draft_basic2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SETUP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set_style('whitegrid')\n",
    "plt.style.use(\"fivethirtyeight\")\n",
    "%matplotlib inline\n",
    "\n",
    "# For reading stock data from yahoo\n",
    "from pandas_datareader.data import DataReader\n",
    "import yfinance as yf\n",
    "\n",
    "# For time stamps\n",
    "from datetime import datetime\n",
    "\n",
    "\n",
    "from pathlib import Path\n",
    "from warnings import simplefilter\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from statsmodels.tsa.deterministic import CalendarFourier, DeterministicProcess\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from statsmodels.tsa.deterministic import CalendarFourier, DeterministicProcess\n",
    "from xgboost import XGBRegressor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LOAD DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[*********************100%***********************]  1 of 1 completed\n",
      "[*********************100%***********************]  1 of 1 completed\n",
      "[*********************100%***********************]  1 of 1 completed\n",
      "[*********************100%***********************]  1 of 1 completed\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'AAPL' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn [3], line 13\u001b[0m\n\u001b[1;32m     10\u001b[0m \u001b[39mfor\u001b[39;00m stock \u001b[39min\u001b[39;00m tech_list:\n\u001b[1;32m     11\u001b[0m     \u001b[39mglobals\u001b[39m()[stock] \u001b[39m=\u001b[39m yf\u001b[39m.\u001b[39mdownload(stock, start, end)\n\u001b[0;32m---> 13\u001b[0m company_list \u001b[39m=\u001b[39m [AAPL, GOOG, TSLA, AMZN]\n\u001b[1;32m     14\u001b[0m company_name \u001b[39m=\u001b[39m [\u001b[39m\"\u001b[39m\u001b[39mNASDAQ\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mGOOGLE\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mTESLA\u001b[39m\u001b[39m\"\u001b[39m, \u001b[39m\"\u001b[39m\u001b[39mAMAZON\u001b[39m\u001b[39m\"\u001b[39m]\n\u001b[1;32m     16\u001b[0m \u001b[39mfor\u001b[39;00m company, com_name \u001b[39min\u001b[39;00m \u001b[39mzip\u001b[39m(company_list, company_name):\n",
      "\u001b[0;31mNameError\u001b[0m: name 'AAPL' is not defined"
     ]
    }
   ],
   "source": [
    "# The tech stocks we'll use for this analysis\n",
    "tech_list = ['NDAQ', 'GOOG', 'TSLA', 'AMZN']\n",
    "\n",
    "# Set up End and Start times for data grab\n",
    "tech_list = ['NDAQ', 'GOOG', 'TSLA', 'AMZN']\n",
    "\n",
    "start = '2010-01-01'\n",
    "end = '2022-10-30'\n",
    "\n",
    "for stock in tech_list:\n",
    "    globals()[stock] = yf.download(stock, start, end)\n",
    "\n",
    "company_list = [AAPL, GOOG, TSLA, AMZN]\n",
    "company_name = [\"NASDAQ\", \"GOOGLE\", \"TESLA\", \"AMAZON\"]\n",
    "\n",
    "for company, com_name in zip(company_list, company_name):\n",
    "    company[\"company_name\"] = com_name\n",
    "    \n",
    "df = pd.concat(company_list, axis=0)\n",
    "df.tail(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# HYBRID MODEL"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Learn from the trend"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 365,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">Close_price</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Companies</th>\n",
       "      <th>TESLA</th>\n",
       "      <th>AMAZON</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-06-29</th>\n",
       "      <td>1.592667</td>\n",
       "      <td>5.4305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-06-30</th>\n",
       "      <td>1.588667</td>\n",
       "      <td>5.4630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-01</th>\n",
       "      <td>1.464000</td>\n",
       "      <td>5.5480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-02</th>\n",
       "      <td>1.280000</td>\n",
       "      <td>5.4570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-06</th>\n",
       "      <td>1.074000</td>\n",
       "      <td>5.5030</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Close_price        \n",
       "Companies        TESLA  AMAZON\n",
       "Date                          \n",
       "2010-06-29    1.592667  5.4305\n",
       "2010-06-30    1.588667  5.4630\n",
       "2010-07-01    1.464000  5.5480\n",
       "2010-07-02    1.280000  5.4570\n",
       "2010-07-06    1.074000  5.5030"
      ]
     },
     "execution_count": 365,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tesla = df.loc[df[\"company_name\"].isin([\"TESLA\"])]\n",
    "df_tesla = df_tesla[\"Close\"]\n",
    "# df_tesla = df_tesla.rename(columns = {'Close': 'Close_tesla'}, inplace=True) \n",
    "df_tesla.head()\n",
    "df_amazon = df.loc[df[\"company_name\"].isin([\"AMAZON\"])]\n",
    "df_amazon = df_amazon[\"Close\"]\n",
    "# df_amazon = df_amazon.rename(columns = {'Close': 'Close_amazon'}, inplace=True) \n",
    "df_amazon.head()\n",
    "tickers = [\"AMAZON\", \"TESLA\"]\n",
    "df_ta = pd.merge(df_tesla, df_amazon, right_index=True, left_index=True)\n",
    "df_ta.rename(columns = {list(df_ta)[0]: 'TESLA'}, inplace = True)\n",
    "df_ta.rename(columns = {list(df_ta)[1]: 'AMAZON'}, inplace = True)\n",
    "# df_ta = pd.concat({'Close': df_ta}, names=[None, 'company_name'], axis=1)\n",
    "df_ta.reindex(columns=tickers)\n",
    "df_ta=pd.concat({'Close_price': df_ta}, names=[None, 'Companies'], axis=1)\n",
    "df_ta.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 366,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<statsmodels.tsa.deterministic.DeterministicProcess object at 0x14ef09420>\n",
      "            const   trend  trend_squared   trend_cubed\n",
      "Date                                                  \n",
      "2010-06-29    1.0     1.0            1.0  1.000000e+00\n",
      "2010-06-30    1.0     2.0            4.0  8.000000e+00\n",
      "2010-07-01    1.0     3.0            9.0  2.700000e+01\n",
      "2010-07-02    1.0     4.0           16.0  6.400000e+01\n",
      "2010-07-06    1.0     5.0           25.0  1.250000e+02\n",
      "...           ...     ...            ...           ...\n",
      "2022-10-24    1.0  3103.0      9628609.0  2.987757e+10\n",
      "2022-10-25    1.0  3104.0      9634816.0  2.990647e+10\n",
      "2022-10-26    1.0  3105.0      9641025.0  2.993538e+10\n",
      "2022-10-27    1.0  3106.0      9647236.0  2.996432e+10\n",
      "2022-10-28    1.0  3107.0      9653449.0  2.999327e+10\n",
      "\n",
      "[3107 rows x 4 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/y9/x007ynb92t57yt7s517ct53w0000gn/T/ipykernel_80750/2753399662.py:39: UserWarning: When passing multiple axes, sharex and sharey are ignored. These settings must be specified when creating axes.\n",
      "  axs = y_test.plot(color='0.25', subplots=True, sharex=True, ax=axs)\n",
      "/var/folders/y9/x007ynb92t57yt7s517ct53w0000gn/T/ipykernel_80750/2753399662.py:40: UserWarning: When passing multiple axes, sharex and sharey are ignored. These settings must be specified when creating axes.\n",
      "  axs = y_fit.plot(color='C0', subplots=True, sharex=True, ax=axs)\n",
      "/var/folders/y9/x007ynb92t57yt7s517ct53w0000gn/T/ipykernel_80750/2753399662.py:41: UserWarning: When passing multiple axes, sharex and sharey are ignored. These settings must be specified when creating axes.\n",
      "  axs = y_pred.plot(color='C3', subplots=True, sharex=True, ax=axs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            const   trend  trend_squared   trend_cubed\n",
      "Date                                                  \n",
      "2010-06-29    1.0     1.0            1.0  1.000000e+00\n",
      "2010-06-30    1.0     2.0            4.0  8.000000e+00\n",
      "2010-07-01    1.0     3.0            9.0  2.700000e+01\n",
      "2010-07-02    1.0     4.0           16.0  6.400000e+01\n",
      "2010-07-06    1.0     5.0           25.0  1.250000e+02\n",
      "...           ...     ...            ...           ...\n",
      "2022-01-06    1.0  2903.0      8427409.0  2.446477e+10\n",
      "2022-01-07    1.0  2904.0      8433216.0  2.449006e+10\n",
      "2022-01-10    1.0  2905.0      8439025.0  2.451537e+10\n",
      "2022-01-11    1.0  2906.0      8444836.0  2.454069e+10\n",
      "2022-01-12    1.0  2907.0      8450649.0  2.456604e+10\n",
      "\n",
