{ "cells": [ { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "# English Premiership 2014-15 Data Analysis" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This notebook demonstrates the power of [IPython Notebook](http://ipython.org/notebook.html), [python](https://www.python.org/), [pandas](http://pandas.pydata.org/) and [matplotlib](http://matplotlib.org/) to analyse a data set. The data set I am using is the English premier league 2014-15 football match results as of 27th April 2015." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Set-up" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's import the modules we need for the analysis." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": true }, "outputs": [], "source": [ "import pandas as pd\n", "import matplotlib as mpl\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import sys \n", "from itertools import cycle\n", "from collections import defaultdict\n", "from datetime import datetime\n", "from __future__ import division\n", "\n", "# enable inline plotting\n", "%matplotlib inline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Print version numbers." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "python version: 2.7.8 |Anaconda 2.2.0 (64-bit)| (default, Jul 2 2014, 15:12:11) [MSC v.1500 64 bit (AMD64)]\n", "pandas version: 0.15.2\n", "matplotlib version: 1.4.3\n", "numpy version: 1.9.2\n" ] } ], "source": [ "print 'python version: {}'.format(sys.version)\n", "print 'pandas version: {}'.format(pd.__version__)\n", "print 'matplotlib version: {}'.format(mpl.__version__)\n", "print 'numpy version: {}'.format(np.__version__)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Data Load" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The data file is located in the *data* sub-directory. The file was downloaded from [english premiership stats](http://www.football-data.co.uk/englandm.php). Here you can find the latest premier league match statistics in a csv data file and the [notes](http://www.football-data.co.uk/notes.txt) file that describes the csv structure. **The csv file used here was downloaded on 27th April 2015.**\n", "\n", "Let's read the premiership data into a pandas dataframe. " ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "collapsed": false }, "outputs": [], "source": [ "DATA_FILE = 'data/E0.csv'\n", "DATA_FILE_DATE = '27th April 2015'\n", "df = pd.read_csv(DATA_FILE)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Check the number of rows and columns." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The dataframe has 337 rows and 65 columns.\n" ] } ], "source": [ "r, c = df.shape\n", "print \"The dataframe has {} rows and {} columns.\".format(r, c)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's display the first 5 rows." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DivDateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGHTR...BbMx>2.5BbAv>2.5BbMx<2.5BbAv<2.5BbAHBbAHhBbMxAHHBbAvAHHBbMxAHABbAvAHA
0 E0 16/08/14 Arsenal Crystal Palace 2 1 H 1 1 D... 1.77 1.72 2.26 2.10 24-1.50 1.81 1.78 2.20 2.10
1 E0 16/08/14 Leicester Everton 2 2 D 1 2 A... 2.10 2.00 1.90 1.80 22 0.25 1.88 1.85 2.10 2.02
2 E0 16/08/14 Man United Swansea 1 2 A 0 1 A... 1.77 1.71 2.30 2.13 25-1.50 2.18 2.08 1.87 1.79
3 E0 16/08/14 QPR Hull 0 1 A 0 0 D... 2.52 2.36 1.65 1.58 24 0.00 1.80 1.73 2.25 2.14
4 E0 16/08/14 Stoke Aston Villa 0 1 A 0 0 D... 2.42 2.31 1.67 1.60 23-0.50 1.95 1.91 2.02 1.96
\n", "

5 rows × 65 columns

\n", "
" ], "text/plain": [ " Div Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG HTR \\\n", "0 E0 16/08/14 Arsenal Crystal Palace 2 1 H 1 1 D \n", "1 E0 16/08/14 Leicester Everton 2 2 D 1 2 A \n", "2 E0 16/08/14 Man United Swansea 1 2 A 0 1 A \n", "3 E0 16/08/14 QPR Hull 0 1 A 0 0 D \n", "4 E0 16/08/14 Stoke Aston Villa 0 1 A 0 0 D \n", "\n", " ... BbMx>2.5 BbAv>2.5 BbMx<2.5 BbAv<2.5 BbAH BbAHh BbMxAHH \\\n", "0 ... 1.77 1.72 2.26 2.10 24 -1.50 1.81 \n", "1 ... 2.10 2.00 1.90 1.80 22 0.25 1.88 \n", "2 ... 1.77 1.71 2.30 2.13 25 -1.50 2.18 \n", "3 ... 2.52 2.36 1.65 1.58 24 0.00 1.80 \n", "4 ... 2.42 2.31 1.67 1.60 23 -0.50 1.95 \n", "\n", " BbAvAHH BbMxAHA BbAvAHA \n", "0 1.78 2.20 2.10 \n", "1 1.85 2.10 2.02 \n", "2 2.08 1.87 1.79 \n", "3 1.73 2.25 2.14 \n", "4 1.91 2.02 1.96 \n", "\n", "[5 rows x 65 columns]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's check the columns, this should align with the csv structure mentioned earlier." ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "['Div' 'Date' 'HomeTeam' 'AwayTeam' 'FTHG' 'FTAG' 'FTR' 'HTHG' 'HTAG' 'HTR'\n", " 'Referee' 'HS' 'AS' 'HST' 'AST' 'HF' 'AF' 'HC' 'AC' 'HY' 'AY' 'HR' 'AR'\n", " 'B365H' 'B365D' 'B365A' 'BWH' 'BWD' 'BWA' 'IWH' 'IWD' 'IWA' 'LBH' 'LBD'\n", " 'LBA' 'PSH' 'PSD' 'PSA' 'WHH' 'WHD' 'WHA' 'SJH' 'SJD' 'SJA' 'VCH' 'VCD'\n", " 'VCA' 'Bb1X2' 'BbMxH' 'BbAvH' 'BbMxD' 'BbAvD' 'BbMxA' 'BbAvA' 'BbOU'\n", " 'BbMx>2.5' 'BbAv>2.5' 'BbMx<2.5' 'BbAv<2.5' 'BbAH' 'BbAHh' 'BbMxAHH'\n", " 'BbAvAHH' 'BbMxAHA' 'BbAvAHA']\n" ] } ], "source": [ "print df.columns.values" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Data Cleanse" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As you can see from the notes file, and the list of column names, the csv contains lots of betting data - starting at 'B365H'; I'm not interested in the betting data so let's drop these columns." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "collapsed": false }, "outputs": [], "source": [ "betting_cols_to_drop = range(df.columns.get_loc('B365H'), df.columns.size)\n", "df.drop(df.columns[[betting_cols_to_drop]], inplace=True, axis=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There are a few other columns I'm not interested in, so let's also drop them." ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "collapsed": false }, "outputs": [], "source": [ "other_cols_to_drop = ['Div', 'HTR', 'HF', 'AF', 'HC','AC', 'HY', 'AY']\n", "df.drop(other_cols_to_drop, inplace=True, axis=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's show the new shape and column names." ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Shape of data set is: (337, 15)\n", "Column names are: ['Date', 'HomeTeam', 'AwayTeam', 'FTHG', 'FTAG', 'FTR', 'HTHG', 'HTAG', 'Referee', 'HS', 'AS', 'HST', 'AST', 'HR', 'AR']\n" ] }, { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGRefereeHSASHSTASTHRAR
0 16/08/14 Arsenal Crystal Palace 2 1 H 1 1 J Moss 14 4 6 2 0 1
1 16/08/14 Leicester Everton 2 2 D 1 2 M Jones 11 13 3 3 0 0
2 16/08/14 Man United Swansea 1 2 A 0 1 M Dean 14 5 5 4 0 0
3 16/08/14 QPR Hull 0 1 A 0 0 C Pawson 19 11 6 4 0 0
4 16/08/14 Stoke Aston Villa 0 1 A 0 0 A Taylor 12 7 2 2 0 0
\n", "
" ], "text/plain": [ " Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG Referee \\\n", "0 16/08/14 Arsenal Crystal Palace 2 1 H 1 1 J Moss \n", "1 16/08/14 Leicester Everton 2 2 D 1 2 M Jones \n", "2 16/08/14 Man United Swansea 1 2 A 0 1 M Dean \n", "3 16/08/14 QPR Hull 0 1 A 0 0 C Pawson \n", "4 16/08/14 Stoke Aston Villa 0 1 A 0 0 A Taylor \n", "\n", " HS AS HST AST HR AR \n", "0 14 4 6 2 0 1 \n", "1 11 13 3 3 0 0 \n", "2 14 5 5 4 0 0 \n", "3 19 11 6 4 0 0 \n", "4 12 7 2 2 0 0 " ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "print 'Shape of data set is: {}'.format(df.shape)\n", "print 'Column names are: {}'.format(list(df.columns))\n", "df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As you can see the dataframe now has 337 rows and 15 columns. Let's describe the data. " ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
FTHGFTAGHTHGHTAGHSASHSTASTHRAR
count 337.000000 337.000000 337.000000 337.000000 337.000000 337.000000 337.000000 337.000000 337.000000 337.000000
mean 1.445104 1.118694 0.700297 0.501484 14.442136 11.272997 4.605341 3.682493 0.080119 0.106825
std 1.199249 1.076331 0.795691 0.711825 5.441004 4.578965 2.436400 1.999276 0.271881 0.309350
min 0.000000 0.000000 0.000000 0.000000 3.000000 1.000000 0.000000 0.000000 0.000000 0.000000
25% 1.000000 0.000000 0.000000 0.000000 11.000000 8.000000 3.000000 2.000000 0.000000 0.000000
50% 1.000000 1.000000 1.000000 0.000000 14.000000 11.000000 4.000000 4.000000 0.000000 0.000000
75% 2.000000 2.000000 1.000000 1.000000 18.000000 14.000000 6.000000 5.000000 0.000000 0.000000
max 8.000000 6.000000 3.000000 4.000000 43.000000 27.000000 16.000000 11.000000 1.000000 1.000000
\n", "
" ], "text/plain": [ " FTHG FTAG HTHG HTAG HS AS \\\n", "count 337.000000 337.000000 337.000000 337.000000 337.000000 337.000000 \n", "mean 1.445104 1.118694 0.700297 0.501484 14.442136 11.272997 \n", "std 1.199249 1.076331 0.795691 0.711825 5.441004 4.578965 \n", "min 0.000000 0.000000 0.000000 0.000000 3.000000 1.000000 \n", "25% 1.000000 0.000000 0.000000 0.000000 11.000000 8.000000 \n", "50% 1.000000 1.000000 1.000000 0.000000 14.000000 11.000000 \n", "75% 2.000000 2.000000 1.000000 1.000000 18.000000 14.000000 \n", "max 8.000000 6.000000 3.000000 4.000000 43.000000 27.000000 \n", "\n", " HST AST HR AR \n", "count 337.000000 337.000000 337.000000 337.000000 \n", "mean 4.605341 3.682493 0.080119 0.106825 \n", "std 2.436400 1.999276 0.271881 0.309350 \n", "min 0.000000 0.000000 0.000000 0.000000 \n", "25% 3.000000 2.000000 0.000000 0.000000 \n", "50% 4.000000 4.000000 0.000000 0.000000 \n", "75% 6.000000 5.000000 0.000000 0.000000 \n", "max 16.000000 11.000000 1.000000 1.000000 " ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.describe()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This provides a useful data summary of the numeric data." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Data Analysis" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's now analyse the data... Think of a question and produce the answer!" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which HomeTeam scored the most goals in a match?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can see from the describe() above that the max FTHG is 8. Let's find out which team did this." ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGRefereeHSASHSTASTHRAR
76 18/10/14 Southampton Sunderland 8 0 H 3 0 A Marriner 20 12 11 1 0 0
\n", "
" ], "text/plain": [ " Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG Referee \\\n", "76 18/10/14 Southampton Sunderland 8 0 H 3 0 A Marriner \n", "\n", " HS AS HST AST HR AR \n", "76 20 12 11 1 0 0 " ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df['FTHG'] == df['FTHG'].max()]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Ok, I remember that. One of the very poor Sunderland results that eventually led to Gus Poyet's demise." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which AwayTeam scored the most goals in a match?" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGRefereeHSASHSTASTHRAR
21 30/08/14 Everton Chelsea 3 6 A 1 2 J Moss 17 12 7 8 0 0
\n", "
" ], "text/plain": [ " Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG Referee HS AS \\\n", "21 30/08/14 Everton Chelsea 3 6 A 1 2 J Moss 17 12 \n", "\n", " HST AST HR AR \n", "21 7 8 0 0 " ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df['FTAG'] == df['FTAG'].max()]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "And the best away performance belongs to Chelsea. Part of their impressive start to the season." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which HomeTeam had the most shots in a match?" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGRefereeHSASHSTASTHRAR
290 21/03/15 Man City West Brom 3 0 H 2 0 N Swarbrick 43 3 16 0 0 1
\n", "
" ], "text/plain": [ " Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG Referee \\\n", "290 21/03/15 Man City West Brom 3 0 H 2 0 N Swarbrick \n", "\n", " HS AS HST AST HR AR \n", "290 43 3 16 0 0 1 " ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df['HS'] == df['HS'].max()]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Man City had an incredible 43 shots, with 16 on target. That's approximately 1 shot every 2 minutes. West Brom had only 3 shots with none on target! That's very low even for a 'well disciplined' Tony Pulis team. However, you can also see that West Brom were given a red card. The [match report](http://www.whoscored.com/Matches/829796/MatchReport) says Gareth McAuley was sent off in the 2nd minute. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which HomeTeam had the most shots in a match with no red cards?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's exclude matches with red cards and then find who had the most shots at home." ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "(278, 15)" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_no_reds = df[(df['HR'] == 0) & (df['AR'] == 0)]\n", "df_no_reds.shape" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGRefereeHSASHSTASTHRAR
90 01/11/14 Arsenal Burnley 3 0 H 0 0 C Pawson 32 6 13 2 0 0
123 29/11/14 QPR Leicester 3 2 H 2 1 R East 32 19 6 5 0 0
193 01/01/15 Man City Sunderland 3 2 H 0 0 R East 32 4 12 4 0 0
\n", "
" ], "text/plain": [ " Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG Referee HS \\\n", "90 01/11/14 Arsenal Burnley 3 0 H 0 0 C Pawson 32 \n", "123 29/11/14 QPR Leicester 3 2 H 2 1 R East 32 \n", "193 01/01/15 Man City Sunderland 3 2 H 0 0 R East 32 \n", "\n", " AS HST AST HR AR \n", "90 6 13 2 0 0 \n", "123 19 6 5 0 0 \n", "193 4 12 4 0 0 " ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_no_reds[df_no_reds['HS'] == df_no_reds['HS'].max()]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There are 3 home teams that share the top spot with 32 shots. And Sunderland are on the receiving end again!" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which AwayTeam had the most shots in a match?" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGRefereeHSASHSTASTHRAR
60 04/10/14 Aston Villa Man City 0 2 A 0 0 C Foy 6 27 1 7 0 0
\n", "
