LSTM predicts AMC closing prices

OpenBB_LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9b013336-5642-4091-8374-644b88c34269",
   "metadata": {},
   "source": [
    "## Long Short-Term Memory\n",
    "\n",
    "Long short-term memory or LSTM is a type of neural network that is very good at finding patterns and predicting sequential data, in particular, time series. LSTMs can remember past data, but also have the ability to forget irrelevant bits of information. In the example given in Part 1, the simple algorithm remembered the previous two closing prices, but forgot everything that happened before two days ago. An LSTM maintains a much longer memory, and decides how important each previous piece of data is in predicting the next point in the sequence.\n",
    "\n",
    "The article in Analytics Vidhya, [\"Introduction to Long Short Term Memory (LSTM)\"](https://www.analyticsvidhya.com/blog/2021/03/introduction-to-long-short-term-memory-lstm/) shows simplified diagrams of LSTMs with three input gates. One accepts new data and the other two decide how much to remember or forget of the past stream. The result is sent to the output gate. Multiple LSTMs may be strung together to improve prediction performance.\n",
    "\n",
    "OpenBB has a built-in LSTM method, but we'll use the [Keras](https://keras.io/) version for this project. We'll also need the [scikit-learn](https://scikit-learn.org/stable/) package, but this should have been installed with your [Anaconda](https://www.anaconda.com/) distribution.\n",
    "\n",
    "Import the libraries required for the project:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2253b4ee-19ba-4460-a94e-7c74b5393de5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# If your numpy version is 1.17 or higher, tensorflow throws a warning. See: https://clay-atlas.com/us/blog/2021/08/20/tensorflow-en-numpy-futurewarning-passing-type-1/\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "5ea64afe-4cb4-4cf3-b1ae-76aa0cfdb427",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import tensorflow as tf\n",
    "# Turn off warnings for tensorflow\n",
    "tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.ERROR)\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, LSTM, Dropout"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51c450a3-8b08-48cb-8c66-a1a98e3974fa",
   "metadata": {},
   "source": [
    "Load AMC stock data into a pandas dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f3d44180-bb3c-47ae-aef5-6c6b4f33ae4f",
   "metadata": {
    "tags": []
   },
   "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>Date</th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Adj Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2/11/2020 0:00</td>\n",
       "      <td>6.80</td>\n",
       "      <td>6.98</td>\n",
       "      <td>6.65</td>\n",
       "      <td>6.94</td>\n",
       "      <td>6.897683</td>\n",
       "      <td>1568500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2/12/2020 0:00</td>\n",
       "      <td>6.93</td>\n",
       "      <td>7.08</td>\n",
       "      <td>6.82</td>\n",
       "      <td>6.86</td>\n",
       "      <td>6.818171</td>\n",
       "      <td>1922100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2/13/2020 0:00</td>\n",
       "      <td>6.75</td>\n",
       "      <td>7.08</td>\n",
       "      <td>6.72</td>\n",
       "      <td>7.00</td>\n",
       "      <td>6.957317</td>\n",
       "      <td>1553400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2/14/2020 0:00</td>\n",
       "      <td>7.01</td>\n",
       "      <td>7.19</td>\n",
       "      <td>6.88</td>\n",
       "      <td>7.10</td>\n",
       "      <td>7.056707</td>\n",
       "      <td>1706000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2/18/2020 0:00</td>\n",
       "      <td>7.09</td>\n",
       "      <td>7.25</td>\n",
       "      <td>7.03</td>\n",
       "      <td>7.04</td>\n",
       "      <td>6.997073</td>\n",
       "      <td>2142800</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Date  Open  High   Low  Close  Adj Close   Volume\n",
       "0  2/11/2020 0:00  6.80  6.98  6.65   6.94   6.897683  1568500\n",
       "1  2/12/2020 0:00  6.93  7.08  6.82   6.86   6.818171  1922100\n",
       "2  2/13/2020 0:00  6.75  7.08  6.72   7.00   6.957317  1553400\n",
       "3  2/14/2020 0:00  7.01  7.19  6.88   7.10   7.056707  1706000\n",
       "4  2/18/2020 0:00  7.09  7.25  7.03   7.04   6.997073  2142800"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(r'AMC.csv')\n",
    "df = df.loc[:, ~df.columns.str.contains('^Unnamed')]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "345a5ea4-1cd9-4233-b4f4-e78964797abd",
   "metadata": {},
   "source": [
    "Normalize the data between zero and one using MinMaxScaler from sklearn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9a3e6f1f-9b48-4640-90bd-0fb8cf70dbea",
   "metadata": {
    "tags": []
   },
   "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>Date</th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Adj Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2/11/2020 0:00</td>\n",
       "      <td>6.80</td>\n",
       "      <td>6.98</td>\n",
       "      <td>6.65</td>\n",
       "      <td>0.081889</td>\n",
       "      <td>6.897683</td>\n",
       "      <td>1568500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2/12/2020 0:00</td>\n",
       "      <td>6.93</td>\n",
       "      <td>7.08</td>\n",
       "      <td>6.82</td>\n",
       "      <td>0.080568</td>\n",
       "      <td>6.818171</td>\n",
       "      <td>1922100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2/13/2020 0:00</td>\n",
       "      <td>6.75</td>\n",
       "      <td>7.08</td>\n",
       "      <td>6.72</td>\n",
       "      <td>0.082879</td>\n",
       "      <td>6.957317</td>\n",
       "      <td>1553400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2/14/2020 0:00</td>\n",
       "      <td>7.01</td>\n",
       "      <td>7.19</td>\n",
       "      <td>6.88</td>\n",
       "      <td>0.084530</td>\n",
       "      <td>7.056707</td>\n",
       "      <td>1706000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2/18/2020 0:00</td>\n",
       "      <td>7.09</td>\n",
       "      <td>7.25</td>\n",
       "      <td>7.03</td>\n",
       "      <td>0.083540</td>\n",
       "      <td>6.997073</td>\n",
       "      <td>2142800</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Date  Open  High   Low     Close  Adj Close   Volume\n",
       "0  2/11/2020 0:00  6.80  6.98  6.65  0.081889   6.897683  1568500\n",
       "1  2/12/2020 0:00  6.93  7.08  6.82  0.080568   6.818171  1922100\n",
       "2  2/13/2020 0:00  6.75  7.08  6.72  0.082879   6.957317  1553400\n",
       "3  2/14/2020 0:00  7.01  7.19  6.88  0.084530   7.056707  1706000\n",
       "4  2/18/2020 0:00  7.09  7.25  7.03  0.083540   6.997073  2142800"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler = MinMaxScaler(feature_range=(0,1))\n",
    "df['Close'] = scaler.fit_transform(df['Close'].values.reshape(-1,1))\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "545724f8-e1cd-468a-bc2c-6cea0db9e4d9",
   "metadata": {},
   "source": [
    "Write a helper function to convert closing prices into a 3D array of samples, timesteps and features in the format expected by LSTM."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6e457a62-fc7b-44f9-9c5c-815032d3c60f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def create_dataset(df, time_steps):\n",
    "    X = []\n",
    "    y = []\n",
    "    for i in range(len(df) - time_steps):\n",
    "        X.append(df['Close'].iloc[i:i+time_steps].values)\n",
    "        y.append(df['Close'].iloc[i+time_steps])\n",
    "    X = np.array(X).reshape(-1, time_steps, 1)\n",
    "    y = np.array(y)\n",
    "    return X, y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "21bbb55e-d707-43d9-9fdd-38bd5d173dd1",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "X, y = create_dataset(df,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4cea5473-d5f6-4185-b98d-a11993aa2c1c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[0.0818887254712121], [0.08056794085789047]], [[0.08056794085789047], [0.0828793139435857]], [[0.0828793139435857], [0.08453029473087502]]]\n"
     ]
    }
   ],
   "source": [
    "print(X[:3].tolist())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d931eb1e-5259-484c-af3f-7d7f04fc4908",
   "metadata": {},
   "source": [
    "Split the dataframe *df* into training (75%) and testing (25%)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "7cea1c8e-59ca-44a5-939e-bca2643dbaa2",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "ind = int(len(df)*0.75)\n",
    "train_df = df[:ind]\n",
    "test_df = df[ind:]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03818643-fd38-4e60-be86-4481cfa50a56",
   "metadata": {},
   "source": [
    "Use the *create_dataset* function to build the training and testing data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "31c1b1b1-146a-4480-b458-658428e295b3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(559, 10, 1) (559,)\n",
      "(180, 10, 1) (180,)\n"
     ]
    }
   ],
   "source": [
    "time_steps = 10 # you can change this value\n",
    "X_train, y_train = create_dataset(train_df, time_steps)\n",
    "X_test, y_test = create_dataset(test_df, time_steps)\n",
    "print(X_train.shape, y_train.shape)\n",
    "print(X_test.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24d01d09-b303-4ab4-82fd-d404891c0784",
   "metadata": {},
   "source": [
    "Make the model using Keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "7e8b6458-166f-4940-9a3e-472bb9504098",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(LSTM(50,\n",
    "               activation='tanh',\n",
    "               dropout=0.2,\n",
    "               return_sequences=False,\n",
    "               input_shape=(time_steps, 1)))\n",
    "model.add(Dense(1,\n",
    "                activation='linear',\n",
    "                use_bias=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b3498ee4-3335-4f45-9ea7-13e25d069703",
   "metadata": {},
   "source": [
    "Define a new function to calculate Root Mean Squared Error (RMSE) and update the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d53becdc-5c78-4a47-a1a0-d0a20b3d174d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def root_mean_squared_error(y_true,y_pred):\n",
    "    return tf.sqrt(tf.reduce_mean(tf.square(y_pred - y_true)))\n",
    "\n",
    "model.compile(optimizer='adam',\n",
    "              loss='mean_squared_error',\n",
    "              metrics=[root_mean_squared_error])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "454f7775-e094-48b7-863c-5ef03503c13a",
   "metadata": {},
   "source": [
    "Train the model by calling the *fit* method. Use batch sizes of 32 and 100 epochs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "1bb7a619-dfad-49ea-9589-544656f5cba2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 559 samples, validate on 180 samples\n",
      "Epoch 1/100\n",
      "559/559 [==============================] - 2s 4ms/sample - loss: 0.0498 - root_mean_squared_error: 0.2038 - val_loss: 0.0098 - val_root_mean_squared_error: 0.0987\n",
      "Epoch 2/100\n",
      "559/559 [==============================] - 0s 425us/sample - loss: 0.0246 - root_mean_squared_error: 0.1505 - val_loss: 0.0029 - val_root_mean_squared_error: 0.0542\n",
      "Epoch 3/100\n",
      "559/559 [==============================] - 0s 425us/sample - loss: 0.0283 - root_mean_squared_error: 0.1620 - val_loss: 0.0046 - val_root_mean_squared_error: 0.0673\n",
      "Epoch 4/100\n",
      "559/559 [==============================] - 0s 408us/sample - loss: 0.0204 - root_mean_squared_error: 0.1394 - val_loss: 0.0033 - val_root_mean_squared_error: 0.0574\n",
      "Epoch 5/100\n",
      "559/559 [==============================] - 0s 411us/sample - loss: 0.0187 - root_mean_squared_error: 0.1260 - val_loss: 0.0027 - val_root_mean_squared_error: 0.0521\n",
      "Epoch 6/100\n",
      "559/559 [==============================] - 0s 414us/sample - loss: 0.0206 - root_mean_squared_error: 0.1352 - val_loss: 0.0027 - val_root_mean_squared_error: 0.0516\n",
      "Epoch 7/100\n",
      "559/559 [==============================] - 0s 409us/sample - loss: 0.0215 - root_mean_squared_error: 0.1384 - val_loss: 0.0025 - val_root_mean_squared_error: 0.0497\n",
      "Epoch 8/100\n",
      "559/559 [==============================] - 0s 376us/sample - loss: 0.0275 - root_mean_squared_error: 0.1562 - val_loss: 0.0041 - val_root_mean_squared_error: 0.0641\n",
      "Epoch 9/100\n",
      "559/559 [==============================] - 0s 393us/sample - loss: 0.0253 - root_mean_squared_error: 0.1482 - val_loss: 0.0038 - val_root_mean_squared_error: 0.0606\n",
      "Epoch 10/100\n",
      "559/559 [==============================] - 0s 376us/sample - loss: 0.0209 - root_mean_squared_error: 0.1397 - val_loss: 0.0035 - val_root_mean_squared_error: 0.0592\n",
      "Epoch 11/100\n",
      "559/559 [==============================] - 0s 340us/sample - loss: 0.0198 - root_mean_squared_error: 0.1365 - val_loss: 0.0031 - val_root_mean_squared_error: 0.0557\n",
      "Epoch 12/100\n",
      "559/559 [==============================] - 0s 365us/sample - loss: 0.0266 - root_mean_squared_error: 0.1545 - val_loss: 0.0041 - val_root_mean_squared_error: 0.0641\n",
      "Epoch 13/100\n",
      "559/559 [==============================] - 0s 297us/sample - loss: 0.0222 - root_mean_squared_error: 0.1445 - val_loss: 0.0039 - val_root_mean_squared_error: 0.0619\n",
      "Epoch 14/100\n",
      "559/559 [==============================] - 0s 270us/sample - loss: 0.0146 - root_mean_squared_error: 0.1133 - val_loss: 0.0021 - val_root_mean_squared_error: 0.0454\n",
      "Epoch 15/100\n",
      "559/559 [==============================] - 0s 312us/sample - loss: 0.0203 - root_mean_squared_error: 0.1359 - val_loss: 0.0032 - val_root_mean_squared_error: 0.0569\n",
      "Epoch 16/100\n",
      "559/559 [==============================] - 0s 294us/sample - loss: 0.0219 - root_mean_squared_error: 0.1428 - val_loss: 0.0040 - val_root_mean_squared_error: 0.0629\n",
      "Epoch 17/100\n",
      "559/559 [==============================] - 0s 287us/sample - loss: 0.0281 - root_mean_squared_error: 0.1584 - val_loss: 0.0042 - val_root_mean_squared_error: 0.0648\n",
      "Epoch 18/100\n",
      "559/559 [==============================] - 0s 275us/sample - loss: 0.0221 - root_mean_squared_error: 0.1397 - val_loss: 0.0026 - val_root_mean_squared_error: 0.0498\n",
      "Epoch 19/100\n",
      "559/559 [==============================] - 0s 262us/sample - loss: 0.0191 - root_mean_squared_error: 0.1339 - val_loss: 0.0017 - val_root_mean_squared_error: 0.0410\n",
      "Epoch 20/100\n",
      "559/559 [==============================] - 0s 269us/sample - loss: 0.0160 - root_mean_squared_error: 0.1184 - val_loss: 0.0038 - val_root_mean_squared_error: 0.0617\n",
      "Epoch 21/100\n",
      "559/559 [==============================] - 0s 274us/sample - loss: 0.0164 - root_mean_squared_error: 0.1225 - val_loss: 0.0015 - val_root_mean_squared_error: 0.0382\n",
      "Epoch 22/100\n",
      "559/559 [==============================] - 0s 271us/sample - loss: 0.0114 - root_mean_squared_error: 0.1030 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0327\n",
      "Epoch 23/100\n",
      "559/559 [==============================] - 0s 287us/sample - loss: 0.0079 - root_mean_squared_error: 0.0855 - val_loss: 9.0537e-04 - val_root_mean_squared_error: 0.0298\n",
      "Epoch 24/100\n",
      "559/559 [==============================] - 0s 264us/sample - loss: 0.0068 - root_mean_squared_error: 0.0796 - val_loss: 0.0020 - val_root_mean_squared_error: 0.0450\n",
      "Epoch 25/100\n",
      "559/559 [==============================] - 0s 286us/sample - loss: 0.0053 - root_mean_squared_error: 0.0694 - val_loss: 0.0010 - val_root_mean_squared_error: 0.0321\n",
      "Epoch 26/100\n",
      "559/559 [==============================] - 0s 270us/sample - loss: 0.0066 - root_mean_squared_error: 0.0778 - val_loss: 0.0013 - val_root_mean_squared_error: 0.0362\n",
      "Epoch 27/100\n",
      "559/559 [==============================] - 0s 266us/sample - loss: 0.0053 - root_mean_squared_error: 0.0678 - val_loss: 7.8821e-04 - val_root_mean_squared_error: 0.0274\n",
      "Epoch 28/100\n",
      "559/559 [==============================] - 0s 273us/sample - loss: 0.0054 - root_mean_squared_error: 0.0685 - val_loss: 7.8856e-04 - val_root_mean_squared_error: 0.0276\n",
      "Epoch 29/100\n",
      "559/559 [==============================] - 0s 250us/sample - loss: 0.0047 - root_mean_squared_error: 0.0663 - val_loss: 8.9104e-04 - val_root_mean_squared_error: 0.0293\n",
      "Epoch 30/100\n",
      "559/559 [==============================] - 0s 269us/sample - loss: 0.0041 - root_mean_squared_error: 0.0607 - val_loss: 6.9811e-04 - val_root_mean_squared_error: 0.0262\n",
      "Epoch 31/100\n",
      "559/559 [==============================] - 0s 250us/sample - loss: 0.0051 - root_mean_squared_error: 0.0662 - val_loss: 8.1491e-04 - val_root_mean_squared_error: 0.0275\n",
      "Epoch 32/100\n",
      "559/559 [==============================] - 0s 279us/sample - loss: 0.0065 - root_mean_squared_error: 0.0724 - val_loss: 6.1105e-04 - val_root_mean_squared_error: 0.0241\n",
      "Epoch 33/100\n",
      "559/559 [==============================] - 0s 276us/sample - loss: 0.0044 - root_mean_squared_error: 0.0623 - val_loss: 6.5641e-04 - val_root_mean_squared_error: 0.0255\n",
      "Epoch 34/100\n",
      "559/559 [==============================] - 0s 277us/sample - loss: 0.0047 - root_mean_squared_error: 0.0612 - val_loss: 6.2135e-04 - val_root_mean_squared_error: 0.0246\n",
      "Epoch 35/100\n",
      "559/559 [==============================] - 0s 276us/sample - loss: 0.0048 - root_mean_squared_error: 0.0639 - val_loss: 8.2745e-04 - val_root_mean_squared_error: 0.0290\n",
      "Epoch 36/100\n",
      "559/559 [==============================] - 0s 282us/sample - loss: 0.0047 - root_mean_squared_error: 0.0652 - val_loss: 0.0012 - val_root_mean_squared_error: 0.0337\n",
      "Epoch 37/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0048 - root_mean_squared_error: 0.0626 - val_loss: 8.5799e-04 - val_root_mean_squared_error: 0.0285\n",
      "Epoch 38/100\n",
      "559/559 [==============================] - 0s 286us/sample - loss: 0.0035 - root_mean_squared_error: 0.0565 - val_loss: 9.5692e-04 - val_root_mean_squared_error: 0.0302\n",
      "Epoch 39/100\n",
      "559/559 [==============================] - 0s 288us/sample - loss: 0.0056 - root_mean_squared_error: 0.0688 - val_loss: 7.5915e-04 - val_root_mean_squared_error: 0.0278\n",
      "Epoch 40/100\n",
      "559/559 [==============================] - 0s 292us/sample - loss: 0.0050 - root_mean_squared_error: 0.0655 - val_loss: 0.0012 - val_root_mean_squared_error: 0.0343\n",
      "Epoch 41/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0031 - root_mean_squared_error: 0.0512 - val_loss: 0.0014 - val_root_mean_squared_error: 0.0365\n",
      "Epoch 42/100\n",
      "559/559 [==============================] - 0s 263us/sample - loss: 0.0049 - root_mean_squared_error: 0.0663 - val_loss: 7.4398e-04 - val_root_mean_squared_error: 0.0265\n",
      "Epoch 43/100\n",
      "559/559 [==============================] - 0s 305us/sample - loss: 0.0051 - root_mean_squared_error: 0.0657 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0329\n",
      "Epoch 44/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0047 - root_mean_squared_error: 0.0640 - val_loss: 9.7720e-04 - val_root_mean_squared_error: 0.0310\n",
      "Epoch 45/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0052 - root_mean_squared_error: 0.0671 - val_loss: 6.7318e-04 - val_root_mean_squared_error: 0.0252\n",
      "Epoch 46/100\n",
      "559/559 [==============================] - 0s 289us/sample - loss: 0.0048 - root_mean_squared_error: 0.0658 - val_loss: 6.7494e-04 - val_root_mean_squared_error: 0.0260\n",
      "Epoch 47/100\n",
      "559/559 [==============================] - 0s 283us/sample - loss: 0.0047 - root_mean_squared_error: 0.0671 - val_loss: 8.1588e-04 - val_root_mean_squared_error: 0.0282\n",
      "Epoch 48/100\n",
      "559/559 [==============================] - 0s 287us/sample - loss: 0.0048 - root_mean_squared_error: 0.0656 - val_loss: 7.5247e-04 - val_root_mean_squared_error: 0.0275\n",
      "Epoch 49/100\n",
      "559/559 [==============================] - 0s 280us/sample - loss: 0.0045 - root_mean_squared_error: 0.0643 - val_loss: 0.0010 - val_root_mean_squared_error: 0.0318\n",
      "Epoch 50/100\n",
      "559/559 [==============================] - 0s 274us/sample - loss: 0.0035 - root_mean_squared_error: 0.0538 - val_loss: 7.6147e-04 - val_root_mean_squared_error: 0.0272\n",
      "Epoch 51/100\n",
      "559/559 [==============================] - 0s 259us/sample - loss: 0.0048 - root_mean_squared_error: 0.0623 - val_loss: 0.0013 - val_root_mean_squared_error: 0.0346\n",
      "Epoch 52/100\n",
      "559/559 [==============================] - 0s 269us/sample - loss: 0.0049 - root_mean_squared_error: 0.0643 - val_loss: 7.7229e-04 - val_root_mean_squared_error: 0.0265\n",
      "Epoch 53/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0036 - root_mean_squared_error: 0.0554 - val_loss: 7.7376e-04 - val_root_mean_squared_error: 0.0272\n",
      "Epoch 54/100\n",
      "559/559 [==============================] - 0s 258us/sample - loss: 0.0053 - root_mean_squared_error: 0.0684 - val_loss: 6.5688e-04 - val_root_mean_squared_error: 0.0240\n",
      "Epoch 55/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0032 - root_mean_squared_error: 0.0555 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0331\n",
      "Epoch 56/100\n",
      "559/559 [==============================] - 0s 272us/sample - loss: 0.0034 - root_mean_squared_error: 0.0537 - val_loss: 5.9527e-04 - val_root_mean_squared_error: 0.0235\n",
      "Epoch 57/100\n",
      "559/559 [==============================] - 0s 265us/sample - loss: 0.0050 - root_mean_squared_error: 0.0651 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0328\n",
      "Epoch 58/100\n",
      "559/559 [==============================] - 0s 256us/sample - loss: 0.0039 - root_mean_squared_error: 0.0573 - val_loss: 7.5873e-04 - val_root_mean_squared_error: 0.0266\n",
      "Epoch 59/100\n",
      "559/559 [==============================] - 0s 262us/sample - loss: 0.0038 - root_mean_squared_error: 0.0583 - val_loss: 5.4792e-04 - val_root_mean_squared_error: 0.0233\n",
      "Epoch 60/100\n",
      "559/559 [==============================] - 0s 266us/sample - loss: 0.0038 - root_mean_squared_error: 0.0580 - val_loss: 7.7730e-04 - val_root_mean_squared_error: 0.0278\n",
      "Epoch 61/100\n",
      "559/559 [==============================] - 0s 275us/sample - loss: 0.0045 - root_mean_squared_error: 0.0618 - val_loss: 9.3643e-04 - val_root_mean_squared_error: 0.0303\n",
      "Epoch 62/100\n",
      "559/559 [==============================] - 0s 262us/sample - loss: 0.0045 - root_mean_squared_error: 0.0617 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0323\n",
      "Epoch 63/100\n",
      "559/559 [==============================] - 0s 304us/sample - loss: 0.0035 - root_mean_squared_error: 0.0551 - val_loss: 0.0012 - val_root_mean_squared_error: 0.0339\n",
      "Epoch 64/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0047 - root_mean_squared_error: 0.0657 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0326\n",
      "Epoch 65/100\n",
      "559/559 [==============================] - 0s 274us/sample - loss: 0.0047 - root_mean_squared_error: 0.0625 - val_loss: 7.3396e-04 - val_root_mean_squared_error: 0.0264\n",
      "Epoch 66/100\n",
      "559/559 [==============================] - 0s 262us/sample - loss: 0.0045 - root_mean_squared_error: 0.0617 - val_loss: 8.5246e-04 - val_root_mean_squared_error: 0.0282\n",
      "Epoch 67/100\n",
      "559/559 [==============================] - 0s 274us/sample - loss: 0.0044 - root_mean_squared_error: 0.0603 - val_loss: 8.1735e-04 - val_root_mean_squared_error: 0.0287\n",
      "Epoch 68/100\n",
      "559/559 [==============================] - 0s 263us/sample - loss: 0.0046 - root_mean_squared_error: 0.0587 - val_loss: 0.0016 - val_root_mean_squared_error: 0.0394\n",
      "Epoch 69/100\n",
      "559/559 [==============================] - 0s 279us/sample - loss: 0.0036 - root_mean_squared_error: 0.0570 - val_loss: 5.9725e-04 - val_root_mean_squared_error: 0.0232\n",
      "Epoch 70/100\n",
      "559/559 [==============================] - 0s 258us/sample - loss: 0.0030 - root_mean_squared_error: 0.0511 - val_loss: 5.7769e-04 - val_root_mean_squared_error: 0.0236\n",
      "Epoch 71/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0034 - root_mean_squared_error: 0.0527 - val_loss: 5.8853e-04 - val_root_mean_squared_error: 0.0230\n",
      "Epoch 72/100\n",
      "559/559 [==============================] - 0s 260us/sample - loss: 0.0054 - root_mean_squared_error: 0.0681 - val_loss: 7.9495e-04 - val_root_mean_squared_error: 0.0280\n",
      "Epoch 73/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0038 - root_mean_squared_error: 0.0566 - val_loss: 6.1188e-04 - val_root_mean_squared_error: 0.0240\n",
      "Epoch 74/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0051 - root_mean_squared_error: 0.0644 - val_loss: 6.6247e-04 - val_root_mean_squared_error: 0.0251\n",
      "Epoch 75/100\n",
      "559/559 [==============================] - 0s 269us/sample - loss: 0.0031 - root_mean_squared_error: 0.0520 - val_loss: 5.8347e-04 - val_root_mean_squared_error: 0.0244\n",
      "Epoch 76/100\n",
      "559/559 [==============================] - 0s 286us/sample - loss: 0.0052 - root_mean_squared_error: 0.0674 - val_loss: 0.0010 - val_root_mean_squared_error: 0.0321\n",
      "Epoch 77/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0040 - root_mean_squared_error: 0.0580 - val_loss: 4.9394e-04 - val_root_mean_squared_error: 0.0221\n",
      "Epoch 78/100\n",
      "559/559 [==============================] - 0s 273us/sample - loss: 0.0037 - root_mean_squared_error: 0.0568 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0328\n",
