0bKMl373t qL1VuZ6Nm YxQegZ6WGS/luRkZ8OkPi3LLR/cvjl3L93H8fhMBr67w6R9qVTw8eSe3BTiY/LxewR60KOW5yU0yJPQIM/S3zTAGcAbcC9d1rDHL3s62zGqi199N6NBs8TrRAhhHWbf/pk8eTKzZs1i3bp1xMXFERcXx9q1a5k9e7aUPxdCiEbKkl37krMLWbU3GoCFk3syMNiylfGsrb2vK6se6EcLD0eTtne1t H9O7sb50kSQghxfTA7I/XBBx gUqm47777KCnRf Ha2try8MMP8 6771q8gUIIIazP2LWvsPYZqZV7oygs0dEj0IOxXRtntqF7Sw/2vHCTydurVDK2RwghrjdmB1J2dnZ88sknvPPOO1y4oJ88Lzg4GCcnp2oeKYQQoqGyVEYqu6CYb/bFAPDw0OBGHWA05rYLIYSwPrMDqczMTLRaLZ6ennTrdqUKUFpaGjY2Nri5mT/bvBBCiPplyEjVdh6p1QdiyS4oIdjbmVH1NPGrEEIIURfMHiM1ZcoU1q5dW275 vXrmTJlikUaJYQQom65WqDYRLFWx5d7ogB4aGiwlLIWQgjRpJkdSB04cICbbirfb3zYsGEcOHDAIo0SQghRtyzRte9MYjbJ2YW4OdgwoWcLSzVNCCGEaJDMDqQKCwuNRSauVlxcTH5 vkUaJYQQom65GeeRqnlGKjxBX2q7W0t37KqZe0kIIYRo7Mz puvXrx/Lli0rt/yLL76gT58 FmmUEEKIumUIpAqKdRSV6Gq0j4iELAA6 8tYWSGEEE2f2cUm3nzzTUaMGMHx48cZPnw4ANu3b fQoUNs27bN4g0UQghhfa4ONmjUKrQ6hfS8InzdHMzeR8Sl0kAqQAIpIYQQTZ/ZGakbb7yRffv2ERgYyPr16/n1119p164d//77L4MHD7ZGG4UQQliZWq2imZMdACk5hWY/XqdTOHUpG4DO/u4WbZsQQgjREJmdkQLo2bMnq1evtnRbhBBC1KPmLnak5BSSmlNk9mPj0vPIKSzBzkZNW29nK7ROCCGEaFjMDqRiY2OrXN qVasaN0YIIUT98XLRZ6RSc83PSBnGR3X0dcVWI4UmhBBCNH1mB1JBQUFVzvau1Wpr1SAhhBD1w8vZHqBGGSnj CgpNCGEEOI6YXYgdfTo0TK/FxcXc/ToUT766CPeeustizVMCCFE3bqSkTI/kAovzUh1aSGBlBBCiOuD2YFUjx49yi0LDQ0lICCA//73v9xxxx0WaZgQQoi61dzFkJGqedc yUgJIYS4XlisI3vHjh05dOiQpXYnhBCijnk5l2akzOzal5pTSGJWAQAhEkgJIYS4TpidkcrKyirzu6IoXLp0iQULFtC fXuLNUwIIUTd8irNSKWY2bXPUPY8yMsJF/saFYMVQgghGh2zv/E8PDzKFZtQFIXAwEDWrl1rsYYJIYSoW57GjJR5XfsiLmUCMhGvEEKI64vZgdTOnTvL/K5Wq/H29qZdu3bY2MidSCGEaKyau9Ssa5 MjxJCCHE9MjvyGTp0qDXaIYQQop4ZuvblF2vJKyrByc60rwhD6fMuAe5Wa5sQQgjR0Jj0Lblx40aTdzh /PgaN0YIIUT9cbbTYG jprBER2pOEU6e1X9FFBRruZCcC0jXPiGEENcXkwKpiRMnmrQzlUolE/IKIUQjpVKpaO5iz8WMfFJziwj0dKr2MWcSs9HqFLyc7fBxta DVgohhBANg0mBlE6ns3Y7hBBCNABeLnb6QMrEghOGbn2dA9zKFSISQgghmjKLzSMlhBCi8TN3LikpNCGEEOJ6ZXKxifz8fLZv386tt94KwEsvvURh4ZU7lhqNhjfeeAMHBwfLt1IIIUSduDKXlPkZKSGEEOJ6YnIgtWrVKn777TdjIPXZZ5/RpUsXHB0dATh9 jQBAQE8/fTT1mmpEEIIqzMnI6XTKZy6JBkpIYQQ1yeTu/atXr2aOXPmlFm2Zs0adu7cyc6dO/nvf//L