ml-finance-python

python scripts for finance machine learning
git clone https://9o.is/git/ml-finance-python.git
alpha_factor_zipline_with_trades.ipynb

(90761B)
      1 {
      2  "cells": [
      3   {
      4    "cell_type": "markdown",
      5    "metadata": {},
      6    "source": [
      7     "# zipline MeanReversion Backtest"
      8    ]
      9   },
     10   {
     11    "cell_type": "markdown",
     12    "metadata": {},
     13    "source": [
     14     "In the chapter 04, we introduced `zipline` to simulate the computation of alpha factors from trailing cross-sectional market, fundamental, and alternative data. \n",
     15     "\n",
     16     "Now we will exploit the alpha factors to derive and act on buy and sell signals using the custom MeanReversion factor developed in the last chapter."
     17    ]
     18   },
     19   {
     20    "cell_type": "markdown",
     21    "metadata": {},
     22    "source": [
     23     "Run the following from the command line to create a `conda` environment with `zipline` and `pyfolio`: \n",
     24     "```\n",
     25     "conda env create -f environment.yml\n",
     26     "```\n",
     27     "This assumes you have miniconda3 installed."
     28    ]
     29   },
     30   {
     31    "cell_type": "markdown",
     32    "metadata": {},
     33    "source": [
     34     "## Imports"
     35    ]
     36   },
     37   {
     38    "cell_type": "code",
     39    "execution_count": 1,
     40    "metadata": {},
     41    "outputs": [],
     42    "source": [
     43     "import sys\n",
     44     "from pathlib import Path\n",
     45     "import pandas as pd\n",
     46     "from pytz import UTC\n",
     47     "from zipline import run_algorithm\n",
     48     "from zipline.api import (attach_pipeline, date_rules, time_rules, order_target_percent,\n",
     49     "                         pipeline_output, record, schedule_function, get_open_orders, calendars,\n",
     50     "                         set_commission, set_slippage)\n",
     51     "from zipline.finance import commission, slippage\n",
     52     "from zipline.pipeline import Pipeline, CustomFactor\n",
     53     "from zipline.pipeline.factors import Returns, AverageDollarVolume\n",
     54     "import logbook\n",
     55     "import matplotlib.pyplot as plt\n",
     56     "import seaborn\n",
     57     "\n",
     58     "from pyfolio.utils import extract_rets_pos_txn_from_zipline"
     59    ]
     60   },
     61   {
     62    "cell_type": "code",
     63    "execution_count": 2,
     64    "metadata": {},
     65    "outputs": [],
     66    "source": [
     67     "sns.set_style('darkgrid')"
     68    ]
     69   },
     70   {
     71    "cell_type": "markdown",
     72    "metadata": {},
     73    "source": [
     74     "## Logging Setup"
     75    ]
     76   },
     77   {
     78    "cell_type": "code",
     79    "execution_count": 3,
     80    "metadata": {},
     81    "outputs": [],
     82    "source": [
     83     "# setup stdout logging\n",
     84     "zipline_logging = logbook.NestedSetup([\n",
     85     "    logbook.NullHandler(level=logbook.DEBUG),\n",
     86     "    logbook.StreamHandler(sys.stdout, level=logbook.INFO),\n",
     87     "    logbook.StreamHandler(sys.stderr, level=logbook.ERROR),\n",
     88     "])\n",
     89     "zipline_logging.push_application()"
     90    ]
     91   },
     92   {
     93    "cell_type": "markdown",
     94    "metadata": {},
     95    "source": [
     96     "## Algo Settings"
     97    ]
     98   },
     99   {
    100    "cell_type": "code",
    101    "execution_count": 4,
    102    "metadata": {},
    103    "outputs": [],
    104    "source": [
    105     "# Settings\n",
    106     "MONTH = 21\n",
    107     "YEAR = 12 * MONTH\n",
    108     "N_LONGS = 200\n",
    109     "N_SHORTS = 0\n",
    110     "VOL_SCREEN = 1000"
    111    ]
    112   },
    113   {
    114    "cell_type": "code",
    115    "execution_count": 5,
    116    "metadata": {},
    117    "outputs": [],
    118    "source": [
    119     "start = pd.Timestamp('2010-01-01', tz=UTC)\n",
    120     "end = pd.Timestamp('2018-01-01', tz=UTC)\n",
    121     "capital_base = 1e7"
    122    ]
    123   },
    124   {
    125    "cell_type": "markdown",
    126    "metadata": {},
    127    "source": [
    128     "## Mean Reversion Factor"
    129    ]
    130   },
    131   {
    132    "cell_type": "code",
    133    "execution_count": 6,
    134    "metadata": {},
    135    "outputs": [],
    136    "source": [
    137     "class MeanReversion(CustomFactor):\n",
    138     "    \"\"\"Compute ratio of latest monthly return to 12m average,\n",
    139     "       normalized by std dev of monthly returns\"\"\"\n",
    140     "    inputs = [Returns(window_length=MONTH)]\n",
    141     "    window_length = YEAR\n",
    142     "\n",
    143     "    def compute(self, today, assets, out, monthly_returns):\n",
    144     "        df = pd.DataFrame(monthly_returns)\n",
    145     "        out[:] = df.iloc[-1].sub(df.mean()).div(df.std())"
    146    ]
    147   },
    148   {
    149    "cell_type": "markdown",
    150    "metadata": {},
    151    "source": [
    152     "## Create Pipeline"
    153    ]
    154   },
    155   {
    156    "cell_type": "markdown",
    157    "metadata": {},
    158    "source": [
    159     "The Pipeline created by the `compute_factors()` method returns a table with a long and a short column for the 25 stocks with the largest negative and positive deviations of their last monthly return from its annual average, normalized by the standard deviation. It also limited the universe to the 500 stocks with the highest average trading volume over the last 30 trading days. "
    160    ]
    161   },
    162   {
    163    "cell_type": "code",
    164    "execution_count": 7,
    165    "metadata": {},
    166    "outputs": [],
    167    "source": [
    168     "def compute_factors():\n",
    169     "    \"\"\"Create factor pipeline incl. mean reversion,\n",
    170     "        filtered by 30d Dollar Volume; capture factor ranks\"\"\"\n",
    171     "    mean_reversion = MeanReversion()\n",
    172     "    dollar_volume = AverageDollarVolume(window_length=30)\n",
    173     "    return Pipeline(columns={'longs'  : mean_reversion.bottom(N_LONGS),\n",
    174     "                             'shorts' : mean_reversion.top(N_SHORTS),\n",
    175     "                             'ranking': mean_reversion.rank(ascending=False)},\n",
    176     "                    screen=dollar_volume.top(VOL_SCREEN))"
    177    ]
    178   },
    179   {
    180    "cell_type": "markdown",
    181    "metadata": {},
    182    "source": [
    183     "Before_trading_start() ensures the daily execution of the pipeline and the recording of the results, including the current prices."
    184    ]
    185   },
    186   {
    187    "cell_type": "code",
    188    "execution_count": 8,
    189    "metadata": {},
    190    "outputs": [],
    191    "source": [
    192     "def before_trading_start(context, data):\n",
    193     "    \"\"\"Run factor pipeline\"\"\"\n",
    194     "    context.factor_data = pipeline_output('factor_pipeline')\n",
    195     "    record(factor_data=context.factor_data.ranking)\n",
    196     "    assets = context.factor_data.index\n",
    197     "    record(prices=data.current(assets, 'price'))"
    198    ]
    199   },
    200   {
    201    "cell_type": "markdown",
    202    "metadata": {},
    203    "source": [
    204     "## Set up Rebalancing"
    205    ]
    206   },
    207   {
    208    "cell_type": "markdown",
    209    "metadata": {},
    210    "source": [
    211     "The new rebalance() method submits trade orders to the exec_trades() method for the assets flagged for long and short positions by the pipeline with equal positive and negative weights. It also divests any current holdings that are no longer included in the factor signals:"
    212    ]
    213   },
    214   {
    215    "cell_type": "code",
    216    "execution_count": 9,
    217    "metadata": {},
    218    "outputs": [],
    219    "source": [
    220     "def rebalance(context, data):\n",
    221     "    \"\"\"Compute long, short and obsolete holdings; place trade orders\"\"\"\n",
    222     "    factor_data = context.factor_data\n",
    223     "    assets = factor_data.index\n",
    224     "\n",
    225     "    longs = assets[factor_data.longs]\n",
    226     "    shorts = assets[factor_data.shorts]\n",
    227     "    divest = context.portfolio.positions.keys() - longs.union(shorts)\n",
    228     "\n",
    229     "    exec_trades(data, assets=divest, target_percent=0)\n",
    230     "    exec_trades(data, assets=longs, target_percent=1 / N_LONGS if N_LONGS else 0)\n",
    231     "    exec_trades(data, assets=shorts, target_percent=-1 / N_SHORTS if N_SHORTS else 0)"
    232    ]
    233   },
    234   {
    235    "cell_type": "code",
    236    "execution_count": 10,
    237    "metadata": {},
    238    "outputs": [],
    239    "source": [
    240     "def exec_trades(data, assets, target_percent):\n",
    241     "    \"\"\"Place orders for assets using target portfolio percentage\"\"\"\n",
    242     "    for asset in assets:\n",
    243     "        if data.can_trade(asset) and not get_open_orders(asset):\n",
    244     "            order_target_percent(asset, target_percent)"
    245    ]
    246   },
    247   {
    248    "cell_type": "markdown",
    249    "metadata": {},
    250    "source": [
    251     "## Initialize Backtest"
    252    ]
    253   },
    254   {
    255    "cell_type": "markdown",
    256    "metadata": {},
    257    "source": [
    258     "The `rebalance()` method runs according to `date_rules` and `time_rules` set by the `schedule_function()` utility at the beginning of the week, right after market_open as stipulated by the built-in US_EQUITIES calendar (see docs for details on rules). \n",
    259     "\n",
    260     "You can also specify a trade commission both in relative terms and as a minimum amount. There is also an option to define slippage, which is the cost of an adverse change in price between trade decision and execution"
    261    ]
    262   },
    263   {
    264    "cell_type": "code",
    265    "execution_count": 11,
    266    "metadata": {},
    267    "outputs": [],
    268    "source": [
    269     "def initialize(context):\n",
    270     "    \"\"\"Setup: register pipeline, schedule rebalancing,\n",
    271     "        and set trading params\"\"\"\n",
    272     "    attach_pipeline(compute_factors(), 'factor_pipeline')\n",
    273     "    schedule_function(rebalance,\n",
    274     "                      date_rules.week_start(),\n",
    275     "                      time_rules.market_open(),\n",
    276     "                      calendar=calendars.US_EQUITIES)\n",
    277     "\n",
    278     "    set_commission(us_equities=commission.PerShare(cost=0.00075, min_trade_cost=.01))\n",
    279     "    set_slippage(us_equities=slippage.VolumeShareSlippage(volume_limit=0.0025, price_impact=0.01))"
    280    ]
    281   },
    282   {
    283    "cell_type": "markdown",
    284    "metadata": {},
    285    "source": [
    286     "## Run Algorithm"
    287    ]
    288   },
    289   {
    290    "cell_type": "markdown",
    291    "metadata": {},
    292    "source": [
    293     "The algorithm executes upon calling the run_algorithm() function and returns the backtest performance DataFrame."
    294    ]
    295   },
    296   {
    297    "cell_type": "code",
    298    "execution_count": 12,
    299    "metadata": {
    300     "scrolled": true
    301    },
    302    "outputs": [
    303     {
    304      "name": "stdout",
    305      "output_type": "stream",
    306      "text": [
    307       "[2019-04-18 18:18:48.293921] WARNING: Loader: Refusing to download new benchmark data because a download succeeded at 2019-04-18 17:27:53.709992+00:00.\n",
    308       "[2019-04-18 18:19:05.859843] INFO: Performance: after split: asset: Equity(3064 [WCN]), amount: 1992, cost_basis: 26.25, last_sale_price: 39.27\n",
    309       "[2019-04-18 18:19:05.860257] INFO: Performance: returning cash: 0.0\n",
    310       "[2019-04-18 18:19:12.050942] INFO: Performance: after split: asset: Equity(3070 [WEC]), amount: 1842, cost_basis: 29.1, last_sale_price: 58.64\n",
    311       "[2019-04-18 18:19:12.051501] INFO: Performance: returning cash: 0.0\n",
    312       "[2019-04-18 18:19:15.749578] INFO: Performance: after split: asset: Equity(798 [DDD]), amount: 3014, cost_basis: 19.18, last_sale_price: 37.08\n",
    313       "[2019-04-18 18:19:15.749994] INFO: Performance: returning cash: 0.0\n",
    314       "[2019-04-18 18:19:24.574877] INFO: Performance: after split: asset: Equity(2349 [QSII]), amount: 1200, cost_basis: 44.6, last_sale_price: 86.7\n",
    315       "[2019-04-18 18:19:24.575377] INFO: Performance: returning cash: 0.0\n",
    316       "[2019-04-18 18:19:29.655546] INFO: Performance: after split: asset: Equity(641 [COG]), amount: 1750, cost_basis: 33.66, last_sale_price: 66.62\n",
    317       "[2019-04-18 18:19:29.655937] INFO: Performance: returning cash: 0.0\n",
    318       "[2019-04-18 18:19:32.026047] INFO: Performance: after split: asset: Equity(126 [ALK]), amount: 1694, cost_basis: 34.75, last_sale_price: 69.79\n",
    319       "[2019-04-18 18:19:32.026520] INFO: Performance: returning cash: 0.0\n",
    320       "[2019-04-18 18:19:32.326564] INFO: Performance: after split: asset: Equity(793 [DCI]), amount: 1616, cost_basis: 35.59, last_sale_price: 70.24\n",
    321       "[2019-04-18 18:19:32.327034] INFO: Performance: returning cash: 0.0\n",
    322       "[2019-04-18 18:19:36.955213] INFO: Performance: after split: asset: Equity(1397 [HSH]), amount: 592, cost_basis: 95.05, last_sale_price: 18.5\n",
    323       "[2019-04-18 18:19:36.955605] INFO: Performance: returning cash: 57.03\n",
    324       "[2019-04-18 18:19:50.093413] INFO: Performance: after split: asset: Equity(798 [DDD]), amount: 1513, cost_basis: 39.93, last_sale_price: 56.95\n",
    325       "[2019-04-18 18:19:50.093863] INFO: Performance: returning cash: 19.96\n",
    326       "[2019-04-18 18:19:59.922411] INFO: Performance: after split: asset: Equity(3132 [WTR]), amount: 2501, cost_basis: 25.15, last_sale_price: 30.37\n",
    327       "[2019-04-18 18:19:59.922817] INFO: Performance: returning cash: 6.29\n",
    328       "[2019-04-18 18:20:11.254267] INFO: Performance: after split: asset: Equity(1733 [LVNTA]), amount: 1040, cost_basis: 61.38, last_sale_price: 113.61\n",
    329       "[2019-04-18 18:20:11.254715] INFO: Performance: returning cash: 0.0\n",
    330       "[2019-04-18 18:20:11.342908] INFO: Performance: after split: asset: Equity(2905 [UAA]), amount: 1218, cost_basis: 52.93, last_sale_price: 102.16\n",
    331       "[2019-04-18 18:20:11.343320] INFO: Performance: returning cash: 0.0\n",
    332       "[2019-04-18 18:20:14.173526] INFO: Performance: after split: asset: Equity(2926 [UFS]), amount: 1452, cost_basis: 43.47, last_sale_price: 86.93\n",
    333       "[2019-04-18 18:20:14.173920] INFO: Performance: returning cash: 0.0\n",
    334       "[2019-04-18 18:20:15.186710] INFO: Performance: after split: asset: Equity(126 [ALK]), amount: 1352, cost_basis: 46.8, last_sale_price: 96.86\n",
    335       "[2019-04-18 18:20:15.187103] INFO: Performance: returning cash: 0.0\n",
    336       "[2019-04-18 18:20:20.444623] INFO: Performance: after split: asset: Equity(198 [APH]), amount: 1274, cost_basis: 48.87, last_sale_price: 98.68\n",
    337       "[2019-04-18 18:20:20.445017] INFO: Performance: returning cash: 0.0\n",
    338       "[2019-04-18 18:21:19.568282] INFO: Performance: after split: asset: Equity(2889 [TWO]), amount: 2970, cost_basis: 19.6, last_sale_price: 9.8\n",
    339       "[2019-04-18 18:21:19.568683] INFO: Performance: returning cash: 9.8\n",
    340       "[2019-04-18 18:21:19.633009] INFO: Performance: after split: asset: Equity(2889 [TWO]), amount: 2600, cost_basis: 30.83, last_sale_price: 9.84\n",
    341       "[2019-04-18 18:21:19.633401] INFO: Performance: returning cash: 15.41\n",
    342       "[2019-04-18 18:21:22.980920] INFO: zipline.finance.metrics.tracker: Simulated 2013 trading days\n",
    343       "first open: 2010-01-04 14:31:00+00:00\n",
    344       "last close: 2017-12-29 21:00:00+00:00\n"
    345      ]
    346     }
    347    ],
    348    "source": [
    349     "backtest = run_algorithm(start=start,\n",
    350     "                         end=end,\n",
    351     "                         initialize=initialize,\n",
    352     "                         before_trading_start=before_trading_start,\n",
    353     "                         capital_base=capital_base)"
    354    ]
    355   },
    356   {
    357    "cell_type": "markdown",
    358    "metadata": {},
    359    "source": [
    360     "## Extract pyfolio Inputs"
    361    ]
    362   },
    363   {
    364    "cell_type": "markdown",
    365    "metadata": {},
    366    "source": [
    367     "The `extract_rets_pos_txn_from_zipline` utility provided by `pyfolio` extracts the data used to compute performance metrics."