" ], "text/plain": [ " Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG Referee HS \\\n", "60 04/10/14 Aston Villa Man City 0 2 A 0 0 C Foy 6 \n", "\n", " AS HST AST HR AR \n", "60 27 1 7 0 0 " ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df['AS'] == df['AS'].max()]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Man City take top spot again with 27 shots, 7 on target." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which team has scored the most goals and which has had most shots (home and away)?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's build a new dataframe giving the total goals and shots for each team." ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "collapsed": false }, "outputs": [], "source": [ "team_goals = defaultdict(int)\n", "team_shots = defaultdict(int)\n", "team_goal_diff = defaultdict(int)\n", "team_goals_shots = {}\n", "for HomeTeam, AwayTeam, FTHG, FTAG, HS, AS in df[['HomeTeam', 'AwayTeam', 'FTHG', 'FTAG', 'HS', 'AS']].values:\n", " team_goals[HomeTeam] += FTHG\n", " team_goals[AwayTeam] += FTAG\n", " team_shots[HomeTeam] += HS\n", " team_shots[AwayTeam] += AS\n", " team_goal_diff[HomeTeam] += (FTHG-FTAG)\n", " team_goal_diff[AwayTeam] += (FTAG-FTHG)\n", "df_team_goals = pd.DataFrame(team_goals.items(), columns=['Team', 'Goals'])\n", "df_team_shots = pd.DataFrame(team_shots.items(), columns=['Team', 'Shots'])\n", "df_team_goal_diff = pd.DataFrame(team_goal_diff.items(), columns=['Team', 'GoalDiff'])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's check the high goal scorers." ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TeamGoals
0 Man City 70
1 Chelsea 65
2 Arsenal 63
3 Man United 59
4 Tottenham 55
\n", "
" ], "text/plain": [ " Team Goals\n", "0 Man City 70\n", "1 Chelsea 65\n", "2 Arsenal 63\n", "3 Man United 59\n", "4 Tottenham 55" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_team_goals.sort(columns='Goals', ascending=False, inplace=True)\n", "df_team_goals.reset_index(inplace=True, drop=True)\n", "df_team_goals.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Man City have scored the most goals. This metric predicts the *current* top 4 teams, though the order isn't correct." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now let's find the low goal scorers." ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TeamGoals
15 West Brom 32
16 Hull 31
17 Aston Villa 26
18 Sunderland 26
19 Burnley 26
\n", "
" ], "text/plain": [ " Team Goals\n", "15 West Brom 32\n", "16 Hull 31\n", "17 Aston Villa 26\n", "18 Sunderland 26\n", "19 Burnley 26" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_team_goals.tail()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Aston Villa's recent revival under Tim Sherwood has improved their position, now they share the bottom with Burnley and Sunderland." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now let's check the top shots." ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TeamShots
0 Man City 593
1 Arsenal 510
2 Liverpool 507
3 Chelsea 491
4 QPR 488
\n", "
" ], "text/plain": [ " Team Shots\n", "0 Man City 593\n", "1 Arsenal 510\n", "2 Liverpool 507\n", "3 Chelsea 491\n", "4 QPR 488" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_team_shots.sort(columns='Shots', ascending=False, inplace=True)\n", "df_team_shots.reset_index(inplace=True, drop=True)\n", "df_team_shots.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Man City have also had the most shots. Liverpool would be in the top 3 if their points tally was more closely related to their shots!" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "And now the least shots." ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TeamShots
15 Swansea 380
16 West Brom 374
17 Aston Villa 363
18 Hull 363
19 Sunderland 345
\n", "
" ], "text/plain": [ " Team Shots\n", "15 Swansea 380\n", "16 West Brom 374\n", "17 Aston Villa 363\n", "18 Hull 363\n", "19 Sunderland 345" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_team_shots.tail()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Sunderland take bottom bottom spot." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now let's check shots per goal." ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TeamGoalsShotsShots_per_Goal
0 Chelsea 65 491 7.6
1 Man United 59 451 7.6
2 Arsenal 63 510 8.1
3 Man City 70 593 8.5
4 Tottenham 55 468 8.5
5 Swansea 41 380 9.3
6 Crystal Palace 42 389 9.3
7 Everton 44 423 9.6
8 Southampton 47 454 9.7
9 West Ham 42 434 10.3
10 Liverpool 47 507 10.8
11 Newcastle 36 401 11.1
12 Leicester 35 387 11.1
13 Stoke 39 449 11.5
14 West Brom 32 374 11.7
15 Hull 31 363 11.7
16 QPR 38 488 12.8
17 Sunderland 26 345 13.3
18 Aston Villa 26 363 14.0
19 Burnley 26 396 15.2
\n", "
" ], "text/plain": [ " Team Goals Shots Shots_per_Goal\n", "0 Chelsea 65 491 7.6\n", "1 Man United 59 451 7.6\n", "2 Arsenal 63 510 8.1\n", "3 Man City 70 593 8.5\n", "4 Tottenham 55 468 8.5\n", "5 Swansea 41 380 9.3\n", "6 Crystal Palace 42 389 9.3\n", "7 Everton 44 423 9.6\n", "8 Southampton 47 454 9.7\n", "9 West Ham 42 434 10.3\n", "10 Liverpool 47 507 10.8\n", "11 Newcastle 36 401 11.1\n", "12 Leicester 35 387 11.1\n", "13 Stoke 39 449 11.5\n", "14 West Brom 32 374 11.7\n", "15 Hull 31 363 11.7\n", "16 QPR 38 488 12.8\n", "17 Sunderland 26 345 13.3\n", "18 Aston Villa 26 363 14.0\n", "19 Burnley 26 396 15.2" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_team_goals_shots = df_team_goals.copy()\n", "df_team_goals_shots = df_team_goals_shots.merge(df_team_shots)\n", "df_team_goals_shots['Shots_per_Goal'] = (df_team_goals_shots['Shots']/df_team_goals_shots['Goals']).round(1)\n", "df_team_goals_shots.sort(columns='Shots_per_Goal', inplace=True)\n", "df_team_goals_shots.reset_index(inplace=True, drop=True)\n", "df_team_goals_shots" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Chelsea and Man United are the most clinical, scoring after every 7.6 shots. Every team should have a Diego Costa. Burnley are the least clinical, needing 15.2 shots per goal. This data also highlights Liverpool's problem - they languish in mid table because of the trouble they are having scoring this season. They've had lot's of practice but not much success." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Of course success is based on a balance of attack and defence, unless you have Luis Suarez in your team ;-). Let's check the goal difference." ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TeamGoalDiff
0 Chelsea 39
1 Man City 34
2 Arsenal 31
3 Man United 25
4 Southampton 21
5 Liverpool 11
6 Tottenham 6
7 Everton 1
8 West Ham 0
9 Stoke -3
10 Swansea -3
11 Crystal Palace -5
12 Hull-14
13 West Brom-14
14 Leicester-16
15 Newcastle-21
16 QPR-21
17 Aston Villa-22
18 Sunderland-23
19 Burnley-26
\n", "
" ], "text/plain": [ " Team GoalDiff\n", "0 Chelsea 39\n", "1 Man City 34\n", "2 Arsenal 31\n", "3 Man United 25\n", "4 Southampton 21\n", "5 Liverpool 11\n", "6 Tottenham 6\n", "7 Everton 1\n", "8 West Ham 0\n", "9 Stoke -3\n", "10 Swansea -3\n", "11 Crystal Palace -5\n", "12 Hull -14\n", "13 West Brom -14\n", "14 Leicester -16\n", "15 Newcastle -21\n", "16 QPR -21\n", "17 Aston Villa -22\n", "18 Sunderland -23\n", "19 Burnley -26" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_team_goal_diff.sort(columns='GoalDiff', ascending=False, inplace=True)\n", "df_team_goal_diff.reset_index(inplace=True, drop=True)\n", "df_team_goal_diff" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As expected the goal difference is a much better predictor of success. Chelsea have the best goal difference. This metric shows the *current* top 4 in order. And 3 of the *current* bottom 4. Who said Chelsea are boring? That would be the Arsenal fans." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which referees brandish the most red cards?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This time we'll use the pandas groupby function to slice the data." ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
HRAR
Referee
A Marriner 0 1
A Taylor 1 4
C Foy 1 4
C Pawson 3 3
G Scott 0 0
\n", "
" ], "text/plain": [ " HR AR\n", "Referee \n", "A Marriner 0 1\n", "A Taylor 1 4\n", "C Foy 1 4\n", "C Pawson 3 3\n", "G Scott 0 0" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_refs_reds = df[['Referee', 'HR', 'AR']].groupby(df['Referee']).sum()\n", "df_refs_reds.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's sum the reds and sort." ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
RefereeHRARTotalReds
0 J Moss 3 4 7
1 M Atkinson 5 1 6
2 C Pawson 3 3 6
3 C Foy 1 4 5
4 L Mason 3 2 5
\n", "
" ], "text/plain": [ " Referee HR AR TotalReds\n", "0 J Moss 3 4 7\n", "1 M Atkinson 5 1 6\n", "2 C Pawson 3 3 6\n", "3 C Foy 1 4 5\n", "4 L Mason 3 2 5" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_refs_reds['TotalReds'] = df_refs_reds['HR'] + df_refs_reds['AR']\n", "df_refs_reds.sort(columns='TotalReds', ascending=False, inplace=True)\n", "df_refs_reds.reset_index(inplace=True)\n", "df_refs_reds.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "J Moss is the top brandisher of reds with 7. M Atkinson stands out for sending off mostly home players, with 5 home and 1 away red. Let's see which teams have been subject to M Atkinson's card waving." ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
HomeTeamAwayTeamRefereeHRAR
143 Newcastle Chelsea M Atkinson 1 0
170 Arsenal QPR M Atkinson 1 0
180 Aston Villa Sunderland M Atkinson 1 0
251 Chelsea Burnley M Atkinson 1 0
286 Everton Newcastle M Atkinson 0 1
297 Liverpool Man United M Atkinson 1 0
\n", "
" ], "text/plain": [ " HomeTeam AwayTeam Referee HR AR\n", "143 Newcastle Chelsea M Atkinson 1 0\n", "170 Arsenal QPR M Atkinson 1 0\n", "180 Aston Villa Sunderland M Atkinson 1 0\n", "251 Chelsea Burnley M Atkinson 1 0\n", "286 Everton Newcastle M Atkinson 0 1\n", "297 Liverpool Man United M Atkinson 1 0" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[['HomeTeam', 'AwayTeam', 'Referee', 'HR', 'AR']][(df['Referee'] == 'M Atkinson') & \n", " ((df['HR'] > 0) | (df['AR'] > 0))]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## What is the average number of goals per game?" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "average number of goals per game is 2.6\n" ] } ], "source": [ "total_games = len(df)\n", "total_goals = df['FTHG'].sum() + df['FTAG'].sum()\n", "goals_per_game = total_goals/total_games\n", "print 'average number of goals per game is {:.2}'.format(goals_per_game)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## What is the average number of goals in first half?" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "average number of goals per fist half is 1.2\n" ] } ], "source": [ "total_fh_goals = df['HTHG'].sum() + df['HTAG'].sum()\n", "goals_per_first_half = total_fh_goals/total_games\n", "print 'average number of goals per fist half is {:.2}'.format(goals_per_first_half)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## What is the average number of goals in second half?" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "average number of goals per second half is 1.4\n" ] } ], "source": [ "total_sh_goals = total_goals - total_fh_goals\n", "goals_per_second_half = total_sh_goals/total_games\n", "print 'average number of goals per second half is {:.2}'.format(goals_per_second_half)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So you are more likely to see a goal in the second half. I guess this is down to teams tiring and leaving themselves more open." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## What does the premiership table look like?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's create a standard table from the raw match data. For comparison see this [premiership table](http://www.premierleague.com/en-gb/matchday/league-table.html?season=2014-2015&month=APRIL&timelineView=date&toDate=1430089199999&tableView=CURRENT_STANDINGS) (as of 27th April 2015)." ] }, { "cell_type": "code", "execution_count": 30, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
TeamPWDLGFGAGDPTS
Position
1 Chelsea 33 23 8 2 65 26 39 77
2 Man City 34 20 7 7 70 36 34 67
3 Arsenal 33 20 7 6 63 32 31 67
4 Man United 34 19 8 7 59 34 25 65
5 Liverpool 33 17 7 9 47 36 11 58
6 Tottenham 34 17 7 10 55 49 6 58
7 Southampton 34 17 6 11 47 26 21 57
8 Swansea 34 14 8 12 41 44 -3 50
9 Stoke 34 13 8 13 39 42 -3 47
10 Everton 34 11 11 12 44 43 1 44
11 West Ham 34 11 11 12 42 42 0 44
12 Crystal Palace 34 11 9 14 42 47 -5 42
13 West Brom 34 9 10 15 32 46-14 37
14 Newcastle 34 9 8 17 36 57-21 35
15 Aston Villa 34 8 8 18 26 48-22 32
16 Hull 33 7 10 16 31 45-14 31
17 Leicester 33 8 7 18 35 51-16 31
18 Sunderland 33 5 15 13 26 49-23 30
19 QPR 34 7 6 21 38 59-21 27
20 Burnley 34 5 11 18 26 52-26 26
\n", "
" ], "text/plain": [ " Team P W D L GF GA GD PTS\n", "Position \n", "1 Chelsea 33 23 8 2 65 26 39 77\n", "2 Man City 34 20 7 7 70 36 34 67\n", "3 Arsenal 33 20 7 6 63 32 31 67\n", "4 Man United 34 19 8 7 59 34 25 65\n", "5 Liverpool 33 17 7 9 47 36 11 58\n", "6 Tottenham 34 17 7 10 55 49 6 58\n", "7 Southampton 34 17 6 11 47 26 21 57\n", "8 Swansea 34 14 8 12 41 44 -3 50\n", "9 Stoke 34 13 8 13 39 42 -3 47\n", "10 Everton 34 11 11 12 44 43 1 44\n", "11 West Ham 34 11 11 12 42 42 0 44\n", "12 Crystal Palace 34 11 9 14 42 47 -5 42\n", "13 West Brom 34 9 10 15 32 46 -14 37\n", "14 Newcastle 34 9 8 17 36 57 -21 35\n", "15 Aston Villa 34 8 8 18 26 48 -22 32\n", "16 Hull 33 7 10 16 31 45 -14 31\n", "17 Leicester 33 8 7 18 35 51 -16 31\n", "18 Sunderland 33 5 15 13 26 49 -23 30\n", "19 QPR 34 7 6 21 38 59 -21 27\n", "20 Burnley 34 5 11 18 26 52 -26 26" ] }, "execution_count": 30, "metadata": {}, "output_type": "execute_result" } ], "source": [ "results = [] # create results list\n", "\n", "for team in df['HomeTeam'].unique():\n", " home_results = df[df['HomeTeam'] == team]\n", " home_played = len(home_results.index)\n", " home_win = home_results.FTR[home_results.FTR == 'H'].count()\n", " home_draw = home_results.FTR[home_results.FTR == 'D'].count()\n", " home_lose = home_results.FTR[home_results.FTR == 'A'].count()\n", " home_goals_for = home_results.FTHG.sum()\n", " home_goals_against = home_results.FTAG.sum()\n", "\n", " away_results = df[df['AwayTeam'] == team]\n", " away_played = len(away_results.index)\n", " away_win = away_results.FTR[away_results.FTR == 'A'].count()\n", " away_draw = away_results.FTR[away_results.FTR == 'D'].count()\n", " away_lose = away_results.FTR[away_results.FTR == 'H'].count()\n", " away_goals_for = away_results.FTAG.sum()\n", " away_goals_against = away_results.FTHG.sum()\n", " \n", " result_d = {} # create dictionary to hold team results\n", " result_d['Team'] = team\n", " result_d['P'] = home_played + away_played\n", " result_d['W'] = home_win + away_win\n", " result_d['D'] = home_draw + away_draw\n", " result_d['L'] = home_lose + away_lose\n", " result_d['GF'] = home_goals_for + away_goals_for\n", " result_d['GA'] = home_goals_against + away_goals_against\n", " result_d['GD'] = result_d['GF'] - result_d['GA']\n", " result_d['PTS'] = result_d['W']*3 + result_d['D']\n", " \n", " results.append(result_d) # append team result dictionary to list of results\n", " \n", "# create DataFrame from results and sort by points (and then goal difference)\n", "PLtable = pd.DataFrame(results, columns=['Team', 'P', 'W', 'D', 'L', 'GF', 'GA', 'GD', 'PTS'])\n", "PLtable.sort(columns=['PTS', 'GD'], ascending=False, inplace=True)\n", "PLtable['Position'] = range(1, len(PLtable)+1) # add new column for position, with highest points first\n", "PLtable.set_index(['Position'], inplace=True, drop=True) \n", "PLtable" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which team has drawn most and least games?" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "team(s) with most draws (15): Sunderland\n", "team(s) with least draws (6): Southampton, QPR\n" ] } ], "source": [ "max_draws = PLtable['D'].max()\n", "teams_most_draws = PLtable['Team'][PLtable['D'] == max_draws].values\n", "min_draws = PLtable['D'].min()\n", "teams_least_draws = PLtable['Team'][PLtable['D'] == min_draws].values\n", "print 'team(s) with most draws ({}): {}'.format(max_draws, ', '.join(teams_most_draws))\n", "print 'team(s) with least draws ({}): {}'.format(min_draws, ', '.join(teams_least_draws))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "##What is the most and least common score?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's add a new column with the score and analyse." ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateHomeTeamAwayTeamFTHGFTAGFTRHTHGHTAGRefereeHSASHSTASTHRARScore
