analytics.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sqlite3\n",
    "import json\n",
    "import pandas as pd\n",
    "DB_PATH = 'data/experiments.db'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[243, 242]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_last_experiments_ids(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_last_experiments_ids(n):\n",
    "    with sqlite3.connect(DB_PATH) as conn:\n",
    "        # SQL query to get the last n experiment IDs\n",
    "        query = f\"\"\"\n",
    "        SELECT experiment_id\n",
    "        FROM Experiment\n",
    "        ORDER BY experiment_id DESC\n",
    "        LIMIT {n}\n",
    "        \"\"\"\n",
    "\n",
    "        # Execute the query and fetch results\n",
    "        cursor = conn.execute(query)\n",
    "        experiment_ids = cursor.fetchall()\n",
    "\n",
    "    # Flatten the list of tuples and return the experiment IDs\n",
    "    return [id_tuple[0] for id_tuple in experiment_ids]\n",
    "\n",
    "\n",
    "def get_experiments_details(experiment_ids):\n",
    "    with sqlite3.connect(DB_PATH) as conn:\n",
    "        # Convert list of experiment_ids to a string format suitable for the SQL query\n",
    "        formatted_ids = ','.join(map(str, experiment_ids))\n",
    "\n",
    "        # SQL query to get the desired information\n",
    "        query = f\"\"\"\n",
    "        SELECT \n",
    "            E.experiment_id, \n",
    "            E.start_time, \n",
    "            E.config_id, \n",
    "            json_extract(EC.agent_config, '$.agent_id') AS agent_id,\n",
    "            json_extract(EC.agent_config, '$.matrices_mode') AS matrices_mode,\n",
    "            json_extract(EC.gridlink_config, '$.grid_shape') AS grid_shape,\n",
    "            json_extract(EC.gridlink_config, '$.n_predictable_cycles') AS n_predictable_cycles,\n",
    "            json_extract(EC.gridlink_config, '$.n_unpredictable_cycles') AS n_unpredictable_cycles,\n",
    "            json_extract(EC.agent_config, '$.n_obs') AS n_obs,\n",
    "            json_extract(EC.agent_config, '$.n_states') AS n_states,\n",
    "            (SELECT COUNT(*) FROM Steps WHERE Steps.experiment_id = E.experiment_id) AS steps_count\n",
    "        FROM Experiment AS E\n",
    "        JOIN ExperimentConfig AS EC ON E.config_id = EC.config_id\n",
    "        WHERE E.experiment_id IN ({formatted_ids})\n",
    "        \"\"\"\n",
    "\n",
    "        # Execute the query and fetch results directly into a pandas DataFrame\n",
    "        df = pd.read_sql_query(query, conn)\n",
    "\n",
    "    # Return the DataFrame\n",
    "    return df\n",
    "\n",
    "def prepare_experiment_data(experiment_id):\n",
    "    with sqlite3.connect(DB_PATH) as conn:\n",
    "        # Query to get the steps for the given experiment_id\n",
    "        query = f\"\"\"\n",
    "        SELECT \n",
    "            step_num, \n",
    "            agent_action, \n",
    "            environment_state\n",
    "        FROM Steps\n",
    "        WHERE experiment_id = {experiment_id}\n",
    "        ORDER BY step_num\n",
    "        \"\"\"\n",
    "\n",
    "        # Execute the query and fetch results directly into a pandas DataFrame\n",
    "        df = pd.read_sql_query(query, conn)\n",
    "\n",
    "\n",
    "        # Deserialize the agent_action binary data and map it to 0 or 1\n",
    "        df['agent_action'] = df['agent_action'].apply(lambda x: 1 if x == b'\\x00\\x00\\x00\\x00\\x00\\x00\\xf0?' else 0)\n",
    "\n",
    "        # Parse the environment_state JSON string and create new columns\n",
    "        df['status'] = df['environment_state'].apply(lambda x: json.loads(x)['status'])\n",
    "        df['score'] = df['environment_state'].apply(lambda x: json.loads(x)['rally'])\n",
    "        df['ball_y'] = df['environment_state'].apply(lambda x: json.loads(x)['ball_y'])\n",
    "        df['paddle_y'] = df['environment_state'].apply(lambda x: json.loads(x)['paddle_y'])\n",
    "\n",
    "        # Calculate the ball_paddle_alignment\n",
    "        df['ball_paddle_alignment'] = df['paddle_y'] - df['ball_y']\n",
    "\n",
    "        # Drop the original environment_state column\n",
    "        df.drop('environment_state', axis=1, inplace=True)\n",
    "        df.drop(['ball_y','paddle_y'], axis=1, inplace=True)\n",
    "        # df.drop(['paddle_y','ball_y'],inplace=True)\n",
    "\n",
    "    # Return the DataFrame with the new structure\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>experiment_id</th>\n",
       "      <th>start_time</th>\n",
       "      <th>config_id</th>\n",
       "      <th>agent_id</th>\n",
       "      <th>matrices_mode</th>\n",
       "      <th>grid_shape</th>\n",
       "      <th>n_predictable_cycles</th>\n",
       "      <th>n_unpredictable_cycles</th>\n",
       "      <th>n_obs</th>\n",
       "      <th>n_states</th>\n",
       "      <th>steps_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>100</td>\n",
       "      <td>2023-11-04 14:06:34</td>\n",
       "      <td>46</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>10</td>\n",
       "      <td>400</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>101</td>\n",
       "      <td>2023-11-04 14:10:05</td>\n",
       "      <td>47</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>10</td>\n",
       "      <td>400</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>102</td>\n",
       "      <td>2023-11-04 14:10:15</td>\n",
       "      <td>48</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>10</td>\n",
       "      <td>400</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>103</td>\n",
       "      <td>2023-11-04 14:10:37</td>\n",
       "      <td>49</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>10</td>\n",
       "      <td>400</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>104</td>\n",
       "      <td>2023-11-04 14:11:06</td>\n",
       "      <td>50</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>10</td>\n",
       "      <td>400</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>1275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>221</td>\n",
       "      <td>2023-11-07 09:24:16</td>\n",
       "      <td>90</td>\n",
       "      <td>radom</td>\n",
       "      <td>None</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>4</td>\n",
       "      <td>800</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>222</td>\n",
       "      <td>2023-11-07 09:24:47</td>\n",
       "      <td>91</td>\n",
       "      <td>random</td>\n",
       "      <td>None</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>4</td>\n",
       "      <td>800</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>50000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>223</td>\n",
       "      <td>2023-11-08 01:34:12</td>\n",
       "      <td>92</td>\n",
       "      <td>None</td>\n",
       "      <td>uniform</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>4</td>\n",
       "      <td>800</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>1718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>224</td>\n",
       "      <td>2023-11-08 01:35:17</td>\n",
       "      <td>93</td>\n",
       "      <td>None</td>\n",
       "      <td>uniform</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>10</td>\n",
       "      <td>400</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>50000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>225</td>\n",
       "      <td>2023-11-08 01:36:35</td>\n",
       "      <td>93</td>\n",
       "      <td>None</td>\n",
       "      <td>uniform</td>\n",
       "      <td>[2,3]</td>\n",
       "      <td>10</td>\n",
       "      <td>400</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>644</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>126 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     experiment_id           start_time  config_id agent_id matrices_mode   \n",
       "0              100  2023-11-04 14:06:34         46     None          None  \\\n",
       "1              101  2023-11-04 14:10:05         47     None          None   \n",
       "2              102  2023-11-04 14:10:15         48     None          None   \n",
       "3              103  2023-11-04 14:10:37         49     None          None   \n",
       "4              104  2023-11-04 14:11:06         50     None          None   \n",
       "..             ...                  ...        ...      ...           ...   \n",
       "121            221  2023-11-07 09:24:16         90    radom          None   \n",
       "122            222  2023-11-07 09:24:47         91   random          None   \n",
       "123            223  2023-11-08 01:34:12         92     None       uniform   \n",
       "124            224  2023-11-08 01:35:17         93     None       uniform   \n",
       "125            225  2023-11-08 01:36:35         93     None       uniform   \n",
       "\n",
       "    grid_shape  n_predictable_cycles  n_unpredictable_cycles  n_obs  n_states   \n",
       "0        [2,3]                    10                     400      3         2  \\\n",
       "1        [2,3]                    10                     400      8         2   \n",
       "2        [2,3]                    10                     400      8         3   \n",
       "3        [2,3]                    10                     400      8         3   \n",
       "4        [2,3]                    10                     400      8         8   \n",
       "..         ...                   ...                     ...    ...       ...   \n",
       "121      [2,3]                     4                     800      8         8   \n",
       "122      [2,3]                     4                     800      8         8   \n",
       "123      [2,3]                     4                     800      8         8   \n",
       "124      [2,3]                    10                     400      8         8   \n",
       "125      [2,3]                    10                     400      8         8   \n",
       "\n",
       "     steps_count  \n",
       "0           3961  \n",
       "1            130  \n",
       "2            354  \n",
       "3            732  \n",
       "4           1275  \n",
       "..           ...  \n",
       "121            0  \n",
       "122        50000  \n",
       "123         1718  \n",
       "124        50000  \n",
       "125          644  \n",
       "\n",
       "[126 rows x 11 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "experiment_ids = list(range(100,226))\n",
    "get_experiments_details(experiment_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "ex_255 = prepare_experiment_data(225)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculate_metrics(df):\n",
    "    # Calculate hits and misses\n",
    "    df['hits'] = (df['status'] == 1).cumsum()\n",
    "    df['misses'] = (df['status'] == -1).cumsum()\n",
    "\n",
    "    # Avoid division by zero for hit_miss_ratio\n",
    "    df['hit_miss_ratio'] = df['hits'] / df['misses'].replace({0: 1})\n",
    "\n",
    "    # Calculate long rallies\n",
    "    df['long_rallies'] = (df['score'] >= 3).cumsum()\n",
    "\n",
    "    # Calculate aces\n",
    "    df['aces'] = ((df['status'] == -1) & (df['score'].shift(1) == 0)).cumsum()\n",
    "\n",
    "    return df[['step_num', 'hits', 'misses', 'hit_miss_ratio', 'long_rallies', 'aces']]\n",
    "\n",
    "def calculate_metrics2(df):\n",
    "    # Calculate hits and misses\n",
    "    df['hits'] = (df['status'] == 1).cumsum()\n",
    "    df['misses'] = (df['status'] == -1).cumsum()\n",
    "\n",
    "    # Avoid division by zero for hit_miss_ratio\n",
    "    df['hit_miss_ratio'] = df.apply(lambda x: x['hits'] / x['misses'] if x['misses'] > 0 else 0, axis=1)\n",
    "\n",
    "    # Calculate long rallies\n",
    "    df['long_rallies'] = (df['score'] >= 3).cumsum()\n",
    "\n",
    "    # Calculate aces\n",
    "    df['previous_score'] = df['score'].shift(1, fill_value=0)\n",
    "    df['aces'] = ((df['status'] == -1) & (df['previous_score'] == 0)).cumsum()\n",
    "\n",
    "    # Select only the relevant columns to return\n",
    "    metrics_df = df[['step_num', 'hits', 'misses', 'hit_miss_ratio', 'long_rallies', 'aces']]\n",
    "    \n",
