{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "77840595-cc42-407f-ad0f-a1395079a69f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ================================================================\n",
    "# Intrusion Detection (NSL-KDD-like dataset)\n",
    "# Dataset: kdd_test.csv (41 features + target column)\n",
    "# Models: Random Forest & Gradient Boosting (multi-class)\n",
    "#\n",
    "# Fixes included:\n",
    "# 1) Robust CSV load + column-name stripping\n",
    "# 2) Safer target-column detection ('labels' / 'label' / 'class')\n",
    "# 3) Merge rare classes (<2 samples) into 'other' so stratified split works\n",
    "# 4) Force dense output for OneHotEncoder/ColumnTransformer so GradientBoosting works\n",
    "# 5) Safer feature-name extraction for both models (RF + GB)\n",
    "# ================================================================\n",
    "\n",
    "import os\n",
    "import time\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.model_selection import train_test_split, RandomizedSearchCV\n",
    "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
    "from IPython.display import display\n",
    "\n",
    "# -----------------------------\n",
    "# SETTINGS\n",
    "# -----------------------------\n",
    "RANDOM_STATE = 42\n",
    "np.random.seed(RANDOM_STATE)\n",
    "sns.set(style=\"whitegrid\")\n",
    "\n",
    "data_path = r\"C:\\Users\\paul_\\OneDrive\\Desktop\\Artificial Intelligence\\kdd_test.csv\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0bbe32c4-0e1e-4998-9697-e3ffaa0c9fb8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Exists: True\n",
      "Is file: True\n",
      "Readable: True\n",
      "\n",
      "First 5 rows:\n"
     ]
    },
    {
     "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>duration</th>\n",
       "      <th>protocol_type</th>\n",
       "      <th>service</th>\n",
       "      <th>flag</th>\n",
       "      <th>src_bytes</th>\n",
       "      <th>dst_bytes</th>\n",
       "      <th>land</th>\n",
       "      <th>wrong_fragment</th>\n",
       "      <th>urgent</th>\n",
       "      <th>hot</th>\n",
       "      <th>...</th>\n",
       "      <th>dst_host_srv_count</th>\n",
       "      <th>dst_host_same_srv_rate</th>\n",
       "      <th>dst_host_diff_srv_rate</th>\n",
       "      <th>dst_host_same_src_port_rate</th>\n",
       "      <th>dst_host_srv_diff_host_rate</th>\n",
       "      <th>dst_host_serror_rate</th>\n",
       "      <th>dst_host_srv_serror_rate</th>\n",
       "      <th>dst_host_rerror_rate</th>\n",
       "      <th>dst_host_srv_rerror_rate</th>\n",
       "      <th>labels</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>tcp</td>\n",
       "      <td>smtp</td>\n",
       "      <td>SF</td>\n",
       "      <td>2429</td>\n",
       "      <td>475</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>57</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.0</td>\n",
       "      <td>normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>udp</td>\n",
       "      <td>domain_u</td>\n",
       "      <td>SF</td>\n",
       "      <td>45</td>\n",
       "      <td>134</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>155</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>udp</td>\n",
       "      <td>domain_u</td>\n",
       "      <td>SF</td>\n",
       "      <td>45</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>255</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1979</td>\n",
       "      <td>udp</td>\n",
       "      <td>other</td>\n",
       "      <td>SF</td>\n",
       "      <td>145</td>\n",
       "      <td>105</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.84</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>14462</td>\n",
       "      <td>tcp</td>\n",
       "      <td>other</td>\n",
       "      <td>RSTR</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.68</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>portsweep</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   duration protocol_type   service  flag  src_bytes  dst_bytes  land  \\\n",
       "0         5           tcp      smtp    SF       2429        475     0   \n",
       "1         0           udp  domain_u    SF         45        134     0   \n",
       "2         0           udp  domain_u    SF         45         80     0   \n",
       "3      1979           udp     other    SF        145        105     0   \n",
       "4     14462           tcp     other  RSTR          1          0     0   \n",
       "\n",
       "   wrong_fragment  urgent  hot  ...  dst_host_srv_count  \\\n",
       "0               0       0    0  ...                  57   \n",
       "1               0       0    0  ...                 155   \n",
       "2               0       0    0  ...                 255   \n",
       "3               0       0    0  ...                   2   \n",
       "4               0       0    0  ...                   2   \n",
       "\n",
       "   dst_host_same_srv_rate  dst_host_diff_srv_rate  \\\n",
       "0                    0.11                    0.02   \n",
       "1                    0.80                    0.02   \n",
       "2                    1.00                    0.00   \n",
       "3                    0.01                    0.84   \n",
       "4                    0.01                    0.68   \n",
       "\n",
       "   dst_host_same_src_port_rate  dst_host_srv_diff_host_rate  \\\n",
       "0                         0.01                         0.04   \n",
       "1                         0.01                         0.00   \n",
       "2                         0.01                         0.00   \n",
       "3                         1.00                         0.00   \n",
       "4                         1.00                         0.00   \n",
       "\n",
       "   dst_host_serror_rate  dst_host_srv_serror_rate  dst_host_rerror_rate  \\\n",
       "0                   0.0                       0.0                  0.87   \n",
       "1                   0.0                       0.0                  0.00   \n",
       "2                   0.0                       0.0                  0.00   \n",
       "3                   0.0                       0.0                  0.00   \n",
       "4                   0.0                       0.0                  1.00   \n",
       "\n",
       "   dst_host_srv_rerror_rate     labels  \n",
       "0                       0.0     normal  \n",
       "1                       0.0     normal  \n",
       "2                       0.0     normal  \n",
       "3                       0.0     normal  \n",
       "4                       1.0  portsweep  \n",
       "\n",
       "[5 rows x 42 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Dataset shape (rows, columns): (22544, 42)\n",
      "\n",
      "Column names:\n",
      "['duration', 'protocol_type', 'service', 'flag', 'src_bytes', 'dst_bytes', 'land', 'wrong_fragment', 'urgent', 'hot', 'num_failed_logins', 'logged_in', 'num_compromised', 'root_shell', 'su_attempted', 'num_root', 'num_file_creations', 'num_shells', 'num_access_files', 'num_outbound_cmds', 'is_host_login', 'is_guest_login', 'count', 'srv_count', 'serror_rate', 'srv_serror_rate', 'rerror_rate', 'srv_rerror_rate', 'same_srv_rate', 'diff_srv_rate', 'srv_diff_host_rate', 'dst_host_count', 'dst_host_srv_count', 'dst_host_same_srv_rate', 'dst_host_diff_srv_rate', 'dst_host_same_src_port_rate', 'dst_host_srv_diff_host_rate', 'dst_host_serror_rate', 'dst_host_srv_serror_rate', 'dst_host_rerror_rate', 'dst_host_srv_rerror_rate', 'labels']\n",
      "\n",
      "Info:\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 22544 entries, 0 to 22543\n",
      "Data columns (total 42 columns):\n",
      " #   Column                       Non-Null Count  Dtype  \n",
      "---  ------                       --------------  -----  \n",
      " 0   duration                     22544 non-null  int64  \n",
      " 1   protocol_type                22544 non-null  object \n",
      " 2   service                      22544 non-null  object \n",
      " 3   flag                         22544 non-null  object \n",
      " 4   src_bytes                    22544 non-null  int64  \n",
      " 5   dst_bytes                    22544 non-null  int64  \n",
      " 6   land                         22544 non-null  int64  \n",
      " 7   wrong_fragment               22544 non-null  int64  \n",
      " 8   urgent                       22544 non-null  int64  \n",
      " 9   hot                          22544 non-null  int64  \n",
      " 10  num_failed_logins            22544 non-null  int64  \n",
      " 11  logged_in                    22544 non-null  int64  \n",
      " 12  num_compromised              22544 non-null  int64  \n",
      " 13  root_shell                   22544 non-null  int64  \n",
      " 14  su_attempted                 22544 non-null  int64  \n",
      " 15  num_root                     22544 non-null  int64  \n",
      " 16  num_file_creations           22544 non-null  int64  \n",
      " 17  num_shells                   22544 non-null  int64  \n",
      " 18  num_access_files             22544 non-null  int64  \n",
      " 19  num_outbound_cmds            22544 non-null  int64  \n",
      " 20  is_host_login                22544 non-null  int64  \n",
      " 21  is_guest_login               22544 non-null  int64  \n",
      " 22  count                        22544 non-null  int64  \n",
      " 23  srv_count                    22544 non-null  int64  \n",
      " 24  serror_rate                  22544 non-null  float64\n",
      " 25  srv_serror_rate              22544 non-null  float64\n",
      " 26  rerror_rate                  22544 non-null  float64\n",
      " 27  srv_rerror_rate              22544 non-null  float64\n",
      " 28  same_srv_rate                22544 non-null  float64\n",
      " 29  diff_srv_rate                22544 non-null  float64\n",
      " 30  srv_diff_host_rate           22544 non-null  float64\n",
      " 31  dst_host_count               22544 non-null  int64  \n",
      " 32  dst_host_srv_count           22544 non-null  int64  \n",
      " 33  dst_host_same_srv_rate       22544 non-null  float64\n",
      " 34  dst_host_diff_srv_rate       22544 non-null  float64\n",
      " 35  dst_host_same_src_port_rate  22544 non-null  float64\n",
      " 36  dst_host_srv_diff_host_rate  22544 non-null  float64\n",
      " 37  dst_host_serror_rate         22544 non-null  float64\n",
      " 38  dst_host_srv_serror_rate     22544 non-null  float64\n",
      " 39  dst_host_rerror_rate         22544 non-null  float64\n",
      " 40  dst_host_srv_rerror_rate     22544 non-null  float64\n",
      " 41  labels                       22544 non-null  object \n",
      "dtypes: float64(15), int64(23), object(4)\n",
      "memory usage: 7.2+ MB\n"
     ]
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# LOAD DATA\n",
    "# -----------------------------\n",
    "print(\"Exists:\", os.path.exists(data_path))\n",
    "print(\"Is file:\", os.path.isfile(data_path))\n",
    "print(\"Readable:\", os.access(data_path, os.R_OK))\n",
    "\n",
    "df = pd.read_csv(data_path)\n",
    "df.columns = df.columns.str.strip()  # IMPORTANT: removes hidden spaces in column names\n",
    "\n",
    "print(\"\\nFirst 5 rows:\")\n",
    "display(df.head())\n",
    "print(\"\\nDataset shape (rows, columns):\", df.shape)\n",
    "\n",
    "print(\"\\nColumn names:\")\n",
    "print(df.columns.tolist())\n",
    "\n",
    "print(\"\\nInfo:\")\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f257fb65-85bc-424d-8b24-89d74df0ac79",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Using target column: 'labels'\n"
     ]
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# TARGET COLUMN (ROBUST)\n",
    "# -----------------------------\n",
    "possible_targets = [\"labels\", \"label\", \"class\", \"target\", \"y\"]\n",
    "target_col = next((c for c in possible_targets if c in df.columns), None)\n",
    "\n",
    "if target_col is None:\n",
    "    raise ValueError(\n",
    "        f\"Could not find a target column. Expected one of {possible_targets}, \"\n",
    "        f\"but got columns: {df.columns.tolist()}\"\n",
    "    )\n",
    "\n",
    "print(f\"\\nUsing target column: '{target_col}'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "50416d57-da6d-4ac0-8877-b9ea56fbaa9d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Class distribution (original):\n",
      "labels\n",
      "normal             11245\n",
      "neptune             6654\n",
      "satan                698\n",