0 16/08/14 Arsenal Crystal Palace 2 1 H 1 1 J Moss 14 4 6 2 0 1 2:1
1 16/08/14 Leicester Everton 2 2 D 1 2 M Jones 11 13 3 3 0 0 2:2
2 16/08/14 Man United Swansea 1 2 A 0 1 M Dean 14 5 5 4 0 0 1:2
3 16/08/14 QPR Hull 0 1 A 0 0 C Pawson 19 11 6 4 0 0 0:1
4 16/08/14 Stoke Aston Villa 0 1 A 0 0 A Taylor 12 7 2 2 0 0 0:1
\n", "
" ], "text/plain": [ " Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG Referee \\\n", "0 16/08/14 Arsenal Crystal Palace 2 1 H 1 1 J Moss \n", "1 16/08/14 Leicester Everton 2 2 D 1 2 M Jones \n", "2 16/08/14 Man United Swansea 1 2 A 0 1 M Dean \n", "3 16/08/14 QPR Hull 0 1 A 0 0 C Pawson \n", "4 16/08/14 Stoke Aston Villa 0 1 A 0 0 A Taylor \n", "\n", " HS AS HST AST HR AR Score \n", "0 14 4 6 2 0 1 2:1 \n", "1 11 13 3 3 0 0 2:2 \n", "2 14 5 5 4 0 0 1:2 \n", "3 19 11 6 4 0 0 0:1 \n", "4 12 7 2 2 0 0 0:1 " ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df['Score'] = df['FTHG'].map(str) + ':' + df['FTAG'].map(str)\n", "df.head()" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "Let's count each score as a percentage of all the scores." ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "most common scores:\n", " Count Count%\n", "Score \n", "1:1 34 10.1\n", "0:1 34 10.1\n", "1:0 32 9.5\n", "2:0 32 9.5\n", "2:1 32 9.5\n", "\n", "least common scores:\n", " Count Count%\n", "Score \n", "0:5 1 0.3\n", "0:4 1 0.3\n", "3:6 1 0.3\n", "4:3 1 0.3\n", "8:0 1 0.3\n" ] } ], "source": [ "Score_df = pd.DataFrame(df['Score'].groupby(df['Score']).count())\n", "Score_df.columns = ['Count']\n", "Score_df.sort(['Count'], inplace=True, ascending=False)\n", "tot_scores = Score_df['Count'].sum()\n", "Score_df['Count%'] = (100*(Score_df['Count']/tot_scores)).round(1)\n", "print 'most common scores:'\n", "print Score_df.head()\n", "print '\\nleast common scores:'\n", "print Score_df.tail()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So 1:1 and 0:1 are most common at 10.1%. And 8:0 (and several others) are least common at 0.3%." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "##What does the score matrix look like?" ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
AwayGoals0123456
HomeGoals
0 8.0% 10.1% 3.9% 1.5% 0.3% 0.3% -
1 9.5% 10.1% 7.1% 3.6% 1.2% - -
2 9.5% 9.5% 6.5% 1.8% 0.3% - -
3 4.7% 3.9% 2.4% 0.9% - - 0.3%
4 1.2% 1.2% 0.6% 0.3% - - -
5 0.6% - - 0.6% - - -
6 - - - - - - -
7 - - - - - - -
8 0.3% - - - - - -
\n", "
" ], "text/plain": [ "AwayGoals 0 1 2 3 4 5 6\n", "HomeGoals \n", "0 8.0% 10.1% 3.9% 1.5% 0.3% 0.3% -\n", "1 9.5% 10.1% 7.1% 3.6% 1.2% - -\n", "2 9.5% 9.5% 6.5% 1.8% 0.3% - -\n", "3 4.7% 3.9% 2.4% 0.9% - - 0.3%\n", "4 1.2% 1.2% 0.6% 0.3% - - -\n", "5 0.6% - - 0.6% - - -\n", "6 - - - - - - -\n", "7 - - - - - - -\n", "8 0.3% - - - - - -" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# create dataframe of scores, with home goals as rows and away goals as columns\n", "df_scores = pd.DataFrame()\n", "df_scores.columns.name='AwayGoals'\n", "df_scores.index.name='HomeGoals'\n", "\n", "# calculate max home goals and max away_goals\n", "max_hg = df['FTHG'].max()\n", "max_ag = df['FTAG'].max()\n", "\n", "# create empty matrix (ensures all possible scores, up to max, are covered)\n", "for hg in range(max_hg + 1):\n", " for ag in range(max_ag + 1):\n", " df_scores.set_value(hg, ag, '-')\n", "\n", "# create the sorted, unique ndarray of scores\n", "scores = df['Score'].unique() \n", "scores.sort() \n", "\n", "# for each score populate the matrix with its percentage occurrence\n", "for score in scores:\n", " score_split = score.split(':')\n", " hg, ag = int(score_split[0]), int(score_split[1])\n", " df_scores.set_value(hg, ag, \n", " '{:.1%}'.format((df['Score'][df['Score'] == score].count())/total_games))\n", " \n", "df_scores" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "##What is the percentage of full-time home wins, home draws and away wins?\n", "\n", "Also known as Spiegelhager's 48/26/26 law. That is Spiegelhager predicts - over a full season - an outcome of 48 home wins, 26 away wins and 26 draws." ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Total games=337, home wins=149, away wins=102, draws=86\n", "Total games=337, home wins%=44.2%, away wins%=30.3%, draws%=25.5%\n" ] } ], "source": [ "home_wins = df['FTR'][df['FTR'] == 'H'].count()\n", "away_wins = df['FTR'][df['FTR'] == 'A'].count()\n", "draws = df['FTR'][df['FTR'] == 'D'].count()\n", "print 'Total games={}, home wins={}, away wins={}, draws={}'.format(total_games, \n", " home_wins, away_wins, draws)\n", "home_wins_pc = home_wins/total_games\n", "away_wins_pc = away_wins/total_games\n", "draws_pc = draws/total_games\n", "print 'Total games={}, home wins%={:.1%}, away wins%={:.1%}, draws%={:.1%}'.format(total_games, \n", " home_wins_pc, away_wins_pc, draws_pc)" ] }, { "cell_type": "code", "execution_count": 36, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Full Time ResultHAD
2015-16 % 44.2 30.3 25.5
\n", "
" ], "text/plain": [ "Full Time Result H A D\n", "2015-16 % 44.2 30.3 25.5" ] }, "execution_count": 36, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# calculate Spiegelhager result series using groupby\n", "spieg = (100*(df['FTR'].groupby(df['FTR']).count())/total_games).round(1)\n", "spieg.sort(ascending=False)\n", "\n", "# display result as dataframe\n", "df_spieg = pd.DataFrame({'2015-16 %': spieg})\n", "df_spieg.index.name='Full Time Result'\n", "df_spieg.transpose()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "##What does the graph comparing the top team performance look like?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Build a dictionary of team points.\n", "The dictionary is keyed on team name and has a value of a list containings the cumulative points from match 1 to the most recent match.\n", "\n", "For each team, the length of the list is therefore the number of games played and the final list entry is the total points accumulated." ] }, { "cell_type": "code", "execution_count": 37, "metadata": { "collapsed": false }, "outputs": [], "source": [ "team_points = defaultdict(list)\n", "\n", "# create mapping dictionary, mapping result to points, for home and away full time results\n", "HFTR_points = {'H': 3, 'A': 0, 'D': 1}\n", "AFTR_points = {'H': 0, 'A': 3, 'D': 1}\n", "\n", "for ht, at, ftr in df[['HomeTeam', 'AwayTeam', 'FTR']].values:\n", " if team_points[ht]:\n", " team_points[ht].append(team_points[ht][-1] + HFTR_points[ftr]) \n", " else:\n", " team_points[ht].append(HFTR_points[ftr])\n", " \n", " if team_points[at]:\n", " team_points[at].append(team_points[at][-1] + AFTR_points[ftr]) \n", " else:\n", " team_points[at].append(AFTR_points[ftr])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Print sub-set of the team_points data to check the data." ] }, { "cell_type": "code", "execution_count": 38, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Team Played Cum Points (last 3)\n", "--------------- ------- --------------------\n", "Chelsea 33 [73, 76, 77]\n", "Man City 34 [61, 64, 67]\n", "Arsenal 33 [63, 66, 67]\n", "Man United 34 [65, 65, 65]\n", "Liverpool 33 [54, 57, 58]\n", "Tottenham 34 [54, 57, 58]\n", "Southampton 34 [56, 56, 57]\n", "Swansea 34 [47, 47, 50]\n", "Stoke 34 [43, 46, 47]\n", "West Ham 34 [43, 43, 44]\n", "Everton 34 [38, 41, 44]\n", "Crystal Palace 34 [42, 42, 42]\n", "West Brom 34 [33, 36, 37]\n", "Newcastle 34 [35, 35, 35]\n", "Aston Villa 34 [29, 32, 32]\n", "Leicester 33 [25, 28, 31]\n", "Hull 33 [28, 28, 31]\n", "Sunderland 33 [29, 29, 30]\n", "QPR 34 [26, 26, 27]\n", "Burnley 34 [26, 26, 26]\n" ] } ], "source": [ "print \"{0:16} {1:<8} {2:<8}\".format('Team', 'Played', 'Cum Points (last 3)')\n", "print \"{0:16} {1:<8} {2:<8}\".format('---------------', '-------', '--------------------')\n", "for team, points_list in sorted(team_points.items(), key=lambda (k, v): v[-1], reverse=True):\n", " print \"{0:16} {1:<8} {2:<8}\".format(team, len(points_list), points_list[-3:])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Extract asummary of key data for the top 5 teams, using the (ordered) premier league table generated earlier." ] }, { "cell_type": "code", "execution_count": 39, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[['Chelsea' 33L 77L]\n", " ['Man City' 34L 67L]\n", " ['Arsenal' 33L 67L]\n", " ['Man United' 34L 65L]\n", " ['Liverpool' 33L 58L]]\n" ] } ], "source": [ "TOP_TEAM_COUNT = 5\n", "top_team_summary = PLtable[['Team', 'P', 'PTS']][0:TOP_TEAM_COUNT].values \n", "print top_team_summary" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now, plot the graphs!" ] }, { "cell_type": "code", "execution_count": 49, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsoAAAIMCAYAAADsJ4rpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3XecXFX5x/HPNyEhEHov0gVBVEB+ID1Dka7YQOlRBGyA\nXRBls6A0FaygAtJEASkCgooCA6goIAgoRVroJJQAoZhA8vz+OHfYm9mZ3dndmbmzu9/36zWvnbn3\n3HOfe6fsM2fOPUcRgZmZmZmZzWtM0QGYmZmZmXUiJ8pmZmZmZjU4UTYzMzMzq8GJspmZmZlZDU6U\nzczMzMxqcKJsZmZmZlaDE2Xrk6Q1Jf1Y0j2SXpY0U9K9kn4u6T1Fx9cISZMlzc1u+w+yjs9LmiLp\nsFbUP1SSpuZiqNxmSbpf0vclLTnIeku5+rqGEN+U7FbI+RkqSeWqc/vhGmXuqiozaZD7qvtaG0Ad\nldfDw0Oo46zcsbwsaencugm5dYPex1BJWjUXx5lFxZFX431YfVu5DTFMbiCOQj6z6nxWVW6/7mfb\nswZwXC0/zzY6zFd0ANa5JH0cOBUYny2qDLq9JrAWsDTwwQJCG6yg5xgG6vPAysAjwA/6qLcTBiav\nxDAfsAZwKLCrpP+LiBcGWddQzh3AUdnf64Gzh1BPp/gUcHHlgaQtgXVz61v1WmtUs1+PCwKHA1/q\nY19F65Q48oqKqfr5V5311ffbrXrf/cVS/b5S1TqzpnOLstUkaRvgdFKSHMAxwErA/MDbgK8DzxcW\n4OBU/7MYjF4fxhFxdkSMzW7nNGEfQxHAahExFngH8EC2fHXgkAFXFnF9RIzJju3oJsU3Emwjac3c\n409XrW/Ja20AmrH/ap+StFwL6h1WJC3QYNHIfS7kb4+2OpbsM6nyvh0LdOdWT6mKp6jPrOo4xkbE\nXn1tEBEfz5cnffGu2LpV59lGNyfKVs9x9Pyz/UFEdEXEkxHxRkQ8EBHHAwdVCud+7rouX0mt5dnP\nypXlB0v6oaTns9upkhaQtJOk2yS9KukOSbtW1Vvzp+WB/OQsqVvSTZKmSZot6ZVsX0dIGpeVKUma\nS2rhA8j/1PtwVqZX1wtJt2ePZ0gan9vnYpL+l627Prd8HUnnSnoii2W6pN9Iemd/x1Hr0AAi4h7S\nLwIVG+X2t5SkkyU9oNRFY6akv0maXHWOana90LxdEd4m6fKsjqcknSZp4fy5yVWZr++6XCw/kfRQ\n9ny/JOk+Sb+StFbdg5R2y9X1tap1J+TWlbJlH5J0o6RnsmN+WtINkr44wPP7COkcH5zVuzRQ6Yox\ntU6sTXmtZeWWy567/2avpZck/VPSfnX2vYGkP2fn9pHs3Iwb4DHPARYgtSrXVeu90M/y/M/w78ie\nj9ey539PSeMlHZu9rmZIulTS8vV3r70k/Sd7fu9VjW4FkjbJ6pkm6XVJT0o6U9IqVeXyr/F1JV0t\n6RXgqoGcuH7O1/uz52ZG9rp4VNLpLYil7hcnDe6zYIqkr0p6ONvmdkk7DfTwB1h+QHWowc9USR+V\n9CdJj2XvkVlKn0WnSlqmquyQX6+SDpR0q9L/u1lZfFfXe/9aB4gI33yb5wYsA8zNbnOA5RvYplL+\n2v6WA1Nyy6fn9lNZdgPwetWy2cAauToezpY/VLW/XsuBybl69sstvyfbR/5WKfeLrEwptyx/TuZU\n9pGrf06lflLrbWXZB3L7/ESNslsAr1bVXYnlVWCLBs7/1Nz2K+eWfzkX9+XZsuVy5efS+9h/mts+\nf/xH5ZaXc8ufr/EcnpaV27+P83dtVubKOrHMAXbu45jHAk9nZf+VWy5SMjsXeCBb9p6qevP7ubmB\n85s/3iOyv8+QfmH5Wvb4NuDXuXJbteC19tbcMVcfyy9qvB5mZrfq5+fIBo75rFz5yv1XgRWACbl1\ntd5rb76++1n+cK6e6fQ+9jK9Y89/lqyaW/5EjbJzgY/nyu8BvFHndfAssFad5/yZ3DbX9nPe3oy/\nn3JHMO+x5uN+DlhnqLHU+czNv48H+1lQ63P7dWC7AXxWPQf8L3tN3QYcBoxp5HjqnJf8+63hz1Tg\np3XKzSW9b8c18fW6e41zXSl74UCO3bf23dyibLWsmrv/UkQ81cJ9zQLeDmyYW7YFcC6wOPDDbNl8\npH9yFfVaEgbSSnEEqV/pYqSEZ03gX9m6/SQtHhHliBgDVH7Gmxo9P+2t3se+zyMl9wLyPydW7s8E\nfpPdP42UeEwlnYcJ2d9nsvs/GcAxvRmHpHXIWj0z/8j+HkNPq+WZwJLA+qTkEuAgSZsOYH//IP3D\n3YT0fALsDT0/AefKlnPnb5ts2VbZ34uBRbPbu0j9YR+vt9OImEN6nQC8KztegM1J3YQgJXiQXlMi\ndWfYlPR8rwS8j5TcNipIif1jwBLAx+g5x6fW24jmvdZ+SPoiC3AJKXFeGJgEXFtjvxNJr7Mlgd1y\ny/dt4FgrArgIuIv0evw6/XcLafT9mX98Fen8nJRbtwWwK+n1VXl9luq0Ki8P7AcsQvb6yxwnaayk\nBUnP0Rjgn8A62fFsS3qvLgF8p07cU0nP34L07mZTj3Ktj5Xb7dmKVYBKV6YZpOdvMeBb2bLFge83\nMZZ6BvtZMBHYiXSuK7/mjAVOaGCfldfOYsA40nOwPnAycP4A469nIJ+p55G+SC9N6mq4HD2fG28D\nds6VHezrtdJlqfJZNzOrewLp/+3uwB8GfpjWFkVn6r513g3YmJ5vvTMa3KbXt+d6y5m3dePo3PJp\n9HzTXi1b9t5c2VNzZadS1ZpVbzn1W5S3I304TaenlSn/bX/j/vbXT/0XZMteARYifXhW9vPzrMya\nVfutd1u2n/M/tZ/tHwQWy8o+mTvGxXJ1HJorf0y2rJRbVqtFeQ7w9tzyW3LLl+nv9ZGtu4OeFsET\ns/P5bkANvO7WqRHzT+hp4XpLtuyDuXKXklradwWWavD1nT/edYFv0NMKORd4gZQ8nJ/bT76Fa8iv\nNVLXh9ezdS8CCzTwepgNLJxbXon31QaO+axcjLvkzuFrpAR9bnWc1H8v1FteiXMOsHq2bKdc2etz\nZc/LlX1PtmzVXNkbq+L/S678u5n3s6Te7dXc9uXc8k0G8PlZ/dzmb7dlZQ7KlTk5t+0YeloqZwHj\nhxJLnc/c/Pt4sJ8F51bV/2iunsX7ieUIUsK4BOk987HsNVWpe/MBHFf+vGyVLRvQZ2pW/tzsGGbX\nKPeVJr5ev5BbdjapFf29wCIDfU59a9/NLcpWy9Tc/UX66BPYJ0mNjKqS39drufuVb+Ov55bN30B9\nDY3kImlz4I/A9sBS9LQ2vlmE9G1/KH6R/Z1ASjI+SvpHGMAZ2bplqraJOrclBrDfyjazgYeAH5E+\nqCsjXlSG+Xo55h0FI3/xy9I07r7c/Vdz9xs9fwdmdSxPSmB/AdwKPCRpvb42jNQPu9JS/jFJY0mt\nMwB/jojHs3KXAqeQko/dSAn55cBTkn7UYJxv7pZ0oesb9Dwv50bEK7UKN/G1tgSp1Q7g0Yh4ra/C\nmWkRMTP3uBJjI++lvMjO4b+ybY8c4PYD+Syo9TkAPZ8Fonb81RdvPZa7vxSNvdfmV+8L5AK4va/A\n64jofbHau7N1+ffXo7kN5pKSV0jnrPp9P9hY6hnsZ0H1uX4id3+pvnYYEcdFxA0R8XxEvBIR59Pz\nyxCkX6aGouHPVEmLkr5Q7Q28hfT+qqyvqHfx5tTs70Ber6eQfuGZS/pV52TSZ8M0VV1nYZ3DibL1\nEhHTgZuzhwK+UqtclpRUVD4U8v/wq7sm1PJGnRjm9rNd5Sf+N/cnaSKwbAP7hJRMVX5GO57U6jaW\n1NpYM6QG6827mvQPRMCe2Q3gnoioJHfTcuX/VOMfa+V2T4P7DLJRLyJiQkS8NSIOi4hnc2Uq+1xI\n0mK55Svn7k9vcH9E6gKR3/+ARMTNEbEOqXVwR9IFYy8Dq9DYT7lnZn/XAL5Jzz/qX+QLRcTnSD9p\nb0z6x/h70j/Gz0oa0D/niHgauKzykL67XTTrtfY8Pe+XlWskdLW8XvV4MK/jvK7s7z511s/K3c/H\n1+9nQZ33fM3PhzpWqXqcfz0/w7zvtdP6eK/1+gISEbOqlw1RPpY3484+U1fMHr5BjZGFmhzLYD8L\n8jGLlGRWPNu7+Dxl+9PfZ39/8vH295m6NT1fBP5Muh5nLGmIxj4N5vUaEbMi4qOkz6HNgY+TvujP\nDxwraYV+j87azomy1XMkPR9Yh0rqkrS8pHFKk5B8ndQPrGJq9vddklZWuqr+mBbGV9nfcpI2kjSG\nNATS2PqbzCOfQLwCzJW0C/P2R8urfPgv1eiHWUQEPWMGv5eUoEFPYkdEPAD8N3u4naTDlEbGmCBp\nPUlHSWpWv72KK7O/Ar6b7e8dpJ8FoacfbjNV/uGvWvUPGUnfVhrVJIDrSC0uldatlejfr+lp1fl6\n9vc54Le5fUzKWmzWAu4nJal/z9XRyH6qfY+ULP8wIu7uo1xTXmtZAven7OHCwNmSVpM0UWkkh4H0\nOx6UiLiC1L+33vtsau7+LkrWBA5oZhh1lm+mNOrFQpL2IvVFh5Q43QncROoPDLB/NkrBQtn5e4+k\n70iq1y+42f5IT1K1r6TNJS1CGm+8MjnQtRExu8VxDPaz4IOS3qs0us1X6Enub4+IGTXKV7w/Gwli\nB0mLZOd/T1Lf8sr+bhzKAUXE/TT+mZpPbGcBr0pal0EMpdlXSJU7kj4s6XOkLxZ3kq7LuDNXdkWs\n8xTd98O3zr2R/rnNon4fr0tyZY/KLX+d9BP8y7ll1+XKTsktr9VfcW5uWSlXNn9V/35Vsbyc7fd/\n2eOHc2UnV++P9G2++ur4N0hJVK0+pj+qcfxn1qs/t90aVdvMApauKrMVPVdo17r1e2U78/ZRXrmf\nsssy79Xb1bdT6pz/mqNeVNVdrhUHcEWN/XRl6x7oI5aTGnytnlO13Q+q1u/Txz5eop+RXaqO6+19\nlOvVR7nJr7U1gKfqHEetUS8erorvzddJA+f0rFzdO+eW71y13/x7bSwpScm/L+cw72fBQN7zZ9aJ\np3JuV80te4Le52QOMDlXx8fo3Ue83jl88zkf4Ofm3Ea2I/1yUi+OZ4G1hxpLbvspubrz7+PBfhY8\nXqPsbGDbfuLYrY99zSU3ykaDx1XObZt/DzX0mUq6GG9ajfX31TlfQ329fqPGvvLndP7BPL++tfbm\nFmWrKyLOII0+cArpg6OS/N5L6qN5fK748aT+Vk+SktXrSQkC9O7zFVV/qSrXSNlzSa3eU0mtibeT\nLph6qpE6IuKvpJ/f783i/Q/pJ/K/1NnfFFISNL1GnPViJCIeJA13Vyl/VUQ8U1XmBtIV2eeQ+lXO\nJrWI3kEa5eDr9K/WuatdMGIa8H+kq+ofJCXvM0ktbh+PiM9U1Vvr2Aa6/BDSVeLP14j1R8A1pERn\nFun5/Dfpy1fNbj81/CK336Cq2wWpz/OZwN2k1uo3SD/HXwGUov+RXeo+x/2Va+ZrLXs9rU+ase/+\nrL6XSa1S5ao4ar0eGn6d1DqWLIarSD8X96orUjec95FG4JhJuujwJOCrdY61v/d8I58FlWV/JI0q\ncw/pdfRf0uv5rFx855NGJriYNMze66TzfAvpM+x7De6vP/2e40hj0e9GOlcvZLE8Rrp+4d0RcW+T\nYqm7/QA/C/J+Tmp1fjjb5g5gt4i4pp84biK9r28k/a+YTTr264F9I+JTAzususfV0GdqpH7ZO5He\ni6+QPoO66Pnf1uzX65+BX5HeuzNJn0NPkn4VmxTN7+JjTaDsW46ZmZlZTUoT91ybPZwSzZmp06zj\nuUXZzMzMzKwGJ8pmZmbWn6F2/zAbltz1wszMzMysBrcom5mZmZnV4ETZzMzMzKwGJ8pmVghJp0qa\nK6k8yO3XlzQlmwynz6muc9usKulESTdJelLSLEmPSPqtpI2qyk7O4uvrtn9V+SmSumrst5Tbptf6\nQRz79lVxXDDUOhvYZ81jkFSuLG+gjl0lnSPpHkkzJL0s6S5Jx0lavEb5BSV1S/pv9lw9I+lCSevU\nKPtNSX+S9EIuzt/UiaOv5/S4XLnFJL2YLd+9Vl1mNrLNV3QAZjb6ZInOgaQLg741yGrWJ43JCmk8\n1zsa2GYT4MvZ/coFGitlt10lfTgiLqtaX30hR34a3pdy9yeTJjqANEtkPc24MKR6Fr73SVo4ImY2\noe56BjO+cbXPAdtXlV03u+0haYOIeAlA0nzAH0jjHlfqXxL4CLCjpFJE3Jar50vAInVirqfemMzp\nTsQLkn4MHAEcJ+m3EVE9LbiZjWBuUTazInyR9PkzNSL+3IT61H8RICVBfwU+TJqVa3nSYP9k8bw5\nNmxEnB0RYyJibOUGLEqaKADShCW1pvdt6RXSkhYEPlC1eAIpgWzF/sZKGhcR1+fOR34M3UbPPaQJ\nUn5CmgxiAdI0049n61Zj3qmuP01PkvxLYCnSMc4BFgJ+VlX3maQvXwOZtOLj+ec3u1VP8HN69nd1\n4EMDqNvMRgAnymbWVpIWIs2gBnBB1br1JV0i6QFJL0l6XdLTki6WtGGuXJl5Z987s1Z3iBqujIgt\nI+LSiJgZEdOBQ3Pr1+wn/I8DC2f3fx4Rs7PuHHPpaU1W/qf82qdAn8u6E7wq6XZJO/az37wPAhOz\n+2fSk5hXtzLP0y1C0tsk/T7r7vCcpJ9JWjhXNt+1olvSkZKmkmZe26xJ3Uf2iYhDIuJfETE7Im4m\nzTJYkT//k7O/ARweETMi4hLSLG4AG0p6R6VwRHwhm030vwOIp98kPyIeBm7OHh44gLrNbARwomxm\n7bYFqTUR0lS2eWuTWktXIyWDY4ClScnh9ZLWzsrV+8m8z6mDI+LlGosXyN1/rN62kkTqOgBpuuFT\n6pXtJ5ZPkabRXYPUErwecJmk1fqoL6+SEAdp6uFKEjdJ0op9bHcjsAPpeBcnJX2XZcdV7TPAMaQu\nKapxHINqNW/0/EsaD7wrW/ZSRDyZK/Of3P15+pUPwvckzZY0U9JfJe1Xp1zldbpF1qJvZqOEE2Uz\na7dKchPAnVXr/klK5lYgJZGLAp/N1i0IHAwQEVuTWncrJud+Oj9ngPEcm7v/0z7K7UBPi+dFEfFU\nFsvUiBhDT0tnVHXXqLY4KfFfHDgvWzYO+Gh/gUpaFtgue3h/RNwFXFxZDezdx+Y3kr50rAPcly0r\nAbvWKLskcAjp/K8E/Lu/2AZD0vL0fPl4Bag8d0sClXP3YtVm+X7hSw9h90F6DsaSvpRtCpwl6cQa\nZSv938cBGwxhn2Y2zDhRNrN2Wy53/9mqddNIieB1pATpJeZtuV0rd38gfWN7UfIjYJ9s0aURcXIf\nmxyS/Q1Si3CvKhvc9eURcVl20dr5ueUrN7DtnvR8bl9c9Rd6jqVaAF+NiOci4r/A93Lr3luj/NUR\n8ZOIeDkinoyI5xuIbUAkrQRcCyxD6ne8f0Q80cimTdj98aTEeHFSUv55elrJv5jFlvdcbt/LNmH/\nZjZMeNQLM+skF5JGRYB5uy5UkqMFem0xCJLGAWeREk+AS+mjRVfSW4Gdsoe3RMQ/hrD7+3L3X83d\nn9DAtvl+yHdLWpd0bh4iXWz2DknrRUStEUAezd3PdzFZskbZ2xuIZdCyLjRXA28hdWPZP+t/XPEc\n8Abpf9RiVZvnR7aYPpj917hg74eSdiV9SRsDbEz9bjjNSNTNbJhwi7KZtdvTuftv/nSejaO7fa7M\nulnXhXpjJA+qn2zWx/QyUpIcpFENPhIRb/Sx2SG5+z+oU6bRePLDizV8DNmQevmf/c8F7iJ1X1k9\nt7xeq/Iqufv51uvqVn2A1xqNa6Ak/R+pG8hbSN0tdouIfMs6ETGbnm45C1f1vV63Ugy4ZRD7byTR\nrb4Ic6ncPp/GzEYNJ8pm1m6V5Eb0XLAFqQWxkjjOAWZKWob64yznE7x3SqrVH3gekhYjtWRWRpk4\nNiIOioi6CWs2Ssfk7OGTpFbvPuNRgxOgDFA+AY4at4q96E3A8ZKWlLQWaczhij81O9B6JG1D6m6x\nJOl8bRsRf6hT/KzKZqTYl5D0EWBStvyfEfHmhX2SFpe0FKlfdcX82TEvlY3LDPBZSWdL2krSxGxS\nkUOBbbP1rwN/r4ql8ny+QYtb282sszhRNrN2+ws9XQ62rCzMJsu4Jnu4Iumn76dJF59V5FsDbwdm\nZ/e/BLyeDV3WV1/fDwCb5R4fWWNmturtJ9MzJNypETGnTt2V5ErA7Vld1/URS8OyVtDKhXoBvL16\n/F/gb9n65SVtV6OaTUhjP99LT1/v6yLid82Ikca6JBxFGgMZUivt36vOff58nUrPaBN7kxLrC0n/\nt2aSXdiZczupK8aluWXvIx3zdGDzbNlYUheWclbP86TRQyqje3RVLtTMqbxOb4qIVzGzUcOJspm1\nVTZEWGX85Op+wftk654HXiB1L6iUmaflNBsybD/gbtJEFn0ODZerI19XXy2zlQT1s9ny/9F7kou8\nn5BGzXiK9NN9rXjqPe4v7i1J3SUCuDUi7qtRpjJiRNB79IvI6rgSeBmYAZwG7DaAWOqtH8jMfH2d\n9+rndw6p5f8Y4AHSeM7PAr8B3hMR1S27jdZ9BXAcaVi9aaQW5GdJswDuGhHH5yuVtDo9I7Wc1sAx\nmtkIoj5+cTQza4msv+2dpNa9HSKibT//jybZxCxbkQ1ZV3A4w5KkY4HDSdOkr+0prM1GF7com1nb\nRcQ9pIvoAqgegcCazyM1DELWp/0z9MwO6CTZbJRxi7KZ2QiV9fmdhFuUzcwGxYmymZmZmVkNHTvh\niCRn8GZmZmbWchFRu4taRHTkLYXWtLqmNDm2UVNfJ8fmY+2Mujq9vk6OrdPr6+TYfKydUddoq6+T\nY/OxDqmuqLfOF/OZmZmZmdXgRNnMzMzMrIbRkiiXXV9H1NXp9TWzrk6vr5l1dXp9zaxrtNXXzLo6\nvb5m1tXp9TWzrtFWXzPr6vT6mlnXcKivpo4d9UJSRL2O1WZmZmbW8STGk2YGfSaCO4uOp5a+cs6O\nHfWiHo+GMXz5i4+ZmdnIJ7ECsHN22xa4B/gWdGai3Jdh16Lslubhyc+bmZnZyCQxFngPsAspOV4F\n+CNwFfCHCJ4pMLx+9ZWjOFG2tvDzZmZmNnJILAXsQEqMdwCeAK4kJcd/j+CNAsMbECfKVjg/b2Zm\nZsOXhIAN6OlSsS5wHSkxviqCxwsMb0icKFvh/LyZmZkNLxKLANuRulTsBLxMT6vxDRHMKjC8pnGi\n3AEkTQHWiIh927ltpxiuz5uZmdloI7EB8F1gY+BvpOT49xHcX2hgLTKiRr3odJL2Ar4IvA2YCfwL\n+DYwlG8knfltxszMzEYUiZ2Ac4DDgfdH8ErBIRWqpYmypLcB5+cWrQ4cBSwOfBLevAryiIj4Qytj\naQdJXwS+BhxMutpzNrAj8H7g1aFUPfTozMzMzOqTOAg4Gtgtgr8VHU8naOnMfBFxX0RsEBEbABuS\nksVLSC2kJ1XWjZAkeVGgG/hMRPw2Il6LiDkRcWVEfC0rNl7S2ZJekvRvSRvmtl9B0sWSpkt6SNIh\ndfYzQdIvJT0raYakmyUtU4lB0hmSnpT0uKRjJI3J1q0h6dpsu2eyOhZt8WkxMzOzDicxRuI44CvA\nlk6Se7RzCuvtgAci4jFSC+lIayXdFJgAXFpnvUgty78GFgUuB34MkCWzVwC3AyuQBuf+vKTta9Sz\nP7AI8BZgCVLr9WvZurNIrdhrkK5M3Z7Ucl/xbWB5YB1gJWDKQA/SzMzMRg6J+YHzgEnAZiO1H/Jg\ntTNR/hgpSYTUonyIpDuyFtDF2hhHqywJPBsRc/soc2NE/CHSFZS/BNbLlm8ELBUR34qINyLiYeB0\n0jmrNjvb15qR3B4RMyUtS7oi9QtZa/YzwPcrdUTEgxFxTUS8HhHPAieT3hRmZmY2CkksAVwNjAO2\n7fSJQYrQlkRZ0njgfcBvskWnAqsB6wNPAd9r3r6IZtwGsevngKUqXR3qmJa7/yowISu/CrBC1pVi\nhqQZwBHAMjXqOJfU//l8SU9IOkHSfFkd44CncnX8FFg6nRctK+n8rEvGi1k9Sw7iOM3MzGyYk1id\nNKLFLcAeEW/+Om057Rr1Yifgn1krJxExvbJC0umkbge9ZMOiVZQjotzfjiIK69JxEzAL+CBwcY31\nfSXfjwEPR8Radda/uW1EvEHqaH+0pFVIYxnel/2dBSxZp1X7WGAO8I6IeEHSB4Af9X1IZmZmNtJI\nbARcBhwbkbqBjiaSSkCpkbLtSpT3pKfbBZKWj4insocfBO6qtVFETGl9aM0RES9KOgr4iaQ3gD8B\nr5P6Zpfoe9SLm4GZkr5KSl5nk/oRT4iIW8n1586e3OeAu0nDz70OzImIpyVdDZwk6ZvAK6RW+xUj\n4gZgIeBF4CVJK5I67JuZmdkoIrEbqXvnARFcXnQ8RcgaXsuVx5K66pVtedcLSRNJyeIlucUnSLpT\n0h2kfrJfaHUc7RARJ5HGUP4GMB14FPgMPRf4VbcqR7bdHGBXUleUh0jD5v2cdNFepVxl2+VIXVhe\nJCXLZVI3CoD9gPHZ8uezcstl67qBd2fbXUFq9fb4zGZmZqOExCGk7q87j9YkeaA8M5+1hZ83MzOz\nYkiMAb4D7ExKkh8uOKSO4pn5zMzMzEYhiQVIvzwvTRr+bUbBIQ0r7RwezszMzMzaRGJp4BrSxf7b\nO0keOCfKZmZmZiOMxJqk4d+uA/aNYFbBIQ1LTpTNzMzMRhCJzYAbgRMjODKCviZDsz64j7KZmZnZ\nCCGxO3AKqRX5D0XHM9w5UTYzMzMb5iQEfAn4PPDeCP5VcEgjghNlMzMzs2FMYizwA9LcFJtG8FjB\nIY0YTpTNzMzMhimJiaTZjxcEtojgxYJDGlF8Md8IJ2lLSfcWHYeZmZk1l8RypBl6nyNNJOIkucmc\nKDeRpKmSZklasmr57ZLmSlq5RfvdWNJVkmZIek7SPyRNBoiIGyNi7aoYt2lFHGZmZtYeEusANwFX\nAJ+IYHbBIY1ITpSbK4CHgD0rCyS9E1ggW9d0kjYlDSZ+HbBGRCwJfBrYsY8YPZW0mZnZMCUxidSS\nPCWCoyMhghkVAAAgAElEQVRak2OYE+VW+CWwX+7x/sA55JJTSbtkrcwvSnpUUldu3apZ6/N+kh6R\n9Iykr/exv+8AZ0XEdyLieYCIuC0iPpbVV5L0WHb/XGBl4ApJMyV9RdLvJH0uX6GkOyXtNrTTYGZm\nZs0msRfwG2CvCM4uOp6Rzoly8/0dWETS2pLGAh8lJc95LwP7RMSiwC7Ap2skppsDawHbAkdJWrtq\nPZIWBDYBLmoksIjYF3gU2DUiFo6I7wBnA/vk6lwPWAG4spE6zczMrPUkJPF14DhgmwiuKTqm0cCJ\ncmucS2pVfi9wN/BEfmVEXB8R/8nu3wWcTxrSJa87ImZFxJ3AHcB6NfazOOk5fGoIsV4BrCVpjezx\nvsD5EfHGEOo0MzOzJpEYB/wc+Ahp+Ld/FxzSqDHihodTt6YAXTVWdUdXTGmwfM2yDQpSonwjsBpV\n3S4AJL0HOB5YFxgPzA9cWFXP07n7rwITa+xrBjAXWB7476CCjfifpAuBfSV1Ax8DPjyYuszMzKy5\nJBYmdbWYC0yKYGbBIY0qIy5RzhLcKa0q31CdEY9KegjYCfhEjSK/An4I7BARsyWdDCw1iP28Kukm\n0jfM6xvdrMays0kJ/V+BVyPiHwONxczMzJpLYkXgd8DNwGcj8K+9beauF61zALBNRLxWY91CwIws\nSd4Y2Iv+R8WoN1LFV4HJkr5cGZZO0nqSfl2n/DRgjfyCiLgp2/93SQmzmZmZFUjincDfgAuATzlJ\nLoYT5RaJiIci4rb8otz9zwBHS3oJ+CbpTUCdsn0tqyS522S3ByU9B/yMeS/Gy297HPCNbMzlL+aW\nnwO8k94XHpqZmVkbSWxHGvr18AiO9/BvxVFEZ557SRERvVpR6y23oZG0L3BgRGzVovr9vJmZmfVD\nYjJwArB7BDcUHM6o0FeO0tIWZUlvy8YLrtxelHSopCUk/UnSfyVdLWmxVsZhfcuGmfss6YpaMzMz\na7Ns+LcpwFGki/acJHeAlibKEXFfRGwQERsAG5JGb7gUOBz4U0SsRfbTQivjsPok7QBMJw0x96uC\nwzEzMxt1JMYDZwI7k4Z/u7fgkCzTtq4XkrYHvhkRW0q6F5gUEdMkLQeUI2LtqvLuejGC+HkzMzPr\nTWJR4BLSZGR7RfBKwSGNOoV1vajyMaAyEsOyETEtuz8NWLaNcZiZmZkVTmJl0tCs/wE+5CS587Ql\nUZY0HngfacDseURq0u7MKwrNzMzMWkBiA9Lwb2cAh0Uwp+CQrIZ2TTiyE/DPiHgmezxN0nIR8bSk\n5Ul9ZHuRNCX3sBwR5daGaWZmZtZaEjuRJvv6TAQXFR3PaCOpBJQaKtuOPsqSzgd+HxFnZ49PBJ6L\niBMkHQ4sFhGHV23jPsojiJ83MzMzkDgI6CZ1tbip6His7xyl5YmypInAI8BqETEzW7YEcCGwMjAV\n2CMiXmgkaCdcw5OfNzMzG80kxgDfBj4C7BTBAwWHZJlCE+XBcqLcGpLKwLkRcUab9+vnzczMRiWJ\n+YGzgFWA90fwbLERWV6njHoxakgqS3o+u4ix0/jiSTMzszaRWAK4GhgHbOskeXhxotxkklYFNiZd\noPj+Psr53JuZmY1gEquRRra4BdgjgtcKDskGyMla8+0H/Bk4F9i/slDSWZJOlXSVpJeBkqSdJd0t\n6SVJj0v6Uq78rpL+JWmGpL9Kemdu3VRJX5J0h6QXJJ0vaf5s3WKSfidpetaqfYWkFdt3+GZmZiax\nEWmM5B9F8OUI5hYdkw2cE+Xm2w+4gHSx4g6Sls6t2xM4JiIWomfsxAMjYhFgXeBaAEkbVNYBSwA/\nAy6XNC6rJ4DdgR2A1YB3AZOzdWOybVfObq8BP27FgZqZmVlvErsBVwGfiuAnRcdjg+dEuYkkbQGs\nCFweEfcDdwN754r8NiJuAoiI/wGzgXUlLRIRL0bE7Vm5g4CfRcQtkZwDzAI2ydX1w4h4OiJmAFcA\n62f1Ph8Rl0bE/yLiZeBYYFLrjtrMzMwqJD4HnArsHMHlRcdjQzPyEmVpClJktyl11tdbXn+7xuwP\nXF0ZBo80E+H+ufWPVZX/MLAzMDW7ALCSCK8CfCnrdjFD0gzgLcAKuW2fzt1/DVgoHYYWlPSzrHvG\ni8D1wKKSPOKEmZlZi0iMkfge8Flg8whuKTomG7p2zczXPhFTgCn9rB/4dv2QtACwBzBG0lPZ4vlJ\nSeq7qDHSRETcCnxA0ljgEHrGln4U+HZEHDuIUL4ErAVsHBHTJa0P3AaoVgxmZmY2NBILkK5NWhrY\nLIIZBYdkTTLyWpSL8wHgDWAdYL3stg7wF1K/5XlIGidpb0mLRsQcYCa8Oc/7acCnJG2sZKKkXSQt\n1EAcC5FamF/MJnbpqlHGrctmZmZNILE0cA2pi+T2TpJHFifKzbMf8IuIeDwipme3aaQL6fYGxtK7\nRXcf4OGsi8RBWTki4p+kC/l+DDwP3J/VX69FOD828veBBYBnSRcM/r7Gdm5ZNjMzGyKJNUn/a68D\n9o1gVsEhWZN5Zj5rCz9vZmY2kkhsBlwCfDOC04qOxwavrxxl5PVRNjMzM2shiY8ApwD7R/D7ouOx\n1nGibGZmZtYACQFfBL5A6o/8r4JDshZzomxmZmbWD4mxpOuASqSRLR4tNqJhRJpAmj9i2PHFfGZm\nZmZ9kJhI6o+8DrCFk+QGScsiXQs8hjR/0eEMhhNlMzMzszoklgPKwAzSbHsvFhtRh5IWoPfkZs8C\nJwOrEjEsRwRxomxmZmZWg8Q6pOHffgd8PILZBYfUyW4FVppnScQcIq4g4pWhVKxujVG3Jg2ljkHv\n28PDWTv4eTMzs+FEYhJpxtyvRnB20fF0BGk8sDnwEBGP9FoX0fQvEupWZdbDJYHtoivm9LPJwPfR\nR47iFmUzMzOzHIm9gN8AezlJzkiHAtOBE4BVe61vTZKcn/Vwx1Ykyf1xojzKZdNo/7GJ9Z0l6Zhm\n1WdmZtYuEpI4AjgO2DaCa4qOqYNcAbyNiI2JuL7VO1O35p31sKuYPs5OlJtI0lRJsyQtWbX8dklz\nJa3cgn2WJR1Qtawk6bFGto+I8yJih9y2cyWtPoSQ8tNpm5mZDQsS8wE/A/YANo3groJDKoa0FNJZ\nSOPmWR7xMBHT2hJCt8aQWvRPjK44Mrpibjv2W0vLE2VJi0m6SNI9ku6WtImkKZIezxLI2yXt2Oo4\n2iSAh4A9KwskvRNYgNYlj61ITIfal9h9kc3MbNiQWJjUYroSsFUETxYcUpGeB84n4vWiAsgS482j\nKwqfGrwdLco/AK6KiHWAdwH3kBK7kyJig+z2hzbE0S6/BPbLPd4fOIdc8ihpl+wLwouSHpXUlVu3\nataqu5+kRyQ9I+nrQwkoq+9gSf+VNEPSj3PrJku6Mbt/Q7b4DkkzJe2eLd9V0r+ybf+aJf+V7TeQ\ndJuklySdD0wYSqxmZmbtJLECcAPwKPC+CGYWHFKxIubSAXlZdA1tpIxmaWmiLGlRYMuI+AVARLwR\nEZXxB0dqq+PfgUUkrS1pLPBRUvKc9zKwT0QsCuwCfFrSblVlNgfWArYFjpK09hDj2gX4P9KXlT0k\n7VBdICK2yu6+KyIWjojfSNoAOAM4EFiC9LPU5ZLGKV39+lvgbGBx0s8kH8ZdL8zMbBiQeCdwE3AB\n8KkI3ig4JOswrW5RXg14RtKZWavjaZIWzNYdIukOSWdIWqzFcbTbuaRW5fcCdwNP5FdGxPUR8Z/s\n/l3A+UD1+IDdETErIu4E7gDWG2JMx0fESxHxGKlj/PoNbncQ8LOIuCWSc0hXn24KbALMFxE/iIg5\nEXExcMsQ4zQzM2s5ie1IIyocHsHxEaOwkUcag/RdpPcXFkK3JqpbHy1q//1pdaI8H/Bu4JSIeDfw\nCnA4cAopiV4feAr4XrN2qHJ5isrlKc16PAhBSpT3pka3CwBJ75F0naTpkl4ADiaND5j3dO7+q8DE\nOvt7AxhXtWwcUN23qNH6qq0CfCnrdjFD0gzgLcDywApUfQkAHmHk/lpgZmYjgMRk4DzgIxH8uuBw\niiEtQBoneiPgLzWLlMuXqVwe27LH1/zhDzCmDOyobmmg27fDfC2u/3Hg8YiotDJeBBweEc9UCkg6\nndSBvhdJU3IPyxFR7m+HUSpNaebjwYiIRyU9BOwEfKJGkV8BPwR2iIjZkk4Glhrk7h4lfenIWw2Y\nOsj6atX/7Yg4tnqFpEnAilWLVwEeaNK+zczMmkZCQBfpV99JEdxbcEhFmgA8COydn15a5fKYKJUq\no0ycDuRHnGjaY3VrHZaatD7EKcAx0RWhcnmg9Q2KpBJQaqhwRLT0Ruogv1Z2fwppoOrlcuu/APyq\nxnZRp76ayzvhBjwMbJPdXx14d3Z/vuyJXTl7PA3YL7u/cfb4nOzxqlnZMbl6rwMOqLPP7bPtNyK1\n5K5F6u5xUK7MXGD13OOzgGOy+5OBG3PrngLem3u8ISlZ3jirfyKpv/NCpJbrR4BDs/sfAmYDRw+n\n580333zzzbeRf4MYD3EWxM0QyxYdTyfeuO66bbjuunNavp8pTGIK05jC/kUfc3ptEPXWtbpFGeAQ\n4Lzswq8HSS2sP5S0PqmbwsOkrgcjSkQ8VL0od/8zwPey0SeuJ11EsFidsn0tIyKulnQ4cCZpWJvp\nwGnZrd62+SHlqoeXmwKcrfSTzIERcZGkA4EfA2sCrwE3AtdHxOuSPpTt61vAVcDFteI0MzMrisSi\nwCWki+m3jqAjRlToQNeTRidrGXVrPuC7wF7RFR0/oYuyTLrj1Jt3u6/5uK1z+XkzM7MiSKxMasi5\nFvhCBG2fBrkjpIat/xBx3zyLy+WJwGpRKv27baF0a0yRk4hU6ytH8cx8ZmZmNiJJbECaBvkM4LBR\nmyQnE6g918GGpG6YbdNJSXJ/3KJsbeHnzczM2kFiNWDn7LYpcGCEuwVafW5RNjMzsxFJYrzE1hLf\nlbibNPHXRqTJsFZ3ktybyuWNVS5/ueX76dYK6tZhrd5PKzlRNjMzs2FFYnmJT0hcTLqI/QRgJmn+\nguUjmBzBhRG8UGigRZGWQ/oqUr1fch8BbmtpCN2qzHq4oLrrxtHx3PXC2sLPm5mZDZbEWNIwpTuT\nhihdFfgTcCXwxwimFRddh5HWIV28eBZwdGX8M5XLAhaMUqnlI36oW9uR5oz4fHTFr1q9v6HqK0dp\nx/BwZmZmZgMisSSwAyk53oE0zv9VwGHATRG8UWB4nexgYAoRZ1ct3430JePAVu5c3ZpMauH/SHTF\nDa3cVzs4UTYzM7OOIPFWYA9SQvcOoExKjo+I4LECQxs+Ij5fZ83lpFb4llG3JgAfBSZFV4yIWQ/d\n9cLaws+bmZn1RWIX0uRZvyZ1qbghgv8VG9XwpnJ5BWD1KJX+UnQsncyjXhRE0paShtU3KkklSf7W\nbmZmbSPxaeB04H0RHBbB1U6SGyCNQ9qkjxKrk0YAsUFyi3KTSJoKHBDR+dMx9kVSCTg3IlZqcr0d\n+byZmVlxJMYAxwEfAHaO4MGCQxpepLcC3wQm06kJ3TDgFuX2iOzWcpLGtmM/ZmZmrSIxgTQywubA\nZk6SByHiASL2zyfJKpd3Urk8pdW7Vrc2ULeObfV+iuZEuYXy3RgkfU3Sb6rW/0DSD7L7i0o6Q9KT\nkh6XdIykMdm6yZL+KukkSc8CUyTtny37kaQXJN0jaZtc3StIulzSc5Lul/TJ3Lr5JX1f0hPZ7WRJ\n49tyUszMbNTLRrT4EyBguwieKzikkeTvwPmt3IG6tTNwNfDPVu6nEzhRbp/zgZ0lLQRvtgrvDpyX\nrT8LmA2sAWwAbA98Mrf9xsCDwDLAt0kfLhsDDwBLAl3AJZIWy+3vUWB54CPAsZK2ztYdmW27Xnbb\nGPhGU4/WzMysBok1gL+RJqPY032RB0DaEmnRXovL5fEql5cGiFJpRpRKLbs+St06CDgDeH90xYif\n9dCJcptERGUWnA9mi7YBXo2ImyUtC+wEfCEiXouIZ4DvAx/LVfFkRPwkIuZGROVDZXpE/CAi5kTE\nhcB9wK6SVgI2A74WEbMj4g7SRRL7ZdvtDRwdEc9GxLNAN7Bv647ezMwMJN4D3Ah8P4KvRjC36JiG\nDWkycBHw1hprPwx8taW779YYdes44CvAltEVN7Vyf51ixI2jXFa5Kf2ES1FqxYVnvwL2BM4F9qKn\nNXkVYBzwVG62yTGkFuGKWiNRPFH1+BFSC/LywPMRkZ9951Fgw+z+8lnZ/LoVBnIgZmZmAyHxQeDn\nwMcj+F3R8Qwr0rtIF+1NImqOT3w+cGGLo1gAWBrYNLri2Rbvq2OMuES5RQlus1wEfE/SiqQrfCtD\nujwGzAKWjIh6365rfQFYserxKsBlwJPAEpIWioiXs3Ur05NYP0ma/vOe3LonB3YoZmZmjZE4jNTi\nuWPEyO/X2nQRdyK9k4hXK4tULq8PrBil0pVRKgUwp6UhdMUrzNsldFRw14vmGi9pQuVG1ReRrEtF\nmdQf+aGIuC9b/hSpU/xJkhaWNEbSGpK26md/y0g6VNI4SbsDawNXRcTjpP5fx2UX7r0L+ATwy2y7\nXwPfkLSUpKWAo0it3GZmZk0jMVbi+6RplTdzkjwEuSQ5Mx8woYhQRpMR16JcsKuqHv+V3i3BvwLO\nIfXxydsPOB64G1gYeCh7DPWHnvsHsCbwDPA08OGImJGt2xP4KamleAZwVERcm637FrAIcGf2+MJs\nWYXHYjQzsyGRWJDUQLM4sHkEM/rZxCqkpYFna42NrHJZUSpFlEq3Are2P7jRxROODFNKnfoPiIgt\ni46lEX7ezMxGD4llgMuB+4FPRjCr4JA6iro1HjiZdOF9xfXRFZ9PBXQpcAIRf1e3SllZWGG3xVlg\npQk8+OOn5ik/b9095ec1kPIC/hFdcfDAjmx46itHcYuymZmZNY3EWqRfWH8NHBXhXynz1K2xwJXA\nK8AB9PyK+1Ku2B5EvJ7dv53UfRJeuncBZj83P/BCVfm8nvLzGmj5aXXKjyotb1HOxvU9HViX9GL4\nOOkb5gWki8+mAntExAtV27lFuQ+S9ie1KPfXj7kj+HkzMxv5JLYgXbh+ZARnFB1Pp1K3tiG18PZ7\nAZ7K5cWBRaJUeqS/sjY4feUo7UiUzwauj4hfSJoPmEia8OLZiDhR0teAxSPi8EaCdsI1PPl5MzMb\n2ST2AH4C7BPBH4uOZ9hIs/CuSEStYWBRubwPsHyUSt9pb2CjR2GJstLsMbdHxOpVy+8FJkXENEnL\nAeWIWLuRoJ1wDU9+3szMRiYJkS5QPwTYNYI7Cg5p+JDmJ42ENZ6IDxcczajVV47S6uHhVgOekXSm\npNsknSZpIrBsRFT6vkwDlm1xHGZmZtZkEvORWpH3BjZ1ktybutVXjnMB6XqxfebZplzeUeXyni0N\nzBrS6ov55gPeDXwuIm6R9H1gni4WERGSajZrS5qSe1iOiHKrAjUzM7P+SSwB7ADskv39B7BlRN2L\nxUYldWsMcBywlbq1WXTV/An/K8CD9J5s7DHg+VbHOFpJKgGlhsq2uOvFcsBNEbFa9ngL4AhgdWDr\niHha0vLAde56MbL5eTMzG56yrhXrATuTkuN3kibPuhL4fQSPFhddZ1K3JpC6VKwE7NbIlM8ql8cA\n46JU8lB6bVbY8HBZIvyYpLUi4r/AdsB/stv+wAnZ398OpN56LdBmZmY2dBILA9uSEuOdgddIiXE3\ncEME/yswvI6mbi1JymueAraLrnitZ6XG1Gg9rvgcsChwTMuDtIa1YxzlQ4DzJI0HHiQNDzcWuFDS\nAWTDwzVamVslzczMmitrNV6Lnlbj9wA3kcZDPjGC+wsMb9jIJhK5nnTeDo+uXFIsbQV8CtirzuY/\nb3mANmDDbmY+MzMzGzqJCcAkelqNJ5Baja8Crong5QLDG7bUrbdHV9zde4XGAssS8eSbi8rlt5K6\nW9zTxhCtimfmMzMzs8ooFR8H3kdKku8kJcYfAu7yLHpDVzNJBoiYAzxZtXQj0q/sTpQ7lFuUzczM\nRoGs3/EFwILAT4GrIzyyQstI43LTUFsHK3IcZTMzMyuYxAqkvrOPA++N4HwnyUOjbo1Rt95ee6XW\nBO5CWqfXqnJ5ssrlT7Q6PmsOd70wMzMbwSTeQep7/FPgeHevGDp1awHgl8B4UjeW3EptSDrfRxFR\nq0vFDcAbrY7RmsNdL8zMzEYoiW2BXwOfj+BXRcczEqhbSwOXAw8Bn4iumHfcY2kZ4F1E/PnNReXy\nREBRKvkCyQ7krhdmZmajjMT+wK+A3Z0kN4e6tSbwN+AaYJ9eSTJAxPR8kpw5FHB3i2HIXS/MzMxG\nkGxM5KOAyUApYnSOqKBuLQUcDSwMfCm6YnqNMicBS9fYvFf5bErqS4EToitOH2A43wHmDHCbYa+s\n8nzAJqThB/9RitJlBYc0YE6UzczMRgiJ8cDPgHcAm0bwdMEhFULdeivw++x2NfBqnaJ/ASbWWN6r\nfHTFXHVr9+jK9TuWRJoo5GgiHpsnhnJ5K+DVKJVujVJp1PRJLqu8NLAjKTneHniUNAThA0XGNVhO\nlM3MzEYAiUWBi0jTTZcieKXgkAqhbm0GXAJ0RVf8rK+y0RWXDKTueZJkgIhAugCYVqP44oyCPKus\n8hhgA3pmdVwHuJZ0QeOXS1F6osDwhswX85mZmQ1zEiuRWu2uBw6LGH0/8wOoW+OAm4Ejoyuuavv+\ny2UBRKnUmclVk5RVXhR4Lyk53gl4kfT6uxL4SylKvftudzDPzGdmZjZCSawPXAF8HzhpNA//Fl3x\nurq1UXRFUV0dDgemA2cUtP+WKKssUktxZbrz/yN1W7kK+HYpSg8WGF5LOVE2MzMbpiR2BM4BPhPB\nRUXH0wlamiRLewFrE3FUnRJnASNiCLiyygsCW9PTpUKkxPh7wHWlKI2Krj3uemFmZjYMSRwIHAN8\nKIK/FR3PiJYu2jsc+BSwCxH/fnNVubwC8FqUSjOKCq/ZyiofTBqp4zZ6ulTcXYqR2aXEXS/MzMxG\niGz4t28BewBbRnB/wSEVQt1aFvhAfxfsNWt3pNExNiXiyap1B5AmHzmvDXG0VHZh3rHAh4ANS1Ea\nla+tPLcom5mZDRMS8wO/AFYDdovgmYJDKoS6tTappfNcYEp0FZfMqFzWSLh4r6zyBOBMYGVgt1KU\nni04pLbpK+d0omxmZjYMSCxBmvDiGWDfCF4rOKRCqFtbAb8BDo+uOLOQGMrlvYDHo1S6oYj9N1tZ\n5SWAy4Cngf1KURpVry1PYW1mZjaMSawG/BW4FdhjFCfJHyONFb13S5Nk6Z1Iu/ZR4kkYGa35ZZVX\nJ03L/Xfgo6MtSe6P+yibmZl1MImNSK19x0bw46LjKYq6NQE4GNg2uuKuFu9uHLDQPPsvl8cBc6JU\nmhulUrnF+2+LssobA78FvlWK0ilFx9OJ3PXCzMysQ0m8nzQm7ycjuKzoeEYzlcvfBe6MUumcomNp\nhrLKuwGnA58oRemKouMpUqGjXkiaCrwEzAFej4iNJU0BPknPzxZHRMQfWh2LmZnZcCHxWeBIYJcI\nbi46HuNbwMyig2iGssqHkoa726kUpVuLjqeTtbxFWdLDwIYR8XxuWRcwMyJO6mM7tyibmdmoIzEG\nOBHYFdgpgocLDmlkk8YDXwNOJmKeyUJULq8HPB2l0rRCYmuysspjge8COwA7l6I0tdiIOkMnXMxX\na+dOgs3MzHIkFgAuADYCNhutSbK69Q51q6v1O9KipGHmNoKaU3/vAKzX8jjaoKzyAsCFwAbA5k6S\nG9OORDmAP0u6VdKBueWHSLpD0hmSFmtDHGZmZh1LYingGuB1YPsInu9nkxFJ3doWuBbaMpHKTsA9\nwAeJ6DUlc5RKJ0apdHUb4mipsspLk87pa8AOpRg5swi2Wju6XiwfEU9JWhr4E3AIcB89/ZOPAZaP\niAOqtnPXCzMzGxUk3gr8njQ+8DcimFtwSEOibm0EbFtj1a3RFX/uo/ziwGRgj+iK61saZB0ql78K\n3BalUq84h6OyymuRWs3PB745UqehHopCL+aLiKeyv89IuhTYOCJuzAV3OlDzasvsor+KckSUWxiq\nmZlZ20lsSppI5KgIfl50PI1St7YGPgvsXmNmvAlArV+LJ9SprlJ+LjApuuLepgU6cNcAjxa4/6Yp\nq7w5cDHwjVKUTi86nk4hqQSUGirbyhZlSQsCYyNipqSJwNVAN3BnRDydlfkCsFFE7FW1rVuUzcxs\nRJP4MHAqsH8Evy86nkapW/sAJwEfi664tuh4GiZtAfyVquRH5fKiwMtRKs0pJrDmK6u8O/ATYN9S\nlP5YdDydrMgW5WWBSyVV9nVeRFwt6RxJ65P6Lz9MGkDczMxsVJAQ8AXgi8COEdxWcEgNUbdEGrLu\nk8DW0RX/KTikxknzk8753dCr//eJpK4vv213WM1WVlnAl4FDge1LUfpXwSENa55wxMzMrI0kxgLf\nB7YGdo4YHj/zq1vjSK3fGwC7RlfqWjkSqFweH6XS7KLjGKqyyvMBPwC2BHYpRemxgkMaFjpheDgz\nM7NRT2IicAnwdmCL4ZIkZ+aSRqKYNBKSZJXLO6tcXglghCTJE0l93dcEtnSS3Bwtv5jPzMzMQGJZ\n0sXr9wC7RzCskrPoijnACUXH0TBpNWBBom73kJWBp4GGEspsyufD6Nx5IFYGrgcOLkXp9aKDGSnc\n9cLMzKzFJNYmDdF1DtD9/+zdd7xcZbX/8c83CSG0kIBSRIqKcFVQinpNaA9KkYSOIgqIIEhRQQUF\nAX/DCCqCFSv3YkEQQUBFIFJEHwTF66UJWLCiF6VIFaQJfH9/PHOSk+TkZCZT9sw56/165cWZPbP3\nXoSQs86z17OWPeJwi9Ap0quAi4Bjsb/e7uWy8hHABygtbvu1B/ETwM+j/VvrRss5I1EOIYQQukhi\nS0p/5GNsvlZ1PM1SXRMbq8iDRVoGuA14L/b353sr59OA7zfbI3mBkc87JKe/dDrcUL1K+yiHEEII\n45l7sXwAACAASURBVJXEmymbq95iMzADLFTXAcAewOyqY2mZ/TjShtiPjfDu52i+1GJZ4GzKEJTN\nYprd+BSJcgghhNBhjfZvRwOHAa+zubXikJrSaP/2YeAtwKyKw1lyw5Jk5bwO8Hen9JRTamosdlZe\nBfg+ZfPim5PTk12JM/S96HoRQgghdJDEJODLwF7AjAFKkidTaqi3BWa45tsrDqk50jJIy4/yieOB\nzZu9XGPk88+AK4G3RpI8vsWKcgghhNAmieWBbSirsLOAW4Atbf5ZaWBNUl3LApcADwOvdW3EsoV+\n9S5K67pPLuL9g5ya2+A2bOTzccnpKx2KLwyw2MwXQgghtKhRWvFiSg3vLOA1wM8pnS3mAL8bpM4W\njZKLtwDnDtwGPmkS8Cz2s3MP5Xwo8MNmSy0AsvKewOeBfZLTFZ0PNPSr2MwXQgghtEliCrAVJTGe\nDSxDSYq/AOxu80iF4bXFNRv4ZtVxLBH76RGO3gc0VTKxwMjnbZPTLzsYXRhwsaIcQgghLILEWswr\np0iUkoo5wKXALYO0ajxmSGthLzTRUDlPckojJc2L1Bj5fBqwGWXk850dijIMkBhhHUIIITRBYimJ\nLSU+LnErcAMlifoW8AKbzW0+avPLQU6SVR9181t/koR0JHB1o1fygr6pnLdq9nJZeXnge8C6lJHP\nkSSHhUTpRQghhHFNYingzcCOlI4Pf6KsGB8E/K/NYNXsjkJ1TQA+BqwP7FpxOK16H/A2YCvsx0d4\n/93AP5q5UFZejfLf+GbgkBj5HBYlSi9CCCGMWxIrUKbmLQd8DfiBzV3VRtUdqmsK8HXg+cAurvn+\naiNqkbQa8Dj2w3MP5fwq4DanNFLiPKKs/FJKkvwV4CMx8jnEZr4QQghhARJrUBKmnwPvsmmpvnWQ\nqK6VKWUGdwHbuOYnKg6pdfbdIxzdBzgTuLGZS2TlrYFzgSOT09kdjC6MUbGiHEIIYdyR2JCSJH8B\nOGWQ640XR3VNA/4HuAg4xrV5bdT6mjRheMu3dmXlfYBPAW9KTj/u1HXD4IsV5RBCCKFBYltKK7TD\nbc6tOp4eeBg4yDX/pOpAmiatBFyBNBP7qfneyvlk4Cyn9KtmLtVo/3YccCCwdXJz54UAkSiHEEIY\nRyT2B04G9rC5pup4eqHRI3lwkmQA+wGkXRdMkhuuABZqDzeSrLwU8CVgY2BGchqT9eehe6L0IoQQ\nwpjXmKR3ArAvMMvmt9VGFFqhnKc6pZbGgWflqZSNmv8G9kpOj3YluDDwoo9yCCGEcUtiMqXbw+uB\nGWM5SVZdE1XX86qOo2XSJKTRcpLzlPOmzV4uKz8fuIbS6m/XSJLDkup6oizpDkm3SLpJ0i8ax1aS\ndKWk30m6QtK0bscRQghh/JGYBvwAWBHY2uaeikPqGtW1LGUF9aNVx9ISaQXgEmC/UT61u1O6oZnL\nZeWXAz+j1KEfltzatL4QhuvFirKBZHtj269uHDsGuNL2esBVjdchhBBCxzTGT18L3EapSX6s4pC6\nRnWtAvwI+BdwcMXhNE9amVI/fQdw1nxv5bybcp4K0Gyf5Ky8HfBD4P3J6ZTokRza1avSiwXrPnam\n9D2k8c9Bmw4UQgihj0lsQllVPMPmiLE0XW9Bqms9yr/rlcBbXfOTFYfUiocpK+CHYi+48vsKYJVm\nL5SVDwC+AeyenM7rXIhhPOv6Zj5Jf6L8j/AMcLrt/5b0oO3pjfcFPDD0eth5sZkvhBBCyyRmURZh\nDrb5TtXxdFNjJfkW4FjX/NWq46lCo/3bh4G3ALOS0+0VhxQGTNWb+TazvTGwA/BOSVsMf9MlU49H\nIyGEENomcTBlNPHOYz1JBnDN9wKbj5UkWTmfo5xf2uzns/JkyirytpT2b5Ekh47qeh9l23c1/vkP\nSd8FXg3cI2k123dLWh24d6RzJZ0w7GW2nbsdbwghhMEjMYHyCH93YAubP1QcUs+45sH5d5XeB9yG\nfcUiPvFxoKlkNytPA75DeWr92uQ0ZmvQQ2dJSkBq6rPdLL2QtCww0fYjkpajNAmvA9sA99v+uKRj\ngGm2j1ng3Ci9CCGEsFgSU4CvAWsBu9jcV3FIYVGkTYC/Yc/tPqKc1wb+6tT8xrusvDYwh1KXfWRy\nGrM16KH7qhxhvSrw3VKGzCTgm7avkHQ98G1Jb6fsdN2zy3GEEEIYgyRWAi4C7gK2sWmqO8IgUl3L\nA2u45sEtL7BvHP5SOQv4MnAk8OtmLpGVNwW+D5ySnD7b8RhDGCYm84UQQhhIEi+krCp+HzjG5tmK\nQ+oa1bU6pdfwD13z0VXH00nKeYJTauq/XVaeTRke847k9N2uBhbGjao384UQQggdJfFqSo/k02w+\nMMaT5A2A64DvMkhzB6SNkb694MQ95Szl/D7lvDJAC0nyocAZwI6RJIde6fpmvhBCCKGTJHahJEwH\n2FxcdTzdorp2B44DXgAc4ZrPWswp/UPagdKi7zDs+RJhp2Tl/DhNLtZl5QnAycAuwObJ6Y+dDjeE\nRYnSixBCCAND4nDgaMqmveurjqebVNdGlNHbP3PN/646npZI7wBuxb5u7qEWSiyGZOUplIR7DWCX\n5HR/ZwMNYfScMxLlEEIIfU9iIvAJYHtgls0d1UbUHtUlYENgNjDVNX+w4pC6qrFp78fAgU6pqXZ2\nWXllykbNvwH7JacnuhhiGMeq7HoRQgghtEViGeBsYGVgM5sHKw5piaiupSiJ8azGryeBSynJ4JjW\nKLfYxynd2czns/KLKBs1vwd8MLm1legQOiVWlEMIIfQtiedSulr8EXi7zZMVh7TEVNdkyoCMqygJ\n8u9d69Nvwq2QVgL2xP7yQm/l/Grg+lZKLrLyaygbFz+cnL7UuUBDGFmsKIcQQhg4EutRVhXPBT5k\n0/dJpeqaAmwJXO+aHxj+nmt+CtixksC6b0UkMWz1rVFucRTwAWiuVCYr7wb8F/C25HRpNwINoRWx\nohxCCKHvSGwGXAgcb3NG1fGMRnWtxbxyigTcAhzimm+rMq5Bk5XfA7wf2Dk53VB1PGH8iM18IYQQ\nBobEG4EvAPvaXF51PKNRXTXgXcBllNXvyxdcSR5PlPME4DPAR5zSPYv7PEBWngh8CtgGmJWc/tLF\nEENYSJRehBBC6HsSQ4/qDwe2s7m54pDmUl1LLaJF22eAk1zzM72OqTLSwUDGC4/SdkrPKucfAw81\nc6msvCzwTUobvM2SU1PnhdArsaIcQgihchKTgM8CWwCzbf6v0njqmgBsyrwuFfe65rFaX9ycMmHv\nVIbKTOw/z30r5+Wc0r9auVxWXoWyUfP3wNuT01OdDDeEZsUI6xBCCH1LYjlKl4P1gC2qTJJV1zTV\n9XXgLsqgi+UpA052ryqmPvJi4CXAzAWS5AnANcr5+c1eKCuvTxnLfSXw1kiSQ7+KFeUQQgiVkVgN\nuAS4FXiHzagT6BqDOvYDXkdJYofb2zU/NsI55wDLjHC5hT6vuiYCB1Fqjf88wjljgnLeBvgPp/T5\nDr1e3ik92sy9s/LmwAXAscnpqx3+VwuhZVGjHEIIoe9IvJTST/irwEnNtH9zzVZdGwBXAwuOM15U\nkv0tRv5+t9DnG7XGC/UDHguU87rAH52SgT8AwzcdtvW6hST5TcDngH2S0xUt/QuEUIFYUQ4hhNBz\nEgk4D3i/zTfme6+u6cB2wC9d828rCG/MafQ0vhR4j1P6XXMnaVPgN3jhVfpWZWVR+im/E9gpOf2y\n3WuG0CltrShL2hy42fajkvYFNgY+azvat4QQQmiZxD6UdmB72fyoUU6xIfN6EW8E/ASI7zMd0hgh\nPbuxmtysfYBzgP9t595ZeRJlFXkmMDO5uTHWIfSDxa4oS7oVeHnj19eBM4A9bW/V1cBiRTmEEMaU\nRvu3Y4F3ALNsfgWgug6hDJq4lNKL+GrX/HhlgY4RjVXkDwL/5ZTuqyKGrLw8ZbLiUsAbk9M/q4gj\nhNG0W6P8tG1L2hX4gu0zJL29syGGEEIYyySWYupfv8I/19oAmGHz92FvnwGc7lqf1gIOqMYq8j1A\nJT2es/LqlI2aNwOHJKdRN2qG0I+aWVH+CWXi0P6U/pb/oJRibNjVwGJFOYQQeqqx4vtySvnDbEqp\nXft/Dy/9MLzhzUsz/U+P8PnfPt+mqY1fYcko50lO6enWTtJywKuxf9yJGLLyyyhPCM4APpLcUslH\nCD3V7orynsBbgANs3y1pLUrD8WZvPhG4HrjT9k6STgAOpCTcAB+0fVmz1wshhNA5EitQWq0N1Qc/\nSUlwTqT0uW0t4VrQ29LqrHXNd3l20i+4d4ND7DavF0bVKLf4iXLe2yk1195OWpWy8nsT0HainJW3\nppRbHJmczm73eiFUqZlE+b22jx56YfuvkjZo4R5HAL8GVhi6BPAp259q4RohhBA6oLFq/GLmTZx7\nDfBzSm3wJ4DfN9Omral71fVySgL2eSY8dapPvyFWFbusUW6xm1O6p4XTzqD8d/pwu/fPyvtS/hzt\nlZw6sjodQpWaKb24yfbGCxy7tZnSC0nPp2wA/AjwvmEryo/Y/uRizo3SixBC6ACJKcBWzEuOp1AS\n4znAVTaPdPyedS1PWST5gGs+t9PXD/NTzpsDP3NKz7Z+spZttwVco/3b8cDbgdnJ6VftXC+EXlqi\n0gtJhwKHAS9qdL4YsgLw0ybv/WnKTuapw44ZeLekt1JKMo60/VCT1wshhNAEibWYV2u8FXALpaRi\nD+CWTq0aL4prflR1vcI1P9jN+4S5I6QPo7TTa338d/tJ8lLA6cArgBnJ6a52rhdCP1nkirKkFYHp\nwMmUOfdDmfYjthechjTS+TsCO9h+p6RESYh3krQK8+qTTwRWt71QF41YUQ4hhNZIvAbYjZIgr0bZ\niD0HuNyeb6paGK8kAc/B/sdiP9uErDyVMo76KUq5RWzUDANntJyzqcl8jQ15qzJsBdr2XxdzzkeB\nfSkbQaZQVpUvtP3WYZ9ZB7h4pDIOSQbqww5l23mxwYYQwjjTqDs+DjiEMg56DvC/djVtwUL3NVaR\nvwx8qKV6ZGkG8H7s3duNISs/n/KU4mfAu5Nb7LQRQkUaC7hp2KHaEifKkt4N1IB7GdaLsZX2cJK2\nAo5qrCivbvuuxvH3Aq+y/ZYRzokV5RBCWAyJpYAvAZsAOy7Qn7g3MdQ1DdjENf+o1/cez5TzLOBK\npxb7E0tLYbfV0zgrv4KyAfBzwKnR/i0Msnbbw70HWL+ZcovRYoC59XCnSHpF4/WfgYPbuG4IIYxb\nElOBb1MWMbasoj+x6lqbsqr4AyAS5S5TzlOdynQ7pzRniS7SfpK8PXAWZRX5vHauFUK/a2ZF+cfA\ndm7zf6xWxYpyCCEsmsQalBKL64B3VdGfWHVtAlwMnOqaP9Pr+483ynloLsEspyY3zJXZB/9HM3WW\nTcjKb6d0snpDcrq2E9cMoWpt1ShL+iqwHmXF4KnGYXe7D3IkyiGEMDKJof7EXwBO6XYHixFjqGsW\ncCZwiGu+sNf3H6+U8zJO6fHmPqxtgHOA12HfuriPj6bR/u1E4M3ArOR0ezvXC6GftFt68dfGr8mN\nX8PLKEIIIfSQxHbA2cDhNpX0J1Zd04HTgJ1d83VVxDBeNFaRDwW+7JSebiFJ3gL4JvCGDiTJk4Gv\nUAbVzEhO97ZzvRAGSVNdL6oQK8ohhDA/if0pLTvfYHNNpbHUNdk1P7X4T4ZWNTbp/dQpPdx4/SHg\ni06p+b1C0lLAmth/aieWrDwd+A7wELB3cmqr53II/WhJB4581vYRki4e4W3b3rljEYYQQlikRvu3\nOrAPsJXNbysOiUiSO0c5rwv802nuSu1M4PfAwwBO6cSWL1r2FbWbJK9N2aR5BXBkcop2g2HcGa30\n4huNf440aro/l6FDCGGMkZgMnAGsD8ywab5nbuhLynkKsLxTuq9xaB/KpszLAZzS8a1fVBOwWx9f\nvQhZeVPg+8ApyemznbpuCIOm2YEjS1M29Bm4vRcdMKL0IoQw3klMAy4EHgHeYtPzx96q64WUHskX\n9PreY4lyXtopPdn4+n3ARKd0amcuLgFXAu/EbnuTXVbeEfga8I7k9N12rxdCvxst55zQxMkJ+B1l\nd/UXgd83BoiEEELoEom1gWuB24A9KkqSX92I4bm9vvegU84a9vUs5j2lBfh0x5JkoNH67W2U79Vt\nycqHAv8N7BRJcgjNdb34FKWP8u0AktYDzqVMgQohhNBhEptQHnt/wqan/YlV1/LAa4HZwO7AAa55\npL0qYRGU88uA04HNG4eupFFWAeDUhSl29p3tnJ6VJ1A2iu4CbJ6c/tiRuEIYcM0kypM87FGO7d9J\naua8EEIILZIY6k98sM13KgjhJGBDSu/8V7vmP1cQw0BzSr9SzrsMe935ckVpMu7MhsqsPIXyZ24N\nYGZyC901Qhjjmhk48jXKeNSzKT2U9wYm2D6gq4FFjXIIYZyROASoAbvbdK0/sepaGljVNf91hPfk\nWp/2De1jynlFYHOndGl3b6QJlMl4q9KB78NZeWXge8Dfgf2S0xPtXjOEQdPuwJFDgHcBhzdeX0Op\nVQ4hhNABEhOAj1JKHbaw+UPH71HXmsCsxq+tKSuI717wc5EkL7FVgBmUlfjuKJv2zgJeALTdojUr\nv4gyBv0i4Jjk1LGuGSGMFYtcUZa0KnAssC5wC3Cy7Yd7FlisKIcQxgGJKZQOA2sBu9jct5hTWrt+\nXWtQeuE+j1InOwe43DV39D6hR6TdgMuwm5vQtwhZ+T8pK8kfTk5f6khsIQyo0XLO0RLly4HrKSvI\nOwLL235bt4Ic4f6RKIcQxjSJlSireXcB+9m0lfyMeI+6JgD/CfzCNcfAiA5SzhsDhzmlg0b/oKYD\nDy/U57gcH/o+99Ai3l/4vDZl5V0pnS32T06XdPLaIQyiJU2Uf2n7FcNe32R74y7FONL9I1EOIYxZ\nEi+krO5eDBxt01YypLp2AW6JzXe9o5wnAy91SjeP/kHdBmyP/bcRjj+v8WrDRby/8HltyMpHAB8A\ndklO13fquiEMsiWtUZaklYa+BiYOe43tBzoYYwghjBsSr6Y89j7Jbn/Ph+p6D/B+Sku30EXKeQKw\nulP6m1N6CpiXJEvrADOwvzXfSfYGI15sUcebfb8FWXkiZdLudpTOFn/p1LVDGMtGS5SnAjcscGzo\ntYEXdiWiEEIYwyR2oYykPsCmrf7Eqmsipdf9NsBM1xzJT/f9J/A+4I3zHZVeSSmjObmCmEaVlZel\ndK6aDmyWnB6sOKQQBkZTI6yrEKUXIYSxRuJw4GjKpr22HnurrmWBbwIrAru75oc6EGJognKe4DSs\nQ0TpRnExcAb29yoLbARZeRXK8JrfAwcmlzHaIYR52hphHUIIoT0SEyU+TWm3uVm7SXLDbsAjwOsj\nSe4u5byOct5v6PV8STIMjZDeqQ+T5PWAn1EmA741kuQQWhcT9kIIoYskhh57r0RJkjv12Psc4Jzo\ne9wTzwKTR/1Enz2ezcqbAxcAxyWnr1QdTwiDKkovQgihSySeS3kk/0dKTXKs6A2QhUos5r6hKcDa\n2Lf3PqrFy8p7Ap8H9klOV1QdTwj9ru3SC0lbSNq/8fVzJb2ghZtPlHSTpIsbr1eSdKWk30m6QtK0\nZq8VQgiDQmI94Drgh8A+kSQPFuW8JWV64UhmAof1MJymZGVl5Q8AnwC2jSQ5hPYtNlGWdAKl5+IH\nG4cmUx4jNusI4NeUThkAxwBX2l4PuKrxOoQQxgyJzYCfACfbHG+zxI/uVJdU13tV10s6F2Fowk9Z\n1Pcn+0fYR/Q2nNFl5UnAF4C9Ke3ffllxSCGMCc2sKO8G7AL8C8Cl8fkKzVxc0vOBWZRWSENL2jsz\n76f0M4FdW4g3hBD6msQbKT2S97M5o61r1TUJ+BKwH/DPDoQXRqGcl1HOLwNwSs84pY4N+uimrLw8\n5c/cusAWyenOikMKYcxoJlF+0sPGZ0paroXrf5rSBH94jdeqtu9pfH0PsGoL1wshhL4kIYn3U/oa\nb2tzeVvXq2t5Sl/edYAtXevcdLawSK8E3rHQUWkm0sTeh7N4WXl14GrK99PZySl+oAqhg5pJlM+X\ndDowTdI7KOUSi10lkbQjcK/tm5i3mjwfl52E/bmbMIQQmiQx9Nh7X2CmzegjjRd3vbqeRynd+Duw\nk2uO5KcHnNI1Tmn+kgrpPcD5wPMrCWoUWfmllDr471J6JP+74pBCGHMW2x7O9qmStqP061wP+JDt\nK5u49kxgZ0mzgCnAVElnAfdIWs323ZJWB+5d1AUa9dFDsu3cxH1DCKFnJJYDzqX8PbeFzcMduOzW\nlNZeH4v2b92lnDcFtnJKn1r4Te0CHATMxP019TArb035c3dkcmpl31AI456kBKSmPru49nCSjgTO\nbdQmL2lAWwFH2d5J0inA/bY/LukYYJrthTZMRHu4EEK/k1gNuAS4FXiHTazoDRjlvBrwcqcROkRI\nE4DlcX+t6GflfYBPAnslpx9XHU8Ig260nLOZgSMrAFdIepDy0+v5w2qMWzGUkZ8MfFvS24E7gD2X\n4FohhFApiZcClwJfBU5qp7NF6C3lLGCKU3rcKd0N3D3iB8v+nL5JkrOygOOAA4HXJqdfVRxSCGNe\n0wNHJL2CktS+AbjT9uu6GlisKIcQ+pREAr4NHGXzjYrDCS1SzjsBuzmlA+Z/QytgP1JNVKPLyktR\nOqBsDOyYnO6qOKQQxox2V5SH3Ev5qft+4LmdCCyEEAaNxD6UzhZ72fyorWvVtRTwGeDLrvnWTsQX\nmnIpkOc7UsosMtIu2H3VXi0rT6VsKPw3sFVyerTikEIYNxabKEs6jLKSvArlf9QDbf+624GFEEI/\nkRBwLKV92NY2bT32Vl0rUv5OfRL4c/sRhtEo51WBFzmlnzXGUs+/cmw/i7QF9mOVBLiArDwF2BKY\nTZlncAlweHJ6utLAQhhnmllRXgt4j+222h2FEMKgkhh67L0JMMPm721dr641Kaua1wBHuOZIfrrv\nxcAM4GeL/ETFSXJWXgvYgZIcJ8om0Uspg7p+mZyiDj6EHltkjbKkqbb/KWllRuh1bPuBrgYWNcoh\nhD4gMZVSj/wM8Cabth57q66NgIuBzwKfjPZvFZEmAVPp8vey0TTGTs+kTLCdBTwPuIySHF+RnO6v\nKrYQxpMlrVH+FuWn2hsYeSjICzoQWwgh9C2J51OSluuAd9l0YuX3xcCRrvnbHbhWGIVy3g6Y4ZTq\n87+h5YHzgN8CR/Yypqy8CmXVeBawLaXsZg6lpOd/k9MzvYwnhDC6prte9FqsKIcQqiTxckpd6BeA\nU6L92+BRzisBz3NKt807qEmU8otbgEOxu9r7OitPADalJMazKYO7fkhJji9LTm2V8YQQ2jdaztnM\nwJGrFmwFN9KxTotEOYRQFYntgLOBw23OrTqe0DzlPAmY5pTuW/SHtClwI11aKcrKy1GS4lmU1eP7\nKU8m5gA/TU5PdeO+IYQls0SlF5KWAZYFnitppWFvTQXW6GyIIYTQHyT2Bz4G7GFzTdXxhJbtAbwG\neO8iP2Hf0K2bZ+U1KQnx3cD3gHpyiq4mIQyo0WqUDwaOoGwuGP6XyiPA57sZVAgh9Fqj/Vsd2BvY\nyub2tq5X13Tgm8AHXPNti/t86JhvAxfOd0RSt1aPh8vKG1M2an4a+FR0qQhh8DVTenG47dN6FM/w\n+0bpRQihJyQmA2cA6wM729zT1vXqWoeyqng5cJRrjg1aXaScXwas7ZTmLPym9gbWxa4v9F4HZeUd\ngG8AhyanC7p5rxBCZ7U1mc/2aZI2AF4KTBl2PMa2hhAGnsQ04DvAPymDRNrqpau6XglcBJzimj/b\ngRDD4i0LrLiI9y4BlunmzbPyQcCJwK7J6afdvFcIobeaWVE+AdgKeBllM8IOwLW239DVwGJFOYTQ\nZRJrU/5euwp4n01bK7+qazbwdeAg1/y99iMMo1HOcqquvCErCzgJeBOwQ3L6fVWxhBCW3Gg554Qm\nzn8DsA1wl+39gVcA0zoYXwgh9JzEJsBPgTNsjmg3SW54BtgxkuSFKef1lfMbh71ep83Xb6YkqQvc\nSM18X2tbVl6a0hnltcCMSJJDGJuaGWH9uO1nJD0taUXgXmDNLscVQghdIzGLsvJ7iM13OnVd13xZ\np6416JTzNGCWUzqncehJYPqwjyzd5utLgB/Pf1NtBJyFtCX2g239C4wiK08HvgvcB7w2OT3erXuF\nEKrVTOnFF4HjKI+WjgT+BdzUWF3uXmBRehFC6AKJQ4AasJvNz6uOZ6xQzhOBbZ3SZY3XKwAfckof\n6E0A2oYyUfaduHtTD7PyCygbNecA709Oz3brXiGE3mhr4MgCF3oBMNX2LzsV3Cj3ikQ5hNAxEhOA\njwK7A7Ns/tDW9eqa5Jo7MdJ6YCnn9YA7nNJTylmUsdD7O6V/9T4YbQwsh31tt26RlV9F2aj5seT0\nuW7dJ4TQW0uUKKtMLlpkFm37xs6EN7JIlEMInSIxBfgapWxsV5tFT21r5np1rUtJmHZ1zeOmNlU5\nTwEmOKXHGq8vBI5zSr+tNrLuy8o7A18BDkxOF1UdTwihc5a0PdwnGSVRBrZuK6oQQugBiZUoSe1d\nwDY2T7R1vbpmUNrJnTAekmTlPNEpDW10/DJl2tz3AJzSHpUF1kNZ+Z2UEsTZyekXVccTQuidRSbK\ntlMP4wghBCSWBk6h9G3vlPWBc4FjbNqqJ1VduwOnA/u55oWHW4wxyvlIYDngw41DBzj1QU2utAzw