vXrLd5AIYQQdcfLxfRJeWPS8sgr0uJgq6att4u1myaEEEI0KCYHUufPn6dbt27G3x0cHFCrrzy8X79 REREWLZ1Qggh6pSXs75rX2pu9Rmp8IRMADr6uaFRS6EJIYQQ1xeTu/ZlZGSUGROVnJxcZr1OpyuzXgghRONjyEilmNC1TwpNCCGEuJ6ZnJFq2bIlJ0 erHT9v// S8uWLS3SKCGEEPWjeWmxiTQTik1IoQkhhBDXM5MDqXHjxjFv3jwKCgrKrcvPz e1117jlltusWjjhBBC1K0rY6SKUBSl0u0URSFcMlJCCCGuYyZ37fvPf/7D vXr6dixI4899hgdOnQA4MyZM3z22WeUlJTwn//8x2oNFUIIYX2epVX7SnQKWfkluDvZVrhdfHo ydmF2KhVEkgJIYS4LpmckfL19WXv3r106tSJF198kdtvv53bb7 dl156ic6dO7Nnzx58fX0t2jitVsurr75KmzZtcHR0JDg4mDfeeKPMXVJFUZg3bx7 /v44OjoyYsQIzp07Z9F2CCHE9cLeRoOrvf4eW1VzSR2KTgOgawt3HO00ddI2IYQQoiExOSMF0KZNG7Zs2UJaWhrnz58HoF27dnh6elqlce 99x5Llixh1apVdOnShbCwMO6//37c3d154oknAHj//ff59NNPWbVqFW3atOHVV19l9OjRREREyOTAQghRA14udmQXlpCaU0Swd8XbHIpOB6BfG t8/gshhBANnVmBlIGnpyf9 vWzdFvK2bt3LxMmTDCOvQoKCuL777/n4MGDgD4btXDhQl555RUmTJgAwDfffIOvry8bNmxgypQpVm jEEI0NV4u9kSn5lU5l1RYaUYqtHWzumqWEEII0aCY3LWvPgwcOJDt27dz9uxZAI4fP86ePXsYO3YsAFFRUSQmJjJixAjjY9zd3enfvz/79u2rdL FhYVkZWWV RFCCKHn5WwogV5xIJWeW8S5pBwA kggJYQQ4jpVo4xUXXnxxRfJysoiJCQEjUaDVqvlrbfeYurUqQAkJiYClBub5evra1xXkXfeeYfXXnvNeg0XQohGrJWnEwCnE7MrXH84Rt tL9jbGa/SculCCCHE9aZBZ6TWr1/P6tWrWbNmDUeOHGHVqlV88MEHrFq1qlb7femll8jMzDT xMXFWajFQgjR IUG6cc9hZWOg7rWoRh9t76 QTI SgghxPWrQWeknnvuOV588UXjWKdu3boRExPDO  8w4wZM/Dz8wPg8uXL Pv7Gx93 fJlevbsWel 7e3tsbeXu6hCCFGR0CB9d70zl7PJzCsuVwL9UFTp CgJpIQQQlzHGnRGKi8vD7W6bBM1Gg06nQ7QVxH08/Nj /btxvVZWVkcOHCAAQMG1GlbhRCiqWjuYk/b5s4AhJVmnwwKirWcuJgJQD8JpIQQQlzHGnRG6rbbbuOtt96iVatWdOnShaNHj/LRRx/xwAMPAKBSqXjqqad48803ad  vbH8eUBAABMnTqzfxgshRCPWN8iTyJRcDkWnM7zTlXGox IyKNYq LjaE jpWI8tFEIIIepXgw6kFi1axKuvvsojjzxCUlISAQEBzJ07l3nz5hm3ef7558nNzWXOnDlkZGQwaNAgtmzZInNICSFELYQGNWNdWJyxzLnB kP6MaUDg71QqVT10TQhhBCiQWjQgZSrqysLFy5k4cKFlW6jUql4/fXXef311 uuYUII0cQZCkn8G59JQbEWB1sNcWl5/HI8AYAHBrWpz YJIYQQ9a5Bj5ESQghRP1p7OdHcxZ4irc44JmrF7ki0OoVB7ZrTvaVH/TZQCCGEqGcSSAkhhChHpVLRt7R636HoNFJyCllb2q3v4WHB9dk0IYQQokGQQEoIIUSFDOXNV 2NZtqKAxSW6OjR0p2BwV713DIhhBCi/jXoMVJCCCHqz