    368    ]
    369   },
    370   {
    371    "cell_type": "code",
    372    "execution_count": 13,
    373    "metadata": {},
    374    "outputs": [],
    375    "source": [
    376     "returns, positions, transactions = extract_rets_pos_txn_from_zipline(backtest)"
    377    ]
    378   },
    379   {
    380    "cell_type": "markdown",
    381    "metadata": {},
    382    "source": [
    383     "## Persist Results for use with `pyfolio`"
    384    ]
    385   },
    386   {
    387    "cell_type": "code",
    388    "execution_count": 14,
    389    "metadata": {},
    390    "outputs": [
    391     {
    392      "name": "stderr",
    393      "output_type": "stream",
    394      "text": [
    395       "/home/stefan/.pyenv/versions/miniconda3-latest/envs/env_zipline/lib/python3.5/site-packages/IPython/core/interactiveshell.py:2901: PerformanceWarning: \n",
    396       "your performance may suffer as PyTables will pickle object types that it cannot\n",
    397       "map directly to c-types [inferred_type->mixed,key->block4_values] [items->['factor_data', 'orders', 'period_label', 'positions', 'prices', 'transactions']]\n",
    398       "\n",
    399       "  if self.run_code(code, result):\n",
    400       "/home/stefan/.pyenv/versions/miniconda3-latest/envs/env_zipline/lib/python3.5/site-packages/IPython/core/interactiveshell.py:2961: PerformanceWarning: \n",
    401       "your performance may suffer as PyTables will pickle object types that it cannot\n",
    402       "map directly to c-types [inferred_type->mixed,key->axis0] [items->None]\n",
    403       "\n",
    404       "  exec(code_obj, self.user_global_ns, self.user_ns)\n",
    405       "/home/stefan/.pyenv/versions/miniconda3-latest/envs/env_zipline/lib/python3.5/site-packages/IPython/core/interactiveshell.py:2961: PerformanceWarning: \n",
    406       "your performance may suffer as PyTables will pickle object types that it cannot\n",
    407       "map directly to c-types [inferred_type->mixed,key->block0_items] [items->None]\n",
    408       "\n",
    409       "  exec(code_obj, self.user_global_ns, self.user_ns)\n",
    410       "/home/stefan/.pyenv/versions/miniconda3-latest/envs/env_zipline/lib/python3.5/site-packages/IPython/core/interactiveshell.py:2901: PerformanceWarning: \n",
    411       "your performance may suffer as PyTables will pickle object types that it cannot\n",
    412       "map directly to c-types [inferred_type->mixed,key->block3_values] [items->['commission', 'order_id', 'sid', 'symbol']]\n",
    413       "\n",
    414       "  if self.run_code(code, result):\n"
    415      ]
    416     }
    417    ],
    418    "source": [
    419     "with pd.HDFStore('backtests.h5') as store:\n",
    420     "    store.put('backtest', backtest)\n",
    421     "    store.put('returns', returns)\n",
    422     "    store.put('positions', positions)\n",
    423     "    store.put('transactions', transactions)"
    424    ]
    425   },
    426   {
    427    "cell_type": "markdown",
    428    "metadata": {},
    429    "source": [
    430     "## Plot Results"
    431    ]
    432   },
    433   {
    434    "cell_type": "code",
    435    "execution_count": 42,
    436    "metadata": {},
    437    "outputs": [
    438     {
    439      "data": {
    440       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+gAAAGoCAYAAADVZM+hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzs3WdglFXWwPH/1LRJT0hCSCOF3ntvoi4K6IKKgNi7rmvbd9VVEVGxrsoKq6LgiigWEAUUBVGkQyBAAgRISO+9Z9rzfpgwSQglCS0D5/cp89Q7cxOY89x7z1EpiqIghBBCCCGEEEKIS0p9qRsghBBCCCGEEEIICdCFEEIIIYQQQog2QQJ0IYQQQgghhBCiDZAAXQghhBBCCCGEaAMkQBdCCCGEEEIIIdoACdCFEEIIIYQQQog2QHupGyCEEEI4AovFwv/+9z9+/PFHLBYLJpOJMWPG8Nhjj6HX6y/ovTMyMpg4cSJ79+4943HffPMNRqORGTNm8OWXX1JeXs599913zvefP38+X3zxBQEBAQAoikJFRQXjx4/nn//8JyqV6ozn33XXXbz11lv4+Picc1uEEEKIy5kE6EIIIUQzzJ49m9LSUj777DPc3d2pqqriqaee4rnnnuPNN9+81M0DIDY2lujoaABuvfXW83rtCRMm8MILL9hfl5aWMmnSJIYPH86IESPOeO6WLVvOa1uEEEKIy5UE6EIIIcRZZGRk8OOPP7J582YMBgMArq6uvPTSS+zZsweAf/7zn0RHR3P33Xc3eT127Fiuv/56tm/fTmlpKffccw979uwhISEBrVbLwoULCQgIYOzYsbz33nv06NEDwP7a29vb3paCggJeeOEFCgsLyc/PJzg4mHfffZc9e/bw22+/sWXLFpydnSkqKqK4uJixY8fy+uuv8+OPPwJQVlbGuHHjWL9+PTU1NcyZM4fs7GxMJhPXXXcdDzzwQLM+k4KCAmpqavD09AQgKSmJV155hZKSEiwWC7fddhtTp07lmWeeAeD222/no48+YsaMGad9jzNmzCAyMpLMzEzmzZvH008/zahRo9i3bx9lZWU8/fTTjB8/nqSkJJ577jmMRiOKojB16lRmzJhxrt0shBBCXHKyBl0IIYQ4i4SEBKKiouzB+Qn+/v5cc801zbpGbW0tX3/9NY899hgvvPACt99+Oz/88ANBQUGsXLmy2W1Zs2YNvXv3Zvny5WzYsAFnZ2dWrVrF+PHjGTt2LHfccUejYHXYsGFUVlZy4MABAFavXs2oUaPw9PTk6aefZsqUKaxYsYJvv/2WrVu3snbt2lPed+3atUyePJmrr76aQYMGMXfuXF566SV69uyJ2Wzmb3/7G08++SQrVqxg6dKlfPrpp8TFxfHaa68B8NlnnxEUFHTG95aTk8NDDz3EunXr8Pf3Jz09neHDh/Ptt9/y5JNP8uqrrwLwySefMHbsWFasWMFHH33E7t27sVqtzf4MhRBCiLZKRtCFEEKIs1Cr1eccAF599dUAhISE4OfnR+fOnQEIDQ2ltLS02de5/fbb2b17N4sXLyYlJYWjR4/Sq1ev0x6vUqmYMmUKK1eupEePHqxYsYJ//OMfVFVVsWvXLkpLS3nvvfcAqKqq4vDhw0yYMKHJdU5McTcajbz88sscO3aMsWPHApCSkkJaWhrPPvus/fiamhoOHjxI7969m/3etFpto+N1Oh2jRo0CoGvXrpSUlAAwfvx4/u///o/9+/czZMgQ/vWvf6FWy5iDEEIIxycBuhBCCHEWPXv2JDk5mYqKikaj6Lm5uTz//PO8//77qFQqFEWx7zOZTI2u0TCRnE6nO+29Gl7DaDQ22f/mm2+yf/9+pkyZwqBBgzCbzY3OOZWpU6dy4403ctNNN1FeXs7AgQOpqKhAURS++uorXFxcACgqKsLJyemM19Lr9Tz//PNMmTKFN954g3/9619YLBbc3d1ZtWqV/biCggLc3d1b9B71ej1abf1XE51OZw+8GyaiGzNmDOvWrWPr1q1s27aNDz74gBUrVhAYGHjGtgshhBBtnTxuFkIIIc4iICCAiRMn8uyzz1JRUQFARUUFs2fPxsvLC2dnZ7y9vYmPjwdsgfvOnTtbfB8fHx/7NXbs2EF+fn6TYzZv3sztt9/ODTfcgK+vL1u3bsVisQCg0Wgwm82nbH/Pnj154YUXmDp1KgAGg4HevXuzePFiwLY2/dZbb2XDhg1nbader+fFF19k2bJlHDx4kIiICPtUe4Ds7Gyuv/56+3tp2K7mvMezefLJJ1m7di3XXXcdL774IgaDgbS0tBZfRwghhGhrZARdCCGEaIYXX3yRBQsWMG3aNDQaDUajkauuuopHH30UgNtuu42nnnqKa665hg4dOjB48OAW3+Opp55i9uzZLF++nG7dutGtW7cmxzz88MO88cYbvPfee+h0Ovr27WsPTkeOHMm8efNOee2bbrqJxx57jIULF9q3vfXWW7z88stMnDgRo9HI9ddfz6RJk5rV1v79+zNx4kTmzJnDl19+yYIFC3jllVdYtGgRZrOZxx57jH79+gFw7bXXcttttzF//vxmvcezeeihh3juuedYvnw5Go2Gq666igEDBrT4OkIIIURbo1LONi9OCCGEEEIIIYQQF5xMcRdCCCGEEEIIIdoACdCFEEIIIYQQQog2QAJ0IYQQQgghhBCiDZAAXQghhBBCCCGEaAPaTBZ3q9WKxeIY+eo0GpXDtFVIfzki6TPHI33meKTPHI/0meORPnM80meOx1H6TKfTNOu4NhOgWywKJSVVl7oZzeLl5eowbRXSX45I+szxSJ85HukzxyN95nikzxyP9JnjcZQ+8/d3b9ZxMsVdCCGEEEIIIYRoAyRAF0IIIYQQQggh2oBWT3G3Wq3Mnj2bxMRE9Ho9c+fOJSwszL5/7ty57NmzBzc3NwAWLFiAu3vzhvWFEEIIIYQQQogrTasD9PXr12M0Glm+fDlxcXHMmzePhQsX2vcnJCSwaNEifHx8zktDhRBCCCGEEEKIy1mrA/TY2FhGjBgBQO/evYmPj7fvs1qtpKam8sILL1BQUMDUqVOZOnXqGa+n0ajw8nJtbXMuKo1G7TBtFdJfjkj6zPFInzke6TPHI33meKTPHI/0WdtWXmPC3VnXaNvl1metDtArKiowGAz21xqNBrPZjFarpaqqipkzZ3LnnXdisViYNWsW3bt3p3Pnzqe9nmRxFxeK9JfjkT5zPNJnjkf6zPFInzke6TPHI33Wdi3ekcaCzSn88uBgvF319u2O0mcXPIu7wWCgsrLS/tpqtaLV2uJ9FxcXZs2ahYuLCwaDgcGDB3P48OHW3koIIYQQQgghxBVsweYUAL6Jy7q0DbnAWh2g9+3bl02bNgEQFxdHTEyMfV9KSgrTp0/HYrFgMpnYs2cP3bp1O/fWCiGEEEIIIYS4oqQVV9t//nhbGlZFuYStubBaPcV9/PjxbNmyhWnTpqEoCq+++iqLFy8mNDSUcePGMXHiRG6++WZ0Oh2TJ08mOjr6fLZbCCGEEEIIIcQV4EheRaPXPx/K49PtaSy9re8latGFo1KUtvH4wWSyOMTaAXCcdQ7CRvrL8UifOR7pM8cjfeZ4pM8cj/SZ45E+a1sqas38kpjPsfxKvonL4tnx0bz661H7/iXTezOsS6BD9Flz16C3egRdCCGEEEIIIYS4EBRFYU1CLm9tTLJv6x/ihUatwmK1jTHnlNdequZdMBKgCyGEEEIIIYRoM55dfYgdqcXUmq2Ntod4u/DJtF78d0sqZbVmerb3uEQtvHAkQBdCCCGEEEII0SbUmq38mphvf92rvQf/vrE7RVVGALoFeTB/ao9L1bwLrtVZ3IUQQgghhBBCiLP5ak8mCdllAJgtVswW62mPPZxbDsAtfdoDMCDUC3dnLWE+rhe+oW2AjKALIYQQQgghhLggKmrNvL0xCTe9huEdfVh3OJ/2Hk6sundQo+OqjBb+teYQZTVmAO4aHModA0PwcdNfimZfMhKgCyGEEEIIIYQ476yKwoYjtunqlUYL6w7bfs4qq8VsVdCqVfZjd6WV8GdyESqgo68rPq5XVmB+ggToQgghhBBCCCHOi+yyGlYn5KJTq1gWm0lxtYlgT2dGRPry1Z5M2hn05FUYmb8pmam92hPi7UJ+RS3vb0rGSatm5d0D0Gmu3JXYEqALIYQQQgghhDgvFu9IY+X+nEbbHh0ZwbAIHyZ1D0CjVnHLkliWxWayLDaToRHebD1eDMCIjj74G5wuRbPbDAnQhRBCCCGEEEKcF/syy9BpVHT0dePV67vg4azFy0UHQLS/AYBXruvM+5uOk1teaw/Op/cL5tGRHS9Zu9sKCdCFEEIIIYQQQpyz5MJKkgurmDWgwxmD7as7t8NFp2F1Qi7VJgvbUooZEOrVaE36lUoCdCGEEEIIIYQQ50RRFN7YcAyAqzu1O+vxIyJ9GRHpS63Zyu60EoZEeF/oJjqEK3f1vRBCCCGEEEJcoQoqajmUW84n21Mxmk9fl/x09mWWUmk0N3hdRmx6KU+PjaJTgKHZ13HSqhnW0Qe1SkbPQUbQhRBCCCGEEOKK8mdSIU98n2B//d8tqdw/NIx7hoQ16/zY9BIe+Ho/AD6uOlx0Gix1ZdMmdD376Lk4PRlBF0IIIYQQQogryKc70uw/j4n2A+DDranszSht1vkHc8rtPxdVmcgsrSGnvJbHR0dicJIx4HMhn54QQgghhBDiimFVFBQFNBcgIdmxgkq+jcvi6bFRrbq+oigs35vF0AgfQr1dznv7SqtNPL4ynvjscu4ZHMrMAR1w02v5fFc67286zn3L9zFnQif+0iXgjNc5ml+JVq3ix/sG4eGkpdpkwWxV8HXTn/c2X2lkBF0IIYQQQghxxXjsu3gmfrzjvFxrR0oxH29L5c0Nx/h+fza3fhbLd/uySSqobNX1kgqqeHtjErd+tvu8tO9ksRmlHMgu57b+HZg1MAQ3vW28dkLXAPp28ATgtyMFpzy31mylpNqEVVHYnlLMuBg//Nz06LVqPF10EpyfJxKgCyGEEEIIIa4Ivybmsz21mPwKI1/vzWRnarF9X5XRwsQPtnAgq+ys17FYFeZvOs4j3x3go62pfB2XxSu/HrXv/9faw2SV1gC2Efvm+iPJFhwbLQrP/HiIGpOl2ecm5lXw2q9HsVhPf78dKcU4adXcPywcF53Gvt3XTc+Ht/Ti+m4B7M0opbjKyJG8ikbnzt+UzPgF2xj0zp8UV5sY1tGn2W0TzScBuhBCCCGEEOKyZ7YqvPRzov31m78l8fC3ByivsWUiP5xXzuGcct7emIRylqD6j6RC/rcrHSetmjsHhbBkem9u7t3evj+7tIaPtqZwrKCS4e9t5u2NSfZge+3BXL6Ny2JnanGj+5TXmPlkexqezrZR7fVH8tmRWnLGdmSUVFNSbeJIXgUzP9/Div3ZZJfVnPb4vRmlDArzxkl76jCwW6A7pTVmrl64nRmf7yG9uNq+b1tKcaNjh4RJgH4htGoNutVqZfbs2SQmJqLX65k7dy5hYWFNjrnvvvsYN24ct95663lprBBCCCGEEEK0xpG8CmrNVl65rjMDQr34xw8HicssY2lsBqMifak22UqNJeSUM/CdPzE4aege6MFNfdqzcn823QLdmTUghOOFVfx+tACDk4b1Dw21rzXvFuTBzAEdsCoK7/6ezJqDeWxLKcZkUfhqTyal1Sb+ProjL/5U/5BgaIQ3fTt4cUuf9rzw02FMFoWpA9ozKsqXWUv3knOKYNtiVVifmI+rXsOzqw9Rc1KJtJyyWjp4uTQ6vqDSSDuDntzyWgaFn77eeJCHc6PXeRW1hHi78GdSIWnF1dzaN5gv92QC4OWqa2EPiOZoVYC+fv16jEYjy5cvJy4ujnnz5rFw4cJGx7z77ruUljYvC6AQQgghhBBCnE+KorA6IZcdqcVc26UdGSW2YLdXsCfernpm/6UTNyzaxafb0/h8VzqPjIhodL6LTsP21GK2102D35xcxIdbU+37r+7k3yQR3IkAt08HT34/VkhRlQmA0VG+/HQoj4wS24h0vxBPYtNL2Xq8mK3Hi/F21bH1eBEAU3oF2dZ2a1S8tTGJb/dlYbQoRPm5EeThxPK9Wad8vwNDvdiZVtJkBP3Fnw6z7nA+K+8eQJXJQoC702k/s3butnXkIV7OpJfU8NWeTHs5NYAbegZicNJIpvYLqFWfbGxsLCNGjACgd+/exMfHN9r/888/o1KpGDly5Lm3UAghhBBCCCFa6If4HOb+YlsXfiCrjOEdfXHTa2hnsAWhAQZboOqm11BptPDv35MJ83El2s+V0hozC27qyfT/xXI035bw7a5BIXy6Ix0fVx3XdG7H9H7Bp733pO6BqFQqPJ21xGeXE+ThxO/HCjmQXU6UnxsLb+rJSz8nMijcm5d+PsLL644AEOrtgn9duzxddORXGEkpsgX1J9a0A4yM9CXIw4ldaSU8d3UMS3ak8cjICKYtiWXJznSKq0zMGhjC7J8TWXc4H4AbP9kFQJ9gj9O2O8rPjVev70KfYA+u/2gHvx8rBMDbRcdTYyPp6OvGfUPdWtgToiVaFaBXVFRgMBjsrzUaDWazGa1Wy5EjR1i9ejXvv/8+H3zwQbOvqdGo8PJybU1zLjqNRu0wbRXSX45I+szxSJ85HukzxyN95niuxD6zWJULUr6spRKyyuzBuZeLjnaezhRUmwjxdsXbuz7A3P5/Y3DVa+n58q8ALJjRl2h/236VSsXSuweRW1aDVVHo1t6TZ67v1qz7ewEPBtQHwjUmC2Ht3AnwcCbM1xVvD2fevbUvAEt3Z9oTst3QO9j+O9PB25X8CiMzB4VSbbLwc3wOlUYLz/6lM3cODW90v5FdAwFQq1WkFVcz/8/j9Iv0ZU1CbqPjXpncjWFdAs/Y9psG2d7/vL/24OnvDgDwn+l9GBjeNtecX25/Z60K0A0GA5WV9aUDrFYrWq3tUt9//z25ubncfvvtZGZmotPpCA4OPutousWiUFJS1ZrmXHReXq4O01Yh/eWIpM8cj/SZ45E+czzSZ47nSuuzb+KyeGPDMZ4cE8m0vqcfXb7QfjqUywtrbeu8X5/Uld+O5BOfXU5WcTU923s26hMNUGu28MuDgympNhPTztBovxoIcrHFOefalyPDvGw/WK2NrhXl50paURX/vaUXXQPq7//cVVEs2JzC/YNCcNZpCHF34t0/konxdj5tWx4YGsYHm1MAuGOJrVRb10B3DuaUAzAizKvZ72N0uDebHxuORgVajbrN/i47yt+Zv797s45rVYDet29fNm7cyIQJE4iLiyMmJsa+7x//+If95/nz5+Pn5ydT3YUQQgghhLhM1ZqtDH9vs/312xuTGBrhQ6i3yxnOujBqTBbe2HAMgP9M7cGgMG/iMkrJrJsePjTi1AnSvF31eLtemjrej47syMz+HYj2NzTaHubjyuuTutpf39KnPeNi/Ag8KZFbQ3cMCiXK343HVybYtz1/TQz/3pjEzrSSRqXVmuN02d7FhdOqAH38+PFs2bKFadOmoSgKr776KosXLyY0NJRx48ad7zYKIYQQQggh2iCrovDot/ubbF+yIw1/g57CShNGi5VHR0bY11ZfSCsP5FBRa+G/N/ekX4htxNqnLtu4m17D1Z3bXfA2tJSfmx4/t7M/HNBq1GcMzk8IaZDBfecTI1CpVPz7xu5Ut6Cmurh0WhWgq9Vq5syZ02hbZGRkk+MeffTR1rVKCCGEEEII0aZZrAqTF+0kt7wWgF8fHIKbk4anViXw40lrn/dnlfHtnf3Ras48IltjsvD0qoM8MiKCXeklpBRV8beREXg4n72kV2m1iU+2pTIg1Iu+HTzt253rRo1HRvpeESPCwV4uDAj14qbe7VGpbPkA9Fo1+ivgvV8OJD++EEIIIYQQosXe3phEbnktET6u/HN8lL0udv8QL7YeL6ZHkAdv39CVrceLmf1zIr8k5tPew5muge7otWoUReGjrankVxj51zW2JbPHCioblTYDsFoVXri201nb894fyVQYLfx9VEd7YArYk9adSP52udOqVSy4qeelboZoJQnQhRBCCCGEEC22oy6Ifn9K90ZTr6/q5E9yYRUPDw/H21VPvxDbaPaLP9kSt901OJQHhobxyi9HWRWfA8B9Q8No5+5ESbWpyX1+TMhlQtcAvFx1WCwKn+xIY3L3QAI8nIjyswXdiqKwKamQv3RpR0y7xmu5b+gRiNWqMKVX0Pn/EIQ4zyRAF0IIIYQQQrSYp7OOAaFOTdZFB3k482KDEe9AD2c+mNqDHxNy+flQHp9uT2PlvmyKq01E+LhyvKiK7LIa2rk7kV9hBGyj3UfzK7lncCiLtqfx4DeN17lvPFqAVq3iu7sG4O2qI6u0htIaM10CGgfnADqNmlsuYVZ5IVpCAnQhhBBCCCFEixRU1JJSVMWoKN9mHT8wzJuBYd6MjPTl2dWHKK4bKf9rryDe3phEpdGWwKyg0hagfzajD7q69eqVRgtf7sm0X6tXew983PRsPFrA5EU7AZjZvwMatYqrOvmft/coxKUgAboQQgghhBDijBRFabSu+43fkjBZrEzv17KR6REdfbilT3si/dzYlVZCn2Db9PfHVsSzdGZfVifkolFhD84B/j66I8Gezry1MYmNjwzF4GQLYcZ9sJWyGjMA3x/IpkeQOz6XqFSaEOeLBOhCCCGEEEKI09qbUcp9y/fx+cw+OGk1hPu4cDS/gpGRvk1qd5+Ns07DU2OjALixZxA5ZTX2fTOX7jnlOWqVilv6BjOld3u06vqHBEMjfPj5UB4AFbUWBoadusa5EI5Ecu0LIYQQQggh7H45nMcXuzNQFAWApbszAFhzMI+bl+zmrd+SKKw04mc499FqN33T8cI3J3U95bENg3OAf10dw68PDrG/HiQBurgMyAi6EEIIIYQQAkVR2Hi0gOfWHAYg0s+V2PRSNiUVAlBea5tO/nVcFgC+52E6uate0+j1ZzP60DXQvVnnOmnVOGnVDO/ow+bkIroHNe88IdoyCdCFEEIIIYQQfBGbyXt/JBPo7kR+pZFHv4sHoEeQBweyy8gsqW50fPf25x4Qa9Qq/jezD1/tyaSy1nLKLOxn8/rErhgtVtQq1dkPFqKNkwBdCCGEEEKIK0St2cru9BKGRfjYt1msCt/EZbFkRxrR/m78b2ZfcspqeGpVAmOi/JjerwNjP9hKalHjAL1ne8/z0qYuAe689JfOrT5fr1Wj18rKXXF5kABdCCGEEEKIK8Rr64+yJiGXr27vR6SfGwAvrD3ML4n5RPi68tzVMWjVKjp4ufDV7f0BMFtta9GLq00YnDTcOySM/ApjkzXhQohzJwG6EEIIIYQQl6mUwioe+e4AV8X4E+7jwpqEXAByymqJ9HOjpNrEL4n5AHx1e79TThPXqlW46jRUmSz4uzkxvV+Hi/oehLiSyFwQIYQQQgjRRHGVka/2ZFJrtgLwbVwW0z7b3agsVkvVmCz8+/cknl6VgKVuVFZcGEazlUO55SzankpueS1fxGbwyq9H7ftTiqooqjKSkFMOwKvXdznjGu6YdrbRdt/zkLldCHF6MoIuhBBCCCGaWLIznWWxmby9MYnrurbjt6MFVJusfLknk8dHRzY5/s+kQuatP8qNPYMYEuFDlJ8b6SXVRNVNowaY8fke0opt65gP55bTLcjjor2flkjIKcdZq7ZPAQcorzEzZ10i+RVGrunSjlv6tG+TSclKq018uDWVzcmFZJfVnvIYnUbFu38k8+4fyfZtA0K8znjdnu09icssw8NJwgchLiT5CxNCCCGEEE1sO14MgFplq3+tVkGYtws/xOdw56BQvFx0FFUZ8XDW8duRfF74KRGLVeHDral8uDXVfp2x0X5UGS2MifGzB+cAiXkVTQJ0RVFQXeKg95u4LN7YcAyA56+JYVL3QPv234/Zyo0l5JQT7uPCkHCf017nYluxP5vX1x/lgWHhfFNXBu2Eazr7A3Btl3b4uel54Ov9mCwW+/47B4Xg5ao74/V7trf1VX7FqYN+IcT5IQG6EEIIIcQVLKOkmj3ppfQN8aSDlwsAvx8t4HhRFU+MiaSDpzNPfJ/Ac+NjiG7nxqyle1kWm0HXAHee/uGg/ToRvq58Mq03i7ansiw2k8Hh3mxPKea3owUAbE+1BfxvTe7KP388xH/+TOGvvdrbz88qreHWz2J59fouDOt46QLfE8E5wMvrjjCxWwC1Zivrj+QT6edKuI8rG44U8Lfv4tn2+IiLkihtf1YZqxNyiPIzsOFIPu/c2A03ff3X+JJqEx9vTcWqwKoDOQA8PDyc6HYGuge64+nSOPh+YnQkvx0tYMvxIgAeGh5x1jacqDE+JKLtPJQQ4nIkAboQQgghxGXqYE45WaU1dA10J8jDqcnotMWqcOMnuwDoF+LJyxM6czi3gtk/J+KsVTM22o92Bj3rHhyMj6tt7bG3i47FO9Kb3Ou/N/fE3VnL46Mj7VPgK41mFvyZQmxGCUkFVbjo1AwO9yHEy4XjRVWkFVfj46rj+bWHKa02UWWysGDz8UsWoFcZLaiA67oF4Oum57Od6ZRUm1iyM51j+ZXMm9SVsdF+DHh7EwCbkgoZG+1nPz+1qAonrZpAD+fz1qZKo5m7v4xrtO14YRVRfm7otWqqTRZu/GQnFbW2EfHM0ho6tTNwx6DQ015zUo9AJvUI5HhhFVUmy2mPa8jXTc8vDw7Gw/nMI+1CiHMjAboQQgghRBszb/1RvtuXjZNWzbgYP3oEeXDP6Khmn19eY+aFnw6zObnIvu3+oWHcMySs0XE/xufYf45NL2XChzsAWzC2ZHpvAtydAOzBOYDRYm10jZ/uH4STVoO7c9OvlW56LU+Ps7X7RHI5J62aNyZ35abFu1mTkENFraVRO4/kV5JbXmu/98WyL7OU19YfRQHGxfihQsVnpLM9tZhlsZkM7+jTKBgHmL8pmRAvZ6L9DQBMXbwbgJ1PjDhvU/U3HClosu3Br/dTY7byt5ERbEsptgfnJ5zcztOJ8HVtUVu8XSVBnBAXmmRxF0IIIYRoY77blw1ArdnK2oN5vL7hGIs2H2/2+d/uy2oU9ALe2qw5AAAgAElEQVR8uDWV9OJqzA0C7Bqz9eRTAXhsVMRpR4GH1U1xdtGp6RJgwM/gdMrg/GSBHs72a4Z5u6BRq/h0RzpfN1gv/frELoBtmrmiKFisClblwmV7rzVbOZZfiVVReHxlAkkFVQB0DnAnzMc23X9lXV+MjvK1n7firgFM6h5ARkkN0/+3p8l1B77zJxuO5J+XNibmVjTZdqLflsVmsiuthO5B7tw/NIxnxkez+r5BzBoYcl7uLYS4+Fo1gm61Wpk9ezaJiYno9Xrmzp1LWFj9E9kvvviCFStWoFKpePjhhxkzZsx5a7AQQgghxJVkSq8gvtuXzevrEpnaPeCMxyqKwraUYhZsTgFg4yNDWbk/mwB3J55bc5i/fmqbzv7TA4Pxc9OTWWob1f76jv6kFVdTVmPC36BnUJj3ae/x/DUxzBzQgUhfN1q7/FqlUjEw1IttKcVc09mfdYdtwezYGH8eG1XLe38kM/CdP5nSK4i1B3NZdc/A8zp6m1FSjbuTlsmLdlJptPDeX7vjrFNTXpf/zM9Nby8DtzezDICRkfUBeoi3C4+PjuSH+NxG1/Vw1lJWYwbgnz8eYmz3IM51HD2ztIaOvq4EeTgzrKMPwZ7O1JqtFFcZeW29bb389d0CmNJgPb8QwnG1KkBfv349RqOR5cuXExcXx7x581i4cCEARUVFLFu2jO+//57a2lquu+46Ro8efckzcgohhBDi/LMqCmU1Zlx1GvRamZjXkKIo1JqtOOs0LT7Xy0VHSbWJ0VG+3No32D6iXlFrxnCGMle700t4bEU8YAvsDU5abhsQglVRsCgKL6xNBGw1zVfuz6aoygTYpjo3d7qzs05DlwD3Fr+nk82Z0JmCCiNR/m7c1r8+i/iMfsGsT8wnIafc/r53pJZwTWd/vt2XzYiOPue0xjs2vYQHvt6Pr5ueSqNtanhGSTXtPZzJrzDaA2rNSU8fvE5KtGZw0jJrQAhf7smwZ5+vNVuJ9HO1j8SPeecP3ruxO72CPVvd3oJKI0Eezrz71+5N9vm46vl4Wyr9zlIiTQjhOFr1P2lsbCwjRowAoHfv3sTHx9v3+fj4sGrVKnQ6HQUFBXh4eEhwLoQQQlyG8sprmfLpLsYv2Maw9zZL+aWT/Hw4jxHvbyGlsKrJvoySar7em0VhpfGU55osViZ0bcfc67oQ5uPKa9fbpn5n163jPp0/6sqAPTM+mn+Mq1+zrlap+EuXAH68dyAAueW19uD8iTFNa5pfDF4uOqL8bXXGOwUY7GvOVSoVz4yPbnTsN3FZJBdW8caGY0z8eCezlu5hTUJuk2s29MvhPNYdymu0zWJVeH+TbalAw88+v8Jonw2w6Nbe9u0PDQ9ncvdAvrmj/ym/z3o4azFZFK7973YWbj5OrdnKuBh/frx3IG9M6orVCqvP0s6zKag04ud26tkDo6P9+GJWP8J9WraWXAjRdrVqBL2iogKDwWB/rdFoMJvNaLW2y2m1WpYuXcr8+fO57bbbmnVNjUaFl5dj/OOi0agdpq1C+ssRSZ85Hukzx9Owz8wWKyaLgou++SO97244yge/JwHQO8STuPRS3vo9mTBfV27uF0JUO9v3hMpaM05aNVrNlTe6frTQVvP72/gc5k6uH/1MK6pi2mex1JqtrDyQw09/G97k3FqzlTA/AwF+ts8xum4EttSs4O7h0mR0F2yj68v32tZz3zXy1EG3l5crKlV90OhvcOLBsdGnPPZS6u1WnyCuc6A7+7PKmPZZrH3bodwKlsdlMWNYBB/9mUxmSTUvTexm319lNPPcmsMA3DIk3L79lbWHOJhT3uhegR7OlJmsZJcbmdQziJFdA+37Hr+m8xnb2d7X9oChqMrEp3WZ7X09nOkc6kPnUB8W70qnzGhp9b+PZdUmCiuNRAW6y7+xF4n8f+Z4Lrc+a1WAbjAYqKystL+2Wq324PyEmTNncvPNN3Pvvfeyfft2Bg8efMZrWiwKJSVNnzC3RV5erg7TViH95YikzxyP9JljMZqt+Pu6sS3RlnwsqaASN72Gd27oxt7MMmb0C24yWlhWY2J9Yj5mKxzKLbcHeO0Mej68qSfXfbiD3xJt64hX789mzX2DUKlUXPvf7QS4O/HZjD4X/X22Rl55LQ9/u5/RUX6o1So8nLRYFYWberdv8VT1TUfqP48nR0bYP9Nl21KoNVsZHeXL78cKiU8pxNtVZ69rbbZYMVsVsFjtf1eGuu54cNleegd78PE02yivVVGwKqBVq0gtsh0b5ed2xr/HEznXPJy1LLutb5v9271vaBgfbU3l9v4deGb1oUb7Bod7syuthKy8ct785QgA03oFEVQ39X1vRqn92D1J+XSsC6SXbEsFwOCksWc+93XVsWJvJgChns4t+jz6BxmabNMp9d9p/dycyCmtafVnvC/T9j6CDfo220+XG/n/zPE4Sp/5+zdvaVCrAvS+ffuyceNGJkyYQFxcHDExMfZ9ycnJvPPOO8yfPx+dToder0etvvKemgshhBDnU3mNGa1GhUsr1jODLZBbuDmFbSnFJOZVMKF7ID8n5FCXB4tas5U7ltlqLQ8I9aJTu/rAQ1EUnliZwL6sMvu2yT0CuX9oGAYnLWqViv8bF8XyuCx2p5WQX2Ekt7wWXzc9hZXG007jbmtMFivPrTlESlE1S3Y2rvP9/qbjjIn247Xru5xy9PpkOWU1pBbbRtArjRb+/XsyVUYLqxqUNbt/WDi/Hyu01yHf9eRIoD5Dt7Ou/vuTj6sOJ62aWrOVuMwyakwWnHUanvw+gc3JRUzsFsDkHraR34eGh5+xbR9M7YFWo6Jvh7a9bvnuwaH0bO9B/xAvVt0zkMmLdjK+kz+vXt+F344WsD2lmOOF9QNGaxJy7WXk0kuq7dtnfr6HP/82nNTi+i/wT4yO5PUNx3DSqvGuW/ves70HN/SoHz1vDm9XPaMiffkjqdC+LabB346fu54jueWnOrVZiuuWIfgZpLyZEFeKVgXo48ePZ8uWLUybNg1FUXj11VdZvHgxoaGhjBs3js6dO3PLLbegUqkYMWIEAwcOPN/tFkIIIa4Y2WU1TPp4Jx19Xeno68ZNfYJaHFytT8xnyc50e3C5tkGgeMKk7gH8EJ/LvswyUKCjnysl1SY+3Z7GvqwyHhoezoGsMnoFezJrQIdGo+yjo/0YHe3H/qwy7v4yju0pxfQPrW+jVVFQX+CcNGarwndxWYyK8m1xErEV+7N57dejAMwa0IEh4T7sSism2MuF7NIaVuzPZuPRAtYczGVS9zMHcSVVJv63K6PRti/3ZDZ6PSzCh0hfV8J9XEgpsgWTJdUmvFx0VNUlLnNukHRPpVIR4OFMWt0o+coDOdzaN9heSu3HhFx7qTOPs5Q8G3iGDO1tiVqlsmeTb+/pzMq7B9iD6RPr1U8kYwPbqPnPh/J4fu1h+7bBYd5sTy2mqMpIRolt/f6SGX3oFujOuBh/jBYr89bb+v2azv54npQIrjnmTerKJ9tSWbQ9jQhfV6Lr1tWDbQlBYaXRnkSupU5khPd0bnm7hBCOqVUBulqtZs6cOY22RUbWr3V65JFHeOSRR86tZUIIIcRlymxVeHxFPFN7BzEqyu+Ux1isChkl1aQUVfHUqoMAJBdWkVxYhdlqJTGvku/3Z3Nzn/ZnLa+UUVLN3F+O0CXAwKe39mZTUiFzfz3KVTF+3Nq3A0+tSuDlCZ3pEmBgw5EC3vzNVrppcvdACquMbE4uwuCk4Y6BIWcNMroFutMjyJ03fjvG9d3qS4JV1lqaVSv7XLyzMYlv4rJ4a2MSS2/r22gWwJmU15jtwflTYyK5pW8wQKMHDPcNDePuL+NYuDmFq2L8cT3Nev09GSXcv3z/ae/VP8STKb3aMy7GD5VKxdRe7Xlro20tf3pxNV4uOoqqbDMOfE4qK9bwnl/FZjCtT+N+P5Jnq5d9pizvjqyDl4v95xMB+rGC+hH0nWkl7EwraXTOLX3bsz21mMySGspqbKPRnnW/h656Da5o0NflR2hNtn2wLS+Y1jcYo8XKrAEhjR5E+RmcMFttlQ5aGvzHZ5fxct30fU+Xy7NPhRBNydxzIYQQ4iJLKqhke2qxPfBWFIWV+7O576s40uumRS/fm8nUxbvtx9w7JBQ3vYZwHxd+P1bIOxuTSC6sYt76Y6QUVaEoCj8dymVXWnGT+609mEu1ycrLEzqj1agZG+PPnueu4tnxMUT4uvLdXQPoGuiOSmULNE74+XCePeibfW2nZo0AatQqXri2EyaLwsr99aP0ZbWm1n9gzZBXXssvh+szds/8fA8LNh+3tz+5sJLS6lO3YeV+WymvD2/paQ/OT6ZSqXhsVEcKKo2Mmr+F7SlFpzxuYV398ROeq8tGrgJW3j2AhTf34qpO/vbPcmL3QIaE20aJT0zLLqhbEuB7UuZupW7x+PCOPmSV1TLk338CcMfAEADi6up1n3ze5cjHVYdWrSKpQYDeUEdfV176Syd7Obh7l++joML2uZ48w+BEeUCzxdrq9ni66Hh0ZMcmQfiJBwlny75/KhuOFNh/dm3lwwMhhOORx3FCCCHERVJjsvCfP49zILt+Terkj3dgtirk1QUPD3+7n+/uGsDS3fVTpB8aHs6dg0K5e3AY6xPz+VeDKbxueg03Ld6Np7OW0rrpsAA9gjzoFezBIyMi+HhbGgCh3vUjkKfzwLBwLFaFJTvTqTVbKak2MbN/h9OO9J9KuI8rge5O5JTX0t7TmazSGsprzND6UtCnlFxYyer4XI4WVLI9xfZg4vVJXfl+fzbbUopZvCOdxTvSmTWgA//blUGPIHfeubF7o3rWseklLNySwoiOPvQ5S63qXsGeRPu7cTS/kke/i2802n6C9qT16R519+oa6N5oBPgEV72GtyZ3Y8T7m+0PZ4oqbQ8SfNwaB3vuddOcT0xtt9TlDxgV5cuSnemYrQrR/m5N6nVfjtQqFe0M+kYj6Ccsmd6bLoHu9pFsdyct5bVmPt6WioqmMwyGhHuz6kAOkX5uTa51rvqE2h6+/BCfS+cW1o7PKavB3UnL/CndpWSxEFcQGUEXQgghzkFibgU/H8qjxmQ567Ff7slk+d4sDuaUM7lHINP7BRPq49oosM4uq2Xou5vJrzDi7qTlh3sHcuegUMA2Ot2hLsge38mfPx4dZk9IVVpjpqOvLTAGOJBdxtLdGayrG1W+tku7Zn/Jf3hEBM/WjfwaLYo9M3ZLvH1DN27p055/jLXV4t6dXsriHWktvs6ZvPdHMp/vzrAH5/+9uSdjo/1454ZubHh4CIum9QKwrwc/kF3OtQu32WcZ7Mss5YGv99PBy5kXmzlD4J0b6kt5/X6soMl+7UmJcYM9bZ/dLX1PvwxBr1UT6O5kH0GvMdt+l04eNX1rak8eHh7OgLqp9yrglwcH0z3Iw35M96CWBYGOrL2ns72W+wlvTupKtyCPRtPMP62ra260KPi46ZvkQhgX48/PDwym11ke0LRGkKczN/duzzdxWfx+tOnvy5mkl9TQPcidbg36Vwhx+ZMRdCGEEKKVjhdWMXPpHgBi/N2Y2D2QUG8Xhkb4ADRKDGW2WPnpoC1Y/uK2vo0yPW9LKeLnQ3nc0COI+5bvs2+/tW9wk+C4a4CBF66JYUy0H656DQ8OC+f5tYeZP6UHod4u1JgtJBdU4eOm44ZFu3jxp0S0ahWzBnRo0XvrFVwfFAR5OJ3hyFOLaWfgqbFRHM23TTF/749kAGYNCEFTNzW5oNJoTwLWXCaLlSP5lRjNVrYeL2Zyj0BWHcgh2t+NfiG2wFWrUeOhUdMtyANvFx3dgtx5eHgElUYz93y1j4e+OdDomg8OC2/2+uBAD2c2PzacdzYmsaGujFpDlhM1zOp0amfg14eGnHVUO8TbhXWH8xne0Rdj3dC47qTa8cFeLtxR97Bm7f2DAFsWcYDXru/CM6sPMazud+9K0CvYk93ptjJk8yZ2ITa9lFFRvk2OC/d1xcdVR1GVias7+Z/yWhdyWcDjYyL5/VgB6w7nMTq6eTNRqowWEvMquLn3mfNLCCEuPxKgCyGEEC2UU1aDWqWyr10GOJJfydt1yb7WPzSE2T8ncii3gjsGhvDXnkHMWZfI8aIq3pzUtVFwDjAk3Ich4T7kldfat237+/BTlvNSqVRMbJBFvE8HT1bfN8j+2k2vpUd7Dyx19dP0GhVLb+tHhK9ri95juE/98V0DWz8q637SdOIaswU3vZZpn8UC9aXFmqPKaOHBb/ZzMKd+icDEbgE8MiICJ23TSYFatYpfHhpyxms+Nz6akS2Yvg/gpFUT5uNCaY2Z6z7cTlGViXUPDsbDWWcf9b61bzA31yVxa86U8xAvF3aklvD82sP2MmknB+gN+RsaPzQZF+PHV7f3uyDTtNuqfiGefLLd9vPgcG/GxZw6+AYI83ahqMp0Sab/a9UqAj2cKWkwU+Zs/vqprfReh2YsSxFCXF4kQBdCCCFaIKWwipuW7La/jvF346be7fl8dwZ3DQpl9s+JXLVgm33/2xuT7IH7pO4BpxzhO8G/Qa1j7RmCs+bQqFX89vBQ3Jw0rSpvplap+PL2fni76M5pdNHjpPJQ1SYrDS9XZbScNiP6yTYeLeBgTjkB7k646TW8fUO3U67rPpO/j+qIh7OWI/mVeDhpuaFnUIvOP2FkpC/fxmWRXle666eDedzSNxhnnRo3vYa/j+7Yos89pEEgZqpLVqbTNP98lUp1RQXnYMuzcMLZkqi1q1v6cbYSdBdKea2Z41lVZJXW0N7z7EtGCusSBXpJ9nYhrjjyVy+EEEI0U1GVkUe+q58e/cCwMIaE+9A10N0e6K1OyLFPu31/SnfmrT9GjcmCm17D/42LPuM6Z5VKxeLpvU85ct4a51rWLOo8BHwuOjX+Bj35dUnwakwWvt6bZd9/OK+82TXdtxwvwsdVxw/3Dmx1TfUZ/Vs21f90Oni5sOLugQx4exMAHnWBVHpJDWE+ri1uX8MEfkaLgkatuuB14x2ds05jryN/tvwBU3u1J6Okht4dzv868+Y4Xmir1/7ToVzuHhwGwP6sMrYcL2JslB+dAupn1TTM+N6r/aVprxDi0pEAXQghhDiL344WsHxPJnsybIH3R7f0Itrf7ZT1pt+Y1M1eA3xQmDer7hnYont1v8wSQqlUKq7rGsCSnemArXTUf/48Tu9gD+Iyy9iXWXbGAD2jpJqHvz1AWY2JiloLU3oFtanA9d4hoXy8LQ1FsSWdO5hTztNjI1t8nYZT1k0WK/oWjJ5fyZbN6kdF7dmnjvfu4MmSGX0uQovOzEWnYeGWFMK8XfhqTyaHcivIKq3h5QmdATCarUz6eCcAS2/r26zRdiHE5UWyuAshhBBnsT4x3x6cT+8XTJ8OnqcMzsE2av3yhM4tTn52OWu4Zn5v3ef4n6k9cdNrKKw08uDX+7h58W6WxWY0Ou9YfiV3LYsjq7SGIA9n7h8axiMjIi5q28/mr71s68yrjBZ+TMjFTa9p9H6bK9q/fraCyaKgP8clDlcKnUZtT5TXln15ez8Acstr+XR7Gi/+ZMtRAfDzoTz+uyUFgONFVfZzzscMFiGE45ERdCGEEOIszFaFjr6ufHhLLzwv0RpWRxbq7cLbN3Tjye8TyCq11XZ20qrxcdWxL7OMw3m2QGVNQi59OnhSWGlkT3opn+/OwM9Nzzd39m+UtK4tcatbP19tspCQXU6vYA9czrIe+lTUKhWzBoTw5Z4MjBbrGRPECccT5efGyEhflsVmnnL/stgMHhgWTrWxvlzj+VrqIoRwLPItQwghxGUpMbeCfVllDAz1QqdVEezZ+mzIlbVm3PTaS5IB+nJxoh748aIq+3prH1c9+7LKAFs2+sO55cxaurfReXcPDm2zwTnYMrqrsI2gF1YaG5Wnayl3Jw0mi0J+RW2LEsQJx/DSXzoxdfFuewK4hgLqktiV1tjqui+4qcdFbZsQou2QAF0IIcRl6bk1h0gtrra/Xnn3gBZn/D6hymTBoJf/Ms9Fw0zwJzJqj4z0tQfooyJ97dPf/dz03DagAyFeLgzr2LbreqtVKtycNJTXmqkyWVo1en7CgFDbWvxtx4sbZXUXlweDk5Z/39iNBX+msD21GIAnx0Ty9sYk+79NxVW2AP1cHigKIRybzJ8SQghx2VkWm0FqcTUDQ+uTj934yS7e3HCsxdeqMVmIzy6nvBmJqMTpNSxv1TfYlpn6pro64X06eHJjXRb8aH83fnpgMNP7dWBEpG+bSgh3Ot4uOgoqjdSarc0uGXcqHevWHCsga9AvU10C3Hl/SncGh3nz+qSuTOsbTP9QL8rraqTH55TjptfYH2IJIa48MhwghBCiTbEqChklNRzOLWdQmDeep5lWbrZYUalU9nWaVkUhqaCSd39PZmdaCSM6+vDaxK5o1Sp+Tczn0x1pfB2Xxf6sMgoqjXw2o0+zvgTHZdpGdRNyys/fm7wCqVUqbugRiFatspc6c9FpWHPfIJx1alz1Gn56YDAuOscLTL1d9WSV2kpjuZ1DgO6i09hL0rk7tf46om1TqVTMn1o/hd3HRUd8TjmKorAjpZgBoV5oZf25EFcsCdCFEEK0KY99F2+f/gnQ3tOZBTf1YEtyER193egf6kWNycJfP91FgLsTn9zaG7VKxUdbU/lke5r9vMdHR+KktQV713Zph6eLlrnrjtgTkr37RzKvXt/FfnxacTW/HM7j7sGhjWoq59XV75Y1oefuuatjmmxr+JDEz63tZ+M+FR9XHbvSSgBwPYcp7gDjO/mzLDZTEoRdQcJ9Xfk1MZ8j+ZXklNdyx6CQS90kIcQl5HiPqYUQQly2fojPYXtqMZO6B9C/bnp6VmkNNyzaxZu/JfHgN/vZcryIVQdyyK8wEp9dTmJdwP1pg+AcoINX4/rBQ8J9WHP/YP5TN3J1vLCKj7elsmDzcSxWhQe/3seHW1MpqlsDCrbEX9/ty8ZVp6F33bRsIU4W4etKZV327XOZ4g4wta5s2+nK+InLT4SPKwrw9V5bhvfB4VKiUYgrmfzrL4QQ4qKqqDWTWVLDL4n5dGrnxuBwbzycbdPYv9uXTZSfG89cFY1Wo+aH+BxeXneEQHcnQrxdKK8x8/cV8QB0amcgMa+CWUv38vUd/VGAx0Z1xMdVh0pFo1HwhgaFefPPq6KYt/4YxwoqAVi8I92+v6zGjNFixWJVePL7BJILq3hrcjcpeyVOq1tgfeb2doZzWzsc4u3CB1N7EOUvNbCvFBG+tioFP8TnApIgTogrnQToQgghLpr3/0jm890ZjbZN7RXE0+Oi+H5/Nol5Fczs3wFtXTA8qXsgk7oHAqAoCrvSSnj42wMA/Pfmnoz5z1YAbl6yG4BhET72L7tnMqVXe35NzCc2vbTJvpJqE/ct2Wd/fV3XdoyK8m3FuxVXim6BBvvPwSfN3GiNgWEygnolCZWM/UKIBiRAF0IIccHUmCxsPV5ETnktK/dnk1JkK3t29+BQ/A16Vifk8u2+bPRaNctiM+kX4sm0uszeJ1OpVPRo70GPIHceHdkRg5OWSD9XkgqqALh/aFizgvMTOvq6EZteiotOTbXJyvhO/vyamE9pdf0U92l9g3l4eHjrPwBxRfBrMGru66Dr6MWl03B2zpwJnS5hS4QQbUGrAnSr1crs2bNJTExEr9czd+5cwsLC7PuXLFnCmjVrABg1ahSPPPLI+WmtEEIIh/JNXBbvbzpuf63TqPjurgEEedhGGben2JLBLYvNJNDdiYU39Tzt1HSwZbn+dHof++sPb+7FrC/24qbXcPfg0Ba17ZEREYR4u5BUUMmqAzn4G2yBVVmtGb1GxY09g3hyTGSLrimuXHMmdEKrVjtEWTjR9iya1gutRk23QPdL3RQhxCXWqgB9/fr1GI1Gli9fTlxcHPPmzWPhwoUApKen88MPP/DNN9+gUqmYPn06V111FZ07dz6vDRdCCNF2KYrCl3sy+eDP47Qz6BkS7sOgcG/GRPs1Kh/04PBwfj9WCICfQX/G4PxUPF10rLx7AIpy+jXnp+Oq13Br32De/yO50fbcslqMFsX+EEGI5vhLl4BL3QThwHpJEkohRJ1WZbyJjY1lxIgRAPTu3Zv4+Hj7vsDAQBYtWoRGo0GtVmM2m3FyOreEKUIIIRzLu38k8+/fkxkS4cOn0/vwr2tiGN/Jv0lt346+bqy+bxAB7k7cNzTsNFc7M3WDWuitMa6TPwBjovwAOFqXOC7IQ/7vEkIIIcTF1aoR9IqKCgyG+oQoGo0Gs9mMVqtFp9Ph4+ODoii88cYbdO3alYiIiLNeU6NR4eXV/LWDl5JGo3aYtgrpL0ckfeZ4GvbZsbwKvt6bxcSeQbw1pSfqswTPXl6ubP7HmIvRzFMa5uXK0ZevBUD19T521tVg79TB+7L+PZS/M8cjfeZ4pM8cj/SZ47nc+qxVAbrBYKCystL+2mq1otXWX6q2tpZnn30WNzc3XnzxxWZd02JRKCmpak1zLjovL1eHaauQ/nJE0meO50Sf/X60gDd/O4Zeo+a+QSGUlVVf6qa1iKJgr2ftoeay/j2UvzPHI33meKTPHI/0meNxlD7z929ejolWTXHv27cvmzZtAiAuLo6YmBj7PkVReOihh+jUqRNz5sxBo9G05hZCCCEcyI7jRbz66xGe/uEgKpWKt27oSqCDr+F2d5ZCJ0IIIYS4uFr17WP8+PFs2bKFadOmoSgKr776KosXLyY0NBSr1crOnTsxGo38+eefADzxxBP06dPnLFcVQgjhqOauOcTh3HJcdRo+ntbLYROsrXtwMEt2pFNjtlzqpgghhBDiCtSqAF2tVjNnzpxG2yIj60vRHDhw4NxaJYQQwmGkFlVxOLecqzv5c+fgUIcNzgF8XPU8IaXVhBBCCHGJyPw9IYQQ52RHagkAD40IJ9jT5RK3RgghhBDCcbVqDboQQrRlxwurSMytuNTNuCDcqEYAACAASURBVCgqas28+NNhdtRlHj8dRVH4M6kQo9l6zvdcFpvB06sSqDSaAdibUUqQpzPtHXjkXAghhBCiLZARdCHEZSGrtIaj+RUUVhp5bf0xANoZ9Dx/TQyDw33Yl1nKpqQiKo1mugW689nOdF65rgupxVWMivLDSetYzysrjWa2Hi8mv6KWtQfzWHswj7HRfvi56fEz6JnWNxgXXX2Szp2pJTzxfQJTegXxz6uiz+neX+zOIK/CyGu/HuWZ8dHsTi9hVIw/KlXra5ELIYQQQggJ0IUQl4n5m5JZf6QAABWgAHkVRh79Lp4wbxdSi+vLfX23LxuAmUv3APDw8HDuGBR62mvXmq1tLoBfHZ/LWxuTGm377WiB/eddaSWUVJvIK6/l0ZER/8/efQc2Xed/HH+mSfduactqy5bZskTZ0wHKEhBQOXCcp57nPsd5p8ghuH4/PbifiAtPFAFRRAFFhRME2XsVKJSySvdKS5M2ye+PQqBSoHSQBl6Pf2iSb755J+9vvuT9/SyOZBcBpb0LKutwZiGfbDhCmtkKwLKEdJYlpAPQu0WdSu9XREREREqpQBeRq8Luk/nU8fcio8DKQ90bcUdcPTYcyebFJQkczy0tToe1q0vdIG/mbjlBvqUEm90BwPYTeeXu84e9afxr5SEyCqzMHd+JpnX8r9j7uRC7w4EBmLf1OAADW0XSrn4QjcP8CPQxcc/s0osOG4/kOJ/z9Y6TBHqXtqYfzipkV0oebesF8WNCGtmFxTgo/Wx8PMsui5lVaGXaykP0aBJOkzp+3P/FNsyW0tnNV/y5GzN/O8y8rScA6NmsDpzu8i4iIiIilaMCXUTcVl5RMUdzinA4HKTkWXisV2OujwmhWUQAJg8D/VtE4GEwcH1MCEE+JmcX7PtuiOFgZiG7U/JYn5zD5qM5WEvsHMs9ReMwPwwGAyU2O/9YmuB8rT0n82ukQD+YUUD9YJ8y3dHP5XA4zv4N3DN7CwfSCwDo2DCYSYNaltn+83Ed+WBtMr8kZuLnaWRYXF3mbC4t5mNCfck5Vcyfv9zJ53/oyItLzr6/YF8TA1tFldnXh2uPsGRPGkv2pAGlPRMAnuzThEAfEw92i2Xe1hM0j/AnxM+LHBXoIiIiIlWiAl1E3NKek/m8/H0Ch7POdl3vEhvKdZEBzttGDwMDros477kGg4FmdfxpVsefyEBvftqXzs0z1lJgtfHKwOsY1DqKxIzSIviWlhEsS0jnyDld5Ksiu9BKiK8nBoMBs6WEMf/ZzIAWEUwd3Mq5TbHNzsPzd2AwwLbjeXibPLD8bnK3P3WLZWynBuftv0VkAG8ObcOWYzlEh/hi8jCwLCGdztHBTBzYknSzhbs/3cLwjzaWeV5avtX5d2JGAe/+msSWY7n0aBJGk3B/vE0GRravT5ifl3O7IB9P1j/VE408FxEREakeKtBFxG2cKrZxNPsULSIDmPj9PnJPnW2x/cctLcoU5xXVtVEYT/Zpwtu/HAJg8e5UWtcNZP/pVuoHuzVib6qZzUdzKLbZ8TRWfiz693tTeWnpPt4Z3pboUF92nMgFYPPRnDLbHcosLNPtPtzfixOnu+m/MKAZvZqVTgZ3MR0bhjj//uGhG51/1wvy4aOx7fkxIQ1vkwet6gby6IKd/PvXJHaeyOP6mBDe+u9B/DyNtK4XyJgODbihUegFX8dDE8OJiIiIVBsV6CLiNib9sI+f92cwd3wnkrIKeaRHI26IDSXnVDHdGodVer93dmjgLNA3Hslh1KxN3NWpAQagXpA3jcP8WHkwk78t3subQ9tU+nVmbzwGwJSf9jsnWgPw9TJiKbFTYrfj72XiQHrpEnHzJnQiv6iEuPpBFFhtZBcWEx1a9XXGG4f78afujZy3/3B9NJ9uPMrKg5msPJhJkI+JGaPiaFGJCx4iIiIiUnm1a1piEZGLWHkwE4Ax/9kMQL/mdWhdN7BKxTmAyeP8VuA5m48TEeCFp9GDJ/o0ITLAi18SM1m4I6VSr+FwOEg3W4kK9D7vsRO5RQz/aAN9pv/G0j2p7DlpxtvkQWyoH/ENgjEYDAR4m6qlOC/Pn3s2YvXjPZy337tTxbmIiIiIK6gFXUSq7Ej2KaICvWt0KbIfE9IotpVOmHZdZACdooOJDfOrtv2/MaQ1XiYPFu9KpWGID3lFJfRoUlr4NwzxZdqIdoz5z2am/HSAQG9TmbHtx3NPUT/I56LrgA/7cAM5p4p5qHssYX5eTPnpAKM71OePXWMZ//lW50zzL3+/D6MB+jaPwFjOhYOa4GEw4G0y0LZeIEeyT9GsFsxWLyIiInItUoEuIlVyqtjGiI830rVRKNNGtKuR11h7OIsXlyRQL8ibmaPjqRfkU+2v0bd56Tre3S/QGt+0jj+/PtadPy/YyeQf99M5JoQQX0/mbz3BmysS+b+R7egSW3as9uHMQrw9PQj19eREnsW5/+siA2gS7sd1kQH4eBppHO7H8dwiGof7kZRZiM0Bzw9oVu3v8VL+b2QcJg/DRS80iIiIiEjNURd3EamShNTS8dJrD2eTVWglq9BKflH1Lre1KrG0a/u7o+JqpDivKB9PI88PaEaB1cbP+9LZkJzNmysSgdJeBLtP5nM4qxAoXa981Ceb+MNnW8k7/Xm8cFNzWkYFYjAYiG8Q7Fx3fGCrSIweBp7u29T5WsG+nlf43YGflxGvGuwFISIiIiIXpxZ0Eam0hNR8Hpy33Xn7lhnrAGhax4+54ztX2+tkFhbTONyPhiE1Mwb7cjQJ98fDAP+3Ogmzxea8P73AyuufbwVg9j0dCPAuPb3mnComt6gYgGCf8k+5N7eMpH+L0i7td3dqSEyY69+niIiIiFx5KtBF3NjulDzMFttFl8GqST/tyyj3/oMZhew8kUe7+kFV2n9RsY1VBzPJKrAS7nflW5TLY/QwEOLrSVZhadE9bURb3v7vIbaes1Ta09/sLjNL++QfDwAQ7HPh93BmvPkTfZrURNgiIiIi4gbUl1HEDRUV29ifZmbCnG08+tVOHA6H87HsQutFnlm9jueeOu++kfH1ALjvi238fcne8x4/VWwj53Rxeykvfb+PF5cksP1EHpHlzH7uKjZ76ef9WK/GdG0URqfoYA5kFDgfP1Ocnym695zMJ9TXk7gqXrAQERERkaubCnQRN/TlthPcPXuL8/b+tNLicG9qPjfPWMc3lVwK7HIdyT5boIf4elI30LtMIb0sIR37ORcPCqwljPx4IzfNWMvoTzZRYndwITa7g/8eKG2hD/Q28ceusTXwDirH9/TY8R5NwoHS935ud/czVv6lO3d1agDATddFaHy3iIiIiFyUuriLXEEOh6NaZshOyizEz9NIwxAf9qcX8EtiBtdFBfBbUhYAr/50gOaRAbSpG1jl17oQS4mdQ5mFztvf/rELBqCo2I7d4cCAgRlrDpOWb6Hu6Ynd3lie6GxdPpRZyMrEDPq3iOB47in+Z8VB2jcIZlSH+oQA6ebSWc/HdW7Ig91inROq1Qb/O7wNKxMzaXR6rPi5E7oNbVeXRTtP0jkmBG+TB0/0bkLn6JDzZngXEREREfk9Fegi1cThcOAAvtl5kn7N6xDyu1m41xzK4uXvE3i4RyNGxNenxGbHZCzbonqq2Mb//ZpEVKA3466PvuBrncgrolmEPx+Nbc+D87bz4bojLNieQs6ps13HfzmQUSMFelahlZEfbyK+QRA2u4PoEB8e793U2ars42nk/htj2ZCcDcDRnFPUDfLhcFYhS/ekMbRtXRbtOgmU9gTo3yKCr7al8OuhLH49lEWgj4nW5mIe+HQTAD2ahtWq4hygeUQAzSMCnLeDfc+eSkfG1+PvN7dw3jYYDPRsGn5F4xMRERER91Sp/pZ2u52XXnqJ0aNHM27cOJKTk8/bJisri5tvvhmLxVLlIEVqO2uJnfGfb2XQzPVM/ekA/16VVObxzAIrLyzeQ25RCa/9nMjy/el0fWc1X2w5Xma7vyzYybytJ3hvzeGLvl5SZiGxoaWttwNbRQI4i/NR7esT6uvpXNqruu09aSbfUsLqQ6Wt9c/2b0bvZucXoDGn40vOKu0Gn1lQ2nJ+6+l4AXacyKPE7mD2pmM0CfcDYMpPB7jn4w0UldgZ3yWaDg2Ca+R9VKdODUPoFB1M43A/GoX5uTocEREREXFTlSrQf/75Z6xWK/PmzePpp5/mtddeK/P4r7/+yn333UdGRvkzPIvUdg6Hg9xTFZvIDGB9cjZ7U83OIvTXQ5nOtcDfW3OY4R9t4FSxnSm3twLg+e9KJ09be7pLOpR26d5+Ig/AuURXeXIKi8kqLKZpHX8AhrWry9C2dZ2PT+gSTYivp3Npr+pmc5QdNx7m51XudpGB3kQEeLHy9Brmuac/j0AfE0sevIFHejSi2Obg+z2pAGW6ywPcd0M0j/ZsXC1DAmpaZKA3790Zz/wJnWtda7+IiIiIuI9KdXHfvHkzPXv2BKB9+/bs2rWrzOMeHh7MmjWLESNGVHifRqOBkBD3aHkyGj3cJla5/HxtOJzF3R9tAGDZYz1ocrorc3ld0gGmfp/Ax78dBmDXyzez50Qud320gTd+OchbI+OZu/U4fl4mHuoVy8guMXy8/iiJ6WaCfEwcyTnFwVwLK/alEXu65bVbk3B+O5RJcLBvucXpvuzSoj4uNtT5vt4a3Z4pI+zsPZlHi4YhhAd6k2ux1chxWuKRXeZ2y+hQgnzLXz5sQrdGvPnjfn45nM3ShHQAoiMDqR/iS47NwburDzNp2X4ARnVqyJjO0ZiMBto1DMFms1d77FJzdF50P8qZ+1HO3I9y5n6UM/dzteWsUgW62WwmIODs+Euj0UhJSQkmU+nuunfvftn7tNkc5OQUXnrDWiAkxM9tYpXLy9dH65J5b83ZIRv/t+IAf7+5BfvTCrjnsy082C2Wjg2D6RQdApQud3amOO/bvA6nzEU0DvJmaLu6fLU9hR92/wjAoz0aM7J9PXJzT/HmkFYUWm3sOZnPqz8dYPwnG7GUlBajXkYD7eoG8NuhTL7eeIT+LSLOi3H3kdICPcLbeN77ig3wIienkBZ1/Jiz+TjLd55wxlpdElNynX/3bhqO3VJMjqX81vo72kQxd8MRJi3eS76ltAXdUFxCTk4hYaayFx/GxtcjOqC00LfZ7PqOuRmdF92PcuZ+lDP3o5y5H+XM/bhLziIiKjY3VKW6uAcEBFBQcHbNX7vd7izORdxVQmq+szhfeP/1jIivx7e7Uun69q/c81npkmbv/5bMQ/N3UFRcuqTWsZwiAF68qTmTBl7n3Ffw71qUzx2X3DDElxaRAdzeJoqGIT7O4hwgOtSXoe1Ku6tvO55XZh8peUW8+uN+dpzIw+hhICLgwuuC33tDDJEBXvzvfw+WWeasOpwZew7wRJ8mF93W5GFgWLt6zuK8c3SwczK5cH8vOjY8O748yEfnEBERERG5tlWqQO/YsSOrVq0CYNu2bbRo0eISzxCp3XJOFfPikgQAXhjQjIYhvtzdqSEAtnLq2ycW7sJsKWHsp5sBaFMvsMzY43s6NeSRHo0Y17kh9YK8ad8g6Lx9mIwePNO3WZn7fD2NRAR40yUmhB8T0iiwllBotXEs5xRv/3KIb3aeZMmeNKICvDB6XHhsdoivJw/3aMT+9AJ2/K7QP9exnFP8ZcHOyxpvfyyniJHx9dj4dC8ahvhecvsw/7MXK+45Z2Z6g8HAzNHxxNcv/WwuNu5eRERERORaUKlfxDfddBNr1qxhzJgxOBwOpkyZwqxZs4iJiaF///7VHaNItcg9Vcy0VYfYc9KMp9HAu6PinEXhrPVHOJJ9ivdHx9PhdKtudKgvsaG+JGeXzkJu9DDgY/KgwGpj89Fc+v77N+e+m52esO2MQB8T994Qg8Ph4C+9LjzRWfcmYTzSoxFN6/izbG8a47uUFrAPdW/EfV9so8/0s68x4Jzu7gOuO7/r+++1jCztRpNZaL3gNhO/38f2E3lsPpZLv+Z1LrnPxIwC8i0lFSrMzwj3PzuJXMvIgPMe/79RcaTmWy56wUFERERE5FpQqQLdw8ODSZMmlbmvadOm5223YsWKykUlUo0OpJkpyC9i1Cel62o3DvcjMbWA7SfymLvlOKesNufs6R0all3Sa8rtrUjKLKRDw2B8PD0I8vFk27Fc5mw5zn8PZNCnWTivD2l9wQK8IjOQ33tDDAC9zlkru13981vc1x4u7VoeFejN/TfGXnK/gae7jL+35nC5Y9kB9qbmA5BfwRnfNx/JAaBP84qv693+nGXSzi3Wz/A2eTiXZBMRERERuZapT6lc1Y5mn+KOjzc6b7eI8Oe9O+MZOHMdT3xddvWBB7ueX/S2iAygxe9afds3DMbDw8CJ3CL+2q8ZHjW0DNhn93R0jn0HKLDaeOeOtnRrFFqhwv/MmO7Dp9ch/z2b3YH1dP/9tPyzreyz1h/heE4RT/dr6hwvfkaa2YKX0UD9IJ8Kvw9fTyNfTujsHIcuIiIiIiLlU4EuVx2zpYTfkrLYdjyPL7edcN7/RO8m3N25dFz5s/2b8fYvBymw2PDzMvJoz8aMbF+/wq8RVz+Iz8Z1rPbYz3VdVAAL7u3M1ztSmLflOH/q3ojujcMq/Hwf08WnmMg6p+t7qtnC35fsZcINMby7+jAA+ZYSXh/SusxzUvMtRAZ6X/ba5I3Cr56lL0REREREaooKdLnqvPbzAZadXnMboG39IF6+uUWZInFI27oMaVvXFeFdltgwP57s05Qn+5w/hORSLlVEp+ZbnH+vTMwk51Sx83NrWy+QlYkZmC0lznH665OzWZaQXmbmdRERERERqT6VmsVdpLbJLLCSklfE+uRsViZmcl1kAB+Pbc/X913Pwoe7XbMtuBO6RGM8XacXWm1c/z+r+G7XSaBsgW4psTn/9vM0ck/nhtgccDDj7HKKjy7YCUBEwPnjyEVEREREpOrUgi5uzeFwsCfVzEPztlN0ej3xxmF+vD6kFQ2CNfGYt8kDmwNKbHZnsf3x+iMMblvXWaDHhPpyJPvsOPXH+zSh8ekLGkdzTnE4q5DbWkc5Hw/yKbvGu4iIiIiIVA8V6OK2ck8V8+TCXexMyS9z/9t3tFFxfpr36XHoRSV2Ek8X6I7T67qn5lvwNnlQP9jHWaB/88D1NAj2pai4tEX9lR/2A+DndfZUcamx7SIiIiIiUjkq0MVt/bQvnZ0p+dzZvj4tIv3p3iScvSfzVZyfw9tUOgu7pcROYnppgX48t4iiYhtp+RaiAr25MTaUdYezAfD3LD0l+Pxu9vZzR7P7eKpAFxERERGpCSrQpdYqsJawbG8a/l4mbmkV6bw/ITUfuwMOZxXi72XkmX5NnROi9Wxa8fW5rwVnWrstJXYOnDOevMBqY9vxPBqF+TIivh7vrDwEgJ+Xsdz9FBbb8PcyUmC1cVubqHK3ERERERGRqlGBLrXWF5uPM/O3ZAB2n8znqb5NWXs4i8e+Ort+efMI/8te8uta4u9dWnCn5BWRmF6Al9GA1ebg378mkVFgJSLACx9PIy/e1Jyf9qXjaTz7WX44Jp7MAivPfbeXAquNMD9PejQJUw8FEREREZEaogJdaq2NR3Kcf3+x5Tg2u4P556xrDnD0nMnN5Hw3xIYC8M3Ok+RbSujeOIw1SVms2J8BQFFx6cR6w+LqMSyuXpnnxjcIpsRW+vj//vcgYX6e+HvplCEiIiIiUlP0a1tqJYfDwcGMAga3iSLfUsIviZllivM/92jE7pP5DGqt7tYXE+BtwtfTg/1pZgCui/RnTVIWhacngTMZL977wGQ8O968wGq7YBd4ERERERGpOhXobmjTkRx8PD1oWy/I1aHUiE1HcliXnE1uUQlx9YO4vW1dZqw+zKcbj+LvZeT7h27Ex+Shru0VFOzjSZq5dEm1YN+yS6TZ7I5LPj++fhDbT+RhKbHjrwJdRERERKTGqEB3Eza7g4Q0M6G+njz85Q7n/U/0bsLA1pGE+Xm5MLrqcyznVJn3d3PLSEweBu69IZqtx3J4vHcTfD1VJF6OEF9PTp5e8zzkdwV6fINLX+Tp07wO20/kAReeRE5ERERERKpOBXot53A4KLDaeOTLHexNNZ/3+DsrD7E+OZtpI9qx/Xgu7eoH4eGGLcsZBVagdKzzGdfHhDgLwgBvEx/f1cElsbm7YN+zX/N6QT7Ov6MCvXmmb7NLPj/knOerBV1EREREpOaoQK/FEtMLuP+Lbc7xwmfcEVePr3ekOG8fyznF9uO5PDB3O4/0aMS9N8Rc6VArLb+ohBcW72F9cumEcB4GeKZvU0Z3bODiyK4ewT5nW82