\nCeAD2F0r88jKEyh/HncENk9Of+rWvUII/WmxXS8k7ccIK8sxcCSE0EmNld+hTgKndPDSD9u0vQqo\nug6lrCpu71p3S8+qopy3BXZ1Su9sHDodmNvRoS+S5OIJ4OrGP7siKy9DmbS3KjAzOT3QrXuFEPpX\nM+3hXsW8RHkZSs/IGyl/gYxK0hTKX2ZLA5OBi2x/sDHE5EDgH42PftCOtkohjFcS83USaHflI4F6\nsAAAIABJREFUt0tuATZzzX+pOpBOUc4vBI52Sgc3Dv0CuG3ofaf0aCWBLU7ZXNPNzhbPoZTr/BXY\nLjl1LSEPIfS3lrpeAEiaBpxne/smP7+s7cckTQKuBY4CXgc8YvtTo5wXm/lCGAck5nYSsIlOAl2k\nnJehrNYf7pTc2Jz3n07p6opD6xtZeV3KD2wXAsdF+7cQxr4l3cy3KI8BL2j2w7Yfa3w5GZgIDDWB\njyQ4hHFOYm4nAZvoJNAFyvm9wH83VoefoDwRnAT82ykNlTD0N+kw4BHss7p5m6w8d6Nmcjq9m/cK\nIQyGZmqULx72cgJlk03Tj7xUxoneCLwI+JLtX0l6A/BuSW8FrgeOtP1QS5GHEAaaxNxOAp2oIe4k\n1TXVNf+z6jiWhHLeBviNU/pb49AzwLLAo07JlDZ5g6F8//g4sBMwq5u3ysp7ULp67JecxvxGzRBC\nc5pZUf7ksK+fBv5i+/+avYHtZ4GNGuOvL5eUgC8xbxf1iY17vH3Bcxu1zEOy7dzsfUMI/akx+GNu\nJwGbvuokoLq2AM5TXf/pWvN/1/WRlwMPAH8DcEqnVRtOW5YBpgKbYd/fjRtkZQHvoUye3T45jcmN\nmiGEeRq5aGrqs83WKEuayrDE2nbLO4AlfQh43PYnhh1bB7jY9oYLfDZqlEMYYySGOgmsQhkh3Ved\nBFTXXsBpwD6u+Yqq42mWct7YKd1UdRyDJitPBD5N2aQ+Kzn9teKQQggVGC3nnNDEyQdLuhu4Fbih\n8ev6Jm/8nMbmP1T6Xm4L3CRptWEf261x7RDCGCbxHOAqypOp7fopSVZdUl1HU1a6txmwJHkZ4GTl\nPLXqWAZJVl6WsmHvZZQeyZEkhxAWstgVZUl/AF5ju+WRr5I2BM6kJOQTgLNsnyrpG8BGlLZzfwYO\ntn3PAufGinIIY4TEusAPgAuA4/qt/Zvq+iilBna2a/7b4j4fukR6DfAY9i3dvE1WXhW4GPgNcFBy\neqqb9wsh9Ld2u178iWEN51th+1ZgkxGOv3VJrhdCGDwSMyiDRGo2/dpJ4Bzg5EHZwKeclwU+Ahzn\nlB5b3OcHyJrAw5Se1V2Rlf+D0v7tLEp3i9Z6pIYQxpVmEuVjgOskXQcM/dRt24d3L6wQwlggMbeT\ngE3fdhJwzbct/lN95XFKGdyTVQfSUfb53bx8Vt4SOB84JjkNTvePEEJlmim9uB74CaWO+FlK/2Pb\nPrOrgUXpRQgDS0LAe4H3ATvbRCeBDlDOUxq9j0OLsvLQRs23JKcfVh1PCKF/jJZzNpMo32R7465E\nNvp9I1EOYQBJTAQ+A2wNzLLpq01Sqmst19xXMTVDOS8P/JwySe9fVcfTNmlZ4CDgc5Q2ol3RaP92\nNHAYMDs5xebxEMJ82up6Afyg0flidUkrDf3qcIwhhDFAYjnKZLOXUnok91VCqrr2BX6hup5bdSyt\nakzW22yMJMmrAhnYlDKxtSuy8iRK6c9ewIxIkkMIrWpmRfkOSneK+dhueoz1kogV5RAGi8R8nQRs\n+qaTgOoScDxwAKWzxa8rDqkpynkCZTDLxY2pemODVKeU8dVotpl/i7LyCsB5jfvsmZwe6cZ9QgiD\nr62uF7bX6XhEIYQxRWK+TgL2wj9cV0V1LQWcTplYN8M1311xSK1YDtgVuJIl7D7Up07oVoIMkJWf\nB1xC2fB4WHL6d7fuFUIY25pZUd6PkVeUv9GtoBr3jRXlEPqcxDrAbOD/AcfY9F0nAdX1NeA5wJtd\n86NVxzOwpD2Bv2H/dITjMxuvzl/E+wuf1yVZeQPgUsoPRx+L9m8hhMVpt4/yq5iXKC9DGfV5I2UM\nbQhhHJFYCticMpxjFvBcyiCRPWyurTK2UdSAv7vmp6sOpBnKeRpwBrBPZR0upBWwFyxVuB8YqXzh\nfuCOxteLer8nZQ9Z+XXAt4D3JKdzenHPEMLYttgV5YVOKCOpz7O9fXdCmnufWFEOoQ9IrA68nrJy\nvA3we8qK3Rzg+n6bsjfolLOALZ3S1dUEoDcCpwIvxR6YYSZZeT/KCPI9kyv6vQshDKS22sONcLHJ\nwG221+tEcKPcJxLlECrQaO/2KhojnYEXUmpkLwUus7lnlNPDElLOqzqlan9vpQMpK/A7Yd9caSxN\narR/+xCwPzArOf2m4pBCCAOm3T7KFw97OYHS9unbto/uXIgj3jcS5RB6RGIlYHtKcvx64G7KivGl\nwHU2A7EZSnW92jX/ouo4WqWcVwKuovRIrq5biPRi4Ans/6sshhZk5cmUWuQNgJ2S0yBt1Awh9Il2\nE+U07OXTwB227+xceIu8byTKIXRJY3LeyykrxrMaX2dKcvwDm79UF13rGu3fTgLeBGzqmh+uOKSW\nKedJTmkg6qj7QVZeEbiA0g3kzcljoL90CKESS5Qoq6wsrGr72gWObw7cZfuPHY90/vtEohxCF0hs\nRdmM+zRlxfhS4GqbgRyNrLqWBr5KKRHZ2TX/o+KQmqKcXwQc5pSOrDqWQZOV16T8uf0JcERyeqbi\nkEIIA2xJJ/N9BvjnCMf/2XgvhDBgJPYGzqeMDl7X5nCbywc4SZ4OXE6jI08/J8nKeTnl/NHGZj0o\n5S2XVxeQpiPNQVq6shiWQFbeCPgZcCbw7kiSQwjdNFp7uFVt37LgQdu3SOrqVL4QQmc1Si2OpSTI\nW9v8quKQOuXrwE3AUa657xIm5Xwo8LVGm7fHgL9TRjY/3RhFfUVlwdkPIv0/7Ccri6FFWfn1lKch\nhyWnC6qOJ4Qw9o2WKE8b5b0pnQ4khNAdjd7HXwQ2BWbY3FVxSJ20n2t+qOoghijnLYDbndK9jUPT\ngKnAE40R1J+vLLiR2NdXHUKzsvJBwInAbsmpJ8NLQghhtET5eknvsP1fww9KOogyFjSE0OckVqCU\nWjwDbGkzpibTVZ0kK+fnATilvzcOJeBfwL2N4x+rJrIRSOrm2OhuabR/OwnYE9giOf2+4pBCCOPI\naJv5VgO+CzzFvMR4U2BpYDfbXV2Vis18IbRHYg3KhqefA++yiY4KbVLOE4DpTun+xuv3A3c6pW9V\nG9liSENPFa7BHpipqll5vo2ayalva9BDCINridvDSRKwNaVHpYFf2f5RV6Jc+N6RKIewhCReDlxC\nedR/qs3ArSQOp7omALsB33GtulVR5bwnsLlTOryqGFomDT1VeBbYE3sgnipk5emUxZr7gH2T0+MV\nhxRCGKNGyzlHK73AJYv+UeNXqzedAlxNWYGeDFxk+4OSVgLOA9YG7gD2tPunxjCEQSexLfBN4HCb\nc6uOp12qawplA9fqwA8om+Kqcj4leRskTwM/BD6DPRBPFbLyCyg9vecA709OMSY9hFCJlkdYt3Rx\naVnbj0maBFwLHAXsDNxn+xRJRwPTbR8zwrmxohxCiyT2Bz4GvNHmmqrjaZfqWhm4CLgTeJtr7nkb\nO+W8PrC2U6quQ8U4kpVfCXwf+Fhy+lzV8YQQxr4l7aPcNttDKz+TKS2RHqQkymc2jp8J7NrNGEIY\nDyQk8WHgeGCrMZIkvwi4jvJD9luqSJIbplFWs0OXZeWdKKvIh0aSHELoB91eUZ4A3Ai8CPiS7Q9I\netD29Mb7Ah4Yer3AubGiHEITJCYDZwDrAzvb3FNxSB2hui4DvuuaT6/k/jmr0dJtsEj7An/Avq7q\nUFqRld8JHAfsmpx+UXU8IYTxY4lrlNtl+1lgI0krApdL2nqB9y1p8L4RhdAnJKYBFwKPUAaJVFm/\n22k7ulZNTa1yfiPwCsoK/aC5i/L0biBk5QnAx4GdgM2T058qDimEEObqaqI8xPbDki6ltJe7R9Jq\ntu+WtDqNfqMjkXTCsJfZdu5upCEMDom1Ke3frgLeZ9N3k+naUVWS3HA5ZUzy4LF/WHUIzcrKy1A2\naq4CzExOD1QcUghhHJCUKH3vF//ZbpVeSHoO8LTthyQtQ/nGUwe2B+63/XFJxwDTYjNfCK2R2ISy\n4ekTNp+pOp6xQDkvD0xzSndWHct4kJWfQ9mo+VfgbclpYEZphxDGlqo2860O/EjSzcD/ABfbvgo4\nGdhW0u+A1zZehxCaJDELuIzS/m3gk2TVtazqOqLRK7lKuwIHVBxDa6S1kD6NKv+9a0lWXpeyYv8T\nYO9IkkMI/aprpRe2bwU2GeH4A8A23bpvCGOZxCFAjbJp7+dVx9Mu1bUKcDFwO6UzTmX9cp3S2cp5\ncJ5iSXOfKlD2gwyErDwD+A5wQnKqZKNmCCE0q6tdL9oRpRchzCMxAfgosDswy+YPFYfUNtW1PqUV\n2NnACVVM3FPOWwGrOKXze33vtknHAb/B/k7VoTQrK+8OfJlSajGn6nhCCAEq7HoRQmifxBTga8Ca\nwEyb+yoOqW2qawvgAuCDrvmrFYbyIDCYP5DbH6k6hGZk5aWBLSkjyHcGXp+cbqw2qhBCaE6sKIfQ\nxyRWBr5Hafn1Vpuqhm50jOoSpWTgNNd8Zc/vX8orJjmlf/f63uNFVn4+MKvxa2vgV5SnB19LTn+r\nMrYQQlhQrCiHMIAkXgj8gNIZ4Bi7uvrdTnLNVl07V1Fq0XAwsBpwQkX3b500BdgG+5KqQxlJVp4E\nvAaYTUmO16B0OjofODA5DfxTkBDC+BSJcgh9SOLVlJXkk2y+WHU8nVZhkgxwJt3t+NMNKwHbI11K\nnzwGzMrPBV5PSY63A/5C6et9KPA/yWlM9fUOIYxPUXoRQp+R2BX4b2B/m75cQRw0ynlNYBmn9Luq\nYxlUjQl6m1BWjGcD/0EZdnMpcFmUVIQQBlWUXoQwICQOB44GdrC5vup42qW6VgfeS9m0V+UK4xbA\nckAkyi3IyitSVotnATtQNj/OAY4FrklOT1UYXgghdF0kyiH0AYmJwCcokys3s7mj2ojap7peRllt\nPIMK+yMDOKVzqrx/y6QdgZuwK1mlzcoCjgT+H3AtJTk+MTn9qYp4QgihKpEoh1AxiWUovYRXpiTJ\nD1YcUttU12uBc4H3ueazK4kh5zdRyi2+XsX9l5h0OHAMZRW354lyY2PeZykt3V6WnP6v1zGEEEK/\niEQ5hApJPJfSKu2PwPY2Az/KV3XtS1kd39M15wpDuZFB65EsrQ8cCMzEvqPXt8/Ky1F+wJkCbJ6c\nHu51DCGE0E9iM18IFZFYj/JI+1zgQzb9+T9jC1TXBOCrwCmu+dc9v3/OSwMTndJjvb53x0gTce/r\nubPyasAlwK3AO5Kjz3QIYXyIzXwh9BmJzYALgeNtzqg6nk5xzc8Cb6swhMMb/zy1whjaU02S/BIa\nA0EotcgD/0NbCCF0QiTKIfSYxBuBLwD72lxedTxjzGeoeONgS6T/AP6IXdnqbVZOwHnA+5PTN6qK\nI4QQ+lGUXoTQIxICjqKseu5kc3PFIY0JyvmVwNNOafB+P6XTgbOwr63i9ll5b+DTwJuT01VVxBBC\nCFWL0osQKiYx1ElgC2CmzcB3ElBdGwHvA/areNLeWsBT0Kc/eEjPp/Qg/gv2FfO9Zx9cRUiN9m/H\nAu8AXpucbqsijhBC6HexohxCl0kMdRJYGnijzcB3ElBd2wNnAe90zedXHU/fkvYDPgVcDpyOfXXF\nEZGVlwK+CGwK7Jic/l5xSCGEUKnRcs5IlEPoIomhTgK3AAfbDHwnAdX1duAjwB6u+aeVxJDzEcC/\nnFJ/bISUlgc2xr5mgePLAP/GfrqSuBaQlVcAzqfUce+ZnB6tOKQQQqhclF6EUAGJoU4CXwVOGvT2\nb6pLwInAXsCWrrnKcdAXAo9XeP8FTaN0+5g/Ubb7JsasvAZlUuLPgXclp75I3kMIoZ9FohxCF0gk\nGp0EbMZKJ4EJwERgpmu+t9c3V84rUzbtPeyU7uz1/UsQWo+yIfPd2POGw9h3Am+vJKYmZOUNKUny\nF4BTov1bCCE0JxLlEDpMYqiTwF42P6o6nk5xzc8AH6wwhMMoI52/WsndpS2ACyibMifBYExRzMrb\nAt8EDk9O51YdTwghDJKuJsqS1gS+AawCGPgv26dJOoEypvUfjY9+0PZl3YwlhG5rtH87FjgI2Nrm\nVxWHNNac5FTpSuiKwD7YV1YYQ0uy8v7AycAbktNPqo4nhBAGTVc380laDVjN9s0qm11uAHYF9gQe\nsf2pUc6NzXxhYEgM7yQw2+auikNqm+pSxW3fUM67Avc6pZ9VGcegabR/OwHYF5iVnH5bbUQhhNC/\nRss5J3Tzxrbvtn1z4+tHgd8AawzF1c17h9ArElOBi4HnAVuOkSR5J+Cixga+Kj0GPFFxDAMlK08G\nvk7p3TwjkuQQQlhyPWsPJ2kd4GrgZcCRwP7Aw8D1wJG2H1rg87GiHPqexHydBGwGvpOA6joMOB7Y\n1TX/ouf3z3ki8GylZRbScpQJiqdgP1NZHC3KytMoHUEeAfZOTv+qOKQQQuh7la0oDwtgecommCMa\nK8tfAl4AbATcBXyyF3GE0EkSLweuA74FHDroSbLqmqC6TqUkiJtXkSQ3nAjsV9G9h/ybslmvJ39H\ndkJWXgu4FrgN2COS5BBCaF/Xu15IWoqywnG27e8B2PNaS0k6g/LYeqRzTxj2MtvO3Ys0hOZJzO0k\nYDPwnQRU19KUSXurU9q/PVBhOJ+krIhWx36KMlFvIGTlTYDvA59MTp+uOp4QQuhnkhKQmvpslzfz\nCTgTuN/2e4cdX932XY2v3wu8yvZbFjg3Si9CX5KY20nAXmDAxIBSXROAQ4GvuOae1wQr5/WBB5zS\nPxb74TCfrDyL8vfsIcnpwqrjCSGEQVPlZL7NgH2AWyTd1Dh2LPBmSRtRWsb9GTi4y3GE0LZG+7c6\n5c/0VjZjZpOUa36WMoyiKrsDv6KsivaetC9wJ/aPK7n/EsrKB1O6W+ycnK6rOJwQQhhzeraZr1Wx\nohz6icRk4AxgfWBnm3sqDil0QnnqdTxwADAb+9cVR9SUrDwB+CjlB4xZyekPFYcUQggDq/LNfCEM\nMolpwA+AqZRBIgOfJKuuyZXHkPO7lPPrKg5jZeBVwIwBSpKnUOrjtwBmRpIcQgjdE4lyCKOQWJth\nnQRsHqs4pLaprvcA36s6DuB/oeLyFfs+7J2x7640jiZl5ZWBK4GJwDbJ6b6KQwohhDEtSi9CWASJ\nTSgdWU61+UzV8bRLdU2kdHLYBpjlmv/S8xhyXg54wikNTG/ifpGVX0h5svF94Ojk9GzFIYUQwphQ\n5Wa+EAaSxNxOAjYD30lAdS0LnEMpH9nMtfkH/PTQx4FrgPMqubv0cuAfNLruDIqs/GrKU4CTktMX\nq44nhBDGi0iUQ1iAxCFAjbJpb+A7Caiu5YGrgNuBPV3zUxWG8wHg8Qrvvy2lu8bAJMpZeVfgv4ED\nktOIPedDCCF0R5RehNAgMV8nAZsxsUlKdQnYGfi+a73/H145J+D3Tulvvb73oMvKhwNHA7skp+ur\njieEEMaiKL0IYTEam/Y+DqwJzLQZM5ukGsnxRRWGsCHwFNBaoiy9BNig8eq32LeO8P6kRRxf9HkD\nICtPBD4BbA9slpzuqDaiEEIYnyJRDuOSxFLATGA2MAtYlVI3u79daWnAmOOUPreEp74E2LPx9XeB\nBRPelwBTFnF8tPP6WlZeFjgbWImSJD9YcUghhDBuRelFGDckVgV2oCTG2wJ/BC4F5gDX2wx8J4ZG\nmcV01/xApXHkfDJwlVO6srkTVP5f79e/kHokK69C6WrxB+DtyenJikMKIYQxL0ovwrjUqDl+JfNW\njdcFfkhJjg+3GYjeuc1SXZOAzwPPAd5QcTjfAP7awuePA+6hbFobl7LyepQf2s4FPpScxvUPDSGE\n0A9iRTmMKRLTge0oifEOwL2U5GMO8FObf1cYXtc0OlucRxlEsadr/mfPY8h5deA+p9T677G0CvAv\n7H91PLABkJU3Ay4Ejk9OZ1QdTwghjCexohzGLAlRNm4NrRpvBPyEsmpcs7mjuuh6Q3U9D7gEuAE4\nzDVX9cPAicD5wOUtn2nf2/FoBkRW3pPyJGDf5NT6710IIYSuiRXlMJAkpgAfAd4IPE1JjC8Frh5P\nm/FU10rAjZSShY9W0f5tbiw5T3BqclqcNAn76S6H1Ley8iTgNZQ/v7sDOyWnm6uNKoQQxqfRcs5I\nlMPAkViZ0u7sb5TBILfb9Ocf5B5QXZu65hsquXfObwWudUp/av4kzQI+BrwSV7b63XNZ+bnA6ylP\nPraj1HBfCnwpOXpMhxBCVaL0IowZEi8CfgB8BzjWprkVzDGsqiS54VlKXXRzpP0ZGuoyxpPkrDwB\n2Jh5ZUEvoUxIvBQ4KpLjEELof7GiHAaGxGsofXFPsDm96njGq5ZKLBY6WZsB92CPiamHC8rKK1Ja\nDw5tJn2IeZtJr0lOVY4PDyGEMIIovQgDT2J34HRgP5s5VcdTBdW1FPA81/yXSuPI+Uzg607px1XG\n0Q+ysigrxUOrxpsC19JIjpNbKEkJIYRQiSi9CANN4j3AUcD2NjdWHU8VVNdU4ALgz8DBFYdzLIyt\nHtStaEzO25qSGM8GRCmn+ATw4+T0WIXhhRBC6KBIlEPfkpgIfAp4HTDTbmmAxcBTXUsDW1ISst0p\nLeCOqCSWnDcA/uCUnnBqsrZWWpXSMu5Q7J5OPczK05hXArEFMLmDl59GacU3p3H938RwkBBCGJsi\nUQ59SWJZ4BxgKrC5zUMVh1SFC4CVKQnZrsDNFbZ/O4wybe/nLZxzH3BRL5LkRgnES5m/BOIayu/d\nqcCjHbzdg8npkQ5eL4QQQp/qao2ypDUp31xXAQz8l+3TJK1EmSK2NnAHsKfthxY4N2qUxymJVYCL\ngduBA23G7Aaoxtjp5V3zQj8IqK5Jro3fXsOLk5WXo5RADCXHppRAzCFKIEIIITSpss18klYDVrN9\ns6TlKY8rdwX2B+6zfYqko4Hpto9pNugwdkmsT0l0zqZ0txhzj7RV14L9dD/rmj9cbVQLU87HA99x\nSr+uOpYhWflFlN+3WcDmwPWUPy+XEiUQIYQQlkBlm/ls301j04/tRyX9BlgD2BnYqvGxM4EMHDPS\nNcL4IbEFpdzgGJuvVR1Pp6muDSkT9Ib66c4BjnLN/dpP9wbgnqY+KU0ETgGuwO7YGOasPJlSYzy0\najyN8vv2FWCv5PRwp+4VQgghLKhn7eEkrQNcDWwA/NX29MZxAQ8MvR72+VhRHkck9gJOA/a2ubLq\neLpBdU0HNgGucc19WU6inJd1arFkQVqW8gRgOmWQyIPtxJCVVwV2oiTGrwV+y7ySipuSl7CHcwjh\n/7d352GOVdX6x79vd9NAN8hwUWZtRBEEFBCRQWQjKNCADXpBQa9M4hVRwAEFUYtSUUBFuc4IjYqA\nIAKCzHLZXpRBZGxmB1Bm+DHIJDTQ6/fHOdWmu1PpJCeppKrez/PUU8lJzjorldrJys4+e5tZHT2f\nHq4cdvEr4MCIeKqojQsREZLqVuuSDq+5miMidzNPG3kSAj4D7A9sFcGsHqdUmQa1MvDI/MVwDMTj\nFD3J/ew05XxEpNTKSXtTgb8BhxHxfJWDZ+WhbxX+l2L1xf9OkR6pEtPMzKyWpASkpu7b7R5lSYtQ\nTGt1QUR8u9x2O5Ai4kFJKwKXRcSa8+3nHuUxTmIS8F1gY2D7CPp1CELTNKj1KU5E3DcG4oJe59Mq\n5bxkpN7M6JCVh75V+ECKdHEvcjAzs/GnlyfziWIM8qMR8Yma7UeX246SdAiwtE/mG18klqCY+WQi\nsGsET/Y4pco0qO0oZnnZLwbijF7n0yzl/E7gikipk1OoNa2c2m3oW4UdUqSbepGHmZmNT41qzgld\nPvZmwAeALSVdX/5sCxwJvEPSnRRjEI/sch7WRyRWAv4PuB/YcYwUyR8GTgRmjKYiubQlsGrT95be\njbRaJw6clScBPwB2BzZ1kWxmZv1kxE7ma5V7lMcmiXUoTsz6EfC1sTD9mwb1AWAAmB4D8ede59N1\n0l7AtURUKmqz8tC3CpOAXVKkUf+ByczMRp9e9iibzSXxdoqTtD4XwVfHQpFcOgvYZDQVycr5OOW8\ndls7R5zYgSJ56FuFByiGW7hINjOzvuMlrG1ESHyQYinhXSPIPU6no2IgngGe6XUeLToe6Elhn5WH\nvlU4DviqFwkxM7N+5aEX1lXl9G9foFiNcXoEt/U4pXFLOa8APBSphcK0WF3zvUQc24kcsvLbgV8A\nn0yRft6JmGZmZlV46IX1hMQiwEyKlRg3GQtFsgb1Rg1qSq/zaNNPKRb8aY60FnAlsBS1k5+3KSt/\nEDgV2NVFspmZjQYeemFdIbEUxcIRzwFbRIy6oQkL0KBmUAxZeBdFATnaTI+UXmrh/gcAA0T8rMpB\ny+nfhr5V2DJFurVKPDMzs5HioRfWcRKrUCw3fDlwYAQv9jilyjSojwGfo5j+7Zpe59MM5SxgX+CX\nkVLry0pLouILRFaeTDHDyboUJ+09WCWemZlZp/V8CWsbPyTWo1iZ7ljgm6N9ZgsNagLFSYjTgc1i\nIO7qcUpNi5RCOU8FFgdaL5SrF8m13yqkFL1Z0MTMzKxdHqNsHSOxDXAx8KkIvjHai+TS/sCGjKIi\nuexJBiBS+lakdP/Cd9IiSBt2KoesvCrFNwp3Aju7SDYzs9HIhbJ1hMQ+FCeL7RzB6b3Op4OOA94Z\nA/FYrxNpRlkkn6+cX9firq8BPt6JHLLyesAVFMt5fyxFGvVDb8zMbHzyGGWrpJz+7cvA+yimf7uz\nxymNe8r51cBdLU0D1yFZeVvgJGD/FGksfWAyM7MxqlHN6ULZ2iaxKHACsDrwrgge6XFK45ZyXhe4\nJVKa06scsvKHgK8A70mR/tCrPMzMzFrheZSt4ySWAS4EpgBvHwtFsgY1XYNaqtd5tKocbvEl4NXN\n76S3Ik3txPGzsrLyEcAhwNtcJJuZ2VjhWS+sZRLTKKZ/uwj4dAStzM3bFRrUjyimQpvfR2Igjmvy\n/vcB7wT+2fkE9WPgaiKOr7N9n/Lah4e5veF+Uey3c4v7PQpsDdxY4VGRlRelWFRmdWAvBOL9AAAg\nAElEQVSTFGnUf2AyMzMb4qEX1hKJDYFfA0dF8D+9zmdIOY1bPREDC/6TD3P/uvdtLRGtCHwKOISI\nF2u2TwBigSnXNE8ew92+wPbTU5rwvi9+8SvLPPXU/zy6004PNbtfzfEWvK1FWXkZ4CyKovsDKdK/\nqsQzMzPrBY9Rto6Q2AE4Edg3grNH/PiDEsV0bZfGQPTfctjS6yl62o8HjqhaiC70cDnvBpwbaeSn\nXsvK0yge64XAwSlaWvHPzMysb7hQtsok9gO+COwUwdUjfvxBTQS+BWwFTI+B+PtI57BQ0iHAfUSc\n1LVD5LxopPR8t+I3IyvP/VYhReqbbxXMzMza4ULZ2iYxATgSmEEx/dtfRzyHQU0BTgFeBrw7BuKJ\nkc6hHyjnCcBVwC6RUk8+KGTlHSnGJO+bIo34twpmZmad5kLZ2iKxGMUiIitR9CQ/OuI5DOoVFEti\n3wF8KAZi9kjn0E+U8zKRUuvLUXdAVt4f+DywU4o04t8qmJmZdYOnh7OWSfwH8Nvy6jt6USQPHZti\ndo09+qpILpZ83m5EDpXzNsp5IkAviuSsPCErf51i5b7NXCSbmdl40dXp4STNBLYHHo6IdctthwMf\ngrnz7h4aERd2Mw9rjcTqwAUUMxocGkHPFrGIgTi5V8deiGWBnZEuIqJrf59yuMUuwCzg/m4dZzhZ\neTGKpahXBDZNkUbFUt5mZmad0NWhF5I2B54GflZTKA8AT0XEMQvZ10MvekBiY4oCeTCCH/Y6H+ud\nrLwcxUl79wB7pkjP9TglMzOzjuvZ0IuIuByo91WxC+A+JLEzxXjgD7lI7h3lPFE5n6ycV+hVDln5\nNcAVwOXA7i6SzcxsPOrVGOWPS7pR0gmSlu5RDlZD4iDgu8C2EZw34scf1CQN6lsa1BtG+thNkbZG\nWmUkDhUpvUQxF3NPVrnLypsAvweOSZEOSZF6NvTGzMysl3pRKP8AWA1YD3gA+GYPcrCSxESJYymW\nc940gmtHPIdBLUHxFf9awN0jffyFkvYGfg6s3NXD5Lzc0OVI6bKyYB5RWfndwDnA3imSv1UwM7Nx\nrasn89UTEQ8PXZZ0PMVX/XWVJ/4NyRGRu5fZ+CMxBTgZWArYLIIRn59Yg1oJ+A1wLfDRGIgXRjqH\nhqTVgIOBLYi4o2uHyXkScLFy3i5Seqhbx2kkKx9E8Vi3SZGu60UOZmZm3SYpAamp+3Z7HmVJ04Bz\na07mWzEiHigvfwJ4c0TsXmc/n8zXRRLzzk8cjPjUaxrUOsB5wHHAV2Ogbyf1nkTEi10/TM6LREoj\n/kEhK08EjgG2Bqan6M1iJmZmZr3QswVHJJ0KbAEsBzwEDFBU8OsBAdwF/HdELNCD5kK5eyReB5xP\n0Zs8EEFPClQNahdgch9PAddV5dRv+wInRko9mSM6K8+z6mGKNC5XPTQzs/GrUc3Z1aEXEbFbnc0z\nu3lMa0xic+AMivmRe/pcxED8spfHr0taGXiYGJEhIKIY9zwFRr5HPyvXfquwa4reFOtmZmb9yktY\njyMS7wW+A3wggot7nU9fKsbNn0bEJV07RM6KlHra8LLyPN8qpOhtPmZmZr3Ss6EXVbhQ7hwJAZ8B\n9gd2iOCmHqfUN7LyyyjH5gJvhpjQzWm+A3T/Skxb4UH+MXEOIz6rRY2VgINTJH/DY2Zm41rPhl5Y\n70lMopgfeROK6d/uHfEcBrUUcBJwWAzErJE+fq2sLGBNisJ4OrARxcIa5wM/AHV1+IGAP27EtJ3P\n7vk0eE+kSPf0OAczM7O+5kJ5DJNYAjiN4nnePIInRzyHQa1CUYReDtw20scHyMqLA1vy7+J4EeA8\n8eL/rM8B974sbu363NHKec1I6XaABD39sGBmZmbN8dCLMUpiaH7i64D9Ihjxacc0qPUoThY7Fvjm\nSE7/lpWnURTF2wObAzdQTEV3PnBzYstFgZ9SzPYwnS42BOW8KHAJMCNSqreku5mZmfWIxyiPMxLz\nzk/cg+nfNKhtKIZbfCwG4vRuHy8rTwY249/F8XLABRR/h0tSzFegSidT9LTvQcRz3c6vH07gMzMz\nswW5UB5HJLYCTgU+EUHP5ifWoKYD/4yB+EO3jpGVl6coiqdTnJB3J0WP8XnAtSnSnOET1ErAg0TM\nvY9yXh54fGhOY+X8cuCf7Vwve5G/ABwRKf2rs4/czMzMOsUn841hEotQnKg31JP6cuC9EVzWy7xi\nIM7vZvysvCNwIvC/FMM79k/RwtLPEfcr5yXIOSKlZ8qtJwKHAjeW108CPtvm9dnAP6CnM1uYmZlZ\nBe5RHoXK5ae3oyiO30GxwuFQT+o1EWO7OMvK+wOHATulSH9saidpMhGza5eJVs4zgbMipXO7l62Z\nmZn1M/coj3ISE4A38e9e4zWASykK409EcH8P00ODmhgD0fXiPCtPAI4GdgA2S5HuampHaT3gKOV8\nKTAZ+Ep5yz4eN2xmZmbDcaHcpySWBt5JURxvBzxK0Wv8WeAPESO/5HE9GtQ04GwN6r0xEHd06zjl\nFG8/A14BbJoiPbbQ3HLeFtg54CPAbsC/gLkn7rlINjMzs0ZcKPeJcvW8tfl3r/EGFHMPnwcMRtBc\n7+kI0qA2BH4NHN3lInm58jh/B96ZIj1fN5+cXwMcFintVW66ArimnPrtMVfFZmZm1goXyj0m8RZg\nT4oCOSgK46OAHMGzPUytIQ1qB4qT3/aNgTi7W8fJyq+hmObtl8Dna2eyUM5TgG8BHyl7h+8BfoQ0\nlYhnIqURX2DFzMzMxg4Xyj0ksSdFUfwNiuEVt/VizuNWaVD7AF8GdoiBuLpbx8nKmwBnAV9MkY4D\nUM6fBb4XKT1NMZTidxT/xy9ESs8jPQbciLQ7Ec2d6GdmZmZWhwvlHiiHWQwAHwS2iOD2HqfUqjuA\nt8ZA/K1bB8jK7wF+cPYMvnPsQZxf8+nhSWBx4OmyF/mUubdIbwAuBj7nItnMzMyq8vRwI0xiMsWK\