      "[2907 rows x 4 columns]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmkAAAHOCAYAAADDgJ4ZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/av/WaAAAACXBIWXMAAA9hAAAPYQGoP6dpAACR2klEQVR4nOzdd3gU1f7H8ff23fQeCC30EnoHQQREERWxg3oVLlfs4s/eu9dyFbGhIhZUVMCCqAgC0osgIL2XkALpbZNs398fYRdCNskm2SSb5Pt6njwkM2dmzh42ySdnzpyjyM3NdSKEEEIIIfyKsr4rIIQQQgghypKQJoQQQgjhhySkCSGEEEL4IQlpQgghhBB+SEKaEEIIIYQfkpAmhBBCCOGHJKQJIYQQQvghCWlCCCGEEH5IQpoQQgghhB9S13cFhBBNy+zZs5kzZ06Vjunbty8fffRRLdXId15++WUWL17Mf/7zH6ZNm1bf1RFCNHAS0oQQdapjx46MHTu2zPalS5cCMHDgQCIiIkrti4+Pr4uqCSGEX5GQJoSoUyNHjmTkyJFltrtC2pQpU+jXr19dV0sIIfyOjEkTQgghhPBD0pMmhPB7qampTJgwge7du3Pffffx+uuvk5SUREREBE8++SRDhgwBIDMzky+//JINGzaQlpaGTqejc+fOXHvttYwePbrUObdt28Zdd93FmDFjePDBB5k9ezYbNmwgJyeHiIgILrzwQv79738TGRlZpj5btmzh66+/Zv/+/VitVrp3717hGLTU1FQ+//xzdu3aRWpqKmq1mjZt2jBy5EhuvPFG9Hq9bxtMCNEoSEgTQjQY6enpPPjgg0RHRzN06FD2799P165dAdi7dy//93//R25uLs2aNWPw4MEUFRWxc+dO/v77byZMmMCTTz7p8Zy33XYbRqOR7t2707FjR7Zt28bChQvZtGkT8+bNw2AwuMt/++23zJw5E4CePXsSGRnJ7t27ueOOO2jVqlWZ8ycnJzN58mTy8/OJj49n6NChmEwm/vnnH/bt28e6dev4+OOPUalUtdNoQogGS0KaEKLBSE9P54ILLuCtt95CqVTicDhQKpUYjUYeeeQRcnNzueOOO5g8ebI79Jw8eZIHHniARYsWuXvVzrVz50769OnDK6+8QlRUFAApKSlMnjyZ5ORkli1bxoQJEwA4cuQI7777Lmq1mrfeeovBgwcDYLFYeOWVV/j999/L1Pmrr74iPz+ff/3rX9x3333u7ZmZmfz73/9m165drFu3josuuqgWWkwI0ZDJmDQhRINy0003oVSW/Ohy/bt48WIyMzO54IILmDp1aqleqdatW/P4448DMHfuXI/nfPzxx90BDaBFixbuhxsOHz7s3v7DDz9gt9u54YYb3AENQKvV8uSTT3q8NZqeng5QppctKiqKp556imeeeYb27dt73wBCiCZDQpoQokHp3LlzmW1btmwBKBWcztWvXz8MBgOnT58mMTGx1L6QkBDatm1b5hhXaDOZTO5tW7duBWD48OFlyut0Oi644IIy2/v37w/A//73P5577jmWLVtGdnY2AIMGDeLKK6/0eJtUCCHkdqcQosFQKpUEBweX2X7q1CkA3nrrLd56660Kz5GWlkabNm3cX3s6H+DujXM4HO5tGRkZAMTGxno8pkWLFmW2TZw4kZMnT7Jo0SJ+//13fv/9dxQKBR06dODCCy9kwoQJ5Z5PCNG0SUgTQjQYCoUChUJRZrsrSA0aNIjw8PAKz3F+KPN0voquD+B0Oj3uV6vL/khVqVQ88cQTTJ48mbVr17JlyxZ27tzJ4cOHOXz4MF9//TVvvfUWAwcO9LoeQoimQUKaEKLBi46OJjExkWuuucbjRLm+EhMTw4kTJ0hNTaVly5Zl9rvGn3nSvHlzbrzxRm688UYcDgf79u3jww8/ZOvWrcycOZNvvvmm1uothGiYZEyaEKLBc61QsGbNGo/7XQHurrvuIi8vr9rXcY15W7lyZZl9drudDRs2lNrmcDi45557uPjii0lLS3NvVyqVdO/enenTpwNnb9cKIcS5JKQJIRq8a665huDgYH7//Xfmzp2L3W5378vJyeG5554jOTkZvV5PaGhota9zww03oNfr+fnnn93LWAHYbDZmzJhBcnJyqfJKpZLw8HDy8/OZMWNGqYcQHA4Hv/32GwDdu3evdp2EEI2X3O4UQjR44eHhvPrqqzz22GN88MEHLFy4kE6dOmGz2fjnn38wmUy0bduWp59+ukbXadmyJU8//TTPP/88zz77LN9++y1xcXHs37+fU6dO0bNnT3bt2lXqmPvvv59//vmHVatWsWPHDhISElCr1Rw5coSUlBRCQkJ44IEHalQvIUTjJCFNCNEoDBw4kHnz5jFv3jw2b97Mli1b0Ov1xMfHc/HFF3PNNdcQFBRU4+tccskltGzZkrlz57Jz506OHTtGx44defjhhzl69GiZkBYTE8Nnn33G3Llz2bx5M1u3bkWhUBAbG8vEiRO59dZbS83RJoQQLorc3FzPjykJIYQQQoh6I2PShBBCCCH8kIQ0IYQQQgg/JCFNCCGEEMIPSUgTQgghhPBDEtKEEEIIIfyQhDQhhBBCCD8kIU0IIYQQwg9JSBNCCCGE8EMS0oQQQggh/JCENCGEEEIIPyQhTQghhBDCD0lIE0IIIYTwQxLShBBCCCH8kIQ0IYQQQgg/JCFNCCGEEMIPSUgTQgghhPBDEtKEEEIIIfyQhDQhhBBCCD8kIU0IIYQQwg9JSBNCCCGE8EMS0oQQQggh/JCENCGEEEIIPyQhTQghhBDCD0lIE0IIIYTwQxLShBBCCCH8kIQ0IYQQQgg/JCFNCCGEEMIPSUgTQgghhPBDEtKEEEIIIfyQhDQhhBBCCD8kIU0IIYQQwg9JSBNCCCGE8EMS0oQQQggh/FCjDmkmk4ljx45hMpnquyoNmrSj70hb+oa0o29IO/qGtKNvSDuW1ahDGoDdbq/vKjQK0o6+I23pG9KOviHt6BvSjr4h7Vhaow9pQgghhBANkYQ0IYQQQjRpVpuFtbt/w2a31ndVSlHXdwWEEEIIIerT9+s+Zl/i32w9+Cejel9Nr3ZDUSrrvx+r/msghBBCCFFPth5cxb7EvwHINWby4/pP+GDx0+w/uR2n01mvdZOQJoQQQogm6VT2SX7ZPLfM9vTcFJZvXyghTQghhBCirlltFr5eMaPcIHb98Dvr/ZanjEkTQgghRJOzcO2H5BfleNw3tv9EYsNbUVBQ4PW8bXq9nsDAQJ8GO5+GNLvdzp133snOnTvZuHEjavXZ058+fZoPP/yQrVu3UlBQQMeOHZk8eTIXXnhhmfPs2rWLjz/+mAMHDuB0OunTpw/33HMP7dq182V1hRBCCNEEbd6/nP0nt3vc1z4ugcFdLyErK4ugoCCioqJQKBQVns/pdGIymcjKyiIyMtJnQc2n/XhffPEFO3fuLLM9MzOTO+64g7Vr13LllVcyffp0bDYbDz/8MEuXLi1Vdtu2bdx1112kpaUxefJkbrvtNvbt28fUqVM5ceKEL6srhBBCiDridDpJSkpiw4YNpKWl1Vs9dh3bzG9/fe1xX4AumIkX3UtRURFBQUEYDIZKAxqAQqHAYDAQFBREYWGhz+rqs560vXv3MmfOHLRaLRaLpdS+OXPmcOrUKT755BN69eoFwBVXXMGUKVOYMWMGI0aMwGAw4HQ6eeONNwgJCeHTTz8lNDQUgJEjR3LzzTczc+ZMZs6c6asqCyGEEKKOfPPNNyxcuBAAtVrNPffcw8iRI+u0Dum5qSxc+2G5+28d8xB6bQAFeRlERUVV+fx6vZ7MzEyCg4NrUk03n/SkFRUV8eyzzzJkyBC6d+9eap/dbmfp0qV0797dHdCg5IXceOON5Obmsn79egD27dvH8ePHufzyy90BDaB169ZcdNFFbN68mczMTF9UWQghhBB15NSpU+6ABmCz2fj0008xm811VgerzcLcP94od/+4gTfTIqqt+2tvetDOV51jKuKTkPbWW29hNBp56qmnyuw7duwYRUVFZcIbQEJCAgB79uwp9a+nst27d8fhcLBv3z5fVFkIIYQQdWTNmjVlthmNRrZt2wZAfn4+K1asYOXKlSQlJdVKHeav/qDcBwU6t+rNkG6X1Mp1a6LGtztXrVrFL7/8wv/+9z8iIyPL7E9PTwcgNja2zL6YmBgAUlNTAdz3qL0pK4QQQoiG4bvvvvO4PTU1ldzcXJ566ilSUlIAWLlyJU8++SR9+vTx2fXX7f6Ng8n/eNynVmq4ccQ9PruWL9UopKWnp/PKK68wfvx4RowY4bGM0WgEwGAwlNmn1+sB3I+3ugbbBQQElFu2uLi40nq5zucaG3f+GDlRNdKOviNt6RvSjr4h7egb0o7VZzab+fPPP90BDcBqtbJgwQK6du3qk2scP72fP7Yt8LhPqVRx/1WvY7c5sNvOTrXhcDhwOBzVup7D4ahw2g5XnvFGtUOa0+nkhRdeIDg4mAcffLDCcpXtc93D9aasN4+1pqamYrfb3V/X51MkjYm0o+9IW/qGtKNvSDv6hrRj1WVnZ7Ny5coy2/fv3++T256F5nx+3l7+gwKX9ZhMZno2kF1qu6eHIL1lMpnIz8/3uE+lUlVpOrFqh7RvvvmGv//+m//9739YLBb3i7HZbADk5eWh0WgIDAx0V/p8rm2upyBcPWgVlQ0KCqq0bnFxcUDJXzVpaWnExsai1Wqr9PrEWdKOviNt6RvSjr4h7egb0o7VV9HUWq1atarRuR0OOx///gk2h9Xj/sv630zvDgM97svLy6v2/6Ver/c4bKs6qh3S1q1bh9Pp5OGHH/a4/7LLLqN58+bMmDEDODs27Vznj1dzhav09HQ6d+5cYdmKnN+VqNVqq9S9KDyTdvQdaUvfkHb0DWlH35B2rLqjR4+Wu0+n05V5WtLhcLBnzx6Sk5Pp1q0b8fHx5R7/7ar3yC7w3LvZs90QhnYv/0GBgoKCak9Iq1QqffY+qHZImz59OgUFBWW2z5w5kyNHjvDuu+9iMBho06YNQUFB7N27t0xZ17YePXoAZ5/q3Lt3L8OHDy9Vds+ePSgUCo9PfgohhBCicbHZbGg0GvfXTqeT999/nz///NO97e677+aSS8qGre/Xfsy+xL89njc6NI6rL/iP7ytcC6od0sob0BcSEgJA//793ctCjRkzhkWLFrFr1y569uwJlNy+XLBgAREREQwdOhSAbt260aZNGxYvXsykSZPcc6WdPHmSNWvWMHz4cMLCwqpbZSGEEEI0EFartVRIO378eKmABiUrHY0ePRqVSuXetmrnz+w8ttHjOTUqLZMvfRS1qvL443Q6qzzvWUVj66ujThZYv/3221m7di3Tp0/npptuIjw8nMWLF3Ps2DFefvlldDqdu+wjjzzC9OnTmTp1Ktdffz1ms5lvv/0WnU7HfffdVxfVFUIIIYQPVWcgvtVaeizZDz/8UKZMUVERBw8epFu3bgBsP7yWP3f8WO45b7vkEUICwiu9tl6vx2QyeZyZoiImk8mnt7zrJKRFRUUxZ84cPvjgA+bPn4/NZqN9+/bMmDGDCy64oFTZgQMH8t577zF79mzef/99DAYDvXr14u6776ZNmzZ1UV0hhBBC+Ijdbq/Wk5Lnh7TyHjJwPViYlHGURRs/K/d8lw+6hTaxnby6dmBgIFlZWUBJYPN2gXWj0ehxztjq8nlI++ijjzxub9GiBf/973+9Okf//v3p37+/L6slhBBCiHpQ3akszg9p586ldi6tVktSxlFm//ZiuefqHj+IwV3HeH1tpVJJZGQkhYWFXi9HqdfriYyMrPYDB57USU+aEEIIIZqm6q7P6ZrSqzJFlgIW/fF+ufu7tu7HjRfdXeXrK5VKgoODfbZYenX4Lu4JIYQQQpynotn3K3LuOPSioiKPZRQqWHvoe8xWz6sRRYU0r1ZA8xfSkyaEEEKIWlPdkHYuT1N+AbTuH0CeyfPtyJZR7Zl62ROolA036khPmhBCCCFqTXVvd57L07rdzRL0hDTXeChdMtXGTaPuR63yvL+hkJAmhBBCiFrji5608293RrTVEt1R57GsSqninqteIjggrMbXrW8S0oQQQghRa2rSk2Y0GoHSIS28jYYWvTzPX6ZSqrnzyheIDGlW7Wv6EwlpQgghhKg1NQlpmzZtAs72xgXFqGnZJ6Dc8jdedA/Nwmu2MLs/kZAmhBBCiFpTk9ude/bsAaCwsBB9qJI2A8sPaGP7T6Rr677VvpY/kpAmhBBCiFpTk54011qY6zevJn5wIEq155n/sxMtXND9smpfx19JSBNCCCFEralJSDt69CjF5kLMEcloDJ4jiynfTso/xeTl5VX7Ov5KQpoQQgghak1NbnempqYwb+VMdMGe40pBmpXDfxrBCRs2bKj2dfyVhDQhhBBC1Jqa9KR1vCiUxPRDHvcV59o5ueXsU5/p6enVvo6/arjT8AohhBDC71WrJ00B7S8MRBfqebelyMGJTYU47DWrm7+TnjQhhBBC1BpPPWlBQUEVHtOqr4GAcM/9SHark+MbCrGZnaW2L1q0yOtF2RsKCWlCCCGEqDWeQppery+3fPPuesJaaT3uczicHN9YiKXQ4XH/9u3bq1dJPyW3O4UQQghRazzd7iwvpMV21RHVwfNyTwBJW4sozin/Hue8efMYOHBgme2bN2/mtddec3/do0cPBg4cyG+//UZOTg6DBg1iypQphIeHV/RS6pyENCGEEELUGk89aVpt2Z6y5j30RLUvP6Cl7iom/1TFtzMTExPLbNu9e3epgObatnv3bvfXa9euJScnh5deeqnC89c1ud0phBBCiFrjTU/ayQ6jyVXH4rA7y5QFOLm1iKxjlmpd//fff/eq3O7du8nNza3WNWqL9KQJIYQQotZUNibtUM8bSW17IceADqdWcKFhBYHhZ/uQUnYWk5dirfb1N27c6HXZ3NxcwsLCqn0tX5OQJoQQQjRyO3fuZOvWraxYsYKQkBBatWrFxIkT6dixY61f21NI0+l0XHHFFbx3BFLbXujefqT5xSSZ+jHqxGzatcol47CZ7OPV60EDsNurNkdHTeZ0qw0S0oQQQohG7Ndff2XOnDnur00mE+np6Wzbto033niDTp061er1i4qKymzT6/UcaTuKRFXZMWhmfTir2tyLZe872I/kV+laKpWq1NdHjhyp0vH+FtJkTJoQQgjRSDkcDhYuXFju/kcffZTi4uJarYOn828J6MpHJ8t/SKDZ/pXYj6RU+VrnP52ZnJxcpeOzsrKqfM3aJCFNCCGEaKTS09MrXXj8iSeeqLXrJyUllfraCZzoNJbf1AnlHhN3fC3xh5ZV63rnhzSrtWpj2d55551qXbe2SEgTQgghmrATJ07U2rn/+eefUl8f63Y1J7peWW755ic20HHX/GpfT60uPYrLYqn6eLbqHFNbJKQJIYQQTVxtjcU6d0qLY13Hk9Tx4nLLPtrDQOed36CowfXOfx1V7UkDKCwsrEENfEtCmhBCCNFIeft0Y7UWQfeC61br8S5XcLLTpeWWe6hnEA92D6jSuQ0GQ5lt57+O6vSK+dP6nxLShBBCiEbK28BRW7f4iouLOd7lchI7X1ZumVcGhvJMv9Aqn7tHjx5ltp0f0ubPr/qtU7ndKYQQQoha521PWlUmfK2KH0ytSOw8rtz9rw8K5Z6EIPfXrVu39vrcQUFBZbZlZ2e7P6/uLVx/moZDQpoQQgjRSHkb0j7//HOfXtdms3HZrBUcaDm83DKvDAzljm6lg9awYcO8vobD4fC43eksWVoqPT3d63OdS3rShBBCCFHrqjLjvivc1JTT6WTq4oNsMnQtt8xLA0JK9aC59OnTh8cee4zRo0dXep0tW7Z43H7XXXexc+fOaoctCWlCCCGEqHVVGQSfl5eH1Wpl9erVfP3112Wmz/Dqeg4nU9fk8HNeWLllrnQe5L7uweXu79OnD/fdd1+l1/K0kgHA6dOneeGFF9i6dWul5/Ckth6iqA4JaUIIIUQjVZWetPT0dF599VVmzpzJ999/z/PPP8/333/v9fFFNgdXL8vkx+Plr2DQfs8PPDU0zutzVmTMmDHl7nM4HHz33Xfl7n/ttdcYN87zWLnaXoGhKiSkCSGEEI1UVXrS/vnnH7Zv315q29dff+1Vz1KexcGVv2ey7nT5two77F5Iq6N/0rlzZ6/rVJH+/ftX6zitVkuXLl3KDWmVrdBQlySkCSGEEI1UVXrSvvnmG4/bJ06cWOGksKeL7Fz2WwbbMssv03Hnd7Q8tppPP/0UhcK76WorCnMjR45Eq9V6dZ7zXXvttYDnedYAfvjhh2qdtzZISBNCCCEaKV/duitvio5ko43LlmSwL9dzj53CYaPL9i9pcWIdkydPJjIy0utrlnc7MyAggCuvvNLrsHe+fv36uc/jiT/1pKkrLyKEEEKIhshXgWPNmjWMGDGi1LZDuVauWJpJerHnqTCUNjMJW+cQmb4PgLi4qo1Fu/jii9FoNKxdu5bCwkLCwsJo3749w4YNo3nz5uzcubNar8XVg6bT6ap1fF2SkCaEEEI0UtUNMuc7f6za5jQzE1dkkWvxPG2HylpM7w0zCc5Ldm/r0KFDla87YsSIMuHQJSYmpsrnA4iKigJAqfT/m4n+X0MhhBBCVNkPP/zAtm3bymxv3rx5tc7nGt/29objjF2SWW5A05ry6Lv2f6UC2sCBA4mIiKjWdctT3dfREHrQXCSkCSGEEI3MiRMn+Oqrrzzu69SpU7XOmZh4ktt/3MULh8ofsK8vzKT3+rcJNKa5t40cOZJ77723WtesTLt27apU/vyHDbp161amTHBw+XO41TUJaUIIIUQjs2bNmnL3RUdHV/l8DoWSh/8uYmFe+QP/A/JT6bvuTQIKM9zb3nrrLaZPn05ISEiVr+mNsWPHVqn8iy++WOrrFi1alClTUFBQ7pJTdU1CmhBCCNHIlLdkEkBgYGCVzmXVBLJj+MP8WRhWbpmQ7GP0XfcWWnNBqe3t27ev0rWqyjW+zFvnT+tR3q3Pa665hqVLl5KWluZxf13xyYMDR44cYc6cOWzfvh2j0UhUVBQjRoxg2rRppboNjx49yocffsju3bsxm81069aNO+64g169epU557p16/jiiy84duwYarWaIUOGcM899xAbG+uLKgshhBCNVkpKSrn7AgICGDp0aLnTapzLGNycXUPvxaIPK7dMTNIWOv/zDSpH+fOk1ZaqDP73NEdbeXOlAXz00Uc8+uij9Zo7atyTlpiYyNSpU9m6dSsTJkzg4Ycfpl+/fixcuJDbb7/dvbbW8ePHmTZtGvv27eOGG27gzjvvJC0tjbvuuqvMUyNLly7l4Ycfxmazcccdd3D99dezbt06pk6dSnZ2dk2rLIQQQjRaR44cqXB/REREhUsquWTFdGPHhQ9XGNDa7ltEt+1zPQa0G2+8sdJr1JS3t1FHjBjhcY62ynoVq9pT52s17kl78803sdlsfPHFF7Rt2xYo6Sbs0qULb731FgsWLGDy5MnMnDkTi8XCl19+6b4HPHbsWCZOnMjrr7/O/PnzgZKJ92bMmEH79u2ZPXu2uytywIAB3HHHHXzyySc89thjNa22EEII0Sj98ssvFe5PSEiosAcJICV+OId73gCK8vtyumybS7Pk8m+rXnfddRVX1Afi4+O9Kjdx4kSP24cOHcoXX3xR7nH1ffeuRj1pZrOZHTt20Lt3b3dAc3GtibV9+3aysrLYtGkTF154YalBemFhYYwfP57jx4+zZ88eANavX09ubi7XXnttqXvFffr0oWfPnixdurTC5SmEEEKIpqyihwag5HZnebP1O1FwpPu1HO41sdyAprYY6bXh3XIDWmhoKO+88w4ajaZqFa8Gb253uia/9aSyhyjKW5WgrtSoJ02j0TB//nyPT0G4bkuqVCr27t0LQPfu3cuUc23bu3cv3bt3d4e1Hj16lCmbkJDArl27OHHiBB07dqxJ1YUQQogm5+GHHy53n0UbyL7+/yY3uku5ZQIKTtH9r49LPcF5rh49evDSSy/VuJ6+dMMNN5S7r7KlpeoiaFakRj1pSqWSFi1a0KpVqzL7vvzyS6Bkjaz09HTAc7eha8Zg1yBHV1lPMwm7tqWmptak2kIIIUSTNHToUI/b88PasO2iJyoMaOFp++i35vVyA1pgYCDPP/+8L6pZJRVNwzFmzBhat25dh7XxrVpZFurXX3/ll19+ITY2lquvvpqFCxcCnp+i0Ov1AJhMJgCMRiPguYvRVbayBWNd57JYLKX+FdUj7eg70pa+Ie3oG9KOvuFP7WizeV7o3MVTHU+3GsSBPrdUOP6s+fF1dNo1HwVlVxmYN28eCoUCpVKJ1Wqt9pCk6rbjJZdcwqZNm0qtU3rJJZcwYMAAunfv7s4EFR3/xx9/lNn+wgsvVHpsdbiyjDd8HtIWL17Mq6++isFg4PXXXycwMBCn0/PSEYB7nzer2bvKVnYPOjU11b18BVDv85w0FtKOviNt6RvSjr4h7egb/tCOq1evLnffk08+SVJSkvtrh0LJ0YRrSGk/svwTOux02PsjLY95Pu9VV13l87tb1WnH++67j71796JQKOjWrRuhoaEAJCcnV3IkdOnSpUxI69SpEwEBAaXayxdUKlWVVknwaUibPXs2c+bMITAwkBkzZriXW3D1inlKpK5trvnUzi17/vINZrO5VNnyxMXFASVpPC0tjdjY2DLnEt6TdvQdaUvfkHb0DWlH3/Cndvztt9/K3dezZ0/352nFDnYOnU5eVPmLnittFhK2fkJk+r5yy1x77bU+e801acdWrVp5HPfu7bGRkZEsX76cgoICxowZQ//+/at1Ll/zSUiz2Wy8/PLLLFmyhOjoaN5+++1Sa4O5QpNrvNm5zh+vdm7Z8+c/caVrT+PVznV+V6JWq61S96LwTNrRd6QtfUPa0TekHX2jvtvx3DtI54uPj3fXbWlSMdPW5JJfQUBrF6Qg+qeXMRRllVvmueeeq5XlnuqjHXv16uVxYv36VuPJbO12O0899RRLliyhffv2fPbZZ2UWb+3WrRtKpZJ9+8qmcdeTn66En5CQUGr7+WWDgoKqvKCqEEII0dhVNF573LhxOJxOXtyWx8QV2eRbyx+GND4O1l0VU2FA69ixI3369KlRfUXlahzSPvroI1atWkVCQgKzZ8/2+ARnZGQkAwYM4M8//yy1VEVubi6LFy+mY8eO7vW0hg0bRkhICAsWLCg1eHDHjh3s3r2byy+/3Kvxa0IIIURTUtEg97b9hnHR4gxm7DJWeI5uBxfz+cXNCNSqPU6F5XLNNddUu57CezW63ZmSksLXX3+NQqHgoosuYv369WXKhIeHM3jwYB544AGmTp3K7bffzqRJk9BqtSxYsACj0chrr73mLq/X63nggQd48cUXuf322xk/fjzZ2dl88803xMXFMWXKlJpUWQghhGiUXOO2z+UEbAOvZfhvOZjKvxuK2lJIwtY53Dd2ICqVCih5utFTGHv++efp3bu3j2otKlKjkLZp0yb3PfAPPvjAY5levXoxePBg9zJPs2bN4rPPPkOhUNC1a1eef/75Mmn9iiuuwGAwMHfuXN5++22Cg4O58MILufvuu4mIiKhJlYUQQohG6fzbnSZ9GPv7TSl5OKCCgNZebeR66xZGTbuOAQMGuLcrlUq+//571q1bR2JiIt26dWPAgAFyN6sO1SikXXfddVVam6tTp07MnDnTq7KjR49m9OjR1ayZEEII0bjs2LGD3377DZPJxMiRI8v8jjw3pGU078WBvrdiV1c8AP/R3sE80TsOhaKzx/1qtZqRIyuYokPUqlqZzFYIIYQQvrN//35efvll992rPXv24HQ6ufjii91lVqxYgVVt4FjC1ZyKv6DC80XrlcwaHs6YlvJUrz+r8YMDQgghhKhdq1atKjPFxooVK0p9vWhvKltHP1NpQBvTQsf6q2IkoDUA0pMmhBBC+DlPyxYdOHAAq9XK7n0HmJOiZ+fQ+ytc2ilAreDVgaHc2ilAxpU1EBLShBBCiAbqxnsf46+uN1IYGgEV5K6RcTo+vjCcGIOq7ionakxCmhBCCNHAOIGUtiM42v0anMryf5Ur7FZeGBTJfd2DpPesAZKQJoQQQjQgJn0YB/reSm605ycyXYJyE7mueCv395heRzUTviYhTQghhPBjWVklyzM5UZDcfiTHu1yBQ60r/wCng1ZHVtB2/y80HzyojmopaoOENCGEEMKP5eTkYAyJY3/f2ygMbVlhWW1xLl23fUF41mEAdLoKwpzwexLShBBCCD9ldTh58e9c/r7oiQqf3ASITv6bzju/RW07u4anhLSGTUKaEEII4Ye2pJu5d30uh4qbV/jkptpSSKed3xKTuqPMPo1GU4s1FLVNQpoQQgjhR/ItDp7/O5/PDhZWWjbi9G66/DMPrbnA4/7Q0FBfV0/UIQlpQgghhB+wO5x8frCQF7blU2B1VlhWY8qn4+4FRKfuqKiTjX79+vm2kqJOSUgTQggh6tm2DAv3b8hhb46t0rJxx9fRbt+iUmPPPLnzzjtp166dr6oo6oGENCGEEKKeJBltvPZPAfMOF1VaVl+YSecdX7uf3KzIa6+9RpcuXXxRRVGPJKQJIYQQdazY5uTNnfl8sNeIyV5JYYedVkdXEn9wCSq7tczu0NBQ+vbty/bt29FqtUyaNEkCWiMhIU0IIYSoI06nkx+OF/PM1jxOFTkqLR+cfZzOO78lKD+l3DLvvPMOYWFhPqyl8BcS0oQQQog68GeKief+zmd3dtnesPNpi3Npt/9nYpO2VPhgACABrRGTkCaEEELUon8yLTyzNY91py1ele+esZXwv75BZfeuvGi8Kp6+WAghhBDVkmS0cefabC76JcOrgHZJSx3f97cQtfELrwPaVVddVdNqCj8mPWlCCCGEDyUW2Hh/j5G5hwqxVD7sjIRwNW8MDuOCZjrmzVvm9XWaN2/OFVdcUYOaCn8nIU0IIYTwgdNFDt7bkctXXoazSJ2Sp/qGMLlzAEpFycizvXv3enWtRx99lF69ehEYGFiTKgs/JyFNCCGEqIGj+Xb+e0jL0owc7BUvFABAoFrBXQlB3JsQRJiu9Kijffv2eXXNoUOHVqeqooGRkCaEEEJUkdPpZO0pM58dLOSXRBMOZ+W/TtUKmNIlkKf6hJQJZ1UREhJS7WNFwyIhTQghhPCSzeFkyUkTM3YV8E9W5VNpACiASR0CeKRXMG1Dyv+1W1DgeZH0802YMMGrcqLhk5AmhBBCVKLA6mDuwUJm7TWS6sUktC7j2+h5ok8IXcM1lZZNTk6utExUVBQjRozw+vqiYZOQJoQQQpQjrcjO7P1GPj1QSK7FiwFnlPScXdvOwBO9Q2gf6v2v2fJC2q233kp+fj5BQUGMHDmSyMhIr88pGjYJaUIIIcR59uVYmbXXyIKjRV49qQmgUcK1bQ081CuYjqGV95ydy2w2s3btWo/7Ro0aJasKNFES0oQQQgjA7nDyR7KJj/YVsuaU2evjYvQKroy2cF/fGOIjAqp83ZSUFJ588kny8vLK7FMqlQQHB1f5nKJxkJAmhBCiSUsy2ph7sIjvjhaRXGj3+ri2wSru6BbExDYqMlKTaRZw9onNpKQkMjMz6dq1K3q9vsLzzJkzx2NAg5InOVUqldd1Eo2LhDQhhBBNjs3hZFWqmTkHClmebMLh3XAzAIbEapneI4hLWupRKhSYTKaSc9psLFy4kHnz5pUqf8cdd3DZZZd5PJfFYmHHjh3lXis3N9f7iolGR0KaEEKIJuNYvo0fjxcz91AhSUbve80AxrXWM717EINidR73z5o1i40bN5bZ/vHHH3PxxRej0Zwdp5aRkcGGDRv44osvKrxmr169qlRH0bhISBNCCNGo5Zod/Hi8mO+OFLElw7uFy10C1AomdQjgrm6BdKjgYQCj0cimTZvK3f/3338zZMgQiouLmTRpktfXv/jii6tUX9G4SEgTQgjR6BitDpacNLHoRDErkk1eP6Hp0jJQxX+6BHJb50DCvVgd4ODBgzid5d8zff311/nggw+45557vK6DWq1m+PDhXpcXjY+ENCGEEI2C0epgTaqZH48X83uSiSJbFQaaAUoFXN5azy0dAxndQodaqfD6WKu18tUHqhLQHn30UVmfU0hIE0II0XAVWB0sTzLx4/FiVqaYKfZmhfPzNDMomdw5kFs7BRIXWL0nKQsLC6t1nCcffPABLVq08Nn5RMMlIU0IIUSDklZkZ2WKicWJJlanmjBVbfw/ULLY+UVxOv7dJZBLWuqr1Gvmya5du2p0vMtTTz0lAU24SUgTQgjh16wOJzsyLfyZYub3JBM7vVzY3JN+URomdQjgmrYGIvS+mX/sxRdfJDU1tUbnePDBB+nbty9BQUE+qZNoHCSkCSGE8CtOp5ODeTY2nraw/rSZP1NMXq+b6UmnUDXj2xi4vr2BzmFVW66pMg8++CDHjh2r0Tlee+01unTp4qMaicZEQprwe06nE7MdTHYnFocTq6PkL2vrmc9tDicOJzgBpxMcgMN5dpsSUCkVKCkZGKxSKlApcH8oFSVf61QKDGoFBpUCjRIUiprd/hDCF06dOsU///xDdHQ0HTt2JDQ01GO5wsJC9u/fj0ajISEhAbW6bn+8FxcXo9frS33fZGZmYjabiY6ORqvVlnusw+nkQK6NTWlm1p+ysOG0mXRTFR/HPE/zACXXtwvgunYGekaWf+3qysjI4I477sDhqH49dTod8+bNq/P/K9FwyDtD1Bqn00mhzUmmyUG2yUGuxUGu2UGexUmexXHmw0mB1UG+xUG+1YnR6sRkd1JsO/Nx5vPq/w1dPUoFBKgU6NUK9CoFAWf+NahLPvSqkjDn+tpwpqxBpSBIoyBYoyBYqyREoyBYoyREW/KvxuGkgqf0RRPldDpRKBQUFBRw4sQJDAYD+/fv59NPP63xuVu0aMF//vMfEhISMBqNLFu2jF9//ZWwsDD69OnDsmXLyjyZqFarsdlshIaGcsMNNzBq1CgMBoPH8+/evZtnnnmm1LYnnniC1157rdSUFJMmTeLGG290v95j+XZWpZr4eVcSm7PBqg+p8WttH6Li6vgArmijp2ekBqUP/tByOp2kpKSwZcsWvvzyyxqf71wPPPCABDRRIUVubm6j/ZVhMplISkqiVatWla6dJsp3fjs6nCXBK7XQTkqhnVNFdk4XO0grsnO6yE6GyUGW2UFmsaNaT1o1diqFkxCNklCdkjCtklBtSYgLPfN5uE5JxJmPcJ2ScJ3CvS1Arai0h89ut1NQUMCOHTtYs2YNHTp04OqrryYwMLCOXuFZaWlpbNu2DZPJhF6vx2az0alTJ2JjYwkNDUWpLJl/yhUS1Gq1Vz2Y3n5vOxwOioqK/GKcT0pKCitWrECj0TBu3DhCQ0P5888/ee+99+q7arXOrtRQENaa/PB48iPakhvZEZvON/8ngfkpRJ3aSV9lJtH2XK6eMIG+ffui03leFeBchYWF6PX6ctfGtFgszJgxg82bN1e5Xs888wxz587l5MmT5Zb5/vvvJaSdQ35nlyUhTZRhdThJKbSTWGDjaL6do7kmjmcayVXoSSlyklpkx1qzOxGimnRKCNcp0DvMqExGNBYjOlsxelsRuSnHURblorEUorEY0ZhL9qscNvfxCoWizISb8fHxtG/fnoSEBPcvttatW9OyZUvMZjMZGRmYzWa0Wi0tW7Z0BysomWX9n3/+ITk5mYMHD2IymSgsLMRsNpOWllat16hUKt0TeObn52M0GunatSuFhYVkZmZitVqJiooiKiqK8ePHExYWVup4s9nM+vXry4QfvV6PRqOhc+fOjBkzho0bNxIQEIBSqWTo0KHYbDaOHj3K33//TatWrWjVqpX7Oi1atECv12OxWFAoSoLyuUv8nMvhcHDy5El+++03tm/fTlRUFCkpKRiNxmq1R0PjUCgpCm5OfngbjKGtyItoT2FIc1BUPiGsNxR2K+EZB4hM20NE+n4MRVkVlp88eTKXX345+/btIyUlBYVCwfr169m7d2+ZsgMHDiQgIIDNmze71+Osjk8//ZTIyEjeeecdVq1a5bHMXXfdxaWXXlrtazRG8ju7LL8Nabm5uXzyySesW7eOnJwcWrVqxcSJExk/frzX55D/8IrlmB3sy7FyMNfG4TwrR/JsHM63cdJor9Jiw8K/qazF7tCmtRjRmAvQmgvQmvPRmAvO7DsT7M4LdcJ74eHh2Gw2CgoK6rsqdcaiDaIoKBZjaEsKwlpTGBJHYXBznCrfDs4PzEshLPMgkWl7CM0+hspe/ac7a9vnn39OeHg4AEeOHOHhhx8uU+b6669n4sSJ5fbgNVXyO7ssv+xnLS4u5r777uPo0aNcd911xMfHs2LFCl5++WWysrKYMmVKfVexQTHZnOzOtrIvx8r+XCv7c2wczLVyuli6w5oCu8aAXWPAFBjtVfmSUOcKbYXuUKcxF5SEPHfQK/lX5fDfX5h1KScnp76rUCucgFUbSHFQbEkgC4k7E8bifDKOzJPAvGTCso4QlnmI0KwjaC2+myi2tgwfPpw77rij1K31Dh06MHDgQLZs2eLedtVVV3HzzTfXRxVFA+SXIW3BggUcPHiQF198kbFjxwIwYcIEpk+fzpw5cxg3bhyxsbH1XEv/VGB1sCvLyq4sKzuzLOzItHIk34a/Dg1TKyBEe3ZgfZCmZJC+wTVYX312gL5epUCnUqBWgkapQKss+Vx95olNBSUD/kue4lTgGtrkdJY8PWZ3cubDid3hxGSxkpSSyt/bt5ORmYVDqcGu0uBQaXGozv9ci0N5zucqTanPnUq//FaqlrOhLsqr8kqbuSTEnQl0GovxzNfGMrdeteYCVDYT8tysf3ECFn0oxQFRmAIiMAdEUHQmlBUHxmDTBtTq9QPzUwjNOkZE+l5Cso81iFB2rg8//BCj0ehxfNnjjz/OqlWrOH36NO3atWPIkCH1UEPRUPnlb5YlS5YQFRVV6n69Uqnklltu4a+//mLp0qXcdttt9Va/RYsWUVBQgN1ux+l0lruormv8j8ViwWw2Y7FY3J/b7SVTZGu12nKfmgJKnfv86zhQkKYOJ0UTQ7o6jGRtDNlqz4/n1wWlw4bBasRgLURvL8bgMBPoKCbQYSIQK4FKG2pLEYriApSWQrQ2E1qlE71OV+rxfFe7mUwmtFqt+wefxeHA5HDgcDiw2WxYrVbsdrv7EXjX/0VlH2azmaKiIvdgdR3Qsoav3aFQngluWuwqHTaNAZtaj12jP/OvAZtah11twKbRY1MbSspoAs586LFpAnCoKx/s7G8cah0mtc7