{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# ES-RNN model \n", "\n", "In this notebook, we demonstrate how to:\n", "- prepare time series data for training a RNN forecasting model\n", "- get data in the required shape for the keras API\n", "- implement an ES-RNN model in keras to predict the next 3 steps ahead (time *t+1* to *t+3*) in the time series. This model uses recent values of load as the model input. The model will be trained to output a vector, the elements of which are ordered predictions for future time steps.\n", "- enable early stopping to reduce the likelihood of model overfitting\n", "- evaluate the model on a test dataset\n", "\n", "The data in this example is taken from the GEFCom2014 forecasting competition<sup>1</sup>. It consists of 3 years of hourly electricity load and temperature values between 2012 and 2014. The task is to forecast future values of electricity load.\n", "\n", "<sup>1</sup>Tao Hong, Pierre Pinson, Shu Fan, Hamidreza Zareipour, Alberto Troccoli and Rob J. Hyndman, \"Probabilistic energy forecasting: Global Energy Forecasting Competition 2014 and beyond\", International Journal of Forecasting, vol.32, no.3, pp 896-913, July-September, 2016." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import os\n", "import warnings\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pandas as pd\n", "import datetime as dt\n", "from collections import UserDict, deque\n", "from IPython.display import Image\n", "%matplotlib inline\n", "\n", "from common.utils import load_data, mape, TimeSeriesTensor, create_evaluation_df\n", "\n", "pd.options.display.float_format = '{:,.2f}'.format\n", "np.set_printoptions(precision=2)\n", "warnings.filterwarnings(\"ignore\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Load data into Pandas dataframe" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/html": [ "<div>\n", "<style scoped>\n", " .dataframe tbody tr th:only-of-type {\n", " vertical-align: middle;\n", " }\n", "\n", " .dataframe tbody tr th {\n", " vertical-align: top;\n", " }\n", "\n", " .dataframe thead th {\n", " text-align: right;\n", " }\n", "</style>\n", "<table border=\"1\" class=\"dataframe\">\n", " <thead>\n", " <tr style=\"text-align: right;\">\n", " <th></th>\n", " <th>load</th>\n", " <th>temp</th>\n", " </tr>\n", " </thead>\n", " <tbody>\n", " <tr>\n", " <th>2012-01-01 00:00:00</th>\n", " <td>2,698.00</td>\n", " <td>32.00</td>\n", " </tr>\n", " <tr>\n", " <th>2012-01-01 01:00:00</th>\n", " <td>2,558.00</td>\n", " <td>32.67</td>\n", " </tr>\n", " <tr>\n", " <th>2012-01-01 02:00:00</th>\n", " <td>2,444.00</td>\n", " <td>30.00</td>\n", " </tr>\n", " <tr>\n", " <th>2012-01-01 03:00:00</th>\n", " <td>2,402.00</td>\n", " <td>31.00</td>\n", " </tr>\n", " <tr>\n", " <th>2012-01-01 04:00:00</th>\n", " <td>2,403.00</td>\n", " <td>32.00</td>\n", " </tr>\n", " </tbody>\n", "</table>\n", "</div>" ], "text/plain": [ " load temp\n", "2012-01-01 00:00:00 2,698.00 32.00\n", "2012-01-01 01:00:00 2,558.00 32.67\n", "2012-01-01 02:00:00 2,444.00 30.00\n", "2012-01-01 03:00:00 2,402.00 31.00\n", "2012-01-01 04:00:00 2,403.00 32.00" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "energy = load_data('data/')\n", "energy.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Data preparation\n", "\n", "For this example, we will set *T=6*. This means that the input for each sample is a vector of the prevous 6 hours of the energy load. The choice of *T=6* was arbitrary but should be selected through experimentation.\n", "\n", "*HORIZON=3* specifies that we have a forecasting horizon of 3 (*t+3*)" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "valid_start_dt = '2014-08-30 08:00:00'\n", "test_start_dt = '2014-10-31 11:00:00'\n", "test_end_dt = '2014-12-30 18:00:00'\n", "\n", "T = 6\n", "HORIZON = 3" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Create training set." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "train = energy.copy()[energy.index < valid_start_dt][['load']]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Scale data to be in range (0, 1). This transformation should be calibrated on the training set only. This is to prevent information from the validation or test sets leaking into the training data." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "from sklearn.preprocessing import MinMaxScaler\n", "\n", "scaler = MinMaxScaler()\n", "scaler.fit(train[['load']])\n", "train[['load']] = scaler.transform(train)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Use the TimeSeriesTensor convenience class to:\n", "1. Shift the values of the time series to create a Pandas dataframe containing all the data for a single training example\n", "2. Discard any samples with missing values\n", "3. Transform this Pandas dataframe into a numpy array of shape (samples, time steps, features) for input into Keras\n", "\n", "The class takes the following parameters:\n", "\n", "- **dataset**: original time series\n", "- **H**: the forecast horizon\n", "- **tensor_structure**: a dictionary discribing the tensor structure in the form { 'tensor_name' : (range(max_backward_shift, max_forward_shift), [feature, feature, ...] ) }\n", "- **freq**: time series frequency\n", "- **drop_incomplete**: (Boolean) whether to drop incomplete samples" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "tensor_structure = {'X':(range(-T+1, 1), ['load'])}\n", "train_inputs = TimeSeriesTensor(train, 'load', HORIZON, tensor_structure)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/html": [ "<div>\n", "<style scoped>\n", " .dataframe tbody tr th:only-of-type {\n", " vertical-align: middle;\n", " }\n", "\n", " .dataframe tbody tr th {\n", " vertical-align: top;\n", " }\n", "\n", " .dataframe thead tr th {\n", " text-align: left;\n", " }\n", "</style>\n", "<table border=\"1\" class=\"dataframe\">\n", " <thead>\n", " <tr>\n", " <th>tensor</th>\n", " <th colspan=\"3\" halign=\"left\">target</th>\n", " <th colspan=\"6\" halign=\"left\">X</th>\n", " </tr>\n", " <tr>\n", " <th>feature</th>\n", " <th colspan=\"3\" halign=\"left\">y</th>\n", " <th colspan=\"6\" halign=\"left\">load</th>\n", " </tr>\n", " <tr>\n", " <th>time step</th>\n", " <th>t+1</th>\n", " <th>t+2</th>\n", " <th>t+3</th>\n", " <th>t-5</th>\n", " <th>t-4</th>\n", " <th>t-3</th>\n", " <th>t-2</th>\n", " <th>t-1</th>\n", " <th>t</th>\n", " </tr>\n", " </thead>\n", " <tbody>\n", " <tr>\n", " <th>2012-01-01 05:00:00</th>\n", " <td>0.18</td>\n", " <td>0.23</td>\n", " <td>0.29</td>\n", " <td>0.22</td>\n", " <td>0.18</td>\n", " <td>0.14</td>\n", " <td>0.13</td>\n", " <td>0.13</td>\n", " <td>0.15</td>\n", " </tr>\n", " <tr>\n", " <th>2012-01-01 06:00:00</th>\n", " <td>0.23</td>\n", " <td>0.29</td>\n", " <td>0.35</td>\n", " <td>0.18</td>\n", " <td>0.14</td>\n", " <td>0.13</td>\n", " <td>0.13</td>\n", " <td>0.15</td>\n", " <td>0.18</td>\n", " </tr>\n", " <tr>\n", " <th>2012-01-01 07:00:00</th>\n", " <td>0.29</td>\n", " <td>0.35</td>\n", " <td>0.37</td>\n", " <td>0.14</td>\n", " <td>0.13</td>\n", " <td>0.13</td>\n", " <td>0.15</td>\n", " <td>0.18</td>\n", " <td>0.23</td>\n", " </tr>\n", " </tbody>\n", "</table>\n", "</div>" ], "text/plain": [ "tensor target X \n", "feature y load \n", "time step t+1 t+2 t+3 t-5 t-4 t-3 t-2 t-1 t\n", "2012-01-01 05:00:00 0.18 0.23 0.29 0.22 0.18 0.14 0.13 0.13 0.15\n", "2012-01-01 06:00:00 0.23 0.29 0.35 0.18 0.14 0.13 0.13 0.15 0.18\n", "2012-01-01 07:00:00 0.29 0.35 0.37 0.14 0.13 0.13 0.15 0.18 0.23" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "train_inputs.dataframe.head(3)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "y_train shape: (23328, 3)\n", "x_train shape: (23328, 6, 1)\n" ] } ], "source": [ "print(\"y_train shape: \", train_inputs['target'].shape)\n", "print(\"x_train shape: \", train_inputs['X'].shape)" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "array([[[0.22],\n", " [0.18],\n", " [0.14],\n", " [0.13],\n", " [0.13],\n", " [0.15]],\n", "\n", " [[0.18],\n", " [0.14],\n", " [0.13],\n", " [0.13],\n", " [0.15],\n", " [0.18]],\n", "\n", " [[0.14],\n", " [0.13],\n", " [0.13],\n", " [0.15],\n", " [0.18],\n", " [0.23]]])" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "train_inputs['X'][:3]" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "array([[0.18, 0.23, 0.29],\n", " [0.23, 0.29, 0.35],\n", " [0.29, 0.35, 0.37]])" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "train_inputs['target'][:3]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Construct validation set (keeping T hours from the training set in order to construct initial features)" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "look_back_dt = dt.datetime.strptime(valid_start_dt, '%Y-%m-%d %H:%M:%S') - dt.timedelta(hours=T-1)\n", "valid = energy.copy()[(energy.index >=look_back_dt) & (energy.index < test_start_dt)][['load']]\n", "valid[['load']] = scaler.transform(valid)\n", "valid_inputs = TimeSeriesTensor(valid, 'load', HORIZON, tensor_structure)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Implement ES-RNN" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We will implement ES-RNN forecasting model with the following structure:" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABMsAAAMwCAIAAACx/XChAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAADsMAAA7DAcdvqGQAAMafSURBVHhe7P0NnFVlvf//D96QAupAipqmM5o3JcXQjejRkqE79UDOeL4JlcVNeSiLI3CsOP0PAnFOUXkAv2TFsQNoVqC/I2PDNyuLGU2zIWtGG7zPGW9QGA0GZFDR5P9mfy6uFvtu9t6zb9Ze+/V88Nhc17XX7L3Xuq611vVZ17oZtHfv3ioAAAAAAAbsIPc/AAAAAAADQ4QJAAAAAMgPIkwAAAAAQH4QYQIAAAAA8oMIEwAAAACQH0SYAAAAAID8IMIEAAAAAOQHESYAAAAAID+IMAEAAAAA+UGECQAAAADIDyJMAAAAAEB+EGECAAAAAPKDCBMAAAAAkB9EmAAAAACA/Bi0d+9elwQAAAAG6IF6lwBCa3SLS6AAGMMEAAAAAOQHY5gAAADIn9gY5o9e+YnloufTh31Sr1GdwWjPndgMMoZZUIxhAgAAAADygwgTAAAAAJAfRJgAAAAAgPwgwgQAAAAA5AcRJgAAAAAgP4gwAQAAAAD5QYQJAAAAAMgPIkwAAAAAQH4QYQIAAAAA8oMIEwAAAACQH0SYAAAAAID8IMIEAAAAAOQHESYAAAAAID+IMAEAAAAA+UGECQAAAADIDyJMAAAAAEB+EGECAAAAAPKDCBMAAAAAkB9EmAAAAACA/CDCBAAAAADkBxEmAAAAACA/iDABAAAAAPlBhAkAAAAAyA8iTAAAAABAfgzau3evSwIAAAAD9EC9Xn70yk8sFyp/+vmPuh+8T4kzzr3wrAs+ZoXZ+vRhn9RreGawq/23Pd2P7Hzx+Zf+usVKRryl9piTz6gdc/6QI0dYSebCNneSdAaHH3/yaWM/lPMMVo1uieVQEIxhAgAAIOK2Prnpzhu+buGlvL7nFUuUNc3UHdd/rf2XP938aLuPvmTbc12P3veLu350rSZwReUpzQz+5Y+tv/7hf5T7DEYVESYAAABKZvvz3T9bMue2xVfes+b/uqJ8e+DOW+695fpgiFJMhZvBh+/5+csv9Spx+BHVNe8699T3jNM/pe1dvdXW9D+7d26zbOF0tf9Wc6d/m+76mSvKk+AMnnDGGJvBI958nL37+p5XijODyBYRJgAAAEpm1/YXbUTx9T2vWkkeKbq784av/+WPrUorShlZc6aVF1PhZvCQwW9S3PW+j02/6IvfePfFnx794cv0T2kV2gT63sfbfm3pwtn54vOWKMTIsOblvMu+qJka23iFzeCHr7hm1LhGe1ff2NV+j6URHkSYAAAAKCdbn9z0wJ23PPb7O10+td/f9t82dKnY8oJPX+2Hv0Iuwxk8f/K/KO566zve6/L7qdCPZPZ0P2KJUOlq/61m8JmH7nf5FMZ95sual2NPOcvl9zv9nA/7gwUvPPWoJRAeRJgAAAAoJ1v+sukvf2ztbF3n8qm9/FLvIYMPG/PRTygYy+GuMKWS+Qym8pbT6yxRqnOD03vqz7/XDP7l/txvt1MuBwsqExEmAAAASuCO67922+Ir//CzlZbd9lyXXc5n/zIZouzXqHGNH/rcv9eOeb/LF1cRZrCEtj/fbTNiJyGLEn7u9E8TWHlBHTL4TS6F0CDCBAAAQAm89mq6y/Zee/VllxqA08/5cAmHLoswg/0q3Fjfnpf7XCqFficYiOce67DEyJq3WwLhQYQJAACAEhjb8Nn3fWz6GedeaFnFQsr6f7Vjzrfy8lXCGdz+/FOWOPLo4y2Rd8eecpbNiL+xkN12yP9LvH4yX9rW3WD3mNUiPf2cD1shwoMIEwAAACWgCOSt73jvkce8xbKHvulwZf2/MrpsMpVSzeD257u3Pddl6eHH11iiEGxGDht2lGWV8HOnf1Y4cFuf3PTMQ/fbvwfuvOWO67+2+dF2lSu8PG/Sl2wahAoRJgAAABAd/grPw4+oLvchPkXL995y/R9+ttL+/eWPrS+/1Kv5OvU94z58xTUROAwRSYP27t3rkgAAAMAAPVCvlx+98hPL9euZh+63e+GMeEvtuM982QqDFC+9/NJ2l4nZ/vxTNkanMMNKvJNGnZ1m1O6BO2+x29LoD0d/+DIrzNanD/ukXsM5g+K/Tt73senZjiVmO3fS71LVBC6133OPdVig6O9568