neWsDO0TQ/Ny/48Dki/NrZ+7NHqvcx57AjltexubA/0VKNyx0Z2kKsAERv+9ymmZmZjZGeGW+PlHO\nZPEr4Glg9wieWcguY145zvg/IqV7svLEWevw22l3s/qST/PWFOkfvc7PzMzMxrZGNeeEkU5mvJJ4\nJfB74Fbg3aOlSNagluh4zJyXrLm6I/DfWXkqcOa6N/PSkk+zrotkMzMz6zUXyiNAYn2KqcpmAgeM\nlpXzNKiNgds1qFd2LGbOG1GcoAdApHTaZVvyHeAy4HFgeor0z+aC6UdIa3cqNzMzM7NaLpS7TGI6\nxQlmB0ZwzGiY1QJAg3o3cC7w4RiISr27ynmich46cfRPFAupAJCV1wKupJgWb68UacGFVCQhLV4n\n9E+AO6vkZmZmZjacrhbKkmZKekjSrJpty0q6RNKdki6WtHQ3c+glif+m6EV+VwS/6nU+zdKgDgK+\nA2wTA3F+B0IeCnwUIFKaE6mYCzkrvw3IwJdSpMEU862UJ70J6XjgPuBjC0SNuJKIFzqQn5mZmdkC\nut2jfCKw7XzbDgEuiYg1gEvL62OKxASJrwGfBjaP4Mpe59QsDWoQ2BfYNAbiug6F/Tbww9oNWXk3\n4Azg/SnST4bZb3FgFrAFEV/vUC5mZmZmTen6rBeSpgHnRsS65fXbgS0i4iFJKwA5Itass9+onPVC\nYlGKIQGvouhJ/n+9zag1GtQbgH/EQDxRKU7OawH/ipTurt2elUXx4egjwA4p0iykNYEvEPH+Ksc0\nMzMza1WjmrMXC44sHxFDcwc/BCzfgxy6QmJZ4GzgYWCrCP7V45RaFgNxU4dCvQ14BLh7aENWnjSJ\np37yIkuuDWySIt1f3nQ3cGSHjmtmZmbWET1dmS8iQtKoOLltYSReDZwP/Ab4TARzepxSNdIBwPFE\nPDvf9r8BQ+PKVyfi8Tq3vykiflS7+UUtcdcyDKwCmvMybp22Wsx8YO6NEc9RDLEwMzMz6xu9KJQf\nkrRCRDwoaUWK3te6JB1eczVHRO52cu2Q2IiiJ/mICL7X63yapUGtEgNx7zA3TwaWAJ6db/ubgKGv\nJ+aZxk057/+qU089/O7ddptne1ZeCX791ARmn7IiF+y7WsxccGYLMzMzsxEgKQGpqfv2YIzy0cCj\nEXGUpEOApSNigRP6RssYZYkZwPHAPhGc0+t8mrXtf2n/tR7hyCMu5aopL/I9Is6uGlM5vx54ItLc\nIRVk5XUopn77IXDkAjNbmJmZmfVQz5awlnQqsAWwHMV45C8CvwZOB15JMTZ114gFTxwbDYWyxMcp\npj6bEcE1vc6nGRqUgM989vcccuBVXL7i0/wcuIiI5hb5mD9escre7Ejp+flvy8pbAacCB6VIp1RK\n3MzMzKwLelYoV9HPhbLEBOAbwHbA9Aju6nFKw5PWBd5CxPEa1CTgu8DGwPYxEPdVDp/zkcAdkdKJ\ntduz8h7A0cCuKdLvqh7HzMzMrBv6bdaLUU1iceDnFL3km0bw+EJ26bWnYe4Udd+n6Ml/WwzEkx2K\n/wXgxaEr5fRvXwT2BFKKdFuHjmNmZmY2otyj3AKJlwPnAHcBe0WwwHCDnpEE7A2cTsRTde8yqGnA\nfTFQbTU75bwVcF+kdHvt9qw8GfgRsA6wY4r0YJXjmJmZmXWbe5Q7QOK1FNO/nQ58oa+mf5MmAd8D\n3kJx4lzdQjkG4u4OHXEF4LnaDVl5KeBXFLNkpBTpmQ4dy8zMzKwn3KPcBIlNgTMpCuQf9zqfBUjb\nAQcAuw7Xm1z5EDkrUv0ZK7LyqhQfIn4HHJgivdSNHMzMzMw6rVHNOWGkkxltJHahmKljr74skgEi\nLgC2ry2SNagNOnyULyvn/5x/Y1ZeD7iCYtnuj7tINjMzs7HCQy+GISHgU8BBwDsiuKHHKTUWMQfm\nTv/2eWBvDWqDGIhOnWz4Y/59UiAAWXlb4GfAR1OkMzp0HDMzM7O+4KEXdY/NROBYijmgt4/gH73I\nY1jS64HbqHnyNKgJFKvmHQisCewQA1HpZDrlvArweKQFxxtn5X2BLwPvSZH+UOU4ZmZmZr3ioRct\nkJgKnAW8DnhrHxbJE4BjKKZ5KzYNah3gAeCnwL3AFlWL5NJBwNa1G7KysvIRwGeBzV0km5mZ2Vjl\noRc1JFYAzgVuAT4cwewep7SgiDlI2zHvVwF/BjaJgfhbh492cO0JfFl5UWAm8GpgkxTpkQ4fz8zM\nzKxveOjF3OOxFsXMDT8BvhRBX/xhNKipwNuB7YFtgDfFQDzWtePlvBdwfaQ0z5jsrLwMRU/7o8AH\nUqR/dSsHMzMzs5HioRcLIbEFkIHDIxjshyJZg9pDg7poqed4cP0HOAz4CzAdur4S4CPAPGOSs/Jq\nFDNbXEexJLWLZDMzMxvzxv3QC4ndgW8Du0Vwaa/zqaFt/8yvfnMKq0wMLiTiG107UM6LAC9GShEp\n/ab2tqz8Zorp8b6WIn2nWzmYmZmZ9ZtxWyiX078dAnwE2CqCWSN6/EGtAmwH3B8Dcd78t8fhnATc\nCJwAfKvL6XyXYtjJrwGy8uIUM35MB3YDPpQi/brLOZiZmZn1lXFZKEtMAr4PvBnYJIL7u37MQU0C\nNqYoPqcDqwIXURTCC4p4CWkLIh7tdm7AYb/ZniXys3m/MrctKIr084BNU6Q/j0AOZmZmZn1l3BXK\nEksCp5dX3xZBV5Z8rmNjip7b84D9gatjIF5suEcXi+Rlz8wbffsglnvlPaTLiuL4FcCFwM+BD6ZI\n3R4LbWZmZtbXxtWsFxIrURSqfwT2j6Bxodpq/GLRj7ViIG5pI7kJwGJEPNvJnGpl5RWAbYHtn5/M\nDi9O4u9Tn+U0imEXf/Ly02ZmZjbeNKo5x02hLLEu8BvgB8BRnZrZQoNaimJRju0pxhw/QTGFW2sF\nr7QnsA4Rn+5EXgBZeQLF8JLpZX6vAS6h+LBwYYrUiUVJzMzMzEatcV8oa+rD2/DC1JNY7InP8KlV\nTgdm1xv2oEEtCkysE2Ke+2flKQA7HLLDj16c8OKMiTHxyskvTL5opcdXuvh7J3yvrUU/luGPE5fg\nr4vcw27PtbN/jdp5l7cFHqboMT5/2wtIzy/GBZHS1RWPYWZmZjYmjOtCWTP2OY51T92Xic8/z4Q5\nc8rN74+BOGuB+w7qFGCnOmHeHwNxVlaeDPyYYiaIF+cwR0JRTKDRiqG/eVfWU3kBuJyi1/iCFOnu\noRuU82bArZE8/tjMzMwM+rRQlnQ38CTwEvBCRGw03+2VCmUJsd6JJ5AG/otb3vuuuPjrF1TJt1yZ\n7kyKoRXvT5FaH0sszQC+SLEE9EeJOLVKTk0dMuelgCdrl6I2MzMzs0KjmrOXs14EkCI6vxyzxGTg\nOG7YY20mPbdWXLPfX6rEy8qvAi4ALgY+1dRJb9IkYoHhHX8DPgFcScQLVXJqwQ+B46GvFlMxMzMz\n63u97FG+C9gwhpkCrd0eZYmlgF9RLMO8e8S8yzG3Kiu/CTgHODpFOrbJJDYGvkTEO6scuxOU82KR\nUtVxz2ZmZmZjUr8Ovfgb8E+KoRc/iogfz3d7y4WyxCspTly7DDgogkrTnWXlHYATgQ+nSAuMaUZa\nEngLEb+db/siwEQielKgKucdgasipUd6cXwzMzOz0aJRzTlhpJOpsVlErE8xpdr+kjavEkzbfOp9\nLHfbH4GZwAEdKJL3ozhxb8cFimRpAtJ5wP3AZ5EWnef2iBdaKZKV84rKeZNOXQdeR7GAiJmZmZm1\nqWdjlCPigfL3I5LOAjaimK1hLkmH11zNEZHrxdIOH/0C650+iF76Wlz47WOq5FXOPXwkMAN4a2LL\n+xCLzVP4RsxBOhJ4HxGdWNlvFYop3a7sxPVI6RsdyMnMzMxszJGUgNTUfXsx9ELSFGBiRDwlaSrF\nSXKDEXFxzX2aGnqhnfY8iTXO351Zu38kLvj2jxd2/0ay8mLAT4GVgRkp0qNIpwEzibioSuz5Kef9\ngKsjpes6GdfMzMzMmtePs14sD5wlaSiHk2uL5GZoy8MnsvxNlzPt+g25fs9t4pKjf7vwvYYLpkWe\n4I3bTOQrX3mJJe4Atk4x9wS43YnoxtLOdwF1T2Q0MzMzs94blQuOSCzKG066gE2+uSF/3n7DuPSI\nOyscaIdnWfnkm/j6opN59OInWWenFGnOwnds41A5LwrM9pzGZmZmZv2hH3uU2yaxLHAWN/3XIzy/\n5Epx+05PN7njBOBVRNxVu/lP/HD207z2OZhwyMax2w+6kHKtHwJnUKyaZ2ZmZmZ9rJezXrRMYjXg\nCuAaYNemi+TCyhTjj+fKyjs/zetOhgn7pEjdLpIBPkkxfZ2ZmZmZ9blRM/RC4s3Ar4EjIvjecDsB\n61BMOXcsEc8PFz8rHwQcDLwrRbq2k7nPk1LOGwO3R0pPdOsYZmZmZtaeUT/0Qtt/9Mus8KGP8uAG\ne0ZwboO7Xga8imJow1RggUI5K08EjgG2BjZNkf7ejZxrbFf+vqrLxzEzMzOzDur7HmW9+4Ons/rF\n72HW7nvGhcecVN64GjCbiPvm22k54FGGeVBZeQpwMrA0sHMK9/KamZmZjWf9ujLfQmm3nf7EqlfM\n4IY90twiufAeigVK5hXx/xoUya8A/hd4Gti2m0Wycv6qcn5Lt+KbmZmZWff1daG83z3XvX7WCf88\nLC45ap4V+4j4BhFnDbPbArLy6yhWrbsE+GCKNOzY5Q75NXBbl49hZmZmZl3U10Mvfs+Zz7zE4k/P\nYbF/VQy3DPDJFGlmJ3KrRzn/B/CY50g2MzMzGz1G7cl8L7DMuh0K9XSK9EiHYg3nB8D3gN91+Thm\nZmZmNgL6ukd5uOq+HynnyZHS7F7nYWZmZmbNa7gatAvl9innXYDfRUoP9zoXMzMzM2vdqJ31YhRY\nmWL8s5mZmZmNMe5RbpFyXhSY7ZP2zMzMzEY/9yhXpJxr/3hXA6v3KhczMzMzGxkulBdCOR8PbF+z\n6S2R0l96lY+ZmZmZjYy+nh6uF5TzR4DnI6UTy02fB+ZOLRep64uVmJmZmVkfGPeFsnJ+G/CGSOm7\n5aaLgLkLnERKD/YkMTMzMzPrqXFXKCvnacAONYXxP4C58x9HSnf1Ii8zMzMz6y9jvlBWzlOBXWuG\nUjwDPDt0e6R0N3D3yGdmZmZmZv1szE0PV85QsR1wUaT0knJeBDga+HSk9FKn8zQzMzOz0asvV+aT\ntC3wbWAicHxEHDXf7U0Xysr51cDDkdLT5fWZwKGR0kMdTtvMzMzMxpC+m0dZ0kTgu8C2wOuB3SSt\n1fT+OU9WzkvUbDoMWHvoSqS0d22RLClVTrr2+OMoXj/n1ul4/Zxbp+P1c26djtfPufV7vH7OrdPx\n+jm3Tsfr59z6PV4/59bpeP2cWzfiDadX8yhvBPwlIu6OiBeAXwAzGu2gnCfWXD0S+M+hK5HSPpHS\n1Q12TxVyHe/xOhmr3+N1Mla/x+tkrH6P18lY4y1eJ2P1e7xOxur3eJ2MNd7idTJWv8frZKzREK+u\nXp3MtzJwT831e4G3DHdn5bwXsB5wYLnp05HSnO6lZ2ZmZmbjXa8K5VYHRp8GnDR3ZxfJZmZmZtZl\nPTmZT9LGwOERsW15/VBgTu0JfZL6czoOMzMzMxtT+mrWC0mTgDuArYD7gT8Cu0XEbSOejJmZmZlZ\nHT0ZehERL0r6GMVy0ROBE1wkm5mZmVk/6dsFR8zMzMzMeqlX08MNS9JMSQ9JmtWBWKtKukzSLZJu\nlnRAxXiLSbpa0g2SbpX0tQ7kOFHS9ZLO7UCsuyXdVMb7YwfiLS3pDEm3lY934wqxXlfmNfTzzyrP\nh6RDy+d1lqRTJC3abqwy3oFlrJslHbjwPRbYf4H/W0nLSrpE0p2SLpa0dIVYu5SP9yVJG3Qgt6+X\nz+uNks6UtFTFeF8uY90g6VJJq7Ybq+a2T0maI2nZirkdLunemv+9bavEK7d/vPz73SzpqOH2byK3\nX9TkdZek66vkJmkjSX8s410j6c0V471R0pXl68o5kpZsMlbd194KbWK4eG21iwbxWm4XDWK12yYa\nvm+12i4a5Ndyu2iUW5ttYrjcTmunXTSI11a7aBCv5XahYeqHCm1iuHjttonh4rXTJoaL1W6baFh7\ntdomWhYRffUDbA6sD8zqQKwVgPXKy0tQjIteq2LMKeXvScBVwFsrxvskcDJwTgce713Ash18Ln4K\n7F3zeJfqUNwJwAPAqm3uPw34G7Boef00YI8K+awDzAIWoxgKdAmweosxFvi/pVg6/TPl5c8CR1aI\ntSawBnAZsEEHcnsHMKG8fGSzuTWIt2TN5Y9TrLbZVqxy+6rAha3+Tw+T2wDwyTb/N+rF27L8H1mk\nvP7yKo+15vZvAJ+vmFsGtikvbwdcVjHeNcDm5eW9gC81Gavua2+FNjFcvLbaRYN4LbeLBrHabRPD\nvm+10y4a5Ndyu2gQq902sdD36FbaRYP82moXDeK12y4WqB/abRMN4lV5r6gXr633imFitdUmhovX\nbpto9afvepQj4nLg8Q7FejAibigvPw3cBqxUMeaz5cXJFEXVY+3GkrQKMB04Hmhque5mwnYkSPGp\ncfOImAnFuPKI+GcnYgNbA3+NiHsWes/6ngReAKaoODF0CnBfhXzWBK6OiOci4iXgd8C7WwkwzP/t\nuyg+bFD+3qndWBFxe0Tc2UpOC4l3SUQMTbN4NbBKxXhP1VxdAvh/7cYqHQN8ptmcmojXVrsYJt5+\nwNeiWCyJiHikYm5IErArcGrF3B4Ahnp8lqaFdjFMvNeW2wF+C7ynyVj1XntXpv02Ufe1vN120SBe\ny+2iQax220Sj962W20WD5wJabBcNYn2E9tpEw/foVttFg/zaahcN4rXbLuavHx6nzTYxTLzHKr5X\n1IvX1nvFMLHaahPDxSuvt/Ve0Yq+K5S7RdI0it6SRiv4NRNngqQbgIcoPpXeWiHct4CDgU7NCx3A\nbyX9SdK+FWOtBjwi6URJ10n6saQpHcgR4H3AKe3uHBGPAd8E/kExa8oTEfHbCvncDGxefgU2Bdie\nFgrHBpaPiKGl1B8Clu9AzG7YGzi/ahBJR0j6B7AHRc9Du3FmAPdGxE1Vc6rx8fIrvxOa/WqzgdcC\nb5N0laQsacMO5Lc58FBE/LVinEOAb5bPw9eBQyvGu6V8PgB2oei9acl8r72V20SnXsubiNdyu5g/\nVtU2URuvE+2iJt5V5aa228V8j3UNKraJYZ6HttvFfI+1cruYL7+22kWd+uEWKrSJDtcjzcRruk0M\nF6vdNlEvXpfeKxYwLgplSUsAZwAHlp8K2xYRcyJiPYpC6m1qc61xSTsAD0fE9XSuN3mziFif4qul\n/SVtXiHWJGAD4PsRsQHwDMWLTSWSJgM7Ar+sEGN14CCKIRgrAUtIen+78SLiduAo4GLgAuB6Ovfh\nZegYQesL7XSdpMOA2RHR9geXIRFxWES8EvgJxYfAdvKZAnyO4mvhuZsrpvYDig9+61H0LH2zYrxJ\nwDIRsTHFB93TK8YD2I0KHx5rnAAcUD4PnwBmVoy3N/BRSX+i6AGa3crO5Wvvryhee2t7k9pqE518\nLW8Ur512US9WlTZRG4/i9ahSu6iTX9vtYr5YT1GxTTR4XttqF3XiVWoXdR5vW+2iTv2w5Xy3t9Qm\nOlWPNBOv1TYxXKx220SdeNMpPvB08r2irjFfKEtahOKF+ucRcXan4kYxDOE8oN3epE2Bd0m6i+Jr\npbdL+lnFnB4ofz8CnAVsVCHcvRSf1K4pr59BUThXtR1wbbNfzQ1jQ+CKiHg0Il4EzqT4e7YtImZG\nxIYRsQXwBMVYtKoekrQCgKQVgYc7ELNjJO1JMfSn7Q8ZwzgFaPoksvmsTvEB6MaybawCXCvpFe0m\nExEPR4limFOVdgFF2zizjH0NMEfSf7QbrBw+tDPFWPuqNoqIs8rLZ1DxsUbEHRGxTURsCPwCaLpn\nr+a196Sa196220SnX8uHi9dOu2git5baRJ14ldpFvfzabRfDPNa220SD56GtdvC9ItQAAAYNSURB\nVDFMvLbbxTB/u7bbRbn/UP3wJjrwPtGBeqRhvCrvFQ1ya+t9oibeBhQf9Dr2XjGcMV0ol+ObTgBu\njYhvdyDeckNfT0lanGKQe9NnqdeKiM9FxKoRsRrFUIT/jYgPVshtisozbyVNBd5JcYJaWyLiQeAe\nSWuUm7YGbmk3Xo3daGEc5jBuBzaWtHj5HG8NVP3K6RXl71dSvDh3onfvHIqvlyh/d+qDWuVPzSrO\ncD8YmBERz3Ug3mtrrs6g/XYxKyKWj4jVyrZxL8UJKW1/yCjffIbsTIV2UTobeHsZew1gckQ8WiHe\n1sBtEXF/xbwA/iJpi/Ly24G2xioOkfTy8vcE4PMUvZDN7Dfca29bbaLJ1/Km28Vw8dppFw1itdUm\n6sWr0i4a5Ndyu2jwPLTVJhbyvLbcLhrEa6tdNPjbtdwuGtQP7baJZuqRVtpE3XhttonhYr2m5m6t\ntIl68a7s9HvFsKILZwhW+aEoou4HngfuAfaqEOutFF9Z3UDxhFwPbFsh3rrAdWW8m4CDO/SYt6Di\nrBcUn6xuKH9uBg7tQF5vpDi790aK3oJKs14AUykG7y/Zgdw+Q1G4z6I4AWKRivH+r4x3A7BlG/sP\n/d/OHvq/BZalONHjTophHUu3GWtvihM87gH+BTwIXFAht72BPwN/r2kX368Y74zyubiBovflFS3G\nqtveKWY3aWXWi3q5/axsrzdSvAkt30a8ufkBiwAnlY/3WiBVeazAicCHO/Q/tyHFGMobgCuB9Sv+\n7Q6g+HblDuCrLcSq+9pboU3Ui7ddu+2iQbyW20WDWO22ibrx2m0XDfJruV00eF7bbRPDvke30y4a\nPNa22kWDeC23C4apHyq0ieHi7dxmmxguXjttYrhY7baJhdZerbSJVn+84IiZmZmZWR1jeuiFmZmZ\nmVm7XCibmZmZmdXhQtnMzMzMrA4XymZmZmZmdbhQNjMzMzOrw4WymZmZmVkdLpTNzCqQNEfSSTXX\nJ0l6RNK5C9nvjZK2ayL+QpdqlrSnpJckrVuz7eZyAZ3KmsnBzGwscqFsZlbNM8DakhYrr7+DYpWo\nhU1Svz7FsrAL0+xk9/cCh7WxXzPajiVpYgfzMDMbUS6UzcyqOx/Yvrw8tEy7ACRtJOkKSddJ+oOk\nNSRNBr4EvFfS9ZJ2kbSEpBMl3STpRkk7DwWX9BVJN0i6cmi59fkE8BuKgn2N+W+s7RGW9J+STiwv\n/0TS98u4f5WUJP1U0q1D96nZ75iyl/q3kpYrt60u6QJJf5L0f5JeVxP3h5KuAo5q949qZtZrLpTN\nzKo7DXifpEUpllu9uua224DNI2IDYIBiydvZwBeAX0TE+hHxy/L64xHxhoh4I3BZuf9U4MqIWI9i\nqfV9h8lhDnA08Lk6t8Uwl6FYMncT4BPAOWWMtYF1Jb2hJodrImId4Hfl4wA4Dvh4RGwIHAx8vybu\nSsAmEfHpYfI1M+t7k3qdgJnZaBcRsyRNo+hNPm++m5cGfibpNRRF6tDrrsqfIVsB762J+UR5cXZE\nDMW8lmJox3BOAQ4rc2kqdWBoLPXNwIMRcQuApFuAacBNFEX4aeX9fg6cKWkqsCnwS2nuw5hcE/eX\nEdHJ4R9mZiPOhbKZWWecA3wD2AJ4ec32LwOXRsTOkl4F5AYxVGfbCzWX59DgdTsiXpL0TeCQ+W+q\nubz4fLfNron9fBPHUhlvAkUP+PrDpPPscHmamY0WHnphZtYZM4HDh3pka7wMuL+8vFfN9ieBJWuu\nXwLsP3RF0tItHLu2wP4JsDXzFusPSVpT0gRgZ1o/OW8CsEt5eXfg8oh4CrhL0n+W+apmqIaZ2Zjg\nQtnMrJoAiIj7IuK7NduGitGjga9Jug6YWLP9MuD1QyfzAV8BlpE0S9INQKqNXyfu/DkM5fECcCzz\nFsqHUJzs9wf+XbTPk3+dy7WeATaSNKvM60vl9vcD+5T53gy8q4lYZmajhjyEzMzMzMxsQe5RNjMz\nMzOrw4WymZmZmVkdLpTNzMzMzOpwoWxmZmZmVocLZTMzMzOzOlwom5mZmZnV4ULZzMzMzKwOF8pm\nZmZmZnX8fw5wwYlCDsrBAAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# define figure size\n", "FIG_SIZE = (12, 8)\n", "\n", "# create line styles to cycle\n", "lines = [\"-\",\"--\",\"-.\",\":\"]\n", "linecycler = cycle(lines)\n", "\n", "# calculate limits\n", "max_played = max(played for _team, played, _points in top_team_summary)\n", "max_points = max(points for _team, _played, points in top_team_summary)\n", "\n", "# plot\n", "fig = plt.figure(figsize=FIG_SIZE)\n", " \n", "for team, _played, _points in top_team_summary:\n", " team_cum_points_list = team_points[team]\n", " plt.plot(range(1, len(team_cum_points_list)+1), team_cum_points_list, \n", " next(linecycler), label=team)\n", "\n", "plt.xticks(range(1, max_played+1))\n", "plt.yticks(range(0, max_points, 5))\n", "plt.xlabel('Match Number')\n", "plt.ylabel('Cumulative Points')\n", "plt.xlim(1, max_played+1)\n", "plt.ylim(0, max_points+1)\n", "plt.legend(loc='upper left')\n", "plt.title('Cumulative Points vs Match Number For Top 5 Teams \\n(at {})'.format(DATA_FILE_DATE), \n", " fontsize=16, \n", " fontweight='bold')\n", "plt.show()\n", "#fig.savefig('PvsPTS_top5_27Apr2015.png', bbox_inches='tight')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can see that Man City caught Chelsea at match 20 and then fell away. We can also clearly see that Liverpool's poor start opened up a big gap." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## What does the graph comparing the bottom team performance look like?\n", "Let's also include key showing events when a new manager took over." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Extract a summary of key data for the top 9 teams, using the (ordered) premier league table generated earlier." ] }, { "cell_type": "code", "execution_count": 45, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[['Crystal Palace' 34L 42L]\n", " ['West Brom' 34L 37L]\n", " ['Newcastle' 34L 35L]\n", " ['Aston Villa' 34L 32L]\n", " ['Hull' 33L 31L]\n", " ['Leicester' 33L 31L]\n", " ['Sunderland' 33L 30L]\n", " ['QPR' 34L 27L]\n", " ['Burnley' 34L 26L]]\n" ] } ], "source": [ "BTM_TEAM_COUNT = 9 # covering the most interesting teams\n", "btm_team_summary = PLtable[['Team', 'P', 'PTS']][-BTM_TEAM_COUNT:].values \n", "print btm_team_summary" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Create function to return the match number for given key event date. This is used to determine the first match number after a new manager arrives." ] }, { "cell_type": "code", "execution_count": 46, "metadata": { "collapsed": true }, "outputs": [], "source": [ "def key_event(event_date, team):\n", " \"\"\"Return match number on or after given event_date and team.\n", " \n", " input: string event_time - date in form 'mm/dd/yy'\n", " string team - team name e.g. 