      "Epoch 79/100\n",
      "559/559 [==============================] - 0s 264us/sample - loss: 0.0036 - root_mean_squared_error: 0.0554 - val_loss: 6.3129e-04 - val_root_mean_squared_error: 0.0247\n",
      "Epoch 80/100\n",
      "559/559 [==============================] - 0s 275us/sample - loss: 0.0035 - root_mean_squared_error: 0.0518 - val_loss: 6.7102e-04 - val_root_mean_squared_error: 0.0254\n",
      "Epoch 81/100\n",
      "559/559 [==============================] - 0s 262us/sample - loss: 0.0045 - root_mean_squared_error: 0.0619 - val_loss: 0.0010 - val_root_mean_squared_error: 0.0320\n",
      "Epoch 82/100\n",
      "559/559 [==============================] - 0s 271us/sample - loss: 0.0046 - root_mean_squared_error: 0.0619 - val_loss: 8.2529e-04 - val_root_mean_squared_error: 0.0290\n",
      "Epoch 83/100\n",
      "559/559 [==============================] - 0s 250us/sample - loss: 0.0046 - root_mean_squared_error: 0.0634 - val_loss: 6.1861e-04 - val_root_mean_squared_error: 0.0246\n",
      "Epoch 84/100\n",
      "559/559 [==============================] - 0s 273us/sample - loss: 0.0040 - root_mean_squared_error: 0.0604 - val_loss: 5.3419e-04 - val_root_mean_squared_error: 0.0223\n",
      "Epoch 85/100\n",
      "559/559 [==============================] - 0s 263us/sample - loss: 0.0031 - root_mean_squared_error: 0.0518 - val_loss: 0.0011 - val_root_mean_squared_error: 0.0316\n",
      "Epoch 86/100\n",
      "559/559 [==============================] - 0s 287us/sample - loss: 0.0048 - root_mean_squared_error: 0.0611 - val_loss: 5.9033e-04 - val_root_mean_squared_error: 0.0245\n",
      "Epoch 87/100\n",
      "559/559 [==============================] - 0s 271us/sample - loss: 0.0038 - root_mean_squared_error: 0.0573 - val_loss: 6.4874e-04 - val_root_mean_squared_error: 0.0251\n",
      "Epoch 88/100\n",
      "559/559 [==============================] - 0s 284us/sample - loss: 0.0049 - root_mean_squared_error: 0.0622 - val_loss: 4.9976e-04 - val_root_mean_squared_error: 0.0219\n",
      "Epoch 89/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0037 - root_mean_squared_error: 0.0562 - val_loss: 5.2337e-04 - val_root_mean_squared_error: 0.0229\n",
      "Epoch 90/100\n",
      "559/559 [==============================] - 0s 286us/sample - loss: 0.0028 - root_mean_squared_error: 0.0499 - val_loss: 8.2758e-04 - val_root_mean_squared_error: 0.0279\n",
      "Epoch 91/100\n",
      "559/559 [==============================] - 0s 268us/sample - loss: 0.0039 - root_mean_squared_error: 0.0581 - val_loss: 4.8593e-04 - val_root_mean_squared_error: 0.0219\n",
      "Epoch 92/100\n",
      "559/559 [==============================] - 0s 266us/sample - loss: 0.0032 - root_mean_squared_error: 0.0519 - val_loss: 4.1142e-04 - val_root_mean_squared_error: 0.0191\n",
      "Epoch 93/100\n",
      "559/559 [==============================] - 0s 269us/sample - loss: 0.0041 - root_mean_squared_error: 0.0588 - val_loss: 7.3350e-04 - val_root_mean_squared_error: 0.0265\n",
      "Epoch 94/100\n",
      "559/559 [==============================] - 0s 265us/sample - loss: 0.0035 - root_mean_squared_error: 0.0524 - val_loss: 4.2536e-04 - val_root_mean_squared_error: 0.0199\n",
      "Epoch 95/100\n",
      "559/559 [==============================] - 0s 275us/sample - loss: 0.0044 - root_mean_squared_error: 0.0608 - val_loss: 7.9969e-04 - val_root_mean_squared_error: 0.0278\n",
      "Epoch 96/100\n",
      "559/559 [==============================] - 0s 262us/sample - loss: 0.0059 - root_mean_squared_error: 0.0668 - val_loss: 6.8287e-04 - val_root_mean_squared_error: 0.0253\n",
      "Epoch 97/100\n",
      "559/559 [==============================] - 0s 261us/sample - loss: 0.0027 - root_mean_squared_error: 0.0490 - val_loss: 4.0627e-04 - val_root_mean_squared_error: 0.0189\n",
      "Epoch 98/100\n",
      "559/559 [==============================] - 0s 250us/sample - loss: 0.0035 - root_mean_squared_error: 0.0555 - val_loss: 4.5173e-04 - val_root_mean_squared_error: 0.0206\n",
      "Epoch 99/100\n",
      "559/559 [==============================] - 0s 269us/sample - loss: 0.0034 - root_mean_squared_error: 0.0545 - val_loss: 5.5961e-04 - val_root_mean_squared_error: 0.0236\n",
      "Epoch 100/100\n",
      "559/559 [==============================] - 0s 286us/sample - loss: 0.0037 - root_mean_squared_error: 0.0571 - val_loss: 4.3464e-04 - val_root_mean_squared_error: 0.0198\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train,\n",
    "                    y_train,\n",
    "                    batch_size=32,\n",
    "                    epochs=100,\n",
    "                    validation_data=(X_test, y_test),\n",
    "                    verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1d721e3-a878-46ad-bce8-1a6852226d78",
   "metadata": {},
   "source": [
    "Evaluate the model with the test data. Print the test loss and RMSE."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "2c927a9e-f4ed-4dfe-abfc-58646ef3f7bd",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss: 0.0004346434033424076\n",
      "Test RMSE: 0.016953053\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_rmse = model.evaluate(X_test,\n",
    "                                      y_test,\n",
    "                                      batch_size=32,\n",
    "                                      verbose=0)\n",
    "print('Test loss:', test_loss)\n",
    "print('Test RMSE:', test_rmse)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "571e1dbf-4b7e-4d95-894e-e1ffcb69bf94",
   "metadata": {},
   "source": [
    "Plot the training and validation loss curves."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "877e60ee-918c-4a71-9454-82821496245f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAHFCAYAAAAaD0bAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAACcfklEQVR4nOzdeViUVfvA8e8M27CDIJsCghvuuG+ZVoqplZmVreZWr69tZv0qs93KtzIzS7NFs9WsTNvUwC1Nccl9y11RARGQfZthnt8fDzMwMiDgMONyf66LS3nmmWfOHAbmnvvc5xyNoigKQgghhBDXEK2jGyCEEEIIYW8SAAkhhBDimiMBkBBCCCGuORIACSGEEOKaIwGQEEIIIa45EgAJIYQQ4pojAZAQQgghrjkSAAkhhBDimiMBkBBCCCGuORIAiSuGRqOp0dfatWsv6XFeffVVNBpNne67du1am7Thcjdq1CiaNGlS5e3nzp3D1dWVe+65p8pzcnJy8PDw4Lbbbqvx4y5YsACNRsOJEydq3JaKNBoNr776ao0fzyQ5OZlXX32VnTt3VrrtUl4vl6pJkybccsstDnns2srJyeHNN9+kS5cu+Pj44ObmRpMmTRgzZgzbt293dPPENcjZ0Q0QoqYSExMtvp86dSpr1qxh9erVFsdbt259SY8zbtw4br755jrdt1OnTiQmJl5yG650DRs25LbbbmPp0qWcP38ef3//Sud8//33FBYWMnbs2Et6rJdeeoknn3zykq5xMcnJybz22ms0adKE2NhYi9su5fVyrTh69ChxcXGkpaUxfvx4XnvtNby8vDhx4gQ//PADnTt3JisrC19fX0c3VVxDJAASV4wePXpYfN+wYUO0Wm2l4xcqKCjAw8Ojxo/TuHFjGjduXKc2+vj4XLQ914qxY8eyePFivv32Wx577LFKt8+fP5/g4GCGDBlySY/TtGnTS7r/pbqU18u1oLS0lGHDhpGenk5iYiJt27Y139a3b18eeughli9fjouLyyU/lqIoFBUV4e7ufsnXElc/GQITV5V+/frRtm1b1q1bR69evfDw8GDMmDEALFq0iLi4OEJDQ3F3d6dVq1Y8//zz5OfnW1zD2pCGaahhxYoVdOrUCXd3d2JiYpg/f77FedaGwEaNGoWXlxdHjhxh8ODBeHl5ER4eztNPP01xcbHF/U+fPs2dd96Jt7c3fn5+3H///WzduhWNRsOCBQuqfe7nzp1jwoQJtG7dGi8vL4KCgrjxxhtZv369xXknTpxAo9Ewffp0ZsyYQVRUFF5eXvTs2ZNNmzZVuu6CBQto2bIlbm5utGrViq+++qradpgMHDiQxo0b88UXX1S67cCBA2zevJmRI0fi7OxMQkICQ4cOpXHjxuh0Opo1a8Z//vMf0tPTL/o41obAcnJyePjhhwkICMDLy4ubb76ZQ4cOVbrvkSNHGD16NM2bN8fDw4NGjRpx6623smfPHvM5a9eupWvXrgCMHj3aPNRqGkqz9noxGo288847xMTE4ObmRlBQECNHjuT06dMW55ler1u3bqVPnz54eHgQHR3N//73P4xG40Wfe00UFRUxefJkoqKicHV1pVGjRjz66KNkZWVZnLd69Wr69etHQEAA7u7uREREMHz4cAoKCsznfPzxx3To0AEvLy+8vb2JiYnhhRdeqPbxly5dyp49e5g8ebJF8FPRoEGDzB9SqhrStNbPGo2Gxx57jLlz59KqVSvc3Nz4/PPPCQoK4sEHH6x0jaysLNzd3Zk0aZL5WE5ODs8884xF/0ycOLHS34Uff/yR7t274+vra/45mf62iCuTZIDEVSclJYUHHniAZ599lrfeegutVo3zDx8+zODBg5k4cSKenp78+++/vP3222zZsqXSMJo1u3bt4umnn+b5558nODiYzz//nLFjx9KsWTOuv/76au+r1+u57bbbGDt2LE8//TTr1q1j6tSp+Pr68vLLLwOQn5/PDTfcQGZmJm+//TbNmjVjxYoVjBgxokbPOzMzE4BXXnmFkJAQ8vLyWLJkCf369WPVqlX069fP4vzZs2cTExPDzJkzAXUoafDgwRw/ftw8FLFgwQJGjx7N0KFDee+998jOzubVV1+luLjY3K9V0Wq1jBo1ijfeeINdu3bRoUMH822moMj0BnL06FF69uzJuHHj8PX15cSJE8yYMYPrrruOPXv21Co7oCgKt99+Oxs3buTll1+ma9eubNiwgUGDBlU6Nzk5mYCAAP73v//RsGFDMjMz+fLLL+nevTs7duygZcuWdOrUiS+++ILRo0fz4osvmjNW1WV9/vvf//Lpp5/y2GOPccstt3DixAleeukl1q5dy/bt2wkMDDSfm5qayv3338/TTz/NK6+8wpIlS5g8eTJhYWGMHDmyxs+7ur5YtWoVkydPpk+fPuzevZtXXnmFxMREEhMTcXNz48SJEwwZMoQ+ffowf/58/Pz8OHPmDCtWrKCkpAQPDw++//57JkyYwOOPP8706dPRarUcOXKE/fv3V9uG+Ph4AG6//fZLei5VWbp0KevXr+fll18mJCSEoKAgjh8/zty5c5k9ezY+Pj7mcxcuXEhRURGjR48G1Oxw3759OX36NC+88ALt27dn3759vPzyy+zZs4eVK1ei0WhITExkxIgRjBgxgldffRWdTsfJkydr9HdDXMYUIa5QDz30kOLp6WlxrG/fvgqgrFq1qtr7Go1GRa/XK3/99ZcCKLt27TLf9sorrygX/mpERkYqOp1OOXnypPlYYWGh0qBBA+U///mP+diaNWsUQFmzZo1FOwHlhx9+sLjm4MGDlZYtW5q/nz17tgIoy5cvtzjvP//5jwIoX3zxRbXP6UIGg0HR6/XKTTfdpAwbNsx8/Pjx4wqgtGvXTjEYDObjW7ZsUQBl4cKFiqIoSmlpqRIWFqZ06tRJMRqN5vNOnDihuLi4KJGRkRdtw7FjxxSNRqM88cQT5mN6vV4JCQlRevfubfU+pp/NyZMnFUD55ZdfzLd98cUXCqAcP37cfOyhhx6yaMvy5csVQPnggw8srvvmm28qgPLKK69U2V6DwaCUlJQozZs3V5566inz8a1bt1b5M7jw9XLgwAEFUCZMmGBx3ubNmxVAeeGFF8zHTK/XzZs3W5zbunVrZeDAgVW20yQyMlIZMmRIlbevWLFCAZR33nnH4viiRYsUQPn0008VRVGUn376SQGUnTt3Vnmtxx57TPHz87tomy508803K4BSVFRUo/Mv/HmaWPu9BBRfX18lMzPT4vju3bstnp9Jt27dlM6dO5u/nzZtmqLVapWtW7danGfqj2XLlimKoijTp09XACUrK6tGz0FcGWQITFx1/P39ufHGGysdP3bsGPfddx8hISE4OTnh4uJC3759AXVI5mJiY2OJiIgwf6/T6WjRogUnT5686H01Gg233nqrxbH27dtb3Pevv/7C29u7UkHtvffee9Hrm8ydO5dOnTqh0+lwdnbGxcWFVatWWX1+Q4YMwcnJyaI9gLlNBw8eJDk5mfvuu89i6CEyMpJevXrVqD1RUVHccMMNfPvtt5SUlACwfPlyUlNTLYYPTMWx4eHh5nZHRkYCNfvZVLRmzRoA7r//fovj9913X6VzDQYDb731Fq1bt8bV1RVnZ2dcXV05fPhwrR/3wscfNWqUxfFu3brRqlUrVq1aZXE8JCSEbt26WRy78LVRV6YMxYVtueuuu/D09DS3JTY2FldXVx555BG+/PJLjh07Vula3bp1Iysri3vvvZdffvmlRsOT9nDjjTdWKrJv164dnTt3thh+PXDgAFu2bLF43f3++++0bduW2NhYDAaD+WvgwIEWQ9mmIdC7776bH374gTNnztT/ExP1TgIgcdUJDQ2tdCwvL48+ffqwefNm3njjDdauXcvWrVv5+eefASgsLLzodQMCAiodc3Nzq9F9PTw80Ol0le5bVFRk/j4jI4Pg4OBK97V2zJoZM2bw3//+l+7du7N48WI2bdrE1q1bufnmm6228cLn4+bmBpT3RUZGBqC+QV/I2rGqjB07loyMDH799VdAHf7y8vLi7rvvBtR6mbi4OH7++WeeffZZVq1axZYtW8z1SDXp34oyMjJwdnau9PystXnSpEm89NJL3H777fz2229s3ryZrVu30qFDh1o/bsXHB+uvw7CwMPPtJpfyuqpJW5ydnWnYsKHFcY1GQ0hIiLktTZs2ZeXKlQQFBfHoo4/StGlTmjZtygcffGC+z4MPPsj8+fM5efIkw4cPJygoiO7du5OQkFBtG0wfGo4fP37Jz8caa/0M6vBqYmIi//77L6C+7tzc3Cw+UJw9e5bdu3fj4uJi8eXt7Y2iKOYg7/rrr2fp0qUYDAZGjhxJ48aNadu2LQsXLqyX5yTsQ2qAxFXH2posq1evJjk5mbVr15qzPkClQlBHCggIYMuWLZWOp6am1uj+33zzDf369ePjjz+2OJ6bm1vn9lT1+DVtE8Add9yBv78/8+fPp2/fvvz++++MHDkSLy8vAPbu3cuuXbtYsGABDz30kPl+R44cqXO7DQYDGRkZFsGFtTZ/8803jBw5krfeesvieHp6On5+fnV+fFBr0S6sE0pOTrao/6lvpr44d+6cRRCkKAqpqanmzAZAnz596NOnD6Wlpfzzzz98+OGHTJw4keDgYPN6TqNHj2b06NHk5+ezbt06XnnlFW655RYOHTpkzthdaODAgXz66acsXbqU559//qJt1ul0lSYHAFVmnKpag+nee+9l0qRJLFiwgDfffJOvv/6a22+/3SJbFBgYiLu7e6XJDBVvNxk6dChDhw6luLiYTZs2MW3aNO677z6aNGlCz549L/q8xOVHMkDimmD6I2nKcph88sknjmiOVX379iU3N5fly5dbHP/+++9rdH+NRlPp+e3evbvS+kk11bJlS0JDQ1m4cCGKopiPnzx5ko0bN9b4Ojqdjvvuu4/4+Hjefvtt9Hq9xTCErX82N9xwAwDffvutxfHvvvuu0rnW+uyPP/6oNMRxYXasOqbh12+++cbi+NatWzlw4AA33XTTRa9hK6bHurAtixcvJj8/32pbnJyc6N69O7Nnzwawukihp6cngwYNYsqUKZSUlLBv374q2zB06FDatWvHtGnT2Lt3r9Vz/vzzT/NssyZNmpCWlsbZs2fNt5eUlPDnn39e5Nla8vf35/bbb+err77i999/rzTsCnDLLbdw9OhRAgIC6NKlS6Uva7PR3Nzc6Nu3L2+//TYAO3bsqFW7xOVDMkDimtCrVy/8/f0ZP348r7zyCi4uLnz77bfs2rXL0U0ze+ihh3j//fd54IEHeOONN2jWrBnLly83/+G/2KyrW265halTp/LKK6/Qt29fDh48yOuvv05UVBQGg6HW7dFqtUydOpVx48YxbNgwHn74YbKysnj11VdrNQQG6jDY7NmzmTFjBjExMRY1RDExMTRt2pTnn38eRVFo0KABv/3220WHVqoSFxfH9ddfz7PPPkt+fj5dunRhw4YNfP3115XOveWWW1iwYAExMTG0b9+ebdu28e6771bK3DRt2hR3d3e+/fZbWrVqhZeXF2FhYYSFhVW6ZsuWLXnkkUf48MMP0Wq1DBo0yDwLLDw8nKeeeqpOz6sqqamp/PTTT5WON2nShAEDBjBw4ECee+45cnJy6N27t3kWWMeOHc1TxefOncvq1asZMmQIERERFBUVmbMi/fv3B+Dhhx/G3d2d3r17ExoaSmpqKtOmTcPX19cik3QhJycnlixZQlxcHD179uS///0vN9xwA56enpw8eZKffvqJ3377jfPnzwMwYsQIXn75Ze655x7+7//+j6KiImbNmkVpaWmt+2bMmDEsWrSIxx57jMaNG5ufi8nEiRNZvHgx119/PU899RTt27fHaDSSlJREfHw8Tz/9NN27d+fll1/m9OnT3HTTTTRu3JisrCw++OADizpCcQVybA22EHVX1SywNm3aWD1/48aNSs+ePRUPDw+lYcOGyrhx45Tt27dXmt1T1Swwa7Nt+vbtq/Tt29f8fVWzwC5sZ1WPk5SUpNxxxx2Kl5eX4u3trQwfPlxZtmxZpdlQ1hQXFyvPPPOM0qhRI0Wn0ymdOnVSli5dWmlWjWkW2LvvvlvpGliZJfX5558rzZs3V1xdXZUWLVoo8+fPr3KmTnU6duxodUaSoijK/v37lQEDBije3t6Kv7+/ctdddylJSUmV2lOTWWCKoihZWVnKmDFjFD8/P8XDw0MZMGCA8u+//1a63vnz55WxY8cqQUFBioeHh3Ldddcp69evr/RzVRRFWbhwoRITE6O4uLhYXMfaz7G0tFR5++23lRYtWiguLi5KYGCg8sADDyinTp2yOK+q12tN+zcyMlIBrH499NBDiqKosxWfe+45JTIyUnFxcVFCQ0OV//73v8r58+fN10lMTFSGDRumREZGKm5ubkpAQIDSt29f5ddffzWf8+WXXyo33HCDEhwcrLi6uiphYWHK3Xffrezevfui7VQU9WcydepUpVOnToqXl5fi4uKiREREKA888ICyYcMGi3OXLVumxMbGKu7u7kp0dLTy0UcfVTkL7NFHH63yMUtLS5Xw8HAFUKZMmWL1nLy8POXFF19UWrZsqbi6uiq+vr5Ku3btlKeeekpJTU1VFEVRfv/9d2XQoEFKo0aNFFdXVyUoKEgZPHiwsn79+ho9d3F50ihKhdy2EOKy89Zbb/Hiiy+SlJQkKw4LIYSNyBCYEJeRjz76CFCHhfR6PatXr2bWrFk88MADEvwIIYQNSQAkxGXEw8OD999/nxMnTlBcXExERATPPfccL774oqObJoQQVxUZAhNCCCHENUemwQshhBDimiMBkBBCCCGuORIACSGEEOKaI0XQVhiNRpKTk/H29q5ymXUhhBBCXF4URSE3N5ewsLCLLh4rAZAVycnJhIeHO7oZQgghhKiDU6dOXXTpEAmArPD29gbUDvTx8bHptfV6PfHx8cTFxeHi4mLTawtL0tf2I31tP9LX9iN9bT+26uucnBzCw8PN7+PVkQDICtOwl4+PT70EQB4eHvj4+MgvVD2TvrYf6Wv7kb62H+lr+7F1X9ekfEWKoIUQQghxzZEASAghhBDXHAmAhBBCCHHNkRogIYQQ9a60tBS9Xu/oZtSKXq/H2dmZoqIiSktLHd2cq1pt+trV1fWiU9xrQgIgIYQQ9UZRFFJTU8nKynJ0U2pNURRCQkI4deqUrAlXz2rT11qtlqioKFxdXS/pMSUAEkIIUW9MwU9QUBAeHh5XVCBhNBrJy8vDy8vLJhkHUbWa9rVpoeKUlBQiIiIu6fUkAZAQQoh6UVpaag5+AgICHN2cWjMajZSUlKDT6SQAqme16euGDRuSnJyMwWC4pCnz8hMVQghRL0w1Px4eHg5uibiamIa+LrUuy+EB0Jw5c4iKikKn09G5c2fWr19f7fl//fUXnTt3RqfTER0dzdy5cy1uX7BgARqNptJXUVFRfT4NIYQQVbiShr3E5c9WryeHBkCLFi1i4sSJTJkyhR07dtCnTx8GDRpEUlKS1fOPHz/O4MGD6dOnDzt27OCFF17giSeeYPHixRbn+fj4kJKSYvGl0+ns8ZSEEEIIcQVwaAA0Y8YMxo4dy7hx42jVqhUzZ84kPDycjz/+2Or5c+fOJSIigpkzZ9KqVSvGjRvHmDFjmD59usV5Go2GkJAQiy8hhBDCkfr168fEiRNrfP6JEyfQaDTs3Lmz3toEsHbtWjQazRU5U+9SOKwIuqSkhG3btvH8889bHI+Li2Pjxo1W75OYmEhcXJzFsYEDBzJv3jz0er25GCovL4/IyEhKS0uJjY1l6tSpdOzYscq2FBcXU1xcbP4+JycHUMevbb1uhel6V9p6GFci6Wv7kb62nyupr/V6PYqiYDQaMRqNjm5OjTk5OVV7+8iRI/niiy9qfd2ffvoJFxeXGvdFo0aNOHPmDIGBgfXaf6ZrO/LnpCiK+d+LtcFoNKIoCnq9vtLPqja/Fw4LgNLT0yktLSU4ONjieHBwMKmpqVbvk5qaavV8g8FAeno6oaGhxMTEsGDBAtq1a0dOTg4ffPABvXv3ZteuXTRv3tzqdadNm8Zrr71W6Xh8fLxNi/cMRsjVgwIkJCTY7LqietLX9iN9bT9XQl87OzsTEhJCXl4eJSUljm5Ojf3777/m/y9ZsoS33nqLrVu3mo/pdDrzB2XA4gN4dZydnVEUxeK+F+Ph4UFBQUGNz68L0/Vzc3MdPtstNzf3oueUlJRQWFjIunXrMBgMFrfVpq8cPg3+wmImRVGqLXCydn7F4z169KBHjx7m23v37k2nTp348MMPmTVrltVrTp48mUmTJpm/z8nJITw8nLi4OJvuBv/PyfPc+/lWGuoU/nr2RtlduJ7p9XoSEhIYMGCA9HU9k762nyupr4uKijh16hReXl5XVB2m6e++oij4+Pig1WrNH6BPnDhBkyZNWLhwIXPnzmXTpk3Mnj2b2267jccff5y///6bzMxMmjZtyvPPP8+9995rvu6NN95Ihw4deP/99wGIjo7m4Ycf5siRI/z000/4+/vzwgsv8Mgjj5gfq2nTpmzbto3Y2FjWrl3LTTfdRHx8PJMnT2b//v3ExsYyb948WrZsaX6cN998kw8//JDCwkLuvvtuAgMD+fPPP9m+fbvV52v6oO/t7W1+7osXL+bVV1/lyJEjhIaG8thjj1m8T3788cfMnDmTU6dO4evry3XXXcePP/4IqJmuqVOncuTIETw8POjYsSNLlizB09Ozyj5XFIXc3Fy8vb0vWuRcVFSEu7s7119/faXXVW2CS4cFQIGBgTg5OVXK9qSlpVXK8piEhIRYPd/Z2bnKNSa0Wi1du3bl8OHDVbbFzc0NNze3SsddXFxs+gfGw02dumcw2v7aomrS1/YjfW0/V0Jfl5aWotFo0Gq15syCoigU6h2zrYS7i1OtZhBVHIoxtd/07+TJk3nvvff44osvcHNzo6SkhC5duvD888/j4+PDH3/8wUMPPUSzZs3o3r27+Tqm/jCZMWMGU6dOZcqUKfz00088+uij9OvXj5iYGIvHrNiHL730Eu+99x4NGzZk/PjxjBs3jg0bNgDw7bff8tZbbzFnzhx69+7N999/z3vvvUdUVFSV2Z0LH2fbtm3cc889vPrqq4wYMYKNGzcyYcIEAgMDGTVqFP/88w9PPvkkX3/9Nb169SIzM5P169ej1WpJSUnh/vvv55133mHYsGHk5uayfv36Ss+7qr6+2Hmmdmo0Gqu/A7X5nXBYAOTq6krnzp1JSEhg2LBh5uMJCQkMHTrU6n169uzJb7/9ZnEsPj6eLl26VPmkFUVh586dtGvXznaNryNXZ/WHalAc3BAhhHCQQn0prV/+0yGPvf/1gXi42uZtb+LEidxxxx0Wx5555hnz/x9//HFWrFjBjz/+aBEAXWjw4MFMmDABgOeee47333+ftWvXEhMTU+V93nzzTfr27QvA888/z5AhQygqKkKn0/Hhhx8yduxYRo8eDcDLL79MfHw8eXl5NX5uM2bM4KabbuKll14CoEWLFuzfv593332XUaNGkZSUhKenJ7fccgve3t5ERkaa62xTUlIwGAzccccdREZGAlwW77/WOHSwb9KkSXz++efMnz+fAwcO8NRTT5GUlMT48eMBNcIeOXKk+fzx48dz8uRJJk2axIEDB5g/fz7z5s2zeNG99tpr/Pnnnxw7doydO3cyduxYdu7cab6mI5kDoCunFlAIIYQVXbp0sfi+tLSUN998k/bt2xMQEICXlxfx8fFVLuti0r59e/P/TTOY09LSanyf0NBQAPN9Dh48SLdu3SzOv/D7izlw4AC9e/e2ONa7d28OHz5MaWkpAwYMIDIykujoaB588EG+/fZbc+1Nhw4duOmmm2jXrh133XUXn332GefPn6/V49uLQ2uARowYQUZGBq+//jopKSm0bduWZcuWmaPGlJQUixdPVFQUy5Yt46mnnmL27NmEhYUxa9Yshg8fbj4nKyuLRx55hNTUVHx9fenYsSPr1q2r9QugPrg6SQZICHFtc3dxYv/rAx322LZyYT3Le++9x/vvv8/MmTNp164dnp6eTJw48aLF3xeOXmg0movOgqp4H9OQXsX7VFUrW1PWanErXsPb25vt27ezdu1a4uPjefnll3n11VfZunUrfn5+JCQksHHjRuLj4/nwww+ZMmUKmzdvJioqqlbtqG8OL4KeMGGCOf13oQULFlQ61rdv3yoLuQDef/99c4HZ5caUASqVDJAQ4hql0WhsNgx1OVm/fj1Dhw7lgQceANSA5PDhw7Rq1cqu7WjZsiVbtmzhwQcfNB/7559/anWN1q1b8/fff1sc27hxIy1atDBPO3d2dqZ///7079+fV155BT8/P1avXs0dd9yBRqOhd+/e9O7dm5dffpnIyEiWLFliUUR9Obj6XoWXMVMGyIiGUqPC5V2+KIQQoqaaNWvG4sWL2bhxI/7+/syYMYPU1FS7B0CPP/44Dz/8MF26dKFXr14sWrSI3bt3Ex0dXeNrPP3003Tt2pWpU6cyYsQIEhMT+eijj5gzZw4Av//+O8eOHeP666/H39+fZcuWYTQaadmyJZs3b2bVqlXExcURFBTE5s2bOXfunN37oSYkALIjUwYIoMRgRFd54pkQQogr0EsvvcTx48cZOHAgHh4ePPLII9x+++1kZ2fbtR33338/x44d45lnnqGoqIi7776bUaNGsWXLlhpfo1OnTvzwww+8/PLLTJ06ldDQUF5//XVGjRoFgJ+fHz///DOvvvoqRUVFNG/enIULF9KmTRsOHDjAunXrmDlzJjk5OURGRvLee+8xaNCgenrGdadRajs4eA3IycnB19eX7Oxsm64DVGIw0uLF5QD888INBPrIDsn1Sa/Xs2zZMgYPHnzZTxe+0klf28+V1NdFRUUcP37cvOH1lcZoNJKTk2NeC+hKNWDAAEJCQvj6668d3ZQq1aavq3td1eb9WzJAduTiVF5UViJTwYQQQthYQUEBc+fOZeDAgTg5ObFw4UJWrlx5Rawcbm8SANmRRqPB1VlLicGIXiqhhRBC2JhGo2HZsmW88cYbFBcX07JlSxYvXkz//v0d3bTLjgRAdubqpAZAJRIACSGEsDF3d3dWrlzp6GZcEa7cQc0rlKuzOgwmQ2BCCCGE40gAZGcuZVPhS2Q1RCGEEMJhJACyM9NaQDIEJoQQQjiOBEB2ZloLSIqghRBCCMeRAMjOzBkgqQESQgghHEYCIDszZYAkABJCCCEcRwIgOzMthig1QEIIcXXr168fEydONH/fpEkTZs6cWe19NBoNS5cuveTHttV1qvPqq68SGxtbr49RnyQAsjPJAAkhxOXt1ltvrXLhwMTERDQaDdu3b6/1dbdu3cojjzxyqc2zUFUQkpKSclnuv3U5kQDIzspngck0eCGEuByNHTuW1atXc/LkyUq3zZ8/n9jYWDp16lTr6zZs2BAPD/vsARkSEoKbm+y4XR0JgOzMnAGSITAhhLgs3XLLLQQFBfHll19aHC8oKGDRokWMHTuWjIwM7r33Xho3boyHhwft2rVj4cKF1V73wiGww4cPc/3116PT6WjdurXV/bqee+45WrRogYeHB9HR0bz00kvo9XoAFixYwGuvvcauXbvQaDRoNBoWLFgAVB4C27NnDzfeeCPu7u4EBATwyCOPkJeXZ7591KhR3H777UyfPp3Q0FACAgJ49NFHzY9VE0ajkddff53GjRvj5uZGbGwsK1asMN9eUlLCY489RmhoKDqdjiZNmjBt2jTz7f/73/9o0qQJbm5uhIWF8cQTT9T4setCtsKwM5kFJoS4pikK6Asc89guHqDRXPQ0Z2dnRo4cyZdffsmTTz5pPv7jjz9SUlLC/fffT0FBAZ07d+a5557Dx8eHP/74gwcffJDo6Gi6d+9+0ccwGo3ccccdBAYGsmnTJnJycizqhUy8vb1ZsGABYWFh7Nmzh4cffhhvb2+effZZRowYwd69e1mxYoV5+wtfX99K1ygoKODmm2+mR48ebN26lbS0NMaNG8djjz1mDpgA1qxZQ2hoKGvWrOHIkSOMGDGC2NhYHn744Ys+H4APPviA9957j08++YSOHTsyf/58brvtNvbt20fz5s2ZNWsWv/76Kz/88AMRERGcOnWKU6dOAfDTTz8xZ84cFi5cSLt27UhNTWXXrl01ety6kgDIzlykBkgIcS3TF8BbYY557BeSwdWzRqeOGTOGd999l7///pshQ4YA6vDXHXfcgb+/P/7+/jzzzDPm8x9//HFWrFjBjz/+WKMAaOXKlRw4cIATJ07QuHFjAN56661KdTsvvvii+f9NmjTh6aefZtGiRTz77LO4u7vj5eWFs7MzISEhVT7Wt99+S2FhIV999RWenurz/+ijj7j11lt5++23CQ4OBsDf35+PPvoIJycnYmJiGDJkCKtWrapxADR9+nSee+457rnnHgDefvtt1qxZw8yZM5k9ezZJSUk0b96c6667Do1GQ2RkpPm+p06dIjg4mP79++Pm5kZERATdunWr0ePWlQyB2ZmsBC2EEJe/mJgYevXqxTfffAPA0aNHWb9+PWPGjAGgtLSUN998k/bt2xMQEICXlxfx8fEkJSXV6PoHDhwgIiLCHPwA9OzZs9J5P/30E9dddx0hISF4eXnx0ksv1fgxKj5Whw4dzMEPQO/evTEajRw8eNB8rE2bNjg5OZm/Dw0NJS0trUaPkZOTQ3JyMr1797Y43rt3bw4cOACow2w7d+6kZcuWPPHEE8THx5vPu/POOyksLKRZs2Y8/PDDLFmyBIPBUKvnWVuSAbIzWQlaCHFNc/FQMzGOeuxaGD16NE888QQ5OTl88cUXREZGctNNNwHw3nvv8f777zNz5kzatWuHp6cnEydOpKSkpEbXVpTKE2E0FwzPbdq0iXvuuYfXXnuNgQMH4uvry/fff897771Xq+ehKEqla1t7TBcXl0q3GY21e6+68HEqPnanTp04fvw4y5cvZ+XKldx9993079+fn376ifDwcLZu3crmzZtZvXo1EyZM4N133+Wvv/6q1C5bkQyQnbk6yW7wQohrmEajDkM54qsG9T8V3X333Tg5OfHdd9/x5ZdfMnr0aPOb+fr16xk6dCgPPPAAHTp0IDo6msOHD9f42q1btyYpKYnk5PJgMDEx0eKcDRs2EBkZyZQpU+jSpQvNmzevNDPN1dWV0tLSiz7Wzp07yc/Pt7i2VqulRYsWNW5zdXx8fAgLC+Pvv/+2OL5x40ZatWplcd6IESP47LPPWLRoEYsXLyYzMxMAd3d3brvtNmbNmsXatWtJTExkz549NmmfNZIBsrPyWWAyDV4IIS5nXl5eDBs2jBdffJHs7GxGjRplvq1Zs2YsXryYjRs34u/vz4wZM0hNTbV4s69O//79admyJSNHjuS9994jJyeHKVOmWJzTrFkzkpKS+P777+natSt//PEHS5YssTinSZMmHD9+nJ07d9K4cWO8vb0rTX+///77eeWVV3jooYd49dVXOXfuHI8//jgPPviguf7HFv7v//6PV155haZNmxIbG8sXX3zBzp07+fbbbwF4//33CQ0NJTY2Fq1Wy48//khISAh+fn4sWLCA/Px8+vbti5eXF19//TXu7u4WdUK2JhkgO3ORWWBCCHHFeOCBBzh//jz9+/cnIiLCfPyll16iU6dODBw4kH79+hESEsLtt99e4+tqtVqWLFlCcXEx3bp1Y9y4cbz55psW5wwdOpSnnnqKxx57jNjYWDZu3MhLL71kcc7w4cO5+eabueGGG2jYsKHVqfgeHh78+eefZGZm0rVrV+68805uuukmPvroo9p1xkU88cQTPP300zz99NO0a9eOFStW8Ouvv9K8eXNADSjffvttunTpQteuXTlx4gTLli1Dq9Xi5+fHV199RZ8+fWjfvj2rVq3it99+IyAgwKZtrEijWBuIvMbl5OTg6+tLdnY2Pj4+Nr32R6sOMT3hMMM6hvH+iI42vbawpNfrWbZsGYMHD663MWShkr62nyupr4uKijh+/DhRUVHodDpHN6fWjEYjOTk5+Pj4oNVKvqA+1aavq3td1eb9W36idmYugpYMkBBCCOEwEgDZmawELYQQQjieBEB2JitBCyGEEI4nAZCdmafBSwZICCGEcBgJgOzMVbbCEEJcY2SujbAlW72eJACyM9MQmF7WARJCXOVMs9QKChy0+am4KplW2664bUddyEKIdiYZICHEtcLJyQk/Pz/zflIeHh5VbslwOTIajZSUlFBUVCTT4OtZTfvaaDRy7tw5PDw8cHa+tBBGAiA7k1lgQohriWmX8ppuqnk5URSFwsJC3N3dr6jA7UpUm77WarVERERc8s9EAiA7k5WghRDXEo1GQ2hoKEFBQej1ekc3p1b0ej3r1q3j+uuvv+wXnbzS1aavXV1dbZKRkwDIzszT4CUDJIS4hjg5OV1yzYa9OTk5YTAY0Ol0EgDVM0f0tQxq2pmrs5qy00sAJIQQQjiMBEB2Vl4ELbPAhBBCCEeRAMjOZAhMCCGEcDwJgOysYhG0LA4mhBBCOIYEQHZmGgIDWQxRCCGEcBQJgOzMNAQGUggthBBCOIoEQHZWMQMkawEJIYQQjiEBkJ05aTVoUYe+pBBaCCGEcAwJgBzANAomGSAhhBDCMSQAcoCytRAplgBICCGEcAgJgBzAVAYkRdBCCCGEY0gA5ACmDJAMgQkhhBCOIQGQA5gyQFIELYQQQjiGBEAO4CQZICGEEMKhJAByAGeZBSaEEEI4lARADmCuAZIhMCGEEMIhJAByAGdt2UKIkgESQgghHEICIAeQWWBCCCGEY0kA5AAyC0wIIYRwLAmAHEAyQEIIIYRjSQDkAE6yErQQQgjhUBIAOYDsBSaEEEI4lgRADiDrAAkhhBCO5fAAaM6cOURFRaHT6ejcuTPr16+v9vy//vqLzp07o9PpiI6OZu7cuVWe+/3336PRaLj99ttt3OpLI+sACSGEEI7l0ABo0aJFTJw4kSlTprBjxw769OnDoEGDSEpKsnr+8ePHGTx4MH369GHHjh288MILPPHEEyxevLjSuSdPnuSZZ56hT58+9f00ak0yQEIIIYRjOTQAmjFjBmPHjmXcuHG0atWKmTNnEh4ezscff2z1/Llz5xIREcHMmTNp1aoV48aNY8yYMUyfPt3ivNLSUu6//35ee+01oqOj7fFUasWUAZIiaCGEEMIxnB31wCUlJWzbto3nn3/e4nhcXBwbN260ep/ExETi4uIsjg0cOJB58+ah1+txcXEB4PXXX6dhw4aMHTv2okNqAMXFxRQXF5u/z8nJAUCv16PX62v1vC5Gr9ebV4IuKjHY/PqinKlvpY/rn/S1/Uhf24/0tf3Yqq9rc3+HBUDp6emUlpYSHBxscTw4OJjU1FSr90lNTbV6vsFgID09ndDQUDZs2MC8efPYuXNnjdsybdo0XnvttUrH4+Pj8fDwqPF1aspZq6aAjp88xbJlJ21+fWEpISHB0U24Zkhf24/0tf1IX9vPpfZ1QUFBjc91WABkotFoLL5XFKXSsYudbzqem5vLAw88wGeffUZgYGCN2zB58mQmTZpk/j4nJ4fw8HDi4uLw8fGp8XVqQq/X89dXKwFoGBLK4MEdbHp9UU6v15OQkMCAAQPM2UFRP6Sv7Uf62n6kr+3HVn1tGsGpCYcFQIGBgTg5OVXK9qSlpVXK8piEhIRYPd/Z2ZmAgAD27dvHiRMnuPXWW823G41qnY2zszMHDx6kadOmla7r5uaGm5tbpeMuLi718qI3FUHrjcgvlR3U189RVCZ9bT/S1/YjfW0/l9rXtbmvw4qgXV1d6dy5c6V0V0JCAr169bJ6n549e1Y6Pz4+ni5duuDi4kJMTAx79uxh586d5q/bbruNG264gZ07dxIeHl5vz6c2pAhaCCGEcCyHDoFNmjSJBx98kC5dutCzZ08+/fRTkpKSGD9+PKAOTZ05c4avvvoKgPHjx/PRRx8xadIkHn74YRITE5k3bx4LFy4EQKfT0bZtW4vH8PPzA6h03JFkGrwQQgjhWA4NgEaMGEFGRgavv/46KSkptG3blmXLlhEZGQlASkqKxZpAUVFRLFu2jKeeeorZs2cTFhbGrFmzGD58uKOeQp3IZqhCCCGEYzm8CHrChAlMmDDB6m0LFiyodKxv375s3769xte3dg1HM22GKitBCyGEEI7h8K0wrkWSARJCCCEcSwIgBzAthCgZICGEEMIxJAByAMkACSGEEI4lAZADyCwwIYQQwrEkAHIAcwZIhsCEEEIIh5AAyAEkAySEEEI4lgRADiArQQshhBCOJQGQA5j3AitVMBoVxzZGCCGEuAZJAOQAThU2tJc6ICGEEML+JAByAOcKvS4BkBBCCGF/EgA5gEUGSAqhhRBCCLuTAMgBtBpwKYuCpBBaCCGEsD8JgBzEtWxHVMkACSGEEPYnAZCDuEgAJIQQQjiMBEAO4lpWCV0sAZAQQghhdxIAOYhrWQ2QzAITQggh7E8CIAcxZYBkCEwIIYSwPwmAHMRUBC2zwIQQQgj7kwDIQVwkAySEEEI4jARADiLT4IUQQgjHkQDIQcw1QDIEJoQQQtidBEAOYsoAyTR4IYQQwv4kAHIQUwZIiqCFEEII+5MAyEFMe4FJDZAQQghhfxIAOYgUQQshhBCOIwGQg8hCiEIIIYTjSADkIDILTAghhHAcCYAcxDwEJgGQEEIIYXcSADmIi9QACSGEEA4jAZCDuDrLLDAhhBDCUSQAchCZBSaEEEI4jgRADiJF0EIIIYTjSADkILIStBBCCOE4EgA5iBRBCyGEEI4jAZCDyGaoQgghhONIAOQgshK0EEII4TgSADmILIQohBBCOI4EQA5iWgdIiqCFEEII+5MAyEGkCFoIIYRwHAmAHERqgIQQQgjHkQDIQWQlaCGEEMJxJAByEFkJWgghhHAcCYAcRDJAQgghhONIAOQgLqbd4CUDJIQQQtidBEAOIhkgIYQQwnEkAHIQUw2QUQGDZIGEEEIIu5IAyEFMGSCQYTAhhBDC3iQAchBTBghAb1Ac2BIhhBDi2iMBkIM4azXm/xeXljqwJUIIIcS1RwIgB9FoNLIatBBCCOEgEgA5kJvMBBNCCCEcQgIgB5LVoIUQQgjHkADIgUwBkBRBCyGEEPYlAZADuZiGwKQIWgghhLArCYAcyJQBKpYaICGEEMKuJAByINkOQwghhHAMCYAcSKbBCyGEEI7h8ABozpw5REVFodPp6Ny5M+vXr6/2/L/++ovOnTuj0+mIjo5m7ty5Frf//PPPdOnSBT8/Pzw9PYmNjeXrr7+uz6dQZ+Yi6FIpghZCCCHsyaEB0KJFi5g4cSJTpkxhx44d9OnTh0GDBpGUlGT1/OPHjzN48GD69OnDjh07eOGFF3jiiSdYvHix+ZwGDRowZcoUEhMT2b17N6NHj2b06NH8+eef9npaNeYqRdBCCCGEQzg0AJoxYwZjx45l3LhxtGrVipkzZxIeHs7HH39s9fy5c+cSERHBzJkzadWqFePGjWPMmDFMnz7dfE6/fv0YNmwYrVq1omnTpjz55JO0b9+ev//+215Pq8ZkCEwIIYRwDGdHPXBJSQnbtm3j+eeftzgeFxfHxo0brd4nMTGRuLg4i2MDBw5k3rx56PV6XFxcLG5TFIXVq1dz8OBB3n777SrbUlxcTHFxsfn7nJwcAPR6PXq9vlbP62JM19Pr9Zj2Qy0stv3jCMu+FvVL+tp+pK/tR/rafmzV17W5v8MCoPT0dEpLSwkODrY4HhwcTGpqqtX7pKamWj3fYDCQnp5OaGgoANnZ2TRq1Iji4mKcnJyYM2cOAwYMqLIt06ZN47XXXqt0PD4+Hg8Pj9o+tRpJSEgg/awW0LJzzz78M/bWy+MIta+FfUhf24/0tf1IX9vPpfZ1QUFBjc91WABkotFoLL5XFKXSsYudf+Fxb29vdu7cSV5eHqtWrWLSpElER0fTr18/q9ecPHkykyZNMn+fk5NDeHg4cXFx+Pj41PYpVUuv15OQkMCAAQP4q+gg2zOSadYihsHXR9n0cYRlX1+YHRS2JX1tP9LX9iN9bT+26mvTCE5NOCwACgwMxMnJqVK2Jy0trVKWxyQkJMTq+c7OzgQEBJiPabVamjVrBkBsbCwHDhxg2rRpVQZAbm5uuLm5VTru4uJSby96FxcX3FzU7i9VNPLLVY/q8+coLElf24/0tf1IX9vPpfZ1be7rsCJoV1dXOnfuXCndlZCQQK9evazep2fPnpXOj4+Pp0uXLtU+aUVRLGp8LhduzjILTAghhHAEhw6BTZo0iQcffJAuXbrQs2dPPv30U5KSkhg/fjygDk2dOXOGr776CoDx48fz0UcfMWnSJB5++GESExOZN28eCxcuNF9z2rRpdOnShaZNm1JSUsKyZcv46quvqpxZ5kgyC0wIIYRwDIcGQCNGjCAjI4PXX3+dlJQU2rZty7Jly4iMjAQgJSXFYk2gqKgoli1bxlNPPcXs2bMJCwtj1qxZDB8+3HxOfn4+EyZM4PTp07i7uxMTE8M333zDiBEj7P78Lka2whBCCCEcw+FF0BMmTGDChAlWb1uwYEGlY3379mX79u1VXu+NN97gjTfesFXz6pU5AyQrQQshhBB25fCtMK5lLpIBEkIIIRxCAiAHKs8ASQAkhBBC2JMEQA5UXgQts8CEEEIIe5IAyIHcqhgCO5NVSEbe5TdtXwghhLhaSADkQKYMkL5CEXR2oZ64GX8xbI71/dCEEEIIcekkAHIga0XQ+85kk19SSlJmAYUl197QWH6xgZeW7mXTsQxHN0UIIcRVTAIgBzJlgIorFEH/m5pr/n9G/rU3DBa/P5WvN51kRsIhRzdFCCHEVUwCIAeythL0v6nlG7ll5pfYvU2OlpxVBMCZ84UObokQQoirmQRADlS+EnT5UNfBihmgvGsvAErJVgOfszlFGI2yQKQQQoj6IQGQA11YBF1qVDh4tuIQmG0DoFKjwufrj3G4wmNcblKz1QyQwaiQfg0OAQohhLAPCYAc6MK9wJIyCyjSlw+H2XoqfML+s7zxxwGe+mGnTa9rS6k5Reb/p2QVVXOmEEIIUXcSADnQhStBH6xQ/wO2zwCdyMgHYO+ZHE6fL7DptW3FlAECSMmWAEgIIUT9kADIgS4sgj6QYjk0ZesaoIrBxcr9Z216bVsoMRhJr/CcU7OlEFoIIUT9kADIgS4MgEwF0C2DvQHbT4NPzioPKBIOXH4B0Nkcy4yPZICEEELUFwmAHMhcA1RqRFHKC6B7NwsEbD8NvmJ9zaZjmWQX6G16/UslAZAQQgh7qVMAdOrUKU6fPm3+fsuWLUycOJFPP/3UZg27FpgCIICcIoO5Rqd3swDA9kNgpjV23F2cKDUqrDmYVudr/bormQ1H0m3VNKBywJMqAZAQQoh6UqcA6L777mPNmjUApKamMmDAALZs2cILL7zA66+/btMGXs1MQ2AA+5KzURQI9HKjRYUhMEWxzVo4an2NOqR2e8dGgDorrC7O5hTxxMIdjP96m03X6jFlgBr5uQOQLDVAQggh6kmdAqC9e/fSrVs3AH744Qfatm3Lxo0b+e6771iwYIEt23dVqxgA7T6dDUBMiDcNPF0BKNIbKbDRfmCm4MLNWcuIruEArD2YRrGh9tc/XbZKc26xgTNZtgtSTBmg2Ag/QBZDFEIIUX/qFADp9Xrc3NwAWLlyJbfddhsAMTExpKSk2K51VzknrQYnrQaAPWUBUMsQbzxcndC5qD8aW9UBmYKLUF8d7Rv5EuzjRn5JKRuP1n7T0XO55UNTx9PzbdI+KB/y6tDYF41GXSDS1ksBCCGEEFDHAKhNmzbMnTuX9evXk5CQwM033wxAcnIyAQEBNm3g1c7FSQ2Adp/JAtQMkEajIcBTDTDTbbQYommLiRBfHVqthv6tgoG6DYOdzSlvk00DoLIsVWN/Dxp6qc8/RYbBhBBC1IM6BUBvv/02n3zyCf369ePee++lQ4cOAPz666/moTFRM6ZC6FOZ6ht9TIgPAAFe6jCY7TNAan1NXJsQQF0PqLbDTGkVMkDHzuXZpH1QngEK8dUR6qsDZCaYEEKI+uFclzv169eP9PR0cnJy8Pf3Nx9/5JFH8PDwsFnjrgWuzk6AAQCtBpoHewGY64BsNRMspaxWxxRY9IhugJebM2m5xew6nUXHCP/q7m4hrUIG6JiNMkBGo2KuUwrx0RHq686u09kyE0wIIUS9qFMGqLCwkOLiYnPwc/LkSWbOnMnBgwcJCgqyaQOvdm4VCqGbBHqic3ECMA+B2aoGpmINkPq4TvRt2RCo/TBYWm6FAOicbQKg9PxiDEYFrQYaersRUtZOmQkmhBCiPtQpABo6dChfffUVAFlZWXTv3p333nuP22+/nY8//timDbzaVZwJFhPibf6/aQjMVhuiXjgEBhDXum51QBUDoOTsQor0lz5T7Wy2es1ALzdcnLSE+akBkGSAhBBC1Ic6BUDbt2+nT58+APz0008EBwdz8uRJvvrqK2bNmmXTBl7tTEXQUF7/AxBgGgKzcQbIlFkB6NcyCGethsNpebUqZq44C0xR4GTGpW+saip2NmWoQsoCNdkRXgghRH2oUwBUUFCAt7earYiPj+eOO+5Aq9XSo0cPTp48adMGXu0qZoBaVsgANbBhAFRsKDXPJgvzK88A+bq70Kms9mdH0vkaXctQajS3qbG/ei1bFEKb6n+CfdQAyFwEnSNDYEIIIWyvTgFQs2bNWLp0KadOneLPP/8kLi4OgLS0NHx8fC5yb1FRxe0wWlXIAAWWTQO3xRCYqWjZzVmLv4eLxW3Nyoqua5oBSs8rQVHUNYy6RKrBky0KoS+sUTL9eza7WBZDFEIIYXN1CoBefvllnnnmGZo0aUK3bt3o2bMnoGaDOnbsaNMGXu1MGSAPVydzRgVsOw0+ucIMMI1GY3FbdKAnUPMgxjQFPtDLlaYNaxc8Vce0BpBp6CvIW4dGo24UK4shCiGEsLU6TYO/8847ue6660hJSTGvAQRw0003MWzYMJs17lqgToNXh7+02vLgpOI0eEVRKgUutVEeXOgq3RZlCoBqOJvLlE0K8tYR1dB030sfAitfA0jNfLk6awn0cuNcbjGp2UU09Ha75McQQgghTOoUAAGEhIQQEhLC6dOn0Wg0NGrUSBZBrAPXsiLoijPAoHwafEmpkbxiA946l0r3rSnTLvBhFWaAmUSXZXFOpOdjNCoWQZg1phlgQd5uRAfaMANkCoB8ytsY5qvjXG4xKdmFtGvse8mPIYQQQpjUaQjMaDTy+uuv4+vrS2RkJBEREfj5+TF16lSMRqOt23hVM2V62jXyszju7uqEh6uaHbrUxRBTK2yDcaHG/u44azUU6ks5m3vxGVemIbAgHzeaBKqLXp4v0HP+EoapFEWxmqUKkdWghRBC1JM6ZYCmTJnCvHnz+N///kfv3r1RFIUNGzbw6quvUlRUxJtvvmnrdl61no5rSacIf4Z1alTptgAvVwoyC8nIL6FJ2VBVXSSbCoz9KmeAXJy0RAR4cOxcPsfO5VusE2RNeQZIh4erM6G+OlKyiziWnk/nsmCutnKKDOZd70N8ygMgU1skABJCCGFrdcoAffnll3z++ef897//pX379nTo0IEJEybw2WefsWDBAhs38eoW7KPjnm4RuJXVAlXUwNM2M8FMw0uhPpUzQFC7Qui0nPIMEEB0WR3QpQyDmabA+7q74O5a3g+mmWCpshq0EEIIG6tTAJSZmUlMTEyl4zExMWRmZl5yo4Qq0NM2M8FSqhkCg4qF0BcvZq6YAartfatun+UUeJPy7TAkAySEEMK26hQAdejQgY8++qjS8Y8++oj27dtfcqOEyhaLIaqLIKr3D7MyBAblhdA1yeKUzwJTM0BRNiiEPpttuQiiiWkITLbDEEIIYWt1qgF65513GDJkCCtXrqRnz55oNBo2btzIqVOnWLZsma3beM0KMC+GWPcAyLTHlrVFEE1MWZyLBTFGo2JeUdqWQ2BVZYDKh8CKLnkpACGEEKKiOmWA+vbty6FDhxg2bBhZWVlkZmZyxx13sG/fPr744gtbt/GaVb4fWN1rgCrusVVVAGEKYk5lFlBsqHpj08yCEgxGBY2mfKXq6ArBU11XbE7NsZ4BMn0viyEKIYSwtTqvAxQWFlZptteuXbv48ssvmT9//iU3TNhmNWhrm6BeqKGXG15uzuQVGziVWUCzIG+r55mGvxp4uOJStoVHIz93XJw0FBuMJGcX0tjfo9ZtTL1gI1QT02KI6XnqYoimoEsIIYS4VHXKAAn7MNUApV/CEJgpALK2CKKJRqMxD4MdrWZFaNMaQBVXZXZ20hIZcGnDYKllgZW1IC3MT9YCEkIIYXsSAF3GbLEh6sVmgJnUpJbHPAPsgqGq2m6ncaHqFmo0rQuUIlPhhRBC2JAEQJexBhWmwStK3eprUqpZBLEicyF0NUHMuVzLGWAml1IIXaQv5XyBXm2jT+U2mmauSQZICCGELdWqBuiOO+6o9vasrKxLaYu4gCkAMhgVcgoN+FYxi6s65iLoKhZBNDFncdKrXs/HvAjihQGQefis9msBmRZB1Llo8XGv/HIMqTATTAghhLCVWgVAvr7Vb0jp6+vLyJEjL6lBopzOxQlvN2dyiw1k5BfXKQAyrwLtV30A1LQGawGlVZEBupS1gMqnwLtbnaVmKoxOzpIhMCGEELZTqwBIprjbXwMv17IAqITohrW7b8VFEC+2x5dpr7H0vBKyC/X4ulcOtqqqATINgZ3JKqRIX4rOpfK2HlU5a54Cb32Gl3kxxBzJAAkhhLAdqQG6zJnXAqrDTLCaLIJo4uXmbA5CqsrkmGaBXRisBHi64q1zRlHgZEZBrdpYMQNkTWiFHeHrWgclhBBCXEgCoMuceUPUOiyGWJNFECsqXxG6ci2PoigVtsGwzABpNJoKCyLWrg4otYptMExMK06XGIyXvCeaEEIIYSIB0GUu0LQYYh0yQDVZBLEiUy2PtensOYUGig1GwHIdIBPTfmLVrSNkTWoV22CYuDk7mZcDkJlgQgghbEUCoMvcpWyIWpNFECtq2tA0E6xyEGMa/vLROVut8WlUNl39bC1rdUy1PdUFaY3KCrhrO7wmhBBCVEUCoMuceUPUOgVANVsE0aS6tYCqKoAub6dp1eqaD9UVG0rN9UZVZYAA2jVWZx/uSDpf42sLIYQQ1ZEA6DJXXgRdlxqgmi2CaBJdYSr8hRubmjJAF06BNzENU9Vm245le1LILtQT4qOjVahPled1jvQHYJsEQEIIIWxEAqDL3KVsiFrTRRBNGvu746zVUKgv5Wyu5VBWeQH0xQKgmgVqiqLwxYYTADzYM9K8uao1nSMaALD3TDZF+qp3qxdCCCFqSgKgy1xdN0TdnnSeAym5ADQJrNkO7S5OWiIaqOdeWAh9sSGwht5l7cytWQC0PSmL3aezcXXWck/X8GrPDW/gTqCXG/pShb1nsmt0fSGEEKI6EgBd5kyZlfMFJZWGpaqSV2xg4vc7KTUq3NYhjGZB3jV+vOgqCqGrWgXaJKBsun5OkYFiw8WzNAs2ngBgaIcwc51TVTQaDZ0j/QDYdlKGwYQQQlw6CYAuc/4eamal1KiQXaiv0X1e+WUfSZkFNPJzZ+rtbWv1eFUVQpv2AbM2BR7A190FZ6261tDFhutSs4tYvicFgFG9m9SoXeY6IAmAhBBC2IAEQJc5V2ctPjp1x5KaLIb4265kFm8/jVYDM++JtbqlRXVMhdAXbopavhO89SEwrVZTPhMst/oA6JtNJzEYFbpFNaBNWPX7y5mYAqDtSedlRWghhBCXTAKgK4BpGOxi22GcySpkypI9ADx2QzO6NmlQ68cybYq67eR5i5ln5TVAVQ9X1aQQukhfyndbkgAYU8PsD0CbMF9cnbSk55WQlCnrAQkhhLg0EgBdAWqyGGKpUeGpRTvJKTIQG+7H4zc1r9NjdY70p3WoD7lFBqYt/xeAghIDecUGoOoaIChfs+hcNQHQr7uSycwvoZGfO/1bBde4XToXJ9o2UqfKyzCYEEKIS+XwAGjOnDlERUWh0+no3Lkz69evr/b8v/76i86dO6PT6YiOjmbu3LkWt3/22Wf06dMHf39//P396d+/P1u2bKnPp1DvTENL1QVAP207xZbjmXi6OvHBPbHVTiuvjpNWwxvD2qLRwE/bTrPleKZ5CryHqxNebs5V3te0bUdVmaqKU99H9ozEuZZtlDogIYQQtuLQAGjRokVMnDiRKVOmsGPHDvr06cOgQYNISkqyev7x48cZPHgwffr0YceOHbzwwgs88cQTLF682HzO2rVruffee1mzZg2JiYlEREQQFxfHmTNn7PW0bM60IerZavbC+n7rKQAev6k5kQGel/R4nSL8uadrBAAvLd1Lcpa6nlCQt1u1m6o2vMgQ2NYT5zmQkoPORcuIi0x9t0YCICGEELbi0ABoxowZjB07lnHjxtGqVStmzpxJeHg4H3/8sdXz586dS0REBDNnzqRVq1aMGzeOMWPGMH36dPM53377LRMmTCA2NpaYmBg+++wzjEYjq1atstfTsrl2jdRC4WV7U6wWAB89l8eOpCyctBqGd2psk8d8dmBLGni6cvBsLu/GHwSqLoA2udh2GBuPpgNwc5sQ/Mpmt9VGpwg1ADp4NpfcoprNiBNCCCGscVgAVFJSwrZt24iLi7M4HhcXx8aNG63eJzExsdL5AwcO5J9//kGvt/6GWFBQgF6vp0GD2hcEXy5uiw3D09WJY+fy2XQss9LtP28/DUDfFg2rnKZeW/6erjw/KAaAHUlZADSspgAaLl4EbcokmQqtayvIR0d4A3cUBXaeyqrTNYQQQgiAqgs66ll6ejqlpaUEB1sWwgYHB5Oammr1PqmpqVbPNxgMpKenExoaWuk+zz//PI0aNaJ///5VtqW4uJji4vI37ZycHAD0en2VgVVdma5Xm+u6aeG2DqEs3HqarxOP0yWifN8so1Hh5+3q8N7tHUJs2t6h7YJZtMWPbWUBUKCnS7XX93NXd4lPzy22et7p8+rsrWBv1zq3s2NjP05lFrL1WAY9mvhVe25d+lrUjfS1/Uhf24/0tf3Yqq9rc3+HBUAmF9aUKIpSbZ2JtfOtHQd45513WLhwIWvXrkWnq3r4Ztq0abz22muVjsfHx+PhUbNtJGorISGhVueHFwM4s2JfKt8vPYNP2QjSoWwNKdlOuDsplJzYzjLr5VN1dpMf7EhywoiGzDPHWbbsWJXnnslX23gmM5dly5ZVuv3IGSdAQ9K/u1iWsrNO7XHN1QBOxG8/TNOigzW6T237WtSd9LX9SF/bj/S1/VxqXxcU1HyZFIcFQIGBgTg5OVXK9qSlpVXK8piEhIRYPd/Z2ZmAgACL49OnT+ett95i5cqVtG/fvtq2TJ48mUmTJpm/z8nJITw8nLi4OHx8qt6lvC70ej0JCQkMGDAAF5faLVIYf34zO09lc94/hnv6RgOwdvEeIIWhncIZektrm7bVxBh8nE/XH2fcLT1oXc2u7Wm5xbyz+y8KDBoG3jwIJ215UKooCs9uXQUYGTawr3nPsdpqkpLDT3M2cbrIlYE332DxGBe6lL4WtSN9bT/S1/YjfW0/tupr0whOTTgsAHJ1daVz584kJCQwbNgw8/GEhASGDh1q9T49e/bkt99+szgWHx9Ply5dLDrs3Xff5Y033uDPP/+kS5cuF22Lm5sbbm6V61tcXFzq7UVfl2s/2KMJO0/tYtE/Z3j0xhYU6Uv5c38aAHd1iai3tj52UwsevbF5tZk5gGBfJzQaMCqQW6KYN0gFyMgrpthgRKOBxgFeuDg71aktbRr54+nqRF6xgRPni4gJuXiAWp8/R2FJ+tp+pK/tR/rafi61r2tzX4fOAps0aRKff/458+fP58CBAzz11FMkJSUxfvx4QM3MjBw50nz++PHjOXnyJJMmTeLAgQPMnz+fefPm8cwzz5jPeeedd3jxxReZP38+TZo0ITU1ldTUVPLy8io9/pVmSPtQfN1dOJNVyF+H0li+N5WCklKiAj3pFOFXr499seAHwNlJa9677MJtO1LKpvA39HLDrY7Bj+kxYsueq0yHF0IIUVcODYBGjBjBzJkzef3114mNjWXdunUsW7aMyMhIAFJSUizWBIqKimLZsmWsXbuW2NhYpk6dyqxZsxg+fLj5nDlz5lBSUsKdd95JaGio+aviVPkrlc7Fibs6q9Pcv92UZJ79dUfHRjUKUOwhsIr9wM6UzQAL83O/5MfoHCHrAQkhhLg0Di+CnjBhAhMmTLB624IFCyod69u3L9u3b6/yeidOnLBRyy5P93WP4PO/j7P6YJr52LBOjRzYIksBnm5AXqWp8MnmAKj6tYRqopNpY1QJgIQQQtSRw7fCELUT3dCL3s0CUBRQFOgZHUBj//qZqVYXgd7W1wIyB0C+l54Big33A+BERgFZBdVvECuEEEJYIwHQFej+7pHm/99xGWV/oMIQ2AX7gSVnqTVAthgC8/NwJSpQ3e5j1+nsS76eEEKIa48EQFegAa2DaRnsTSM/dwa1q7z4oyNVtRq0LWuAADo0VrcH2SUrQgshhKgDh9cAidpzcdLy2+PXoaBc0oyq+hBYxX5gpiGwRrYKgML9WLozWQIgIYQQdSIB0BXK1fnyTN6ZMkAZFYbAig2lpOWqAZEtiqBBDYBA3RPsYquHCyGEEBe6PN9FxRXL2hDY2Wz1/27OWhp41n4XeGtah/rg4qQhI7+E0+cLbXJNIYQQ1w4JgIRNmWaBZeSVmPdpO1Nh+MtWmRqdixOtyrbl2HU6yybXFEIIce2QAEjYVEBZhqek1EhOoQGouAaQbep/TDo09gOkEFoIIUTtSQAkbErn4oS3m1pall62HYYtF0GsqGIdkBBCCFEbEgAJmzMvhlhW+Jycbbs1gCoyLYi450w2hlKjTa8thBDi6iYBkLC5CxdDrK8hsOhAT7zdnCnSGzl09srf7FYIIYT9SAAkbE7dD6x8Jpit1wAy0Wo1tA8vWxBRCqGFEELUggRAwuYCvdUMUEZeMYqi1FsGCMoLoXcmZdn82kIIIa5eEgAJmzOtBXQur4ScQgP5JaUAhPratggayuuAJAMkhBCiNiQAEjYXUGExRNMaQAGeruhcbL9thykAOnQ2l/xig82vL4QQ4uokAZCwuYYV9gOrz+EvgCAfHaG+OowK7D0jO8MLIYSoGQmAhM1V3A8sObt+1gCqyFwHJOsBCSGEqCEJgITNBVoZAquvDBBAbIQfYFkHdCqzgOd+3suqM7JJqhBCiMpkN3hhcwFlQ2AFJaUcTVPX57H1FPiKyrfEyKbEYOTzv48xa9VhivRGtBotb5cacXGpt4cXQghxBZIASNicl5szbs5aig1Gdp9W63LqMwPUrrEvGo266erNH6zj2Ll8821GRUNydhFNdW719vhCCCGuPDIEJmxOo9GYh8HSyrbDqM8AyMvNmeZBXgAcO5dPgKcr793VgehATwBOnS+st8cWQghxZZIASNQL035gJvVZBA1wc9tQNBq4r3sEq57uy/DOjYlooAZdpzIlABJCCGFJhsBEvQj0dDX/39VJS6Bn/Q5BTRrQgvF9o/FwLX9JhzfwAODU+YJ6fWwhhBBXHskAiXphGgIDCPXTodXW/2ysisEPQLi/mgFKkgyQEEKIC0gAJOqFaT8wgDDf+qv/qY55CEwyQEIIIS4gAZCoFxUzQPVZAF2dCP+yITDJAAkhhLiABECiXgRUCIAa1XMBdFUalw2B5RQZyCoocUgbhBBCXJ4kABL1ItCrwhCYgzJA7q5O+LgoACRlyjCYEEKIchIAiXrR0KII2jEBEEBAWfJJAiAhhBAVSQAk6kXgZTAEBhDopmaATmZIACSEEKKcrAMk6oWvuwuN/d0pLCmlcVkxsiOYMkCnJAMkhBCiAgmARL3QajUse7IPpaUKOhcnh7UjUCc1QEIIISqTAEjUGx+d47dgD5AhMCGEEFZIDZC4qpmGwFKyCykxGB3bGCGEEJcNCYDEVc3HBXQuWowKJGfJgohCCCFUEgCJq5pGU3FPMBkGE0IIoZIASFz1wstmoZ2UAEgIIUQZCYDEVS/ctCmqBEBCCCHKSAAkrnoRDdQMUJLMBBNCCFFGAiBx1TPVAMkQmBBCCBMJgMRVzxQAncosQFEUB7dGCCHE5UACIHHVMwVAecUGzhfoHdwaIYQQlwMJgMRVz83FiRAfdUXEkxn5Dm6NEEKIy4EEQOKaYC6EljogIYQQSAAkrhHhZQGQTIUXQggBEgCJa0RkgGSAhBBClJMASFwTTENgsiu8EEIIkABIXCNkCEwIIURFEgCJa4JpCCwlp4hiQ6mDWyOEEMLRJAAS14QAT1c8XJ1QFDh9vtDRzRFCCOFgEgCJa4JGo5Gp8EIIIcwkABLXDKkDEkIIYSIBkLhmRAV6AnAkLc/BLRFCCOFoEgCJa0arUG8A9iXnOLglQgghHE0CIHHNaBPmC8CBlByMRtkVXgghrmUSAIlrRnSgJ27OWgpKSjkhm6IKIcQ1TQIgcc1wdtISE+oDyDCYEEJc6xweAM2ZM4eoqCh0Oh2dO3dm/fr11Z7/119/0blzZ3Q6HdHR0cydO9fi9n379jF8+HCaNGmCRqNh5syZ9dh6caVpEyYBkBBCCAcHQIsWLWLixIlMmTKFHTt20KdPHwYNGkRSUpLV848fP87gwYPp06cPO3bs4IUXXuCJJ55g8eLF5nMKCgqIjo7mf//7HyEhIfZ6KuIKUR4AZTu4JUIIIRzJoQHQjBkzGDt2LOPGjaNVq1bMnDmT8PBwPv74Y6vnz507l4iICGbOnEmrVq0YN24cY8aMYfr06eZzunbtyrvvvss999yDm5ubvZ6KuEKYCqH3J+egKFIILYQQ1yqHBUAlJSVs27aNuLg4i+NxcXFs3LjR6n0SExMrnT9w4ED++ecf9Hp9vbVVXD1aBnuj1UBGfglpucWObo4QQggHcXbUA6enp1NaWkpwcLDF8eDgYFJTU63eJzU11er5BoOB9PR0QkND69SW4uJiiovL3wxzctT6EL1eb9vAqrQEQ8ZJvAtPS8BmB6Y+rtjXzhp1NtiRc/nsSsrkhpYNHdW8q4q1vhb1Q/rafqSv7cdWfV2b+zssADLRaDQW3yuKUunYxc63drw2pk2bxmuvvVbpeHx8PB4eHnW+7oUa5uyl19F36KJrREJCY5tdV1QvISHB4ntfRQtoWfrXPxQelWEwW7qwr0X9kb62H+lr+7nUvi4oqPlWRw4LgAIDA3FycqqU7UlLS6uU5TEJCQmxer6zszMBAQF1bsvkyZOZNGmS+fucnBzCw8OJi4vDx8enztet5Fw0HH0HnT6LAQMG4OLiYrtri0r0ej0JCQmV+jplwwm2rTiEwTuUwYNjHdfAq0hVfS1sT/rafqSv7cdWfW0awakJhwVArq6udO7cmYSEBIYNG2Y+npCQwNChQ63ep2fPnvz2228Wx+Lj4+nSpcsldZibm5vVgmkXFxfbvuj9GgHgWpqPRlOKi4vtskuiahf+HNs39gfgQGqu/FGzMZv/zogqSV/bj/S1/VxqX9fmvg6dBTZp0iQ+//xz5s+fz4EDB3jqqadISkpi/PjxgJqZGTlypPn88ePHc/LkSSZNmsSBAweYP38+8+bN45lnnjGfU1JSws6dO9m5cyclJSWcOXOGnTt3cuTIEbs/v0rc/VGcygKtvDTHtuUa1rpsKvypzEKyC2VsXwghrkUOrQEaMWIEGRkZvP7666SkpNC2bVuWLVtGZGQkACkpKRZrAkVFRbFs2TKeeuopZs+eTVhYGLNmzWL48OHmc5KTk+nYsaP5++nTpzN9+nT69u3L2rVr7fbcrNJowCsIsk+hyTsLDZs6tj3XKD8PVxr5uXMmq5ADKTn0iK778KkQQogrk8OLoCdMmMCECROs3rZgwYJKx/r27cv27durvF6TJk0u6/VdFM8gNNmnIO+so5tyTWsd5sOZrEL2JUsAJIQQ1yKHb4VxzfFSC7w1MgTmULIitBBCXNskALIzpSwAkgyQY1VcEVoIIcS1RwIgezNngKwv9ijsw5QBOpKWR5G+1MGtEUIIYW8SANmZZIAuD6G+Ovw9XDAYFQ6fzXN0c4QQQtiZBED2Zs4ASQDkSBqNxjwMJnVAQghx7ZEAyM4kA3T5KC+EljogIYS41kgAZG+mAKggHYxSe+JIrWUmmBBCXLMkALI3z4YoaNAoRsg/5+jWXNNMGaADKbmUGIwObo0QQgh7kgDI3rROFDuXbbCaKzPBHCkq0IuG3m4U6kv5bP0xRzdHCCGEHUkA5ABFLn7qf6QOyKGctBpeGBwDwKxVhzmZke/gFgkhhLAXCYAcwBwASQbI4W6PbUTvZgEUG4y8uHTvZb2NihBCCNuRAMgBiiUDdNnQaDS8cXs7XJ21rD+czm+7UxzdJCGEEHYgAZADFDmr689IBujyEBXoyWM3NAPg9d/2k12gd3CLhBBC1DcJgBxAMkCXn//0jaZpQ0/S84p5+89/Hd0cIYQQ9UwCIAcocpEM0OXGzdmJN4e1A+C7zUlsO5np4BYJIYSoTxIAOUD5LLA0h7ZDWOoRHcBdnRsD8O6fB6s8T19q5KvEExw7J3uICSHElUoCIAcocvZT/5OXCjLr6LLy1IAWOGs1bDqWyZ7T1leI/nTdMV7+ZR/jvvqHUmP9//w+XnuU2WuOyAw1IYSwIQmAHKDYNARWWgKF5x3bGGEhzM+dW9qHAlhdHDG3SG8+fuxcPr/sPFOv7dl6IpO3V/zLu38e5HCaZJyEEMJWJAByAKPWFUXnp34jhdCXnXF9ogH4Y08KZ7IKLW77cuMJsgr0OGk1AHyw6jCG0vrbRuOj1UfM/1+xV2rGhBDCViQAchTTpqhSCH3ZadvIl57RAZQaFRZsOG4+rmZ/1O/fuL0tAZ6unMwo4Oft9ZMF2n06i78Ole8XJwGQEELYjgRADqKYAiDJAF2WHr4+CoDvt5wit0hdF2jBhhNkF+ppFuTF3V3CGd+3KQCzVh+ul81UZ69Rsz83xgThpNWwPyWHpIwCmz+OEELUxIGUHB6av4W9Z6zXR15pJAByFMkAXdb6tQiiaUNPcosNLNp6ipwKtT9P3NQcJ62GB3pE0tDbjdPnC/lx2ymbPv7B1Fz+3HcWjQZeGBxD96gGAKzYJytVCyEcY+rv+/nr0Dnm/X384idfASQAchDJAF3etFqNuRboiw0n+Hz9cXKKDDQL8mJIO7VI2t3ViUf7qVmgj1YfoUhfarPHn7NWzf4MahtCsyBvbm4bAth/GKzUqPDCkj28F3/QLjPehBCXp4OpuWw8mgHArtNZjm2MjUgA5CiSAbrsDevYiEAvV85kFfLh6sMAPFmW/TG5p1sEob46UrKLWLS1PAukKAp5xYY6Pe7x9Hx+25UMwIR+6hYdA9uoAdD2pCzO5hTV6bp1kXg0g+82J/Hh6iM8sXBHvQz1CSEufws2lmd9jp3LJ6foyt8ySAIgB5EM0OVP5+LEgz2aAOpyTc0rZH8qnvNo2T5i7688xIhPEun37hpiXlpB21f+5P7PN5lriGrq47VHMCpq7U/bRuqSCcE+OjpF+AEQv89+QfPKA+Wvzz/2pDD2y63k1zGwu1ocScvj993Jsi7TJfpw1WHu/iSR8/kljm6KuIjz+SUs2aFO9nB1VsOGvVWsk3YlkQDIUSQDdEV4sGckbmW/8E/2b462QvbH5O4u4TT2dyerQM/m45mcyCiguCxTsuFIBqO+2FrjbNCZrELzrDJTYGViHgazUwCkKAoJ+9UAaOx1UXi4OrH+cDoPzNtMVkHt37QKS0ovOky49mAas1YdJs2OWa7aMBoVxizYymPf7bCYoSdqJ7dIz4erj7DleCZfJp6o8rwSg5FTmdUX/q/cf5ZbPlzPrlNZtm2kMFv0zymK9EbahPnQv1UQALskABJ1JRmgK0MDT1fmPtCZV25tzeC2oVbPcXXWsmB0N166pTUf3BPLokd6sO7/buDnCb3w0Tmz7eR5Rs3fUqMg6P2EQxiMCr2aBtA50t/iNtMw2KZjmXb51Pxvai5nsgrRuWh5Jq4l347rjp+HCzuSsrj7k0TS84oveo0ifSkr9qby6LfbiX09nvavxTMj4VClQCiv2MBzP+1m1BdbmZFwiOvfXcO0ZQcuu+zAhqPpJJW9IS/fU7tAVFEUjFJHBaiZxZKy9bO+2XSSYoP1wPjJ73fQ5501bDySXuW1PlxzhL1ncpj88x6pU6sHhlIjX208AcCoXk1o39gPUJfpuNJJAOQopgCoJA+KZYXfy9kNMUGM7h1lNftj0izIi7HXRTE0thHdowOICPCgU4Q/347rgY/OmX9Onmf0F9UHQRuPpPPTttMAPB3XstLtkQGetAr1odSoWAxN1VVGXjF/HTpHWq71bMvKsuzPdc0a4u7qRMcIf374T0+Cfdw4dDaPOWuOVnntvGIDk3/eTdc3VzL+m238sSeFYoOREoORWasOE/f+Otb8q+6Ft+3keQZ/sJ5F/5xCo1GHGov0Rj5Zd4w+76xhRsIhUrOL6m3IKSOvmDUH0/hg5WHGfbmVZ37cVWWt0/cV6rxWHjhb4zfcjLxiek5bzb2fbarXhTOvFH/sLg8e0/NK+H1X5dmN205msrys6P+Hf6zPskzJLjRnfvan5PDz9tO2b6yDGEqN3PvpJvrP+IvsQsfV2yTsP0tydhEBnq7c2iGM9o3VYfndV0EGyNnRDbhmuXqBiwfoC9QskJuXo1sk6kG7xr58M64793++ma0n1CBowehueLpZ/uoV6Ut5YckeAB7oEVEp+2Nyc5sQDqTk8Oe+VO7qEm4+rigKZ3OKOXg2l0OpuRw6m8ux9Hw8XJ0I9dUR6utOmJ8ORVEDjm0nz3MsPR+AFsFeLH/yeoviboCEsiBrQOsg87EWwd68eXs7xn31D7/vTmbKkFaV7gcwI/4QC7eob1ohPjpuiw3jtg5hJGUW8Ppv+0nKLGD0gq10ivBj56ksjAo08nNn+l0d6BHdgDUH05j+5yH2p+Qwa9VhZq06TKCXG+0b+9K2kS8RDTxIzirkREY+SRkFJGUWEOqr47bYRtzaIZQgb91FfzZ7z2TzxPc7OHYuv9JtseF+PNAj0uJYZn4JCfvUPnF11pKRX8L2pPN0bdLgoo+1cEsSqTlFpOYU8WXiScZeF3XR+9hSkb6U6X8eJD2vGGcnLc5aDU5aDd46F25o2ZAuTRpY/TnWh5wiPevKhg9vjw1j6c5k5m84zh2dGqHRlLdh+p+HzP9feSCNIn0pOhcni2vFl/08XJw06EsVpscfZEj7UDxcL++3NqNR4ZVf9+HqrOXFIa0snrfJon9OkXhMnXU17+/jTBrQwt7NBOCLsuzPfd0j0Lk40a6RLxqNOlyfnldMoJebQ9plC5f3q+RqptGoWaDzx9U6oICm5bcdigedD0T0cFz7hM20b+zHN2O788A8NQh6cN5mFozpho/OxXzOh6sPcyKjgGAfN569OabKa93cNoT3Vx5i3eF0cov0HE7LY/meFJbtSa20bUdNOGk1HDqbx4q9qQxpXz7El5pdxO7T2Wg0cGNMsMV9+rQIxFvnTFpuMVtPZNIjOsDidn2p0bxH2v/uaMfdXcLN2bO2jXy5vkVDZq06zPy/j7M9KQuAOzo24tWhbcx9cmNMMP1aBPHnvlQ+/usoe89kk55XzOp/01hdljm6UFpuMbtOZ/PmH/vp3SyQ4Z0ac1uHMKuZO6NR4fmfd5uDn6YNPc2p/SU7zjB7zRHu6tIYN+fyN9wlO85QUmqkbSMfmgd5s2THGeL3pV40ADKUGvlmU5L5+xnxBxncLoRQX/dq71cVo1Eho0idllxs1FBYUkpJaSldmzTAu8JrqqLP1x/j8yrWbpn711ECvdy4uW0wg9uG0jWqAS5OlQcH9KVGtp08z7pD52js78F93SPq1P6V+9Xhr2ZBXrxyaxuW701lX3IOW0+cp1vZelcbjqSTeCwDVyct3jpnMvJLWH84nQGtLV+Lf5bVwz15U3O+33qK0+cL+WzdcZ7s37xObatKicHI6ex8TmUWcPp8IWm5RZzLLeZcbjFpucV4ujnx0i2tiQnxqdH1/j6SztebTgLQtKFXpb7MKdIzI748AJy3/hijejWhgaer7Z5UDexLzmbL8Uycy9Y9A/DWuRAd6MnRc/nsPp1V6e/DlUQCIEfyDlEDoIp1QGf3wXd3gZMrPL4N/Or2R0ZcXjqEq0HQg/M2sz0piwc+38xXY7rh5+HKv6k5fPKXusjia7e1tQiMLtQi2IuoQE+Op+fT63+ryS0qH1Jz0mqICvSkZbA3zYO9aNrQiyJ9KSnZRWVfhehLjXRo7EeXJv50ivBn/oYTzFp1mI/WHGFwuxDzJ9FV/6qvydhwPxp6W37Cc3N24uY2Ify47TS/706uFACtO3SOjPwSAr1cubNz40oBiJebMy8MbsWdnRvzxYbjXN+8IYPaVa6v0mo1DGoXyqB2oRSWlLI/JYe9Z7LZfTqb1JxCGvm5ExngSWSAB439Pdh9OoulO86wPSmL9YfTWX84nQMpOUwe3KrStX/bnczeMzl4uTmzclJfQnzVjFGRvpTEoxmkZBfxw9ZTPNizCaBm2BZtVYOYEV0jCPR0ZcmOM/y57ywvDLb+Cd4kfv9ZUnOKCPRyJbyBBzuSsnj9t/18/EDnKu9TlbScIh6ct5mDZ51hR6LFbT2jA/h2XPdK/Z2RV8zcstfXQz0jCfVzx1BqxGBUSMosYOX+s6TnFfPNpiS+2ZSEq5OW6IaetAzxpkWwNw08XdlwJJ11h86RU+H1FuLrVqc3v2V71OGuIe1C8fd05Y5OjVi45RTz/z5Ot6gGKIqayQE166DRqGtxLduTYhEAnc8vYfPxTABu7RBGZIAnjy/cwdy/jnJPt3CCfS6eBbyYLzaeZM42JyZuWsnFRmDv+2wz34ztTuuwiwdB35QFPwBv/rGf61sE0tjfw3xszpqjZOSXEN3QE52zE/tTcvjkr6NWX8v1acGGEwAMbhdq0Z8dwv04ei6fnaeyJQASdWStEHrrPPXf0hJY/Sbc8Yn92yXqRYdwPxY+0oMHPt/M7tPZ3PuZGgQ9v3gPBqNCXOtg80yvqmg0Gga1DWHO2qPkFhnwcnPmplZBDGobSt8Waq1ObYzu1YTP1x/jQEoOaw6mmf+Ymep/LvzEbXJLhzB+3Haa5XtSefXWNha3LS6rwxga2whnK5kEkxbB3ky7o32N2unu6kTnSP8qhwZBDdZG9mzCifR8fvjnFHPWHuWz9ccY2DaEThHl9ysxGM1vsP+5Ptoc/IBpWYOmvPTLPj5ac4S7uoSjc3Fi56ksDp3NQ+ei5bYOYbg4aXBz1pKUWcDBs7nVfvJfUDaEcG+3CAa3C+WWD/9m+d5UVv97tlZvHmm5Rdz72SaOnsvHSaPg5+GGh5sTnq7OHE/PJ/FYBov+OcW93Sw/NH24+gh5xQbaNvLhlVvbVAqQSgxGEo9lsHxPCvH7z5KZX8K/qbn8m5pbqQ3+Hi409vdgz5lsJv+8h/inGuDrXnXAfqHsQj3rDqkFzaaM46heUSzccor4/amcyizg0NlcdiRloXPRMuGGpiRlFPDFhhOs3H+WYkOpOSu36t80So0KMSHeRAZ4EtHAg/kbjrMjKYv34g/yzp0datwua/46dI63lh8E1P7SuWgJ9/cgvIEHIb46Gnq50dDbjUAvNz5ee4Rdp7O57/NNfDO2u3n5CmtSsgvNNXzNgrw4kpbH5J/38NWYbmg0Gk5lFjC/LFs3ZXArNBoYs+Afvkw8wdjrogiyQWBXE0fS8vilbD2yUb2bWNzWobEfP28/c8UXQksA5EjeZW92pqnwxbmwe1H57bsXQa/HIKSdbR5PUeDQCmgQDQ0rF9nWSvphUIyXfp2K8tLg3z/g5AbocA8062+7a18m2oT5sug/Pbnvs80cSMkxFzh6uTnz+tC2NbrGozc0w1vnQvMgL65rHlipLqI2/D1deaBHJJ+uO8ZHq49wQ8sgCkpK2VC24uuAVtbfoHs1DcDfw4WM/BI2HcukexP1D352oZ6V+9Uhqjs6Napzuy5Fk0BPnr05htTsIn7ecYb/+3EXfzzRx9xP320+yanMQgK93Bjbp3Itzt1dw/l47VGSs4v4fksSo3pHmRe5HNw21PyG36d5ICsPpBG/72yVAdCBlBy2HM/ESavhvu4RhPq6M/a6KD5dd4yXf9lHz+jAGgWtablF3PupGvyE+uoYF53HyDv64eKituXz9cd4448DvLXsADfGBJk/rSdlFPDtZjXb8PzNrawOB7o6a+nboiF9WzTkrWEKZ7IKOXQ211xPdjanmM6R/twQE0RsuB/6UiODPljP8fR83vrjAG/fWbMgFsqHv5oHedEi2BuAliHeXNcskL+PpPPlxhPm1YYf6tWEIG8dgZ5uhPjoSM0pYv2hdPqXBeWm4a+4stmRGo2GF4e0ZvjHG/lx22lG9YqqUTbGmuwCPc/+tAuAXkFG3ht9AyF+nlVm+no2DeCh+VvYeSqL+z7bxLfjetCusfUgaOHmJIwKdI9qwLQ72jHog/WsP5zO91vV4PV/K/6lpNRI72YB3Bij1t91ivBje1IWs9cc4bUa/p24mMSjGYT46ogK9Kx0W7GhlCe/Vxc+7dM8kI7hfha3VyyEVhSl2gzo5UxmgTnShRmg3T+os8ICmkObYYACK1+13ePt+h4W3gOzu8NPY+DcoYvfx5rkHTCnJ8zuBnOvg40f1X09o/x0SJwD8wfB9Bbw+0TY8yP88jiUXp0L7rUI9uaH//QgxEdnnt3x3M0tLTIR1fF0c+a//ZrSv3XwJQU/JuOui8LVWcv2pCwSj2Ww/vA5SgxGIgM8aBZkvTjfxUlrHrYyrVoNsGxvKiWlRmJCvGkTVvWnYHt4+dbWNPR24+i5fGauVFfyzi3SM2u1us3IxP7NrRbLujk78eiN6hpMc9YeJSOv2PwcR3QtLzyPa62+8cbvr/q1/1XZGjc3tymv+XnypuaE+eo4fb58hfHqnMst5r7PNpuDn6/HdCHwgpfK6N5RdAj3I7fIwEtL95pnzE2PP4i+VKFP80Cuax540cfSajWEN/DgplbBTOjXjJn3dGThIz14ZmBLOkf646TVoHNx4u3hatCz6J9TrD9c8/WQzMNf7S2HPEeXZRjmbzjO/hR1aHL89U3NbTJlRpftVe9fUGIwF1IPbFMepHeO9GdI+1AUBSb/vLvOGYpXft3L2ZxiogI8GNbESKCXW7Vv8r7uLnw9thudIvzIKTJw3+eb2GllXSJ9qdE8k/CBHpFEN/Ti/waqHyLf/OMAv+w8wx+7U9Bo4MUhrdFoNGg0Gp4pmxX63ZYkTp+/+IbIF5sx+cvOM9z72SYGfbDO6hID0/88yL7kHPw9XJh+V4dKz71VqA/OWg2Z+SWcPl/72sPLhQRAjlQxA6Qo8M989fsuY+DGl0DrDEdWwrG/bPN4W0zDaQrsXQxzusPihyH9SM2vYSiBpRPAWDYtM3UPxE+BGa3gmztrFwhlHoOPe8OfkyFpo9qusE6g84PcZDi4rObXKs6F4+u56ED9ZSK6oRc//KcnHSP8uK1DGPd3j7z4nepJkI+OEWUzyj5afYSEsgxO/1bB1f7Rv6XsTWzFvlTztPElO9RAYXinxvXZ5Brx83DlzdvVT8ufrjvKrlNZfLbuGJn5JUQHeloEMxe6q3M4jfzcScstZuyX/5BfUkpUoKe5SBfgplZBaDWw90yO1QL0rILy1XMf6tXEfNzTzZlXblOHDT9bf4wnv9/BlCV7mLbsALNWHebjtUf5aPVhZq48xHvxB7n3s00cScsjxEfH94/0ILKBR6XHctJqeHt4O5y1GuL3n2X53lT2nM7m113JaDTw/KCqC+vroltUAx7qqb5mn1+8p0ZrXGUX6llXFixduKL6DS2DaBLggWlVgbHXReFfoeDXFDAllA2DrTt0jmKDkcb+7rQOtczyPH9zDDoXLbtOZ3PbRxsYOnsDP28/XeVaQxf6Y3cKS3cmo9XAO8PbUtNRZW+dC1+N7U7XJv7kFhkY9cWWSos4Juw/S1quOnPKtK7X6N5RdI70J6/YwJPf7wRgRJdwWlV4Xr2aBdIzOgB9qcJHq6v/e30wNZd+09fywpI9VtedOpNVyItL9wJQpDcy5sutbKgQBK07dI7P1qtDcO/c2cFqLZXOxYmYUDWDdyVPh5cAyJEqZoBObYGze8HZHWLvVWeFdRmj3r7yFTBe4tohZ7apmRsnVxj5C8Tcog5h7fkBZneFJeMho+p1XczWvQtp+8EjAB7dCkPeg8bd1GsdSYCEl2vWnuwz8NVQyEuFBk3h5rdh4l54ZA10Ga2es/Xzmj+/3yfBl7fAnp9qfh8HiwjwYMmE3sy6t2O1awzZw3/6RuOs1bDxaAa/71aDmKrqf0y6RwXQ0NuN7EI9G49lkFYIO05lo9XA0I5h9mj2RcW1CeG2DmEYFXjqh53mP+z/N7Cl1ZlOJq7OWh4rywKZPsnf3SXcIiAM8HKjS9kMsAQrq3P/ULZ6bkyIN12bWNYuxbUOpn+rYPSlCr/sTObbzUl8su4YMxIO8faKf5kef4iZKw/z4eojlsFPQOXhCpOYEB8mlG3O+/Iv+3jtt30A3B7bqF6ycc/eHENjf3fOZBXy9vJ/L3p+wv6z6EsVWgR70bxs+MtEq9UwqixI9PNwqTQ02TnCnyBvN3KLDGw4ks6fZdPfB7YJqRSkhzdQf6+GdWyEq5OWXaeymPTDLnpOW834r7fx4arDrP73rNW1pdJyinhxqbocxaM3NCP2gqGfi/Fyc2bB6G60b+xLVoGeh7/6x2LrmK8T1eHIe7qGm7eUcNJqePfO9uYV5z1dnZgUV3nK+zMD1WM/bjvNifTKSzdA+ezGkxkFfLc5idd/32/xHI1GhWd+2EVukYEO4X7cGBOkBkELtvL34XTS84qZ9IM69Pdgj8hq/wZcyoKIpUaFzccy2Hois9b3tSWpAXKkihmgf8qKn9sOB/eyP5bXPws7v1MDl/1L1NvqylRc3WYYRPdTv5J3wNr/qXVBuxaqQ3DtR8D1z1hOyzdJ2QV/z1D/P+Q9aNhC/eo6Do6uga9vhwO/qdkYN+/K9zfJT1fPzUpS65FGLwfvCr9oXcbA3zPh+F/qMF3Di6x/UVKgPi7A/qXQ/q6a9IiooLG/B8M6NuLHbacpNhjxdXehSzUFx6D+4R7SLpQFG0/wx+5UCtPVP+DXt2hYo3V47OXV29qw8Wi6ecp7bLjfRYvNAe7s3Jg5a49wKrMQJ62G4Z0r1zTFtQ5my/FM/tx3llG9y9+0S42KeZrzqF5NKr1JazQa3h/RgRV7U8ku1JNXbCCvyEBesYFSo4JT2To9zloNnm7O3Nc9wmKWUFUevbEZy/amciQtj/S8YlydtPW2foynmzNvD2/P/Z9v5utNJ7mpVRD9WgZVeX757C/rwfF93SPJzC+hR9OASjMhtVq1+P/LxJP8sjPZvIimKYtyoVahPrw/IpYpQ1qxaOspvtl0kpTsIlbsS7XYSqaBpystgr1oGexNixBvVuxN5XyBnjZhPjx+Y3NQapY1qsjTzZlPHuzMrR9u4N/UXP7vp13Mvq8TR8+phepaDdx7wbT36IZevHhLa15aupdnBra0+vvTObIBN7RsyJqD53jpl70sGN2t0tpN321JMheQF+mNLNh4ghBfHeP7qn/P5/19nMRjGXi4OjFzRCxhfjomfLOdVf+mMfbLrcSEeJOeV0yLYC+mDKl+xlmHxr58t9n6zvD5xQZKFQUvV2fzh7tiQykbj2bw595UEvafJSO/hN7NAvh2nOOWe5EAyJG8yn55CzNh31L1/13HVLi9IfR6Ata+BaumQsyt4FyHdSAKMtUhL1CDFZOwjnDfIjU7tPZtOPwn7PpOLb7ucC/0e658Gr6hBJY+CkYDtB5aVqNUQXQ/CGgGGUdg/y/Q8QHrbSnMgq+HQfoh8GmsZqO8L/iU4RcBLW6GQ8vVwHDQ29U/v6OrwFA2BHFsrdrWuvTTNe6//ZqyePtp80as1c3gMrmlvRoAJfybhqui/qG74zIY/qqogacrU4e25b/fbgfU4aCaFG26lAUPTy3axeB21hdXHNgmhDf+OMCWE+r2JKZhm+V7UziVWYivuwtDY60Xg3vrXCwWs7QFN2cn3h7ejjvnJqIoMLJnJOFWhsxspXezQO7tFsHCLUk8/NU/vD28vdWff3aB3lwrNKS99aDF1VnLJCsroJsMbhfKl4kn+XVXMooCgV6u1c4KBAj0cuPRG5rxn+uj2XriPPuSs9mXnMO+5GyOpOWRWVbEv+lYeSbC1UnLjLtjcXXWor/I3nVVCfV155MHO3HPp5tYtieV2WuOkFG2rcuNMcE08qu8BtSDPSIZ3qlRtYs4Pj+oVVmdXjr/W36AKUNam29Lyy3i7RVqJu75m2MwGBXe+OMA/1v+L0HebsSE+PDun+rsx5duaW0ufp7zQCce/XYHKw+cZdfpbFydtcy6t+NF6wtNGaC9Z3IwGhVzoPPDP6d4aeleig1GtBrwcXfB192FzLwScitkw3x0zjTyc3doEbUEQI7k0QC0Lmo9TWkxhMZCowvWBun5qDoUdP64OhQ28C11EcXa2PkdGIrU2WSNu1a+vVFnuP8Hy0Bo5zfq8FiXMdDnabU+6ewecG8Ag9+rfA2NRg2aVk+FnQutB0Al+fDdCEjdDZ4N1eCnqnWOuo1TA6Cd36n1UNWtlG3K/oBaRJ6UCNF9q+8TUUl0Qy/u6hzOon9O1biGp1OEP6G+OlKyi8hHg5ebM3EXGTpzhEHtQnl9aBs0UGndouoM69iYFsHeVmfKgDrc0irUhwMpOXy6/hhGo0LCgbPmbNOIruG1XprgUnWObMCUwa3YdCxDzWLUs1dubU1OoZ4/9qQw6YddnMwoYGL/5uY3tXWHzvHWsgPoSxVaBnvTLKia7HA1ujRpQENvN87lqnvQDWgdXOPVq52dtPRsGkDPpuU/+8KSUo6k5XHwbC6Hy2a9ncos4JHro2kZUrc2VtQ5sgGvD23L5J/38F7CIfMQ1wM9ql7b7WIrWLcM8Wb6XR147LsdfLb+OC1DfLizs/q7OvX3A+QWGWjf2JcHezbBSavhbE4Rn60/zrM/7SbYR0dJqZH+rYK5p0L9m5uzE3Pu78TERTtYsTeVV26t2YKOzYO8cHdxIq/YwLH0PJoFebPu0DmLPdmMCmQV6MkqUGtGG3q7MbBNMAPbhNAjOqDaYWh7kADIkUyrQeeU7V/TdWzlc9y8IG4qLPkPbJoDRdlw6wfgVMO1N4zG8uG1LmOrD55MgdDpf2DV6+oQ1Oa5sP0rdV0igCHT1cyUNR3ugdVvwMm/4fwJ8G9iefvqN+DUJtD5woNLILCZtauoom9Uh8cyj6mzwkx1QRcylMDBFer/A1tC+kG1FkkCoDp5c1hbnuzfnDArn1Ct0Wo13NI+1FxbM7itbWam1YeRZYsa1tbF6mfiWgdzICWHj9eW19A5azX0axlkrsmxt3F9ohnXJ9ouj6VzceLDezsS3sCDuX8d5YNVh9WtTno3YXr8IfNsLW+dM5MH170Y26lsGOyrsjqauCqGv2rK3dWJdo19q5yubgv3dotgX3I232xKokhvJKKBB9c3r+LvZw3d0j6MQ6m5zFp9hBd+3kNUoCd5xQZ+26UWbr81rJ05MJw8qBVpucX8sjOZM1mFBHq58r/h7SplXFydtcy5vzPZBXp8PWr23uLspKVtIx+2njjPrlPZlBgUJny7nVKjwrCOjZh2RztyCvVkF+rJKtSjc3aiTZiPw+sdK5IiaEfzKhszd/Otusanwz0wdDZonGDnt/D9/WrdS00cW6MGEW4+0K6GtTGNu8BDv6oZmkad1f3KjAZodSu0uaPq+/k2hqjr1f/vWmR5W/oR2PKp+v/h8y6+tpFWqwZsoGbAqprddWIdFGergWTfZ9VjhxOqv7aokrOTtsbBj8kt7ctrOm6PvTyKn+1pWMdGeLg64evuwu2xYXx0X0e2vzyAzx/qgp/HtTEUq9VqeH5QDNPuUN98l+w4w20fbWDdoXO4OGkY0zuKdf93Q7U1QjVheq1565zp1bTmmTxHevmWNubZgw/1amKTAGBi/xYMbBNMSamR/3y9jSll+wiO6hVlsQijVqvh3Ts7cFNMEDoXLe/e1aHavbtqGvyYmIbB4venMmbBVvKKDfSIbsDbw9ujc3EiyEdH82BvujZpQLvGvpdV8AOSAXI830aQvF0NclyrnuFBxwfAIxB+HKUOUX11G9z3gzqMVh3T1PoO99Z+w9XofhDVV52OfnqrWo90seG32PvUzNGuhWpAYjo/4WU1iGo2AJoPqNnjd7xfzRqd3QunNlvfG23/r+q/MUOg2U2g0cK5f9UCa8/K2ysI22vf2Jd7ujbmxMkkukT6Obo5dtck0JMdLw/ASaOpUd3U1ezebhE09ndnwjfbyS02MKR9KM8ObFnt7LXa6BbVgHeGtye8gYfFPm2XM1dnLV+N6cY/J87bLGjTajXMuDuW4R9vNK/YHeqrszp7zNVZy7xRXckvNlTahPlSmRZENM3KaxbkxScPdDHPcLvcSQDkaH2eUbMX/Z6/+Lktb1YzM9/epQYk05ur09o1WkADTs4Q0Qva3K4WERfnlK+lY214rSY0GjW4iBlSs/Nb3Qp/PK3WLCVtgsiecHwdHPxDzWANfLPmj+3uD+2Gw45vYMtnlQMgY6m6cjRAq9vU88O7qzVAhxMgdmTNH0vUmUajYeptrVm27MQVuyLspbpS3oztoU/zhqx8ui+5RYYqF9K8FHdXs37T5Urn4lSjhShrw9PNmc9GdmHo7A1k5pfwyq1t8KomwLF18APqlhgmgV6ufDGqa62zSI50ZYRpV7OwWHVK+cUyOSbh3WBsPPhFqhkVfYFa+FuSC4Xn1UDj54fh3Wbw1e3q+jxN+th2y4rquHqqs8RAnVFmLIUVL6jfdxlT+3Z0fVj9d/8vkJNseVvSJihIVxdObHKdesyUXZJhMCEcJthHVy/Bj7AU3sCDXx7tzfeP9KjR0g62FhngQXSgJ56uTsx7qGu9zjisD5IBuhI1bAmPb1dXS1YUNchRjGqB9MHlsG8JZBxWv6Du2Z+66nCvWqu0bymEtFdnj7n5Qr/Jtb9WWKya1Tm1GX4cDQ/9Vj7F3TT7q+Xg8qLwZgPKC7gNRbZ4NkIIcdkKb+DhsMBDo9Hw6+PXUWIw0sDzyqt3kwDoSuXkbH0KeaNOcMMLcHafmjXRaNXhIXuK7A2+EZCdBMufU4/1/T/wrOP499DZ8NmN6gyy5c/CrTPVwM8UALW6tfzckHbq+kp5qWiSNl3S0xBCCFE9LzdnqLqu+rImQ2BXI40GQtrCjVPghsmgtXN9glarFnWDupKqfxR0e6Tu1wtsrs4cQwPbvlALu5N3qMsHuHhC0xvKz9VozMNgmqMyDCaEEMI6CYBE/TAFQKCuY+R8iR8RWsTBTS+p/1/2f+rsMFCDHZcLpm2XBUDaIytr/zintsLsHlfUnmJCCCFqT4bARP0IaKoWdxfnqhuv2sJ1k9Td5/ctUbe/AMvhL5PofqB1RpN5FI+QszW/vqEYlo5Xt/P49XF1Ze7qFmsUQghxxZIMkKg/XcfBdU/VfuuOqmg0aj1QcNkiik6u0Dyu8nk6X4joCUBwzu6aX3/DB2rwA+rsup8fhlL9JTZaiKuMoUT9sCDEFU4CIHFlcfWEe76FsE7qwoy6KvasadYfgJDs7VWvIl1RxlFYN139/4CpahCVvB3WvWujhgtxFcg4Ch90gDk9ID/D0a0R4pJIACSuPP6R8Mia8poga1rcDEBQ7j6cvhkKaf9Wfa6iwLJn1A1po2+AXo/DLe+rt617F05tsTw/Jxk2fqjOtLvWnT8BG2bh9NUtdDs2Ux3yFFen/Az49k51+Y3MY+rvjBBXMAmAxNUpKIbSuGkYNK5okzbC3N6w8lXre6jtWwJHV4OTm1q3pNGo+7K1u1tdX+nnR6A4D84dgl8ehZntIf5F+Owm9b5XiqIcOH/y0q+Tn6EGhnP7qNmAhJfQntpEaPZ2nL4bDoVZl/4Y9U1RYMMsmN5C3abFHkOdNclE2oK+EFL3wvH1kHZA/XkZjZd+zYX3qIGPd5i6qvu+n6+s178QF5AiaHHVMnZ9mDWnXemvj0d7+E/4+311dleHe9RMT+OuYCiEFWULNPaZpBZvmwx+F05uVLf1+KQPZB4Hyt7EvMPUT8I/jlIDo4r7nl1uCjJh08ewea66PUrjrtDzMbWAvDZLJCgK7Poe/nwBCjPVYxotRPamNOoGDOtm4Ja8Xd2n7sGllqubn0yEVa+pNVahHaBRF7UdjTrVfBX02shLg5Rd6mO4+1neVpAJSyfAoeXq9xs+gBMb4M75anaxtlJ2wdr/qcOy1z2lrtF1oRMb1EDa3R/6v6IO0V7q68VQDOf2qXvlpe6F9EPq4qdZpzC/Tk00TuDZUJ1N2e8F8KnFPnlGIywZD6e3qEPDI5fCnh/VIPj3Seq6X16XtsnpJSnIVLO0TW+oerbp0TWwYSa0v0f9/a9p3xdk4FtwQl11n0vY4iHtAPz7O8TeDz7X3obBlyuNotjrY8mVIycnB19fX7Kzs/HxqaLGpI70ej3Lli1j8ODBuLhcOXumXIks+vpovLooY/ap8hNcvdQ/RumHoEFT+O9GcNFZXuT4evjyVsxvKC2HwHUToVFniH8JNs1Wj7cdrhZoXzglH9TswvkTkH4YirLUAu0GUbZ/whcqyITE2bD5E3WrlAv5RUKPCeDbWA1MMo+qNR5F2RDZC1oMhMjr1D7JPAa/PwXH1qr3DWoDPcar/eEZgF6v5+/Fc+mX9D6aggy1UH3kL2rAtfIVdVHOqkT1Veu5mt10aUGBoqjbo2z9XH08ox6c3aHtHdB5lBoMndmuBq3ZSWoRfbdHYMfX6nN284WhH5Zv5XIx+iJY9w78PVNd7wrUvfiGf65ucmxq05ZP1aDRaLB8znFT1WCwps8t8xic3IjxxEbyDq3DuyQVTcVrVqTzA89AyE9XX3MVuXioAXDvJ8DN2/IxclOgtATcfNTfD2dX9XW+cRZoXdTgp8l1aiH0Zzeqq7zH3AIjvnHMB4B//4DfJkJ+mhpUj/i6coCxaxH8MqG8/1vdCrd8UP3CrAWZsGEmyuZP0RgKUXS+aKJvUCddNOsP3sE1b+PuH9VZpYZC9edyy/vqa1JYsNV7Y23evyUAskICoKtDpb4uyVe35zi6Wn0jL0gvP/nBpZYLKla08ztI2a2+iQbFWN627Uv4Y5L6xzWgmeXq3IoRsk+rwc+Fb1QBzdX1ipoPUN+Q0vbD2f3qvznJ6vYfLQepQUHFN6mLKciEQ3+qm+AeWanOZgMIbqtmqcK7w9Z5apBgyuJUx8VT3YT25Eb1D7izDvo+p9ZJOZW/fs193TUal2/vUN+QfMMh76z6hqrRQqeR6ifws3vh9D9w5p/yWXcAQa3V67a9s3y7E1DfmLOS4My2sq/t6vceDdSNhL2Dwb2B+nM9u7f8fp5BajtMAluqQYRRD/5N4K4v1a1Wzp+En8ao7QF1axX3BpRnUTTqzzWoFQS3Ue97Zps6HJp+SD2l6U1qFqIkV73vsE8gqo8aNO5aqJ7T9k4187L5E7VP0ED7EerrKrxb5WxccR4cWqGueJ6UqPblhdz91Z9tSDtoGKMuGhrQXA1+TAGJoUR9rZ87CGunqdvKmPqn6zjIP6fWs6XtUwPBipzc1No4gDs+g/Z3l9+Wugc+vUHtz2GfQocRldtnjb5I3arm4DI1UxfZC5reqP78a5yZyYQVz8PuRZbHvYLh7q8horv6/YZZkFBWKxjeXf25GQ3qcx86W82IVVSUDYlz1A8OZR8aDBpXnJUSy/PCOqrBf8zgqttdqleHyjfPVb/X+Zb3b/t7YPA76rG6Mpaqr41Nc9RAt8to9edp7UPYhRRFneSh87PMeld07hCsn65e2/TaD2qlvs5q8hi1JAHQZUICoKtDtX1tNKqfXo/9Bd6h0P6uuj/Q8fXww4PqZrRVcfFQ/9C4eJT/Ea4JJ9eyzWxj1E/yhVnqv0U5ambGzVv9pO7mowZaSYnl2QhQ3xj7Pqf+sdZWKPkrKVA3q93+tfrHO6CZmgULaKYGH0dWweF4NSNgEtVX/fRq5Q+mRV9nHVezZnmp6o3RN8DAN9U/oBfKSoJNc2H7l+qmvqAGWVrTz0tR/9AbCmvWX87u0O5Odf+70Fg4vRX++UKtVzHtDdfqNhj6keWbT6le3UNu46yaPYahSG2bZ5BaN9b6NjV79uMoSC1besG0HYzGSc329Jig9vX5k7B6qjqMZOLZUA14Y25Vr713sRrIVnzeTq4Q1onS8O5sTdXSecgoXBpE1i7zoihw4Fe1Hi7zWOXbNU7qMJIpcAY1eL3xJXWI+EJ/vQtr3lD7sv+r6uvbybXsZ+ik9qvRoH4V56hB6pHVoM+vfC2vYPW14ualBjgFGWqQbihWszq+jdWg2s1bDWzyUtW29XpC3X/wp9HqBwitizp8nXEEEj9Sr93jUYh7Q/3ZLPkPnCubFBF1vdonxTnq71Te2fLnHtIOw/WTWXaomMEdQnA+vkb9nUjZadluv0g1MxTYXA2U/SLVAGHpBEjaqJ7T5xm4/v/UgGL9e+qHI98I6Paw+th5aWqgUZyj7vXYuBuEd1VX0b/w51uSr34oS/xI/Z2/sA/7PK0G1daGA42l6s9//XtqAAtqwN/7SfWDDqjtWPs/dcX9in9LzK8HJ/W12nWc+jeh4t+VqhRkqnVkPmFVvl6vyQBozpw5vPvuu6SkpNCmTRtmzpxJnz59qjz/r7/+YtKkSezbt4+wsDCeffZZxo8fb3HO4sWLeemllzh69ChNmzblzTffZNiwYTVukwRAVwe79nXuWfVTrXJBsalXsPqH0Tus/A9FUbaagTocr9YmKAoEt1Y/SQa1Vt8Mj6+Ff5epw1K1FdxW/aPWcpD6SbWuQxOKor5hHFur/iFudWvN/3hlHlM/RTePU78u1obC87BtgRoMmQKnirQu6vYuYZ3U4cfA5mowmHe27CsNGkSrWQh3f