V9wu6d2379w/9I+qRN/fmo+ktzm8Gq0bnfxhb9IsIEAABA/uQ7wrxt8ZUulYH0oWM4I8w8zqDisbt+dK1do3jCGWPGNl5h5ZnLe4QZjHgzkT4q3vrkph09z/3ljy02jzJqXGNW47REmEXAWbIAAAAIr8OPqHapDBTuxjaFk8cZvHftd/0tcN75wX+ywtIaNvxol8rAIYMPSz/9saecpXjyoi9+w99eqLN1XXEei4LMMYYJAACA/Mn3GGYiP2h26dzvWUmGwjmGmSi3GWxbd4PdAkdx2oc+9++5XaOY9zHMpFpv+s6257oyXyCJgmfP1rzr3Hdf/Gkr7xdjmEXAGCYAAABQ3oLh5diGz0b+FjiawWNr3ZMwd764xRIICSJMAAAAoIz96ec/CoaXhXsQZaj4p6QgbIgwAQAAUDJZXadXjgo9g4/9/s7uB+9TolTh5eFHDHep4nrpr27o8pDBb7IEQoIIEwAAACXjn70R98SOyCjoDCq87GxdZ+l31jeWZPTy8CNdhPnKrh2WKILtz3f3dD9i6fSPb0HxEWECAACg9F5+qfeZh+53mSjK+wzq03x4OWpcY+2Y91u6VLZ2Pbx75zaXGTAFz3dc/zW9Jn6mwsv7199k6UMGH1Y75nxLIyQOXrBggUsCAAAAA7T1Rr08+HoWj8r467N/6et9UYmerof7tvf0dD/a/osf733jjTefeKpNEEfT//XZJxRanHHuR11RCl3tv326s23rk5vs3/bnn7KHeQwaNGhHz2ZffuQxxx/6psPtT/o1+pD/1WvJZ1Bx1+9uuf6Nv72u9BFvPu5NQ4b52Yn7t/OF51J9UaIc5u6oY97y+Mbf6Jfo3+ZH/rR7x1+fe/RPf2i+cWTNGakexPLsw3/UAtHPPmnUWFeUQNP0dD+if0/8YcPWJx/y9fXIvT9/+J7/t+flXTbZ6A99fGTNmZbOhM1g1XFTYzkUBE8rAQAAQP5k+bQSUdjQ1vQ/9uQJ74xzLzzrgo+5TK7sqRguk1r6p/zHyeF5HoWYQf8QlExk/tSTHOZOgifreudd9sWBnLW7/fnu39/233ZEIClF4O+sz3rklqeVFAFnyQIAAKCUFIeMbfisH4lS5KB0Xk59zPAeMIMPy3QAMzeFmMFhw4/W57hMWqkGEvPo9HM+POajnzjizcdZVt9Y865zB3hR6PDjay764jf0sfqoEW+pdaX7l96ocY0f+ty/l/zEYCTFGCYAAADyJ/sxzPKS2yhfuYj23AljmEXAGCYAAAAAID+IMAEAAAAA+UGECQAAAADIDyJMAAAAAEB+EGECAAAAAPKDCBMAAAAAkB9EmAAAAACA/CDCBAAAAADkBxEmAAAAACA/iDABAAAAAPlBhAkAAAAAyA8iTAAAAABAfhBhAgAAAADygwgTAAAAAJAfRJgAAAAAgPwgwgQAAAAA5AcRJgAAAAAgP4gwAQAAAAD5QYQJAAAAAMgPIkwAAAAAQH4QYQIAAAAA8oMIEwAAAACQH0SYAAAAAID8IMIEAAAAAOTHoL1797okAAAAMEAP1LsEEFqjW1wCBcAYJgAAAMpb93OvuFTURXhOK6cSI48xTAAAAJSx7u7uMWPGtLe319TUuKKIivCcVk4lVgLGMAEAAFDGrrvuuurq6tmzZ7t8dEV4TiunEisBESYAAADKVWtra1NT01VXXdXR0aG0K42iCM9p5VRiheAsWQAAAJSr+vr6KVOmTJs2bd26dbNnz+7q6nJvRE6E57RyKrFCMIYJAACAsrRs2TK9Tp06Va8NDQ01NTWrV6+OvRM1EZ7TyqnEysEYJgAAAMpPb29vbW3tunXrxo0bN2jQvj5tVO8WE+E5rZxKrCiMYQIAAKD8LFu2bOrUqYpMlLYRMMUkDQ0NCxcujL0fHRGe08qpxIrCGCYAAADKTHd3d319fXt7e3V1tSuK6e3tHTNmzLp16+rq6lxRmYvwnFZOJVYaxjABAABQZqZNmzZlypS4yERUMn/+fL3r8uUvwnNaOZVYaYgwAQAAUE6ampq6u7tnzZrl8lVVwXvD2MmWmsayZS3Cc1o5lViBOEsWAAAAZaO3t7e+vn7p0qV28Z6xm8S4TFVVR0eHptm+fbvLl6cIz2nlVGJlYgwTAAAAZaO1tbWmpiYYmSSqq6vTBOX+0IsIz2nlVGJlYgwTAAAA5aS3tzfu4r244S+TOFnZifCcVk4lViDGMAEAAFBOEkMOu2wvTgQikwjPaeVUYgViDBMAAAAAkB+MYQIAAAAA8oMIEwAAAOWtcu4HE+E55aY+kcFZsgAAAChvSW8SE0kRntPKqcTIYwwTAAAAAJAfRJgAAAAAgPwgwgQAAEB5S/qgi0iK8JxWTiVGHqc7AwAAAADygzFMAAAAAEB+EGECAACgvPG0kgjgaSWRwVmyAAAAKG88rSQCeFpJZDCGCQAAAADIDyJMAAAAAEB+EGECAACgvPG0kgjgaSWRwenOAAAAAID8YAwTAAAAAJAfRJgAAAAobzytJAJ4WklkcJYsAAAAyhtPK4kAnlYSGYxhAgAAAADygwgTAAAAAJAfRJgAAAAobzytJAJ4WklkcLozAAAAACA/GMMEAAAAAOQHESYAAADKG08riQCeVhIZnCULAACA8sbTSiKAp5VEBmOYAAAAAID8IMIEAAAAAOQHESYAAADKG08riQCeVhIZnO4MAAAAAMgPxjABAAAAAPlBhAkAAIDyxtNKIoCnlUQGZ8kCAACgvPG0kgjgaSWRwRgmAAAAACA/iDABAAAAAPlBhAkAAIDyxtNKIoCnlUQGpzsDAAAAAPKDMUwAAAAAQH4QYQIAAKC88bSSCOBpJZHBWbIAAAAobzytJAJ4WklkMIYJAAAAAMgPIkwAAAAAQH4QYQIAAKC88bSSCOBpJZHB6c4AAAAAgPxgDBMAAAAAkB9EmAAAAChvPK0kAnhaSWRwliwAAADKG08riQCeVhIZjGECAAAAAPKDCBMAAAAAkB9EmAAAAChvPK0kAnhaSWRwujMAAAAAID8YwwQAAAAA5AcRJgAAAMobTyuJAJ5WEhmcJQsAAIDyxtNKIoCnlUQGY5gAAAAAgPwgwgQAAAAA5AcRJgAAAMobTyuJAJ5WEhmc7gwAAAql95Xeji0dtz96e2t3q9L2z70HACgrNdU11YdV1x1Xd8kZlzSc2eBKExBhAgCAglBUOfuXsxVhujwAICoUZ666ZJVeXT6ACBMAAORZ7yu9C+9auLpjNSOWABBhs86ZtfSjS11mPyJMAACQT4oqp90+remRJpcHAERX3XF17TPaXSaGCBMAAOTTgtYFC+9a6DJVVcMGD734bR+6+G0fHDZ42BGDhyrr3gAAlJXnd/U8vu3J9i1/vuWh211RTNxIJhEmAADIm9bu1sa1jf7k2NNGnPL/O3+WXi0LAIgAxZn/ec8yvbp8VVXLlJZxNeMsTYQJAADyQ4Fl/Y31/tY+Ciy/e+E3GbQEgEia+rN/8UFm9WHV27+63dI8DxMAAORHa3erDy8VWF519hWElwAQVd8c/+8uFTvCuLpjtaWJMAEAQH7c/ujfr8z5wEnnjjnunS4DAIic44eNvOwdl7hMYBdAhAkAAPIj+OjL9590jksBACLqs3WfdKnYaSyWIMIEAAD5EXz6JXf3AYDKRIQJAADyo7u326ViZ0+5FAAgooIX2/uDjESYAAAAAID8IMIEAAAAAOQHESYAAAAAID+IMAEAAAAA+UGECQAAAADIDyJMAAAAAEB+EGECAADg7y4cPfm8mgkrvnOTywOR0PSTO9Sw1byfe3qLK0JhEGECAIDQ+ezHZqsvmPhvzmeuUeTz4B8ectOhAF7asUuvj/75Ccua+1rutypQN90VlVTYfk+B7Nj+kmZQzd5mVv+UVonK3RTIWN/O3XpV835x6zYrMba1ueyCK0KyVLWJ0+9RJFy+GzoiTAAAEDqPPPi4Sx2o7e4/3XT9LV/4+FfUz37i4S5XisJ78tGnLHH3L+6zRGmF7fcUglr4pHFXfOdr16vZu6LYKqASlTMQly+2tdn81PNPPfGMlZSWHdxRJPznPz5sJWWHCBMAAITUme867fu3ftv/W7j8K42XX3zEUcP0lvrZX5r8bwxmFs0pZ5xsiYs//iFLFNq1//49G8lx+QMV//cUmQJItXAbT1azv3bVAlsFPjjh/SpJHIhDzrSd0esJJx//rve9w0oK6omHu+xE9FTD72e882161YZu7AXvtpKyQ4QJAABC6qjqI9Tn8/8+NPEDV//HlWtbb1CHW++qk/2Vz32dkcziOLf+vfd2r9c/1YIrKjAbo7MQK1Hxf0+RNa/9lc37lXOnqdlrfm0V+Pp3v6pQU0HRkGGH25QYoP/52VI1pFvuusHlC2z3rpetZu2s3UQzvvwZ/Z5fPLDmbW+vdUXlhggTAACUk6OGH6EOtw8yv/fNVVYOhNATD3flNtLur4OdMOkjlvAUaiooKt/wA5FHhAkAAMrPFf/66RNOPl6Jtrv/xDAmQmvKRTO/8PGv/PgH/+vy2Ttq+BEuBZSJQXv37nVJAACAARi0cJBLVVXdO3W9S+XkvJoJeh37gXcvuenrVpKo6Sd3fOdr1yvxmS9eNuPLn7FCb8f2l1ruuOfuX9zn75KiT7v44x9KelLlNV/61m/W//bKudM+9fl/eu7pLT/579t+3Xy3ncn2wQnv/z9TJr4r9QVaD/7hoV/d3rrxt+2bn3pe2SOOGnb2+8d84KPnJv0i+802XwqMv/fNVfbzGi+/+Or/uNKmCf4YTdP04zvsx+iT9Zmf/OdL33LScTalPq15za/sPiX6zI9P+9i59e+1t4I0mZbDjt6XgvdPSrM0ki58/RIFS0pcu2qB/xYFTt9bnHwMWb/2Fw+scZnYn2tGtGwfeuAxW7BywsnHa1kF50g0zcc/8DmXSXDjHctt7C7p7wnSQrv7l/epXuzr7Ls+csm4pFU5kAaQni1M+2QrydCcz1xjbWPh8q8kraak9OOb1/5Kc211ba0x1e/PtmGIWvv/d2OzX6qa+AMXnlt/0fmJYXDiCnjmu07Tj5k46SPB6vYuHD1Zn/n9W7+tn2rfohpReWKz9zJvVMY3V/sWKxSrff28//nZUlcUWP6J1CS+/t2vukzstsZ3/O9vdu3sC06vTxt/8fkTJn0kuGSu/ffvrbv55y5zoOBn2lZCM76yeVniXGS1YP1qcm/3vg2yWvVPb1hn1a0FdcknLoz7hbk5b/W+Rm72zt8XWjKGCQAAytKoMWdaIu65GqJ+1RUNcxJvwjl/5rc/+7HZic8kUO/QEurbKbxRL9B3WNX1/MLHv6KemWWD9Dnqm+pdTW/hpegP9SepvshfeaVf+KXJ/+Z/XrDfGfwx6h36H6NXpadPnKW/1T99vmbQxwb6qKunLUj8neqs23IIRhFiS0PdaJfvz+5dL1vC38Q1Pb8AjWJp/Xh9abBcC83mSH10V6Qv6nNflJT/GWl+jy0czZ0qwn+dfZcqS718KwnKrQEUlL+DkX5wcPmkoR+phXnT9bf4utZc2O9PnOscGoY+RB8VXKqaWB+yfu2vLOvpByfeBVdfpN+mxZt0Ydpn9u3a7b/Fl/tmbyVe5o0qPav9uOWQhm8tRuudfm1wTkWfpmhWWyEFwK6oqmpnb8qnoQQ/M9VTVSTbBetXE/0MWyn8bGpB6RfOmXJN4mZq4BjDBAAA+VHkMUyxycSO0Bt1mNQJU//siKOGXf0fV77v/DFHDT9CHayWn99rIxhxgxVi4xWa3rqqjZdf/JFLxinxq9tb1VtVQm+tbb0h7mC/H+U44eTjPznj0lNO23d30ycff+onK26zgDPxi2wURdOrr6nvsmGE55/d+vCDj/spk/4Y9bx/uPTH1jvUx+rz9a4Sn7iiceRxR9/Xer+6mLG/rrr17h8GhzIslD37/WPOGPW2d77n7VYY/JGJQ2RJF74CEvX7lQgOx2lRxz3gQb9THW4l4v5cofjmp7foZ5x6Zo1+sEo05b2/2WiLV8sheJ8VfZder1t0g83v92/9dqx4Hz/0lPT3iK99pW04zr5Oi0jfZeWJg965NYBM5DyGKQoJfDygeZk6c3KaoVS/QNQqPjf7Uzauq9pXw7BoLW6us20Y/vO1cPwIoQKeW1f97D3/MDo4d35K0cTnffDsocOGBKtbEkeebUFZFej101/4uH6V/kqfb2tZ4tYgq0YlqcYw/Yoc3IxoixEX4P35jw/bn8fV5oWjJ79j9OlaCKeccbLmVCU9W178+a2/ts/UEvCnJ9hnaiErRLS3rJnJ0ceO8Kttqt+Zw4L1f2ILVsvkn6/+tJaVfuF/X/sjq+jcGmdQ4hgmESYAAMiP4keYl11whXWSgl3Da/efiubPqPTs9DMl4vphvoupfti3f3hNsFfnP+3L3/hiwycvskL5dfPd82fui3z0I+df9+Vg7KEgZ86Uayw2iPur4Gmlwa5nUKofE4ydJO7P/SfHfWMq6uzayaiJCznzCDORnXCoROLCT8ov3rjOtCTt93upfo//wMQw0mIqW4BxPy+HBpChgUSYqvGFV33HfphRpaQ6F9rWBYUQNzQtiYuEfaX8vP2n/QbJqRqGb2D9fohfKxPDSL/WaDkHT6IWW1CiCHnRd78aPEriP3CAjSqrCDOO6uKKhjmplnBSSbdO0u+qlOp35rBg44LSK/710/6X+4pOrItscZYsAACIjhNjN/sJUkfQOpfqTiV2Rn2EcO9vNloiznfXfDMuzmn4lPuTJx464CS9n96wzhIK8+K6m8qqi2zp5jXxJxCaD054f9LwMijux+hjNVOWToxO6y8+zxJ/+t2DlkhPPXiFEErsSH3mXrbua7nfD5dlEgmIH8Pp25X8yQ1Z8bWvMCDx0lz9JL/Q7HcmyrwBJKVeu/r0wX9W/vyzW5OWp6caV4y3cPlXNDtWokDo6mkLFBHFnTKqJW+xxz9f/fcQwvs/UyZa4g/3tFsijQE2DP9L1EQTI+EPTfyAtWHF+ZrSCoM0p0tu/HowvJRPzrjUEt0HDpinkt9GZdb8cJ3N1+wFMxKXcFKXfOJCS6hZWmIgBrhgbYMT/OVayCpUQn+idmuF+UKECQAAylVweMc81PGoJc774NmWiGO951Q9qsS4yJcE/0RdRhui1KfF9YaNCs+MPcbdn+UYx3f600j8MceecIwlfB/a8z8j7iKxfqX6hdnSMlk461olFCRM/lyjFWYuw8s70/O17zv3cd53/hhLJF67azJsAKlcG7uGMPjPyhX3xpVnGGSKgodb7rohLs780uR/CwaZD97vPs3PYJCPmbdufsESmYhrGL7tzZlyTZof739JYhM1fsVMWuMnnnx8Yvxm559LVr9f8tKoRPNrZ6ErJEs6gJxe3JnkuRnggj1j1NtcKsAXJl7wOUBEmAAAoLwdcdQwlwr0rq6etuC8mgmJ/xKD0hz4LuN7/mG0JRKdsD/kyzyW6JfvamdL0dGK79w05zPXXDh6cn4Xhfdf875nJ6D++7WzUw3y/Lr57mu+9K3Pfmy2/w2Ktdx7+eBr319YGEc/zFpLfufdG3bkUJfqz5Bhh7tUZnycab9fi/prn/+GvSU+YL54zCf8sg3+s3cTZd4w/CiZIk/VmipRtZk4Oud/SdxQsHfMcW+2xB9/94AlBqjQjUrz+B9X77tGWkv+XxclP+9A0zT95A4txssuuML/DH8+fF4Uf8EOBBEmAAAoSz5ye8fo0y2RuaQDj/mVdNCgJBRCfPwDn7vp+lsUOVgQmHf3Bc6PTdoDfuLhLnW+58fu75qvUdPc5NBaMvf173713u71wX9WfuXcaXHliYOlmVCY990137Qgc/NTz2d78MKPQ5psG8bV/3GlH0pVJao2J4274sdZPuoztxlPqjiNyp8fO3/Z1UkPnSjE9fd3tSlLIo8LduCIMAEAQFm6r9VdbpR0IPH7t347rk8f/NfvNZAD92inG3PIdrQqv5p+coed4Hfmu05TePDz9p/6hWAnDA9cv+fHaoKvff4b1vlWCHrjHcv9b1A12TRF89ADj+k1OO5dXhRI2Fii/PmPD1vC8ws26T8FqG66XBuGDaVeu2qBv+rve4tXXZvsATCp+JN7Mx/vTao4jarf82M1OwpxtRzUor78jS/eevcP/c+4cu40N1FR5GvB5gURJgAAKD/qX9o9XcTf5EZOOcOdR/rk4/m5BCspHzT6MDLR5v2X7ZV2bOEnK27Tq7q/S278usKDVOevDkS/58f+4Z52iwTU557x5c8UboEMPXLfsyIkVe2r2dhPLehIZqENOyI+hDjjnW7A3IcZ/RpIw1CsdfV/XKloyi421proL1L14U2qX/LClr9aYoCD/EVoVGot/Z4f2/TjOyzx7R9e0/DJiwp3ckTRFmxeEGECAIDy46OaD054f7BXd/KpJ1ri7l/cZ4lCUHfWBsF+s/636oZaYZA63Hbant2tsYSsF66AqhCxpfR7fqz4G7SkujwyX0aNOdMSqW7h62+m+oELz7VEOXr+2a2W8Mvz1DNrLNF2V6bXlw68YWi9+9zsT1na3yrm3f/wLktYq0jkb+M89oIBDaEXoVH1e36s+NA6VePPl6It2LwgwgQAAOXkwT889NmPzbZulsK8z39lipUb9XrtHL+2u//06+a7rTBO5uM8aXz6Cx+3hGJdS3iKOed96VuWzuSesUXwbKyjHKSFk/SGLlnRnGZ1/9ieLS+6VIz+/LpFBzwTPyiHcTlF/lb7Cu+bfuIGlzx9jp3PqWZTf9H5VhhOauSXXXBF0lvpaL5s9F5z4aOa950/xg55/Oj7t6ZaXEnLM28YWukSF2nibUu1YO2X3HT9LYnfqA+3H69qyteoY1aNKnP9nh8bx4eaRvOuunCZA538trdaIs0ZEImKv2AHgggTAACElLq/P/7B//p/Cg/Uzf3Cx79iw4Pqb313zTcTT0u78t/c5U/zZ357zmeuUd9LnUX9u6/l/hXfuUmfMOWimcraNDmbMOkj1uFTrKtgQJ1v+xYlrmiYY7+w8fKLCz2y0S+7Xm7zU89r6VknWK9Ka+HE3h+QG/7rRzaSfMknLnzqiWdsCfh/PjrypzHre/3D+lQvfkEl5cflvvnV/6u/0gf6WUhj6szJlvjO164P1r6q/kuT/81+bdyDAQvKGkncLXYyoSpTHV085hOaC2v/mgW1NM2XTTB/2dWWEM3O57+671CLZlDN25azzbiWgLL6Q5XbxCbbhqGaSlykdrvUE04+3rdz/RJ/8CX4S/TqP1zLxK+kOcutUWXOzo+Vi/7pgza/wX/2llz88Q9ZYt6XvmWBn5q9Kss3tkRaRHa3JG06NKU+TT9bCXs3laIt2LwYtHfvXpcEAAAYgEELB7lUVdW9U91dNHNzXurnK5ixH3i34oRUVz2p17Vw1rWpenhy4x3Lg0f61W+2cZt799/8M8h+jL5xyU1ftxKjDqW/10iiz3zxssSH/qsfaZ3y79/67VTBZ5ofow6lPYkh6Z8n/Z0KG6ZPnJW4KNQZPfv9Y2woOO6Lkn6O/+or50771Of/yQr9T03qy9/4YsMnL7K0ohEbEYqjOMfGXoIfa9RTV7QQt3j9jCf9PcZiqqS1r7lWswne8Mbk1gAKJ1WtmVRz4VtXUvqrXzywxmWybxipalDTf3fNN+PGzVJNLAquvvGDryWOs6VZyKnqOttGlWrtS1r79ntSCW5APvux2UljWv8zEtfWpp/c4Y8UmOCMp9lKZLtg06wmksnmKBPnrf77sto7f19oyRgmAAAIHbuDSBz1ZdUPU1dJ3Tv1xlKFl3Ju/XvXtt6gCEfTu6JYD8z+/Na7fxjXD7O7aCT9UtH36jXxDo36kBualiR+i3qW6rElhpdit6LRBx597AgrSZTmx/g7DCW9P23S36mlpEWhcNeGTUSfrIWgQruyy/4qyEr8SarGf6O/l5IkLpOg4A1ptDSuXbXAX5Wqr1BaS0mRkn2dv0mPd9TwI5b9aJEWpk1gf+LPMEz6e4xCr5XNyzTLwWWotEpUnhiYSW4NoHCs1hYu/4pm2VecqKWpvaWaC4UQatuJM64P0V/pA11RTLYNQzWo+tJH+XI/fWK4aBOr7hJ/vFaZxOnFPjau1Rlf13GNJNtGlWrtS1r7fjaTGjL07yvg//xsqebL/7ltAVQRDZ9yh1cS19aGT16kyvV/ooQfC5U0W4lsF2ya1UT88kn8hQPEGCYAAMiPPI5hAgDKAmOYAAAAAIBCIcIEAAAAAOQHESYAAAAAID+IMAEAAAAA+UGECQAAAADIDyJMAAAAAEB+EGECAAAAAPKDCBMAAORH9WHVLlVVtWtPn0sBACoJESYAAMiPmuoal6qqen7XVpcCAETU49uedKnALoAIEwAA5EfdcXUuVVX18yd+41IAgIj67dO/d6nALoAIEwAA5MclZ1ziUlVVtzx0e/DYNgAgYp7f1bP2odtdpqpqyugpliDCBAAA+TGuZlzwRNl/2/CfLgUAiJZde/r+856l/pJ7bfy1C7A0ESYAAMiP6sOqV12yymVil2JO/dm/PL+rx+UBAJHw+LYnFV62b/mzy1dVLf3oUn+zt0F79+61FAAAwMDN/uXsZb9f5jIxl73jkovf9sHTRpzi8gAG7Plntx5/4rEuAxTFrj19z+/a+vMnfvPzJ34dvGH4rHNmKcJ0GSJMAACQd2NWjOnY0uEyQBH0VlX9/Vk5FeCVqqplVVWfr7C5RijVHVfXMqUl+LQqzpIFAAB51j6jfdY5s1wGKDSFlz+IvVaO1liQ+QuXA0plat3UuPBSiDABAED+Lf3oUsWZwRv/AIXy+woLt7qrqh6pqjpsfyIywnyMQA1M/xCgqHLdpHWrLlkVF14KZ8kCAIACWt2x+q6n7mrtbu3uVXcYyDc1q6b9Z8k2VFVVwjGN1VVVZ8aGMTW/mve5rrjsab7qYv9CSEtbbUwLvLIpmNS/muqaKaOnNJzZkBhbGiJMAAAAlKv6+vopU6bcfvvtep09e3ZXV5d7I6Kampo0m+3t7b29vTU1NY2NjaNHj16wYIF7u2xpvjQv1dXVqkG9utJw6O7uHjNmjBa4fpuWuStFakSYAAAAKEvLli1TbNnS0mJZizanTp1q2ehRkKN5XLp06bhx7sGDKhk+fHi5Rz5+vm688UZlV636+0OPwkC/TUGmElrIvrEhDa7DBAAAQPlRWLJw4cL58+e7fCwymT17tgUDkbR69WoFOT68lOrqai0BzbXLl6dly5bZfCnIbG1t7egI0Z2om5qa1NIaGhqmTJmihLLuDaTGGCYAAADKz4IFC3bs2KGYRGn1+xUDKDFt2jS9hm0QLC8U3owZM2bdunV1dfsuVVS6paXFTiitra3VcrAlUHa6u7vr6+v9fCmKvvHGG0MyVOgHV7WcRT+1sbFx+/bt7m2kQIQJAACAMmNhSXt7u4VYgwa5Pm1cGBYlwYhaFFUqDLOTY1tbW8s38pk2bdrJJ5+suVPiggsumDp1qmpw/vz5YQiY9aseeOABNSeXr6qKzIWvBcVZsgAAACgzikamTJli4WWQShSc2EhmlCiivvHGG4OnBAeNiynHsEexscyadcDjcxVFh6EGky7zVatWXXfddXrL5ZEMESYAAADKSVNTk7r4cWGJZ3f6idj1cqkiaq9MIx+7kjZuvkISMOu3aZnbYLia0+rVq5XQT73qqqvK/cLXQiPCBAAAQNnojd3gRwFVMCyJO6NS70ZpGLO1tTUxovZnCBulFapZFFQuli1bptek9/5VDaqWSxgwxw2udnR0+B+jQmVDdTuisOE6TAAAAJSNpqamG2+8MXhpXFKNjY2XXHJJZJ5corg6bqCv3GmOamtrVY/jArfGDUq8BrKY6g988o0NqPphVTXCSnj4as6IMAEAAFBOMgy3oheVRYmiteCNi5JSCLpq1apUIWjhLFu27K677goGt3ERpsSFoAgiwgQAAEB5a9r/tJLKoQhH0ZfdS7YcKXqMO9E3kar1uuuuK/6TS8aMGaPQNxjZdnR06KcGl3Z37G7GDGMmRYQJAACA8uafVlI5FAUpwizfh7IkjjBP2/+0EpePYSC6HHGnHwAAAKDMKO5S9OUyZSjDuJHwshwRYQIAAABASv5pJcgEESYAAADKW94vwuzq6uqMcflkbJpSXYm3bt268j1Ftux0BJ5Wgn4RYQIAAKC85f2ZFn19ffNi2traXNGBFFjaBKWKMKtjXCYSVq1axa1Zo4EIEwAAADjAqFGjxo8fr8Ty5cutJM7KlSsVhY4dO9YmA+ARYQIAAKC8NTU1uVT+TJ8+feTIkQoj16xZ44r2a25u7uzsHDp06MyZM11R0TU2Nra2troMCqyhoYHx1cwRYQIAACCdnp4exTOiyMoVJVi+fLkmmDNnjkIyV1RE+mqXyh8FkAoylVi7dq2WgBWKZlAlSuhdTWOFJVHW95JNNG3atNDeTaeurq58Hz1afESYAAAASGfkyJGTJk1SIi7W8tra2jZs2GAhWWmDrvwaG6NE8FxZLYTyPT9WNSVJK9Fo1myakhwpQDQQYQIAAKAfkydPrq2tVdSxcuVKV7SfH9ObMGHCqFGjrDAy7DzYzs5OBV2WaG5uLu35sQPR0tKyePHiOXPmuHwCzZ0mSKzlCpfz00rUYNasWWM3hdKCXb58+YYNG9x70UWECQAAgP5ZTKVAK+7uqYpJVKL4c+LEia6o6PL+tBJPweTcuXOVUGzgA+xJkyaVfKh21apV48aNc5mMqRLt4tKkMaTqcf369UpEbCx64HJ4WokWstqMAsu1a9cqzhStOwovVahQ000UUUSYAAAA6J9iSDsvNNg/7unpUUyiaKS0MUnen1YSNHbs2FGjRilg0IwrBlO2hLG0l9vTSqymlGhubk48V9ZukKu5s3ODiyxKTyux8NJOHZ80adKiRYuWLFkyd+5cpRXhu4miiwgTAAAAGVFwov6xIhN/pp+N7CnyjN75sUH+XFkFDMU5P1ZLVV+n5axXpV1pPih6tABSdWclRjGnvstfc4uBUMW1tbWptSi2nDx5staO2tpaLXalV6xYMWHCBDddRBFhAgAAICM+vlq5cqXiTItJ1HUueUxSiKeVBPkTgxXspblNTl7oK7R4Z8yYMW/ePCX0evnllyee1Dp79uyc59oHzHZxqehLQ3KD3HDK9mklmzZt0qtCSq0dVhIU7cMxQoQJAACATKlzrH6zRUGKSRSNhCEmKcTTSjwffSWeJBxHIbe4TK70CRs2bFDQfnPMunXrtIRVGPe9vTEukyV/pMAPY2oGNZulOj/WROlpJVqYeq3YWJ0IEwAAAFlQcKKuc1vsgRbFPD/Weu3Fp+jLLr/UjNfW1vb09CSOKJr169dnez+YRCNHjlyyZImCPR+fKK3lrAWex+FTfaDmRYt0zZo1nbEb5HJ+bB7Z0KU/mbzSEGECAAAgC4p8/Ll/RXsMvWIhu3WKyxeLvlHRlx/0s1eV+PNm806xX+LNYOzKPYWCls0Lu0euomILmDk/No1sn1ZSX1+vVzXaefPmFfq06hAiwgQAAEAWFF/ZLWGUVnCSamhR5fmKiPQ5M2bM8BcNJirQ00r8cOWk/Y8nUWid9FxZ/UKbWf2JpdOEoHZPWpfJTGLst3Tp0gHOtQ1aqpr0YzRTRTg/1r5L9ai4vXAheiFk+7QStRNF7EqoJajpqhVVVJxJhAkAAIBMKTCwyy/nzp1rF2TG3ZLUU6861VsZ0ocrGpkzZ45CsvQncBboaSUWP48aNSr4eBJFDorNfPBp5sVYeGlpu3QzKTvB2GUyY/FYMM7M7WklcV544QVL6GdbonDWrFmjWEu1qVYhSkgw7orS00pEbWbJkiV2IKa5uVnznsORhdzYiuMypUCECQAAgIyo52pB14QJE2pra+2UUfVlC9Sd1RcpFKmvr1+xYkXx70CjqEDzpaDOBqM8XxI8V9buyqO3xo8fb2lbOPnS0tKiWCW/l7xuiNHHqioV6Wl23BsH0lsKDoOhYG4UzQZvX6Tlo6WnUNy9HUVasGq6c+fOtYpTc1JQHTwwUQidnZ36FjUYly8FIkwAAABkxJ8fa2N6CqhsaFFxYLaDcp663akGdvRF6qAH73mTSt6fVqKAygYhLZa2Qk/hrhX6c2X1C01cOi8s1vVn6poFCxYM5M6rmkELdRTpWTCsbNKKUKHdadblc6VvCValQnGV6GcU6PBEfmX7tJIgtZZFixYtWbLE4kzVptYXe8toIahk4DG8HQAayMqYL0SYAAAA6J9iy/Xr1yuhwMDHCZMnT1YcaIONViKKuxobGzds2KBOsxLGvZdAn5lmvCXDOC3N5+dGYY/iAcUGwfNjg+wkYcWZmk1XVBgKSBTg6WfY9Z/ejh07cn5aidhYtD5Ws6m5SHpxaaFZxOUjqyg9rSSRFrLiTFvOtmpYuSiGV8kAw0JV6IwZM/RRaplaJV1piRBhAgAAoB82POJjElcaYyNgCsn8YNSkSZPszEDFh0oYe6tcaDbtZ6cKcdWJtwniAr+k5syZc3mASubNm+cyMTZZHIvbFV5qecadqTtANiiqWbMhaPEXl+otKxGFK5rMBjY7OztjNZzyClK9FTc01y/7qAwPIkSDr8eBj1jG0cJUa1QQmzjkXnxEmAAAAOiH4hxFGsGYxLOxPiX86Xnq46pEEYumV8LEpq1QFiJ6KlEE6zIxNlmQlrbiUr0qZtDErjQfFNvYCcD6VT66U8J+RvBcWSUWL15sE6tcaUl1TyB9bKq3UlFQqu+NO2ARTtk+rSQVv8BtTdES1kKwBd5vDL9hwwbVgsskmDlzZvolaZ+f98g2KSJMAAAApKOOqQ1tKQjxXeQgG8ZUzzjbUax8KdDTSvJFAfb4AJWcffbZLhNjk3la4PPmzVPAkGZIav78+TlcGWh1pFf9pLjAVSUWoiiMtBL9sHXr1s2NjT8vWbJEacnXwQIFPOvXr58wYULJT+nMRLZPK0kVbKtmLWHVqsm0tONieB/hx1GtpfrYTKje03x4fhFhAgAAIB0FFRZdJMZCRmGnTWDRSHrqZKuz69kwjsvE5NAJ1le7VPnT7GshaFEHr3dNlNvTSppj92rSxyaORYsdKYg7VzaP9LGaNQXPc+bMUVilr5s8ebJ7L1pPK1G4qNkMXqNr8aFmX2nF9hZXK+HXGh/DF2JQV41K1ap6L86IMREmAAAAikq9bU9ZdXxdZj+brDIpOFH4kTQCHLgXXnhh7Nix+vCkQ6P63unTp2uCTZs2FagW/Mcq3GppaSnOSZslYfFkY2PjjP0Uc2r2FeMVqHLTsMFPfXWaYxZ5NGjv3r0uCQAAAOSJutfq165YscLlU1C3W9FO0msRgxSKqI8+c+bMpOOoTU1NYThRds6cOQrS+h3IVdSxaNGipKNJikAuv/xyu0rTFaWwbNkyvc6aNcuyhdPW1rZ48eIlS5bEBaVdCU+z1I8PBjCqqTRzoT9XLK0WkvjJIdTR0VFdXZ357WQ1d4qfNXdKWImWjGZTsWVivadawnGam5v1mZrM5VNLbIf6fH2LSvJ1knN6RJgAAADIv2JGmIMGhaJPq5BJYYDiw7POOkvZVL15O18xGIx5CkgUHigIsbMo4yjC8RdPLliwwL8WVKr4R/Gk3nKZqqqNGzequoP1qFlIGkUHqU61HCxqmjZt2gUXXBCZE2U9G7ZNWt0m1RLW8lS5y8RkGMPHRZj6Ky1nJbQypvkZecRZsgAAACgI61tXDnX31enfsGGDoutNmza50gTq/afq6OstxaV6V4su0QsvvOCmi9mxY4dLlYJ+pGbWU1AdV9JveCmaWT/KF1VaLOIy2VDAOTNAy0rNw2Vi6uvr3aRpKVJV41GN5PYzckCECQAAgPwrWnc2VNTvvzmm31HZpLTQ5qYW/MwcbvMTQkOGDHGpcMvX00qyosagqNKrqamJK0l/Vq1nMbyNqxcHESYAAADyTx3ivr6+NWvW6LUz9VMWJkyYMPC+b8ifVlIIs2bNuuqqq1ymbHV3d2cYJpVWtk8rCZWWlhYbG3f5wiPCBAAAQP7ZeZJr1669/PLL4y4nC7LRGJfJ1boIPa0kc5nfeGYgbCx6gOey9vT09CWcMt3c3NzW1uZP9YzS00pCxQbAXaYoDi7C9cEAAACoQAoyFT2ef/75n/70pwcPHuxKc6I4Z/LkyWUx3hUxI0eObGlpsctKH3vsMdXj8OHD7a2gM844Q3Vt4Wiizs7OOXPm7N69e/v27Yo229vbb7755l/+8pcTJ04MPhIztFpbW/U6btw4y+ZXb2+vlvDpp5+evnlr4Xd3d3/0ox91+dS0YFURWu8sq/pKWmWFQ4QJAACAQlHXViHKAMPLfjU1NZ155pkuUxlWr16tuS5QzBNn1KhRCm/uueeeF154IdV9biVVeCknnnii3lWYun79en2OPk0NY+bMmZnES2FQXV19zjnnFOjaVy0ZLROL4RV7K5s0IFQMr6Axk1UpLsIsPp5WAgAAgPIWkqeVFJMizLvuumvVqlUuX/6i+rSSTHTFng7a09Oj6H369OkDHKvXRw0ZMqSEg8NEmAAAAChvRJgRUMkRZsRwpx8AAAAASKkkTyspX0SYAAAAKG8V+LSSqVOnzp8/32VQYGX9tJLi4yxZAAAAAEjJ7o3KHVIzxBgmAAAAACA/iDABAABQ3pqamlyqYmiWZ8+e7TIosIaGBm5BlDkiTAAAAJS3xsZGl6okEbsycNq0aaG9m05dXV1NTY3LoD9EmAAAAACA/CDCBAAAAICUeFpJVogwAQAAUN4q8GklmuWlS5e6DAqMp5VkhaeVAAAAAEBKPK0kK4xhAgAAAADygwgTAAAA5a0Cn1bS2to6bdo0l0GB8bSSrBBhAgAAoLzxtJII4GklkUGECQAAAADIDyJMAAAAoMxUV1f39va6DAqMp5VkhQgTAAAA5a0Cn1ZSV1e3bt06l0GB8bSSrPC0EgAAAABIiaeVZIUIEwAAAAivPa+/sXXnnl2v/q1vz9+27X7ttdf3Ku3eixl8yKDhQw4dfMhBQwcfdOyRg0fE0u495AMRZlaIMAEAAFDempqaInairKJKBZPPbn9VsaUSrjRjgw8+6K0jDntr9Zv06oowAB0dHdXV1dxONkNEmAAAAChvgwZFp0+7ZeerCiz/8sLLe/72hisamLcOP+zM44Ycd+SbXD6spk2bdsEFF/DYyQhgAB0AAAAosT2vv/HMtlf+359fvPPhbQ9v6ctXeCnPbH9Fn6lP/ssLu10RUEhEmAAAAEApbdn5qoLA1se353BCbIb0yb97cse6jh7izBzwtJKsEGECAACgvJXvRZiKLW3csnCxZdCuV/+mOFPfuK2vGF8XGTytJCtchwkAAAAU257X33hw8648Xm+ZrbcfN/S9Jx/pMkgr53vJdsZs2rRJ6aExZ5111vjx4+3dqCLCBAAAAIpqW99r9z25ozjjlmmMGHLouaccNWLooS6PFHKIMPv6+lauXLlhwwaXDxg7duzcuXNdJoqIMAEAAFDeyutpJVt2vnrXY72lGrpMNO604TzUJL1sn1ai8HL58uVtbW1Dhw6dMGGCQkoV9vT0dHV1tbS01NbWEmECAAAA4VVGTyt5+Pm+BzfvCk94ad570pFvP36oy5RIlJ5W0tzcvHLlSoWXixYtUjzpSvfr7OwcNWqUy0QRd/oBAAAAikHh5f1P7wxbeCn6VQ88+5LLYMDswsvx48cnhpcS7fBSiDABAACAgrPw0mXC58HNuwgyU8n2aSV9fX0uVZGIMAEAAFDewn8RZsjDSxO7ty1Py0wi26eV2NBl0tv8VAIiTAAAAJS3devWuVQoPbPtFQVvLhNuv3tyh36tyxTXqlWronERptTX1+u1r69vzpw5PT09Vlg5iDABAACAQtn16t8UtoXw2stU9Gu39ZX4MSrlrra2dvr06Up0dXXNmDFj5cqVFRVnci9ZAAAAlLfQPq1kz+tvtD6+fevOPS5fJoa96eDGupEug+yfVmIUXi5evNjHlqNGjZo0aVKhb/OjL5W+vj5FuaW6pRARJgAAAMpbaJ9W8sCzL5XL+bFx3n7c0PeefKTLFEWUnlYS1NbWtn79+s7OTstOnDjRhjfzTl+xcuVKhZdDYyyynTlz5vjx422CouEsWQAAACD/tvW99siWcr1xzsNb+jhXNi/Gjh27aNGiJUuWKKFsc3Pz8uXL7S2jUFAlAz+NVh+iwFJfdPPNN6+I0TeqUCGum6JYiDABAACA/Avnoy8zd9fj212q4mX7tJJEtbW1c+fOteHEDRs2BOPJrq4ulQz8ASeTJk1SKOufwDly5Eh9o15bWlqspGiIMAEAAFDeQngR5jPbXim7yy/j7Hr1bzy8xGT7tJJU/PmxAx+xTJT0bNixY8f6E3SLhggTAAAA5S1sTyvZ8/obZXr5ZZz7n3rJpQovSk8rSWXo0KGWsBFLG73ctGmT0m1tbUqnGczUWytXrnSZjOnTEj/QvqgQUa4hwgQAAADy6Zntr2zbHYWLGPf87Q2GMXOQatjQXxJp57IqwlTQ2NzcrPT69euVFhXGJomnQDGH0Uj9lT9v1ujzly9fri9y+QIgwgQAAEB5a2pqcqlwKN8b/CQq5jBmaDU0NGQ1vqr4bc6cORs2bHD5/fGh3eNn4sSJI0fuexjM+PHjb7755rlz5yq9aNEipSWPjxhRMKmY9uyzz3b5GAtT9S32GwqBCBMAAADlrbGx0aVC4JltERnANEUbxpw2bdoA76ZTOHV1dTk8DFPxpFrmjP3mzZunONOeiukmKiR9l35AbW2tAlpXFGMn5dbX11u2EIgwAQAAgLz5y4svu1RUPLP9VZdCZmbOnKkw0k5P7YlRQrHlohh/NWbhKL6dM2eOgsy5c+cGv86GUlWSx5HSRCF9Oi0AAACQoUGDwtKn3fP6G+s6Xijrh5QkNek9xw4+pLBDU9OmTbvgggvCebOfpqam3t7e3H6bgjq9pokq29raFi9evGTJkrgLJhUKzps3z2WSGT9+vEJZlwnQH+oD9Wl6N+5UWPuusWPH2qm5BcIYJgAAAMpbeJ5WsnXnnuiFl/LM9ldcqiIN5Gklii3ThJdpKERUHOgpnrRHXHoTJkxwk+6naHblypWKISdNmqQJEq+0tNsIFfQUWSHCBAAAQHkLz9NKnumN5gmlRThRthKeVpIVxaVjA2pqauJK4sY87cLLDRs2zJw5c+LEiUnD2paWFpUX9BRZIcIEAAAA8mPrzmhGmFt37nEphNXKlSs7OzsXLVqk4NMVHairq6unp0fhZW5jqpkjwgQAAEB5C8nTSna9+jf9c5lo2fO3N6I6a5nI9mklxafoccOGDdOnT48b2AzSWytWrEh66WZ+EWECAACgvIXkaSXb+6LzkJJEz2wr7KWYEXtaSeZsRNGukMyZf8qlZVMZOXJkoQcwhXvJAgAAoLyF5F6yDz/fd//TO10mtQfuvOUvf2x1mZjDj6i+6IvfcJmwetcJw0afeITLFECY7yVbaDNmzOjr67M79yReYOlpmlTx4fLly9va2lJFmP7RKcXBGCYAAACQB9tfft2l0npl1w691k/5yqVzv2f/lL1t8ZWKPGPvh9T23RnNXSQ1NTUVdHx17ty5igBbWlo2bdrkipJJM/x41llnKTTVBEnZE1OKhjFMAAAAlLfGxsYw3E629bHtmTzVo23dDZsfbVeEOfz4v594aYXv+9j0t77jva4oZEYMOfQf33m0yxRAmMcwFyxY4F/RL8YwAQAAUN5C8rSS7btzvw5zbOMVev3L/S2WDaFC3+mHp5VExsHE4gAAAMDAPbKlb8/f+j89cPMjf3rpr1tq6847/IhqVxSz9cmHtj3X9fbz/9Hlq6pab/rOn+748cP3/D/92/Ny33GnnuXeiJ1Vq5KdLzx3183/ZRMMG3HsUce8xb0du9rzd7den/StO67/2p83/K+9dfAhg9984qnujbQOPmjQqLcMc5kK09q677rZcePGWRbpEWECAACgvDU1NZ155pkuUzoPPrvrbxlcgJYmwlS5wkgrVxz4+p5XPjZniWJOFSrUVDx54tvfYxMrONz+fPfre1696Ivf0AR665Hf3eH/1m4mVD/lK+++6FOxv735tLM/aH+o0PTIo4+3v1Lk2fHLn8bFrqlo1gp6p58wq66uPuecc/Tq8kiLs2QBAABQ3hr3s7EmWbZsmSsqYuEDf7jHCnNz2LCjXCoWIr78Uu85l/6zZYcfX3Pqe8ZtfrRdUaWVyIi31I77zJctffo5H9br050bLfvKrh0KNe06T736G9W2rbtBr/6v3vqO955wxpi4G9umoZkds5+/802+Cmtrawt6N52BKOjTSqKHO/0AAACgvDU1NVnCRwIdHR3d3S4YK1rhswefeNTIEyydRtI7/UiwvPWm77z80vbgI0we+/2dna3r/K2Ablt8pYJDu3pTFHm23PhtRaGjP3yZsjaxgsy4h6Dccf3XDj9iuI8wxY92xv2YpCaecbif2erqalsCvb29AykULUNLBAtRvogwAQAAgDxYe//WPX97w2VSSxVhKqrc9lyXPbzE0lYelD7CDJZYkGlp/1eKMF9+qdcKgzKMMD899niXqjBNTU0Kj7kRUYY4SxYAAADIg2FvOtilcqKQUiGiy1RVHX5EtT0tM/gv82eZnH7Oh+1PlP7Dz1b602tHvKXWf5r/l0l4OcC5K2vBkWr0iwgTAAAAyINDDxnkUtmzKyTfsj/CHH78yS+/1Bu86jJno8Y16nXX9hf1qvAy6dBoJgYfTOCAjNBQAAAAgDwYMeRQl8pS603f2fxou0JBP0Rpl1P+/rb/tqw889D9d1z/NZfpz22Lr/TRqSXsk+2GQPq6fW/EPPb7O4PZNIZW8BgmssJ1mAAAAEAePPDsSw9u3uUyqdnNdVwmJnhX2KDgZZNxt+1Jfx2mwtE//GylvSVxl1nqb10q9VcnevtxQ9978pEuU2E6Ojq4C1HmiDABAACAPHhm2yutj293mcj5h1OOOvWYIS4DpMZZsgAAAEAeHHvkYJeKomOPfJNLAWkRYQIAAAB5MPiQg6J6w9XBB0d21jLR1NS0evVql0F/iDABAACA/Hjr8MNcKlqiPTzbL55WkhUiTAAAACA/jj0impHYW4dziiwyRYQJAAAA5EdUx/q4CBOZ416yAAAAQN787skdf3lht8tEwluHHzbu9OEuU5F4WklWiDABAACAvNmy89U7H97mMpEw7rThbx0RzetLUQicJQsAAADkzXFHvilKt13VvFT4bX6QLSJMAAAAIJ/edcIwlyp/pxx9+OBDKj1k4GklWSHCBAAAAPLp1GOGRGMYc/DBB739uKEuU8F4WklWiDABAACAPIvGMOaZxw1hABPZ4k4/AAAAQP79vz+/uG33ay5Thoa96eB/HHU0EaYsWLDAv2alM2bTpk19fX0jR44cOnToWWedNX78ePd2RBFhAgAAAPm3re+1/9f5osuUoQ+/fcRxPAYzJoenlSikXL58eVtbm8sHjB07du7cuS4TRUSYAAAAQEHc/9TOh7f0uUxZeftxQ9978pEugywpvFy8eHFnZ+fQoUMnTJigkFKFPTHr16+vra0lwgQAAACQi3I8V5bzYweoubl55cqVCi8XLVqkeNKV7qfIc9SoUS4TRbQbAAAAoFAuOH24S5WJwQcfdO4pRxFeBmX7tJJNmzbpdfz48YnhpUQ7vBSaDgAAAFAow9508IffPsJlysF7Tz6Cyy/jZPu0kr6+sjw1Ol+IMAEAAIACUsD23pPK45rGtx839NRjhrgMcmVDlxs2bLBspSHCBAAAAArr7ccPVfDmMmHF3X3ypb6+Xq99fX1z5szp6emxwspxcA7PdQEAAACQlbdUv+m1v+19cVdI7/pDeJlGdXX1Oeeco1eX78/w4cOHDh3a3t7e29u7fv363bt3n3DCCSpxb0cd95IFAAAAiuSBZ196cPMulwmNd50wbPSJR7gM8qSrq2vx4sV+DHPUqFH+ySUF0tfX19nZad9YW1tbqlsKEWECAAAAxbNl56t3PrzNZUpt8MEHvffkI7j2snDa2trWr1+vwM+yEydOnD59uqXza82aNfoiJYYOHeqDzJkzZya9n21BEWECAAAARbWt77W7Ht++69W/uXyJjBhy6Hu4c2wGmpqaent7p06d6vLZ6+rqWrt2raJNpceOHTt37lwrF0WDemvSpEkjR450RTlZuXJlTU2NPtxOx1VMu3jxYqVXrFhhExQNd/oBAAAAimrE0EMb60aW9t4/px4z5MNvH0F4mYlsn1aSqLa2VlHl+PHjlVacGbz9j4LPDRs2DPwBJ9OnT9fn+6s9R40apRJ9kYW1xUSECQAAAJTAe08+8h9HHT3sTQe7fLGMGHLouNOG/8MpRw0+hFigqPz5scEIs3DsOsziP5yTVgUAAACUhg1mvvekI4sTZ+676vKkIz/89hFvHXGYK0IR+QFGi/ps9HLTpk1Kt7W1KZ1mMFNvrVy50mUyYx/lv7RoiDABAACAUnr78UMLHWdabNlYd4y+i6HLbDU0NGR1Eaa/r08cf8Kq3X1HEaaCxubmZqXXr1+vtKgwNkk8hYupPjYVfZ3Cy+LfUZY7/QAAAABh8cy2V57pffUvL+x2+RRefP6Zo49/q8ukpsBy+NBDTj368LcOP4zAsmjmzJmj1wkTJtiFl6L4UKHj4sWLlYi7naziQJUvWbIk/U1fFYi2tLRoMpdPS9+yYcMGRa2TJk3yv6FoiDABAACA0FGQufWl15KGmrtf2jnn0rP/46Y7UwWZCiyPPXLwW4e/icCyJBRh+qFIu