rUKdw2FGfCW+uMKe2FKGxFqG2Fp0T6gpRW4tQW4tRW4ua3G3YqKgoLr74Yq666ir0ej1bt27l559/Zu/evSQkJDBq1Cjat29PfHx8mWNdT4suXrqcD+Z+i1UXhFUbjEUXTHDz1tj0IVj0YVgCwsl1aChQBWB21FF0dtgJLDhFeMZBQnKOE5Z5GK2lZLzea6+9hkKhID4+njfffJOtW7fWTZ2q4NFHHyU0NBSDwUBMTAxBQUGYTKZyxxwqlUpGjx5dx7UUjYXfhTSj0ciJEycYPnx4mae8EhISADwO+KwrWVlZfPHFF/VybZtaT15ke3IjO5Af3o6CsNY41L6f/8cTpc2MzpSLrjgHfVE2uuKcM1+f+TDlobYWlttDYjnz0VgpnQ6UNjPYzED1xyQ5FCqs2kCs2kBsrn81AWe3aQLdn5d8BGHTBuJUNpyxLU6lCqs+BKs+hKIqHKe0W0pCm6X4bHizmVBZi1HbilFbTahsJtS2YlQ285nPS/5VWU0l2+xmlM76uc0/ZMgQNm3ahEKhoEOHDhw+fBiA0aNHM3HiRKKjK74dPWDAALr16U+myUGWyUGmycF6k52fdxeQeebrzGI7mWaHu0yRrRtc8lL5J3UCCkomF6wlLQJUDIrV0lFrQnP6IF2C7ChtFmav/tFdZvLkyVx22WWlnsh86qmnKj13bm4uq1evZtu2bQAEBgZW60nMyvTp04du3boxcuRIoqK8+2NECF/wu5CWnp6O0+n0eDszKCiIwMDAGi+/URO+Wp/NG3aVlpyoTuRFticnuivG0BY+e0LKE40pn4DCdAIK0tAXZaAvysZQmIGhMAONtbjWriugXbt2TJw4kR49erB161ZmzJhRpoxOp6NTp06EaEOIjo4mKiqKQYPiOXToMPsTU9CGxaCLak5QbCuyzCVz02WYHGSY7GSf+SWeXmQl0+TE4mx4NxwdKi0WlRaLPqxG59HgQKtwYFCB02LCYSpEr1KgVzoJDdQTrNcSERKE0mGjMC+HrLRTFBvzMRgMFBYUUFRUhFKpxOGwg9OJWq3GbrPBmdn8IsIj0Ot1hIeH0yKuOVGRUe5b7xdcehcKhQK700nQmXkAj9qcPL3fiWlPlnt+QJPdSbG9ZDypa97AQpujWmtU1hUF0D5ETUKEmp4RWnpGaugbpSGy1NJLce7Pxo0bV+NrhoWFMWHCBCZMmFBmX1FREU8++SQnTpxwb4uMjOTee++lWbNmREdHY7PZUKlUHD58mLVr1xIfH0+XLl2Ii4sjJyfH/TtHiPridyHN1WUcEOB5DIROp6O4uOLA4Hp02mKxlPrXF3bs2OGzc53PoVSTeyaQ5UZ1xBjSwudPSQFoTXkEFJwisOA0gfmpBBacIqDglASxKhgyZAg9evSgoKCA4cOHU1hYMobmjz/+YPny5eUe16JFC3Q6HYGBgfTt25eRI0eWeYpp4MCBfPfdd17XpV+/vvTr1/e8raozH2U5nU4KrE4yTM6SMGd2km12kGUu+TrL5CTHUrI988ycd4WN6E6jFSVWp7LkNSmDIOCc+bpcXb75dkpiRwQERYCriBejCRJdn9iBJCCp8T3tGa1X0CVUTedQFd3CVLQLVpEQriJYc/4fkVZMpvp5sESpVPLaa6+5v3Y4HKWmj7HZSt7Udruddu3a0a5du1JlXSs7eDMVR238rmmKmko7VuXJVb8Lad449xvNk9TUVPeYL6Da8zWdz+l0snPnTp+cy6U4IIrM5r3IjulCXkR7345JcjowGNMJyT1JYH4KwbmJBOanNrhBuXVJqVTSv39/wsPDsVqtGAwGYmNj3T1XnrgCGsAll1zCJZdcgtPpJD8/H7vdTlhYWLnv2YyMjFp5Hd7QAs3PfKA98xHsuWyxHXKtCnKsCnKsJZ9nn/nIsSrIsyrIPme7qa7GN4lao1I4aa130sbgoLXBSasz/7YNcBB+/t+OZsg9Dbn1UVE/4qvfNU1dY25HlUpV6g+CyvhdSHP1oJX314vJZCIuLs7jPhfXfovFQlpaGrGxsRWuG+etzMzMas2BpFar0Wq16HQ6FDoDuRHtSQ7vQmpwPEZdWI3r5RJoyiEqP5FwYyoRhacILUpDc+70CEogLBCFovRM366xfwqFgqCgIPegY5vN5v5wDbJ3OBxoNBoMBgMajQaNRlPqgQadTofD4cBut7uPtdvtWK1WrFYrWq3W/QElf8W6Hqpw1cN1Pp1Oh9VqxeFw4HQ6USqV7g+VSoVGo0GlUqFUKt3HusKQ61/XxKPnfqjVagICAggODiY0NJSgoCDatm1Ly5Yta329Tl+/J/1Vkc1JzpneOXcvnclBtuXs19lneu2yzQ7yrE6ZILmOBWsUxOoVhCuttAzVER+sIT5ISZsgFS0DlTQzKFErJWx7o6l8X9c2acey/C6kNW/eHIVCQXp6epl9RqORoqIiYmJiKjzH+V2JWq3WJxPjtWzZkgULFmAymVCpVO430bkh59yg4QobuRYnv5408VtiMWtOmX0yriRUq6BflJYBMVoGRGsZFKslWNMC6F7zk59HJhj0PV+9J/2VHogIgvZelnet85pjLhnwnm0u+XCt75prLr32a47JRnaxlSKnigIJeAAEqRVE6pVEnfmI1KuIdn+uJEqvKtlnKNkWoFae870d06jfj3WlsX9f1xVpx7P8LqQFBgYSHx/Pvn37yuzbs2cPAD179qzrarmp1Wqv1gHMszhZnFjMouMlwaym85SFaBQMb65jRHMdA2O0Pls8WAh/oFCULEwfpFHSyoslHc//w8Fkc5JvdZBvcWC0loy5M1pLPje6Pred/bz4zGB8k81J0XmfW+xOzHYw20sG7dtqeY5BvQoMZ+YD1KvOzg2oP2ebQV3yEaE7G7jOD2AGtfw8EKKx8buQBnDZZZcxa9Ysli1b5p4rzeFwMG/ePLRaLZdcckk919CzIpuDJSdNfH+smBXJphr9cA/RKBjWXEffKC0Xt9BJKBOiAnq1Ar1aRYzB91OR2B0lQc11V98JOM88yel0up7pPLP9nDKc8/XZ/SUbFIozgUxFrd9iF0I0XH4Z0iZOnMjvv//Oiy++yMGDB2ndujXLly9n69at3H///X41T43F7mRFiolvDhexPMWEuZq3MlUKGBijZVgzHaNa6BgQrZXxIEL4AZVS4eE5WfneFELUPr8MaXq9no8++ohZs2axZMkSCgsLadOmDc8//7xP5tbxhb3ZVuYdKeS7I8Vkm6s3ICZSp+Sy1nouaKbjkpa68+YTEkIIIURT5pchDSA8PNyrGafrUpbJzk/Hi/nmSBHbM6s390+PCA2XtdZzaUs9vSM1qKS3TAghhBAe+G1I80fTN+Ty68nKJzY8X69IDRPiDVwVb6BdiDS5EEIIISoniaEKnu4Xwu9JJq+e1OwYqua6dgaubWugQ6jvVw0QQgghROMmIa0KuoRpuK1TIJ8d9Dxjf8tAFTe2N3BlGwO9IjXy1JYQQgghqk1CWhU90SeYhceKKLCWdKcFqBVc0VrPLZ0CuSBWK2PMhBBCCOETEtKqKNqg4uFewfx8oph/dQzk6rYGwnQVryUqhBBCCFFVEtKq4f7uQUzvUc5K1EIIIYQQPtDou4BUKt/PPdYUx5rVRjs2VdKWviHt6BvSjr4h7egb0o6lKXJzc2t5ZTohhBBCCFFVjb4nTQghhBCiIZKQJoQQQgjhhySkCSGEEEL4IQlpQgghhBB+SEKaEEIIIYQfkpAmhBBCCOGHJKQJIYQQQvghCWlCCCGEEH5IQpoQQgghhB+SkCaEEEII4YckpAkhhBBC+CEJaUIIIYQQfkhCmhBCCCGEH5KQJoQQQgjhhySkCSGEEEL4IQlpQgghhBB+SEKaEEIIIYQfkpAmhBBCCOGHJKQJIYQQQvghCWlCCCGEEH5IQpoQQgghhB+SkCaEEEII4YckpAkhhBBC+CEJaUIIIYQQfkhCmhBCCCGEH5KQJoQQQgjhhySkCSGEEEL4IQlpQgghhBB+SEKaEEIIIYQfkpAmhBBCCOGHJKQJIYQQQvghCWlCCCGEEH5IQpoQQgghhB9q1CHNZDJx7NgxTCZTfVelQZN29B1pS9+QdvQNaUffkHb0DWnHshp1SAOw2+31XYVGQdrRd6QtfUPa0TekHX1D2tE3pB1La/QhTQghhBCiMharub6rUIaENCGEEEI0aftPbuftHx5h/8nt9V2VUiSkCSGEEKJJcjgdLPt7Pt/8+Q5GUx7frnqXbYfW1He13NT1XQEhhBBCiLpmLM5nwZpZHD+9373N6XSyaONnGE35XNjjChQKRT3WUEKaEEIIIZqYY6f28f3ajykozvW4f8X279Fp9AzuOqZuK3YeCWlCCCGEaBLsDjsrtn/P+j1LKizXqWUv+nQYVke1Kp+ENCGEEEI0ernGLOav/oDkzKMVlhvZawIje0/A6XRSUFDg9bxter2ewMBAlErfDfeXkCaEEEKIRu1Iym4Wrv2IIrOx3DIGXSDXD7+Tji174nA4yMrKIigoiKioqErHpjmdTkwmE1lZWURGRvosqElIE0IIIUSj5HA6+HPHT6zZtbjCcnGR8UwaeR9hQVEAFBYWEhQUhMFg8Oo6CoXCXbawsJDg4OCaVfwMCWlCCCGEaHTyC7NZsPZDEtMOVVjugoSxXNz3etSqs5HIZDIRFRVV5Wvq9XoyMzMlpAkhhBBCeLLz6EYWbfwMm91abplAfQg3jLibds27etxfnek3fD1lh4Q0IYQQQjQKBUW5LNr4GYeSd1ZYLj62MzdedC9BhpA6qln1SEgTQgghRIO3L/FvFm/6gkJTQYXlhve4nIv7XOfTpzBri09C2pEjR5gzZw7bt2/HaDQSFRXFiBEjmDZtWqn7skePHuXDDz9k9+7dmM1munXrxh133EGvXr3KnHPdunV88cUXHDt2DLVazZAhQ7jnnnuIjY31RZWFEEII0QiYrSZ+3fwl/xzdUGE5gzaQ6y8seXqzoahxSEtMTGTq1Kmo1WquvfZamjVrxu7du1m4cCFbt27ls88+IyAggOPHjzNt2jR0Oh033HADgYGBLFy4kLvuuov333+fvn37us+5dOlSnnvuObp06cIdd9xBfn4+3377Ldu3b+fLL78kIiKiptUWQgghRAOXnHGUBWs+JMeYUWG59s0TuHb4NIIDwuqmYj5S45D25ptvYrPZ+OKLL2jbti0A11xzDV26dOGtt95iwYIFTJ48mZkzZ2KxWPjyyy9p0aIFAGPHjmXixIm8/vrrzJ8/H4Di4mJmzJhB+/btmT17NjqdDoABAwZwxx138Mknn/DYY4/VtNpCCCGEaKCcTifr9yxhxfbvcTgd5ZbTqnWM6Xc9g7pcXO/rcFZHjW7Ims1mduzYQe/evd0BzWXcuHEAbN++naysLDZt2sSFF17oDmgAYWFhjB8/nuPHj7Nnzx4A1q9fT25uLtdee607oAH06dOHnj17snTpUqzW8p/WEEIIIUTjlZV/mk+WvMQf2xZUGNBax3TgnvEvMbjrmAYZ0KCGPWkajYb58+fjcJRtpOzsbABUKhV79+4FoHv37mXKubbt3buX7t27u8Najx49ypRNSEhg165dnDhxgo4dO9ak6kIIIYRoQJxOJ1sPrmLp1m+x2i0Vlh3T93qG9RiHUuH/DwdUpEYhTalUluoZO9eXX34JQL9+/UhPTwfwOOg/JiYGgJSUFAB3Wdd2T2VTU1MlpAkhhBBNRLG5kB/Xf8KBpB0VlosOjePGi+4hNrxlja/pdDqr3APndDprfN1z1coUHL/++iu//PILsbGxXH311SxcuBDA4/IKer0ewL2AqdFYsq5WQEBAuWWLi4srvL7rXBaLpdS/onqkHX1H2tI3pB19Q9rRN6QdfaO8djx+ej8/b/6MQlN+hcf37XAhY/rcgFql8XpR9PIolUqKioq8XhbKpbi4GKVSWeH1XVnGGz4PaYsXL+bVV1/FYDDw+uuvExgYWGGydO3zJq26ylY2t0lqaip2u939dVpamjdVF5WQdvQdaUvfkHb0DWlH35B29A1XO1rtFrYcW8bR9IonptWpDQzucDltIrtwKvW0z+qh0WgIDw9Hq9V6tcC6xWIhJycHq9XqHvJ1PpVKRbt27byug09D2uzZs5kzZw6BgYHMmDGDbt26AWd7xTwlS9c213xq55bVarWlyprN5lJlyxMXFweUpPG0tDRiY2PLnEt4T9rRd6QtfUPa0TekHX1D2tE3zm3HlOyjLN5Wee9Zl1Z9ubTvRIIMoT6vj8PhoLi42OteObVaTUxMjE8nyfVJSLPZbLz88sssWbKE6Oho3n77bTp16uTe7wpNrvFm5zp/vNq5ZUNCSi/X4ErXnsarnev8rkStVlul7kXhmbSj70hb+oa0o29IO/qGtGPN2exWVu/5ia2H/qywnEat5crBt9Gnw7BarY+noVd1qcYhzW6389RTT7Fq1Srat2/PzJkzyzwg0K1bN5RKJfv27StzvOvJz549S2YATkhIcG/v0KFDmbJBQUFV6ioUQgghhP87knaUxds/pNhS8bJOLaLacv2FdxEZ0vhXIKpxn9xHH33EqlWrSEhIYPbs2R6f4IyMjGTAgAH8+eef7qc4AXJzc1m8eDEdO3akc+fOAAwbNoyQkBAWLFhQavDgjh072L17N5dffnmDne9ECCGEEKXlmh3ctz6HN/Y4cDjsFZYd2WsC08Y92yQCGtSwJy0lJYWvv/4ahULBRRddxPr168uUCQ8PZ/DgwTzwwANMnTqV22+/nUmTJqHValmwYAFGo5HXXnvNXV6v1/PAAw/w4osvcvvttzN+/Hiys7P55ptviIuLY8qUKTWpshBCCCFqyGw2s3v3boqLixk8eDAajaZa51l8opiHN+eSXuwAIrg8bDLdir/AZCkqVS46LI4bLryLZhGtfVD7hqNGIW3Tpk3upyg/+OADj2V69erF4MGD3cs8zZo1i88++wyFQkHXrl15/vnny0xce8UVV2AwGJg7dy5vv/02wcHBXHjhhdx9992ybqcQQghRj3JycnjhhRc4ceKEe9uYMWO49dZbK32wD8BqtfLPoWO8dVTL0qzSD1r8ltuS4ObTaJn1MWZrMQqFghE9x3NRr6tQKVW+fil+T5Gbm+vbmdf8iMlkIikpiVatWslgzhqQdvQdaUvfkHb0DWlH32hq7fjNN9+wYMGCMtsjIiJ44YUXaNWqVbnHZmdn8+8Pf2ZDxGCs+pByy93WKp3ORT9x5eBbaRXTodxyjV3DXi9BCCGEEHVq507P85ZlZ2fzyy+/lHtcaqGdKxYnszrukgoDmkYJrSJacfvlzzbpgAYS0oQQQghRBQcPHix33x9//OFx+zeHCxm8KI19zsgKz90xwMFvY0J5uG8UGlWtLIrUoEgLCCGEEMIr3qxNee6al6eL7Ny/IYc/ks0VHqNTweM9Arg8MJPWERJNXKQlhBBCiAbA6XRy9OhR0tLSSEhIICwsrM7rUFBQ8RxmUPJggd1u56dUeGm3lQJrxcEuNPMQ710YwyVdI0lK8lVNGwcJaUIIIYSfczgcfPDBB6xcuRIoWVfyySefpE+fPnVaj8TExErL3P/im2xoPoqcmK4VllPZTHTYvZBmJzfz2QYY+NFHvqpmoyFj0oQQQgg/d+DAAXdAg5JpLD7++GOvbj9WR3FxMb/++ivz5893T0J/4sQJnnnmmXKPcSjVHO9yOb/1uLPSgBaevo8BK1+i+cnNuKanf/vtt31V/UZDetKEEEIIPzdv3rwy206fPs2KFSvYvXs3Wq2WsWPHlllOsToKCwt56KGHOH36NAALFy7k/vvvZ8aMGeUekxfelgN9/0VxUMUrAShtZjruXkizk5s4f+2ggwcPkpmZWeEUHk2NhDQhhBDCz7nWuT7fuRPJr1u3jldffbXG61v/97//dQc0AJvNxnvvveexrE2l43i38aS0u6jS84an76fLjq/QmfLKLfP666/zxBNPMGjQoCrXuzGS251CCCGEH8vIyPCqnNlsZsWKFRWWKSws5JNPPuHxxx/nk08+wWg0lto/ZcoUj4HQarWW2ZYVm8CW0c9UGtBU1mK6bP+SnpverzCgubz66qts3LixzPa8vDzeeecdJkyYwIQJEyp9rY2B9KQJIYQQfmzr1q1el12yZAnTpk3zuM/pdPLUU0+5l3M6cOAAv/32G99//z1qtZply5aRk5NT6TWKAqM50uN6smMTKi0bnfw3Hfd8j9Zc+VOh53r33XcZOnRoqW2ffvopa9eudX/9/vvvU1BQwNVXX12lczckEtKEEEIIP2axWKpU/pFHHmHatGl07NiR5ORk3n//fY4ePeqxNwzguuuu4+233+bDDz+s8LwOpZrEjpdwsuMlOFUVL6iuL8yg8z/zCM88XKW6u5hMplLzrRUUFJQKaC5z585l/PjxqFSNc11Pud0phBBC+LEvvviiSuUPHz7MG2+8QXFxMW+++SYHDhwoN6C5vPPOOxXuz4ztztaRT5LY5fJKA1qrw38w8M+Xqx3QXEwmk/vzLVu2lFvu6NGjNbqOP5OeNCGEEMJPFRYWVuu4jIwM3nnnHfetzcqUV644IJLDPW/06tZmYF4yXXZ8RXBechVqWr7ExEQCAgKIi4vj008/LbdcUVGRT67njySkCSGEEH7Ebrdz9OhRjh49yuzZs6t9ns2bN1f7WIdCSVLHMZzodFmlPWcKu5W2B36l1ZGVKPB+3ja9Xl+qt+x8jz/+OAAtW7asMIjV1lxx/kBCmhBCCOEnbDYbb775Zo0CVk3lRHbkUK+JFAc3q7Ts0EAjzh9eQWfOr/J1Jk+ezLBhw7jnnnvIyyv/qc/k5Ip75qrb29gQyJg0IYQQwk9s27at3gKayRDOngH/YeewByoNaPrCDJ6KOMaSGzpzx6Rrqnytxx57jLFjxxIUFMTMmTOrWeMSb775pldrijZEEtKEEEIIP7Fs2bI6v6ZdpeVY1/H8dfELZMZVvBaowm6l7f7FDFz5EsNjS26Dnj9VRmXmzp3LkCFD3F+Hh4fTv3//qlf8HN99912NjvdXEtKEEEIIP7F9+/Y6vV568z5sGf0sJztdilNZ8TQW0SnbGLziOdocWobSaadnz54AREVFERMT49X1rrjiCkJDQ8tsf/DBB1Eqqx9Jfvvtt2of688kpAkhhBBNTH5Ya/4Zeh/7Bv4HsyG8wrL6wgx6bnyPhL8/c68YcP311xMQEACAUqlk6tSpqNVnh7l36tSJd955h+joaPe2+Ph4brjhBo/XCAgIIDIysqYvq9GRBweEEEKIJsKsC+FowtWktxpYaVmF3UrrI8tpfWgZKocNgLFjx3LdddcRFRVVquygQYN455132LFjB8HBwQwbNgyVSsW7777Ltm3bUKlU9O/fH42m/CdFDQZDzV5cIyQhTQghhPADtTmVhE2t52THMSS3H4VDpa20fFTqDi53HOS2G0ayf38ALVq0YOjQoWi15R/bokULWrRoUWqbwWBg2LBhXtVRQlpZEtKEEEIIP3D+Yue+YFdpKep/DftiB1OsqHi+M4DA/BQ67fyO0OxjXHHPPfTt25e+ffv6vF6e6PX6Gh1vt9sb3fJQEtKEEEIIP5Cbm+uzczlREH3VPfyp60a2ufIeOpXNRNt9i2lxfK17Qtrhw4f7rD7eOHdMW3VYLJZKe+OsViunT58mLi6uQQQ6CWlCCCGEH/AmpHXu3JmDBw+Wu98JZMV251jCBIpoDpUFNKeDuBPriT/wK1rL2UlhFy5cWOH4sdpQ09B09OhRunfvXu7+pUuX8tFHH7mv9cILL1RY3h/I051CCCGEH6ho1n2ADz/8kH79+nncN23aNIjvzT8jHmXP4LsoCm5e6fXiCxMZuPIlOu2aXyqg/fDDD3Ue0KDmY/Kefvrpcvd9+eWX7oAGJbdGX3zxRex2e42uWdskpAkhhBB+oKKQ9tlnn9G8eXMuvvhimjUrvRrAzfc+wnzDIFb3up28sDaVXichXM3Pl0ax/e7BBBSml9o3efLkersNWFtzxO3du5cff/yxzHaLxcK+fftq5Zq+Irc7hRBCCD9Q3iLin3/+OeHhJXOZRURE8MYbb/DXX3+RWWhhS0gv7jqpwOYsf6Fyl65hah7tHcyEeAMKhQKABQsWsGLFCnJycujSpUu5PXV1ITw8nMzMTJ+fd/Xq1eXu+/XXX+nRo4fPr+krEtKEEEIIPzBv3jyP24OCgkp9bQgK5kjcEN7YmU9OTuW3CFsGqnixfwhXtz0bzly0Wi3jxo2rfqV9qH///ixdutTn512+fHm5+4KCgvz6qVC53SmEEELUs1OnTnncrlQq3ePD7A4nXx4qpP8PaTyxJY+cSh4KCNcpeHlACFuuieGadgFlApq/GT16dJ1fc+XKlUydOpXdu3fX+bW9IT1pQgghRD2bO3eux+0OhwOn08lvJ028uC2fQ3m2Ss+lVcKd3YJ4tHcwQZqG0xcTFxdXL9fNzc3l1VdfZe7cufXywERFGs7/nhBCCNFIbd682eP23Ij2jF2SyS1/ZnsV0K5ta2DLNbG8OCC0QQU0KJkW44ILLvDpOdPS0rwqV1RUxLFjx3x6bV+QnjQhhBCiHq1Zs6bMNmNwc453HU9W856Qbqn0HP2jNbw6MIwBMZUv+eTPrrzySsLCwvjtt9/KLRMQEEDbtm3Zu3dvqe2elqyqaE658yUmJtK5c2fvK1sHJKQJIYQQ9einn35yf24MacHxLpeT1byXV8e2DVbxfP9QxrfR+/2YM2+oVCr+9a9/cfvttwMl85k9/fTT7N+/313m6quvJjk5ucyxnpaVMpkqf+rVJSwsrOoVrmUS0oQQQoh6dOLECYqCYjje5UoyWni3TmbLQBUP9wrmlo4BqJUNP5yVR6VS8fzzz/Pnn3+SmppKz549GTBgAO+8806Zsp4mw7VarV5fy2ar/HZyXZOQJoQQQtSTHacK2Nd3Mukt+4Gi8jFkzQOUPN47hFs6BqBqxOHsXDqdjssuu6zUNk+9hgUFBcyfPx+LxcLo0aOJi4urUkjbtWsXQ4cOrXF9fUlCmhBCCFGHrFYrizbt4n+7izlkaAOtBlR6TIhWwQM9grmjayCBDeyBgNqgVHpug2+//RYoWdrqzjvv5Pjx416fc+nSpdx5550+qZ+vSEgTQggh6sjejCKmfreFA4HtIaDysKVTwbSuQTzYM5hwnYQzF2/W+Tx3rc6GSkKaEEIIUcv+ybTw1q4Cfkk0QVDHSsurHVZu6xbGgz2DaRHon7Ph1yeLpfInXqvjvvvuY/DgwXTr1o0+ffrUyjWqQkKaEEIIUUv+ybTw2j8FLE3y7ilDhcNG3In1LL7/CtpHh9Zy7RquDh06sG7dOp+fNykpiaSkJACmTp3KlVde6fNrVIX0nQohhBA+tiXdzI3LM7nolwzvAprDTvMT6xm44gV6Hf1NAlolhg0bVuvX+Omnn3A4HLV+nYpIT5oQQgjhAw6nk99PFjNjVwFbM7x7qlBhtxJ3Yj2tD/+BzpwPwMR//7s2q9koREZG1uj4+++/n3fffbfCMtnZ2eTm5hIREVGja9WEhDQhhBCiBkx2Jz+dVvHD7jwO5tm9OkbhsNE8cROtDy1Fb8p1b9dqtVxxxRW1VNPGJTo6moyMjGod623Iq62xb96SkCaEEEJUQ2qhnW+PFPHJ/gJOF+uAygOaymYi7vha4o6vw1CcXWb/s88+W+70EqI0nU5X7WODgoK8KleVedZqg4Q0IYQQogoO5Fp5e1cB3x8rxl75TBAAGJROog/8QcujK9FaCj2WmT9/fo2CR1Oj0Wiqfay3IU160oQQQgg/53Q6+SPZzDu7C9iY5v0v7hAN/KdrEBdYj/DeT4vLLTdjxgwJaFXkbdA638iRI71ep1NCmhBCCOGnrA4nPxwr5t3dBezL9X5tR21xDi2PrqJ54kau7/csWVnF5Za94ooraNeunS+q26QEBARU67gbbrgBnU5H69atOXnyZIVl6zuk+fTG9549exgyZAhbtmwps+/06dM899xzjBs3juHDh/Pvf/+btWvXejzPrl27uOeeexg9ejSjRo3ioYce4tixY76sqhBCCFGufIuDWXuN9Pk+jTvX5Xgd0AzGNDrv+JrBK56n9dGVaGzFfPDBBxQXew5pl19+OVOmTPFl1ZuM6vSkXXDBBTRv3hyA//73v5WWr+8xaT4LaSdPnuTRRx/Fbi87cDIzM5M77riDtWvXcuWVVzJ9+nRsNhsPP/wwS5cuLVV227Zt3HXXXaSlpTF58mRuu+029u3bx9SpUzlx4oSvqiuEEEKUcarIzgt/59Fj4Wme3JJHcmHlDwMoFTAs3M4LLdIYuPJFmp/chNJxNtQlJyeXu4bk+PHjUalkRYHqyM/Pr/Ix1113nfvzoKAgFi1aVGH5+u5J88ntzlWrVvHKK6+U22Bz5szh1KlTfPLJJ/Tq1Qso6d6dMmUKM2bMYMSIERgMBpxOJ2+88QYhISF8+umnhIaWTOY3cuRIbr75ZmbOnMnMmTN9UWUhhBB1xOl0kpycTFJSEh07diQ6Orq+q1SK0+lkY5qFuYcK+el4MVYv5y81qBRM6hDAXZ01kJ3K8uWbUZRT9siRIx63h4eHV6/Sosrvox49etC2bdsy2+Pj48vtBGrwIe3//u//2LBhA23btmXw4MH88ccfpfbb7XaWLl1K9+7d3QENQK/Xc+ONN/Lf//6X9evXM2bMGPbt28fx48e59dZb3QENoHXr1lx00UUsX76czMxMoqKialptIYQQdcDpdDJ37txSPRZt2rQhIiKCW2+91eMvzdp04sQJHnjgAQAcChWXPjaDBekGdmZ5f1srLkDJ3QlB3NIxkDCdEpPJRFI2ZX7/nevQoUNltoWFhaHVaqv8GkSJUaNGsWTJEq/LP/300x63X3bZZXz44Yce9zX4250nTpzg7rvv5quvvqJ169Zl9h87doyioiK6d+9eZl9CQgJQMpbt3H89le3evTsOh4N9+/bVtMpCCCHqyNatW8vcUkpMTGTHjh383//9H4sXl//Eo6+lpKTwwAMPYNUEkNjxUjaPeZGn9qu9DmidQ9V8MCyMf65rxr3dgwnTnf0Vmp1dds6zyvhbj2JD07ZtW69XA3j66afLfXr24osvZtCgQaW2XX311UyePJmOHTvWuJ41UeOetPnz51f4l0B6ejoAsbGxZfbFxMQAkJqaCkBaWprXZYUQQvi/Tz75pML9n332GZdddlmN5rxycTqdfPnll/z0009AyV2YBx54gHbt2uF0Onn5myXs73sbGc1741B734M1JFbLvQlBXNZaj1Lh+YZmVXp0XKo7hYQooVKpePbZZ909o+V55JFH6N+/f4XneeKJJzAajTgcDkJCQnxc0+qrcUirrKvWaDQCYDAYyuzT6/UAmEwli88WFpZM8OfpsVpX2fKekDmX63yue8n1fU+5oZN29B1pS9+QdvSN2m7Hw4cPe7Vsz759++jcuXONrzdz5kw2b97s/vrkyZNMf/gxLnvyXT49bOJo1Divz6UAxrXUcndXPf2iSgKkxWz2WPbYsWPs3LmzyvW1Wq3u31eieu/HZs2a8d1337FixQpWrFjhcWxZv379vGpntbokEtX2/4krz3ij1udJczrLn47ZtU9x5i8Tb8p6s1xGampqqadMXT10omakHX1H2tI3pB19ozba8bfffmP16tVelT1w4EC157xyKSoqKhXQjMFxpLS9kPSW/Vi32wF413MWqnZyZayNa5vZaGkogmJISvJctri4mGeffbbadR44cCBJ5Z28CavO+7Fz58507tyZ1NRUFi9eTGJiIj179mTChAl+1cYqlapKc+LVekgLDAwEPCdT17bg4GDgbA9aRWW96R6Oi4sDStJ4WloasbGxMjizBqQdfUfa0jekHX2jNtoxJSWFhx56qErHHDp0iKuuuqra13Q6nUyaNAm7SkNms56kxg8nL6pqY4mCCtO5v3c4zi0/E2UJIr7jKPd8Wp44HA5uuummatcZSsZau35fCd+8H1u1alVmfFlDVushzfUGdI1NO9f549XOLXt+13dFY9vOd35XolarrVL3ovBM2tF3pC19Q9rRN3zVjocOHeLRRx+t8nHbtm3DZDJ5vVSPi8PhYPny5bz59U+k9rie060GYdeUHVpTkfD0/bQ8+icR6fvZuOLs3ZxffvmFyMhIPv30UwBOnTpFYWEhaWlpbN++nb1791bpOp60atXKJ2PxGhv5vj6r1kNamzZtCAoK8viGdm3r0aMHcPapzr179zJ8+PBSZffs2YNCofD45KcQQoj65XA4ePzxx6t9/OTJk1m0aBHZ2dkkJibStm3bCkObxe5k+mdLWJIbTN6oZ6p0LbXFSLOTf9Hs5GaCCsp/GC0rK4sJEyZU6dxVIQFNVKbWQ5parWbMmDEsWrSIXbt20bNnT6Dk9uWCBQuIiIhg6NChAHTr1o02bdqwePFiJk2a5J4r7eTJk6xZs4bhw4dX+S8tIYQQtevQoUN88sknOBxezgJbjnMDkVKpdC8PeK5tGRYWHC1i/tEictV9oArTZgbmJdPi2Bpik7eictTv/Fcff/xxvV5fNAx1ssD67bffztq1a5k+fTo33XQT4eHhLF68mGPHjvHyyy+XmrvkkUceYfr06UydOpXrr78es9nMt99+i06n47777quL6gohhPDSnDlz+PXXX70q++ijj5Kens4XX3xRaVmHw8GsWbMYMmQIeeiYuyOVLw8aSVVUbXoEpc1CbPIWmp/cRHDOiXJXBKhLr732mldDd4Sok5AWFRXFnDlz+OCDD5g/fz42m4327dszY8YMLrjgglJlBw4cyHvvvcfs2bN5//33MRgM9OrVi7vvvps2bdrURXWFEEJ44YsvvvA6oD333HP06dOHhQsXelXeptKRHpvA1T+fYEtxME6FEqoQ0AzGNOKOr6PZyc1obKWnbpozZw7/+c9/vD5Xdeh0OszlTNkRGRlZq9cWjYdPQ9q0adOYNm2ax30tWrTwasV5gP79+1c48ZwQQoj6tW/fvkoXp3YJCAigd+/eABUuJu5QqsmJ6szp1oPIatYDh0oLJvC2+0tpsxCTuo1P/z2GzO3b+HzlKo/lQkNDufrqq92T3vrKY489Rnp6OpGRkfTo0YPbbrvNYzlP84YK4Umd9KQJIYRoeKxWK2q12j2XJZRMd7Fr1y6ee+45r87RuXNn7r//fvc5hg4dypdffnn2fEBuZEfSWg0gI65vlZ/OBAgoOE3zE+tplvQXzzx0P4NbhUKr8Xz++Wcey2s0GiZMmMDvv//us4lLp0+fTp8+fbyaeF1CmvCWhDQhhBClpKenl7orolAomD9/Pvn5+VW6TfjOO++UGabSrFkzevbsyeajp0hv0Z/TrQdjCqzC6P8zVNZiok7vonniRkKzjqLASUJCAoMHD3aXGTp0KBs3bix13LhxJasOhIaG8tVXX7Fu3TrefffdKl//XO+++26ZW5sVzfNVUW+iEOeSkCaEEMLNbDbzzDOlp7RwOp3ccMMNVTpPhw4dygS0XLODxYnFbBl4D1va2qpVv5CsozQ/uYmYlG2o7KWXDzp/Qtyrr76a7du3u3vLAgMDGT9+vHu/RqNh1KhRjBo1qtRxO3bs4IUXXii1bcaMGbRo0YJt27ah1WqJjo5GoVDQqlUrzGZzmVntVSoVnTt35uDBg6W2n/+0qhAVkZAmhBDC7aOPPvLJMlH/+9//ACiwOvgt0cTPJ4pZmWLCUo1ZOvTGDGJStxN3Yh364pxyy50/lrljx4688847/PnnnyiVSi699FKvpnHq06cPixYtIj8/n8LCQmJjY91LErqmjPLG008/zWOPPUZqaslcbP369ZNZCkSVSEgTQggBlEwavmqV58H23oiOjmbgwIH8a/K/+SPZzIKjRfyeZKLIVv66zOXRmnKJTdpCTMo2gvKSK3124K677vK4tnNsbCyTJk2q8vUBQkJCCAmp2pQf5woODmbWrFmYTCZ0Ol2psX1CeENCmhBCNHF5eXl8+umnLF++vNrnePPtd0jRNePXk8X0+D6dDFPVu8w05gKiU7YTm/I3IdnHUVB5uOvYsSNXXXVVlXq46poscSSqS0KaEEI0YQUFBTzyyCPVOtahUJIb1YmWV0xl9EYNeZbMKp9D4bATkb6PZic3EXl6D0qnvdJjmjVrxqRJk7jwwguld0o0ahLShBCiCfv555+rVN6hVJMd05WMuL7kteyDSaFhVzbgRa/XuXpFauhrT+bw/JmoTfleHdO5c2def/31Kl1HiIZMQpoQQjRRNpuNnTt3VlrubDDrQ2azntWaywygS5iaK9oYuL6dgc5hGiCGY72eZ+XKlfz222+VHv/YY49V67pCNFQS0oQQoon6448/yt1n1QSQ2awnmc17kRvVsdrBLC5AyfXtAriufQA9IjRl9rdr14527doxceJEvv32W5YsWVLuuSIiIqpVByEaKglpQgjRBBmNxlIz/wMUBUaTFdudrGbdyYvsiFNZvUlXw3UKxrYyMLF9AMOba1F6MW4sODiYadOmccEFF/DUU0+VPWd4eLXqIkRDJiFNCCGaoJ9++gmHQkl+RDuyo7uQ2bw3RSHNq32+KL2Sq+INXN3WwOAYLWpl9Qb0t2vXzuP2AQMGVLtuQjRUEtKEEKIJSSyw8XuSiXdPx5Jx6X+x6oKrfa4WASquaqvnkpZ6hjXTVTuYnctgMHDppZeybNmyUtsHDhxY43ML0dBISBNCiEbMYnfy58kCPl61k615aoxhrUt2RCVU63yuYHZVGwMDY7S1MgXG5MmTycvL46+//kKtVjNx4kT69evn8+sI4e8kpAkhRCNzusjO8mQTy5NN/JlixmhzgqIdhFXvfN3C1FzaSs/YVvpaC2bnMhgMPP744xQVFaHX6z2uJCBEUyAhTQghGjiL3cn2TAsrks0sTzGxM8tao/MpFdAnUsOVbQxc2cZA+9D6+VUREBBQL9cVwl9ISBNCiAbG6XRyNN/G8mQzq1JNrDtlodhe9fUxz6Vx2hjTJogrWuu5rLWBcJ30XglR3ySkCSFEA5BebGdruoXv96axJs1GNtWbt8zN6SAoL4mQ7BNEn97JnCfuomO7SN9UVgjhExLShBDCD+UVW/hszS7+PGXhVGBrjhS6xoFpznxUndpSREj2UaJO7SIybQ86c8lyTP/+97/p2C7eJ/UWQviOhDQhhPADBVYHf6VZ2JRmZsHuUyTZA0B5Zt6ywuqfNzA/hYi0vUSe3kNo9jEU56yxef3119OhQwd69epVw9oLIWqDhDQhhKgHuWYHG06b2ZJuYVWqmb05Vs4OKwuGag4JU9rMhGccJCJ9L5Fp+9AXZ5faP3bsWC666CI6d+6M2WwmKSmpRq9DCFF7JKQJIUQdSDLa2JFpZWOamU1pFvZknxvKasDpIDj3JBFp+wjLOkRI9nFUDpvHoj/88AMqVfWWehJC1D0JaUII4WMmm5P9uVb+ybTyV7qZrRkWjubbfXZ+rSmXiPT9JcEs8yBaS+X3Q9VqtQQ0IRoYCWlCCOGlgoIC/v77b2w2G4MGDUKtVmMwGDhRYGdnVkkv2Y5MCzuzrFgcvruuxlxAWOYhwtMPEJJznMCC06XGlnkjMDDQdxUSQtQJCWlCCOGFqVOnkpWVhV2lxRjSgrxlh8gPb4OpeVeMCr1Pr1UyruwAYVlHCM06SnDuySqHsvM9+eSTPqqdEKKuSEgTQojzOJ1OMjMzUev0fLVsPd+s/Yf8FhdTkNCagtBWoPTtbUOVtZjwzIOEZh0lLPMQgfmpKJ0lXXHR0dFknBPQ4uLieO211/jmm29YunRpheft378/gYGBTJ06lZCQEJ/WWQhR+ySkCSGavOzsbH5b8jvfrdhATkAz8iLaYgxtTUFYaxzqntC3p0+vpzYbCcs6TGj2ccLT95W5fdm3b1+mT59OaGhouee48847ufPOO7Hb7WRlZREdHU1xcTGHDx/GYDDQsWPHWl9jUwhRuySkCSGaBKfTydatW1m5ciWbtmylKKgZxtAWFITHYwxpgTF0IPYRF9XKtQ0FpwnJTSQ84yAhOccxGDPKvX35+uuv07lzZ6/PrVKpiImJAUrWupQ5z4RoPCSkCSH8ktPpJC8vjz179rB+/XocDgctW7bkxIkTKJVKWrduTcuWLSs9T6HVwbb0Yp587wsy9TEYQ0dSePkknKrqzdpfGY0pn5CcE4TkHCc49yRBecloLcYKj0lISOChhx4iIiKiVuokhGiYJKQJIfzO4cOHeeSRR8ps37Jli/vzv//+G5PJxH/+8x+USiV2h5NEo52dWRb2ZtvYn2tlb46VxAJ7SZ9V5wk+r6fCYSMoL4XgnBOEZR0hKC8JQ2EGFd1kVCgUvPjii7Rs2ZK0tDSCg4Np0aKFz+smhGj4JKQJIfzKqVOnPAY0F4dSjSkggsLgOD48bOfNl3/GEtWWwoAorM7aHYOlK8omKC+JVpZ0nEl7Cck9gcpurfS40aNHc/HFFxMSElIqkIWHh9dmdYUQDZyENCGE38jNzeWuu+4CwKbWUxQUS2Fwc4qCm1EU3IzCoGaYAiNB4WHNJF/M3n8OjSmf4LwkQrLP3LbMT6Znm2a8+OKL6HQ6Pv/8c37++XCF51Cr1TzwwAMMGzbMt5UTQjQJEtKEEHXG6XRiMpkwGAzur1ONVpbvPMLeLBNL/t5P0ZB7KQxujsUQVmf10pryCMpLJjA/he4hTsb37cS1Fw1Eq+1a7jFTpkyhf//+fPnll6SnpxMbG8vVV19N9+7dCQoKkicrhRA1JiFNCFGrnE4nx9NzePi1d0mx6SkOiqEoKIaioFhMgVE4VFogtOSjXWyt18dgTCcoL4k4jFw/tAe/fzoTdWEWo0eN4pZpt1Rp8H6PHj343//+V4u1FUI0ZRLShBA1ZrY7STLaSDTaOVFgI7HAzrF8G8fyS7YV2pzQ8/Y6rZPSZiaw4BRBeSkEFJw6c8syBbXNxPXXX8/NN98MwL2jPq3TegkhhLckpAkhKmV1OEkttJNSWBLCjhfYSTLaSCm0cyzPQmqxE2eFzzTWHoXDjsGYVhLGCtMJzE8lKC8JfVG2x7nI2rZty/jx4+uhpkIIUTUS0oSoI1arFaezJDTk5uYSERGBVqut51qBxe4ktcjOqSI7p4vspBY5SCm0kVpY8m9KoZ1TRY5KxuXXfkAL0yroHKYh3uBg65KFGAozCcxPQV+U5V5CqSKDBw9m6NChDBgwwD0mTggh/JmENCFqycaNG1mzZg1//fVXheXuvvtuLrnkEoxGI/n5+QQFBflsncV8i4NT7gDm4HRRSW9YcqHdHcwyiisLYHVLW5xLgDGNwIJT3Hn1GHo2C6VLmJoYg9I9GN84cDLfffcdv/76T4XniomJ4dFHH6VDhw51UHMhhPAtCWlC+Mi8efNYuHBhlY+bNWsWs2bNKrP9lVdeISEhocx2i91JltlBWpGdlAIL3/z6B/tOnsKiC6FF1z4oQmM4XWTndJEVs9O3C4H7ispajKEwnYCCdAyFGeiLswgoOE1gwWnUNhM333wz119/c7nHq9VqRowYQWZmJps3by61r23btrz99tu1/RKEEKLWSUgTogLFxcUYjUaysrKIi4sr1cNVVFTEzz//zPz586t9fidgV+uwagKxaQOxagOxaoOw6oK46atNWHR7sOqCsOhCCIlrS2axDZPivOWMggZBt5JPUyxAhmty1foNaGpLIYbCTAyF6eiLstEVZxNgTCOgIA2tOb/UDdI2bdrQqX8noCOXXHIJHTt29Ooad999N4GBgaxatQqVSsWYMWP4z3/+UyuvRwgh6pqENNGomM1mdu7cidFoZODAgQQFBeF0OsnJycFqtRITE4NCocBsNuNwOLDb7QQEBKBUlkyOmpqayrp167Barfz1118kJSVVek0n4FBqsGkDsGkM2DQBWDUB2DQBZ7YFYD2z3bXfpg0oCWPaQJxK774N803A+QGtHintFnTFOeiKczEUZpz5yERflIW+KBONtdjjcR9++CFFRUUsX74cgEsuuYR27dpVqw5arZb77ruPO++8E5vNJmPNhBCNioS0Jshut5OTk0NQUBBFRUWkpqYSERFBXFxcvdaroKCAnTt38vXXX3P69OlS++Li4rjyyiu59NJLUSqV5Ofns2XLFt5///0qXcMJOJVq7CoNDpUWh1Jz5vOSr+0qrftzm7oN9g6dsan12NV67Bo9NrUem8aA/cy/NrUBm8ZQa4t11ye1xYiuOBddcS764hx0xTnoi7PRFedgKMxAa8rz+nGBVq1aMXDgQK666ip3b2T79u19VleNRoNG0/j+D4QQTZuEtAbMai25rXXuLyen08nhw4f53//+R0ZGBgBXX301I0eOJC4ujhdffJFdu3ZVeN5JkyZx6aWXEhYWVqX62O12lEolFouFJUuW8Pfff7N37173/pCQEPr06cOaNWvO1pczoUnpCkpnwlJY61IhKk2hYdvvu3jmjwMlZc6Ut3e/tqScSlsSts45z7mB69zze1xSqInRmvLcvWA6U+45/+agNeWjM+V4tSbluXr16sXzzz8vM+0LIYSPKHJzc/3pwS6fMplMJCUl0apVK/R6vU/O+dWhQhYcLUKpUKBSgFLBmX8VpT537VOe+zVOHHY7RmMBDocDm9VKUXExZrMZm9VKQUEBVquVyMhINBoNgQEBREVHE2AwYDKZSEtLIynpJPn5BaXq5GkuqDKcnsp42KZQ4nR/qHAqFOgCAgkJDSMjKxunQonNebacLiAQvSGA7Lz8c447c6xSiRNlyb8K5TnnPrvPoVJLaPIhjbkArSkPrTkfrSkfrTkfnSkPXXFuyXZTHjpzPkqHrcrn1uv1mEwmoOQ249ixYzEYDERFRdG7d2+io6N9/XLKVRvf202RtKNvSDv6hrRjWdKTVkXHC2ysO22p4Vl0Z/41ACElU0xpgciSrUddxYqARAdQeGZDFERFQVQNL19dzVqVvy+ivirVuCkcdjSWArSmgpJ/zUY05gI0lsJSQUxjLkBrMVYrfHmjU6dOvP7669JLJoQQdchvQ1pubi6ffPIJ69atIycnh1atWjFx4sR6nSncbreTkZnF2ZAlhPdCNApCtAoM2IgO1BJlUBOhUxKhVxKuU5Z8rlMSqVcSqoHsxCPs2rKB35ctKXWeESNGcOGFF9K7d29UqrNPcB45coRt27bRqlUrBg0aVGqf0+lk165dPPfcc6XOdeedd9K7d2/uvPPOcus9fvx4pkyZIgFNCCHqmF/e7iwuLmbatGkcPXqU6667jvj4eFasWMHff//NXXfdxZQpU7w6j6+7Ts1mMwNems/J9qNrfC7RsCiAII2CYI2CQJUTncNKTGgAYToVYVoloVoFoVolIVolYVoFYTolYVrlmX9L9qmUigbTne90Ov06lDWUdvR30o6+Ie3oG9KOZfllT9qCBQs4ePAgL774ImPHjgVgwoQJTJ8+nTlz5jBu3DhiY2PrvF46nY6Q0LA6v66omE7pRKcEvRKCdGr0agUGlcL9r+Gcr/VKMKiV6NUKAtQKivNzCQ3QERkShN5V9kz5klCmJFirIFCtQOnHocXX/DmgCSFEU+GXIW3JkiVERUVx6aWXurcplUpuueUW/vrrL5YuXcptt91WL3WLjwjkeF4yKBQlg+BRnBkQX/I1nPnXtV+hAM7Zf94vP68WpS7zC7PsMed3h7p+yZbXTapQKD3vczpQOOwonE4UTjsKp6P0h8MOFew7/7jY6CiiIsJJP52KEict4+KICA1Bq1Zht1lxWM1EBBoI1mvc4en8gKU/91+VAr2KM2FKiV5V00ARXINjhRBCiNrjdyHNaDRy4sQJhg8fXuaXr2uJnHOndahrd3bWYXzz1WodGx0djUajITw8nAEDBhATE4NaraZNmzbYbDa+//57kpOTGTRoEN27dyc1NZUjR46QlJREYWEhKpWKESNGEBUVRVxcHC1atHCPO3I4HOTn55c8FRoYWObaaWlppKam0qJFC2JiYiqsZ3FxsXt6j88++4zDhw8TFxdHu3bt6NevH1qtFofDgcPhQKlUolKp0Ol0hIWFodfrcTqdWCwWdLpzx+61rVabCSGEEE2V34W09PR0nE6nx9uZQUFBBAYGkpqaWg81K9G5c2eP2wMCAgAwGAzceOONDBs2DIPBUKVenunTp5f6ukuXLowaNcqrY5VKZYXzmsXGxnp9i9hgMLhnbn/ggQeqPE5AoVCcF9CEEEIIUVV+F9KMRiNwNvScT6fTUVzsebkZF9dcThaLpdS/vhAUFMQLL7xA8+bNsdlsBAQElBtczGazz65bn2qjHZsqaUvfkHb0DWlH35B29I2m0o5VeSjC70KaN1zrLJbn9OnTOBwOAFQqFZmZmT69fnBwsDtMugJhY1cb7dhUSVv6hrSjb0g7+oa0o2809nZUKpXEx8d7Xd7vQpqrB6288GMymSpdY7IqDSCEEEII4Y/8bj2e5s2bo1AoSE9PL7PPaDRSVFRU6cB3IYQQQoiGzu9CWmBgIPHx8ezbt6/Mvj179gDQs2fPuq6WEEIIIUSd8ruQBnDZZZeRlpbGsmXL3NscDgfz5s1Dq9VyySWX1GPthBBCCCFqn18uC2UymbjttttITk7mxhtvpHXr1ixfvpytW7dy//33c8stt9R3FYUQQgghapVfhjSAnJwcZs2axbp16ygsLKRNmzbcdNNNjBs3rr6rJoQQQghR6/w2pAkhhBBCNGV+OSZNCCGEEKKpa3AhzemUjj8hhPBEfj76hrSj8BcNKqSZTKZSy0XIN1L1FBYWkpSURHZ2trsNpS2rLi8vj0WLFrlnx5Y2rB6j0cihQ4dIT093rxQibVl1xcXFFBYWYrPZ6rsqDZrZbC71/nO9J0XVFBUVkZmZSWFhYX1XpUHzuxUHyvPpp5+yatUqFAoFQUFB3H777XTr1g29Xo/T6azSQuZN2SeffMKvv/6KyWTCbrczevRopkyZQrNmzeq7ag2KzWbjzTff5I8//uCll15izJgx8h6shjlz5vDTTz9hNpux2+1ccMEF3HvvvfJ+rKI5c+awfPlyHA4HGo2G22+/nYEDBxIYGCg/H6vgs88+Y9OmTahUKqKiopg6dSqtWrVCqVRKO1bBnDlzWLZsGSaTCafTyfXXX88VV1xBZGSktGMV+X1PWkZGBvfddx8LFiyga9eutGjRglOnTvF///d/zJ49G0D+w71w4sQJbr/9dhYsWMDw4cOZMGECffr0YdGiRTz33HPY7fb6rmKDolarOX36NACrVq3i5MmT9VyjhsVkMvH888/z/fffM27cOKZMmcKQIUPYvHkza9asqe/qNRiJiYn85z//4bvvvqNHjx60b9+eoqIinn32WX744QdAfj56IyUlhWnTpjF//nxiY2NxOp1s3ryZ++67jx9//BGQdvTG4cOHmTx5MvPnz6dv374MHz6c5s2bM2vWLN544w1A2rGq/L4n7a+//mLv3r1Mnz6dMWPGEBAQQGZmJg899BDz5s0jPj6e8ePH13c1/d6SJUtISkri4Ycf5oILLiAoKAiHw8Err7zCr7/+yooVK7j00ktxOByVLmDf1NntdqxWK2azmfDwcFavXk2vXr1o3rw5Wq22vqvXIBw4cIA///yTW265hZtuuomgoCCcTid79uxx96LJX9yVW7FiBUePHuWhhx5i1KhRBAQEkJiYyOTJk/n999+59NJLiY2Nre9q+r3Vq1eTmJjIww8/zLBhwzAYDBw9epTp06fz9ttvExcXx7Bhw+Q9WYkff/yRzMxMnnzySQYMGEBQUBDFxcX83//9H6tXr2bLli0MHDhQ2rEK/Pq3scViYf369YSEhHDVVVcREBCAzWYjKiqKBx98kPbt2zNr1iwOHjxY31X1azk5OSxZsoSOHTty6aWXEhQUhN1uR6lUctFFFwG4ey8koFVOpVKhVqtJTU1l3LhxJCQk8NNPP3HkyJH6rlqDsW3bNsxmszugQclf2F27diU0NNT9tShfbm4uS5YsITIykiuuuML987FNmzaMGjWKrKwsAgMD67uafs9oNLJy5UqaNWvGmDFjMBgM2Gw22rdvz2OPPUZgYCBvvPEGBQUF8p6swKlTp/jll1/o168fI0eOJCgoCJvNhsFg4MorrwRg8+bNgHxvV4Xf/UY+d5CmVqslMzOTgIAA8vLycDgcqFQqAHr16sX1119PYWEh33zzDSaTqb6q7JfOb0eFQuH+5Wez2dztGBcXB0B4eDggA7bPV96g4cTERAwGAz179uTSSy8lMTGRFStWUFBQAEg7nu/8djx69CghISHk5eW5t33yySfcdNNN3Hnnnbz88svuH+jirHPbMSQkBK1Wi1Kp5MSJE0DJH1kmk4mTJ0/Sp08f9Hq9u7y8Jz0LCAigsLCQiIgIoPTPx+HDh3PdddeRlpbG559/Dkg7lkehUKDT6YiOjgbAarW629H1+yU4OBiQNqwKvwhp+/fv56WXXqKoqMg9QNP1n9i5c2dOnTqFxWIps2/48OGMGDGC5cuXe1yQvak5vx1dP9CNRiOBgYHk5ubicDhQq8/e5U5KSgJw/zCXv3A8vx/Pp9frycrKIiYmhlGjRjFgwACWLFnC7t27AWlH8NyOrvdkp06d3IOK09PTueWWW1i4cCEtWrRApVLxyy+/uG+RWK3Wen4l9au872u73c7w4cNJTEzk559/JjExEZPJ5H4fnjx5kunTpzN79mwyMzOb/Hvyr7/+4rXXXuPtt9/m+++/d//sKyoqIi4ujn379mG1Wt0/H13tfPnll5OQkMB3331HWloaCoWiSYeM89vRNR43Pz8fm83G8ePHAdBoNO5j0tPTAUr1mgvv+MWYtN9//51ff/2VLl26cP311wNn/xNbt25NUVERCxcu5O6778bpdLpvyUVFRTF69Gg2bNjgHqjYlO91e2pHgNjYWF5++WUiIyPdbedqp7///huA3r17l9relJXXjuc6efIkarWaqKgoQkNDGT9+PK+88gp//PEHxcXF7Nixg3/9619NejyQp3Z0vf+aN2+OxWJh6dKldO7cGaVSydtvv03btm0JCAhg6dKlfPLJJ8yePZuWLVvSoUOH+nwp9aq896NGo+Haa68lMTGRn376iWXLlrnvPlx88cXodDqOHz/Op59+ytatW3n88cdp3759Pb6S+pGRkcF///tftmzZQmxsLLm5uRQWFtKsWTPeffdd2rRpQ7t27di4cSM//vgjN954Y6nfM61atWLcuHHs3buXzz//nMcff7yeX1H9KK8dY2Njee+99+jUqRMffPABLVq0KHPs9u3bgZJOF1E19dqT5vprxJWyf/75Z1JSUlAoFO65foYPH06LFi348ccfSUtLQ6VSYbfb3cf279+f3r17s379epKSkprkXznltaNSqXS3Y4cOHdxdzlASgh0OBzt27CAmJoZ27drVfcX9TEXvx/Pn7zKZTKjVamw2G2q1mhEjRjBs2DD++OMPnnzySY4dO1bqL8mmxJvv6969e9OmTRvmzZvHK6+8wkUXXURCQgIGgwGAiy66iGuuuYajR4+yZ88eoOnNV+XN93VsbCyvvPIKs2fPpnfv3litVp577jkee+wxnnnmGT755BOmTZvG/v37Wb16danzNgVGo5G3336bI0eO8NRTT/H222/z559/Mm3aNLKzs5k1axYA48aNQ6PR8Ouvv5KTk+PusXS950aPHk18fDwbNmwgPT29yf0hW1E75uTk8OGHH2KxWOjevXuZ3zNGo5GdO3cSHx9Pq1atgKb1Hqyper/dmZKSwj///ENwcDCHDx92P+6sVqtxOBzExcVx2WWXUVBQwKeffgqUDNx2hbHg4GD69+9fqpu1qX0DQcXtWJ7Dhw9z5MgR+vTp4/7rpym23bnKa8fzH6hITEwkICCAuLg4bDYbX331FatXr0ahUBAfH89dd93lHuPSFFX2fR0bG8uYMWMwmUyYTKZS7z+n04ler6dHjx5otVr32LSm+FBLZd/XTqcTtVqNVqtl5cqVJCQkMGrUKPfYH5VKxfDhwwkODmbDhg1A0/oeP3ToEGvXruXKK69k3LhxtGnTBoCrrrqK3r17s3r1alJSUmjfvj1jx47l0KFDpb7nXWEtPDycYcOGkZubS05OTn2+pHpRWTuuWrWK7OxsoGwA2717N6dPn+aCCy5wB7im9B6sqXr9qadQKFi5ciWFhYU8+eSTBAcHs3jxYve4HtfcXZMmTaJ9+/b8/PPP7ttzdrvdvb9Lly4A7r/Cmxpv29HF9U20YsUKAPcTnlAy2/auXbvYsmVL3VTej3jTjq4fLhaLhWbNmrFo0SKuu+465s+fz6233sp1111HYmIiu3btwmw2A03vr8aqfF+3adOG4uJicnNz3ce79nfr1g2Hw0FMTEydvwZ/UJX3Y0ZGBlDyQJVKpcJkMpUa/+cKGq73ZFNx5MgRbDZbqeEcTqeTqKgo90NTrtkB7r33XgICAli4cKF7jLPNZnO3Y9u2bbHZbE0yYFSlHV3t43A4cDqdrFq1CoBhw4a5z+f6PeNq56b2M7Iqaj2k5eXlcfLkyVJ/fbje9KmpqSxYsIB+/foxatQoHnjgAfLz85k/fz5QMubCZrMRFBTE1KlTCQgI4NVXXyUjI8M9DYLT6WT16tVotVrCw8Mb7X92TdpRpVKVaheFQkFhYSEbN24kLCyMnj174nQ6SUxMZP78+UyfPp033nijUS7nUdN2dAWInJwcdu/ezcyZM+nXrx+zZs1i2rRpjB8/nm7durFgwQL3bbrG+EPdl9/XOp2Ob775hgMHDgAlvUQWi4VffvkFm81Gjx496v4F1pGavh9dZV3hYeXKlUDJgy1KpRKz2cyiRYvIzc1l6NCh6HS6On6FdcNTOwLu3wlGoxGg1NCFgoICNBqN+xZcWFgY99xzDzk5Obz55pvuXkpXr+Vff/1FSEiIexWHxqgm7egKay5KpZLc3Fy2bt1Ks2bN6Nq1K3a7nZMnTzJ//nzuu+8+XnjhBff5hGe1+uDArFmz+OWXX9ype+zYsdx2222lbgMNGzaM4cOHA3DFFVfw3XffsXz5ckaOHMno0aPd5UaPHk1KSgqff/45DzzwALfccgutW7fm+PHjbNy4kUsuuaTRDi72RTue/0DA6dOnOXLkCKNGjUKr1bJ06VK++eYbDh8+zE033cT9999f56+ztvny/dirVy/3slq9evVyPx3bpk0bxowZw7vvvltqndnGxJftOGbMGBITE5k3bx6PPPIIN954I23btiUjI4Nvv/2Wvn37MmDAgDp/jXXBF+3omnx6yJAhDBkyhI0bN/Luu+9y+eWXY7fbOXToEF999RW9e/cu1ZPRmHhqx1tvvZXIyEiaNWvGlClTSj0w4fo5eOLECcLCwtzzeanVaq677jqOHz/O999/z4MPPsgNN9xAVFQUBw4cYNeuXYwbN87jwPjGwBfteP7vmcOHD5OamsrVV18NwB9//NHof8/4miI3N9fnfxKYTCZefvllduzYwfjx4wkKCmLXrl2sXr2a7t278+STT7r/s41GY6n/3O3bt3PXXXfRu3dv3nrrrVLfQCaTiU2bNjFjxgzS09MJDg6mqKiICy+8kMcff5ywsDBfv5R65et2PHc1gb///pt77rmHQYMGodVqWbduHUOHDuXxxx9vdE8k+rodXed0OBwYDIZS3fuuvx6VSiUhISH19pprQ219XxcVFfHXX3/x/vvvk5ycjF6vR6FQMGDAAJ544olGN7bP1+1otVrRaDQcOnSIjz76iA0bNqBUKomKiiI7O5shQ4bw5JNPNql2TEhI4NlnnyU+Pt7jsceOHWPSpEkMHDiQ9957Dyi5faxSqcjOzmb58uV8/PHH7vnT8vLyGDlyJI899liT+r6uTjvC2VkCli1bxrPPPsvYsWMpLCxs1L9nakuthLRdu3Zxzz338K9//Yubb76ZwMBA7HY733//Pe+//z6DBw/m7rvvpm3btu7/TFf3sUKh4PHHH2fVqlXcf//93HzzzWXSeXJyMklJSaSnp9OxY0e6devm65fgF2qzHZcsWeLuao6Pj+ehhx5i4MCB9fZaa1Ntvx+bitpux4yMDNLS0igoKCA6OrrR9ozXZjvabDbWrFlDVlYWhYWFDBw4kISEhPp8ubXGm3a85557iI+PL7Pc3ezZs/n000957LHHuOaaa9wB7Vz79u0jKSmJjIwMevfuTffu3ev6JdaJ2mzHuXPnup+gbdOmDY888kij7RmvLbUyJu3gwYNYLBYmTZpEYGCgewbnyy+/nEmTJrF27VqWLFniHvjq+iHjusf94IMPAiXrgJ08eRKFQlFq8HvLli0ZMmQIV111VaMNaFA77eh6dL99+/bodDqeeuop5s+f32gDGtROOza16SCg9r+vo6Oj6d69O0OGDGm0AQ1qrx1dY6hGjx7NDTfcwJQpUxptQAPv2vG3335zL4HnCro2m43NmzcTFhbmHu/oGrebnZ1NUVERUPLgyqWXXsott9zSaAMa1E47ZmZmAiUPrRgMBp5++mkWLFggAa0aahzSVqxYwZIlS1i3bp37B25WVhZQMlM2nH10PigoiJtuuon27dvz+++/u5/UdHENzI6JieHf//43ycnJ7sehz/8rp7Gpq3ZUq9XY7XY6d+7M6tWrG93i9HXVjo19Ogj5vvaNumzHc3slG9vAdl+2Y3JyMnv37qVHjx507NgRKFkHdeXKlTz00EP8+eefjfaPsLpqx0ceeYRff/2VPn36sHLlSvfanaLqqv2bZs2aNVx11VW8+uqrvPrqqzz88MPce++97N69m0GDBgElgwZtNlup9B0WFsbUqVPJyMhg7dq1FBUVlfrh4nqD3HHHHcTExPDtt9826ukg6qMdXb8YG1PQkPejb0g7+kZ9t2NjuR1fG+3oChsXX3wxDoeD7du387///Y9nn30Wh8NB//79G9XPRqj7drTb7QwYMAC9Xt/o/xCrbdV6Jy5evJiXXnqJHj168PTTT/PWW29x7733sn37dt58802USiWdOnVi/fr17gnuzv2hccEFFzBgwAA2bNjgXj/N5dxbclOnTiUiIqLUDMaNibSjb0g7+oa0o29IO/pGbbXjtm3bgJIB85999hkPPvggu3fv5rXXXmPu3Lk0a9asbl9oLauPdvzyyy/lwQAfqXJIMxqNLFiwgObNm3P77bczcuRIBg4cyL/+9S8mTJjAgQMH2LdvH8OGDWPHjh1s27atzJI6Op2OsWPHcurUKVJSUoDSS7645qWZMGECv//+u7srtTGRdvQNaUffkHb0DWlH36itdoSz7Tdjxgy++OILJk+ezOLFi7nwwgvr/oXWMmnHhq/KIa2wsJDDhw8zfvx42rRpg8PhcP9l16tXLwACAgIYPnw4sbGxfPvttyQnJwNn07lCoaB169bo9Xp3GvfUvdzYxlWcS9rRN6QdfUPa0TekHX2jttrR4XBw8uRJoGTeuWXLljF58uQ6fnV1R9qx4atySDt9+jSAe3ZwpVLpvufseiomPz+fzp07M2HCBA4dOsTPP//snsHY9Qbp1KkTVqu1wrl7Gsu4Ck+kHX1D2tE3pB19Q9rRN2qjHZ1OJ0qlknv+v717D635f+A4/tzFmQ2zzZaG2drkOpOM1JBbLqEURaNI7sP+0SiSJJeIUCh/+YOj/TOSGmVmyx+WnaWdJttkMztzv2zDnDn2+2Ptk+Hni/PGubwe5Y9zzud8PufzOrVezvv9+bxzcjh37hx5eXn06tXrb5/aX6Uc/d8vl7T09HRGjRpFenq6tQ5c1x+LrqtEhg0bRlhYGDNnzmT69Onk5+dz5coVoPMn0vb2dgoLC/F4PAH5U/3PUI5mKEczlKMZytGMP5Fj1/snTJgQ0Ld4+ZJy9H+/vCxUWFgYJ0+etJbB6eJ2u3E4HERGRpKamgpAUlISubm5PHr0iBMnTtDQ0EBGRgYfPnwgPz+fzMxMa8HWYKMczVCOZihHM5SjGcrRDOXo/7xecaDrDsROp5P169czfvx4Dh8+TEhIiPWvvr6eCxcucPHiRXr06EFYWBhZWVnk5eUF3BIbv0s5mqEczVCOZihHM5SjGcrR/3i9wHrXhNby8nLa29vJysrqdl+Ujx8/kpSUxLZt28jOzqa1tZWoqCiSk5O9PXRAUY5mKEczlKMZytEM5WiGcvQ/Xpc06Pzp9Pr169a4NnQ29sbGRkpKSnj79i0bN24kKSnJxOEClnI0QzmaoRzNUI5mKEczlKN/MVLSampqaGhoICsri5iYGJ49e0ZFRQXnzp3j/v37rF69GiBoF6b+WcrRDOVohnI0QzmaoRzNUI7+xauS1vUlNjQ08P79ewYMGIDD4cBut1NSUsLEiRMpKChgwIABQGBfMu4N5WiGcjRDOZqhHM1QjmYoR//kVUnr+hIfPnwIQFVVFZcvXyY2Npbjx49ba4LJjylHM5SjGcrRDOVohnI0Qzn6JyPDnQkJCQDU1dWxdu1ali5damK3QUc5mqEczVCOZihHM5SjGcrRv3h9Cw6AO3fu4HQ6yc7OxmazmfhcQUk5mqEczVCOZihHM5SjGcrRvxgpaZpgaIZyNEM5mqEczVCOZihHM5Sjf/nlZaG+R1+4GcrRDOVohnI0QzmaoRzNUI7+xUhJExERERGzVNJEREREfJBKmoiIiIgPUkkTERER8UEqaSIiIiI+SCVNRERExAeppImIeOnz58//+iOISAAysiyUiMi/Vl5ezoYNG755PiQkBJvNRu/evUlJSWHKlCksXLiQyMhII8e9ceMGBQUFHD9+3Mj+RES6qKSJSMCZM2dOt8dut5vnz59TWVlJeXk5drudo0ePkpaW5tVxKioq2LZtG4MGDfJqPyIi36OSJiIBZ8+ePd99/sWLF+zfv5/S0lJycnI4e/Ys/fv3/+3jaJhTRP4kzUkTkaARHx/PgQMHyMjI4NWrV5w8efJffyQRkf/LyALrIiL/2pdz0srKyn647b1791ixYgWhoaEUFhYSExMDQFtbGxcvXqS4uJgHDx7Q2tpKZGQkKSkpzJo1i8WLFxMe3jkAsX79ehwOR7f9JiYmcunSJetxS0sLdrudoqIiGhsbCQ8PJzU1lQULFrBgwQLCwsIMJiAigUbDnSISdEaMGEFiYiJNTU3cvn2b2bNn8+7dO9atW0d1dTV9+vRh9OjRRERE0NDQgNPpxOl0UlVVZQ2lTpgwgfDwcMrKyoiKimLKlClW2QN4/PgxmzZtwuVyERcXR2ZmJp8/f6aiooJ9+/Zx8+ZNDh06ZJU+EZGv6a+DiASl1NRUmpqaqKurA+Ds2bNUV1czYsQITp06RVRUlLXttWvX2LlzJ1evXiU3N5d+/fqxatUqxowZQ1lZGXFxcd3mwXk8HvLy8nC5XCxatIjc3Fx69uwJdM6L27p1K7du3eL06dNs2rTpr563iPgPzUkTkaDUp08fAF6/fg1AVFQUkyZNYvPmzd0KGsCsWbPo27cvHR0duFyu/9x3SUkJtbW1DBkyhK1bt1oFDTrnxe3Zs4eQkBDy8/N59+6dwbMSkUCiX9JEJCi53W4AQkM7/6+6cuXKb7b5+PEjjx49oqqqyrqSs729/T/33TUnbvz48d+ddzZ48GCSk5Opq6ujsrKSiRMn/u5piEgAU0kTkaDU3NwMQN++fa3nnj9/TkFBAQ6Hg/r6el69ekVHR+e1VSEhIQDW4x958uQJAHa7Hbvd/lPbioh8TSVNRIJOR0cH1dXVAAwdOhSA4uJidu7cidvtJjo6mpEjR5KSkkJaWhpjx44lNzeXxsbGn9q/x+MBYPTo0QwcOPCH23pznzYRCWwqaSISdBwOB83NzURERDBu3Dja2trYu3cvbreb5cuXk5OT880wZUtLy0/vPyEhAYAZM2aQnZ1t9LOLSPDQhQMiElQ8Hg+nT58GYPbs2URHR1NbW2sNf65Zs+abguZ0Oq3Xv1xloGsI9GuZmZkA3Lx587uvNzc3s2TJEtasWUN9fb13JyQiAUslTUSCxosXL9ixYwd3794lISGBjRs3AhAbG2ttU1xc3O09NTU17Nq1y3rcdcEBQEREBACtra3d5qrNnDmTgQMHUlFRwZEjR2hra7Ne+/DhA7t37+bhw4e0tLQwePBgo+coIoFDKw6ISED4csWBLxdY7+jooK2tjadPn1JbW8unT59ISkri0KFDpKamWttt376doqIiADIyMoiPj+fJkyfcu3cPm81Gv379cLlc7Nq1i/nz5wPw5s0b5s6di8fjIT09nUGDBln3S6upqWHLli28fPmS2NhYhg8fTmhoKJWVlTQ3NxMfH8+pU6dITk7+WxGJiJ9RSRORgPBlSfuazWYjJiaGtLQ0pk6dyrx587DZbN22aW9vJz8/nytXruByufj06RP9+/cnMzOTZcuWUVpayrFjx5g2bRoHDx603ldYWMiZM2doamoiOjqaCxcuWCsPvH79mvPnz1NaWmotC5WYmMjkyZNZsmQJcXFxfywPEfF/KmkiIiIiPkhz0kRERER8kEqaiIiIiA9SSRMRERHxQSppIiIiIj5IJU1ERETEB6mkiYiIiPgglTQRERERH6SSJiIiIuKDVNJEREREfJBKmoiIiIgPUkkTERER8UEqaSIiIiI+SCVNRERExAf9DypJ4cMMVYfZAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = df_ta.copy()\n",
    "\n",
    "# Create trend features\n",
    "dp = DeterministicProcess(\n",
    "    index=y.index,  # dates from the training data\n",
    "    constant=True,  # the intercept\n",
    "    order=3,        # quadratic trend\n",
    "    drop=True,      # drop terms to avoid collinearity\n",
    ")\n",
    "print(dp)\n",
    "X = dp.in_sample()  # features for the training data\n",
    "print(X)\n",
    "# Test on the years 2016-2020. It will be easier for us later if we\n",
    "# split the date index instead of the dataframe directly.\n",
    "idx_train, idx_test = train_test_split(\n",
    "    y.index, test_size=200, shuffle=False,\n",
    ")\n",
    "X_train, X_test = X.loc[idx_train, :], X.loc[idx_test, :]\n",
    "y_train, y_test = y.loc[idx_train], y.loc[idx_test]\n",
    "\n",
    "# Fit trend model\n",
    "model = LinearRegression(fit_intercept=False)\n",
    "model.fit(X_train, y_train)\n",
    "\n",
    "# Make predictions\n",
    "y_fit = pd.DataFrame(\n",
    "    model.predict(X_train),\n",
    "    index=y_train.index,\n",
    "    columns=y_train.columns,\n",
    ")\n",
    "y_pred = pd.DataFrame(\n",
    "    model.predict(X_test),\n",
    "    index=y_test.index,\n",
    "    columns=y_test.columns,\n",
    ")\n",
    "\n",
    "# Plot\n",
    "axs = y_train.plot(color='0.25', subplots=True, sharex=True)\n",
    "axs = y_test.plot(color='0.25', subplots=True, sharex=True, ax=axs)\n",
    "axs = y_fit.plot(color='C0', subplots=True, sharex=True, ax=axs)\n",
    "axs = y_pred.plot(color='C3', subplots=True, sharex=True, ax=axs)\n",
    "for ax in axs: ax.legend([])\n",
    "_ = plt.suptitle(\"Trends\")\n",