+vX3/uzGiy2v7vq9iRtUrNdUH6O0aDWf53dp75xmgKpDvepgV3FGiZ9kZpx2PKp+r1HANy1QH2jvdCZ7Wo26NDyypkXkwbR0OYONRMY1glcdLZ5XRtK1P4+sU792Qa3Ub8CW6hvnKUGNSAtzlW/r6rGp9QA8wao2YTa8Gmkvj79m6gfQk78XfMg1ySgOQybC427qN8X58HS/6pv8BUNmKoO95noC9Vgd9Mc69cNbKnup9jqNvSlpZX7Oves+jP7d5n6u2HKkFnj5qO2MWZI+bGkzeoaY1k1mJDg2VANqAzFav/oi9Sg0BSkuftD14fVwGL9DDXgBrV/W9ysDsX6lH1ln1LPMW2SXTGQBzXoanKdmh0uzlGPtRikZqnPHSzLUu+zzBw3aKp+2Gh6o/paMX1Ay09XA9D0Q+rvTP459Xydrzo8HtJO/X0OaQ+h7dUfy7UWAC1atIgHH3yQOXPm0Lt3bz755BM+//xz9u/fT0RE5Y0+jx8/Ttu2bXn44Yf5z3/+w4YNG5gwYQILFy5k+PDhACQmJtKnTx+mTp3KsGHDWLJkCS+//DJ///033bt3r1G7JAC6OlwVfZ1+GA4uV/+AuPurBb06P3X9I0Ox+kfH9OXmrf6x8m9i92barK9L9WpWqCKNRg0gL6zPcgRjqRoMaTRVDx2A+uZ4dBX0ngh+4dVfs1SvBgD//g4HV6iBcuuhal1ZaGylYO2ye11nHlczBsU56huqoUQNCowGNXDVOquF4VoXNXiNGVz5eRmK1eDzxN+AogaO7g3Uf51c1Exk1in1TTwnWS2ev+6pykMxiqJmN1a/oWaghs6pemjuzDZI3qm+Kbv5qL9TOj81uC4bjrxoXxfnqVmtU5vVbExWkvpVlKW+yd+5wPpq8sW5ajszj6lBjmeQOmzp4qFmZk5vUYvrS0sq3xfU3/Gej0HsferaaKD2+46v1fXMcpOr+GGhPsfu46H7f9Qs28YP1WHaio8V0g7i3oTovpb3VRQ1oPlnPuxcaL22sCoap8oZpcCW8Ji6zMg1FwB1796dTp068fHHH5uPtWrVittvv51p06ZVOv+5557j119/5cCBA+Zj48ePZ9euXSQmJgIwYsQIcnJyWL58ufmcm2++GX9/fxYuXFijdkkAdHWQvrYf6Wv7kb6ugdQ9asB1Yc1eLdW5r0vy1WDmUgrD9UVqBjbvrJqtcdGp/7p6WgRpVu+3/xc105OTrNYh5iSr2elOI9WMzYV1hbln1Vqlkxug00PqTLmLzRAtzoM9P6iZxMwT6ocz0wc0d391okfDGDWrGNhCDWTPHSybtbhH/Qpsbl5zzREBkMOmwZeUlLBt2zaef/55i+NxcXFs3LjR6n0SExOJi7MsWhs4cCDz5s1Dr9fj4uJCYmIiTz31VKVzZs6cadP2CyGEuEyFtHPs45uyMpfCRacWx9flfjUtSDfxDlaXZ6gNNy/oMkb9qqnQ8iGvy4HDAqD09HRKS0sJDracThgcHExqqpUaACA1NdXq+QaDgfT0dEJDQ6s8p6prAhQXF1NcXD6Om5Ojjn/q9Xr0etsukGa6nq2vKyqTvrYf6Wv7kb62H+lr+7FVX9fm/g5fCFFzQYpQUZRKxy52/oXHa3vNadOm8dprr1U6Hh8fj4eHR9WNvwQJCQn1cl1RmfS1/Uhf24/0tf1IX9vPpfZ1QYGV1f6r4LAAKDAwECcnp0qZmbS0tEoZHJOQkBCr5zs7OxMQEFDtOVVdE2Dy5MlMmlQ+xTMnJ4fw8HDi4uLqpQYoISGBAQMGyPh9PZO+th/pa/uRvrYf6Wv7sVVfm0ZwasJhAZCrqyudO3cmISHBYop6QkICQ4daX4m1Z8+e/PbbbxbH4uPj6dKli7nDevbsSUJCgkUdUHx8PL169aqyLW5ubri5VV4zwcXFpd5e9PV5bWFJ+tp+pK/tR/rafqSv7edS+7o293XoENikSZN48MEH6dKlCz179uTTTz8lKSnJvK7P5MmTOXPmDF999RWgzvj66KOPmDRpEg8//DCJiYnMmzfPYnbXk08+yfXXX8/bb7/N0KFD+eWXX1i5ciV///23Q56jEEIIIS4/Dg2ARowYQUZGBq+//jopKSm0bduWZcv+v717j6m6/v8A/jzcDocTMS7C4UgiFIVAmoE1kSKlEUg2CrowUKg/HNdAV2mBQS7DP5o5N6PhjP6AhmOiI28FZpS2wgFHUUhrIZrCyGlyiAT1vL5/ND+/3/mifhEO5wjn+djOxnm/3+ec13nu6Hntcz6XfQgM/PeChL29vTh79v/OCRIUFIR9+/Zh1apV2Lp1K/R6PbZs2aKcAwgAoqOjUVtbi5KSEqxbtw4PPvggduzYMeZzABEREdH0Z/OdoHNzc5Gbm3vLuS+++GLUWGxsLNra7nzW0dTUVKSmplqiPCIiIpqGxnARDyIiIqLphQ0QERER2R02QERERGR32AARERGR3WEDRERERHaHDRARERHZHTZAREREZHdsfh6ge9HNC6zezTVFxuratWsYGhrCwMAAT60+yZi19TBr62HW1sOsrcdSWd/83r75PX4nbIBuwWg0AgAeeOABG1dCREREd8toNMLDw+OOa1QyljbJzphMJly4cAHu7u5QqVQWfe6bV5o/d+6cxa80T+aYtfUwa+th1tbDrK3HUlmLCIxGI/R6PRwc7ryXD7cA3YKDgwMCAgIm9TXuv/9+/oOyEmZtPczaepi19TBr67FE1v9ry89N3AmaiIiI7A4bICIiIrI7bICsTK1Wo7S0FGq12talTHvM2nqYtfUwa+th1tZji6y5EzQRERHZHW4BIiIiIrvDBoiIiIjsDhsgIiIisjtsgIiIiMjusAGyok8//RRBQUFwdXVFZGQkfvjhB1uXNOWVl5djwYIFcHd3h6+vL5KTk3Hq1CmzNSKCsrIy6PV6aDQaPPPMMzh58qSNKp4+ysvLoVKpUFRUpIwxa8s5f/48MjIy4O3tDTc3Nzz22GNobW1V5pm1ZVy/fh0lJSUICgqCRqNBcHAw1q9fD5PJpKxh1uP3/fffY9myZdDr9VCpVNi9e7fZ/FiyHR4eRkFBAXx8fKDVavHCCy/gjz/+mHhxQlZRW1srzs7Osm3bNuns7JTCwkLRarXS09Nj69KmtOeee06qqqrkxIkTYjAYJCkpSWbNmiWDg4PKmo0bN4q7u7vs3LlTOjo65NVXXxV/f38ZGBiwYeVTW0tLi8yePVvmzp0rhYWFyjiztoxLly5JYGCgZGVlyc8//yzd3d3S1NQkv/32m7KGWVvGhx9+KN7e3rJnzx7p7u6Wuro6ue+++2Tz5s3KGmY9fvv27ZPi4mLZuXOnAJBdu3aZzY8l2+zsbJk5c6Y0NjZKW1ubLF68WObNmyfXr1+fUG1sgKzkiSeekOzsbLOx0NBQWbt2rY0qmp76+/sFgDQ3N4uIiMlkEp1OJxs3blTWXL16VTw8POSzzz6zVZlTmtFolJCQEGlsbJTY2FilAWLWlrNmzRqJiYm57TyztpykpCR54403zMZeeuklycjIEBFmbUn/3QCNJdu//vpLnJ2dpba2Vllz/vx5cXBwkAMHDkyoHv4EZgUjIyNobW1FfHy82Xh8fDx+/PFHG1U1PV25cgUA4OXlBQDo7u5GX1+fWfZqtRqxsbHMfpzy8vKQlJSEZ5991mycWVtOQ0MDoqKi8PLLL8PX1xfz58/Htm3blHlmbTkxMTE4ePAgTp8+DQA4duwYDh8+jKVLlwJg1pNpLNm2trbi2rVrZmv0ej0iIiImnD8vhmoFFy9exI0bN+Dn52c27ufnh76+PhtVNf2ICFavXo2YmBhEREQAgJLvrbLv6emxeo1TXW1tLdra2nD06NFRc8zacn7//XdUVFRg9erVeO+999DS0oI333wTarUaK1asYNYWtGbNGly5cgWhoaFwdHTEjRs3sGHDBqSlpQHg53oyjSXbvr4+uLi4wNPTc9SaiX5/sgGyIpVKZXZfREaN0fjl5+fj+PHjOHz48Kg5Zj9x586dQ2FhIb755hu4urredh2znjiTyYSoqCh89NFHAID58+fj5MmTqKiowIoVK5R1zHriduzYgerqanz55ZcIDw+HwWBAUVER9Ho9MjMzlXXMevKMJ1tL5M+fwKzAx8cHjo6Oo7rV/v7+UZ0vjU9BQQEaGhpw6NAhBAQEKOM6nQ4AmL0FtLa2or+/H5GRkXBycoKTkxOam5uxZcsWODk5KXky64nz9/dHWFiY2dicOXNw9uxZAPxcW9Lbb7+NtWvX4rXXXsOjjz6K5cuXY9WqVSgvLwfArCfTWLLV6XQYGRnB5cuXb7tmvNgAWYGLiwsiIyPR2NhoNt7Y2Ijo6GgbVTU9iAjy8/NRX1+Pb7/9FkFBQWbzQUFB0Ol0ZtmPjIygubmZ2d+luLg4dHR0wGAwKLeoqCikp6fDYDAgODiYWVvIokWLRp3O4fTp0wgMDATAz7UlDQ0NwcHB/KvQ0dFROQyeWU+esWQbGRkJZ2dnszW9vb04ceLExPOf0C7UNGY3D4Pfvn27dHZ2SlFRkWi1Wjlz5oytS5vScnJyxMPDQ7777jvp7e1VbkNDQ8qajRs3ioeHh9TX10tHR4ekpaXxEFYL+f9HgYkwa0tpaWkRJycn2bBhg/z6669SU1Mjbm5uUl1draxh1paRmZkpM2fOVA6Dr6+vFx8fH3nnnXeUNcx6/IxGo7S3t0t7e7sAkE2bNkl7e7tyCpixZJudnS0BAQHS1NQkbW1tsmTJEh4GP9Vs3bpVAgMDxcXFRR5//HHlUG0aPwC3vFVVVSlrTCaTlJaWik6nE7VaLU8//bR0dHTYruhp5L8bIGZtOV999ZVERESIWq2W0NBQqaysNJtn1pYxMDAghYWFMmvWLHF1dZXg4GApLi6W4eFhZQ2zHr9Dhw7d8v/ozMxMERlbtv/884/k5+eLl5eXaDQaef755+Xs2bMTrk0lIjKxbUhEREREUwv3ASIiIiK7wwaIiIiI7A4bICIiIrI7bICIiIjI7rABIiIiIrvDBoiIiIjsDhsgIiIisjtsgIiIbkOlUmH37t22LoOIJgEbICK6J2VlZUGlUo26JSQk2Lo0IpoGnGxdABHR7SQkJKCqqspsTK1W26gaIppOuAWIiO5ZarUaOp3O7Obp6Qng35+nKioqkJiYCI1Gg6CgINTV1Zk9vqOjA0uWLIFGo4G3tzdWrlyJwcFBszWff/45wsPDoVar4e/vj/z8fLP5ixcv4sUXX4SbmxtCQkLQ0NCgzF2+fBnp6emYMWMGNBoNQkJCRjVsRHRvYgNERFPWunXrkJKSgmPHjiEjIwNpaWno6uoCAAwNDSEhIQGenp44evQo6urq0NTUZNbgVFRUIC8vDytXrkRHRwcaGhrw0EMPmb3GBx98gFdeeQXHjx/H0qVLkZ6ejkuXLimv39nZif3796OrqwsVFRXw8fGxXgBENH4TvpwqEdEkyMzMFEdHR9FqtWa39evXi4gIAMnOzjZ7zJNPPik5OTkiIlJZWSmenp4yODiozO/du1ccHBykr69PRET0er0UFxfftgYAUlJSotwfHBwUlUol+/fvFxGRZcuWyeuvv26ZN0xEVsV9gIjonrV48WJUVFSYjXl5eSl/L1y40Gxu4cKFMBgMAICuri7MmzcPWq1WmV+0aBFMJhNOnToFlUqFCxcuIC4u7o41zJ07V/lbq9XC3d0d/f39AICcnBykpKSgra0N8fHxSE5ORnR09LjeKxFZFxsgIrpnabXaUT9J/S8qlQoAICLK37dao9FoxvR8zs7Oox5rMpkAAImJiejp6cHevXvR1NSEuLg45OXl4eOPP76rmonI+rgPEBFNWT/99NOo+6GhoQCAsLAwGAwG/P3338r8kSNH4ODggIcffhju7u6YPXs2Dh48OKEaZsyYgaysLFRXV2Pz5s2orKyc0PMRkXVwCxAR3bOGh4fR19dnNubk5KTsaFxXV4eoqCjExMSgpqYGLS0t2L59OwAgPT0dpaWlyMzMRFlZGf78808UFBRg+fLl8PPzAwCUlZUhOzsbvr6+SExMhNFoxJEjR1BQUDCm+t5//31ERkYiPDwcw8PD2LNnD+bMmWPBBIhosrABIqJ71oEDB+Dv72829sgjj+CXX34B8O8RWrW1tcjNzYVOp0NNTQ3CwsIAAG5ubvj6669RWFiIBQsWwM3NDSkpKdi0aZPyXJmZmbh69So++eQTvPXWW/Dx8UFqauqY63NxccG7776LM2fOQKPR4KmnnkJtba0F3jkRTTaViIitiyAiulsqlQq7du1CcnKyrUshoimI+wARERGR3WEDRERERHaH+wAR0ZTEX++JaCK4BYiIiIjsDhsgIiIisjtsgIiIiMjusAEiIiIiu8MGiIiIiOwOGyAiIiKyO2yAiIiIyO6wASIiIiK7wwaIiIiI7M5/ACLuwYcngs1gAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'], label='Training loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Training and Validation Loss Curves')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14c9096c-6919-46d6-9d4c-3c3ddec60d09",
   "metadata": {},
   "source": [
    "Plot the training and validation RMSE curves."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "300cdf26-a590-4372-8553-5a302cb15069",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkkAAAHFCAYAAADmGm0KAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAADDwUlEQVR4nOzdd1wT9/8H8Ndls/dUhiiKAyfuWjeu2mm17tnWr11q+2u11lZta7faaZeK1t1tW2vFbeseOFFxACogSzaEjPv9cbkjIQkkEJKI7+fjwQO4XO4ux3rz/rw/7w/DsiwLQgghhBBiQOToCyCEEEIIcUYUJBFCCCGEmEBBEiGEEEKICRQkEUIIIYSYQEESIYQQQogJFCQRQgghhJhAQRIhhBBCiAkUJBFCCCGEmEBBEiGEEEKICRQkkXsSwzAWve3bt69e51m0aBEYhqnTc/ft22eTa3B2U6ZMQWRkpNnHc3JyIJPJ8NRTT5ndp6ioCK6urnj44YctPm9CQgIYhkFqaqrF16KPYRgsWrTI4vPxMjIysGjRIiQlJRk9Vp/vl/qKjIw0+N53c3ND586d8cUXX6D6wgr89ybDMEhISDB5vAEDBoBhGKP7WVpaig8++AAdOnSAp6cnPDw80Lx5c4wePRr79+83eQ5Tb+bOW93BgwcxevRoNGnSBDKZDF5eXujVqxdWrlyJ0tJSa24RIVaTOPoCCKmLw4cPG3z+9ttvY+/evdizZ4/B9jZt2tTrPDNmzMDQoUPr9NzOnTvj8OHD9b6Ge11AQAAefvhh/Pbbb7h79y58fHyM9tm8eTPKy8sxffr0ep1r4cKFeOmll+p1jNpkZGRg8eLFiIyMRMeOHQ0eq8/3iy307t0bH3/8MQDuOpctW4YXXngBRUVFeP3114329/DwwKpVqzBlyhSD7Tdu3MC+ffvg6elpsF2j0SA+Ph7nzp3D//3f/6Fbt24AgJSUFPzxxx84ePAg+vbta/CcpUuXon///kbnbt68ea2v56233sKSJUvQq1cvvP3222jevDnKyspw6NAhLFq0CFeuXMHy5ctrPQ4hdcYS0ghMnjyZdXNzq3W/0tJSO1zN/WXy5MlsREREjfts376dBcB+/vnnJh/v3r07GxQUxKpUKovPu2bNGhYAe+PGDSuutgoA9q233rL6ecePH2cBsGvWrKnTeRtKREQEO2LECINthYWFrJeXFxseHm6wfe/evSwAdsaMGSwA9sqVKwaPv/HGG2zTpk3ZYcOGGXxt9+zZwwJgV69ebfIaNBqN0Tl+/PHHOr2erVu3sgDY6dOns1qt1ujxoqIi9p9//qnTsauj3wvEHBpuI41Wv3790K5dOxw4cAC9evWCq6srpk2bBgDYsmUL4uPjERISAhcXF7Ru3Rrz5s0zSt+bGj6JjIzEQw89hB07dqBz585wcXFBTEwMVq9ebbCfqeG2KVOmwN3dHVevXsXw4cPh7u6OsLAwvPzyy1AqlQbPv3XrFkaNGgUPDw94e3tj/PjxOH78uEVDFTk5OZg1axbatGkDd3d3BAYGYsCAATh48KDBfqmpqWAYBh9//DGWLVuGZs2awd3dHT179sSRI0eMjpuQkIBWrVpBLpejdevWWLduXY3XwRsyZAiaNm2KNWvWGD2WnJyMo0ePYtKkSZBIJEhMTMQjjzyCpk2bQqFQoEWLFnj22WeRm5tb63lMDbcVFRXh6aefhp+fH9zd3TF06FBcuXLF6LlXr17F1KlTER0dDVdXVzRp0gQjR47EuXPnhH327duHrl27AgCmTp0qDB3xw3amvl+0Wi0+/PBDxMTEQC6XIzAwEJMmTcKtW7cM9uO/X48fP44+ffrA1dUVUVFReP/996HVamt97aZ4enqiZcuWuHPnjsnHBw8ejLCwMIPvXa1Wi7Vr12Ly5MkQiQz/ROTl5QEAQkJCTB6v+v71sWTJEvj4+OCzzz4zOYTp4eGB+Ph4AFXfx6Z+LqoPq/Jfo1OnTmHUqFHw8fFB8+bNsWLFCjAMg6tXrxod47XXXoNMJjP4Hty1axcGDhwIT09PuLq6onfv3ti9e7fB83JycvDMM88gLCwMcrkcAQEB6N27N3bt2lXHu0LsjYIk0qhlZmZiwoQJGDduHLZv345Zs2YB4IYHhg8fjlWrVmHHjh2YPXs2tm7dipEjR1p03DNnzuDll1/GnDlz8Pvvv6N9+/aYPn06Dhw4UOtzVSoVHn74YQwcOBC///47pk2bhuXLl+ODDz4Q9iktLUX//v2xd+9efPDBB9i6dSuCgoIwZswYi64vPz8fADdc8ddff2HNmjWIiopCv379TNZIffnll0hMTMSKFSuwYcMGlJaWYvjw4SgsLBT2SUhIwNSpU9G6dWv8/PPPeOONN/D2228bDXGaIhKJMGXKFJw6dQpnzpwxeIwPnPgA9tq1a+jZsydWrlyJnTt34s0338TRo0fxwAMPQKVSWfT6eSzL4tFHH8UPP/yAl19+Gb/++it69OiBYcOGGe2bkZEBPz8/vP/++9ixYwe+/PJLSCQSdO/eHZcvXwbADaHy1/vGG2/g8OHDOHz4MGbMmGH2Gv73v//htddew+DBg7Ft2za8/fbb2LFjB3r16mUU+GVlZWH8+PGYMGECtm3bhmHDhmH+/PlYv369Va+bp1arcfPmTbRs2dLk4/zXZd26ddBoNACAnTt34tatW5g6darR/nFxcZBKpXjppZewYcMGZGZm1noNWq0WarXa6K0mmZmZOH/+POLj4+Hq6mrBK7Xe448/jhYtWuDHH3/E119/jQkTJkAmkxkFWhqNBuvXr8fIkSPh7+8PAFi/fj3i4+Ph6emJtWvXYuvWrfD19cWQIUMMAqWJEyfit99+w5tvvomdO3fi+++/x6BBg4Rgk9wDHJ3KIsQWTA239e3blwXA7t69u8bnarVaVqVSsfv372cBsGfOnBEee+utt9jqPyYRERGsQqFg09LShG3l5eWsr68v++yzzwrb+OGGvXv3GlwnAHbr1q0Gxxw+fDjbqlUr4fMvv/ySBcD+/fffBvs9++yzdRrqUavVrEqlYgcOHMg+9thjwvYbN26wANjY2FhWrVYL248dO8YCYDdt2sSyLDeMEhoaynbu3Nlg6CM1NZWVSqW1DrexLMtev36dZRiGffHFF4VtKpWKDQ4OZnv37m3yOfzXJi0tjQXA/v7778Jjpobbqg/9/f333ywA9tNPPzU47rvvvlvrcJtarWYrKyvZ6Ohods6cOcL2mobbqn+/JCcnswDYWbNmGex39OhRFgD7+uuvC9v479ejR48a7NumTRt2yJAhZq+TFxERwQ4fPpxVqVTCPXv66adZqVTK/vnnnwb76g+F8V8Xfp8nn3yS7devH8uyLDtixAijr+2qVatYd3d3FgALgA0JCWEnTZrEHjhwwOQ5zL3dvHnT7Gs5cuQIC4CdN29era+bZau+j019Tap/nfmv0Ztvvmm07+OPP842bdrUYNiQHyr+448/WJblhuZ8fX3ZkSNHGjxXo9GwHTp0YLt16yZsc3d3Z2fPnm3RayDOiTJJpFHz8fHBgAEDjLZfv34d48aNQ3BwMMRiMaRSqVBwmpycXOtxO3bsiPDwcOFzhUKBli1bIi0trdbnMgxjlLFq3769wXP3798PDw8PoyLgsWPH1np83tdff43OnTtDoVBAIpFAKpVi9+7dJl/fiBEjIBaLDa4HgHBNly9fRkZGBsaNG2cw9BEREYFevXpZdD3NmjVD//79sWHDBlRWVgIA/v77b2RlZQlZJADIzs7GzJkzERYWJlx3REQEAMu+Nvr27t0LABg/frzB9nHjxhntq1arsXTpUrRp0wYymQwSiQQymQwpKSlWn7f6+asXRnfr1g2tW7c2Gp4JDg4WiqF51b83arJ9+3ZIpVLhnn333Xf4/PPPMWLECLPPadasGfr164fVq1cjLy9PyG6aM23aNNy6dQsbN27Eiy++iLCwMKxfvx59+/bFRx99ZLT/Bx98gOPHjxu9BQUFWfSaGsoTTzxhtG3q1Km4deuWwXDYmjVrEBwcLGQfDx06hPz8fEyePNkgM6bVajF06FAcP35cGLbv1q0bEhIS8M477+DIkSNWZ0KJ41GQRBo1U7UTJSUl6NOnD44ePYp33nkH+/btw/Hjx/HLL78AAMrLy2s9rp+fn9E2uVxu0XNdXV2hUCiMnltRUSF8npeXZ/KPiKV/WJYtW4b//e9/6N69O37++WccOXIEx48fx9ChQ01eY/XXI5fLAVTdC354IDg42Oi5praZM336dOTl5WHbtm0AuD9A7u7uGD16NABuaCY+Ph6//PILXn31VezevRvHjh0T6qMsub/68vLyIJFIjF6fqWueO3cuFi5ciEcffRR//PEHjh49iuPHj6NDhw5Wn1f//IDp78PQ0FCjYZf6fF8BwAMPPIDjx4/jyJEj+OGHHxAZGYnnn38e//77b43Pmz59Ov744w8sW7YMLi4uGDVqVI37e3l5YezYsfj0009x9OhRnD17FkFBQViwYAEKCgoM9o2KikJcXJzRm1QqNXt8/h+QGzduWPS668LU12TYsGEICQkRhlTv3r2Lbdu2YdKkScI/EXx916hRo4SAlH/74IMPwLKsMNy9ZcsWTJ48Gd9//z169uwJX19fTJo0CVlZWQ32uohtUQsA0qiZKvjcs2cPMjIysG/fPoPpytV/uTuSn58fjh07ZrTd0l+u69evR79+/bBy5UqD7cXFxXW+HnPnt+YX/uOPPw4fHx+sXr0affv2xZ9//olJkybB3d0dAHD+/HmcOXMGCQkJmDx5svA8U8W0ll63Wq1GXl6eQQBi6prXr1+PSZMmYenSpQbbc3Nz4e3tXefzA1yNTdOmTQ0ey8jIEGpcbMXLywtxcXEAgO7du6N79+7o0KEDZs2ahaSkJLOF1Y8//jiee+45vP/++3j66afh4uJi1Xnbtm2Lp556CitWrMCVK1eMsmHWCgkJQWxsLHbu3ImysrJa65L4fzqqT36oqfbH1O8GsViMiRMn4rPPPkNBQQE2btwIpVJpUJ/Ff80+//xz9OjRw+Sx+X9m/P39sWLFCqxYsQLp6enYtm0b5s2bh+zsbOzYsaPG10ScA2WSyH2H/+XIZ0t433zzjSMux6S+ffuiuLgYf//9t8H2zZs3W/R8hmGMXt/Zs2eN+ktZqlWrVggJCcGmTZsMGhOmpaXh0KFDFh9HoVBg3Lhx2LlzJz744AOoVCqDoR1bf234/jwbNmww2L5x40ajfU3ds7/++gu3b9822FY9y1YTfqi3euH18ePHkZycjIEDB9Z6jPqIjo7Gq6++inPnzmHLli1m93NxccGbb76JkSNH4n//+5/Z/fLy8oSh0uouXboEgMuQ2cLChQtx9+5dvPjii0bNMAEuI7xz504AXFCiUChw9uxZg31+//13q887depUVFRUYNOmTUhISEDPnj0RExMjPN67d294e3vj4sWLJjNkcXFxkMlkRscNDw/H888/j8GDB+PUqVNWXxdxDMokkftOr1694OPjg5kzZ+Ktt96CVCrFhg0bjGZdOdLkyZOxfPlyTJgwAe+88w5atGiBv//+G//88w+A2qdaP/TQQ3j77bfx1ltvoW/fvrh8+TKWLFmCZs2a1TqzyBSRSIS3334bM2bMwGOPPYann34aBQUFWLRokVXDbQA3tPPll19i2bJliImJMahpiomJQfPmzTFv3jywLAtfX1/88ccfSExMtPqaASA+Ph4PPvggXn31VZSWliIuLg7//fcffvjhB6N9H3roISQkJCAmJgbt27fHyZMn8dFHHxllgJo3bw4XFxds2LABrVu3hru7O0JDQ00GB61atcIzzzyDzz//HCKRCMOGDUNqaioWLlyIsLAwzJkzp06vyxqvvPIKvv76ayxevBijR482qD3TN3fuXMydO7fGY+3duxcvvfQSxo8fj169esHPzw/Z2dnYtGkTduzYgUmTJhndr5SUFJPtJJo2bWq0r74nn3wSCxcuxNtvv41Lly5h+vTpQjPJo0eP4ptvvsGYMWMQHx8PhmEwYcIErF69Gs2bN0eHDh1w7Ngxk8FwbWJiYtCzZ0+89957uHnzJr799luDx93d3fH5559j8uTJyM/Px6hRoxAYGIicnBycOXMGOTk5WLlyJQoLC9G/f3+MGzcOMTEx8PDwwPHjx7Fjxw48/vjjVl8XcRDH1o0TYhvmZre1bdvW5P6HDh1ie/bsybq6urIBAQHsjBkz2FOnThnNkDE3u6160z7+fH379hU+Nze7zVTTS1PnSU9PZx9//HHW3d2d9fDwYJ944glhpo3+LC9TlEol+8orr7BNmjRhFQoF27lzZ/a3334zmv3Fzwr66KOPjI4BE7O/vv/+ezY6OpqVyWRsy5Yt2dWrV1vUTLK6Tp06sQDYDz/80OixixcvsoMHD2Y9PDxYHx8f9sknn2TT09ONrseS2W0sy7IFBQXstGnTWG9vb9bV1ZUdPHgwe+nSJaPj3b17l50+fTobGBjIurq6sg888AB78OBBo68ry7Lspk2b2JiYGFYqlRocx9TXUaPRsB988AHbsmVLViqVsv7+/uyECROMZneZ+3619P6a+75k2arZkmvXrmVZ1vJGj9Vnt928eZN944032N69e7PBwcGsRCJhPTw82O7du7Off/65wQzJ2ma3LViwoNbXxLIsu3//fnbUqFFsSEgIK5VKWU9PT7Znz57sRx99xBYVFQn7FRYWsjNmzGCDgoJYNzc3duTIkWxqaqrZ2W05OTlmz/ntt9+yAFgXFxe2sLDQ7HWNGDGC9fX1ZaVSKdukSRN2xIgRwj2tqKhgZ86cybZv35719PRkXVxc2FatWrFvvfUWNa+8hzAsayKPSQhxSkuXLsUbb7yB9PT0Gv8LJ4QQUn803EaIk/riiy8AcOl/lUqFPXv24LPPPsOECRMoQCKEEDugIIkQJ+Xq6orly5cjNTUVSqUS4eHheO211/DGG284+tIIIeS+QMNthBBCCCEmUAsAQgghhBATKEgihBBCCDGBgiRCCCGEEBOocLuOtFotMjIy4OHhYbK9PSGEEEKcD8uyKC4uRmhoaK2NeSlIqqOMjAyEhYU5+jIIIYQQUgc3b96stZ0KBUl15OHhAYC7yZ6enjY9tkqlws6dOxEfH1/jStmk/uhe2w/da/uhe20/dK/tx1b3uqioCGFhYcLf8ZpQkFRH/BCbp6dngwRJrq6u8PT0pB+6Bkb32n7oXtsP3Wv7oXttP7a+15aUylDhNiGEEEKICRQkEUIIIYSYQEESIYQQQogJVJNECCHE5rRaLSorKx19GQ1OpVJBIpGgoqICGo3G0ZfTqFl6r6VSKcRisU3OSUESIYQQm6qsrMSNGzeg1WodfSkNjmVZBAcH4+bNm9Qzr4FZc6+9vb0RHBxc768JBUmEEEJshmVZZGZmQiwWIywsrNZmffc6rVaLkpISuLu7N/rX6miW3GuWZVFWVobs7GwAQEhISL3OSUESIYQQm1Gr1SgrK0NoaChcXV0dfTkNjh9WVCgUFCQ1MEvvtYuLCwAgOzsbgYGB9Rp6o68oIYQQm+FrRWQymYOvhNzP+ABdpVLV6zgUJBFCCLE5qs8hjmSr7z8KkgghhBBCTKAgiRBCCGkA/fr1w+zZsy3ePzU1FQzDICkpqcGuiViHgiRCCCH3NYZhanybMmVKnY77yy+/4O2337Z4/7CwMGRmZqJdu3Z1Op+l+GCMf/Py8kKPHj3wxx9/GOyXkJAAhmHQunVro2Ns3boVDMMgMjJS2KbRaPDee+8hJiYGLi4u8PX1RY8ePbBmzRphnylTppi8x0OHDm2w11sfNLvNyVSoNMgurECB0tFXQggh94fMzEzh4y1btuDNN9/E5cuXhW38bCmeSqWyaIFVX19fq65DLBYjODjYqufUx65du9C2bVsUFBTgq6++whNPPIFTp04ZBGlubm7Izs7G4cOH0bNnT2H76tWrER4ebnC8RYsW4dtvv8UXX3yBuLg4FBUV4cSJE7h7967BfkOHDjUInABALpc3wCusP8okOZm/zmbiwY8PYNM1+tIQQog9BAcHC29eXl5gGEb4vKKiAt7e3ti6dSv69esHhUKB9evXIy8vD2PHjkV4eDhCQ0PRoUMHbNq0yeC41YfbIiMjsXTpUkybNg0eHh4IDw/Ht99+Kzxefbht3759YBgGu3fvRlxcHFxdXdGrVy+DAA4A3nnnHQQGBsLDwwMzZszAvHnz0LFjx1pft5+fH4KDgxETE4N3330XKpUKe/fuNdhHIpFg3LhxWL16tbDt1q1b2LdvH8aNG2ew7x9//IFZs2bhySefRLNmzdChQwdMnz4dc+fONdhPLpcb3PPg4GD4+PjUer2OQH+JnYybnOvnoNTSzBBCyL2PZVmUVaod8sayrM1ex2uvvYYXX3wRycnJGDJkCCoqKtClSxds27YNhw4dwtNPP42JEyfi6NGjNR7nk08+QVxcHE6fPo1Zs2bhf//7Hy5dulTjcxYsWIBPPvkEJ06cgEQiwbRp04THNmzYgHfffRcffPABTp48ifDwcKxcudKq16ZSqfDdd98BgMkM2fTp07FlyxaUlZUB4Ibhhg4diqCgIIP9goODsWfPHuTk5Fh1fmdGw21OxlXGfUmUtAQQIaQRKFdp0ObNfxxy7otLhgi/U+tr9uzZePzxxw22vfLKK9BqtSgqKkL79u3xzz//4Mcff0T37t3NHmf48OGYNWsWAC7wWr58Ofbt24eYmBizz3n33XfRt29fAMC8efMwYsQIVFRUQKFQ4PPPP8f06dMxdepUAMCbb76JnTt3oqSkpNbX1KtXL4hEIpSXl0Or1SIyMhKjR4822q9jx45o3rw5fvrpJ0ycOBEJCQlYtmwZrl+/brDfsmXLMGrUKAQHB6Nt27bo1asXHnnkEQwbNsxgvz///BPu7u4G21577TUsXLiw1mu2N8okORkhk0RBEiGEOI24uDiDzzUaDd5991107NgRUVFR8PT0xM6dO5Genl7jcdq3by98zA/r8UtoWPIcfpkN/jmXL19Gt27dDPav/rk5W7ZswenTp7Ft2za0aNEC33//vdk6qmnTpmHNmjXYv38/SkpKMHz4cKN92rRpg/Pnz+PIkSOYOnUq7ty5g5EjR2LGjBkG+/Xv3x9JSUkGb88995xF12xvDs8kffXVV/joo4+QmZmJtm3bYsWKFejTp4/JfX/55ResXLkSSUlJUCqVaNu2LRYtWoQhQ4YY7Pfzzz9j4cKFuHbtGpo3b453330Xjz32WJ3Pa09CJqnxrwtJCLkPuEjFuLhkSO07NtC5bcXNzc3g808++QTLly/HsmXL0KxZMwQFBWHu3LmorKys8TjVh7MYhql1IWD95/BNEvWfU71xoqXDjGFhYYiOjkZ0dDTc3d3xxBNP4OLFiwgMDDTad/z48Xj11VexaNEiTJo0CRKJ6fBBJBKha9eu6Nq1K+bMmYP169dj4sSJWLBgAZo1awaAu5ctWrSw6BodzaGZpC1btmD27NlYsGABTp8+jT59+mDYsGFmI/EDBw5g8ODB2L59O06ePIn+/ftj5MiROH36tLDP4cOHMWbMGEycOBFnzpzBxIkTMXr0aINxYmvPa09uuiCpkjJJhJBGgGEYuMokDnlryK7fBw8exCOPPIIJEyYgNjYWUVFRSElJabDzmdOqVSscO3bMYNuJEyesPk7fvn3Rrl07vPvuuyYf9/X1xcMPP4z9+/cb1ETVpk2bNgCA0tJSq6/JGTg0SFq2bBmmT5+OGTNmoHXr1lixYgXCwsLMFp2tWLECr776Krp27Yro6GgsXboU0dHRBr0dVqxYgcGDB2P+/PmIiYnB/PnzMXDgQKxYsaLO57UnV91wW6WWgUZru6JDQgghttOiRQskJibi0KFDuHz5MmbOnImsrCy7X8cLL7yAVatWYe3atUhJScE777yDs2fP1ilAfPnll/HNN9/g9u3bJh9PSEhAbm6u2fqpUaNGYfny5Th69CjS0tKwb98+PPfcc2jZsqXBc5RKJbKysgzecnNzrb5ee3DYcFtlZSVOnjyJefPmGWyPj4/HoUOHLDqGVqtFcXGxwRjq4cOHMWfOHIP9hgwZIgRJdT2vUqmEUlnVvKioqAgANyugvgvo6ZOLqgKj4rIKiEU0y60h8V87W34NiWl0r+3HkfdapVKBZVlotdpah5GcEX/Npt7rv54FCxbg+vXrGDZsGFxcXPD000/jkUceQWFhocF+/L0w97n+turnMnXu6tvGjh2La9eu4ZVXXkFFRQWefPJJTJ48GcePHzd7/829puHDhyMyMhLvvPMOvvzyS6NzyuVyyOVy4XN+WI//fPDgwdiyZQvee+89FBYWIjg4GP3798dbb70FkUgErVYLlmWxY8cOobaK16pVK1y8eNH0F0XvPpm7h6ZeI8uyUKlUEIsNh12t+blgWFvOkbRCRkYGmjRpgv/++w+9evUSti9duhRr16416gNhykcffYT3338fycnJwhiqTCZDQkKCQf+GjRs3YurUqVAqlXU+76JFi7B48WKj7Rs3bhRWG7YFlgXmHBGDBYMlXdTwooW0CSH3EIlEguDgYISFhUEmo19gjvDYY48hMDAQ33zzjaMvxWEqKytx8+ZNZGVlQa1WGzxWVlaGcePGobCwEJ6enjUex+GF26YKzixJE27atAmLFi3C77//blRkZskxrT3v/PnzDRpiFRUVISwsDPHx8bXeZGu9cXo3SpQaxPXojehgL5semxhSqVRITEzE4MGDLeqgS+qO7rX9OPJeV1RU4ObNm3B3d4dCobDruR2BZVkUFxfDw8OjQWugzCkrK8M333yD+Ph4iMVibN68Gfv27cM///xj879NjmbNva6oqICLiwsefPBBo+9DfiTIEg4Lkvz9/SEWi43GcLOzs40aVFW3ZcsWTJ8+HT/++CMGDRpk8FhwcHCNx6zrefk0Y3VSqdTmv4TcZBKUKDWo1DL0x8ROGuLrSEyje20/jrjXGo0GDMNAJBJBJGr8XWb4YR/+NdubWCzG33//jXfffRdKpRKtWrXCzz//jPj4eLtfS0Oz5l6LRCIwDGPyZ8CanwmHfQfLZDJ06dIFiYmJBtsTExMNhsGq27RpE6ZMmYKNGzdixIgRRo/37NnT6Jg7d+4UjlnX89qTq4wbPy2jKW6EEEJq4OLigl27diE/Px+lpaU4deqUUdNLUncOHW6bO3cuJk6ciLi4OPTs2RPffvst0tPTMXPmTADcENft27exbt06AFyANGnSJHz66afo0aOHkA1ycXGBlxc3LPXSSy/hwQcfxAcffIBHHnkEv//+O3bt2oV///3X4vM6Gj/DraxSXcuehBBCCGkoDg2SxowZg7y8PCxZsgSZmZlo164dtm/fjoiICADcysz6vYu++eYbqNVqPPfccwbdOSdPnoyEhAQAXJv1zZs344033sDChQvRvHlzbNmyxaBNfG3ndTS+oSRlkgghhBDHcXjh9qxZs4R1bKrjAx/evn37LDrmqFGjMGrUqDqf19H44bZSCpIIIYQQh2n8VXX3IDeqSSKEEEIcjoIkJyQMtympJokQQghxFAqSnBDNbiOEEEIcj4IkJ+RGNUmEEHLP6devH2bPni18HhkZabBuqCkMw+C3336r97ltdRxiiIIkJ0SZJEIIsZ+RI0caNSbmHT58GAzD4NSpU1Yf9/jx43jmmWfqe3kGFi1ahI4dOxptz8zMxLBhw2x6ruoSEhLAMIzwFhQUhJEjR+LChQsG+02ZMgUMw5hsqzNr1iwwDIMpU6YI27Kzs/Hss88iPDwccrkcwcHBGDJkCA4fPizsExkZCbFYDB8fH4jFYuEa3n///QZ7vQAFSU7JVc63AKCaJEIIaWjTp0/Hnj17kJaWZvTY6tWr0bFjR3Tu3Nnq4wYEBNh0bc+aBAcHm1wVwtY8PT2RmZmJjIwM/PXXXygtLcWIESNQWVlpsF9YWBg2b96M8vJyYVtFRQU2bdqE8PBwg32feOIJnDlzBmvXrsWVK1ewbds29OvXD/n5+Qb7LV68GJcuXcLt27eRmZmJzMxMvPDCCw33YkFBklOiFgCEEGI/Dz30EAIDA43azpSVlQnLYOXl5WHs2LFo2rQpXF1dERsbi02bNtV43OrDbSkpKcJaYm3atDFa+QEAXnvtNbRs2RKurq6IiorCwoULhVXrExISsHjxYpw5c0bIpPDXXH247dy5cxgwYABcXFzg5+eHZ555BiUlJcLjU6ZMwaOPPoqPP/4YISEh8PPzw3PPPSecyxyGYRAcHIyQkBDExcVhzpw5SEtLM1ocvnPnzggPD8cvv/wibPvll18QFhaGTp06CdsKCgrw77//4oMPPkD//v0RERGBbt26Yf78+Uaranh4eCAoKAjBwcHCm5ubW43XW18UJDkhagFACGk0WBaoLHXMG8tadIkSiQSTJk1CQkICWL3n/Pjjj6isrMT48eNRUVGBLl264M8//8T58+fxzDPPYOLEiTh69KhF59BqtXj88cchFotx5MgRfP3113jttdeM9vPw8EBCQgIuXryITz/9FN999x2WL18OgGuE/PLLL6Nt27ZCJmXMmDFGxygrK8PQoUPh4+OD48eP48cff8SuXbvw/PPPG+y3d+9eXLt2DXv37sXatWuRkJBgFCjWpKCgABs3bgRgej20qVOnYs2aNcLnq1evxrRp0wz2cXd3h7u7O3777TcolUqLz20vDm8mSYxV1STRcBsh5B6nKgOWhjrm3K9nADLLMg3Tpk3DRx99hH379qF///4AuD/qjz/+OHx8fODj44NXXnlF2P+FF17Ajh078NNPP2HhwoW1Hn/Xrl1ITk5GamoqmjZtCgBYunSpUR3RG2+8IXwcGRmJl19+GVu2bMGrr74KFxcXuLu7QyKRIDg42Oy5NmzYgPLycqxbt07ItHzxxRcYOXIkPvjgA2Exdx8fH3zxxRcQi8WIiYnBiBEjsHv3bjz99NNmj11YWAh3d3ewLIuysjIAwMMPP4yYmBijfSdOnIj58+cjNTUVDMPgv//+w+bNmw0aQ0skEiQkJODpp5/G119/jc6dO6Nv37546qmn0L59e4PjzZs3z+he//nnn+jXr5/Z660vCpKcUFWfJMokEUKIPcTExKBXr15YvXo1+vfvj2vXruHgwYPYuXMnAECj0eD999/Hli1bcPv2bSiVSiiVSotrjpKTkxEeHi4ESAC3IHt1P/30E1asWIGrV6+ipKQEarUanp6eVr2W5ORkdOjQwWAoqnfv3tBqtbh8+bIQJLVt2xZisVjYJyQkBOfOnavx2B4eHjh16hTUajX279+Pjz76CF9//bXJff39/TFixAisXbsWLMtixIgR8Pf3N9rviSeewIgRI3Dw4EEcPnwYO3bswIcffojvv//eoMD7lVdewRNPPAF3d3eIRNxAWJMmTSy+L3VBQZITotlthJBGQ+rKZXQcdW4rTJ8+Hc8//zy+/PJLrFmzBhERERg4cCAA4JNPPsHy5cuxYsUKxMbGws3NDbNnzzYqWDaHNTH0xzCMwedHjhzBU089hcWLF2PIkCHw8vLC5s2b8cknn1j1OliWNTq2qXNWHyJjGAZarbbGY4tEIrRo0QIAF1hmZWVhzJgxOHDggMn9p02bJgzzffnll2aPq1AoMHjwYAwePBhvvvkmZsyYgbfeessgSPL390dUVBQ8PT2FIKmhUU2SE3KTU+E2IaSRYBhuyMsRb2YCBXNGjx4NsViMjRs3Yu3atZg6daoQVBw8eBCPPPIIJkyYgA4dOiAqKgopKSkWH7tNmzZIT09HRkZVwKg/xR0A/vvvP0RERGDBggWIi4tDdHS00Yw7mUwGjabmvw1t2rRBUlISSktLDY4tEonQsmVLi6/ZEnPmzMGZM2fw66+/mnx86NChqKysRGVlJYYMGWLxcdu0aWNw/Y5CQZITEobbKtUm//sghBBie+7u7hgzZgxef/11ZGRkGGQxWrRogcTERBw6dAjJycl49tlnkZWVZfGxBw0ahFatWmHSpEk4c+YMDh48iAULFhjs06JFC6Snp2Pz5s24du0aPvvsM6PgIzIyEjdu3EBSUhJyc3NNFjuPHz8eCoUCkydPxvnz57F371688MILmDhxojDUZiuenp5C1sfU3yuxWIzk5GQkJycbDO3x8vLyMGDAAKxfvx5nz57FjRs38OOPP+LDDz/EI488YrBvcXEx7ty5g6ysLOGtqKjIpq+nOgqSnBA/3KZlAaW65tQnIYQQ25k+fTru3r2LQYMGGfTzWbhwITp37owhQ4agX79+CA4OxqOPPmrxcUUiEX799VcolUp069YNM2bMwLvvvmuwzyOPPII5c+bg+eefR8eOHXHo0CGjQuUnnngCQ4cORf/+/REQEGCyDYGrqyv++ecf5Ofno2vXrhg1ahQGDhyIL774wrqbYaGXXnoJycnJ+PHHH00+7unpabauyt3dHd27d8fy5cvx4IMPol27dli4cCGefvppo+t96623EBMTgyZNmiAkJAQhISF49dVXbf569DEspSrqpKioCF5eXigsLLS6qK42FcpKxLzF9c84+cYg+Lk3fIOw+5VKpcL27dsxfPhwk1NYie3QvbYfR97riooK3LhxA82aNYNCobDruR1Bq9WiqKjIrnUy9ytr7nVN34fW/P2mr6gTEosYyERc7ErF24QQQohjUJDkpHQjbiilXkmEEEKIQ1CQ5KTkuq9MKfVKIoQQQhyCgiQnpesCQF23CSGEEAehIMlJ8UESZZIIIfcimhNEHMlW338UJDkpuVC4TZkkQsi9g++FY2knakIaAr+uXH1nd9KyJE6qqnCbMkmEkHuHRCKBq6srcnJyIJVKG/20eK1Wi8rKSlRUVDT61+poltxrfuHd7OxseHt7m2xgaQ0KkpxU1XAbZZIIIfcOhmEQEhKCGzduGC2p0RixLIvy8nK4uLiYXS+N2IY199rb2xvBwcH1PicFSU6Kn91WRkESIeQeI5PJEB0dfV8MualUKhw4cAAPPvggNUltYJbea6lUWu8MEo+CJCclp+E2Qsg9TCQS3Rcdt8ViMdRqNRQKBQVJDcwR95oGUJ2UXEyF24QQQogjUZDkpGTUTJIQQghxKAqSnBQ1kySEEEIci4IkJ0XNJAkhhBDHoiDJSQmz2yiTRAghhDgEBUlOima3EUIIIY5FQZKTkvGz26hPEiGEEOIQFCQ5KQVlkgghhBCHoiDJSVFNEiGEEOJYDg+SvvrqKzRr1gwKhQJdunTBwYMHze6bmZmJcePGoVWrVhCJRJg9e7bRPv369QPDMEZvI0aMEPZZtGiR0eO2WOPFlvgFblUaFpVqrWMvhhBCCLkPOTRI2rJlC2bPno0FCxbg9OnT6NOnD4YNG4b09HST+yuVSgQEBGDBggXo0KGDyX1++eUXZGZmCm/nz5+HWCzGk08+abBf27ZtDfY7d+6czV9ffcj1vjKUTSKEEELsz6FB0rJlyzB9+nTMmDEDrVu3xooVKxAWFoaVK1ea3D8yMhKffvopJk2aBC8vL5P7+Pr6Ijg4WHhLTEyEq6urUZAkkUgM9gsICLD566sPsQiQSbgvD9UlEUIIIfbnsAVuKysrcfLkScybN89ge3x8PA4dOmSz86xatQpPPfUU3NzcDLanpKQgNDQUcrkc3bt3x9KlSxEVFWX2OEqlEkqlUvi8qKgIALcqsUqlstn18scEAFepGJVqLQpLKxDoRmsRNwT+Xtv6a0iM0b22H7rX9kP32n5sda+teb7D/vLm5uZCo9EgKCjIYHtQUBCysrJsco5jx47h/PnzWLVqlcH27t27Y926dWjZsiXu3LmDd955B7169cKFCxfg5+dn8ljvvfceFi9ebLR9586dcHV1tcn1VifSVgJgsGvvAaR4NMgpiE5iYqKjL+G+Qffafuhe2w/da/up770uKyuzeF+HpycYhjH4nGVZo211tWrVKrRr1w7dunUz2D5s2DDh49jYWPTs2RPNmzfH2rVrMXfuXJPHmj9/vsFjRUVFCAsLQ3x8PDw9PW1yvTyVSoXExET4erojP6cUHbt2R88o08EbqR/+Xg8ePBhSqdTRl9Oo0b22H7rX9kP32n5sda/5kSBLOCxI8vf3h1gsNsoaZWdnG2WX6qKsrAybN2/GkiVLat3Xzc0NsbGxSElJMbuPXC6HXC432i6VShvsB8NNzn15lBqGfvgaWEN+HYkhutf2Q/fafuhe209977U1z3VY4bZMJkOXLl2M0maJiYno1atXvY+/detWKJVKTJgwodZ9lUolkpOTERISUu/z2pKbrg8AzW4jhBBC7M+hw21z587FxIkTERcXh549e+Lbb79Feno6Zs6cCYAb4rp9+zbWrVsnPCcpKQkAUFJSgpycHCQlJUEmk6FNmzYGx161ahUeffRRkzVGr7zyCkaOHInw8HBkZ2fjnXfeQVFRESZPntxwL7YOXHVBUqmSZrcRQggh9ubQIGnMmDHIy8vDkiVLkJmZiXbt2mH79u2IiIgAwDWPrN4zqVOnTsLHJ0+exMaNGxEREYHU1FRh+5UrV/Dvv/9i586dJs9769YtjB07Frm5uQgICECPHj1w5MgR4bzOwlXGfXkok0QIIYTYn8MLt2fNmoVZs2aZfCwhIcFoG8uytR6zZcuWNe63efNmi6/PkVzllEkihBBCHMXhy5IQ86gmiRBCCHEcCpKcmFCTREESIYQQYncUJDkxoSaJhtsIIYQQu6MgyYlRJokQQghxHAqSnFhVTRJlkgghhBB7oyDJifHDbaVKyiQRQggh9kZBkhPjWwBQJokQQgixPwqSnBjVJBFCCCGOQ0GSExNqkmh2GyGEEGJ3FCQ5MaEmiTJJhBBCiN1RkOTE+OG2CpUWGm3ty7EQQgghxHYoSHJi/HAbQEuTEEIIIfZGQZITk0lEEIsYADTDjRBCCLE3CpKcGMMwVTPcqFcSIYQQYlcUJDk5N379NsokEUIIIXZFQZKT4xtKVs8kVaq1YFkq5iaEEEIaCgVJTs7NRBuAO0UV6PruLszdesZRl0UIIYQ0ehQkObmqmqSq4bb9l3NQWK7C/is5jrosQgghpNGjIMnJucn5mqSqTNLpmwUAgPzSSlSoqFaJEEIIaQgUJDk5U5mk0+l3hY+zCivsfk2EEELI/YCCJCdXNbuNyySVKNW4cqdYeDyTgiRCCCGkQVCQ5OSE2W26FgBnbxVAf4WSzMJyR1wWIYQQ0uhRkOTkhEySrgVAkq4eiXc/ZpKKKlT47sB1ChAJIYQ0KAqSnFz1TNLp9AIAgIeCC57ux0Bhw5F0vLs9GZ/tTnH0pRBCCGnEKEhycvo1SSzLCkHS4DZBAO7Pwu3kzCIAwLXsUgdfCSGEkMaMgiQnpz+77dbdcuSWKCEVMxjUmguSMgruvyDpanYJACA9v8zBV0IIIaQxoyDJyen3SeL7I7UJ8USknxsAIKvItkFSUYUKH+64hJtOGoBotSyu53JBUlZRBfWJIoQQ0mAoSHJy+pmkJN1QW6dwH4R6KwDYvqHk+iNp+GrfNSzadsFmx7Sl2wXlqFBphc9v3XXOYI4QQsi9j4IkJ+dukEnimkh2CveGl4sUCin35bNlXRJf53Pwaq7RorrOICW72ODztDwKkgghhDQMCpKcnKuucPtumQoXbnMFyx3DvMEwDEK9XAAAGTac4ZaWxwVJlWotDtRzbbjfk27jUlaRLS5LwNcj8aguiRBCSEOhIMnJuelaABSWq1Cp0cLXTYZwX1cAQLAXN+Rmy0xSql5mZufFO3U+zqn0u3hpcxJmrT9li8sS8EGSRMQAoCCJEEJIw6EgycnxmSReJ10WCQBCdJkkWzWULFGqkVuiFD7fnXwHKo22hmeYdzmLGxa7nltq07ohPkjqGukLAEin4TZCCCENhIIkJ8dnknidwr2Fj0N0mSRbNZTkh9p8XKXwd5ehqEKNYzfy63Ss1LyqHkZHrtftGNWxLCsESQNiAgFQJokQQkjDoSDJySkkYugSRwCAjmE+wschuhlumTbqlcQXQTfzdxP6MO28kFWnY+lneI5cz6v/xQHIKVGiqEINEQP0bRXAnSe/DFr9xewIIYQQG3F4kPTVV1+hWbNmUCgU6NKlCw4ePGh238zMTIwbNw6tWrWCSCTC7NmzjfZJSEgAwzBGbxUVhoGENed1JJGIgauUyyYxDNA+zEt4rCqTZJsgic/+RPq5Ib6tLki6eAcsa30Qol/bdPiabYKkq3e4LFKYryua+btBLGKgVGuRozdESAghhNiKQ4OkLVu2YPbs2ViwYAFOnz6NPn36YNiwYUhPTze5v1KpREBAABYsWIAOHTqYPa6npycyMzMN3hQKRZ3P62iuujYA0YHu8FRIhe1VNUk2Gm7L5QKbCD839GruD1eZGJmFFTh/27oZaizLIl1vuO12QblNmlNezeGCpBYB7pCKRUKvKBpyI4QQ0hAcGiQtW7YM06dPx4wZM9C6dWusWLECYWFhWLlypcn9IyMj8emnn2LSpEnw8vIyuQ8AMAyD4OBgg7f6nNfR3HQNJTvpDbUBVZmku2UqmzSUFDJJ/q5QSMXopxvS2nnRuiG33JJKlFZqwDBAbBPu63TYBkNufD1SiyB3ABBm+VGvJEIIIQ3BYUFSZWUlTp48ifj4eIPt8fHxOHToUL2OXVJSgoiICDRt2hQPPfQQTp8+bZfzNhR+hpt+0TYAeLlI4aIbirPFkBsfbEToljyJb8MFlzsvWNcKgC8AD/VywYMt/QHYpi5JCJICDIMkyiQRQghpCJLad2kYubm50Gg0CAoKMtgeFBSErKy6FQsDQExMDBISEhAbG4uioiJ8+umn6N27N86cOYPo6Og6n1epVEKprKp9KSrihqBUKhVUKlWdr9cU/nj8+9FdQrHtrAj9W/oZnSvES47ruWW4lVeCpl6yOp+zvFIjrAPXxFMGlUqFB5r7QCJicPlOMa5mFSLCz9WiY13XdcUO93VBnC6wO3ItD5WVlUL7grrgg6RIXwVUKhWa6DJpqTkldf4aVL/XpOHQvbYfutf2Q/fafmx1r615vsOCJF71P5osy9brD2mPHj3Qo0cP4fPevXujc+fO+Pzzz/HZZ5/V+bzvvfceFi9ebLR9586dcHW1LHiwVmJiIgDAB8DkJsCR/buM9pGoRABE2HHwKPIv1X2WV0YpAEjgKmZxaF+isD3KQ4QrhSJ8/ut+DAi17Pi707lrYkpykZOcDTEjRkZhBX749W/4K2p9ukllaiC7mPt2vXb6EDLOAbl5DAAxzl7PwPbtN+t2YB3+XpOGR/fafuhe2w/da/up770uK7N89MFhQZK/vz/EYrFR9iY7O9soy1MfIpEIXbt2RUpKSr3OO3/+fMydO1f4vKioCGFhYYiPj4enp6fNrhfgotzExEQMHjwYUqm0xn33V5zHldMZCIpsheF9o+p8zp0X7wBnz6BFsBeGD68KMvP90rH4z0u4xfph+PBuFh1r149ngdtZ6N2xFR7r0wxbso7hZHoBXCI6YHiXJnW6vtPpBcDxYwjykOOJh7mh0oiMIiRcOYISyDF8eL86Hdeae03qh+61/dC9th+61/Zjq3vNjwRZwmFBkkwmQ5cuXZCYmIjHHntM2J6YmIhHHnnEZudhWRZJSUmIjY2t13nlcjnkcrnRdqlU2mA/GJYcu4kPl8W6U1xZr+u4VcANJUb6uxscZ0i7UCz+8xJO3SxAQYUWAR7G96C69LvcsF3zQA9IpVL0auGPk+kFOJ5WgHE9Iut0fan53DFbBFVdX7NALjjNLalEpZaBm7zu384N+XUkhuhe2w/da/uhe20/9b3X1jzXobPb5s6di++//x6rV69GcnIy5syZg/T0dMycORMAl72ZNGmSwXOSkpKQlJSEkpIS5OTkICkpCRcvXhQeX7x4Mf755x9cv34dSUlJmD59OpKSkoRjWnLeewnfBqC+67fxfY0iq9UdhXq7ILaJF1gW2Hc526Jj8YXbfAF4jyg/AFzxdl16LgGG0/95Xi5SeLlw3+w3bbj0CSGEEAI4uCZpzJgxyMvLw5IlS5CZmYl27dph+/btiIiIAMA1j6zeu6hTp07CxydPnsTGjRsRERGB1NRUAEBBQQGeeeYZZGVlwcvLC506dcKBAwfQrVs3i897L7FVQ8nqgY2+ATGBOHe7EHsuZePJuLAaj1NYpkJBGVcUx88+6xzuA5lYhMzCCqTllSHS3/gctama/u9hsD3CzxVnbxUiLa8MMcG2HfYkhBByf3N44fasWbMwa9Ysk48lJCQYbastE7F8+XIsX768Xue9lwhLk9SzoSQ//T/S37gIfWDrQHy6OwUHruSgUq2FTGI+AZmWzwVbAR5yYfjLRSZGxzBvHEvNx5HrefULkvQySQDXffvsrUKbNKskhBBC9Dl8WRJSPyGe3HBbfRpKVqg0yNAFWaYySe1CveDvLkdppabWBW+FXku+hsFWj+ZVQ251uT5+OK1FoGGQFEG9kgghhDQQCpLucZ4uErjK6tdQ8tbdMrAs4C6XwM/NuNeSSMRgQAzXfXvPpZrrkswN2/WI8gXAdd62ti7pWk4JWJarQfJ3N7w+6rpNCCGkoVCQdI9jGAbBXvUbcksV1mxzNdsrakAM1x5h96WaF7w1VwDeOdwHMokId4qUBovfWkIYagt0N7o+Pkii4TZCCCG2RkFSIxDKL3RbULdMkrBmm4mhNt4D0f6QiUVIyyvD9dxSs/ul6wKg8GpBkkLK1SUBwMm0u1Zd3zUz9Uj657l5twwabd2baRJCCCHVUZDUCPCZJH5ZEWtVrdlmvnO4u1yC7rohs701DLnVFHDxWZ/sYuuuk5/+Hx1kHCSFeLlAKmag0rB1fv2EEEKIKRQkNQKhuiApo6COw20WZJIArhUAAOxONh0klVWqkV2sNHssvt7pbmmlVdfHD7c1DzQOksQiBk11DTXTqS6JEEKIDVGQ1AgE17OhZNX0f8uCpOOp+SgsN14gkJ9h5uUihZercUdTH12QlGdFkFSh0gg1U6aG2wCuDQB3fvPDgIQQQoi1KEhqBPheSRl1CJIq1Vrcumu62Lq6CD83tAh0h1rL4mBKjtHjfDBj7ji+dcgkJV68g0qNFk28XdDUx8XkPuG+3HZqA0AIIcSWKEhqBPiu21l1mN12u6AcWhZwkYotWpeNzyaZagXAZ3LCzQzb+bpyQVK+FUHSL6duAQAe69TE7My7CF/ufNQGgBBCiC1RkNQI8Ou33S1TobzSuoaSqUJfI/PT//XxQdK+yzlGs8nMTf/n+bpbN9yWU6zEgZRcAMBjnZuY3S+M2gAQQghpABQkNQKeiqqGktbO8ErLtaxom9clwgeeCgnySyuRdLPA8Fg1rP8GVGWSLB1u23YmAxoti45h3mhuph6JOx913SaEEGJ7FCQ1AgzDVC10a+UMNz77E2FizTZTpGIR+rbiZ7ndMXistlYCfCaptFJj0RIq/FDb4zVkkYCqTNLdMhWKKowLygkhhJC6oCCpkeCH3G5bGSSlWTj9X9+g1lyQlHAoFSl3igFwBeB8CwJzQZKHXAKpmBvSu1tWczbpclYxLmQUQSpm8FD70Br3dZdLEOTJ1VMdvmb92nCEEEKIKRQkNRLNdNP3+Z5ClrKkkWR1I2JD0Ku5H8oqNZi5/iRKlGrculsGLQu4ysQIcDddAM4wDHx0Q255JTUHSb+c5rJI/VoFCrPiavJ456YAgIT/Ui1+HYQQQkhNKEhqJFqHeAIALmYWWfycazklQh2PNZkkiViEz8Z2QrCnAtdySvHaT2eFAvBw35oLwIU2ADVkkjRaFr+fzgAAPFHLUBtvQo8IiEUMDl/Pw6Usy+8BIYQQYg4FSY1Em1BdkJRRVOMCtLxSpRozfzgJtZZFzyg/oabJUv7ucnw5vjOkYgZ/ncvER/9cAVB7sMUHSTW1ATh8LQ9ZRRXwcpGiv242XW2aeLtgSFtuEd61h1Iteg4hhBBSEwqSGolWQR4QMdz0+hzd0iDmsCyL134+i5TsEgR6yPHp2I4WTf+vrkuED94Y0QYAkKzLYNU2bOdjQZDED7U91D4EconY4uuZ2rsZ9/xTt61e+oQQQgipjoKkRsJFJkaUbpp8bUNuq/9LxZ9nMyERMfhqfGcEeliXRdI3qWcEHulYVVhtbvo/z6+WIKmsUo0d57MAVNUZWSouwgdtQz2hVGux+fhNq55LCCGEVEdBUiNiSV3SsRv5WLo9GQDwxojWiIv0rdc5GYbBe4/HIibYAwDQMcy7xv1rG27750IWyio1iPRzRefwmo9l6lr4bNIPh1Oh1mitej4hhBCij4KkRqRNSFVdkinZRRV4buMpaLQsHu0Yism9Im1yXleZBL/O6o3dL/cVaqPMqS1IOnItHwAwPDakTkOAD7UPgZ+bDBmFFdh58U7tTyCEEELMoCCpERGKt81kkr7YexU5xUrEBHtg6eOxdQpCzHGRiWvsis3jgyRzS5Nk6jqG8y0NrKWQijG+ezgAagdACCGkfihIakT4TNKN3FKUVaqNHt9/JQcA8HJ8K7jKJHa9Nl5tS5PwHcP55ph1Mb5HBCQiBsdS83H+dmGdj0MIIeT+RkFSIxLgIYe/uxwsy3Ws1peWV4q0vDJIRAx6Nvdz0BVWLU1ibrgtq5DLJIV4172YPMhTgRHtQwAA64+k1fk4hBBC7m8UJDUy5obcDqbkAgA6h/vAXe6YLBKgl0kqq4RWa9jPqbhChWIllwEL9qx7kARAmHF3Kv1uvY5DCCHk/kVBUiPDD7klGwVJ3FBbn2h/u1+TPr5PkpYFCssNF6Pls0ieCgnc6hnIRQdys+1u5JbSLDdCCCF1QkFSI6PfeZun1mhxSLfwa5+WAQ65Lp5ULIKHgguA8qstTZKpC5JCvetej8Rr4u0CF6kYKg0rLL1CCCGEWIOCpEamTQiXQbmUVSwMZ525VYjiCjW8XKSIbeLlyMsDYL6hZGYhV7QdbOUSKaaIRAyaB3Iz5FKsXPSXEEIIAShIanSa+btDIRWhrFKDNF0GhR9q693CD2KR7ab915W5pUn4TFJ9Zrbpa6FrSXCVgiRCCCF1QEFSIyMWMWgVbDjkxhdt94l27FAbz2wmqYAPkuqfSQKA6CAuq0ZBEiGEkLqgIKkREjpvZxaiqEKFpJsFAIAHWji2aJtnrus230jSFsNtAITmlhQkEUIIqQsKkhohvi4pObMYh6/lQaNl0czfDWG+rg6+Mo654bYsXU1SqK2G2wK5IOlaTolRuwFCCCGkNhQkNUL6M9ycZeq/vtqG22yVSYrwc4VUzKCsUoMMXQBGCCGEWIqCpEaoVbAnGAbIKqrAjvNZAJynHgkAfFyNgyT9RpK2qkmSikWI9ONmuNGQGyGEEGtRkNQIucslQnCQW1IJiYhBjyhfB19VFT8TS5PYspGkvuggqksihBBSNxQkNVKtdXVJANAp3BseCqkDr8aQqUySraf/86gNACGEkLpyeJD01VdfoVmzZlAoFOjSpQsOHjxodt/MzEyMGzcOrVq1gkgkwuzZs432+e6779CnTx/4+PjAx8cHgwYNwrFjxwz2WbRoERiGMXgLDg629UtzKH6GG+BcQ20A4OcmB2A6k1SfhW1NaR5IQRIhhJC6cWiQtGXLFsyePRsLFizA6dOn0adPHwwbNgzp6ekm91cqlQgICMCCBQvQoUMHk/vs27cPY8eOxd69e3H48GGEh4cjPj4et2/fNtivbdu2yMzMFN7OnTtn89fnSHzxNuBcRdsA4OPGZbXKVRqUV2oAQCistlU9Eo9fwy0luwQsSzPcCCGEWM6hQdKyZcswffp0zJgxA61bt8aKFSsQFhaGlStXmtw/MjISn376KSZNmgQvL9PLa2zYsAGzZs1Cx44dERMTg++++w5arRa7d+822E8ikSA4OFh4CwhwrmxLfbVv6g0XqRihXgq0b+rt6Msx4C6XQCbmvvX49dv4TFKwp22H26IC3MAw3GK6uSWVtT+BEEII0bFdhayVKisrcfLkScybN89ge3x8PA4dOmSz85SVlUGlUsHX17BwOSUlBaGhoZDL5ejevTuWLl2KqKgos8dRKpVQKpXC50VFXDdrlUoFlUpl7ml1wh+vPsf1kovw6/96wFUmhlajhlZjq6uzDR9XKe4UK5FTWIZANwlu3+WWUAn0kNr0fooBNPV2wc275bicWQBvheH3gS3uNbEM3Wv7oXttP3Sv7cdW99qa5zssSMrNzYVGo0FQUJDB9qCgIGRlZdnsPPPmzUOTJk0waNAgYVv37t2xbt06tGzZEnfu3ME777yDXr164cKFC/Dz8zN5nPfeew+LFy822r5z5064ujZMk8bExESbHOe0TY5iWxKNGACDHfv+Q5o3i5Tb3Ofpl85ie9YZm57LEyIAIvy+9yjykk0PudnqXpPa0b22H7rX9kP32n7qe6/Lysos3tdhQRKPYQwXXGVZ1mhbXX344YfYtGkT9u3bB4WiqtZl2LBhwsexsbHo2bMnmjdvjrVr12Lu3LkmjzV//nyDx4qKihAWFob4+Hh4enqafE5dqVQqJCYmYvDgwZBKnWdWmi1tuXMCt6/no0WbDhjeMRRvnN4DQI2HB/UROmXbyjnxFVz4NxWKwEgMH97a4LH74V47C7rX9kP32n7oXtuPre41PxJkCYcFSf7+/hCLxUZZo+zsbKPsUl18/PHHWLp0KXbt2oX27dvXuK+bmxtiY2ORkpJidh+5XA65XG60XSqVNtgPRkMe29H8PLigtVCphVLLoLiCayQZ5u8BqdS235YtdQv+Xs8rM3s/G/O9djZ0r+2H7rX90L22n/rea2ue67DCbZlMhi5duhilzRITE9GrV696Hfujjz7C22+/jR07diAuLq7W/ZVKJZKTkxESElKv8xLLVS1NohTWbPNQSOBuw0aSvGhdZirljnEbgAsZRSihUgJCCCEmOHS4be7cuZg4cSLi4uLQs2dPfPvtt0hPT8fMmTMBcENct2/fxrp164TnJCUlAQBKSkqQk5ODpKQkyGQytGnTBgA3xLZw4UJs3LgRkZGRQqbK3d0d7u7cH8tXXnkFI0eORHh4OLKzs/HOO++gqKgIkydPtuOrv79VNZRUIUO3ZputFratju+VlF2sRFGFCp66xpo7L2ThmR9OItpThNENcmZCCCH3MocGSWPGjEFeXh6WLFmCzMxMtGvXDtu3b0dERAQArnlk9Z5JnTp1Ej4+efIkNm7ciIiICKSmpgLgmlNWVlZi1KhRBs976623sGjRIgDArVu3MHbsWOTm5iIgIAA9evTAkSNHhPOShufrrp9Jsu3CttV5KqQI8pTjTpESV7NL0DncB6VKNRZtuwAASC9lqIcSIYQQIw4v3J41axZmzZpl8rGEhASjbbX9MeODpZps3rzZkksjDchXl0m6W6pqsEaS+qIDPbgg6Q4XJH22OwUZuuBMqWFQWK5GgEzWYOcnhBBy73H4siTk/uSrq0nK08sk2XrdNn38jLmrOSW4nFWMVf/eAABIRNxMylt3yxvs3IQQQu5NFCQRh+CDpLtlKr3FbRsuk8TXJV25U4w3fjsHtZbFkLZBaNeEm/l2867lfTMIIYTcH6wKko4dOwaNpqp1c/WhL6VSia1bt9rmykijVhUkVeJ2AZfFaaiaJKBqhtuBKzk4nnoXrjIx3hrZFk29uezVrQLKJBFCCDFkVZDUs2dP5OXlCZ97eXnh+vXrwucFBQUYO3as7a6ONFrertwMM5YFbuSWAgBCvRsuSOKH27S6uH72oGiEersgzEcXJNFwGyGEkGqsCpKqZ45MFVHTLCFiCalYBC8XLlDS6CKX4AasSfJzk8FHF5i1CvLA1N7NAABNKUgihBBihs1rkmy1pAhp/PghN6DhGknyGIZB35YBUEhFWPp4O0jF3Lc+BUmEEELMcXgLAHL/8nWTCUNtDVm0zVs2uiOWVKqFZpKAXpBUUAGtloVIREE+IYQQjtVB0sWLF4Uu1izL4tKlSygp4ZZ7yM3Nte3VkUaN77oNNOxQG08kYgwCJIALzhiwqFRrkVOiRJBnwwdrhBBC7g1WB0kDBw40qDt66KGHAHDDGSzL0nAbsZif3nBbqB0ySaZIxSL4yIF8JXAzv4yCJEIIIQKrgqQbN2401HWQ+xC/NAnQsNP/a70OOYt8JYNbd8sRF+mwyyCEEOJkrAqSaG0zYku+esNt9qhJMnsdcu79zXxqKEkIIaSKVbPb8vPzcevWLYNtFy5cwNSpUzF69Ghs3LjRphdHGjf92W0NuSRJbfzk3PAxdd0mhBCiz6og6bnnnsOyZcuEz7Ozs9GnTx8cP34cSqUSU6ZMwQ8//GDziySNk2GQ5LhMkp/u1DfzqQ0AIYSQKlYFSUeOHMHDDz8sfL5u3Tr4+voiKSkJv//+O5YuXYovv/zS5hdJGif9IMnRNUkAZZIIIYQYsipIysrKQrNmzYTP9+zZg8ceewwSCVfa9PDDDyMlJcW2V0garSY+LhCLGIR4KeBRbWq+PfnpapIyCyug1mgddh2EEEKci1VBkqenJwoKCoTPjx07hh49egifMwwDpVJps4sjjZu/uxxbnumBH6Z3c+h1eMoAmUQEjZZFZmGFQ6+FEEKI87AqSOrWrRs+++wzaLVa/PTTTyguLsaAAQOEx69cuYKwsDCbXyRpvOIifdEi0MOh1yBigCa64T6a4UYIIYRnVZD09ttv4/fff4eLiwvGjBmDV199FT4+PsLjmzdvRt++fW1+kYQ0NH55EqpLIoQQwrOqT1LHjh2RnJyMQ4cOITg4GN27dzd4/KmnnkKbNm1seoGE2IMQJNEMN0IIITpWL0sSEBCARx55xORjI0aMqPcFEeIITbwpk0QIIcSQVUHSunXrLNpv0qRJdboYQhwlTJdJunWXMkmEEEI4VgVJU6ZMgbu7OyQSicEit/oYhqEgidxzqobbKJNECCGEY1WQ1Lp1a9y5cwcTJkzAtGnT0L59+4a6LkLsig+SsouVqFBpoJCKHXxFhBBCHM2q2W0XLlzAX3/9hfLycjz44IOIi4vDypUrUVRU1FDXR4hd+LhK4SbjAiMaciOEEAJYGSQBQPfu3fHNN98gMzMTL774IrZu3YqQkBCMHz+eGkmSexbDMAjzdQVAxduEEEI4VgdJPBcXF0yaNAmLFy9Gt27dsHnzZpSV0R8Xcu9q6sMFSbeoLokQQgjqGCTdvn0bS5cuRXR0NJ566il07doVFy5cMGgsSci9JsyXbwNAw22EEEKsLNzeunUr1qxZg/3792PIkCH45JNPMGLECIjFVORK7n1hukwSzXAjhBACWBkkPfXUUwgPD8ecOXMQFBSE1NRUfPnll0b7vfjiiza7QELshWqSCCGE6LMqSAoPDwfDMNi4caPZfRiGoSCJ3JNoaRJCCCH6rAqSUlNTa93n9u3bdb0WQhyKzyQVlqtQVKGCp0Lq4CsihBDiSHWe3VZdVlYWXnzxRbRo0cJWhyTErtzlEvi4coHRLcomEULIfc+qIKmgoADjx49HQEAAQkND8dlnn0Gr1eLNN99EVFQUDh8+jNWrVzfUtRLS4KguiRBCCM+q4bbXX38dBw4cwOTJk7Fjxw7MmTMHO3bsQEVFBf7++2/07du3oa6TELsI83XF2VuFSM+jIIkQQu53VmWS/vrrL6xZswYff/wxtm3bBpZl0bJlS+zZs6fOAdJXX32FZs2aQaFQoEuXLjh48KDZfTMzMzFu3Di0atUKIpEIs2fPNrnfzz//jDZt2kAul6NNmzb49ddf63Vecv9o5ucGALiRV+rgKyGEEOJoVgVJGRkZaNOmDQAgKioKCoUCM2bMqPPJt2zZgtmzZ2PBggU4ffo0+vTpg2HDhiE9Pd3k/kqlEgEBAViwYAE6dOhgcp/Dhw9jzJgxmDhxIs6cOYOJEydi9OjROHr0aJ3PS+4fEX7ccFsaBUmEEHLfsypI0mq1kEqrZvyIxWK4ubnV+eTLli3D9OnTMWPGDLRu3RorVqxAWFgYVq5caXL/yMhIfPrpp5g0aRK8vLxM7rNixQoMHjwY8+fPR0xMDObPn4+BAwdixYoVdT4vuX808+e+n1NzabiNEELud1bVJLEsiylTpkAulwMAKioqMHPmTKNA6Zdffqn1WJWVlTh58iTmzZtnsD0+Ph6HDh2y5rIMHD58GHPmzDHYNmTIECFIaqjzksYhUhckZRSWo0KlgUJK3eQJIeR+ZVWQNHnyZIPPJ0yYUOcT5+bmQqPRICgoyGB7UFAQsrKy6nzcrKysGo9Z1/MqlUoolUrh86KiIgCASqWCSqWq8/Wawh/P1sclxqrfa08ZA3e5BCVKNa7fKUJ0kLsjL69Roe9r+6F7bT90r+3HVvfamudbFSStWbPG6oupDcMwBp+zLGu0rSGOae1533vvPSxevNho+86dO+Hq6lqPqzXEsBpINGWQsSwSExNtdlxSM/177S0Ro0TJ4JfEg4j1ZR14VY0TfV/bD91r+6F7bT/1vddlZZaXU1gVJNmSv78/xGKxUfYmOzvbKMtjjeDg4BqPWdfzzp8/H3PnzhU+LyoqQlhYGOLj4+Hp6Vnn662OObMJkj9fwB2P9vD43z8GNWDE9lQqFRITEzF48GDhXu8sPotb57PgF9kawx+IdOwFNiKm7jVpGHSv7Yfutf3Y6l7zI0GWcFiQJJPJ0KVLFyQmJuKxxx4TticmJuKRRx6p83F79uyJxMREg7qknTt3olevXvU6r1wuF2qx9EmlUtv+YLh6AwAk2nLbH5uYpX+vowK5Ibb0uxV0/xsAfV/bD91r+6F7bT/1vdfWPNdhQRIAzJ07FxMnTkRcXBx69uyJb7/9Funp6Zg5cyYALntz+/ZtrFu3TnhOUlISAKCkpAQ5OTlISkqCTCYTWhO89NJLePDBB/HBBx/gkUcewe+//45du3bh33//tfi8DqXgslJSDc2ucpRIP36GG7UBIISQ+5lDg6QxY8YgLy8PS5YsQWZmJtq1a4ft27cjIiICANc8snrvok6dOgkfnzx5Ehs3bkRERISw+G6vXr2wefNmvPHGG1i4cCGaN2+OLVu2oHv37haf16HkHgAAiYbWDnMUfoYb9UoihJD7m0ODJACYNWsWZs2aZfKxhIQEo20sW3sh7ahRozBq1Kg6n9eh5Fz/JykFSQ4TqWsomVFYQW0ACCHkPmZVM0liB3wmSVsBsFoHX8z9yddNBg8F9/9DGq3hRggh9y0KkpyNLkhiwAKVNNzjCAzDCJ23b1BdEiGE3LcoSHI2UhewIt0oqLLYsddyH+OLt6kuiRBC7l8UJDkbhhGySRQkOQ5fl5RKQRIhhNy3KEhyRnKuDQCjtLzhFbGtSBpuI4SQ+x4FSc5IRpkkR+ODpNRcKtwmhJD7FQVJTohV8EESZZIcpZmuJimrqALllRoHXw0hhBBHoCDJGVEmyeG8XaXw5NsA5NOQGyGE3I8oSHJGfBsAyiQ5jH4bAFqehBBC7k8UJDkhVle4TZkkx6oq3qa6JEIIuR9RkOSM+BYAlRQkORItdEsIIfc3CpKcEd8CoIKCJEcShtuoVxIhhNyXKEhyRpRJcgoR1FCSEELuaxQkOSFWTi0AnAGfSbpTpERZpdrBV0MIIcTeKEhyRrQsiVPwdpXB21UKgJpKEkLI/YiCJGdELQCcBi10Swgh9y8KkpwQtQBwHvxCtzcoSCKEkPsOBUnOSBhuK3HsdRC9NdwoSCKEkPsNBUnOiG8BoCoFNFQw7EjNaKFbQgi5b1GQ5Iz4TBJAbQAcjK9JuppTAo2WdfDVEEIIsScKkpyRWAYNw82qQgUVbztSq2APeCokyC+txP4r2Y6+HEIIIXZEQZKTUolduA+oeNuhFFIxxnQNAwAkHEpz8NUQQgixJwqSnJRaCJIok+RoE3tEgmGAA1dycD2HiukJIeR+QUGSk1KLKJPkLML9XDGgVSAA4IcjlE0ihJD7BQVJToqG25zLpF6RAICfTtxCqZJmHBJCyP2AgiQnpRZzTQxRUejYCyEAgD4t/NHM3w3FSjV+OX3b0ZdDCCHEDihIclIqsYL7gDJJTkEkYjCxRwQAYN2hVLAstQMghJDGjoIkJyVkkqhw22mMimsKV5kYKdklOHw9z9GXQwghpIFRkOSkVFS47XQ8FVI83rkJAGAdtQMghJBGj4IkJyW0AKBmkk5lUs9IAMDOi1m4XVDu2IshhBDSoChIclI0u805tQzyQK/mftCywK+nbjn6cgghhDQgCpKcVFWfJMokOZthsSEAQHVJhBDSyFGQ5KRUVLjttHo08wUAnEy7i0q11ux+N/PLaFFcQgi5h1GQ5KTU1ALAabUIdIefmwwVKi3O3S4wuc+vp2+hz4d78fX+a/a9OEIIITZDQZKTEjJJVLjtdBiGQTddNunI9XyT+/x8kms4+dfZTLtdFyGEENtyeJD01VdfoVmzZlAoFOjSpQsOHjxY4/779+9Hly5doFAoEBUVha+//trg8X79+oFhGKO3ESNGCPssWrTI6PHg4OAGeX11pabCbafGB0lHbxgHSSVKNY7e4OqVkrOKUFBWaddrI4QQYhsODZK2bNmC2bNnY8GCBTh9+jT69OmDYcOGIT093eT+N27cwPDhw9GnTx+cPn0ar7/+Ol588UX8/PPPwj6//PILMjMzhbfz589DLBbjySefNDhW27ZtDfY7d+5cg75WawmF2xoloFY69mKIke7N/AAAJ1PzodYY1iX9m5IDlYarRWJZ4JiJQIoQQojzc2iQtGzZMkyfPh0zZsxA69atsWLFCoSFhWHlypUm9//6668RHh6OFStWoHXr1pgxYwamTZuGjz/+WNjH19cXwcHBwltiYiJcXV2NgiSJRGKwX0BAQIO+VmsJLQAAyiY5oZhgD3i5SFFaqcH5DMMh0d3J2QAAsYgBYH5IjhBCiHOTOOrElZWVOHnyJObNm2ewPT4+HocOHTL5nMOHDyM+Pt5g25AhQ7Bq1SqoVCpIpVKj56xatQpPPfUU3NzcDLanpKQgNDQUcrkc3bt3x9KlSxEVFWX2epVKJZTKqoxOURH3h1GlUkGlUtX8Yq2kUqkARgRW6gpGVQZVaT4g87LpOQiH/9rV5WsYF+GN3ZdycOhqNtoGc99fWi2LPZe4IGlU5ybYcuIWDl/Ltfn3yL2oPveaWIfutf3QvbYfW91ra57vsCApNzcXGo0GQUFBBtuDgoKQlZVl8jlZWVkm91er1cjNzUVISIjBY8eOHcP58+exatUqg+3du3fHunXr0LJlS9y5cwfvvPMOevXqhQsXLsDPz8/kud977z0sXrzYaPvOnTvh6upa6+utiwrI4YIy/Lf7bxS6RjbIOQgnMTHR6ud4lDMAxPjz6GU0KUoGAKQVA3mlEsjFLFprUwFIcCmrCD/+vh1uxjH8faku95rUDd1r+6F7bT/1vddlZWUW7+uwIInHMIzB5yzLGm2rbX9T2wEui9SuXTt069bNYPuwYcOEj2NjY9GzZ080b94ca9euxdy5c02ed/78+QaPFRUVISwsDPHx8fD09DR7vXWhUqmQmJgIuYcfkH8XD3RtDzbiAZueg3D4ez148GCTmciahN8uwm9fH0F6uQxDhvaHWMRgxe6rAK6jf0wwxj/WAT/c/A/Xckrh0zIOg1oHNsyLuEfU514T69C9th+61/Zjq3vNjwRZwmFBkr+/P8RisVHWKDs72yhbxAsODja5v0QiMcoAlZWVYfPmzViyZEmt1+Lm5obY2FikpKSY3Ucul0Mulxttl0qlDfeDoeCG2CTqMoB++BpUXb6O7cN94SGXoFipxtXccrRr4oV9V3IBAIPaBEMqlaJncz9cyynF8bRCDGvfpCEu/Z7ToD8zxADda/uhe20/9b3X1jzXYYXbMpkMXbp0MUqbJSYmolevXiaf07NnT6P9d+7cibi4OKMXvXXrViiVSkyYMKHWa1EqlUhOTjYarnM4uQf3ngq3nZJYxCAu0gcA1wogq7ACFzKKwDBAv1bcRIAeUVzwfqSBlzA5kZqPq9klDXoOQgi53zh0dtvcuXPx/fffY/Xq1UhOTsacOXOQnp6OmTNnAuCGuCZNmiTsP3PmTKSlpWHu3LlITk7G6tWrsWrVKrzyyitGx161ahUeffRRkzVGr7zyCvbv348bN27g6NGjGDVqFIqKijB58uSGe7F1wQdJ1FDSaXXTtQI4ej1PKNjuGOYNf3cu68i3CmjIfkkXM4ow+pvDGP/9EWhpGRRCCLEZh9YkjRkzBnl5eViyZAkyMzPRrl07bN++HREREQCAzMxMg55JzZo1w/bt2zFnzhx8+eWXCA0NxWeffYYnnnjC4LhXrlzBv//+i507d5o8761btzB27Fjk5uYiICAAPXr0wJEjR4TzOg25rtaJ1m9zWt2juKaSx1LzodYFKANjqmqPAjzkaBHojqvZJTh2Ix/xbW3ftHTD0TRoWeBOkRKXsorRJtS2NXKEEHK/cnjh9qxZszBr1iyTjyUkJBht69u3L06dOlXjMVu2bCkUdJuyefNmq67RUVi5O/cBBUlOK7aJF1xlYhSUqbD3MpdJGhBjWFPXI8oXV7NLcOS67YOkUqUavydlCJ8fvZFHQRIhhNiIw5clITUQMklUk+SspGIRukRwdUksC4R4KdA6xMNgn4asS9p2JgMlSrXw+VFqXEkIITZDQZIzo8Lte0J33TpuANA/JtCoHUVD1iVtPMoNRw/VZaiOpebXmEUlhBBiOQqSnBjLZ5KocNupdY+qmhygX4/E4+uSbL2O27lbhTh3uxAysQhLHmkLhVSE/NJKpNAsN0IIsQkKkpwZZZLuCe2beqGpjwtCvRTo1dzf5D49dAXe+uu4FZZzdUyFZXVrsb/xWBoAYFhsMAI9FcKwX0O3G6iuRKnGlDXH8P3B63Y9LyGENDSHF26TGsj4IIkySc5MLhFj+0t9wLKAi0xscp8eUX5YfyQd+65kI2CfHHsvZeNk+l1otCzCfF2wdmo3RAW4W3zO4gqVULA9rls4AG5Y77+reTh6PR+TekbW+3VZanfyHey7nIN9l3MQ7uvaIDP47jUvbT6N1LwybHmmBxRS098TpHbHU/NRWKbCoDamGwwT0tAok+TMFDW0ACgvAFQVdr0cYp6nQgovF/NdXPm6pOs5pfhgxyUcS82HRstCIRXhZn45Rn19GKfT71p8vt+TMlBWqUGLQHd009VE8bVRR2/k2bUu6eytQuHjV348g5v5lq+L1BgVlnEB7JmbBUi6WeDoy7lnabQspiUcx9M/nKj391RZpRrXc2gYmliPgiQnxsrMNJMsuAksawP8NNX+F0XqJMBDjsFtgqCQitC/VQDefqQtDr7aH/++NgDtm3ohv7QS4747ij2X7tR6LJZlsUFXsD22W7hQKN4hzBtyiQi5JZW4llPaoK9H3zldkOQqE6OoQo0XNp1GpVrbIOdiWRbv/30Jz6w7gZNppoPKEqUaX+27itd+OovUXPvdB96FjKqg8fztwhr2JDW5mV+G4go1WBZmv9aWWvjbBQxctt/uQ9H3I62WRXZR4/kHnoIkZ6bfAkA/M3BjP6AqBa7sACrol/C94rtJcUheMhRrpnbDxJ6RCPN1hb+7HJue7oG+LQNQrtLg6XUnsflYeo3HSbpZgOTMIsgkIjzRuWo9OIVUjE7h3gC4bJI9aLQszuuCgi/HdYanQoKkmwX4cMcli557I7cU5ZUai8/39f7r+Hr/Ney8eAdPrDyEGWtP4HIWV7NXolTjy71X8cAHe/DhjsvYcuIm4pcfwLKdl606R32d1wuSLmTQUHld6U9AsCbLasrha7lgWWDH+azadyb18sGOS+i2dDf2X8lx9KXYBAVJzoxvJslqAFV51fbMs7rtWiDtsP2vi9RZ9fYAAOAml+D7yXEY1aUpNFoW8345V+MsOD6L9FBsCLxdZQaP8cN6R+zUL+l6TgnKKjVwlYnxYMsAfPxkBwDA9//eQOLFmrNi3xy4hv4f70O7Rf9gxGcH8fqv57D1xE1kFJSb3H/v5Wx8+A8XfPVu4QcRA+xKvoOhnx7AjLUn0OeDPfjon8soKFMhyt8NPaP8UKnR4rM9VzF4+f5ar8dW9AOjey2TVKGyXzBZG/21CE/XY9iyRKlGRiGX2TiY0jj+cDurwnIV1h5OBQAcupbr2IuxEQqSnJnMHYDuj6p+XVLW2aqPUw/a9ZJIw5CKRfhoVHs81onLDH2z/5rJ/e4UVWCbrmB7Qk/jZXT4ZVKOXq9bXVJxhQo5xUqL9+frkdqFekEsYhDfNhjTH2gGgKtPulND2v2/q9wvUY2WxYWMImw8mo5XfzqLfh/vw3cHrkOjtw7d9ZwSvLjpNFgWGNstDOund0fi3L4YERsCluWCpbtlKkT6uWLZ6A7YOedBbHy6O1aO74wQLwVu3S3H0+tOYML3R+s9dFMb/cDoWk4JyirVNextSKtlHbb+3vHUfLR5cwfm/3LWKXptpWRXzeq9mFFU5wDuml6wdS2nFJmFpoNwUn8/nbyFChU31J6W2zhqEylIcmYMY9x1W6sFss5X7XM/Bkm5V4Ed84GcK46+EptiGAYvDowGwwC7L2Ub/CfNW/NfKio1WsRF+KBzuI/R453DfSATi5BdrERqnuW/pArLVPj4n8vosXQ3en+wB/t0S6zU5pwuIIht6iVse21oDNqEeKKwXIW/z2WafW7KHe71fT2hM74a3xnPPhiF9k29UKnW4t3tyRjzzWHcyC1FcYUKz/xwEsUVanSJ8MGih9uCYRg0D3DHl+M744/nH8DYbuH45MkO2DW3Lx7v3BQSsQgMw2BYbAh2v9wX/+vXHFIxg3+v5uKJlYcwafWxeg/hmFKqVOO6rg7KTSaGlgWSMy1r4XEzvwyxi/7Bgt/O2fy6LPHX2UxoWWDTsZtY81+qQ65Bn35wo9ayBrVe1qj+c/TfVapLaghaLYv1R9KEz1Pz7F8P2BAoSHJ28mrF23dvAJXFgEjXvSHzLDfTzVY0auD6PtvMnCu8bfsZeMl/At/2A458BRz4yLbHdgLN/N0wuDU33XnVv4Z9h4orVNhwlPsl9Gzf5iafr5CK0THMGwCXTapNiVKNz3en4IEP9+CLvVdRWqlBpVqLZ344aVFNwdlbBQC4XlE8mUSEAbqmmpfvmA4QCstUyNZlrHq38Mfw2BDMH94avz/XG+8/Hgt3uQQn0u5i2KcHMO67o7iaXYIgTzlWju8MucRwSn1sUy+893gsnujCBUfVucokeG1oDPa83A9j4sIgFjE4cCUHj311CDPWHrcqQ1Gh0uCjfy6ZzfRdyioCywKBHnKhyailf9wTL95BaaUGP528hcLyuvXOqg/9Id53tyc7dLiEZVkhuInydwMAnE4vMLnvtweuYXrCcbN1Z3xtk1jEZeX/pSG3BvHv1VzcyC0V7nNaXplTZCTri4IkZ1e9DQA/1BYcC/i1AMACaYdsd76kDcC6R4DN47isVV3lXAY+bQ982Q3Irr2It1ZaDbD7bWDLeC5IBIDs5Pof1wk9/WAUAODnU7eRW1I19LX52E0UV6jRPMDNZGdvXnehcWXNQdJ/V3Px4Id78UniFRRXqNEqyAMrx3dGfJsgVKq1eHrdCRyoIVBSa7RC/U1sEy+Dx1oFc8H9pSzTQRI/lBLqpYCHoqp1AsMweKpbOHbM7oPeLfxQodJyXcUlInwzMQ6BnooaX1NNwnxd8cGo9tj7cj+MjmsKsYjBruRsbD1x06LnZxaWY8y3R/Dl3mt47+9LuG2idur8be5+tGvihba6hYYtrUvi625UGtaiWY62VFShwqUs7toHxARCo2Xx/MbTuHXXMUMmmYUVKK3UQCJi8EhHbgjaVF1SiVKNj3dewe5L2WaDej7YGtKW++fj36v2bZFxv1h3mPsHbkzXMIgYoFylsWro3llRkOTs5NUaSvJF28Htgcg+3Mep/9rufDePcu+v7QaOf1f341z4FdCqgYI0YNVg4Oquuh+rLB/Y8CRw8GPu83ZPcO9zr3DBUyMTF+GDjmHeqFRrhV88Ko0Wq/+7AQB4uk8URCLjAnAeX7x99Ib5ddxOpt3FjLUnkF9aiSh/N3w2thP+fqkPhsWG4ItxnTGodVWg9G+K6YxCSnYJlGotPOQSRPq5GTwWowuSrmQVm6yxuaIbamsR5GH0GAA09XHF+und8faj7dA6xBPLR3cUMmT1Fe7nig9HdcDsgdEAuPtUmxOp+Rj5+X84o/eH2lQAyWeN2oZ6om0oFzjygVNt9If/7D0L61TaXWhZIMzXBV+N74zYJlxbimd/OGnXmYE8PvsT6e+Grs24YeUkE5mk/ZdzhHYT5oLRq7qAfHRcGFykYuSWKIXvP3s4f7sQY745jL4f7cW1Rtqr6dbdMiGwn9Y7Ek18XADAqiF/Z0VBkrOrXpOUpatXCGkPRD7AfWzLuqTsi1UfJ75Z92zN5b+5926BXIC34Ung6LfWH0ejAtYM54I2iQvw+PfA498BEgWgUQJ3U+t2fU6MYRg83YfLJq0/koYKlQZ/nMlAZmEFAjzkeLRTkxqf3znCGxIRg8zCCtzMN852XMoqwtQ1x1Cu0qBPtD/+nt0HD3cIFQIvmUSEr8Z3xqDWgVCqtZi+9rjJoTu+P1K7Jl5GQVukvxtkYhFKKzUmMy58JqlloPku4wzDYGKPCPz9Uh+MaB9S42uuC74J5/EagkkA2HA0DWO/O4LcEiVigj3wVNcwANwf6Or4gKhtqBfaNeF+dq/cKYZSXXOgkV1cgVt3q+7T/is5VhV819eJVC5A6xrpC4VUjK8ndoGvmwwXMorwfz+dqfPSOXWVohumjQ50R/um3hAxwO2CcqP+OzsuVAWTZ00ESRUqDdJ1jSjbhHoKX3N7zHIrKKvEG7+dw8gv/sXRG/lIyyvD02tP2P1e2sPGo+nQskCv5n5oEegh/NPUGOqSKEhydtXXbxOG2zpUZZKyznHZlvrSaqqGxoJiAXUF8MvTgNrKleuLMoHMJAAM8MxeoON4rl3B3/8H/PWyddmfS38COcmAiy8wYxfQ/klAJAb8W3KPN9IhtyFtg9DUxwX5pZX46eQtfHuAq0+a0iuy1mUuXGUSoUbo3e0XhWEUAEjLK8XEVcdQVKFG53BvfDOxi1GND8AFSl+O74wBMVyg9Na2C0aBxNnbBQAM65F4UrEIUQHcL0pTQ2580XZ0kOVLsdhahzBvocg9zcx/vD8cScOCX89DpWExon0IfpnVC2N1y8D8dzUXKk3VkLRSrcEV3R/3dk080cTbBd6uUqi1LK5k1ZxB4LMkrYI8EObrggqV1mQQZimtloVSA5RXaqBUa6DSaGucNXcslfv90S2SCyKaeLvgy3GdIRYx+PNsJuLeTcT0hOP47fRtlCgbPnjjMy7Rge5wl0vQUpdx1B9yU6o12HupaoLB+duFRt+jN3JLoWUBT4UEAe5y9Inm1lbkZ1baQkGZCleLgMPX83DgSg72XsrGmv9uoP/H+7D+SDpYFniofQiaeLvgem4pnt90CmqNbZqtarQsFv52Hp/svGyT49WFUq3BluPckPUk3YzbCD9XANzvm3sdBUnOjq9JqigCiu8AJXcAMEBQG8AjSBcssEC6Dfol3U0F1OVclmb8Vi4wyToH7H3XuuNc2cG9b9IF8GoKPPIlMGgxd93HvweSt1l+rGPfc++7TgeC21VtD4jh3udYESTtfhtYEcvdRycnEYuEqfTv/30Jl7KK4SYTY0J342n/pozRZTv+uXAHQ1ccxITvj2LbmQxMWHUUOcVcRmTNlG5wlZlfvlEuEWP56I5QSEW4lFWMU9Vmg/GZpFgTQRKgN+RmonibzyRFmxlusweFVCwEeHyQUN0mXU+qZ/tG4YuxneAqkyC2iRd83WQoVqoNiolT7pRArWXh5SJFE28XMAyDdroht9qKt0/pjtM5whvD2nFZs7/rOORWVKHCsM//w6vHJGj/9m60emMHohf8jZg3d+D3pNtG+yvVGmEYMU4XJAFAz+Z++GJsJ8QEe0ClYbH7UjZmb0lCl7cTMe/ns2b7WdkCH0Q312Ua+Sap+vf70NU8lCjVCPCQQyJikF9aaZS15IftooM8wDAMerfggqSjN/Jt0hVeq2Xx+NdH8PkFCSatOYlJq49hasJxLP7jIu6WqdAqyAObn+mBL8Z1xreTusBFKsbBlFy8u902/9xtO3MbPxxJw+d7rjqsJ9f2c5nIK61EsKcCg3STTqoySTTcRhqafk0Sn0XyjwZkuhoQPpt0wwZDbvxQW0ArwDMUePgz7vP/PrWu7okPkloN5d4zDPDAbKDLZO7zjNOWHefORSDtX4ARA12qLcESyAdJVvwHdW4rUJAOpNuw0L0BjY4Lg6dCIvzn/lS3cHi5ml8fTt+YruH4aWZPDI8NhojhZp68uOk0buaXI8LPFeumd7PoWF6uUoxsHwoA2HCkqhN4pVorTG1v38Tb5HNbBXMBfvVMUlG5CneKuILOFjUMt9mD/pBbdZmF5biYWQSGAZ7pEyU0AhWJGCEjsf+KYSYD4LJI/L5tdUNu52sJkvh6pE5hPhiiWyB4z6XsWofpTHn3z2RcN9GjplKtxcp914yyLedvF0Kp1sLPTYbmAYa1ZcNiQ7Bj9oNInPMgXhwYjSh/NyjVWmw+fhP9Pt6Hd/+6iLulVmaaa8GybFVwE8j9/uPr0fTrtv7RDbUNaxcsZJqqBwp80XYL3eLRrYI84O8uQ1mlxiYtIC7fKcbNu+UQMSyiA93QOsQT7Zp4okuED94a2QZ/vfgAeuhmObYN9cLyMVyz1TX/pWJTLZ31S5VqjP7mMBaZyOICXJ3i8sQU4fMfDqcZ7WMPfN3kuO7hwuzSCF2QRJkk0vDkerPbsvSKtnlCXZINirf5oavAttz71iOBThMBsMDPTwPX9tZ+jMoyroUAALQcZvhYYBvufZ7p6dNGjuuySDHDAa9qdTgBrXXXbOHMOa0GKOKaMArvnZybXILxPbjMkVjEYJous2SpuEhffDW+Cw682h/PPBgFD4UETbxdsH56dwR6WD5LjL+GP89lCn8Qr9wpRqVGCy8XKcJ8XUw+j88kXc4yLFzm/wCGeCngqbAs6GsoXfkgyUQmae8lbrirU5g3/NzlBo/1bRkAANinNyTGB0J89kj/45qKt9UardCUs1O4NzqFeSPIU44SpdrqYaH9V3Kw5cRNMAzwfBsNkt4YgLOL4vHfvAGQSbiMYPWlUo7r6pHiIn1MdoQHuEzM3MEtsfvlvvhxZk90i/RFpVqL7w7ewIMf7sUXe1Js1q07t6QSheUqMAyEIdtOup5g524XQq3RQqNlhQ7qQ9oGCxlB/cWWgaqibX5YVySqyibZYsiNn0Ha0pPF9hd64++X+uDPF/rg5//1wtTezYxaUgxtF4K5g7lSgYW/na+xTcfuS9k4diMfCYdS8edZ435jP564hfT8Miik3Dl+P3Pb7vVOl7OKcTq9AFIxg6e6hQnbI/nhttx7vw0ABUnOTr9wW5jZFlv1OJ9JumODuiQ+kxTYumrb0Pe5VgPFGcAPjwLrR3EZHnOu7+NqmbzCgKC2ho/5teDe56YYPc1IRRFwdgv3cdenjR8PaKU7loUz3Iozudl2wD0TJAHA9AeaoWukD+YMikYTb9PBSG2a+rji9eGtcWrhYOx9pR/CfF2ten6Hptx09kq1Fj+fugWg6o9R+6ZeZv+w8m0ArueUGgxtXNUtvuvIoTZelwgfMAw3LFC9KJifrTPARLuFPtFckHQhowjZxRXCxwBXIMxrp2uNkJxZZLYO5fKdYpSrNPCQS9A8wB0iEYOhumzS3+csH3IrrlBh/s/c74hJPcIR7cXCTS6Bp4Ib/otvww2F/HTylsHz+CxaV72hNnMYhkHXSF9sebYH1kztitYhnijWTcMf8dlBm3Qz57M/4b6uQv1diwB3eMglKKvU4MqdEpxIzUdeaSW8XaXo1sxXuM/nzGWS9DKWfJB00AZB0lHd8j8tvCwPBF4Y0AIPtQ+BWsvi7b/M/y7V7+f01rYLyNNrB1Kh0uDzPdzv0f8bEoOYYA9UqLT48aRl7SxsZVcy9zPyYHSAwT9eYb6uYBigWKlGvo0zjfZGQZKz028myWeSQvQySe4BVfU5af/V71x88MNnfABu/bjpiUCPWYBIClxNBL7uDWx7ESg18Uvmim5WW6th3DCbPn9uyjXyr3NNK2tyZjNQWQL4twKaPWj8uE+kdTPcCvR+eRSb7wLtbPzd5fhxZi88PyC63seSikWQSaz/kWcYBuN1tVAbjqaDZVmc0xVtV++PpC/ESwEPhQRqLWsw9blqKMWxQ20A4KmQorVuWFC/LqlCpcG/uj+iA2KCjJ4X4CEXXvvBK7lQa7RIzqzqkcSL8HWFu1wCpVqLazmmhx74OpuO4d7CLMEh7bggKTH5jsVFvku3X0JGYQXCfV0xd1ALo8dHdWkKAPg96bYQtGq1LE6kVc1ssxTDMOjfKhB/vfAAPn2qIwI85LiWU4pRXx/Ckj8u1mtmHp/94YfIAC4D1D6Mu6+nb94VZrUNjAmCVCwSMknn9Iq31Rotbui6n+sHSQ/ogqQzNwtQVFH3zItWywoLSbfwtDxIYhhG1zWeyzCaWrqHZVmh9YaHQoL80kq8ue2C8PjGo+nILKxAiJcC47uHY1LPSADcRAN7LmvDB0kDWxv+jCikYoToeprd63VJFCQ5O75wuyiDCy4AbmabPn7IrT51SWolkHeV+ziojeFjrr7A0PeA544CrR/mZqqdWstllvQ7amu1wJV/uI9bDjU+h2dTbhq/VsX1TzKHZauG2rrOMA62AN0MN13gkGPBkFuhXpBUdO8ESc7i4Y6hcJdLcCO3FIev5QmZpJqCJIZh9IbcquqSrmZzf7haOnBmmz5TdUmHr+ehQqVFiJcCrUNMZ7z4Ibf9V3JwPbcUFSot3GRiNNPrGSUSMWgTUnNTST5I6qTXB6pbpC983WQoKFNZ1Mfp35Rcocblw1HtTRbk94kOQJCnHHfLVNijmxWWkl2CwnIVXKRigwyYpUS6Zo+75vTFqC5NwbLA6v9uYOiKg3Xu2M0H0S2qfX90CuOG3E6nF2DnBe6P81BdMNkq2ANSMYOCMpXQSiEtvwwqDQtXmRihXlVZ2FBvF0QFuEHLAoev1X2JkivZxbhbpoKLVIRwt9r31+fvLkf7pt4AYDBDj3cjtxQZhRWQiUVYPaUrxCIGf53NxI7zmSirVOOrfdzv6hcGREMhFePRTqHwUEiQlleGAzZqb1BeqcHiPy6YHZbMLVEiSVfwbyrb2ljqkihIcnZ8JomfxeXZBHDzM9zHFk0lc68ArAZQeAEeZnrS+DUHxvwATPsHcPXjZr7tWlT1eOZpbvadzL0qcNMnEnHHAKoCMlNuHAByL3PH6fCU+f2EuiQLZooU6BVJFhnP8CE1c5dL8GgnroB79X83hBlr5ma28Ux13q4aAnH8cBtQFSQdS60aKtqTzP3h6h8TaHY4sW8rLkg6mJIjBI2tQzyNekbVVrx9+qauaFtvLT6JWCQMj/GNJbVaFkev5+HN38/j1Z/OYFniFWw+lo69l7LxGj/M1jNCKBSuTixi8FgnLpvED7nx2bPOEd6QmljSxVJerlJ8/GQHJEztilAvBdLzyzDuu6NY+Nt5lFrZMuBqtaJtHl+8/fe5TNwuKIerTCwU0MslYuF7jR9yE2bI6YYw9T1gg7qkI7oAq0uED+py6wa04gKLPSaCJD6L2SXCB10jfTGzL9c37Y3fzmPFrhTkllQiws8VT8ZxX09XmUTIFFpSwK3Rsth2JgMFZeaHwjYcTcOa/1Lx8tYzJrOZey9lg2W5iQrBXsY1jpH+3LA+ZZJIw5JX++9Ovx6Jxwck2RdMD4FZQr9o28wfBUF4D+DRldzHR1cCl3Wz2fj3zQcAErnp51pSl8R3+m4/piqTZooww83KTFJxFpetIlYZ140bctuVnA2VhoWvm6zWOil+hhtfvF2mBu7olipwZI8kffww06WsIhSWq8CyrPCHq6blXzqFecNDIcHdMhU267I47Uxk1oQ2ACaKtwvKKnFdNwxXvaM4P+S240IWlu28jAc/2osx3x7BusNp2HriFj7bnYJ5v5zD1ITjuF1QjqY+LnhtaEyNr3VUF24CxN7L2cgpVuJEquX1SJbo1yoQ/8x5EOO7c72kfjiShqGfHrAqY5Nioo4I4IYjAaBU1wG8X6sAg55hsdXqkoSibRPDunxd0tYTN/He38kG9T6W4jN83SONF5q2BJ99+fdqrtEsxoO6obYHdEHgiwOjER3ojtySSqFn2uxB0QaB7UTdBIs9l7NxM7/mwGT9kTS8uOk0/u+ns2b3+U3XLiKrqMJkdmq37h8JU8PRgG0ySc5Q9E1BkrMzCpLaG+/j5l9VR8TPLLPWHd14t37Rdk1aDuHqlADg91ncEJZ+PZI5fJBkLpNUeBu4tJ37uJuJgm19AVYESfo1SRqlbZpv3mfahHqis+4PFcD9UTKXZeFVH267o2tjE+zp+JltvAAPOZr5u4FlgZNp+bhypwS3C8ohl4jQq7m/2edJxCIhI8HX9bQ1MWTFB04XMgqN6kX45ojN/N3g4yYzeKx3c394yCXIKVbisz1XcetuOdzlEoyOa4qXB7fEuO7hGBATiJhgD0T6uWLFmI5wk5vvewVw2buOYd7QaFn8nnTbqqJtS3kopHj3sVhsmNEdTbxdcDO/HGO/O4I3fz9fa2+iwjKVsN5X9SDJ312OcL1JB3yrBF6srhUF37+Lz0g1NxEk9WsVgF7NubUBv9l/HX0+3Iv3/75kcZExV4+kC5Ka1e3etQ31RKCHHGWVGoPFhdUarZCl4r+/5BIxPnqyA/iEWHSgOx7uYDjjNyrAHX2i/cGywPqjNWeT/jrHlRzsTr5jst9Vyp1igxmZm48ZFoQr1Rqha/mg1qb/keBnuNU1k5RVWIGHPv8XJ9Mc+7uagiRnJ682JBFiIkgCgFbDufdHVtYtS8JnkqrXI9Vk0CIus1WWB2weq1syhQGi480/h68jMhcknV7PDftFPFB7wMYHSbkptc9wK6w264OG3OpkvF4zS1Odtqvj+9dkFFagqFyFrDLut7yzZJF4XXXZgGM37mK3blZbr+Z+cJHV3N2cr0vitQ01vifNA9wgl3BLtFRfpsFUPRJPJhFhQs8IiBjuPJ8+1RHHFwzCh6M64IWB0Vj6WCxWT+mKHbMfxL7/62/QCLImT+iGZb4/eAMZhRWQiBihWaMt9W7hj3/mPIhxuqzSusNp+Hp/ze0/ruZwwXSIlwLuJgI+PtsmFTPoXy3Lp59JMuy1ZPy9JpeIsWFGd6yaHIfYJl4oq9Tg6/3X8MAHe/DprpRaC89TskuQX1oJF6lYWH7GWiIRV/wOGA65nblViGKlGl4uUoPMZMcwb8wZ1BJSMYOFD7WB2MT6jXw2aevxm2ZbMtwtrRQyiFoW2HzceEbcr6e534/8Pzm7L2UbzP48ej0fpZUaBHrIDVpe6KtPJim3RInx3x/BhYwiLPj1vF2L0aujIMnZVR9uMpVJAoDuz3KzvW6fqNtabsJwmxVBkkQOjFoDSN2qGkSGdeMyW+b46YIkc8Nt/LXHjqr9/PwMN3VFzTPcWLYqk8Rn5u6hGW7OZET7EHi5cBkgvvC0Jl4uUoTq6hWuZJcgs5z7xd7SCab/6+MzKcdT84VCWlPFqNU9qBckycQik8GfRCxCa754u1qPIqGJpJkg5bWhMbj67nCsndYNj3RsUmvQZomH24dCJhYhS/dHr20Trxo7r9eHu1yCpY/F4t3HuG7564+kGSzlUh1fR2SuyShfP9YnOsAoE9ky2B1SMYPCchXS8sqqljYx873GMAwGtg7Ctud74/tJcWjXxBNllRos33UF/T/eh60nbkJj5o8z3x8pLtKnXrVcfKCnX7zNz2rr3cLPKBB6YWA0Lr89zOD7Tt/A1kFo4u2Cu2Uqk72VAG6oVctygSbABVT6NUdaLYvfk7g2KS8MiEaXCB9otCx+OlXVOmKP3s+IucW2+aVJCspUNdY+VVdQVokJ3x/FtZxShHop8P3kuBoX9G5oFCQ5O4kCEOl+gSm8AO9w0/u5B+oaPwI4+Il156goAgp1hc2WDrfx/KOB4R9VfW5qVpvB/rrhtpIs7rz6tBrg9inu47DutZ/b0hluZXnccisAt1QKcE/1SnImCqkYn4/thBcHtLAoiACAlsLyJCW4o8u8O8P0f33dm3HFzmdvFQi9fqpnKkwJ9XYRZunFhHiY/YPJZxtO6fUR0mpZYXaQftF2dbb+A+HlKsXgtlV1JF0j6lZTY40nu4TB312O7GKl0CnbFHNF27ynuobhvcdj8f7jxrWZcokYMboauB0XslCh0kImFiHMp+a6OYZhMKhNEP54/gF8Ma4TwnxdcKdIiVd/OouHPv/XYCiMxwdJ5orkLfVAtD+kYgapeWW4rgvq/r3KDWM90MJ0IFTT94NYxAiZuw1mhtz4JpzTHmgGXzcZsooqsFevKeqx1HzcLiiHh1yCga0DhSWOthy/CZZlwbKs2an/+lxlEgR6cLWp1ddGvJZTgpc2n8b6I2kGawEWV6gwefUxXMoqhr+7HBue7oGmPtb1dbM1CpKcHcNUZT+C29dcVN3rBW4Jj+v7gNsnLT8HH2B4hAIudfiF2XEc0O1ZwDuCK7auicILcNP98ak+5JadDKhKAZlHVbPI2lgyw40fanMP4rJPAGWS6uHBlgGYG9/KZLrflFZ6QRKfSXKGRpL6wnxdEOQph0rDQstywwyW/nLmg6nqhdf6+ExVwqFULPnjIpRqDa7nlqC4Qg2FVCTcI3vhZ0IBVV3HG5JMIsI4XUfmdTXMvjJXtM2TiEUY2y0cgZ6mO8bzw1O/nuKGi6IC3Iy6XpvDMAweah+KXXP7YsHw1vBUSJCcWYQJ3x81WCSaZavqkXpE1e/eucslQoC+51I2SvTWA+Trkaz1ZFxTSEQMTqcXCL27eBUqDfZf4QKiEbEhwveB/hIpv+mG2obHhkAhFWNEbAjc5Vx7gSPXuZq9W3fLIZOI0LtFzUFi1RpuhkNu721Pxu9JGXjjt/Po/u4uLPj1HE6n38X0hBM4c6sQPq5SbJjRHc38reyt0AAoSLoX8HVJ5obaeD4RQPvR3McHl1l+fL5o25p6JH0MAwz/EJh91nj5EFPM1SXdOs69b9KZyxJZgg+malrDjR9q8wrj1qQDqCbJjvi6hpNpd1FYyQVJjl6zrTq+kzTPkiwS78UB0Vj4UBvMHtTS7D4PtQ/FtN7csjKr/7uBUSsP47fTXDazfZP6Tb+viz4t/BEd6A4fV2mdC4+tNa57BMQiBsdu5Bv98eYJmaQ61qzxdXKXdS0qTBVt10YuEePpB6Ow///6o0+0Pyo1WszenCTU+OjXI8WaWbfQGsKQ2+VsHL2eB7WWRbivK8L96pZBCfRQIF6XKay+PtyR63koq9QgyJNrhvqULku073I2bheUo0KlEYq6H+vM/S53k0swsgP3e3PL8XShZq93c79ah2n5ITf9TFJ2cVXmKtLPFaWVGmw4mo7HvjqEY6n58JBLsG5ad7v/42AOBUn3AoWuMM5c0ba+3rO595f+tHxdM6Eeycqhtroy1yvp1gnufdOulh+Lv+YcCzJJ3vpBEmWS7KVVkG6hW129SZCnXKhrcibd9IKFmqb+V+cml2C6bujCHLGIwZsj2+D7SXHwdpXi3O1CfLGX+/5viKLp2kjEIvz6XG/sfaUfvF3NX7ctBXsphOVWTGWTSpVq3NbNtNLvtm2N6s1N6zOs6+Mmw/IxHeHvLsOlrGJ89A/3j5h+PVJdOthXxw9bH72ej791PbH4qf91xbfr+PXUbYMidH6obVDrIDAMg6gAd/SM8oOWBbYcS8eeS9korlCjibcLuun90zBWlwXcfj4L23T1SjUNtfEi/Y0zSb+eug2NlkXncG/sfaUfNj3dAyPah0AiYuAmEyNhWtda+6/ZEwVJ94JeL3Cz1/gZbDUJjAFiHuI+/m+FZccX1myrYybJWuaKt/lMkjVBkiUz3PQzSXyjTBpus5vmgW4GQ3N1/QPY0Ho19wfDcC0BaqoRqo9BbYLw90t9DAIyRwRJADfUY68AiTepJ/fH+7fTt1FYbrgkCN8vyt9dZtQOwVItgzwg08vK1Tdj6e8uxwdPcP+crvr3Bv5NyRWCJFtl4Jr5u6GZvxvUWha/6Iqj+9RxqI3Xq7kfIvxcUaxU488z3O86/VqiQW2qAhy+hmnLiZv48QT3u/KRjqEGtU+xTbzQOoRbv5FvDGtJTWL1TBLLstiqO8fouDAwDIOezf3w5bjOOL5gEA682h9dIuyT2bQUBUn3gvajgbGbam6sqK/PXO792a3A3Vq6r7Ks/YMkYbhNL0gqL+C6bANA0zjLj2XJDDchkxROw20OIJeIEaVXWxAd6Pg6A1NaBLpj/fTu2DCju8X1VnUR4uWCTU/3wGtDY/B4pybo18ryrNW9rlszX8QEe6BcpTFaaJfv4t68HkG0TCJCjN4yMuYKwK0xsHUQJvTgAomXf0wSGmPWt2hbH98KQMty1Qs9m9fv2CIRg6e66gq4dUNu3DpxSrjKxOipd+3xbYPg6ybDnSKlMAz2WCfDsgmGYYShOQBoE+KJUAsW3I6s1gbgVHoBruWUwkUqxoj2his7+LjJ4OdupgmxA1GQ1Bg16QJE9eP6De18A7h5DKg006uiNIeb/cWILC+Wri8+k5R3jVvvDagqNPdpVnMLgeosmeHGL0miX5NUUQhU3tvt8u8l+vUFzjazTV/vFv52aU8gFjH4X7/mWDamo0HX6MaOYRhM1GWTfjicCq2WhVbLYsPRNCz6g6uN5Nsl1BU/5CYWMcLSGPW1YHgbRAW44U6REnfLVFBIRRa1wLDUQL2GjO2beNkkw/dkXFNIxQzO3CzAhYxCJF7khvL6tjTsVC6XiPGkXiF/uyaeJidWPNqxiTC8ONBMA8nq+Lqq3JJKFFeohEzV8NgQeDhJM9naODxI+uqrr9CsWTMoFAp06dIFBw/W3ONn//796NKlCxQKBaKiovD1118bPJ6QkACGYYzeKioMV1q29rz3nAd02aTkbcCqwcB7TYEvugK/PGtY5MwXbftGAdLa/zOwCZ8Irq2Bqgwo1k3Fr0s9Eq+2ztv6NUlyT66vE0BDbnYUoxckOVvRNrGvRzs2gYdCgtS8MiQcSsXobw5jwa/nUVyhRvumXvhfv+b1Oj4fJEX4ukIusU0A6iIT49MxnSDRZRjjInxtUo/E6xrpKzTPrG89Es/fXY54XQ3YpmPpSNQtIzLIRC3RU92qWss82tH05BsvVymefTAKwZ4Kg9mRNfFUSOGnGzq9lFWMP85wv+9Hx1n2fGfg0CBpy5YtmD17NhYsWIDTp0+jT58+GDZsGNLT003uf+PGDQwfPhx9+vTB6dOn8frrr+PFF1/Ezz//bLCfp6cnMjMzDd4Uiqopo9ae957U7EFg2EdAi8Hc1HdWyy1ie3Yz8G1/4NxP3H72LtoGALGUyxgBVXVJdalH4vFBkqlCdWUJUK7rTeMVxuWyPXVpXuqVZDf8Gm4A0CLAOYfbiH24ySV4sgs3dLPkz4s4kXYXrjIx3nyoDX6d1RtBZqb3W2p4+xAMjAnEc/1b2OJyBbFNvTBvGPe75mHdbC9bkUlEeDKuKeQSkdFyI/UxThf8/HTyFpIziyBiTM/cbObvhnHdwxET7IHHO5sPYF6Ob4Ujrw8Uumlbgq9LWrnvGkorNYj0czWoyXN2DdNm1ULLli3D9OnTMWPGDADAihUr8M8//2DlypV47733jPb/+uuvER4ejhUrVgAAWrdujRMnTuDjjz/GE088IezHMAyCg4ONnl/X896TGAbo/gz3BgDFd4Css8Chz4AbB4CfpwM3j3LDTgC3sK09+bXgapLyrnJDg0KQZEU9Ek+Y4WYiSOKzSAqvqpoujxDuvBQk2U2HMC/IJCL4STXwdMKZbcS+JvaMQMKhG9Cy3EzCJY+2q3WxZEt5KqRYNaUO/2xZYEafKIztFl7rGnl18eZDbfD68NY2bQfRM8oPkX6uwvppcZG+ZmdhLn3MxOLpNhDp54ZT6QVCl+4ndQXb9wqHBUmVlZU4efIk5s2bZ7A9Pj4ehw4dMvmcw4cPIz7ecF2wIUOGYNWqVVCpVJBKuV++JSUliIiIgEajQceOHfH222+jU6dOdT4vACiVSiiVVStFFxVxfT5UKhVUKpW5p9UJfzybHlfhC0T2A8L7QLT/fYgPLQeOfSs8rPZrCdbGr6MmIt8oiAFocq5Ae+cSpBUFYCUKqP1aAdZeh09zSAGwuVegVlYY9Fhi8m5AAoD1DINad1yxRwhEADQFtxrmXhMjPgoxts3shpNH/qV7bQfO/n3d1EuGH6bFQa1h0TPKFwzDOO21VicTGd5XW99rVW3rUFppdFwTfPgPl7Ef0Mrf7ve5qU9VZlDEACNjg+p8Dba619Y832FBUm5uLjQaDYKCDMdHg4KCkJVlum19VlaWyf3VajVyc3MREhKCmJgYJCQkIDY2FkVFRfj000/Ru3dvnDlzBtHR0XU6LwC89957WLx4sdH2nTt3wtW1YdqmJyYmNshxgU4IipqDzmnfQKbh/sPYn5yDkhvbG+h8xsJzy9EJQO6lw7h9B+gMIF8ehn//2WX9wVgtHmKkEKsrcOC3BJQoqmZNRObsRgcAWUopjm3nXl/r7DK0BJB2/jDOFXAp+Ya710Sfu5TutT3dC/f67xr6wN5LnPVee6kACSOGhgXEWRexfftFu57/bg4DgPvHtZWXFqf/24PT9Txmfe91WZnlk3YcOtwGwCjtxrJsjak4U/vrb+/Rowd69OghPN67d2907twZn3/+OT777LM6n3f+/PmYO3eu8HlRURHCwsIQHx8PT8/6zcaoTqVSITExEYMHDxayY7Y3HCiYAO0fLwAiCR58dIrlXa5tgEn3AX5YjUBRIfz9VUA64N1uMIYPsqAXlKnj5XcF0g+hXxgLbdeqY4j2nABuAYHRcRg+hNsuOp4B7PwTkb4yBA8ebId7TQB7fV8TgO61Pd0L9zqifT5KKzUY0Mr0enANqcmtQvxw9SgA4H9DOmJYO/OlMLWx1b3mR4Is4bAgyd/fH2Kx2Ch7k52dbZTl4QUHB5vcXyKRwM/PdF8JkUiErl27IiUlpc7nBQC5XA653LiHg1QqbbAfjIY8NgAgoAUw7W8ADqjgD+LqiJjCmxCn/wcAEId3h7iur7dlPJB+COIbeyDu9b+q7cVcPySxb0TVsX24olFRSZZwfxv8XhMB3Wv7oXttP858rx9oWXt37IbSOtQbfm4yuMrFGBIbCqkNZhzW915b81yHzW6TyWTo0qWLUdosMTERvXr1Mvmcnj17Gu2/c+dOxMXFmX3RLMsiKSkJISEhdT4vaQBu/rrlVlhu1h0AhHWr+/FaDObe3zgIqPTaPeh32+YJs9uoBQAhhDQkN7kEO+c8iD+ef8BmLRnsyaEtAObOnYvvv/8eq1evRnJyMubMmYP09HTMnDkTADfENWnSJGH/mTNnIi0tDXPnzkVycjJWr16NVatW4ZVXXhH2Wbx4Mf755x9cv34dSUlJmD59OpKSkoRjWnJeYgcMU9VUEgA8m1Q1eqyLoLaARyigLgfS/q3art8jieehO0/JHUBbta4RIYQQ2/Nzl9t9CRxbcWhN0pgxY5CXl4clS5YgMzMT7dq1w/bt2xERwXVkzczMNOhd1KxZM2zfvh1z5szBl19+idDQUHz22WcG0/8LCgrwzDPPICsrC15eXujUqRMOHDiAbt26WXxeYid+LYDbfBPJOkz918cwQIuBwOkfgJRdQItBgLoSKNYNq3pVNUuDeyDAiLmO5CU59TsvIYSQRsvhhduzZs3CrFmzTD6WkJBgtK1v3744deqU2eMtX74cy5cvr9d5iZ346zV7q0sTyeqiB3NB0tVEAO8DRbcAsIDExXCpE5EY8AgGim6Doa7bhBBCzHD4siTkPqY/3GaLICmqH7fcSd5VIP+GXj1SUy7TpM9DV5dEQRIhhBAzKEgijsMvqCuSAiEd6n88hRcQ1p37+OquqnokLxNt9nXF20xxHbpuF98BDnwMVFg+jZQQQsi9x+HDbeQ+FhAD9F/AFW3banHdFoOAtP+AlEQglOuyblC0zfPUrY9UnAnAyrWS9r8PnFgNqJXAgAX1ulxCCCHOizJJxHEYBuj7KtBpvO2OGc23AjjADbsBhkXbPA8+k1SH4basc9z72yfrcIGEEELuFRQkkcYlqB0XAKnLgcu6ZVZMZpJ0bQCsDZJYFsjRraOQeYb7nBBCSKNEQRJpXPhWAACg0q3P42UiSPKoY01ScSag1NUileVS4TchhDRiFCSRxofvvs2rMZOUZV02KOeS4eeZZ627NkIIIfcMCpJI4xPVj2sWCXDvPUx08tYFSYyqDBKN5StCC0NtvMwzdbtGQgghTo+CJNL4uHhXtQLwDAXEJiZxSl0AhTe3u+qu5cfmM0m651KQRAghjRcFSaRxih7Evfc2MbONp2sDoFDlW35cPpPU9jHufRYNtxFCSGNFQRJpnOKmAx3Gci0GzNE1lLQ4k8SyQHYy93H70dz7wptAaV49LpQQQoizoiCJNE4u3sBjX3P1SeboZrgpLA2SSnOAigKAEXGNKn2juO1ZNORGiIFbJ4C7qY6+CkLqjYIkcv/SFW+7VFoYJPH1SD6RXE0Tv5QKzXAjpErOFWBVPLDxKUdfCSH1RkESuX/pgiSLa5L4eqSAGO69ECRRJokQwY39AKsBcpIBZbGjr4aQeqEgidy/dIXb7koLeyXx9Uj8wrzB7bn3VLxNSJX0w1Uf80sDEXKPoiCJ3L+axoGVusFd+f/t3XlcVOX+B/DPmRl2QUGUTUFw3xdwwSXXVDS7praYGlZmaJrmrZtLpZWlt+s1f92Srl6zzDVTS01NMEVzSWNR3HHBHVBRFpFt5vn98Z0zwzBnkGVgwL7v18vXHM6cOXPmYWS+8zzf5/ukQkre/+jjLfUk3b0A5GZWzjUyVpMIAVwpEiTdSbLdtTBmBRwksb8uJ3foOtDiuqojXz76eDknSe5JcvE09EYh9WQlXOBjpjAPEDpbXwWrTPevAkWX+uEgidVwHCSxvzRdl9chIEF1aS+QUkKg8+AOrdUGAJ7NjPstJW8/vA/8vgTILOPacI+ry/uh+aItul/4zNZXwirT1SOmP985b5vrYMxKOEhif211AnCzTmfaPlxCb5I81FbHH7B3Me63lLy97U0gei4Q9YH1rrWmOrER+H4EpIfpqJd9mhcFfpzJ+Uie+t5W7kliNRwHSewv70L9IbSRuBHIuKF8kGGorYXpfjl5u2iQlBQNnP5Zvx0FaAutd7E1iRDUm7Z5AqArgJDoz4109ZBtr4tVHrknqdNLdJt+EdBpbXc9jFUQB0nsL+++SxB0/t0BXSFw9L/KBxmStpub7pd7km6fBQpygYKHwI63jffn3gduxFr9mqs9nRbY8Q71pgFA6BToOr8GAJCuHLThhZVR+iVg93tAThmWrvmrykmnaf8A0PZZQO0AFOZSVXrGaigOkhgDoOv2Bm38uVJ5ppqhJ6ml6X43X8DZk+rCpJ2inpN7l6mad9OBdMyFqEq77mor6gPg2HIAEjBoATDoE4iAXgAAVU0KknbOBA79Bzjwb1tfifUlRQMJa4HCfOuc79ofdFu3KeDqBdRtTD/zkBurwThIYgyAaPIkJWTnZQLx35sfUHz6v0ySAB/9kNupLcDvn9P24AXGRXCTdlfORVubTgeseRb4T0jFek7ysoHYb2l7eCQQOhkAIPxDISBBSr8IZNaAvKScdODiHto+u710tbRqgpx04MdXgTUjgZ8mAV91prwxXQVnHsr5SAGhdOvZlG45SGI1GAdJjAG0HlvoFNo+EgloC4z3PbwHZKfQdr1m5o+Vh9wO/QfQ5gGN+wGthgNNBtD+W8eBrNRKu3SrOb2FArq7ScAfFoYdS+PUFiA/G/BoDLQvsjSFY21kOAXQdvLvyo+9FAOc3109ApKz22kIFqB1yNJOl/9c968BK4cAp7eWfNzFvZU7I/LcLmBpN+Dkj4CkBpzr0mvbPAFY9gT1LpW37eV8JH99kFRXDpJqyAy33Azg7I6KB4vsscJBEmOyds8DLvUoh+LQf4z7b+v/yLs1ABxczR8nB0kA5WEMWUQ9TLXqAz4daL/cI1FdaQuA3+Ybf/7j6/IvKSH3xHUcS+1QxB1X/XBl8gHzx91LBr4fDqx9Ftj0qu0LdJ7cTLeSmm7P/lL+c/25ArhyEIgpoQTCpRj963/O+kFibibw02Rg3fNAdirNPpsQBUxPBPq9Bzi4ASmJ1Lu0/19lP3/BQ+BGHG37d6NbuVRGdehJ0mmB/AclHxP1AbB+tH6YmDHCQRJjMjtHoO9s2t7zEXBmG20XLyJZnDzDDQB6vmXMxQCApk/SbXUfcov7jpKUXepRD1DufeDPb8p+ntvnKDdFUgMdXjS7+04t/XClUk9S/GpjscmTm4BlvSt3Xbwrh4Ad/6CaVsVl3wYu66uw93yLbs9uL/9zXdAHyamJVHNLyflddJuSaMzvsYbU08CyPkDCGgAS0H0q8Pp+wC+Yylk88Q4w7TigT6zHoS8fHVAUdyMO0BUAtbwB90DaJw+33a0GQdLOd4GFASUvRn3xN7qVg+PSEALqnyai5/n5FU/uz80suVYbswkOkhgrKvhl/YeFADa9Rn/8LeUjydwDgWaDgUa9gJ7TTe+Tk7cv/lZ9SwHkZQP7/knbvd8Fev2dtg99ST0EZSH3IjUdCLh6m91916UZlQJIv2g6rKQtpCAJAHpMB2o3pKDtfwOAo8srp2flh5doNmPRHjTZmZ8pGd+3I9BtEg3H3jpOw2ZllX3bdH0/pV40gIbaZH+uLPvzKJBO/gj8rz+1t1sD4OWdwMD59IWgKGcPIOwzwL0RkJdRtkABMOYj+Xcz9h7WbUK32anKgWhVyX9A7y1dAQ0FK8m8SdXCAQpQs2+X7tw346E6tRl1H5yHevPLpsP0ZfXzZODrHsaAmlULHCQxVpQkAYMXAk2eBAofAuteMH6oWepJUqmAFzcA47cDdk6m9/kFA07ulO9w/VjlXnt5HYkEHqTRB2SncKDdcxSkPEgzBi6loS0Ajq+n7U7jFA8p1LhAeLWlH5KLzHK7EE1FJp3rUm/e6/uB5kMAbT6VVDj0RflemyUHFgEP9B+EsSuBuxdN7z+p/zBtPYKWn2moH0I6t6Psz3Vpb7GfY8yPyUoxTp8H6MO8Ij0T2ny0vbYKmp8jgIIcIKgvtamcVK1EpaIvCUDZexGL5yMBgKMbzfIEbLvQ7YVo+r8MWM6FM6kULoy9eo9S5P+H6spBKntRnoC+4CHl4gFA3KqyP55VGg6SGCtOrQFGfQPUb03fguVeAEs9SSVRqSmRG6iepQAe3AUO/h9t93sf0NgDajugxzTad/AL02/HOi3la22JMP+2fX4XBR4u9Y09aApEQA/aKLqocNx3dNt+NKBxoJ6NF9YCffTDn4f+Q2u/WcPdi8DhpbTt3oiSs4v2JmXeovwhAGg9nG5bDKXb8gy5yT0D9VvR7WWFIOnSPrr1aQ94t6UJAHLAWVYP70O9ejiC7kTTz0+8A4zdBLjUffRjO44FVHbAzTjgZkLpnk+nBa4dpW05H0lWHWa4FU2WvxmnPJQoB0ka/Zec0gTDBbmUAA/gQr1BEJAo4D5ajpymq4fpdw7Q/yNb5+MxAw6SGFPi6Ea9Qy71jfuUZraVhhwwJFXDIOnAIiA/i/KqWo8w7u84ll57xlWqRA7QcMS3T1FxxePrgO+eMp21F6cfauswmgItC0RAT9qQv9Vn3gLO/0rbcqVmgHr1es0AXH0p+CrrEJAlu9+noZfG/YHn9Xk6pzYbE49P/wxAAA060zI0ANBCX5U9+WDZeniEMOa69J1DuVrpl8yH7eQgKaivaW9OWXslHt4Hvn8GqutHka92RuFzaygxW6Uu3eNdPIFWf6Pt2FIO+aWdpiE6e1fAq43pfbae4VaYZ3xvqe0pIFbK95KHC/XlKnBxL5CfU/K5z24HcjMg3BrglN9o6Pq9T/t3zTT+zktL/v0DVICzIvlvzKo4SGLMkjoNgRfX0x9/nw40bFYejfvTbcoJGlaxJOMGsG408EVHSratbPevAsf+R9tPfkjDLTI7JyBUX2DzwGLg+AYgsgdw9RBgXwuo5UUJ7d8OpSAn86axp6yj8lCbTDTsRjk+6ZfoNSesofwf/1DzIU21HdBlAm3/EVnx3KSLe4Fzv1CwMngB4N3GWKYgei6d/5Q+GCsaNHoEUc+i0JYtCT/1JA1b2jlTEr9fJ9pftDdJCGM+UuO+VK3azoUSnstSeDM3A1g9ArgZB+HkgYNNZkM0HVT6x8tCXqHbExtL16Mh98I07Ey9sEUZZrjZKEi6uJe+BLj6GOuWFR9yy82k3xMAdJ4A1Pan4bniw6TF6YfadO1eACQVdN2mUk+o0AIbxwN3yjDEKP/+5d5G+YsJszkOkhgriV8w8NZJ4NUKzE6rVY8SgAHKjyhOCOqFWdqNuvnTL1Eu1IO75X/O0tjzMeX8BPY2DgkW1flVwLEOfVhvmUiFNht0ASJ+B17ZRXlLd5OAb4dQRWqho0BHHmKxxNHNWDYh+YAx2btoL1JRwS8DGkdKnC6+ynxZaAuBXbNou8trxoCs72zqZbi8n/JBrv0BQDIOtcnKM+QmD7U16knDiIG96eeieUm3z1IdLo0j5T45ugHtnqX7SpvAnZsBfP8MLYHj5IHCMZuR6exf+ussKqA7lQgoeAAk/mB+/8P79Lpi/gWsfQH47WPa76+Q72SY4WajnKQz+qG2lsOAwCdou3iQdP0YvXfrBFAF/eZhtL+kIbf71wy9P7r2o2mfJAFPLaH/I7kZwE8RpQvqH9w1Duk/pS9Ge2lfzait9hfAQRJjj+JUhz7gKsLSkFvGDapyvXUKBSF+wZQnc/8KsDG8YrNlSnI9Vv8BKAFPfqR8jIMr0DWCtiU10GcWzY7yCKSelfG/0HBU+iVjj9QjepEMGumH3A78m+ojOdSmApxKnD0omRyg3qRHycsC1r0ILPSnnrk/ltG3+tiVlBzt5AH0mWk8vo4/0GUibW/XT/f3D6UPzKLkIOnCntLP+pOHXeTexCB9kHQ5xvgBKg+1+IcaZ53JQ25ntlouGSDLzQC+H6EPkNyB8K3mw15lIUnG3qRjRYb8dDqa8bioGfVY7Z0PnN9Jz2/vSoFIcYYg6WL5Zncm/ggc/qp8PYjaAmNtq5ZPG99zN2JN85Lk4Tc5yJOHVs/tsrw47/H1AATNaK0TYNxv5wg89x31BF4/VroeIblXsX5ryunyC6GgzdJMvPK4/qfpkJ41XT4ApF+unHNXAxwkMVYVmujrJZ3/FfgmDFjaHfi8DQ2tXYiiIpQDPgRe2Q2M1g/xJR8Adv7D9Dw56UD0PGDV38xnZJWWEMCv+oTo9qMB3w6Wj+31d2DwP4EJ0RRYFB1OcQ+goEmui2Pvat77YkkjWsfNMAzT7lnA3tny8XKwdmZ7ydPwM24A3wymIbXcDOoN2PkO8GUwzTwCqOeo+NBpr79ToCb0H4ptRsCMT3uaRl+QY/qBk5NOuUrFA9r8B8Zclyb6IKlBF+oxyk41lpYoOtQm8+1AvY/afH19oxL88nfgxp/0ml7aSonfFdX+eUpiTjtFH/aZtygw2j2HEozdG9Gw4OCFwKtRwDtJQP2W5udxa0Dn0RVQ4F8Wd5KAza/Re/XEhrK/huTfqd6Xsyf1jtUJoN7P4nlJhvIFXek2oAfgWBvIuaM8I1WnAxL0s9o6jjW/382XcukAIGruo2tOycN6QX3otq2+F9FaQ26ZN2lYfNVwIO2sdc4pu5lAuYn/fYK+eD2GOEhirCr4daJCe4UPKa8n7RRV9tbm0TfHiANUY0mtoQ+bkf8DIFHy7tHl1DsS8xnwf+1pfbhL+4AfwmmGTVmd2QpcO0IfXv3fL/lYjT3QLcKYS1Nc7QbAyzso2XfwAipOWBr+oZSXJOsUXvLxXq0psBJayxWRbx2nmkCpJynp/LlVwIB5NMyitgcgqIdF7qUpytnDWONKUhmTl4uSJGNv0rEVNFy5rC/wWRANOW561bTHI/kgBTm1GxprBtk5Ag31H8aXYyiwkod/gooESYDxOmO/tbxUxvVY/YepBLy40biOYEU5uQNtRtL2rllAZHf6MNc40ZDQmwn0Hu02CWjYxbz0hUylAjz1r72sM9xi/mksLvrr7LKXRJCH2loMpcR1STL2Jsltri2gXhbA2JOktjP2/CpVWb96iHo/7V2ph0pJ6BTqocy6SYteWyIEcHEfbctBcpsR9B688Wf5vwgVdWAxJYNDWL+auNzblZepH+6Ns+75qwGbB0lLly5FYGAgHB0dERwcjAMHLBRa04uJiUFwcDAcHR0RFBSEr7/+2uT+5cuXo1evXnB3d4e7uzsGDBiAo0ePmhwzb948SJJk8s/b27zwHWNWo1LTMMjT/wGe/RYYtwWY8BswNY56aYonLDcfDAyYS9s736XgaO8n9MfIqw3VE0pNpKUUyqIwz/iYHtPMh5TKw82XAhILtZEUOboZl2zx6VC6D/duk+g29jvzb+dyD13WLaBeS+C1PRTo9HwLCN8GvJtMw4Ph28yTi4uev80o6mmqVV/5GMOQWxTNDLwZB0AAkGhW3B9F/h4Zhtr6mS7PIg+5XYqhnoqCB9TbUXyIrM1I+iBOv2T8wC9KCOrZAahHsGFn5WsuL3nI7cafwMN0mgH5+n7aX2y5mRKVZ4Zb2lkaagMoyMy5W7b3uk5LvY4A0KpIIGMIkvQJ8SknqGfQsQ7lYcmay0NuCnlJ8fqevTbPWO79tHOkop0A1fiSC1UWl36JZpCq7Ki3C6D3ntyrdHKTpVdYOhnXjeU1ACBhHfWwlkZWKvXKykPpxQlhzM+r5UUzHL8fXvrSETWETYOkDRs2YPr06ZgzZw7i4+PRq1cvhIWF4epV5TfU5cuXMWTIEPTq1Qvx8fGYPXs23nzzTWzaZHwj7du3D6NHj8bevXtx+PBh+Pv7Y+DAgbhx44bJuVq3bo1bt24Z/iUmJlbqa2UM9ZpTcnLrZ+iDs0EwLWFi6QOnx3Sg7XPUe5Jzl5YLGfUN8PoBYLj+w/jof2lRzuKEoG/exXM5ji6jb8G1vIEeb1rz1ZWdnGckL/vxKM0G05BJ7n0afrl/jeodfRMGrH2ego2gPsCrvxqn7svsXegD0tnD8vntnIBRK6iukCUB3Sm52qU+DYsMjwRmnKFhJ4DKI1zVD+XI6/XJQ22ywD50m/y7MZE/qLfp7EIAcKhlnNm37U36vRV1ZhsNFWmcaJq/tfl10g+LShRQT9hTvjIY5ZnhFrMQgKA8p5EraF/896YFSEty7Q+aVehYG2j0hHF/8bwk+Xfl3820/ZsMoMDl7gXj2o0A9eie/om2H5V/1/Jpar/CXMsBnjxs27CraS+sPOR24oeKzeg8sJh6MwN6GJPxS1t/689v6P0V/aFyOYQ756l91PbAxBj6f5GbQakAlbmcUBWzaZC0ePFivPrqq5gwYQJatmyJJUuWoGHDhoiMVE7O/Prrr+Hv748lS5agZcuWmDBhAl555RUsWrTIcMyaNWswefJkdOjQAS1atMDy5cuh0+mwZ49pqXeNRgNvb2/Dv3r16lXqa2WszCQJePoLoPdM4G9LgTeOUu+CSgU0Gwh000/R/3ky5eLIrh4BlvcDPgsEvupCxSKzUmkWTYx+8dL+75d+aKyydI0AZl4rfR6TSg10fZ22d80ClrQBfp1Fwx8QNDw15kf6YKwsajsKwt5JouGmDi9ST1rX1yn41RXS9O+b8fQhIqmMs6pkvh0o/ykvw1jZuvhQm6zPbBqOzc2g88oFNQvzqWQBAHSfAtT2s/5rlSTgxR+At5MouV9jX77zlLWgZOop4zBOn1mUKyQPPW6fblpUVC6fEPutafKwXECy+RDT664TQHlSugIqgCnnI8lDoDJHN+Pv7dwvNOx1JBJYPYp6nuo2pTpaJZEkGoKWVPR6rhwyP6Z4PpKsxVOUu3Y3yRhwaAto7bkL0ZaHX4u6f81YvbvvbJrRCdAXpUc9vmgpjLxM5Rmd8tqWgb0BNx9g7I/6mX33K5YzWc1Y6HeufPn5+YiNjcXMmTNN9g8cOBCHDim8mQAcPnwYAweaVvIdNGgQVqxYgYKCAtjZmRewy8nJQUFBATw8TL9BJiUlwdfXFw4ODujatSs+/fRTBAUFWbzevLw85OUZ/3NmZlL9kIKCAhQUWHcGknw+a5+Xmav+ba0Ber5NmzpBf9xlvWdDk3wAUsoJ6DZNgHboEqj3fQLVmZ+Nx9w5D0R9ABH9IVDHH1JeBkT9NihsNQqo4tes2NZqp7JdR5sXoNn7KaT8bAhJBdGwG0SLp6BrPhRw8wN0MG2jqhS2GJqUREh3L0B8/wwkADrfYGg1tcxeozqgO1TndwIP7wEACvx7WmgHCXjmf9Cs6AvpZjy0O2dCN/gzqI4ugzr9EoRLfRR2mWz2WKu9ryU7wKFOxd4rdQJhB0DcTUJhKc6j3rsAKgC6ln+D1qMZPXfvOdCc/QXSnfPQHvgcuu7TIZ3dBvWh/4OUahwFEPVaQNc0DKrTP0ECUNhsCIRS2yf+AO3FGKiuHqHj/DqbHadqOgjqi3sgfpsPKXqeyX2FPd6CKKTZeiW2dd0WUHUYC3X8Kogd/0DhK1GASv+xq9NCc3k/PX9AL9PnVztB3XQQVGd+hm7bdEClhpR6ElIh5SBqO0+EbuCnJbajKuYzqHUF0DXqBa1fV8CzFTTR8yDdvYDCpD0QxQOzolJPwq5Iz58ubhW0LZ8xOUR9ZhtUAAqbhdG1qxyBFzZAvWYEVLfioTvwObRDPy/xGsvKWu/rsjzeZkHSnTt3oNVq4eXlZbLfy8sLKSnKBfdSUlIUjy8sLMSdO3fg4+Nj9piZM2fCz88PAwYMMOzr2rUrVq1ahWbNmiE1NRXz589H9+7dcerUKdStq1y6f8GCBfjwww/N9u/evRvOziXMyqmAqKhqWKH5MVVT29rFfSz6pH0AzdVDkCK7QoKAgIQrdXvjgtdQ1M0+C/+7+1H3QRJwj75pH3Ibiju7frXZNVe0rWsHvgPX3JtIc22LfDs34DaA28cB2L6L37XeK3ji3jxo9MHPeW0DnNthPhwamFMXchZWtoM39hw8AcDyCvX1fV5B6KV/Qx37DU6matHy1o9QAzjuMRRX9ljO46wO72u1Ng9PAZBy7iJ66wbka1wtHuuWcwV9z22DgIR9oiuyirSdX71RCHmwFNi/CHlHvkGtPKojVKiyR4ZTANwfXITq9lmob5/V73fEzvN50F0wbX//DFd0BJB37Ds4F6RDK2mw83gKdImmxznm2+NJSQ2VrhA6qHG3VnOk1m6PVLcOyL5WC7hmeryltrYv6IIB6h9hl5qIO0uH4s9Gk6BT2aPOg4vonZuBArUzdibchDhuWhfJO7cRugJQ3Yo37CtQOcFO9xDqY8sQn6bCDY/uis/plHcbA05T7tRBuyeQrm/HtrVDEXQ7Crd3fIKjQZYrire8+QOaAUh3bgz3nEtQJR9A9Jbv8NChnr5t7mLQrQQISIi+aoe8W8a2qOsShp6Ih+7ERvyq6wWt2tHCs5RfRd/XOTmPqKZehM2CJJlULB9DCGG271HHK+0HgM8++wzr1q3Dvn374Oho/EWFhYUZttu2bYvQ0FA0btwY3333HWbMmKH4vLNmzTK5LzMzEw0bNsTAgQPh5uZWwissu4KCAkRFReHJJ59U7B1j1vNYtHViLWDrZEgQ0DV6AtoBH8PPqzWKDsAU3EmC6uSPQC0vdJETcqvYY9HWpXHSHfiZksybDJqIxkrDMreDgGU0jdypzVAMGTzkEScdAu1eHdSHPkf765SIK+q1QOuxC9BaZf5nvLq1tUj+EFLmdTzZCIADIN1PBu5dAeycIHw7QfiFAG6+UG+kPB/Rajh6PTOx2EnCoFt/FupLv6FWXiqEkzt0Ia9BhEyAm7MHtA/vQ3cxCqrzuyBdPQyp40sY3Hu4+cXcawUsXQHnApotJ/kFY/BTCscB0IU0h+7BbYhGvVDHwRV1ABRf5ro0bS218oDY9DJ8MmIxNGMVtM+ugurYOeA8oG7SF2FDFWpM6QZBe0AD5GdD+HSE8OkAeARBG7MQ6oOLEXzjO7R/cjTN/CxG/ct0qKCFLrA3uj1XJOfvTlPgv1HwzkjAkO5tzHP3AEAIaJbSrFe3QbMg4ldBSt6P/nVToHuCZqGqjv0POAWIBl3Q/2+jiz0+DCJyHTT3LmNww4cQHRTKaSi1UeIPkPKyoAu2PDHAWu9reSSoNGwWJHl6ekKtVpv1GqWlpZn1Fsm8vb0Vj9doNGY9QIsWLcKnn36K6OhotGtX8swZFxcXtG3bFklJlsfMHRwc4OBgXlDQzs6u0v4IVea5maka3dadxgDOdQA7Z6iC+kCl9AfGpxXgQ8mjpVzFq9LU6LYujY4vAvnZQHYqNI1Clf/g+7Sh5PnsFKib9oe6NO3R/z3g5p9UPwuANHA+7BwsTL3XqzZt7dkUyLwOzbY3LB/j6kOzEyUVVH1nQ6V03cO/ojphPu0hdQqH2qGW8f1sV4/avuOLAOh9rvher9eE8pIyrwMAVAGhys8FAIHKPTVKSmzrVkMBx43AutFQJe+Hau0oQ3kDVeO+Fp7fDhigkPDd/z0g5Tiki3tgt2k8MHEfFbyVpZ4CTlBytqrvHNNz+7QCgvpCurQXdgnfKReSvR5LNa3sXKBpOQSQACTvh/rEBqj7zqacyCTqOVK1GqZ87cHhQPQ8aI6vATqPV26Tov5cSflmANSOrkDHMSUeXtH3dVkea7PEbXt7ewQHB5t1m0VFRaF7d+U3ZmhoqNnxu3fvRkhIiMmL/te//oWPP/4Yu3btQkhIyCOvJS8vD2fOnFEcrmOsRmgxlOqslGVqNqs8XSdScryl34ckASP+C/R7H2gWpnxMcWoNzfTy7UQzq5oMePRjqotWT1MCs5M7XX/rEUDPGVROwLsdVXTPukXHtn3W8iw6N19gxDJaV9ChVvmupWi9JIBmtlWFoN5UgsLJncoq3NTXFFJaEqgkKjVNGqjjT0PoW16nauZJ0cCa52iNRV0hnde/q/nj5ckPcauUK8fLCdvNB9PkjpZP0USDjKtA8n5j8VSAEsyVtH+RfqfXjwJpZ0p+Pcm/AzveNv68axYVwKwmbDrcNmPGDIwbNw4hISEIDQ3FsmXLcPXqVUREUHXdWbNm4caNG1i1ijL0IyIi8OWXX2LGjBl47bXXcPjwYaxYsQLr1q0znPOzzz7D+++/j7Vr16JRo0aGnqdatWqhVi36T/X2229j2LBh8Pf3R1paGubPn4/MzEyEhz+ioB1jjFlLUB/zWU2P4uoFTHzEwqvVUcgrFNipLXyDz39As7jSLykX8rS2Rj0NvS1mM9sqU4NgYPwOKryYnUI1oDwsTxiyyNkDeO574JtBwPldwL+bUZkQWeN+wLAvlB/bdCAFWPevUimN4PHG+3Q64KQ+SJKLido5AW1H0kzM+NW0xI7QUl0vj0Dl53D1ojXwzm6ndSkHW0gyT78MbBhHQV3rZ2gI9mYcsG068OKGavGlz6YlAJ5//nksWbIEH330ETp06ID9+/djx44dCAigtXBu3bplUjMpMDAQO3bswL59+9ChQwd8/PHH+OKLLzBy5EjDMUuXLkV+fj5GjRoFHx8fw7+iZQKuX7+O0aNHo3nz5hgxYgTs7e1x5MgRw/MyxhizMksBEkA9FgHdaZkPB8uJ3VbT9EkqIBnUt+TaWZXBqxXwyk6g+VCg/9zyBwK+HYChi2k75y719nSdBEyJpWK1dRoqP06lBjrrywH8Ose0ptG1P6hKuIObaU+lvPzK6a3GZXLkwqqWyAtWH19nWrZBlptJC3k/TKcleIZHAsOXUt2lpF/LtxRNJbB54vbkyZMxefJkxfu+/fZbs329e/dGXJzl0ufJycmPfM7160tZTIsxxtjjx9UbeOukfrkaG/AIAkavrfh5Oo6hxbcLc6knprS1z7q+DiTtpvy2Nc8Cr+6m9fjkCt8tnjJd1Nu3E1C/FZB22pATZ3GoTda4P+DqS0HXuR10fTKdlpbxuX2WctFeWEc9VvVb0hqRez6idSuD+tDvyoZsviwJY4wxVuUcXE0DgZqq7Sjq6SlLcViNA/DCGhoyy04FVo8EstOM1cSLL/AsSaaL+dbxf/RCymqNMQFbLmoJUL7R2ucpSNM40nW4FckH7j6NlirKzaBht4pUHLcCDpIYY4yxvxrH2lShvnZDWl5kWV/gwW1KLFfKlWv3vLEQZothpRsmlAOri3sp3yh+DfBVN1r7UG0PPPNfwC/Y9DFqDQ29qeyA8zv1CzjbDgdJjDHG2F+Rmw8wdhPlZ+lLIqDl08r5Yy6elGekcXzkFH0D90b6gEtQkvnPk2k5Hr9gWoPS0pJEXq2APu/S9uGvbNqbZPOcJMYYY4zZSL3mNJNs1d8ot6ntKMvHDvk3LeZclmHKjuNoId+sW9R71GcW0P1N6jEqSY/pVEagy0SbznLjIIkxxhj7K/PvBry8gxalLb4gc1EqFaAqYx5Xy2GUxK0rBML+ScnZpaG2A3opr4BRlThIYowxxv7q/ILN84OsQeMAjNts/fNWEc5JYowxxhhTwEESY4wxxpgCDpIYY4wxxhRwkMQYY4wxpoCDJMYYY4wxBRwkMcYYY4wp4CCJMcYYY0wBB0mMMcYYYwo4SGKMMcYYU8BBEmOMMcaYAg6SGGOMMcYUcJDEGGOMMaaAgyTGGGOMMQUcJDHGGGOMKdDY+gJqKiEEACAzM9Pq5y4oKEBOTg4yMzNhZ2dn9fMzI27rqsNtXXW4rasOt3XVsVZby5/b8ud4SThIKqesrCwAQMOGDW18JYwxxhgrq6ysLNSuXbvEYyRRmlCKmdHpdLh58yZcXV0hSZJVz52ZmYmGDRvi2rVrcHNzs+q5mSlu66rDbV11uK2rDrd11bFWWwshkJWVBV9fX6hUJWcdcU9SOalUKjRo0KBSn8PNzY3/01URbuuqw21ddbitqw63ddWxRls/qgdJxonbjDHGGGMKOEhijDHGGFPAQVI15ODggLlz58LBwcHWl/LY47auOtzWVYfbuupwW1cdW7Q1J24zxhhjjCngniTGGGOMMQUcJDHGGGOMKeAgiTHGGGNMAQdJjDHGGGMKOEiqZpYuXYrAwEA4OjoiODgYBw4csPUl1XgLFixA586d4erqivr162P48OE4d+6cyTFCCMybNw++vr5wcnJCnz59cOrUKRtd8eNjwYIFkCQJ06dPN+zjtraeGzduYOzYsahbty6cnZ3RoUMHxMbGGu7ntraOwsJCvPfeewgMDISTkxOCgoLw0UcfQafTGY7hti6f/fv3Y9iwYfD19YUkSfjpp59M7i9Nu+bl5WHq1Knw9PSEi4sLnn76aVy/ft06FyhYtbF+/XphZ2cnli9fLk6fPi2mTZsmXFxcxJUrV2x9aTXaoEGDxMqVK8XJkydFQkKCGDp0qPD39xfZ2dmGYxYuXChcXV3Fpk2bRGJionj++eeFj4+PyMzMtOGV12xHjx4VjRo1Eu3atRPTpk0z7Oe2to709HQREBAgxo8fL/744w9x+fJlER0dLS5cuGA4htvaOubPny/q1q0rtm/fLi5fviw2btwoatWqJZYsWWI4htu6fHbs2CHmzJkjNm3aJACILVu2mNxfmnaNiIgQfn5+IioqSsTFxYm+ffuK9u3bi8LCwgpfHwdJ1UiXLl1ERESEyb4WLVqImTNn2uiKHk9paWkCgIiJiRFCCKHT6YS3t7dYuHCh4Zjc3FxRu3Zt8fXXX9vqMmu0rKws0bRpUxEVFSV69+5tCJK4ra3n3XffFT179rR4P7e19QwdOlS88sorJvtGjBghxo4dK4TgtraW4kFSadr1/v37ws7OTqxfv95wzI0bN4RKpRK7du2q8DXxcFs1kZ+fj9jYWAwcONBk/8CBA3Ho0CEbXdXjKSMjAwDg4eEBALh8+TJSUlJM2t7BwQG9e/fmti+nN954A0OHDsWAAQNM9nNbW8/WrVsREhKCZ599FvXr10fHjh2xfPlyw/3c1tbTs2dP7NmzB+fPnwcAHD9+HL///juGDBkCgNu6spSmXWNjY1FQUGByjK+vL9q0aWOVtucFbquJO3fuQKvVwsvLy2S/l5cXUlJSbHRVjx8hBGbMmIGePXuiTZs2AGBoX6W2v3LlSpVfY023fv16xMXF4dixY2b3cVtbz6VLlxAZGYkZM2Zg9uzZOHr0KN588004ODjgpZde4ra2onfffRcZGRlo0aIF1Go1tFotPvnkE4wePRoAv68rS2naNSUlBfb29nB3dzc7xhqfnRwkVTOSJJn8LIQw28fKb8qUKThx4gR+//13s/u47Svu2rVrmDZtGnbv3g1HR0eLx3FbV5xOp0NISAg+/fRTAEDHjh1x6tQpREZG4qWXXjIcx21dcRs2bMDq1auxdu1atG7dGgkJCZg+fTp8fX0RHh5uOI7bunKUp12t1fY83FZNeHp6Qq1Wm0W+aWlpZlE0K5+pU6di69at2Lt3Lxo0aGDY7+3tDQDc9lYQGxuLtLQ0BAcHQ6PRQKPRICYmBl988QU0Go2hPbmtK87HxwetWrUy2deyZUtcvXoVAL+vremdd97BzJkz8cILL6Bt27YYN24c3nrrLSxYsAAAt3VlKU27ent7Iz8/H/fu3bN4TEVwkFRN2NvbIzg4GFFRUSb7o6Ki0L17dxtd1eNBCIEpU6Zg8+bN+O233xAYGGhyf2BgILy9vU3aPj8/HzExMdz2ZdS/f38kJiYiISHB8C8kJARjxoxBQkICgoKCuK2tpEePHmalLM6fP4+AgAAA/L62ppycHKhUph+XarXaUAKA27pylKZdg4ODYWdnZ3LMrVu3cPLkSeu0fYVTv5nVyCUAVqxYIU6fPi2mT58uXFxcRHJysq0vrUabNGmSqF27tti3b5+4deuW4V9OTo7hmIULF4ratWuLzZs3i8TERDF69GievmslRWe3CcFtbS1Hjx4VGo1GfPLJJyIpKUmsWbNGODs7i9WrVxuO4ba2jvDwcOHn52coAbB582bh6ekp/vGPfxiO4bYun6ysLBEfHy/i4+MFALF48WIRHx9vKH1TmnaNiIgQDRo0ENHR0SIuLk7069ePSwA8rr766isREBAg7O3tRadOnQzT1Fn5AVD8t3LlSsMxOp1OzJ07V3h7ewsHBwfxxBNPiMTERNtd9GOkeJDEbW0927ZtE23atBEODg6iRYsWYtmyZSb3c1tbR2Zmppg2bZrw9/cXjo6OIigoSMyZM0fk5eUZjuG2Lp+9e/cq/n0ODw8XQpSuXR8+fCimTJkiPDw8hJOTk3jqqafE1atXrXJ9khBCVLw/ijHGGGPs8cI5SYwxxhhjCjhIYowxxhhTwEESY4wxxpgCDpIYY4wxxhRwkMQYY4wxpoCDJMYYY4wxBRwkMcYYY4wp4CCJMcYqQJIk/PTTT7a+DMZYJeAgiTFWY40fPx6SJJn9Gzx4sK0vjTH2GNDY+gIYY6wiBg8ejJUrV5rsc3BwsNHVMMYeJ9yTxBir0RwcHODt7W3yz93dHQANhUVGRiIsLAxOTk4IDAzExo0bTR6fmJiIfv36wcnJCXXr1sXEiRORnZ1tcsw333yD1q1bw8HBAT4+PpgyZYrJ/Xfu3MEzzzwDZ2dnNG3aFFu3bjXcd+/ePYwZMwb16tWDk5MTmjZtahbUMcaqJw6SGGOPtffffx8jR47E8ePHMXbsWIwePRpnzpwBAOTk5GDw4MFwd3fHsWPHsHHjRkRHR5sEQZGRkXjjjTcwceJEJCYmYuvWrWjSpInJc3z44Yd47rnncOLECQwZMgRjxoxBenq64flPnz6NnTt34syZM4iMjISnp2fVNQBjrPysskwuY4zZQHh4uFCr1cLFxcXk30cffSSEEAKAiIiIMHlM165dxaRJk4QQQixbtky4u7uL7Oxsw/2//PKLUKlUIiUlRQghhK+vr5gzZ47FawAg3nvvPcPP2dnZQpIksXPnTiGEEMOGDRMvv/yydV4wY6xKcU4SY6xG69u3LyIjI032eXh4GLZDQ0NN7gsNDUVCQgIA4MyZM2jfvj1cXFwM9/fo0QM6nQ7nzp2DJEm4efMm+vfvX+I1tGvXzrDt4uICV1dXpKWlAQAmTZqEkSNHIi4uDgMHDsTw4cPRvXv3cr1WxljV4iCJMVajubi4mA1/PYokSQAAIYRhW+kYJyenUp3Pzs7O7LE6nQ4AEBYWhitXruCXX35BdHQ0+vfvjzfeeAOLFi0q0zUzxqoe5yQxxh5rR44cMfu5RYsWAIBWrVohISEBDx48MNx/8OBBqFQqNGvWDK6urmjUqBH27NlToWuoV68exo8fj9WrV2PJkiVYtmxZhc7HGKsa3JPEGKvR8vLykJKSYrJPo9EYkqM3btyIkJAQ9OzZE2vWrMHRo0exYsUKAMCYMWMwd+5chIeHY968ebh9+zamTp2KcePGwcvLCwAwb948REREoH79+ggLC0NWVhYOHjyIqVOnlur6PvjgAwQHB6N169bIy8vD9u3b0bJlSyu2AGOssnCQxBir0Xbt2gUfHx+Tfc2bN8fZs2cB0Myz9evXY/LkyfD29saaNWvQqlUrAICzszN+/fVXTJs2DZ07d4azszNGjhyJxYsXG84VHh6O3NxcfP7553j77bfh6emJUaNGlfr67O3tMWvWLCQnJ8PJyQm9evXC+vXrrfDKGWOVTRJCCFtfBGOMVQZJkrBlyxYMHz7c1pfCGKuBOCeJMcYYY0wBB0mMMcYYYwo4J4kx9tjibALGWEVwTxJjjDHGmAIOkhhjjDHGFHCQxBhjjDGmgIMkxhhjjDEFHCQxxhhjjCngIIkxxhhjTAEHSYwxxhhjCjhIYowxxhhTwEESY4wxxpiC/wdTwU36smX6wwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['root_mean_squared_error'], label='Training RMSE')\n",
    "plt.plot(history.history['val_root_mean_squared_error'], label='Validation RMSE')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('RMSE')\n",
    "plt.title('Training and Validation RMSE Curves')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7632ae62-9648-4742-8fdc-426cfd66001d",
   "metadata": {},
   "source": [
    "Use the model to predict closing prices on the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "373a668a-c8d3-49bf-a79d-6b7079038f3a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "y_pred = model.predict(X_test,\n",
    "                       batch_size=32,\n",
    "                       verbose=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe7029fd-c279-4f0c-beff-adfba63142a0",
   "metadata": {},
   "source": [
    "Inverse transform the predicted values using the same scaler that we used before to get back the original scale of the close price."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "dc70f2a4-e847-45b1-b260-f64e32ae596c",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "y_pred = scaler.inverse_transform(y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ee8ac75-ef68-4b80-a35e-abebce0d4297",
   "metadata": {},
   "source": [
    "Inverse transform the test data. Since it is 1-dimensional, select only the first column of the transform."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "78ca4554-8fff-4401-a14d-dbec5625dabe",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "y_test_inv = np.transpose(scaler.inverse_transform([y_test]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f19f1f37-07fd-4a0b-92f2-054f239da41f",
   "metadata": {},
   "source": [