0OsQj5RYuzYsTNnzgyeuZrHCHPNmjWbNm3SF/X09OjTFMem/8wCOXjBggUuCQAAACAcRgw9VPHh6BP3PaxyxJBDhg85tGpQ1bA3HTyoqurm5f/5yJ/ue3HLs+d8+BIFkwcfNOioww85etjgE6rfVDPi8DFvPWL0icP2/dXQQ/WW+zgMQFNT0+9///u6ujqX78+YMWMUQ+7evbu3t9diy8GDB7/73e/+/Oc/f+mllyrtpovZvHnzPffc89GPfnT48OGuKJnHHnusu7tbk7l8MgosBw0apK/WV3R2duqTFWGm/9hCYAwTAAAAKButra3Tpk1T6KL0qlWrGhoarByFY2NyuY3MKbzUa5rb7aQaw1SIOG/ePJdJZvz48TNnznSZBPpefaw+pN/R0bwjwgQAAADKRn19/SWXXLJw4UKFlwo1t2/f7t5AwQwkwuxXqghTIWLw1j4bN25UNngB58iRI/sNHWfMmKFp5s6d6/JFwWnZAAAAQHloamrq7u6eOnVqe3t7Q0PDuHHjChT2oOSGDh06NqCmpiauJJORSU0WDFOLgwgTAAAAKAO9vb02dFldXa14QyVKq0Qxp02AAsn2aSXhse8a0BQP2CwcIkwAAACgDKxevVqB5bhx41y+qkqh5vz582fPnu3yKIy6ujoL6cuO3VTWZYqFCBMAAAAIu97e3uuuu07xpGXHjBljN/tZsGBBR0dHU1OTlaPs2E2A/NNNcqM/VzDpMvs1Nzd3dnZOmjTJ5YuFO/0AAAAAYadIcseOHUuXLrVsbW1tS0uLDay1trY2NjZyy5/CUQCveL5wJ8rOmDGjr69vwoQJSqe5wFLTpLon7YYNG5YvX66/VZMYOXKkos2NGzfqVeHlxIkT3UTFQoQJAAAAhFp3d3d9fX17e3t1dbWVBCNMUYQ5evRo7vpTILZgC7d4u7q61q5dq1cFh9OnT8/tvNbm5maLKhWI6nNGjRqlNlP8U2SFCBMAAAAINYUKF1xwQTDCiYswe3t7VaIQtEwvFwy5QkeYEcN1mAAAAEB4tba2dnd3z5o1y+VjguOZYrf8Wb16tcsDpcMYJgAAABBqvb29wXgSRdbR0aHlz/hwhogwAQAAAAD5wVmyAAAAQJmpr6/v7u52GSBMiDABAACAMtMb4zIosKamJq5xzRwRJgAAAFBmqquriTCLpqOjgxHjzBFhAgAAAADygwgTAAAAKDPr1q2rq6tzGSBMuJcsAAAA0L+urq6WlhYlJk2aNHToUCtMo6enZ/369UpMmDBh5MiRVohyxNNKssIYJgAAANC/vr6+5hiX74+fXglXhPJUV1dHeJk5IkwAAACgzDQ2Nra2troMECZEmAAAAED54V6yRcPTSrJChAkAAAAAKfG0kqwQYQIAAAAA8oMIEwAAACgzq1atGjdunMsAYUKECQAAAJSZ6hiXQYE1NDRMnTrVZTIzZ86cxsbGzO88HCVEmAAAAACQEk8ryQoRJgAAAFBmZs+e3dTU5DJAmBBhAgAAAGWmN8ZlUGA8rSQrRJgAAABAFjoz09XV5f4AZY6nlWRl0N69e10SAAAAQAoKGufNm+cy2ViyZEltba3L5Mm0adMuuOCCbG8/g9wsWLDAv2Zozpw5XV1d06dPnzhxoiuqGIxhAgAAAFkYmZmhQ4e6PyiApUuXNjQ0uAwQJoxhAgAAAP3zY5g333xzJtFjV1fXnDlzlCjEGCaKqaOjo7q6OqvbyTKGCQAAAABIgqeVZIUIEwAAACiNvr6+zs7O5hglXGkGFixYwN1NEU5EmAAAAEAJtLW1zZkzZ968eWvXrl2/fr0Sa9asce/1Z8eOHTytpGh4WklWiDABAACAYmtubl68ePGoUaOWLFly8803r1ixQq9jx451byNMeFpJVogwAQAAgKLq6upau3btxIkTZ86c6W8CNHToUG4IhAggwgQAAACKSuGl4slJkya5fPbmz5/PwzARTkSYAAAAQPH09PS0tbXV19cP5IGZ1TEugwJraGggns8cESYAAABQPHbP2PHjx1sW4cfTSrJChAkAAAAUz6ZNm0bGKNScN2/enJiVK1f29PS4KTKwLMZlgDAhwgQAAACKp6+vb+jQoXYv2dra2vr6+lGjRimrOLOrq8tN1J/eGJdBgfG0kqwQYQIAAAD9Gzly5PgYl++Pnz7uektFmNLS0rJkyZLp06dPnDhRrytWrNBbijltmkzs2LHDpVBgPK0kK0SYAAAAQP8UMc6MyfAOPZrMptcfuqL9enp6Jk2aFCxXWnGmyu0qzX5xm58ytXLlyra2NpeJKCJMAAAAoKgUfI4dO9Zl9rOSDE+UnTVr1lVXXeUyKBN9fX3Nzc1ZXXBbjogwAQAAgOKpra11qWQUhLhUf7i7adHwtJKsEGECAAAAxXPMMcfErsSMjyStJPGUWpTcwJ9W0tPTsyFG6e7ubktnfmOn8kKECQAAABSP3Sso8WI8Cz9GjRpl2fRWr169YMECl0E5WLly5dq1a5VQ1Ssd4QsyiTABAACA4rGLMBVgBIewlF6/fr3KMx/DfOqpp1wKBTbwp5WoWm+++Wa7Y/CkSZOUlsmTJ9u7EUOECQAAABSV3WB2zpw5ijM3bNig13nz5qlE5W4KhAlPK8kKESYAAABQVEOHDl2yZMnEiRPthMnOzs5JkyYtWrQow+egIPzOPvvs8ePHp7+rU1QN2rt3r0sCAAAAKBPd3d3cTrY47JLXgV/42tfXd/nll0+fPn3ixImuKIoYwwQAAADKD+Fl0fC0kqwQYQIAAABASgN/WklFIcIEAAAAykxTU9Ps2bNdBggTIkwAAACg/HB306IZ+NNKKgoRJgAAAACklK+nldi9giN/aIAIEwAAAACKYfz48fYE1ObmZiVcabQcPPC77mZOy3HdunXPPvvsqFGjXFFaWuhr16597LHHxowZ44oAAACAinfmmWeec8451dXVLo9Cam1t1eu4ceMsOxCKg3p7e9vb2zdv3qwY58QTT3RvREhRn4e5fPlyBY1jx46dO3euK0rLgvva2tolS5a4IgAAAAAooo6ODgXz3E42Q5wlCwAASqmvr6+np0evLt8fm15cHgAKjKeVZIUIE2GhHkPm3QvJdnoA5SXbdTzb6REeGzZsmDFjxpw5c1y+P21tbVlNjzDL9mBBttNHWGtr67Rp01wGCBMiTISFuguXX3555lc8q2+h6desWePyAKLF1vHm5maX78+8efM0/cqVK10eQOj19fVp7y9tbW2uqD/aMmj6qN4fJVs8raRoeFpJVogwAYRLZ0zmh6i7urqymh4AytFrb+zZuvvZ7p2PPrKtfeOWlt8998vfPH1b8N9dz65v77nnzy9ufHx757ZXejS9+0sAA5avp5VUCCJMAOGyfPnyefPmZX58euXKlZp+/fr1Lg8AUaEoUbGiQspYPLmuvefeR7Z1dO98TIU79/S+9sZrwX8vv963dffmzbu6/rJjk0JQTa+wUwGn4lL3cYiW6urq3t5elwHChAgTAIBQY/CqAllgefez61XjCikVT7o3sqGwUwGn4tJfdK9VC9FnujdQJtKv+38d8eTc789k3UcIEWGGi21KtJnIpBuhKV9+ndtaRBMtAaJmsHPPdnUQfTNQvcc1AxWqJWgCdUHoPkaJat9ijNjqz+BVpbCNvypd9ajqVuW6NwZMLcQ2I2onrgihlNW6f9RxQ1n3i6OhoWHq1Kkug/6Ud4TZF7tfucuULdudaHPgNyXaTGTSjdCU1t3U37LDiABaAoyiSoWLagZ37wsgf6Vq9c1A9R7XDFSolqAJ1AVR98JGKh7f3qm33Meh3FjnksGrCqRq+sOWFtVabpWeCX2yNinaZbCzCCHW/TDjaSVZKdcIU4Hl4sWL7f5jjY2N5XjzQDtGpZVfmxJtCLQ5GMimRDsM25Swzyg7tAQYtQTVmgJLRZUKF9UMVKfuvWwo4PzLjk3qQcb6KI+6UoSeHWNSA7DOZW61n5SahD5Tn8xmIbS0F7Cqz237n62XX+/TzkLfuHPPdleE0mHdR/SUIMJUcLghM11dydeHvr6+efPm6XXRokUrVqyYOXNmc3OzAk73duhpU/L49s4/bGnRaq+VP7+bEjs2qc93RQgxWgKMWoIdt1at5bF/GTsc3vGL7rX6cIY0Q06VxeBVZbLVv73nnuLElkH6xtjxrHaXRymw7pcLnlaSlRJEmIobl2emszN559gejzZ37tza2tqRI0eOHz9e6ba2tlQRaXhoRxI7+e1Xf9mxqXCbEnUl9fnqVjJ8EVq0BBgfW+b3uHUcfbj6FvoifZ0rQmgweFXJVAuKLgq6+vdL3057KAnW/fLC00qyUoIIc+jQoaMyo+jR/U1AX1/f+vXr6+vr9TmuqKpq7NixmljlLh9KdpjqkW0dRRtM0HdpU8LV3mFDS4DZvKur0LFlkL7oN0+vq8DDDdprdHZ2Njc3r1y5cs2aNeG5ep/Bqwpn+4Li134iaw/sI4qGdR+RV4IIU6HjoswobnR/E9DV1aXuwvjx411+P02casyz5GxTUrTDVEH6xvaee//84kaXR0nREmB27tmuvoWqo/gDF3a4oWhHN8JgcczGjRsVW7a0tMyYMUNxpnuvdBi8SqQKmpOZkB9QzkT3zke1EShh7SfSPoITXoqAdR+VYNDevXtdsvCWL1++YcMGhYJz5851RWmtXLmyubm5trZ2yZIlrih2iqzK161b5/L7qcegXc7NN9/s8qER60reW/L+3KEHHfq+4+qPHDzc5cPn8ssv7+vrGzp0aNKx60Tqi2j6SZMmTZ482RWFGy0hQwoAVLmZ1+y8efM6OzsnTpw4ffp0V7SfymXTpk1qVzU1NSFpKttiN3Yqec/ynUeffcKwWpcJJWsJ2W4TEltCW1vbqFGj/GkvtmfRbkU7FyspvpC0AW/MyPOOHXKiy5SC7dldJmOq0xDu9DOhQO6RbR0uEzKnHnXWacNHuUzhaZ3V3l+JzNd0uyRq5syZiYMN4ce6X746Ojqqq6u5nWyGyi/CVCTZ0tKyYsUKl99Pn6zP187GdyPCQHuRv/RuCs+m5MwRdTVHnuEyIWMRpstkrFwiTFpC5vIVYS5evFjldr79kCFDtN1QA1u0aFEJ4wp5ZN+DKx9zmVIrclcyW9YSXCZjSY81xNHWZsKECaXadIRta2BKu02wCFO7b4UNriitTZs26U/KNMIMc3hpirll8BFmtsoxwmTdR+UovwhTheo1BktMW1ubOpShijAf3975lx2bXCY0QtuntAhTvcOzzjrLFaWl5qTpyyLCpCVkJS9xhdqGNgjqgviD4iqZM2eOtg+JW4/iiJ0j3RG2W/mNOGzk2cfVu0zIWEtQJ/Lss892RWlp76DpM4kw1RJGjRrV72SFEOboooTbBIswtbYmHj5Oyo4pl2OEGf7w0hTtHAcfYWpvnuHhP9v7l12EybqPilKWz8PUlsWlApIWltCfX9wYwqBC9Ks2bmlxmfCpqakZm5nwHEpIj5ZQEmoec+fODZ5zpRL1YOxCbldURAovO1/cGMI7xW97pSfkzSDv2wQ1AAWiJTnTKeTRhbYJPN+ooLbufvYvvWHcHSTSnqvIN/5ReOnW5P64PygrrPsRwNNKshKdCNOEJOrQpjnMDx0Kf58yMmgJOZs4ceKKzKQ67J24NbCAs/gRpoWXW3dvdvmQqagNgmrfxsqK308ti8ErdTTDvMnKgWp83rx5Ybi308uv92k7ELYzJNPQr+VOMHnBuh8NPK0kK+UXYQ4ZMsSlDqS9SHC8ooQe2dYe/rVUfUoFPy4TUepSqCvpMqVASxgIrelaozOR+XElO/m2+Meh1LcIbXhpIh9kNsdog7B48WI1g7lz5xa5GTB4VSqq987OzpL3C197Y48WbBmFl6Jf295zr8uUA/UDtYInKu3D0ln3UZnKL8Ksra3VRsR6ikGbNm3K8Az+gure+Wh4buORnoKfqB6v0u5EO5W1a9eqqbiioqMlhFBLS8vYop9frZZQFos32kedtIPYuHGjtgwKNtQAitzjZPCqVNRVWL9+ffEPKiXq3vGYVjGXKR9quo+Uz4MTtcdva2sr4X4/Ees+Klb5RZh20/kNGza4fIw2KOo31NeX+H4V2n+Uy5Eqow5lOe7z0lAzWLly5bx587Ia2so7WkIINTc3K66YMGGCyxeFYsvwnxzl6ddG9QD23Llz7THL69atq6mpWbx4sXqi7r0CY/CqhLQ7sLtJu3yJqMv+9EvlccAxUffOx8ol5LDYcubMmVrfg0o1/MC6HzENDQ1Tp051GfSn/CJMhQ3qI65fvz54EHrt2rUq147E5UuhHDcl0t5zT8mf0JhHy5cvV8NQP3L69OmlijBpCWGjboc6mtpKqOdRzK1EeR3+N50vboxqM/AmT548fvx4bStcvsAYvCoVOz920qRJLl86j2zrKLs9QlC5hByhGr0U1v2Iqaur42GYmSvLO/2ofzBy5Ei7dl+7ECU2bNigvmMJx6ykTHtm2u0pHHKZ8mfDFKU9X5qWECrqcyiWaGtr0yaiyDd3KccDDfrBlXDXn7PPPlsNowj9UQavSqWnp2ft2rWZPwCjcLbufrbczxDRHq0sTvW3Nbq0XUGPdR8VrqgRpnbq2txnfi7rWWedlWr6JUuW6K1Nmza1tLRo/6FsaQcwtQsJ+Z080tDOLzKnxpX8bChaQqh0dnbOmTNH3Q5tIoocXnbvfLRMu5XqTerHu0xEWU+0CBEmg1elsnLlSvUNJk6c6PIl8tobe8rriolUymJQK1QRJut+9PC0kqwUNcJUD2/y5MmZ9/Ns+lR7CJUvWrRIHcfp06eXNq7QLqSMLrVKqjPq95UtDlpCqLS1tS1evHjChAnaUBS5z1Hu3Ur9+HIch8+c3Rmu0DsOBq9Kxc6PVd/A5UunZ/fmnXt6XaacKVgKf0tQhFnyo8yGdT+SeFpJVsryLNmw6d7xWLn3xrTziOQZkkVGSwgPdTWWL18+fvz4kgxiqCWU9dFr/fgnIjHwIhtiXCbWMFauXNnW1lboy/MYvCoVxZZ20XXJz4+Vp3aW63mSicLfErR29/T0XB4zY8aMxYsXB9f9omHdB4QIc6C0KSnfU+2DenY/G+1Ri0KjJYSKdSwKHUUkpaUXgZaweVdXNDYICjOWL1/e2NioHuecOXP0qgikCBflMniVlYkTJ958881Llixx+f6MHz9e069YscLl91OAobhC7xb5rPiktu5+NhptwBSnJQyEVvbp+9XX13d1dWndnzdvnnu7WFj3ARm0d+9el0ROHt/e+ZcdETnYf8Kw2ncefbbLFJ0943RojJWk1xeTZnp1JUeNGqWupMsXGC0hX1RxagwKDidPnuyK0lIHQjGDeqjqVbii2F2F7e4+Lh9Q6NMjI9MSStsMpN91PI5Nr0Ri/aox2FtW+xl+4ED87rlfRia6OPSgQz940qUuE2KqYm0NVLlz584NVvGcOXNU6Sp0+WJp77mnfC/LT+rYISeMGXm+y+RPtnv/zKfXjmBD7E6Q48ePd0WFx7ofVR0dHdXV1dxONkMHL1iwwCWRvdfe2PPgi/e9sfcNly9zr7ze99YjTj140MEuX1y2txg8eLDL90dTpp9+/fr16lUU50g2LSGPWlpaXnvttdNPPz3D23dt3Lixt7dXkcOYMWNcUaxw8+bNek2kPcQZZ5zhpsu3KLUENYPjh5506EGZrpJ51+86HsemF5cPOPHEE9VCZPjw4Zl/YM627n72qZ2Pu0z5U3s+/JChRw4e7vJhtWLFivb29vr6eq37jwVs2rRJm5Q9e/YorZZQhAYg2hQ8sq09MjsF0/faSycfeXredw222mZeL5lPrw6Adija7F96aZHCJNb9CDvuuOPUf3AZ9IcxzAHZvKsrYpcvnjmirubIQnW+i6yYY5i0BJjunY+W+92egmgGuWHwqiQWL17c2dnpMgF9sZFwSxfhBGmjSCOSd+N859FnnzCs9Be4Zm7lypUbNmy4+eabXb7AWPcBw3WYA7J5V9RuKhWlY2/FREuA6dn9nEtFQjTuV1Fkr72xp9xvI5lInWbNl8uE1dy5cxVIJKqtrR01apSli3Z9ZsQ2BV5PGYZPdoZ8EbDuRxtPK8kKEWbuXn69L3qbEs1UNG7vUUy0BJide7ZHrCW89sZrNINsqQ1oublMhJRjaFFC0dspmLKbr56entpi3VWYdT/aeFpJVogwc7c1Ws+m96I6X4VDS4CJZJ+SZpAtBq8Q4SN05XXUqbOzs62t7eyzi3THMtZ9wCPCzN32V15wqWiJ6nwVDi0BJpJL7LnInQFeaAxeYeee7S4VReE86tQXe9RtV5d7uoayzc3Nixcvrq2tLdpTkVn3AY87/eTurmfXR/IgZWRuTl20O/3QEmB+8/RtkTxF6oMnNZbwjrLlRZsCbRBcJnIuOHHC4Ydk9DyJsAne7KcI8nLHr3XL7ux56q8uEzNj6d8f4NSx4eG25gdcJuaU0W/98NTzXGb/BGMnjq4b/3ZXFHPvbX/s/O3jH/z0uW9798muKEunHnXWacMzutF3MamKL7/8ckv7uh47duz06dOLU/Ws+5HH00qywtNKcrRzz/auHY+4TLS8sfeNEYeNjMCmZPz48aeffnqhdy20BBi1hKjeHumYw4+nGWRo2ys9W/qecZnIOfyQIdVvOtplykq+nlDS1NT0hS98oa6u7rjjjnNFyTzz0pMvDeCJiAoOb/+/vxlaPeTyBZe898JR9u/V3a/efcv97xrn7u38l/ane57epkDxI9PO17vVI4/40682PfPI828/51SbYEvXi5sf23riGccdV3uMlZhnHn5ef6hwdMTxOT53YfDBg48fepLLhIaqePLkybWxx1bJRz/60U9/+tPnn39+Xqq+tbVVr+mfVMG6H3k8rSQrRJhJZLIpefHlLRE+Mf2IwdUR2JRovzKQ8LK3t/cXv/jFmWee6fIp0BIir6Oj4/e//30ltwSagXR3d2ub0G/34sWXn1dLcJlc3bn63l/f+Ls//rLT/zvpHccPPWqIvfvEn5669du/CL4bjCvEJlBActLb3+KKYixuOWTwwXEhR+aOGvzmNx8+0mUqkm0HFGQ+8MADijNTtYfndnX1vfaSy2RJ1ffbW+9XBPiPnx/nimJUmz68lLhAUa/bt+x49tEtvqkULsI8eNDBbz3ibS4TMv7JtyNHjszXYQXRXuDcc8/dsWPHhRde6IoS5GXdT1y741bkFbPXBN/Vv/deeMB48s0Lf/bgXY8Gm4rRH8ZtKLLCuo9sEWEmoQizvr5eiXHjDti+Bz23q7v31QNOX8mB7e+DW4rgvr/n6b/evOBnwXcfaXsybquRdJNhf6idzal1OR5lfNPBhx875ASXqVSvvPLKtGnTVqxYcc4556Q5XJ2XlpC+Q5nYTuIqN1XH8d7b/njHf99dPfKInHsStARRS/jEJz5x1113qSWkCTCe2vn4QEYtJHGVj6todR3uu709OEGwnci6ZXfevfYPcR0OSdXnyJA6lCEcsigyRZgXXXTRU089pRgjTTMY4OCV9S/3vPr69G/+k41c6Z9WbW0ffLU+/fDzihzGThw94Qv1KlQb+OMvDtg7bHt+R9eDz448+c1xEWaqkCNz4Ry8KjIFlrNmzbL2oGzSfsJj2x98PdcT5u9a+4e+HS9fNvdil08hMVDc8uQLKjnhtGOtpHAR5ht7/3bKUQeceRt5WuunTp36y1/+cvbs2Vr91QbcGwEDXPdFG+pHN3b5cWn7p52437lrH/HI759U3al52Lta8bVTCO4ItKnXa+LWXvuLodVDco4wWfelqanp97//fdLaRyLu9JOEtiPt7e3qSYwZMybVo28Gft2dgsO25gcaZ394xtLJ9k9plaiHYRPsfHGXXke9/zQ/QV/vbv2VvVtQr/Pso9ggtprBlClTGhsbtVNJdYvqAbYEVbfqdOtTf/W1rH/qO65beqeboqpK9a5X7XXsXSWefOAZxRL2bkHREqSmpqarq+uCCy6ora1VS3ClCQa4rFShqnRVvW8G+qeGce9tf3RTxFqCwgb/rtL6E7/FQEGpV6FmcNRRR6kZpDkyO5BmoO7jb350n6r18vkfc0UxdePfrup2mQQjT3qzdhNqG0VoCZG84Dw3CjL77SfkQG2g56m/KoRw+Wzs3vmKXo88ephlkV/aEaxatWrdunULFy5UpSd2CQa+C9BarH5g3PWxWvfjrqQNOmfiaL0+fn/B78fGui88rSQrRJjJqTOhTcnSpUu1Kamvr09sUgNc2W5e+DO97usjnvRmKxGlVZLm4nv1PvXaseFhyxbOzoEdh4sSdSPSdysH0hJy61Cqhaj/oV6I/twVFQwtwVNL2L59e29v7/Dhw5N2KAfSEu5cfa8q9IOfPjeuJ6GGcd6l73GZBOdd+m69dj1Y8Ps60gw8bQTSxxUDWVYPbNh3RbdVa1aGVu8bvtgVOxRVUPQyg4L9hMbGxmA/Iedgw44sH3PSCMtmTh2DJx94ZtT7Twt2Kgokkvczy5AdaZoyZUriAceBrPtP/Okp7QJyqD47oGAHFwqKdR/ZIsJMZ9y4cdqUXHLJJYmbkoFsYbUn6OvdbeFiVobFuhE2olVQjFzFSdOtHEhLyLlDOeTIw/RqfZGCoiUEVVdXq0PZ0tJy3XXXqSV0dBxwr8icW0LP039V1/CU0W/N9taO1hcpQlxBMwiyuEJSDWXkTM1g5MlvziFCsJ2C7SBQZNZPiDvHIeetQdzqrI3Ditlr/D87Nh30mx/dZ2/ZKVFpDkghj2bNmrV3797e3t5BgwblZQTbDhSe9t6sb1Jq3QDrEgChQoTZv6RjFwPpctmmJM1pD6nYvscOVxdUJR+hTCXV4eqBtIScO5RFOxuKlpBILSHp6dM5twQ7wan2XSdaNnM2iF2EuIJmkChuKEM7CCvPuRlYbY7MfvDqiT891fnbx3M4QpGDVUt+qv2gWbZsmRUuWLDAFQUKldCSMYmFwUN1KlTW+EIl6vdLX9jU1KQ10QQLp+2ndPpCVZ+xm/ylKlRCc2qChfr9onRLS4vvJxx60KE2wQBp7zAjcFa8Kw2wqycUWyr9q1X3WiGKQ10C7Qv8AceBdAbUu1PXLofOwG9+dJ/+sAhHFtgLSENDw9SpU10G/SHCzEj6sYts9e14OemuIj31P9qaH9Af5hCaIl+SHq7OTc4dymKeDYVU+j19OnN2vCCH8MA6lMEn4KHIbChDCR9c5dwPSzwlwY9c2T9Xup92B1auLqbCjOI0g8tn/R81e+N7Wmr/rihQqIT2mMYXanFZifanCtF9obLGF6onN3+/9IVKKM43wUJtpY2/LVOqwpP384cJkhb6P5Fg4VP7qWOgrOjdXTtzPKXwLW/bd7vOF57eZtkMaV8wduLovt7dwYto0h97GvgBStX7oP38NnCAhVqJXNGgQX4PG+ZCdQjVK1CDUeLnt/za3s2BuoUuFXPn6nv9iq9/cddGqQPg31JPIO4qGxSOKpqHYWYueYSZ+eYg88KCruRShEKLLbu7u5W495cb7a0cxJ3mum7ZnX5joX9xd2vo/O3jVm53AWmcte9QZRFkXrOZFxauakzRCtWZUF9q9erVii5y7klk26Es1dlQmddv5oWFrqCiFe7YsUPd5euuu66+vj7nlhB3Xty9t/3RtwH9C97pR3qe+vtZc8fG7vrj3igwP1Yj/hBbqlEdVxQoTD9SFDemZKNMEix0o1QHjlO5okChEjbGJXGF2m6LL9SMWIkEC/cNscUonb4wNmi3z1FHHaVmYBftv7or93GMOH7wKul9X/xNoZTWxsEKi8CCKOOKkhUqoa6Y8YViJeqoiSuK9ds8K9GfqONu0hfq0xR5mmChwlqjidMXKsQ1+oQ0hfpw16wXLAgWLo0ZPXq0tgPqHqgl6K+OGZHjcx0UKw6tHqIQwuUzZoeeE6/KTnVZzUAOUNrTcbUc9u6ntL01wEItOle0d6+WavgL29vbVenaJK5bt+7iyz5k7w7ch6eeZ2v3Bz99risK0AbB3h158pvVReRmbwinQVpDXBL90UZEfQhtTebPn3/Eu9/I+Vi1uobaLiTGih0bHlbwoA2KjWZoq6F+w6j3n2axxM0Lf6ZdhUKL4I4h6Uf1PP1XhaPaBg3kqPaFNZNcCgfq7e1V11adCetb/HHnhtxaQlz9Bt25+l71MKzvKIoxtBexhmGVqy5I8LCltRz1OOPGt4N/6IqyR0tIRS1BGwQ1hquuukp91pxbwrpldypu9NXtWV0HW0hwfbdGElfpqT7KLt8ayKHuw7qPvf322y19wQUXWA9bsaIPF2Pd/n299qSF2mz6cNFCgmChRSN+Sv/nwUIf1qrQAomkhaoUX6iPDRb6OMcKxU8pvlAfawnRx1oiaaE+1l7VDPSbB9gMJNWuIW6DELe+J25JEluOidvF5IYNQhxV/ezYEyy0O/CtaOOWlm2v9Fg6W1ZNiXvwffca3fGyX4sTN+/WToIlWvGHHnV4XItKWpiVIwdX/8NbPuoylSpu+68G8Junb8t53be6i+vgia3dfmVP7N0lbjTi2omnKekWDpCqW/Wu6nZ5pMVZshlRk9IupLGxUV2rlpYWf/wyN9ocqBfoMhn7yLR92wW7N4yX5prMgVz5bUcokUidifr6+htvvNHO6fJ91hxYJ6An3GdD0RJSUUsYEzupwYYsBtIS7EzpbI9Dq6Og1V9dTJePSdMS1Kd0qeypGSjSUwfa+A2gChfsZ6FgqkJFZVpKxv+5L4wbU1LWBAv1V8Z34pMWqiL0VyauUFljheLyMa4o9rGeK0pRqI9VA9B+QfHnunXrNCMqGcgqY7sGdSJdPjPakqgldP15s8vvl+r2kmwQ8sU6BtOmTZsyZYq2A8FWpBjMpbKnWGLU+09TvKFQwRXFQot+uw12LXdwGFOfo79S9OLyFn707s7h9nJBNANFGtr+a8UPbv8Hsljs7sE5PHREQWPiRiPVwDXdwgHSBj94tBHpEWH2Q7sQO8NKiXxtSqxDGYwQMqHQQv0P7XVcfr+40/e9NMFnv/J1l4Io8UcZ4joTA2oJOXUo7Vhmcc6GoiUksqBCjUGxllqCDzlybgnH1hyt18QK7VftO09QpSeGpklbFDcazS81A4UWagbz588PbhAOOWiwJXLwzg+crtd7b/uTZTNnLcHvU2xnsTUhIFEb036BDcLA2e5AMcbJJ5/c3t6ujoF7Y7+DBw1oQZ136XtmLJ2snfuK/afE28hV+tMQ7FhDsJOgnYU+RyX+c1SokoG0AankYMOOLV533XXrYoKHnAay7qum7Ihhtv0BO6wQHHuwkrj9gm0cbF+TG9Z9ZIsIM524Aat8bUrszKW25gcsmznbcASvy7KORdwmyQ6D2Q0DcjOQuYueYGeiq6srrjNRkg6lHbb0uxDrWCQOh3b9ebM6mi6TE1pCkD/KkPRchpyXlarPDh5lO4xpt7b/892PWVZsE/HcEwecnqeP1VYihyfseRy9DvLNYPTo0YnNYCCDV2oJYyeO1qod91CKxAOLcawlBA9SaNuiSg8Ogt0Ze+aqPZ89Z7QEyeT8hSPflHsz8BRPKhr0/+LOb7QoVG3G5WPsT1xmP/8J+jeQk2O9IwbQyMuXrfg2ZK3YMjhkbQay7osdPli39M7gjqDfI49JDyuo5Dc/us/3DPWBdt51XGvJCus+skWEmZzvQySe/WIGuCmxe4uvmL0mGBz2uylxg1eBs6EsWNUmybLSseHhzt8+PmpgNxplU+LZUYY0nYmSdCg5G6rItEFYvXq1upVK2JBFfluCun3WJwie2tDvGVM2VKV695sRtSiVqDPh+yh6Sx+rQtt65IZm4PUbXQxw8MoGnYYedbgfdNI/VZ8K3RTJJG0J+pPgINjW2AW6A+liCi1BDWDhwoUKMNQAgged44w4LPcjvOEX7blLyrb/Rx11VKrtvwxw3RetoeoSaIvtV1t1BtRdTL/1rn3nCXoN7jsUrKpLoJ6hfYg+8IMDvt0067408LSSbHCnnySWLVt23XXX+au3XemBtu5+tr1noM+esiv1XSYm2I1QH1HbBW0mgndrSLyUX6zQZWLh6wDPgTlzRF3NkWe4TKVSLDFt2rTu2I2d4oYpgvLSEhQNqnfoMrGzZ4NHmpPesEdBqQLIuH6n9iUulfAhuaEliNpAY2OjNgVLA3fySDTwlmCrvMvExFW66jeuWtWrUDwZt5WwQpeJ3XR0IOGl0AxEzWB27PGnRdgghNY7jz77hGG1LoO07np2/cuv53h/6TA79KBDP3jSpS5TMZqammr2304sFdZ9IIgIM4lMNiXac2j/4TKRM2bkeccO2TdKVskUYaolqCuZ6iiDoSVEnoKK1tbWfo9cbnulZ+OWFpeJFpqBWDPod4Pw2ht7fvP0OpeJnAtOnMBQRoYe2dbevfPvZ7BHxrFDThgz8nyXQQDrPhDEWbJJqA+RPrwUrWkRvu65As+BSaR+ZJpBbI+WEHk1sbubukxqEb426cjBw12qglkz6HeDcOhBg6PaD9OGji5m5oYfdoxLRcvIIfvOyUQi1v3Ia2pq8g9PRr+IMHN3RER7XUcOrtaG0mWQAVoCRMsqkgG5mgF9i6wcG9EuOMebshLVxUUzSIN1P9o6eFpJNogwczdyyFtcKlqOOZwjlNmhJcAMf1MERy1oBtli8Apy6EGDo3fdmiIojjelwboPeESYuYvqRc/HDmVTkh1aAkzNUfsePxMxbz6co9fZYfAK5oRhKW82W6beErk5yi/WfcAjwsxdJM+5P3JwNddcZYuWABO9lqBmQN8iWwxewWjdidJC07ywNUiPdT/aGnhaSTaIMAfkbdVnuVRUcEZcbmgJMBFrCTSD3DB4BROlDcJbhtYognIZpMC6H2F1dXVpnoKLOESYAxKxc9MPPejQE4+I5gmfhUZLgIlSS6AZ5IzBK5gThtVGoyVoaxDJqwDyjnUfMESYA3LoQYNrjozONnfkkBOjtGUsJloCTJTOkopYV6nIGLyCiUZLOOmI02kDGWLdjyqeVpIVIsyBOjUqm5JDDzr05CNPcxlkj5YAc+aIfp6mWy4i06RLgsErGLWEI8v8YblqybSBzLHuRxVPK8kKEeZAHXrQ4FOPikI/LLYX5M4uuaMlwKglaBm6TNk6dsgJNIMBYvAKZtTRZ7tUeXrn0WfTBrLCug8QYeZBzVGnl/vxqkMPOpTxioGjJcCU+zAmzSAvGLyCOXLw8PK9jEK/nCvxssW6DxBh5sGhBw0u9w6lepMcqRo4WgJMubeEk444nQHMvGDwCubMEWPKMeRQmMHBptyw7kcPTyvJChFmfhw75MRjy/YekiMOG1lz5Bkug4GhJcBoSZbpMWz9bA5d5wuDV/DGjDzfpcrEoQcdSpiRM9b96OFpJVkhwsybUfs2xIe6TPnQb47MjUlCgpYAU3YdSok1gzH0KfOIwSsYLdKzj6t3mXKgpkuYMRCs+6hkRJh5oz5ZOXYoFQ5xOlx+0RJgtJ8eM/I8lykT6ljQp8w7Bq9gtHKVy4G8miNPj8Ady0qOdT9KeFpJVogw86mMdh5Gu5Bjh5zoMsgfWgKMlmoZ9dLUDDhNuhAYvIKnVUwrmsuElX6h2oDLYABY96OEp5VkhQgzz7TzKJcOJbuQgqIlwLzz6LPLYoetH8mZUYVTRkedtEFg8KqgtL0Nc5DJHiG/WPdRmYgw808dyvCvovqF9CYLjZYAc/Zx9SEPMo8cXD1m5HmcGVVQDF7B00IO5/OT9atoAHnHuo8KdPCCBQtcEvlz7JATXn5990t7el0+ZBRUnDmijt5kEdASYEYOOWHHq9tefr3P5cNE4eX7jqunGRTB0Ycf//obr/W++leXDxm6mMX05sNHjjhs5OZdYTnp7tCDDj3rze+tOYrz5AuCdT8CqqurzznnHL26PNIatHfvXpdEvj2yrb1752MuExrajpzKMw+Li5YAs3FLy7ZXelwmHNTHZfSyyB7f3vmXHZtcJjROPeqs04aPchkUy84929t77i35sacjB1crwAj5qRYRwLqPykGEWVjdOx99ZFuHy5TaoQcdqoiihjt5lAItASZUhxs4ylAq217p2bilxWVKTRsERRfhP6U/wkq7WeBklmJi3UeFIMIsuJAcodR2ZNTRZ3O/0BKiJcCE4XADRxlKjsErBJWkPaj2tR1gd1BkrPtlqqmpqbe3d+rUqS6PtIgwi+G1N/aoQ7l5V5fLF522IO88+uzDDxnq8igRWgKMehidL27cWaILdNWx4PGnIcHgFYK6dz761M7HixB72DEmNQBqv1RY98uO3bmG+9dkiAizeBRXPNG7qchHrTgFIoRoCTDFH8ykWxlCDF4hTkHjTDYC4cG6X16IMLNChFlsRetTshcJOVoCpGjD2moGRwwezgh2aDF4hThbdz/bs/u5fG0cbAtwwrCakUNOoOpDhXW/XBBhZoUIszQKfYTypCNOrznqdDYi4UdLgOzcs13NoEBxpvUszxxRx2mx4VfoDQL9y3KkLcO2V17IbfugSh9x2EhFlQSWIce6H34dHR3V1dU1NTUuj7SIMEtp6+5ntUHJ19MLrB85cshb2IiUHVoCTH47GUcOrj7m8BNOPKKWccvywuAVktKWQbuJ3a/1bX/1BcsGtxWqaL1qZbd/hx0yRLGlElR6GWHdR2QQYYaCupXbX3lh6+7NLp8NH04cO+RE+pHljpYAUSfyry/3qBOZ6qDD1mdfOPbEY1zmQNYMhr/pmGOHnsCgZbnLfPAqsUmoJTB4VTm6u7sZWomSTNb9VDsC1n2EARFmuOzcs/2lPb0coQQtAWbr7mdV9S/t2WFt4PU39jz79HNf/Me51/+/xepbqCUcctBg1wwOHvLmw0ceMbiaZhA9qvo0g1d/fW775y+6evWvvndK7SlsECpQb29vbW1te3s7QWb0pFr3+3b2TXn/v6y449rjTty3sru9AOt+IfG0kqwQYYYdByaRiFZRyWbPnr1s2bKGhoZ169a5IlQ2mkSFowFUICq9+LjTT1YOcv8jlBRIjBkzRq8uD9AqKltra2tTU9PSpUs7OjqUdqWoYDSJCmcNoLq62hKuFJFGpSP8iDBD7brrruvt7Z09e7bLA7SKyrZw4cL58+fPmjVLEcW0adNcKSoYTaLCqQFcddVVSqxatYoGUCGodIQfEWZ42aEpjkwjiFZRyZYtW6ZXuwikoaGhpqZm9erVsXdQoWgSFU67g+7ubjWA9vZ2NYBx48ZxCl/kUemloqXNRZhZ2Iuw0lZj1apVSqxbt079BitEhaNVVKzt27dXV1e3tLQoPWvWLL12dXWpRK+x91FxaBIVTg2grq7OGoBRifp1NIAIo9JRLrjTT0gtW7bs9ttv10bEsvX19VOmTOHYSYWjVVSyBQsW7NixY+nSpUoPGuQ23XZ+1KpVq/ZNgQpDk6hw2iPcddddcTd6Uat44IEHuPtLVFHpKBv7wkyEDEemkYhWUclUyzU1NWoDlvWbbpWovL293bKoHDSJChdX0XV1db4xqFzBhqURJVR6aWkJ20lkyATXYYbRsmXLpk6dOm7cOEvrVduOhoaGhQsXxt5HJaJVVLJp06ZNmTKlurra5fdTyfz587nTQwWiSVQ47QW0/VeMYdneGEurE0wDiCQqvbQ6Ojq4jX/miDBDR833xhtvVBfB5fdbunRpa2ur2rfLo5LQKiqZ3ddh1qxZLh+rd5faf5cXTWNZVAKaRIVLtUcw42K4+0vEUOkoL0SYocORaSSiVVSs3t7ehQsXrlq1Klj7wdBCOHpdUWgSSLVH8NQArrvuOsUkLo/yR6WjvBBhhgtHppGIVlHJWltba2pqxsXOjk6lrq5OE/CYigpBk6hwagBxewRpb28Pxh5Kz58/nwYQGVR6GPC0kqxwL9kQ6e3tra+vV/CQpuvQ0dGhabbHbk6NSkCrgNpA3HHr2bNnB48ymMTJEFU0iQpHzVYgKh3lhQgzRJqamm688cZ+7zfd2Nh4ySWXcBylQtAqkMg/mgIwNAkAQHiwTwoXjkwjEa0CcQgnEIcmUeHq6+tXrVpVU1Pj8qgAVHqRNTU1qaPFofwMcR1muCRGCPZcijgEEhWFVgEASEMdX3EZVAYqvch4WklWiDABoMwkjmCjwtEkKlx1dTXBRqWh0hFmRJgAUGbi7igI0CQAAOFxMI9nDbnq6upzzjnHZYAYWgUAwGtoaDjuuOMOO+wwl0cFoNKLzLpeenV5pMW9AQCgzCS91RMqGU0CABAeRJgAUGa4cSji0CQAAOHBdZhhN3v2bJcC9qNVFFnffi6fjJsi7TQAypRW7RkzZjQ2Ni5fvtwVJVizZo0muPzyy4u/HdD3tra2ugwqA5VeZE1NTatXr3YZ9IcIM+ySPpcCFY5WUWTqL6rXKBs2bHBFCRYvXqwJ0vQ+Ue66urpWxijhihKohWgCRRoca4iYoUOHTp8+XQlVcdIG0NPTs379eiU0mSa2wmLitqKhYtuKtrY2l0/gtydqOa4oe1R6MfG0kqwQYe7bK2gTIGk6BNoQ2DQuj8qgGldnIn29a4JUHQ5ExsiRIydOnKhEqt5Ac3NzZ2enupUzZ850RYXEFXclUVtbq1pWXac6jqCdiFqIJlC6yDEGTaIIxsYosXjxYisJUqtQA9AE48ePd0WobLatSBVA2rZC72r/4oqACCHC3NcP0HquHUb6ToMm2LhxoytCZVDbUKtQ1acKMhVbagI1D5dHdE2fPl39ANsauKL91EVYu3atEpMmTSpOXMGjKUpl7ty5eu3q6tK6byVBahtqIaNGjbLjEcVEkygOO4SkVT6uAShUKOYxJoRfml2G0GAQeUSY+6IIW8P3jVEmCyT8hsDOkCkyjkyXkHqKdsQ66R7C7zkmTJhQW1trhcVBqygJv6GIG7JWeKnGoKZS/LgCRaYu46RJk5RI3CaoYSjqsD2FXl0posV3GOxoghUqYceYSlj1q1atGjdunMsgHFLtMnyDGeBBSSq9yBoaGqZOneoy6A8R5j5pAome/VdWaEtRkj0HR6ZLy+pdzaA5duZbkMUVjFdUDtW1nf8WPEfOxxXWmSgObvVUQpMnT7ahieBpL77LWPzjTYYmUTTaCKiKVeO+w+CPMZXw/NjqGJdBOCTdZYhvMAPsPFDpRVZXV1dTU+My6A8RpqPeoToNiYGEHafUNsJCUFQaHzlol6DmYYVil2PpXcYrKoqd+OTPkfNxRdHOjzXc6qm0bJugNuCHJrQ1ULokx5sMTaKY7GRpawB+X1DMY0woF3G7DGlra7MGY2dDAFFFhOlYqKBEMJDQhkD8WyXBkemSGzt2rDqOwSPWPq5gvKLS+K2BHXuyuGLgh6JRXvzQhG0TtMtYv369tQ29xiZBlAVPlrZ9QcmrXjuFpqYml0FoBHcZ2lAED0oOvPNApRcZTyvJChHm36mbKD6Q8ImZJTo/1nBkOgzsiHXb/qspNmzY0NnZyXhFZfIbisWLF1tcwdhFBVKvUWGGtgPaLCyP3US0VMebUBJ2srQagGiDUKDzY9WuXKo/vTEugzDxuwzFlvk9KEmlFxlPK8kKEeYBrKdogYQ2BD09PZwfC/GHIRVUqFVoP2ElJTz0gBKyo07qWarTUIjzY/Wxambi8gm41VPJqdJtf6FtglpCCY83GZpE8anSLXHWWWdZIr/UrmbMmOEyKGe2rdiwYYMdlOT8WFQCIswDBDsN/qwnewsVTt3H2tpadfrnzZvHeAVG7n+CmU/khfqUc+bMUbfSNDY22pkUcbjVUxgowPAxRt4PNGgjo8ag2l++fPmaNWuUdm+kQJMosrbYLb5s9bcTIK08jtWjXl0+M5pen2n7GleEcuY7k6rQ8bE7RVk5EGEHL1iwwCURozV/48aNmzdvfu211/71X/81vxsC7YS0fYmj8sGDB9sEiaqrq8855xyXQUmdfvrpv/zlL1VlahWf//zn09TawHV1dfX29g4fPtzlD0SrKK2bb75Z/UtFF1qj29vbP/rRjyZtDKpEvZvVNmT9+vVqYGpdEydOHDNmjJrc2rVrN23aVKBz8DAQqt/bbrtNewqld+/eff7551v5wCkmWbRoUUtLi2XVitQwaAbhoZVU4Z+qfubMmVr31RK6u7uT1o52GUuWLLn00ksz3F9ok6JV3m5TrC2MPnny5Mn2Vnrjxo0788wzDzvsMJdHmKjBrFy50s5oVfdS7cHKB4hKLzLrenH/3gwxhhlPGwKxtPbolsgX7Tbc2ESAXfadCkemw8MPI2ivn+qIdV6oBar7Ii6fgFZRQoot7U6Ac+fOVRdQlRV8akWQVm0FBi6TmUmTJim00McqLrVrddSFVcPz9yE03Oqp5FTv6jLq1UYv1SrEvXcgtZY5c+a4TGb0yWoDikzUGNTMVqxYYc0gVUsTmkQxqSJU9fuurhs71l+Rq4p2bw+A1nTtX1TdqvpjjjnGlWZAvV46vqGlfYGqVa3FdhlxDy/x1Iq0rci8d0GlFxlPK8kKEWY8O93FDkZqh6GNgpXHUXmanX0q9snacwTV19e7txFiqnHtJNSV1E5C2VR7CFETiosHsqWGp52QyyBMVC+24ltcYSfVp4kusqXPdKn9rL3F9Tm41VPJaSVVd1D9xcmTJ1szUMNI2jXs7u7OdnVW0KLPDJ6ArR2HpGlmNImiUdWrIvzqr4SdAGmdh9gkuZs4caJ6BbbWIxrUWtRmrMFYm9m3w0i2Lqv9pOpzAmWHCPMACgxE+3XtMOxS7FSBREtLS9INRL9qamrUKQlKfxIdR6bDQB1EG6+YMGHC3Llz1UK0J9A+w70doGlUnm2HMkifvH79elpFOPmxC7uti1qCbSisPDbJPkpb1hLG3sqW/WFi5IkSUmwZvFBf7UErrGpKWwmbYIC0X3CpAO079BUDj2EwEFr+dtqRHWOyQjUAUSJ43FmNwV9Hffnllyst/VZfzmv6ggULeI5CCGmdDR6UDO4yYu8PCJVeZDytJCtEmH+nTb/tDGxDYPci94V5oW1NtvsPjkyHgYJGG6+wuMIOQ6phFOJwo3Y8sSMPSbqYHq2iJNQM2mJjF9ZFMLah8N0IM2/ePPUpNbE2IErIjFzvCWltLKuLOVFQFknqNXi7L/9AI7ESUYOxrYT9iXHvoTypBlWbfl/g2U4heK5sfX29WsX42PlQeldpyTmA7NeOHTt4cEUIab8Q12D8LiM4gKHthprWxo0blV67dm1sU9H/kDiVXmQ8rSQrRJh/p/VZ6/zYwIOtbJ+hHUYwkNAuRNsCW/Nj3Yl97K1+6fMLt4NBgfjxCn/QWnsLO2Id3EOocm0MXGltgyydbQhqjS0YwCAktMrb2EXibYQTT3xSyaJFizSZxM6F33c1nb2VFTUGfam2SHFHHHg0RQmpRlQv6iMGYwyfVYfS9wtfeOEFTWlZJYy9lYNNmzZp+6MvcvkD0SSKQNtnreOqBRu7DrIdhBI+MNC6r92EXbW1b4gzxvYgqBCpGkziLkP89kHtx9LqVNhbQDkiwnT8hiDYuU8aSCjYUFYT2yEo495Ly3c7UEZUy3boIa6Xb3sI1alajpWIprQgRM1DafF3g8yEPk1/rhaorokrQmioNtUMVDVxYxfiNxR2uFoJTaZCCwaUMPsmzcCGDRvUqESfpm1LfX19YneWWz2VilZtVY0S2gLERQtac21oQk3FSlRxi2LX1GlKO9Ag9la2tHHo7OycMGGCyyegSRSaevypjjEZG5tSIng6AyqW7dCVSGww2h34XYaVKKuNg/U/7QCl0BNAWSPC3MdvCBI79z6Q8J2GuXPnrlu3Tr1MdRqUMPZWetb11EfZ6Ja6C1aeHkemS0sNQx0L9RuChx5EtW9tQw1DE1jJzTffvGLFCqU1sdKSGBukouahnY12PIkBTCJaRZHZCqsqVoXq1ZUG+HjDbyhy1t3dvT5GX6qG98ILL7g3UGq2kiqhlTTxkIE1DyUUhYoV5oW+d/HixWoMmWwcUCDqG2iTrt29IklXlEDbf02g2MDlU9DGZM2BtLK793I1f/78qVOnugxCIM1BSbH+g63aVpIDKr3IGhoaWOCZI8J0nQa9qsfgz4/11GmwDUFz6vvKZkURi75O5s2bN2PGjH47IhyZLiHVjo1XJI0r1FrseMRA9hCehbLW2PpFqygy1bUdMkg1FKnmYRP0W4NqUep5BMVtBNTY1E+1ruqECRP0rjYUPQeeAcGtnkpCtTBp0iRVUNzxJm/s2LFqAJpAOxRXlIImiOPeSKDNgj3sJP1VfDSJMqI63ZjAvZcrHlwRKlqjzzrrLG0KtEFIutqqcMmSJZpAk6VZ/dOj0ouMp5VkZdDevXtdslLZVXZKqDOXqvtoIejI2D1mrUT9wg0bNqhDadkM6UP8tkYdRwsq1G+w8yUQNqplVdkxxxyT6qC1qs83Hos2Nf3ll1+udpLVaINFHcGWkFsDQ6gsXrxYMYm6ES4fo5qNCxe1x0ozKqIWpQBDrSt4JeegQWy6y4Z2H9rLrIid3eDNmzcv7jQWrfvBKjbaMmg3YbWfJrwUmkQ42bZdW/L01ZfUmjVrVPsZniSFCNBOX5sLbSvsdGugrLFPylHSAECbBoUcLhNz9tlnp+k7yowZM9SDTBNIzJ49m1Miy0jSCFOtInGcU11GC0r1rrqbilGDTSV9hEmrKAtJI8wcqDGonxrspBJOlJGkEabWem0rXCZGlWsbBKN3Ve9tbW2TJk3K5HAVTSKcihZh2g3GOb2lrGUbYVLpRdbU1NTb28uJshniLNl8UjwZJ9hjSEq9B/Uk4uLSIJ5LEQHaW6ii49guRBGI4pCxY8emPxIRh1ZRUY455hiXQlRo1zDqQHHhpcWlc+fO5dpLZEIdXx5cUWmo9CLjaSVZIcLMJ4sTgvo9/dWHGZZFJA0dOnR8AjukvWHDBtW+XmMPTfw7ldhwqKxZs8Y+B5VJuzS1FmswhhHsaFu7dq3Cy0WLFinydEX9oUmEWdx4dYHs2LHDpVAxqHSEFhEmUEpjx46dHuNGNvezAY1gGmVn5MiRWR08StoN7erqamtrm3DgYyo4LaqMDB06NKsAQzXe3NysGs/wTDlDkwgnO0awNnaz+jQnKw0cd3yJADuMmPnmgkpHmBFhllhn7BEIaQ5Uc2S6HGW+h1D0ODEZlathWDpxJJxWURZqamrUEhQtKM6Mu61LUppmzpw5ml49Uf2JXtesWTNv3jxFGmoGbiKUmyFDhqgZZH6sweKQ8Qk3Nkc5si38hg0bGhsbtXZndcgpK7NmzbrqqqtcBuUs8yMRVHqR8bSSrHBvgBwl3nujX+pkxE3c1ta2fPly9SSmZ/zURITf5ZdfPmrUqMTbQmYl/Z1+UC60gqselVCUGHevl0Tqfa5du9amN9piTJo0yZ9T7XGrpzKiLb+2CWoA2iyo79jvnZ+0Z9HqrxqPq3QzYcKEpIcbaBJhpjagqleFckIK0pszZ46airb52h2oZ0iDQfkiwsyRNgH2CAHt77u7uzMJEefNm6cdTE1Njf5K2w79lSJMpfu9DT3KiwUV6gUec8wxqtncxiKIMCPDxq8UYGS+mmt6OyCV6jxJbhxaXrS/aGlpUbWmfzKN0WTaNbhMAu0ykp7zQpMAIkBb/ubmZvUPtfGfNGkSnUOUL/ZJuVMAsH79evUGtMtftGiRK01N06uToa6GZfVXZ599duLoRByOTJcjxYd2VuSECRNyizDVWtIcuaBVVDjCCcShSVS41atXa5exYMECl0cFoNKLjKeVZIV9UtjRb0AiWkWFowEgDk2iwinYuOuuu1atWuXyqABUepFZME9InyHu9AMAZYYRbMShSQAAwoMIEwDKDI+mQByaBAAgPA5mtDfkqqurzznnHJcBYmgVAACvLoYHJFYUKr3IrOvFAs8QV24AQJnhVk+IQ5MAAIQHESYAlBlu64I4NAkAQHhwHWbYzZ4926WA/WgVAACvqamJ/UKlodKLTAt89erVLoP+EGGG3bJly1wK2I9WAQAI6u7udilUDCq9mDo6OljgmSPCBIAywxV3iEOTAACEBxEmAJQZHk2BODQJAEB48LSSsOO5FEhEqwAAeGeeeSbPUag0VHqRWdeLBZ4h7j4HAGWGR1MgDk0CABAeRJgAUGZ4NAXi0CQAAOHBdZhhx62okYhWAQDwWltbp02b5jKoDFR6kfG0kqxw1PMAva/0dmzpuP3R21u7W5W2f+69UlkQ+wcE0SoqHA0AcWgSFa5bAUdV1VSXQ0UoVqXXVNdUH1Zdd1zdJWdc0nBmgyutPHbnGu5fkyHGMP9OUWX9jfX6t+z3yxRndvd2lz68BIBEF7r/AYcmAaAw1B9Wr3h1x+rGtY1jVoxR2r0BpEaEuY8iydm/nK01h9UGQBngRsKIQ5OocIdVVb3ikqgUpah09ZMVZKrP7PJACpwluy+8nHb7tKZHmlw+bH5P1wEJaBUAgKDeqioeo1BpSlfpdcfVtc9od5nK0NHRUV1dXVNT4/JIiwizakHrgoV3LXSZqqphg4de/LYPXfy2Dw4bPOyIwUOVdW8AAAAAleT5XT2Pb3uyfcufb3nodlcUM+ucWUs/ykOSkFylR5it3a2Naxv99ZanjTjl/3f+LL1aFgAAAIDizP+8Z5leXb6qqmVKy7iacS4DBFR0hKnAsv7Gen/tpQLL7174TQYtAQAAgERTf/YvPsisPqx6+1e3Wzrympqaent7p07lls0Zqeg7/bR2t/rwUoHlVWdfQXgJAAAAJPXN8f/uUrGhmtUdlfKIyI6Oju7ubpdBfyo6wrz90b+fUP6Bk84dc9w7XQYAAADAgY4fNvKyd1ziMgf2pQGvoiPM4LNJ3n8St+YEAAAA0vls3SddKnY+oEsBARUdYfob/Ah39wEAAACQqKGhgYswM1fREWZ3799Ppz5+2EiXAgAAAJBM8K4lwdGaaKurq+NhmJmr6AgTAAAAAJBHRJgAAAAAkFJTU9Pq1ZVy49yBI8IEAAAAgJR4WklWiDABAAAAAPlBhAkAAAAAyA8iTAAAAABIiaeVZIUIEwija770rfNqJnz2Y7NdHplhuSFDF46erKay4js3uTyAvGIVQ8TwtJKsEGFGijbl2qCn+af+t5sUhWFVoD3rg394yBXlZNfOPr0+8uDjls2EvvHaf//eZRdcYXWthLID/BllJ4fl9tzTW6wnFMm1476W+6096N+vm+92pQmeeLjLplHCFUXdSzt26fXRPz9hWeMXV9NP7nBFJRW231MIqXZbn/3YbG3B0jRahFxZrGLR3v4DJUSEGSlxm/JE1v9G4VgVaM/65z8+bCXFoa7YFz7+lXU3/3zzU89biRLKqpBDyOm9uHWb9YR+s/63VhIlTz76lEtVVc2f+e1UAeTuXS/HJSqTX1x3/+I+S5RW2H5PIaTabT3y4OPagqnRXnbBFQpLXCnKXNiadLS3/8gvnlaSFSLMaPr+rd9O+u/q/7jSTYGcqINuxztTHX89451v0+sRRw0be8G7raQIFEOqK6bECScf/+VvfNHq+jNfvExZFfZ73KGM/Lr5bjv+ncehtqOPHaH6UqLx8outJMK+981VLoVkTjnjZEtc/PEPWaLQrv3376k9a6vi8gcq/u8pIb+f0r9rVy3wW7DNTz1/9bQFER7FrShhW8UqavuPAeJpJVkhwoymd73vHUn/veWk49wUyMnuXS/b8c6+nbutJM6ML3/m3u71v3hgzdveXuuKCmzH9pduuv4WJbSbvKFpScMnL7K61i9R9sq504YdOdSmjICtm1+wRB6H2rRSqL5Ua9E+/vLBCe/Xa9vdf6q0E6ezcm79e9US9O9DEz/gigrsuae36NW2KomK/3tKyO+n9E8zri3YLXfd8OVvfNHe/c7XrueM2QgI2ypWIdt/oPiIMIHy9tQTz1ii8fKLjxp+hKWNsp/6/D99/btfdXmUJ/WQFBbu2P6Sy+fk/0yZaIn/uHqpJYB8eeLhrgIduWj45EULl3/F0tf++/cGuBYAAIqDCBOIiGFHRGesEkF2ke3Cq77j8jk5+tgRdhrY5qeeZywI+TXloplqoj/+wf+6fF59aOIHbAT+pR27Wu64xwoBoMh4WklWBu3du9clK8+ghYNcqqrq3qnrXaqczfnMNW13/0mJe7v7n50LR0/WDvuIo4atbb0hbuxLHdD5M7+txLWrFpxb/14lnni4S30ITfyLB9bYBD+/9df2XSecfPzZ7x/zyX++NM0puJr+7l/et/G37Xamiv3JRy4Z9673vcMmCLrmS9/6zfrfXjl32qc+/0/PPb3lJ/99m/7c/lD9jP8zZWLSvzIP/uGhX93eqi+yG96k+W02R0rYstJX/PSGdXYPUv3VJZ+4cMKkjwQXi3r5dq1jIv0qP07Y9JM7vvO167WgVjYvC37pfS333/G/v9m1s88WmjnzXaeNv/j8uC8ymVelZll9OyX0s2+56wYrTM+q/vu3fltL0n6Y3eQgbsZ3bH9pzQ/X6S0tTM2RluRnvnhZmrN/4xa+/ckHPnpu+rOhMmwbl11whX1soi9/44sNn7zI0sHlZo3H/540jccWiOZuxpc/44r2V+XYD7x7yU1f16JYv/ZXt//0F/ZRqrhPXNGYar5s4g0/vyfVLW1vvGN5VidR20zZL3FFGVOn/3uL9117qepWkPnxD3xOaVVN3FrvW5G1CisM0sJsXvsrLUw/U/o9H7jw3PqLzk9svcFFpxXte99cZZWiENefihZczTVN04/vsNVcv00LNrjO6tOa1/zKvlef+fFpH7ONUhxNdvcv7tvR+1JwsWv6iz/+oaQ1dV7NBL3GLVW/WfCbPvELJ6lgbdoq8+ifn3j2qed9c7UVIa7taXlaXSTlPzPp7wkq2qY1PVuY9slWkrlMtnV+OWjV+5+fJRmEj1sOaTat/Ypbf9Mv0sy/1xq5FvtDDzxmE4t9eKq9Z9wWTB+uiSdO+kjSiYvQEiK5iknS7b9XhAVbps5bva9+zd75lRtKIBXGMCvXp7/wcb1q2/df875nJUabRUVTSmjH4HcAdtmbJtb2XR0CxZ/WJxBt5RV6TZ84S3sOKwlS4Wc/NlvTa7Nr21mxP9Eexb4ojr/hrfZn2kNoSv+H+hD9lTbZlo2z4js36V1N73c89kX6kMQ/8RfyaX7tF/q9pv5KnfI5U67RzFqJ7OxNeXZW8A69dn2mfvCLW7dZibl62gL9eL/QjL5RX3RFwxz9BleUPe2xtM9TQj9bu7fgb07Flmffrt1a/vbDrDw444o8J4274qbrb7GFqT/RZF+a/G9Ja1nT66vjFr79iRasFm/SX5VV20iz/JNeE6vfrzYZ/D36llSNx7497pZI/mPV+9Gi0JLxH6WK089O2no1UzZxsBMWp1T3a1XHVL0oJTS/6kNbYSYUpmolUmMIzpQas8JIzWxik/CLTm+pzfhmr+qwhARXc/U49ZbVgl6Vtu2JtRB9i/9efZRabGIlqo40md6NW+wqUU1pk+Xy/fFVE7wHb3rB2tRS1VLSl/qmIpoja3vBIb7dfenagP/MNL/HFk6Gq4/JedNacgoGFEUokbhaadtiu6TgctBkWgeTNs70NH3c+usXqWW9bL/3e99cpc9R2/ATi324WnvizXJVF1ZHwc2OWldizRanJUR1FROb98Rb4lXUKgbk3SDGME3ExjC/f+u+Ecg4J7/trXEHVrX1tL1FcODCPkS78+AonD/EeOa7TtOfKKT55IxLTzntZEUpt676mX2pCm9oWhL8Cu2Dta+1Lawd6Rx53NFK39d6v9/yJh449D/AJmi8/OKPXDJOiV/d3qq/UkJvJY67+rGaD054vx9q0277h0t/bPMYNwjg58i+SD/+n6/+tH5ez5YX//vaH9neK3hIXkGggsYnH39Ke1ll/a+So48d4RdUcMgoeLTywtGT3zH69Pf8w+hTzjh56LAhKtEX+XHg4NiO8VWZyXC0ZlPdbktrdj79hY+nP3hvh5Ztxm36d77n7fo92mX6yrJv18K86J8+qERwnDNxpNT/Wt8wlNay+smK22xJJo48ZNs2tLNXn8C3gS9/44v2LRJs2P6XmAwbT9Jj7VaVtpSU1bsfn/Yx1d2f//jwj75/qxUmjvjZWKv+ShVqx/XV0q5bdIMaoQrnL7tan5DtYWybqbifl6G4BukXe9xy8KtD4hz5T9CfaHmeO27fShRsvSqPG7G3P1Fj2Nn7kr5LiUs+ceHzz259+MHHfTNIuppre+JXWLUZLUm9q8QnrmhU81DbUOcy9tdVt979w+A3qnkolD37/WPOGPU2NWYrDLbAhcu/EjfMkrTS/XKIG45TuUvt9x9XL7WKDi5GdSv1jfoZb3tHrW+fwSYd/Nn2mdY2lNCSjxXv46sg1e8p5qY1E4UewxQ/WVwT9eX69vqLz9Pi1cL5wz3ttjVLbJzp+UWkVdX2F9ryt/z8Xq3ydv6Ol+33XvOlb21+eovaxqln1lhlqbXf+5uNtuTjtqu+ftX4/+1b/+IHtG0UNNhii9YSorqKSdLfGbZVLIQqcAyzqampt7eXE2UzxBhmNGkLnvgvcdRCuy5LaFOu7akSv26+2/aan//qlKR7ZW2ptdH09yzVPlgbZW1/9ZZ2Btq/2mTmhv/6kd8QazLtfvQn+qeN8nfXfFObWr2lLqN2XbHJD6A/1ATaJainbn+lhL7a3oq7Gkc7XesEKyL6+ne/6s9+2ffzbvy6dt5Kq+dqhXH0aTZH9vP0uuxHi+wt9SosIVoaetfv0o4/8Vj7VfqXdEHFUe9ES0A7VP0k+yt9kUrst9m+J2f6TO3dbXlqdrQotGtc8Z2brE5T0ZTqvqgbpF9lv+fbP3THoa0N6DO1MPXh+qeEr+W4+vJtRhMEb2arhLL6Cr2lZqOeQWxyJ9u2oTrVu1rsllVF2PT6l3RvnXnjSc9+pAJa/UirOy0ufZq9+//d2GwJo1DfOluaQHNkhfqTRbGTqPVRW59/QVkrLwktK3/ygqrACtPwa5aWp5qKaseWp7Vev0h/8O0bY5MfQItCb2kadZ1toSWe36gJgjWlJawV1mpfbcb+XH9lzUPfrl6p/eHGAzc1ah5axb7+3a9aY7Z/aoF+XVY8bInc+M+0f+pYW0VrOxlsfvpGzazmxa8F+qef7RuMYhVLiL17VLX7cz+9/llJGkXbtIZH0s2s3/hoDVXV2zSqES0QLQelNUfN2QzX26dpaagpWok+U58cF17m8L1f/+5X1ZJVQb6y9BV+yas5BUOsp554Rp+gxOdmf8rvzpTQ9Kpuy5qitQRWMfvDCK9i6BdPK8kKEWYFGXrkvqGzIO0zrMemTbniTwUkduKHQgVtvmOTxFMIp21lcIsvV/6b6/YF9zH6NAucFERpo2yFnu0sLW2DY4m0HY/bEzR8yv2qJx46YLPu9+Wf/8oUS3j6qZd84kIl1GFVd9kKgxLnSN0FFSqhnUHSP8kj+22SPhrsl/Z/CgBsHyb65dr5Kc78cep7b6he1JsP9tu0tC0glMQD0hfvf3xZZ/sjljA/vWGdJRIbhrIWXEnzmr/3twbeNvqVeePpV/A6T6MlYz2MRw48scqf9xW36LSQbcFm8pBxLRz1NYP/dsTOENZrfHlObWbCpI/44xr9Nm+/ZqlGgk3FqNA+StWU9MfYyuUyKcTVlNqMb8ZKxP15/cXnWeJPv3vQEunpN9uREVuGeaGF9oNv7YuoVaeptpNxfHvY9dLfz6jP2cBXnwGuHVoCcU3Ryp9/dmvS8rzwR5eCbOOj5ZBYEVoOVvU5b0PSyOP32hiX9O1K/gSsNEreEoRVzBJx8rgDAsoUEWY03du973lTcf+Sbql9d/N7i1ctvOo7dtDOR4yJzhj1NpcK0AbXOtwPPfCYlchDHY9awgdRcd53/hhLJF7/YPSxLrWfL4nrGW/87b4BDe2NEjvB4k/pibs20iSdI1+Y9E8KwT90JGead+32br37h8E4U9Wa9HIROfHk4+MCQjlh/wKMi5HEzhGS4HWP2hPbCUjqZCRd+Cq04MomMwNvG/3KvPH0yw9cB71j9Ol6taPsmbA1KxPr1/4q7uwDW3R6jStPPCshE6r0f77605ZO1TY8W7MksT0YOxYjSRvw/9n/iJQ0Emvq2BOOsYTvfHu+jQUvfs5EsPkNkJ0GqYQ/ByRzObfnoGJuWpPSEohrilauTnlceR6DzD/+7gGX2s9vfHwjjPOefxit1+BK+kTsqSpx/4LDULY3vOn6W378g//V51thnBy+NxPB6xKPPnaEJRbOujbNlXslbwlBrGJBeVywQJkiwqx06m7++7WzLW2n/Vw5d1rixrFf1uEOdqP9/tIHeHH01RaX2vcOhO3b9HpezYTEf+ro2GSlpX5J00/umPOZay674Ar/2+wUxDzycabv/ajbl6aPEidpvJ2GjyusU5WUj1rVmbNEMdtGMfnQKO6+HerCWnczzVLyEs81SCXzKeMoXLSwX4vXV0pStmbZGEVSfpb//MeHLTFwSUP6TKjrtuI7N2kVu3D0ZL+K5bcJaRXudzupRaousn6G/w36597Lh5KvPsOOzPTBSEOGHe5S+eM/0298FBAGF7X/l7h1/drnvxEMgO2f3drUfOMHX7NFp7+dNO4K1WPiaZA5fK+nTfE1X/rWZz8220+sH+DeC9BmfGHsKaDapc6f+W3tMhTxJkYmxW8JrGKmTPdQGCCeVpIVIkwccG6k+PPQisDi0uLQ/qAQ3Z0MqWOh/ordiy/bY9s5UAfl69/9qh/MHOAVMgOUbdRqitk28kWRm42BLJx1rQ8y1Rn65lf/r6UzWbkaPnlR3NkHFuDpNa48w/PHkrpq3hWWWL38gAvMspVzNJh36vjaDW+1igUPdeWRutf+5L1PJbulzY7Y/UUVM6yL3TXUlZZCQVcfbVvimqKVKySIK8/hYGUqfnlm+5m2ShqLCuIEC/Xh/nIDtSLVo+JP1WnSy+3SC36v6BMUKNqNSTMZ7tPG5Pu3ftvWfe0yFLWqeSusSjWymkoeWwKrWFA57qEK4gdVVfuvTJw9e/aY/VpbW6NXWFdXV1NTY2+hX0SY2HfIMLjD0z7MpQrPzqpNutfPQWIXPPjvFw+syWN3JyvqW6hjoV2y5vTL3/jirXf/0P8quw62QK74V3cyZGn3xI92uvOIsorw89s2isZ+sOr66mkL7Mi6OkO2fi1c/hV/kmfJvet977DOq9pG5kPciZ583B3sz3lANS+0EVPHVwl1TLWcf97+U7+K2WzmhbaN1rFOdfLef837nq1rH5zw/mtXLfC/Qf9sgqIp09UnFX+8JmltJka2wX/Be7T+z8+Wxr2rf3F38fGngWhbbVGi6jTpg5oy/17FRV/7/Dfs2OJnvnjZjXcs95MpjLRpEmklXXLT1zWx/sSqUmFV3GO0+pWvlsAqFidiq1jutlRVVbvk/PnzV+3nI7HoFSJDRJiVzh8y1E7CzqvU5lv7ktibWbCNfvCore9x+j5oHO0mbV8y8GOBNgZrW/wQavqxW57f/uE1DZ+8qGhhxlEJl1nmnQ8afRiZaPP+k7t8hF/MtlFMCtUUTGotUA/M2qQo23j5xeqwprqUsVT8zSr++9qUN5W1/lOaNWvr5hcsUdrBzJ+suE2v+rVLbtx318dCtPx+T97TttRu+6Ft6ddjN2G28ryL6uqTxq2rfmYJf78xOfltb7VEmo1PzrSV1rZaUaLiTGW1PP1mPIfv/cM97RZeqvHM+PJnsjrWqYn1J2tbb7CRVW1h/D3bi9kSWMWCoreKDcj+CLO6urpuPx+PRa8QGSLCrHT+kKG6m/+66ErrUCrm1AY09n68pPtUf4D57Pe7y99l1JgzLRG8iWiQ301+4MJzLZGzt8d685qR9BeVlYr2i5aIu7lcofnvzeMx5jjqB1ib0Y4/aZvRb7ARvOBdMYrZNorJzka+5BMXqgfmR0vs3vrhGb309JOsz6q+769udycFxbE1WmtW3JWlnnX41AaK3LbjWPddvb0CHVVRM05/8p74u4JlcrXtQER19UnFBx4nnHy8v8OKqK7tmObG37an2mENnOJMO1rkN6c5fK8/EJPqur5+6Uv9OSn+04rZEljFgiK2ig3I/vASiEOEWdGChwzV3dSe4/Nf3fe0D3Uo/2te8nNl1aGMO41Wu1j/qMngvR8Ve1hgowAjcVD0iYe77HPUN62/6HwrzJn/3usW3ZB0l69C3z8YiAEeNY/7DVoIwUdu5mbOZ67Rkkw8g0vfNe9L37J0QfeC9nxFSWwzWuz+NwRvK5pz2zjlDDdQlurQchhs+Pk9eWlsRfDJf77UDhCsi92aP5GvtaULViSuWaom63f6NlBaz8Z+TNCv9z+3cIA0p3YkLpObWz7/7FaX2s/ac1JnvNNdpZy4CqdSzE1raWk9uuZL3/rO16637OwFM+LCG7vVp6om1cNd9QlJ9whJqbVocxq38urPrZEH5fy9PVtedKkYTaN9lssEqAYTH2iceLvm4rcEVjGJ0iqWB7Pc/0AcIsxoejDhhuz2L7jv1N7LDhmecPLx/pBhwycvsg2rIslUoxbqjF52wRXar+gD9XpFwxxthVWuP4wbx5g6c7Il1EXQntv+RP+07/zS5H+z3cnVCQ9RzIG+14bI9EsmjbtCn29fpH/aPaiPYncFtIkHQj/Vjl5r+fz4B/+rz9dMpXnmpPFndincsp2cFr7+yi+EAbLbUahSNI/6WP3TLH/8A5+zejkz40eK5WbCpI9YlKJlot+gBe6XvG8bjZdfnJe2cfKpJ1pCTdf+RDObqqEWnx1Z1yxr4fs7HGqZaAb1TwskaaezX3brzkKMguoz0weHqjXbIKiTHVyztPA1RxaXqoGpDcQmLxk/GKtVwLZyelV6/syUF7llTjNrfWhtZHbvetmWgP/nN6paVrZx0GLxda0J/IJK6tQz3ZlX3/zq/1VL1vR+FtIo2qY1Q7YF8DcWzo1tu+yfZkTzpfXIBsll4fKvJJ4VGXy4a3Djo3/6EPuEzB/qs3XzC6ro6RNnHdDOp1xjyzN4gm623+tv8aXK9dsrfbjfQsZRA7jp+lv0rv9kJb7yua/bu8EbhhWtJbCK2Y8p1SoGlJ1D3P+IlqQ3QBd1Fpfc5PZS/zXPHTL0TysxV/7bNNvWL5x17drWG+I2oNrNPPzg49opxu1X1Mucf92XXWY/7Q/ULdDWXF+kz7SP9dQp0QY6Xxen/euifReVqTui79K+2e5JEJT5HfbT++SMS7XLUcLfkl5LNdVZPUbz+NMb1mmh6V/wzvii5Zlm15iJM975Nluw2vevSzjArN+WWC/5pRby3TXftPtY6J8tnKDPfPGyxIdW59Y2FBGpC2K17FvglXOnFe6CnKyoGTza+YTvExtbLEpoHhUYa1m9Lcs7Tn39u1/VP5fJN/WVb//pL+wXJqX2s/Cq7+jHJ12zrIGVvJv1yX++VF1A/UKtTcEVSg3p7PePiauRbPmTEvU5iR+lBulrZ/aCGVdPW6CE1oLgimCnWSaNJd53/hh1mrX89a79rXzkknHpDygUc9Oaibib5eQm1UM+tHy0k9Isu3yAGt43fvA1RQ5aekk3PuKvqeuXIrcfff/WpO1cG+rg8sz2e1Wb2gzqM/XhvpZN0l3Ax6d9THWa9JNV78G2UbSWwCrm3ogp/ioGlB3GMCPFnw2Sio+yntt/xbx2e3F7bnV/1WVXQhvWljvusULv+BOPXXLj1zWBHUoUbdm//I0v/s/PlibtZWoTvLJ5mb7FdgBGaZWoPOkG2n5kcPogbdn1mhgu6tu1E/r+rd/WDtv/NlEPWCXaVfhdlPG3qPEnXgb5zkHi7U8bPnmRPs3/PCWCx7btD/Uj/SOzjZaPlpL/K/1C/apb7/5hw6fc6GLcF6VfCEEK3uxmg5pTV7T/869dtWDJTV9PrBdbhklbi//9lg3yvzCxx6Y2c0PTEs1g4m9QjSSGlyaHtiH/uujKYPPTN4694O9fmlvjSbpAUlWlsQ+JW1D3tdxvq5XNuP1TLegH25LROqVQPDZtkaSfC7G+si3PpJNpArUitXn18/z8KqGsCpM2sH6/VNLUlG9piWuf2G+Iq0R1Fte23qCW4xuGPlmLXYXv/od3Ket/uWclcZXuvzG4WUhs8EFHVv999s+tf69WRtW+/zrVu5aSVn97Kmxi29PSW/ajRf5P9KoF68/GT/p7TNE2rYWWarelRad50epzy103JA0vjTY+WryJ7dP+XCtg5mdwWCtSs/GLyKpDH6Jwwkq8bL9Xm0HNiya2bPCT7c+DzUwNSXsHtQrfnm1zmvSGYcVpCVFdxSTp75TIrGJASQzau3evS1aeQQsHuVRV1b1Ti32r6/Ly4B8esnFR7VHSj9cBlWnH9pcmjbtCMaS6L3GHM8w1X/qWxZ/qJqY/fA4AQJidt3qCS1VV7Z1fuaEEUmEMEwDy4KGORxVeKhG8p1HQGaPcMfLdfS9bAgAAIHqIMAEgD/p27Y5LxLn9p7/Q6xFHDcv2OkwAAIAyQoQJAHngH729cNa1TT+5w9+rcMf2l+5ruX/OZ67ZHLubjj0QCAAAIKqIMAEgD95y0nELl++7VvmlHbu+87Xr/QNLLh7ziaunLbBbEV45d1pBnxwDAABQckSYyEiaWxoCMB+a+IFb7/7hZw68r6/Y7SX1FnfJAgAAkce9ZB3uJQsAAAD0i3vJIj3GMAEAAAAA+VHREWb1YdUuVVW1a0+fSwEAAAAAclLREWZNdY1LVVU9v2urSwEAAABI5vFtT7rUgX1pwKvoCLPuuDqXqqr6+RO/cSkAAAAAyfz26d+71IF9acCr6AjzkjMucamqqlseuj14SAYAAABA0PO7etY+dLvLVFVNGc1DnpFERUeY42rGBQf3/23Df7oUAAAAgIBde/r+856l/t4l6kWrL21pIKjS7/Sz6pJVLhO7FHPqz/7l+V09Lg8AAAAgdvmlwsv2LX92+aqqpR9dGrxrJuBV9PMwzexfzl72+2UuE3PZOy65+G0fPG3EKS4PAAAAVJ5de/qe37X150/85udP/Dr45IVZ58xShOkywIGIMPcZs2JMx5YOlwEAAACQQt1xdS1TWhjARCoVfZas1z6jfdY5s1wGAAAAQDJT66YSXiI9xjD/rmNLR+Paxu7ebpcPid6qKlZhxKFVQGgG8GgMlYzah1fgxmB3MGk4s8HlgRQYw/y7uuPquq7q0poztW5qWB4gqy3FD2KvgEergNAM4NEYKhm1D68wjUFRpd0zVj1k9ZMJL5EJxjBDbfbs2cuWLWtoaFi3bp0rQsWjVUBoBvBoDJWM2odHY0B4MIYZXq2trU1NTUuXLu3o6FDalaKy0SogNAN4NIZKRu3DozEgXPYirMaNG7dq1Sol1q1bV1NTY4WocLQKCM0AHo2hklH78GgMCBXGMENq2bJ9j+icOnWqXhsaGrSxWL16dewdVC5aBYRmAI/GUMmofXg0BoSOizQRJtu3b6+urm5paVF61qxZeu3q6lKJXmPvoxLRKiA0A3g0hkpG7cOjMSCEuNNPGC1YsGDHjh1Lly5VetAgV0fTpk3T66pVq/ZNgcpDq4DQDODRGCoZtQ+PxoAQcg0R4dHd3V1fX9/e3l5dve+RRn5j0dvbO2bMmHXr1tXV1cUmRAWhVUBoBvBoDJWM2odHY0A4cR1m6EybNm3KlCm2pQhSyfz58+2gFCoNrQJCM4BHY6hk1D48GgPCiQgzXJqamrq7u2fNmuXyVVV22oOxa7g1jWVRIWgVEJoBPBpDJaP24dEYEFpuMB1h0NvbW19fr63DuHHjXFGCjo4OTbN9+3aXR9TRKiA0A3g0hkpG7cOjMSDMGMMMkdbW1pqamjRbCqmrq9ME3IS6ctAqIDQDeDSGSkbtw6MxIMwYwwyX3t7euJPpZ8+evTRwzoNJnAwRRquA0Azg0RgqGbUPj8aA0CLCDDt/WzDAo1VAaAbwaAyVjNqHR2NASHCWLAAAAAAgP4gwAQAAAAD5cfCCBQtcEqFUXV19zjnnuAwQQ6uA0Azg0RgqGbUPj8aAkOB0bQAAAABAfnCWLAAAAAAgP4gww2727NkuBexHq4DQDODRGCoZtQ+PxoCQ4CzZsOPG00hEq4DQDODRGCoZtQ+PxoCQYAwTAAAAAJAfRJgAAAAAgPzgaSVhx42nkYhWAaEZwKMxVDJqHx6NASHB6doAAAAAgPzgLFkAAAAAQH4QYYYdN55GIloFhGYAj8ZQyah9eDQGhARnyYYdN55GIloFhGYAj8ZQyah9eDQGhARjmAAAAACA/CDCBAAAAADkB08rCTtuPI1EtAoIzQAejaGSUfvwaAwICU7XBgAAAADkB2fJAgAAAADygwgz7LjxNBLRKiA0A3g0hkpG7cOjMSAkOEs27LjxNBLRKiA0A3g0hkpG7cOjMSAkGMMEAAAAAOQHESYAAAAAID94WknYceNpJKJVQGgG8GgMlYzah0djQEhwujYAAAAAID84SxYAAAAAkB9EmGHHjaeRiFYBoRnAozFUMmofHo0BIcFZsmHHjaeRiFYBoRnAozFUMmofHo0BIcEYJgAAAAAgP4gwAQAAAAD5wdNKwo4bTyMRrQJCM4BHY6hk1D48GgNCgtO1AQAAAAD5wVmyAAAAAID8IMIMO248jUS0CgjNAB6NoZJR+/BoDAgJzpINO248jUS0CgjNAB6NoZJR+/BoDAgJxjABAAAAAPlBhAkAAAAAyA+eVhJ23HgaiWgVEJoBPBpDJaP24dEYEBKcrg0AAAAAyA/OkgUAAAAA5AcRZthx42kkolVAaAbwaAyVjNqHR2NASHCWbNhx42kkolVAaAbwaAyVjNqHR2NASDCGCQAAAADIDyJMAAAAAEB+8LSSsOPG00hEq4DQDODRGCoZtQ+PxoCQ4HRtAAAAAEB+cJYsAAAAACA/iDDDjhtPIxGtAkIzgEdjqGTUPjwaA0KCs2TDjhtPIxGtAkIzgEdjqGTUPjwaw/+/vTtIamPZ0gCMV2BYATB9E2AFhnEPQBF33LZWAKwAsQLECoAevwjEoMeYFQCTO4W3AuwV0Ocqy9llJGRJpISQvm/gyCpJBY46ZOZfVapiRjiHCQAAQBkSJgAAAGU4mf4Gf/+7akxS+3/+9+C//6tamLJ//VU1GGwqlVA37apQCUOabiUogxk19Q4hTLUYVMKQplUJ7zZPUAlDmmKf8D7FoBLoIWG+wXtMI6ZKlzEklUAy35WgDIakQyBRCSQqgcXjKlkAAADKkDABAAAoQ8IEAACgDAkTAACAMiRMAAAAypAwAQAAKEPCBAAAoAwJEwAAgDIkTAAAAMqQMAEAAChDwgQAAKAMCRMAAIAyJEwAAADKkDABAAAoQ8IEAACgDAkTAACAMiRMAAAAypAwAQAAKEPCBAAAoAwJEwAAgDIkTAAAAMqQMAEAAChDwgQAAKAMCRMAAIAy+ifM79+/H3Z1Op1q1cd0d3fXbrdbrVb8j6pVAAAATEb/hHlxcRHBLETIrFZ9TKenp/FfOD4+bjQa1SoAAAAmw1WyAAAAlCFhAgAAUIaECQAAQBkSJgAAAGVImAAAAJQhYQIAAFCGhAkAAEAZBRLmjx8/vn//np6f2Wq1zs/PY7F6bTh3d3edTid9PP6Ndmyzem0I+RdIPz22Vr0AAADAFH16fn6umjXNZjOiWjTW1tYeHh7Syl6R666uruKdfQPhwcHByclJtdDP4+Pj6elp5MloVKtqtre34+Obm5vVcj/xc4+Pj3t/gfzZ/B9ZXl5+enpKrxbz97+rxrz6119Vg8FUAsl8V4IyGJIOgUQlkKgEFs/45zDb7Xaj0Yh/XzvfGC+trKy8dkYxXt3a2op/+8bLEPE13hD5s1ruET/3tV8gfTZlSwAAAKZjzHOYkesODw+rhaWlzc3N7e3tz58/R/v+/r4eC5eXl6+vr3tPRUb4TMkw3hCvhtXV1ViMlRcXFzl2xqu3t7fxa6TFuoiX+QfljcTH6xfZxgfTppzDHIeDUkNSCSTzXQnKYEg6BBKVQKISWDzjJMyIcJHuUjuS29nZ2d7eXlpMIuDFG/K3Mftu5NOnT/HZ/f39b9++9QbIeno8OTk5ODhI7awecWM7l5eXEXHTYngRgIOEOQ5dxpBUAsl8V4IyGJIOgUQlkKgEFs84V8menp6mRt94GWJ9/bzl4+Nj7wWrkQlvb29brVbf85Ox2aq1tHRzc1O1auq/w4t4GSKRxi9QLQAAADAVIyfMTqeTT05GtuyNl1kEv6q1tHR1dVW1fokP9s2WSeTGHBp7v8kZa/JltLGdF/EyiZVCJgAAwDSNnDDreW9/f79q9RMBMmfIHEqHlz+bv1SZ1be2u7tbtXpEyPz27Vu1AAAAwISNnDDv7+9TY7l7c53Ufk0+wxkp8bV7xo6hft1s3xOYAAAATN/45zD/GC9DurtsMiBhphvAttvtVqt1+MtrjzkJ9VvFRtBNbQAAAN7XyAkzB8Vhol2+0jX0XuwaayJSbm1traysNBqNSJXHx8eRM5MBCXOk3wEAAIDpGDlhlhIBcmdnJyLlgCQJAADABzJywhzptGHvecsk1jebzZwtt7e3z87Obm9vn56enn8ZcJOe/Du8tn0AAACmb/yEOcztYesJsP69zfpFsEdHR9fX15En4w1DxlcJEwAAYAaNnDDrDxH5Y8Cr33i2/p3Mi4uL1IhU2Wq1Unt4OazGL1DwFrUAAAC8xcgJ88uXL1VraanT6VStfiL75Te8eKZIjqb12Dm8jY2NqjXwd7i7uxv8GwIAAFDQyAnz4OCgai0tHR4eVq1+6q/u7+9XreFEOh1wFW5+zGY4Pj6uWr+LeNloNP54lhUAAIBSxvkeZg54kd92dnZ6U1ysiXiZzx/G+1+cw8ynLuM9vfeSjTWx2QGXv8bvUL9QNt6c2lmk04iXLqAFAACYppETZjg7O8v32okst7KyEnkyGiESY6vVisjXbrfTG+KdJycnqZ3VT2nGm8/Pz6MRWTGyZWxqcLxM4neoWt3fYWtrK35i+gWazWaOl/XbCwEAADBRn56fn6tmTYS0lPrW1tYeHh7SyrqIghHqqoXXRcCLKNg35sXHe89eZpFLj46O0nW20X56ekrr6yJS1i/EfSFtIX50OsP52kbe5O9/V4159a+/qgaDqQSS+a4EZTAkHQKJSiBRCSyecc5hhkhuEdgGP7Ly4ODg+vr6tbOI6Qkl1cLvtre349V8LW5sKjVeiO33nh1NUrKtf2X0tY0AAABQSv9zmHfdu7BGKouw91pETB67N4y9v79PV6XGR9bW1jY2NiIfDhPq0g+6ubmJdvrs169f8088Pz+PzcbKAVE2tNvt9Av0/elDbmQcDkqRqASS+a4EZTAkHQKJSiBRCSye/gmToegySFQCyXxXgjIYkg6BRCWQqAQWz5hXyQIAAMALEiYAAABlSJgAAACUIWECAABQhoQJAABAGRImAAAAZUiYAAAAlCFhAgAAUIaECQAAQBkSJgAAAGVImAAAAJQhYQIAAFCGhAkAAEAZEiYAAABlSJgAAACUIWECAABQxqfn5+eqCQAAAG/gHCYAAABlSJgAAACUIWECAABQhoQJAABAGRImAAAAZUiYAAAAlCFhAgAAUIaECQAAQBkSJgAAAGVImAAAAJQhYQIAAFCGhAkAAEAZEiYAAABlSJgAAACUIWECAABQhoQJAABAGRImAAAAZUiYAAAAlCFhAgAAUIaECQAAQBkSJgAAAGVImAAAAJQhYQIAAFDGp+fn56rJR/Pjx492u/3z589quZ/9/f21tbVqgYWRauM///lPtJeXl4+OjuLf9BLzLXb99+/fr66u7u7uoh1rNjc3NzY2vn37pitYKLH3z8/PoxKikSthd3c3KiG9gUUTZRDjwv39fQwHZ2dn1VqACZAwP7CYR+7s7FQLrzg5OTk4OKgWWBitVuv4+LhaUAYLIxJF7PfHx8dq+Xcxp5QuFkEEiSiDKIYULF+InHl5eelww6KJCUOz2cydw9PTk8OOwOS4SvZ9RF+/1dXpdKpVo+s7e3jBEDLjIgpGGTQajddSwRhiU6enp9UCH8TOzk5UQkwBq+XRRWeSZ5CRIvb29iJPbm9vp1dDvBo9T7XATLq7u+uODG8aGg4PD9vtdhogYgiIYgjppRA/4o+HJnl3sQejDIoMDVEJsZ3Y6QVHGWZT6j2iB6iW4f1ImO8jOvoY5pNq1RvEVPLpFU5ZzLj7+/uogZhNDnO8YEgxusTWYmZZTxfMuMh+UQlvSYBp+hg7/bbr8vLy7Ozs+vo62ukNoX5mmxmUh4abm5tq1ehyJUQNPDw8pHp4fn4+OjrKb4gAk9rMpjw0pL05tvPz8/X19XTAwogw32LcT73HW45PQSkS5pyIONFX9TILI4aWNLr4Cu6i2dvbOzk5iUhZP2EVYjF/5yoSbMFjGcymr1+/poMLURL1UaDVasWa1HasYRFEvGw2m+mAY3QOl5eX1QsAEyZhwvyImUSaOEa29MXLRTNgp9fPXUiYc+9bV7Xwu3waUxksgpQtoxgeHh6ic3DQGZgad/qZtvT1mPv7+3SiKWZ+X758SS+FVqtVtYYQW2g0GtHY29tzbPJjydexpHt+RiOG/8+fP3dfrL5El9ojyTf4iXqILTSbzfPz81h0p5+ZFWWQCiDtuJgC7u/vd1/5R+zEFyckx/P4+Li+vp7aMdd0cnsGvRgaYr/v7u6ml8JIQ8MA8SNWVlZS2+g/g+JPNXXaeWiIfLi6utp98Z+jSK8dO+grdndssN6HfPr0KTWe3OlnjqRx5OfPn+nq9xfjSNSMPp93EGMMU1P/QlRf6dsyQ8qpMqah1So+iIh8ad/1FYNB9b5RRHJIM4ZcD3kuEj8urWHWDA6QR0dH1fveJvcVUSHVKmZJ/PGmHfSa6+vr6q1vk8eg8ToZJm3w0PD2v99qQ92EWa3i45vOOAIjcZXsVMWgnuRjh9GoVnXl9cy3GA/SHq+Wa7URxrsfw/HxcbomKl8Ix+yLfZ12elqcRIcQVZHvLVw/sM3siB2d+oS8x19UQkjr3yjfNqbUBilrEkMDc2/wODI4f8KkVEmT6co33njLsaV8XiJ6kIODg2+/xDZLHfBm0vLVsCOdvu4VezxtJyqhWuUc5oeS9lT8LVfLhaTbvaSNxzzDiYsZl3v1+h9yQbkYnNaYcbn3Ljuap20GXcH8iX2adm7xcQTG4BzmPHjs3nr+/Jfj4+P0bD3Pvlsc+QY/TmAuuOgBml2NRiM6gegKOp3OcvfRNTFVzafIWEB3vx6HE2Xgi9kATI6E+YFFnIiJQtjc3IzpY7pMonrt12O1000jmG/tdjtNHPf396Me0koW08XFRTrSlO8hFCURncPXr1/VxoJLT8qNho4CgImSMD+wdM1buL29ve56eHh47t4qIM8ems1m/uINcymmjOkEZqQI5yXY3d3tHm6qxJqokEib0RVsbW2lzMkCysehoip0FABMlIQ5i2JGGHPB9X6GmSDG7CHfMDBNLlObD+f8/Lza8b+L8qje8esGP9Fwfewci3xY7fvfHR4eVu/4Jf780/Gm5Ln71b50dUO6rsEhp49rwNAwuJ+PniQdh1ruPnnfCcyPLnZ3teN/F+WRhgOA9yVhzqKYAsZcMP7tNeQpiJhQ5qPUNzc3qcGHU+31HlEGaRoRjZg7RiN2dzphxVx6S4ewt7f38PCQbicYZZOSBh/Ua5UQqnf0iI+k62MjWJ6dnbmx5ByodnmP2NcSJjALPj175vJ7iFTQbDajcXR01PdR2rHy58+f1ULN/v5+/cuWA3Q6nUajEY14/x8ftsZ7iX2UTj7c3t72zvxixhCl0lsJq6ur6QhCTBzTE5Zj8fPnz90X/19+ZndkjI2NjdQwv5xN6Unor/21RpH0PVQUuzXfc3KwqKX19fXUfvKw9VmV++34i+77aMRRh4boAXZ2dlLqiHg5ZLXw7mKGkI4eXl9f9x49fG1oCIOfqJn6maATmD/xZ76yshINsz5mwj83lGXqijytZLD6s0yqVcye/PCA8Z5Wkj8+pJhfVp9kxqQdNNG/1jyhfOOjcZicsk8riR2dd7pHFn0snlbCqGKfpp1r1scscJXs3Hr0ZO0FkM5MDm91dbVqMZMmeoVbDhvOXSyC+tnLo6OjdNUDAEyBhPnO+l7lUsTV1VVqfPnyJTWYZeNFi1arVR0s6icfBc+nL/peks0iiLyRjjpFvHTUafblQ4Tjid3daDRSrxLZ0h/+xzXRo07MJTXDLJAw30c+hzD2NCI+OOCz+cb08YN88WaW5bl++sIkCyv1CTEzGLtPaDabr1VRbDbf4McdoWZZHhpSBz6eKKGIl6mQXvs+JzPu7ZMEFk3UTCqb6PBDWgnvRcJ8H3mS1+l00rf5oztIt2wZUnxwfX09phHx8TwCxUZiXhITzfwMg+HvDMS7yJetRgBI+zFCQrr3Dwsl34Ep3QMs/S0Pf9wh3h9dwdbW1s7OTvQkL/qEWJmKKuYfnmozy3IZ5BEh7dm0ckg5XsbWNjY24uN9vSXEMmn54qMYGtKeGm9oiE/Fx7NqbfcQRibEzo0836uPI/Yv7yNdO8f09b2l5/B34BjmsHSRe0UwUfmr+XUxSFQvv03vVbLMrHyLl7qR7gT2x2NJES/jp1TvZlb1vepk+Nu9xCBSfWYIbvcys4oMDVE21ScH2tvbqz7AB5fvIlk3uTtKwgDOYb6b6PpfhMzuBQ7VhTF/FLOQvhORZHt7O6aSLo6afbHHeycBfY8+jCGX0/B1xXuJSV7vvVhG2nFRSK/1CbGd6BPiDaPefJjpi367d2gY/lKU4d8Zmw3VAjMmdk3v0DD8zk1+uFRywcQQ0DuOlJpRwEg8D/OddTqddCFcjBwxBRx1/AixhRhF0lUQMSbFFoIO5cM5717tHHsw9l0oNfOL8kjbLLVBJip6g5AqITqE2HHVC6N40Sd0C0oBfDB5aBijEvLe/6ORNsu7SJUw9h9yqqLBYptjzD2YWf+MIm8eR+CNJEwAAADKcJUsAAAAZUiYAAAAlCFhAgAAUIaECQAAQBkSJgAAAGVImAAAAJSwtPR/z1QRmP6iyr8AAAAASUVORK5CYII=\n", "text/plain": [ "<IPython.core.display.Image object>" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image('./images/es_rnn.png')" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Using TensorFlow backend.\n" ] } ], "source": [ "from keras.models import Model\n", "from keras.layers import Input, GRU, Dense, Lambda\n", "from keras.callbacks import EarlyStopping" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "LATENT_DIM = 5 # number of units in the RNN layer\n", "BATCH_SIZE = 48 # number of samples per mini-batch\n", "EPOCHS = 10 # maximum number of times the training algorithm will cycle through all samples\n", "m = 24 # seasonality length" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Create custom layers in Keras\n", "In this section we define two custom layers:\n", "- ***ES*** layer: This layer implements the Expomemtial Smoothing, normalization and de-seasonalization for input data.\n", "- ***Denormalization*** layer: This layer takes the normalization and de-seasonalization coefficients from ES layer and multiply them to output of RNN for de-normalization and seasonalization.\n", "\n", "\n", "There are 3 methods you need to implement in your custom layer:\n", "- build(input_shape): this is where you will define your weights.\n", "- call(x): this is where the layer's logic lives.\n", "- compute_output_shape(input_shape): in case your layer modifies the shape of its input, you should specify here the shape transformation logic. \n", "\n", "You can check [Keras documentation](https://keras.io/layers/writing-your-own-keras-layers/) for more details about creating custom layer." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "from keras import backend as K\n", "from keras.layers import Layer\n", "from keras import initializers\n", "\n", "# Exponential Smoothing + Normalization\n", "class ES(Layer):\n", "\n", " def __init__(self, horizon, m, batch_size, time_steps, **kwargs):\n", " self.horizon = horizon\n", " self.m = m\n", " self.batch_size = batch_size\n", " self.time_steps = time_steps\n", " \n", " super(ES, self).