D2zQG4nFXI5axCAB65qZ0UmRBCCCGQQEoIIUQlOvi68sNDA4hNzQPAx82ewe2967lVQgghRMMggZQQQohK9Q3yNFbwE0IIIcQVMkZKCCGEEEIIIcwkgZQQQgghhBBCmEkCKSGEEEIIIYQwkwRSQgghhBBCCGEmCaSEEEIIIYQQwkwSSAkhhBBCCCGEmaT8OaAoCgBZWVn13BIhhADIBoqBDKCodFle6bJsQD6rREVq87rJAvKvekxh6e9ZgEzALOrTta9NaLivz raCvI axwMMYEhRqiMSqlui tAfHw8gYGB9d0MIYQQQgghRAMRFxdHy5YtK10vgRSg0 lISEjA1dUVlUruCFwrKyuLwMBA4uLicHNzq /mNClybq1Hzq31yLm1Ljm/1iPn1nrk3FqPnFvrqezcKopCdnY2AQEBqNWVj4SSrn2AWq2uMtoUem5ubvIGthI5t9Yj59Z65Nxal5xf65Fzaz1ybq1Hzq31VHRu3d3dq32cFJsQQgghhBBCCDNJICWEEEIIIYQQZpJASlTL3t6e fPnY29vX99NaXLk3FqPnFvrkXNrXXJ rUfOrfXIubUeObfWU9tzK8UmhBBCCCGEEMJMkpESQgghhBBCCDNJICWEEEIIIYQQZpJASgghhBBCCCHMJIGUEEIIIYQQQphJAilhJHVHLEvOpxBCiKZCvtOEKE8CqevYokWLuPfee1mwYAGRkZGoVKr6blKTsXjxYqZMmcLjjz/O/v370Wq19d2kJqWgoIBNmzaRkZFR301p0uTCybrk/FpWfn4 mzdvBuTcWlpJSQlpaWn13QwhzFIXnwMSSF2H0tLSGDt2LB999BGenp58//33jB49mq   qq m9bo7d27l65du/Lpp5/Srl079u/fz/Tp0zl27Fh9N63JWLhwIf7 /ixevJj4 Hh0Ol19N6nJ PTTT5k2bRqvvvoqZ86cqe/mNDlLly5lzpw5LFiwgDNnzsjNKwv673//i7OzM59  ik6nU7OrQV99NFHhIaGctttt/H4448TGRkJSLBqCT///DOHDx u72Y0SQsXLmT8 PE88sgjHDhwgKKiIqscRwKp69Dhw4eJjIxkx44dfPbZZ5w5c4b /fuzbNkytm7dWt/Na7TCw8P573//y7hx4zh27BhvvfUWhw4d4sKFC1y8eLG m9ckvPvuuyxevJilS5fy888/06lTJ9Rq RirrZiYGG6  WY   wzgoOD eOPP5g8eTKffPJJfTetSQgLC6NXr1588skn2NjY8N133zFx4kT27t1b301r9P7  28CAwNZvHgx69atY8uWLfKZYCEpKSncdtttLFmyhOeff56RI0cSFhbGwoULASRYrYUTJ04wZMgQJk2axKJFi8jMzKzvJjUZubm53HnnnXzyySf06tWLsLAwpk fzhtvvGGV48mnzXXEcOc Li4OnU6Hm5ubcd3zzz Pv78/H374YX01r9ELDAzkgQce4Pnnn8fBwQHQn sxY8bQoUOHem5d46bT6cjMzOS3335jwYIF3H333WRkZHDo0CHi4 MpKSmp7yY2alu2bEGr1XLkyBFee 019u/fT fOnfnoo4/4888/67t5jVpUVBTvvvsugwcP5uDBgyxevJjz58 Tm5vLrl27ALmzX1ORkZE88cQT2NnZERUVxV133UVaWhoZGRkUFhbWd/MavUOHDhEVFcX27du59957ee211wgICDB v8nrtmZSU1NZtWoVbdu25eOPP2bt2rX8888/9d2sJiMiIoITJ07w888/8/rrr3Pw4EHuuecevv/ e9auXWvx40kg1cRt27aNv// m9TUVONdutzcXGxtbUlJSTFu1717d8aPH09KSgqrV6 ur Y2KlefW0VRcHNzY8yYMTRv3hzQj5Pq3r07x48fZ/To0dx///3GCydhOkVRUKvV5ObmcvbsWQYMGMCnn35Kjx49ePLJJ nXrx/z58 v72Y2SoqiUFhYyG //UabNm1wdnYmPz8fgL59 xIfHy/ntoYMF5lOTk7Y2dkxd 5cXFxcKCgoAGDw4MEcOHAAkDv75jKc24CAAObOnUtOTg7Hjx/nzTffZNCgQYwYMYJhw4ZJxq W4uPjycnJKRMw5eTkEBQUxMWLF