    "print(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Learn from the residual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 367,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Close_price</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th>Companies</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">2010-06-29</th>\n",
       "      <th>TESLA</th>\n",
       "      <td>1.592667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AMAZON</th>\n",
       "      <td>5.430500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">2010-06-30</th>\n",
       "      <th>TESLA</th>\n",
       "      <td>1.588667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AMAZON</th>\n",
       "      <td>5.463000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-07-01</th>\n",
       "      <th>TESLA</th>\n",
       "      <td>1.464000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Close_price\n",
       "Date       Companies             \n",
       "2010-06-29 TESLA         1.592667\n",
       "           AMAZON        5.430500\n",
       "2010-06-30 TESLA         1.588667\n",
       "           AMAZON        5.463000\n",
       "2010-07-01 TESLA         1.464000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "Date        Companies\n",
       "2010-06-29  TESLA          1.592667\n",
       "            AMAZON         5.430500\n",
       "2010-06-30  TESLA          1.588667\n",
       "            AMAZON         5.463000\n",
       "2010-07-01  TESLA          1.464000\n",
       "                            ...    \n",
       "2022-10-26  AMAZON       115.660004\n",
       "2022-10-27  TESLA        225.089996\n",
       "            AMAZON       110.959999\n",
       "2022-10-28  TESLA        228.520004\n",
       "            AMAZON       103.410004\n",
       "Name: Close_price, Length: 6214, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# The `stack` method converts column labels to row labels, pivoting from wide format to long\n",
    "X = df_ta.stack()  # pivot dataset wide to long\n",
    "display(X.head())\n",
    "y = X.pop('Close_price')  # grab target series\n",
    "display(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 368,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            Companies  Days\n",
      "Date                       \n",
      "2010-06-29          0   180\n",
      "2010-06-29          1   180\n",
      "2010-06-30          0   181\n",
      "2010-06-30          1   181\n",
      "2010-07-01          0   182\n",
      "...               ...   ...\n",
      "2022-01-10          1    10\n",
      "2022-01-11          0    11\n",
      "2022-01-11          1    11\n",
      "2022-01-12          0    12\n",
      "2022-01-12          1    12\n",
      "\n",
      "[5814 rows x 2 columns]\n",
      "Date        Companies\n",
      "2010-06-29  TESLA          1.592667\n",
      "            AMAZON         5.430500\n",
      "2010-06-30  TESLA          1.588667\n",
      "            AMAZON         5.463000\n",
      "2010-07-01  TESLA          1.464000\n",
      "                            ...    \n",
      "2022-01-10  AMAZON       161.485992\n",
      "2022-01-11  TESLA        354.799988\n",
      "            AMAZON       165.362000\n",
      "2022-01-12  TESLA        368.739990\n",
      "            AMAZON       165.207001\n",
      "Name: Close_price, Length: 5814, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Turn row labels into categorical feature columns with a label encoding\n",
    "X = X.reset_index('Companies')\n",
    "# print(X)\n",
    "# Label encoding for 'Industries' feature\n",
    "for colname in X.select_dtypes([\"object\", \"category\"]):\n",
    "    X[colname], _ = X[colname].factorize()\n",
    "\n",
    "# Label encoding for annual seasonality\n",
    "X[\"Days\"] = X.index.day_of_year # values are 1, 2, ..., 12\n",
    "# print(X)\n",
    "# Create splits\n",
    "X_train, X_test = X.loc[idx_train, :], X.loc[idx_test, :]\n",
    "y_train, y_test = y.loc[idx_train], y.loc[idx_test]\n",
    "\n",
    "print(X_train)\n",
    "print(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 369,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date        Companies\n",
      "2010-06-29  TESLA        -39.035961\n",
      "            AMAZON         9.738623\n",
      "2010-06-30  TESLA        -38.811373\n",
      "            AMAZON         9.735532\n",
      "2010-07-01  TESLA        -38.587264\n",
      "                            ...    \n",
      "2022-01-10  AMAZON       190.087642\n",
      "2022-01-11  TESLA        304.666422\n",
      "            AMAZON       190.268135\n",
      "2022-01-12  TESLA        305.267215\n",
      "            AMAZON       190.448747\n",
      "Name: Close_price, Length: 5814, dtype: float64\n",
      "Date        Companies\n",
      "2010-06-29  TESLA          1.592667\n",
      "            AMAZON         5.430500\n",
      "2010-06-30  TESLA          1.588667\n",
      "            AMAZON         5.463000\n",
      "2010-07-01  TESLA          1.464000\n",
      "                            ...    \n",
      "2022-01-10  AMAZON       161.485992\n",
      "2022-01-11  TESLA        354.799988\n",
      "            AMAZON       165.362000\n",
      "2022-01-12  TESLA        368.739990\n",
      "            AMAZON       165.207001\n",
      "Name: Close_price, Length: 5814, dtype: float64\n",
      "Date        Companies\n",
      "2010-06-29  TESLA        40.628628\n",
      "            AMAZON       -4.308123\n",
      "2010-06-30  TESLA        40.400040\n",
      "            AMAZON       -4.272533\n",
      "2010-07-01  TESLA        40.051264\n",
      "                           ...    \n",
      "2022-01-10  AMAZON      -28.601649\n",
      "2022-01-11  TESLA        50.133566\n",
      "            AMAZON      -24.906136\n",
      "2022-01-12  TESLA        63.472775\n",
      "            AMAZON      -25.241746\n",
      "Name: Close_price, Length: 5814, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Pivot wide to long (stack) and convert DataFrame to Series (squeeze)\n",
    "y_fit = y_fit.stack().squeeze()    # trend from training set\n",
    "y_pred = y_pred.stack().squeeze()  # trend from test set\n",
    "print(y_fit)\n",
    "print(y_train)\n",
    "# Create residuals (the collection of detrended series) from the training set\n",
    "y_resid = y_train - y_fit\n",
    "print(y_resid)\n",
    "# Train XGBoost on the residuals\n",
    "xgb = XGBRegressor()\n",
    "xgb.fit(X_train, y_resid)\n",
    "\n",
    "# Add the predicted residuals onto the predicted trends\n",
    "y_fit_boosted = xgb.predict(X_train) + y_fit\n",
    "y_pred_boosted = xgb.predict(X_test) + y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 370,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/y9/x007ynb92t57yt7s517ct53w0000gn/T/ipykernel_80750/2752424835.py:5: UserWarning: When passing multiple axes, sharex and sharey are ignored. These settings must be specified when creating axes.\n",
      "  axs = y_test.unstack(['Companies']).plot(\n",
      "/var/folders/y9/x007ynb92t57yt7s517ct53w0000gn/T/ipykernel_80750/2752424835.py:8: UserWarning: When passing multiple axes, sharex and sharey are ignored. These settings must be specified when creating axes.\n",
      "  axs = y_fit_boosted.unstack(['Companies']).plot(\n",
      "/var/folders/y9/x007ynb92t57yt7s517ct53w0000gn/T/ipykernel_80750/2752424835.py:11: UserWarning: When passing multiple axes, sharex and sharey are ignored. These settings must be specified when creating axes.\n",
      "  axs = y_pred_boosted.unstack(['Companies']).plot(\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "axs = y_train.unstack(['Companies']).plot(\n",
    "    color='0.25', figsize=(11, 5), subplots=True, sharex=True,\n",
    "    title=['TESLA', 'AMAZON'],\n",
    ")\n",
    "axs = y_test.unstack(['Companies']).plot(\n",
    "    color='0.25', subplots=True, sharex=True, ax=axs,\n",
    ")\n",
    "axs = y_fit_boosted.unstack(['Companies']).plot(\n",
    "    color='C0', subplots=True, sharex=True, ax=axs,\n",
    ")\n",
    "axs = y_pred_boosted.unstack(['Companies']).plot(\n",
    "    color='C3', subplots=True, sharex=True, ax=axs,\n",
    ")\n",
    "for ax in axs: ax.legend([])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Add more features to the Gradient boosting (ensemble of weak predictions from decision trees)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.8 64-bit",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.8"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}