    "    return metrics_df\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='step_num', ylabel='hit_miss_ratio'>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAioAAAGwCAYAAACHJU4LAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACB5ElEQVR4nO3dd3gU5drA4d+W7KZXEmog9N57lV7Egp6jfIgKNiwgKvaOFY69o3gEPDawoUiVXqRJr1IDCYQQSnrf3fn+mGRLsqkkmU14bq9c2Xnn3dlnB5N98ladoigKQgghhBAeSK91AEIIIYQQRZFERQghhBAeSxIVIYQQQngsSVSEEEII4bEkURFCCCGEx5JERQghhBAeSxIVIYQQQngso9YBXAmbzUZcXBwBAQHodDqtwxFCCCFEKSiKQmpqKvXq1UOvL77NpFonKnFxcURGRmodhhBCCCHKITY2lgYNGhRbp1onKgEBAYD6RgMDAzWORgghhBClkZKSQmRkpP1zvDjVOlHJ7+4JDAyUREUIIYSoZkozbEMG0wohhBDCY0miIoQQQgiPJYmKEEIIITxWtR6jIoQoO6vVSm5urtZh1EheXl4YDAatwxCiRpFERYirhKIoxMfHk5SUpHUoNVpwcDB16tSRtZ2EqCCSqAhxlchPUiIiIvD19ZUP0gqmKAoZGRkkJCQAULduXY0jEqJmkERFiKuA1Wq1JylhYWFah1Nj+fj4AJCQkEBERIR0AwlRAWQwrRBXgfwxKb6+vhpHUvPl32MZByRExZBERYiriHT3VD65x0JULElUhBBCCOGxJFERQgghhMeSREUI4bFOnTqFTqdjz549WocihNCIJCpCCI8VGRnJuXPnaNeundahMH36dDp16qR1GEJUnZwMSDwFaRc0DUMSFSGER8rJycFgMFCnTh2MRllJQYgqt20WfNgRvvu3pmFIoiKEwGazMWPGDBo3boyPjw8dO3bk559/RlEUhg4dyogRI1AUBYDLly/ToEEDXnrpJQDWrVuHTqdjyZIldOjQAW9vb3r16sWBAwdcXmPTpk30798fHx8fIiMjmTp1Kunp6fbzUVFRvPbaa9x5550EBgYyadKkQl0/+a+1YsUKOnfujI+PD4MHDyYhIYFly5bRunVrAgMDue2228jIyCjx/eXLv+7q1avp1q0bvr6+9OnThyNHjgAwb948XnnlFfbu3YtOp0On0zFv3rzK+KcQQnuKAqtfg9WvqsdePlrHU30lJycrgJKcnKx1KEJ4tMzMTOXQoUNKZmam2/Ovv/660qpVK2X58uXKiRMnlLlz5ypms1lZt26dcubMGSUkJET54IMPFEVRlFtuuUXp0aOHkpubqyiKoqxdu1YBlNatWyt//vmnsm/fPuW6665ToqKilJycHEVRFOX48eOKn5+f8v777ytHjx5V/vrrL6Vz587KxIkT7TE0atRICQwMVN555x3l+PHjyvHjx5Xo6GgFUHbv3u3yWr169VI2bdqk7Nq1S2nWrJlyzTXXKMOHD1d27dqlbNiwQQkLC1NmzpxZqvfnfN2ePXsq69atUw4ePKj0799f6dOnj6IoipKRkaE8/vjjStu2bZVz584p586dUzIyMsp1r4XwaJZcRflyiKK8HOj4+mdphb9MWT6/JVER4ipQ3IdnVlaW4uvrq2zevNml/J577lHGjRunKIqi/Pjjj4q3t7fyzDPPKH5+fsrRo0ft9fI/5OfPn28vu3TpkuLj46MsWLDAfq1Jkya5XH/jxo2KXq+3x9SoUSNlzJgxLnWKSlRWrVplrzNjxgwFUE6cOGEvu//++5URI0aU+v25u+6SJUsUwB7fyy+/rHTs2LHQ/StIEhVRbSWfdU1QXg5UlHP7K+elyvD5LR2/Qlzljh8/TkZGBsOGDXMpz8nJoXPnzgDccsstLFy4kJkzZzJr1iyaN29e6Dq9e/e2Pw4NDaVly5YcPnwYgL1797Jv3z6+++47ex1FUbDZbERHR9O6dWsAunXrVqqYO3ToYH9cu3ZtfH19adKkiUvZ9u3bS/3+3F03f6+ehIQEGjZsWKq4hKi2Dv0OP97pOPavAw9tAd9Q7WLKI4mKEFe5tLQ0AJYsWUL9+vVdzpnNZgAyMjLYuXMnBoOBY8eOles17r//fqZOnVronHMS4OfnV6rreXl52R/rdDqX4/wym81mf20o/v0VdV3Afh0haiRFgSXTYMccR1mvh2DEm+AhqyxLoiLEVa5NmzaYzWZiYmK45ppr3NZ5/PHH0ev1LFu2jGuvvZbRo0czePBglzpbt261Jx2JiYkcPXrU3lLSpUsXDh06RLNmzSr3zbhRmvdXGiaTCavVWoGRCaGx3CyY1Rsun3SU3f4rNBuiXUxuSKIixFUuICCAJ554gsceewybzUa/fv1ITk7mr7/+IjAwkFq1ajFnzhy2bNlCly5dePLJJ5kwYQL79u0jJCTEfp1XX32VsLAwateuzfPPP0+tWrUYM2YMAE8//TS9evViypQp3Hvvvfj5+XHo0CFWrlzJJ598oun7mzBhQqmuExUVRXR0NHv27KFBgwYEBAQUapERotq4dAI+7uJaNmUH1Crcras1mZ4shOC1117jxRdfZMaMGbRu3ZqRI0eyZMkSoqKiuOeee5g+fTpduqi/1F555RVq167NAw884HKNmTNn8sgjj9C1a1fi4+P5448/MJlMgDr2Y/369Rw9epT+/fvTuXNnXnrpJerVq6fp+2vcuHGpr/Gvf/2LkSNHMmjQIMLDw/nhhx8qMWIhKtHOr12TlFot4ZkYj0xSAHSKkrc4QjWUkpJCUFAQycnJBAYGah2OEB4rKyuL6OhoGjdujLe3d4Vee926dQwaNIjExESCg4Mr9NrVUWXeayGuiNUCP98Fhxc5yq55BgY+U+XjUcry+S1dP0IIIURNl50KH7SHzERH2V3LoVHvop/jISRREUIIIWqyc3vhiwGuZY/shZAoTcIpK0lUhBBXZODAgVTjHmQhai5Fgb8+hFUvO8oie8IdC8FUuqUAPIEkKkIIIURNY8mGb/8FpzY6ykbOhF4PahdTOUmiIoQQQtQk6ZfgneagOK37c99aqN+l6Od4MElUhBBCiJri1F8w71rHsd4LHt0HgVWzFEBlkHVUhBBCiOrOZoVV012TlOYj4Nkz1TpJAWlREUIIIaq37DT4ajgkHHSU3fgZdB6vXUwVSBIVIYQQorpKiYP3WruWPbQVIlq7r18NSdePEKJamjFjBt27dycgIICIiAjGjBnDkSNHXOrMnj2bgQMHEhgYiE6nIykpqdB1jh49yo033kitWrUIDAykX79+rF27torehRBX4Mgy1yTFJxSePFGjkhTQOFGJiopCp9MV+po8ebKWYQkhqoH169czefJktm7dysqVK8nNzWX48OGkp6fb62RkZDBy5Eiee+65Iq9z3XXXYbFYWLNmDTt37qRjx45cd911xMfHV8XbEKLsrLnwxyPww/85yjr8Hzx+BPxqaRdXJdF0r58LFy64bJt+4MABhg0bxtq1axk4cGCJz5e9foQonath/5kLFy4QERHB+vXrGTDAdRXOovYjunjxIuHh4WzYsIH+/fsDkJqaSmBgICtXrmTo0KFljuNquNdCQ5mJMKsfpJxxlI39Flpfr11M5VBt9voJDw93OZ45cyZNmzblmmuu0SgiIQSAoihk5lpLrljBfLwM6Mq5OVpycjIAoaGhpX5OWFgYLVu25H//+x9dunTBbDbzxRdfEBERQdeuXcsVhxCV5tIJ112PAabuhtAm2sRTRTxmMG1OTg7ffvst06ZNK/IXVXZ2NtnZ2fbjlJSUqgrPLsdi45bPN3M4PpW7+kTx7LU1qy9QCIDMXCttXlpR5a976NUR+JrK/mvJZrPx6KOP0rdvX9q1a1fq5+l0OlatWsWYMWMICAhAr9cTERHB8uXLCQkJKXMcQlSavQtg4STHcUhjmLQOfIK1iqjKeMxg2t9++42kpCQmTpxYZJ0ZM2YQFBRk/4qMjKy6APPEXM5g75lkciw2luw/V+WvL4QobPLkyRw4cID58+eX6XmKojB58mQiIiLYuHEj27dvZ8yYMVx//fWcOyc/38IDWLJhwR2uSUrPB2DKjqsiSQEPalH56quvGDVqFPXqFb0wzbPPPsu0adPsxykpKRokK7L5mqj5fLwMHHp1hCavW1ZTpkxh8eLFbNiwgQYNGpTpuWvWrGHx4sUkJiba+8k/++wzVq5cyddff80zzzxT5niEqDBpF+CjzpCT6ii74zdoOkizkLTgEYnK6dOnWbVqFb/++mux9cxmM2azuYqiEuLqpdPpytUFU5UUReHhhx9m4cKFrFu3jsaNG5f5GhkZGQDo9a6Ny3q9HpvNViFxClEu5w/BrN6uZY8dgqD62sSjIY/4TTR37lwiIiIYPXq01qGUSHazF8IzTJ48me+//57ff/+dgIAA+3TioKAgfHx8AIiPjyc+Pp7jx48DsH//fgICAmjYsCGhoaH07t2bkJAQJkyYwEsvvYSPjw9ffvkl0dHR1eL3kaiBFAW2fwnLnnSU1esMExaD2V+7uDSk+RgVm83G3LlzmTBhAkajR+RNxZI8RQjPMGvWLJKTkxk4cCB169a1fy1YsMBe5/PPP6dz587cd999AAwYMIDOnTuzaNEiAGrVqsXy5ctJS0tj8ODBdOvWjU2bNvH777/TsWNHTd6XuIrlZMDX17smKQOfg3vXXLVJCnhAi8qqVauIiYnh7rvv1jqUMpPWFSG0U5oloKZPn8706dOLrdOtWzdWrKj6GU5CuEg+C++3cS27dzU06KZNPB5E80Rl+PDhpfqF4ymqUahCCCGqg7M74cvBjmODGaYdqpGrzJaH5l0/1Y0inT9CCCEqgqLA+rddk5SmQ+DZWElSnGjeoiKEEEJcdbJSYO61cH6/o2zUW9BjEpRzdeaaShKVMpKuHyGEEFfk8kl1fRRnD22DiFbaxOPhJFEpI0lUhBBClNupTTDPaeq7Xzg8vBO8g7SLycPJGJUykjEqQgghysxmgxXPuyYp7f4N0/6RJKUE0qIihBBCVKaMy/DFAEiOdZTd/CV0uFW7mKoRSVTKSLp+hBBClFrCP/BZT9eyR/dDcENt4qmGpOtHCCGEqAxHV7gmKbVawPPxkqSUkbSoXIHqtFCdEEKIKmKzwqKpsOdbR1mP+2HkTNBL+0BZyR0rI8lNhPAMM2bMoHv37gQEBBAREcGYMWM4cuSIS53Zs2czcOBAAgMD0el0JCUlFbrOrl27GDZsGMHBwYSFhTFp0iTS0tKq6F2IGictAf7T2DVJue1HuPYtSVLKSe5aGcmsHyE8w/r165k8eTJbt25l5cqV5ObmMnz4cNLT0+11MjIyGDlyJM8995zba8TFxTF06FCaNWvGtm3bWL58OQcPHmTixIlV9C5EjXJuL7zTHLKTHWVPHIcWI7SLqQaQrh8hRLW0fPlyl+N58+YRERHBzp07GTBgAACPPvooAOvWrXN7jcWLF+Pl5cWnn36KPu+v3c8//5wOHTpw/PhxmjVrVmnxixrmwC/ws9PmupG9YOJiMHhpF1MNIYlKGUnXj7gqKArkZlT963r5lnv58ORk9a/Y0NDQUj8nOzsbk8lkT1IAfHx8ANi0aZMkKqJk1lz4aSL8s9hRNvA5uOYpWQq/gkiiUkZKEY+FqFFyM+DNelX/us/FgcmvzE+z2Ww8+uij9O3bl3bt2pX6eYMHD2batGm8/fbbPPLII6Snp/PMM88AcO7cuTLHIa4yyWfh/ba4fBrctQwa9dEspJpIxqgIIaq9yZMnc+DAAebPn1+m57Vt25avv/6ad999F19fX+rUqUPjxo2pXbu2SyuLEIXEbIP322BPUoze8EysJCmVQFpUykimJIurgpev2rqhxeuW0ZQpU1i8eDEbNmygQYMGZX7+bbfdxm233cb58+fx8/NDp9Px3nvv0aRJkzJfS1wldn4Nf0x1HLccDWO/lVk9lUQSlTKSNEVcFXS6cnXBVCVFUXj44YdZuHAh69ato3Hjxld0vdq1awMwZ84cvL29GTZsWEWEKWqS3Cz47t9waqOjbNTb0HOSdjFdBSRREUJUS5MnT+b777/n999/JyAggPj4eACCgoLsA2Lj4+OJj4/n+PHjAOzfv5+AgAAaNmxoH3T7ySef0KdPH/z9/Vm5ciVPPvkkM2fOJDg4WJP3JTxU4in4sKNr2f0boW4HTcK5mkiiUkbS8yOEZ5g1axYAAwcOdCmfO3eufR2Uzz//nFdeecV+Ln/asnOd7du38/LLL5OWlkarVq344osvuOOOOyo9flGNnFwP/7vBcRxYHx7eBV7e2sV0FZFEpcwcmYokLUJopzTjxaZPn8706dOLrfO///2vgiISNdLmj+HPFxzHXe6E6z+SqcdVSBIVIYQQoqCcdJgzEuL3Ocpu/hI63KpdTFcpSVTKSFpRhBCihrt4DD7p5lr28C4Ia6pNPFc5SVTKSPIUIYSowY6ugO+dWk3qdoJ7V4NBPi61IpO+hRBCCIC1b7omKf2mwf3rJUnRmNz9MnLu+pGdlIUQogbISobP+0PSaUfZuAXQcqR2MQk7SVTKSFamFUKIGiT+AHze17Vs2mEI1GCvK+GWdP0IIYS4Oh1c6JqkNB8OLydJkuJhpEWljKQ9RQghqjlFgeXPwrZZjrLhb0CfKdrFJIokiUoZSc+PEEJUYxmX4eOukHnZUTZxKUT1Lfo5QlOSqFwBSVqEEKIaObsTvhzsONYZ4Ilj4BemXUyiRDJGpYxkpo8QnmHGjBl0796dgIAAIiIiGDNmDEeOHLGfv3z5Mg8//DAtW7bEx8eHhg0bMnXqVJKTk+119u7dy7hx44iMjMTHx4fWrVvz4YcfavF2RGXb/Z1rktL5dnjpkiQp1YC0qJSV5ClCeIT169czefJkunfvjsVi4bnnnmP48OEcOnQIPz8/4uLiiIuL45133qFNmzacPn2aBx54gLi4OH7++WcAdu7cSUREBN9++y2RkZFs3ryZSZMmYTAYmDJFxivUCIoCC++HfQscZTd+Bp3HaxeTKBNJVIQQ1dLy5ctdjufNm0dERAQ7d+5kwIABtGvXjl9++cV+vmnTprzxxhvcfvvtWCwWjEYjd999t8s1mjRpwpYtW/j1118lUakJ0hLgnRa4/IU5aT3U66RVRKIcJFEpI2lQEVcDRVHItGRW+ev6GH3QlXNX2vwundDQ0GLrBAYGYjQW/asvOTm52GuIauLUXzDvWsexf22YsgO8A7WLSZSL5onK2bNnefrpp1m2bBkZGRk0a9aMuXPn0q1bt5KfrAHXlWmFqJkyLZn0/L5nlb/uttu24evlW+bn2Ww2Hn30Ufr27Uu7du3c1rl48SKvvfYakyZNKvI6mzdvZsGCBSxZsqTMMQgPsnUWLH/GcdzvMRg6XbNwxJXRNFFJTEykb9++DBo0iGXLlhEeHs6xY8cICQnRMiwhRDUzefJkDhw4wKZNm9yeT0lJYfTo0bRp04bp06e7rXPgwAFuvPFGXn75ZYYPH16J0YpKY7PCt/+Ck2sdZbd+A21u0C4mccU0TVT+85//EBkZydy5c+1ljRs31jCiksVcztA6BCEqnY/Rh223bdPkdctqypQpLF68mA0bNtCgQYNC51NTUxk5ciQBAQEsXLgQLy+vQnUOHTrEkCFDmDRpEi+88EK5YhcaS4qFDwq0pj20DSJaaROPqDCaJiqLFi1ixIgR3HLLLaxfv5769evz0EMPcd9997mtn52dTXZ2tv04JSWlqkIFICkjh+cW7q/S1xRCCzqdrlxdMFVJURQefvhhFi5cyLp169z+kZOSksKIESMwm80sWrQIb2/vQnUOHjzI4MGDmTBhAm+88UZVhC4q2pHl8MNYx3F4a7hvDZg8+/9hUTqarqNy8uRJZs2aRfPmzVmxYgUPPvggU6dO5euvv3Zbf8aMGQQFBdm/IiMjqzTeC6nZJVcSQlSJyZMn8+233/L9998TEBBAfHw88fHxZGaqg4BTUlIYPnw46enpfPXVV6SkpNjrWK1WQO3uGTRoEMOHD2fatGn28xcuXNDyrYmyWPOGa5Iy7FWYvFWSlBpEp2i4HbDJZKJbt25s3rzZXjZ16lT+/vtvtmzZUqi+uxaVyMhI+0j+ynb0fCrD399gP67lb2bHC0Mr/XWFuFJZWVlER0fTuHFjt60K1VFRs4Pmzp3LxIkTWbduHYMGDXJbJzo6mqioKKZPn84rr7xS6HyjRo04depUueKqiffaI1ly4MtBcP6Ao2z8z9B8mHYxiVJLSUkhKCioVJ/fmnb91K1blzZt2riUtW7d2mXtA2dmsxmz2VwVobklS+YL4TlK+htr4MCBJdaZPn16kYNrhQe7eAw+KTAz9OFdENZUm3hEpdK066dv374uS14DHD16lEaNGmkUkRBCCI+270fXJCWyJ7yQIElKDaZpi8pjjz1Gnz59ePPNN7n11lvZvn07s2fPZvbs2VqGVSTZ50cIITSiKLBkGuyY4ygb/S50v1e7mESV0DRR6d69OwsXLuTZZ5/l1VdfpXHjxnzwwQeMH++ZezBI148QQmggJwM+6gxp8Y6yCYuhcX/tYhJVRvOVaa+77jquu+46rcMQQgjhieJ2w+yBrmWP7ofghpqEI6qepmNUqpvCLSrSxCKqFw0n+V015B5XoG2zXZOUZsPgxYuSpFxlNG9RqU5kjIqorvJXY83IyMDHp+yrv4rSy8hQV692twKuKCVFgR/vgMN/OMrGfA6dxmkXk9CMJCpCXAUMBgPBwcEkJCQA4OvrW+5dioV7iqKQkZFBQkICwcHBGAwGrUOqnjKT4O2mYLM4yu5ZBZHdNQtJaEsSlTKQFl1RndWpUwfAnqyIyhEcHGy/16KMojfC105jFg0mdTxKgNzPq5kkKkJcJXQ6HXXr1iUiIoLc3Fytw6mRvLy8pCWlvNb9B9a96Thufwvc9AXo5X5e7SRRuQLSwiKqI4PBIB+mwnNYLWorSozTtim3/g/a3KhdTMKjSKJSBpKYCCFEBUo9D++2cC27fwPU7ahNPMIjSaJSBjLrRwghKsg/S2G+0ywe3zCYvB38amkXk/BIkqgIIYSoOooCy56G7V84yrrfC6PeBr0s7SUKk0SlDKTrRwghroAlGz7vDxedNqMd/zM0H6ZdTMLjSaJSBgXzFMlbhBCilBJPwYcFxp5M3g7hLTUJR1Qf0s4mhBCicu35wTVJCWkMT5+WJEWUirSolIHs4SGEEGVgs8Kvk+DAz46yftNgyEsgKyOLUpJEpQwkTRFCiFLKToOPOkO600rIE5dAVD/tYhLVkiQqQgghKtb5gzCrj2vZ1N0Q2kSbeES1JmNUyqBgz490BQkhhBNFgS2fuSYp9TrDs2ckSRHlJi0qZSKJiRBCuGXJge9vhZNrHWXDXoW+j2gXk6gRJFERQghxZTIuw7stwZrjKLt3DTToql1MosaQRKUMpKdHCCEKiNkGc4a7lj12CILqaxOPqHFkjEoZSJ4ihBB5bDZY+6ZrktJkEDx3TpIUUaGkRUUIIUTZ5KTD3FFwbq+j7PoPoetEzUISNZckKmVQaNaPNmEIIYR2Us7Be61cyx74C+q00yYeUeNJolIGMh1ZCHFVO7YKvvuX49gcBA/vAP8I7WISNZ4kKkIIIYpntcDyZ+DvLx1lbW+Gmz4Ho1m7uMRVQRKVMpD2FCHEVScrGb4YoO5+nO/fc6HdzZqFJK4ukqiUQX7Pj04nU5WFEFeBy9HwUSfXsod3QVhTTcIRVyeZnlwO+Xt+SrIihKixDvzimqQENYSnT0mSIqqctKiUgZLX+aOTJhUhRE1lyYbfHlQTlXzd7oFRb4FBPjJE1ZP/68oiv+tH2yiEEKJypF+Ej7tCVpKjbPwv0HyoZiEJIYlKOegkUxFC1DQJ/8BnPV3LHj0AwZHaxCNEHhmjUgbS2SOEqJF2zHFNUuq0h2fPSpIiPIK0qJSBY9aPDlBkATghRPWWmwnzb4MTaxxl/Z+AQc+B3qBdXEI4kUSlDOyDaTWOQwghrlhKHLzfFhSbo+zuFdCwl3YxCeGGJCrlIGNUhBDVWtxumD3QcawzwONHwD9cs5CEKIqmY1SmT5+OTqdz+WrVqlXJT9SI9PQIIao1RYFNH7gmKVH94bmzkqQIj6V5i0rbtm1ZtWqV/dho1DykIuXnKTrp/BFCVDfZqfD19WprSr7hb0DvydJMLDya5lmB0WikTp06paqbnZ1Ndna2/TglJaWywnLrmy2nAcfPtDSwCCGqhcRT8GFH17IH/oI67TQJR4iy0Hx68rFjx6hXrx5NmjRh/PjxxMTEFFl3xowZBAUF2b8iI6tu6lzMpQxWHT4PQKC3V5W9rhBCXJHTW1yTFJ8QePq0JCmi2tA0UenZsyfz5s1j+fLlzJo1i+joaPr3709qaqrb+s8++yzJycn2r9jY2CqLNT3HYn/8+hj5ARdCeDhFgVXTYe5IR1mbG+GJY+ATrFVUQpSZpl0/o0aNsj/u0KEDPXv2pFGjRvz444/cc889heqbzWbMZnNVhlhILX8zTSP8NY1BCCGKlZkIswdBYrSj7MZPodN4GY8iqh3Nx6g4Cw4OpkWLFhw/flzrUAqRGT9CiGrhwlH4tLtr2dTdENpEm3iEuEKaj1FxlpaWxokTJ6hbt67WoRTi2DnZpVAIITzH8dWuSUpIY3junCQpolrTNFF54oknWL9+PadOnWLz5s3cdNNNGAwGxo0bp2VYxdIhK9MKITyMzQZ/PArf3uwo63oXPLwLTL6ahSVERdC06+fMmTOMGzeOS5cuER4eTr9+/di6dSvh4Z638JB0/QghPFL6Rfi0B2RccpSN/Q5aX6ddTEJUIE0Tlfnz52v58uUi49CEEB4jfj983s+17PEjEFC6tamEqA48aoxKdSCr0gohPMKhRa5JSr0u8OJFSVJEjeNRs348mXT9CCE8gtUCv94LBxc6yvo/AYNfkCZfUSNJolJK7mb9SO4ihKhSKefgww5gzXGUTfgDGg/QLiYhKpkkKmWkQ/5oEUJo4MwO+O8Qx7FOD0+dVJfEF6IGu+IxKmfOnOHMmTMVEYtHk64fIYRm9vzgmqQ0Hw4vXpIkRVwVypWo2Gw2Xn31VYKCgmjUqBGNGjUiODiY1157DZvNVtExeoT8PEUnzSlCiKpiyYFvboLfHnCUjXgTxv8EepkLIa4O5er6ef755/nqq6+YOXMmffv2BWDTpk1Mnz6drKws3njjjQoNUgghrjpJMfBBe9ey+9ZC/S7axCOERsqVqHz99df897//5YYbbrCXdejQgfr16/PQQw/VyERFcdP3465MCCGu2KlNMG+049gvHB7ZJ6vMiqtSudoOL1++TKtWrQqVt2rVisuXL19xUJ7I0fUja6kIISrRttmuSUrH2+CJY5KkiKtWuRKVjh078sknnxQq/+STT+jYseMVB+XJZIiKEKJS5GbCl4Nh2ZOOsjGz4KZZ8otHXNXK1fXz1ltvMXr0aFatWkXv3r0B2LJlC7GxsSxdurRCA/QU0ssjhKg0l07AxwXGnkzeDuEttYlHCA9SrhaVa665hqNHj3LTTTeRlJREUlISN998M0eOHKF///4VHaOHyFvwTbp9hBAV6dgq1yQloo26FL4kKUIAV7DgW7169WrkoNmSyMq0QogKs+EdWPOa47jPwzD8de3iEcIDlTpR2bdvH+3atUOv17Nv375i63bo0OGKA/M0zl0/0l0shLgi2anqeJSLRx1lY7+D1tdpF5MQHqrUiUqnTp2Ij48nIiKCTp06odPp3E7P1el0WK3WCg3SE9hn/WgahRCi2ks4DJ/1ci179AAER2oTjxAertSJSnR0NOHh4fbHVytZmVYIUW6H/4AFtzuOGw+AO36XVWaFKEapE5VGjRrZH58+fZo+ffpgNLo+3WKxsHnzZpe6NYXM+hFClJuiwMqXYPNHjrKh06HfY5qFJER1Ua7BtIMGDeLcuXNERES4lCcnJzNo0KCa2fWj5M/6cS7TJhYhRDWSmQSf9oS0eEfZnYugyTWahSREdVKuREVRFLddIJcuXcLPz++Kg/Jo0vMjhCituD0wu0BC8sRx8A/XJBwhqqMyJSo333wzoI7TmDhxImaz2X7OarWyb98++vTpU7EReghpPBFClMneBbBwkuO4/a1w82yZNihEGZUpUQkKCgLUFpWAgAB8fHzs50wmE7169eK+++6r2Ag9RH43j/yKEUIUS1Fg0cOw+xtH2fUfQteJmoUkRHVWpkRl7ty5AERFRfHEE0/U/G4eN2TWjxCiSOkX4b02YM12lN27Bhp01S4mIaq5co1Refnllys6Do+nuOn8cVcmhLhKxWyFOSMcxz4hMHUP+ARrFZEQNUK5l9D/+eef+fHHH4mJiSEnJ8fl3K5du644MI/j1PUjjSpCCBfbv4SlTziOe02GEW/ILwshKkC5Vhn66KOPuOuuu6hduza7d++mR48ehIWFcfLkSUaNGlXRMXoU+b0jhLCz2eD7sa5Jyi3zYOSb8stCiApSrkTls88+Y/bs2Xz88ceYTCaeeuopVq5cydSpU0lOTq7oGD2CdPIIIVykxMGrIXB0uaPswc3Q9ibtYhKiBipXohITE2Ofhuzj40NqaioAd9xxBz/88EPFRedBHLN+5K8kISpFZiLs/g4OLwabhy8aeWwVvNfacRzaBJ49C7XbaheTEDVUuRKVOnXqcPnyZQAaNmzI1q1bAXUPIHcbFdYk0porPILNChvegaVPwfmDWkdz5dIS4D9R8PtDsGA8nFijdURFW/8WfPcvx/GgF2DqbjD7axeTEDVYuQbTDh48mEWLFtG5c2fuuusuHnvsMX7++Wd27NhhXxSupnE766dm52TCk62bCRveUh8nxcBt86v29RUFLh0Hn1DwC7uyayXFwgftXMtSz13ZNSuD1QJfDYM4p8kC4xZAy5HaxSTEVaBcicrs2bOx2WwATJ48mbCwMDZv3swNN9zA/fffX6EBegrnpETWUhGait7gSFIAslOq9vUVRd0B+J/FoDfCQ9ugVrPyXevSCfi4S+HyrbOg8TUQ4iEbnLqLc8rO8r9vIUSplbnrx2Kx8PrrrxMf79hg6//+7//46KOPePjhhzGZTBUaoKeRJEVo6tIJ+Pp61zJrjvu6FSH5LJzdpbYm5Pt9spqkANgscOFw+a59/pDrh39Uf6jfTX2ccAh+9ZBVrg/84hpnvc7wfLwkKUJUkTInKkajkbfeeguLxVJy5RpEenmE5tIvuX5gtrlR/X7mb1jyRMUPQI3bDe+3gS8HwR9T1bKlT8Ge71zrWbILP9eZ1QK/T4F518GhRWrZ2V0wq7ejTpsx6o7CIVGOMk/o/ln6FPx8t+N41