    "Plot the actual and predicted prices over the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "a9ff93b0-6460-4274-9384-5b7af09d7f85",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAioAAAHFCAYAAADcytJ5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAACi5klEQVR4nOzdd3hU1dbA4d+ZyaRX0gMhgZCE3os06UhTUVEUC4oF26der72iYvdar/1SbNgbVUBEQHrvvQRIJ6TXycz5/jiZgZA2k8yksd7nyTMnM6fsnLQ1e6+9tqKqqooQQgghRCOka+gGCCGEEEJURQIVIYQQQjRaEqgIIYQQotGSQEUIIYQQjZYEKkIIIYRotCRQEUIIIUSjJYGKEEIIIRotCVSEEEII0WhJoCKEEEKIRksCFVGt999/H0VR6Ny5c63PkZSUxIwZM9ixY4fjGlaNoUOHMnTo0Hq5VnWio6NRFMX64e3tTb9+/fjyyy/r5fpz585FURROnDhhfa629+aVV17ht99+c1jbLE6cOIGiKMydO9em/Y8dO8b9999PXFwcHh4eeHp60qlTJ5555hkSExOt+916661ER0c7vL2OMGPGjHI/F66urrRp04YHH3yQrKwsm87RUD/j33//PZ06dcLDwwNFUertd3r+/PkoikJgYCDFxcWV7mP5favqvnz55ZfWe/73339XeH3NmjVcd911tGzZEldXV/z8/BgwYAAff/wx+fn5DvxqhL0kUBHVmj17NgB79+5l48aNtTpHUlISL7zwQr39UWtMBg4cyPr161m/fr01cJg6dSoff/xxg7Tno48+4qOPPrL7OGcFKvZYuHAhXbt2ZeHChdx1110sXLjQur1gwQImTJjQoO2z1x9//MH69etZtGgREydO5IMPPmDs2LHYsqpJbb+PdZGens7NN99MTEyMte1xcXH1cu1Zs2YBcPbs2Wp/Dn18fFi9ejVHjx6t8Nrs2bPx9fWt9Ljnn3+eSy+9lMTERF566SWWL1/Od999x4gRI5gxYwbPPPOMQ74OUUuqEFXYvHmzCqjjx49XAfXOO++s03nmzJnj2AZWYciQIeqQIUPq5VrViYqKUsePH1/uuczMTNXX11dt165dlceVlpaqRUVFdb7+nDlzVEA9fvx4nc/l5eWlTp06tc7nudDx48dt+tk4duyY6uXlpfbo0UPNysqq8LrZbFZ//vln6+dTp05Vo6KiHNxax3j++edVQE1PTy/3/M0336wC6j///FPlsfn5+c5uXpX++ecfFVC///57h53Tlq8nOTlZdXFxUYcPH666u7uro0aNqnS/qKgodezYsWqrVq3Up556qtxrR44cURVFUe+8804VUFeuXGl97YcfflAB9fbbb1fNZnOF8+bk5KhLly617wsTDiU9KqJKlncxr732GgMGDOC7776joKCgwn6JiYncddddREZG4urqSkREBJMmTSI1NZW///6bPn36AHDbbbdZu15nzJgBVN2FXVnX/QsvvEC/fv1o0aIFvr6+9OzZk1mzZtn0DvRCEydOJCoqCrPZXOG1fv360bNnT+vnP/74I/369cPPzw9PT0/atm3LtGnT7L4mgL+/P/Hx8SQkJADnhj7eeOMNZs6cSZs2bXBzc2PlypUAbNmyhSuuuIIWLVrg7u5Ojx49+OGHHyqcd8OGDQwcOBB3d3ciIiJ48sknMRqNFfar7H4XFxfz4osv0qFDB9zd3QkMDGTYsGGsW7cOAEVRyM/P54svvrB+/84/R0pKCtOnT6dVq1bWYYwXXniB0tLSctdJSkriuuuuw8fHBz8/PyZPnkxKSopN9+3tt98mPz+fjz76CD8/vwqvK4rC1VdfXe05ioqKePLJJ2nTpg2urq60bNmS++67r8Jwy19//cXQoUMJDAzEw8OD1q1bc80115T72S8pKWHmzJm0b98eNzc3goODue2220hPT7fp66nMJZdcAmD92Rg6dCidO3dm9erVDBgwAE9PT+vPXW2+jwCqqvLRRx/RvXt3PDw8CAgIYNKkSRw7dqzatt16660MGjQIgMmTJ1f4GZg/fz79+/fH09MTHx8fRo0axfr168udwzLktW3bNiZNmkRAQAAxMTE13pcvvviC0tJS/vWvf3H11VezYsUK6z26kE6n45ZbbuGLL74o97s9e/ZsIiMjGTlyZIVjXnzxRQICAqzD3Bfy8fFh9OjRNbZTOI9LQzdANE6FhYV8++239OnTh86dOzNt2jTuuOMOfvzxR6ZOnWrdLzExkT59+mA0Gnnqqafo2rUrGRkZLF26lMzMTHr27MmcOXO47bbbeOaZZxg/fjwArVq1srtNJ06cYPr06bRu3RrQ/jn/3//9H4mJiTz33HN2nWvatGlceeWV/PXXX+X+eB04cIBNmzbx/vvvA7B+/XomT57M5MmTmTFjBu7u7iQkJPDXX3/Z3X4Ao9FIQkICwcHB5Z5///33iYuL46233sLX15fY2FhWrlzJmDFj6NevH5988gl+fn589913TJ48mYKCAm699VYA9u3bx4gRI4iOjmbu3Ll4enry0UcfMW/evBrbU1paytixY1mzZg0PPfQQw4cPp7S0lA0bNnDy5EkGDBjA+vXrGT58OMOGDePZZ58FsHahp6Sk0LdvX3Q6Hc899xwxMTGsX7+emTNncuLECebMmQNoP08jR44kKSmJV199lbi4OBYtWsTkyZNtum/Lli0jNDTU+s/cXqqqMnHiRFasWMGTTz7J4MGD2bVrF88//7x1aM7NzY0TJ04wfvx4Bg8ezOzZs/H39ycxMZE//viDkpISPD09MZvNXHnllaxZs4bHHnuMAQMGkJCQwPPPP8/QoUPZsmULHh4edrfxyJEjAOV+NpKTk7npppt47LHHeOWVV9DpKn9vacv3EWD69OnMnTuXBx54gNdff52zZ8/y4osvMmDAAHbu3EloaGil53/22Wfp27cv9913H6+88grDhg2z/gzMmzePG2+8kdGjR/Ptt99SXFzMG2+8wdChQ1mxYoU1wLG4+uqruf7667n77rttyv2YPXs24eHhjB07Fg8PD+bNm8fcuXN5/vnnK91/2rRpvPrqqyxdupSxY8diMpn44osvuP322yvcv+TkZPbs2cPkyZPx9PSssS2igTRwj45opL788ksVUD/55BNVVVU1NzdX9fb2VgcPHlxuv2nTpqkGg0Hdt29fleeqbuinqmGamrruTSaTajQa1RdffFENDAws12Vry9CP0WhUQ0ND1SlTppR7/rHHHlNdXV3VM2fOqKqqqm+99ZYKVDrcUJOoqCh13LhxqtFoVI1Go3r8+HF16tSpKqA++uijqqqeG/qIiYlRS0pKyh3fvn17tUePHqrRaCz3/IQJE9Tw8HDVZDKpqqqqkydPVj08PNSUlBTrPqWlpWr79u0rDP1ceG8s3+fPP/+82q+lqqGf6dOnq97e3mpCQkK55y33be/evaqqqurHH3+sAurvv/9ebj9LV3xNQz/u7u7qJZdcUu0+57vw5+ePP/5QAfWNN94ot9/333+vAupnn32mqqqq/vTTTyqg7tixo8pzf/vttypQbqhJVc/9nH/00UfVts0y9JOSkqIajUY1MzNT/frrr1UPDw81MjJSLSwsVFVV+14B6ooVKyqcozbfx/Xr16uA+p///Kfc86dOnVI9PDzUxx57rNp2r1y5UgXUH3/80fqcyWRSIyIi1C5dulh/HlVV+3sREhKiDhgwoMLX/dxzz1V7nfOtXr1aBdQnnnhCVVVtiK9NmzZqVFRUhWGa84dahwwZok6aNElVVVVdtGiRqiiKevz4cfXHH38sN/SzYcOGcucXjZMM/YhKzZo1Cw8PD66//noAvL29ufbaa1mzZg2HDx+27rdkyRKGDRtGhw4dnN4mS++Hn58fer0eg8HAc889R0ZGBmlpaXady8XFhZtuuolffvmF7OxsAEwmE1999RVXXnklgYGBANZhq+uuu44ffvih3MwSWyxevBiDwYDBYKBNmzb88MMP/N///R8zZ84st98VV1yBwWCwfn7kyBEOHDjAjTfeCGjvmC0f48aNIzk5mYMHDwKwcuVKRowYUe7dsF6vt6m3YsmSJbi7u9d6KGvhwoUMGzaMiIiIcm0cO3YsAKtWrbK20cfHhyuuuKLc8VOmTKnVde1l6QGz9EJZXHvttXh5ebFixQoAunfvjqurK3fddRdffPFFpUMiCxcuxN/fn8svv7zc19y9e3fCwsIqnVFSmbCwMAwGAwEBAdx000307NmTP/74A3d3d+s+AQEBDB8+vMZz2fJ9XLhwIYqicNNNN5Vrd1hYGN26dbO53ec7ePAgSUlJ3HzzzeV6K7y9vbnmmmvYsGFDheHia665xubzW4afLV+XoijceuutJCQkWL9nlZk2bRrz588nIyODWbNmMWzYsEY7C0zUTAIVUcGRI0dYvXo148ePR1VVsrKyyMrKYtKkScC5mUCgzQSozTCOvTZt2mQdJ/78889Zu3Ytmzdv5umnnwa0oQV7TZs2jaKiIr777jsAli5dSnJyMrfddpt1n0svvZTffvuN0tJSbrnlFlq1akXnzp359ttvbbrGoEGD2Lx5M1u2bGHfvn1kZWXx/vvv4+rqWm6/8PDwcp+npqYC8Mgjj1gDHcvHvffeC8CZM2cAyMjIICwsrMK1K3vuQunp6URERFQ5pFCT1NRUFixYUKGNnTp1qtDGyoYVbGkjQOvWrTl+/Hit2mi5vouLS4UhN0VRCAsLIyMjA4CYmBj+/PNPQkJCuO+++4iJiSEmJob33nvPekxqaipZWVm4urpW+LpTUlKsX3NN/vzzTzZv3syOHTs4c+YM//zzDx07diy3z4U/F1Wx5fuYmpqKqqqEhoZWaPeGDRtsbvf5LPetsnZGRERgNpvJzMws97ytX1Nubi4//vgjffv2JTg42Pp36KqrrkJRFGsQU5lJkybh7u7OO++8w4IFC7j99tsr3c8yjFyXny3hfJKjIiqYPXs2qqry008/8dNPP1V4/YsvvmDmzJno9XqCg4M5ffp0ra/l7u5u7dE434V/NL/77jsMBgMLFy4s946zLlNmO3bsSN++fZkzZw7Tp09nzpw5REREVEicu/LKK7nyyispLi5mw4YNvPrqq0yZMoXo6Gj69+9f7TX8/Pzo3bt3jW25MIkvKCgIgCeffLLKJNH4+HgAAgMDK01KtSVRNTg4mH/++Qez2VyrYCUoKIiuXbvy8ssvV/p6RESEtY2bNm2qVRsBLrvsMj744AM2bNhQqzyVwMBASktLSU9PLxesqKpKSkqKtecMYPDgwQwePBiTycSWLVv44IMPeOihhwgNDeX6668nKCiIwMBA/vjjj0qv5ePjY1ObunXrZv0+V6Wy5M7K2PJ9DAoKQlEU1qxZg5ubW4XXK3uuJpaex+Tk5AqvJSUlodPpCAgIKPe8rV/Tt99+S0FBAZs2bapwDoBff/2VzMzMSl/z9PTk+uuv59VXX8XX17fK36Hw8HC6dOnCsmXLKCgokDyVRkp6VEQ5lsSzmJgYVq5cWeHj3//+N8nJySxZsgSAsWPHsnLlSuswRGUsfwAr6/WIjo7m0KFD5Yo4ZWRklJupANofNxcXF/R6vfW5wsJCvvrqqzp9vbfddhsbN27kn3/+YcGCBUydOrXcNS78OoYMGcLrr78OwPbt2+t07erEx8cTGxvLzp076d27d6Ufln+Iw4YNY8WKFdZeGNC+j99//32N1xk7dixFRUU1Flxzc3Or9Ps3YcIE9uzZQ0xMTKVttAQqw4YNIzc3l/nz55c73paEX4B//etfeHl5ce+991Ya2Kqqyq+//lrl8SNGjADg66+/Lvf8zz//TH5+vvX18+n1evr168eHH34IwLZt26xfc0ZGBiaTqdKv2RJA1idbvo8TJkxAVVUSExMrbXeXLl3svm58fDwtW7Zk3rx55Wbf5efn8/PPP1tnAtXGrFmz8PHxYcWKFRX+Dr355psUFxfzzTffVHn8Pffcw+WXX85zzz1X7s3NhZ599lkyMzN54IEHKp1BmJeXx7Jly2r1NQgHabj0GNEYLViwQAXU119/vdLX09PTVTc3N3XixImqqqrq6dOn1fDwcDUkJER999131RUrVqg///yzeuedd6r79+9XVVWrleDh4aEOHDhQXblypbp582Y1MTFRVdVztRkmTZqkLl26VJ03b57avXt3NSoqqlwy5IoVK6z7LVu2TP3222/VXr16qbGxsTUmjFYnKytL9fDwUFu1aqUC6sGDB8u9/uyzz6q33Xab+vXXX6t///23+ttvv6nDhg1TDQaDumfPnmrPXVkdlQtZkmnffPPNCq/99ddfqpubmzp69Gh13rx56qpVq9Rff/1VfeWVV6yJgqqqqrt371Y9PDzUjh07qt999506f/589bLLLlMjIyNrvDdGo9H69Tz22GPqkiVL1EWLFqnPPfec+u2335Y7LiQkRJ0/f766efNm9cCBA6qqqmpSUpIaFRWltm/fXv3oo4/UFStWqIsWLVI//PBDdfz48eqpU6dUVdV+BuLi4lQ/Pz/1v//9r7p06VL1wQcfVFu3bm1zjZ0FCxaonp6eanR0tPrWW2+pK1asUFesWKF+8MEHao8ePdTu3btb970wmdZsNquXXXaZajAY1BkzZqjLly9X//Of/6je3t5qjx49rHVrPv74Y/Xaa69V586dq/7111/q4sWL1UmTJqmAtZZGaWmpOnbsWLVFixbqCy+8oC5ZskT9888/1blz56pTp05Vf/nll2q/jqrqqFxoyJAhaqdOnap8rTbfx7vuukv19PRUH330UXXBggXqX3/9pX7zzTfqPffcU2MScGXJtKqqqt98840KqOPGjVN///139YcfflD79Omjurq6qmvWrLH761ZV7WcaUO+5555KXy8pKVHDwsLKfc9t+X27MJnW4tlnn1UBdeDAgers2bPVVatWqUuWLFFnzJihhoeHqw899FCNbRbOI4GKKGfixImqq6urmpaWVuU+119/veri4mKdZXLq1Cl12rRpalhYmGowGNSIiAj1uuuuU1NTU63HfPvtt2r79u1Vg8GgAurzzz9vfe2LL75QO3TooLq7u6sdO3ZUv//++0pn/cyePVuNj49X3dzc1LZt26qvvvqqOmvWrDoFKqqqqlOmTLH+kbrQwoUL1bFjx6otW7ZUXV1d1ZCQEHXcuHHl/gBXpa6Biqqq6s6dO9XrrrtODQkJUQ0GgxoWFqYOHz7cOhvLYu3ateoll1yiurm5qWFhYeqjjz6qfvbZZzbdm8LCQvW5555TY2NjVVdXVzUwMFAdPny4um7dOus+O3bsUAcOHKh6enqqQLlzpKenqw888IDapk0b1WAwqC1atFB79eqlPv3002peXp51v9OnT6vXXHON6u3trfr4+KjXXHONum7dOruKAR49elS999571Xbt2qlubm7WAO3hhx8u93VW9vNTWFioPv7442pUVJRqMBjU8PBw9Z577lEzMzOt+6xfv1696qqr1KioKNXNzU0NDAxUhwwZos6fP7/cuYxGo/rWW2+p3bp1U93d3VVvb2+1ffv26vTp09XDhw9X+zU4I1CxfH01fR9VVfs96tevn+rl5aV6eHioMTEx6i233KJu2bKl2vZUFaioqqr+9ttvar9+/VR3d3fVy8tLHTFihLp27dpafd2qqqoPPfRQjbOvnnjiCRVQt27dqqpq3QIVVVXVVatWqZMmTVLDw8NVg8Gg+vr6qv3791fffPNNNScnp8Y2C+dRVLUW1bKEEEIIIeqB5KgIIYQQotGSQEUIIYQQjZYEKkIIIYRotCRQEUIIIUSjJYGKEEIIIRotCVSEEEII0Wg16RL6ZrOZpKQkfHx8bC7LLIQQQoiGpaoqubm5Nq011qQDlaSkJCIjIxu6GUIIIYSohVOnTtW4sG2TDlQsa52cOnUKX19fh57baDSybNkyRo8ejcFgcOi5RUVyv+uX3O/6Jfe7fsn9rl+1ud85OTlERkbatIhnkw5ULMM9vr6+TglUPD098fX1lR/0eiD3u37J/a5fcr/rl9zv+lWX+21L2oYk0wohhBCi0ZJARQghhBCNlgQqQgghhGi0mnSOihBCiObJbDZTUlJSq2ONRiMuLi4UFRVhMpkc3DJxocrut8FgQK/XO+T8EqgIIYRoVEpKSjh+/Dhms7lWx6uqSlhYGKdOnZIaW/Wgqvvt7+9PWFhYnb8HEqgIIYRoNFRVJTk5Gb1eT2RkZI3FwCpjNpvJy8vD29u7VscL+1x4v1VVpaCggLS0NADCw8PrdH4JVIQQQjQapaWlFBQUEBERgaenZ63OYRk2cnd3l0ClHlR2vz08PABIS0sjJCSkTsNA8h0UQgjRaFhyHFxdXRu4JaKuLIGm0Wis03kkUBFCCNHoSG5J0+eo76EEKkIIIYRotCRQEUIIIZoxRVH47bffGroZtSaBihBCCOEg69atQ6/XM2bMGLuOi46O5t1333VOo5o4CVQqoaoqKTlFnClq6JYIIYRoSmbPns3//d//8c8//3Dy5MmGbk6zIIFKJb7akMDgN1fz2wm5PUIIIWyTn5/PDz/8wD333MOECROYO3duudfnz59P7969cXd3JygoiKuvvhqAoUOHkpCQwL/+9S8URbEmoc6YMYPu3buXO8e7775LdHS09fPNmzczatQogoKC8PPzY8iQIWzbts2ZX2a9k//ElWgb5A1ASqFknQshRENSVZWCklK7PwpLTLU67vwPVVXtauv3339PfHw88fHx3HTTTcyZM8d6jkWLFnH11Vczfvx4tm/fzooVK+jduzcAv/zyC61ateLFF18kOTmZ5ORkm6+Zm5vL1KlTWbNmDRs2bCA2NpZx48aRm5trV9sbMyn4Vom4UC1QOVMExUYTBoOhgVskhBAXp0KjiY7PLW2Qa+978TI8XW3/Nzlr1ixuuukmAMaMGUNeXh4rVqxg5MiRvPzyy1x//fW88MIL1v27desGQIsWLdDr9fj4+BAWFmZXG4cPH17u808//ZSAgABWrVrFhAkT7DpXYyU9KpUI9nHDz8MFFYVjZwoaujlCCCEauYMHD7Jp0yauv/56AFxcXJg8eTKzZ88GYMeOHYwYMcLh101LS+Puu+8mLi4OPz8//Pz8yMvLa1b5MdKjUglFUYgN8WZLQhaH0/Lo2rpFQzdJCCEuSh4GPftevMyuY8xmM7k5ufj4+tSphL6Hwfay77NmzaK0tJSWLVtan1NVFYPBQGZmprWkvD0s6+ac78Iqr7feeivp6em8++67REVF4ebmRv/+/Wu98nRjJIFKFWKCtUDlSFpeQzdFCCEuWoqi2DX8AlqgUuqqx9PVpV7W+iktLeXLL7/kP//5D6NHjy732jXXXMM333xD165dWbFiBbfddlul53B1dbUuH2ARHBxMSkoKqqpaE2x37NhRbp81a9bw0UcfMW7cOABOnTrFmTNnHPSVNQ4SqFQhNsQLgCPp+Q3cEiGEEI3ZwoULyczM5Pbbb8fPz6/ca5MmTWLWrFm88847jBgxgpiYGK6//npKS0tZsmQJjz32GKDVUVm9ejXXX389bm5uBAUFMXToUNLT03njjTeYNGkSf/zxB0uWLMHX19d6/nbt2vHVV1/Ru3dvcnJyePTRR2vVe9OYSY5KFWJDtITaw9KjIoQQohqzZs1i5MiRFYIU0HpUduzYga+vLz/++CPz58+ne/fuDB8+nI0bN1r3e/HFFzlx4gQxMTEEBwcD0KFDBz766CM+/PBDunXrxqZNm3jkkUfKnX/27NlkZmbSo0cPbr75Zh544AFCQkKc+wXXM+lRqUK7skDl5NkCiowm3O0YqxRCCHHxWLBgQZWv9ezZ05pn0rNnT2vtlAtdcskl7Ny5s8Lzd999N3fffXe555566inrdo8ePdi8eXO51ydNmlTuc3unWTc20qNShWBvVzz1KmYVjsnwjxBCCNEgGjRQKS0t5ZlnnqFNmzZ4eHjQtm1bXnzxRcxmc0M2C9ASuMI8te3Dac2ncI4QQgjRlDTo0M/rr7/OJ598whdffEGnTp3YsmULt912G35+fjz44IMN2TQAwjxVjuUqHEqVQEUIIYRoCA0aqKxfv54rr7yS8ePHA1rW87fffsuWLVsasllWYR7auN7hVEmoFUIIIRpCgwYqgwYN4pNPPuHQoUPExcWxc+dO/vnnnyqXui4uLqa4uNj6eU5ODqAVwLmwCE5dGY1G69DPodRch59flGe5v3Kf64fc7/ol99t2RqMRVVUxm821TgOwJI9aziOcq6r7bTabUVUVo9GIXl9+Qoo9vwsNGqg8/vjjZGdn0759e/R6PSaTiZdffpkbbrih0v1fffXVcuskWCxbtgxPT0+Hty+8bCp6QkY+vy9cjEFSj51u+fLlDd2Ei4rc7/ol97tmLi4uhIWFkZeXV+fqqs1pYb6m4ML7XVJSQmFhIatXr6a0tLTcawUFti9P06CByvfff8/XX3/NvHnz6NSpEzt27OChhx4iIiKCqVOnVtj/ySef5OGHH7Z+npOTQ2RkJKNHjy5XAMcRjEYjy5Ytx9fdhZyiUmJ7DaJjuGOvIc4xGo0sX76cUaNGySKQ9UDud/2S+227oqIiTp06hbe3N+7u7rU6h6qq5Obm4uPjY63oKpynqvtdVFSEh4cHl156aYXvpWVExBYNGqg8+uijPPHEE9ZFnLp06UJCQgKvvvpqpYGKm5sbbm5uFZ43GAxO+eVXFG0l5S0JWZw4W0S31oEOv4Yoz1nfS1E5ud/1S+53zUwmE4qioNPpal3+3jL8YDmPcK6q7rdOp0NRlEp/7u35PWjQ72BBQUGFHyK9Xt+oxhQthd9k5o8QQghR/xo0ULn88st5+eWXWbRoESdOnODXX3/l7bff5qqrrmrIZpUTaw1UZOaPEEKIhjdjxgy6d+9u/fzWW29l4sSJ9d6OEydOoChKhYUSHa1BA5UPPviASZMmce+999KhQwceeeQRpk+fzksvvdSQzSqnXXDZ4oSy5o8QQohq3HrrrSiKYh3uaNu2LY888gj5+c6tbv7ee+8xd+5cm/atr+DCkRo0R8XHx4d33323yunIjYGlRyUhI1/W/BFCCFGtMWPGMGfOHIxGI2vWrOGOO+4gPz+fjz/+uNx+RqPRYflKlS2G2JxIllENgrxd8fc0YFbhaLr0qgghhKiam5sbYWFhREZGMmXKFG688UZ+++0363DN7Nmzadu2LW5ubqiqSnZ2NnfddRchISH4+voyfPjwCosTvvbaa4SGhuLj48Ptt99OUVFRudcvHPoxm828/vrrtGvXDjc3N1q3bs3LL78MQJs2bQBtMUNFURg6dKj1uDlz5tChQwfc3d1p3749H330UbnrbNq0iR49euDu7k7v3r3Zvn27A+9c1WT15BooikJsiDebT2RyJC2PThHNO3IVQohGRVXBaHvNDQDMZu2YEj3UZdaPwVOb/lkHHh4e1uJmR44c4YcffuDnn3+2FkAbP348LVq0YPHixfj5+fHpp58yYsQIDh06RIsWLfjhhx94/vnn+fDDDxk8eDBfffUV77//Pm3btq3ymk8++SSff/4577zzDoMGDSI5OZkDBw4AWrDRt29f/vzzTzp16oSrqysAn3/+Oc8//zz//e9/6dGjB9u3b+fOO+/Ey8uLqVOnkp+fz4QJExg+fDhff/01x48fr7elbiRQsUFsqA+bT2TKzB8hhKhvxgJ4JcKuQ3SAvyOu/VQSuHrV+vBNmzYxb948RowYAWgF0L766iuCg4MB+Ouvv9i9ezdpaWnW0htvvfUWv/32Gz/99BN33XUX7777LtOmTeOOO+4AYObMmfz5558VelUscnNzee+99/jvf/9rLfMRExPDoEGDAKzXDgwMJCwszHrcSy+9xH/+8x+uvvpqQOt52bdvH59++ilTp07lm2++wWQyMXv2bDw9PenUqROnT5/mnnvuqfX9sZUM/dggTmb+CCGEsMHChQutxer69+/PpZdeygcffABAVFSUNVAA2Lp1K3l5eQQGBuLt7W39OH78OEePHgVg//799O/fv9w1Lvz8fPv376e4uNgaHNkiPT2dU6dOcfvtt5drx8yZM8u1o1u3buWqwFfXDkeSHhUbxIX6AHBYelSEEKJ+GTy1ng07mM1mcnJz8fXxqVvBN4P9S7MMGzaMjz/+GIPBQERERLmEWS+v8r0zZrOZ8PBw/v777wrn8ff3t/vaoA012ctSu+zzzz+nX79+5V6zDFFZ1vNpCBKo2KBdqNajcvJsgcz8EUKI+qQo9g+/mM1gMGnH1XNlWi8vL9q1a2fTvj179iQlJQUXFxeio6Mr3adDhw5s2LCBW265xfrchg0bqjxnbGwsHh4erFixwjpcdD5LTorJZLI+FxoaSsuWLTl27Bg33nhjpeft2LEjX331FYWFhdZgqLp2OJIM/dgg2NtNZv4IIYRwqJEjR9K/f38mTpzI0qVLOXHiBOvWreOZZ55hy5YtADz44IPMnj2b2bNnc+jQIZ5//nn27t1b5Tnd3d15/PHHeeyxx/jyyy85evQoGzZsYNasWQCEhITg4eHBH3/8QWpqKtnZ2YBWRO7VV1/lvffe49ChQ+zevZs5c+bw9ttvAzBlyhR0Oh233347+/btY/Hixbz11ltOvkMaCVRsoCgKcSGW4R8JVIQQQtSdoigsXryYSy+9lGnTphEXF8f111/PiRMnCA0NBWDy5Mk899xzPP744/Tq1YuEhIQaE1ifffZZ/v3vf/Pcc8/RoUMHJk+eTFpaGqCtTv3+++/z6aefEhERwZVXXgnAHXfcwf/+9z/mzp1Lly5dGDJkCHPnzrVOZ/b29mbBggXs27ePHj168PTTT/P666878e6co6gNOfBURzk5Ofj5+ZGdne2U1ZMXL17MuHHjMBgMPPXrbuZtPMl9w2J49LL2Dr2WqHi/hXPJ/a5fcr9tV1RUxPHjx2nTpk2tV082m83k5OTg6+srixLWg6rud3XfS3v+f8t30EYy80cIIYSofxKo2Ehm/gghhBD1TwIVG8WWBSoJZTN/hBBCCOF8EqjYyLLmjyozf4QQQoh6I4GKjWTmjxBC1J8mPM9DlHHU91ACFTvEhloSaiVPRQghnMFSCbWkpKSBWyLqqqBAW0yyrjPdpDKtHWLLZv4cTpMeFSGEcAYXFxc8PT1JT0/HYDDUanqx2WympKSEoqIimZ5cDy6836qqUlBQQFpaGv7+/tbgs7YkULGDzPwRQgjnUhSF8PBwjh8/TkJCQq3OoaqqtdS7oigObqG4UFX329/fv9wKzbUlgYodLpz5I2v+CCGE47m6uhIbG1vr4R+j0cjq1au59NJLpcBePajsfhsMhjr3pFhIoGKHIG9XAjwNZBYYOZqeR6cIv4ZukhBCNEs6na7WlWn1ej2lpaW4u7tLoFIPnH2/ZfDODoqiECszf4QQQoh6I4GKnawzf1JyYO37sPO7Bm6REEII0XzJ0I+dLAm1hhMrIfVZ7cnCLLjk7oZrlBBCCNFMSY+KnSxTlAee+f7ck388Dtu+aqAWCSGEEM2XBCp2ig31IVY5TV/TDlQU6DpZe2H+/8Genxu2cUIIIUQzI4GKnYK8XZnutgyA3KjRcNWn0Os2QIVf7oKDSxq2gUIIIUQzIoGKnZSCs1zBagB2troRFAXGv631rJhL4YepcHRlA7dSCCGEaB4kULHX1jm4UsIeczTrS+O053Q6uPIjaD8BTMXw3RQ4uaFh2ymEEEI0AxKo2KO0BDb/D4BZpWM5lJZ/7jW9C0yaDTEjwFgA31wLSTsapp1CCCFEMyGBij32/Qa5yZR4BLPIfAlH0i5Y88fFDSZ/DVEDoTgHvroK0vY3SFOFEEKI5kACFVupKmz4CABjz2mUYLCu+VOOqyfc8B1E9ITCs/DlRMg4Wv/tFUIIIZoBCVRsdXIDJG0HvRue/e8kwNOAqsKRtEpK6bv7wk0/Q0gnyEuBryZCSUG9N1kIIYRo6iRQsVVZbwpdr0PxDraupHz4wuEfC88WcMtv4BUCWSchYV39tFMIIYRoRiRQsUVmAhxYqG1fci8AcWVr/lS7OKF3CMSO1rZPrHZmC4UQQohmSQIVW2z6DFQztB0KoR0BrKsoH6ppFeU2g7XHE/84sYFCCCFE8ySBSk2Kc2Hbl9p2WW8KnFtFucqhH4voQdpj0g4oynFCA4UQQojmSwKVGuh2fadNNQ5sB+1GWZ+3rKJ88mwBhSWmqg4Hv1YQ0AZUkxSBE0IIIewkgUp1VDO6zZ9p2/3u1irQlgnydiPQy7XqmT/ns/SqnFjjpIYKIYQQzZMEKtUIy9mBknkc3P2g+5QKr1t6VQ6m1jT8I3kqQgghRG1IoFKNtmlLtY1et4KrV4XX48MsCbU25qkk74CibMc1UAghhGjmJFCpSspugvP2oyp66HtXpbtYe1RSaghU/FpCi7bazCHJUxFCCCFsJoFKFfRluSlqh8u1hNhKxIdZaqnUEKjAuV6V41JPRQghhLCVBCqVyU1F2fszAOa+d1e5m6U6bVJ2ETlFxurPGX2p9ih5KkIIIYTNJFCpzLYvUEwlnPWMQW3Zu8rdfN0NRPi5Azb0qlh6VFJ2QWGWgxoqhBBCNG8SqFTmknsxXfY6B8Mm1rhrXJglT6WGKcq+4VotFtUMJ9c7oJFCCCFE8yeBSmXcvDH3vp00v2417hofauPMHzivnooM/wghhBC2kECljmye+QPn6qlIQq0QQghhEwlU6sjmWipwXp7KbijMdGKrhBBCiOZBApU6ign2RlEgI7+EM3nF1e/sEwaBsYAKCZKnIoQQQtREApU68nDVE9XCE4BDNg3/yLo/QgghhK0kUHEAm9f8AWhjWfdHAhUhhBCiJhKoOIBdeSpRljyVPVBw1omtEkIIIZo+CVQcwK6ZPz6hEBSPlqeyzrkNE0IIIZo4CVQc4FyPSh6qqtZ8gNRTEUIIIWwigYoDRAd6YdAr5BWXkpRdZMMBklArhBBC2EICFQdwddHRNkhbSdm2mT9lCbWpkqcihBBCVEcCFQexrvljS0KtdzAEt9e2E9Y6sVVCCCFE0yaBioPEh5b1qNgSqMC54Z/jMvwjhBBCVEUCFQeJs2dxQjg3/CMJtUIIIUSVJFBxEMvMn8OpeZjMdsz8SdsL+Wec2DIhhBCi6ZJAxUEiAzxxN+goLjVz8mxBzQd4BUFIR21b8lSEEEKISkmg4iA6nUJsiB2F30DqqQghhBA1kEDFgezPU5GEWiGEEKI6Eqg4UHyYNvPHpinKcG7dn/T9kJfupFYJIYQQTZcEKg5k7VGxdejHKxBCOmnbkqcihBBCVCCBigNZZv4cP5NPcanJtoPaWKYpy/CPEEIIcaEGDVSio6NRFKXCx3333deQzaq1MF93fNxdKDWrHD+Tb9tBklArhBBCVKlBA5XNmzeTnJxs/Vi+fDkA1157bUM2q9YURSE+1M6ZP1EDAQXSD0BemvMaJ4QQQjRBDRqoBAcHExYWZv1YuHAhMTExDBkypCGbVSeWNX9snvnj2QJCO2vbCeuc1CohhBCiaWo0OSolJSV8/fXXTJs2DUVRGro5tXauRyXP9oMiumuPafsc3yAhhBCiCXNp6AZY/Pbbb2RlZXHrrbdWuU9xcTHFxcXWz3NycgAwGo0YjUaHtsdyPnvPGxPkAcDBlBybj9UFxaMHzCl7MDn462gqanu/Re3I/a5fcr/rl9zv+lWb+23PvoqqqjYsTON8l112Ga6urixYsKDKfWbMmMELL7xQ4fl58+bh6enpzObZLM8IT2/R4r83+pbipq/5mOCcPQw4+gZ5bqGs6Pimk1sohBBCNKyCggKmTJlCdnY2vr6+1e7bKAKVhIQE2rZtyy+//MKVV15Z5X6V9ahERkZy5syZGr9QexmNRpYvX86oUaMwGAx2HXvJa3+TkV/Cz9P70bWVX80H5KVheK8jKgqlj50Eg0ctW9101eV+C/vJ/a5fcr/rl9zv+lWb+52Tk0NQUJBNgUqjGPqZM2cOISEhjB8/vtr93NzccHNzq/C8wWBw2g9jbc4dH+bDuqMZHMsopFeboJoP8I8Az0CUggwMWUchokctW9v0OfN7KSqS+12/5H7XL7nf9cue+23P96XBk2nNZjNz5sxh6tSpuLg0iripzuxe80dRzq2knLbfSa0SQgghmp4GD1T+/PNPTp48ybRp0xq6KQ5jqVB7MNWOmT/WQEVm/gghhBAWDd6FMXr0aBpBmoxD2b3mD0BIB+0xVQIVIYQQwqLBe1Sao7hQbRXllJwisgtsnIIlQz9CCCFEBRKoOIGPu4GW/trMnUNpNvaqWHpUcpOgMNNJLRNCCCGaFglUnMTSq2Lzmj/uvuAXqW1Lr4oQQggBSKDiNHav+QPnelUkoVYIIYQAJFBxGrtXUQbJUxFCCCEuIIGKk5xfS8XmWU2WQEVm/gghhBCABCpO0y7EG50CmQVG0vOKaz4Ayg/9NLMp20IIIURtSKDiJO4GPVGBXgAcSrGx8FtQHCh6KMqC3BTnNU4IIYRDmM0qOUWySrMzSaDiRNaZP7Ym1BrcITBG207b66RWCSGEcJRPVh+l64xl/HP4TEM3pdmSQMWJ4utSoVYSaoUQotFbsT8NgGX7pBfcWSRQcaI465o/MvNHCCGaG1VVrSUo9iRmN3Brmi8JVJzI0qNyODUXs9nOmT9SS0UIIRq11JxicotKAdifnIvJ1r/zwi4SqDhRdJAXBr1CfomJxKxC2w6yBioHwGxyXuOEEELUyeHzlkgpNJo4fsbGiRPCLhKoOJFBryMmWEuoPWzrmj8t2oCLO5QWQuYJ5zVOCCFEnRxKLR+Y7E3KaaCWNG8SqDhZnLVCrY2Rtk4PwfHatuSpCCFEo3W4LD9Fp2ifS56Kc0ig4mTxtVrzR/JUhBCisbP8XR/YLgiAPYnSo+IMEqg4WVyt1vyRxQmFEKIxU1WVw2laT/nE7i0B2JuUbfuSKcJmEqg4mWXmz5H0PEpNZtsOCumkPcrQjxBCNEqWGT96ncKYzmEY9Ao5RaWczrRx4oSwmQQqTtYqwAMPg56SUjMJZwtsO8jSo3LmMJTauE6QEEKIemMZ9okO9MTLzcXae743SfJUHE0CFSfT6RRrKX2bK9T6RoCbH6gmLVgRQgjRqFiGfSwBSqcIX0DyVJxBApV6EBtqZ4VaRYFQqVArhBCNlWXGT2yI9ka0c0s/APZIj4rDSaBSD6xr/tg188eSUCuLEwohRGNj+Xsea+1R0QIVqaXieBKo1APrmj92zfyRHhUhhGiMVFXlcGr5oZ8O4T4oCqTnFpOWU9SQzWt2JFCpB5YelRMZBRQZbSyLL7VUhBCiUUrNKSa3WJvxEx3kCYCnq4u1Ern0qjiWBCr1INTXDV93F0xmlWPp+bYdZBn6yToJxXb0xAghhHCq82f8uLnorc+fS6iVPBVHkkClHiiKYn+FWs8W4B2mbacdcFLLhBBC2Mvyd9wy7GPROUISap1BApV6EmfvzB84b+aPDP8IIURjcaRsarJlxo9Fp5Zaj4oM/TiWBCr1xNqjUquEWglUhBCisbhwxo9Fp3CtR+V0ZiFZBSX13q7mSgKVemLpUTmUJmv+CCFEU1XZjB8LP08DkS08ANgnvSoOI4FKPbH8QJ86W0h+caltB8kUZSGEaFRScoqsM37aBHlVeN3SqyJ5Ko4jgUo9aeHlSrCPG3Cu9HKNguMBBfLTIS/deY0TQghhE0tvSnSgJ64uFf+Fdm4ppfQdTQKVemStUGtrnoqrFwREa9sy/COEEA2uqhk/Fp1aWirU2tmjYjbDqc3yprQSEqjUo9rN/OmkPcrwjxBCNDhLj8qFibQWlloqx87k2zbMbzbDvvnw6WCYNRK+mABmGwuDXiQkUKlH8WFlqyjLmj9CCNEkHU4rvxjhhUJ83AnxcUNV4UBKNcM/ZjPs/Q0+GQQ/3Aype7Tn0w/A3l8d3OqmTQKVemTtUbFrirIlUJEeFSGEaEjVzfg537kKtZUEKmazFoh8MhB+nKq9CXXzhUsfhQEPaPusflPbTwDgUpuDSktL+fvvvzl69ChTpkzBx8eHpKQkfH198fauPMoU57oK03KLycwvIcDLteaDQs4b+lFVUBQntlAIIURVLDN+XKqY8WPRuaUfKw+mly+lbzbBvt9g1ZuQXvbG080XLrlH+/AIgKJs2PqF1quyfz50mujUr6epsDtQSUhIYMyYMZw8eZLi4mJGjRqFj48Pb7zxBkVFRXzyySfOaGez4O3mQqsAD05nFnIoNZd+bQNrPigwBnQGKMnT1v0JiHJ+Q4UQQlRwyDLjJ8ir0hk/Fp0iLAm1ZT0quSnw5cTzAhS/sgDlbi1AsXD3055b9brWq9LhCtDJwIfdd+DBBx+kd+/eZGZm4uHhYX3+qquuYsWKFQ5tXHNkLfxma56K3gBBcdq2DP8IIUSDOZxafX6KhWXo51BqLsWlJtj4qRakuPvB0KfgoV0w7MnyQYpFv7vB1VvLWTm0xOFfQ1Nkd6Dyzz//8Mwzz+DqWn7YIioqisTERIc1rLmSNX+EEKJpqmnGj0WrAA/8PAyUmstyWk5t0l4YPROGPg4e/lUf7NkC+t6pba96Qxvyv8jZHaiYzWZMpopTp06fPo2PT/XfPHHezJ8UG4u+gZTSF0KIRsCyBEpcaPU9KoqiWHtV9p46A4lbtRda9bXtQv3vB4MnJO+AI3/WtrnNht2ByqhRo3j33XetnyuKQl5eHs8//zzjxo1zZNuapfN7VFRbI2UppS+EEA1KVVWOWHpUQmp+U965rPBbxrFtUFqoDftYhvFr4hUEvadp26tev+h7VewOVN555x1WrVpFx44dKSoqYsqUKURHR5OYmMjrr7/ujDY2KzHB3ugUyC40kpZbbNtBlh6VM4fAZHRe44QQQlTK1hk/FpYeFUPSFu2JVn3tS4wd8AC4uMPpzXDs71q0uPmwO1CJiIhgx44dPProo0yfPp0ePXrw2muvsX37dkJCQpzRxmbF3aAnuuyH3OZ6Kn6tteQqUwlkHHVi64QQQlTG1hk/FpaZPxE5u7QnIm0c9rHwCYWeU7Xt1W/ad2wzU6s6Kh4eHtx2223cdtttjm7PRSE+1Idj6fkcSs3l0rjgmg/Q6SC4PSRu0fJUQto7v5FCCCGsbJ3xY9EmyAtPVz3dlEPaE/YGKgADH4StcyBhLZz4B6IH2X+OZsDuHpVXX32V2bNnV3h+9uzZMvRjI7unKMO54R9LmWUhhBD1xtYZPxZ6ncLAkBJaKWdQ0UHLXvZf1K8l9LhJ2171hv3HNxN2Byqffvop7dtXfEffqVMnKfZmo/gwS0KtHTN/WvXWHhPWOaFFQgghqmPrjJ/zjfROACDVIwbcajkrdtC/QOcCx1edm+Z8kbE7UElJSSE8PLzC88HBwSQnJzukUc2dpUflcGouZrON2dxtLtUeT2+GknwntUwIIcSFzp/xU90aPxfqXjbss0uJr/3F/VtDtxu07Yu0V8XuQCUyMpK1a9dWeH7t2rVEREQ4pFHNXXSgJ656HQUlJhKzCm07KKCNllRrLoWT653bQCFE46OqkLIbUmUl9fqWnH1uxk90YM0zfiwi87Wh+pUF0baXo6jM4IdB0cOR5ZC4rfbnaaLsDlTuuOMOHnroIebMmUNCQgIJCQnMnj2bf/3rX9x5553OaGOz46LXEVOWkGXzzB9FOdercmyVk1omhGhU8s/Arh/gl+nwVhx8Mgg+HgC/3gPFduS4iTo5nGbfjB8AjEV4nNkNwNriGE6dtfFNaWVatIUu12rbF+EMILtn/Tz22GOcPXuWe++9l5KSEgDc3d15/PHHefLJJx3ewOYqPtSb/ck5HEzNZWTHUNsOanMp7Pgajq92buOEEA3DZNTyEI6ugCMrtMqk5zN4QmkR7JwHJ9fBNbPO5a8Jp7HM+LEnP4XknShmI5mKPyfVEPYkZdM60LP2jbj0Edj1PRxcrPWshXWp/bmaGLsDFUVReP3113n22WfZv38/Hh4exMbG4ubm5oz2NVtxYbWY+WPpUUneCQVntTUhhBBNn9kMS5+E7d9AyQV/E0K7QLvh0G4kRPaD01vg1+mQeQJmjYahT2pDAzp9gzT9YmD5O93Ohoq0Vqc2ApDk0wUKFfYkZjOuS8X8TpsFxULnq2HPz1qvynVf1v5cTUyt6qgAeHt706dPH0e25aISbymlb+vQD4BvuFaC+cwhbV59h8ud1DohRL3aOQ82ls2a9AyEmOEQM0J79LmgxzV6INz9Dyz8F+z9BVbO1Hpgrv5MS7wUDmcZ+rGrR6UsUCkK6wVpsDcpp+4NufRRLVDZNx/y0sD74iiyalOgcvXVVzN37lx8fX25+uqrq933l19+cUjDmjtL5vix9HyMJjMGvY3jnm2GaIHK8dUSqAjRHBRlw58ztO1hz8Dgf9dcat3DHybNhtjRsPgRLcH+40Ew4W3oMsnZLb6o1GrGj6pqMzQB73YDYZeRvUnZqKqKoii1b0xIBwjvrg0JHl4OPW6s/bmaEJv+O/r5+Vlvrp+fX7UfwjYt/T3wdNVTYjKTkGHHdGNJqBWiefn7dchPh8BYrRKprevBKAp0vwHuXgOt+kBxNvx8u5Z4W+SAd+8CqOWMn6wEyEsFnQutOw1Ap8CZvBLb13erTtxl2uPhpXU/VxNhU4/KnDlzAC2ynDFjBsHBwXh61iEpSKDTKcSG+rDzVBYHU/JsH/uMHgQocOYg5KaAT5hT2ymEcKK0A7DpU2177Ovg4mr/OVq0hduWaDU21rwFu77Telhu+A5COzq2vRchS36KXTN+LIXZwrvh4eVNTLA3h9Py2JOYTaive90aFHuZtqLy0ZVa8rXeULfzNQF2TU9WVZXY2FgSExOd1Z6LSnzZeOdBexJqPVtAeFdtW2b/CNF0qSoseUyrjRQ/HtqNqP259AYY/jTculirt5SVAHPGXLSVTB3pSK3yU8rue2Q/ADq31EYb9iQ6oKcrogd4BUNxzkVTU8uuQEWn0xEbG0tGRoaz2nNRsa75Y09CLZwb/jkuwz9CNFn7F2i/w3o3uOxlx5wzqj9MXwWt+mq5L19eqU1zFrV2yLoYof0zfmilTTixBCrbT2XWvUE6HbQbVda4i2P4x+6Cb2+88QaPPvooe/bI4nh1FV+bKcoAbYZqj9KjIkTTVFIAS5/Wtgc+CC3aOO7cni3glt+0WUPGApg3GfbIJIfaOmRdjNDGHpXivHOLx5b1qFzSVislsen4WUpKzXVvVNxo7fHwsrqfqwmwO1C56aab2LRpE926dcPDw4MWLVqU+xC2s0xRPpGRT5HRZPuBrS/RFqnKOglnjzupdUIIp1n7HmSfBN9W2qJzjubqpeWodLoazEb4aRpsmeP46zRzqqqeN/RjY49K0jZQzdr31q8lAB3CfAnwNFBQYmLX6ay6NyxmuPY/4Myhi+J/gN11VN555526Ta8SVsE+bvh7GsgqMHIkLc/aPVgjN29o2RtObdB6VRz5bkwI4VyZCbD2XW37spng6qSJCS6ucM3/wN0Pts6BhQ9B4VkY9LA2Y0jUKDm7iDx7Z/xYhn0iz9UZ0+kU+scEsnh3CmuPZNA7uo5v6t39oHV/OLFG61XpN71u52vk7A5UbrjhBkpLS/Hysn1hJlE5RVGIC/Vh0/GzHE7LtT1QAWg7pCxQWQW9pjqvkUIIx1r2tFYGP3owdJzo3Gvp9DDhHW04aM1/YMWLWlXr0TMlWLFB7Wb8aPVTLMM+FgNigrRA5egZHhwZW/fGxY7WApVDS5t9oGLz0M+ZM2cYP3483t7e+Pr6MmDAAI4dO+bMtl0UzlWozbPvQGtC7Wpt9oAQotFTjq/SkmgVPYx9o36CBUWBEc/B6LKE3fX/hfn3g6nU+ddu4g6n2jnjx2yG05YZP33LvTSwXRAA209mUlDigHtvqady4h8osaMWVxNkc6Dy5JNPsnXrVl544QXefPNNzpw5w/TpzTuKqw+1WvMHtGxyFw+tUFT6ASe0TAjhSIpain5Z2cKtfe+s/xonA+6HKz8ERQfbv4Yfp4KxqH7b0IioqorRZK7246C9M34yjkBhJri4a2s0nSc60JMIP3eMJpXNJxww+ycoDvyjwFTc7AuA2jz0s3TpUmbPns24ceMAGDduHJ07d8ZoNGIwNP+CM85iy5o/qqqSlltMSnYRXVr6odMp4OKmJdUeW6n9kIZ0qK8mCyHKqKrKiYwColp4ar+X1WiT/ifKmUPaWj5DG2il+R43afkNP02DAwth2TMw/q2GaUsDKjWZueaT9ew8lWXT/jYn0lp6UyJ6VijepygKA9oF8dPW06w7coYhccF2tLgSiqL1qmz6TKtS235c3c7XiNnco5KUlESPHj2sn7dv3x5XV1eSkpKc0rCLhaVLMTGrkNwiIynZRaw+lM6sf47zxM+7uObjdXR7YRn9XlnBlR+u5cHvd2Aylw31nD/8I4Sod4t2JzPsrb/5cOWR6nfMS6N98q/a9ojntbV6GkqHy+G6r7TtrXO12YMXmQ3HztocpAR6udK3jY3Jr9ZE2r6VvjywXSAA6446qBZZrKWc/vJmnQJgc4+Kqqq4uJTf3cXFBbPZAXPCL2L+nq6E+rqRmlNM35dXUFjFNGW9TkFVVRbsTMLfw8CLV3ZCaTsEVqCNUZpKQV/rxbCFELWw+fhZAH7dnsj/jag6QVK/ciY6cyHm8O7oetxcX82rWvwY7Y3O8dVaku3l7zV0i+rVkj3JAEzq1YpnJ1Q/BOfpqrd90dhTleenWAyI0fJU9iRlk1VQgr9nLZZMOF/0IDB4Qk6iVrslrEvNxzRBdgUqI0aMKBesFBQUcPnll+Pqeu5mb9u2za4GJCYm8vjjj7NkyRIKCwuJi4tj1qxZ9OrVy67zNGW9ogJYvDuFQqMJvU4hKtCTuBAfYkO9iQ31IS7UmzZBXizbm8oD323nqw0JBHgaeHhEN3Dz0xYjS9kJLS+eeyZEY5CYVQjAsTP5JGTkE1XZFNa0A+h2zQPAfNlr6GxddNDZhj6pBSrbv9amLAdENXSL6oXJrLJ0byoAl3eLwM/DQakLhVnn8gVbVR6ohPq60y7EmyNpeWw4lsGYzuF1u6bBHdoMgUNLtNk/F3ug8vzzz1d47sorr6zTxTMzMxk4cCDDhg1jyZIlhISEcPToUfz9/et03qbmlau6cHWPVrRq4UGbIC/cXPSV7nd5twiyC40889se3v/rCP6erkyLHggHF2t/cCRQEaJeJWadS0b9+2A6UwdUEqgkrAUgzacTAS1711fTahY1QPsnd3yVtpjhFR80dIvqxdaETM7kFePr7kL/toGOO/HpLdpjQBvwrjr/ZEBMIEfS8lh7xAGBCmhVag8t0eqpXPpI3c/XCNUpUKmr119/ncjISOvqzADR0dEOv05j5+/pysiOoTbte9MlUWQVlPDWskO8uHAfvXp3oxuLtYRaZ1S4FEJUKTGzwLr998E0pg6IrrhTWTn1bI9oAuqpXTYb9pQWqOyYB4P/DQHRDd0ip7MM+4zsGGp7bRRbWPNT+lW724CYIL5cn8Dao2ccc93YsnL6pzdrNXI8m1+F+AZNapg/fz6XXXYZ1157LatWraJly5bce++93HnnnZXuX1xcTHFxsfXznBxtJUqj0YjRaHRo2yznc/R5HeGuQVGcyS1i7vqTPLE9gCUGUE9uoLQwT5sN1AQ15vvdHMn9rrvcolJyis7Vw1h/LIO8giLcDOV7RPXJu9EB2R6tG9/9Du+Fvs1QdMf/xvz3G5gmNI9clap+vlVV5Y89KQCMbh/s0O+H/tRGdIApohfmas7bu7UvOgWOpedzKiOXMF/3ul3YMxSXkE4oaXspPfgHaudr63a+WqjN3xN79lVUteFShd3dtW/Qww8/zLXXXsumTZt46KGH+PTTT7nlllsq7D9jxgxeeOGFCs/PmzcPT08nlaFupMwqzDuqY3O6wha3ewhScvin3VNk+LRv6KYJcVFILoDXdrrgoVdx1UN2icLdHUx08D/vT6pqZvyu6biYi/mr/avkerRsuAZXISD/MJceegkzOlZ0fJ0CN9t6d5uihFx4e48LbjqVl/uYMDiqQ0U1M37X3biYi1gZP5Mcz9bV7v7WLj2n8hVubGeib3Dd/wV3SPqRuNQFnA64hK3R99b5fPWhoKCAKVOmkJ2dja+vb7X7NmiPitlspnfv3rzyyisA9OjRg7179/Lxxx9XGqg8+eSTPPzww9bPc3JyiIyMZPTo0TV+ofYyGo0sX76cUaNGNdo6MZeZzNz/7U7WHevEFfr1tPPOpd+4pjmXvinc7+ZE7nfd/X0oHXZuJyrYl64tfflhayIFfm0YN+68Nwtnj+GyoxhV70aee1ijvd/mb9eiO/YXI1y2YRrX9HNVqvr5fmPpIeAEIzuGc+WEro67YOoeXHYUobp6MejqO7WlC6qx1+UQn605QYFPJOPGda7z5ZVTgfDlAloWHiB0zGhtwcJ6VJu/J5YREVs0aKASHh5Ox47lp4Z16NCBn3/+udL93dzccHOrOLRhMBic9svvzHPXlcEAH93Uiy/+2xey15O8YzmFg56kTVD16zAVGU24uega5eKSjfl+N0dyv2svNVfrum4V4MnwDqH8sDWRNUcyyt/PDG0WiBrcHlXRN977PfxpOPYXut0/oBvyKATGNHSLHOL8+62qKkv3pQEwrmuEY78PydpsV6VVHwxuNQ/lDI4L4bM1J9hw7CwuLi51/1sc3R88AlAKMzGk7ICo/nU7Xy3Z8/Ntz/2vU6BSVFRkHb6pjYEDB3Lw4MFyzx06dIioqItjmpwjuBv0TLnhZvjkPTqrhxn7+d/cNaorWQUlZOSXkJFXwtl8bftsfjEZeSUUlJjoHRXA99P7o6+hmqYQonKWqckt/d0Z2C4IF53C8QunKadoibSE1v1ds1O16g3tRsGR5bD6Lbjq44ZukcPtS87h5NkC3Fx0DI2vY1XYC1nqp1QxLflCvaNa4KrXkZxdxPEz+bQNtnEtoaro9NBuJOz+UatS20CBirPYPUJnNpt56aWXaNmyJd7e3taFCZ999llmzZpl17n+9a9/sWHDBl555RWOHDnCvHnz+Oyzz7jvvvvsbdZFzSe0HSbfSAyKiZa5O3nsp128svgAn646xk9bT/PXgTR2nsri1NlCCkq0gnJbEjKZvzOxgVsuRNOVZAlUAjzwcTfQO1qb0/P3wfRzO5XN+FFDO9V7++xmKeu/6zvIONqwbXECSxLtkLhgvNwcPJhgXYiw+hk/Fh6uenpG+QOw1tFVag8tc8z5GhG7A5WZM2cyd+5c3njjjXKF3rp06cL//vc/u87Vp08ffv31V7799ls6d+7MSy+9xLvvvsuNN95ob7MuboqCvu0QAG4OPcHg2CAmdo9g2sA2PHpZPK9d3YXPbu7Fz/cM4O9HhvJQ2RLj7yw/jNEklYWFqI3ETC1QifD3AGBofAgAKw+mndspZTcAamPvUQFo1Uv7Z6eaYdUbDd0ah1tSFqiM7RLm2BPnpcNZ7Q07rWyvkzOwrErtuiMOmqbcboS24GTaXsg65ZhzNhJ2Bypffvkln332GTfeeCN6/bmEoa5du3LggP2r+E6YMIHdu3dTVFTE/v37q5yaLGpQFqiMdD/AV7f3493re/Dc5R25b1g7ru/bmtGdwugVFUB0kBd3XdqWIG9XTp4t4IctzesHWoj6Yu1RsQYq2nDC+qMZFBlN2iq62drvlxrSBHpUAIY+oT3u/gHOHG7YtjjQkbRcjqTlYdArjOjg4FlNlt6U4PZ2reE0oGzdn/XHMjCbHTD51rPFuaGnw82rV8XuQCUxMZF27dpVeN5sNje+GgEXk+jB2mPyLq3oTzU8XV24f5j2PXx/xWHtj6oQwmZGk5mUHK0qrSVQiQ/1IczXneJSMxuOZUDqXm1nv9baisVNQcueEDe22fWqLNmt9aYMaheEr7uDk5lPbtAeq1jfpypdW/nj5aonq8DIvmTbZ8BUK66s+NvFHqh06tSJNWvWVHj+xx9/LLe6sqhnvuEQFA+o2iKFNbihX2ta+nuQmlPM1xsSnN8+IZqR1JwizCq46nUEeWszERVFsfaq/H0w/VwibVgTGPY5n6VXZc9PkH6oYdviINZhH0eUrD+fsRB2fqdttxli16EGvY5+bS2rKTuqSm1ZnsqxVVrbmgm7A5Xnn3+e+++/n9dffx2z2cwvv/zCnXfeySuvvMJzzz3njDYKW7W5VHs8vrrGXd1c9DxYttrrhyuPkFskvWFC2MqSnxLu747uvJlzljyVvw+mQaqWn9LoZ/xcKKI7xI8v61V5vaFbU2cJGfnsS85Br1MYZeNSJTbb9hXkp4FfJHS0f+27ATFaoLL2iIMSakM7gW9LKC206Q1rU2F3oHL55Zfz/fffs3jxYhRF4bnnnmP//v0sWLCAUaNGOaONwlZth2qPh5eCDQWHr+7ZkrZBXmQWGJn9zwmnNk2I5iQpuyyR1s+j3PMD2wXiolM4kVFA8eld2pNNrUcFzutV+RnS7M89bEwsvSmXtG1BgJdrDXvbobQE1pYtOTDoIdDbP6Q0sJ2WULvp+FlKSh0wsUFRzq39c2hp3c/XSNSqgPBll13GqlWryMvLo6CggH/++YfRo0c7um3CXjHDwMUDsk5C8s4ad3fR6/jXqDgA/rfmGJn5Jc5uoRDNQlLZqsktA8oHKpZpynpMuJQVe2tyPSoA4V2h/QRAhdVvNnRr6sQSqDhkpeLz7foOck6Ddxh0v6lWp4gP9SHQy5VCo4kdp7Ic0664suEfG9+wNgV2ByqnTp3i9OnT1s8t6/N89tlnDm2YqAVXL4gdqW3vn2/TIeO7hNMh3Jfc4lI+Wd38aicI4QynL5iafL6h8SG0UZLRm0vA1RsC2tR38xxjyGPa477fa0zQb6ySs4vYeSoLRYHLOjlw2MdUCmve1rYH/B8Yalf4VKdT6G8d/nFQnkqbS0Hvpr1hTT9Y8/5NgN2BypQpU1i5ciUAKSkpjBw5kk2bNvHUU0/x4osvOryBwk4dysZJ9823KZrW6RQevUzrVfli3QnSymYyCCGqZpma3KqSQGVYfAgdlZMAmIM7gM5RK9/Vs/BuENYVzEZtCKgJWrovFYDeUQGE+NRxleLz7f0VMo+DRwvofVudTmUZ/lnvqMJvrl7Q+hJtO2GtY87ZwOz+DdqzZw99+2rTsH744Qe6dOnCunXrmDdvHnPnznV0+4S94i4DvStkHIZ028aWh8WH0CsqgCKjmf+uPOLkBgrR9FnK51fWoxIX6k1vD63qc7JHbL22y+G6lxXf3PFNw7ajlpbu1QIVhw77mM2w5i1tu/+9WmBQB5aE2u2nMikoKa1r6zSWQOXURsecr4HZHagYjUbrwoB//vknV1xxBQDt27cnOTnZsa0T9nP3hbbDtO19v9t0iKIoPDI6HoBvN53k1NkCZ7VOiCZPVdVy5fMvpCgK/TyTANhW3Kpe2+ZwXa4FnQGStkPqvoZujV1ySmDrySwAxnR2YDXaAwu1N4FuftD3rjqfrnULT1r6e2A0qWw67qAhNkspf0uNlyauVnVUPvnkE9asWcPy5csZM2YMAElJSQQGBjq8gaIWOmrBI/tsy1MB6B8TyODYIIwmlXf/bD4VKYVwtKwCo3XNrHC/yocTooxaSfUl6U38b6JX4LnkzJ3zGrYtdtp1VkFVoVukv7UoX52p6rnelH53OaSQn6IoDGxnqafioOGfVn20cvpZCZCb4phzNiC7A5XXX3+dTz/9lKFDh3LDDTfQrVs3AObPn28dEhINLH4cKHptzQc7Fhez9Kr8uv00R9JyndU6IZo0y7BPkLcr7gZ9xR3y0nEvSsesKqzMCub4mfx6bqGDWYZ/dn6vJZE2ETvPavVtxjqyN+XIn9qMSoMn9LvHYae15Kk4LKHW3RcsyzY0g14VuwOVoUOHcubMGc6cOcPs2bOtz99111188sknDm2cqCXPFueKv9k4/APaO4/RHUMxq/D28qorUp7NL+GPPcnMmL+X++dtI1UScMVFJPGCNX4qKCv0luoSQSHuWvG3pix2FHgGaYXNjq5o6NbYJLOghCPZDg5U1POmaveepvU2OYhl5s++5BzHlYloXTb80wzyVGqVjq7X6yktLeWff/5h7dq1pKenEx0dTUhIiKPbJ2rLMvxj4zRli3+PjkdRYPHuFPYkZgPlA5Mx766m50vLufvrbcxdd4KFu5K566utsl6QuGgkVZNIC1hL5xe06ACUldNvyvQG6DpZ224iSbV/7k/HjEL7MB+iAuuW7Gp14h/tn77eTZuS7EAhPu7EhXqjqtoihQ4RWZZQezH2qOTn5zNt2jTCw8O59NJLGTx4MBEREdx+++0UFEgSZqPRfgKgaElwWSdtPiw+zIeJ3VsC8MB32ysEJgdStCGh+FAfbr4kCj8PAztPZfHc73tQm0lxISGqc+GqyRWkaoGKb7S29tmGYxkUljTxQL77DdrjwSVNoqaKZVryZR0d+ObZ0pvS82bwceBwUpkBMdrwj8PW/bH0qKTsgpKm/b/Z7kDl4YcfZtWqVSxYsICsrCyysrL4/fffWbVqFf/+97+d0UZRG94hEDVA296/wK5DHxoZi4tO4Vh6frnAZGr/KD6+sSdbnxnJ0n9dyksTO/PfKT3QKfDDltOyuKG4KFQ3NRmw9qgExfQk3E9bTXnjicb/z71aYV20D1NJo6+pklNktCalOqzI26nNcHwV6Fxg4IOOOecFLHkqqw6lO+ZNn18k+ESAuRQSt9b9fA3I7kDl559/ZtasWYwdOxZfX198fX0ZN24cn3/+OT/99JMz2ihqq4P9s38AogK9ePf67tw+qE25wOSFKzsztks4gWWrxQIMjg3m8THtAXhhwT7HTa8TopFKrKJ8PgClxXBGqwaqhHWxLlK4+pCD3iU3JGtNlcY9+2fF/lSMJpVQD5XYEG/HnNQy06fb9eDf2jHnvMDAdoG4G3ScOlvI3qScup9QUc7LU2nawz92ByoFBQWEhlaMUkNCQmTop7HpcLn2eGqj3VPUJnSN4NkJHSsEJpW569K2XN4tglKzyr3fbCU5u/ksLy7EhSwrJ1c69JN+UHsH6+4Pfq0YGh8MwKrDzSBQ6XKt1qOQtA3S9jd0a6q0aJdWz6t7CwcNRSfvgkN/aNN9Bz3smHNWwtPVheHttcB24S4H1SSz5qk07YRauwOV/v378/zzz1NUdG6mR2FhIS+88AL9+/d3aONEHfm1hJa9AdXu4R97KIrC69d0oX2YD2fySrhbkmtFM1VkNHEmrxioIlApy08hrAsoCgPbBWHQK5w8W0haU4/fvYIgTqub1Vh7VXKKjNbeq+5BDliNGGDNf7THTldDYIxjzlmF8V0iAFi8O9kxwz+WHpXTm7SKuk2U3YHKe++9x7p162jVqhUjRoxg5MiRREZGsm7dOt577z1ntFHURS1n/9jL09WFz2/pjb+ngZ2ns3n2N0muFc1Pcrb2Bs3DoMff01Bxh7L8FMuKyd5uLvSOagHA/iylXto4f2cSN8/ayKFUJ9RC6j5Fe9zVOGuqrNifSonJTNsgL8IdUeMt/eC5Eg+DnZ+DOax9MO4GHSfPFrAn0QHDP6FdwOAFRdk2L6nSGNkdqHTu3JnDhw/z6quv0r17d7p27cprr73G4cOH6dSpkzPaKOrCkqdyYi3kO2jaWxUiW3jy3xt6olPgx62n+UqSa0Uzc25qsjuKUkngkbJLewzrbH3KMvyzOV1Hcalz39WuPJjGQ99tZ83hM0z/ait5xQ4OJmJHazVV8lLh6F+OPbcDWIZ9xnUOpbJvj93+eQdQtVmUoR0dcMLqnT/8s2i3A4Z/9C7Qqpe23YTzVGpVR8XDw4M777yT//znP7z99tvccccdeHg4qESxcKwWbbQVUFUTHFzk9MsNig3iibFacu2LC/ax0VE1AYRoBKz5KQGeFV9U1XNDP6HnApXxXcPxdNVzKl/h3z/uwmR2Tk/j/uQc7v9mG2YV9DqF42fyefKX3Y7t2dQboOt12nYjq6ly/rDP2M4OmO1TmAV7ftG2nZibciGHD/9Y1/1punkqLrbsNH++7cMGlkUKRSPS8Qrtnd6++dDzFqdf7s7BbdmTmMP8nUncN