__init__(**kwargs)\n", "\n", " # initialization of the learned parameters of exponential smoothing\n", " def build(self, input_shape):\n", " self.alpha = self.add_weight(name='alpha', shape=(1,),\n", " initializer='uniform', trainable=True)\n", " self.gamma = self.add_weight(name='gamma', shape=(1,),\n", " initializer='uniform', trainable=True)\n", " self.init_seasonality = self.add_weight(name='init_seasonality', shape=(self.m,),\n", " initializer=initializers.Constant(value=0.8), trainable=True)\n", " self.init_seasonality_list = [K.slice(self.init_seasonality,(i,),(1,)) for i in range(self.m)]\n", " self.seasonality_queue = deque(self.init_seasonality_list, self.m)\n", " self.level = self.add_weight(name='init_level', shape=(1,),\n", " initializer=initializers.Constant(value=0.8), \n", " trainable=True)\n", " super(ES, self).build(input_shape) \n", "\n", " def call(self, x):\n", "\n", " # extract time-series from feature vector\n", " n_examples = K.int_shape(x)[0]\n", " if n_examples is None:\n", " n_examples = self.batch_size\n", " x1 = K.slice(x,(0,0,0),(1,self.time_steps,1))\n", " x1 = K.reshape(x1,(self.time_steps,))\n", " x2 = K.slice(x,(1,self.time_steps-1,0),(n_examples-1,1,1))\n", " x2 = K.reshape(x2,(n_examples-1,))\n", " ts = K.concatenate([x1,x2])\n", " \n", " x_norm = [] # normalized values of time-series\n", " ls = [] # coeffients for denormalization of forecasts\n", " \n", " l_t_minus_1 = self.level\n", " \n", " for i in range(n_examples+self.time_steps-1):\n", " \n", " # compute l_t\n", " y_t = ts[i]\n", " s_t = self.seasonality_queue.popleft()\n", " l_t = self.alpha * y_t / s_t + (1 - self.alpha) * l_t_minus_1\n", " \n", " # compute s_{t+m}\n", " s_t_plus_m = self.gamma * y_t / l_t + (1 - self.gamma) * s_t\n", " \n", " self.seasonality_queue.append(s_t_plus_m)\n", " \n", " # normalize y_t\n", " x_norm.append(y_t / (s_t * l_t))\n", "\n", " l_t_minus_1 = l_t\n", "\n", " if i >= self.time_steps-1:\n", " l = [l_t]*self.horizon\n", " l = K.concatenate(l)\n", " s = [self.seasonality_queue[i] for i in range(self.horizon)] # we assume here that horizon < m\n", " s = K.concatenate(s)\n", " ls_t = K.concatenate([K.expand_dims(l), K.expand_dims(s)])\n", " ls.append(K.expand_dims(ls_t,axis=0)) \n", " \n", " self.level = l_t\n", " x_norm = K.concatenate(x_norm)\n", "\n", " # create x_out\n", " x_out = []\n", " for i in range(n_examples):\n", " norm_features = K.slice(x_norm,(i,),(self.time_steps,))\n", " norm_features = K.expand_dims(norm_features,axis=0)\n", " x_out.append(norm_features)\n", "\n", " x_out = K.concatenate(x_out, axis=0)\n", " x_out = K.expand_dims(x_out)\n", "\n", " # create tensor of denormalization coefficients \n", " denorm_coeff = K.concatenate(ls, axis=0)\n", " return [x_out, denorm_coeff]\n", "\n", " def compute_output_shape(self, input_shape):\n", " return [(input_shape[0], input_shape[1], input_shape[2]), (input_shape[0], self.horizon, 2)]\n", " \n", "class Denormalization(Layer):\n", " \n", " def __init__(self, **kwargs):\n", " super(Denormalization, self).__init__(**kwargs)\n", "\n", " def build(self, input_shape):\n", " super(Denormalization, self).build(input_shape) \n", "\n", " def call(self, x):\n", " return x[0] * x[1][:,:,0] * x[1][:,:,1]\n", "\n", " def compute_output_shape(self, input_shape):\n", " return input_shape[0]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Create ES-RNN model\n", "Since Denormalization layer has inputs from two previous layers, we need to use functional API of Keras to create the model." ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "scrolled": false }, "outputs": [], "source": [ "model_input = Input(shape=(None, 1))\n", "[normalized_input, denormalization_coeff] = ES(HORIZON, m, BATCH_SIZE, T)(model_input)\n", "gru_out = GRU(LATENT_DIM)(normalized_input)\n", "model_output_normalized = Dense(HORIZON)(gru_out)\n", "model_output = Denormalization()([model_output_normalized, denormalization_coeff])\n", "model = Model(inputs=model_input, outputs=model_output)" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "model.compile(optimizer='RMSprop', loss='mse')" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "__________________________________________________________________________________________________\n", "Layer (type) Output Shape Param # Connected to \n", "==================================================================================================\n", "input_1 (InputLayer) (None, None, 1) 0 \n", "__________________________________________________________________________________________________\n", "es_1 (ES) [(None, None, 1), (N 27 input_1[0][0] \n", "__________________________________________________________________________________________________\n", "gru_1 (GRU) (None, 5) 105 es_1[0][0] \n", "__________________________________________________________________________________________________\n", "dense_1 (Dense) (None, 3) 18 gru_1[0][0] \n", "__________________________________________________________________________________________________\n", "denormalization_1 (Denormalizat (None, 3) 0 dense_1[0][0] \n", " es_1[0][1] \n", "==================================================================================================\n", "Total params: 150\n", "Trainable params: 150\n", "Non-trainable params: 0\n", "__________________________________________________________________________________________________\n" ] } ], "source": [ "model.summary()" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "earlystop = EarlyStopping(monitor='val_loss', min_delta=0, patience=20)" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train on 23328 samples, validate on 1488 samples\n", "Epoch 1/10\n", "23328/23328 [==============================] - 8s 341us/step - loss: 0.0324 - val_loss: 0.0194\n", "Epoch 2/10\n", "23328/23328 [==============================] - 4s 153us/step - loss: 0.0029 - val_loss: 0.0184\n", "Epoch 3/10\n", "23328/23328 [==============================] - 4s 156us/step - loss: 0.0023 - val_loss: 0.0163\n", "Epoch 4/10\n", "23328/23328 [==============================] - 4s 155us/step - loss: 0.0019 - val_loss: 0.0140\n", "Epoch 5/10\n", "23328/23328 [==============================] - 4s 157us/step - loss: 0.0016 - val_loss: 0.0120\n", "Epoch 6/10\n", "23328/23328 [==============================] - 4s 154us/step - loss: 0.0015 - val_loss: 0.0103\n", "Epoch 7/10\n", "23328/23328 [==============================] - 4s 154us/step - loss: 0.0014 - val_loss: 0.0091\n", "Epoch 8/10\n", "23328/23328 [==============================] - 4s 156us/step - loss: 0.0013 - val_loss: 0.0081\n", "Epoch 9/10\n", "23328/23328 [==============================] - 4s 154us/step - loss: 0.0012 - val_loss: 0.0074\n", "Epoch 10/10\n", "23328/23328 [==============================] - 4s 151us/step - loss: 0.0012 - val_loss: 0.0068\n" ] }, { "data": { "text/plain": [ "<keras.callbacks.History at 0x17909381828>" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.fit(train_inputs['X'],\n", " train_inputs['target'],\n", " batch_size=BATCH_SIZE,\n", " shuffle=False,\n", " epochs=EPOCHS,\n", " validation_data=(valid_inputs['X'], valid_inputs['target']),\n", " callbacks=[earlystop],\n", " verbose=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Evaluate the model" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "look_back_dt = dt.datetime.strptime(test_start_dt, '%Y-%m-%d %H:%M:%S') - dt.timedelta(hours=T-1)\n", "test = energy.copy()[test_start_dt:test_end_dt][['load']]\n", "test[['load']] = scaler.transform(test)\n", "test_inputs = TimeSeriesTensor(test, 'load', HORIZON, tensor_structure)" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "predictions = model.predict(test_inputs['X'], batch_size=BATCH_SIZE)" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[0.54, 0.69, 0.72],\n", " [0.48, 0.61, 0.63],\n", " [0.46, 0.55, 0.57],\n", " ...,\n", " [0.58, 0.59, 0.68],\n", " [0.6 , 0.69, 0.77],\n", " [0.69, 0.78, 0.76]], dtype=float32)" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "predictions" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "text/html": [ "<div>\n", "<style scoped>\n", " .dataframe tbody tr th:only-of-type {\n", " vertical-align: middle;\n", " }\n", "\n", " .dataframe tbody tr th {\n", " vertical-align: top;\n", " }\n", "\n", " .dataframe thead th {\n", " text-align: right;\n", " }\n", "</style>\n", "<table border=\"1\" class=\"dataframe\">\n", " <thead>\n", " <tr style=\"text-align: right;\">\n", " <th></th>\n", " <th>timestamp</th>\n", " <th>h</th>\n", " <th>prediction</th>\n", " <th>actual</th>\n", " </tr>\n", " </thead>\n", " <tbody>\n", " <tr>\n", " <th>0</th>\n", " <td>2014-10-31 16:00:00</td>\n", " <td>t+1</td>\n", " <td>3,717.48</td>\n", " <td>3,437.00</td>\n", " </tr>\n", " <tr>\n", " <th>1</th>\n", " <td>2014-10-31 17:00:00</td>\n", " <td>t+1</td>\n", " <td>3,543.91</td>\n", " <td>3,466.00</td>\n", " </tr>\n", " <tr>\n", " <th>2</th>\n", " <td>2014-10-31 18:00:00</td>\n", " <td>t+1</td>\n", " <td>3,483.05</td>\n", " <td>3,374.00</td>\n", " </tr>\n", " <tr>\n", " <th>3</th>\n", " <td>2014-10-31 19:00:00</td>\n", " <td>t+1</td>\n", " <td>3,409.18</td>\n", " <td>3,315.00</td>\n", " </tr>\n", " <tr>\n", " <th>4</th>\n", " <td>2014-10-31 20:00:00</td>\n", " <td>t+1</td>\n", " <td>3,340.66</td>\n", " <td>3,142.00</td>\n", " </tr>\n", " </tbody>\n", "</table>\n", "</div>" ], "text/plain": [ " timestamp h prediction actual\n", "0 2014-10-31 16:00:00 t+1 3,717.48 3,437.00\n", "1 2014-10-31 17:00:00 t+1 3,543.91 3,466.00\n", "2 2014-10-31 18:00:00 t+1 3,483.05 3,374.00\n", "3 2014-10-31 19:00:00 t+1 3,409.18 3,315.00\n", "4 2014-10-31 20:00:00 t+1 3,340.66 3,142.00" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "eval_df = create_evaluation_df(predictions, test_inputs, HORIZON, scaler)\n", "eval_df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Compute MAPE for each forecast horizon" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "h\n", "t+1 0.03\n", "t+2 0.08\n", "t+3 0.11\n", "Name: APE, dtype: float64" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "eval_df['APE'] = (eval_df['prediction'] - eval_df['actual']).abs() / eval_df['actual']\n", "eval_df.groupby('h')['APE'].mean()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Compute MAPE across all predictions" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.07506279816956267" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "mape(eval_df['prediction'], eval_df['actual'])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Plot actuals vs predictions at each horizon for first week of the test period. As is to be expected, predictions for one step ahead (*t+1*) are more accurate than those for 2 or 3 steps ahead" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "<Figure size 1080x576 with 1 Axes>" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plot_df = eval_df[(eval_df.timestamp<'2014-11-08') & (eval_df.h=='t+1')][['timestamp', 'actual']]\n", "for t in range(1, HORIZON+1):\n", " plot_df['t+'+str(t)] = eval_df[(eval_df.timestamp<'2014-11-08') & (eval_df.h=='t+'+str(t))]['prediction'].values\n", "\n", "fig = plt.figure(figsize=(15, 8))\n", "ax = plt.plot(plot_df['timestamp'], plot_df['actual'], color='red', linewidth=4.0)\n", "ax = fig.add_subplot(111)\n", "ax.plot(plot_df['timestamp'], plot_df['t+1'], color='blue', linewidth=4.0, alpha=0.75)\n", "ax.plot(plot_df['timestamp'], plot_df['t+2'], color='blue', linewidth=3.0, alpha=0.5)\n", "ax.plot(plot_df['timestamp'], plot_df['t+3'], color='blue', linewidth=2.0, alpha=0.25)\n", "plt.xlabel('timestamp', fontsize=12)\n", "plt.ylabel('load', fontsize=12)\n", "ax.legend(loc='best')\n", "plt.show()" ] } ], "metadata": { "kernelspec": { "display_name": "Python (dlts)", "language": "python", "name": "dlts" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.6" } }, "nbformat": 4, "nbformat_minor": 2 }