FswaR14+RT5FCFExSpX18+QIUNYv349UVFRFRyO58ofJCztKUITlmz4cqDjuN9j0GQgHPpdPf77S+g8Xv1rvySXTqhLvdfvAgYv93UuHIXZTq934R9Y/Sps/8JRFtZMHafyyz2QGg99priJO0dNrpJj1ePsVPCrBXOd1lvqdjeMfk8dqT7kRUhPULu3SkqAKlNuFnzSHZJjHGV3/q7ecyFElSpXojJq1CieeeYZ9u/fT9euXQvt+XPDDTdUSHCeyLnnR1pZRJVQFFj6pDpoFqBeFxj8EsTvc60Xv19NDOp3BUMRP9qnt8DcvMGfXSeqm+UVlHgaPu3uWnZ2p/qV76lo19adg78WTlRys9QWmYxLjrJze1yTlAFPwaDnHD9YIVFw7bvq66edV5MVo5lCLkfD3h/ANwy63VP0+y2P+P3weT/Xskf2urb2CCGqTLl+uh966CEA3nvvvULndDodVmv5mqBnzpzJs88+yyOPPMIHH3xQrmtUFuekRFpVRJU69Bvs+tpxfNsC0OuhbkcY+R9Y/rRavuhh9XuP++HatwpdhovHHEkKqC0rBaXEwYcdHMdtxqiv7+zJk+AbCr0nw5rX1bKzO+GDDhBYH8b/CDoDvN0UcjOKfl8j/wO9Hihc7us0i+jEWojZrCYmXSdAs6Fqt9GXgxx1QptC86FFv05Z7JgDix9zHDcZCLf9BMaaPfZOCE9WrkQlf8ZPRfr777/54osv6NChQ8mVtZC/4JtkKaIqndsHP010HE9aD/4R6mOdTv2gX/Ma5KQ56lx2k4AkxcAn3VzLTm2EV2tBaGO4bw3kZLguYjbsVWjYxzVRefyoYzrygCfVJOHnu/Je47T6dWIt/Hgn9h+akCi4+b/wlVMycdNs6DjW/Xv2CwOdHhQb/OBUJ7/76Nt/udZfMB4adIfu90LbMYWvZ7PBiufg0jHoMgHauGnxVRR1LMrBXx1lN34KnW93H6MQosqUa8G30mrfvj2xsbEl1ktLS2P8+PF8+eWXhISEFFkvOzublJQUl6+qkGu1cde8vwHXlWlzLDau/3gT51OyKuy1bDaFe7/eQa83V/PhqmMVdl1RDSWfhS/6O45v/R/U61S4nnOSAnB8FbzXBr6+Qe0KSkuAD9o7zjdxao2w5cLFo+qYkHdbOMr7TIW+j0BYUzAFqGWPHYSA2q6vFdywcDw/3oE9SandHib/rV7H6K2W3fZT0UlKvoZ9CpfF7S6cpABYstSka8M7hc9lp8Kb9WDbLPW+rH2zcJ2sFLWOc5Jy95+SpAjhISo1UTl16hS5ubkl1ps8eTKjR49m6NDim29nzJhBUFCQ/SsyMrKiQi3W8QTHB0GzCNfVJ/efTWbryUsFn1JuZxIzWXX4PPEpWXy//XSFXVdUM9mp6mDOfENedszyKajrRPV7badF01LOQvR6iN0K7zR3lPeaDP0eLXyN+bc5Hne8DYa/pj72DYXHD8MzsRDUoPDzQhqDl6/7uBr2hvvXq90mvqEwdQ88uh9aDHdf31lSTPHnb/kaarVwLSuYsJ3bCzMagCXTUXbhMCT84ziO2QozI127qKYdhoY9S45RCFElKjVRKY358+eza9cuZsyYUWLdZ599luTkZPtXaVprKoLNabW3d27pWKmvlevUrWaT0bpXJ5sVfrkPctPV4+Yj1Fk+RRn9Pjy8Cwa/WPic85orzYbBiDeKHxTabBiM+cy1zBwA3oHu6/uFqS0tT54Eg9M4jubDYeJS0BscZYF13bfAuJM/MLfltfDAX67nHtikdvH0zqsTmJdAJUY76uycB18McByHNnU8/qyn2jqz8V2YM8JR3vp6ePESBNYrXYxCiCpRgUPlyy42NpZHHnmElStX4u3tXWJ9s9mM2exmBkAVqR1oxqDXFRqnUpFL6VusjovJEv0eJOEwLLhDnYVyw4fQdHDlvdaOOXB0mfpYZ4CbZxc/OEqvV7tWMpOKrhPaFMZ+q14nJAqm7FCTiI+cpjPX7wrj5pd9IJZvqPq976Pqirntb4Gbv7yyAV0971e7Xkx+YM1VW2fSEuCelY4xMl0nQMdx6tiV/BlIZ3fB9i9h7/eOa934mToo9v02jjLnqdcAt8yTXY+F8FCaJio7d+4kISGBLl0c0xytVisbNmzgk08+ITs7G4PBUMwVqkZVJgy5VueBypKpeARLtjrQ8lLemKFDi8qXqCT8A9/donYz3PARtBpduM7pzbD0CcfxQ1vAJ7h016/fRR2k6h3kOghVb4R7V4GX0x8DtfK6g1qOhiNLwC8CJi65smm+g5+HHvc5BvteKVPesgcGL7h7ufs6RpO6c3E+59lAoLbqRPVVHzvPkHI2aV3p1p8RQmhC00RlyJAh7N+/36XsrrvuolWrVjz99NMekaQ40xUxMdndhoXlZXXq76mWXT/RG2HhA+pf60UN/qxu9v2oLumez5KtrjUSULf001YzE9Uuh3yHFxdOVC4ed11j5PZfIbxl6ePU6RyDVAMbQMoZ9fEj+xytHgXd8DFE/xuaDamY1VYrKkkpC50O6nZS12hx9vAutaUpX68H1Naqi0fUY3OgWsc/vKoiFUKUg6aJSkBAAO3aue6a6ufnR1hYWKFyT1AVU5MX74uzP1aqY9/PP4sdH5DHV1b/ROXcXlhUYCGzvd+rXyGN1S6UklohbDbXtTlAXRtkzRvqX/KtroWMy/BJV8f50e+pyUN53TxbjbHPVAiqX3Q9vzBoVwN2PE+/6Hgc2kRtJfEOKlyv/+Ow5HF1jMv1H6ndZkIIj1ZhiUpSUhLBwcEuZV988QW1a9d2/4RqpKR8oaLyiYTULL7c6BgQWO3SlKwU2Pa547gS1tupUrmZ6i7B+Rp0V/fVyZcYrbaUlPQX+aHf4OBC17LEU44dkJ84Bt87ddV0vh2633MlkavdHfldHleDgc/AsqfUdVJGzij6r4qOY0ueGi2E8Cjl+nPiP//5DwsWLLAf33rrrYSFhVG/fn327t1rL7/tttsKLa9fknXr1nngqrSu+/wU1QV0pVIyXTd6tHlS30/6RXWWxOZPIDvNfZ2ts1yPbdV848o93zumyXaZAB3/r3CdTe+p9yW9iCnqZ3Y6FkQD6Plg4TorX4a4Xepj/9rqxneibLrcAc+fg1EzZVVGIWqYciUqn3/+uX0Nk5UrV7Jy5UqWLVvGqFGjePLJJys0QE+iK+IXYEW1qFgKtEB4UJoCWz5RN6X783k48LP7OilnXY9PrIH1b0NsXivEibXwdnN4t7XrvjGe6MwOWDLNcTzwGff1tn6m3petnxU+l3oe/us06PaWr6GOmy5N5xkq96x0DCIVQghRvq6f+Ph4e6KyePFibr31VoYPH05UVBQ9e9a8hZKqaqiI89Rk9YWr5nVLJe2C47G7abDHVjntR6MDFDi7Q/3a/gU8eRyOLFV3xgU4vkadDuuJctLh+1sdx7d+o66t0XI0HFiojrvZ8onrczIvQ1ayuoqrXq9Oqf3xDsf5a55Wx0XE71fXGwmJUleEdXb3CghpVElvSgghqqdytaiEhITYF1tbvny5fUVZRVHKvSGhJyspX6iofMJ1arLrQnOaOn8I9nzrOFbc/BsfWep43KDAzruZiXDhKGyfXfw1PMWu/zl2/O31kGNvmMC6cNcSddG0gnbMgZkN1RYUm01dcCx2m3ouuJFjwbY67dUxKQUXMfvXV9CwV6W8HSGEqM7KlajcfPPN3HbbbQwbNoxLly4xapQ6pXL37t00a9asQgP0JJXd9Z2erX546/Nex0PSFPVD2FnBQbIZl2HHV+rj/k8UXn7cZlU/uF3KPHT8yunNsNypm6ffNPf1wvM27/MrMJA2bjf884frWih3/u469dcnWJ3WXDdvleOeD0D7f19x6EIIUROVq+vn/fffJyoqitjYWN566y38/dX9b86dO8dDDz1UoQF6gpKmCVfENOKMHAu3f6X+Be5l0JNtsXnOyrTZBTZ/XPs6nFynTmvtfg9s/thxLrQJxGwpcAEFtn7qWmTzwBaVrBT4xmmq7m0/FT2j5/Zf1KnLF4/Aqumu53680/H4joXq7sTujFugJjZXMg1ZCCFquHIlKl5eXjzxxBOFyh97rJj9SKqx/Hwhv0WlMlpW4pIcG6fd2KkeP+444xldP8dWwb68GV46PSh5rSmnN0HCQTVRSTufd96gdpOcWFP09byDISvJM1tUdsxxbGDX/4niN88Lqq9+bS1m87xr3yl+BdvAuuqXEEKIIpWr6+frr79myZIl9uOnnnqK4OBg+vTpw+nTNXfH36JXpr1ylrypyLX8TTwytEWFXfeKHXJa/6NuJ9dzlhw4/Afs+U49HvUfdQO7/PU7ggpsQKc3OhYX2/wR/DqpUkJGUeCbm2B6kLo/T2mcXAerXnYc955cuue1/ze0+xeMnOla3rA3dL3L/XOEEEKUWrkSlTfffBMfH7XPfcuWLXz66ae89dZb1KpVq0a2qlRFw0b+jB+jXm8fo5JjsZGckVv5L16cy3kL0PV/Qt1LxplihaMrHMf5s3i6TIS7lsHdy1zrj5ypJjf5Di2q8HABdepzfqvO4UUl/wNmJsL/bnQc3/l70UvOF+RXC/49B3oVWB/llnlXtm+OEEIIoJyJSmxsrH3Q7G+//ca//vUvJk2axIwZM9i4cWOFBugZ8hZ8K6rLpwISmfwZP0aDzqXlpuvrK9l3JunKX6A8tn8Jp/NmpwQ3hPQLructWbD7G/XxkJcdiYxeD436qBvdOWs5St2QL19lzPyx5MB/C4z5OLIUjq8GaxHdTdu+cDwe/KK602559LhfXbDtnlUQUKd81xBCCOGiXImKv78/ly6p0zf//PNPhg0bBoC3tzeZmZnFPbVaq8xJP/mbEXoZ9C4JkcWm8E98aiW+cjHO7HA8bjwAcrOKrhvcsHCZ3mlTyUHPQ1ADaHmto6wyxqmsea1w2fzb4NubYd98dQDsjrnqTsYAR/+EdTPUxzoDdL+3/K997VvwxFGI7F5yXSGEEKVSrrbpYcOGce+999K5c2eOHj3KtdeqHz4HDx4kKiqqIuPzCAV7DgomLFe6e7KiKLy86CAABr2ucMuNoi6nn2uzYTZW0Y7SuZlw4Bf18bXvqDNXQpu4r+sTUngnYFAH3zbqp85saTFCLetwCwRHwpwR6sBcm63iNobLTFLHvoDr7sH5Lp2APx4FW67a8vHgFvj+Fsf5iUvUqcNCCCE8Rrk+IT799FN69+7NhQsX+OWXXwgLCwNg586djBs3rkID9ASOWT+V06YSfTGdg3HqFOC6Qd6FBu1mWawMe389LV9YzvML97u9xvztMbz8+wG2nixiz5my2vC2+oEO6gBZgP7ToO+j6kqtzoZOd10nJJ9OBxMXw1MnHWuGAIS3cjxe86o6LfhK2WzwmdOCaXf+VrjOpvcc7yntPGz+0HFuxAxo1PvK4xBCCFGhytWiEhwczCeffFKo/JVXXrnigDxZkUNUrnCMSlauYwG1D/+vc6FpydEX0zlxIR2AdUcKjBMB4pOzeOZXNYHZdPwiqx8feGUBgbq7b7781hD/CBj2SuHEokGPoq+j04GXt2uZ3ul/u03vq8vJ+4Wrg2Cbjyi8YFxpnFgNqefUx23GQK3mJT/nr7xExRyobmonhBDC45Q6Udm3bx/t2rVDr9ezb9++Yut26NDhigPzJCUlIlc6ljZ/fErdIG9C/UxcTs9xOZ9jcSQyiqJwJjGD+dtjCfA2MqFPFMmZjplB+avbXpHkM45un+s/VLt2nDmPPRnyEtRuU7br6wv8b5cUC388oj7euwCmHSzb9XLS4bu8lV29fOHWvD2HutwJ0RsgtKmayBTljt8crUZCCCE8SqkTlU6dOhEfH09ERASdOnVCp9O5rMiaf6zT6Wrcfj/295nXpOJtqthxIrk2x4wfAG8v1x65bKdExabAJ2uOM/9vda+lhqG+NAjxdTpfAVOQNr3veFxwiXhQN9XLF1yOTfQMXkW/XlZS2a+35HHH4+udunNuyFsx13m9lpH/geVPO46v+wAaeOjmiEIIIUo/RiU6Oprw8HD745MnTxIdHW3/yj8+efJkpQWrFfsYlbzvgd5ezBrvWFPkSnOD/DVUvPIGlfqajHxxh+PDMzPHkfjZFIXEDEeLS2q2hV92nXE5f8XypyGHREGzYYXPG7zU5eWHvwGtry/79fUGGP+L49h5mrK1jOvGpMbD3h/Ux3XaQ4dbC9dpmDf2JLw19LzfUW4wQYexZXs9IYQQVarULSqNGjVy+/hq4jyYdlT7ugxrU5uVh85f8XX3xiYB6oyffCPa1mF0h7os2XeOJfvP2cttCuRaHcnIuaQs5m0+5XL+ipzZAYd+Vx8PeFLdPM+dFsOLX2K+JM2HqlOB//6va3lZpixbc+FDp0G64+a7r9dlAtRqAeEt1TEzzUdAzFZ1oK/J1/1zhBBCeIRyL50ZFxfHpk2bSEhIwFZgN92pU6decWCepOQxKiVUsOZCdqrb1U4vpGbzxtLDAHh7uXYp6d3MMlIUhUtOY1gup2e7nL/iFpXtXzoeh0Rd2bVKkn7R8TigHqTGqa0rc0fDzV+o664U5+BCddE5gK4Ti66v1zuW9Qe4bYE6NVpfRVO9hRBClFu5EpV58+Zx//33YzKZCAsLc2lp0Ol0NS9RyV+ZtkB5qSYrZ6fBJ93VD+EudzrGTeS5kOpINKYOcZ2p4u76l9JzXBKVzFzX8UC2K21Syc5bXK7jOGjUt/i6V8p5PMqtX8NXed1MpzepS/N3v6fo52Ymwa/3qY8D67uOTSmJTqcu7iaEEMLjlStRefHFF3nppZd49tln0VfUYl3VQFHLqBTbiHH5pJqkAJz6q9Dp/Bk/9YK8Gdamtss5fSkyoRUH1a4ng16H1aZcWdfP/p/hSN5mk00GVc420c4a9VU3A6zVEiJau56z5rh9CqDe8F+ckphr366U8IQQQmivXIlKRkYG//d//3f1JClFfPiX6nPcZT+bwhfKn/FjMBS+mN6lpcp9QpQ/NTnQ20hiRi5p2RYyciz4msrxT3v4D8dj5wXaKkufhyGsGTToDl5+6mDXC2o3WLGDai+fhOOr1MeRPd2viiuEEKJGKFemcc899/DTTz9VdCweyzHrx31mUmwjhvP4HTeZRsEZP84ynGb7mAzF/1O9MNqxlkmX11byT3wZV3tVFDift37JyJkQ0ar4+hXBywfa3awuqa/XwwMboU3eLsZrXoO4PYWfk5sJHzvt4jz2u8qPUwghhGbK1aIyY8YMrrvuOpYvX0779u3x8nJdF+O9996rkOA8TcEWlKISFxcus1jcJSqua6g4i0107DQ8un1dft191n5s1Ouw5PXzRIb60LtpmP1cVq6NQ3EptKoTWHJ8+Ta8A5eOqY/9axdft7IYvBw7HFtz1PVPpmx3rbPza8fjPg+Dv5t1XoQQQtQY5U5UVqxYQcuWLQEKDaataUqcSFNcBeeunwL1si1WbvvvNgAMblpUrE4DTh4b1sIlUQkPMFMvZS9N9XHE6HoWmiFkLetglXN7HI+j+pXtuRUpw2kmUOZl13NpCY7F2sJbwfDXqy4uIYQQmihXovLuu+8yZ84cJk6cWMHheKaiph+XKiezFT1G5XhCmv1xl4bBhZ6aa3V0Gzm/Vo/GoYRzmU+z1b2VDmX/hV7vuvBamaYpZ6XAP4vVxzd9oe7po5UmgyBWTd5cxqkoimOZfFC7p4QQQtR45RqjYjab6du3kqeueqCiWouKH6Pi1PVTIHnIH59i1Ot4fUy7Qk91bhVxbjEZ1yMSXapjobnG3mmYDa7TbcvUoLLaaTNJcxm6iypDv8fUFW/BNVE5t1f9AjWZaTqo6mMTQghR5cqVqDzyyCN8/PHHJVesIYpqnChNi0pWWmKRF8ofY1Iv2MdtEhTg7Rj7o9fpqMdFAknDZDCgsznWX/Ex6gjy9eLJES3tZWXq+kmKzXsRIzQZWPrnVQYvb2g7Rn1sy0tUslNh9jWOOrd+XehpQgghaqZydf1s376dNWvWsHjxYtq2bVtoMO2vv/5aIcF5ioJ7/RQ6X0ROkJx4kaCFd7m5kqq4gbQAjw1rztO/7Oeefo3x3/Yem73/Q45iYEfaQppbjjtdVr3O5EHNOBiXzNL98S4bRhbr4nE4tkJ9fNMXnrGkvD7v/ydrDqRfgm2zHOcGPQ/eQdrEJYQQosqVK1EJDg7m5ptvruhYPF5ZZ/1ciPkHl4/UlLOw8EF1Kf3BL9hbVNxNTQYY3Ko2fz+vzsBRvtsDgElnpWPObvrkOO2Rkxxrv65JUdcUuZxeys391juN9dBqtk9BPsGOxzvnwoa8Bd3qdYZrntIkJCGEENooV6Iyd+7cUtX766+/6NatG2azuTwv4zFKap0o6rzNai1cuPd79XujvszZqk6tLapFxZnOaaVWv8z4Iq/bLDwcaM77q44Sn5LFjJvbu9bLSoG08xDaVF27JCNvZk3D3pW/ZH5pGc1Quz2c36+up5JPBtAKIcRVp1KXlh01ahRnz54tuaKHs3f9lHGzH1sxq6vGX05k9T8JAAT7ehWukJkEF444Zg05Xys3o3D9PMmpyfbH205ecj2ZcRneawOfdINf7oYTa+HEavVcrwfVxCWPoihEX0znYprrpodVJjTK9bjlaGjYS5NQhBBCaKfcuyeXRqnHSVQTZV2ZVrG5aVHJk5XjmA306o0FZvxkXIb320FuOrS9CfpMVTfqy3fhSF5Aevv4FHevaS14/y9HQ07epoPx+2H7bMe50CYuVd/98yifrD2OXgdLpvandd0qng2UneZ6/O+vqvb1hRBCeISrZLOeK1TUrJ8Snua26yf/XN7S+o3CfGka7u96MvGUmqQAnD8Em953PR+7Vf1uKvA8QHFasv9yWg4/7zzDmn/Oq0ljilPrls3q2Cm576NQx7WLaN9ZtWXGpsCxhAJJQ1VIjHY8HjlTXW5fCCHEVUfTRGXWrFl06NCBwMBAAgMD6d27N8uWLdMyJLfyF3wr6+7JSl53jU0p/MT8RMXgbotkl4XObJBTRKLQ68FCRa3r+Nkfp2ZbeOKnvdw9bwc7Dx+HH+9wVEyMhlMbAbgQ3IkHv93JyA82MH97DHtik9hw9IK96rdbT/Pm0sPsP5PMhdRsvtl6miX7zmG7oq2aS1Avbz+fwAZu36cQQoirQ6V2/ZSkQYMGzJw5k+bNm6MoCl9//TU33ngju3fvpm3btlqG5lahISolLKSS3w1jQY8J19YVa97UZLczfqxO40IUm/udhMOaQ4exsP4/LsXXto1gT62GfLs1xqU88/yxIuNccCaEZQfUAbpz/oqmUZify/nt0ZftX01q+dmX8v/+vp70aVqryOtekWGvQkiUY5NCIYQQV6VKbVEp6YP8+uuv59prr6V58+a0aNGCN954A39/f7Zu3VqZYZVZ6uXztNWdQq9Y1d2Qzx+ElDgAQkkh8cTfWHJzCp3LSjwHgMVNPpiaqSYjbmf8pJxzPLZZIG63+ljn9M9lNIPedTVaAN+L+7mtW/1C5ZZLMYXKAB7NeYi9qY4upByLrfAg3PywMnNZcyTBfnzsfBoHzibb14OpUMGRMPRlqNep4q8thBCiVM4mZXI8IVXTGDxmMK3VauWnn34iPT2d3r17u62TnZ1NdrajtSElJeWKYyxJ4oVzjFwxiBvNFrak9oHVq+GvDwGoF/4ftpifxxxtYdcH39ClYyfY/BEAybf8TPc9zwFgoXBCMf/vGKAxRkOBXDHtAiyc5DhOOu147Dxw1mgGXeHrsnMukYkJwHh7US2SGXRA3cwvR2fGpDjuYRL+rDvkWI7/1KWiZxSdvJjucvzyooMAjO5Ql09v61Lk84QQQlQfCSlZfLzmON9sVT9/OjYI4vcp2m1WW64WlcGDB5OUlFSoPCUlhcGDB9uPU1NTadKkSaF6zvbv34+/vz9ms5kHHniAhQsX0qZNG7d1Z8yYQVBQkP0rMjKyPOGXyaW4E5h16gydNqbzjhk3QKOk7fZzoZnRLuey/1llf/yDtfC+NLq8cS//173Ae0g8VXQwziuydpkAQQ2g9Q0Q2QvMjnOmpOMuT4vSOVpo/vAa5XJuh61FkS/n4+UmEXLj+HkNBtsKIYSoMInpOcxYdpiWLyyjx5ur7UkKuG6gq4VyJSrr1q0jJyenUHlWVhYbN24s07VatmzJnj172LZtGw8++CATJkzg0KFDbus+++yzJCcn279iY2PLE36ZOLcKBfl4uYyc1VOgy8OpxUPJ26cmxhbOfOtgCtKhEBXmy7geDV1PWAvfVwD+PQeMTjNfuk5QR/eO/QbuWaG2sORfW7Giw0b+dKVGOrW75pCtEf/LdSRNk3IeIw33S+Z/NaEb3l6l+98j11YJXT9CCCEqVVq2hY9WH6P99BV0fm0lX6w/SbbF8fu8RW1/Pvy/Tux5ebiGUZax62ffvn32x4cOHSI+3rFCqtVqZfny5dSvX3h8RHFMJhPNmjUDoGvXrvz99998+OGHfPHFF4Xqms3mql/l1rn7SlFwnqusUxwDZHUATsf5uyZb0bud3axHKdztA0UnKgZz0efAsYEfYEo8zjHznVwikHtznuBd0+cAZGLifKYevNV6KTgGzdYONHM+xdElVD/Ehzt6R/HR6mO0qhPAP/FF91HGXMogJSuXQG83C9cJIYTwGFm5Vr7deprP1590u6Bnw1BfHhrYlJu7NMBk9IwVTMqUqHTq1AmdTodOp3Pp4snn4+Nzxbsq22w2l3EoHsep1UTnkoIojlVkwT5Tx4Yem5uGKx0KRndTk/O7fvRGe7IDqC0mvqGQedl9XJ3Gw5ZPHNV1NmqTxBjDX/ay7yxDiCeMt3LHEumdwT9ebSFbfQ+v3tiO+7/ZCcA9/RrTqk4gLSICGNG2Nga9jpEfqC1ln97WhRd+209ihiMxstgUur2+ipWPDSg0Y0gIIYS2LFYbP+88w8drjnM2KbPQ+dqBZh64pinjejTEu5Rd/lWpTIlKdHQ0iqLQpEkTtm/fTnh4uP2cyWQiIiICg6H0b/LZZ59l1KhRNGzYkNTUVL7//nvWrVvHihUryhJWpXIdEKy4tLDoCiYtLl0/apJhweCSztgUHXqdkteiooP4A/D3f8EvHLrcAYsfVSsafRyryAIE1oOhr8DSJ9QF2goa8hL4hLjujQO0NqpTiQ/YovjVNgCAz6w3MveW7uh+3AOoCUfL2gH25wxqGQGAXq+jbb0gEtMdLTkt6/jz4MCmvLn0H3o1CWXrSTVxyrHY+G5bDK3rBjCgeThh/tV7fychhKjOrDaFxfvi+HDVsUITIQBCfL24b0AT7u7b2COTE2dlSlQaNWoEOBYru1IJCQnceeednDt3jqCgIDp06MCKFSsYNmxYhVy/YhTsuHEeo+LU9aO4Jir5rSs29ChOK7DkYsRMbl6Lih7WzYB/FuddxKmFpcd9sOm9vMeToHZb9av1de7DNJqh2dBCiUof1O66DFwTh/AAMw8NbMYbSw8zsU8UDUN9ua5DXZIzc+naKMSlboifiaVT+5OZa6FZRADNIgIY07k+yRm5DHt/g73e7A0nARjaujb/ndDNfZxCCCEqhdWmsPLQeT5YddRtd72vycDdfRsz6Zom1aqrvtSJyqJFixg1ahReXl4sWrSo2Lo33HBDqa751VfVbP+WAsmIruD0a3ddPzo9uCQqBnui0jDU17GMPagbEQKEt4KO4xyJSuMBpYvPzboq+c63HA/71cfXdahL23qBtKsfxI2d6xHub0an0/FJMVOM29Rz3esnIsCbYB8THRoEse9Mssu58ylZpYtXCFHlrDaF+X/HcDAuhc6RwdzSrfJnT4rKY7MpbDh2gQ9WHWNPbFKh80a9jjt6N2LyoGbUqqYt3aVOVMaMGUN8fDwRERGMGTOmyHo6nQ5rMXvcVDdKwdajImb9FOz6MSSeAMCKgSTFMW4jTgmjhe4sOhTe+ncH+NZpHIol7wPe4OWadBhMpQvW3boqAPU603rY3bBfbf24tVukfTG+iADv0l3bDZNRz6Ip/ej2+iqXQVm5lbEAnBDiimTlWvli/UneX3XUXvb9thjm/x3LjJvb08Kp+1dUvEtp2Rj0OoJ9S/n7vBiKorAt+jIfrDpq734vaGy3SKYObU794Oq/T1qpExXn7p6K6vqpDpSCA2aL6PpRx684jsNT1MXQrDojafgyJPtt/MnkXuNSWnCWCH+T2i/ovDz+hX/U7waTa6JiLGUyUVSLipcf4QHeGPU6LDaFqAoe8Fpw5Pg/8an8e9Zm5t3dA3+zprs0CHHVS8nK5d0VR/h6y2m353eeTmT4+xsY3b4u79zSER+TgaxcKzGXM2gY6ovJoGd3bBI6HXRqEMzJi2nsjU2mY2QQzSIcyU2u1cb8v2OJvpBOryahDG9bp6reYpXLzLHy085Y3lt5FJNBT6MwX968qT3NIvxdVmRPzszls7XH+SKvWxygR1Qoz41ujb/ZiLeXngYh7peIKEhRFPadSeaDVUdZe+SC2zrXd6zHY0Ob06TgRrfVXLk/RVavXs3q1atJSEhwSVx0Ol3169IpC5fBtE6PUcBSeLZSXIs7YR+cUNRp2/kzgOwzk3OdVoKN3aZ+N5ghuJE6kyfjMkT2KF1sPiHuy3tOIsjHiw1PDSLHYqNhWOl+MEqrR+NQtke7ZvU7TieyOyaR/s3Di3iWEKI4+88k88WGE/iZjPxfj0g+W3eCS2nZ9GtWi2nDW5b4/ITULKYvOsjS/fEu5QHeRubd1YP9Z5KY/odjzaol+8+xZP85/EwG0nNK1yo+aUATbunagC82nOTnnWfs5XP+iqZ7VAiv3tiOVnUCStxOpTpQFIXt0Zd55Y9DHDrnuip6Qmq2fbyeyagnx2IjzM/EpfTCS0psP3WZMZ86ZmN+Nr4L17av6/I6O08nEn0xnV5NwsjMtfLhqmMs2X+u0LVAHRP4+PAWtK4b