V1W0POzs4MHTqUuXPn8uSTT3LjjTfy7rvvcunSpfpuWqNmSBZkZGSQmpqKt7e3cd3s2bMZNmwYr776qsWPK4FUExUeHk737t154IEHmDlzJiNHjmTp0qUAzJgxg8jIyHJ3mocPH46npydHjhyRO/xVqOjcfvnllwDY2toCkJ6ezq5du3jzzTfZu3cvy5YtIyYmhk8//ZScnJz6bH6jcHWQaviyjo6Opn379nz33Xf873//Y XKlXz//fc89thjxjtPonrX3gCwt7fHw8ODU6dOkZWVhaOjI6C/2z9p0iRyc3NZtWpVPbe68bhw4QKgD450Oh2 vr58 eWXdOnSBcB4Nz82NpYhQ4bUWzsbo6vPraIoODg4MHbsWHr06EGvXr3YtWsXr776Ko8//ji2trY8  yzbNiwoX4b3UgYzi1cCVRbtGiBr68vd911F/Pnz8fT05MLFy6wZMkSRowYwSuvvAIg41SrsXfvXk6dOmX87je8bgcMGADAF198wZ49e9i4caNce5np119/5aeffiI2NtZY1CszM5MWLVpw6tQp43aBgYFMnToVtVpt Z5XimiS5s6dq9x5551KSkqKcvLkSeWRRx5RfH19lQ0bNiiKoihPPfWUEhgYqERFRZV53JQpU5RbbrlFURRF0el0dd3sRqGyc7tp0yZFURRFq9UqiqIoxcXFZR73n//8R nVq5eSnp5e101uNE6ePKl069ZNadGihRIUFKT06tVLWbp0qaIoipKXl6d07NhRcXV1VebOnWt8TE5OjrJgwQKlU6dOSmpqan01vcGr6Nx 8cUXiqIoSnx8vOLq6qoMHjxYefnll5W2bdsqPXr0UHbs2KH0799fmT9/vnweVOPIkSNKaGio0r9/f XIkSOKolz5LFCUsp nGRkZSrdu3ZS///67ztvZGFV0bktKShRF0Z/XDRs2KC //LISHR1tfExUVJQycuRI5f7771fy8/Prpd2NQXXnNiwsTFm fLnSqVMn5d1331UKCgqU Ph45euvv1ZUKpUSExNTn81v0Pbt26f07NlTCQ4OVlq3bq0MHTpU2b17d5ltDNcJjzzyiNKmTRvl1KlT9dHURuf06dNKz549lZYtWyohISFKu3btlP/85z KouivFVq1aqW8 OKLSm5urvExKSkpysyZM5U77rhDycvLs1hbJCPVBKWmprJ27VomTpyIl5cXXbp04ZVXXuHWW2/lkUceAeDtt9 mqKiI1157jcTERONjFUXBx8cHRVEkbV Bqs7t3LlzAVCr1Wi1WmxsbAD93TpFUUhPTycgIAAXF5f6fAoN2qJFi jQoQPHjx9n06ZNDBgwgHnz5vHLL7/g6OjI008/TU5OTpm7ds7OzrRp0waNRkN2dnY9tr5hq jczp8/n40bN9KiRQvWrVvHDTfcwK5du5g9ezb79 /npptuws3NjaSkJPk8qMKff/7JrFmzcHJyorCwkF9//RWdTlem6IEhiwJw8uRJEhISaNeunXG9vHYrVtm51Wg0xup8Q4cO5cknn6R169aA/nssKCiIwMBALly4YMwCirIM59bR0dF4bhVFKXNu /TpQ7t27cjNzWX27NnY29vTokUL vTpQ0BAAHv27Knvp9HgKIrCd999x7Rp0xg7diw7d 7k22 /JTc3l6 //pqsrCzjtobPiE8  YTU1FRWrlxp7P4bHx9fL 1vDL7  mt8fHwIDw9n69atPPTQQyxcuJBly5bh6OjIE088wZIlSwgLCzM xsvLC09PT5KSkow9LyxBAqkmSK1WExwcTFJSknGZv78/Tz/9NMXFxSxYsABHR0dWrFjB33//zdSpU1mzZg1vv/0227dv584775SLpkpUd24NVWE0Gg2gD6KKiop4//332bFjB3PmzDEGWKKsqoLUhx9 GIC5c fSp08fDh48WGa8WU5ODu7u7vj6 tZX8xu0qs7t/7d370FRnWcYwJ/lYiEBEa