28b8+8fVPVUTiGakHNVaStZ4ycnCQozteUrgttbn24V4MnHU7oz7YstLN2XxtO/7ubVq7tU3iNTS6k5RUybu5n8EhP92wby4MhYbvrfRhbsTOKSti24sV+Uw65Ftxtgw0dwcLFWU8WzhePOXQd/7tOGfWKCvYgN8abO68Dv+x1MxRDSEVr2dEQTbXLh8E+XVnVcT8iSUNuEe1RsClQmTpxo08kURcFkkiTKRqfDlfDXTDj2t/YuwcPfqZfTkmu7cjgtj/3JOdzz9Va+n96/8nVRhGhCqi2fb+lNCYoDQ/lAZkBMIFNjzcw9rOe7zafw93S19jzWVUFJKbd/sZnk7CJigr345KZe+HkaeGxMPK8sPsALC/bRPdKfThF1X0APgPCuWu5D6m6tpkrfOx1z3jpaXDZUMr5rhGOCwF3fa49dr8Mx40i28XR1YUT7UBbtTmbR7mQHBCp9AAUyT0BuKvg4qLZMPbJp6MdsNtv0IUFKIxUcp73DMxu1aXb1wMNVz2c397Im1364ss7vb4RocNWWz0/R1vg5Pz/lfN0CVV66Qstz+GTVUT5dZfuCoVUxmVUe/G4HexJzCPRyZc6tffErS/K9c3BbRnYIoaTUzP3ztpNbZKzz9awsSbWNZPbP+cM+47uE1/2EWSchYS2gaNOy69m4sq9h0e6kug//uPtpvULQZNf9qVWOimiCaln8rS4iW3gyc6L2R3vexpMUl0ogK5q2asvnV5KfcqHrerey9qS8uuQAP2w+Vaf2vLJ4P8v3peLqouOzW3rTOvBc7oyiKLx1bTda+ns4Pl+lXE2Vhp9NYhn2aRfiTVyoA4q87f5Re4weBH6t6n4+O1mGf06dLXTM7J8mvkChzYHKX3/9RceOHcnJqXjTsrOz6dSpE6tXr3Zo44QDWaYpH10BxXn1dtkxncII83UnI7+EP/bYV3ROiMbEbFZJKqtKW3mPiqWGStWBCsDdQ2KYPqQtAE/8sqvWvxdfrT/BrH+OA/Cfa7vRKyqgwj7+nq68f0MPXHQKC3cl881G29f9qpZ3MMRepm3vbPheFcuwz7gu4XUf9lFV2Fk27NPt+jq2rHYswz8AC3cn1f2ETXyBQpsDlXfffZc777wTX1/fCq/5+fkxffp03nnnHYc2TjhQaGcIaAOlRXB4Wb1d1kWvY0o/reT0V+tlurJous7kF1NiMqNTIMzvgmTakgI4WzaUE9a1xnM9MaY9k3tHYlbhgW+3270Q3coDaTw/fy8Aj14Wz+XdIqrct1dUgHWZixcX7rOuil5nluGfnQ1bU+X8YZ8JXR0w7JO8U1sGwcX9XHXvBmAZ/nHI7B9Lj0ryTjA2vcqDNgcqO3fuZMyYMVW+Pnr0aLZubdoLHzVrigIdr9S2nVz87ULX94nERaewJSGTfY5Yw0KIBmCZmhzq645Bf8GfzrT9oJrBK0RbELQGiqLw8lWdGdMpjBKTmTu/2MKu01k2tWNfUg73z9OmIV/XuxX3Dq25Wuodg9ucl6+yzTH5KrGjwTMQ8lLg2Mq6n6+Wyg/7+NT9hJYk2vixWn5HAxnWPhgPg94xwz/+UeAdpuUpJm5zTAPrkc2BSmpqKgZDJZUYy7i4uJCenu6QRgknsQz/HFpWr1F1iK87l3XWlkX/eqP0qoimqdphn9TqE2kr46LX8e713RkQE0h+iYlb52xm28lMTmcWVPmxPzmH27/QpiEPiAlk5kTbpjmfn69yIqOAJxyRr+LiCl0avqbK+cM+dWYqhd1li+t2bZhhH4tya//UdfiniS9QaNP0ZICWLVuye/du2rVrV+nru3btIjzcAT8ownkiempLf2ef0qpKth9fb5e++ZIoFu1K5rftiTwxtj2+7lUHvUI0RolZ2qKrlSbSWmb8VJNIWxl3g57PbunNjZ9vYOfpbK7+aJ1Nx8UEe/Hxjb1wdbF9PoS/pysfTOnBdZ+sZ9GuZC5pG8jNl9Sxvkr3G2Djx7B/IWQcrXItnLScIlJyiujayr9u17uAw4d9jv0N+WlaT1G7EXU/Xx2N6xLOot3JLN6dzBNj2tct/ybyEq02TBMs/GbzT/m4ceN47rnnyi1GaFFYWMjzzz/PhAkTHNo44WCKcm7MtR5n/wD0a9OC2BBvCkpM/LL1dL1eWwhHsC2Rtovd5/V2c2HObX3p3zYQd4Ouxo9OEb7lpiHbo2frAOuso5cWOCBfJawrtBupDSks/JeWiHqB4lITkz5Zz8QP13IwxbHrDzlt2KfzNVoV3gZ2/vDP7rp+r86f+dPEFii0uUflmWee4ZdffiEuLo7777+f+Ph4FEVh//79fPjhh5hMJp5++mlntlU4QocrtKqSBxbVa1VJRVG4uX8Uz/2+l682JDB1QLRDK3MK4WynreXzLwhUzGZI1RJb7e1RsWjh5cq3d11Sl+bZ7PZBbdhwLIM/96fxwHfbWfR/g/FwrWUxRkWBcW/CR/3h+Cpt2KRr+boj3206xcmzWm/U6kPpxIc5IKAoY1nbxyG1U4rz4MBCbbvr5LqfzwEswz+W4m916pEK6woGTyjKgjOHIMQxBQfrg809KqGhoaxbt47OnTvz5JNPctVVVzFx4kSeeuopOnfuzNq1awkNbXoV7y46kf20P6YlubDu/Xq99FU9WuLpqudoej7rZQ0g0cQkVVU+PytB+33Su0JQbAO0zD6WfJVQXzeOpeczc9G+up2wRVu49BFte+mT2jICZQpKSvngr3PFHjedOFu3a50np8jImsNlRd4cMexzYCEYC6BFDLTsVffzOYjla1u0q46zf/SGc19XE8tTsavgW1RUFIsXL+bMmTNs3LiRDRs2cObMGRYvXkx0dLSTmigcSqeDYWU9Xxs/hby0eru0j7uBq3q0BOBrWVlZNDHnyudfsCChpdBbcPtGMVxgC39PV/5zbXcAvtl4kj/3pdbthAMehKB4yE+HP1+wPj133QnO5BXjVdZjs/nEWcwOWpTRMuwT6+hhn66T67Vkfk2GxYfgYdBzOtMBwz9NdIHCWlWmDQgIoE+fPvTt25eAgIpFhkQjFz9Wi6yNBbDm7Xq99E1lyXtL96aSmlMx30mIxiivuJTsQm1Kb8SFPSp1yE9pSINig7hjUBsAHv95F+m5xbU/mYsrTCiro7V1DpzaRHahkU9XHQPg+cs74WHQk1Vg5Ei6YwpOWoZ9HDLbJzdFS6SFc6tDNxIernrr7J9FZTOcaq1101ygUEroX4wUBYY/q21vmQXZ9Zfc2iHclz7RAZjMKt9uclCVTCGcLLmsN8XX3QWfC2espTbNQAXg0THxtA/zISO/hMd+2lm3oYXogdD9Rm174b+YveoQ2YVGYkO8uaZXK3q09gdg4/G6D/9kFzp42Gf3T1odnMh+0KJN3c/nYA4b/mlVtkDh2WP12pteVxKoXKzaDoXowWAqgVVv1OulLb0q3246idHUtLLPxcXptC2LEdYykbYhubnoef+GHri66Fh5MJ2v6jokO+ol8GgBqXsoXfcRAP8eHY9ep9C3jZa4v9kBgYrjh32+0x4bWW+KhcOGfzz8IaSDtt2E1v2RQOVipSgw/Blte/vXWg2EejKmcxhB3q6k5hTXfWxciHpgSaRtdeGMn6IcLZkW7Cr21pjEhfrwZNmU5ZcX7edwah2mEHsFwuiXALhP+ZHh4cVc1kmbZNE3WgtUNh0/W+dicw4t8pa6Tws2dQbodHXdz+cEHq56hncoG/7ZVcfhH2ueStMZ/pFA5WLW+hJoNwpUE/z9Wr1d1s1Fz+Q+kQB1fwcnRD2wlM+v0KNimZbs2wo8mm6+3q0Dork0LpjiUjMPfrejTiudJ0VfzWa1PZ5KMW94fm0tQ9CjdQAuOoWUnCLrVO/acPiwjyWJNnZ0vZVrqA3LFOxFdV37x5qnIj0qoqmw9Krs/lFbr6Se3NC3NToF1h3N4EiaY4tACeFoSVUN/STv0B6baG+KhaIovDWpKy28XNmXnMPbyw7V+lwfrDzCkyXTKMWFoMQVWtVatF6BLq20tXPqkqfi0GEfs1n72wfQrXHUTqnK+cM/u07XYfjH0qOStKPJLFAogcrFLqK7VgQOFVa+XG+XbRXgyfCyZcy/3iBJtaJxOzc1+YJAxfJuPHpQPbfI8UJ83Xn1ai0h+LM1x+xe0Rng+Jl8fthymiNqK9K73qU9ueQxrZgaOCRPxTLs45DelIS1kJMIbn4Qe1ndz+dE5w//LK7L7J+AaPAO1aoJJ213TOOcTAIVUVZXRYH9C+r1B/eW/lpS7c9bT1NQ0nDLxAtRk0rL5yft0H5f9K7QbUrDNMzBLusUxvV9IlFV+PcPO8kusG+V5Xf/PITJrDIsPpjwCc9qq/bmJMLfrwLn5anUsvBbdqGR1Ye1xW8dUo3WkkTbaSIY3KvdtTFwyPCPojS5PBUJVIRWStlSMvqvmfV22UHtgogO9CS3uJQFu1Lq7bpC2KPUZCalrOZPuWTarXO0xw5XaEmkzcSzEzoSHehJcnYRT/9m+yrL+5NzmL9TW+X336PjwdUTxpfVadrwMSTvondUCxRF63lJy7W/jtLyfakYTSqxId7E1nXYx1h4bs2zRlIyvybD4kNwc9FxOrOQo3WpR9PE8lQkUBGaoY+DzgWO/AkJ6+vlkjqdYp2q/M3GU5WtZyZEg0vNLcZkVjHoFYK93bQni3O12hsAvW9ruMY5gZebC+9e3wO9TmHhrmQ+XX3MpuTa/yw7hKpqQzKdW2q5KMSOhI4TtYT9BQ/gV5xEfFmAsfl4ZtUnq4IlEJrQNcLuYys4uASKc8CvNbTuX/fz1QMPVz29o7Wk7bVH6rAMSeR5gUoTWKBQAhWhadEWetykbf/1UqWroDrDpF6tcHPRsT8llxOOKVgphENZEmnD/TzQ6cpKq+/+EUryICgOogY2YOuco3ukPw+N0NYtem3JAQa8+hdv/HGA05kFle6//WQmf+5PRafAw6Piyr845jVw89WGyd7ryoelL3CFbh3bj9mXZ5GeW8zaI1rezJXdHRCo7PpBe+x6rba0SBMxICYIwHovaiW8K7h4aOsyZRx2UMucp+l8d4TzXfoY6N20BLOjf9XLJf09Xbmim/ZHZ21K/fw4Fpea6lYrQlxUzk1NLsthUFXYUjbs0+vWRrUujCPdO6wdT4xtT7ifOxn5JXz091EufWMld3yxhdWH0sut2fPWsoMAXNOzFTHB3uVP5BsON/6oFZkEYvK28L7rf3lw5wRY9G8tgLHhjdGiXUmYzCrdIv2JDvKq2xd39jgcWa5tN5FhH4uB7bRAZcOxDEy1XTfp/AUKm0CeigQq4hy/ltDndm37r5n11qtyXVlNlT2ZCqX1UKn2/nnbGfXOav4+2HRKSIuGU2ExwqRtkLJLC+q73dCALXMuvU7h7iExrHlsGJ/c1IuB7QIxq/Dn/lRumb2JEW+v4n9rjvHHnmTWHsnAoFd4cGQVq0e3vgRu+R0e3EX+JY9wWg3CR82Hzf+Dz4bCJ4O0PJaCqpNsfy8b9rmymwN6U/6cAeZSiBkBwfF1P1896hzhi4+bCzlFpeypS5XayD7aY+JWxzTMiSRQEeUNehgMntof44OL6+WSPVsH4O9hoNCksLMu9QFssPbIGZaXVcP9dXuiU68lmodzgUpZj4qlN6XTxEZdIMxRXPQ6xnQO45s7LuHPh4dw64BofNxcOH4mn5mL9nP319sAuLFfFK0CPKs/WUAUXmOe5Rbvz7mx5ElSWk/QAr7UPfDHE/BOJzi1qcJhCRn5bD+ZhU6BCd3qONvn1CbY9xsoOmsV3abERa+jX1steXttLaaQW0X01B4TtzmgVc4lgYoozzsY+t2tbf81E8y1r1BpK71OYWA77Rdv1aE6/OLVwGxWeXXJuaJ2Kw+kyVpDokblir0VZcOen7UXejWvJFpbtAvxZsYVndjw1Ahevqoz7cO0xFhvNxfuHRZj83l6twlkrbkLc8OfhUcOwri3ILiDtqJ72VTm883fofWmDGwXRIhPHaYRqyosfUrb7nEThHaq/bkakOXv5bq6JNRahn7S9kFJ5blHjYUEKqKigQ9oBZDS9sGOefVyySGx2rjrqsPOC1QW7EpiT2IO3m4uBHgayCkqdcgCaaJ5s+SotAzw0BIwjQUQ3P7cFM+LkJebCzf2i2LJg4NZcP8gFv7fILsCiL5ttH+0m45naEsP9L0Tpnyv9XIc/QvSD1r3VVWV33ZovZ9Xdm9Zt4bv+w1Ob9Z6jYc9XbdzNSBLnsrmE2cpMtbyzaRvBHiHaTOyUnY5sHWOJ4GKqMgjAIY8pm2veEFbeM3JBsdqf7j2JefWqr5CTYpLTby5VPvjd8/QGEZ20KriLpNFEUU1VFU916Pi5w5b52ov9Lqt2SbR2kNRFLq08rM7udVS+G13YjaFJWX/aAOiIH6ctr3pM+u+e5NyOJqej5uLzrrAYa2UFmu5KQADHwSfsNqfq4HFhngT7ONGcamZ7SezancSRYGWluGfxp2nIoGKqFzfuyCwHeSnw+o3nX65IG83Ir205N1VB9Mdfv6v1idwOrOQUF83pg1sw6iO2h+85ftS67ySq2i+sguN5Jf9I21VsFfLpXBxb/TrwjR2kS08CPN1x2hS2X7qvHoq/aZrjzu+hcIsAH4v600Z2SEUH3dD7S+66XPIPKH1Igz4v9qfpxFQFIUBMWXDP3XJU2nZNPJUJFARlXNxhcte0bY3fAwZR51+yY7+WsDw9yHHBirZBUY++OsIAP8eFY+Hq57BscG4uehIzCpkf7JMVRaVsyTSBnq54rbjS+3JTlc36ZWSGwNFUehjXffnvEAlejCEdARjPmz/GpNZtRZ5u6IutVMKzsLqN7Tt4c+Aax2nNzcCAx1RT8WSpyI9KqLJih0N7UZqi1cte8bpl+sQoCW2rjmU7tBpyh+tOkJ2oZG4UG+u6dUKoCxY0X7Rl8vwj6iCJT8lzs8Ee37RnmxmlWgbimWBwk0nzksIVZRzyfybPmPj0TRSc4rxdXdhaHxw7S+2+k0tETq0M3RvHusy9S/rUdl5OpvcIvvWZLKK6KE9Zh6vdmp4Q5NARVRNUbReFZ2LNlX5yAqnXi7KG/w9tCTX7aeyHHLOxKxC5qw9AcATY9uj153LK7AM//y5XwIVUTlLfsqVujVQWqi922/Vp4Fb1TxY8lS2JWSVn33X5VqtxyorgcP/aDOsxnUJx81FX7sLZRzVhn1Am46sq+V5GpnIFp60buGJyayyqbaTAjwCoEXZbK2kxjv8I4GKqF5wvJavAtq0PlMtI3cb6JRz0+4cVYztP8sOUlJq5pK2LRgWH1LuteHtQ1EULaEvObvQIdcTzUtSdhGgMjyvrKaQJNE6TGyIN/6eBgqNpvKFy1w9oedUANonfAPUcbbPihe0XuF2IyFmeF2a3OhY/l7Wad0f6/CPBCqiKRvyGHgGQvoB2DLbqZcaGqcNx/ztgITafUk51qJuT47tgHLBP5hgHzd6ttZyDf6U4R9RicTMQnophwgpOqatjdL1uoZuUrOh0yn0jirLUzlxQY9AnzswK3r6sYcB3qn0a1O7wnrKqY2w73dt2vOoplfcrSaWdX8ck1DbePNUJFARNfMI0BLQAFa+DPl1iN5rMLjsHcLepBzScuo2Tfm1Pw6gqjChazjdIv0r3ccy/CPTlEVlErMKmeJSNuTZ+Rrw8G/Q9jQ3lgCkwtCFfyQ7vAYB8GjAqnOLQdpDVdH9+Zy23eNmCO1Yl6Y2SpaZPwdScjmTV1y7k5zfo9JIZ0BKoCJs03OqlohWlA1/v+K0ywR6u9G1lbZEfF1m/6w5nM7qQ+kY9AqPXlb1Wh6WQGXDsQxyapuQJpqt3Mw0Jug2ap9IEq3DWWf+nMgst8hhbpGRt7KGAtDt7B+1SvSMyNqILmkrGLyadHG36gR6u1mrA68/Wss3kGFdtDzE/DTIPu3A1jmOBCrCNjq9tlw7aMM/qXuddqmhcVp2f23rqZjNKq8uPgDATZdEERVY9VTEmGBv2gZ7YTSpTqnfIpqu4lITQwpX4KYYKQ3udO6dp3CYThG+eLrqyS40cijtXJmApXtTWVcaxxFdG3SmItj2pX0nLi2mY9IP2vagh8CnDoXiGjlLldpaD/8YPLQkcWi0wz8SqAjbtRkMHa4A1awtIOakbsKh7bWk1zWHazdN+fediexLzsHHzYX/G17Faq7nGdVBZv+IipIzC5mi14Z99H2mSRKtExj0Omue2PnLWWhF3hROxNysPbH5f2Aqtfm8ui3/w6vkDKp3GPS/z5FNbnQcmlDbSGf+SKAi7DP6JW210+Or4cAip1yiWyt/61o89k5TLjKaeGvpIQDuGRZDCy/XGo+xDP/IIoXifHmHVtNOl0Qh7iiSROs0lnoqG8sClbTcImsRs7iRt2mJ/NmnbF/NPf0QurVvA2Aa+lSzKO5WnT7RLdDrFE6eLeDU2VouLtjIZ/5IoCLsExB9rvz0sqfB6Ph1efQ6hcGx2vDPygP2TVP+cv0JErMKCfN1Z9rANjYd06N1AIFeruQUlda+HoFodrz3aQtybvAaBu6+Ddya5qtP9LmZP6qqsnBnMmYVerT2p3Voi3OrVG/8tOaT7f0NPh+GUpRNlkc0apfmv9SBj7uBbmV5fbUe/rHM/EnaDuZaLnLoRBKoCPsN+hf4hGvrZmz4yCmXsFShtGeaclpOkbVU/sOj43A32FbYSa9TGNFBG26SKrUCgJICIpL/BGB/6BUN3JjmrUdrfwx6hdScYk6eLeD3spL5Ey21U/rcDooeEv6B5CpW+TWVwtKn4cepUJKHOWogG2L+3WyKu9XEkqdS6+Gf4PZa0nFJHpw57MCWOYYEKsJ+bt4w8gVte/VbTlkH6NKyhNp9ybZPU35hwT5yi0rp2sqPa3q2sut6ozpqK6nKIoWisMTE9j+/xdVcyElzMKaI3g3dpGbN3aCnayt/AH7Ycoqdp7LQ6xTGdw3XdvCNgI5XatubKulVyU2FL6+E9f/VPh/wAKYpP1Ns8HN+4xuJc/VUMmr390unh4ju2nYjTKht0EBlxowZKIpS7iMsrOkuvX1R6XIttOqrLR72vxFwfI1DTx/k7WbtzrRlmvKf+1JZtDsZvU7h1au7lCuVb4tB7YJwN2iLFO5LzqlVm0XTVVhiYvHuZO77Zhs9X1pO+nqtIup88wC6tpYFCJ3Nkqfy+erjgPb7GOTtdm6HS+7RHnf9WL6O08kN8OmlWm+Lqw9c92VZmXyX+mp6o9Azyh93g44zecUcTsur3Uks6/5IoFJRp06dSE5Otn7s3r27oZskbKHTweSvIKInFGbCVxNh61yHXmJIWcn7mqYN5xWX8tzvewC4Y1AbOkXY/07KsqIywJ/7HFO+XzRuFwYn936zjUW7kzEYsxmm3wnAZZPvZ0hcHRbDEzaxrPtTUpbMfuWFKyW36qP9IzUVw9Y52ozDDR/D3PGQl6INXdy18lzPy0XGzUVvzfWp9WrKjXjmT4MHKi4uLoSFhVk/goPlj0KT4RMGty3WKnaaS2HBg7DkcbumEVbHkqeyuoZpyv9ZdpCk7CIiW3jw4MiapyNXxTJNefn+lFqfQzR+RpOZlxbuKxecFBpNtArwYPqlbVk0KhMDpRDSidgufRu6uReFXtEB1tnf7gYdoztd0LN+/qrKm2fBz7drJRLMpdDparhjBQTV/ne/ObAM/9Q6T8USqKTsccokibpo8EDl8OHDRERE0KZNG66//nqOHTvW0E0S9jB4wDWzYFhZif2Nn8C8a6Ewq86ntkxTzi0qZdvJys+381QWX6w7AcDMiV3wdK19l+/wDiEoCuxJzLGumiual8z8Em6ZtYlZ/xyn0GgisoUH04e0Zf79A1nz2DCeHNeByNNl0+67XNOwjb2I+Lob6BCmzawa2SEUb7dKfo87XQVeIZCbBHt+1oZ3xrwGk2ZreXMXOUs5/Y3HMmpVfwr/1tpUcLMRUvc4uHV106ADef369ePLL78kLi6O1NRUZs6cyYABA9i7dy+BgYEV9i8uLqa4+Nx6Bjk5Wi6B0WjEaHRs+XPL+Rx93mZrwEMoLWLQz78P5ehfqP8bQem1X0NgO5sON5YUg6pWuN+D2gWyYFcKf+1PoUcrn/LHmMw8/vMuzCpc0TWcAW386/T98nPT0TPSn60ns1i6J4mb+rWu9bkau4vx5/tIWh7Tv9nOybOFeLnqeeOazozqEGJdrLK0tBRyU3A5vgYFMLa/Ehx0fy7G+22v6/u05L0VRdzav3UV90mHrvft6Fe9iuoVguma2aiRl0BpxR7ci/F+x4d44uvuotWfSsigexXrm1VHH94D3dE/MZ3cjDm0m83H1eZ+27OvojaiKQ75+fnExMTw2GOP8fDDD1d4fcaMGbzwwgsVnp83bx6enp710URRA7+CE/Q79i4exrOU6D3ZEn0/6b6dy++kmvEuTiYg/xgBBcfwLziOX2ECOtWESTFg0rlqH4qBPLMrKSWuqDpXInwMJAX05WTgEABWJCrMP6nH00Xlqe4mfAx1b7/lnPF+Zu7tKMXfmov9WQpzD+koMim0cFO5M95ERCV1wNqm/UGXxHmc9WrHmrjn6r+honqqmbCcHZz1bEeJQWrbXOh/B3TsztQxPtLE6Fb2/2uPT/6V9im/cipgINuipzuhhecUFBQwZcoUsrOz8fWt/nvZqAIVgFGjRtGuXTs+/vjjCq9V1qMSGRnJmTNnavxC7WU0Glm+fDmjRo3CYHDAf8CLSV4q+p+mokvcgqroMY98AdU3EiV5G0rSNpTkHSjFuTWfpwql135FQtAQxv93HUVGM69d1YlrerZ0SNOPn8ln9HtrMegVNj4xFB/35vm9v1h+vlVV5YsNJ3l1yUHMKvSO8ue/N3QnsIqKxfo5o9ElbcM0+lXMfe50WDsulvvdWFys9/urDSd5cdEB+rdtwZe32T+tXjmyHJfvb0ANbEfp3RtsPq429zsnJ4egoCCbApVGNYeruLiY/fv3M3jw4Epfd3Nzw83NrcLzBoPBaT+Mzjx3sxXQCm5dBAseQNn1Pfrlz1Tcx+AJ4d21iogte2EM6cKfqzcwcuhADGopGAugtAiMhcz8bRvJGVk82uYE0Ynz0S/6F2/7f0iR0Uz/toFM7htl7b6vq7hwf2KCvTians/aY1lc3i2i5oOasOb88200mXn+9718u+kkANf2asXMqzrj5lJFEbCzx7QZD4oOfddJ6J1wX5rz/W6MLrb7fWl8CCw6wNaTWZjQ2Vz00qq1ljyuZBzBYCoAd/tmUNpzv+35vjRooPLII49w+eWX07p1a9LS0pg5cyY5OTlMnTq1IZslHMHgDld9qq3KueEj8A7RssojtMCE4PagP+/Hz2ikxLAP/CLhgh9gr64tWbTiMHr3cbwfchQlbS8Tc19nucujvHxVZ4cFKRajOoZxdNVRlu9LbfaBSnOVmV/CPd9sZcOxsygKPDW2A3cMblP9z8run7XHNkO0n1chmpiYYG9CfNxIyy1mW0ImA8oq1trMKwj8oyArQSun33aoU9pprwad9XP69GluuOEG4uPjufrqq3F1dWXDhg1ERUU1ZLOEoyiKtsT6I4fg7n/g8veg11QI61w+SKmBZZryyqM5nB3zX0pwYZR+G5902EPbYMdn+4/qqP2TWnlQFilsio6k5TLxo7VsOHYWL1c9/7ulN3de2rb6IEVVYfeP2naXa+unoUI4mKIo58rp13Xdn0a0QGGDBirfffcdSUlJlJSUkJiYyM8//0zHjh0bskmiEep63jTlmxfm86ZRW8l22Il3nVK+v3tkAEHeruQWlbLxmCxS2JQUGU3c8PlGEjIKaBXgwS/3DmREWX2caqXugTMHtZXBO0xwfkOFcBLLNOU611NpRBVqG7yOihA10esU69o/e5NymGUeR27YJSjGfPh1usMKzJ1/vRHttX9uf9m5erNoWMfS80nPLcbH3YXf7xtIfJhPzQcB7P5Je4wbbfe4vBCNiWW4Z9fpLHKKajE9O0J6VISoFcvwD8CN/drgc/3n4OYLpzfDP+84/HoDY7Vf9s0npEelKUnIyAe0sfpA74qJ95Uym7UCYgCdJzmpZULUj5b+HrQN8sKswg+bT9l/gvBuoOi0wno5yY5vYC1IoCKahKFxIfh5GGgV4MGjY+K1Korj3tJeXPWaw6P/PtHaQnR7k7LJK3Zsj41wnhMZBQBEB9pRV+n0Jsg+pS1qF3eZk1omRP2569K2APx35RGyC+3sVXHzhuAO2nYjWfdHAhXRJAR4ubLykaEsfnAwvpbaJl2vg44TtfU+frkLSgocdr1wPw9aBXhgVmH7yUyHnVc418mzWo9K68BKqrlVxZJE22GCtiSEEE3cpF6tiA3xJqvAyMd/1yKPr2XjWklZAhXRZLTwcj0XpIA2q2jCO+AdBhmHYbljK4laViPdfEIClabixBk7e1RMRtj7m7Ytwz6imXDR63h8THsA5qw9bv/aZdaEWulREaLuPFvAxI+07c2fw+E/HXZqa6ByXPJUmgpLjkqUrT0qx1ZBwRnwDIK2Q5zYMiHq14gOIfRt04LiUjNvLz9k38GWQCVpm5bD1cAkUBFNX7sR0LdsXYrf74MCBwQWqsolISYAtp/KlHoqTUCR0URyjrY8vc09KnvKZvt0mgj6i6eCqWj+FEXhqXFarsnP206zPznH9oNDOoKLOxRlaxWbG5gEKqJ5GDkDguIgLwV+uweyE2t3nqJs2PQ5fDyQtl9051P3D9AZC9ibZMcvuWgQpzMLUFXwcXOhRRVr+ZRjLIT9C7RtKfImmqHukf6M7xqOqsJrSw7YfqDeAGFdte1GkFArgYpoHlw94erPQOcCh/6AdzrC/0bB+g8h+3T1x6oqnN6q9cb8pz0sfgTS9gJwGev5zfVZDu7dXg9fhKgLS35K60BP25ZVOLQUSvK0ZRta9XVy64RoGI+OjsdFp7DqUDprj9hRrbYRFX6TQEU0HxE9YPI30Lo/oGjTTpc+Be90qjxoKc6FzbPg08Hwv+Gw/WttMcTg9jDmdbjpZ/Jdg4jTJXLFxhvh4JIG+9JEzU6U5adE25qfYpnt0/lq0MmfQtE8RQd5cdMl2rI0ry7Zj9ms2nagtZR+wwcqjWr1ZCHqLH6M9pGTDPvnazM6Tq7XghZL4NKqDwS207r9S/K04/RuWp5Cr9ug9SXajCLgyFWLKfn2JvroDsG318OQx2HIE/KPrRFKKKuhEmVLfkphFhxerm3LsI9o5v5veDt+2nqaPYk5LNiVxJXdW9Z8kKVHJXmXNjuuAXO45K+taJ58w6HfdJi2BB7eD2PfhKiBaD0tm2Hnt1qQEhgLl70C/z6gDR1F9bcGKQAdYmO5zfwcc0tHa0+seh2+naz9oxONSsJZy9RkG3pUDiwEUzEExUNoZye3TIiGFejtxj1DYwB4c+lBiktNNR/Uoq22nISpGFL3OrmF1ZNARTR/vuHQ7y64bbEWkIx9Ey65D25dBPdvhv73adOcK+HqoqNTZBAzSm9lQ9eXtUz4w8vgs6EN/ssryrNMTW5tS4+KZW2fLteWC0yFaK6mDWxDqK8bpzML+Wp9Qs0HKMp56/407PCPBCri4uITpgUtY16B6EE2/ZOy1FP5sXQwTFsKfq0h8zj8b+S5NWJEgzKazJzO1Ipa1dijUpwHJ/7RtjtNdG7DhGgkPFz1PDwqDoAP/jpCdoENpfXPr6fSgCRQEaIGvcvW/dmScBYiusP0VdB2qJZ4+9M0WPdBg7ZPQGJmISazirtBR4hPDYsRnlwPZqMWcAa2q58GCtEIXNOzFXGh3mQXGvlo1ZGaD2jVW/sd8Q51fuOqIYGKEDXoGRWAomjJmmk5Rdow0U2/wIAHtB3+fg2KpM5KQ7LM+Ilq4YVOV0Mv2bG/tce2Q2TYR1xUXPQ6nhhrKa1/gsSaSuvHj4X/2wojHLs8ib0kUBGiBr7uBjqE+QLnrfuj08OoF7VkzJI82DGvAVsoTp49V0OlRkdXao8xw5zYIiEap2HxIVzStgUlpWbeXmZnaf0GIoGKEDboUzb8s/nEeeX5FQX63qltb/qsUayJcbGyeTHC3FRrMT/ayNo+4uKjKApPjtVK6/+y/TR7k7IbuEU1k0BFCBv0Lkuo3ZJwwTpC3W4AN184exSO/dUALRNgx2KEx1drj2FdwCvIya0SonHqFunP5d0iUFV44ufdjX4tMwlUhLCBZebPvqQccovOy5Z384buN2rbGz9rgJYJsKMqrTU/ZahT2yNEY/fs+A74eRjYnZjNp6uONnRzqiWBihA2CPNzJ7KFB2YVtp/MKv+iZfjn8LJGsdLoxcZkVjl1VksKrLYqrapKoCJEmRBfd2Zc0RGA91Yc5kBK450QIIGKEDbqE1U2/HPiguGfwBhoNwpQYdP/6r9hF7mUnCJKTGYMeoUIf4+qd8w4CjmnQe8KrQfUXwOFaKQmdm/JyA6hGE0qj/y4s9EOAUmgIoSN+rTRApVNFwYqoJXrB21hw+K8emyVSDijDftEBniir25q8rGy2T6R/bTVtoW4yCmKwitXdcbPw8CexBw++btxDgFJoCKEjSwzf3acyqKk9IJ3HjEjtLUxirNh1/cN0LqL1wlbFyOUYR8hKgjxdeeFKzoB8P5fjXMISAIVIWwUE+xNgKeBIqO54pQ+nQ76WKYqf67lQ4h6YdOMH1MpHF+jbbeV+ilCnO/K7hGM6th4h4AkUBHCRoqi0KssT2VzZcM/PW4Egxek7z83DVY4XYItPSrJO7TeLjc/bRkEIYSVoii83IiHgCRQEcIOfdtYCr9lVnzR3Q+6Xa9tb5KpyvXFpqnJlvyUNoO1qsJCiHJCfMoPAe1PbjxDQBKoCGEHa+G3E2dRKxve6XuX9nhwMWSdrMeWXZxUVbWtR+XYKu1RyuYLUaXGOgQkgYoQdugc4Ye7QUdmgZGj6fkVdwhpr5VmV82weVb9N/Aik55bTKHRhE6BVgFVBCol+XBqo7Yt+SlCVMkyBOTvaWBvUg4fN5IhIAlUhLCDq4uO7pH+QBV5KnBuqvK2L8BYw+qkok4SyhYjjPD3wNWlij9nJ9eDqQT8IrWZWUKIKp0/BPRBIxkCkkBFCDtZyulXGajEjQH/1lCYCbt/qseWXXxOnLEhP8WyWnLbIdpCkkKIal3RLYLRjWgISAIVIex0Lk+lkoRa0JI1+9yhbW/6VKYqO5Fd+Sky7COETRRFYWYjGgKSQEUIO/Vs7Y9OgZNnC0jNKap8px43g4s7pOyGkxvqt4EXkRpn/OSlQ+pubbvNkHpqlRBN3/lDQHsSsyufPFBPJFARwk4+7gY6hPsC1Qz/eLaALtdq2zJV2WlOluWotK6qR+V4WW9KaBfwDq6nVgnRPFzRLYKvb+/Hpzf3QmnAYVMJVISohT41Df/AuaTa/fMhJ7keWnVxUVWV4zXlqFjL5ktvihD2UhSFQbFBDRqkgAQqQtRK72hL4bcqelQAwrpoq/SaS2HL7Hpq2cUjq8BIblEpAK1bVNKjoqrnBSqSnyJEUyWBihC1YOlR2Z+cQ26Rseod+5UVgNs6F0zV7CfsZslPCfN1x8O1kmqzZ49B9inQGSCqfz23TgjhKBKoCFELob7utG7hiVmFbSezqt6x/QTwCob8tHPTZIVD1Djjx1I2P7IfuFYzfVkI0ahJoCJELVmGf7ZUN/yjN0Dna7TtXd/XQ6suHjUHKn9rj22H1kt7hBDOIYGKELVkGf5ZfzSj+h27Xqc9HlgExblObtXFI6Fs6CeqskRas+ncCtayvo8QTZoEKkLU0qVx2nTXrSczScutop4KQERPCIyF0kLYv7CeWtf8VVtDJXkHFGWDmx+Ed6/XdgkhHEsCFSFqqaW/B91a+aGqsGxvatU7Kgp0naxty/CPw1Q79GMZ9mkzGPQu9dcoIYTDSaAiRB2M6RwOwB97Uqrfscsk7fH4Kqmp4gC5RUYy8kuAGgIVyU8RosmTQEWIOhjbOQyA9ccyyCz7x1mpFm0g8hJQzbBHFiqsK0tvSqCXKz7uhvIvlhScW7ZAAhUhmjwJVISog+ggL9qH+WAyqyzfX83wD5xLqpXhnzqrdtjn5HowlYBvSwhsV88tE0I4mgQqQtTRuC42Dv90ukorPpayG1L31UPLmq9qE2nPr0bbwKW/hRB1J4GKEHVkGf755/CZ6qvUeraA2NHa9u4f6qFlzdfJjGoWI5T8FCGaFQlUhKij2FAfYoK9KDGZ+etAWvU7W4d/fgSz2fmNa6aq7FHZ+xuk7NK221xav40SQjiFBCpCOMDYstk/i3fXMKMnboxW2yPnNJxcVw8ta54qzVHZ9zv8NE3b7nUr+ITWf8OEEA4ngYoQDjCmbPhn1aF0CkpKq97R4A6drtS2d35XDy1rfgpLTKTkaAX2rD0q++ZrQYpqgq7Xw/i3G7CFQghHkkBFCAfoFOFLZAsPioxm/j6YXv3OluJv+34HYzUVbUWlTp7VelN83F3w9zTA/gXw021gLtXu7cSPQFfJaspCiCZJAhUhHEBRFOvwz5KaZv+0HgC+raA4Bw79UQ+ta14SzstPUQ4sgh9v1YKULtfCxI8lSBGimZFARQgHsQz//LU/lSKjqeoddTroeq22vUtm/9jLkp8ywW0b/DhVC1I6T4KJn0iQIkQzJIGKEA7SvZU/Yb7u5JeY+Ofwmep3tgz/HF4GBWed37hm5ERGPiN0W7kj+YWyIOUauOpTWdNHiGZKAhUhHESnU6y9KjUO/4R0gLAuYDbC3l/roXXNR8Dpv/jY8C56tRQ6XQ1XfSZBihDNmAQqQjiQpfjb8n0plJTWUCfFuqKyDP/Y7NBSHsh4EVfFREb0eLj6cwlShGjmJFARwoF6R7cgyNuVnKJS1h/LqH7nzpNA0cGpDXD2eP00sCk7+hfq9zfhSikLTf0wTZSeFCEuBhKoCOFAep3C6E5ar8ofe2oo/uYbDm2GaNu7f3Ryy5q4kxvhuxtRTCUsMfXhSR4g2K+SdX6EEM2OBCpCOJhl+GfZ3lRMZrX6na3DP9+DWv2+WQUlLN9Xw4yi5ihlN8y7FowFZIRfygPG/6NloC+KLDgoxEVBAhUhHOyStoH4eRjIyC9h0/EaZvR0mAAuHpBxBJK2Vbvrg9/t4M4vtzDy7VX8sScFtYbAplnIOApfXQVF2RB5CYs6vI4Rl/Kl84UQzZoEKkI4mEGvY1RHbZ2ZGod/3Hyg/Xhtu5qk2hNn8ll1SKt4ezqzkLu/3spNszZyKDXXIW1ulLJPw5dXQn66NkNqyvccy9KCswqLEQohmi0JVIRwAsvwzx97UzDbOvyz+ycwGSvd5dvNJwEY2C6Q/xveDlcXHWuPZDD2vTXMmL+X7ILKj2uy8s/AlxMh+xQEtoObfgUPf2tV2igJVIS4aEigIoQTDIoNwtvNhdScYrafyqp+55hh4BkEBWfg8PIKL5eUmvlpy2kApvaP5t+j41nx8BAu6xSKyawyd90Jhr61km82JtScE2MjVVX5Zdtplu2toR6MMxRlw9dXQ8ZhbamBm38D72DgXFXaaBn6EeKiIYGKEE7g5qJnRIcQwIbhH70Bul2vbW//qsLLy/alkJFfQqivG8Pba+eMbOHJpzf35ps7+hEX6k1mgZGnf93D5R/8U3NejA1+2HKKh3/YyV1fbeWbjQl1Pp/NSgpg3vWQvFML3m75DfwjATCZVU5laoFKawlUhLhoSKAihJOMPa9KbY2Jrz1v0R4PLYXc8r0Y8zZqwz6Te0fioi//KzuwXRCLHxjMjMs74uvuwr7kHK77dD0frjxS63bvT87hud/3Wj9/5rc9LNyVVOvz2ay0BH64BU6uAzdfuPkXCIq1vpyUVYjRpOKq1xHu5+H89gghGgUJVIRwkiFxIXgY9JzOLGRPYk71OwfHQ2Q/UE2wY5716eNn8ll3NANFgev6RFZ6qItex60D27DykaHc0Ffb582lB/lte6Ldbc4rLuW+b7ZRXGpmaHwwU/q1RlXhX9/vYHVZMq9TmE3w63Q4slybBTXlBwjvVm6Xhbu0nqnIFh7odTI1WYiLRaMJVF599VUUReGhhx5q6KYI4RAernqGxmu5FUtqGv6Bc70q27+y1lT5bpPWmzI0LphWAdUPdwR6u/Hq1V2569K2ADz20y67hoFUVeXpX3dz7Ew+Yb7uvH1dd166sjPju4ZjNKlM/2or205m2nw+m5UWw893wN5fQGeAyV9DVP9yu/yy7TSv/3EAgBv6tnZ8G4QQjVajCFQ2b97MZ599RteuXRu6KUI4lGWRwkW7kyk11bD2T8eJ4OoNZ49BwlqKS038uFVLorXnn/MTY9ozplMYJSYzd321heNn8m067vvNp/h9RxJ6ncJ/p/SghZcrep3CO9d1Z3BsEIVGE7fN2ezYKdFF2fD1NWVBigtc8z+IHVlulz/3pfLoT7sAuG1gNLcPauO46wshGr0GD1Ty8vK48cYb+fzzzwkICGjo5gjhUCM6hOLnYSAho4C5605Uv7ObN3S+Rtve9iXL9qZy9oIkWlvodArvTO5Ot1Z+ZBUYmTZ3M5n5JdUesz85h+fna3kpj14WT+/oFtbXXF10fHpzL3q09ie70MjNszZy6myBze2pUm4KzBkPJ9ZoAdqNP0KnieV22Xgsg/vmbcNkVrm6R0ueHd9RKtIKcZFp8BW97rvvPsaPH8/IkSOZOXNmtfsWFxdTXFxs/TwnRxv3NxqNGI2OrSNhOZ+jzysq11zvt5sOHr8slqd+28d/lh1kRHwQrQKqTgRVuk7BZdsXqPt+57e0awGY1LMlqtmE0Wx76XwXBT6e0p1Jn27k+Jl87vxyM3Nv7Y2bi/be5Pz7nVdcyr1fb6W41MyQuCBuuySywvfBoMBnN/ZgyqxNHE7L56b/beS7O/sQ5O1m7y3RZBzG5dvrULJPoXoFUzr5Oy0n5bzr7k3K4fYvtlBcamZ4fDAzr+yAyVSKqQmuINBcf74bK7nf9as299uefRW1Aetwf/fdd7z88sts3rwZd3d3hg4dSvfu3Xn33Xcr3X/GjBm88MILFZ6fN28enp4yXVE0TqoK/92n50iOQgd/M9Pbm6myU0BVGXbgaXyLTvOM8Ta+MY3kuZ4mWtQyHkgqgPf26CkyKfQOMnNTu/LXVlX48rCObRk6/F1VHu1qwttQ9fmyS+DdPXrOFiu09FS5v5MJTzvf7gTkH6Hf0bdxM+WR5xbK+phHKXAr32OUVgjv7dWTZ1SI8VG5u4MJV7191xFCNF4FBQVMmTKF7OxsfH19q923wXpUTp06xYMPPsiyZctwd3e36Zgnn3yShx9+2Pp5Tk4OkZGRjB49usYv1F5Go5Hly5czatQoDIZq/nILh2ju97tD33wm/H97dx4dVX33cfw9k2UCIQkgkElIgMiWSkJiNiAHWSUFhGqpFSyWqOBCRaH6uGB7CqfYE7QVH2PA3QhoxSpQ0EeUeAjBjTYEImOkiGWtECIoJATI+nv+GBhJE1RgtsDndc49k9x7c+d3v15mPv7u79678GO2HbFCt2TGJtrPuq618z4o+D2TAgrZ1/NGbvp5ygW9d3zyYaYt3cymQ1YyE3tz94iernof7dSPzRu3E2C18MyUdFK7//Dp18yrjjPx+X/yVXUtyys68dKUVNr8yBRh2fEeASv+jKXhBI1RV2Kb+BrDQjs1Wae88iSTnv8nx+pO8hN7GK9OTSMspHUfExf78e1vVG/vOp96nz4j8mP4LKiUlJRQUVFBamqqa15DQwMbNmwgLy+PmpoaAgKafvjZbDZstub/axkUFOSxg9GT25bmLtZ6941uz4zhvXni/S945J1/MTzeTkTblvezJnEiDQVzSbDuZnrf6guux/Cf2HnkugRmr3CQW/hv4rq0Y1xCJF9Vw/8W7wDggZ/2ZWCvHzcOppc9giVTM5j07EY27TnCrDccPH1TCrbAHwgrm5fAW7Ocl2D3GoX1ly9jtbVrssqR47VMXbKZr46cpMdlbVkydQAdw86zO8kPXazHt79Svb3rXOp9Lv9dfDaYduTIkTgcDkpLS11TWloakydPprS0tFlIEWnt7hx2Ob26tOPQsVpy1mw763prd9XzXkMaABnfvu2W974xoxt3DHVetvzgmw7Wf/E1+V8EUFvfyIj4Ltx21eXntL1+0RG8eHM6tkAr6/5VwdSXN3Gspr7llY2Boj/D6rudISV5Mtz4mnPw8Bmqa+q5Ob+YLw4eIzLcxtKpA+h8EYUUETk/PgsqYWFhJCQkNJlCQ0O57LLLSEhI8FWzRDzGFhhAzoREAJYV72PjzsMtrvfXf+zl9YbhAFgdb0LdCbe8/4M/jWdMgvOy5duWbuHrkxbs4TYe/2US1vO4gVpGXEdezE6nbXAAH355iMnPb+TwsZqmK1XuhzeyofDUQPmr7oNrFzofG3CGmvoG7nylhNJ9R2jfNoilUwcQ21HjzkTEDy5PFrmUpPfoyK8GOO+J8vBKBzX1TS9h2fn1MT7ZeZhPTD/qw2Kg5ih8vtot7336suXk2PbO3zE8OTGJDqHB573Nwb078dptA+nQNohP/3OUXz77CV8dOeF8CvTHT0FeOny+CixWGPsXGPkH/nsk8WdfHeXavI/4YMch2gYHkH9zOn0iwy5kV0XkIuJXQWX9+vVnveJH5GLx4Oh4OofZ2Pl1NYsK/91k2bLifQAM7WsnMPXUnWo3L3Hbe4cEBfD8lDSuT+lKdp9GUrq1v+BtJsW25407M4mOCGHn19X8Ke85avIyYe3vofYYxGTA7esh47Ymf1db38iCtdu5buFH/Ku8io6hwbwwJY0ru+l+SiLyHb8KKiKXgog2Qcwd3w+AReu/5MsK551ea+obePPMO9FeORmwwJ4P4fC/z7a5c9Y5zEbOz/uRfJn77kzQq0s7Vk7pyQthz7Ko/g/Yvv2CupCOztM8t77X7Lk9n311lJ/lfUjuui+pbzRckxjF2t8OIbNXp7O8g4hcqhRURHxgbKKdkfFdqGswzF7hoLHR8N6pO9Haw0MY3rczRMRAr1O3k9+y1LcN/j4N9fDJIiIXD+bquiIasbC0/mquqv4z69tmgfW7j5nTvSjXntGLsvBXKSycnHL+N48TkYuagoqID1gsFv54XQJtgwMo3v0ty4r38do/nA8gvCE9lsCAU/80U37tfC39qzMQ+JvdH8GzQ+C92VBbBV1TqbnlfQouf5DyujZMW7yJVaXOpzif2YvScKoXpeC3Q7imf5SPd0JE/JnPb6Evcqnq2r4N/5PVlz++/Tl/+r/Pqa5twGqBiemx363UZwy07QTHDsKOtRA/1ncNPtPX2+H9ubD9HefvbTrC1XPhyl/TxmrlhSmN3P/mp6wq3c/MZaW8V1bOe2UHaWg0XBYazB9PPZVZROSHqEdFxIeyM3vQPyaC6lrn1T/D+naha/szngUUGAxJk5w/u3FQ7XmrPACr74FFA50hxRIAaVPh7hJIzXad5gkOtPLEDcncnNkDgHcc5c5elP7OsSgKKSLyY6lHRcSHAqwWciYk8rO8j2hoNPwqo1vzlVKmwCd5zh6VygMQ7oMv+ZOV8HEufLIQ6k49OTl+nLMXpVPvFv/EarUwZ/wVRLcP4a1PDzB9WE/GJiqgiMi5UVAR8bF+0RHkTrqS3YerGRHfwm3sO/eF2AGw7x/w6V+dN027UCe+JbC+2nnX2O/TUAclL8P6+XD8kHNeTAZkzYNuA3/wbSwWC7cP6cntQ3peeJtF5JKkoCLiB37wVEjKFGdQ2bwUBt/b7KZp36vuJJQ74D/FzumrTQQd2cs1gPlsBrTt6BxjcuZr244QHAafvgbfnLo0+rJeMHIO/GT8ub2/iMgFUFARaQ2uuA7WPATf7oK/9IZ2dgiLbPnVFgYHy1yhhANbobGuxc1aTANUf+2czia0Mwx7CFKym936XkTE0xRURFoDWzsYcDt88Ph3weKg48f/fdtOEJMOManQNY26Lv159/11jB4ygKC6o3D8GzjxTdPX499A5z6Qcbsz/IiI+ICCikhrMfIPMOBOqDoAVQfhWLnzsuXTP59+PXHUOcA1Jh1i0pxT++5NT9fU1dFoDXYOzA1qYQCviIifUFARaU3adXFOunhGRC4Ruo+KiIiI+C0FFREREfFbCioiIiLitxRURERExG8pqIiIiIjfUlARERERv6WgIiIiIn5LQUVERET8loKKiIiI+C0FFREREfFbCioiIiLitxRURERExG8pqIiIiIjfUlARERERvxXo6wZcCGMMAJWVlW7fdl1dHcePH6eyspKgoCC3b1+aUr29S/X2LtXbu1Rv7zqfep/+3j79Pf59WnVQqaqqAiA2NtbHLREREZFzVVVVRURExPeuYzE/Js74qcbGRvbv309YWBgWi8Wt266srCQ2NpZ9+/YRHh7u1m1Lc6q3d6ne3qV6e5fq7V3nU29jDFVVVURHR2O1fv8olFbdo2K1WomJifHoe4SHh+tA9yLV27tUb+9Svb1L9fauc633D/WknKbBtCIiIuK3FFRERETEbymonIXNZmPOnDnYbDZfN+WSoHp7l+rtXaq3d6ne3uXperfqwbQiIiJycVOPioiIiPgtBRURERHxWwoqIiIi4rcUVERERMRvKai0YNGiRcTFxRESEkJqaioffPCBr5t0UdiwYQPjx48nOjoai8XC3//+9ybLjTHMnTuX6Oho2rRpw7BhwygrK/NNYy8COTk5pKenExYWRpcuXbjuuuvYvn17k3VUc/d5+umn6d+/v+umV4MGDWLNmjWu5aq1Z+Xk5GCxWJg1a5ZrnmruPnPnzsVisTSZ7Ha7a7kna62g8l9ef/11Zs2axe9+9zu2bNnCVVddxZgxY9i7d6+vm9bqVVdXk5SURF5eXovLH3vsMRYsWEBeXh7FxcXY7XZGjRrleqaTnJuioiLuuusuNm7cSEFBAfX19WRlZVFdXe1aRzV3n5iYGObPn8+mTZvYtGkTI0aM4Nprr3V9WKvWnlNcXMxzzz1H//79m8xXzd2rX79+HDhwwDU5HA7XMo/W2kgTGRkZ5s4772wyLz4+3jz00EM+atHFCTArV650/d7Y2GjsdruZP3++a97JkydNRESEeeaZZ3zQwotPRUWFAUxRUZExRjX3hg4dOpgXXnhBtfagqqoq07t3b1NQUGCGDh1qZs6caYzR8e1uc+bMMUlJSS0u83St1aNyhtraWkpKSsjKymoyPysri48//thHrbo07Nq1i/Ly8ia1t9lsDB06VLV3k6NHjwLQsWNHQDX3pIaGBpYtW0Z1dTWDBg1SrT3orrvu4pprruHqq69uMl81d78dO3YQHR1NXFwckyZNYufOnYDna92qH0robocOHaKhoYHIyMgm8yMjIykvL/dRqy4Np+vbUu337NnjiyZdVIwx3HvvvQwePJiEhARANfcEh8PBoEGDOHnyJO3atWPlypVcccUVrg9r1dq9li1bxubNmykuLm62TMe3ew0YMIAlS5bQp08fDh48yCOPPEJmZiZlZWUer7WCSgssFkuT340xzeaJZ6j2njFjxgy2bt3Khx9+2GyZau4+ffv2pbS0lCNHjrB8+XKys7MpKipyLVet3Wffvn3MnDmTtWvXEhISctb1VHP3GDNmjOvnxMREBg0aRM+ePVm8eDEDBw4EPFdrnfo5Q6dOnQgICGjWe1JRUdEsKYp7nR49rtq73913383q1aspLCwkJibGNV81d7/g4GB69epFWloaOTk5JCUl8eSTT6rWHlBSUkJFRQWpqakEBgYSGBhIUVERubm5BAYGuuqqmntGaGgoiYmJ7Nixw+PHt4LKGYKDg0lNTaWgoKDJ/IKCAjIzM33UqktDXFwcdru9Se1ra2spKipS7c+TMYYZM2awYsUK1q1bR1xcXJPlqrnnGWOoqalRrT1g5MiROBwOSktLXVNaWhqTJ0+mtLSUyy+/XDX3oJqaGrZt20ZUVJTnj+8LHo57kVm2bJkJCgoyL774ovn888/NrFmzTGhoqNm9e7evm9bqVVVVmS1btpgtW7YYwCxYsMBs2bLF7NmzxxhjzPz5801ERIRZsWKFcTgc5sYbbzRRUVGmsrLSxy1vnaZPn24iIiLM+vXrzYEDB1zT8ePHXeuo5u4ze/Zss2HDBrNr1y6zdetW8/DDDxur1WrWrl1rjFGtveHMq36MUc3d6b777jPr1683O3fuNBs3bjTjxo0zYWFhru9GT9ZaQaUFCxcuNN27dzfBwcEmJSXFdTmnXJjCwkIDNJuys7ONMc5L3ObMmWPsdrux2WxmyJAhxuFw+LbRrVhLtQZMfn6+ax3V3H1uvfVW1+dG586dzciRI10hxRjV2hv+O6io5u4zceJEExUVZYKCgkx0dLSZMGGCKSsrcy33ZK0txhhz4f0yIiIiIu6nMSoiIiLitxRURERExG8pqIiIiIjfUlARERERv6WgIiIiIn5LQUVERET8loKKiIiI+C0FFREREfFbCioi4jE333wzFosFi8VCUFAQkZGRjBo1ipdeeonGxkZfN09EWgEFFRHxqNGjR3PgwAF2797NmjVrGD58ODNnzmTcuHHU19f7unki4ucUVETEo2w2G3a7na5du5KSksLDDz/MqlWrWLNmDS+//DIACxYsIDExkdDQUGJjY/nNb37DsWPHAKiuriY8PJw333yzyXbfeustQkNDqaqqora2lhkzZhAVFUVISAg9evQgJyfH27sqIh6goCIiXjdixAiSkpJYsWIFAFarldzcXD777DMWL17MunXreOCBBwAIDQ1l0qRJ5OfnN9lGfn4+119/PWFhYeTm5rJ69Wr+9re/sX37dl555RV69Ojh7d0SEQ8I9HUDROTSFB8fz9atWwGYNWuWa35cXBzz5s1j+vTpLFq0CIBp06aRmZnJ/v37iY6O5tChQ7z99tsUFBQAsHfvXnr37s3gwYOxWCx0797d6/sjIp6hHhUR8QljDBaLBYDCwkJGjRpF165dCQsLY8qUKRw+fJjq6moAMjIy6NevH0uWLAFg6dKldOvWjSFDhgDOQbulpaX07duXe+65h7Vr1/pmp0TE7RRURMQntm3bRlxcHHv27GHs2LEkJCSwfPlySkpKWLhwIQB1dXWu9adNm+Y6/ZOfn88tt9ziCjopKSns2rWLefPmceLECW644Qauv/567++UiLidgoqIeN26detwOBz84he/YNOmTdTX1/P4448zcOBA+vTpw/79+5v9zU033cTevXvJzc2lrKyM7OzsJsvDw8OZOHEizz//PK+//jrLly/nm2++8dYuiYiHaIyKiHhUTU0N5eXlNDQ0cPDgQd59911ycnIYN24cU6ZMweFwUF9fz1NPPcX48eP56KOPeOaZZ5ptp0OHDkyYMIH777+frKwsYmJiXMueeOIJoqKiSE5Oxmq18sYbb2C322nfvr0X91REPEE9KiLiUe+++y5RUVH06NGD0aNHU1hYSG5uLqtWrSIgIIDk5GQWLFjAo48+SkJCAq+++upZLy2eOnUqtbW13HrrrU3mt2vXjkcffZS0tDTS09PZvXs377zzDlarPuJEWjuLMcb4uhEiIj/Gq6++ysyZM9m/fz/BwcG+bo6IeIFO/YiI3zt+/Di7du0iJyeHO+64QyFF5BKiflER8XuPPfYYycnJREZGMnv2bF83R0S8SKd+RERExG+pR0VERET8loKKiIiI+C0FFREREfFbCioiIiLitxRURERExG8pqIiIiIjfUlARERERv6WgIiIiIn5LQUVERET81v8DeEGlw/AO4MUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y_test_inv[-50:], label='Actual')\n",
    "plt.plot(y_pred[-50:], label='Predicted')\n",
    "plt.xlabel('Days')\n",
    "plt.ylabel('Close Price')\n",
    "plt.title('Actual vs Predicted Close Price for AMC')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "e83e38ef-77b0-4069-bace-a5bdf9746d08",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE =  0.8364303522075167\n"
     ]
    }
   ],
   "source": [
    "mae = np.mean(np.abs(y_test_inv - y_pred))\n",
    "print(\"MAE = \",mae)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ddeec8a4-0cc7-41ff-be56-c8d891b323fb",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.7.12"
  },
  "toc-showmarkdowntxt": true
 },
 "nbformat": 4,
 "nbformat_minor": 5
}