6PZ8TVCuROWVV17h1VdfpVu3btStW7dG/E9YpILjUIro+qmvnIfz512qntHVpdWwibBvvePped+NOgU2fwLnDxR+zZ6T1IG1Yz4rW6z+ETD+F3W35V/udaytkrexX71KagJ886b2zN8eQ8s6ATz5s2OtHekCEqLs9sYmceec7SRnOlpbF+xwLG65KyaJb7fF8P7YTgxorm4KeupSBgadjoZhvsRezuDxn/YW+uOhcS0/Zt3ehVZ11A+0ro1CuKlLA9798wj/c2ptKW2SAuoA+tlOrQXO/j6VyKgPNzKoZThz7yrlH1se6PSldN5afsRtohBgNpKabXEpy8lbMM05Sakf7MOgVuH8svMsmbmu9/eh73YxdXAzrmkZwYerj7nsXF+Uvs3CeGJ4Szo3LOKP0xqmXInK559/zrx587jjjjsqOh7PU8yCb3qK/4G26fSFdlzOb1GpoyTA8eOFnxTRBtreVM5ggeZD1e/rZqhdScbK759sFuHPC9e1IT3bwgerjtnn6d/79Q7mT+pNj8ahlR4DQHq2BbNR734hPSEKUBSFxfvOseXkJXo2DuXGTmVbrLKi7Y5JZMKc7aRkWUqsezk9hwlztpfqul0aBvP+2E5u1zgK8vHi1RvbcUvXSF754yA7TicC0DEymFMX00nOzMVk0HNH70Z8u/U02RYbz4xqxfGENJcWFIB29QN5/9ZObD91mecXOv4AW3vkAo//uJcQXy8Gt4qgT7OSd1xPTM9h1voTpGTmMqJtHQa1UpdMsNkUNh6/yMkLafRrVgsfkwFvL0OFDxJNycplzqZoPljlftf5cT0a8tiw5kQEeJOSlcvqw+f5Yv1JsnKt9v3SvL30PDy4OXf3bYyPSe2Wf31Me37fc5b1Ry4QHmjmi/VqkvfRmuN8tMbN54GTro1CeHJES3o1CavAd1o9lCtRycnJoU+fPhUdS/VQ5IJvhdkwFGptMqNm2f2yN4HVzZ5GhgqaMnbt27BuJvSeUjHXKwU/s5ENTw2i6XNLAbAp8PGaY3xzT88KuX5WrpWDccnUD/Yl2NeLt1cc4XJ6Dnf3bcwvu84wb/Mp6gZ58+djAwioRlPvROWz2RT2nEkC1HEWe84kcfNnm+3nv98Ww7A2tfE1Vf2Yql0xidz51XbSCvxlfnuvhtzdt7F988+3/t2BHacSefiH3aW67vA2tXl9TDsiAkse49a+QRA/PdCbPbFJNAjxJTzATI7FxqX0bOoGqX/sPDWyJSaD3v477fHhLZix9B+8vfQ8PbKVfe2k5rUDuKlzfX7accYe+y+71KTmv5uiaV8/iOeubU3PxqFYFYW9sUn4mY20qhPAztOJPPHTXpfNUef/HYu/2UizCH+3s1ry1Q/2IcDbSOeGwYT5mZk8qJk9QShKZo6VOX9Fszc2iVA/E92jQnlv5VG3i6L1bBzK86Nb06FBsEt5oLcXN3VuwE2dGwCQbbGSlmUpci2pGzvV58ZO9VEUhZTMXH7Y7roVzLA2tZk2rAXboy+z4egFJvSJYkCLq7sLXaeUZYvjPE8//TT+/v68+OKLlRFTqaWkpBAUFERycjKBgZXTP/fPtj9ptewW9SC0CfhFQOxWAI4RSXOK3m/opD4Kw+TNXPP2OnvZ3YZlvOT1DRmhbfD1C3CMS8nXoDvcu4rqLOqZJfbH3RqF8PODhZPaH/+O5b+bTtIw1I/Pxndxu1RzalYuT/y0l8vpOTw8uDn/23KaVYfPY9TruLlLfX7ccabQcwB+fagPXa6SJlFRsgNnk7nu400l1ps6uBlThzRHAf6Ovoxer6N7VCgGdytIl9Nfxy+yJzaJvs1qYbXZuOOr7WQU6GqZ0LsRz17bushFuNKzLfywPYbXlxy2lzkvEzCuRyTPjGxNkK/2yfr7K4/y4Wr3rRJVZVib2oTlJSFp2RauaRHO0fOpvL3iCMdK2GyvXpA3jw1rwZjO9fGqpJba5MxcFu2No0WEPz0ah9bsoRROyvL5Xeo/H6ZNm2Z/bLPZmD17NqtWraJDhw54ebn+QLz33ntlDNlzKQVbTZxaVIpLUgBsOgPmAv9zH1HUrNvXS+9+757SzvDxYPf1b8yXG6MByHEzTiUr18pTv6hjWY6eT+NAXHKhxEJRFN5ecYQVB9VxP3c6NXNbbEqRSQpAdq6MjRFq98Edc7Zx4GxKkXVeH9OOF35TuyncNb93igzmnVs6uMxuKY+Nxy5wx1eO/4ffXnGkUJ27+zbm6VEtMRuLbwXwMxu5t38TRrarw8kL6fRrVgu9XkfMpQwiAs0etcroQ4Oa0rpuAO3qB7HuyAX7vS7Jnb0b8djQFmw+cYnJ3++yl3dsEMRrY9qx8dhF1v6TQNNwfxbviyt2XM3KQ+rvkPl/l34T2/uvacJD1zSrkmQvyMeLO3o1qvTXqc5Knajs3u3a3NipUycADhxw/R+vRmeDBcaolMSmM1AvyJs7ezficnoOKw+dx2bLS1xSzkJmYuEn9Z5cMbFq6LlrWxNVy4/nFx6wDyxzVrDpesuJS3SoH+QytmR79GWXAX5lcTAumd5Nr75+XKHKyrXy2uJDfLctxqU81M/E5bwBjmM61eONm9rjZzaSlWt1aZ1wtic2iaHvbeCF0a25oVM9Lqfn0KSWf5GbtVmsNvafTcbHZKBl7QA2Hb/okqC4c1//xjw5olWZN4BrEOLrMrW1omfzVQSz0cDIduqMltt7NeLfXRuw5p8EXvjtAJfTc+jYIIjrOtTj6y2n0OnUPccGtgi3f46M7lCXUe2u5WBcCk3C/fDLW+6gQ4NgJg9SN7P9z787AOpCk4kZOYT5mZm3OZpP154oMb7bejZk2rAWmI16/jx4nq6NQoiqJfuVeZpydf14iqro+jm0dTltlo9VD0Iag28YnN1RquceMbai5QuOrp22Ly2nvWU/802vF/2k6clFn6tGtp28xNjZahfZgVdG2NdT+ev4Rcb/d1uh+mM61WNku7pE1fIlxNdEzzdXl/ga/ZrVYtPxiwDc3KU+e2OTOHFB3dOiVZ0AWtUJ4L1bO6GvwKb7fG8uPcz87TG0bxDEQwObkWOx0btpGEv2neOPfXF0bBDMY8NaVPjr1kQJqVnM2XQKo17H2O6RJGXk0izCv8TxBQUpisKvu87y+E97Xcp7RIXy1cRu+HgZ2Hj8Is3C/YkMdXyop2Vb+HbraWYu+8de1rVRCDtPu/lDIs8P9/WiZ+NQ+/9bNpvCjztieebX/cXG+NqYdmw9eYkl+85x/zVNeGJ4y0rrUvBkNpuCxaZU6u686dkWsnKtGA16/jwYT4/GoSzdH8/e2CSmDG5Gu/pBlfbaomRl+fyWRKUEh7Yso82K/1MPQqLAJxTidhX7nHwHTe1p+5yjb7zVi8voYDnIj2an/XjCmkHiacdU4hqSqCSm59D5tZX2Y5NRz8vXt+HLDSddBsq50yzCn+N5fcfdGoXYZyL8q0sDoi+msSsmiW6NQvjvhG7c/Nlmgny9+PH+3ny46hifrHVtul/52ACaF7PiZlaulaPnU2kY6osOHW+t+IeMHCsPDmxa5Eqdby497HZKZi1/s8vid/Mn9So0Qj/HYsOmKFXePG+zKew7q/6/1aF+ECsPn+ev4xfp1STMZQ2H0jiXnMkna46j1+l4eHCzUg3WdCfHYuOdP48UOb111bRraBZRuoWrjiekMurDjeRaHb/ODHodS6f2p2Wd0nXbHDibzJH4VG7oVA+jXsfR82k8+O1Otxu6AbSpG8h//tWB4xdSeWzBXrd18k2/vg139I6q0PEuQlRnkqhUoMKJSgjElW7U/Z7+X9BpyP/Zj7u9vpKo9H38bH7VUWnst/DTRMjbbbmmJCoAA95aS8xl90nJ89e2ZuvJS6z+J6HI53dtFMLnt3dl1IcbCfIxsvSR/sQnZ7Hy0Hlu7FSf8ADXUfWL9sYx1c2MiLl3dbfvCO1MURSGvreeExfSMep1eHsZ7DMvxnaL5D//7kCOxcZve86SlWvlho71mLX+hH1KYWm8fH0b7urbGIAVB+O5/5udeBl03NItkgBvI10ahjCiElbwzMq1Ens5g8hQX3KsNv49azNHz7sfODjz5vaM6Vwfo15X7NRuq03h8/UnCo2vmDuxu336aEGbjl3kg1VHCfA28q+uDUjKyKVn41ByrQrXfrSxxPdxe6+GPDG8ZZHLjl9Oz+Her/9mV0ySS/m7t3Tk5i71r7gr2mpTWHX4PM/+uh9FUUjMyC3xOc4LIL4wujV39W0sCYoQBUiiUoEObl5K2z/HqQfBjdRE5dye0j25QNLx49+x/PL7LywwvuwovO1Hdcfj+P3gHQTPxFBTOM/+Keif10aSkWPl1T8O8tueOLd1Vj9+DU3D/bHa1P2nS+rCWfPPee6eV7hbbmTbOsy6vUuhD637/rfDPtDOnU6RwQR4G9l47GKxr1uSJ0e0xGzUFzkOYsWjA0r9V39pHIpLsScBOl3hNQuLU9TiXAkpWQx7f4PLImTOnh3Vipu61CcpI5eoMD8ycizc/lXxA1mdPT6sBe+vOoqtiFj9TAZC/EzMGt+V9g2CyMq18tbyI8z5K9ql3oTejXjhujaV1p1yPiWLL9afLPS6oC58OK6Huo/W2aRMageYZU0fIYogiUoFOvjXEtquvE09CG6kJhPx+4p/Uj43rSOW09swzh3uKLj9VzAHwMZ3occkaDakAqL2DEUlKhufGmQfI7Dp2EVu/6rwmJWZN7fn/3o0LNPrZeZYeX7hfmoHebPt5CWXv7Jb1Qngj4f72T/Anlu4n++3lT8p3PvScNYfu8CiPXFc0zKcF/NmM6x+/BrWH7nAq4sPlXAFV2/9uwO3VsAutgv+juHpX4ofJwHqmhOJGTmFpsaCOiVz9p3daFc/CEVR+H57jMsCXgAPXNOUz9eXPFixJA8NbMq0YS0wGvQkpucQ6OPFntgkHluwp8jWOHdLvHeMDOabe3pU2db1SRk5PPXzPtYdvcALo1tzR69GNXsigRAVTBKVCuSSqJSVu26cMzvhv4MdxxP+gMYDynd9D3fnnO1sOHrBZexGfpdKvmyLlRcWHiDU38SmYxc5GJfC0NYR/HdC9yt67bikTJ7+ZZ9La8hb/+7AzZ3r88na4/YVJ4N8vEjLtmDN+1N+zsRu3Pe/nfZjd/a8NMylK0JRFA7GpVDL30ydIHW8xvqjF9yuHPrIkOZFritxa7cGdGukruJ7Xce69sXH0rMt/L4nDr1O3XTMpij4m40uH4xWm8J9/9vBmiK60iYNaMKEPlG8seQQvZqEcWfvKE5eSOPxn/ayu0C3Sb5xPRqy4egFl8Wv+jevxew7uuFjMhBzKYOxs7dwLjmryHvVpm4gT49qxdM/7yPA28iJC2nYFIgM9eGPKf2K7NLJzLGyeF+cy5YMRfljSj/aN5CBkUJUJ5KoVKADf/1Bu5W3l+/J7hKVs7vgy0GO40cPQPCV/yXtiS6kZvPX8Ys0Dffn+k824eNl4O8Xhha5o/LxhDRWHT7PzV3qExFw5evJ7D+TzPWfuC70VS/ImzinD9bDr45k84mLLNl/jkeGNKdRmB8X07Lp9rpj0b23/93B/oFZMEkpiqIo7D2T7LL52OrHr6FxmB9boy/RINiXdUcTeOn3g26ff0PHesz8V3uSM3PpPWNNofN9mobx/X29AIi+mM6gd9a5nP9jSj9WHj7PigPxvHNLxyI/yPMXD+vcMJjlB+Lt69+487+7exRaITM1K5dP155w27ry+e1dGdG2tktClZCaxbmkLDo0CCpVC8TpS+mcTcqkUZgf76w4wsLdZ+3nZt7cnrHdI6UlQ4hqSBKVCnRg0yLarSp6TyNFZ0CnqM3Qb+WOZar5D7xteU3W7hKVuD0w+xr18cj/QK8HKjhiz3QmMQOz0VBoAGxli7mUwYC317o9t/25IUXOWFm4+wxL9p3jmVGtaRbhz6mL6YT4msq8ANTifXGs+SeBJ4a3dLsp5M7Tlxk3e5vbhfFKcvT1USw7cI5H5u+xlzUN9+P3Kf2KTAZLsic2iXvm/e2yoVrXRiF8c0+PIpeXVxSFo+fTiAz1YfPxSxw5n8o9/RpXysym1YfPE3M5gztlBo0Q1ZokKhWopETFpjeht6m/1KfmTOFt81eYlbymcneJyrl98EV/9fG4+dByVEWHLAp4b+VRPirQ3bLskf4esy16eraFkR9uIPZy4f1FilNw59bp17dhYt4MoyuhKAoP/7CbPw+e5/2xnbi2fR1ptRBCVKhKWUL/qlVSHqfXQ94fw1b0KEoJW6TrnGYBVNQGhKJY04a1oGm4n73l4dt7enpMkgLqkug/3t+bY+fTaBbhz4xl//DHXsdMqOnXt+H6jvX4bN0JejYOZdI3OwFckpQFk3rRs4J2VdXpdHxyW5cKuZYQQlwpSVRKUGJ7k87RvG1Fz0FbI7rqj5GqD8TthFPnREUviUpVGdG2Do8NbUGL2v70a17yNvNVrW6Qj32X2o/HdSbc38y6Iwl8clsX2tRTk6oXr1N32x7XI9K+42qA2ciaJwZWeZeaEEJUFUlUrpDilKgo6Hgx9y7uNy6m2bAptHX3BL3TLTfKh0tV8fYy8MjQ5lqHUWovXd+Gl65v4/bc7b0acTAuhd5NwnhmVCvplhFC1GiSqJSohEGOTi0kVvQcUqL4OOQZVvW9xn39sGbQYSzkZkA9aV4XZde2XhCLpvTTOgwhhKgSkqiUpIS+H8XoDXnbu1hRkxZjcbMR9Hq4eXZFRSeEEELUaLK+8xW63H2a/bG60DvopSleCCGEqBCSqJSgpNnb2fV72R/nt6jI+g5CCCFExZBEpSQlTfspMEYFJFERQgghKookKlcoINixpPhFRV2mXBIVIYQQomJIolICheJbVIJr1eGfa3/mzpxnOKqoe/YYZIyKEEIIUSEkUSlJKXYYaNVjGCcCe9qPfUwVv8eJEEIIcTWSRKWCJGY4NnF7bFgLDSMRQgghag5JVEpUul1tbU4tL50igyspFiGEEOLqIolKCUq7t7St2u5BLYQQQnguSVQqSEnrrQghhBCi7CRRKUkpExBpURFCCCEqnqaJyowZM+jevTsBAQFEREQwZswYjhw5omVIhZU6UZFMRQghhKhomiYq69evZ/LkyWzdupWVK1eSm5vL8OHDSU9P1zKscpE8RQghhKh4mu6evHz5cpfjefPmERERwc6dOxkwYIBGUTnEJWWS9vf3WochhBBCXLU8aoxKcnIyAKGhoW7PZ2dnk5KS4vJVmb5dspp+ORsLn/BWl8onrJm9qE/TMABa1g6o1JiEEEKIq4mmLSrObDYbjz76KH379qVdu3Zu68yYMYNXXnmlymKyZrpJhGq3g7HfwK7/Qbt/24vfvKk98/+O5abO9assPiGEEKKm0ykeMq/2wQcfZNmyZWzatIkGDRq4rZOdnU12drb9OCUlhcjISJKTkwkMDKzwmGZ8+R3Pnn3ItfCm2dBxbIW/lhBCCHG1SElJISgoqFSf3x7RojJlyhQWL17Mhg0bikxSAMxmM2azuQojc5PDyYaDQgghRJXRNFFRFIWHH36YhQsXsm7dOho3bqxlOKUkiYoQQghRVTRNVCZPnsz333/P77//TkBAAPHx8QAEBQXh4+OjZWgqd71i0qIihBBCVBlNZ/3MmjWL5ORkBg4cSN26de1fCxYs0DIsJ5KoCCGEEFrSvOvHk7mNT+dRM7qFEEKIGk0+dctMWlSEEEKIqiKJSjF00vUjhBBCaEoSleJI148QQgihKfnULTNpURFCCCGqiiQqxZKuHyGEEEJLkqgUQ2b9CCGEENqST90ykxYVIYQQoqpIolIs6foRQgghtCSJSnHyun4U59skXT9CCCFElZFP3VJQXFpRpEVFCCGEqCqSqJSKzu1DIYQQQlQuSVSKo9gKl0nXjxBCCFFl5FO3FKTrRwghhNCGJCrFcOz149z1I4mKEEIIUVUkUSlOXp6i6GTWjxBCCKEFo9YBVAcZvvUJDG8IWUlQr7PW4QghhBBXDUlUiqGgDqZVdAa48zdtgxFCCCGuQtKPURzFzRgVIYQQQlQZSVRKQwbQCiGEEJqQRKU4brb6EUIIIUTVkUSlWPkLvkmLihBCCKEFSVSEEEII4bEkUSmOdP0IIYQQmpJEpVh5mYoMphVCCCE0IYlKqUiiIoQQQmhBEpVi6KTvRwghhNCUJCrFyVvwTZGuHyGEEEITkqiUiiQqQgghhBYkUSmWdP0IIYQQWpJEpViy148QQgihJUlUSkPGqAghhBCakESlOIp0/QghhBBakkSlWNL1I4QQQmhJEpXSkK4fIYQQQhOaJiobNmzg+uuvp169euh0On777TctwylMun6EEEIITWmaqKSnp9OxY0c+/fRTLcMoBWlREUIIIbRg1PLFR40axahRo0pdPzs7m+zsbPtxSkpKZYQFwMZ/4ng8eYaao0ieIoQQQmiiWo1RmTFjBkFBQfavyMjISnutTZvWYdJZAdAFN6y01xFCCCFE0apVovLss8+SnJxs/4qNja28F7NZ7A+Dxn5Rea8jhBBCiCJp2vVTVmazGbPZXKWvmerbgAAv7yp9TSGEEEKoqlWLSlVSZMaPEEIIoTlJVIqgk8XehBBCCM1p2vWTlpbG8ePH7cfR0dHs2bOH0NBQGjb0lAGskqgIIYQQWtE0UdmxYweDBg2yH0+bNg2ACRMmMG/ePI2iyiNdP0IIIYTmNE1UBg4c6MFjQfLikuXzhRBCCM3IGJUSKNL1I4QQQmhGEpUieG5LjxBCCHH1kESlSDLrRwghhNCaJColkTEqQgghhGYkUSmKdP0IIYQQmpNERQghhBAeSxKVkkjXjxBCCKEZSVSKoti0jkAIIYS46kmiUgTZ60cIIYTQniQqJZAF34QQQgjtSKJSFJn0I4QQQmhOEpUiKOSNUZHBtEIIIYRmJFEpkSQqQgghhFYkUSmKLPgmhBBCaE4SlaLYJ/1Ii4oQQgihFUlUSiSJihBCCKEVSVSKJAu+CSGEEFqTRKUoMkRFCCGE0JwkKiWRMSpCCCGEZiRRKZI0qQghhBBak0SlCLLXjxBCCKE9SVRKoEjXjxBCCKEZSVSKIgu+CSGEEJqTRKVI0vUjhBBCaE0SlRJJoiKEEEJoRRKVIknXjxBCCKE1SVSKkj9GRQbTCiGEEJqRRKVIMkZFCCGE0JokKiWRFhUhhBBCM0atA/BYV+n0ZEVRiE+Px2Qwse/CPuLS41gfu55g72DCvMOo71+fOn517F/Hk46TnptOgFcAEb4RRPhGkJKTwo7zOzh48SD+Jn+CTEHU8qlFuG84tXxqkWXJ4mzaWQ5cPIBVsbLp7CaCzcH4Gn0J8wkj1DuUUO9QwnzCOJt2FovNws7zO/E1+trjCvUOxcfLh5iUGIx6Iz5GHwJNgQSbg0m3pJOUlUSQOQij3sjRxKOEeIfga/Ql0BRIoDmQEHMIAaYAgsxBBJgCCDYH4+flR7Y1m5TsFMwGM75e6uvFpsbiY/TB2+BNoDkQP6MfAaYA/Lz8AIhPjwdAp9Ohy2uBi06JxqbY8DZ442P0wd/kj6/RF18vX7VuXr0saxZxaXGYDCa8Dd74evna63kbvNHlJco51hwuZV5i38V9KPZuSfU6GbkZnM84j7fBG5PBhI/RB18vX3yMPvgYfTDoDOh1evuXTqfDoDOgQ+dSrtfpScpOwmqzYjKYMBlM+Bp98TZ6YzaYMRvM6HV69blVnMDbFBtWxaoe5L39bGs2SdlJmAwmvPRe9piNOmOVx+dpFEVBQcGm2FAUBRs2bIqNtJw00nLTyLHmkJSdhE2xYdQb8dJ72b+bDCb74/zj/McGvUHrt1ZqWZYs9f3n3Qebom40a1NsZFuzSc1JxapYsdgsWGwWcqw5pOamAqBH/TnR6/T2n5NCx04/PzqdDj2Ox/nnAPs5nU5nP+d8Defy/OfkXzf/3zH/e378LmUK2LA56uaX5dfLqwvq75Esa5b9OP+a9u8FyvO/+Zn86BjesTL/uYoliUqJau4vvIuZF9Hr9ASYAjh86TCLTixiwZEFWodVKnHpcWWqfzHzYiVFUnn0Oj3eBm8yLBlah1JI/i9Sg86AQWfAZDDZkxmz0YyP0QeT3oSCQqYlEz16xweewYuM3AzOpZ8j15aLSW/Cy+CFoqh1dTqdywdkpiWT+PR4cm25pY5Ph5qMGfQGe5x6nR6j3ohRb7Qfu3zXq9+NOiN6vd7lg0ev07t+eBX4YNKjx6A3FL6u0+vr0GHUG1FQsNgs9g9Pi81CljWL1JxU9fWdEgejXv0VnV/fqlixKTYycjNIzkm2f8jmf6XmpJKWm+b4oKlgOnT2f8P8hNX+n67AdzWTdnwg553Lv47zhzrg9gPeZrORY8shLScNq2J1+QAt+CHs/GFrsVmwKJZKuQdXo3Zh7fjhuh80e31JVEpSw/KUuLQ4ZmyfwbrYdaV+TpOgJrQIaUEtn1rEp8cTlx5HQkaCy4d/ZEAkFzMvkmnJtJcFmYPoFN4Jo97IpcxLXMq65FKnWXAzUnNS6Ve/HwadgZahLbmcdZmLmRdJzErkctZlErMSScpOole9XuRac2lfqz3JOcl4G7xpENCApOwksq3ZtA1rS2J2IklZSSRlJ3Ep8xKtw1pjVaykZKcQYAqglk8tkrKTSM1J5XLWZVJzUknKTiItJ43knGT1l3xOGik5KbQObU2GJcN+LjIgEqPeSGpOqv1DIv8vNIAQc4jLXyeKouCl9yLCN4JMSyZpuWlk5GaQacl0+yFSx68OudZcMiwZ9vtjU2xuk5SowCjCfcNd/tqy2CxE+Eag1+nt18iyZJFhyUBRFPsHXP5j57+ynb8UFAJNgYDaYpF/nYIx2z9kUT8MtEymjDpjoQ8lBQWLYsFilQ+rgvy9/O2tT/5e/vZWhVxbLrnWXPV73pfFVvi+5thyyLHlaBR9xQkwBeBt8MagV5PT/JZZs8HsaKHI+zlRlLxWGafH+T97+T8Lzj9j+f85t+LYW0GcW7ucW0nyXsu5haRg0lewJQZwnyQWSCDBkTD6efm5bRV1vp7Ld52ORkGNKv8fpBiSqBTF6UOoNM6mneWZDc+QlptGs+BmDGs0jO51uhPiHcKiE4v45tA3+Hn50TSoKS1DW9IkqAlNgptwIukEs/fNxqg3Us+vHg0DG1LPvx71/OtR27c2IeYQDHoDq2NWcz79PP4mf2p517J3kYR4h9j/6srIzeDgpYOYDCZ2nd/FyeSTmA1mTiaf5MDFAxh0BtJy04p9H02CmvBcz+foHNGZTEsmQeagIutabBbSc9PtdRRFIS03jcSsROr61cXL4OX2eZmWTHs3QnWlKArZ1mzSctMINgfb/w3Key3nXxr5fzFnWjLJtGSSnpuOj9GHcN9we3dTVVKUvA8na469BcD5F7PVZiXbmk22LZscaw6ZlkyyLepxtiUbnU5HkCmIXFsuOdYcsq3ZeBu97cljrjXXfn2j3oifl5+9bq41F5PBRIh3COG+4faY8n9h598Pq82KRbGQZclyicuqWB3HTs38BeN3Pra3dhTxAeXcnZBf5vI6Ba7n/N1qs7q0tBh1RnuLi6+XL156L3uCkJ88uGud0ev0andp3nOMeiNGnREvgxch5hC8DF4uLUD5H1L5rSFl+be3Jy7WXCyKxf7vZVWsjpYNpw90cJQ5t3oUPAY33Rh5r5nfzZf/Xn291C5I+4c2FPoQz///Ir/cz8sPk8Fkbw1D5+jScf4AF57PIxKVTz/9lLfffpv4+Hg6duzIxx9/TI8ePbQOK0/x/zPvPL+TlOwU1sauZc+FPQAcTzrO8lPLAYjwjSAhI8Glflm4+2uxoEBTIAGmAM6mnS31dbvW7sodre/gmshruJx1mfTcdBoGNHTpgzYZTMXHpje6JDI6nY4AUwABpoBin+dj9Cl1nJ5Kp9PhbfTG2+hdIddyptfp8Tf542/yv+JrVwSdTmfv1vFUBr0BAwaPjrE60ul09tYXSp/fCFGhNE9UFixYwLRp0/j888/p2bMnH3zwASNGjODIkSNERERoHV6xs34OXDzAxOUTC5UPihzEkctH7F0k+Sa2nUi2NZsTSSeITo7mQuYF+7n/a/l/eBu9OZt2ljOpZ7iQeYGLmRcLJSk96/S0d6Gk5KRgU2yk5KSQkpNir2PUGfE3+ZOUncR97e8jJSeFtNw0bmhyA0a9kVZhrezN+6AmU0IIIYQn0jxRee+997jvvvu46667APj8889ZsmQJc+bM4ZlnntEkptT0JHIzjxNnNJCis5KR5n7g5r4L++yPvfReWGwWZvSfwegmowFIzErkn8v/EJ0cTbhvOMMaDXN5fkZuBnFpcfh6+VLPv16h6+facrmQccE+nqJjeEf7rJH886k5qSRmJZKSk0JydjIRvhG0CWtTEbdBCCGE0JxOsc91rHo5OTn4+vry888/M2bMGHv5hAkTSEpK4vfff3epn52dTXZ2tv04JSWFyMhIkpOTCQwMpKIs++sbnjr+VqnrD4wcyEeDPiItN63Ebg8hhBDiapeSkkJQUFCpPr81Hc148eJFrFYrtWvXdimvXbs28fHxherPmDGDoKAg+1dkZGSlxKXT6THbFPVLZ3BMu3TzFeAVwPBGw+3jM4QQQghRcTTv+imLZ599lmnTptmP81tUKtrIPuMZ2Wd8hV9XCCGEEGWjaaJSq1YtDAYD58+fdyk/f/48derUKVTfbDZjNsuofiGEEOJqoWnXj8lkomvXrqxevdpeZrPZWL16Nb1799YwMiGEEEJ4As27fqZNm8aECRPo1q0bPXr04IMPPiA9Pd0+C0gIIYQQVy/NE5WxY8dy4cIFXnrpJeLj4+nUqRPLly8vNMBWCCGEEFcfTacnX6myTG8SQgghhGeoNtOThRBCCCGKI4mKEEIIITyWJCpCCCGE8FiSqAghhBDCY0miIoQQQgiPJYmKEEIIITyWJCpCCCGE8FiSqAghhBDCY0miIoQQQgiPpfkS+lcif1HdlJQUjSMRQgghRGnlf26XZnH8ap2opKamAhAZGalxJEIIIYQoq9TUVIKCgoqtU633+rHZbMTFxREQEIBOp6vQa6ekpBAZGUlsbKzsI1SJ5D5XDbnPVUPuc9WRe101Kus+K4pCamoq9erVQ68vfhRKtW5R0ev1NGjQoFJfIzAwUH4IqoDc56oh97lqyH2uOnKvq0Zl3OeSWlLyyWBaIYQQQngsSVSEEEII4bEkUSmC2Wzm5Zdfxmw2ax1KjSb3uWrIfa4acp+rjtzrquEJ97laD6YVQgghRM0mLSpCCCGE8FiSqAghhBDCY0miIoQQQgiPJYmKEEIIITyWJCpufPrpp0RFReHt7U3Pnj3Zvn271iF5lA0bNnD99ddTr149dDodv/32m8t5RVF46aWXqFu3Lj4+PgwdOpRjx4651Ll8+TLjx48nMDCQ4OBg7rnnHtLS0lzq7Nu3j/79++Pt7U1kZCRvvfVWoVh++uknWrVqhbe3N+3bt2fp0qUV/n61MGPGDLp3705AQAARERGMGTOGI0eOuNTJyspi8uTJhIWF4e/vz7/+9S/Onz/vUicmJobRo0fj6+tLREQETz75JBaLxaXOunXr6NKlC2azmWbNmjFv3rxC8dTkn4lZs2bRoUMH+4JWvXv3ZtmyZfbzcp8r3syZM9HpdDz66KP2MrnPFWP69OnodDqXr1atWtnPV8v7rAgX8+fPV0wmkzJnzhzl4MGDyn333acEBwcr58+f1zo0j7F06VLl+eefV3799VcFUBYuXOhyfubMmUpQUJDy22+/KXv37lVuuOEGpXHjxkpmZqa9zsiRI5WOHTsqW7duVTZu3Kg0a9ZMGTdunP18cnKyUrt2bWX8+PHKgQMHlB9++EHx8fFRvvjiC3udv/76SzEYDMpbb72lHDp0SHnhhRcULy8vZf/+/ZV+DyrbiBEjlLlz5yoHDhxQ9uzZo1x77bVKw4YNlbS0NHudBx54QImMjFRWr16t7NixQ+nVq5fSp08f+3mLxaK0a9dOGTp0qLJ7925l6dKlSq1atZRnn33WXufkyZOKr6+vMm3aNOXQoUPKxx9/rBgMBmX58uX2OjX9Z2LRokXKkiVLlKNHjypHjhxRnnvuOcXLy0s5cOCAoihynyva9u3blaioKKVDhw7KI488Yi+X+1wxXn75ZaVt27bKuXPn7F8XLlywn6+O91kSlQJ69OihTJ482X5stVqVevXqKTNmzNAwKs9VMFGx2WxKnTp1lLfffttelpSUpJjNZuWHH35QFEVRDh06pADK33//ba+zbNkyRafTKWfPnlUURVE+++wzJSQkRMnOzrbXefrpp5WWLVvaj2+99VZl9OjRLvH07NlTuf/++yv0PXqChIQEBVDWr1+vKIp6T728vJSffvrJXufw4cMKoGzZskVRFDWh1Ov1Snx8vL3OrFmzlMDAQPt9feqpp5S2bdu6vNbYsWOVESNG2I+vxp+JkJAQ5b///a/c5wqWmpqqNG/eXFm5cqVyzTXX2BMVuc8V5+WXX1Y6duzo9lx1vc/S9eMkJyeHnTt3MnToUHuZXq9n6NChbNmyRcPIqo/o6Gji4+Nd7mFQUBA9e/a038MtW7YQHBxMt27d7HWGDh2KXq9n27Zt9joDBgzAZDLZ64wYMYIjR46QmJhor+P8Ovl1auK/VXJyMgChoaEA7Ny5k9zcXJf336pVKxo2bOhyn9u3b0/t2rXtdUaMGEFKSgoHDx601ynuHl5tPxNWq5X58+eTnp5O79695T5XsMmTJzN69OhC90Luc8U6duwY9erVo0mTJowfP56YmBig+t5nSVScXLx4EavV6vIPBFC7dm3i4+M1iqp6yb9Pxd3D+Ph4IiIiXM4bjUZCQ0Nd6ri7hvNrFFWnpv1b2Ww2Hn30Ufr27Uu7du0A9b2bTCaCg4Nd6ha8z+W9hykpKWRmZl41PxP79+/H398fs9nMAw88wMKFC2nTpo3c5wo0f/58du3axYwZMwqdk/tccXr27Mm8efNYvnw5s2bNIjo6mv79+5Oamlpt73O13j1ZiKvB5MmTOXDgAJs2bdI6lBqrZcuW7Nmzh+TkZH7++WcmTJjA+vXrtQ6rxoiNjeWRRx5h5cqVeHt7ax1OjTZq1Cj74w4dOtCzZ08aNWrEjz/+iI+Pj4aRlZ+0qDipVasWBoOh0Ajo8+fPU6dOHY2iql7y71Nx97BOnTokJCS4nLdYLFy+fNmljrtrOL9GUXVq0r/VlClTWLx4MWvXrqVBgwb28jp16pCTk0NSUpJL/YL3ubz3MDAwEB8fn6vmZ8JkMtGsWTO6du3KjBkz6NixIx9++KHc5wqyc+dOEhIS6NKlC0ajEaPRyPr16/noo48wGo3Url1b7nMlCQ4OpkWLFhw/frza/v8siYoTk8lE165dWb16tb3MZrOxevVqevfurWFk1Ufjxo2pU6eOyz1MSUlh27Zt9nvYu3dvkpKS2Llzp73OmjVrsNls9OzZ015nw4YN5Obm2uusXLmSli1bEhISYq/j/Dr5dWrCv5WiKEyZMoWFCxeyZs0aGjdu7HK+a9eueHl5ubz/I0eOEBMT43Kf9+/f75IUrly5ksDAQNq0aWOvU9w9vFp/Jmw2G9nZ2XKfK8iQIUPYv38/e/bssX9169aN8ePH2x/Lfa4caWlpnDhxgrp161bf/5/LPPy2hps/f75iNpuVefPmKYcOHVImTZqkBAcHu4yAvtqlpqYqu3fvVnbv3q0Aynvvvafs3r1bOX36tKIo6vTk4OBg5ffff1f27dun3HjjjW6nJ3fu3FnZtm2bsmnTJqV58+Yu05OTkpKU2rVrK3fccYdy4MABZf78+Yqvr2+h6clGo1F55513lMOHDysvv/xyjZme/OCDDypBQUHKunXrXKYZZmRk2Os88MADSsOGDZU1a9YoO3bsUHr37q307t3bfj5/muHw4cOVPXv2KMuXL1fCw8PdTjN88sknlcOHDyuffvqp22mGNfln4plnnlHWr1+vREdHK/v27VOeeeYZRafTKX/++aeiKHKfK4vzrB9FkftcUR5//HFl3bp1SnR0tPLXX38pQ4cOVWrVqqUkJCQoilI977MkKm58/PHHSsOGDRWTyaT06NFD2bp1q9YheZS1a9cqQKGvCRMmKIqiTlF+8cUXldq1aytms1kZMmSIcuTIEZdrXLp0SRk3bpzi7++vBAYGKnfddZeSmprqUmfv3r1Kv379FLPZrNSvX1+ZOXNmoVh+/PFHpUWLForJZFLatm2rLFmypNLed1Vyd38BZe7cufY6mZmZykMPPaSEhIQovr6+yk033aScO3fO5TqnTp1SRo0apfj4+Ci1atVSHn/8cSU3N9elztq1a5VOnTopJpNJadKkictr5KvJPxN333230qhRI8VkMinh4eHKkCFD7EmKosh9riwFExW5zxVj7NixSt26dRWTyaTUr19fGTt2rHL8+HH7+ep4n3WKoihlb4cRQgghhKh8MkZFCCGEEB5LEhUhhBBCeCxJVIQQQgjhsSRREUIIIYTHkkRFCCGEEB5LEhUhhBBCeCxJVIQQQgjhsSRREUIIIYTHkkRFCCGEEB5LEhUhRJlNnDiRMWPGaB2GEOIqIImKEEIIITyWJCpCiCL9/PPPtG/fHh8fH8LCwhg6dChPPvkkX3/9Nb///js6nQ6dTse6desAiI2N5dZbbyU4OJjQ0FBuvPFGTp06Zb9efkvMK6+8Qnh4OIGBgTzwwAPk5OSUKp6BAwcydepUnnrqKUJDQ6lTpw7Tp0+3nz916hQ6nY49e/bYy5KSklxiXLduHTqdjhUrVtC5c2d8fHwYPHgwCQkJLFu2jNatWxMYGMhtt91GRkbGFd5BIcSVMmodgBDCM507d45x48bx1ltvcdNNN5GamsrGjRu58847iYmJISUlhblz5wIQGhpKbm4uI0aMoHfv3mzcuBGj0cjrr7/OyJEj2bdvHyaTCYDVq1fj7e3NunXrOHXqFHfddRdhYWG88cYbpYrr66+/Ztq0aWzbto0tW7YwceJE+vbty7Bhw8r0/qZPn84nn3yCr68vt956K7feeitms5nvv/+etLQ0brrpJj7++GOefvrpst04IUTFKteey0KIGm/nzp0KoJw6darQuQkTJig33nijS9k333yjtGzZUrHZbPay7OxsxcfHR1mxYoX9eaGhoUp6erq9zqxZsxR/f3/FarWWGNM111yj9OvXz6Wse/fuytNPP60oiqJER0crgLJ79277+cTERAVQ1q5dqyiKuj09oKxatcpeZ8aMGQqgnDhxwl52//33KyNGjCgxJiFE5ZKuHyGEWx07dmTIkCG0b9+eW265hS+//JLExMQi6+/du5fjx48TEBCAv78//v7+hIaGkpWVxYkTJ1yu6+vraz/u3bs3aWlpxMbGliquDh06uBzXrVuXhISEMr471+vUrl0bX19fmjRp4lJWnusKISqWdP0IIdwyGAysXLmSzZs38+eff/Lxxx/z/PPPs23bNrf109LS6Nq1K999912hc+Hh4RUWl5eXl8uxTqfDZrMBoNerf3spimI/n5ubW+J1dDpdsdcVQmhHEhUhRJF0Oh19+/alb9++vPTSSzRq1IiFCxdiMpmwWq0udbt06cKCBQuIiIggMDCwyGvu3buXzMxMfHx8ANi6dSv+/v5ERkZecbz5CdG5c+fo3LkzgMvAWiFE9SNdP0IIt7Zt28abb77Jjh07iImJ4ddff+XChQu0bt2aqKgo9u3bx5EjR7h48SK5ubmMHz+eWrVqceONN7Jx40aio6NZt24dU6dO5cyZM/br5uTkcM8993Do0CGWLl3Kyy+/zJQpU+ytIVfCx8eHXr16MXPmTA4fPsz69et54YUXrvi6QgjtSKIihHArMDCQDRs2cO2119KiRQteeOEF3n33XUaNGsV9991Hy5Yt6datG+Hh4fz111/4+vqyYcMGGjZsyM0330zr1q255557yMrKcmlhGTJkCM2bN2fAgAGMHTuWG264wWWK8ZWaM2cOFouFrl278uijj/L6669X2LWFEFVPpzh35gohRCWaOHEiSUlJ/Pbbb1qHIoSoJqRFRQghhBAeSxIVIYRHiImJsU9rdvcVExOjdYhCCA1I148QwiNYLBaX5fYLioqKwmiUiYpCXG0kURFCCCGEx5KuHyGEEEJ4LElUhBBCCOGxJFERQgghhMeSREUIIYQQHksSFSGEEEJ4LElUhBBCCOGxJFERQgghhMf6fwpOsQ8Toli1AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "\n",
    "# Reset the index so 'step_num' becomes a column\n",
    "ex218_metrics = calculate_metrics(prepare_experiment_data('218'))\n",
    "ex219_metrics = calculate_metrics(prepare_experiment_data('219'))\n",
    "ex222_metrics = calculate_metrics(prepare_experiment_data('222'))\n",
    "\n",
    "# Add an 'experiment' column to distinguish between the two datasets\n",
    "ex218_metrics['experiment'] = '218'\n",
    "ex219_metrics['experiment'] = '219'\n",
    "ex222_metrics['experiment'] = '222'\n",
    "\n",
    "# Concatenate the two DataFrames into one\n",
    "combined_metrics = pd.concat([ex218_metrics,ex219_metrics,ex222_metrics])\n",
    "\n",
    "# Now plot using seaborn\n",
    "sns.lineplot(data=combined_metrics, x='step_num', y='hit_miss_ratio', hue='experiment')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='step_num', ylabel='hit_miss_ratio'>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAioAAAGwCAYAAACHJU4LAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACCHklEQVR4nO3dd3xUVdrA8d+UzKRXEmog9N57lV7Egu6qL6KCDQuIir1jhbV3FFfAtWFFkSq9SJNepQYIJYSS3jNz3j9uMiWZVJLMJHm++8lm7rln7jxzMZknp+qUUgohhBBCCA+kd3cAQgghhBCFkURFCCGEEB5LEhUhhBBCeCxJVIQQQgjhsSRREUIIIYTHkkRFCCGEEB5LEhUhhBBCeCyjuwO4ElarlbNnzxIQEIBOp3N3OEIIIYQoAaUUycnJ1KtXD72+6DaTKp2onD17lsjISHeHIYQQQogyiImJoUGDBkXWqdKJSkBAAKC90cDAQDdHI4QQQoiSSEpKIjIy0vY5XpQqnajkdfcEBgZKoiKEEEJUMSUZtiGDaYUQQgjhsSRREUIIIYTHkkRFCCGEEB6rSo9REUKUnsViITs7291hVEteXl4YDAZ3hyFEtSKJihA1hFKK2NhYEhIS3B1KtRYcHEydOnVkbSchyokkKkLUEHlJSkREBL6+vvJBWs6UUqSlpREXFwdA3bp13RyRENWDJCpC1AAWi8WWpISFhbk7nGrLx8cHgLi4OCIiIqQbSIhyIINphagB8sak+Pr6ujmS6i/vHss4ICHKhyQqQtQg0t1T8eQeC1G+JFERQgghhMeSREUIIYQQHksSFSGExzpx4gQ6nY5du3a5OxQhhJtIoiKE8FiRkZGcO3eOdu3auTsUpk2bRqdOndwdhhCVJysN4k9AygW3hiGJihDCI2VlZWEwGKhTpw5Go6ykIESl2zITPugI3/7brWFIoiKEwGq1Mn36dBo3boyPjw8dO3bk559/RinF0KFDGTFiBEopAC5fvkyDBg148cUXAVizZg06nY5FixbRoUMHvL296dWrF/v27XN6jQ0bNtC/f398fHyIjIxkypQppKam2s5HRUXx6quvcscddxAYGMjEiRMLdP3kvdayZcvo3LkzPj4+DB48mLi4OJYsWULr1q0JDAzk1ltvJS0trdj3lyfvuitXrqRbt274+vrSp08fDh06BMDcuXN5+eWX2b17NzqdDp1Ox9y5cyvin0II91MKVr4KK1/Rjr183B1P1ZWYmKgAlZiY6O5QhPBo6enp6sCBAyo9Pd3l+ddee021atVKLV26VB07dkzNmTNHmc1mtWbNGnX69GkVEhKi3n//faWUUjfddJPq0aOHys7OVkoptXr1agWo1q1bqz///FPt2bNHXXPNNSoqKkplZWUppZQ6evSo8vPzU++99546fPiw+uuvv1Tnzp3VhAkTbDE0atRIBQYGqrffflsdPXpUHT16VEVHRytA7dy50+m1evXqpTZs2KB27NihmjVrpq666io1fPhwtWPHDrVu3ToVFhamZsyYUaL353jdnj17qjVr1qj9+/er/v37qz59+iillEpLS1OPPfaYatu2rTp37pw6d+6cSktLK9O9FsKj5WQr9cUQpV4KtH/9s7jcX6Y0n9+SqAhRAxT14ZmRkaF8fX3Vxo0bncrvvvtuNXbsWKWUUj/++KPy9vZWTz/9tPLz81OHDx+21cv7kJ83b56t7NKlS8rHx0f98MMPtmtNnDjR6frr169Xer3eFlOjRo3UmDFjnOoUlqisWLHCVmf69OkKUMeOHbOV3XfffWrEiBElfn+urrto0SIF2OJ76aWXVMeOHQvcv/wkURFVVuIZ5wTlpUClzu2tmJcqxee3dPwKUcMdPXqUtLQ0hg0b5lSelZVF586dAbjpppuYP38+M2bMYObMmTRv3rzAdXr37m17HBoaSsuWLTl48CAAu3fvZs+ePXz77be2OkoprFYr0dHRtG7dGoBu3bqVKOYOHTrYHteuXRtfX1+aNGniVLZ169YSvz9X183bqycuLo6GDRuWKC4hqqwDv8OPd9iP/evAg5vAN9R9MeWSREWIGi4lJQWARYsWUb9+fadzZrMZgLS0NLZv347BYODIkSNleo377ruPKVOmFDjnmAT4+fmV6HpeXl62xzqdzuk4r8xqtdpeG4p+f4VdF7BdR4hqSSlYNBW2zbaX9XoQRrwBHrLKsiQqQtRwbdq0wWw2c+rUKa666iqXdR577DH0ej1Llizh6quvZvTo0QwePNipzubNm21JR3x8PIcPH7a1lHTp0oUDBw7QrFmzin0zLpTk/ZWEyWTCYrGUY2RCuFl2BszsDZeP28tu+xWaDXFfTC5IoiJEDRcQEMDjjz/Oo48+itVqpV+/fiQmJvLXX38RGBhIrVq1mD17Nps2baJLly488cQTjB8/nj179hASEmK7ziuvvEJYWBi1a9fmueeeo1atWowZMwaAp556il69ejF58mTuuece/Pz8OHDgAMuXL+fjjz926/sbP358ia4TFRVFdHQ0u3btokGDBgQEBBRokRGiyrh0DD7q4lw2eRvUKtit624yPVkIwauvvsoLL7zA9OnTad26NSNHjmTRokVERUVx9913M23aNLp00X6pvfzyy9SuXZv777/f6RozZszg4YcfpmvXrsTGxvLHH39gMpkAbezH2rVrOXz4MP3796dz5868+OKL1KtXz63vr3HjxiW+xr/+9S9GjhzJoEGDCA8P5/vvv6/AiIWoQNu/ck5SarWEp095ZJICoFMqd3GEKigpKYmgoCASExMJDAx0dzhCeKyMjAyio6Np3Lgx3t7e5XrtNWvWMGjQIOLj4wkODi7Xa1dFFXmvhbgilhz4+U44uMBedtXTMPDpSh+PUprPb+n6EUIIIaq7zGR4vz2kx9vL7lwKjXoX/hwPIYmKEEIIUZ2d2w2fD3Aue3g3hES5JZzSkkRFCHFFBg4cSBXuQRai+lIK/voAVrxkL4vsCbfPB1PJlgLwBJKoCCGEENVNTiZ88y84sd5eNnIG9HrAfTGVkSQqQgghRHWSegnebg7KYd2fe1dD/S6FP8eDSaIihBBCVBcn/oK5V9uP9V7wyB4IrJylACqCrKMihBBCVHVWC6yY5pykNB8Bz5yu0kkKSIuKEEIIUbVlpsCXwyFuv73s+k+h8zj3xVSOJFERQgghqqqks/Bua+eyBzdDRGvX9asg6foRQlRJ06dPp3v37gQEBBAREcGYMWM4dOiQU51Zs2YxcOBAAgMD0el0JCQkFLjO4cOHuf7666lVqxaBgYH069eP1atXV9K7EOIKHFrinKT4hMITx6pVkgJuTlSioqLQ6XQFviZNmuTOsIQQVcDatWuZNGkSmzdvZvny5WRnZzN8+HBSU1NtddLS0hg5ciTPPvtsode55ppryMnJYdWqVWzfvp2OHTtyzTXXEBsbWxlvQ4jSs2TDHw/D9/9nL+vwf/DYIfCr5b64Kohb9/q5cOGC07bp+/btY9iwYaxevZqBAwcW+3zZ60eIkqkJ+89cuHCBiIgI1q5dy4ABzqtwFrYf0cWLFwkPD2fdunX0798fgOTkZAIDA1m+fDlDhw4tdRw14V4LN0qPh5n9IOm0veyWb6D1te6LqQyqzF4/4eHhTsczZsygadOmXHXVVW6KSAgBoJQiPdtSfMVy5uNlQFfGzdESExMBCA0NLfFzwsLCaNmyJf/73//o0qULZrOZzz//nIiICLp27VqmOISoMJeOOe96DDBlJ4Q2cU88lcRjBtNmZWXxzTffMHXq1EJ/UWVmZpKZmWk7TkpKqqzwbLJyrNz02UYOxiZzZ58onrm6evUFCgGQnm2hzYvLKv11D7wyAl9T6X8tWa1WHnnkEfr27Uu7du1K/DydTseKFSsYM2YMAQEB6PV6IiIiWLp0KSEhIaWOQ4gKs/sHmD/RfhzSGCauAZ9gd0VUaTxmMO1vv/1GQkICEyZMKLTO9OnTCQoKsn1FRkZWXoC5Tl1OY/fpRLJyrCzae67SX18IUdCkSZPYt28f8+bNK9XzlFJMmjSJiIgI1q9fz9atWxkzZgzXXnst587Jz7fwADmZ8MPtzklKz/th8rYakaSAB7WofPnll4waNYp69QpfmOaZZ55h6tSptuOkpCQ3JCuy+Zqo/ny8DBx4ZYRbXre0Jk+ezMKFC1m3bh0NGjQo1XNXrVrFwoULiY+Pt/WTf/rppyxfvpyvvvqKp59+utTxCFFuUi7Ah50hK9ledvtv0HSQ20JyB49IVE6ePMmKFSv49ddfi6xnNpsxm82VFJUQNZdOpytTF0xlUkrx0EMPMX/+fNasWUPjxo1LfY20tDQA9HrnxmW9Xo/Vai2XOIUok/MHYGZv57JHD0BQfffE40Ye8Ztozpw5REREMHr0aHeHUizZzV4IzzBp0iS+++47fv/9dwICAmzTiYOCgvDx8QEgNjaW2NhYjh49CsDevXsJCAigYcOGhIaG0rt3b0JCQhg/fjwvvvgiPj4+fPHFF0RHR1eJ30eiGlIKtn4BS56wl9XrDOMXgtnffXG5kdvHqFitVubMmcP48eMxGj0ibyqS5ClCeIaZM2eSmJjIwIEDqVu3ru3rhx9+sNX57LPP6Ny5M/feey8AAwYMoHPnzixYsACAWrVqsXTpUlJSUhg8eDDdunVjw4YN/P7773Ts2NEt70vUYFlp8NW1zknKwGfhnlU1NkkBD2hRWbFiBadOneKuu+5ydyilJq0rQrhPSZaAmjZtGtOmTSuyTrdu3Vi2rPJnOAnhJPEMvNfGueyeldCgm3vi8SBuT1SGDx9eol84nqIKhSqEEKIqOLMdvhhsPzaYYeqBarnKbFm4veunqlHS+SOEEKI8KAVr33JOUpoOgWdiJElx4PYWFSGEEKLGyUiCOVfD+b32slFvQo+JUMbVmasrSVRKSbp+hBBCXJHLx7X1URw9uAUiWrknHg8niUopSaIihBCizE5sgLkOU9/9wuGh7eAd5L6YPJyMUSklGaMihBCi1KxWWPacc5LS7t8w9R9JUoohLSpCCCFERUq7DJ8PgMQYe9mNX0CHm90XUxUiiUopSdePEEKIEov7Bz7t6Vz2yF4IbuieeKog6foRQgghKsLhZc5JSq0W8FysJCmlJC0qV6AqLVQnhBCiklgtsGAK7PrGXtbjPhg5A/TSPlBacsdKSXITITzD9OnT6d69OwEBAURERDBmzBgOHTrkVGfWrFkMHDiQwMBAdDodCQkJBa6zY8cOhg0bRnBwMGFhYUycOJGUlJRKehei2kmJg/80dk5Sbv0Rrn5TkpQykrtWSjLrRwjPsHbtWiZNmsTmzZtZvnw52dnZDB8+nNTUVFudtLQ0Ro4cybPPPuvyGmfPnmXo0KE0a9aMLVu2sHTpUvbv38+ECRMq6V2IauXcbni7OWQm2ssePwotRrgvpmpAun6EEFXS0qVLnY7nzp1LREQE27dvZ8CAAQA88sgjAKxZs8blNRYuXIiXlxeffPIJ+ty/dj/77DM6dOjA0aNHadasWYXFL6qZfb/Azw6b60b2ggkLweDlvpiqCUlUSkm6fkSNoBRkp1X+63r5lnn58MRE7a/Y0NDQEj8nMzMTk8lkS1IAfHx8ANiwYYMkKqJ4lmz4aQL8s9BeNvBZuOpJWQq/nEiiUkqqkMdCVCvZafBGvcp/3WfPgsmv1E+zWq088sgj9O3bl3bt2pX4eYMHD2bq1Km89dZbPPzww6SmpvL0008DcO7cuVLHIWqYxDPwXlucPg3uXAKN+rgtpOpIxqgIIaq8SZMmsW/fPubNm1eq57Vt25avvvqKd955B19fX+rUqUPjxo2pXbu2UyuLEAWc2gLvtcGWpBi94ekYSVIqgLSolJJMSRY1gpev1rrhjtctpcmTJ7Nw4ULWrVtHgwYNSv38W2+9lVtvvZXz58/j5+eHTqfj3XffpUmTJqW+lqghtn8Ff0yxH7ccDbd8I7N6KogkKqUkaYqoEXS6MnXBVCalFA899BDz589nzZo1NG7c+IquV7t2bQBmz56Nt7c3w4YNK48wRXWSnQHf/htOrLeXjXoLek50X0w1gCQqQogqadKkSXz33Xf8/vvvBAQEEBsbC0BQUJBtQGxsbCyxsbEcPXoUgL179xIQEEDDhg1tg24//vhj+vTpg7+/P8uXL+eJJ55gxowZBAcHu+V9CQ8VfwI+6Ohcdt96qNvBLeHUJJKolJL0/AjhGWbOnAnAwIEDncrnzJljWwfls88+4+WXX7ady5u27Fhn69atvPTSS6SkpNCqVSs+//xzbr/99gqPX1Qhx9fC/66zHwfWh4d2gJe3+2KqQSRRKTV7piJJixDuU5LxYtOmTWPatGlF1vnf//5XThGJamnjR/Dn8/bjLnfAtR/K1ONKJImKEEIIkV9WKsweCbF77GU3fgEdbnZfTDWUJCqlJK0oQghRzV08Ah93cy57aAeENXVPPDWcJCqlJHmKEEJUY4eXwXcOrSZ1O8E9K8EgH5fuIpO+hRBCCIDVbzgnKf2mwn1rJUlxM7n7peTY9SM7KQshRDWQkQif9YeEk/aysT9Ay5Hui0nYSKJSSrIyrRBCVCOx++Czvs5lUw9CoBv2uhIuSdePEEKImmn/fOckpflweClBkhQPIy0qpSTtKUIIUcUpBUufgS0z7WXDX4c+k90XkyiUJCqlJD0/QghRhaVdho+6Qvple9mExRDVt/DnCLeSROUKSNIihBBVyJnt8MVg+7HOAI8fAb8w98UkiiVjVEpJZvoI4RmmT59O9+7dCQgIICIigjFjxnDo0CHb+cuXL/PQQw/RsmVLfHx8aNiwIVOmTCExMdFWZ/fu3YwdO5bIyEh8fHxo3bo1H3zwgTvejqhoO791TlI63wYvXpIkpQqQFpXSkjxFCI+wdu1aJk2aRPfu3cnJyeHZZ59l+PDhHDhwAD8/P86ePcvZs2d5++23adOmDSdPnuT+++/n7Nmz/PzzzwBs376diIgIvvnmGyIjI9m4cSMTJ07EYDAwebKMV6gWlIL598GeH+xl138Knce5LyZRKpKoCCGqpKVLlzodz507l4iICLZv386AAQNo164dv/zyi+1806ZNef3117ntttvIycnBaDRy1113OV2jSZMmbNq0iV9//VUSleogJQ7eboHTX5gT10K9Tu6KSJSBJCqlJA0qoiZQSpGek17pr+tj9EFXxl1p87p0QkNDi6wTGBiI0Vj4r77ExMQiryGqiBN/wdyr7cf+tWHyNvAOdF9MokzcnqicOXOGp556iiVLlpCWlkazZs2YM2cO3bp1K/7JbuC8Mq0Q1VN6Tjo9v+tZ6a+75dYt+Hr5lvp5VquVRx55hL59+9KuXTuXdS5evMirr77KxIkTC73Oxo0b+eGHH1i0aFGpYxAeZPNMWPq0/bjfozB0mtvCEVfGrYlKfHw8ffv2ZdCgQSxZsoTw8HCOHDlCSEiIO8MSQlQxkyZNYt++fWzYsMHl+aSkJEaPHk2bNm2YNm2ayzr79u3j+uuv56WXXmL48OEVGK2oMFYLfPMvOL7aXnbz19DmOvfFJK6YWxOV//znP0RGRjJnzhxbWePGjd0YUfFOXU5zdwhCVDgfow9bbt3iltctrcmTJ7Nw4ULWrVtHgwYNCpxPTk5m5MiRBAQEMH/+fLy8vArUOXDgAEOGDGHixIk8//zzZYpduFlCDLyfrzXtwS0Q0co98Yhy49ZEZcGCBYwYMYKbbrqJtWvXUr9+fR588EHuvfdel/UzMzPJzMy0HSclJVVWqAAkpGXx7Py9lfqaQriDTqcrUxdMZVJK8dBDDzF//nzWrFnj8o+cpKQkRowYgdlsZsGCBXh7exeos3//fgYPHsz48eN5/fXXKyN0Ud4OLYXvb7Efh7eGe1eBybP/GxYl49Z1VI4fP87MmTNp3rw5y5Yt44EHHmDKlCl89dVXLutPnz6doKAg21dkZGSlxnshObP4SkKISjFp0iS++eYbvvvuOwICAoiNjSU2Npb0dG0QcFJSEsOHDyc1NZUvv/ySpKQkWx2LxQJo3T2DBg1i+PDhTJ061Xb+woUL7nxrojRWve6cpAx7BSZtliSlGtEpN24HbDKZ6NatGxs3brSVTZkyhb///ptNmzYVqO+qRSUyMtI2kr+iHT6fzPD31tmOa/mb2fb80Ap/XSGuVEZGBtHR0TRu3Nhlq0JVVNjsoDlz5jBhwgTWrFnDoEGDXNaJjo4mKiqKadOm8fLLLxc436hRI06cOFGmuKrjvfZIOVnwxSA4v89eNu5naD7MfTGJEktKSiIoKKhEn99u7fqpW7cubdq0cSpr3bq109oHjsxmM2azuTJCc0mWzBfCcxT3N9bAgQOLrTNt2rRCB9cKD3bxCHycb2boQzsgrKl74hEVyq1dP3379nVa8hrg8OHDNGrUyE0RCSGE8Gh7fnROUiJ7wvNxkqRUY25tUXn00Ufp06cPb7zxBjfffDNbt25l1qxZzJo1y51hFUr2+RFCCDdRChZNhW2z7WWj34Hu97gvJlEp3JqodO/enfnz5/PMM8/wyiuv0LhxY95//33GjfPMPRik60cIIdwgKw0+7Awpsfay8QuhcX/3xSQqjdtXpr3mmmu45ppr3B2GEEIIT3R2J8wa6Fz2yF4IbuiWcETlc+sYlaqmYIuKNLGIqsWNk/xqDLnH5WjLLOckpdkweOGiJCk1jNtbVKoSGaMiqqq81VjT0tLw8Sn96q+i5NLStNWrXa2AK0pIKfjxdjj4h71szGfQaaz7YhJuI4mKEDWAwWAgODiYuLg4AHx9fcu8S7FwTSlFWloacXFxBAcHYzAY3B1S1ZSeAG81BWuOvezuFRDZ3W0hCfeSRKUUpEVXVGV16tQBsCUromIEBwfb7rUopej18JXDmEWDSRuPEiD3syaTREWIGkKn01G3bl0iIiLIzs52dzjVkpeXl7SklNWa/8CaN+zH7W+CGz4HvdzPmk4SlSsgLSyiKjIYDPJhKjyHJUdrRTnlsG3Kzf+DNte7LybhUSRRKQVJTIQQohwln4d3WjiX3bcO6nZ0TzzCI0miUgoy60cIIcrJP4thnsMsHt8wmLQV/Gq5LybhkSRREUIIUXmUgiVPwdbP7WXd74FRb4FelvYSBUmiUgrS9SOEEFcgJxM+6w8XHTajHfczNB/mvpiEx5NEpRTy5ymStwghRAnFn4AP8o09mbQVwlu6JRxRdUg7mxBCiIq163vnJCWkMTx1UpIUUSLSolIKsoeHEEKUgtUCv06EfT/by/pNhSEvgqyMLEpIEpVSkDRFCCFKKDMFPuwMqQ4rIU9YBFH93BeTqJIkURFCCFG+zu+HmX2cy6bshNAm7olHVGkyRqUU8vf8SFeQEEI4UAo2feqcpNTrDM+cliRFlJm0qJSKJCZCCOFSThZ8dzMcX20vG/YK9H3YfTGJakESFSGEEFcm7TK80xIsWfaye1ZBg67ui0lUG5KolIL09AghRD6ntsDs4c5ljx6AoPruiUdUOzJGpRQkTxFCiFxWK6x+wzlJaTIInj0nSYooV9KiIoQQonSyUmHOKDi321527QfQdYLbQhLVlyQqpVBg1o97whBCCPdJOgfvtnIuu/8vqNPOPfGIak8SlVKQ6chCiBrtyAr49l/2Y3MQPLQN/CPcF5Oo9iRREUIIUTRLDix9Gv7+wl7W9ka44TMwmt0Xl6gRJFEpBWlPEULUOBmJ8PkAbffjPP+eA+1udFtIomaRRKUU8np+dDqZqiyEqAEuR8OHnZzLHtoBYU3dEo6omWR6chnk7fkpyYoQotra94tzkhLUEJ46IUmKqHTSolIKKrfzRydNKkKI6ionE357QEtU8nS7G0a9CQb5yBCVT/6rK428rh/3RiGEEBUj9SJ81BUyEuxl436B5kPdFpIQkqiUgU4yFSFEdRP3D3za07nskX0QHOmeeITIJWNUSkE6e4QQ1dK22c5JSp328MwZSVKER5AWlVKwz/rRAUoWgBNCVG3Z6TDvVji2yl7W/3EY9CzoDe6LSwgHkqiUgm0wrZvjEEKIK5Z0Ft5rC8pqL7trGTTs5b6YhHBBEpUykDEqQogq7exOmDXQfqwzwGOHwD/cbSEJURi3jlGZNm0aOp3O6atVq1bFP9FNpKdHCFGlKQUb3ndOUqL6w7NnJEkRHsvtLSpt27ZlxYoVtmOj0e0hFSovT9FJ548QoqrJTIavrtVaU/IMfx16T5JmYuHR3J4VGI1G6tSpU6K6mZmZZGZm2o6TkpIqKiyXvt50ErD/TEsDixCiSog/AR90dC67/y+o084t4QhRGm6fnnzkyBHq1atHkyZNGDduHKdOnSq07vTp0wkKCrJ9RUZW3tS5U5fSWHHwPACB3l6V9rpCCHFFTm5yTlJ8QuCpk5KkiCrDrYlKz549mTt3LkuXLmXmzJlER0fTv39/kpOTXdZ/5plnSExMtH3FxMRUWqypWTm2x6+NkR9wIYSHUwpWTIM5I+1lba6Hx4+AT7C7ohKi1Nza9TNq1Cjb4w4dOtCzZ08aNWrEjz/+yN13312gvtlsxmw2V2aIBdTyN9M0wt+tMQghRJHS42HWIIiPtpdd/wl0GifjUUSV4/YxKo6Cg4Np0aIFR48edXcoBciMHyFElXDhMHzS3blsyk4IbeKeeIS4Qm4fo+IoJSWFY8eOUbduXXeHUoB952SnQiGE8BxHVzonKSGN4dlzkqSIKs2ticrjjz/O2rVrOXHiBBs3buSGG27AYDAwduxYd4ZVJB2yMq0QwsNYrfDHI/DNjfayrnfCQzvA5Ou2sIQoD27t+jl9+jRjx47l0qVLhIeH069fPzZv3kx4uOctPCRdP0IIj5R6ET7pAWmX7GW3fAutr3FfTEKUI7cmKvPmzXPny5eJjEMTQniM2L3wWT/nsscOQUDJ1qYSoirwqDEqVYGsSiuE8AgHFjgnKfW6wAsXJUkR1Y5HzfrxZNL1I4TwCJYc+PUe2D/fXtb/cRj8vDT5impJEpUScjXrR3IXIUSlSjoHH3QAS5a9bPwf0HiA+2ISooJJolJKOuSPFiGEG5zeBv8dYj/W6eHJ49qS+EJUY1c8RuX06dOcPn26PGLxaNL1I4Rwm13fOycpzYfDC5ckSRE1QpkSFavVyiuvvEJQUBCNGjWiUaNGBAcH8+qrr2K1Wss7Ro+Ql6fopDlFCFFZcrLg6xvgt/vtZSPegHE/gV7mQoiaoUxdP8899xxffvklM2bMoG/fvgBs2LCBadOmkZGRweuvv16uQQohRI2TcAreb+9cdu9qqN/FPfEI4SZlSlS++uor/vvf/3LdddfZyjp06ED9+vV58MEHq2Wiolz0/bgqE0KIK3ZiA8wdbT/2C4eH98gqs6JGKlPb4eXLl2nVqlWB8latWnH58uUrDsoT2bt+ZC0VIUQF2jLLOUnpeCs8fkSSFFFjlSlR6dixIx9//HGB8o8//piOHTtecVCeTIaoCCEqRHY6fDEYljxhLxszE26YKb94RI1Wpq6fN998k9GjR7NixQp69+4NwKZNm4iJiWHx4sXlGqCnkF4eIUSFuXQMPso39mTSVghv6Z54hPAgZWpRueqqqzh8+DA33HADCQkJJCQkcOONN3Lo0CH69+9f3jF6iNwF36TbRwhRno6scE5SItpoS+FLkiIEcAULvtWrV69aDpotjqxMK4QoN+vehlWv2o/7PATDX3NfPEJ4oBInKnv27KFdu3bo9Xr27NlTZN0OHTpccWCexrHrR7qLhRBXJDNZG49y8bC97JZvofU17otJCA9V4kSlU6dOxMbGEhERQadOndDpdC6n5+p0OiwWS7kG6Qlss37cGoUQosqLOwif9nIue2QfBEe6Jx4hPFyJE5Xo6GjCw8Ntj2sqWZlWCFFmB/+AH26zHzceALf/LqvMClGEEicqjRo1sj0+efIkffr0wWh0fnpOTg4bN250qltdyKwfIUSZKQXLX4SNH9rLhk6Dfo+6LSQhqooyDaYdNGgQ586dIyIiwqk8MTGRQYMGVc+uH5U368exzD2xCCGqkPQE+KQnpMTay+5YAE2ucltIQlQlZUpUlFIuu0AuXbqEn5/fFQfl0aTnRwhRUmd3wax8CcnjR8E/3C3hCFEVlSpRufHGGwFtnMaECRMwm822cxaLhT179tCnT5/yjdBDSOOJEKJUdv8A8yfaj9vfDDfOkmmDQpRSqRKVoKAgQGtRCQgIwMfHx3bOZDLRq1cv7r333vKN0EPkdfPIrxghRJGUggUPwc6v7WXXfgBdJ7gtJCGqslIlKnPmzAEgKiqKxx9/vPp387ggs36EEIVKvQjvtgFLpr3snlXQoKv7YhKiiivTGJWXXnqpvOPweMpF54+rMiFEDXVqM8weYT/2CYEpu8An2F0RCVEtlHkJ/Z9//pkff/yRU6dOkZWV5XRux44dVxyYx3Ho+pFGFSGEk61fwOLH7ce9JsGI1+WXhRDloEyrDH344Yfceeed1K5dm507d9KjRw/CwsI4fvw4o0aNKu8YPYr83hFC2Fit8N0tzknKTXNh5Bvyy0KIclKmROXTTz9l1qxZfPTRR5hMJp588kmWL1/OlClTSExMLO8YPYJ08gghnCSdhVdC4PBSe9kDG6HtDe6LSYhqqEyJyqlTp2zTkH18fEhOTgbg9ttv5/vvvy+/6DyIfdaP/JUkRIVIj4ed38LBhWD18EUjj6yAd1vbj0ObwDNnoHZb98UkRDVVpkSlTp06XL58GYCGDRuyefNmQNsDyNVGhdWJtOYKj2C1wLq3YfGTcH6/u6O5cilx8J8o+P1B+GEcHFvl7ogKt/ZN+PZf9uNBz8OUnWD2d19MQlRjZRpMO3jwYBYsWEDnzp258847efTRR/n555/Ztm2bbVG46sblrJ/qnZMJT7ZmBqx7U3uccApunVe5r68UXDoKPqHgF3Zl10qIgffbOZcln7uya1YESw58OQzOOkwWGPsDtBzpvpiEqAHKlKjMmjULq9UKwKRJkwgLC2Pjxo1cd9113HfffeUaoKdwTEpkLRXhVtHr7EkKQGZS5b6+UtoOwP8sBL0RHtwCtZqV7VqXjsFHXQqWb54Jja+CEA/Z4NRVnJO3l/19CyFKrNRdPzk5Obz22mvExto32Pq///s/PvzwQx566CFMJlO5BuhpJEkRbnXpGHx1rXOZJct13fKQeAbO7NBaE/L8PklLUgCsOXDhYNmuff6A84d/VH+o3017HHcAfvWQVa73/eIcZ73O8FysJClCVJJSJypGo5E333yTnJyc4itXI9LLI9wu9ZLzB2ab67Xvp/+GRY+X/wDUszvhvTbwxSD4Y4pWtvhJ2PWtc72czILPdWTJgd8nw9xr4MACrezMDpjZ216nzRhtR+GQKHuZJ3T/LH4Sfr7LfjzqTZi4Brx8Cn2KEKJ8lanrZ8iQIaxdu5aoqKhyDsdz5Q0SlvYU4RY5mfDFQPtxv0ehyUA48Lt2/PcX0Hmc9td+cS4d05Z6r98FDF6u61w4DLMcXu/CP7DyFdj6ub0srJk2TuWXuyE5FvpMdhF3lpZcJcZox5nJ4FcL5jist9TtLhj9rjZSfcgLkBqndW8VlwBVpOwM+Lg7JJ6yl93xu3bPhRCVqkyJyqhRo3j66afZu3cvXbt2LbDnz3XXXVcuwXkix54faWURlUIpWPyENmgWoF4XGPwixO5xrhe7V0sM6ncFQyE/2ic3wZzcwZ9dJ2ib5eUXfxI+6e5cdma79pXnyWjn1p39vxZMVLIztBaZtEv2snO7nJOUAU/CoGftP1ghUXD1O9rrp5zXkhWjmQIuR8Pu78E3DLrdXfj7LYvYvfBZP+eyh3c7t/YIISpNmX66H3zwQQDefffdAud0Oh0WS9maoGfMmMEzzzzDww8/zPvvv1+ma1QUx6REWlVEpTrwG+z4yn586w+g10PdjjDyP7D0Ka18wUPa9x73wdVvFrgMF4/YkxTQWlbySzoLH3SwH7cZo72+oyeOg28o9J4Eq17Tys5sh/c7QGB9GPcj6AzwVlPITiv8fY38D/S6v2C5r8MsomOr4dRGLTHpOh6aDdW6jb4YZK8T2hSaDy38dUpj22xY+Kj9uMlAuPUnMFbvsXdCeLIyJSp5M37K099//83nn39Ohw4diq/sDnkLvkmWIirTuT3w0wT78cS14B+hPdbptA/6Va9CVoq9zmUXCUjCKfi4m3PZifXwSi0IbQz3roKsNOdFzIa9Ag37OCcqjx22T0ce8ISWJPx8Z+5rnNS+jq2GH+/A9kMTEgU3/he+dEgmbpgFHW9x/Z79wkCnB2WF7x3q5HUfffMv5/o/jIMG3aH7PdB2TMHrWa2w7Fm4dAS6jIc2Llp8ldLGouz/1V52/SfQ+TbXMQohKk2ZFnwrqfbt2xMTE1NsvZSUFMaNG8cXX3xBSEhIofUyMzNJSkpy+qoM2RYrd879G3BemTYrx8q1H23gfFJGub2W1aq456tt9HpjJR+sOFJu1xVVUOIZ+Ly//fjm/0G9TgXrOSYpAEdXwLtt4KvrtK6glDh4v739fBOH1ghrNlw8rI0JeaeFvbzPFOj7MIQ1BVOAVvbofgio7fxawQ0LxvPj7diSlNrtYdLf2nWM3lrZrT8VnqTkadinYNnZnQWTFICcDC3pWvd2wXOZyfBGPdgyU7svq98oWCcjSavjmKTc9ackKUJ4iApNVE6cOEF2dnax9SZNmsTo0aMZOrTo5tvp06cTFBRk+4qMjCyvUIt0NM7+QdAswnn1yb1nEtl8/FL+p5TZ6fh0Vhw8T2xSBt9tPVlu1xVVTGayNpgzz5CX7LN88us6Qfte22HRtKQzEL0WYjbD283t5b0mQb9HCl5j3q32xx1vheGvao99Q+Gxg/B0DAQ1KPi8kMbg5es6roa94b61WreJbyhM2QWP7IUWw13Xd5RwqujzN30FtVo4l+VP2M7thukNICfdXnbhIMT9Yz8+tRlmRDp3UU09CA17Fh+jEKJSVGiiUhLz5s1jx44dTJ8+vdi6zzzzDImJibavkrTWlAerw2pvb9/UsUJfK9uhW80qo3VrJqsFfrkXslO14+YjtFk+hRn9Hjy0Awa/UPCc45orzYbBiNeLHhTabBiM+dS5zBwA3oGu6/uFaS0tTxwHg8M4jubDYcJi0BvsZYF1XbfAuJI3MLfl1XD/X87n7t+gdfH0zq0TmJtAxUfb62yfC58PsB+HNrU//rSn1jqz/h2YPcJe3vpaeOESBNYrWYxCiEpRjkPlSy8mJoaHH36Y5cuX4+3tXWx9s9mM2exiBkAlqR1oxqDXFRinUp5L6edY7BeTJfo9SNxB+OF2bRbKdR9A08EV91rbZsPhJdpjnQFunFX04Ci9XutaSU8ovE5oU7jlG+06IVEweZuWRHzoMJ25flcYO6/0A7F8Q7XvfR/RVsxtfxPc+MWVDejqeZ/W9WLyA0u21jqTEgd3L7ePkek6HjqO1cau5M1AOrMDtn4Bu7+zX+v6T7VBse+1sZc5Tr0GuGmu7HoshIdya6Kyfft24uLi6NLFPs3RYrGwbt06Pv74YzIzMzEYDEVcoXJUZsKQbXEcqCyZikfIydQGWl7KHTN0YEHZEpW4f+Dbm7Ruhus+hFajC9Y5uREWP24/fnAT+ASX7Pr1u2iDVL2DnAeh6o1wzwrwcvhjoFZud1DL0XBoEfhFwIRFVzbNd/Bz0ONe+2DfK2XKXfbA4AV3LXVdx2jSdi7O4zgbCLRWnai+2mPHGVKOJq4p2fozQgi3cGuiMmTIEPbu3etUduedd9KqVSueeuopj0hSHOkKmZjsasPCsrI49PdUya6f6PUw/37tr/XCBn9WNXt+1JZ0z5OTqa01ElC35NNW0+O1Loc8BxcWTFQuHnVeY+S2XyG8Zcnj1Onsg1QDG0DSae3xw3vsrR75XfcRRP8bmg0pn9VWyytJKQ2dDup20tZocfTQDq2lKU+v+7XWqouHtGNzoFbHP7yyIhVClIFbE5WAgADatXPeNdXPz4+wsLAC5Z6gMqYmL9xz1vZYVcW+n38W2j8gjy6v+onKud2wIN9CZru/075CGmtdKMW1QlitzmtzgLY2yKrXtb/kW10NaZfh467286Pf1ZKHsrpxlhZjnykQVL/wen5h0K4a7HieetH+OLSJ1kriHVSwXv/HYNFj2hiXaz/Uus2EEB6t3BKVhIQEgoODnco+//xzateu7foJVUhx+UJ55RNxyRl8sd4+ILDKpSkZSbDlM/txBay3U6my07VdgvM06K7tq5MnPlprKSnuL/IDv8H++c5l8SfsOyA/fgS+c+iq6XwbdL/7SiLXujvyujxqgoFPw5IntXVSRk4v/K+KjrcUPzVaCOFRyvTnxH/+8x9++OEH2/HNN99MWFgY9evXZ/fu3bbyW2+9tcDy+sVZs2aNB65K67zPT2FdQFcqKd15o0erJ/X9pF7UZkls/BgyU1zX2TzT+dhaxTeu3PWdfZpsl/HQ8f8K1tnwrnZfUguZon56u31BNICeDxSss/wlOLtDe+xfW9v4TpROl9vhuXMwaoasyihENVOmROWzzz6zrWGyfPlyli9fzpIlSxg1ahRPPPFEuQboSXSF/AIsrxaVnHwtEB6UpsCmj7VN6f58Dvb97LpO0hnn42OrYO1bEJPbCnFsNbzVHN5p7bxvjCc6vQ0WTbUfD3zadb3Nn2r3ZfOnBc8ln4f/Ogy6vekrqOOiS9Nxhsrdy+2DSIUQQpSt6yc2NtaWqCxcuJCbb76Z4cOHExUVRc+e1W+hpMoaKuI4NVl74cp53RJJuWB/7Goa7JEVDvvR6AAFZ7ZpX1s/hyeOwqHF2s64AEdXadNhPVFWKnx3s/345q+1tTVajoZ987VxN5s+dn5O+mXISNRWcdXrtSm1P95uP3/VU9q4iNi92nojIVHairCO7loGIY0q6E0JIUTVVKYWlZCQENtia0uXLrWtKKuUKvOGhJ6suHyhvPIJ56nJzgvNudX5A7DrG/uxcvFvfGix/XGDfDvvpsfDhcOwdVbR1/AUO/5n3/G314P2vWEC68Kdi7RF0/LbNhtmNNRaUKxWbcGxmC3aueBG9gXb6rTXxqTkX8TsX19Cw14V8naEEKIqK1OicuONN3LrrbcybNgwLl26xKhR2pTKnTt30qxZs3IN0JNUdNd3aqb24a3PfR0PSVO0D2FH+QfJpl2GbV9qj/s/XnD5catF++B2KvPQ8SsnN8JSh26eflNd1wvP3bzPL99A2rM74Z8/nNdCueN356m/PsHatOa6uasc97wf2v/7ikMXQojqqExdP++99x5RUVHExMTw5ptv4u+v7X9z7tw5HnzwwXIN0BMUN024PKYRp2XlcNuX2l/gXgY9mTlWz1mZNjPf5o+rX4Pja7Rprd3vho0f2c+FNoFTm/JdQMHmT5yLrB7YopKRBF87TNW99afCZ/Tc9os2dfniIVgxzfncj3fYH98+X9ud2JWxP2iJzZVMQxZCiGquTImKl5cXjz/+eIHyRx8tYj+SKiwvX8hrUamIlpWzCfaN067vVI8ft532jK6fIytgT+4ML50eVG5ryskNELdfS1RSzueeN2jdJMdWFX4972DISPDMFpVts+0b2PV/vOjN84Lqa1+bi9g87+q3i17BNrCu9iWEEKJQZer6+eqrr1i0aJHt+MknnyQ4OJg+ffpw8mT13fG38JVpr1xO7lTkWv4mHh7aotyue8UOOKz/UbeT87mcLDj4B+z6Vjse9R9tA7u89TuC8m1ApzfaFxfb+CH8OrFCQkYp+PoGmBak7c9TEsfXwIqX7Me9J5Xsee3/De3+BSNnOJc37A1d73T9HCGEECVWpkTljTfewMdH63PftGkTn3zyCW+++Sa1atWqlq0qldGwkTfjx6jX28aoZOVYSUzLrvgXL8rl3AXo+j+u7SXjSFng8DL7cd4sni4T4M4lcNcS5/ojZ2jJTZ4DC8o9XECb+pzXqnNwQfH/gOnx8L/r7cd3/F74kvP5+dWCf8+GXvnWR7lp7pXtmyOEEAIoY6ISExNjGzT722+/8a9//YuJEycyffp01q9fX64BeobcBd8K6/Iph0Qmb8aP0aBzarnp+tpy9pxOuPIXKIutX8DJ3NkpwQ0h9YLz+ZwM2Pm19njIS/ZERq+HRn20je4ctRylbciXpyJm/uRkwX/zjfk4tBiOrgRLId1NWz63Px78grbTbln0uE9bsO3uFRBQp2zXEEII4aRMiYq/vz+XLmnTN//880+GDRsGgLe3N+np6UU9tUqryEk/eZsRehn0TglRjlXxT2xyBb5yEU5vsz9uPACyMwqvG9ywYJneYVPJQc9BUANoebW9rCLGqax6tWDZvFvhmxthzzxtAOy2OdpOxgCH/4Q107XHOgN0v6fsr331m/D4YYjsXnxdIYQQJVKmtulhw4Zxzz330LlzZw4fPszVV2sfPvv37ycqKqo84/MI+XsO8icsV7p7slKKlxbsB8Cg1xVsuVHacvrZVitmYyXtKJ2dDvt+0R5f/bY2cyW0ieu6PiEFdwIGbfBto37azJYWI7SyDjdBcCTMHqENzLVay29juPQEbewLOO8enOfSMfjjEbBmay0fD2yC726yn5+wSJs6LIQQwmOU6RPik08+oXfv3ly4cIFffvmFsLAwALZv387YsWPLNUBPYJ/1UzFtKtEXU9l/VpsCXDfIu8Cg3YwcC8PeW0vL55fy3Py9Lq8xb+spXvp9H5uPF7LnTGmte0v7QAdtgCxA/6nQ9xFtpVZHQ6c5rxOSR6eDCQvhyeP2NUMAwlvZH696RZsWfKWsVvjUYcG0O34rWGfDu/b3lHIeNn5gPzdiOjTqfeVxCCGEKFdlalEJDg7m448/LlD+8ssvX3FAnqzQISpXOEYlI9u+gNoH/9e5wLTk6IupHLuQCsCaQ/nGiQCxiRk8/auWwGw4epGVjw28soBA2903T15riH8EDHu5YGLRoEfh19HpwMvbuUzv8J/dhve05eT9wrVBsM1HFFwwriSOrYTkc9rjNmOgVvPin/NXbqJiDtQ2tRNCCOFxSpyo7Nmzh3bt2qHX69mzZ0+RdTt06HDFgXmS4hKRKx1Lmzc+pW6QN6F+Ji6nZjmdz8qxJzJKKU7HpzFvawwB3kbG94kiMd0+Myhvddsrknja3u1z7Qda144jx7EnQ16E2m1Kd319vv/sEmLgj4e1x7t/gKn7S3e9rFT4NndlVy9fuDl3z6Eud0D0OghtqiUyhbn9N3urkRBCCI9S4kSlU6dOxMbGEhERQadOndDpdE4rsuYd63S6arffj+195japeJvKd5xIttU+4wfA28u5Ry7TIVGxKvh41VHm/a3ttdQw1JcGIb4O58thCtKG9+yP8y8RD9qmenmCy7CJnsGr8NfLSCj99RY9Zn98rUN3znW5K+Y6rtcy8j+w9Cn78TXvQwMP3RxRCCFEyceoREdHEx4ebnt8/PhxoqOjbV95x8ePH6+wYN3FNkYl93ugtxczx9nXFLnS3CBvDRWv3EGlviYjn99u//BMz7InflaliE+zt7gkZ+bwy47TTuevWN405JAoaDas4HmDl7a8/PDXofW1pb++3gDjfrEfO05TtpRy3ZjkWNj9vfa4TnvocHPBOg1zx56Et4ae99nLDSbocEvpXk8IIUSlKnGLSqNGjVw+rkkcB9OOal+XYW1