      "smurf                540\n",
      "ipsweep              479\n",
      "guess_passwd         411\n",
      "portsweep            371\n",
      "mscan                310\n",
      "warezmaster          279\n",
      "apache2              228\n",
      "back                 227\n",
      "processtable         211\n",
      "nmap                 206\n",
      "warezclient          107\n",
      "teardrop             104\n",
      "snmpguess             99\n",
      "mailbomb              94\n",
      "saint                 93\n",
      "snmpgetattack         43\n",
      "httptunnel            41\n",
      "pod                   35\n",
      "buffer_overflow       17\n",
      "multihop               9\n",
      "ps                     9\n",
      "sendmail               8\n",
      "xlock                  4\n",
      "loadmodule             4\n",
      "rootkit                4\n",
      "xterm                  3\n",
      "land                   2\n",
      "imap                   2\n",
      "ftp_write              2\n",
      "named                  2\n",
      "phf                    1\n",
      "perl                   1\n",
      "xsnoop                 1\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# CLASS DISTRIBUTION\n",
    "# -----------------------------\n",
    "print(\"\\nClass distribution (original):\")\n",
    "print(df[target_col].value_counts())\n",
    "\n",
    "plt.figure(figsize=(10, 4))\n",
    "df[target_col].value_counts().plot(kind=\"bar\")\n",
    "plt.title(\"Class Distribution (Original)\")\n",
    "plt.xlabel(\"Class\")\n",
    "plt.ylabel(\"Count\")\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a6477b7e-e1f4-4f4a-82af-2175f64a19a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Rare classes (<2 samples) to be merged into 'other':\n",
      "['phf', 'perl', 'xsnoop']\n",
      "\n",
      "Class distribution (after merge):\n",
      "labels\n",
      "normal             11245\n",
      "neptune             6654\n",
      "satan                698\n",
      "smurf                540\n",
      "ipsweep              479\n",
      "guess_passwd         411\n",
      "portsweep            371\n",
      "mscan                310\n",
      "warezmaster          279\n",
      "apache2              228\n",
      "back                 227\n",
      "processtable         211\n",
      "nmap                 206\n",
      "warezclient          107\n",
      "teardrop             104\n",
      "snmpguess             99\n",
      "mailbomb              94\n",
      "saint                 93\n",
      "snmpgetattack         43\n",
      "httptunnel            41\n",
      "pod                   35\n",
      "buffer_overflow       17\n",
      "ps                     9\n",
      "multihop               9\n",
      "sendmail               8\n",
      "rootkit                4\n",
      "loadmodule             4\n",
      "xlock                  4\n",
      "other                  3\n",
      "xterm                  3\n",
      "imap                   2\n",
      "land                   2\n",
      "ftp_write              2\n",
      "named                  2\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# MERGE RARE CLASSES -> 'other'\n",
    "# -----------------------------\n",
    "min_samples_per_class = 2  # must be >=2 for stratified split to work reliably\n",
    "vc = df[target_col].value_counts()\n",
    "rare_classes = vc[vc < min_samples_per_class].index\n",
    "\n",
    "print(\"\\nRare classes (<2 samples) to be merged into 'other':\")\n",
    "print(list(rare_classes))\n",
    "\n",
    "df[target_col] = df[target_col].replace(rare_classes, \"other\")\n",
    "\n",
    "print(\"\\nClass distribution (after merge):\")\n",
    "print(df[target_col].value_counts())\n",
    "\n",
    "plt.figure(figsize=(10, 4))\n",
    "df[target_col].value_counts().plot(kind=\"bar\")\n",
    "plt.title(\"Class Distribution (After Merging Rare Classes into 'other')\")\n",
    "plt.xlabel(\"Class\")\n",
    "plt.ylabel(\"Count\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "143c6b93-4d89-49d6-99ee-94659fb190a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Train set shape: (18035, 41)\n",
      "Test set shape:  (4509, 41)\n",
      "\n",
      "Train label distribution:\n",
      "labels\n",
      "normal             8996\n",
      "neptune            5323\n",
      "satan               558\n",
      "smurf               432\n",
      "ipsweep             383\n",
      "guess_passwd        329\n",
      "portsweep           297\n",
      "mscan               248\n",
      "warezmaster         223\n",
      "apache2             182\n",
      "back                182\n",
      "processtable        169\n",
      "nmap                165\n",
      "warezclient          86\n",
      "teardrop             83\n",
      "snmpguess            79\n",
      "mailbomb             75\n",
      "saint                74\n",
      "snmpgetattack        34\n",
      "httptunnel           33\n",
      "pod                  28\n",
      "buffer_overflow      14\n",
      "ps                    7\n",
      "multihop              7\n",
      "sendmail              6\n",
      "xlock                 3\n",
      "rootkit               3\n",
      "loadmodule            3\n",
      "xterm                 3\n",
      "other                 2\n",
      "named                 2\n",
      "ftp_write             2\n",
      "imap                  2\n",
      "land                  2\n",
      "Name: count, dtype: int64\n",
      "\n",
      "Test label distribution:\n",
      "labels\n",
      "normal             2249\n",
      "neptune            1331\n",
      "satan               140\n",
      "smurf               108\n",
      "ipsweep              96\n",
      "guess_passwd         82\n",
      "portsweep            74\n",
      "mscan                62\n",
      "warezmaster          56\n",
      "apache2              46\n",
      "back                 45\n",
      "processtable         42\n",
      "nmap                 41\n",
      "warezclient          21\n",
      "teardrop             21\n",
      "snmpguess            20\n",
      "saint                19\n",
      "mailbomb             19\n",
      "snmpgetattack         9\n",
      "httptunnel            8\n",
      "pod                   7\n",
      "buffer_overflow       3\n",
      "multihop              2\n",
      "ps                    2\n",
      "sendmail              2\n",
      "loadmodule            1\n",
      "xlock                 1\n",
      "rootkit               1\n",
      "other                 1\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9gAAAGACAYAAABWaMrCAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAADA7UlEQVR4nOzdeVxN+f8H8NetRGTNUtbsKSoRWQsx9hnLyJKtZN+JyCAKJULZIvtu7Ps+zNDYZQ0l2ZJ9ydJ6fn/06367bpH63BPN6/l4zGNy7u28zq3bued9PptCkiQJRERERERERJQlWtl9AEREREREREQ5AQtsIiIiIiIiIgFYYBMREREREREJwAKbiIiIiIiISAAW2EREREREREQCsMAmIiIiIiIiEoAFNhEREREREZEALLCJiIiIiIiIBGCBTURE9AOSJCm7D+GHOIbUfrTj+RnwZ0ZEJC8W2ERElO2uXbsGV1dX2NnZwdzcHM2aNcOkSZPw8OFDlef17NkTPXv2zKaj/B9/f39UrVpV5T9zc3O0aNEC3t7eePfuncrz3dzc0LRp0wzvPywsDN26dfvm87Zv346qVavi0aNHmcr5mq1bt8Lb2zvdLLkdO3YM48ePF7KvzLyW7H79mfHl75CIiDRPJ7sPgIiI/tvWr1+PGTNmoG7duhgzZgyKFy+OBw8eYPny5Th8+DBWrlwJMzOz7D7MNG3evBlAcivhx48fce3aNSxbtgwnTpzAxo0bUbhwYQDA4MGD0atXrwzv98CBA7h8+fI3n2dnZ4fNmzejePHimXsBX7F48WLUqVNHlqyMWLVqlbB9Zea1ZPfrz4wvf4dERKR5LLCJiCjbXLx4EV5eXujRowfc3d2V2+vWrYtmzZqhY8eOmDBhAnbv3p2NR5k+S0tLlX83aNAA9erVQ48ePTBnzhx4enoCAMqWLauR/CJFiqBIkSIa2Xd2ZmlaZl5LTnr9RESkOewiTkRE2SYoKAj58+fH6NGj1R4rUqQI3Nzc0KJFC8TExKT5/a9evYKHhweaNGmC6tWro06dOhgyZIhKN96HDx9i0KBBqFu3LiwsLODg4ICTJ08qH4+NjYWHhwcaN26M6tWro2XLllixYkWmX5OFhQXs7e2xc+dOfPr0CYB61+0bN26gd+/eqFWrFmrWrIk+ffogJCQEQHL384CAAABA1apV4e/vr/w6ICAAnTp1Qq1atbBo0aJ0uy1v3rxZ2d2+d+/euHnzpvKx9L6nadOmcHNzU379+PFj7NixQ/nctL7v9OnT6N69O2rVqqXsgRAVFaWSZWpqipCQEDg4OKBGjRqws7PDsmXLvutn2rNnT5w7dw7nzp1D1apVcfbsWZw9exZVq1bFpk2b0KRJE9SvXx///PMPgOSu0R07doSlpSXMzc3x66+/Yv/+/en+DNzc3NCnTx9s27YNv/zyC6pXr4727durvE8y8z0AcPnyZfTo0QOWlpaws7PD6tWr0adPH+XPOi0ZeU++efMGkydPRv369VGjRg106dIFwcHBysfT+h0SEZHmscAmIqJsIUkS/vnnH9SrVw96enppPqdly5YYOnQo9PX10/z+AQMG4PTp0xgzZgyCgoIwePBgnDlzBpMnTwYAJCUlYcCAAfj48SN8fHywaNEiFCpUCIMHD0ZkZCQAwMvLCydPnsT48eMRFBSEZs2awdvbG9u3b8/0a2vYsCHi4+Nx7do1tcdiYmLQr18/FC5cGAsWLICfnx8+ffoEZ2dnvH//Hr///js6d+4MILlQ/v3335Xfu3jxYvzyyy+YO3cumjVrlmb206dP4e/vj5EjR2Lu3Ll4+/YtevXqhVevXmX4+AMCAlCsWDHY2tqm2y16165dcHJyQokSJTB37lxMmDABly9fhoODA16+fKl8XlJSEkaOHInWrVsjMDAQtWrVgq+vL/7+++8MH8+UKVNgamoKU1NTbN68WWXIgJ+fH8aPH4/x48fD0tIS69evx+TJk9GsWTMsXboUs2fPRq5cueDq6oonT56km3H9+nUEBQVh+PDhWLhwIXR0dDB8+HC8ffs2098THh6OPn36AADmzp2LYcOGITAwEBcvXvzq6/3WezI2Nha9e/fGsWPHMGrUKAQEBMDQ0BD9+vVTFtkZ+R0SEZF47CJORETZ4vXr14iNjUXp0qUz9f3Pnj2Dnp4exo8fj9q1awNI7lr+6NEjbNq0CQDw8uVLhIeHY+DAgbC1tQUAmJubIyAgALGxsQCAc+fOoX79+mjTpo1yH3nz5lWOn86MYsWKAQBevHih9lhYWBhevXqFnj17olatWgCAChUqYNOmTYiJiYGRkREMDQ0BqHdBNzc3R//+/ZX/vnHjhtr+ExMTERAQoPzelBb1VatWpdlTIC2mpqbQ1dVFkSJF1I4BSC6aZ8+ejfr168PPz0+53crKCq1bt8aKFSvg6uoKIPlGyODBg5U3CmrVqoUjR47gr7/+QqNGjTJ0PJUqVVLeZPnyeLp27YqWLVsq//3w4UM4OTlhyJAhym2lS5dGx44dcenSJZQsWTLNjPfv32P79u3K7vx58+aFo6Mj/v33X/zyyy+Z+p6lS5dCX18fy5cvV95EqlChArp27frV1/ut9+SuXbsQGhqKLVu2wMLCAgDQuHFj9OzZE76+vti2bds3f4dERKQZLLCJiChbaGkld6JKTEzM1PeXKFECa9asAQA8efIEkZGRCA8Px6VLlxAfHw8AKFq0KCpVqoQ//vgDZ86cQePGjdGwYUNMmDBBuZ+6deti06ZNiI6ORpMmTWBra6tSnIlWuXJlFClSBIMGDUKrVq1ga2uLevXqYdy4cd/83ipVqnzzOSVLllQpqIoVKwZLS0ucOXMmwwX2t0REROD58+dq+ytbtixq1qyJs2fPqmyvWbOm8uuUou/jx49CjqVq1aoq/07pev3+/Xvcv38f9+/fV7bqprwv0lKkSBGVsfIpNzlSuvln5nv+/fdf2NraqvTQqFmzJkqVKvXV1/St92RwcDCKFSsGMzMzJCQkKLc3adIEPj4+ePv2LQoWLPjVDCIi0gwW2ERElC0KFSqEfPnyfbXb7sePHxEXF4dChQql+fju3bsxd+5cREVFoVChQjAxMUGePHmUjysUCqxYsQKLFy/GkSNHsGPHDuTKlQv29vaYOnUqChUqBHd3dxgaGmL37t3w8PAAkFwETZ48Gaamppl6bdHR0QD+V3Clli9fPqxfvx6LFy/G/v37sWnTJujp6aF9+/Zwd3dH7ty5091v0aJFv5md1nMMDAxUxkZn1Zs3b9LNKlq0qMqYbwAqvxMg+eaKqPWZDQwMVP794MEDTJ48Gf/++y90dHRQoUIFZRH+tcwvhykoFAoAya31mf2eV69eqR0f8L8eDun51nvyzZs3eP78ebqz6z9//pwFNhFRNmGBTURE2aZhw4Y4e/YsYmNj0ywst2/fDi8vL2zYsEGlFRQALly4gPHjx8PR0RHOzs7KYtbHx0dljGuJEiUwdepUTJkyBaGhoTh48CCWLVuGggULwsPDA7q6uhg0aBAGDRqEJ0+e4MSJE1i0aBHGjBmDAwcOZOp1nTlzBnnz5k23AKpQoQJmz56NxMREXL16Fbt27cLGjRtRunRplS7gmfHlGtxAcsGVMgN2eoXjhw8fMpyRcsMjrS7wz58/z1L3+qxISkpC//79kStXLmzZsgWmpqbQ0dFBWFhYtsxEb2hoqDIePcXLly9Rvnz5dL/vW+/J/Pnzw9jYGL6+vml+f2aHXRARUdZxkjMiIso2Tk5OePPmjco43hQvX77E8uXLUa5cuTTHkF6+fBlJSUkYPny4srhOTEzEmTNnACQXW5cvX0b9+vVx9epVKBQKVKtWDaNGjUKVKlXw9OlTfP78Gb/88otyhuaSJUuiR48eaNOmDZ4+fZqp13Tr1i0cPXoUnTp1SvOmwcGDB2FjY4Pnz59DW1sbNWvWxNSpU1GgQAFlZkr3+cyIjIxUTuAGAFFRUbh8+TLq1q0LAMqxzKlbtO/du6dslU7xtWMoX748ihUrhj179qhsf/jwIa5cuQIrK6tMH396MvIzef36NSIiItC5c2eYm5tDRye5HeHUqVMAvt4arQnW1tY4deqUcrw/kPz++NqM3hl5T9apUwdRUVEwMDBAjRo1lP8FBwdj+fLl0NbWBpC19xEREWUOW7CJiCjbWFpaYsSIEZg3bx7Cw8PRoUMHFC5cGHfv3sWKFSvw4cMHBAYGKltdUzM3NwcATJs2DZ06dcK7d++wbt06hIaGAkjuXm5qaoo8efJg3LhxGDZsGIoWLYozZ87g1q1b6NWrF/LkyQMzMzMEBAQgV65cqFq1KiIiIrBjx450J7ZK7cqVKwCSux5/+PAB165dw6pVq2BsbIwRI0ak+T1WVlZISkrCkCFD0L9/f+TLlw8HDhzA+/fv0aJFCwBAgQIFAAB79+6FhYUFypQpk+Gfae7cuTF48GCMGjUKiYmJmD9/PgoVKoTevXsDAGxsbKCnp4dZs2Zh5MiR+PDhAwICAtS64RcoUAA3b97EuXPnlD/rFFpaWhg9ejQmTJiAUaNG4bfffsPr168REBCAggULom/fvhk+XiB54re4uLivdskvUKAALl++jODg4HSfZ2BggFKlSmH9+vUwNDREgQIF8M8//2D16tUAvj6eWhMGDhyI/fv3o1+/fnBycsK7d+8wf/58KBSKNN/TADL0nuzYsSPWrVuHvn37YuDAgTAyMsKZM2ewbNkyODo6IleuXADUf4dfdtUnIiLxeGuTiIiy1aBBg5RF9MyZM9G/f3+sXbsWjRs3xq5du9Kd2Ktu3bqYPHkyLl++DBcXF8ycORMlS5ZUriF98eJF5M6dGytWrEDlypXh5eUFZ2dnHDt2DNOmTUPHjh0BQPn1ihUr4OTkhEWLFqFz586YOnXqN4/dwcEBDg4O6Nq1K8aMGYPDhw/D2dkZmzZtQv78+dP8nuLFi2P58uXInz8/3N3dMWDAANy4cQP+/v6wsbEBALRo0QI1atSAm5sbgoKCvuvnWbVqVXTp0gVTp07FuHHjULZsWWzYsEHZRTx//vxYsGCBssifP38+Bg0ahOrVq6vsx8nJCS9evICzszOuX7+ultOxY0csWLAAkZGRGDJkCGbNmoWaNWvizz///OYY4y95eHhg6NChX31Ojx49kCtXLri4uChbpNOyaNEilChRAm5ubhg5ciSuXLmCxYsXo0KFCrhw4cJ3HVdWlStXDkFBQYiNjcXw4cPh5+cHFxcXFCtWDPny5Uv3+771nsybNy/Wr1+PWrVqYfbs2XBxccHhw4cxZswYlQn8vvU7JCIi8RSSqFlGiIiIiDIhLi4OHTt2xN69e7P7UIQKDg5Grly5lMvIAcDbt2/RoEEDjBs3Dr169crGoyMiIk1gF3EiIiLKVgsXLkS9evWy+zCEu3HjBhYsWIDRo0fDzMwMr1+/xooVK5A/f360bds2uw+PiIg0gC3YRERElK1u376NihUrKiclyymSkpKwZMkS7Nq1C1FRUcibNy/q1KmDMWPGoFy5ctl9eEREpAEssImIiIiIiIgE4CRnRERERERERAKwwCYiIiIiIiISgAU2ERERERERkQA5azYRmVy+fBmSJCFXrlzZfShERERERESkQfHx8VAoFKhZs+Y3n8sW7EyQJAnfOzecJEmIi4v77u/7XnLlyJmV03LkzMppOXJm8TX9+DlyZuW0HDmzclqOnFk5LUfOLL6mHz9HzqycliNnVk7LyWzW99R/bMHOhJSW6xo1amT4ez5+/Ihbt26hUqVKyJs3r6YOTbYcObNyWo6cWTktR84svqYfP0fOrJyWI2dWTsuRMyun5ciZxdf04+fImZXTcuTMymk5mc26du1ahvfPFmwiIiIiIiIiAVhgExEREREREQnAApuIiIiIiIhIABbYRERERERERAKwwCYiIiIiIiISgAU2ERERERERkQAssImIiIiIiIgEYIEtE4VCAT09PSgUiuw+FCIiIiIiItIAnew+gJwmKUmClpZ6Ea2npwdTU9Pv/j4iIiIiIiL6ObDAFkxLSwHf9RfxKPp9hr+ndIn8GNujlgaPioiIiIiIiDSNBbYGPIp+j/DHb7P7MIiIiIiIiEhGHINNREREREREJAALbCIiIiIiIiIBWGATERERERERCcACm4iIiIiIiEgAFthEREREREREArDAJiIiIiIiIhKABTYRERERERGRACywiYiIiIiIiARggU1EREREREQkAAtsIiIiIiIiIgFYYBMREREREREJwAKbiIiIiIiISAAW2EREREREREQCZHuBHR8fDz8/P9jZ2aFmzZro3r07Ll26pHz81q1bcHR0hKWlJezs7BAUFKTy/UlJSViwYAEaNWoECwsLODk5ITIyUuU539oHERERERERUVZle4G9ePFibNu2DZ6enti5cycqVKgAFxcXREdH4/Xr1+jbty+MjY2xbds2DBs2DPPnz8e2bduU379o0SJs2rQJnp6e2Lx5MxQKBVxcXBAXFwcAGdoHERERERERUVbpZPcBHDt2DG3btkXDhg0BAG5ubti6dSuuXLmC+/fvQ1dXF1OnToWOjg4qVqyIyMhILFu2DJ06dUJcXBxWrFgBV1dX2NraAgD8/PzQqFEjHDlyBG3atMGWLVu+ug8iIiIiIiIiEbK9BbtQoUI4ceIEHj16hMTERGzevBm6urqoVq0aLly4AGtra+jo/O8+gI2NDSIiIvDy5UuEhobiw4cPsLGxUT5eoEABmJqa4vz58wDwzX0QERERERERiZDtBba7uzt0dHTQrFkz1KhRA35+fpg3bx7Kli2Lp0+fwtDQUOX5xYsXBwA8efIET58+BQAYGRmpPScqKgoAvrkPIiIiIiIiIhGyvYt4eHg4ChQogIULF6JEiRLYunUrxo8fj3Xr1uHz58/Q1dVVeX7u3LkBALGxsfj06RMApPmct2/fAsA395FZkiTh48ePKtsUCgX09PQyvc9Pnz5BkqRMf3/KPlL/X5PkysppOXJm5bQcObP4mn78HDmzclqOnFk5LUfOrJyWI2cWX9OPnyNnVk7LkTMrp+VkNkuSJCgUigw9N1sL7MePH8PV1RWrVq1C7dq1AQA1atRAWFgY/P39kSdPHuVkZSlSiuK8efMiT548AIC4uDjl1ynPSSl0v7WPzIqPj8etW7dUtunp6cHU1DTT+4yIiBD2prp//76Q/fxIWTktR86snJYjZxZf04+fI2dWTsuRMyun5ciZldNy5Mzia/rxc+TMymk5cmbltJzMZH3ZaJuebC2wr169ivj4eNSoUUNlu4WFBU6dOoWSJUvi2bNnKo+l/LtEiRJISEhQbitbtqzKc0xMTAAAhoaGX91HZuXKlQuVKlVS2ZbRuxrpKV++vJAW7Pv378PY2DhLrek/UlZOy5EzK6flyJnF1/Tj58iZldNy5MzKaTlyZuW0HDmz+Jp+/Bw5s3JajpxZOS0ns1lhYWEZ3n+2FtgpY6dv374Nc3Nz5fY7d+6gXLlysLS0xKZNm5CYmAhtbW0AQHBwMMqXLw8DAwPkz58f+vr6OHv2rLLAfvfuHW7evAlHR0cAgLW19Vf3kVkKhSJLLeBpEflm0tPTE3582Z2V03LkzMppOXJm8TX9+DlyZuW0HDmzclqOnFk5LUfOLL6mHz9HzqycliNnVk7L+d6s72lIzdZJzszNzVG7dm2MHz8e//77L+7fv4958+YhODgY/fv3R6dOnRATEwN3d3eEhYVh+/btWL16NQYMGAAguZne0dERvr6+OHbsGEJDQzFq1CgYGhqiefPmAPDNfRARERERERGJkK0t2FpaWli0aBHmzZuHCRMm4O3bt6hSpQpWrVoFS0tLAMDy5cvh5eWFDh06oFixYhg3bhw6dOig3Mfw4cORkJCASZMm4fPnz7C2tkZQUJCyj7yBgcE390FERERERESUVdk+i3jBggUxZcoUTJkyJc3Hzc3NsXnz5nS/X1tbG66urnB1dU33Od/aBxEREREREVFWZfs62EREREREREQ5AQtsIiIiIiIiIgFYYBMREREREREJwAKbiIiIiIiISAAW2EREREREREQCsMAmIiIiIiIiEoAFNhEREREREZEALLCJiIiIiIiIBGCBTURERERERCQAC2wiIiIiIiIiAVhgExEREREREQnAApuIiIiIiIhIABbYRERERERERAKwwCYiIiIiIiISgAU2ERERERERkQAssImIiIiIiIgEYIFNREREREREJAALbCIiIiIiIiIBWGATERERERERCcACm4iIiIiIiEgAFthEREREREREArDAJiIiIiIiIhKABTYRERERERGRACywiYiIiIiIiARggU1EREREREQkAAtsIiIiIiIiIgFYYBMREREREREJwAKbiIiIiIiISAAW2EREREREREQCsMAmIiIiIiIiEoAFNhEREREREZEALLCJiIiIiIiIBPghCuydO3eidevWqFGjBtq0aYMDBw4oH7t16xYcHR1haWkJOzs7BAUFqXxvUlISFixYgEaNGsHCwgJOTk6IjIxUec639kFERERERESUVdleYO/atQsTJ06Eg4MD9u7di9atW2P06NG4fPkyXr9+jb59+8LY2Bjbtm3DsGHDMH/+fGzbtk35/YsWLcKmTZvg6emJzZs3Q6FQwMXFBXFxcQCQoX0QERERERERZZVOdoZLkoT58+ejd+/e6N27NwBgyJAhuHTpEs6dO4dz585BV1cXU6dOhY6ODipWrIjIyEgsW7YMnTp1QlxcHFasWAFXV1fY2toCAPz8/NCoUSMcOXIEbdq0wZYtW766DyIiIiIiIiIRsrUF+969e3j8+DHatWunsj0oKAgDBgzAhQsXYG1tDR2d/90HsLGxQUREBF6+fInQ0FB8+PABNjY2yscLFCgAU1NTnD9/HgC+uQ8iIiIiIiIiEbK1Bfv+/fsAgI8fP8LZ2Rk3b95E6dKlMWjQIDRt2hRPnz5FlSpVVL6nePHiAIAnT57g6dOnAAAjIyO150RFRQHAN/dhYGCQqWOXJAkfP35U2aZQKKCnp5ep/QHAp0+fIElSpr8/ZR+p/69JcmXltBw5s3JajpxZfE0/fo6cWTktR86snJYjZ1ZOy5Ezi6/px8+RMyun5ciZldNyMpslSRIUCkWGnputBXZMTAwAYPz48Rg6dCjGjh2LQ4cOYfDgwVi5ciU+f/4MXV1dle/JnTs3ACA2Nlb5Q0nrOW/fvgWAb+4js+Lj43Hr1i2VbXp6ejA1Nc30PiMiIoS9qVJuXshBrqycliNnVk7LkTOLr+nHz5EzK6flyJmV03LkzMppOXJm8TX9+DlyZuW0HDmzclpOZrK+rCnTk60Fdq5cuQAAzs7O6NChAwCgWrVquHnzJlauXIk8efIoJytLkVIU582bF3ny5AEAxMXFKb9OeU5KS/K39pGVY69UqZLKtoze1UhP+fLlhbRg379/H8bGxllqTf+RsnJajpxZOS1Hziy+ph8/R86snJYjZ1ZOy5EzK6flyJnF1/Tj58iZldNy5MzKaTmZzQoLC8vw/rO1wDY0NAQAtS7clSpVwl9//YVSpUrh2bNnKo+l/LtEiRJISEhQbitbtqzKc0xMTJQZX9tHZikUiiwV6GkR+WbS09MTfnzZnZXTcuTMymk5cmbxNf34OXJm5bQcObNyWo6cWTktR84svqYfP0fOrJyWI2dWTsv53qzvaUjN1knOTE1NkS9fPoSEhKhsv3PnDsqWLQtra2tcvHgRiYmJyseCg4NRvnx5GBgYwMTEBPr6+jh79qzy8Xfv3uHmzZuoXbs2AHxzH0REREREREQiZGuBnSdPHvTr1w8LFy7E3r178eDBAyxevBinT59G37590alTJ8TExMDd3R1hYWHYvn07Vq9ejQEDBgBI7gfv6OgIX19fHDt2DKGhoRg1ahQMDQ3RvHlzAPjmPoiIiIiIiIhEyNYu4gAwePBg6Onpwc/PD9HR0ahYsSL8/f1Rt25dAMDy5cvh5eWFDh06oFixYhg3bpxyvDYADB8+HAkJCZg0aRI+f/4Ma2trBAUFKQehGxgYfHMfRERERERERFmV7QU2APTt2xd9+/ZN8zFzc3Ns3rw53e/V1taGq6srXF1d033Ot/ZBRERERERElFWZ6iJ+/vx5fPjwIc3H3r17h3379mXpoIiIiIiIiIh+NpkqsHv16oXw8PA0H7t58yYmTJiQpYMiIiIiIiIi+tlkuIv4+PHjERUVBQCQJAlTp06Fvr6+2vPu37+PokWLijtCIiIiIiIiop9Ahluwf/nlF0iSBEmSlNtS/p3yn5aWFiwtLTFz5kyNHCwRERERERHRjyrDLdhNmzZF06ZNAQA9e/bE1KlTUbFiRY0dGBEREREREdHPJFOziK9du1b0cRARERERERH91DJVYH/69AlLlizBiRMn8OnTJyQlJak8rlAocPToUSEHSERERERERPQzyFSB7eXlhW3btqFOnTqoVq0atLQyNRk5aYBCoYCenh4UCkV2HwoREREREdF/SqYK7MOHD2PUqFHo37+/6OOh75CUJEFLS7WQ1tPTg6mp6Xd9DxEREREREWVdpgrshIQEmJubiz4W+k5aWgr4rr+IR9HvM/T80iXyY2yPWho+KiIiIiIiov+mTBXYDRs2xKlTp2BjYyP6eOg7PYp+j/DHb7P7MIiIiIiIiP7zMlVgt27dGlOmTMGrV69gYWEBPT09tef89ttvWT02IiIiIiIiop9GpgrskSNHAgB27tyJnTt3qj2uUChYYBMREREREdF/SqYK7GPHjok+DiIiIiIiIqKfWqYK7FKlSok+DiIiIiIiIqKfWqYK7ICAgG8+Z+jQoZnZNREREREREdFPSXiBra+vj+LFi7PAJiIiIiIiov+UTBXYoaGhats+fvyIixcvYurUqfjjjz+yfGBEREREREREPxMtUTvKmzcvGjVqhCFDhsDHx0fUbomIiIiIiIh+CsIK7BRGRkYIDw8XvVsiIiIiIiKiH1qmuoinRZIkREVFYdmyZZxlnIiIiIiIiP5zMlVgm5iYQKFQpPmYJEnsIk5ERERERET/OZkqsIcMGZJmga2vrw87OzsYGxtn9biIiIiIiIiIfiqZKrCHDRsm+jiIiIiIiIiIfmqZHoMdFxeH7du34+zZs3j37h0KFy6M2rVro0OHDsidO7fIYyQiIiIiIiL64WWqwH737h169eqF0NBQlCxZEsWKFUNERAT27t2L9evXY8OGDcifP7/oYyUiIiIiIiL6YWVqma45c+bg6dOnWLduHY4fP47Nmzfj+PHjWLduHV6+fIn58+eLPk4iIiIiIiKiH1qmCuxjx45h5MiRqF27tsr22rVrY/jw4Th8+LCQgyMiIiIiIiL6WWSqwP7w4QPKlCmT5mNlypTBmzdvsnJMRERERERERD+dTBXYFSpUwIkTJ9J87NixYyhXrlyWDoqIiIiIiIjoZ5OpSc6cnZ0xevRoxMXFoV27dihatChevHiBPXv2YOvWrZg6dargwyQiIiIiIiL6sWWqBbt169YYPnw4du7ciV69eqF169bo1asXdu7ciSFDhsDBwSFTBxMREYGaNWti+/btym23bt2Co6MjLC0tYWdnh6CgIJXvSUpKwoIFC9CoUSNYWFjAyckJkZGRKs/51j6IiIiIiIiIsipTLdgfP37E4MGD4ejoiCtXruDt27eIioqCg4MDChYsmKkDiY+Px9ixY/Hx40flttevX6Nv376wt7eHh4cHrly5Ag8PDxQqVAidOnUCACxatAibNm3CzJkzUaJECcyePRsuLi7Yu3cvdHV1M7QPIiIiIiIioqz6rhbsW7du4bfffsOqVasAAAUKFEDjxo3RuHFjzJs3D927d0d4eHimDsTf3x/58uVT2bZlyxbo6upi6tSpqFixIjp16oQ+ffpg2bJlAIC4uDisWLECw4YNg62tLUxMTODn54fo6GgcOXIkQ/sgIiIiIiIiEiHDBfbDhw/Rp08fvH37FpUqVVJ5TFdXFxMnTsSHDx/QvXt3PH369LsO4vz589i8eTO8vb1Vtl+4cAHW1tbQ0flfQ7uNjQ0iIiLw8uVLhIaG4sOHD7CxsVE+XqBAAZiamuL8+fMZ2gcRERERERGRCBkusAMDA1G4cGHs2LEDLVq0UHlMT08Pjo6O2LZtG/LmzYslS5Zk+ADevXuHcePGYdKkSTAyMlJ57OnTpzA0NFTZVrx4cQDAkydPlIX8l99XvHhxREVFZWgfRERERERERCJkeAx2cHAwBg4ciEKFCqX7HAMDA/Tt2xfr16/P8AFMnToVlpaWaNeundpjnz9/hq6ursq23LlzAwBiY2Px6dMnAEjzOW/fvs3QPjJLkiSV8eIAoFAooKenl+l9fvr0CZIkZei5Wcn6npxv7Sf1/zUlp+XImZXTcuTM4mv68XPkzMppOXJm5bQcObNyWo6cWXxNP36OnFk5LUfOrJyWk9ksSZKgUCgy9NwMF9jPnz/P0PrWVapUyXAX8Z07d+LChQvYs2dPmo/nyZMHcXFxKttSiuK8efMiT548AJLHYqd8nfKclOLzW/vIrPj4eNy6dUtlm56eHkxNTTO9z4iIiAz/orOS9T05GXH//n1h+/ov5ciZldNy5Mzia/rxc+TMymk5cmbltBw5s3JajpxZfE0/fo6cWTktR86snJaTmawvG23Tk+ECu0iRInj27Nk3n/fq1auvtnKntm3bNrx8+RJ2dnYq26dMmYKgoCCULFlSLTPl3yVKlEBCQoJyW9myZVWeY2JiAgAwNDT86j4yK1euXGpj0TN6VyM95cuX/64WbDlyvubTp0+4f/8+jI2Ns9Ry/1/LkTMrp+XImcXX9OPnyJmV03LkzMppOXJm5bQcObP4mn78HDmzclqOnFk5LSezWWFhYRnef4YLbGtra2zfvh1t2rT56vN27tyJatWqZWifvr6++Pz5s8q2Fi1aYPjw4WjdujX27duHTZs2ITExEdra2gCSu6qXL18eBgYGyJ8/P/T19XH27Fllgf3u3TvcvHkTjo6OyuP+2j4yS6FQZKkFPC2afjNpKkdPT0/4z+K/kCNnVk7LkTOLr+nHz5EzK6flyJmV03LkzMppOXJm8TX9+DlyZuW0HDmzclrO92Z9T+Nmhic569mzJ86ePYtZs2alOXY5Li4O3t7e+Pvvv9GjR48M7bNEiRIoV66cyn9A8ljuUqVKoVOnToiJiYG7uzvCwsKwfft2rF69GgMGDACQ3Ezv6OgIX19fHDt2DKGhoRg1ahQMDQ3RvHlzAPjmPoiIiIiIiIhEyHALdo0aNTBhwgTMmDEDu3btQr169VC6dGkkJibiyZMnOHv2LF6/fo0RI0agUaNGQg7OwMAAy5cvh5eXFzp06IBixYph3Lhx6NChg/I5w4cPR0JCAiZNmoTPnz/D2toaQUFByj7yGdkHERERERERUVZluMAGgB49esDExARBQUE4duyYsiU7X758aNiwIZycnGBhYZGlA7p9+7bKv83NzbF58+Z0n6+trQ1XV1e4urqm+5xv7YOIiIiIiIgoq76rwAaAWrVqoVatWgCA169fQ0tLCwULFhR+YEREREREREQ/k+8usFMrXLiwqOMgIiIiIiIi+qlleJIzIiIiIiIiIkofC2wiIiIiIiIiAVhgExEREREREQnAApuIiIiIiIhIABbYRERERERERAKwwCYiIiIiIiISgAU2ERERERERkQAssImIiIiIiIgEYIFNREREREREJAALbCIiIiIiIiIBWGATERERERERCcACm4iIiIiIiEgAFthEREREREREArDAJiIiIiIiIhKABTYRERERERGRACywiYiIiIiIiARggU1EREREREQkAAtsIiIiIiIiIgFYYBMREREREREJwAKbiIiIiIiISAAW2EREREREREQCsMAmIiIiIiIiEoAFNhEREREREZEALLCJiIiIiIiIBGCBTURERERERCQAC2wiIiIiIiIiAVhgExEREREREQnAApuIiIiIiIhIABbYRERERERERAJke4H95s0bTJ48GY0bN4aVlRW6deuGCxcuKB+/desWHB0dYWlpCTs7OwQFBal8f1JSEhYsWIBGjRrBwsICTk5OiIyMVHnOt/ZBRERERERElFXZXmCPHj0aISEhmDt3Lv7880+YmZnB2dkZ4eHheP36Nfr27QtjY2Ns27YNw4YNw/z587Ft2zbl9y9atAibNm2Cp6cnNm/eDIVCARcXF8TFxQFAhvZBRERERERElFU62RkeGRmJ06dPY+PGjbCysgIAuLu749SpU9i7dy/y5MkDXV1dTJ06FTo6OqhYsSIiIyOxbNkydOrUCXFxcVixYgVcXV1ha2sLAPDz80OjRo1w5MgRtGnTBlu2bPnqPoiIiIiIiIhEyNYW7MKFCyMwMBDVq1dXblMoFJAkCW/fvsWFCxdgbW0NHZ3/3QewsbFBREQEXr58idDQUHz48AE2NjbKxwsUKABTU1OcP38eAL65DyIiIiIiIiIRsrXALlCgAGxtbaGrq6vcduDAATx48AANGzbE06dPYWhoqPI9xYsXBwA8efIET58+BQAYGRmpPScqKgoAvrkPIiIiIiIiIhGytYv4ly5evIiJEyeiWbNmaNq0KWbOnKlSfANA7ty5AQCxsbH49OkTAKT5nLdv3wIAPn/+/NV9ZJYkSfj48aPKNoVCAT09vUzv89OnT5AkKUPPzUrW9+R8az+p/68pOS1HzqycliNnFl/Tj58jZ1ZOy5EzK6flyJmV03LkzOJr+vFz5MzKaTlyZuW0nMxmSZIEhUKRoef+MAX20aNHMXbsWFhYWGDu3LkAgDx58ignK0uRUhTnzZsXefLkAQDExcUpv055Tkrx+a19ZFZ8fDxu3bqlsk1PTw+mpqaZ3mdERESGf9FZyfqenIy4f/++sH39l3LkzMppOXJm8TX9+DlyZuW0HDmzclqOnFk5LUfOLL6mHz9HzqycliNnVk7LyUzWl4226fkhCux169bBy8sLzZs3h6+vr/LgDQ0N8ezZM5Xnpvy7RIkSSEhIUG4rW7asynNMTEwytI/MypUrFypVqqSyLaN3NdJTvnz572rBliPnaz59+oT79+/D2Ng4Sy33/7UcObNyWo6cWXxNP36OnFk5LUfOrJyWI2dWTsuRM4uv6cfPkTMrp+XImZXTcjKbFRYWluH9Z3uBvWHDBkyfPh09e/bExIkToaX1v2Hh1tbW2LRpExITE6GtrQ0ACA4ORvny5WFgYID8+fNDX18fZ8+eVRbY7969w82bN+Ho6JihfWSWQqHIUgt4WjT9ZtJUjp6envCfxX8hR86snJYjZxZf04+fI2dWTsuRMyun5ciZldNy5Mzia/rxc+TMymk5cmbltJzvzfqexs1sneQsIiICM2bMQPPmzTFgwAC8fPkSz58/x/Pnz/H+/Xt06tQJMTExcHd3R1hYGLZv347Vq1djwIABAJKb6R0dHeHr64tjx44hNDQUo0aNgqGhIZo3bw4A39wHERERERERkQjZ2oJ96NAhxMfH48iRIzhy5IjKYx06dMCsWbOwfPlyeHl5oUOHDihWrBjGjRuHDh06KJ83fPhwJCQkYNKkSfj8+TOsra0RFBSk7GZuYGDwzX0QERERERERZVW2FtgDBw7EwIEDv/occ3NzbN68Od3HtbW14erqCldX10zvg4iIiIiIiCirsrWLOBEREREREVFOwQKbiIiIiIiISAAW2EREREREREQCsMAmIiIiIiIiEoAFNhEREREREZEALLCJiIiIiIiIBGCBTURERERERCQAC2wiIiIiIiIiAVhgExEREREREQnAApuIiIiIiIhIABbYRERERERERAKwwCYiIiIiIiISgAU2ERERERERkQAssImIiIiIiIgEYIFNREREREREJAALbCIiIiIiIiIBWGATERERERERCcACm4iIiIiIiEgAFthEREREREREArDAJiIiIiIiIhKABTYRERERERGRACywiYiIiIiIiARggU1EREREREQkAAtsIiIiIiIiIgFYYFOmKRQK6OnpQaFQZPehEBERERERZTud7D4A+vElJUnQ0lIvovX09GBqavpd30NERERERJRTscCmb9LSUsB3/UU8in6foeeXLpEfY3vU0vBRERERERER/VhYYFOGPIp+j/DHb7Mlm13RiYiIiIjoZ8ACm34YmemK/rXv+14s5ImIiIiIKCtYYNMP43u7ogOZ746eVlEuVyEPsJgnIiIiIsqJWGDTD0WuruhyjSvP7gniWMgTEREREcmHBTb9Z8lRzMs5QVx2tsqzkCciIiIi+g8V2ElJSQgICMDWrVvx7t071KpVC1OmTEG5cuWy+9Aoh/svtMrnxO71ct404A0KIiIiopzhP1NgL1q0CJs2bcLMmTNRokQJzJ49Gy4uLti7dy90dXWz+/CIhMhJrfJydq+X66bBf2XIAG8YEBER0X/Vf6LAjouLw4oVK+Dq6gpbW1sAgJ+fHxo1aoQjR46gTZs22XyERD+XnFTIy5nFmxNZyyIiIiL60f0nCuzQ0FB8+PABNjY2ym0FChSAqakpzp8/zwKb6Acl5/rrcmXx5kTms9KS01rlc+LQBPZoICKi/xKFJElSdh+Eph0+fBjDhg1DSEgI8uTJo9w+YsQIfP78GUuXLv2u/V26dAmSJCFXrlxqjykUCryNiUNCYlKG96ejrYWC+rr43l/F92bltBw5s370HDmzclqOnFk5LUfOrMwWZ3LlyJmV03Iyk5XePhITE6Gtra3xYl6urJyWI2cWX9OPnyNnVk7LkTMrp+VkNis+Ph4KhQJWVlbffO5/ogX706dPAKA21jp37tx4+/b7W5JSfhHp/UIK6mduTHdm3kyZycppOXJm/cg5cmbltBw5s3JajtxZP2qOnFk5LUdUlkKhgJaWloCj+XGyclqOnFl8TT9+jpxZOS1HzqyclpPZLIVCkeHPqv9EgZ3Sah0XF6fSgh0bGws9Pb3v3l/NmjWFHRsRERERERHlDPLcJshmRkZGAIBnz56pbH/27BkMDQ2z45CIiIiIiIgoh/lPFNgmJibQ19fH2bNnldvevXuHmzdvonbt2tl4ZERERERERJRT/Ce6iOvq6sLR0RG+vr4oUqQISpUqhdmzZ8PQ0BDNmzfP7sMjIiIiIiKiHOA/UWADwPDhw5GQkIBJkybh8+fPsLa2RlBQkNrEZ0RERERERESZ8Z9YpouIiIiIiIhI0/4TY7CJiIiIiIiINI0FNhEREREREZEALLCJiIiIiIiIBGCBTURERERERCQAC2wiIiIiIiIiAVhgExEREREREQnAApuIiIiIiIhIABbYRERERERERAKwwKYfgq2tLaKjowEAAQEB+PTpUzYfEf0oAgMDle+N/4rY2Fjh+4yLi8O9e/eQkJCA+Ph44fuXy5MnTyBJktr2hIQEXL16VXjex48fcePGDYSGhmrk9/Jf9OLFi+w+BPp/48ePx5EjR3LcZ25O+bt98OBBmtvj4+Ph5+cnNEvuc2tOMmjQIISHh2f3YdAPRCe7DyCn6dWrV4afu2bNGmG5nz9/xsGDBxEeHg5nZ2fcuXMHlSpVQpEiRYRlAMDjx48REhKCuLg4tcd+++23TO/39evXiI6ORokSJbBw4UJ0794denp6WTjSjEtKSkJwcDDu3LkDLS0tmJmZoXbt2lneb0BAQIafO3To0CznpXblyhWsXbsWd+7cgba2NkxNTdG3b19UrlxZaA6QfLEcHx+v9sFcsmRJIfsPDAzEL7/8ImRf6Tl//jy2bduGN2/eoFGjRnBwcICOzv9Oj2/fvsWwYcOE/s1u3LgR3bp1U9t+4cIFuLu749ChQ0JyJEnCnDlzsHbtWsTHx+PQoUPw8/ND7ty5MW3aNOTKlUtIDgDExMRg//79Kn9LLVu2RO7cuYVlNGvWDKdPn1Y7tz169Ag9e/ZESEiIkJz4+HjMmDED27ZtU76/9fT00KtXL4waNUpIRoq4uDisWLECrVq1Qrly5eDu7o79+/fDysoKvr6+KFy4sNC8Z8+eYePGjbh79y50dXVRpUoVdO/eHQUKFBCWUa1atXR/T+3atcPly5eF5Bw/fhxNmzZV2/7gwQNMmDAB69evF5KjaTt37szwc7PyWfulz58/Y+LEiYiLi0O9evXQrFkzNG3aFAYGBsIyUjx58gQFChSAvr4+/v33Xxw+fBhWVlZo27atsAxN/t2eP38+w8+1trbOUlaK3r17Y926dShVqpRy2/Xr1+Hm5oZHjx4JPRfJdW5N8ezZM2zZsgX37t2Du7s7zp07hypVqqBixYpCc16+fJnmTRZR1yhA8ue2yM+5L02YMCHDz505c6aQzJMnT2L58uWIiIjA5s2bsW3bNpQtW1bo+SeFpq7DsxMLbMFSnwTl8uLFC3Tt2hUvXrxAXFwcunTpghUrVuDatWtYs2aNsJPVtm3bMHnyZCQmJqo9plAosvRHZ21tDUdHRxQtWhSSJKFTp07Q0kq7g8WxY8cynfOlZ8+eoV+/frhz5w4KFCgASZLw/v17WFlZYenSpcifP3+m9719+3aVf0dFRSFXrlwoU6YMdHR08ODBA8THx6N69epCC+zjx49j6NChMDc3R/369ZGUlIQrV66gY8eOWLlypbCT1pUrVzB+/Hi1O+ySJEGhUODWrVtCciwsLHD8+HH07dtXyP6+lPLzqlu3LgDA09MTu3fvxtKlS1GoUCEAyRdt33NxlRFeXl7IlSsXOnfuDCC51drX1xfr1q2DjY2NsJy1a9di165dmDJlCqZNmwYAsLe3h4eHBwwMDDB27FghOeHh4ejduzc+fPgAY2NjJCUlYcuWLVi0aBFWr14NQ0PDTO97/fr1WLFiBQCke3549+6d0AumuXPn4uDBg5g4cSIsLS0hSRIuXboEf39/6OnpYeDAgcKyfH19sWvXLjRq1AinT5/Gjh07MHz4cJw4cQI+Pj7CLpgA4ObNm+jevTuKFCkCMzMzfP78GRs2bMCqVauwatUqmJiYZHrff/75J3bv3g0g+fc0ZMgQtRs4z549E1rIjxgxAgsXLkTjxo2V21atWoV58+YJ+Txu2rQpFApFhp6blc8mNzc3lX8rFAplcaijo4P3799DW1sbhQsXFnqBO3/+fCQmJuLChQs4efIkVq1ahSlTpsDCwgL29vZwdnYWknPkyBGMGjUKS5YsQbly5dCvXz+UKVMG27dvx9u3b9GjRw8hOZr8u+3Zs6fy9/I1oj//evbsiQ0bNsDAwAD+/v4ICgqClZXVd93ET092nFsBIDIyEl26dIG+vj6io6MxatQoHDhwABMnTlS+vqw6deoUJkyYgFevXqlsF32NAgAdOnSAr68vhgwZgnLlykFXV1fYvoHkmxwpJEnChQsXULRoUZiamkJHRwehoaGIjo5Gs2bNhOSdPn0aQ4cORZs2bRASEoKkpCQkJiZi4sSJSExMRKdOnYTkAJq9DgfkO4d/SSF960xBP7yxY8ciJiYGfn5+qF+/Pnbv3o0CBQpg9OjR0NbWRmBgoJAce3t71KtXD+PHj4e+vr6QfaZ49+4ddu7ciXfv3iEgIAB9+/ZFvnz50nyuyGJ06NChiIqKwuzZs1GhQgUAwN27dzF+/HhUrVpV2IXt6tWrceLECcyZM0fZMvDu3TuMGzcOVapUwejRo4XkAEC7du1gZ2eHMWPGqGz39vbGpUuXsHnzZiE5nTt3hpaWFgYMGJDmCbBOnTpCcoYNG4ajR4+iQIECMDY2VrtLnNVW5U6dOuGXX35B//79AQAhISEYOnQoDAwMsG7dOujr6+PFixdo1KiR0A/kw4cPY+zYsZg2bRpKly6NCRMm4N27d3B1dVUW3SK0adMGI0eORPPmzVGzZk3s3r0bZcqUwdGjR+Hl5YUTJ04Iyenbty90dHTg6+uLggULAgBevXqFsWPHIm/evFm6GPz06ROCgoIgSRIWLlyY5vkhX758aNGihbCbnPXr14eXlxeaNGmisv3IkSPw8vLCX3/9JSQHABo3boyZM2eiQYMGmDZtGu7du4dVq1bh2rVr6N+/P4KDg4VlOTg4oEqVKpg6dSq0tbUBJLegu7m54dmzZ1i3bl2m9/3mzRt4e3sDAHbs2IFWrVohT548Ks/Jly8ffvvtN1SvXj3zLyKVNWvWYO7cuVi0aBFKlSoFNzc3XLt2DS4uLhg0aFCWL3T9/f0zfHEm6rNp//79WLZsGWbOnKm84REREYEJEyagTZs26Nmzp5CcLyUkJODKlSvKGyWSJAk753Xo0AGNGzfGiBEjsGzZMmzbtg0HDx7EgQMHEBAQgAMHDgjJ0eTf7ePHjzP8XFHnoaSkJLi6uuLq1avInTs3oqOjMXbsWDg4OAjZf3acW4HkLtVFihSBp6cnrKyssHv3bpQsWRJubm6IiorK0nkoxS+//IIKFSqge/fuabYui7pGAZKLuCdPnqR7rhB57TB37lw8fvwYM2fOVJ7fEhMTMXnyZCgUCnh6emY5o2vXrmjZsiX69Omjct0QFBSEHTt2YO/evVnOSKHp6/DU5/C3b99i/fr1aNKkCWrWrAkdHR1cu3YNhw8fhpOTk9jeaRJp1MuXL6ULFy5I586dk86dOyedPXtW+vvvv6WAgABhGQ0aNJBu3LghSZIkWVpaSg8ePJAkSZJu3bolWVtbC8upUaOGdO/ePWH7S0+TJk2k9+/fazxHkiSpZs2a0rVr19S2X7lyRejPrn79+tKtW7fUtt++fVuqU6eOsBxJkiRzc3MpIiJCbXtERIRkbm4uLKd69erSzZs3he0vPW5ubl/9L6tS/82kCA8Pl2xsbCRHR0cpLi5Oev78uWRiYpLlrC8dPXpUMjc3l6pVqyaNGDFCevHihfAMc3Nz6eHDh5Ikqb7WBw8eSNWrVxeWY2lpKd25c0dt+61bt6SaNWsKy/H395c+fvwobH/pqVWrlhQWFqa2PTw8XLK0tBSaVaNGDSkqKkqSJElq0aKFFBgYKEmSJD18+FAjWeHh4Wrb7969K9WoUUNYjpubm2zn8fXr10sWFhaShYWF1KVLlzTfhz+TJk2aSFeuXFHbfvXqValBgwZCs0JCQqSlS5dKTk5OkqWlpWRhYSH17t1bWrJkiRQSEiIsp0aNGtKjR48kSZIkR0dHydPTU5IkSXr8+LHQ952cf7dySUpKksaNGydVq1ZN6O/kS3KdWyVJkurWrSvdvXtXkiTVz6WwsDDJyspKSIaFhUWa5zpN2L59+1f/E8na2jrd97ioz1pLS0spMjJS+XXq6waR15GSJN91uCRJ0uDBg5Wfr6mtWbNG6tWrl9AsdhHXoH379mHixImIjY1VditKuYtSqlQpDBkyREjOhw8f0h2vnJCQICQDAExMTBAZGYny5csL22daPnz4gMjISJiZmWk0BwAKFCiQ5oRP2traKmNwsyouLg4fP35U2/7y5UthGSnMzMwQHBwMY2Njle3Xr18XOrbJyMhIlsmyRHaPTUuRIkXw8OFDlClTRrmtQoUKyrv548aNU+u+KUqzZs2wYMECjBgxAi1atNDIuMdSpUrh6tWrKF26tMr2kydPqrzmrCpevDiePn2qNs4/JiZGaJfgoUOH4tOnTwgJCUlz7L+osY+dOnXC/Pnz4evrq2wlkCQJq1atQvv27YVkpChbtiyuXbuGV69eITIyEo0aNQIAHD16VO33llXly5fH3bt3lS0FKSIjI4W2UKX83Wp6jgYA6N69O3LlygUPDw+MHDlSI3NNpAgNDcXq1asRERGB+fPn4+jRo6hUqZJyiIkIb968SbPFLSkpCZ8/fxaWAwBdunSBlpYW7O3tsWjRItSqVUt491Yg+bP2/fv3iImJwZUrV+Dk5AQgeax8ylAcETT5d9usWTP8+eefKFy48De7nWalq2la420VCgV0dHQwZswYlWFeWf18PH/+vLIlr27durh+/Xq6zxV1bgWSW1yTkpLUtsfExCh71mSVjY0Nbty4oXau04QOHTpoPCOFjo4Onjx5onY9Fx4ejrx58wrJyJ8/P6Kjo1G2bFmV7Xfv3lX2UBNFrutwILnr+7hx49S2N27cGL6+vkKzWGBr0JIlS9C2bVu4uLgox0U/e/YMHh4eGDZsmLAca2trrF+/HpMmTVJui4+Px8KFC4WMY0nh5OQEDw8PPHz4EBUqVFD7EBZ18i1atCjev38vZF/fMnz4cEyePBmzZ89WdsV7+PAhvLy8hN0AAZK7D/3xxx+YPHkyqlevDkmScPHiRUyfPh3t2rUTlgMA7du3x+zZsxEREYE6deoou8CsXr0aDg4OKpPpZGUs3+DBgzFjxgx4eHigQoUKQifL+lLKZCgRERGYOHGi0MlQWrdujUmTJmHs2LFo2LChshhMmWBq1KhRiIqKynIOkHyTKq2LMkmSMGbMGJVu/aK6lDk7O8PDwwPR0dGQJAnBwcHYtGkT1q5d+10Tp3zL+PHj4eHhATc3N5X3nYeHB3r37o0nT54on5uV4uqvv/6Cq6srYmJi1Io2kePqXrx4gRMnTqBp06YwNzeHjo4Obt68icePH8PCwkJlQsusDlPo168fRo8eDS0tLdjY2MDExAQLFy7EwoULMWPGjKy+FJX5A9q0aYPJkyfj+fPnqFWrFrS0tHDjxg3MmTNH6OdSSEgIxo0bp5E5Gr5W3AwcOBBFixZV/lvkmLrr16+jW7dusLS0xPXr1xEXF4dbt25hxowZCAgIUOuWnFl169bFtGnT4OPjo7zBEh4eDg8PD9jZ2QnJSDFhwgQEBwfj9OnTuHTpEurWrQsbGxvY2NgIvQFna2uLyZMnQ19fH/r6+mjQoAHOnDmDqVOnCn1Nmvy77dChg3LIQ4cOHTI8dOB7pR5vm5qFhcVXH8+Mnj174vTp0zAwMPjqGHPRY5YbNmyIxYsXqxQ1r1+/xuzZs4XNQeLh4YHOnTvjn3/+QenSpdV+X6Inlz158iSCgoJw7949jU4K1rZtW7i7u2PkyJEq15P+/v7o3r27kIx27drBy8sLXl5eUCgU+PDhA06ePInp06ejdevWQjJSyHUdDiQ3BJw5cwblypVT2X706FHhc2hxDLYG1ahRA7t27UKFChXQu3dvODs7o3Hjxjh8+DCWLFmiNglWZoWHh6NHjx4oXrw47t27h7p16+LevXt4//491q1bl6VJa1L72n5EnnxnzpyJTZs2wdbWFuXKlVO7ky/ypGhra4uXL18iMTER+fLlg46ODt6+favS2yBFVl5fTEwMRowYgdOnTyv3K0kSWrZsCR8fH6EtBhn9fYu4yH327Fmak94B4grELydDOXDgAGbPno2///5byGQosbGxmD59Onbt2oWlS5eifv36Ko8fP34c48ePR0xMTJZf0/bt2zN8USbyjvjmzZuxePFiPH36FABgYGCAfv36CZ04LvX7LvVrTPmISd2LJys/x7Zt26JMmTIYMWJEmmP/RX1Iyj1ra2hoKB49eoTGjRtDV1cXp06dgo6Ojtr7MTNSbux86+Ne5Hlck3M0ZMe4aADo06cPLCwsMGrUKJVxid7e3jh37hy2bdsmJCc6OhrOzs4IDw9X3vB79+4dzM3NERgYKLwFCUhuUQwJCcGZM2dw9uxZXLlyBcWLFxd2g+Lz58+YN28eHj58CBcXF1haWsLf3x+RkZHw8PBId86V75Udsy3LLTY2VtiM1Y8fP0bJkiWhUCi+OcZcZAESHR2NXr164c2bN3j//j0qVKiAx48fo1ChQmqzpmeWp6cn1q1bhyJFiqjNBaFQKITefDt9+jQGDhyINm3aYP/+/di3bx/+/PNPLFu2DNOnTxc6KVhcXBw8PT2xY8cOJCQkQJIk5M6dG46Ojhg7dqyQGz/x8fFwc3PDvn37APzv89vOzg7z5s1T+3lmhVzX4UDytdD06dPRpk0b1KhRQ3lz4siRI/D19UWrVq2ytP/UWGBrUO3atbFjxw6UKVMGU6ZMQdmyZeHs7IwnT56gXbt2uHjxorCsZ8+eYcOGDbh16xaSkpJQuXJldO/eXWj3QrlOvmktuZJC9Elxx44dGX6uiIInIiICt2/fhkKhgKmpqdAWArl962cnqkCUYzIUIPkCUKFQpHnh8u7dO5w6dUrocjJAclfQN2/eKLvz79+/H/Xq1RO+LFOKV69eQZIkjXRFP3fuXIafm5XiqkaNGtixYwcqVaqU6X386OLj4xEaGooKFSoIKTyyY3KmGjVqYMuWLahWrZqQ/X1NUlIS3rx5o1xe6PLly6hevbrwXjW1a9fG1q1bUb58eZUC+8GDB/j111+FLT0GJBe8Z86cwd27dwEkL3tmY2OjsVbT+Ph4XLp0Cf/88w9Onz6N0NBQVK1a9bs+I/9rQkNDcefOHWVXZ0mSEBcXh5CQECE9T4Dkz6XJkyejfPnyGDRoEIDkgqRRo0aYPHmy0JvzDx48UOsSDCS/NwICAoQvT/jp0yfs3btX5br1119/FTaJbu3atTFx4kR07NhRyP6+Rs5JwVJ8+PABERERAICKFStqZGnbyMhI5e+nSpUqqFSpUpqFb1bIfR2+b98+rF27VnktXq1aNfTv3x+2trZZ3ndq7CKuQebm5ti0aRNcXV1RqVIlnDhxAs7OzggLCxP+wV+8eHGMHDlS6D6/9LULL5Hjwo4fPy5sX98i17gZOzs72Nvbw97eHs2bNxc2xuhrnjx5gvDwcFhbW+PDhw/Ciyq5fnaXL1/GunXrVE7o2traGDhwILp06SIsJ/Ud2fPnzyM8PBxt27bF06dPUa5cOeHF9dWrV+Hi4oKOHTti/PjxAIDZs2cjPj4eK1euzNI40m8tKXbv3j3l16KGdoickfVrjI2N1ZZd0ZSnT59i/fr1uH37NnR0dFC5cmU4ODgIX7ImKipK2eWvatWq6NSpE8LCwlCwYEGsWrUqy0VqdiwfKdccDZGRkXB2dkbz5s2Vf0cDBgxAsWLFsHz5chgZGQnLypUrF2JiYtS2P3nyRPjFrba2NkqVKoWYmBjkypULJUqU0EhxvWrVKpw5cwbnz59HYmIi6tSpg86dO6Np06ZZWlovLVeuXMHatWtx584daGtrw8zMDH369BE+Zl6Ov9s1a9Yoi+jUvUMUCoXQ9XtnzJiBkJAQlc+6SZMmwdfXF35+fsr3vAhyrrkNAHp6evj999+F7jM1bW1toePGv+b27dvw8fFR296iRQssWLBAeN7nz59x5MgRhIeHw9nZGdeuXUOlSpXU1jDPrGbNmmHbtm0oV66cSnfq6OhotG/fHmfPnhWSA8g7fh1IHibVpk0bjeewwNagIUOGwNnZGUWKFEHHjh0REBCANm3aICoqSugYhk+fPmHVqlW4ePFimpPJZHV8YIq3b99i8eLFuH37trJbsCRJiI+Px927d4W2yMtJjnEzLi4uOHXqFAYOHIjcuXOjcePGaNasGRo3bixsUooUcXFxGD9+PA4cOAAtLS0cOnQI3t7eeP/+PQICArK8pmBqx48fV3k/pOSHhIRg9erVQjLkmAwl9T779euHK1euQKFQoEGDBvD19cX9+/exatUqoRecPj4+aNGihcpFy5EjRzB58mTMnDlTuTZpZmTHeq1xcXHYvHlzmu+HlGUwRHB1dcX06dMxatSoNOeCEHURfefOHTg6OiJPnjwwNzdHYmIitm/fjvXr12Pjxo1Ci4KZM2fi/fv3KFKkCA4dOoTHjx9jw4YN+PPPPzF79uwsvRe+9PDhQ/j6+uLu3buIjY1Ve1xUDyG55mjw8vJCpUqVVNZrPnjwICZMmICZM2cKvbi1t7fHnDlz4Ofnp9wWHh4OLy8voeOI4+LiMHbsWBw5ckSlcGvSpAnmzZsntNUyMDAQtra28Pb2RsOGDYV/HqU4fvw4hg4dCnNzc9SvXx9JSUm4cuUKOnbsiJUrVworSuX6u123bh0GDBiAIUOGoEmTJti+fTvevHmDMWPGCFuLGEj+uQUEBMDS0lK5rXnz5ihcuDBGjRoltMDW9Jrbqce/f4uI69auXbti9erVcHd311jPjxRyTgr24sULdO3aFS9evEBcXJxyjqdr165hzZo1mZ6bZv/+/fj7778BJPd6mjZtmlqPvsePH2vkZynHtWSKixcvplsviRxSxAJbg2rVqoVDhw4hLi4OhQsXxoYNG7Bx40YYGRkJXcfSw8MD+/fvR+PGjVGsWDFh+/3StGnTcPr0aTRs2BD79+9HmzZtEB4ejps3bwpdxzm9iaBSiJxo4/Tp0xg6dCjatGmDK1euICkpCYmJiZg4cSISExOFjZvp0aMHevTogdjYWPz777/466+/4OPjg3HjxqFevXpYunSpkBwAWLx4sXKm24EDBwJI/mCbOHEiZs+ejWnTpgnJ8fPzw9KlS1G8eHE8f/4cJUqUwIsXL5CYmCj07qAck6GkmDt3LoDkQjdlxtlx48Zh7Nix8PHxUT4uwo0bN1TWsQSSZwft379/lru0iRxGkVEzZszA9u3bYWZmhpCQENSsWRORkZF4+fIl+vTpIywnZb3ywYMHq431FnnDwMfHBzY2NiqzEcfGxsLV1RW+vr5C/2b//fdfrF69GqVLl4afnx8aN24MKysrFC5cWHj3xnHjxuH58+do1aqVsHGcaVmwYAGePXuW7o1KUb+nS5cuYevWrSoTmxUpUgRjx45Fjx49hGSkGD9+PPr164f69etDkiR07NgRMTExMDExSXNm2szy8/PD1atXsWjRIlhbWyMxMRHnz5+Hp6cn/P39VSZDzKqUeUFiYmKUvevKlCkjrJtuCj8/Pzg7O6sdu7e3N2bPno3NmzcLyZHr7/bJkyfo3LkzdHV1YWJigmvXrsHe3h5ubm6YNWuWsHPehw8f0rwpXrhwYeGTwc6dOxeurq7o2bOncs3tyZMnC1tzW+6eNM+fP8eePXtw8OBBlC1bVm1GalGNT4C8k4LNmjULlSpVwp49e5RzdHh7e2P06NHw9vZGYGBgpvZbs2ZNbNq0SVl0PnnyROXGqEKhQN68eeHt7Z31F5GKXNeSQPINxblz56JgwYJqw68UCgUL7J9JiRIllF9XrFhRZaZvUY4cOQJvb2+hg/PT8s8//8DHxwe2trYIDQ2Fs7MzTExM8McffyAsLExYzowZM1QunBMSEnD//n3s2LFD+HJJKRcrffr0waFDhwAAo0aNQoECBbBy5UqhE1MAyV2W9PX1UbRoUZQuXRpPnz7FpUuXhGbs27cPU6dOVVk2pk6dOpg+fTpcXV2FFdi7du3CH3/8gR49esDOzg4bNmxA3rx5MWTIEKFjy93c3NCrVy/Ur18fsbGxGDRokHIyFNEn+hMnTmDOnDlqS3ZNmTJFebNCFH19fTx48EDtZ/X06dMsTyCS1oVMbGwsIiIioKurizJlyghvUTx69ChmzZqF1q1bo0WLFpg+fTrKlCmDUaNGCe0qLPKi6GsuXryIzZs3q9wAyZ07NwYPHgxHR0ehWfHx8cpWjuDgYIwYMQJA8thi0cuU3Lp1C+vXr9f4MogiZyT/Gh0dHbx+/Vpt+chPnz4Jz9LX18emTZsQHByMmzdvKsclNmrUCFpaWsJy9u7dC09PT5Uxgfb29tDW1oaHh4fQAluhUMDb2xvr1q1TTpikq6sLBwcHTJw4UVhr1YMHD9L8PHVwcMCGDRuEZADy/d3my5dPuQyqsbExwsLCYG9vj4oVK37XnAffUrNmTSxduhQzZ85U9tiSJAmrV69GjRo1hOUAgJaWFnx9feHm5oY9e/Zg06ZNMDc3F7Z/uSeVkyRJ+NCu9IwcORJPnz5Vvsc7dOignBRMdNf6f//9F4GBgSrDUgoWLAhXV9fv6iXwJSMjI+Xna8+ePREQEKCRCRW/JNe1JJDc82TQoEHKz1hNYoGtQXJ1xdPS0oKpqamQfX3Nhw8fUKVKFQDJNwtCQ0NhYmICR0dHZauSCOm12JiYmGDXrl1C16CVa9zMsmXL8O+//+LSpUtITEyEhYUF6tWrh1GjRgn9AAOQZjclIPnk+e7dO2E5L168UF4AmpiY4OrVq2jZsiVGjRoFd3d3YSewEiVKYOfOnSqToXTr1k3oZCgpXr16lWYvEH19feEX7L/88gumTp0KDw8PmJubQ6FQ4Nq1a5g2bRqaN28uLCcxMRHe3t7YtGmTskuUnp4e+vXrp9YKnBVv3rxRdmOsUqUKbt68iQoVKmDAgAEYOXKksJuLqcd6x8XFaWTNXiD5AjouLk5te1rbssrU1BRbt25F8eLF8fr1a9ja2iIuLg7Lli0TtgpEivLly+Pjx49C95kWucbV2drawtPTE35+fsqxgg8fPsSMGTOU64mLVq9ePdSrV08j+waSh6p8uYwMkPy7Ez3/wNKlS7Ft2zaMHz8etWvXRlJSEs6fP4+FCxeiRIkS6Nevn5AcMzMzBAcHKyd0THH9+nUhyy2mkOvvtnbt2liyZAkmT54MExMTbNmyBf3798eFCxeEzYgOAKNHj0bPnj1x4cIFmJmZQaFQ4MaNG3jz5o2QoSNyrrmd2rfmCRExdrpevXqwtbWVpUjMlSsX5syZg+HDh6tNCibahw8f0p3zIeWmT1atXbtWyH4yQq5rSSB5qKvoZdPSwwJbg+TqiteiRQvs2LFD45OcGRkZ4fHjxzAyMoKxsTFCQ0MBJE9U8fbtW41mA8nrEv/xxx9C9ynXuJk5c+ZAS0sL9vb2GDBggEZbjypWrIgzZ86oTQC2d+9eoSf7ggUL4sOHDwCAcuXKKXsxlCxZEtHR0cJyJkyYAHd3d7XJUN68eYMhQ4Zg4cKFwrJq1KiB/fv3Y8CAASrb16xZI/wm1pgxY/Dw4UM4OTmpFLnNmzcX2t10wYIF2L9/P/744w+Ym5sjKSkJFy5cQEBAABITEzF8+HAhOUWLFsXLly9RsmRJlC1bFnfu3AGQ3JXxxYsXQjJSbNy4EcuWLcPTp09x6NAhLF++HMWKFRPavcvGxgY+Pj5YsGABChUqBCD5Boyvr6/woQnjx4/HwIED8fr1a7i4uMDQ0BBTp07F0aNHERQUJDRrypQpmDp1Knr27InSpUurtbxm5cI25W9VX1//q8slKRQKYbMsjx8/Hk5OTmjZsqXKklZmZmbCezxFRERg2rRpyvF7XxLV7b1KlSo4ePCgWq+Z/fv3q7XUZ9XmzZsxZcoUla6YpqamKFKkCPz9/YUV2O3bt8fs2bMRERGBOnXqQEdHB9euXcPq1avh4OCAnTt3Kp+blQtguf5uR44cib59+2Ljxo3o1q0bFi9ejDp16uDTp08q8wFkVfXq1bF3715s3rwZd+7cgY6ODtq2batcmjWr5FxzO7W05glRKBRQKBTQ0tLC9evXs5zh6ekJMzMzWQrsFPnz54eFhYVKN2tA3NwgQPI5ev369So3rePj47Fw4cIsLVtarVo1/PPPPzAwMJB1qKZc15JA8tDda9eupXkDUzQW2BokV1e8AgUKYMWKFTh58mSak/6IuuvYsmVLjBs3TjnGaeTIkbC0tMTRo0dlebPu27dP+IlSrnEzO3fuRHBwMM6cOQNHR0fo6+vDxsZG+Z/IsUnDhg3DyJEjcefOHSQmJmLHjh24d+8eDh8+rDI5T1bVq1cPPj4+8PT0RPXq1bFkyRJ0794dhw4dyvJMlhcvXsTDhw8BJP/szMzM1Fqrw8PDcebMmSzlfGn06NHo27cvLl++jISEBCxevBhhYWG4efOm8EJHT08PS5cuRUREhPLCqWLFimotPFm1detWzJw5U6W7abVq1VC8eHFMnz5dWIFta2uLKVOmYObMmbCysoKXlxeaN2+O/fv3C50cbs+ePZgzZw569+6N5cuXA0i+qeTr64vcuXPDxcVFSM7YsWPRtWtXNGnSBMbGxlAoFIiIiECBAgWELQ2XwtzcHKdPn8b79++VRWLv3r0xYsQI4Uu23b17F2FhYXB3d1d7LKtj2B89eqSckFBTF+ZfKlKkCLZt24bg4GDl31GlSpVQr1494ZPxTJ06FU+ePMHYsWOFThb5pUGDBmHw4MEIDQ2FlZUVFAoFLly4oFyrVaSXL1+m2dXYwsICUVFRwnKmTp0KIPlm5ZfDPFKfWxUKRZYKbLn+bitXroyjR4/i48ePyJcvH7Zu3Yrdu3fDyMgILVu2FJYDJA/5ETnPTWoZbakUfZP0yx6cKUMB582bJ+wGs7GxMW7fvi20h0R6rl69ipEjR6r9zYieGwRIvqnYo0cPnDt3DvHx8Zg6dSru3buH9+/fZ+k9PmPGDOV5Tc7u/Jq8lvxSq1atMG3aNFy/fj3Neklk6zbXwdagDh06YOLEiRpfJuBbE6aJ6uoRFxeH2bNnw9zcHO3atcPUqVOxadMm5M+fHwsWLBDWZa5p06Zqkxd9+PAB7969w6hRo4R2R4+Pj4ebmxv27dsH4H/LbdjZ2WH+/Pka6XkQHx+PK1eu4NChQ9iyZQsSEhJw8+ZNoRmnTp3C0qVLleMEK1euDBcXF/zyyy/CMqKjozFgwAD89ttv6N69O7p27YqbN29CoVDAzc0NvXv3zvS+L126hO7duwOA2l3uFHnz5oWTk5PQVksgeW3TFStWqPzsnJyclHf05fDkyRNhd7xr166NzZs3q11khIWF4ffffxe2du/79+8xfvx4NGjQAN27d8eAAQNw6tQp6OjowNvbW9hkJR06dECvXr3QoUMHlfVGd+zYgcWLFwubrVySJHz8+BG7du3C3bt3IUkSqlSpgnbt2mmsuEpreTjRY+VtbW1hZ2eHXr16pTnWPzuW9NKUz58/Z3k+g9TMzc2xevVq1KxZU9g+03P06FEEBgbizp07yvees7Oz8OLt119/RdeuXdGtWzeV7Rs2bMDq1auVc5P8TD58+CDr362mnThxAkuWLFEuO5Yya77IoURA8o3X06dPqxU1jx49Qrt27YSu856elMn8du3aleV9ubu7Y8eOHTAxMYGxsbHa9ZzIIrJTp07Q1dWFs7Oz8iZpaqKXsXz27Bk2bNigsoZ49+7dUbp0aSH7nz59Onr37p3mcEPR0rqWTLkhkdVryS99bciV6BshLLA16MqVKxrripcdnjx5AkNDQ5XX8ebNG+jr6+PmzZvCxhL7+/urtTzkypULVlZWGvuZPXjwQGXSGk2Mm0lKSsLly5dx5swZnD59GteuXUOpUqXQpEmTr3an/NHFxsYid+7c+Pz5M/7++2+UKFFC6LhyExMT/PPPPyqzBOcEjx49gre3t9qyd3FxcXj16pWwmy6enp549eoVZs2apXK3dsKECdDR0cH06dOF5KTl5s2bKFq0qJCujCksLS2xZ88elClTRqXAfvjwIdq0aYOrV68KyenYsSNmzJghfAx0WmJiYuDs7IyQkBAoFAocPnwYXl5eGlkeztLSEnv37hV2IfY1CQkJePnypdr7OyQkRFhLgZzLR9ra2mLZsmXKuUhygp07d2Ly5Mno2bOnSmv5+vXr4erqKnxCv5zi+vXrmDp1Ku7evZvm+G5RF+pHjx7FsGHD0Lx5c5Ux8idOnIC/v3+WlwT7888/sXv3bgDAuXPnULNmTbWbes+ePcOnT59w8uTJLGVlxIMHD9C2bVsh53G5Gp+A5Jtv27dv18i1Y3aoVasWdu3aJcvnRApNX0vKjV3ENUiTXfFSSxnjkR5RLWHNmjVTu7tZqFAhREZGomfPnggJCRGSI9fss6lFR0fjzZs3ypaj+Ph4oS1HgwYNwvnz5/Hx40dYWFigadOm8PLy0ljXpZRluiIiIjB//nwcPXoUlSpVUplZPKuSkpLg7++PYsWKoXv37siTJw8WL16M5s2bCz0ppoz1T+3Vq1fCuw4ByRfmO3bswPXr1/H582e1lnORd7w9PT0RERGBVq1aISgoCE5OToiIiMCRI0eyPNN76plEExMTcfHiRZw/fx41atSAtrY2bt68iadPnwpdrxVILnYiIyNVJnX88OEDIiMjhd0cK1q0KO7du6c2u+ilS5eEFvKPHz/W2JrAX5o7dy4UCoUsy8M1btwY//77Lzp37ixsn2kJDg6Gq6srXr58qfZYnjx5hBXYci0fCSRfsM+dOxezZ8/WeGvolStXsHbtWty5cwfa2towMzNDnz59hK6/DiR3iXzz5g2WL1+u7KptYGCA4cOH/5TF9atXr+Dj45PuOVzU5LLu7u7InTs3JkyYoNE5dhYuXIihQ4diyJAhym19+vRBQEAAFi9enOVzuL29vcpNKENDQ7VeH1WqVBE+MdSX162SJOH9+/dYvHixsCGHck7UZWhoiM+fP8uS9ebNGwQGBqY7gbKIlTbs7Oywbt06DB06VPhksumJiYlRfl6kDK0V2ZvvW0RnscDWoICAAHTq1CndrniifNml+ktZKeTXr1+vnKlSkiR06tRJrSX+3bt3wv8ATp48iTt37qR58hDZJfjLlqMGDRrA19dXeMuRjo4O3N3dYWtrq5HCMLXr16+jW7dusLS0xPXr1xEXF4dbt25hxowZCAgIQJMmTYTkzJs3D1u3blVpAW3fvj0CAwOhpaWlNlFYZr1//x4+Pj5wdHREpUqV4OTkhHPnzsHY2BiBgYFCl3Hw9vbGqlWrULVq1TS7eYl04cIFLF68GNbW1jh16hTs7e1hbm4OPz8/nDx5Um2Suu/xZTffL7t5ie6uBiS3hE2ZMgVxcXFqF7Uibyg6ODjAw8NDOYHVvXv38Pfff2P+/PlC19t2cXGBu7s7nJ2dUbZsWbVzuMhznpzLw9WpUwdeXl74+++/Ub58ebVlwESdX+fOnYvq1aujZ8+eGDp0KHx9ffHkyRMsWLBA6I0quZaPBJI/l65cuYK6devCwMBAbfyeqOLt+PHjGDp0KMzNzVG/fn0kJSXhypUr6NixI1auXKkys7MIffr0QZ8+ffDq1StIkgQDAwOh+5fTH3/8gStXrqB169Yandzq/v37+PPPP4Xf8PhSeHg45s2bp7a9bdu2WLZsWZb3X6hQIZW/x5RJCjUtretWSZKQL18+zJkzR1jO58+fcfDgQdy7dw9OTk64c+cOKlWqJPw6bNCgQfD09MSMGTNQvnx54fM/pObq6oqrV6+iQYMGGuvZ9+TJE+zbtw+rV6+GgYGB2k0kUec6IHlI44QJE9RWSNDE+HW5eg4CLLA16u3bt3BxcdF4F4sv71alTBaxcuXKNFvPv0fHjh3x+vVrSJKEhQsXomXLlmpLUOTLlw8tWrTIUk5qnp6eWLduHYoWLap2ASN6IXi5Wo78/f3x+fNn5c2J8PBw/PXXX7C0tEStWrWEZKTw9fWFk5MTRo0apRwr6Onpifz58wstsHfv3o05c+agfv36ym19+vRB+fLl4eHhIazAnjFjBi5cuIA+ffrg+PHjuHTpEnx8fLBv3z74+PjA399fSA6QXCTOmDEj3aXiRIqNjVWeGypUqIDbt2/D3Nwcv/322ze7tn2L3OuNAsk3XH799Vf06dNHoy06Li4ueP/+PVxdXREbG4sBAwZAR0cHXbt2FfaeA6CcTOr8+fNqc0KI/tCXc3m4FStWoHDhwrh27RquXbum8pjI8+vt27exdetWVK1aFaampsibNy969uyJvHnzIigoCPb29kJy5Fo+EgDq1q0rtBdQevz8/ODs7Ky23rW3tzdmz56NzZs3Z2n/31oi6d69e8qvf7ahbGfOnEFgYKDGj7tGjRp4/Pixxgvs4sWL4/79+2qtuvfv3xfeiyLlc+Pvv/9WjveuXLkybGxslGtwi7J69eo0hwJWqVJF2DJnL168QNeuXfHixQvExcXh999/x4oVK5Sz12e1O/eXM21LkpTuXCMiPy8uXLiApUuXauRGeYoGDRqgQYMGGtt/al5eXjA3N0f37t01eu0AaLbn4JdYYGuQXF3x0vojq1+/PkqWLIklS5bAzs4u0/vW09NTXnApFAo4Ozunu/6eKHv27IGHhwccHBw0mgPI13J0/vx5DBkyBPPnz0elSpXw+++/Q0tLCx8/fsScOXPQqlUrYVnXr1/HlClT1LZ369YNmzZtEpbz5s0bGBkZqW0vV66c0BlHT548iYULF6JixYpYsWIFGjRogHbt2qFKlSrCuzDGxsbKcgENAGXKlMGdO3eUy96lfAAnJSUpl6wQ5dWrV4iIiFDO8Jx6LGzqrodZ8fbtWzg5OQmfBT0to0ePxqBBgxAWFgZJklChQgXhrS4iutlllJzLwx0/flzo/tKjra2t/J0YGxvjzp07qFevHmxsbODt7S0sR87lI0VPqJieBw8eoFOnTmrbHRwcsGHDhizvP60lktIi+kaSHPT09NK8WSXa9OnTMXDgQFy9ejXNOXZEdalu27YtPDw8MGXKFOXN+IsXL2LatGnCJ7x79+4dnJ2dce3aNRQoUABJSUmIiYmBmZkZVq5cKbRXlxyfs7NmzUKlSpWwZ88eZUOAt7c3Ro8eDR8fHwQGBmZp/zNmzNBoS3V6SpQoIXSt9bTUrl0b1tbWwm+spCU6OhqLFy9GhQoVNJ6lyZ6DX2KBrUFydcVLT6VKlYR2dxg6dCgSEhIQHR2t0UlrdHR0NHpnLjW5Wo7mzp2LZs2aoUaNGti2bRv09fVx+PBhbNu2DUuXLhVaYOfKlQsxMTFq2588eSL05oiJiQm2bt2qtqTGrl27hN7V//jxo7KQP3PmDPr27Qsg+UIq5X0oSqNGjXDixAlZxh527NgR48aNw6xZs2Bra4uePXuiZMmSOH36NKpWrSosZ9++fZg4cSJiY2OVF9UpFwWlSpUSVmC3aNECJ0+e1EiBff78edSsWRM6Ojpptr6lLgJEtVzJdQ4C5F0eLoWmW6lMTExw5MgRZa+Wixcvonfv3nj69KmwDED+5SMvXryoXAf7yzV8Rf0tmZmZITg4WO1v6fr160Lm7Uj5DPrymiQn+O2337B8+XJMnz5do8XPwYMHERkZiUWLFqk9ltWlxlIbNGgQ7ty5gwEDBihfjyRJsLW1VevhkFXe3t74/Pkzdu/erewVEhoaCldXV8yZMwceHh5Z2v/3TOYqohfWv//+i8DAQJXrnoIFC8LV1VVlnpLMSt3TbefOnWjdurVar8uPHz9iy5YtWc5Kbfz48Zg2bRpGjRqV5s0dEcOX+vXrh3z58sHW1hb29vZo2LChxuYksbGxwY0bN2QpsDXZc/BLOe/s+gORqyteWmJiYrBq1SqUKFFC2D7lmrTG0dERixcvhqenp9rJSjS5Wo5u3rwJHx8f6Ovr459//oGdnR3y5MkDOzs7oa05QPKkJXPmzFFZ8zo8PBxeXl5Z6s3wpWHDhsHFxQWXLl2CpaUlFAoFrl27hitXrmDhwoXCcipWrIi//voLRkZGiIqKQuPGjQEAW7ZsEXKxGRAQoPy6cOHCmDVrFi5dugRjY2O1Dy6Rf7P9+vWDjo4OFAoFzM3NMXToUCxevBhGRkaYPXu2sJwlS5agbdu2cHFxQZcuXbBixQo8e/YMHh4eQicUdHV1RZs2bXD48GGUKVNG7eI2KxdMPXv2xOnTp2FgYPDV1restrjJfQGYwsrKCps3b8aKFStQrlw5XLlyBZUrV4a7u7vw5eHevXsHJycnXL9+XaOtVC4uLhg6dCh0dXXRpk0bLFiwAP3798ft27dhY2MjJANIPg99/vwZUVFRaNeuHVq1aoWRI0cql48UKTAwEHPnzkXBggXVWpBEFtjt27fH7NmzERERgTp16kBHR0fZrdXBwQE7d+5UPjczn7u9evXCvn37YGhoiF69eiEgIEDjc05oUupiKSEhAZcuXcLJkydRrlw5tXO4qJ4pa9aswYgRI9C3b1+NzrGTO3duLFq0COHh4bh9+zYAoGrVqhqZIPXYsWPw9/dXmSU/ZT6DUaNGZbnAfvTokfJrSZJw4cIFFC1aFKamptDR0UFoaCiio6OFTb754cOHdBsVEhISsrz/V69eKSc2mzBhAipXrozChQurPOfWrVuYO3eu0PlBgORJlFMaG1KIHL7077//4u+//8apU6fg4eGBmJgY2NjYwN7eHk2aNBE69tvDwwOdO3fGP//8g9KlS6tdO4i87pKz5yCX6dKgiIgIlC9fXuM5X44DSaFQKDB9+vQ0u5plxu+//668wE1r0hpRY+ru378PBwcHfPz4EcWKFVN7bSInV7h06RL69u2LevXq4fTp02jXrp1Ky5Gobkw2NjZYv349ypQpg7p162LGjBlo1aoVQkJCMHjwYJw+fVpIDpB8c6Vfv34ICQmBJEnInz8/YmJiYGJigpUrV6JQoULCskJCQrBmzRrcuXMHOjo6qFixIvr16yd0aaOTJ09i2LBhiI+PR5s2beDr64uZM2di/fr1WLhwIWxtbbO0/6ZNm2boeQqFQuh7Ty41atTArl27UKFCBfTu3RvOzs5o3LgxDh8+jCVLlmD79u1CcsaMGYNDhw6hWrVqaV5wZmVG15QuwFpaWnj8+PFXn5uVdZxT38H+1gWg6OJNLu7u7rh69SrmzJmj1kplZWWV5Yvo1G7cuAFtbW2YmJjgwoULCAwMRKlSpTBixAhh5yG5lo8Ekod9derUCSNGjBC2z7Rk9PyZ2Yvp2rVrY+zYsWjUqBGaNWuGbdu2qRUGKeSawTcr3NzcMtxaLerGmJWVFXbv3i3rMkaaVrt2bWzevFmteA8PD0fHjh2FrRQDJPfqe/z4MWbOnKlsSElMTMTkyZOhUCjg6emZ5Yz+/fujbNmymDRpknJJR0NDQ4wePRqfPn3C8uXLs7T/nTt3Kt97qXuGpXydUl7Z2tpi6dKlWX49KVq0aIFKlSrBwcEhzRsImuh9dfXqVaxfvx579uyBQqHAjRs3hO07Zd6lIkWKqF07iL7uWr58OZYtW4ZZs2Yp65lhw4bh9OnT+PTpk9BhlCywNahhw4ZYtGiRxtdx27Fjh9q2XLlywdLSUujJ39zcXDlpTY8ePTB06FDUq1cP27Ztw59//omNGzcKyenatStevnyJFi1apHnyEN3yHxoaiqCgINy6dQtJSUmoXLkynJychLYcjRgxArGxsShYsCAOHz6Mv//+Gw8fPsTkyZNRpkwZocvwpAgODlZZ27tRo0Zqd/N/Fq9fv0Z0dLTywjMkJAT6+voaW+ZMLjdv3sSqVatw9+5d6OrqokqVKnBxcVGb9TsrateujR07dqBMmTKYMmUKypYtC2dnZzx58gTt2rUTtk5wzZo1MW/evCzf8PhRyHEBmCJ1L4q0iDzn2djYwN/fX60r/blz5zBq1ChhN/vevXuH2bNnq8z+f/bsWRgbG2PZsmXCZv+vVq2a2vKRABAZGYn27dsLLQosLCywe/dujXQ9l5OXlxfWrl371aJUE5P55SSTJ09GqVKlhE6smCK9RpO0iPz99O7dG5UrV8akSZNUtk+fPh03btwQWnzUqVMHGzduVPsMv3fvHjp37oxLly5lOSM8PBw9evRA8eLFce/ePdStWxf37t3D+/fvsW7dOiENAefPn0dSUhJ69+4Nf39/lZnrFQoF8ubNiypVqghd9tXCwgJ79+4VuoJKWp4/f46zZ8/i33//xdmzZ/Hw4UOUK1cO9evXT3Oen8yqXbs2Jk6cKMvksgCwatUqGBsbw87ODsuWLcOSJUtgZGQEHx8foT1X2UVcg3R1dWUZ46RQKNId+7Fq1SphXVPkmrTm5s2b2LJli9BW0PRs27YNTZs2FdolNy1TpkzBlClTcPv2bcyePRv6+vrYtWsXdHR0vqtbakY4OjrC3t4ezZo1Q7169YTu+0t79uyBtbU1DA0NsWjRIuzfvx9WVlbKNUJFKVy4sEoLS8rND9HrFkqShAULFijX9gaSx1k1b94cgwYNEpYDJI8nd3Z2hqWlJerWrYvExERcunQJbdu2RWBgoLButObm5ti0aRNcXV1RqVIlnDhxAs7OzggLCxP6oZ8vXz6hNwZS+57xcqK6gG7atAkbN25UOa9qa2vD2dkZnTt3Flpgf9mLICEhAa9evUKuXLmUKwGIkpCQkOYSNQYGBmnO3ZBZM2fOVJv9f/bs2UJm/8+u5SNr1aqFa9eu/fQFtru7Ozp16oR3796hV69eaoXBzyb1+NfU3efTImoom4GBARYuXIgjR46kOcdOVlrKU0+e9fbt23R/N6In8Rs5ciR69eqFkJAQWFlZQaFQ4MKFCwgNDRWyJFhqOjo6ePLkSZqt5aLG+lasWBG7d+/Ghg0bYGRkhKSkJLRq1Qrdu3cX1viUcqNyzZo10NPTw8qVK5VzW1SqVAm9e/cW+jkLAJaWlrh9+7ZGC+xWrVrh/v37MDIyQu3atTFo0CDUr19f2NK1qWlra8u6WkHqmsjFxQUuLi4ayWGBrUHt27dHv3798Ouvv6JcuXJqXR+ycqLPyNiPmzdvCh37IdekNWXKlEFcXJzQfabHz88PkydPhqWlJezt7WFvb6+Rk1aRIkXULijHjBkj/MQLJBdUf/75J7y9vVGpUiU0a9ZMOcGaSIsWLcKSJUuwatUqREVFYcGCBfj9999x9uxZ+Pr6ZnmJuBRyrlvo5+cny9reQPIF2MCBA9W6m86YMQM+Pj7Cum4PGTIEzs7OKFKkCDp27IiAgAC0adMGUVFRQifXGzBgAObNmwcvLy/hM3pnpdt3ZslxAZgirZm9Y2JiMH78eOGz7ZqZmWHjxo1qrVQbNmxAtWrVhOVocvb/7Fg+Eki+6Jw2bRquX7+OChUqqN3UFlW8ySHlBvbo0aNhbm4udL4Wubm5uaFRo0YwMDCAm5tbus8TOfnYhQsXlDd6RV8DpW7Ja9q0KebMmaNyo02SJCxduhQLFy4UOra3Zs2ayptX//zzDyRJQpUqVTBp0iRYWloKywGSZ0d3d3fHyJEjUb16dUiShIsXL8Lf3195czurAgIC4OzsjJEjR6psj4mJgZeXl7BrFADQ0tJC9+7dUaVKFTRs2FB5w7x79+5YvXq10OVYu3TpgsmTJ+Py5cswNjZWu44U8R7PnTs3tLS0ULhwYRQvXhwlSpQQOrwwta5du2L16tVwd3fXyMSE37rplprIczi7iGvQ11pgs9r1Kr2xH6mlzDQpauzH8ePHMXToUEyaNAlNmjTBL7/8AhsbG9y+fRsWFhbCxiSePXsW3t7eGDFiRJp3hkW3Sly9ehWnTp3CX3/9hRs3bqBixYrKolRk9/6UibOKFCmCnTt34sCBA7CyskL//v01clJ59OgRTp48iZMnT+LcuXMoWLAgmjVrhsmTJwvZf7NmzTB27Fi0atUK3t7euHz5MjZt2oQLFy5g1KhR+Pvvv4XkDBw4EBEREWjZsmWa6xaKXFbBzs4OM2bMUFnbG0guFjw8PIQucWRubo49e/aotYZFRETg119/xdWrV4VlRUdHIy4uDmXKlEF4eDg2btwIIyMj9OrVS9hNnr59++LChQuQJAkGBgZqf7eix6/HxcUpC5yoqKg0l4zLihkzZuDgwYPpXgCKnCAuPSmzB584cULYPi9fvoxevXrBxMQkzVYqUb1eLC0tcfDgQRgaGqJJkybo27cvevXqhQcPHuC3334T0gUU+N9FdFrDiV68eCF0Mh5NfqZnl9q1a2Pbtm0/fat8TuXq6ooDBw5g8ODBGDRoEB49egRXV1fcvn0bw4YNg5OTU3YfYqbExcXB09MTO3bsQEJCAiRJQu7cueHo6IixY8dm+pooPDwcr169AoB0e2fcuXMHPj4+QoePdO/eHSYmJmrXVx4eHggLC8vSHCRfkus89Pr1awQHByM4OBhnzpzBs2fPYGFhARsbG6HDliZOnIg9e/agYMGCKFu2rNq1Q1Z7pWl6Tot098cC++eVHWM/5Ji0xszMTNlSmfokK8eYsCdPnmD+/PnYs2cPJEkSlrVp0yZ4eHhgxYoVMDAwwK+//op69erh5s2bcHR01NiM8tHR0QgODsbJkydx6NAhABDW4lujRg0cOXIEhoaG+PXXX9G8eXMMHToUjx8/RuvWrYV9eNWuXVu5bmGHDh3g4eGhXLcwLCxM6IzllpaW2LFjh9rkhPfv30f79u2FFr3du3dHhw4d8Pvvv6ts37lzJ9asWSOsBXvChAlwd3dXa1V+8+YN3N3dhf385BpH/PLlSwwfPhy1atXC6NGjASSvqVqtWjXMnz9fWFdXTV0Afo+bN2+iW7duQi8EgeSbiitWrMDdu3eVrVR9+vQR2krVqVMn/P777zAyMsKAAQNw8OBBGBsbw9fXF2fPnsXWrVuF5KQ3BvvRo0do164dLl++LCQnp3J2dkbDhg3VZiT+WfXq1QsLFy5E/vz5Vba/fPkSzs7O39Wa9aUnT55k+LkiGwL27dsHDw8PlC1bFhERETAzM4Onp6dGhuScOHECS5YsUenm7OzsjObNmwvPApJn+o6IiACQ3KU7q0uJ/vXXXxg4cOA313nv1KkTvLy8spSVmoWFBXbs2KG21FR4eDg6d+7805+H7t27h82bN2Pjxo2Ij48Xeh3+rWGSIlfskBO7iMvg3r17uH37NnLlyoWKFSsKm1k89dgPKysrjY/3fvfuHTZt2gRHR0ckJibC399fOWmNk5OTsAJ75cqVattevnyJwoULC5+kKykpCVevXlVO5JByEqxXr57Q8curV6/GpEmTUK9ePcyfPx+VK1fGihUrcOrUKUydOlVogb1//37lpBQPHjyAkZER6tWrBx8fH6GvydDQEBEREYiLi8Pt27cxdepUAMld50SO05Fz3UJNr+2d+uKuVq1a8PT0xL1791CrVi1oaWnhxo0bWLlyZZaX+7l48SIePnyozDQzM1MrsMPDw3HmzJks5aSmyWUHU/Py8kJCQgJ+/fVX5baVK1fCw8MDPj4+wi6arly5gkmTJmH8+PFCLwDT8uVFvyRJeP/+PTZv3ix8DDaQ3Hti3rx5wveb2vDhw5Wz/7dt2xbGxsYqs/9nxZ9//ondu3cDSP5ZDRkyRO1G8rNnz37qpafkkjdvXvj4+GDJkiUwNjZWmztD1JwGmnTy5EnlUqjnzp3D4sWL1YZxREZGfnMFgm9p2rRptkw+1rhxY9StWxdHjhyBjo4O2rVrp5Hi+ujRoxg2bBiaN2+ONm3aICkpCefPn8eIESPg7+8vbPmsFB8/fsSePXuUxXzlypXRunXrLA0xsrOzw/Hjx5GUlAR7e3ts3bpV5eZbSuPT99wsyYjChQvj5cuXagX2y5cvNb7cbGqi5qV58+YNgoODcfr0aZw5cwZPnz5F9erVMXDgwAyvvJJRP2sB/S1swdaguLg4jB07FocPH1ZuUygUaNKkCebNmyf0j+7KlStYu3Yt7ty5A21tbZiamqJv375CioIUEyZMwIULF7BkyRLcu3cPo0ePxowZM7Bv3z7kypUrS5PWpPbl7LPOzs7KQj4wMFDoGGkrKyt8/vwZtWrVQu3atWFjY4OaNWsKPyHWqFEDhw8fhpGREbp06YI6depg7NixePLkCVq2bCm0ZdTExARaWlqwt7fHoEGDhI6rTC0oKAhLly5F7ty5UaRIEezatQvr16/H7NmzMWzYMDg7OwvJad26NcaPHw9bW1v4+/vj9evXmDx5skbuDJ8+fRouLi4wNzdPc23vrM6QLVdXpZSxXyn7Sus0nzdvXjg5OQktjC9evIiLFy8iPj5eJVPkGsF169bF6tWr1X6W169fh4uLC4KDg4Xk2NjYICgoCGZmZkL29zVpvS90dHSUy2aJXu7x6NGjWLlypcrs9YMHD0bt2rWF5mhq9v83b94oJ9bcsWMHWrVqpTbHSb58+fDbb7+hevXqWcpK7VsF1s+4jF9OaD0KCwvDgAEDIEkSoqKiUKJECZWb8SlFVa9evdR6DH2Pc+fOKb++ffs2AgICMHjwYNSsWRO5cuXC1atXsXDhQgwePBhdu3bN0mtKkTIUSldXF15eXrh+/ToWLFiAunXrwtPTU+jY+Q4dOsDe3l7tXB0QEIC//voLf/75p7CsqKgoODo64uXLlyhfvjwSExMRGRkJAwMDbNiwQchN+qZNm8LHx0flvCZJEgIDAxEQEKC8KSOCh4cHLl68CD8/P+X5LSwsDGPGjEG1atUwa9YsYVlyzEuTstxm3bp10axZM9jZ2aFYsWJZ3m+K7JiYUHYSacysWbMkW1tb6fjx49L79++lN2/eSEeOHJFsbW0lX19fYTnHjh2TqlWrJjk4OEgzZsyQPD09pc6dO0vVq1eXzp8/LyynXr160qVLlyRJkqSJEydKAwYMkCRJkkJDQ6XatWsLy3Fzc5Ps7e2lsLAw6fDhw1L16tWl3bt3SwMGDJCGDh0qLEeSJGnw4MGStbW1VK9ePWnkyJHS5s2bpQcPHgjNkCRJsrW1la5cuSI9f/5cqlatmnTmzBlJkpJ/d02aNBGadeTIEWn69OlSq1atJDMzM8nBwUHy8/OTgoODpdjYWKFZx44dk1atWiW9evVKkiRJ2rVrl7RlyxahGcuWLZPq1KkjHT9+XAoJCZHMzc2lZcuWSX369JEcHByEZkmSJF25ckUaPXq01LZtW+m3336TxowZI926dUt4jlyqVq0qPX/+XOM5S5culapWrSrVqVNHatKkicp/TZs2FZZTq1Yt6caNG2rbb926JdWsWVNYTuvWraXg4GBh+/tR7N69WzI1NZXGjBkjrVmzRlq5cqU0dOhQqVq1atL+/fuz+/C+m5ubm/T+/XtZshYsWCD5+/sr/5s7d640YsQIqWbNmtLq1atlOQb6uiZNmkgvX77UeE6HDh2kw4cPq20/fvy41LJlS2E5VatWlcaNG6fyHg8NDZXatWsn1apVS1iOJElSjRo1pPv376ttj4iIkMzNzYVmDR8+XOrRo4fK7+r58+dS9+7dpdGjRwvJGDt2rGRmZiYtXLhQSkpKkh48eCA5ODhIFhYWUlBQkJCMFG/evJE6dOggmZiYSNbW1lKdOnUkExMTqX379sLfjwMGDJBatGghzZ07VzIzM5PmzJkjDR06VKpataq0efNmIRnHjh2TPn/+/M3njRs3Tnrx4sV3779q1arK76tatWq6/5mYmHz3vn8ULLA1qGHDhtJff/2ltv348eOSra2tsJy2bdumWbDPmjVL6tKli7AcCwsLKSoqSpIkSbKzs1NeUERGRgq9sJWrkE+RmJgohYSESIsXL5Z69uwpmZubS3Z2dpKbm5uwDB8fH8ne3l5q166d1KRJEykhIUHat2+fZGNjI/Rmy5eePn0q7dixQxo7dqxkbm4uWVhYaCxLk1auXCmdOHFCkiRJCgwMlKysrKQ2bdqkWWjRt2niArRRo0bSvHnzhO/3S4MHD5acnJykd+/eKbe9f/9eGjRokNS/f39hOTNmzJDMzc2lYcOGSb6+viqFlb+/v7AcuTVv3lxau3at2valS5cKLQyyW2xsrNAbzF+zdu1aaeTIkbJkaUJ0dLTk7+8vjR49Wnrx4oW0f/9+KSwsLLsP64dmbm4u3bt3T217WFiY0M/ZQ4cOpbk9Li5O8vb2FpYjSZLUrFmzNK9ZT5w4ITVo0EBoVq1ataSrV6+qbQ8JCZHq1KkjLGfv3r2StbW11KlTJ8nKykrq2bOnFBkZKWz/qSUmJkp//fWXtGzZMikwMFD666+/pISEBOE5tWrVks6dOydJkiT99ttvUkhIiCRJkjR37lxp8ODBwvO+pmbNmhpplMoJOAZbg2JiYtKcmbN8+fLKWQ5FePDgATp16qS23cHBARs2bBCWU7FiRfz1118wMjJCVFQUGjduDADYsmVLlrv8pfbx40fljMBnzpxRTr6ip6en7A4jkpaWFszNzWFiYgIzMzMcO3YM27dvx549e4R1jxszZgwMDQ3x8OFD9OjRA9ra2nj58iW6dOmisdmIo6Ojcfr0aeUYGi0tLTRo0EDY/t+8eYPAwEDcvXsXsbGxao+LHLsn17qFABAaGoo7d+4gKSkJwP+6XoWEhGDGjBkay9WU9+/fw8fHRznkwsnJCefOnRM+5OLt27eydOVyc3ND9+7d0bhxY2XX6fv376NQoUIICgoSlnPkyBEYGBjg+vXruH79uspjCoVCaNd6ExOTdLse58qVSzmR4ODBg7M8udrTp0/RqFEjte3Nmzf/5kR1P6Jbt27B3d0dt2/fVv7Nfvm4pjVp0gRz587VeI4mREZGokuXLtDX10d0dDRGjhyJAwcOYOLEiQgKCoKVlVV2H+I39erVK8PPFfW5VLVqVaxZswaTJ09W/k0mJCRg6dKlQpfETG+puVy5cqnNFZJVbdu2hYeHB6ZMmaJcVurixYuYNm0aWrZsKTRLW1tbbVgHkLw8lMhlWuUavw4kX0va2tpmeSjZt8g5L823SBxlnC4W2BpUpUoVHDx4EAMHDlTZvn//fqFj6szMzBAcHAxjY2OV7devXxda+Gpy0prU5CrkgeSLr5Qi9NKlS9DV1UXDhg0xffp02NnZCcvR0tJSO/Fp6kTo5eWFM2fO4N69eyhevDjs7Owwa9Ys1KtXT+jYcldXV1y9ehUNGjQQuhROWk6ePIk7d+6kWciLLHTWrFmjLKJTj11WKBTCx6fKZcaMGbhw4QL69OmD48eP49KlS/Dx8cG+ffvg4+MjbO6EWrVq4dq1axpf7qdMmTI4cOAA9u3bhzt37kBHRwfdunVDu3bt0rxgyyyRS7J9y4QJEzB37lx0795deWEbEhKCdevWoWvXrihYsCDWrFkDXV3dLN9cqlevHg4dOoT+/furbP/rr780MqGaps2YMQM6OjqYMmUKPD094ebmhgcPHmD9+vXw8fGR5RjOnTunNjnYz2LWrFmwt7eHp6enspj28/ODm5sb5s6di3Xr1mXzEX5bqVKlZM8cN24cnJ2d8ffff8PU1BSSJOHatWv49OkTVq9eLfvxiDBo0CDl0oApNw2k/1/udcyYMUKzrKyssGjRIvj4+CgnKIyPj8fixYuFnYdSj19ftWoVrl+/junTp+Pw4cPCx6/LqUyZMrhz5w6MjIxgbGysvImYlJSEDx8+ZPPRUQoW2Bo0aNAgDB48GKGhoSrrjR45cgS+vr7Cctq3b4/Zs2cjIiICderUgY6ODq5du4bVq1fDwcFBZQKBrLQw2dra4uTJkyqT1rRu3RpdunT5KQt5IHlSj5IlS6Jp06ZwcXGBtbW10GXNUjt+/LjKpBQAlC2jIj+QL168iFatWqFJkyYanaDpwoULWLp0KerUqaOxDADw9PTEunXrULRoUbUbBKJbEtetW4cBAwZgyJAhaNKkCbZv3443b95gzJgxwmdQlcvJkyexcOFCVKxYEStWrECDBg3Qrl07VKlSBY6OjsJyWrVqhWnTpuH69euoUKGC2u9KVOt2yprXDg4OKttjY2Oxc+dOjbeix8XF4erVq0JvuOzbtw8TJ05UeU329vaoUKECtmzZgo0bN6Jy5crw8fHJVIGdumW6RIkSmDdvHq5fvw4rKytoa2vjxo0b2Lt3r7CJCeV0/fp1rF69Gubm5ti2bRuqVKmC7t27w9DQEFu2bEGrVq2EZX3ZUipJEmJiYnD79u3vakX9kVy+fBnr1q1T6Rmhra2NgQMHokuXLtl4ZBmXHROx1a5dG3v37sWWLVtw9+5dAMnXE926dUPx4sVlPx4RoqKisGjRIoSHh+POnTuQJAlVq1YV3rABAGPHjkXXrl3RvHlzVK9eHQqFAlevXkVMTIywNaOHDRuGX3/9FX/88Qf09fVhY2ODRo0awdXVFW3atMGFCxeE5MitY8eOGDduHGbNmgVbW1v07NkTJUuWxOnTp1G1atXsPjz6f5xFXMOOHj2KwMBA5cmqSpUqcHZ2FtrdJrsWUdckTc0++6Xbt2/LckLy8/PD0qVLUbx4cTx//hwlSpTAixcvkJiYiDZt2gi94fKlV69e4dy5czAzMxM6A3vLli0xZ84cjc+yXLduXYwePVqtoNKE6tWr48CBAyhTpgycnZ3RrVs32Nvb459//sGsWbOwd+9ejR+DaJaWljh48CAMDQ3RpEkT9O3bF7169cKDBw/w22+/4dKlS0JyvnYeEnnuqVatGjp06IBp06apLE344sULNGrUSFjOzZs3MWnSJFm6Hpubm2PPnj1qrf8PHjxA27ZtcfXqVURFReGXX37J1IoDGV1WRaFQ/HQzYVtYWODgwYMwMjKCm5sbzM3N0b17dzx8+BBdunQRNqs8kDw84csu+rly5YKVlRXatWsHbW1tYVlysba2xvr161GlShXUrFkTu3fvRpkyZRASEgIXFxeVmbN/BufPn//q4ynLm5K6hg0bYtGiRTA3N5cl7/Hjx9iwYQPu3r2rvD7u2rWrsOuUw4cPp9nFPj4+Hn5+fsK72Mtp1apVMDY2hp2dHZYtW4YlS5bAyMgIPj4+MDU1le04Up8zSBVbsDXM3t4e9vb2Gs0IDQ3V6P6zQ+HChVG4cGHlvy0sLDSSU758eSxZsgStWrVCuXLlMHHiRBw4cABWVlbw9fVVOYas2LVrF/744w/06NEDdnZ22LBhA/LmzYshQ4YIPzHduXMHw4YNg6enJ0xMTNC+fXu8ePECurq6CAwMhI2NjZCc8ePHY9q0aRg1ahRKly6ttka5iLUYgeTlijTdSp4iX758SEhIAAAYGxsjLCwM9vb2qFixYpbXUM0ucg25kOs8JEkSTp06hb59+8Lf3x+FChXSSM7MmTNl63pcunRpnDhxQmWuASC510vKcjUPHjxQWc/1e8jZ3V1uFSpUwPnz59G+fXuUK1dOufTO+/fvhY7lBJJ7VxkaGqqd6xISEnDjxg3ZChORGjZsiMWLF6vc5H39+jVmz54t7LNCTj179lRbmlChUEChUEBLS0ttPoXM+vTpE1atWpXmsoTAz7F++Jd0dXVVblpqWqlSpeDq6qqx/cs5fl1ucs5LQ5nDAlvD0lsXFhA7dhRIXmA+PDwc1tbW+PDhAwwMDITuPyfy9fXFrl270KhRI5w+fRo7d+7E8OHDceLECfj4+AjrevbixQvlxBcmJia4evUqWrZsiVGjRsHd3R0jRowQkgMA3t7eKFeuHCpUqIADBw4gISEBJ0+exIYNGzBv3jxs2rRJWNbdu3eVk9ClkCRJaIulo6MjFi9eDE9PT+Hrk3+pdu3aWLJkCSZPngwTExNs2bIF/fv3x4ULF5AvXz6NZmuKnEMu5KBQKBAUFITJkyfj999/x9KlS1GhQgXhOXJ2PR40aBDc3Nxw7do11KxZE0lJSQgJCcGhQ4fg4eGBiIgITJgwId0Lxv8yR0dHuLu7A0i+oP7111+RJ08eXLp0CZaWlkKzmjVrhtOnT6vd6Hj06BF69uyJkJAQoXlycHNzQ69evVC/fn3ExsZi0KBBePz4MQoVKqRca/xn8mUPjISEBNy/fx/z5s0TWlR5eHhg//79aNy4sdD1gbNT+/bt0a9fP/z6668oV66c2pwWIoffyDlJak4kZ23xNVmddDMnY4GtQYGBgZg7dy4KFiyodnEucuxoXFwcxo8fjwMHDkBLSwuHDh2Ct7c33r9/j4CAAOTPn19ITk508OBBzJ07F2ZmZpg2bRrq1KmDgQMHokGDBmqTAGVFwYIFlZNPlCtXDmFhYQCSW3mjo6OF5QDJY+q2bt0KAwMD/P3337C1tUWJEiXQuXNnoWO9Z86cCRsbGzg4OEBPT0/Yfr/UqlUrODg4oFatWihWrJjaCV1kl9aRI0eib9++2LhxI7p164bFixejTp06+PTp0085PhWQb+4EuUiShKJFi2Lt2rUYP348HBwc4OfnJ7xbXFJSkvLCuXz58rhz5w5q166NZs2aYenSpUKz2rVrB319faxYsQJz586Fjo4OqlatisWLF6NRo0Y4f/482rVrl+nPjK/NUv6ln2UYUYpOnTqhYMGCKFSoECpWrAhvb28sXboURkZGmDx5cpb3v379eqxYsQJA8nuvU6dOai3Y7969E9ZjR24lSpTAzp07sXfvXty6dQtJSUno1q0bfv31V+jr62f34X23tCY8K1euHPLmzQtPT0/s2rVLSM6RI0fg7e0t9EZbdluyZAkAYOXKlWqPKRQKoQW2nJOk5jRy1RYZwVHG6WOBrUHr1q3DoEGDhLZOpmXx4sUIDQ3F6tWrlTOW9+rVCxMnTsTs2bMxbdo0jeb/zN68eaMsMk6fPo3OnTsDSO6i/vnzZ2E59erVg4+PDzw9PVG9enUsWbIE3bt3x6FDhzLd7TM9Wlpa0NXVRWJiIv79919l686HDx+EzrIcHR2NoKAgjY+9cXNzQ4ECBdC5c2eNFvIAULlyZRw9ehQfP35Evnz5sHXrVuzZsweGhobClymRU+ohF69evUJUVJTGx85rSkqhqKurCz8/P/j5+WHgwIHCZ7mVs+sxkLzUU5MmTdJ8zNraOktjR2fMmJFjWxp27tyJ1q1bK3u3tGnTBm3atMHHjx+xZcsWtW7336tjx454/fo1JEnCwoUL0bJlS7WL2nz58v20vQsWLFiADh064Pf/a+/Ow2rM3z+Av0+bLWuMMqi0TCZpQSqDUl+DxJQlmaI0M5bCVLYwWZKmRcukieZbURpkKc3UWIYKZSkqmbSoLBM1X8U02Urn/P5wdX7OnMygp/P0HPfrulwXz3Ocz33onPPcz+fzue85c9gOpUMNHDgQVVVVjD2fjIyMRPe6SoIktxtKqkiqNJJUbgG8bIFZVlYm/Hn/+43EuLg4zlZj72iUYHcgSfWFTUtLw+bNmzF27FjhMWNjY/j6+mL16tWUYP+DoUOHoqioCPX19bh9+7awP+yvv/4q7DPIhNWrV2Px4sU4ceIE5s+fj7i4OGFP6nXr1jE2DvCyqNWuXbvQv39/PH36FBMmTEBtbS1CQkIYXTJpYGCA0tLSDk+wi4uLkZSU9MbF/Nqra9euqKmpQV5eHmRkZDB58mTOzk4BktuTLyl/v2Pu4eEBNTU1RmYrXyXJpccAUFBQgISEBJSVlUFWVha6urpwdnaGlpZWu5/bzs6OgQg7j/r6euENUG9vb2hpaYnVy7hx4wZCQkLanWB369ZNOCPE4/Hg6ura4Tf6JCktLQ1RUVEwMjKCra0tpkyZwsmZ61b37t0T+bNAIMBff/2FqKgoRlsITp48GcnJyfj6668Ze873ycCBAzm77YptksgtGhsb4enpiXPnzom0K502bRr8/f2FNzQ74rtQWlCC3YEk1Re2trYWQ4cOFTuuoqKChoaGDh2b67744gt4enpCRkYGJiYm0NHRQWRkJCIjI4X9kJnQugzv+fPnUFBQwI8//ohz585h4MCBjBfG2bhxIzw9PXH37l2sX78e/fr1g6+vL27evIn//ve/jI0zd+5c+Pj4ID8/H2pqamLtzZj6AhgyZEiHzBi25U2/VLhEknvyJcHf319s24utrS0+/PBDJCcnMzbOPy09/uabbxgbB3hZhMzd3R0jR46EmZkZ+Hw+CgoKYGdnh7i4uHa3BPP29saGDRugqKgIb2/v1z6Ox+Mx+rnXUc6ePSus6C0QCIQrj17V2r+XSe7u7mhsbERSUpJwVkdXVxdTpkzhbB/sEydOoLCwEKmpqQgJCYGfnx8sLS1hZ2cHMzMztsN7a5MmTRJbrSEQCNCjRw/s2LGDsXF69eqF2NhYZGVltdmWkI3WYVwiqSKp0kgSuYWfnx+qqqoQHR0trAty9epV+Pr6IiQkhPGJIWlEbbo60KFDhxAUFAQ7O7sO7QtrZ2eHefPmYe7cuSIl83fu3IkzZ87g6NGjjIwjrUpKSvD7779jwoQJUFBQwNmzZyEnJ8f4xQWfz8eFCxdELsyY7KX7T+rq6tCnTx88fPiQsf1OkmrLdOnSJQQEBGDlypVQV1cXq3LK5Bext7c38vLy4OPjI/al8p///IeTXypGRkY4dOgQNDQ0sHz5cnTv3h0BAQG4e/cubGxsUFBQwHaI7yQ3NxcVFRWYPn06ampqoKqq2mE97DuajY0NzM3NxZa5BwQE4OrVqzh48GC7nt/JyQmRkZHo1asXnJyc/vGxTPWg7Wi5ubng8/lYuHAhIiIi0Lt3b+E5Ho+H7t27Q1tbm9GfiYqKCixcuBCPHz+Gmpoa+Hw+bt++jYEDB2Lv3r3Ciu9c1dLSgnPnziEtLQ2nT59G7969kZGRwXZYb6WttmLy8vLQ1tZmdMZUWt5HbMnIyICXlxeePn0qcpzpIqnSSBK5hbGxMSIjI8W2Jl24cAFeXl7Iyclp9xjSjhLsDiSpBCQjIwNff/015syZg6SkJHzxxReorKzEyZMnERoaik8//ZSRcaTZkydPUFVVBVlZWairqzM+G/HHH3/A1dUV5eXl6NWrl3DZmpGREXbv3s1oIbrhw4e/ttKtjY0N8vPzGRtLEnR1ddHS0gJAtGJlR3wRS+OXyujRo5GcnIxBgwbBxMQEGzZswGeffYaSkhI4Ozvj4sWLbIf4VhobG/HFF1+goKAAPB4PJ0+ehJ+fH27fvo09e/Ywuh8sKysLMTExqKysxMGDB3HkyBEMHTqU8eV5+vr6OHbsGNTU1ESO37p1CzNnzuRkdWpJuXz5MoyMjCTSXsjFxQVycnIIDg4WJvT19fVYtWoVunfvjp07d3Z4DB2ptrYWaWlpOH78OH777TcYGxu3WfCKkPaaPHkyNDU1X1sklfZmv54kcgtTU1PExcWJjVVeXo758+f/a795QkvEO5SkCkZYWFggIiICu3fvhqysLGJiYqClpUXJ9Rtobm7G9u3bceTIEWG7g27dumHBggXw8PBgbJytW7dCQUEB6enpwpZC5eXlWLt2LbZv397u5WSHDx9GamoqgJeJp5ubm9jMzR9//IFevXq1axw2SPICT1ZWts2bHf3790dzc7PE4mCSpPbkS0pISAiAl1V8Z8yYAQBYs2YNVq1ahYCAAOH59srOzoa7uzusra1RUFAAPp+PlpYWrF+/Hi0tLZg1axYj4wAvbyJduHBBLMG+fv16h1R6f/HiBerq6oQ3rgQCAZqamlBYWCiRuiFMMjY2Fhb5rKqqQnh4OH799VdoaGgwXl+goKAASUlJIrPl/fr1w5o1azB//nxGx5KUxsZGHD9+HD/99BNyc3Px4Ycf4rPPPkNYWBgnl+lKsv3Ts2fPcPz4cVRWVmLRokUoKyuDpqYm44VLpZGkiqRKI0nkFi4uLvD19UV4eLhw1WNjYyPCwsLg6OjY4eNLA0qwpcSECRMwYcIEtsPgnJCQEBw/fhzr16+HgYEBBAIBrl69ioiICHTr1k1Ylb29cnJyEB8fL9KvV0tLC5s2bcKXX37Z7ue3srLClStXhH9WVlYWqxiura3NuYtn4OUebBUVFbHjz58/xy+//MLoWNL4pfLNN9/Aw8Ojw/fkS0pGRgZ27NghcmE2bNgwbNq0ibH3KwBERETAy8sLzs7OOHHiBICXBdV69eqFuLg4RhPsGTNmICgoCFVVVTA2NoacnByKioqwd+9e2NvbIyUlRfjY9r6HL1y4gNWrV6Ourk7sXNeuXTn3GXH9+nU4ODjAwMAA169fR1NTE27cuIHt27dj586dr63M/i4++OAD1NTUiBWea2xs5OTNSwAwMzODvLw8Jk+ejL1797arWn1nIKn2Tw8ePMC8efPw4MEDNDU1Yc6cOYiNjRW+bzU1NTtsbGkgqSKp5N1kZmaiqKgIlpaWUFNTg5ycHG7duoXHjx/jxo0bwgkdgNlWqdKElohLic7SdJ5rzMzM4OfnJ3YRdurUKfj5+SEzM5ORcczNzREaGgpDQ0OR49evX8dXX33F6NJjLy8vbNmyhdOVYF81fPhw2NraYuvWrSLLQB88eIDx48czukR8/vz5KCoqgoyMjNiXyqBBg0SWqHP5S6V1T76srCzbobw1fX19/PzzzxgyZIhIzYmKigrMmjWLsT3lhoaGOHbsGIYOHSoyzt27dzF9+nRGl22/aYV8Jpb/zZkzB0pKSnBycoK7uzuCg4Nx7949fPfdd/D394eVlVW7nl/SFi5cCAMDA3h4eIj8PwUEBODy5cs4cuQIY2OdOXMG27dvx7p160RuhGzZsgVz5swRWTHGldnf5ORkTJkyRWoqoxsaGkqk/dOqVavQ2NiI0NBQmJmZITU1Fb169YKnpydkZWURHR3doeNzXVpaGvz8/GBra9uhRVLJu3mb7S6UY7SNZrClQGdqOs81TU1NbVZg19DQwJ9//snYOCtWrICPjw+CgoKEF9N3796Fn58f3NzcGBsHeFkUrLKykvHq5GwRCAQ4e/YsXFxcEBERgT59+oicY5KZmRknK+f+m9aljBUVFXB1dUVFRQVnlzLq6ekhPT0dixcvFjkeHx/PaF/anj17ttmhoby8XGSJMBMk2X+2tLQUhw4dwkcffYSPP/4Y3bt3h5OTE7p3746YmBjOJdi//fYbNm/eLHbcwcGB8Qr5y5YtA/DygvLv9SACAgIQGBjIuSJNtra2ePLkCQ4cOIDS0lLIyclBS0sL06ZN4+RNWkm1f7p48SKio6NFbkz07t0bq1evxoIFCzp8fK5rLegYExMjdo7H41GCzbKxY8fCwMBA7MbH8+fPkZmZSdtP3wAl2FJAkk3npc2sWbMQHh6O4OBgYSVGgUCAPXv2CPd3MiE8PBx1dXWwtbVFjx49ICcnhz///BMCgQAFBQXYtm2b8LHtvTBTUFCQSMEfSeHxeIiJiYGPjw/mzJmD3bt3C5fa/70dS3u5u7ujpKQEZWVl4PP5AET3p3KhhdHf/X0p49y5czm9lNHT0xMuLi7Iz8/HixcvEBUVhZs3b6K4uLjNi7V3ZWNjAz8/P/j5+YHH4+Hx48fIysqCr68vpk2bxtg4kiYrKytMnNTU1FBWVgZTU1OYmJggICCA5ejenry8PBobG8WO37t3j/FZWSb373YW9+/fh6OjI+rq6qCuro6WlhYkJSVh165d+PHHHzlXGV1S7Z8eP3782p+vFy9eMDKGNJPkTUXy9hYsWNBmsdybN29i9erVlGC/Aem5Cn+PSaLpvLR68OABMjIyMGnSJIwcORJycnIoLi5GdXU19PX1Re5Et+fi6uuvv2Yg2jczY8YMfPHFF5g5cyZUVVXF9mJz7WdFIBCgf//+SEhIwNq1a2Fvb4/Q0FBGZytbxcfHw9/fXzgL9WovbEm1VGPat99+C01NTfz000/C2fmAgAB4enoiMDCQc0sZjYyMkJSUhJiYGKiqqqKgoABaWlrYsGED9PX1GRvn66+/Rk1NjXCvta2tLQQCAczNzRl/P7cuMy4vL2+z5zuTs6E6Ojo4deoUnJ2doa6ujitXrmDhwoWoqalhbAxJsrKywo4dOxAaGio8VlFRAT8/P5ibmzM61uXLl+Hq6iqWWDU2NiI8PBwbNmxgdDxJ+Pbbb6GiooJDhw4JL6YfPHiAlStXIigoiNHe0ZJSXl4OFxcXkWNMrywYM2YMEhMTsXHjRuGx5uZmREZGwsjIiJExCJGkPXv2CG+yCgQCjBs3rs3HScvqyI5Ge7ClwKJFi2BnZ4fp06ezHQrneHt7v/Fj21vp+004OTkhODi4Xa2GJNUeTlKGDx+O8+fPQ0lJCQAQGhqKmJgYeHl5ITAwkNHXM3nyZEydOhVubm6wsLDA0aNH8ejRI3h5eWH27NlwdnZmbCxJ+eSTTxAdHY2PP/5YZI9qSUkJFixY0Gbf2M7M19cXCxcubHNrB5MsLS1x+PBh/PXXXyguLgafz4e2tjZ69uyJGTNm4NKlS4yNNXPmTHTp0gV2dnZttgi0tbVlbKwzZ87A3d0dGzduhIWFBT799FOYmJigtLQU+vr6+O677xgbSxJa27YVFhZCIBCgZ8+eaGxshI6ODuLi4kS2lLyLiooK1NfXA3g5q/P3ntsAUFZWhsDAQE62Uxs9ejTi4uKgp6cncvzatWv48ssvGf05lwRJtX+qqKjA559/jg8++ACVlZUYO3YsKisr0dDQgMTExDeuq/A+mTRp0huvOuNyjROuWr58OSwtLcHn87F+/XqsX79epKsKj8dD9+7dYWJiwtmijpJEM9hSYOrUqdi6dSuuX7/eYU3npZUkkua30VoFtz2kbenV3+8Benh4QE1NDT4+PoyPde/ePcyePRsKCgrQ0dFBUVERrKyssG7dOnz77becTLClbSljSkqK2OwUU9LT03Hu3DkAQHV1NXx9fUUS3nPnzqG6uprxrQm3bt3C4cOHxapTd4RJkybh0KFDkJWVhYqKCmJiYhAbGwtLS0usWLGiw8dnmqKiIg4cOIALFy6I3AgZP3682PLgd3H37l0sWbJE+H/+upomTFaVlyRZWVmxVU4A0KVLl3Z/F7FBUu2fNDQ0kJqaiv3790NFRQV8Ph+mpqZYunQpVcZ+DVtbW+H76M8//0RiYiIsLCxgaGgoLBh48uRJLFq0iOVI309ZWVlYs2YNhgwZgvXr12PatGkdWolf2lGCLQW++eYbAC+Xd/wdFYsQl5KSgmnTpkFBQUGk/U1buPxvV1lZidLSUsjLy0NDQwPq6upsh/RO/P39xXpT29ra4sMPP0RycjKjY/Xo0UOYdKqpqeHmzZuwsrKChoYGqqurGR1LUqRtKaO5uTn27dsHd3d3xoswGRoa4sCBA8KbOvfu3RMp8tJ6B5/pvcp6enqorq6WSIINvOy73WrMmDGcb80EAKampjA1NWX8ec3NzXHmzBnw+XxYWVmJLKUG/v9nor0z5WwxMjLC999/j8DAQOHPenNzM6KiosS6XnCBpNo/NTQ0ICIiAo6OjnB3d8eiRYtw9OhR5OfnIzo6mpLsNixfvlz4ezc3N3h4eIi1KU1ISMCvv/4q6dAIgKFDh8LNzU34/bB9+/Y2V1QBnW9yqjOiJeLkvaOjo4Ps7GwoKSl1uuXUry7hfVdNTU1YtWoVTp48KTzG4/FgYWGBsLAwsRUOXJGbm4uKigpMnz4dNTU1UFVVFatw2V5ubm5QVFSEj48P0tPTkZSUhIMHDyItLQ0BAQE4f/48o+NJwuuWMv7111/Yt28f55YyOjg4ID8/HzweD0pKSmIXAEwtLXRyckJkZKRElsJVVVVhyZIlsLa2brMwE5M3+h49eoTo6GiUl5fj+fPnYue5VshLR0fntSsK5OXloaysjJkzZ2LZsmXtXnmQnJyMqVOnis34PnnyBElJSZxc4VJRUYF58+ahR48eGDFiBHg8Hq5du4bGxkYkJCR0SK2LjiSp9k/e3t7Iy8vDrl27UFlZCS8vL/j5+SEtLQ3y8vKIiIhgZBxpZWBggGPHjkFVVVXk+O3btzFjxgxObrfguuLiYnz//fdoaGhAbm4u9PX1X3uNlZCQIOHouIcSbEI6ESYS7ICAAPzyyy/YtGkTxowZg5aWFuTm5mLbtm2wsbERtsfgitY9lgUFBeDxeDh58iT8/Pxw+/Zt7Nmzp1371f+utTiOs7MzHBwcYGNjg4aGBjx9+hSurq7w9PRkbCxJ+uOPP7B//37hElotLS3Mnz8fgwcPZju0t/Zv/Tm52JYwKioK4eHhbZ5j+kbfl19+iWvXrmHcuHFtzk5wbWZi7969CAkJwfz58zFq1CgAQGFhIfbt24d58+ahd+/eiI+Ph4uLi9hs2Zuor6/Hs2fPAPz/vvy+ffuKPKa4uBienp64du1a+18QC+7du4fExESUl5dDIBBAW1sb8+bN4+QsrKRumpuZmSEyMhKGhobYsGED6urqsGvXLpSWlsLR0RG5ubmMjCOtJk+eDBcXFzg4OIgcj4mJwZEjR5Cens5SZAR4uZXoyJEjYp915M1Rgk1IJ8JEgj1+/Hhs27YNEydOFDmekZGBLVu2IDMzs51RStbWrVtRXFyMoKAgzJgxA6mpqWhubsaqVaugpqaGkJAQRsd79uwZnjx5gn79+qGurg4//fQTlJWVMWXKFEbHkaSKigo8fvxYWP0zNjYWFhYWnN02IG1MTU2xYMECuLi4tLkflkmGhobYvXs3Y8We2DZ37lzMmjUL9vb2IseTk5ORlJSE/fv34/Tp0wgMDMSJEyfe+vlTUlKwbt06YVeBtmbBBQIBJk6ciN27d7/z6+jsLC0tER8fjw8//JDtUDoFAwMDHD9+HMrKyrCwsICLiwsWLFiAO3fu4LPPPsPVq1fZDrFTO3jwIHx9fWFtbQ09PT0IBAJcuXIFp06dQnBwMKZOncp2iIS0C+3BJkTKNDY2ii27AgB1dXVhNVwuycjIwI4dO0RuOgwbNgybNm3CkiVLGB+va9euwiRHSUmJk8s+X3Xu3Dm4ublh0aJFwgT7l19+QUREBH744QdOth+Ttl7lz58/h42NTYcn1wAwcOBA9OjRo8PHkZSSkhKYmJiIHR81ahQ2bdoEAPj4449x//79d3r+zz77DB9++CH4fD4WLlyI7777TqSKeOsebG1t7Xd7ARxRX18vfL+Rl0XOMjMzoaKigvv372PChAkAgKSkJGhoaLAcXednb28PRUVFJCQk4OTJk+DxeBg+fDi+//57sckBQriIEmxCOhEmqhNra2vj+PHjYslneno6J2cs6+vrMWDAALHjioqKePr0KQsRcUtYWBi++OILkQrRhw4dQmhoKIKDg3HgwAEWo3t78fHxwiRaWnqVT58+HWlpaVi8eHGHj7V27Vps3boVHh4ebe73HjRoUIfHwKTBgwcjIyND7EbYmTNnoKysDAC4c+eOSGGyt9VaBC4+Ph5GRkaQk6NLp/fdihUrsHz5cjQ3N2P69OlQU1ODv78/EhMTERkZyXZ4nGBtbQ1ra2u2wyCkQ9C3BCGdCBM7NpYuXYply5ahpKQERkZG4PF4yMvLEy694ho9PT2kp6eLJR/x8fGcK8DDhoqKCoSFhYkdnz17NucKWgHAvn37sHjx4jZ7lVtaWrId3jtRUlJCZGQkTp06BXV1dbEEjul90a21Bl7VuvxZ0oUd22vp0qVYt24dioqKYGhoCD6fj8LCQpw4cQJbtmxBVVUVvL29MXny5HaPZWxsjEuXLuH69et49uyZ2Oc1F/f/k3czceJEZGVloba2Vrjve9q0aZg7dy7NYL8haVuJRMirKMEmREKePXsGGRkZKCgooKKiApmZmTA0NBRplXTixIl29x00NzfHd999h+joaGRmZgqL1oSEhHByH7GnpydcXFyQn5+PFy9eICoqCjdv3kRxcTFiYmLYDq/T69evH4qLi8X29ZeXl0ukQjbTpLFXeV5eHvT19QEANTU1HTqWv78/TExMYG9v/9r+6FxiY2MDRUVFxMbGIiQkBHJycvjoo48QFRWF8ePHIzc3FzY2Nowkv9HR0QgJCUHPnj3FWgfyeDxKsN8zffv2FSkC1foeJv9OGlciEfIqSrAJkYDc3Fy4ubkhPDwcmpqamDNnDmRkZPDkyRPs2LFDWNDjgw8+YGQ8KysrWFlZMfJcbDMyMkJSUhJiYmKgqqqKgoICaGlpYcOGDXRB8wZsbW2xZcsWNDQ0YOTIkeDxeCgqKkJYWBhsbW3ZDu+tSWOvckm2PKmtrUVMTAwnK0S/joWFBSwsLNo8x2Sf74SEBKxcuRJLly5l5PkIeV9J40okQl5FCTYhEhASEgJLS0vo6enhyJEjUFRUxMmTJ3HkyBHs3r2b8YqZBQUFSEhIQFlZGWRlZaGrqwtnZ2doaWkxOo4k+Pr6YuHChQgMDGQ7FE5atmwZHj58iK1bt+LFixcQCASQk5ODk5OTyL5srhg1ahR27doFHx8f6OjoICkpCV999RXy8vI4Xbzr6dOnKCsrQ3Nzs8jSY6ZndAwMDFBaWipVCbakPu8aGhpgY2PD6HMS8j6SxpVIhLyKEmxCJKC4uBiBgYFQVFTE+fPnYW5ujq5du8Lc3BwBAQGMjnXmzBm4u7tj5MiRMDMzA5/PR0FBAezs7BAXF8e55VcpKSli+0XJm5OVlYWPjw+8vLxQVVUFOTk5qKmpSaRidUfw9PSEs7Mz9u/fDwcHB0RFRcHY2FjYq5yLMjMzsXr1ajQ2Nort62V6X/TcuXPh4+OD/Px8qKmpQV5eXuT8Z599xthYkiDJz7tRo0YhPz+fk/3j24uJApyEtJLGlUiEvIoSbEIkoFu3bmhqakJTUxPy8vKEe48ePHggtpevvUJDQ+Hq6govLy+R4wEBAQgKCsLBgwcZHa+jmZubY9++fXB3d4eioiLb4XDW1atXUVpaCjk5OdTV1cHExASysrJsh/XWNm/ejPnz58PExAQ9evTAoUOHON+rPDg4GKNHj8bKlSsZ/zz4u9bPhbbqF/B4PM4l2B39eZeSkiL8/YgRI7Bp0yaUlZVBTU1N7P3DtX87ALh8+TIMDQ3FbrT8HRc/K0jnNXr0aKlciURIK56AibLFhJB/tHLlSjx//hy9e/fGyZMnce7cOdy9exc+Pj4YMmQIQkJCGBtLX18fx44dg5qamsjxW7duYebMmSgsLGRsLElwcHBAfn4+eDwelJSU0KVLF5Hzp0+fZikybmhoaICrqyuKiorQq1cv8Pl8NDY2QldXF3FxcZwrdBYYGIizZ8+ioqICmpqasLS0FG6/4Co9PT0kJydDU1OT7VA4p6M/71orRP8bLlZgBwATExPExMRAV1eX7VDIe6S1k4GzszMcHBxgY2ODhoYG4UokT09PtkMkpF1oBpsQCdi0aRM2bdqE0tJSBAUFQVFREceOHYOcnBy8vb0ZHUtXVxcXLlwQu+C8fv06J9uHjBs3DuPGjWM7DM4KCAjAs2fPkJqaCm1tbQAv26OsXr0aO3bswJYtW1iO8O2sWbMGa9aswe+//46srCxkZWVhz5496N27NywtLeHj48N2iG9NTU0N9fX1bIfBSR39eVdSUtLu5+jMlJSU8Ndff7EdBnnPaGlp4ddff8WTJ0+kZiUSIa+iGWxCWNLc3Pyvy/LexYEDBxAUFIRZs2bB2NgYcnJyKCoqwt69e2Fvby9S+IeLSxrJ2zExMUFERIRYJeXLly/Dw8MD2dnZLEXWfrW1tbhw4QKysrJw4sQJAC/rHXDN2bNnERQUBA8PDwwbNgwKCgoi5wcNGsRSZJ0ffd61j7+/Pw4cOICJEydCVVVVbIUQtR4jHamyshKlpaWQl5fHsGHDMGzYMLZDIoQRlGATIiFXr16Fmpoa+vXrh5SUFPzyyy8wMjLCV199xWgBGWlc0lhSUoKysjLw+XwAgEAgQFNTEwoLC4X72UnbRo8ejYMHD4rN5lVUVMDOzo5zWwbS09Nx8eJFXLp0CXfu3IGKigpMTU2Fv5SUlNgO8a29+p599bNAIBBw6n3KBmn8vJOkSZMmvfYcj8ejLTikQzQ1NWHVqlU4deqUSA9sCwsLhIWFid1kJIRraIk4IRJw4MABbNmyBbGxsVBSUoK3tzdMTU0RFxeH5uZmRmcJpG1JY3x8vDCJ5vF4Il/GXKuIzgZdXV3s378fGzduFDn+448/Yvjw4SxF9e48PT0hIyMDKysrhIWFcfI1/F18fLzYsbq6OvTt2xcyMjIsRMQd0vZ5J2lnzpxhOwTyHgoNDcW1a9fw/fffY8yYMWhpaUFubi62bduGiIgIsaKFhHANzWATIgFTp06Fo6MjPv/8c4SHh+P06dNITU3F2bNnsXnzZrrI+QeTJ0/G1KlT4ebmBgsLCxw9ehSPHj2Cl5cXZs+eTf0y/0V+fj4WLFgAHR0dGBkZgcfjIS8vDyUlJfjhhx9gamrKdohv5ddff8XFixeRk5ODO3fuYMSIETAxMYGJiQmMjIw4OfPR0NCAoKAgODo6QlNTE66urrh06RLU1NQQHR0tVT2rSeeUm5uLiooKTJ8+HTU1NVBVVe2QLUyEAMD48eOxbds2TJw4UeR4RkYGtmzZgszMTHYCI4QhdGucEAn4/fffhUvxsrOzMWHCBACApqYmHjx4wGZond69e/cwe/ZsKCgoQEdHB0VFRfjoo4+wbt06HD58mO3wOj1DQ0MkJiZi8ODBOH/+PM6ePYvBgwdj3759nEuuAcDKygobN25Eeno6Tp8+jXnz5qG6uhqLFy+GsbEx2+G9E39/f1y8eBFycnI4c+YMrly5gsDAQKiqqiIwMJDt8Dq1oqIizJ49G/r6+hg+fLjYL/LPGhsbYW9vDycnJ2zZsgUPHz5EcHAwbGxsUFNTw3Z4REo1NjZCVVVV7Li6ujoVfCRSgZaIEyIBSkpK+OOPPyAvL4/r16/Dw8MDwMvljf3792c5us6tR48eePHiBYCX1ZZv3rwJKysraGhooLq6muXoOj9fX18sXLgQoaGhbIfCqNraWmRnZyM7Oxs5OTmQkZHhbLX5rKwsREZGQkNDA7GxsRg3bhxsbGygra0NR0dHtsPr1DZu3IguXbrA29tbrEAX+XchISHg8Xg4deoUZsyYAeBlpf5Vq1YhMDCQ0RaShLTS1tbG8ePHsWTJEpHj6enpUFdXZykqQphDCTYhEmBtbY1Vq1ahW7duUFZWhrGxMdLT0+Hr64vZs2ezHV6nNmrUKOzatQs+Pj7Q0dFBUlISvvrqK+Tl5aFHjx5sh9fppaSkwMXFhe0wGOPn54ecnBxUVlbigw8+gLm5Ob799luYmppycnk4ADx58gQqKioAgJycHOH/V7du3dDS0sJmaJ3erVu3cPjwYZFq4eTNZWRkYMeOHSLbEIYNG4ZNmzaJJT+EMGXp0qVYtmwZSkpKRLYunTp1CsHBwWyHR0i7UYJNiAR4eXlBWVkZd+/exeeffw5ZWVnU1dVh7ty5WLFiBdvhdWqenp5wdnbG/v374eDggKioKBgbG+Pp06dwdXVlO7xOz9zcHPv27YO7uzsUFRXZDqfdrly5gqlTp8LCwgK6urpsh8MIDQ0NZGZmQkVFBffv3xduIUlKSuJk73pJ0tPTQ3V1NSXY76i+vh4DBgwQO66oqIinT5+yEBF5H5ibm+O7775DdHQ0MjMzIRAIoK2tjZCQEOqDTaQCFTkjhHRqjo6OMDMzw4QJEzBixAjU1dXhp59+grKyMn0RvwEHBwfk5+eDx+NBSUlJbBktteFhX1ZWFpYvX47m5mZYW1sjODgY/v7+SExMRGRkpFghIPL/qqqqsGTJElhbW2Pw4MFiVdep9/U/c3R0xPjx47F48WIYGhoiNTUVQ4YMwaZNm1BeXo4ff/yR7RAJIYRzKMEmREKysrIQExODyspKHDx4EEeOHMHQoUPpAvBfBAYG4uzZs6ioqICmpiYsLS1haWkJPT09tkPjhJ07d/7jeSZbxJF39/DhQ9TW1gr7OhcWFkJRUZFmsP9FVFQUwsPD2zxHva//3dWrV+Hi4gJTU1NkZ2fDxsYGN2/eRHFxMWJiYjB27Fi2QyRSIiUl5Y0fS9dFhOsowSZEArKzs4WzLOnp6UhLS8Phw4fxww8/wNfXF7NmzWI7xE7v999/R1ZWFrKysnD58mX07t0blpaW8PHxYTs0QghLTE1NsWDBAri4uKBr165sh8NJJSUliI2NRXFxMfh8PrS0tLBo0SLo6+uzHRqRIq03D/8N3Rgj0oASbEIkYN68eZgyZQqcnZ1FluHFxMQgOTkZP//8M9shckJtbS0uXLiArKwsnDhxAgBQXFzMclSdX0FBARISElBWVgZZWVno6urC2dmZ9q0SzjMyMkJqaioGDx7MdiicMXHiRCQlJWHgwIHYuXMnXF1d0a1bN7bDIoQQqUF9sAmRgNLSUmEf7FdNnjwZd+/eZSEi7khPT4ePjw8+/fRTYWGU7t27IzAwEOfOnWM7vE7vzJkzmD9/Pqqrq2FmZoYxY8agrKwMdnZ2yMvLYzs8Qtpl+vTpSEtLYzsMTmndjgAAkZGRVMyMEEIYRlXECZGAnj17ora2FkOHDhU5Xl5ejt69e7MUFTd4enpCRkYGVlZWCAsLw/Dhw9kOiVNCQ0Ph6uoKLy8vkeMBAQEICgrCwYMHWYqMkPZTUlJCZGQkTp06BXV1dcjJiV7W+Pv7sxRZ5zVmzBg4Ojqif//+EAgEmDVrllhxuFZUBJEQQt4eJdiESICNjQ38/Pzg5+cHHo+Hx48fIysrC76+vpg2bRrb4XVqO3fuxMWLF5GTk4M5c+ZgxIgRMDExgYmJCYyMjDjb+1hS7ty50+Yef3t7e6oQTDgvLy9PuFe4pqaG5Wi4ITQ0FCkpKWhoaMDOnTsxZcoU9OjRg+2wCCFEatAebEIkoLm5GevWrRMuZeTxeBAIBDA3N0d4eLhY6yTSttY92NnZ2Th58iR4PB4KCgrYDqtTmz9/PmxsbODg4CBy/Oeff0ZsbCyOHj3KUmSEEDbY29tj586dGDBgACZNmoTU1FQoKiqyHRYhhEgNmsEmRALk5eWxY8cOrFixAjdu3ACfz4e2tjY0NTXZDo0zamtrkZ2djezsbOTk5EBGRgbjxo1jO6xOb8aMGQgKCkJVVRWMjY0hJyeHoqIi7N27F/b29iKtU6g1CuGip0+foqysDM3NzXh1zoDH42H06NEsRtY5FRcXo6GhAQMGDMD9+/fR1NTEdkiEECJVaAabENKp+fn5IScnB5WVlfjggw9gbm6OSZMmwdTUlJaHvwFqjUKkWWZmJlavXo3Gxkb8/XKGfqbbZm9vj+rqaqirqyM3NxeGhoaQl5dv87Hx8fESjo4QQriPZrAJkQAdHR3weLzXnqeLwNe7cuUKpk6dCgsLC+jq6rIdDueUlJSwHQIhHSY4OBijR4/GypUr0bNnT7bD4YSQkBDEx8ejoaEBubm5UFZWph7ihBDCIJrBJkQCjh49KpJgv3jxArdu3UJycjLWrVuHGTNmsBgdIYRwk56eHpKTk2m7zTtycnJCZGQkevXqxXYohBAiNWgGmxAJsLOza/O4jo4Ojh07Rgk2IYS8AzU1NdTX17MdBmclJCSwHQIhhEgdmsEmhEXV1dWwtramStiEEPIOzp49i6CgIHh4eGDYsGFidRkGDRrEUmTcQNuXCCGEeTSDTQiL0tLS0Lt3b7bDIIQQTvrqq68AAMuWLRNJFAUCARU5ewPbt2//x+1LhBBC3h4l2IRIwKRJk8Qu/h4/fow///wTnp6eLEZGCCHc1VaV67q6OvTt2xcyMjIsRMQttH2JEEKYRwk2IRLQ1kWMvLw8jIyMMGbMGBYiIoQQ7tPR0UFQUBAcHR2hqakJV1dXXLp0CWpqaoiOjmY7PM4yMjLCN998w3YYhBDCSXR7lxCWNDc34/Lly4iOjkZqaioePXrEdkiEEMIp/v7+uHjxIuTk5HDmzBlcuXIFgYGBUFVVRWBgINvhcRZtXyKEkHdHM9iESEBubi5yc3MhLy8PdXV1AMDt27fx7NkzqKio4NGjR+jSpQvi4+OhpaXFcrSEEMINWVlZiIyMhIaGBmJjYzFu3DjY2NhAW1sbjo6ObIfX6b1u+1JDQwM8PDxYjIwQQriLEmxCJEBPTw98Ph/h4eHo168fAODRo0dYvXo1Ro4ciSVLlsDHxwfBwcHYvXs3y9ESQgg3PHnyBCoqKgCAnJwcuLi4AAC6deuGlpYWNkPjBFtbW7Eq4rR9iRBC2ocSbEIk4MiRI4iNjRUm1wDQp08feHl5wcXFBcuXL4erqyvmzZvHYpSEEMItGhoayMzMhIqKCu7fv48JEyYAAJKSkqChocFydJ3f8uXL2Q6BEEKkDiXYhEjAixcv0NzcLHb8+fPnePbsGQBAQUEB1JaeEELe3IoVK7B8+XI0Nzdj+vTpUFNTg7+/PxITExEZGcl2eJxQXFyMPXv2oLy8HAoKCtDW1saXX36JoUOHsh0aIYRwEk9AV/SEdDgPDw/cuXMHISEhUFVVBQBUVVVhzZo1UFZWRlhYGIKDg1FYWIgff/yR5WgJIYQ7Hj58iNraWujo6AAACgsLoaioSDPYbyAnJweurq4wMDCAvr4+WlpacPXqVZSXlyM6OhomJiZsh0gIIZxDCTYhElBfX4/Fixfj+vXr6NWrFwQCAf766y/o6+sjIiICv/32Gzw8PLB7924YGxuzHS4hhJD3gI2NDaysrLBy5UqR49u3b0deXh6OHj3KUmSEEMJdlGATIiECgQCXLl3CjRs3ICsrCx0dHWEy/fDhQ8jJyaFnz54sR0kIIeR9MXLkSPz000/ClVWtqqqqMHPmTFy7do2lyAghhLtoDzYhEsLj8WBiYtLmkru+ffuyEBEhhJD32YgRI3D58mWxBLuwsBCamposRUUIIdxGCTYhhBBCyHsiJSVF+PtRo0Zh27ZtqKysxKhRoyAjI4PffvsNcXFxcHNzYy9IQgjhMFoiTgghhBDynmgtBvdveDwebty40cHREEKI9KEEmxBCCCGEEEIIYYAM2wEQQgghhBBCCCHSgBJsQgghhBBCCCGEAZRgE0IIIYQQQgghDKAEmxBCCCGEEEIIYQC16SKEEELeE0VFRYiPj0dubi7q6+sxYMAAmJqaYvHixRgyZAgAwMnJCQCQkJDAZqiEEEIIJ9EMNiGEEPIeSExMxLx581BXVwcvLy/88MMPWLJkCXJzczFr1iz89ttvbIdICCGEcB616SKEEEKk3JUrV+Dk5ITPP/8cGzZsEDlXX18POzs79OrVC6mpqTSDTQghhLQDLREnhBBCpFxMTAx69uwJT09PsXP9+vXDunXrcPPmTTQ2Noqdr6+vR0REBDIzM/G///0P3bt3x5gxY+Dt7Y3BgwcDAO7evYvt27fj6tWrePbsGXR0dLBs2TJMnDgRAPD8+XN8++23OH36NOrr6zF48GDMnTsXixYt6tgXTgghhEgYJdiEEEKIFBMIBDh//jwmTZqEbt26tfmYKVOmvPbvLl68GH/++Se8vLwwYMAA3LhxA+Hh4fDx8UFsbCz4fD4WL16MAQMGIDAwEHJycoiPj8eyZcuQnp4OVVVV+Pn54fz581i7di369++Ps2fPIiAgAH369IGdnV1HvnxCCCFEoijBJoQQQqTYw4cP8fz5c+Fs89v4448/0K1bN6xduxajR48GAIwdOxa///47Dhw4AACoq6tDRUUFlixZIpyxHjlyJHbu3Innz58DAC5fvgwzMzNYW1sLn6N79+7o27cvEy+REEII6TQowSaEEEKkmIzMy3qmLS0tb/13Bw4ciPj4eADAvXv3cPv2bVRUVODq1atobm4GAPTv3x+ampr45ptvkJOTgwkTJuCTTz6Bt7e38HnGjh2LAwcOoLa2FhYWFpg4cSLc3NwYeHWEEEJI50IJNiGEECLF+vTpgx49euDevXuvfcyTJ0/Q1NSEPn36iJ1LTU1FSEgI7t+/jz59+kBHRwddu3YVnufxeIiNjUVUVBROnTqF5ORkyMvLw8rKCps3b0afPn2wYcMGKCsrIzU1FVu2bAEAGBoawsfHBx9//DHjr5kQQghhC7XpIoQQQqTcJ598gkuXLgmXbP/d0aNHYWpqivz8fJHjeXl5WLt2Lf7zn/8gKysLly5dwt69e2FgYCDyuIEDB2Lz5s04f/48UlJS4OrqipMnTyI0NBQAoKCggKVLl+KXX35BRkYGfHx8cPfuXXh5eXXI6yWEEELYQgk2IYQQIuUWLVqER48eCRPeV9XV1eG///0vVFVVxRLn/Px88Pl8rFixAsrKygBeLjXPyckBAPD5fOTn58PMzAzXrl0Dj8fD8OHD4eHhAW1tbdTU1ODZs2f49NNPERsbCwAYNGgQPv/8c1hbW6OmpqZjXzghhBAiYbREnBBCCJFyBgYGWLlyJcLCwlBRUQFbW1v07dsX5eXliI2NxePHjxEdHQ0ejyfy90aOHAkA2Lp1K2bNmoWGhgbs27cPJSUlAF4uLf/444/RtWtXrFmzBsuXL0f//v2Rk5ODGzduYMGCBejatSt0dXWxc+dOyMvL46OPPkJVVRWSk5Px6aefSvzfghBCCOlIPIFAIGA7CEIIIYR0vKysLCQmJuLGjRt49OgRlJWVYWpqiiVLlmDQoEEAACcnJwBAQkICACAxMRFxcXGora1F//79MXbsWFhZWcHNzQ3R0dGYOHEibt26hR07duDKlStoaGiAmpoanJycYG9vDwBobGxEWFgYTp8+jf/9739QUlLCtGnTsHLlSpH93IQQQgjXUYJNCCGEEEIIIYQwgPZgE0IIIYQQQgghDKAEmxBCCCGEEEIIYQAl2IQQQgghhBBCCAMowSaEEEIIIYQQQhhACTYhhBBCCCGEEMIASrAJIYQQQgghhBAGUIJNCCGEEEIIIYQwgBJsQgghhBBCCCGEAZRgE0IIIYQQQgghDKAEmxBCCCGEEEIIYQAl2IQQQgghhBBCCAMowSaEEEIIIYQQQhjwf4uoGxe76eLdAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# TRAIN / TEST SPLIT\n",
    "# -----------------------------\n",
    "X = df.drop(columns=[target_col])\n",
    "y = df[target_col]\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X, y,\n",
    "    test_size=0.2,\n",
    "    stratify=y,\n",
    "    random_state=RANDOM_STATE\n",
    ")\n",
    "\n",
    "print(\"\\nTrain set shape:\", X_train.shape)\n",
    "print(\"Test set shape: \", X_test.shape)\n",
    "\n",
    "print(\"\\nTrain label distribution:\")\n",
    "print(y_train.value_counts())\n",
    "\n",
    "print(\"\\nTest label distribution:\")\n",
    "print(y_test.value_counts())\n",
    "\n",
    "plt.figure(figsize=(10, 4))\n",
    "y_train.value_counts().plot(kind=\"bar\")\n",
    "plt.title(\"Class Distribution, training set\")\n",
    "plt.xlabel(\"Class\")\n",
    "plt.ylabel(\"Count\")\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "plt.figure(figsize=(10, 4))\n",
    "y_test.value_counts().plot(kind=\"bar\")\n",
    "plt.title(\"Class Distribution, test set\")\n",
    "plt.xlabel(\"Class\")\n",
    "plt.ylabel(\"Count\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "239afe78-dfdc-4fbd-8654-ac7878040881",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Categorical columns used: ['protocol_type', 'service', 'flag']\n",
      "Number of numeric columns: 38\n"
     ]
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# CATEGORICAL / NUMERIC COLUMNS\n",
    "# -----------------------------\n",
    "expected_cats = [\"protocol_type\", \"service\", \"flag\"]\n",
    "categorical_cols = [c for c in expected_cats if c in X_train.columns]\n",
    "\n",
    "missing_cats = [c for c in expected_cats if c not in X_train.columns]\n",
    "if missing_cats:\n",
    "    print(\"\\nWARNING: Missing expected categorical columns:\", missing_cats)\n",
    "    print(\"Your dataset may have different column names. Adjust 'expected_cats' accordingly.\")\n",
    "\n",
    "numeric_cols = [col for col in X_train.columns if col not in categorical_cols]\n",
    "\n",
    "print(\"\\nCategorical columns used:\", categorical_cols)\n",
    "print(\"Number of numeric columns:\", len(numeric_cols))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "370b2742-0530-41ec-8dce-2479d4d5904a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# -----------------------------\n",
    "# PREPROCESSOR (DENSE OUTPUT)\n",
    "# GradientBoostingClassifier often fails with sparse matrices.\n",
    "# This forces OneHotEncoder + ColumnTransformer to output dense.\n",
    "# -----------------------------\n",
    "try:\n",
    "    categorical_transformer = OneHotEncoder(handle_unknown=\"ignore\", sparse_output=False)\n",
    "except TypeError:\n",
    "    # for older scikit-learn\n",
    "    categorical_transformer = OneHotEncoder(handle_unknown=\"ignore\", sparse=False)\n",
    "\n",
    "numeric_transformer = StandardScaler()\n",
    "\n",
    "preprocessor = ColumnTransformer(\n",
    "    transformers=[\n",
    "        (\"cat\", categorical_transformer, categorical_cols),\n",
    "        (\"num\", numeric_transformer, numeric_cols),\n",
    "    ],\n",
    "    sparse_threshold=0.0  # force dense output overall\n",
    ")\n",
    "\n",
    "# -----------------------------\n",
    "# MODELS + PIPELINES\n",
    "# -----------------------------\n",
    "rf_clf = RandomForestClassifier(\n",
    "    random_state=RANDOM_STATE,\n",
    "    n_jobs=-1\n",
    ")\n",
    "\n",
    "gb_clf = GradientBoostingClassifier(\n",
    "    random_state=RANDOM_STATE\n",
    ")\n",
    "\n",