ASgQBQSUEoogkddQmRkjT5mJBTWirCWOMxZrYeGkETUxiSI0ZrI2xsZg0caRD0VabeEFEVExjjNHUCyAQxI6igS1Xkcvy9g zB1bxspFvL/L8ZpyBsxff88yye9493/ed559/HgAQExODt99 G/v27cPixYvh4uKCs2fPory8HMHBwVbeA9vWu3dvPPDAA9i0aRMefPBB5OTkYM ePQBMJ Qbf96 fTvuvfde Pj4QK/XY bMmVi4cKHJARZdcaNsjZ9TvXr1MpkPId/P/auoqMCQIUO4KMJ1GLPNyMjQss3JyQFgOm vtLQUwcHBaGho0LaVlJTAw8MD0dHRFq/b1l26dAmHDh3Cs88 i TkZAwcOBA//vGP8ctf/hIHDhwwWejLwcEBra2tcHJywooVK7Bu3TpkZWXhqaeewqxZs1BVVWXFPbFNjY2N2LhxI2JiYtCzZ0/4 voiKSkJL730EhYsWIDy8nLMnz8fvr6 ePfdd/HNN99oj21tbYWnp2fXDqHssnNbZDPq6uokNjZWEhMT5cKFC9r2pqYmWbp0qQwaNEjq6 tFRCQ7O1vi4 PlgQcekBEjRmjD06hzN8p22bJlMnjwYO2U8Y4dOyQpKUlGjhwpfn5 snPnTmuVbRf0er1ERETIO  8Y7L9 PHj0r9/f0lOThYRkf3798vYsWPF19dX3n77bUlJSREvLy9Zs2aNiHBIamdulu1rr70mIu1Deurr66W vl6WLl0q999//zVDgMlUS0uL1NTUiIhIcXGxjBkzRubMmSPV1dUicu1rMi4uTv74xz9KWlqauLu7S0REhJSWllq8bntgbrbNzc3S0NAgqampEhQUJLm5uZYu2W7cLFvjsLMDBw5IQECAPPnkk5KVlSVvvvmmeHl5yYIFC6SpqYnvuZ3YtWuX6PV6EWl/jWZmZkpYWJjU1tZ2 pjLly Ls7Oz6HQ6iY6OlpKSEovVay Mw6Wjo6Nl3rx5ItKeb01NjQQGBkpiYqKIiOTl5cnYsWMlODhY0tLSZNmyZdK7d2/ZuHFjl9bERsoOdRx3f73b0tLSZNiwYbJ582aT2z/99FMJCwuTf//73ybbOzYF3VlXZPvFF1 IiMjFixfl eef1 b40I3drEnt AXA6dOnZe7cufL4449LZGSkbNu2zVpl2wVzvgAoLi6W1atXi7 /v/j5 cmePXusVbbdMTaib731lowePbrTD yysjJxdXUVnU4n3t7esmXLFgtXaZ9uJds9e/bIb3/7Wxk2bJj4 flJdna2pcu0SzfK1niQ pe//EUmTJggo0aNkoiICL7nXsfVTWVbW5uWb1JSksTGxnb6uOzsbPHw8JCAgAC 597E5cuX5YUXXpDY2FjtCyjj8dn69evFxcVFa2JPnjwps2fPlpiYGAkPD1fyumUjZWdWrFghP/vZz SFF16QI0eOSHNzs4i0v4g6/hFHRUVJXFycHDt2TNuWk5MjTk5OUlBQcM39u7uuzrbjY0nNFwCcRH5FV34B0NjYKJmZmZKenq6uYDtzq3/HxveI2tpaeeihh2Tq1Kna2TzjwdS5c ckMDBQ1q5dq6RWe9OV2V68eFFefvll fOf/6ykVnvTFdkaPweNGfPs9BXmfrY3NzfLmDFjOn1fNRgMkpOTIytXruyq8u4InR2fGrdlZGRIRESErFq1yuT2//znPxIcHHzN51xdXZ2yOjlHyk7U1NQgJiYG6enpCA8PR35 Pp5  mmkpqYCaJ wqNPptCUgX331VZSVleH3v/89iouLUVNTg zsbDz88MPw8vLS7t/dqcq242O7u7feegtPPPEE5syZg6 //hotLS0A2pfNNb4O586dC09PT2RkZJiMa3ZxccGJEyfQq1cvk flJPKuy9bDw0P7fcqUKZgxY4aF98Q23SzfjozvEe7u7njuuedQWFiIbdu2oaysDHPnzkVBQQG8vb1x8uRJbW5ad9aV2RYVFaFfv35ITU1FYmKipXfF5nRVtvPmzUNxcbE273fIkCGW3A2bZE62RlVVVdDr9YiMjARwZV5veno6Lly4AAcHB0ycOBHz58 