qs/zA+Su+7u6YBECb8ZNnRNs6jO5Ql0V7zrFo7zlbuVVBtsWejJxLyGDuxhNO56/I6W1w4Hft8YAntM3zXGkxvOgl5ovTfKg2Ffjv/zqXl2bKsiUbPnAYpDt2nut6XcZDrRYQ3lIbM9N8BJzarA30Nfm6fo4QQgiPUOalM8+ePcuGDRuIi4vDmm833SlTplxxYJ6k+DEqxVSwZENmssvVTi8kZ/L64oMAeHs5dynpXcwyUkpxyWEMy+XUTKfzV9yisvUL++OQqCu7VnFSL9ofB9SD5LNa68qc0XDj59q6K0XZP19bdA6g64TC6+v19mX9AW79QZsara+kqd5CCCHKrEyJyty5c7nvvvswmUyEhYU5tTTodLrql6jkrUybr7xEk5UzU+Dj7tqHcJc77OMmcl1IticaU4Y4z1Rxdf1LqVlOiUp6tvN4IOuVNqlk5i4u13EsNOpbdN0r5Tge5eav4MvcbqaTG7Sl+bvfXfhz0xPg13u1x4H1ncemFEen0xZ3E0II4fHKlKi88MILvPjiizzzzDPoy2uxriqgsGVUimzEuHxcS1IATvxV4HTejJ96Qd4Ma1Pb6Zy+BJnQsv1a15NBr8NiVVfW9bP3ZziUu9lkk0EVs020o0Z9tc0Aa7WEiNbO5yxZLp8CaDf8F4ck5uq3KiQ8IYQQ7lemRCUtLY3/+7//qzlJSiEf/iX6HHfaz6bghfJm/BgMBS+md2qpcp0Q5U1NDvQ2Ep+WTUpmDmlZOfiayvBPe/AP+2PHBdoqSp+HIKwZNOgOXn7aYNcLWjdYkYNqLx+Hoyu0x5E9Xa+KK4QQolooU6Zx991389NPP5V3LB7LPuvHdWZSZCOG4/gdF5lG/hk/jtIcZvuYDEX/Uz0/2r6WSZdXl/NPbClXe1UKzueuXzJyBkS0Krp+efDygXY3akvq6/Vw/3pok7uL8apX4eyugs/JToePHHZxvuXbio9TCCGE25SpRWX69Olcc801LF26lPbt2+Pl5bwuxrvvvlsuwXma/C0ohSUuTpxmsbhKVJzXUHEUE2/faXh0+7r8uvOM7dio15GT288TGepD76ZhtnMZ2VYOnE2iVZ3A4uPLs+5tuHREe+xfu+i6FcXgZd/h2JKlrX8yeatzne1f2R/3eQj8XazzIoQQotooc6KybNkyWrZsCVBgMG11U+xEmqIqOHb95KuXmWPh1v9uAcDgokXF4jDg5NFhLZwSlfAAM/WSdtNUf5ZTup4FZghZSjtY5dwu++OofqV7bnlKc5gJlH7Z+VxKnH2xtvBWMPy1yotLCCGEW5QpUXnnnXeYPXs2EyZMKOdwPFNh049LlJNZCx+jcjQuxfa4S8PgAk/Ntti7jRxfq0fjUMK5zCeZ2t5KBzL/Qq93XnitVNOUM5Lgn4Xa4xs+1/b0cZcmgyBGS96cxqkoZV8mH7TuKSGEENVemcaomM1m+vat4KmrHqiw1qKix6g4dP3kSx7yxqcY9TpeG9OuwFMdW0UcW0zG9ohEl2xfaK6xdwpmg/N021I1qKx02EzSXIruoorQ71FtxVtwTlTO7da+QEtmmg6q/NiEEEJUujIlKg8//DAfffRR8RWricIaJ0rSopKREl/ohfLGmNQL9nGZBAV428f+6HU66nGRQFIwGQzorPb1V3yMOoJ8vXhiREtbWam6fhJicl/ECE0Glvx5FcHLG9qO0R5bcxOVzGSYdZW9zs1fFXiaEEKI6qlMXT9bt25l1apVLFy4kLZt2xYYTPvrr7+WS3CeIv9ePwXOF5ITJMZfJGj+nS6upClqIC3Ao8Oa89Qve7m7X2P8t7zLRu//kKUMbEuZT/Ocow6X1a4zaVAz9p9NZPHeWKcNI4t08SgcWaY9vuFzz1hSXp/735MlC1IvwZaZ9nODngPvIPfEJYQQotKVKVEJDg7mxhtvLO9YPF5pZ/1cOPUPTh+pSWdg/gPaUvqDn7e1qLiamgwwuFVt/n5Om4Gjvt0FgElnoWPWTvpkOeyRkxhju65JaWuKXE4t4eZ+ax3Gerhrtk9+PsH2x9vnwLrcBd3qdYarnnRLSEIIIdyjTInKnDlzSlTvr7/+olu3bpjN5rK8jMcornWisPNWi6Vg4e7vtO+N+jJ7sza1trAWFUc6h5Va/dJjC71us/BwoDnvrThMbFIG029s71wvIwlSzkNoU23tkrTcmTUNe1f8kvklZTRD7fZwfq+2nkoeGUArhBA1ToUuLTtq1CjOnDlTfEUPZ+v6KeVmP9YiVleNvRzPyn/iAAj29SpYIT0BLhyyzxpyvFZ2WsH6uRKTE22Ptxy/5Hwy7TK82wY+7ga/3AXHVsOxldq5Xg9oiUsupRTRF1O5mOK86WGlCY1yPm45Ghr2cksoQggh3KfMuyeXRInHSVQRpV2ZVlldtKjkysiyzwZ65fp8M37SLsN77SA7FdreAH2maBv15blwKDcgvW18iqvXtOS//5ejISt308HYvbB1lv1caBOnqu/8eZiPVx9Fr4NFU/rTum4lzwbKTHE+/veXlfv6QgghPEIN2aznChU266eYp7ns+sk7l7u0fqMwX5qG+zufjD+hJSkA5w/Ahvecz8ds1r6b8j0PUA5L9l9OyeLn7adZ9c95LWlMcmjdslrsOyX3fQTqOHcR7TmjtcxYFRyJy5c0VIb4aPvjkTO05faFEELUOG5NVGbOnEmHDh0IDAwkMDCQ3r17s2TJEneG5FLegm+l3T1Z5XbXWFXBJ+YlKgZXWyQ7LXRmhaxCEoVeDxQoal3Hz/Y4OTOHx3/azV1zt7H94FH48XZ7xfhoOLEegAvBnXjgm+2MfH8d87aeYldMAusOX7BV/WbzSd5YfJC9pxO5kJzJ15tPsmjPOaxXtFVzMerl7ucT2MDl+xRCCFEzVGjXT3EaNGjAjBkzaN68OUopvvrqK66//np27txJ27Zt3RmaSwWGqBSzkEpeN0wOekw4t65Ycqcmu5zxY3EYF6KsrncSDmsOHW6Btf9xKr66bQS7ajXkm82nnMrTzx8pNM4fToewZJ82QHf2X9E0CvNzOr81+rLtq0ktP9tS/t/d25M+TWsVet0rMuwVCImyb1IohBCiRqrQFpXiPsivvfZarr76apo3b06LFi14/fXX8ff3Z/PmzRUZVqklXz5PW90J9Mqi7YZ8fj8knQUglCTij/1NTnZWgXMZ8ecAyHGRDyana8mIyxk/Sefsj605cHan9ljn8M9lNIPeeTVaAN+Le7m1W/0C5TmXThUoA3gk60F2J9u7kLJyrAUH4eaFlZ7NqkNxtuMj51PYdybRth5MuQqOhKEvQb1O5X9tIYQQJXImIZ2jcclujcFjBtNaLBZ++uknUlNT6d27t8s6mZmZZGbaWxuSkpKuOMbixF84x8hlg7jenMOm5D6wciX89QEA9cL/wybzc5ijc9jx/td06dgJNn4IQOJNP9N917MA5FAwoZj39ymgMUZDvlwx5QLMn2g/Tjhpf+w4cNZoBl3B67J9DpHxccA4W1EtEhm0T9vML0tnxqTs9zABf9YcsC/Hf+JS4TOKjl9MdTp+acF+AEZ3qMsnt3Yp9HlCCCGqjrikDD5adZSvN2ufPx0bBPH7ZPdtVlumFpXBgweTkJBQoDwpKYnBgwfbjpOTk2nSpEmBeo727t2Lv78/ZrOZ+++/n/nz59OmTRuXdadPn05QUJDtKzIysizhl8qls8cw67QZOm1M5+0zboBGCVtt50LTo53OZf6zwvb4e0vBfWl0ueNe/q97vvcQf6LwYBxXZO0yHoIaQOvrILIXmO3nTAlHnZ4WpbO30PzhNcrp3DZri0JfzsfLRSLkwtHzbhhsK4QQotzEp2YxfclBWj6/hB5vrLQlKeC8ga47lClRWbNmDVlZWQXKMzIyWL9+famu1bJlS3bt2sWWLVt44IEHGD9+PAcOHHBZ95lnniExMdH2FRMTU5bwS8WxVSjIx8tp5KyefF0eDi0eKnefmlPWcOZZBpOfDkVUmC9jezR0PmEpeF8B+PdsMDrMfOk6Xhvde8vXcPcyrYUl79rKgg4redOVGum07poD1kb8L9ueNE3MepQUXC+Z/+X4bnh7lew/j2xrBXT9CCGEqFApmTl8uPII7acto/Ory/l87XEyc+y/z1vU9ueD/+vErpeGuzHKUnb97Nmzx/b4wIEDxMbaV0i1WCwsXbqU+vULjo8oislkolmzZgB07dqVv//+mw8++IDPP/+8QF2z2Vz5q9w6dl8pheNcZZ2yD5DVATgc5+2abEHvcnazHlWw2wcKT1QM5sLPgX0DP8AUf5Qj5ju4RCD3ZD3OO6bPAEjHxPl0PXhr9ZKwD5qtHWjmfJK9S6h+iA+3947iw5VHaFUngH9iC++jPHUpjaSMbAK9XSxcJ4QQwmNkZFv4ZvNJPlt73OWCng1DfXlwYFNu7NIAk9EzVjApVaLSqVMndDodOp3OqYsnj4+PzxXvqmy1Wp3GoXgch1YTnVMKouyryIJtpo4VPVYXDVc6FEZXU5Pzun70RluyA2gtJr6hkH7ZdVydxsGmj+3VdVZqk8AYw1+2sm9zhhBLGG9m30Kkdxr/eLWFTO09vHJ9O+77ejsAd/drTKs6gbSICGBE29oY9DpGvq+1lH1yaxee/20v8Wn2xCjHquj22gqWPzqgwIwhIYQQ7pVjsfLz9tN8tOooZxLSC5yvHWjm/quaMrZHQ7xL2OVfmUqVqERHR6OUokmTJmzdupXw8HDbOZPJREREBAZDyd/kM888w6hRo2jYsCHJycl89913rFmzhmXLlpUmrArlPCBYObWw6PInLU5dP1qSkYPBKZ2xKh16ncptUdFB7D74+7/gFw5dboeFj2gVjT72VWQBAuvB0Jdh8ePaAm35DXkRfEKc98YBWhu1qcT7rFH8ah0AwKeW65lzU3d0P+4CtISjZe0A23MGtYwAQK/X0bZeEPGp9paclnX8eWBgU95Y/A+9moSy+biWOGXlWPl2yyla1w1gQPNwwvyr9v5OQghRlVmsioV7zvLBiiMFJkIAhPh6ce+AJtzVt7FHJieOSpWoNGrUCLAvVnal4uLiuOOOOzh37hxBQUF06NCBZcuWMWzYsHK5fvnI33HjOEbFoetHOScqea0rVvQohxVYsjFiJju3RUUPa6bDPwtzL+LQwtLjXtjwbu7jiVC7rfbV+hrXYRrN0GxogUSlD1p3XRrOiUN4gJkHBzbj9cUHmdAnioahvlzToS6J6dl0bRTiVDfEz8TiKf1Jz86hWUQAzSICGNO5Polp2Qx7b52t3qx1xwEY2ro2/x3fzXWcQgghKoTFqlh+4Dzvrzjssrve12Tgrr6NmXhVkyrVVV/iRGXBggWMGjUKLy8vFixYUGTd6667rkTX/PLLKrZ/S75kRJd/+rWrrh+dHpwSFYMtUWkY6mtfxh60jQgBwltBx7H2RKXxgJLF52JdlTznW46DvdrjazrUpW29QNrVD+L6zvUI9zej0+n4uIgpxm3qOe/1ExHgTbCPiQ4NgthzOtHp3PmkjJLFK4SodBarYt7fp9h/NonOkcHc1K3iZ0+KimO1KtYducD7K46wKyahwHmjXsftvRsxaVAzalXRlu4SJypjxowhNjaWiIgIxowZU2g9nU6HpYg9bqoalb/1qJBZP/m7fgzxxwCwYCBB2cdtnFVhtNCdQYfizX93gG8cxqHk5H7AG7yckw6DqWTBulpXBaBeZ1oPuwv2aq0fN3eLtC3GFxHgXbJru2Ay6lkwuR/dXlvhNCgruyIWgBNCXJGMbAufrz3OeysO28q+23KKeX/HMP3G9rRw6P4V5e9SSiYGvY5g3xL+Pi+CUoot0Zd5f8VhW/d7frd0i2TK0ObUD676+6SVOFFx7O4pr66fqkDlHzBbSNePNn7FfhyepC2GZtEZScGXIZlv4U869xgX04IzRPibtH5Bx+XxL/yjfTeYnBMVYwmTicJaVLz8CA/wxqjXkWNVRJXzgNf8I8f/iU3m3zM3MveuHvib3bpLgxA1XlJGNu8sO8RXm066PL/9ZDzD31vH6PZ1efumjviYDGRkWzh1OY2Gob6YDHp2xiSg00GnBsEcv5jC7phEOkYG0SzCntxkW6zM+zuG6Aup9GoSyvC2dSrrLVa69CwLP22P4d3lhzEZ9DQK8+WNG9rTLMLfaUX2xPRsPl19lM9zu8UBekSF8uzo1vibjXh76WkQ4nqJiPyUUuw5ncj7Kw6z+tAFl3Wu7ViPR4c2p0n+jW6ruDJ/iqxcuZKVK1cSFxfnlLjodLqq16VTGk6DaR0eoyCn4Gylsy3ugD1wTGnTtvNmANlmJmc7rAQbs0X7bjBDcCNtJk/aZYjsUbLYfEJcl/ecSJCPF+ueHERWjpWGYSX7wSipHo1D2RrtnNVvOxnPzlPx9G8eXsizhBBF2Xs6kc/XHcPPZOT/ekTy6ZpjXErJpF+zWkwd3rLY58clZzBtwX4W7411Kg/wNjL3zh7sPZ3AtD/sa1Yt2nuORXvP4WcykJpVslbxiQOacFPXBny+7jg/bz9tK5/9VzTdo0J45fp2tKoTUOx2KlWBUoqt0Zd5+Y8DHDjnvCp6XHKmbbyeyagnK8dKmJ+JS6kFl5TYeuIyYz6xz8b8dFwXrm5f1+l1tp+MJ/piKr2ahJGebeGDFUdYtPdcgWuBNibwseEtaF030OX56qBMicrLL7/MK6+8Qrdu3ahbt261+I+wUPnHoRTS9VNfnYfz552qntbVpdWwCbBnrf3pud+NOgUbP4bz+wq+Zs+J2sDaMZ+WLlb/CBj3i7bb8i/32NdWyd3Yr14FNQG+cUN75m09Rcs6ATzxs32tHekCEqL0dsckcMfsrSSm21tbf9hmX9xyx6kEvtlyivdu6cSA5tqmoCcupWHQ6WgY5kvM5TQe+2l3gT8eGtfyY+ZtXWhVR/tA69oohBu6NOCdPw/xP4fWlpImKaANoJ/l0Frg6O8T8Yz6YD2DWoYz584S/rHlgU5eSuXNpYdcJgoBZiPJmTlOZVm5C6Y5Jin1g30Y1CqcX7afIT3b+f4++O0OpgxuxlUtI/hg5RGnnesL07dZGI8Pb0nnhoX8cVrNlClR+eyzz5g7dy633357ecfjeYpY8E1P0T/QVp2+wI7LeS0qdVQcHD1a8EkRbaDtDWUMFmg+VPu+ZrrWlWSs+P7JZhH+PH9NG1Izc3h/xRHbPP17vtrGvIm96dE4tMJjAEjNzMFs1LteSE+IfJRSLNxzjk3HL9GzcSjXdyrdYpXlbeepeMbP3kpSRk6xdS+nZjF+9tYSXbdLw2Deu6WTyzWOgny8eOX6dtzUNZKX/9jPtpPxAHSMDObExVQS07MxGfTc3rsR32w+SWaOladHteJoXIpTCwpAu/qBvHdzJ7aeuMxz8+1/gK0+dIHHftxNiK8Xg1tF0KdZ8Tuux6dmMXPtMZLSsxnRtg6DWmlLJlitivVHL3L8Qgr9mtXCx2TA28tQ7oNEkzKymb0hmvdXuN51fmyPhjw6rDkRAd4kZWSz8uB5Pl97nIxsi22/NG8vPQ8Nbs5dfRvjY9K65V8b057fd51h7aELhAea+XytluR9uOooH65y8XngoGujEJ4Y0ZJeTcLK8Z1WDWVKVLKysujTp095x1I1FLrgW0FWDAVam8xoWXa/zA1gcbGnkaGcpoxd/RasmQG9J5fP9UrAz2xk3ZODaPrsYgCsCj5adYSv7+5ZLtfPyLaw/2wi9YN9Cfb14q1lh7icmsVdfRvzy47TzN14grpB3vz56AACqtDUO1HxrFbFrtMJgDbOYtfpBG78dKPt/HdbTjGsTW18TZU/pmrHqXju+HIrKfn+Mr+tV0Pu6tvYtvnnm//uwLYT8Tz0/c4SXXd4m9q8NqYdEYHFj3Fr3yCIn+7vza6YBBqE+BIeYCYrx8ql1EzqBml/7Dw5siUmg972O+2x4S2YvvgfvL30PDWylW3tpOa1A7ihc31+2nbaFvsvO7Sk5r8bomlfP4hnr25Nz8ahWJRid0wCfmYjreoEsP1kPI//tNtpc9R5f8fgbzbSLMLf5ayWPPWDfQjwNtK5YTBhfmYmDWpmSxAKk55lYfZf0eyOSSDUz0T3qFDeXX7Y5aJoPRuH8tzo1nRoEOxUHujtxQ2dG3BD5wYAZOZYSMnIKXQtqes71ef6TvVRSpGUns33W523ghnWpjZTh7Vga/Rl1h2+wPg+UQxoUbO70HWqNFsc53rqqafw9/fnhRdeqIiYSiwpKYmgoCASExMJDKyY/rl/tvxJqyU3aQehTcAvAmI2A3CESJpT+H5Dx/VRGCZt5Kq31tjK7jIs4UWvr0kLbYOvX4B9XEqeBt3hnhVUZVFPL7I97tYohJ8fKJjU/vh3DP/dcJyGoX58Oq6Ly6WakzOyefyn3VxOzeKhwc3536aTrDh4HqNex41d6vPjttMFngPw64N96FJDmkRF8fadSeSajzYUW2/K4GZMGdIcBfwdfRm9Xkf3qFAMrlaQLqO/jl5kV0wCfZvVwmK1cvuXW0nL19Uyvncjnrm6daGLcKVm5vD91lO8tuigrcxxmYCxPSJ5emRrgnzdn6y/t/wwH6x03SpRWYa1qU1YbhKSkpnDVS3COXw+mbeWHeJIMZvt1Qvy5tFhLRjTuT5eFdRSm5iezYLdZ2kR4U+PxqHVeyiFg9J8fpf4z4epU6faHlutVmbNmsWKFSvo0KEDXl7OPxDvvvtuKUP2XCp/q4lDi0pRSQqAVWfAnO8/7kNKy7p9vfSu9+4p6QwfD3Zv/8Z8sT4agCwX41Qysi08+Ys2luXw+RT2nU0skFgopXhr2SGW7dfG/dzh0MydY1WFJikAmdkyNkZo3Qe3z97CvjNJhdZ5bUw7nv9N66Zw1fzeKTKYt2/q4DS7pSzWH7nA7V/a/xt+a9mhAnXu6tuYp0a1xGwsuhXAz2zknv5NGNmuDscvpNKvWS30eh2nLqUREWj2qFVGHxzUlNZ1A2hXP4g1hy7Y7nVx7ujdiEeHtmDjsUtM+m6HrbxjgyBeHdOO9UcusvqfOJqG+7Nwz9kix9UsP6D9Dpn3d8k3sb3vqiY8eFWzSkn2gny8uL1Xowp/naqsxInKzp3OzY2dOnUCYN8+5//wqnU2mG+MSnGsOgP1gry5o3cjLqdmsfzAeazW3MQl6Qykxxd8Uu9J5ROrGz17dWuiavnx3Px9toFljvI3XW86dokO9YOcxpZsjb7sNMCvNPafTaR305rXjys0GdkWXl14gG+3nHIqD/UzcTl3gOOYTvV4/Yb2+JmNZGRbnFonHO2KSWDou+t4fnRrrutUj8upWTSp5V/oZm05Fit7zyTiYzLQsnYAG45edEpQXLm3f2OeGNGq1BvANQjxdZraWt6z+cqD2WhgZDttRsttvRrx764NWPVPHM//to/LqVl0bBDENR3q8dWmE+h02p5jA1uE2z5HRneoy6h2V7P/bBJNwv3wy13uoEODYCYN0jaz/c+/OwDaQpPxaVmE+ZmZuzGaT1YfKza+W3s2ZOqwFpiNev7cf56ujUKIqiX7lXmaMnX9eIrK6Po5sHkpbZbeoh2ENAbfMDizrUTPPWRsRcvn7V07bV9cSvucvcwzvVb4k6YlFn6uCtly/BK3zNK6yPa9PMK2nspfRy8y7r9bCtQf06keI9vVJaqWLyG+Jnq+sbLY1+jXrBYbjl4E4MYu9dkdk8CxC9qeFq3qBNCqTgDv3twJfTk23ed5Y/FB5m09RfsGQTw4sBlZOVZ6Nw1j0Z5z/LHnLB0bBPPosBbl/rrVUVxyBrM3nMCo13FL90gS0rJpFuFf7PiC/JRS/LrjDI/9tNupvEdUKF9O6IaPl4H1Ry/SLNyfyFD7h3pKZg7fbD7JjCX/2Mq6Ngph+0kXf0jk+v7eXvRsHGr7b8tqVfy4LYanf91bZIyvjmnH5uOXWLTnHPdd1YTHh7essC4FT2a1KnKsqkJ3503NzCEj24LRoOfP/bH0aBzK4r2x7I5JYPLgZrSrH1Rhry2KV5rPb0lUinFg0xLaLPs/7SAkCnxC4eyOIp+TZ7+pPW2ftfeNt3phCR1y9vOj2WE/nrBmEH/SPpW4miQq8alZdH51ue3YZNTz0rVt+GLdcaeBcq40i/DnaG7fcbdGIbaZCP/q0oDoiynsOJVAt0Yh/Hd8N278dCNBvl78eF9vPlhxhI9XOzfdL390AM2LWHEzI9vC4fPJNAz1RYeON5f9Q1qWhQcGNi10pc43Fh90OSWzlr/ZafG7eRN7FRihn5VjxapUpTfPW62KPWe0/7Y61A9i+cHz/HX0Ir2ahDmt4VAS5xLT+XjVUfQ6HQ8NblaiwZquZOVYefvPQ4VOb10x9SqaRZRs4aqjccmM+mA92Rb7rzODXsfiKf1pWadk3Tb7ziRyKDaZ6zrVw6jXcfh8Cg98s93lhm4AbeoG8p9/deDohWQe/WG3yzp5pl3bhtt7R5XreBchqjJJVMpRwUQlBM6WbNT9rv6f02nI/9mOu722nKjUPfxsfsVe6ZZv4KcJkLvbcnVJVAAGvLmaU5ddJyXPXd2azccvsfKfuEKf37VRCJ/d1pVRH6wnyMfI4of7E5uYwfID57m+U33CA5xH1S/YfZYpLmZEzLmzu21HaEdKKYa+u5ZjF1Ix6nV4exlsMy9u6RbJf/7dgawcK7/tOkNGtoXrOtZj5tpjtimFJfHStW24s29jAJbtj+W+r7fjZdBxU7dIAryNdGkYwogKWMEzI9tCzOU0IkN9ybJY+ffMjRw+73rg4Iwb2zOmc32Mel2RU7stVsVna48VGF8xZ0J32/TR/DYcucj7Kw4T4G3kX10bkJCWTc/GoWRbFFd/uL7Y93Fbr4Y8PrxlocuOX07N4p6v/mbHqQSn8ndu6siNXepfcVe0xapYcfA8z/y6F6UU8WnZxT7HcQHE50e35s6+jSVBESIfSVTK0f6Ni2n751jtILiRlqic21WyJ+dLOn78O4Zffv+FH4wv2Qtv/VHb8Th2L3gHwdOnqC4cZ//k98+rI0nLsvDKH/v5bddZl3VWPnYVTcP9sVi1/aeL68JZ9c957ppbsFtuZNs6zLytS4EPrXv/t8020M6VTpHBBHgbWX/kYpGvW5wnRrTEbNQXOg5i2SMDSvxXf0kcOJtkSwJ0uoJrFhalsMW54pIyGPbeOqdFyBw9M6oVN3SpT0JaNlFhfqRl5XDbl0UPZHX02LAWvLfiMNZCYvUzGQjxMzFzXFfaNwgiI9vCm0sPMfuvaKd643s34vlr2lRYd8r5pAw+X3u8wOuCtvDh2B7aPlpnEtKpHWCWNX2EKIQkKuVo/1+LaLv8Vu0guJGWTMTuKfpJeVy0juSc3IJxznB7wW2/gjkA1r8DPSZCsyHlELVnKCxRWf/kINsYgQ1HLnLblwXHrMy4sT3/16NhqV4vPcvCc/P3UjvImy3HLzn9ld2qTgB/PNTP9gH27Py9fLel7Enh7heHs/bIBRbsOstVLcN5IXc2w8rHrmLtoQu8svBAMVdw9ua/O3BzOexi+8Pfp3jql6LHSYC25kR8WlaBqbGgTcmcdUc32tUPQinFd1tPOS3gBXD/VU35bG3xgxWL8+DApkwd1gKjQU98ahaBPl7sikng0R92Fdoa52qJ946RwXx9d49K27o+IS2LJ3/ew5rDF3h+dGtu79Woek8kEKKcSaJSjpwSldJy1Y1zejv8d7D9ePwf0HhA2a7v4e6YvZV1hy84jd3I61LJk5lj4fn5+wj1N7HhyEX2n01iaOsI/ju++xW99tmEdJ76ZY9Ta8ib/+7AjZ3r8/Hqo7YVJ4N8vEjJzMGS+6f87AnduPd/223Hrux6cZhTV4RSiv1nk6jlb6ZOkDZeY+3hCy5XDn14SPNC15W4uVsDujXSVvG9pmNd2+JjqZk5/L7rLHqdtumYVSn8zUanD0aLVXHv/7axqpCutIkDmjC+TxSvLzpAryZh3NE7iuMXUnjsp93szNdtkmdsj4asO3zBafGr/s1rMev2bviYDJy6lMYtszZxLjGj0HvVpm4gT41qxVM/7yHA28ixCylYFUSG+vDH5H6FdumkZ1lYuOes05YMhfljcj/aN5CBkUJUJZKolKN9f/1Bu+W3le3JrhKVMzvgi0H240f2QfCV/yXtiS4kZ/LX0Ys0Dffn2o834ONl4O/nhxa6o/LRuBRWHDzPjV3qExFw5evJ7D2dyLUfOy/0VS/Im7MOH6wHXxnJxmMXWbT3HA8PaU6jMD8upmTS7TX7ontv/buD7QMzf5JSGKUUu08nOm0+tvKxq2gc5sfm6Es0CPZlzeE4Xvx9v8vnX9exHjP+1Z7E9Gx6T19V4HyfpmF8d28vAKIvpjLo7TVO5/+Y3I/lB8+zbF8sb9/UsdAP8rzFwzo3DGbpvljb+jeu/O+uHgVWyEzOyOaT1cdctq58dltXRrSt7ZRQxSVncC4hgw4NgkrUAnHyUipnEtJpFObH28sOMX/nGdu5GTe255bukdKSIUQVJIlKOdq3YQHtVhS+p5HSGdAprRn6zexbmGL+A29rbpO1q0Tl7C6YdZX2eOR/oNf95RyxZzodn4bZaCgwALainbqUxoC3Vrs8t/XZIYXOWJm/8zSL9pzj6VGtaRbhz4mLqYT4mkq9ANTCPWdZ9U8cjw9v6XJTyO0nLzN21haXC+MV5/Bro1iy7xwPz9tlK2sa7sfvk/sVmgwWZ1dMAnfP/dtpQ7WujUL4+u4ehS4vr5Ti8PkUIkN92Hj0EofOJ3N3v8YVMrNp5cHznLqcxh0yg0aIKk0SlXJUXKJi1ZvQW7Vf6lOyJvOW+UvMKrep3FWicm4PfN5fezx2HrQcVd4hi3zeXX6YD/N1tyx5uL/HbIuempnDyA/WEXO54P4iRcm/c+u0a9swIXeG0ZVQSvHQ9zv5c/953rulE1e3ryOtFkKIclUhS+jXWMXlcXo95P4xbEGPUsVska