    "rf_pipeline = Pipeline(steps=[\n",
    "    (\"preprocess\", preprocessor),\n",
    "    (\"model\", rf_clf)\n",
    "])\n",
    "\n",
    "gb_pipeline = Pipeline(steps=[\n",
    "    (\"preprocess\", preprocessor),\n",
    "    (\"model\", gb_clf)\n",
    "])\n",
    "\n",
    "\n",
    "# -----------------------------\n",
    "# HYPERPARAMETER SEARCH\n",
    "# (You can increase n_iter/cv later once everything runs end-to-end.)\n",
    "# -----------------------------\n",
    "rf_param_dist = {\n",
    "    \"model__n_estimators\": [100, 200, 300],\n",
    "    \"model__max_depth\": [None, 10, 20, 30],\n",
    "    \"model__min_samples_split\": [2, 5, 10],\n",
    "    \"model__min_samples_leaf\": [1, 2, 4],\n",
    "    \"model__max_features\": [\"sqrt\", \"log2\"],\n",
    "}\n",
    "\n",
    "gb_param_dist = {\n",
    "    \"model__n_estimators\": [100, 200],\n",
    "    \"model__learning_rate\": [0.01, 0.05, 0.1],\n",
    "    \"model__max_depth\": [3, 5],\n",
    "    \"model__subsample\": [0.6, 0.8, 1.0],\n",
    "}\n",
    "\n",
    "rf_search = RandomizedSearchCV(\n",
    "    estimator=rf_pipeline,\n",
    "    param_distributions=rf_param_dist,\n",
    "    n_iter=15,\n",
    "    scoring=\"f1_macro\",\n",
    "    cv=3,\n",
    "    verbose=1,\n",
    "    n_jobs=-1,\n",
    "    random_state=RANDOM_STATE\n",
    ")\n",
    "\n",
    "gb_search = RandomizedSearchCV(\n",
    "    estimator=gb_pipeline,\n",
    "    param_distributions=gb_param_dist,\n",
    "    n_iter=15,\n",
    "    scoring=\"f1_macro\",\n",
    "    cv=3,\n",
    "    verbose=1,\n",
    "    n_jobs=-1,\n",
    "    random_state=RANDOM_STATE\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "473cb4b3-e178-47d7-90f2-82d9fca21245",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "========================\n",
      " TUNING RANDOM FOREST \n",
      "========================\n",
      "Fitting 3 folds for each of 15 candidates, totalling 45 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\paul_\\anaconda3\\Lib\\site-packages\\sklearn\\model_selection\\_split.py:776: UserWarning: The least populated class in y has only 2 members, which is less than n_splits=3.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Best RF hyperparameters:\n",
      "{'model__n_estimators': 200, 'model__min_samples_split': 2, 'model__min_samples_leaf': 1, 'model__max_features': 'log2', 'model__max_depth': 30}\n",
      "Best RF CV F1-macro: 0.6209557029492677\n",
      "RF tuning time: 51.70 seconds\n"
     ]
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# TRAIN / TUNE RANDOM FOREST\n",
    "# -----------------------------\n",
    "print(\"\\n========================\")\n",
    "print(\" TUNING RANDOM FOREST \")\n",
    "print(\"========================\")\n",
    "\n",
    "start_time = time.perf_counter()\n",
    "rf_search.fit(X_train, y_train)\n",
    "rf_time = time.perf_counter() - start_time\n",
    "\n",
    "print(\"\\nBest RF hyperparameters:\")\n",
    "print(rf_search.best_params_)\n",
    "print(\"Best RF CV F1-macro:\", rf_search.best_score_)\n",
    "print(f\"RF tuning time: {rf_time:.2f} seconds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ffb772de-0f69-42c1-8d7a-a4e13175d096",
   "metadata": {},
   "outputs": [],
   "source": [
    "# -----------------------------\n",
    "# TRAIN / TUNE GRADIENT BOOSTING\n",
    "# -----------------------------\n",
    "print(\"\\n===========================\")\n",
    "print(\" TUNING GRADIENT BOOSTING \")\n",
    "print(\"===========================\")\n",
    "\n",
    "start_time = time.perf_counter()\n",
    "gb_search.fit(X_train, y_train)\n",
    "gb_time = time.perf_counter() - start_time\n",
    "\n",
    "print(\"\\nBest GB hyperparameters:\")\n",
    "print(gb_search.best_params_)\n",
    "print(\"Best GB CV F1-macro:\", gb_search.best_score_)\n",
    "print(f\"GB tuning time: {gb_time:.2f} seconds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7bac66d3-9a97-4a46-bab0-763b031c2863",
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'RandomizedSearchCV' object has no attribute 'best_estimator_'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[14], line 5\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# -----------------------------\u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;66;03m# EVALUATION\u001b[39;00m\n\u001b[0;32m      3\u001b[0m \u001b[38;5;66;03m# -----------------------------\u001b[39;00m\n\u001b[0;32m      4\u001b[0m best_rf \u001b[38;5;241m=\u001b[39m rf_search\u001b[38;5;241m.\u001b[39mbest_estimator_\n\u001b[1;32m----> 5\u001b[0m best_gb \u001b[38;5;241m=\u001b[39m gb_search\u001b[38;5;241m.\u001b[39mbest_estimator_\n\u001b[0;32m      7\u001b[0m rf_train_pred \u001b[38;5;241m=\u001b[39m best_rf\u001b[38;5;241m.\u001b[39mpredict(X_train)\n\u001b[0;32m      8\u001b[0m rf_test_pred  \u001b[38;5;241m=\u001b[39m best_rf\u001b[38;5;241m.\u001b[39mpredict(X_test)\n",
      "\u001b[1;31mAttributeError\u001b[0m: 'RandomizedSearchCV' object has no attribute 'best_estimator_'"
     ]
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# EVALUATION\n",
    "# -----------------------------\n",
    "best_rf = rf_search.best_estimator_\n",
    "best_gb = gb_search.best_estimator_\n",
    "\n",
    "rf_train_pred = best_rf.predict(X_train)\n",
    "rf_test_pred  = best_rf.predict(X_test)\n",
    "\n",
    "gb_train_pred = best_gb.predict(X_train)\n",
    "gb_test_pred  = best_gb.predict(X_test)\n",
    "\n",
    "def print_model_performance(name, y_tr, y_tr_pred, y_te, y_te_pred):\n",
    "    print(f\"\\n========== {name} PERFORMANCE ==========\")\n",
    "    train_acc = accuracy_score(y_tr, y_tr_pred)\n",
    "    test_acc  = accuracy_score(y_te, y_te_pred)\n",
    "    print(\"Train accuracy:\", train_acc)\n",
    "    print(\"Test  accuracy:\", test_acc)\n",
    "\n",
    "    print(\"\\nClassification report (Test):\")\n",
    "    print(classification_report(y_te, y_te_pred))\n",
    "\n",
    "    return train_acc, test_acc\n",
    "\n",
    "rf_train_acc, rf_test_acc = print_model_performance(\n",
    "    \"Random Forest\", y_train, rf_train_pred, y_test, rf_test_pred\n",
    ")\n",
    "\n",
    "gb_train_acc, gb_test_acc = print_model_performance(\n",
    "    \"Gradient Boosting\", y_train, gb_train_pred, y_test, gb_test_pred\n",
    ")\n",
    "\n",
    "plt.figure(figsize=(10, 4))\n",
    "rf_train_acc.value_counts().plot(kind=\"bar\")\n",
    "rf_test_acc.value_counts().plot(kind=\"bar\")\n",
    "gb_train_acc.value_counts().plot(kind=\"bar\")\n",
    "gb_test_acc.value_counts().plot(kind=\"bar\")\n",
    "plt.title(\"Class Distribution\")\n",
    "plt.xlabel(\"Class\")\n",
    "plt.ylabel(\"Count\")\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c8014daf-12fe-4aeb-8780-c272799d89e9",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'rf_test_pred' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[13], line 6\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# -----------------------------\u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;66;03m# CONFUSION MATRICES\u001b[39;00m\n\u001b[0;32m      3\u001b[0m \u001b[38;5;66;03m# -----------------------------\u001b[39;00m\n\u001b[0;32m      4\u001b[0m labels_sorted \u001b[38;5;241m=\u001b[39m \u001b[38;5;28msorted\u001b[39m(y\u001b[38;5;241m.\u001b[39munique())\n\u001b[1;32m----> 6\u001b[0m rf_cm \u001b[38;5;241m=\u001b[39m confusion_matrix(y_test, rf_test_pred, labels\u001b[38;5;241m=\u001b[39mlabels_sorted)\n\u001b[0;32m      7\u001b[0m gb_cm \u001b[38;5;241m=\u001b[39m confusion_matrix(y_test, gb_test_pred, labels\u001b[38;5;241m=\u001b[39mlabels_sorted)\n\u001b[0;32m      9\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mplot_confusion_matrix\u001b[39m(cm, classes, title):\n",
      "\u001b[1;31mNameError\u001b[0m: name 'rf_test_pred' is not defined"
     ]
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# CONFUSION MATRICES\n",
    "# -----------------------------\n",
    "labels_sorted = sorted(y.unique())\n",
    "\n",
    "rf_cm = confusion_matrix(y_test, rf_test_pred, labels=labels_sorted)\n",
    "gb_cm = confusion_matrix(y_test, gb_test_pred, labels=labels_sorted)\n",
    "\n",
    "def plot_confusion_matrix(cm, classes, title):\n",
    "    plt.figure(figsize=(10, 8))\n",
    "    sns.heatmap(cm, annot=False, fmt=\"d\", cmap=\"Blues\",\n",
    "                xticklabels=classes, yticklabels=classes)\n",
    "    plt.xlabel(\"Predicted label\")\n",
    "    plt.ylabel(\"True label\")\n",
    "    plt.title(title)\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "plot_confusion_matrix(rf_cm, labels_sorted, \"Random Forest - Confusion Matrix (Test)\")\n",
    "plot_confusion_matrix(gb_cm, labels_sorted, \"Gradient Boosting - Confusion Matrix (Test)\")\n",
    "\n",
    "print(\"\\nTraining/tuning time:\")\n",
    "print(f\"Random Forest:     {rf_time:.2f} seconds\")\n",
    "print(f\"Gradient Boosting: {gb_time:.2f} seconds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "adbb8f7f-bc20-46da-98c8-7f90a4483c3a",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'best_rf' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[12], line 10\u001b[0m\n\u001b[0;32m      7\u001b[0m     ohe_names \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlist\u001b[39m(ohe\u001b[38;5;241m.\u001b[39mget_feature_names_out(categorical_cols))\n\u001b[0;32m      8\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m ohe_names \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mlist\u001b[39m(numeric_cols)\n\u001b[1;32m---> 10\u001b[0m rf_model \u001b[38;5;241m=\u001b[39m best_rf\u001b[38;5;241m.\u001b[39mnamed_steps[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel\u001b[39m\u001b[38;5;124m\"\u001b[39m]\n\u001b[0;32m     11\u001b[0m gb_model \u001b[38;5;241m=\u001b[39m best_gb\u001b[38;5;241m.\u001b[39mnamed_steps[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel\u001b[39m\u001b[38;5;124m\"\u001b[39m]\n\u001b[0;32m     13\u001b[0m rf_feature_names \u001b[38;5;241m=\u001b[39m get_feature_names_from_pipeline(best_rf, categorical_cols, numeric_cols)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'best_rf' is not defined"
     ]
    }
   ],
   "source": [
    "# -----------------------------\n",
    "# FEATURE IMPORTANCE (SAFE)\n",
    "# -----------------------------\n",
    "def get_feature_names_from_pipeline(pipeline, categorical_cols, numeric_cols):\n",
    "    preprocess = pipeline.named_steps[\"preprocess\"]\n",
    "    ohe = preprocess.named_transformers_[\"cat\"]\n",
    "    ohe_names = list(ohe.get_feature_names_out(categorical_cols))\n",
    "    return ohe_names + list(numeric_cols)\n",
    "\n",
    "rf_model = best_rf.named_steps[\"model\"]\n",
    "gb_model = best_gb.named_steps[\"model\"]\n",
    "\n",
    "rf_feature_names = get_feature_names_from_pipeline(best_rf, categorical_cols, numeric_cols)\n",
    "gb_feature_names = get_feature_names_from_pipeline(best_gb, categorical_cols, numeric_cols)\n",
    "\n",
    "# Sanity checks (prevents silent length mismatch bugs)\n",
    "print(\"\\nFeature-name sanity checks:\")\n",
    "print(\"RF names vs importances:\", len(rf_feature_names), len(rf_model.feature_importances_))\n",
    "print(\"GB names vs importances:\", len(gb_feature_names), len(gb_model.feature_importances_))\n",
    "\n",
    "rf_importances = pd.DataFrame({\n",
    "    \"feature\": rf_feature_names,\n",
    "    \"importance\": rf_model.feature_importances_\n",
    "}).sort_values(by=\"importance\", ascending=False)\n",
    "\n",
    "gb_importances = pd.DataFrame({\n",
    "    \"feature\": gb_feature_names,\n",
    "    \"importance\": gb_model.feature_importances_\n",
    "}).sort_values(by=\"importance\", ascending=False)\n",
    "\n",
    "print(\"\\nTop 15 Random Forest features:\")\n",
    "display(rf_importances.head(15))\n",
    "\n",
    "top_n = 15\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.barplot(data=rf_importances.head(top_n), x=\"importance\", y=\"feature\")\n",
    "plt.title(f\"Random Forest - Top {top_n} Feature Importances\")\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "print(\"\\nTop 15 Gradient Boosting features:\")\n",
    "display(gb_importances.head(15))\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.barplot(data=gb_importances.head(top_n), x=\"importance\", y=\"feature\")\n",
    "plt.title(f\"Gradient Boosting - Top {top_n} Feature Importances\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e0b25b2e-62b8-44cf-b22a-5451efb4a7ce",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:base] *",
   "language": "python",
   "name": "conda-base-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}