3SP227OOPP8ann34KoPPjU O2 Ph4DBs2DJ999hn27t2r3d7c3IyzZ89qx2Hy/fxIpYt8KWvRqEvl5eVJYGCgydKYixYtEn9/f21ek/Ebo47 9a9/yahRo8TX11f8/f3F29tbdu3aZbG67QGzVae6ulomT54sgYGBsnTpUgkLC5OQkBBZvnz5Nfc1Zrxz506JjIyURx99VE6fPi3V1dWyaNEiiYmJ0cbuE7NVzZx8rycuLk68vb2lR48eEhoaKmfOnOEoAGG2KqnKlm4v261bt8rIkSOlvr5e3n//fXF1dZWHH35YvvvuOwtUbvu  uorefDBB0Wn00lcXJx2KYPO/qaNZwO/ OILefTRR2X48OHy1VdfSUVFhbz  usSHR0t586ds1jtbKRsnPEFs3nzZunXr5/JdXJKS0tl vTpMnLkSG2b8UV39TVLDh06JH// 98tVLV9YLbqsUlVh9mq9UPzFbky9ywzM1O8vb3F399fsrKyLFKzvWC26jBbdW4n29mzZ4u/v7 EhoZKnz595JNPPrFIzfagrq5OFi9eLAkJCZKWlib33HOPrF //paGTx49elR 8pOfyODBg2XIkCHi7e0tn332mQWqbsdGygZlZ2dLfn6 yTcVf/3rXyUsLOyaCzhu375d/P39tQtrck7OjTFby2CTqg6zVeuH5ttRXV2duLq6yvz589UXbEeYrTrMVp3bzbaxsVHGjx8vLi4ukpKSYpmi7Uhzc7Ps27dPPv/8cxERmTJlikycONFkDvrVOmbc1NQkp0 ftngDZcQJHDYkLy8PISEh M1vfoOpU6fipz/9KTIzMwEAsbGxuHDhAvbt26ddrA0A7rvvPkRGRmLfvn1oaWnhnJzrYLbq7d69GwcPHkRlZaWWVX19PXx8fHD8 HHtfn5 fkhISMClS5ewbt06AO3jmDuOifbw8MDo0aPx1FNPWXAvbBOzVasr8jUyGAxwc3ODXq/HypUrLbcTNorZqsNs1emqbEUELi4uePnll3H /Hm8 uqrlt0RG9QxWxGBs7MzoqOjERUVBQB47bXXUFRUhG3btuHSpUsArn29dvw869GjBwICAhATE2O5neiAR4Y2QESwadMmzJo1C/Hx8cjPz8fmzZsxcOBAbNy4EefOnUOfPn2QmJiINWvW4OjRo9pjvby84O7ujqqqKjg7O1tvJ2wUs1WPTao6zFYtFfkaJ V394VQmK06zFYdVe 5sbGx1yyW1N10lu3mzZsBtL/ DAYDQkJCMG3aNGRmZuLQoUMAbHthNH7C2oDGxkacOnUK06ZNw4IFC Dp6YlRo0Zh8uTJKCwshLu7O4ArXXrPnj2xevVqFBYWao83GAzw9PTk1ds7wWzVYZOqDrNVi/mqw2zVYbbqqMrWlhsAS7lRth9//DG   477b7GvFJSUtDU1ISsrCxUVlYCAEpKSrTnsyVspGzAXXfdhcmTJ Oll17CXXfdpb2Qhg4dira2NrS0tGjLar733nv49ttv8cQTT2Dt2rVYvHgx/vGPfyA Pp5/sJ1gtuqwSVWH2arFfNVhtuowW3WYrTo3y/buu /W7uvg4KANM33xxRexa9cufPjhh3jssccQFxeHqqoq2zses9BcLLpFbW1t2sTGhQsXykMPPSQipgsdnDhxQp599lmZPHmyjBo1Snbv3m2VWu0Ns 16 fn5UltbKyLtkz9zcnIkKChIqqqqtGyzs7MlKipKQkJC5L333pNFixZJ7969ZcuWLdYq3eYxW7WYrzrMVh1mqw6zVedm2XZGr9eLu7u76HQ6mTRpkpSXl1usXnOwkbKCq1d0MS6X2fGAvrW1VcaNGydpaWnXfR7ji5LaMVvrYJOqDrNVi/mqw2zVYbbqMFt1rpft1cduO3fuFGdnZxkxYoQcOHDA4nWaw8naZ8S6gw0bNqCpqQnDhg3DhAkToNPpICJoa2uDo6OjNsmu4wTFhoYGVFRUYMyYMQCA6upqrFu3DjNnzkS/fv0AQDvV3J0xW8sREZNT6gaDAY6Ojmhra4ODgwN0Oh0MBgM //xzbTW4jrkPHz4c69evR11dHfO9CrNVi/mqw2zVYbbqMFt1fki2Vw/XGzx4MN555x0kJSVZtPYfgnOkFNq2bRsGDRqEP/3pT0hPT0d8fDx 97vfAYB2oA8AGRkZiIyMxN69e7XHHj58GI6OjggKCsLq1avh4 ODHTt2cAWu7zFb9TZs2ID3338fubm5AKA1qQaDAQBuuUlNTU01mUzKDx1mqxrzVYfZqsNs1WG26qjINiQkxC6aKACcI6XK1q1bJSwsTFauXCkGg0EqKirko48 kh49esh///tfERE5f/68REVFiZeXl6SlpUlzc7P2 BdffFH69 8vQ4cOlf79 /PCmR0wW7W2bt0qAwcOlPvvv19Gjx4t/fr10y7Q2PGq7Zs2bZLRo0dLbm6uti0nJ0dCQkJEr9dLWlqauLq6yvjx46WystLSu2GTmK1azFcdZqsOs1WH2arDbK9gI6VIVlaWJCQkSE1Njbbt66 /lqCgIG0sbUNDgyxfvlzOnz9v8tiWlhaJj48Xd3d3SU1NtWjd9oDZqsMmVR1mqxbzVYfZqsNs1WG26jDbdmykusjBgwfl3Llz2u 1tbXS2Nhocp ysjLp27evnDlz5qbPl5eXJw0NDV1epz1itpbDJlUdZqsW81WH2arDbNVhtuow23ZspG7T7t27xc/PT 655x4ZNGiQPPfcc1JUVKTd3nGVlw0bNkhYWJgYDAaTzryjq1cu6c6YrXpsUtVhtmoxX3WYrTrMVh1mqw6zvT42UrehvLxcoqKiJDk5WYqLiyUzM1P8/f3lySeflLKyMhG5crBvHCs6c ZMmT59ujVLthvMVi02qeowW7WYrzrMVh1mqw6zVYfZ3hyXKbsNBQUFOHbsGH71q19h6NChmDJlCv7whz gsrISK1asAHBllRL5/krXX375JSZNmgQAKC8vx9NPP42CggKr1W/LmK06Z8 exZIlS/DMM88gJycH7777Lvbs2YNFixbhzJkz2v2MK 7s378fI0aMgIODA5ydnTt9Tpu70riVMFu1mK86zFYdZqsOs1WH2d4ia3dy9iwjI0MiIiKksLBQ29ba2ipvvPGGDB8 XPbu3attLywslJCQECktLZUlS5aIq6urjB071mR8KbVjturs2rVLXF1dpbi4WNuWlZUl48aNk1mzZmnbWlpaREQkNDRUPvzwQxEROXPmjEyfPl1OnTpl2aLtBLNVi/mqw2zVYbbqMFt1mO2t4Rmp2zBixAicPHnS5MyHo6MjYmNjMXjwYPzzn//Utm/fvh0FBQUIDw/Hxo0bsWPHDhw8eBA9e/a0Ruk2j9mqo9frERISon2LBAA///nP8cgjj2D//v3Iy8sDADg5OaGoqAitra0YN24ckpOTERwcjG //RY Pj7WKt mMVu1mK86zFYdZqsOs1WH2d4aNlK3YeTIkZgwYQJWrVqF vp6bft9992H/v37o7S0FK2trQAAZ2dn9OzZE2vWrEFpaSnGjRtnrbLtArNVh02qOsxWLearDrNVh9mqw2zVYba3ho3UbVqxYgXy8/PxySefoLm5Wdvu6 uLEydOwMnJCQAQHx P6upqPPPMM9Yq1e4wWzXYpKrDbNVivuowW3WYrTrMVh1me4usPbbwTvDKK6 Ij4 PrF /Xurr66W2tlYeeeQRWbZsmbVLs3vMVo2jR4 Kk5OTrF27VpqamrTtr7zyigQEBGi/2 NVxq2N2arFfNVhtuowW3WYrTrM9uacrN3I3Qlef/11/O9//0NycjI  OADVFRU4O6778aqVausXZrdY7ZqhIWFYeHChVi fDmcnZ0xdepUtLW14fDhwyZn9vr06WPFKu0Ts1WL arDbNVhtuowW3WY7c3pRL5fP5puy XLl3Hq1CkcOXIEP/rRjzjMrAsxW3XmzJmDLVu2wNfXV2tSMzMzMXz4cGuXZveYrVrMVx1mqw6zVYfZqsNsr4 