jQ33x8zRSWGyjY8NgvEyXXtP83KJek8SJiIiIiNQc/dqupRwOB//7y0EKi23UD/LmuQHN6RIbitEABoOB5OxCNh/NpWVkAAlpZtLMpS3QG5Kz3apA/35vqrM4B+gcHaLivJqdO+7cx+TBH7o05L01yTQO96vQ86NDzi6rpi7uIiIiIiI1RwV6LVJiszPlpwOM6diA7FPFbDySw1N9mzK2nIL1X3e0o9hm54O1ySSkmfnb4r0AFJ5eNqs2czgczFp/lBlrDgPQKMyX5wc0Z+pPB3isdxPXBncVCvY5+zX3MXlw/42xDGodVeF5CyICzm6nLu4iIiIiIjVHBbqLZZgtnMizsP14LpkFxXy3O5X1ydnE1Q/C02hgxO/Wpj7D2+SBt8mDQa2jmLP5uPP+mhh9bi2x4+FhwMMA87eewNNoIMDLxC2tIi97X/tSzTz73R5O5BYBEOBtZOKt19GmXhAL7ru+ukMXSsfvn2H0KD1Czu22finnzpavLu4iIiIiIjWn0r+27XY7EydOZN++fXh5eTF58mRiY2Odj8+fP5+5c+diMpl4+OGH6du3b7UEfDVxOBw8s2gPu0/ml7k/zWzl5/0ZFRojfG7rKIClpHpb0B0OB7e8t5aODUMY3CaK/zlnErebWkZUaEK6QquND9cm89O+dE7mW4gM8CK+fhBZhVb+d3hbGoVVrKu1VI6P59ljqKpL0/l7qwVdRERERKSmVLpA//nnn7FarcybN49t27bx2muvMWPGDADS09OZPXs2X331FRaLhbvuuovu3bvj5XV1LAW2dGcKj8/fDsCz/Zsxqn39y97H6kOZ/H1JAgVWGx1OL3W19Xiec83p3k3DmTSo5SX3E3hOgR7m50liRgFvLE/k2f6Xnp27IhLSzJgtNlYdzGTVwcwyj53ILaLhOeOTL+T15QdYuift7O0hra/aNdxrI+/TF3kaBFe81fz3gnxM5BWVaIk7EREREZEaVOkCffPmzfTs2ROA9u3bs2vXLudjO3bsoEOHDnh5eeHl5UVMTAwJCQnExcVVPWIXyyywOotzgDeWJ/LzvnQGtopk2AW6o5fnq+0pGD0MvHRLCwa1jsLoYaCo2IbRw8AviZn0bBKGTwWKIb9ztnn1tla8vzaZXxIzqq1AX7I7tczt3k3DGdW+Po9+tZN0s/WSBfqS3allivNP7+lAq6jAaolNKsbHVHqMNI/wr/Q+Zt4Zz5I9qef12BARERERkepT6V/bZrOZgIAA522j0UhJSQkmkwmz2Uxg4NkizN/fH7P5/DW8z2U0GggJqf1dnc0O6BAdwujODXl+YelFiS3Hckk1W5jQq+kFn7c/NZ/dJ/IY0CqSORuOsvpQFuNuiGFcj/MnRRsVHlDOHi6tV+u6JGQW8q8Vifj6e5NbVEyJzUH9CrRyn2vuxqMs23OSl25rzZbjeTQI8eF4ThGdY0P5cML17E8t7ZJ/ygErD+cQHx1Mo/DS4i81r4h7/7OJfw5pQ5i/F2+sSOT6RqH8Z8L1FFhKCKngxGTVyWj0cItjq6bc2r4BC3ak8OzAVpX+HDqH+NG5eUQ1R3Zh13rO3JFy5n6UM/ejnLkf5cz9KGfu52rLWaUL9ICAAAoKCpy37XY7JpOp3McKCgrKFOzlsdkc5OQUVjacKybAAPMfvJGcnEI+HBPP7pP57ErJZ/n+dNIyzOWOGS8qtjH2w/XkFZUQ6G3iVLGNHk3CeKRbbLW85zPdjwvNRcQEeQPw864Unvi69ALCxqd7Obd1OBxMWrafxbtT6dEkjFdva3Xe0lkfr04iKauQxVuPcTDdzL03xNCraTgNQ3zIySnE2146zv3rzcdYeTCThiE+LLy/CwAr9qZyIM3MMwu24+tpxNPDwMs3t6Agv3RSuBxrSZXf7+UKCfFzi2OrJr03qrT3irt8DsqZ+1HO3I9y5n6UM/ejnLkf5cz9uEvOIiIq1ov44jOQXUTHjh1ZtWoVANu2baNFixbOx+Li4ti8eTMWi4X8/HwOHjxY5vGrRXyDYO7q1JA+zcKxOyA5u/wD48eEdPKKSgvTfEsJJXYHz/VvhsmjeuZc/+q+61n84A0AxNUvHdt9NPtUudtmnypm8elu66sPZfHPZfvLPL5wRwpJWaXvY/qqJOwOqB/kQ+u6gQT5lK6nHezryYAWdVh5ekz68Zwi5/P3njRj9DBwLKeII9mneGVQS6ICvavlfYqIiIiIiFzNKt2CftNNN7FmzRrGjBmDw+FgypQpzJo1i5iYGPr378+4ceO46667cDgcPPnkk3h7X71FWoPTXchP5FpoHlG2e7rD4eDLbSdoEu7H8Lh6/M9/DzLl9lbUvYxlri4lxNfT+feZdaoLrTbnfTPWHOZIViFTB7cmu7AYgPtvjOGjdUf4eX86o47Vo2PDEIqKbUz56cDZ2E//G1TOuOP7boxh1cFMrDYHDuDTDUfxNnmwMyWP1lGBPN67MfWDfYgIuHrzLiIiIiIiUp0qXaB7eHgwadKkMvc1bXp2DPadd97JnXfeWfnI3EhUQOm46mcW7aZxuB9f/KETdocDT6MHe07mk5Bm5tn+zRgZX4+h7erW6EzYnsbSThGbjuY47/t43REApgI5p0oL9I4Ng/Ht2Zh//5rEn+btYObovRbefAAAIABJREFUOAynV1G/I64eX+9IcT6/Qcj5FxOaRwSw/M/dMFttvPrjfqb/muR8bGCrSOIbBFf7exMREREREbmaVbqLu5wV5u+Fp7G0uE3KLOTJhbu4beZ6SuwO5m87gZ+nkYGtIjEYDFdsmaqNR3LOuy8lr4iH5u8AINTPk/Fdovnp4a4A7DqRz66UPAAe6h7LgBZ1CPPz5PXBrc7rFXCGj6eROv5e3NIyssz9l1q7XURERERERM6nNZOqgYfBwNf3Xc+6w9m8+tMB1h7OBkrHgS/bm8bwuHoEeLv+o162t3S5s7EdG9Dk9KzrIX6ehPp6kpBmJtzfCz9PI6F+Xkwd3LrC+232u+W7HA7HBbYUERERERGRC3F91XiVqBvkQ6u6ZWfmO5RZgM0BberVjnW/96cXEOxj4qm+ZZeD69EkjO92p9K0jh8hfp4XePaFNavjz7KHbyTU15Ovd6TQr3md6gpZRERERETkmqECvRrVCyo7Idpnm46dvr/6JoSriGf6NsUOxNcP4tONR1m+PwOAk3lFBJYz4duD3WL5bncqBzMKaRxWuTUEw06vbz4ivn6l4xYREREREbmWabBwNQr8XTf2XSn5ALSpe2Vb0Ed3bMDYjg1oXTeQlpFnx4/vTMk/L0Yobf1/8abmwNll2kREREREROTKUgt6NTIYDLx8awvsDpzri3dvHIbPFZoYrjx3dWpIid3BJxuOYimxk5pvKXe7YXH1GNy2LobqWZpdRERERERELpNa0KvZ7W3qOsdgXx8TwpN9mrg0Hi+TBw90jeWZ0+POGwRfuLu90cOAhyp0ERERERERl1ALeg0I8Dax8elerg6jjGFx9bixUSj+Xkq5iIiIiIhIbaRq7RpS9wpPViciIiIiIiIVpy7uIiIiIiIiIrWACnQRERERERGRWkAFuoiIiIiIiEgtoAJdREREREREpBZQgS4iIiIiIiJSCxgcDofD1UGIiIiIiIiIXOvUgi4iIiIiIiJSC6hAFxEREREREakFVKCLiIiIiIiI1AIq0EVERERERERqARXoIiIiIiIiIrWACnQRERERERGRWkAFuoiIiIiIXFO00rTUVirQRURErnJ2u93VIUgFFRUVYbFYXB2GXAZ9v9xPXl4e2dnZrg5DpFzGiRMnTnR1ELXNl19+yYEDB6hTpw5+fn6uDkcuwOFwYDAY2LBhA4mJiTRq1Mh5n9Run376KatXr8ZmsxEdHe3qcKQCzpwXg4ODCQwMdHU4UgHLly/nyy+/pEePHjo3uonZs2czZ84cmjRpQp06dVwdjlTA+++/z4oVKwgICKBu3bquDkcqYMGCBTz99NPUqVOH1q1buzocuYT58+ezd+9e6tSpg7+/v6vDuSLUgn6OvLw8/vjHP7J9+3aSkpL497//zdatWwFdHa2NzvzYnDNnDqtWrSIvL08/QGu53NxcHnjgAecFlZkzZ7J582ZXhyUXkZ+fz/3338+WLVtISkpi9uzZpKenuzosqYBdu3Yxd+5ckpKS8PDwoKSkxNUhyQWkpqbSv39/MjMzmThxIi1btnQ+pm64tVNhYSHPPfcc2dnZ3HTTTeTl5TkfU85qp/Xr1/Pggw+yY8cOAgICiI+Pd3VIchHZ2dlMmDCBrVu3cvDgQT7++GNSUlJcHdYVoQL9HBaLhdjYWCZPnsxjjz1Gu3btmDlzJgAeHvqoaqPvv/+e5ORkDAYDP/zwg6vDkUtIT08nNjaWSZMmcdttt9G2bVu8vb1dHZZcRHZ2Ng0aNGDq1KmMGTOGjIwMgoODXR2WXMSZC8qBgYHcfvvtnOkoZzKZXBiVXExoaCidOnWiffv2zJw5k8mTJ/P5558D6MJzLWWz2QgODmb48OEsXryY9evXs2jRIkA5q6327NnDvffey6RJkxg0aJAuNtdyeXl5xMbGMnXqVB555BGys7OJiIhwdVhXxDVbdZ65uvnFF184T6jHjx/n8OHDFBUVYTQaufXWW/H392fx4sVlniNXXnn5AmjdujUvvPAC3bp1IzExkUOHDpXZXlynvJzl5OSUuWK9du1avLy8ymwvrlNeznJzc+nfvz9QOjThl19+Yfr06Xz55ZeAehe52oXOjXl5eWzbto1//vOfpKen89hjj7F+/XpXhSnnKC9nBQUFxMTE8P777xMTE8OQIUNYsWIFH374IaDvmatd6DfjkSNHWLt2LW3atKF///4sXryYTz75BFDOXO3cnC1cuBCA8ePH07VrV6xWKxs2bHBebNbvD9cr7zuWl5eHr68vM2fO5J133mHt2rW89957fPPNN8DV/R27Zgv0M1c3165dy/vvv4/dbqd9+/YYDAa++OILAHx9fenevTsnTpzQ+D0XKy9fAFFRUXTp0oUWLVoQHBzMypUry2wvrnNuzmbOnIndbqdz584MGTIEgI0bN+Ln50eLFi2Aq/tE6y7K+561a9eO3r17A9CnTx+WLl1Kly5dWLhwIVarVb2LXKy8nHl4eJCdnU2rVq1Yvnw5JpOJjRs3cv311wP6Mepq5eUsNDSUFi1aMHLkSO68807i4uL4y1/+wrZt2yguLtb3zMXKy1nLli3x9fVl6dKl9O7dmw4dOvCnP/2JTZs2KWe1wLk5+/DDD53nRqvVipeXFx07duTXX38ts624zoV+f9x1113s3buX3NxcvvnmG+Li4vj888+xWCxX9Xfs6n1nF3Bud5aNGzcSGhpK3bp1mTRpEgCPP/4433zzDfv378fDw4OUlBRCQkL05XWRC+VrypQpAHh6egLQsGFD2rZtS1JSEmvXrnVJrFKqvJzVq1fPmTObzQZAcnIy48aNIyEhgccff5zVq1e7JF659PfszMWT+Ph46tSpg8VioVu3bs7eD3LlXShnr776KlA6d8Ann3zC8uXL+eSTT2jWrBnTp08H9GPUVS6Us8mTJwPQo0cPBg8ejNlsBuDQoUN06tTJ+f+cXHmX+s340EMPYbFYOHDgAACHDx+mdevWypkLXer/M6PRCECTJk3w9/fn1KlTLolTSl3q/7KQkBDMZjP33XcfYWFhlJSU0LVr16t+eKTBcY1cSj958iTTp08nMzOTfv360atXLzw9PcnJyaF+/frcdNNNfP7558TGxvLBBx+QmprK0aNHKS4u5vHHH9dEEldYRfL1xRdfEB0dTUlJCSaTifT0dFauXEmHDh1o2rSpq9/CNedycuZwOHj44YdJSkoiNjaWu+++29lKK1fO5eRs+fLlrFu3jpSUFCwWC/fddx9du3Z19Vu45lQkZ3PmzCEmJoY9e/Y4ZyhOSkri2LFj9OzZ08Xv4NpzOd+zJUuWsHz5cgoLC7HZbPzpT3+ic+fOrn4L15zL+c346aefkpiYyPHjx7FarTz66KPccMMNrn4L15zL+Z4BrFy5knnz5jF58mTCwsJcHP2153K+Yy+99BIGg4G8vDzy8vK4//776datm6vfQo26Zgr0d999l+LiYu644w4WLVpEdnY2Tz31lHO6/nfeeYeEhATee+89SkpKKCoqYtOmTfTp08e1gV+jKpKv/fv38+6772r4QS1RkZzt27ePGTNmUFRUxIQJExg2bBhjxoxxceTXrss5L1qtVjIyMtixYwe33nqriyO/dl0qZ2+//Tb79+9nxowZzuecuYgprnG537Mz8wcMGDDAxZFfuy4nZzabDavVyrZt23TR0oUu53fjGStWrKBfv36uCvmaVpF87d27l5kzZ2K1WklLS2Pnzp0MHDjQxZFfGVd1gf7VV1+xYcMGoqOjOX78OI888gjR0dEkJyczb948oqKiGD9+vHP7Ll268Prrr9O3b18XRn3tqky+3njjDV1EcaHK5GzKlCkMGDDAOQ5MriydF92Pzo3uR98z96OcuR+dG92LvmMVZ5x4Zv2Vq8xbb73Fzp07ue+++1i2bBlLlizBy8uL7t274+vri9FoZPfu3bRr1w4fHx8AWrVqRYMGDdTVxQUqm6/69esrXy5Sle9YeHi4cxyYXDk6L7ofnRvdj75n7kc5cz86N7oXfccuz1Xb5y0/P5/Ro0fTpk0b7r77biIjI1m8eDG33347rVq1Ijw8HIvFgp+fn7OLdI8ePVwd9jVL+XI/ypn7Uc7cj3LmfpQz96OcuR/lzL0oX5fnqpzF3W63c/PNNxMXFwfA0qVL6dWrF4888givvvoqSUlJ/Pbbb+Tk5GC32zV+2cWUL/ejnLkf5cz9KGfuRzlzP8qZ+1HO3Ivydfmu6jHoAGazmQkTJjBjxgwiIiKYMWMGubm5ZGRk8NxzzxEREeHqEOUcypf7Uc7cj3LmfpQz96OcuR/lzP0oZ+5F+aqYq7aL+xmpqal069aN/Px8Jk+eTPPmzXn66ae1RmUtpXy5H+XM/Shn7kc5cz/KmftRztyPcuZelK+KueoL9I0bN/L++++ze/duhg4dypAhQ1wdklyE8uV+lDP3o5y5H+XM/Shn7kc5cz/KmXtRvirmqu/i/tVXX5Gens59992nJZ3cgPLlfpQz96OcuR/lzP0oZ+5HOXM/ypl7Ub4q5qov0M/MBCjuQflyP8qZ+1HO3I9y5n6UM/ejnLkf5cy9KF8Vc9UX6CIiIiIiIiLu4KpcZk1ERERERETE3ahAFxEREREREakFVKCLiIiIiIiI1AIq0EVERERERERqARXoIiIiIiIiIrWACnQRERERERGRWkAFuoiIiIiIiEgtoAJdREREREREpBZQgS4iIiIiIiJSC6hAFxEREREREakFVKCLiMg1x2azMWvWLO644w6GDh3KoEGDePPNN7FarTX+2seOHaNDhw6X3O7LL7/k888/B+CLL77g/fffr/JrJyYmMnToUIYOHUqfPn3o1KmT8/Ynn3xS5f1Xtx07dvDSSy8BsHPnTh577DEXRyQiIlKzTK4OQERE5EqbOHEiubm5/Oc//yEwMJDCwkKeeeYZXnzxRd58801XhwfA5s2bad68OQBjx46tln02a9aMRYsWAfD111+zbNkyZs6cWS37rgmJiYmkpqYC0K5dO6ZNm+biiERERGqWCnQREbmmHDt2jO+++47Vq1cTEBAAgJ+fH6+88gpbtmwB4Pnnn6d58+bcf//9593u168ft99+O+vWrSM3N5cHHniALVu2sHv3bkwmEzNmzCAqKop+/frxr3/9i3bt2gE4b4eGhjpjycjI4KWXXiIzM5P09HQaNGjAO++8w5YtW1ixYgVr1qzBx8eHrKwssrOz6devH6+//jrfffcdAHl5efTv35+ff/6ZoqIiJk2aREpKCsXFxdx222089NBDl/XZTJ8+nW3btpGWlsZ1113H888/X2584eHh9OvXj+HDh7N27VpSUlIYOnQoTzzxBAUFBbzwwgskJyfj4eFBmzZtmDRpEgBTpkxh+/btFBQU4HA4mDx5Mp06daKgoIDJkyezZcsWjEYjAwYMYOzYsUybNo38/HxeeOEFhg0bxj//+U8WL15Mfn4+r7zyCgkJCRgMBnr27MlTTz2FyWSiXbt2PPjgg6xZs4a0tDQeeOAB7rrrLtLT03nuuefIzs4GoHfv3jzxxBNVOJJERESqn7q4i4jINWX37t00a9bMWZyfERERwS233FKhfVgsFubPn8/jjz/OSy+9xPjx4/n222+pV68eCxcurHAsS5YsoX379sybN4/ly5fj4+PDokWLuOmmm+jXrx8TJkzg7rvvdm7fvXt3CgoK2LlzJwCLFy+md+/eBAcH89e//pURI0bw9ddfs2DBAn777TeWLl1a4VjOOH78OAsXLuStt966YHxnFBYWMmfOHObOncvHH3/M0aNH+emnnygoKGDRokUsWLAAgKNHj7J9+3bS0tKYN28eS5cuZfjw4XzwwQcATJs2DYvFwtKlS/nmm2/YsmULR44c4bHHHqNz585MnTq1TIyTJ08mJCSE7777jq+++op9+/bx8ccfA2C1WgkNDWXu3LlMmzaNqVOnOvPVsGFDFi5cyOeff05ycjL5+fmX/fmIiIjUpFrXgr59+3beeustZs+eXe7jq1atcv6H7nA42Lx5M4sXL6Zp06ZXMkwREXFTHh4e2O32Ku3j5ptvBiA6Opo6derQsmVLAGJiYsjNza3wfsaPH8+mTZuYNWsWhw8f5sCBA8THx19we4PBwIgRI1i4cCHt2rXj66+/5tlnn6WwsJCNGzeSm5vLv/71L6C0eE5ISGDQoEGX9d7at2+PyWSqUHz9+/cHICoqivDwcHJzc+nUqRNvv/0248aNo1u3bowfP57Y2FhiY2MJDg5m7ty5HD16lPXr1+Pv7w/Ab7/9xgsvvIDRaMRoNPLZZ58Bpd3wy7Nq1Sq++OILDAYDXl5ejBkzhv/85z88+OCDZeJq06YNVquVwsJCevbsyYMPPkhKSgrdunXj6aefJjAw8LI+GxERkZpWqwr0Dz74gG+//RZfX98LbtOrVy969eoFwIcffkjHjh1VnIuISIXFxcVx6NAhzGZzmVb01NRU/vGPfzBt2jQMBgMOh8P5WHFxcZl9eHl5Of/29PS84Gudu4/yJqB788032bFjByNGjOCGG26gpKSkzHPKM3LkSIYPH86oUaPIz8+nS5cumM1mHA4Hc+fOdf4fmpWVhbe390X3VR4/P78Kx3fu/s98ZtHR0fz000+sX7+edevWce+99zJp0iQ8PDx49dVXuffee+nfvz9NmjTh22+/BcBkMmEwGJz7SklJwcfH54Ix2u32Mtvb7XZKSkrOi+vMNg6Hg7i4OJYvX87atWtZt24do0aN4oMPPqBt27aX/RmJiIjUlFrVxT0mJobp06c7b+/bt49x48Yxbtw4/vKXv5Tpinby5EkWLVrEo48+6opQRUTETUVFRTF48GD+9re/YTabATCbzUycOJGQkBB8fHwIDQ1l165dQGnhvmHDhst+nbCwMOc+1q9fT3p6+nnbrF69mvHjxzNs2DDCw8P57bffsNlsABiNxjJF57nxx8XF8dJLLzFy5EgAAgICaN++PbNmzQJKx6aPHTuW5cuXX3bcFY3vQubMmcMLL7xAjx49+Otf/0qPHj3Ys2cPa9asoW/fvtx11120bduWn3/+2bmvrl27snDhQux2O1arlccee4yNGzde8DPo0aMHn332GQ6HA6vVyvz58+nWrdtF43rrrbd49913GTBgAC+++CLNmjXjwIEDlf9wREREakCtKtBvueUWZ7c6gH/84x+8/PLLzJ49m169evHhhx86H5s1axYTJkwo04ohIiJSES+//DLNmjVjzJgxDB06lFGjRtGsWTMmT54MwLhx40hPT+eWW27hb3/7GzfeeONlv8YzzzzDp59+ytChQ1m0aBFt2rQ5b5s///nPvPHGGwwePJiHH36Yjh07cuTIEaC0x9jcuXPLnWV91KhR7N27l+HDhzvve+utt9i+fTuDBw9m1KhR3H777QwZMuSy465ofBcybNgwbDYbgwYN4o477iA/P59x48YxZswYNmzYwODBgxk+fDjR0dEcO3YMu93Oo48+iqenJ0OHDmXYsGH07t2bm2++mfbt23P06NHzLsb//e9/Jysri8GDBzN48GAaN258yQnxxo8fT0JCArfffjsjRoygYcOG3HbbbVX6fERERKqbwXGpvnRX2LFjx3jqqaeYP38+nTp1onXr1kBp98LGjRszdepU7HY7AwcOZNGiRRftAiciIiIiIiLiLmrVGPTfa9y4Ma+//jr169dn8+bNzu6B+/fvp3HjxirORURERERE5KpRqwv0iRMn8txzzznHqL366qsAJCUlER0d7crQRERERERERKpVreviLiIiIiIiInItqlWTxImIiIiIiIhcq2pNF3e73Y7NdvmN+UajoVLPEzlDx5BUlY4hqSodQ1JVOoakqnQMSVXpGLo4T09jhbarNQW6zeYgJ6fwsp8XEuJXqeeJnKFjSKpKx5BUlY4hqSodQ1JVOoakqnQMXVxERGCFtlMXdxEREREREZFaQAW6iIiIiIiISC2gAl1ERERERESkFlCBLiIiIiIiIlILqEAXEXEzDoeDErtmSRURERG52qhAFxFxI4ezCnlg7naGfbiBxPQCV4cjIiIiItVIBbqIiBuw2R3M3niUuz/dzOGsQmx2B3+ct43NR3NcHZqIiIiIVBMV6CIitVxSZiF/nLuNaauS6NY4jHkTOjPrrvZE+Hvzl6928vO+dFeHKCIiIiLVwOTqAEREpHwldgdzNh1j5m+H8fU0MnlQS25uGYHBYADggzHxPP3Nbv62eC+ZBVZGd2zg4ohFREREpCpUoIuI1EKHMguY9MN+dp/Mp2/zOjzXvxnh/l5ltgn29eTfI9vxj6UJvPXfg6SZrTzas5GzgBcRERER91KlLu7bt29n3Lhx592/YsUKRowYwejRo5k/f35VXkJE5JpSYnfwyfoj3DN7C8dyTvHqbS15fXCr84rzM3w8jbw2uDUj4uvx6cajTPxhHyU2+xWOWkRERESqQ6Vb0D/44AO+/fZbfH19y9xfXFzM1KlTWbBgAb6+vowdO5a+ffsSERFR5WBFRK5miRkFTPphH3tTzfRvUYdn+zcjzK/8wvxcRg8Dz/VvRmSANzPWHCaroJjXhrTC30udpERERETcSaV/vcXExDB9+nSeffbZMvcfPHiQmJgYgoODAejUqRObNm1i4MCBF92f0WggJMTvsuMwGj0q9TyRM3QMSVVV9Rgqsdn5YHUS0/+bSIC3iWmj2zOwbd3L3s9Tt7YkOiKAf3y7m0e/2sUH4zpRJ8C70nHJlaPzkFSVjiGpKh1DUlU6hqpHpQv0W265hWPHjp13v9lsJjAw0Hnb398fs9l8yf3ZbA5ycgovO46QEL9KPU/kDB1DUlVVOYYS0wuYtKy01XxAiwie7d+UUD+vSu/vpqZh+A5tzfPf7WXke2uZPqId0aG+l36iuJTOQ1JVOoakqnQMSVXpGLq4iIjAS29EDSyzFhAQQEFBgfN2QUFBmYJdRERKW80/WpfMuM+2cDLPwmuDWzF1cCtCK9Cl/VJ6NAnnvTvjMFtKuP+Lbew5mV8NEYuIiIhITav2Ar1p06YkJyeTk5OD1Wpl06ZNdOjQobpfRkTEbR1INzNhzjbeW5NMv+Z1mD+hM/1bVO88HW3rBfHR2Pb4enrw0Pzt/JaUVa37FxEREZHqV20F+nfffce8efPw9PTk+eef5/7772fMmDGMGDGCqKio6noZEXETK/an8z//PUhRsc3VodQaJTY7H6xNZtxnW0k3W3h9SGtevb0VIX6eNfJ6sWF+fHRXB2JC/Xjqm90s3n2yRl5HRKDQauNQZgFrD2eRbra4OhwREXFTBofD4XB1EADFxTaNQReX0DFUvUpsdqb/msSczccB6BwdzP8Ma4ufl9HFkdWcihxD+9LMTPphH/vTC7ilZQTP9GtGiG/NFOa/Z7aU8Ny3e9hwJIdHejRiQpdorZVey+g8VLvZ7A4yCqyczCsiNd/CyTwLJ/MtnMwr4mS+hdR8C3lFJc7tg31MvD28Le3qB12xGHUMSVXpGJKq0jF0cRUdg641eESk2mQWWPnb4r1sOZbLne3r06puAJOX7efRBTuZNqItAd7X3imn2GZn1vojfLz+KME+Jt4c0po+zetc0RgCvE28c0dbXvlhH++uPky62crTfZti9FCRLgJQYC1xFt2pp4vuc2+nmq3Y7GXbMwK9TdQN8iYq0Jv4+kHUDfKhbqA3gT4m3lyRyCNf7uCNoa3p2ijMRe9KRETc0bX3a1lEasTOE3k8990e8opKeGXgdQxqXTq0xc/LxIuL9/LIlzuYNqLdFWs1rg32pZp5Zdk+DqQXMLBVJE/3bUqwi96/p9GDSYNaEhHgzWebjpFZYGXSoJZ4m6p9KhKRWqXE7iDDbLlgy/fJPAv5lpIyzzEaIDLQm7pBPsQ3CKZukDd1A72JOl2ERwV6X/SC4wdj2vP4Vzt5auFuJt56Hbe0iqzptykiIlcJdXGXa56OoapxOBws2J7C//73IFGB3rwxpDUtIgPKbLPmUBbPfrub6FBf/j0yjjr+VZ+pvDb5/TFUbLPz0bojfLL+CCF+XrwwoDm9m4W7MMKy5mw+xtu/HKJDw2DeGtqaIJ9r56JJbaXzUPXLKSzm5R8SWH84G9vvfukE+ZiICiwtus+0fJ9pDa8b5EMdf68q9zAxW0p4+pvdbD2WyzP9mnJnhwZV2t+l6BiSqtIxJFWlY+jiKtrFXQW6XPN0DFVeUbGN134+wJI9aXRvHMakQdddsNjbkJzN09/sJjLQm3dHxREV6H2Fo6055x5De1PzmfTDfhIzCritdSRP9nFdq/nF/JiQxsvf7yMm1JdpI9pdVflwRzoPVa99aWb+umg3mQVW7uzQgOhQ3zJFuL/XlelAaCmx8/cle/klMZMHbozhwW6xNTb/g44hqSodQ1JVOoYuTgW6SAXpGKqc47mneHbRHg6kF/DHrrHc3zUGj0v88Nx+PJfHv95FsI+Jd++Mo0Gw7xWKtmaFhPiRlmHmo3XJ/GfDUcL8S1vNezatPa3m5dl4JJu/LtqDv5eRaSPa0bSOv6tDumbpPFR9fkxIY9Ky/aVzPgxtQ+u6FftBVFNK7A5e++kAi3adZER8Pf7ar1mNzP+gY0iqSseQVJWOoYuraIGuwYcictnWJGXxh8+2kpJn4e3hbfljt9hLFucA8Q2CeXdUHAVWGw/O3c7hrKvjJL7zeC7jPtvCx+uPMrB1FHPHd6r1xTnA9TGhvD86HrsD/jh3O9uO5bo6JJFKs9kdTF91iBeXJNAqKoBP7+no8uIcwORh4MWbmzO+SzRfbU/h70v2Yi2xuzosERGppVSgi0iF2R0OPlibzJNf7yIq0JtP7+lA9yaXN0Nx67qBvHdnPCV2B3+at53E9IIairbmFdvszFidxKj312G2lPDO8La8fOuFu/nXRi0iA/hobHvC/Dz584IdrDiQ4eqQRC5bXlFZNYD7AAAgAElEQVQxTyzcxacbjzEyvh7vjoojvBbNdWEwGHi0Z2Oe6N2En/dn8MTCXRRYSy79RBERueaoQBeRCskvKp3w6P3fkrm1VSQfj21Pw5DKdVFvFuHPzNHxmDwMPDR/O3tO5ldztDUvMaOACZ9v5eP1RxkaX5+54ztf9sWK2qJ+sA8fjm3PdZEBPP/tHr7cdsLVIYlU2MGMAsZ/vpVNR3J48abmPDegOZ7G2vnz5u7ODZl463VsOZrDw/N3kF1odXVIIiJSy9TO/8FEpFY5kG7mD59vYe3hbP7arymvDLwOH09jlfbZKMyPmaPj8fcy8siXO9h+3D26V9vsDmZvPMofPttCRoGVt4a24fU72hHo496rVob4evLuqDh6NAnjjeWJvLs6iVoyRYnIBf33QAb3ztnKqWI7M0fHMyyunqtDuqTb2kTx5tA2HMos5IG520nJK3J1SCIiUouoQBeRi/phbxr3ztlGUbGdmXfGcWeHBtU2C3HDEF9mjo4n3N+LRxfsZOOR7GrZb005lnOKh+ZvZ9qqJLo3DmPu+E61avm0qvLxNPLG0DYMa1eXWeuP8s9l+ymxaays1D52h4OZaw7z7Ld7aFrHn9n3dCCufpCrw6qwnk3D+feIdmQXFvPAF9s4mOG+Q31ERKR6qUAXkXKV2Oy8tSKRfyxNoHVUALPHdSS+QXC1v07dIB9mjo6nQYgPTy7czZqkrGp/japyOBx8syOFuz/dwoH0Aibeeh1vDGlNqF/tGeNaXUweBv52U3P+2DWG73an8vSi3Zwqtrk6LBEns6WEZ77ZzYfrjjCk7f+zd9/hUVVbG8DfKUkmvfcO6T2hS1OagkgVJDQbdj97QxRBESxYrl7lKihCUIoNkCagFOkQICG9kUoK6Zm0mcyc7w+Eq1cI6Wdm8v6ex0eTycxZwcWeWWfvvbYzvpgZCUcL/TsmMMrDGl/cEwGtADy8OQGJl2rFDomIiHQAC3Qi+odyZTMe+z4Rm89dQmyMOz6fEQGHbmy45GBujP/MjISvnRle2JqsU43KyutVeG5rMt7el4kQV0tsurcf7gx17razjHWBRCLBw7f4YOFYf5zIreJeWdIZuZUNuP+7cziWW4UXR/nhtXEBMJbr70cZf0cLrImNhLVCjie+T9TJG5RERNSz9PddjYi6RUJRDeZuOIe0UiWWTQjCc7f1hbwHGi5d3QMd7GyJV39JwZ7Usm6/5s38lnEZs745g9P51Xj+tr747O5wuFgpxA6rx0yLcMV7k0KRVV6PBZsSUFjdKHZI1IsdyanAfd+eQ01jCz67Oxwzo90M4kaZu7UpVs+KgredGZ7fmozdqaVih0RERCJigU5EAK4s4950tgiPbEmEmZEUa2dH4/Zgpx6NwVIhx7/vDkeUhzUW70rDtgvFPXr9q2qb1Hh9Vxpe+SUV7jam2DA3BrNi3Nt01ruhGelnj8/uDkdNoxoPbjyP9FKl2CFRLyMIAtaezMdzPyfDw8YU6+dGo5+njdhhdSl7c2P8Z2YEotytsHhXOjadLRI7JCIiEgkLdCJCo1qD13el4YMD2Rjqa4f1c2Pg52guSixmxjJ8PDUMg31ssWxvJjb38AfVk7lViF0Xj33pl/HwLd74KjYKPvZmPRqDrol0t8aaWVEwkknxyJYEnW/mR4ajQaXBwh2p+PxILm4PdsKaWZEGu4rFwkSOf00Lx61+9vjgQDZW8SQFIqJeSbZkyZIlYgcBAFqtgKYmdbufp1AYdeh5RFf19hwqqGrEkz9cwJn8ajw21AcvjfaDibxzR6h1llwmxegAR+RU1OO7s0UwkUsR1Q0N6v6qUa3BRwdzsPJANpwsTPDR1DDcEezUplnz3pBDNmZGGBPgiD9yKrDp7CV425mhr4M4N3EMUW/IofYqrL4yNp0rrMHTI/vgqRG+Onu+eVeRSyUYHeCIy/UqbDp7CeX1Ktzia8dxiHoEc4g6iznUOnPztjU0ZYFOvV5vzqE/sivw1E8X0KjW4L3JIZgU5qIzezplUglG+TugsLoRG88WQRAE9PO07pb4LlyqxVM/JeFEbhVm93PH8onBcLVu+yxdb8khcxM5xgU5IqGoBt/FF8FKIUeYq/4cbaXLeksOtdXJ3Cr8348XUK/SYOWUUIwPMezGjH8llUgwvI8dWrQCNp69hJyKBozsaw+ZtPXfnzlEncUcos5iDrWurQW6vJvjICIdpNEKWH08D1+dyEegkwXenRQMd2tTscP6B7lMiqXjg2Ail2LNiXw0qrV4eqRvl31QV2u0WHM8D9+cKoCThQlWzYwwuL2tXc1KYYRPpofj9V1pWHkgGxUNKjw21KfXFE/UvQRBwLfxRfj0cA762Jvj/ckh8LDRvbGpu0kkEjw+zBc2pkb46GAOan9OwsrJITA35sc2IiJDx5GeqJepabzSAO14bhUmhjrj5dF+UBiJu6S9NTKpBIvGBUAhl+Hb+EI0t2jw4mi/Tjdsyyqvxxu70pBxuR53hTrjudv6wsKEQ2JbKIxkWHFXCN77LRNrTxagol6FhWMDIL/JDB9Ra5rUGry9LxN7UsswOsABi28PhJmx7o5NPWF2Pw/YmBrhzV8z8NiWRHw8LQx2Zt135CUREYmPn0aJepH0MiVe2p6CsrpmvDLGD9MiXPVi5lMqkeCFUX2hMJJi/elCNLVo8dq4gJsu+bwejVbAd/GFWHU0F5YmcqycHIKRfg7dELVhk0slWDjGH/ZmxlhzIh+VDWqsmBis0zd7SHeV1DbhxW0pSC9T4vFhPrhvoKdejE09YUKIM6wVRnj5lxQ8tCkBn04Ph1s7tuAQEZF+MexuK9RhCUU1+PhgDlQtWrFDoS6yM7kUD248jxaNFqtnRWJ6pH6dISyRSPDkcF88fIs3diSX4vVdaWjRtC8/i2oa8diWBHxy+CKG+tph0739WJx3gkQiwSNDffDyaD8czanE499fQE0j955R+8QXVGP+hnMoqG7Eh1NDcf8gL70am3rC0D52+OzucFQ1qLFg03lkldeLHRIREXWTDs+ga7VaLFmyBOnp6TA2NsayZcvg7e197fFly5bh7NmzMDe/0uX3888/h6WlZecjpm4lCAJ+SCjGBweyodEKcLYyQWyMu9hhUScIgoBPD19E3JlC9PO0xvKJwXq7RFIikeChId5QyKX45PBFqFq0WD4xGMby1u81CoKAbRdK8NHBHEgkwJI7AjEhxIlFQBe5O8oNdmZGeG1XGh7alIBPpocZ7FFY1HUEQcD35y/hw4M58LRR4P3JofCx691HGrYm0t0aX86KxFM/XsDDmxLw0dRQRHbz6RZERNTzOjyDvn//fqhUKmzevBnPP/883nnnnb89npycjDVr1iAuLg5xcXEszvVAk1qDN3/NwHu/ZWGIjy2iPayx9kQ+6lUtYodGHaTRCnh7bybizhRieqQr/n13hN4W5381b4AnXhzlh0PZFXh+azKa1Job/mx5vQrPbU3G2/syEeJqiU339sOdob2nI3RPGRXgiE+nh6NM2YwHN55HNmf4qBWqFi2W7c3A+79nY6ivHdbOjmZx3gZ+DuZYMysKtmZGeOKHCziaUyl2SERE1MU6XKDHx8dj+PDhAICoqCgkJSVde0yr1SIvLw+LFy/GrFmz8MMPP3Q+UupWxbVNeHhzAnYkl+KhIV74YEoonhrhi6pGNb6LLxI7POoAVYsWi3amYltSCR4c7IWXR/sZVBOvmdFueH1cAE7mVeHpn5KueyPpt4zLmPXNGZzOr8bzt/XFZ3eHc2a3G/XztMGX90RCIwAPbUpAQlGN2CGRDrqsbMYjWxKwPakUCwZ74f3JIWzQ2A5u1gqsnhUJXzszPL81CbtSSsUOiYiIulCH3xGVSiUsLCyufS2TydDS0gK5XI6GhgbMnTsX999/PzQaDebPn4+wsDAEBQXd8PVkMglsbNp/91wmk3boefRfx3Mq8PTm81BrBHwxNwajAp0AAMNszTE22BnfxhfiwRF9YWeu/zOv12OIOdSgasGzG8/hSFYFXh0fhPtv8RE7pG4xf3gf2For8OKPF/DM1mR8Na8/rEyNUNOoxps7UrA9sRjh7lZ4f3oE+jpa3PwFO8gQc6ijBtqY4ftHBuOBdWfwxA8X8K+ZURgd7CR2WDqvt+TQufwqPLHxPBpULfgsNhrjQpzFDkkv2diY4buHBuOJjWfxxu50NEOCBcP79Iocou7TW8Yh6j7Moa4hEQRB6MgTV6xYgcjISEyYMAEAMGLECBw+fBgAoNFo0NjYeK2Af++99xAQEIApU6bc8PXUag2qqxvaHYeNjVmHnkdX9v9tOFOIf/9xEd52Zlg5ORRetn8/bzanoh6x6+IRG+OBZ27tI1Kk3cvQcqi2SY1nfkpGckktFo0LwKQwF7FD6nYHM8vx6s5U9LE3x30DPfHRwWxUNKjx4GAv3D/QE3JZ9/bDNLQc6gpVDSo883My0krr8OpYf0wOdxU7JJ1m6DmUX9WIX9PK8PWJfLhameD9yaHo62Audlh6T9Wixeu70vB7ZjmeG+OP2Ej+PaOOM/RxiLofc6h1jo5t2/Ld4U+tMTEx1wry8+fPIyAg4Npjubm5mD17NjQaDdRqNc6ePYvQ0NCOXoq6QaNag0U70/DJ4Yu4zd8B38yO/kdxDgB97M0xPsQZ358vQmldswiRUnuUK5vxyOZEpJXV4Z27QnpFcQ4At/o7YOXkUORWNmDhjlSYG8vxdWwUHhri3e3FOV2frZkxVs2IwEBvWyzbm4mvT+Sjg/eDSQ8JgoDU0jqsOnIR93xzBtO/Po0vj+VhiI8tvpkTzeK8ixjLpVg+MRh3BDvho98ycTyXe9KJiPRdh2fQr3Zxz8jIgCAIWL58OQ4fPgwvLy+MHj0aq1evxp49e2BkZITJkycjNja21dfjDHrPKahqxIvbk3GxogFPDPPFvAEerTbMulTThOlfn8adoc54bVzADX9OXxlKDhXVNOLJHy6gol6FlZNDMdDbVuyQelxCUQ3OFdZgVox7j57HbSg51B1aNFq8+WsGdqeWYUaUG56/rW+Hzq83dIaQQy1aAecKq3EoqwIHsypQWtcMmQSI9rDGrX4OGOlnzx4Q3aRJrcGCzQkorW3GhnkxcLY0ETsk0kOGMA6RuJhDrWvrDHqHC/SuxgK9ZxzNqcTru9IglQBv3xmMQT5tK+JW/p6FH85fwub7+sPbwDrtGkIOZZfX48kfLkCl0eJf08IQ5moldki9iiHkUHfS/nnU34YzhRgd4IA3xwfd9Gi83kZfc6hJrcGJ3CoczCrHkZxK1DS1wEQuxWBvW9zqb49hfexhY2okdpi9QqVai6mrjsPP0RxfzIzg6iFqN30dh0h3MIda19YCnW1TewmtIODrE/n48lge/B3N8d7kELhb/3NJ+408MNgL25NK8J+jeVhxV3A3RkrtlVRci2d+SoKRTIov7omEH5eOko6RSiR4emQf2Jsb41+HclDTeAHvTw5l5249VdOoxpGcShzMKsfx3Co0t2hhpZBjeB87jPRzwGAfW5j24AoWuqKPowUWjfPHop1p+OxILp4eaZh9Y4iIDB0/HfUCyuYWvLE7HYezKzA+2AmvjvVv9/JfOzNjxPbzwNcn8nFvqQeCnHmuvS44lVeFF7Ylw87MGJ/NCG/XTReinja3vwfszIzw5q8ZeHhzAj6ZFgYHCy7F1QcltU1Xlq5nV+BcQTU0AuBkYYzJYS641d8e0e7WnLHVAeOCnHCusAYbzhQiyt0KI/0cxA6JiIjaiUvcDVxORT1e3JaCopomPDuyD2ZGu7W637w1yuYWTFlzCsEulvh0engXRyoefc2hA5nlWLQzFd62Zvh0OgsdMelrDonl2MVKvPJLCmxNjfDp3RHXbVDZ2+haDgmCgJyKhj/3k5cjtVQJAPC1N8Otfva41c8Bwc4WHX4/oa53NYdULVos2HQehdVNiJsXzRu31Ga6Ng6R/mEOtY570Am/ZVzG0j3pMDWSYcVdwYjxsOn0a8adLsAnhy/iPzMj0M+z86+nC/Qxh35JKsGyvRkIdbHER1PDYM09nqLSxxwSW3JJHZ75KQkSAB9PC0OIS+9elaMLOaQVBCQV1+FQVjkOZlUgv6oRABDuanmtyZuh9SAxJH/NoaKaRsyNOwtPG1OsmRXFng/UJrowDpF+Yw61jgV6L6bRCvj8SC7Wny5AmKsl3r0rBE5d1NG1Sa3B9K9Pw9lSga9iIw1i9kTfcui7+EJ8dDAHg7xt8P7kUO711AH6lkO6Iq+yAU/9eAFVjWq8NykEg33sxA5JNGLlkFqjxZmCahzMrMCh7ApU1Ksgk0owwNMGt/rbY2Rfe67O0RP/m0OHssrxwrYUzIhyw0uj/USMjPQF38uos5hDrWOTuF6qulGN13am4mReNaZGuOCF2/y69M65wkiGB4d4Y8W+TPyRU4kRfe277LWpdYIg4ItjefjqRD47YZNB8LYzw1exUXjqpyQ883My3rgjAOODncUOy6C1aLS4WNmAtFIlTuZV4UhOJepVGpgaSTHU90qTt6G+drBU8OOBvhvp54A5/TzwbfyV/ejjgpzEDomIiNqA78AGJL1UiZe2J+NyvQqLxvpjSoRrt1xnUqgzvj1TiM+PXMRQXzueadwDtIKAD37PxpbzlzA5zAULx/rzz50MgoOFCb68JxLPb03G4l3pqKxXY05/D7HDMghNag2yyuuRXqZEWqkS6WVKZJfXQ6W5snDO1tQIYwIccau/PQZ42cKEN/wMzpPDfZB4qRZv781EoJMFtygQEekBFugGYldKKZbvy4S1Qo4v74ns1nOw5TIpHrnFG4t2puHXtDJMCOGMV3dq0Wix9NcM7Ektw9z+HnhqhK9BbC0gusrCRI5Ppodj8a40fHwoBxX1Kjw5whdS5nmbKZtbkF6mvPZPWqkSeZUN+LMWh5VCjkAnC8yMdkeQkwUCnSzgaWvKG30GTi6TYvnEIMyNO4tXfknF2tlR7T7FhYiIehYL9HZSNrfg58RiuFgpEOxsAXdrhajFUotGi48P5WDzuUuI9rDGionBsDc37vbrjgl0xLpTBfjiWB7GBjrCiMfrdIsmtQYLd6TiSE4lHh/mg/sGerI4J4NkIpdi+cRgrPw9C3FnClHRoMLr4wJ4dNd1VDaors2IX/2nsLrp2uOOFsYIdLLAbf4OCHSyQJCzBVwsTTh29FIuVgq8OSEIT/+UhJW/Z+O12wPEDomIiFrBAr2dypUqrD6eh0a1FsB/ZyWCnS0R7Hzlg1BPFe0V9Sos3JGKc4U1iI1xx1MjfHvsw6xUIsHjw33xzE9J2HqhBDOi3Hrkur2JsrkFz29NxrnCGrwyxg/TI/lnTIZNJpXgpdF+sDc3xhfH8lDVoMY7d4XAzLh3zvgJgoDSuuZrxXhamRIZZUqUKVXXfsbdWoFAJwtMCnNB4J8z4z1xk5b0yy2+dnhgkCe+PlmAKA8rTAx1ETskIiK6AXZx7wBVixbZFfVILVUirbQOqSVKZJXXo0V75Y/SSiFHkJMFgl2uFO3BzpZwtera2YsLl2rx8i8pqG1qwaJx/qI0VhIEAY9sSUReZQO2Lhiot93EdbHjZFWDCk/9mITM8nq8OT6QzX10nC7mkL77ObEY7+zPhLXCCF62pnCxMoGrlQKuViZwsVLA1UoBFysTvR13/peVlSkScyuQ8Zf94ullStQ0tQAApJIrTfWuLk8PcrZAgKMFm7nRNTcbhzRaAU/+kIgLxXX4Zk40/BzMezA60gd8L6POYg61jses9TBVixZZ5fVIK61DSumVD1hZ5fXQ/Fm0WyvkCHa2RJCzxZWi3cWyw0sOf0osxsrfs+BoYYL3JoUg0Mmiq3+dNksoqsGCTQl4fJgP7h/kJVocnaErOXRVSW0TnvzhAkrqmvHuXSEY2qf3Hj2lL3QthwzFidxK7EktQ3FtM0pqm1CqVF0bU6+yVsivFet//berlQlcLBWwNpWLtrRbEAQomzWoalSjqkGFqgb1n/+t/sf3LtU0oV6lAQAYySTwczBHgJPFtYLc39Gce4epVW0Zh8rrVZizPh6WJnKsnxvTa1en0PXxvYw6iznUOhboOqD5L0V7aokSqaV1yK5o+HvR/pdZ9mBnCzi3UrSrWrR47/csbLtQgsHetlh2ZxCsTY168le6rmd/TkJCUS22LhgAK4X48bSXLuVQXmUDnvjhApTNLfhoahiiPazFDonaQJdyyJBptALK61UoqW1CcW0zimubUPI//25q0f7tOaZGUrhY3qCAt1LAwdy4zY3SBEFAvUpz3QL7Rt9r0V7/LdbcWAYbUyPYmRnBxtQIPk4W8LFWIMjJAr72ZuzrQe3W1nHoTH41nvghEWMDHfHWhCD2JqBr+F5GncUcah0LdB3V3KJF1mUlUkuvFOyppUrklNdf67Rra2r031n2P2fcnS1NUKZU4eXtKUguqcN9Az3x6FAfnem+m3lZiTnrz2L+QE88OdxX7HDaTVdyKL1Uif/78QIA4NPp4Qh0Fm9lBLWPruRQbycIAmqaWq5bwJfWNaO4thnVjeq/PUcmlcDZwvjPZfNXinYTuRTVV4vuvxbejWqoNdd/yzQzksHWzAi2fxbcVwpvY9j943tGsDUz/seRZswh6qz25NDXJ/Kx6mguFo7xwzT2N6E/cRyizmIOta6tBTo3r/UwE7kUoa5WCP3LMWhXz6pNLVUitaQOaWVKrDtVcK1otzMzQotWQItGwLuTQjDK30Gk6K/P39ECtwc7YdPZIsyKdoODhYnYIemdc4U1ePbnJFiayPHvu8N5Vi1RB0gkEtiYXimCg5yv/ybYqNb8Zda9CSV/Fu4ltU04U1CDy8oyaIUrM++2ZsawNTWCo4UxAhzN/yzAr3zvajFu++f1uPyc9Ml9gzxxrqgGKw9kI8TF8oZ/X4iIqOdxBl1HNak1yLxcf22mvbapBU8M90Efe91s6lJY3Yi7157BlHAXvDLGX+xw2kXsHDqaU4mXf0mBq5UJPp0eDhcrhWixUMeInUPUdVo0WrRohR4vuJlD1FntzaHqBjXmxMXDSCbFhnkxsDDhnI0uaNEKWPl7FoxkUowPdkKws0WPbUPgOESdxRxqHWfQ9ZzCSIZwNyuEu1nd/Id1gIeNKaaEu2DrhRLM7e8BDxtTsUPSC7+mluGNPekIcDTHv6aFwdaMxyMRiUkuk0LOyXDqBWzMjLB8YjAe2ZKIpXvS8d6kEO5H1wGrjlzEjwnFkEsl2HS2CD52phgf7Iw7gp3gZs0b+ES9AbvQUJdZMNgLcqkEXxzLEzsUvfDD+Ut4fVcaIt2s8PmMCBbnRETUoyLdrfF/w31xMKsCG88WiR1Or/d7xmWsP12I6ZGu+PWxwXh1rD9szYyx6mguJq85hYc3ncdPicWobVLf/MWISG/JlixZskTsIABAqxXQ1IEBR6Ew6tDzqOuZGcvRoNbgp4Ri3OpvD3tz/Sg4ezqHBEHAN6cK8PGhHAzvY4eVk0NhZszFLPqM4xB1FnOIOqujORTuaomMsnp8n1CMQd62cLZkHxkx5FY04NmtyQhwtMDyicEwM75yPO9dYS64K8wZtmZGuFBcix3Jpdh4tgjpZfWQSyVwt1Z0WdNgjkPUWcyh1pmbt2185R506lK1TWpMWXMake5W+GhqmNjhtElP59DPicVYvi8T44OdsPj2AMh5nJLe4zhEncUcos7qTA7VNqkxb8M5aLQCNsyLgY0OHOHamzSoNLjv23OoalQjbm70DXvRCIKAtDIldqeU4de0MlQ2qGGlkGNMgCMmhDghws2qU9sUOA5RZzGHWtfWPeisDKhLWSmMMG+AB47kVCKhqEbscHROi0aLr0/kI9zVCkvGB7I4JyIi0VkpjLBiYjAqG1RYsjsdWt2Yu+kVBEHAsr0ZyKtqwNt3BrXaKFYikSDY2RLP3dYXOx8ZjH9NC8MtvnbYlVKKBZsSMOWr01h1NBe5lSyQiPRZh6sDrVaLxYsX45577sG8efOQl/f3fcdbtmzBtGnTMHPmTBw4cKDTgZL+mBXjDntzY3z2x0XoyAINnbEv4zJK6ppx3yBPSNmMh4iIdESIiyWevbUvjl6sxPpTBWKH02tsPFuEfemX8fgwXwz0tm3z8+RSCW7xtcNbE4Lw62NDsHR8ILxsTPHNyXzMWHsG9357DpvPFqGyQdWN0RNRd+jwxtf9+/dDpVJh8+bNOH/+PN555x2sWrUKAHD58mXExcXhxx9/RHNzM2bPno2hQ4fC2Fg/9iRT55gayfDgYC+891sWjuVWYaivndgh6QRBEBB3uhC+9mYY1od/JkREpFvujnTFucIarDqai3A3K/TztBE7JIN2rrAGnxzKwa1+9pg/wKPDr2NmLMOEEGdMCHHGZWUz9qZdxq6UUqw8kI2PDmZjsI8dJoQ4YURf+x4/QpKI2q/DM+jx8fEYPnw4ACAqKgpJSUnXHktMTER0dDSMjY1haWkJLy8vpKWldT5a0htTwl3gZq3A539c5FK5Px3PrULm5XrM6+/B2XMiItI5EokEi8b5w8PGFIt2pqGinrOv3aVc2YyFO1LhbmOKN+4I7LIj7hwtTDCnvwe+nd8PG+/th7kDPJFVXo9FO9Nw+6oTWLInHafyqqDR8rMZka7q8Ay6UqmEhYXFta9lMhlaWlogl8uhVCphafnfTfDm5uZQKpWtvp5MJoGNjVm745DJpB16HnW/58YE4IUfE3G8sBZ3hruKHc4N9VQOfXcuCS5WCtwz2AfGcu49NyQch6izmEPUWV2VQzYAPp8dg+lfHMfSvRlYe++ALusSTleoNVo8+n0iGlQaxN0/AB7ObWsc1V79bczQ388Rr94ZgtN5ldiWUIzdSSXYmVwKZysT3BXhhsmRbghyuXJ9jkPUWcyhrtHhAt3CwgL19fXXvtZqtZDL5dd9rL6+/m8F+/VoNAK7uBuYYV7W8HMwxwd70zHIzXk9bQEAACAASURBVFJnG6L1RA4lF9fi5MVKPDOyDxqUTWDGGhaOQ9RZzCHqrK7MIWeFDC+N9sNbv2Zg5e5UPDLUp0tel6744EA24vOrsWxCEBxNZD3ydz/Q1hQv3doHTw31xpGcSuxKKcXaY7lYc+Qi/B3NMT7YCfeP6IuWRq6aoI7je1nrur2Le0xMDA4fPgwAOH/+PAICAq49FhERgfj4eDQ3N6Ourg7Z2dl/e5x6B5lUgseG+aCgugnbk0vFDkdU608XwtJEjikRLmKHQkREdFOTwlwwMdQZX53Ix4ncSrHDMRi/ppZh09kizIpxx+3BTj1+fYWRDGMCHfHh1DDseWQwXhzlB4Vcik8OX8SirUk3fwEi6nYdLtDHjh0LY2NjzJo1CytWrMDChQuxdu1a/Pbbb3B0dMS8efMwe/Zs3HvvvXj22WdhYtK2g9nJsAzvY4dwVyusOZ6HJrVG7HBEkVfZgAOZ5bg7yhXmxh1etEJERNSjXh7thz4OZnh9VzrK6prFDkfvZZfXY9neDES6WeHpEb5ihwMbMyPMjHbD17Ojcd9AT+xOLkFWef3Nn0hE3Uoi6Mg5WGq1hkvcDVR8QTUe3ZKIp0b4Yt4AT7HD+YfuzqG392ZgV0optj80CPbmPMnAEHEcos5iDlFndVcO5VY0YP63ZxHgaIH/zIzQ2e1quk7Z3IJ7vz2HepUGG+ZGw8FCtyauqhvVmPLVKQzxtsWKu0LEDof0FN/LWtftS9yJ2qqfpw2G+Nhi3akCKJtbxA6nR5XXq7AzpRQTQ11YnBMRkd7xsTfDq2MDkHCpFp8fyRU7HL0kCAKW7klHUXUjVkwM1rniHABsTI0wf7A39meUcxadSGQs0KlHPDHMFzVNLYg7Uyh2KD1q09kiaLQC5vbv+PmmREREYroj2AnTI10Rd6YQh7IqxA5H76w/XYiDWRV4amQfRHtYix3ODT1wiw/MjWVYczxP7FCIejUW6NQjAp0tMCbAERvjC3vNuarK5hb8mHAJo/wd4GlrKnY4REREHfbsrX0R5GSBpXvScammSexw9MapvCp8fuQixgY6IjbGXexwWmVjZox7YtzxW0Y5si5zFp1ILCzQqcc8OtQbqhYt1p7MFzuUHvFzYjGUzRqd3HdPRETUHiZyKVbcFQwBAhbuSIWqRSt2SDqvpLYJi3amwdvWDK+NC4BEovvnyc+Ocb8yi36Cs+hEYmGBTj3G284Md4W54MeEYoO/+65q0WLj2SL097JBiEvbGkIQERHpMg8bUyy+PRApJXX45HCO2OHoNFWLFgt3pEKt0eK9SSEwM5aJHVKbWJsaYdafs+iZl5Vih0PUK7FApx61YIg3pBLgSwPf37QntQyXlSrcO4B7z4mIyHDc5u+A2f3csfncJexPvyx2ODrrw4PZSCquw+I7AuFjbyZ2OO0Se3UW/XjvWPFIpGtYoFOPcrY0wYwod+xOKUVOhWHub9IKAuLOFCDA0RyDvG3FDoeIiKhLPTncF2Gulli+LxOlPB/9H3Ykl+DHhGLMH+CBUf4OYofTbldn0X/PLEdGGWfRiXoaC3TqcfcN9ISpkQyrDPS4lj+yK5Bb2Yj5Azz1Yr8ZERFRexjJpHhzfBDUGi2W7kmHVhDEDklnpJcp8c7+LPT3tMZjw3zFDqfDZve7uheds+hEPY0FOvU4GzMjzO3vgYNZFUgqrhU7nC4lCALWnSqEm5UJRgc6ih0OERFRt/C0NcWzt/XF6fxqbDl3SexwdEJtkxovbU+BtUKOtycGQy7V35v0VgojxMa440BmOdI5i07Uo1igkyhi+7nD1tQInxnYLHpCUS0uFNdiTn8PvX5jJiIiupmp4S4Y1scO//7josFuW2srrSBg8a50lNU1Y8VdIbAzMxY7pE6b3c8DFiY8F52op7FAJ1GYG8tx/2AvnMmvxsm8KrHD6TLrThfAWiHHpDAXsUMhIiLqVhKJBIvGBUAhl+KNXelQa3rv0WtfncjH0YuVeO62vohwsxI7nC5hqZAjNsYdB7MqOIuuZ8rqmtHMoxD1Fgt0Es30CFe4WJrgsz8uQjCA/WtZ5fU4klOJe6LdoTDSj+NUiIiIOsPB3BivjgtAWpmy1+5XPnaxEquP5WFCiBPujnQVO5wuFRvDWXR9k1BUg0lrTmHs58fw8vYU7EopRU2jWuywqB1YoJNojOVSPHyLN1JLlTiQWS52OJ224UwhFHIpZkS7iR0KERFRjxnl74CJoc745mQ+Ei8ZVm+ZmymqacTru9Lg52iOhWP8Da45rKVCjtkxV/oGpZdyFl3X1Tap8fquNDhbmmBCiDMuFNfijd3puH3VcTz2fSI2nS1CcW2T2GHSTbBAJ1FNCHGGr50ZVh3NRYtWf2fRS2qbsCe1DJPDXWBjaiR2OERERD3q+dv6wtnSBG/sTkODSiN2OD2iSa3BK9tToRUEvDcpxGBXz82KcYeliRyrOYuu0wRBwNt7M1GmVGH5nUF4ZYw/djw8CN/MjsL8gZ6oqFfhgwPZmLT6FOasj8fqY3nIKFMaxCpWQ8MCnUQlk0rw6DAf5FY2YldKqdjhdNjGs0WAIGB2Pw+xQyEiIupxFiZyLBkfiKLqJnx8KFvscLqdIAh477cspJUp8eb4IHjYmIodUrexVMgR288dh7I5i67Lfkosxu+Z5XhimA9CXa/0QZBKJAh1tcLjw3yx5b7++PGBAXhqhC9MjWRYfTwPc+LOYvKaU/jgQDbiC6r1erLMkLBAJ9Hd5mePEBdLrD6Wp5cNLWqb1Pg5sRhjg5zgZq0QOxwiIiJRxHjYYN4AD/ycWII/sivEDqdbbb1Qgl+SS/HAYC8M72svdjjdLvbPWfQvOYuuk7Iu1+PDA9kY4mOLOf1vPFnkZWuKeQM8sSY2CrsfHYzXxvmjr4M5fkq4hEe3JOKOVcexZHcafs8sR6O6d6yE0UVysQMgkkgkeGKYD5744QJ+TLikd7PQP5wvRqNai/kD9CtuIiKirvbILT44nluFZXszsOnefrA1gOPG/ldySR3e/z0Lg71t8fAQb7HD6REWJnLM7ueOL47lIa20DkHOlmKHRH9qVGvw6o5UWCqMsGR8IKRt7INgb26MyeGumBzuigaVBifyqnAoqxx/5FRiZ0oZTORSDPSywUg/ewzva28QRwfqC86gk04Y6G2LAV42WHuyABX1KrHDabMmtQabzxVhiI8t/B0txA6HiIhIVMZyKd4cH4S65hYs35dpcPtbqxvUeHl7ChzMjfHWnUGQSQ2rKVxr/rsXvXd269dVHxzIRm5lA5aOD+xwEW1mLMMofwcsHR+EXx8bglUzIjA1whVZ5fVYtjcTd6w6gYc2nUfc6QIUVDV28W9A/4sFOumMZ2/tg0a1Bi9vT9Gbs1R3ppSiskGNewd6ih0KERGRTvBzNMdjQ31wMKsCvyTrb3+Z/6XRCnhtVyqqGlR4566QXtcU1sJEjjn93XE4uwKppXVih0MA9qaVYduFEtw3yBODvG275DXlUgn6e9ng+dv6YtuCgdgwLwYPDfFGg0qDTw5fxLSvT2PmN2fw+ZGLSC6pg9bAbsLpAhbopDP8HS2w+PYAJFyqxbv7s3T+rrtGK2DDmUKEulgixsNa7HCIiIh0xux+HojxsMaHB7JRVGMYM25fHsvFybxqvDTaDyEuvXOJ9z3R7rBSyLH6GPeii62wuhHL92Uiws2q27ZaSCQSBDpZ4KFbvPHt/H7YtmAgnr+tL+zNjLD+VAHu+/YcJn55Eu/sz8SJ3Eq06MkEm65jgU46ZVyQEx4Y5IltSSXYcu6S2OG06kBmOQqrmzB/gIfBnXtKRETUGTKpBEvGBwIAlu5Oh0bPu0MfyqrA1ycLMDncBZPDXcUORzQWJnLM6eeBP3IqkVLCWXSxqDVavLYzDVKJBMvuDIJc1jMlnZu1ArNi3LFqZiR+fWwIlo4PRJirFXYml+L/fkzC+E+PYGdyKbvBd1KH/m82NTXh//7v/zB79mw89NBDqKys/MfPPProo5g1axbmzZuHBQsWdDpQ6j0eGeqDEX3t8dHBbJzKqxI7nOsSBAHrTxfAy9YUI/0cxA6HiIhI57haKfDiKD+cK6rFt2cKxQ6nw/KrGvHG7jQEO1vgxVF+YocjupnRbldm0dnRXTSrjuQiuaQOr43zh6uVOCcIWZsaYUKIM96bFIJ9jw/BionBMDWSYcmedNz99Wlsv1DCGfUO6lCBvnHjRgQEBOC7777DlClT8Pnnn//jZ/Lz87Fx40bExcVhzZo1nQ6Ueg+pRIKl4wPhbWeGhTtSUVite0vjTudXI7VUibn9PXpVgxgiIqL2mBDihFH+Dlh1NBcZZfp3hvbV3jhyqQTv3BUCEzkXn16dRT+SU4lkzqL3uGMXKxF3phDTI10xKsBR7HAAAAojGcYEOmLb47dg5eRQWCnkeGtvBqZ/fRo/JxbrTW8pXdGhUSY+Ph7Dhw8HAIwYMQLHjx//2+Pl5eWora3Fo48+itjYWBw4cKDzkVKvYmEixwdTQiEB8NzWZCibW8QO6W/iThfCzuzKnUMiIiK6PolEgoVj/GFtaoTFu9PQ3KI/H9RbNFq8uScD2eX1eOvOILhZizNTqYtmRrvBWiHHGs6i96hyZTOW7E6Hn4M5nhnZR+xw/kEikWCknz3WzYnGR1NDYWtmjOX7MjH1q9P44fwlqPTo77+YbnoO+vfff49169b97Xv29vawtLzSHMPc3Bx1dX+/e6ZWq/HAAw9g/vz5qKmpQWxsLCIiImBvb3/D68hkEtjYmLX7F5DJpB16Huk+GxszfBobjfvXncFb+zKxanYMpN0wW93eHEq+VIsTeVV4YWwAnB14tBpxHKLOYw5RZ+lyDtnYAO9OC8eCuHisPVOIV+4IEjukm6qsV+HpH8/hVG4VXro9EOOjPMQOqdu1J4dsADw4zBcf7s9EvlKNCDbL7XZarYCnf05GY4sGn8ZGw0UHj/f9aw5NtDXHndEeOJJVjk8PZOPd37LwzekCPDK8D2b284CJkUzkaHWXROhAq+wnn3wSDz/8MCIiIlBXV4fY2Fjs2LHj2uNqtRpqtRpmZlf+Bz399NOYN28e+vfvf8PXVKs1qK5uaPcvYGNj1qHnkf7Ycq4I7/+ejfsHeeLxYb5d/vrtzaFFO1Jx9GIlfnloECwVN73HRb0AxyHqLOYQdZY+5NA7+zPxU0IxPp8Rgf5eNmKHc0OZl5V4YWsyyutVWDQuoNeslmtvDtWrWjB59SmEuVrh42lh3RgZAcDak/n4/EguXhvnr7ONCm+UQ4Ig4FR+NdYcz8P5olo4mBtj3gAPTItwhaIXFeqOjm07/aFDS9xjYmJw6NAhAMDhw4fRr1+/vz1+7NgxPPPMMwCA+vp6ZGZmok8f3VuGQfphRpQbpoS7YO3JAvyaWiZqLEU1jdifcRlTI1xZnBMREbXD0yP7wNPWFEv2pOvc1rWrDmSW48GN56HSCPjynsheU5x3hLmxHHP6e+DoxUokF9eKHY5BSyiqwRdHczEu0BGTwlzEDqfdJBIJBnnb4st7IvGfmRHwtjPFRwdzMHnNKWw4U4hGtUbsEHVKhwr02NhYZGZmIjY2Fps3b8aTTz4JAHjvvfeQmJiIkSNHwtvbGzNnzsSDDz6I5557DnZ2dl0aOPUeEokEL432Q5S7Fd7am4HUUvEaknx7pghSiQSxMe6ixUBERKSPTI1kWDo+EOXKZqz8PUvscP5GEASsOZ6Hl7anoI+9OdbPjUaoq5XYYem8q3vRv+Re9G5T26TGazvT4GylwMKx/np9tK9EIkE/Txv8Z2YkvrgnAn4O5vjXoRxMWn0K604VoEHFQh3o4BL37sAl7nQzlQ0q3LvhHLSCgHVzY+Bgbtwlr9vWHKpqUOGu1adwe5AjXr89sEuuTYaB4xB1FnOIOkufcuiLo7lYcyIf79wVjNE60IW6Ua3Bm3vSsT+jHBNCnPDq2IBe2a29ozn0zcl8fHYkF2tnRyGMNzW6lCAIeOWXVBzKrsBXsyJ1/qZRR3IooagGa07k40RuFawVV1ZlzIhyg4WJ4a1U7dYl7kRisDMzxsopoahtasFL21J6vBPklnOX0Nyixdz+nj16XSIiIkPy4GAvBDtbYMW+TJQrm0WNpaS2CQs2nsdvGeV4aoQvltwR2CuL886Y8ecsOs9F73o/JRbj98xyPDHMR+eL846KdLfGp9PDsXZ2FMLdrPD5kVxMXnMKa47noa5JN7fCdDeOQKRXAp0ssGR8IC4U12LF/kz01AKQRrUG35+/hJF97eFrr5tdcomIiPSBXCbFmxOC0NSixZu/ZvTYe/n/Ol9Yg/kbzqGopgkfTQ3DvAGeer18WCzmxnLMG+CJYxercOES96J3lczLSnx4IBtDfGwxp7/hnyIQ5mqFj6aGYd2caES5W+OLY3mYtOYkvjiai9omtdjh9SgW6KR3Rgc4YsFgL+xILsXGs0U9cs2tF0pQ09SC+QM5e05ERNRZPnZmeGpEHxzPrcKPCcU9fv1tF4rx2PeJsFTI8c3saAztw15JnTEjyg02pkacRe8ijWoNFu1Ig6XCCEvGB0Lai24chbhY4oMpodgwNwb9PW2w5kQ+Jq0+hVVHLqK6sXcU6izQSS89dIs3bvWzx78O5eBEbmW3XqtFo8V3ZwoR5W6FCDfDXF5ERETU02ZEuWKwjy0+PpSDvMqe2T/fohWw8vcsLNubiX6e1lg7Owo+XBnXaWbGMszr74HjuVVI5Cx6p31wIBu5lQ1YOj4QdmZd03NJ3wQ6W+D9yaH4bn4MhvjYYu3JAkxefQr//uMiqhpUYofXrVigk16SSiRYOj4IfezN8eqONORXNXbbtfamX0ZJXTPmD+DsORERUVeRSCRYfPuVhmxv7E5Hi7Z7l7rXNKrx9I8XsPncJczu546Pp4XDSmHUrdfsTe7mLHqX2JtWhm0XSnDfIE8M8rYVOxzR+TtaYMVdIdh4bz8M62OH9acKMGn1KfzrUA4q6g2zUGeBTnrLzFiGlVNCIJUAz29N6pYzVQVBQNzpQvSxN+PyNyIioi7maGGCV8b4I7mkDmtP5nfbdXIq6nHfd+dwrqgGr98egGdv7Qu5tPcsG+4JZsYyzB/ggROcRe+wwupGLN+XiQg3Kzw8xFvscHRKXwdzvD0xGFvu64/b/B3wXXwhJq85hczLSrFD63Is0EmvuVub4t1JISiobsJrO9Og6eK778dyq5BVXo95Azx61f4fIiKinjI20BF3BDvhq+N5SC6p6/LX/yO7Ag98dx4NKg1WzYjApDCXLr8GXXF3lBtsTY2w+hhn0dtLrdFi0c40SCUSLLszCHIZy7Tr8bE3w5sTgrDlvv64d6An7Lvo2GVdwv/zpPf6edrgxVF9cfRiJT4/crFLX3v9qQI4WRjj9iCnLn1dIiIi+q+XRvnB3twYi3eloUmt6ZLXFAQB604V4PmtyfC0McW6OdGIdLfuktem6zM1kmHeAA+cyKtCQlGN2OHolVVHcpFSUofXxvnD1Uohdjg6z9vODA8N8TbIPfos0MkgTI90w/RIV6w/XYjdqaVd8ppJxbU4W1iDOf09YMS7mERERN3GUiHHkvGByK9qxCeHO3+zvUmtweu70vDvPy5idIAjVs+KhAuLnh5xbRade9Hb7NjFSsSdKcT0SFeMCnAUOxwSGasOMhgv3NYXMR7WWPZrRpcskVt/uhCWJnJMDudSOCIiou42wMsWs/u54/vzl3DsYsdPaCmra8bDmxPwa9plPDbUB8snBkFhJOvCSKk1V2fRT+ZVcxa9DcqVzViyOx1+DuZ4ZmQfscMhHcACnQyGXCbFO3cFw8HcGC9uS8ZlZXOHXyu3sgEHM8sxI8oV5sbyLoySiIiIbuTxYb7wtTfDW79mdOjM46TiWsz/9hzyKhuxcnIIHhjsBQl7yPS4u6PcYGdmhC+5F71VWkHA4t3paFBrsHxiMG8kEQAW6GRgbM2MsXJKKJTNLXhxWwqaW7Qdep0NZwphJJNgZrR7F0dIREREN2Iil+Kt8UGoblTjnf2ZEIS2N3/dmVyKRzYnwEQuxVezozDSz6EbI6XWXJlF98Sp/GqcL+Qs+o2sO1WA0/nVeHFUX/jam4kdDukIFuhkcPwdLbB0fBCSS+qwfF9Gu97cgStLjXallOKuMBeD7AxJRESkywKdLfDwLd74LaMcu1PLbvrzGq2Afx3KwZI96Yhws8K6OdHwczDvgUipNXdHul6ZRede9OtKKKrBF0dzMS7QkScL0N+wQCeDdJu/Ax6+xRu7UsrwbXxRu5678ewlaLQC5vTz6KboiIiIqDXzB3giws0K7/+ehZLaphv+XF1TC579OQkbzhRiRpQbPp0eDhtTox6MlG5EYSTD/AGeOJ1fjXOcRf+b2iY1XtuZBhcrBRaO9ec2DPobFuhksB4c7IXRAQ749HAOjrax2YyyuQU/JlzCKH8HeNqadnOEREREdD0yqQRLxwdCqwWW7kmH9jqr4fIqG3D/d+dwKr8aC8f44aXRfjw7WsdM/3MWnR3d/0sQBCzbm4nL9Sq8fWcQLEzY64j+jqMYGSypRII37ghEXwdzvLYzFbmVDTd9zs+JxahXaTB/oGcPREhEREQ34mFjiudu64MzBTXYdPbvq+GO51bivu/OoaapBZ/PCMe0SDeRoqTWKIxkuHcgZ9H/6qfEYhzILMcTw3wQ6moldjikg1igk0EzNZLhgymhMJJK8fzWZNQ1tdzwZ1UtWnwXX4QBXjYIdrbswSiJiIjoeiaFuWBEX3t89sdFZJXXQxAEfBdfiGd+SoKrlQLr5kQjxsNG7DCpFdMiuBf9qszLSnx4IBtDfGwxpz+3UtL1sUAng+dqpcC7k0JwqaYJr+5MhUZ7/aZxu1NLUV6vwvwBHDCJiIh0gUQiwaJx/jA3lmPxrjS8+WsGPjqYgxF97bFmVhTcrBVih0g3cXUW/Ux+Nc4WVosdjmga1Ros2pEGS4URlowPhJT7zukGWKBTrxDtYY2XRvvhRG4VPj188R+PawUBcacLEehkgUHetiJESERERNdjZ2aMReMCkHm5HjuSS/HQEC+8OykEZsY8M1pfTItwhb25MVb34nPRP/g9G7mVDXhzfCDszHhKEN0YuxJQrzE1whVZl+vxbXwh/B3NcWeo87XHDmdVIK+qEW/fGcROmkRERDpmpJ89Fo7xg6OFCYb3tRc7HGqnq7PoHx7IRnxBNfp59q5tCXvTyrAtqQT3D/LEQE4E0U1wBp16lWdv7YP+XjZ4e18GLlyqBXClm+b60wVws1ZgVICjyBESERHR9UyLdGNxrsemhrvAwdy413V0L6xuxPJ9mYhws8LDQ7zFDof0QKcK9H379uH555+/7mNbtmzBtGnTMHPmTBw4cKAzlyHqMnKZFCsmBsPJwgQvbk9BWV0zzuRV4UJxHeb084BcytlzIiIioq52dRY9vqAG8QW9Yy96daMar+5IhVQiwbI7g3gMILVJh7Nk2bJl+OCDD6DVav/x2OXLlxEXF4dNmzbhq6++wocffgiVStWpQIm6io2pEVZOCUWjSoMXtiXjs4PZsDE1wqQw55s/mYiIiIg6ZMqfs+hf9oK96CkldZgXdxZZ5fVYMj4QrlZsaEht0+ECPSYmBkuWLLnuY4mJiYiOjoaxsTEsLS3h5eWFtLS0jl6KqMv5OZjjzQlBSCtV4mh2BWZGu0FhxGYzRERERN3l6iz62ULDnUUXBAE/JxZjwabzAIA1s6IwglszqB1u2iTu+++/x7p16/72veXLl2PChAk4efLkdZ+jVCphafnfc6TNzc2hVCpbvY5MJoGNjVlbYv6f50k79Dyiyf3NUN7cgo2nC7BgZF/YsKMmdRDHIeos5hB1FnOIOquncuj+4X0Qd6YQX50qwKgwV4Nqztuk1mDJjhT8eLYIw/zs8cHdkbAz7z2fLzkOdY2bFugzZszAjBkz2vWiFhYWqK+vv/Z1fX393wr269FoBFRXN7TrOgBgY2PWoecRAcD0UGc8cIsPamoaUa1qETsc0lMch6izmEPUWcwh6qyezKH5/T2w8kA2tp0pwK3+Dj1yze5WWN2IV35JRXqZEg8O9sJDQ7whVbegurr3fL7kONQ6R8fW6+GruqVTQUREBOLj49Hc3Iy6ujpkZ2cjICCgOy5F1GmGdOeWiIiISNdNiXCFj50pXtyegnf2Z0LZrN9F7NGcSszfcA6Xaprw0dRQPDrUBzI2HqYO6tJz0NeuXQsvLy+MHj0a8+bNw+zZsyEIAp599lmYmJh05aWIiIiIiEgPmcilWD83Bl8czcPGs4U4lFWBF0f7YZSezaZrtALWHM/DmhP5CHA0x7uTQuBhYyp2WKTnJIIgCGIHAQBqtYZL3EkUzCHqLOYQdRZziDqLOUSdJVYOpZbW4e29mUgvU+JWP3u8OMoPTpa6P7FX3ajG67vScCK3ChNDnfHyaL9e33CY41DrRF3iTkREREREdDPBzpb4Zk40nhrhi+O5VZj5zRl8f/4StLoxh3hdKSV1mL/hLOILqrFwrD8W3x7Q64tz6jos0ImIiIiISDRyqQTzBnhi0739EOZqifd+y8KCjQnILq+/+ZN72NY/j1ATBGD1rChMizCsTvQkPhboREREREQkOg8bU3w6PRxLxwciv6oBc+PO4j9Hc9HcohU7NDSpNXjr13S8vS8T/TxsEDc3BqEubVuyTNQeXdokjoiIiIiIqKMkEgkmhDjjFh87fHQoG1+dyMf+9Mt4dZw/YjxsRImpqKYRL2+/coTaA4O98PAQb3Zpp27DGXQiIiIiItIpNmZGWDo+CP+eHg61VsAjmxOxbG8GapvUPRrH0Yv/PULtwymheIxHqFE3Y4FOREREREQ6aZCPLTbf2w/z+ntgR1IJZqw9g/3pl9HdB1FpBQFfHsvFsz8lwdnSBOvnRmN4X/tuvSYRwAKdiIiIiIh0mMJIhqdG9sG6OTFw/M40wAAACHRJREFUtjTBwh2peG5rMkpqm7rlejWNajzzUxJWH8/HhFBnfB0bxfPNqcewQCciIiIiIp0X6GyBr2dH45mRfXAmvxr3fBOPzWeLoNF23Wx6WumVI9TOFFRj4Rg/vMEj1KiHsUAnIiIiIiK9IJdKMKe/Bzbd1w8R7lZYeSAbCzadR9blzh/Jtu1CMR7ceB6aq0eoRbrxCDXqcSzQiYiIiIhIr7hbm+KTaWF4a0IQiqqbMHfDWXz2x0U0qTXtfq3mFi2W/ZqBZXszEe1hjQ08Qo1ExGPWiIiIiIhI70gkEtwR7ITBPrb416EcfHOqAL9lXMarYwPQ36ttR7JdqmnCy9tTkFamxAODPPHwLezSTuLiDDoREREREektG1MjvHFHID67OxwCgMe+T8Sbe9JR09j6kWzHLlZi/oazKKxpxAdTQvHYMF8W5yQ6FuhERERERKT3BnrbYuP8frh3oCd2pZRi5jdnsDet7B9HsmkFAauP5eGZn5LgZGmC9XNiMIJHqJGOYIFOREREREQGQWEkw5PDfbF+bgxcrBRYtDMNz/ychOI/j2SraVTj2Z+T8OXxPEwIccLXsVHwtOURaqQ7uAediIiIiIgMSoCTBb6OjcKW85ew6shFzFx7BnP7e2BXSinKlCq8MsYP0yJc2aWddA4LdCIiIiIiMjgyqQSxMe641c8e7+7PwpoT+XCyMMbqWZEIc7USOzyi62KBTkREREREBsvVSoGPpobiXFEN+tibw8bUSOyQiG6IBToRERERERk0iUSCGI+2Hb1GJCY2iSMiIiIiIiLSASzQiYiIiIiIiHQAC3QiIiIiIiIiHcACnYiIiIiIiEgHsEAnIiIiIiIi0gESQRAEsYMgIiIiIiIi6u04g05ERERERESkA1igExEREREREekAFuhEREREREREOoAFOhEREREREZEOYIFOREREREREpANYoBMRERERERHpABboRERERERERDpALnYAHaXVarFkyRKkp6fD2NgYy5Ytg7e3t9hhkZ6ZMmUKLC0tAQAeHh5YsWKFyBGRvkhISMDKlSsRFxeHvLw8vPLKK5BIJPD398cbb7wBqZT3P6l1f82h5ORkPProo/Dx8QEAxMbGYsKECeIGSDpLrVbj1VdfRVFREVQqFR577DH4+flxHKI2u14Oubi4cByiNtNoNHjttddw8eJFyGQyrFixAoIgcBzqAnpboO/fvx8qlQqbN2/G+fPn8c4772DVqlVih0V6pLm5GQAQFxcnciSkb1avXo3t27fD1NQUALBixQo888wzGDRoEBYvXozffvsNY8eOFTlK0mX/m0MpKSm4//778cADD4gcGemD7du3w8bGBu+//z6qqqowdepUBAUFcRyiNrteDj3xxBMch6jNDhw4AADYtGkTTp48ea1A5zjUeXp7SyM+Ph7Dhw8HAERFRSEpKUnkiEjfpKWlobGxEQ888ADmz5+P8+fPix0S6QkvLy98+umn175OTk7GwIEDAQAjRozAsWPHxAqN9MT/5lBSUhIOHjyIOXPm4NVXX4VSqRQxOtJ1d9xxB55++ulrX8tkMo5D1C7XyyGOQ9QeY8aMwVtvvQUAuHTpEhwcHDgOdRG9LdCVSiUsLCyufS2TydDS0iJiRKRvFAoFHnzwQXz11VdYunQpXnjh/9u5f5fU/jiO469DOoQSDTYWqOCQZiD9BUE4dSNoiaChaA0LKySclCiaG/oDXCKwLWp0KWmxIKEf0BKF5FQaVOK5y719u5fgW/fbl3PO9fmY1Ok1vHlzXvr5mGSG8CHxeFwu1z8HkEzTlGEYkiSPx6OHhwerosEhfp+haDSqxcVF5XI5dXd3a2Njw8J0sDuPxyOv16tarabZ2VklEgn2ED7lvRliD+GzXC6XlpaWlMlkFI/H2UNfxLEF3ev1ql6vv75vNpu/POwA/8bv9+vbt28yDEN+v1+dnZ26u7uzOhYc6O39qnq9ro6ODgvTwImGhoYUiUReX5fLZYsTwe5ub281OTmpkZERDQ8Ps4fwab/PEHsIf2JtbU17e3tKp9Ov10cl9tB/4diCHovFVCgUJEmlUkmhUMjiRHCa7e1tra6uSpIqlYpqtZq6urosTgUn6u3tVbFYlCQVCgUNDAxYnAhOMz09rZOTE0nSwcGBwuGwxYlgZ9VqVVNTU1pYWNDY2Jgk9hA+570ZYg/hM3Z2drS5uSlJam9vl2EYikQi7KEvYJimaVod4k/8/Bf38/NzmaaplZUVBYNBq2PBQZ6fn5VKpXRzcyPDMJRMJhWLxayOBYe4vr7W/Py8tra2dHV1pXQ6rZeXFwUCAWWzWbW1tVkdETb3doZOT0+VyWTkdrvl8/mUyWR+ucYFvJXNZrW7u6tAIPD62fLysrLZLHsIH/LeDCUSCa2vr7OH8CGPj49KpVKqVqtqNBqamZlRMBjkeegLOLagAwAAAADwN3HsEXcAAAAAAP4mFHQAAAAAAGyAgg4AAAAAgA1Q0AEAAAAAsAEKOgAAAAAANkBBBwCgxTw9PWlwcFBnZ2c6OjqyOg4AAPiBgg4AQIva39/X5eWl1TEAAMAPLqsDAACA/1+9XlcymdT9/b16enpUqVSUz+fldrsVDocVjUatjggAQMujoAMA0ALy+bxCoZDm5uZ0fHysYrGo0dFR+Xw+yjkAADbBEXcAAFrAxcWF+vr6JEn9/f1yufiOHgAAu6GgAwDQAgKBgEqlkiSpXC6r0WjIMAw1m02LkwEAgJ8o6AAAtICJiQlVKhWNj48rl8vJ7XYrEokol8vp8PDQ6ngAAECSYZqmaXUIAAAAAABaHb+gAwAAAABgAxR0AAAAAABsgIIOAAAAAIANUNABAAAAALABCjoAAAAAADZAQQcAAAAAwAYo6AAAAAAA2MB3C/4Nbin84C4AAAAASUVORK5CYII=\n",
    441       "text/plain": [
    442        "<Figure size 1008x432 with 2 Axes>"
    443       ]
    444      },
    445      "metadata": {},
    446      "output_type": "display_data"
    447     }
    448    ],
    449    "source": [
    450     "fig, axes= plt.subplots(nrows=2, figsize=(14,6))\n",
    451     "returns.add(1).cumprod().sub(1).plot(ax=axes[0], title='Cumulative Returns')\n",
    452     "transactions.groupby(transactions.dt.dt.day).txn_dollars.sum().cumsum().plot(ax=axes[1], title='Cumulative Transactions')\n",
    453     "fig.tight_layout();"
    454    ]
    455   },
    456   {
    457    "cell_type": "code",
    458    "execution_count": 43,
    459    "metadata": {},
    460    "outputs": [
    461     {
    462      "name": "stdout",
    463      "output_type": "stream",
    464      "text": [
    465       "<class 'pandas.core.frame.DataFrame'>\n",
    466       "DatetimeIndex: 2012 entries, 2010-01-05 to 2017-12-29\n",
    467       "Columns: 1717 entries, Equity(0 [A]) to cash\n",
    468       "dtypes: float64(1717)\n",
    469       "memory usage: 26.5 MB\n"
    470      ]
    471     }
    472    ],
    473    "source": [
    474     "positions.info()"
    475    ]
    476   },
    477   {
    478    "cell_type": "code",
    479    "execution_count": 26,
    480    "metadata": {},
    481    "outputs": [
    482     {
    483      "data": {
    484       "text/html": [
    485        "<div>\n",
    486        "<style scoped>\n",
    487        "    .dataframe tbody tr th:only-of-type {\n",
    488        "        vertical-align: middle;\n",
    489        "    }\n",
    490        "\n",
    491        "    .dataframe tbody tr th {\n",
    492        "        vertical-align: top;\n",
    493        "    }\n",
    494        "\n",
    495        "    .dataframe thead th {\n",
    496        "        text-align: right;\n",
    497        "    }\n",
    498        "</style>\n",
    499        "<table border=\"1\" class=\"dataframe\">\n",
    500        "  <thead>\n",
    501        "    <tr style=\"text-align: right;\">\n",
    502        "      <th></th>\n",
    503        "      <th>amount</th>\n",
    504        "      <th>price</th>\n",
    505        "      <th>txn_dollars</th>\n",
    506        "    </tr>\n",
    507        "  </thead>\n",
    508        "  <tbody>\n",
    509        "    <tr>\n",
    510        "      <th>count</th>\n",
    511        "      <td>54596.000000</td>\n",
    512        "      <td>54596.000000</td>\n",
    513        "      <td>54596.000000</td>\n",
    514        "    </tr>\n",
    515        "    <tr>\n",
    516        "      <th>mean</th>\n",
    517        "      <td>6.639039</td>\n",
    518        "      <td>54.526810</td>\n",
    519        "      <td>-125.974292</td>\n",
    520        "    </tr>\n",
    521        "    <tr>\n",
    522        "      <th>std</th>\n",
    523        "      <td>3164.246839</td>\n",
    524        "      <td>70.158189</td>\n",
    525        "      <td>46190.665141</td>\n",
    526        "    </tr>\n",
    527        "    <tr>\n",
    528        "      <th>min</th>\n",
    529        "      <td>-193113.000000</td>\n",
    530        "      <td>0.260000</td>\n",
    531        "      <td>-260289.136266</td>\n",
    532        "    </tr>\n",
    533        "    <tr>\n",
    534        "      <th>25%</th>\n",
    535        "      <td>-798.000000</td>\n",
    536        "      <td>23.870000</td>\n",
    537        "      <td>-50045.888868</td>\n",
    538        "    </tr>\n",
    539        "    <tr>\n",
    540        "      <th>50%</th>\n",
    541        "      <td>7.000000</td>\n",
    542        "      <td>40.440000</td>\n",
    543        "      <td>-401.450000</td>\n",
    544        "    </tr>\n",
    545        "    <tr>\n",
    546        "      <th>75%</th>\n",
    547        "      <td>790.250000</td>\n",
    548        "      <td>65.459000</td>\n",
    549        "      <td>48254.080538</td>\n",
    550        "    </tr>\n",
    551        "    <tr>\n",
    552        "      <th>max</th>\n",
    553        "      <td>312268.000000</td>\n",
    554        "      <td>2846.059938</td>\n",
    555        "      <td>110457.269868</td>\n",
    556        "    </tr>\n",
    557        "  </tbody>\n",
    558        "</table>\n",
    559        "</div>"
    560       ],
    561       "text/plain": [
    562        "              amount         price    txn_dollars\n",
    563        "count   54596.000000  54596.000000   54596.000000\n",
    564        "mean        6.639039     54.526810    -125.974292\n",
    565        "std      3164.246839     70.158189   46190.665141\n",
    566        "min   -193113.000000      0.260000 -260289.136266\n",
    567        "25%      -798.000000     23.870000  -50045.888868\n",
    568        "50%         7.000000     40.440000    -401.450000\n",
    569        "75%       790.250000     65.459000   48254.080538\n",
    570        "max    312268.000000   2846.059938  110457.269868"
    571       ]
    572      },
    573      "execution_count": 26,
    574      "metadata": {},
    575      "output_type": "execute_result"
    576     }
    577    ],
    578    "source": [
    579     "transactions.describe()"
    580    ]
    581   }
    582  ],
    583  "metadata": {
    584   "kernelspec": {
    585    "display_name": "Python 3",
    586    "language": "python",
    587    "name": "python3"
    588   },
    589   "language_info": {
    590    "codemirror_mode": {
    591     "name": "ipython",
    592     "version": 3
    593    },
    594    "file_extension": ".py",
    595    "mimetype": "text/x-python",
    596    "name": "python",
    597    "nbconvert_exporter": "python",
    598    "pygments_lexer": "ipython3",
    599    "version": "3.5.6"
    600   },
    601   "toc": {
    602    "base_numbering": 1,
    603    "nav_menu": {},
    604    "number_sections": true,
    605    "sideBar": true,
    606    "skip_h1_title": true,
    607    "title_cell": "Table of Contents",
    608    "title_sidebar": "Contents",
    609    "toc_cell": false,
    610    "toc_position": {
    611     "height": "calc(100% - 180px)",
    612     "left": "10px",
    613     "top": "150px",
    614     "width": "252.628px"
    615    },
    616    "toc_section_display": true,
    617    "toc_window_display": true
    618   }
    619  },
    620  "nbformat": 4,
    621  "nbformat_minor": 2
    622 }