5zmAVQXhsQiiJNHdaCpuF+tpaHb+7u6TFJCmhLov94X2+OnE+hWYQ/05f8wx+77TOhpl3bhms71uPTNcfo2TiUiV9vB3BKUn6Y2Iue5bSrqk6n4+Nbu5TLtYQQ4kpJolKMYtubdPbmbQt69lsb0VV/hGR9IC4nnDomKnpJVCrLiLZ1eHRoC1rU9qdf8+K3ma9sdYN8bLvUfjS2M+H+ZtYciuPjW7vQpp6WVL1wjbbb9tgekbYdVwPMRlY9PrDSu9SEEKKySKJyhZRDoqLQ8UL2ndxnXEizYZNp6+oJeodbbpQPl8ri7WXg4aHN3R1Gib14bRtevLaNy3O39WrE/rNJ9G4SxtOjWkm3jBCiWpNEpVjFDHJ0aCGxoOeAiuKjkKdZ0fcq1/XDmkGHWyA7DepJ87oovbb1glgwuZ+7wxBCiEohiUpxiun7UUZvyN3exYKWtBiLmo2g18ONs8orOiGEEKJak/Wdr9Dl7lNtj7WF3kEvTfFCCCFEuZBEpRjFzd7OrN/L9jivRUXWdxBCCCHKhyQqxSlu2k++MSogiYoQQghRXiRRuUIBwfYlxS8qbZlySVSEEEKI8iGJSjEURbeoBNeqwz9X/8wdWU9zWGl79hhkjIoQQghRLiRRKU4Jdhho1WMYxwJ72o59TOW/x4kQQghRE0miUk7i0+ybuD06rIUbIxFCCCGqD0lUilWyXW2tDi0vnSKDKygWIYQQomaRRKUYJd1b2lpl96AWQgghPJckKuWkuPVWhBBCCFF6kqgUp4QJiLSoCCGEEOXPrYnK9OnT6d69OwEBAURERDBmzBgOHTrkzpAKKnGiIpmKEEIIUd7cmqisXbuWSZMmsXnzZpYvX052djbDhw8nNTXVnWGVieQpQgghRPlz6+7JS5cudTqeO3cuERERbN++nQEDBrgpKmchx35zdwhCCCFEjeVRY1QSExMBCA0NdXk+MzOTpKQkp68KdekYjWKXFSz31pbKJ6yZrahP0zAAWtYOqNiYhBBCiBrErS0qjqxWK4888gh9+/alXbt2LutMnz6dl19+ufKCykgsWFa7HdzyNez4H7T7t634jRvaM+/vGG7oXL/y4hNCCCGqOZ3ykHm1DzzwAEuWLGHDhg00aNDAZZ3MzEwyMzNtx0lJSURGRpKYmEhgYGD5B3VmO3wx2LnshlnQ8Zbyfy0hhBCihkhKSiIoKKhEn98e0aIyefJkFi5cyLp16wpNUgDMZjNms7nyAnOVwsmGg0IIIUSlcWuiopTioYceYv78+axZs4bGjRu7M5wSkkRFCCGEqCxuTVQmTZrEd999x++//05AQACxsbEABAUF4ePj487QcrloUpEWFSGEEKLSuHXWz8yZM0lMTGTgwIHUrVvX9vXDDz+4Myw7V8N3JFERQgghKo3bu348m6tExaNmdAshhBDVmnzqlpq0qAghhBCVRRKVokjXjxBCCOFWkqgUSbp+hBBCCHeST91SkxYVIYQQorJIolIU6foRQggh3EoSlSJJ148QQgjhTvKpW2rSoiKEEEJUFklUiiJdP0IIIYRbSaJSJC1RsSiH5ES6foQQQohKI5+6JaCcunukRUUIIYSoLJKoFCW368cpUZE8RQghhKg0kqgUSTn8fy7p+hFCCCEqjXzqloB0/QghhBDuIYlKUWyzfhy7fiRREUIIISqLJCpF0hIVKzLrRwghhHAHo7sDqApiVDh+YZHU986Eep3dHY4QQghRY0jzQFFyu34s6Fna5TOYuAbMAe6NSQghhKhBJFEpkn2MioxMEUIIISqfJColoCRNEUIIIdxCEpWiuNrrRwghhBCVRhKVItkXfJNZyUIIIUTlk0SlBKTrRwghhHAPSVSKIj0/QgghhFtJolIkh64f9wYihBBC1EiSqJSAdP0IIYQQ7iGJSlFk1o8QQgjhVpKoFCmv60eHTqb9CCGEEJVOEpUSkHYVIYQQwj0kUSmKdP0IIYQQbiWJSpEc9vqRnh8hhBCi0kmiUgIy60cIIYRwD0lUiiJdP0IIIYRbSaJSJFnwTQghhHAnSVRKQLp+hBBCCPdwa6Kybt06rr32WurVq4dOp+O3335zZzgFSdePEEII4VZuTVRSU1Pp2LEjn3zyiTvDKIJ9wTeZ9iOEEEJUPqM7X3zUqFGMGjWqxPUzMzPJzMy0HSclJVVEWBpLNvx0JyALvgkhhBDuUqXGqEyfPp2goCDbV2RkZMW9WOxesGYDcEbVqrjXEUIIIUShqlSi8swzz5CYmGj7iomJqbgXs1psD5/KnijDaYUQQgg3cGvXT2mZzWbMZnOlvuYFY10yMVXqawohhBBCU6VaVCqXjEwRQggh3E0SlcIohxk/yKQfIYQQwh3c2vWTkpLC0aNHbcfR0dHs2rWL0NBQGjZs6MbI7GSxNyGEEMJ93JqobNu2jUGDBtmOp06dCsD48eOZO3eum6LKI10/QgghhLu5NVEZOHAgylNXf82LS5f3TVpWhBBCiMomY1SKIV0/QgghhPtIolIoD23pEUIIIWoQSVQKI7N+hBBCCLeTRKVYkqEIIYQQ7iKJSqGk60cIIYRwN0lUCmPr+tFIu4oQQghR+SRRKYbM+hFCCCHcRxKVQknXjxBCCOFukqgURjl3+sisHyGEEKLySaJSDGlXEUIIIdxHEpVCSYoihBBCuJskKoXJv+CbDKoVQgghKp0kKsWSBEUIIYRwF0lUCiVdP0IIIYS7SaJSmHxdP9KwIoQQQlQ+SVSKIe0qQgghhPtIolIoSVGEEEIId5NEpTC29d7yZv0IIYQQorJJolIM2etHCCGEcB9JVAolXT9CCCGEu0miUhjbrB+NTjb7EUIIISqdJCrFkK4fIYQQwn0kUSmUdP0IIYQQ7iaJSmGUrdPH4f+FEEIIUZkkUSmGdP0IIYQQ7iOJSqGk60cIIYRwN0lUClNg1o/7QhFCCCFqKklUCpVvU0IhhBBCVDpJVIqVO5hW8hUhhBCi0hndHYDHUjVzjIpSitjUWEwGE3su7OFs6lnWxqwl2DuYMO8w6vvXp45fHdvX0YSjpGanEuAVQIRvBBG+ESRlJbHt/Db2X9yPv8mfIFMQtXxqEe4bTi2fWmTkZHAm5Qz7Lu7DoixsOLOBYHMwvkZfwnzCCPUOJdQ7lDCfMM6knCHHmsP289vxNfra4gr1DsXHy4dTSacw6o34GH0INAUSbA4mNSeVhIwEgsxBGPVGDscfJsQ7BF+jL4GmQALNgYSYQwgwBRBkDiLAFECwORg/Lz8yLZkkZSZhNpjx9dJeLyY5Bh+jD94GbwLNgfgZ/QgwBeDn5QdAbGosoC0KqMtNbKOTorEqK94Gb3yMPvib/PE1+uLr5avVza2XYcngbMpZTAYT3gZvfL18bfW8Dd62hQazLFlcSr/Enot7UHn/beq066Rlp3E+7TzeBm9MBhM+Rh98vXzxMfrgY/TBoDOg1+ltXzqdDoPOgA6dU7lepychMwGL1YLJYMJkMOFr9MXb6I3ZYMZsMKPX6bXnVnLmblVWLMqiHeS+/UxLJgmZCZgMJrz0XraYjTpjjV+gUSmFQmFVVpRSWLFiVVZSslJIyU4hy5JFQmYCVmXFqDfipfeyfTcZTLbHecd5jw16g7vfWoll5GRo7z/3PliVFdD+W8q0ZJKclYxFWcix5pBjzSHLkkVydjIAerSfE71Ob/s5KXDs8POj0+nQY3+cdw6wndPpdLZzjtdwLM97Tt518/4d877nxe9UpsCK1V43ryyvXm5d0H6PZFgybMeKfN/zled98zP50TG8Y0X+cxVJEpVCOY9RqY4upl9Er9MTYArg4KWDLDi2gB8O/eDusErkbOrZUtW/mH6xgiKpOHqdHm+DN2k5ae4OpYC8X6QGnQGDzoDJYLIlM2ajGR+jDya9CYUiPScdPXr7B57Bi7TsNM6lniPbmo1Jb8LL4IVSWl2dTuf0AZmek05saizZ1uwSx6dDS8YMeoMtTr1Oj1FvxKg32o6dvuu170adEb1e7/TBo9fpnT+88n0w6dFj0BsKXtfh9XXoMOqNKBQ51hzbh2eONYcMSwbJWcna6zskDka99is6r75FWbAqK2nZaSRmJdo+ZPO+krOSSclOsX/QlDMdOtu/YV7CavufLt93LZO2fyDnnsu7juOHOuDyA95qtZJlzSIlKwWLsjh9gOb/EHb8sM2x5pCjcirkHtRE7cLa8f0137vt9SVRKUbeGJW8H6aq7mzKWaZvnc6amDUlfk6ToCa0CGlBLZ9axKbGcjb1LHFpcU4f/pEBkVxMv0h6TrqtLMgcRKfwThj1Ri6lX+JSxiWnOs2Cm5GclUy/+v0w6Ay0DG3J5YzLXEy/SHxGPJczLhOfEU9CZgK96vUi25JN+1rtScxKxNvgTYOABiRkJpBpyaRtWFviM+NJyEggITOBS+mXaB3WGouykJSZRIApgFo+tUjITCA5K5nLGZdJzkomITOBlKwUErMStV/yWSkkZSXROrQ1aTlptnORAZEY9UaSs5JtHxJ5f6EBhJhDnP46UUrhpfciwjeC9Jx0UrJTSMtOIz0n3eWHSB2/OmRbsknLSbPdH6uyukxSogKjCPcNd/prK8eaQ4RvBHqd3naNjJwM0nLSUErZPuDyHjv+le34pVAEmgIBrcUi7zr5Y7Z9yKJ9GLgzmTLqjAU+lBSKHJVDjkU+rPLz9/K3tT75e/nbWhWyrdlkW7K177lfOdaC9zXLmkWWNctN0ZefAFMA3gZvDHotOc1rmTUbzPYWityfE6VyW2UcHuf97OX9LDj+jOX9z7EVx9YK4tja5dhKkvtaji0k+ZO+/C0xgOskMV8CCfaE0c/Lz2WrqOP1nL7rdDQKalTx/yBFkESlMKXs+jmTcoan1z1NSnYKzYKbMazRMLrX6U6IdwgLji3g6wNf4+flR9OgprQMbUmToCY0CW7CsYRjzNozC6PeSD2/ejQMbEg9/3rU869Hbd/ahJhDMOgNrDy1kvOp5/E3+VPLu5atiyTEO8T2V1dadhr7L+3HZDCx4/wOjicex2wwczzxOPsu7sOgM5CSnVLk+2gS1IRnez5L54jOpOekE2QOKrRujjWH1OxUWx2lFCnZKcRnxFPXry5eBi+Xz0vPSbd1I1RVSikyLZmkZKcQbA62/RuU9VqOvzTy/mJOz0knPSed1OxUfIw+hPuG27qbKpNSuR9OlixbC4DjL2aL1UKmJZNMayZZlizSc9LJzNGOM3My0el0BJmCyLZmk2XJItOSibfR25Y8Zluybdc36o34efnZ6mZbsjEZTIR4hxDuG26LKe8Xdt79sFgt5KgcMnIynOKyKIv92KGZP3/8jse21o5CPqAcuxPyypxeJ9/1HL9brBanlhajzmhrcfH18sVL72VLEPKSB1etM3qdXusuzX2OUW/EqDPiZfAixByCl8HLqQUo70MqrzWkNP/2tsTFkk2OyrH9e1mUxd6y4fCBDvYyx1aP/Mfgohsj9zXzuvny3quvl9YFafvQhgIf4nn/XeSV+3n5YTKYbK1h6OxdOo4f4MLzeUSi8sknn/DWW28RGxtLx44d+eijj+jRo4ebo3JembYw289vJykzidUxq9l1YRcARxOOsvTEUgAifCOIS4tzql8arv5azC/QFEiAKYAzKWdKfN2utbtye+vbuSryKi5nXCY1O5WGAQ2d+qBNBlPRsemNTomMTqcjwBRAgCmgyOf5GH1KHKen0ul0eBu98TZ6l8u1HOl1evxN/vib/K/42uVBp9PZunU8lUFvwIDBo2OsinQ6na31hZLnN0KUK7cnKj/88ANTp07ls88+o2fPnrz//vuMGDGCQ4cOERER4e7wilxHZd/FfUxYOqFA+aDIQRy6fMjWRZJnQtsJZFoyOZZwjOjEaC6kX7Cd+7+W/4e30ZszKWc4nXyaC+kXuJh+sUCS0rNOT1sXSlJWElZlJSkriaSsJFsdo86Iv8mfhMwE7m1/L0lZSaRkp3Bdk+sw6o20Cmtla94HLZkSQgghPJHbE5V3332Xe++9lzvvvBOAzz77jEWLFjF79myefvppt8SUnJrAmZjdBBoNnNdb0Rnjic86z9kU52xlz4U9tsdeei9yrDlM7z+d0U1GAxCfEc8/l/8hOjGacN9whjUa5vT8tOw0zqacxdfLl3r+9QrEkW3N5kLaBdt4io7hHW2zRvLOJ2clE58RT1JWEomZiUT4RtAmrE153g4hhBDCbXRKlXIwRjnKysrC19eXn3/+mTFjxtjKx48fT0JCAr///rtT/czMTDIzM23HSUlJREZGkpiYSGBgIOVlyV9f8+TRN0tcf2DkQD4c9CEp2SnFdnsIIYQQNV1SUhJBQUEl+vx262jGixcvYrFYqF27tlN57dq1iY2NLVB/+vTpBAUF2b4iIyMrJC6dTo/ZqjBbFQarHpQXJr3ZPv3S4SvAK4DhjYbbxmcIIYQQovy4veunNJ555hmmTp1qO85rUSlvI/uMY2SfceV+XSGEEEKUjlsTlVq1amEwGDh//rxT+fnz56lTp06B+mazGbNZRvULIYQQNYVbu35MJhNdu3Zl5cqVtjKr1crKlSvp3bu3GyMTQgghhCdwe9fP1KlTGT9+PN26daNHjx68//77pKam2mYBCSGEEKLmcnuicsstt3DhwgVefPFFYmNj6dSpE0uXLi0wwFYIIYQQNY9bpydfqdJMbxJCCCGEZ6gy05OFEEIIIYoiiYoQQgghPJYkKkIIIYTwWJKoCCGEEMJjSaIihBBCCI8liYoQQgghPJYkKkIIIYTwWJKoCCGEEMJjSaIihBBCCI/l9iX0r0TeorpJSUlujkQIIYQQJZX3uV2SxfGrdKKSnJwMQGRkpJsjEUIIIURpJScnExQUVGSdKr3Xj9Vq5ezZswQEBKDT6cr12klJSURGRhITEyP7CFUguc+VQ+5z5ZD7XHnkXleOirrPSimSk5OpV68een3Ro1CqdIuKXq+nQYMGFfoagYGB8kNQCeQ+Vw65z5VD7nPlkXtdOSriPhfXkpJHBtMKIYQQwmNJoiKEEEIIjyWJSiHMZjMvvfQSZrPZ3aFUa3KfK4fc58oh97nyyL2uHJ5wn6v0YFohhBBCVG/SoiKEEEIIjyWJihBCCCE8liQqQgghhPBYkqgIIYQQwmNJouLCJ598QlRUFN7e3vTs2ZOtW7e6OySPsm7dOq699lrq1auHTqfjt99+czqvlOLFF1+kbt26+Pj4MHToUI4cOeJU5/Lly4wbN47AwECCg4O5++67SUlJcaqzZ88e+vfvj7e3N5GRkbz55psFYvnpp59o1aoV3t7etG/fnsWLF5f7+3WH6dOn0717dwICAoiIiGDMmDEcOnTIqU5GRgaTJk0iLCwMf39//vWvf3H+/HmnOqdOnWL06NH4+voSERHBE088QU5OjlOdNWvW0KVLF8xmM82aNWPu3LkF4qnOPxMzZ86kQ4cOtgWtevfuzZIlS2zn5T6XvxkzZqDT6XjkkUdsZXKfy8e0adPQ6XROX61atbKdr5L3WQkn8+bNUyaTSc2ePVvt379f3XvvvSo4OFidP3/e3aF5jMWLF6vnnntO/frrrwpQ8+fPdzo/Y8YMFRQUpH777Te1e/dudd1116nGjRur9PR0W52RI0eqjh07qs2bN6v169erZs2aqbFjx9rOJyYmqtq1a6tx48apffv2qe+//175+Piozz//3Fbnr7/+UgaDQb355pvqwIED6vnnn1deXl5q7969FX4PKtqIESPUnDlz1L59+9SuXbvU1VdfrRo2bKhSUlJsde6//34VGRmpVq5cqbZt26Z69eql+vTpYzufk5Oj2rVrp4YOHap27typFi9erGrVqqWeeeYZW53jx48rX19fNXXqVHXgwAH10UcfKYPBoJYuXWqrU91/JhYsWKAWLVqkDh8+rA4dOqSeffZZ5eXlpfbt26eUkvtc3rZu3aqioqJUhw4d1MMPP2wrl/tcPl566SXVtm1bde7cOdvXhQsXbOer4n2WRCWfHj16qEmTJtmOLRaLqlevnpo+fbobo/Jc+RMVq9Wq6tSpo9566y1bWUJCgjKbzer7779XSil14MABBai///7bVmfJkiVKp9OpM2fOKKWU+vTTT1VISIjKzMy01XnqqadUy5Ytbcc333yzGj16tFM8PXv2VPfdd1+5vkdPEBcXpwC1du1apZR2T728vNRPP/1kq3Pw4EEFqE2bNimltIRSr9er2NhYW52ZM2eqwMBA23198sknVdu2bZ1e65ZbblEjRoywHdfEn4mQkBD13//+V+5zOUtOTlbNmzdXy5cvV1dddZUtUZH7XH5eeukl1bFjR5fnqup9lq4fB1lZWWzfvp2hQ4fayvR6PUOHDmXTpk1ujKzqiI6OJjY21ukeBgUF0bNnT9s93LRpE8HBwXTr1s1WZ+jQoej1erZs2WKrM2DAAEwmk63OiBEjOHToEPHx8bY6jq+TV6c6/lslJiYCEBoaCsD27dvJzs52ev+tWrWiYcOGTve5ffv21K5d21ZnxIgRJCUlsX//fludou5hTfuZsFgszJs3j9TUVHr37i33uZxNmjSJ0aNHF7gXcp/L15EjR6hXrx5NmjRh3LhxnDp1Cqi691kSFQcXL17EYrE4/QMB1K5dm9jYWDdFVbXk3aei7mFsbCwRERFO541GI6GhoU51XF3D8TUKq1Pd/q2sViuPPPIIffv2pV27doD23k0mE8HBwU5189/nst7DpKQk0tPTa8zPxN69e/H398dsNnP//fczf/582rRpI/e5HM2bN48dO3Ywffr0AufkPpefnj17MnfuXJYuXcrMmTOJjo6mf//+JCcnV9n7XKV3TxaiJpg0aRL79u1jw4YN7g6l2mrZsiW7du0iMTGRn3/+mfHjx7N27Vp3h1VtxMTE8PDDD7N8+XK8vb3dHU61NmrUKNvjDh060LNnTxo1asSPP/6Ij4+PGyMrO2lRcVCrVi0MBkOBEdDnz5+nTp06boqqasm7T0Xdwzp16hAXF+d0Picnh8uXLzvVcXUNx9corE51+reaPHkyCxcuZPXq1TRo0MBWXqdOHbKyskhISHCqn/8+l/UeBgYG4uPjU2N+JkwmE82aNaNr165Mnz6djh078sEHH8h9Lifbt28nLi6OLl26YDQaMRqNrF27lg8//BCj0Ujt2rXlPleQ4OBgWrRowdGjR6vsf8+SqDgwmUx07dqVlStX2sqsVisrV66kd+/eboys6mjcuDF16tRxuodJSUls2bLFdg979+5NQkIC27dvt9VZtWoVVquVnj172uqsW7eO7OxsW53ly5fTsmVLQkJCbHUcXyevTnX4t1JKMXnyZObPn8+qVato3Lix0/muXbvi5eXl9P4PHTrEqVOnnO7z3r17nZLC5cuXExgYSJs2bWx1irqHNfVnwmq1kpmZKfe5nAwZMoS9e/eya9cu21e3bt0YN26c7bHc54qRkpLCsWPHqFu3btX977nUw2+ruXnz5imz2azmzp2rDhw4oCZOnKiCg4OdRkDXdMnJyWrnzp1q586dClDvvvuu2rlzpzp58qRSSpueHBwcrH7//Xe1Z88edf3117ucnty5c2e1ZcsWtWHDBtW8eXOn6ckJCQmqdu3a6vbbb1f79u1T8+bNU76+vgWmJxuNRvX222+rgwcPqpdeeqnaTE9+4IEHVFBQkFqzZo3TNMO0tDRbnfvvv181bNhQrVq1Sm3btk317t1b9e7d23Y+b5rh8OHD1a5du9TSpUtVeHi4y2mGTzzxhDp48KD65JNPXE4zrM4/E08//bRau3atio6OVnv27FFPP/200ul06s8//1RKyX2uKI6zfpSS+1xeHnvsMbVmzRoVHR2t/vrrLzV06FBVq1YtFRcXp5SqmvdZEhUXPvroI9WwYUNlMplUjx491ObNm90dkkdZvXq1Agp8jR8/XimlTVF+4YUXVO3atZXZbFZDhgxRhw4dcrrGpUuX1NixY5W/v78KDAxUd955p0pOTnaqs3v3btWvXz9lNptV/fr11YwZMwrE8uOPP6oWLVook8mk2rZtqxYtWlRh77syubq/gJozZ46tTnp6unrwwQdVSEiI8vX1VTfccIM6d+6c03VOnDihRo0apXx8fFStWrXUY489prKzs53qrF69WnXq1EmZTCbVpEkTp9fIU51/Ju666y7VqFEjZTKZVHh4uBoyZIgtSVFK7nNFyZ+oyH0uH7fccouqW7euMplMqn79+uqWW25RR48etZ2vivdZp5RSpW+HEUIIIYSoeDJGRQghhBAeSxIVIYQQQngsSVSEEEII4bEkURFCCCGEx5JERQghhBAeSxIVIYQQQngsSVSEEEII4bEkURFCCCGEx5JERQghhBAeSxIVIUSpTZgwgTFjxrg7DCFEDSCJihBCCCE8liQqQohC/fzzz7Rv3x4fHx/CwsIYOnQoTzzxBF999RW///47Op0OnU7HmjVrAIiJieHmm28mODiY0NBQrr/+ek6cOGG7Xl5LzMsvv0x4eDiBgYHcf//9ZGVllSiegQMHMmXKFJ588klCQ0OpU6cO06ZNs50/ceIEOp2OXbt22coSEhKcYlyzZg06nY5ly5bRuXNnfHx8GDx4MHFxcSxZsoTWrVsTGBjIrbfeSlpa2hXeQSHElTK6OwAhhGc6d+4cY8eO5c033+SGG24gOTmZ9evXc8cdd3Dq1CmSkpKYM2cOAKGhoWRnZzNixAh69+7N+vXrMRqNvPbaa4wcOZI9e/ZgMpkAWLlyJd7e3qxZs4YTJ05w5513EhYWxuuvv16iuL766iumTp3Kli1b2LRpExMmTKBv374MGzasVO9v2rRpfPzxx/j6+nLzzTdz8803Yzab+e6770hJSeGGG27go48+4qmnnirdjRNClK8y7bkshKj2tm/frgB14sSJAufGjx+vrr/+eqeyr7/+WrVs2VJZrVZbWWZmpvLx8VHLli2zPS80NFSlpqba6sycOVP5+/sri8VSbExXXXWV6tevn1NZ9+7d1VNPPaWUUio6OloBaufOnbbz8fHxClCrV69WSmnb0wNqxYoVtjrTp09XgDp27Jit7L777lMjRowoNiYhRMWSrh8hhEsdO3ZkyJAhtG/fnptuuokvvviC+Pj4Quvv3r2bo0ePEhAQgL+/P/7+/oSGhpKRkcGxY8ecruvr62s77t27NykpKcTExJQorg4dOjgd161bl7i4uFK+O+fr1K5dG19fX5o0aeJUVpbrCiHKl3T9CCFcMhgMLF++nI0bN/Lnn3/y0Ucf8dxzz7FlyxaX9VNSUujatSvffvttgXPh4eHlFpeXl5fTsU6nw2q1AqDXa397KaVs57Ozs4u9jk6nK/K6Qgj3kURFCFEonU5H37596du3Ly+++CKNGjVi/vz5mEwmLBaLU90uXbrwww8/EBERQWBgYKHX3L17N+np6fj4+ACwefNm/P39iYyMvOJ48xKic+fO0blzZwCngbVCiKpHun6EEC5t2bKFN954g23btnHq1Cl+/fVXLly4QOvWrYmKimLPnj0cOnSIixcvkp2dzbhx46hVqxbXX38969evJzo6mjVr1jBlyhROnz5tu25WVhZ33303Bw4cYPHixbz00ktMnjzZ1hpyJXx8fOjVqxczZszg4MGDrF27lueff/6KryuEcB9JVIQQLgUGBrJu3TquvvpqWrRowfPPP88777zDqFGjuPfee2nZsiXdunUjPDycv/76C19fX9atW0fDhg258cYbad26NXfffTcZGRlOLSxDhgyhefPmDBgwgFtuuYXrrrvOaYrxlZo9ezY5OTl07dqVRx55hNdee63cri2EqHw65diZK4QQFWjChAkkJCTw22+/uTsUIUQVIS0qQgghhPBYkqgIITzCqVOnbNOaXX2dOnXK3SEKIdxAun6EEB4hJyfHabn9/KKiojAaZaKiEDWNJCpCCCGE8FjS9SOEEEIIjyWJihBCCCE8liQqQgghhPBYkqgIIYQQwmNJoiKEEEIIjyWJihBCCCE8liQqQgghhPBY/w+A8NOuLbuvmAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "\n",
    "# Reset the index so 'step_num' becomes a column\n",
    "ex218_metrics = calculate_metrics2(prepare_experiment_data('218'))\n",
    "ex219_metrics = calculate_metrics2(prepare_experiment_data('219'))\n",
    "ex222_metrics = calculate_metrics2(prepare_experiment_data('222'))\n",
    "\n",
    "# Add an 'experiment' column to distinguish between the two datasets\n",
    "ex218_metrics['experiment'] = '218'\n",
    "ex219_metrics['experiment'] = '219'\n",
    "ex222_metrics['experiment'] = '222'\n",
    "\n",
    "# Concatenate the two DataFrames into one\n",
    "combined_metrics = pd.concat([ex218_metrics,ex219_metrics,ex222_metrics])\n",
    "\n",
    "# Now plot using seaborn\n",
    "sns.lineplot(data=combined_metrics, x='step_num', y='hit_miss_ratio', hue='experiment')\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}