NFFE3xiZVHWarFvNVh9mqw2zVYbbqMNvrYyNFRERERERkJq7aR0REREREZCY2UkRERERERGZiI0VERERERGQmNlJERERERERmYiNFRERERERkJjZSREREREREZmIjRUREREREZCY2UkRERERERGZiI0VERN3Gr3/9azz  OPWLoOIiO4ATtYugIiIqCvodLob3r506VKkpaVBRCxUERER3cnYSBER0R3h/Pnz2s9/ 9vfkJKSgsLCQm2bm5sb3NzcrFEaERHdgTi0j4iI7gheXl7aPw8PD h0OpNtbm5u1wztGz9 PJKSkjBv3jx4enpiwIAB OCDD9DQ0IAZM2bA3d0dAQEB2L59u8n/dfz4ccTExMDNzQ0DBgxAQkICKisrLbzHRERkTWykiIioW/voo4/Qt29fHDp0CElJSZg9ezZ 8YtfIDo6GkeOHMGkSZOQkJCAS5cuAQCqq6sxceJEhIeH4/Dhw9ixYwcuXLiAuLg4K 8JERFZEhspIiLq1sLCwrBkyRIEBgZi8eLFcHFxQd  fZGYmIjAwECkpKSgqqoK33zzDQBgzZo1CA8Px5tvvong4GCEh4cjPT0dubm5KCoqsvLeEBGRpXCOFBERdWv33nuv9rOjoyP69OmD0NBQbduAAQMAABcvXgQAHDt2DLm5uZ3OtyopKUFQUJDiiomIyBawkSIiom7N2dnZ5HedTmeyzbgaYFtbGwCgvr4ejz32GFJTU695Lm9vb4WVEhGRLWEjRUREZIaIiAhkZWVhyJAhcHLixygRUXfFOVJERERmmDNnDvR6PaZNm4Yvv/wSJSUl2LlzJ2bMmAGDwWDt8oiIyELYSBEREZnBx8cH fn5MBgMmDRpEkJDQzFv3jz06tULDg78WCUi6i50wku8ExERERERmYVfnREREREREZmJjRQREREREZGZ2EgRERERERGZiY0UERERERGRmdhIERERERERmYmNFBERERERkZnYSBEREREREZmJjRQREREREZGZ2EgRERERERGZiY0UERERERGRmdhIERERERERmen/coFijNJB5NIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1000x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 날짜 필터링\n",
    "filtered_df = dexcom_df[(dexcom_df.index >= '2020-02-19 22:43:26') & (dexcom_df.index <= '2020-02-20 22:59:59')]\n",
    "filtered_fl = fl_df[(fl_df['time_begin'] >= '2020-02-19 22:43:26') & (fl_df['time_begin'] <= '2020-02-20 22:59:59')]\n",
    "\n",
    "ax = filtered_df.plot(y='glucose', figsize=(10, 3), legend=True)\n",
    "print(filtered_df.iloc[0])\n",
    "\n",
    "for idx in range(len(filtered_fl)):\n",
    "    start_time = pd.Timestamp(filtered_fl.iloc[idx]['time_begin'])\n",
    "    end_time = start_time   pd.Timedelta(minutes=10)\n",
    "    plt.axvspan(start_time, end_time, color='yellow', alpha=0.3)\n",
    "\n",
    "plt.title('Glucose Value with Highlighted Interval')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Glucose Value (mg/dL)')\n",
    "\n",
    "# 그래프 출력\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}