'Liverpool'\n", " output: integer match_number, starting at 0 (none if no natch)\n", " \"\"\"\n", " DATE_FORMAT = '%d/%m/%y'\n", " # create data frame for given team and reset index (starting at 0)\n", " df_team = df[(df['HomeTeam'] == team) | (df['AwayTeam'] == team)].reset_index(drop=True)\n", " \n", " # convert Date column to datetime (uses numpy datetime64)\n", " df_team['Date'] = pd.to_datetime(df_team['Date'], format=DATE_FORMAT)\n", " \n", " # convert event date to numpy datetime64, for comparison\n", " event_date = np.datetime64(datetime.strptime(event_date, DATE_FORMAT))\n", " \n", " # find match\n", " for match_date in df_team['Date'].values:\n", " if match_date >= event_date:\n", " # match found, return match number (the index)\n", " return int(df_team[df_team['Date'] == match_date].index.tolist()[0])\n", " \n", " # match not found\n", " return None" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Create event data structure using a dictionary with key of team and value of tuple with match number and event description. This is used for plotting annotations." ] }, { "cell_type": "code", "execution_count": 47, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "{'Aston Villa': (25, 'Tim Sherwood takes over'),\n", " 'Crystal Palace': (19, 'Alan Pardew takes over'),\n", " 'Newcastle': (22, 'John Carver takes over'),\n", " 'Sunderland': (29, 'Dick Advocaat takes over'),\n", " 'West Brom': (20, 'Tony Pulis takes over')}" ] }, "execution_count": 47, "metadata": {}, "output_type": "execute_result" } ], "source": [ "key_event_dict = {}\n", "# use key_event() function to determine match at which event took place\n", "# dates given are from wikipedia\n", "key_event_dict['Crystal Palace'] = (key_event('01/01/15', 'Crystal Palace'), 'Alan Pardew takes over')\n", "key_event_dict['West Brom'] = (key_event('02/01/15', 'West Brom'), 'Tony Pulis takes over')\n", "key_event_dict['Newcastle'] = (key_event('26/01/15', 'West Brom'), 'John Carver takes over')\n", "key_event_dict['Aston Villa'] = (key_event('14/02/15', 'Aston Villa'), 'Tim Sherwood takes over')\n", "key_event_dict['Sunderland'] = (key_event('17/03/15', 'Aston Villa'), 'Dick Advocaat takes over')\n", "key_event_dict" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now plot the graph of bottom team performance with annotated key events." ] }, { "cell_type": "code", "execution_count": 50, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuYAAAIMCAYAAAC5Rmm3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3XmcFMXdx/HPFwTxQEBQvMV4n0HjlUSRaGK8jyeXt5g8\nURPjEZPHqDECGs1lPKLGqFFUknhFTdR4H4tHNF5gEu8LUbkFAQ9A2N/zR1UzxTCzO7s7Q8/O/t6v\nV792prq6+tc9Pb011dXVMjOcc84555xz+eqWdwDOOeecc845r5g755xzzjlXF7xi7pxzzjnnXB3w\nirlzzjnnnHN1wCvmzjnnnHPO1QGvmDvnnHPOOVcHvGLeoCRtKOkSSS9J+lDSHEkvS7pC0g55x1cJ\nScMkNcfpyHaWcZKkEZJOrEX5HSVpfBJDNs2T9JqkCyX1b2e5Q5PyhncgvhFxymX/dJSkpqJ9+7US\nef5TlGeXdq6r7LHWhjKy4+GtDpRxTbItH0paJZnXK5nX7nV0lKRBSRyj8oojVeJ7WDytk3MssyU9\nK+k0ST07UPYB8TgdLqlPifkdPo6XBknflvS0pI/j/7dHJR1QwXIjKvisO3QucK4jlsk7AFd9ko4C\nLgOyk3c2WP2GwEbAKsCBOYTWXkZhG9rqJGAd4G3gohbKrYcB/bMYlgHWB04A9pG0rZl90M6yOrLv\nAM6Mf8cA13agnHpxLHBL9kbSzsDmyfxaHWuVqvbxuDxwKvCjFtaVt3qJI1VPMWWxrAhsHaeNgWHt\nLO8A4Ij4ehQwq2h+NY7jmpJ0IeH8CIX980Xgi5JOMLNLWli8+DuuonnFeZ1bqrzFvMFI2hX4I6FS\nbsDZwNrAsoST+enAjNwCbB+1nqVVS5xgzexaM+sep+uqsI6OMGA9M+sObAG8HtM/Axzf5sLMxphZ\nt7htZ1Upvkawq6QNk/ffK5pfk2OtDaqx/mLHSlqtBuV2KpKWqzCrJeeFdJqQVyyEH+t7Uji2DpHU\nq4NhGC0fb3X5nZe0BYVK+X+AtQj/27LP5zeS1ii3vJmNTD9XFm9wOKroM3+kFtvgXEu8Yt54fkHh\nZHuRmQ03s4lmtsDMXjezXwJHZ5mTS3YPp4WUSi+6BHiMpN9JmhGnyyQtJ2lPSc/Fy4vPS9qnqNyS\nl+rbcglf0khJT0iaImm+pI/iuk6T1CPmGSqpmdDyA5BeOn8r5lmiK4uksfH9zPRysaS+kubGeWOS\n9E0ljZb0XoxlqqSbJW3Z2naU2jQAM3uJcMUjs12yvgGSLpD0ukKXlzmS/ilpWNE+KtmVRYt37dhY\n0u2xjEmSrpTUO903SZFpeQ8nsVwq6c34ec+W9Iqkv0jaqOxGSvsnZf2kaN6vknlDY9r/xMvU0+I2\nT5b0iKST27h/3ybs42NiuasAWdeW8WVircqxFvOtFj+7V+OxlHVNOKLMureW9EDct2/HfdOjjdu8\nEFiO0GpeVqnvQivpaResLeLn8Un8/A+W1FPSufG4minpNkmrl1+9DpH0Qvx8X1aJrlOSdozlTJH0\nqaSJkkZJWrcoX3qMby7pPkkfAXe1Zce1sr/2i5/NzHhcTJD0x1rFYsG9wPSY1B1YqWhdwyQ9Hr/P\n8xTOERco6Q4Xj9PseBPwVvr5VngcD4rbOiFu+wdxX+xbIp5s+ZEKXWcmxfhulNRf0vaSHovH+MuS\nvl3B7tgteX29mU0ys9eBW2PassC3KihnUahlZ0hrKfxveytu60xJdylcaUvz7Sbpzvi9+DDmfUfh\nf8P6RXnTY2LnuFz2HT8p5vlRLGu2pAclbVxURrXOia4emZlPDTIBqwLNcVoIrF7BMln+h1pLB0Yk\n6VOT9WRpjwCfFqXNB9ZPyngrpr9ZtL4l0gmXarNyjkjSX4rrSKcs39Uxz9AkLd0nC7N1JOUvzMon\ntE5naQck6/x2ibw7AR8XlZ3F8jGwUwX7f3yy/DpJ+o+TuG+Paasl+ZtZctv/kCyfbv+ZSXpTkj6j\nxGd4Zcx3ZAv776GY5x9lYlkI7NXCNncHJse845J0ESrPzcDrMW2HonLT9TxVwf5Nt/e0+Hca4Z/3\nT+L754Drk3xDanCsbZBsc/G2XF3ieJgTp+LP56cVbPM1Sf7s9cfAGkCvZF6p79qi47uV9LeScqay\n5LY3sWTs6blkUJL+Xom8zYTWyyz/N4EFZY6D6cBGZT7zackyD7Wy3xbF30q+01h8W9O43wc2rXYs\nhO/GV5J1vVOU//IWYnoLGFhUbqnjsKXvfHYcb0bhvFFqXaeWOX+X+n/xNOEYL/7ch7Syb05N8p6W\npF+QpP+pte9Jme9LeoxvXPSZpdu6APhmiZhKnScmA/3LHBPF351m4KES++oVoHs1zok+1f/kLeaN\nZVDyeraZTarhuuYRTtKfS9J2AkYD/YDfxbRlCP9UM+VaJ9pyCf80Qr/gvoQK1obAuDjvCEn9zKzJ\nzLpRuLw53gqXJz/Twrr/TPgxIeCQJD17PQe4Ob6+klDRGU/YD73i32nx9aVt2KZFcUjalNiqG/0r\n/j2bQmvWKKA/MJhQmQU4WtLn27C+fxEq+zsSPk+AQ2FRN5/0/NCU7L9dY9qQ+PcWoE+ctiL0Z363\n3ErNbCHhOAHYKm4vhD6ia8fX18S/OxH2iwGfJ3zeawP7EirTlTLCD4l3gJWBgyjs48vKLUT1jrXf\nEX44Q2jZ2wDoDexC+EdcbAXCcdYf2D9JP7yCbc0Y8FfC5f5ehG5srXVPqPT7mb6/i7B/zk/m7QTs\nQzi+suNzaJlW89UJrbgrEY+/6BeSuktanvAZdQOeBTaN27Mb4bu6MvCbMnGPJ3x+y7Nkt6VylLRo\nZtPYOGNdIOsaNpPw+fUFfh7T+gEXVjsWQsXrXsK+nQ4clmT4IvDdZB2DCcdNdlPtopjjcZp13VjU\nfS5O11ZwHF8Ut5e4zX3jPsjugTlLUvYdTvUifL/XI5xDIZwrHwcGEBoiMofSsnHJ64MlrSFpA+B/\nkvR23TRf5KJYzgfAlwhXnjYGXiYci5eocAXrPsL5cDVCN9L+wLlx3qokn1eRcYTtPylJG0q4F2Zl\n4KmYtiGwfXxdrXOiq1d5/zLwqXoT4Yub/cKeWeEyS7RmlUtn8Rbzs5L0KRR+va8X076S5L0syTue\nota6cumUbzH/MnAPobVhAYu3NiwEtm9tfa2Uf2NM+4hww9VqyXquiHk2ZMmWjlLTwFb2//hWln8D\n6BvzTky2sW9SxglJ/rNj2tAkrVSL+UJgsyT96SR91daOjzjveQotnr+O+3MbQBUcd5uWiPnS+P5T\nYK2YdmCS7zbCP/B9gAEVHt/p9m4OnEGhlbWZ8E93BeCGZD1pi3mHjzXCP/RP47xZwHIVHA/zgd5J\nehbvxxVs8zVJjHsn+/ATwg+C5uI4Kf9dKJeexbkQ+ExM2zPJOybJ++ck7w4xbVCS99Gi+B9L8m/D\n4ueSctPHyfJNSfqObTh/Fn+26fRczHN0kueCZNluFFo/5wE9axRL9v5KoFvMe26S98SkjD5J+jtJ\n+jVJOeu2cAyWOo6z78C0bP1x3vnJur5T4ti5Lsn7ZJL+pZi2UZJ2dyv7RsCjrRwPt7dhX1+TLHdE\niW1tacqO59WA3xPuDZpbIt+lZY7PXUucD8cnec9J0r9VjXOiT/U/eYt5YxmfvF6phT6dLZJUyWg9\n6bo+SV5nrWOfJmnLVlBeRSMExdahe4HdCS0NWcvBoiyE1pmOuDr+7UU4CX6L8I/XgKvivFWLlrEy\n08ptWG+2zHzgTeBiwok/a43Khr370BYfpSW9KW0VKvdK8vrj5HWl+++7sYzVCf8crgaeAd6U9NmW\nFrTQjz67EnCQpO7AN+L7B8zs3ZjvNsI/vHmEluNfA7cDkyRdXGGci1ZLuDF6AYXPZbSZfVQqcxWP\ntZUJ3XcAJpjZJy1ljqaY2ZzkfRZjJd+llMV9OC4u+9M2Lt+Wc0Gp8wAUzgWidPzFN1W+k7weQGXf\ntWW15A2VBoxtKfAyzJa88XObOC/9fk1IFmgm/HCGsM+Kv/cdjoXQ8v0IYT9+Bzi4lZhmAbNL5CmO\nq1IrU7gvbWLc5iXWW2Zd45PXc5N1Z8fJ/GR+i8e4mRnhR+DvCPt8LvAC8Icy8bRHuq1ZrCXP75K6\nAQ8SWrk/A/RI5mdK3exrlP7upLEv8X+0yudEV4e8Yt5AzGwqhUtfAv6vVL5YCcpkX/y0glHc1aOU\nBWViaC6Vnsi6TCxan6QVgIEVrBNC5S27jP5LQqtid0LLQcmQKiw3dR+hFViEf37ZP8CXzCyrTE5J\n8t9f4h95Nr1U4TrTy8q9zGwDMzvRzKYnebJ1riipb5K+TvJ6aoXrw0KXknT9bWJmT5nZpoTWzz0I\n/Sw/JFQgflVBEdml9vWBnxEqYVD4YZSt5weELgLbEy5z302o6B4nacc2xjwZ+Hv2lpa7sVTrWJtB\n4fuyTokKZCmfFr1vz3GcGh7/lrukPi95ncbX6rmgzHe+5PmhjHWL3qfH8zQW/65d2cJ3bYkfPGY2\nrzitg9JYFsUdz6lrxrcLKDHyVUdjiT9Wb06Ssm5r5WLqS+EG0fS80Nqx1NJxnJ0z1ogV0kxr56By\nx0NbjpNFzOxDMzvJzNYys+XNbEtC16LMg+0pN5Fu66stHHN3E7rvZd3x/gsMiueJ/UuUW6zU9re6\nT6p5TnT1xyvmjeenhEtcACfEO+FXl9RD4aFDpxMug2bGx79bSVon9pk7u4bxZetbTdJ28eQ+kkKL\nYmvSCstHQLOkvYG9yuTPKrYD1MIQWqnYIpP1w/wKhb59o5I8rwOvxrdflnSiwsgtvSR9VtKZkm6o\nZH1t8I/4V8B5cX1bAD/MwkryVEtWwRhU9GMASecojLpjwMOESkPWkl+qn2mx6ym0FJ0e/74P/C1Z\nxy4KI7dsBLxGqBQ/mZRRyXqK/ZZQOf+dmb3YQr6qHGuxwnh/fNsbuFbSepJWUBhppC39xtvFzO4g\n9M8u9z0bn7zeW8GGhJbZqoVRJv0LCqOyrCjpEAoVzqnAv4EnKFS6jlQY9WXFuP92kPQbhXGtl4Z7\nKVScDpf0RUkrEcb7z/o1P2Rm80su3QGx73Y62kj2XUu/8ycojJLTFzgvSU/zZMepgMGSiu8faOk4\nfii+7Q8Ml9Q7XlkaFtPnExo2akrSQZK2icfAKpK+T6Gf+qsUfny3S9zWrHK/kcKISKsojDa0iaST\nJWXz04r0fOCjeC/C6dRAjc6Jrp7k3ZfGp+pPhH+m8yjfL+7WJO+ZSfqnhC4NHyZpDyd5RyTppfqb\nNidpQ5O86agTRxTF8mFcb9Yv760k77Di9RFuICq+i38B4QSVvU/7CF9cYvtHlSs/WW79omXmAasU\n5RlCYVSWUlOLIy8U7zuSUVnK5B3I4qNhFE+/L7P/S47KUlR2U6k4gDtKrGd4nPd6C7GcX+Gxel3R\nchcVzT+shXXMppWRh4q2a7MW8i3Rx7zKx9r6wKQy21FqVJa3iuJbdJxUsE+vScreK0nfq2i96Xet\nO6FCk34vF7L4uaAt3/lRZeLJ9u2gJO09ltwnC4FhSRkH0XKf33QfLvrM23jebK5kORYfFaR4mg5s\nUs1YykxzWHwkmj+0kPdNFr9v5Gsl8qTHQkvH8aaEH8+l1rMQOCUpZ1gyr9zIUOuUOB4qOW/eUyaG\nqcBn27iv0+MzPcY3oXBvR8n9GvMtQ+hKUzz/leL9V+H2P5zkHVEcGx08J/pU/5O3mDcgM7uKcHnt\n94STQ1bZfpnQx/aXSfZfEoaZyvrqjSFUSGDJfnJW9JeifJXkHU1o1R9PaC0dS7jBblIlZZjZ44RL\ndy9T6Fv4DcINY6XWN4JQ6ZpaIs5yMWJmbxD6c2b57zKzaUV5HiGMLHAdoV/sfMI/recJ/R8raTEp\nte9KZzSbAmxLGPXhDcKPhTmEFsWjzOz7ReWW2ra2ph9PGHVjRolYLya0Kr0XY/mEcCn3TMp0oyrh\n6mS9RlE3FkKf9VHAi4QWwgWE7g13AEOt9ZGHyn7GreWr5rEWj6fBhJEeXovlfUhoEW4qiqPU8VDx\ncVJqW2IMdxH69S9RloVuTfsSWkTnEG5SPR84pcy2tvadr+RckKXdSxj16CXCcfQq4Xi+JonvBsJo\nFLcQhp/7lLCfnyacw35b4fpa0+o+tvAsiP0J++qDGMs7hPtPtjGzl6scSzZl67ke+IKZvbooo9mx\nwFGEc8EcwrnoDcK5YlsL3RyzvLcQbiqcQPg+FW/zCMofxy8Rbsi9Ki7/adwHDxGGl/11Bdve2vFQ\nibsI9018ELd1POH/3WAze77CMlqMJ36Ogwnd3bLz7QeE88AfCX3KMbMFwH6E7iSzCeeni1j8yaRt\n+T60lrej50RX52TWnvOFc84555xzrpq8xdw555xzzrk64BVz55xzzjnn6oBXzJ1zzjnnnKsDXjF3\nzjnnnHOuDnjF3DnnnHPOuTrgFXPnOkjSZZKaJTW1c/nBkkbEh0G1+Cj7ZJlBkn4t6QlJEyXNk/S2\npL9J2q4o77AYX0vTkUX5R0gaXmK9Q5Nllpjfjm3fvSiOGztaZgXrLLkNkpqy9ArK2EfSdZJekjRT\n0oeS/iPpF5L6lci/vKSRkl6Nn9U0STdJ2rRE3p9Jul/SB0mcNxfni3lb+kx/0dZ9Uw2SrkliWDdJ\nPykeVyeWWCY9Ro8snl/heqt6bHZFktZNzkW7tLOMDSUtiJ/D9q0v4ZxLLZN3AM51ZrFi9V3COLM/\nb2cxgwljf0N4gFAl4/DuSOFJd9mYp2vHaR9JXzOzvxfNLx4bNX3i3+zk9TDCw5MgPJW1nGqMtVr8\n1Mt9JfU2szlVKLuc9oy3XewHwO5FeTeP0zclbW1mswEkLUN4IMpOSfn9ga8De0gaambPJeX8iMKj\n1ItjK6cjY0JXW/H425mTCI9uf5swznNLy1YjBtd261E4FxnhuRZtYmavxR/YBxOePjqklUWccwlv\nMXeuY04mfI/Gm9kDVSiv+PHY5RjwOOEpfn2B1QkPHiHGc9aijGbXmlk3M+ueTUAfwoNIIDycIn1k\nd7qOmpG0PHBAUXIvQoW1FuvrLqmHmY1J9sdZaZY2FDcXuJTwgKnlCI+RfzfOW4/FH2X/PQqV8j8B\nAwjbuBBYEbi8qOxRhB97x7YhnqPSzzdONXkkeGvMLItlGTObUCpLC4u35TNwtdWRz+LK+HcnSTtU\nIxjnugqvmDvXTpJWJDyxEODGonmDJd0q6XVJsyV9KmmypFskfS7J18TiT7scVeEl/X+Y2c5mdpuZ\nzYlP9jshmb9hK+EfBfSOr68ws/mxe0wzhRYupV0jSu8C/SB2z/hY0lhJe7Sy3tSBwArx9SgKFbbi\nVvTFuplI2ljS3bH7yPuSLpfUO8mbdmkYKemnksYTntz3hSp1eTjMzI43s3FmNt/MnmLxVuB0/w+L\nfw041cxmmtmtFFojPydpiyyzmf0wPr33VSpXtQqtpH6SFsb98+ckfXRMm56kHZrsy4Nj2mJdWbL9\nTWgtBxiUzH+rRAg9YleKt+Nn/LiKume1YVt6SXogWd/JMb2bpOMkPR3X8Ymkf0v6kaTuMc9ASXPj\ncncVlbtPUuZPKojhDIWuTh9L+kjSU5KOSvJ8NinvsqLlv5fMG5akHyLpEUmzFLpHvSrp55KWK1o+\nW/ZhSXvFbf4knpv+L8l3DeEJnpnhxd+TGOetkt6L63xf0nOS/qBwZSgzhvCEVgg/Mp1zlTIzn3zy\nqR0TsAfQHKc9i+YdFNMXFk3NhEexbxLzPZyU0Vy0zBFtjGetpIxXWsgn4JWYbx6wekwf1EIsC2Oe\nocm8SUme5qS89SqM955k+S0JjxPP3q9ZlLcpWcfUEut9iMKTjNMYpxVtx5Ci+WeWWMfCdh4PZyTl\nnhbTehIeW94MzCzKf1GS/6gS5aVx3lRmndn86YRHk88hXElp07FTotxnY7lvJ2njk325WUy7PEkb\nGNOuSdLWAXZp4bh6My4zrJXjajqwUisxL/a5AssSHpPeTHhs+fdivm7A30rEkr2/PSlzVLL8oCT9\nzxSO94EtxLQ88GQL67q4xD6fBiyTpD8S02cBy8e0i1so8ymgV4lj5IMSeZuBQ4u2tdTndCbhytC0\nMutdmMWWrPfmOO/djhyLPvnU1SZvMXeu/bJWPAP+XTTvWeCrwBqE7hl9gOPivOWBYwDM7EuE1uvM\nMCt0RbiujfGcm7z+Qwv5vkqhRfevZjYpxjLezLpRaMk1W7z7S7F+hFbvfoSKCkAP4FutBSppIPDl\n+PY1M/sPcEs2Gzi0hcUfBVYBNiX8wIBQKdunRN7+wPGE/b828N/WYmsPSasT+p0DfARkn11/INt3\ns4oWS/v1r9KB1RvhM+hOuALxeeAaSb/uQJlZy+laktaRtDahkt1M+HyyrjnZ1ZWXzWxKyeBi1yEg\n69YyPjmuPlNikZ6EyvwAwo8lgJWBvdoQf0/gr4RjfSFwtJllLdHfBPaLr88lfEZ9KFzx2EfSgfH1\nhfFvN+BogNginS3/j3LbHZ0AZDdAHke4d2BV4KYsTdLW8XV25aw/4f4F4n7P9vXNZvaxpB0pnEtG\nEbqxrQCcEtO2JXSfKrZS3N5+FI5ViFeozOwo4EtJ+ojkczoL2CzGBvB/hPPaqjG+XxD2cyq7V2Z1\nSWuWiMc5V4JXzJ1rv9WS19OL5k0hVDwfJlTIZgO/T+ZvlLzuUDcEBRcDh8Wk28zsghYWOT7+NeB3\npYqscNW3m9nfLdzkeEOSvk65BRIHUzj/3FL0FwrbUsyAU8zsfTN7FfhtMu8rJfLfZ2aXmtmHZjbR\nzGZUEFubxMrTQ4RKykLgSDN7r5JFq7D6XxIq4v0IlaaTKHQJOjnG1h5ZxTyrhGcV8Nvi350lDQA2\nLspfDVeZ2aNmNpPFj4lKjqvMScDeFD6PtLvYvsnr04EZhO9nOlrM7gBm9jyFHwffjt019qHQBeuP\nrcSRrutSwhWNaYQfB4uti/Djdl58fXD8e1D8axQq7mmZRxG6jHwMpD/EdmdJkwlXiGYD1ybp6X5t\n6ZicQLhykMV3GrAbMMXMzjCzeUX5309eD2yhXOdcwivmztXGTYRWpU0Il9SLR5tYrtRCbSWpB+GG\nwqwF7TZaaLGWtAGwZ3z7tJn9qwOrfyV5/XHyulcFy6b9yF+UtDmhsvNmTNtC5YeOTG8ofCd53b84\nIzC2gljaTdImhK4jGxO6rBxuof945n0KlZm+RYunI69Mbc/6zex0M3vKzGab2Qdm9jvgwTi7G4XW\n2rZ6hEIL6M5xAriEUIkdwuKjbVSzYt6R4yqzfPw7k9C1I7Vq8rp4BJlsWjnJk7WarwL8D4XK8ruE\nrjItqXhdZvYBoYsNwH6xZT67h+UVM/tnO+PPvGFm2TmozfvVzKYRbkieDmwDjCD8IH8t9nXv3cLi\nflOvcxXyirlz7Tc5eb2oK4LCONa7J3k2j11BylU02zX6icKoJn8ntF4ZofXu62a2oIXFjk9etzRk\nXSU+bccy2RCTWydJo4H/ELoDpV0byrWar5u8Tlv7iq9aAHxSaVxtJWlbQreatQjdV/Y3s/TKAWY2\nn0I3p95Fl/Q3z7IBT7dj/ZVUdlodk70UM/uIQkxZxXwu4T6AJwjdgo5I1tFUSbEVrr5dx1WRxwg/\niAYADxRdOci6nhiwky05mk13M0t/3N4BvEGoXJ5M4YftqKSiW066rrXKrCu9eXRU/Lsi8BMK54xR\nSZ6068xhZcosNRLKov3aQtwtbk+88rAasAXhR0p2xW0nCo0DmQHJ68k45yriFXPn2i+ruAjYKklf\nQOEf3EJgjqRVKT/OeVqh3DIbFaIlkvoC9xFuQAU418yObqmioDCKzLD4diKFfq5l42mh1boj0gp3\nqda+zCEsScAvJfWXtBFhzO/M/dUOtBxJuxJaifsT9tduZnZPmezXZIsRYl9Z0tcJ/agBnjWzF5Ky\n+8VuIn2SMpaN2zwgGf3iOEnXShoiaQVJfSWdQOheAKEi9mRS7niVH2GnlKwVfFPClZ+n4w+NR2J6\n1qXi37HbSWuy42qApDUqjKG9HgC+TTie1iFUzrOW5jvjXwG/k7SVpJ5xFJYDJd0padHVgPiduji+\n3Z7QwtwMXFVBHOm6rpa0gaQektZSGNHmMSUPYSIcw+/E/Nlwlwso3LOQlgnwc0lfkLRsPK72kPQX\nSS3do9GS9Fy0maSe2Zt4/J1H6Do1nTDE6u1J/uJuU9m5Y0qFXbucc3jF3LmOeIzCJeHsUj8WHo6T\ndSdYk/CPdjKhgpNJWzvHEkbUgFDR/DRWoFrqU3sA8IXk/U+Toc2ayyw/jMIQiZeZWfHNWpmsMidg\nbDbUWguxVCy28maVBiOM7rFYax+QXbJfXdKXSxSzI6Gf7ssU+uo/bGZ3lsjbrjAryHMmoVUTQsvg\nk0X7Pt1flxFa1iFs+3TCj6JuhD7HxxSVPZbQteW2JG1fwjZPBb4Y07oTugQ1xXJmELpdiLBvh2c3\n9haptBU67WeuZBseSdLTfMWK92N2XK0IvBv30yhqxMz+ROhOBuFm5/vjD9obKYzb/zlgHOFqwCRC\nn/Y9WdLVLH6z7oNWeoz2YhcBz8TXXyUMgTmP0B1rNKGSu+jziD8Csv7f2Q+wu9MbTM3sScIxBWHM\n/McIV4amA3cRutq0+uM+kX5Or1OonH8TyIaLHELofncy4TiYHLcje3aDAfcuKlDqRuE4va8NsTjX\n5XnF3Ll2MrMPKYxfXtyv+7A4bwZhmLLRSZ7FWobNbCKhW8CLhApCJU8/tORvuWmRWCE+LqbPZcmH\n2qQuJYzqkg1bVyqecu9bi3tnQgumAc+Y2Ssl8mStg8aSo7NYLOMfhGEnZxIeZrJ/G2IpN7/Sbcjy\ntDaFjOEH0B7A2YSKzzxC5edmYAczK+4HX2nZdxBGw3iK0L3h01juPcA+ZvbLrMB4FaZffPtsBdsH\noe/8vGSdj8X0Z1j8OC2umJfbjyMIfZKnltiWan0mi+U1s/Mp3BS5BaFP+AqE4+UHwL8IP2rmEoaD\nvDumL/aZxO962kJ+JRUws08IffHPIIxS8hHhx/wbwK2EVv3iH0+jWHz/XF00HzM7jnCOGUM4v8wn\nNAA8RPhfGogeAAAgAElEQVQxUnz1pqXvQrq/5hHOU8/GWNP5M4ALCMfbNEJL/mzCcXKImaWt57sQ\nbvg0KtxXzrkgG/fXOdcOsb/0vwktVF81s6XWnaIrUXgQ0xBCo2JbWgMdIOmLhJbOD4FtzOz1nEPq\ndCT9hdAaPQlYt5V7Obq0ZF/908x2ai2/c67AW8yd6wAze4lw06VR6BPqasdHd2ifPQnH6IleKW8b\nSQ9KmkyoaBrhfg6vlJchaUPgG4R9dXLO4TjX6XiLuXOu7sU+27vgLeZuKZP0FqH71STgiviwHeec\nqwmvmDvnnHPOOVcHlmk9Sz4k+S8G55xzzjlXc2ZWH10lzawuJ+LIUVUqa0SVY+sy5dVzbL6t9VFW\nvZdXz7HVe3n1HJtva32U1dXKq+fYfFs7VJZVM7aOTH7zp3POOeecc3XAK+bOOeecc87Vga5SMW/y\n8uqirHovr5pl1Xt51Syr3surZlldrbxqllXv5VWzrHovr5pldbXyqllWvZdXzbI6Q3l1oW5HZZFk\nVi8d8Z1zzjnnXEOqpzpn3Y7KUo6P1uLq5cvjnHPOOVdNna5iDl4x68r8h5lzzjnnGlVX6WPunHPO\nOedcXfOKuXPOOeecc3WgU3Zlcc4555xztSGxEfBboEfesbTgGDPezjuIavOKeRchqQkYbWZXVbHM\nYcB3zGznapXpnHPOufxIdAOuAh4BHs05nJbMyDuAWvCKeZVJOgQ4GdgYmAOMA84xs8ervJ5mYAMz\ne7PCRSxOpcoaAfwUmAssAF4EfmRmT1YhVOecc851Hv8LdAd+ZkZz3sF0Nd7HvIoknQxcAPwcWBVY\nG7gU2K9M/u4dXWUHl88YcL2Z9QZWAR4Dbq1S2c4555zrBCRWJ9RhjvZKeT68Yl4lkvoAI4Hvm9nf\nzOwTM1toZv8ws5/EPCMk/VXSaEmzgFMlfSRp5aScbSRNldRd0gaSxkj6QNI0SdfHPI/E7M9LmiPp\nG5L6SrozLjtD0h2S1qw0/DhhZguA64DVJPWXdKqk1yXNlvSCpANa2AcXSZogaZakZyTtlMzrJun0\npKxnJK0V520i6X5J70t6WdI3Kt3vzjnnnKuai4ArzPhv3oF0VV4xr57PA72A21rJtx9ws5n1IdxY\n0QR8M5l/OKH1eiFwNnCPmfUF1gQuBjCzITHvVmbW28xuhkV9wtaJ0yfAJW3dCEnLAsOACWb2PvA6\nsJOZrUT44fEnSQPLLP4U8FmgH/AX4GZJPeO8HwEHAXvGso4CPpa0AnA/8CdCa/1BwO8lbdrW2J1z\nzjnXPhL7AFsD5+QdS1fmFfPq6Q9MN7PWLv3808xuBzCzuYTW6cNgUdeWg4DRMe98YJCkNc1svpn9\ns1yhZjbDzG4zs7lm9iFwLrBLG+L/pqSZwATCF/PAWO5fzWxyfH0T8BqwQ5kY/mxmM82s2czOB5Yl\n9LWH0Gftp2b2Wsz7HzObAewDvGVm18blxhG60XiruXPOObcUSKxI6Hp7jBmf5B1PV9ZwFXMJq8bU\njlW/DwyQ1No+fbfo/d+BzSQNAr4CzDKzZ+K8UwhdTJ6S9F9JR5Xfbi0v6XJJ42M3mTFAH0mV9kO/\n0cz6mdlAM/uymY2N5R4haaykmbHivgXhR0ipGH4s6cXY9WYm0AcYEGevBbxRYrF1gR2y8uNyhwDl\nWuWdc845V11nAw+Z8VDegXR1DTcqi1nVbohsqyeAeYSW5lvK5FliZBQzmyvpZkKr+SaEFvRs3hTg\naABJXwQekDSmzEgsPwI2ArY3s6mSBgPPESr2rf3QMErcSCppXeAKYFfgCTMzSWPL5N0Z+D9gVzN7\nIabNSPK+A2xAGPElNQEYY2a7txKjc84556pMYjvgYGDzvGNxDdhinhczmwWcCVwqaf/Ygt1D0p6S\nfhWzlfvRcB2hz/V+FLqxEG/qXCu+/YBQgc66ykwB1k/KWJHQr3xWvJl0eIn1lFt/ufQV4jqnA91i\ni/0WZfL2Jgy1OF1ST0lnAisl8/8InB1vaJWkrWKcdwIbSTos7q8ekraTtEmZ9TjnnHOuCiR6AFcC\nPzLj/bzjcV4xr6rYr/pk4AxgKqE1+PsUbggtOZZ4HOO8GXjWzN5JZm0LPClpDqHLywlmNj7OGwFc\nG7t/fB24EFiOUIn+J3B3iXWVazkvF9eLhBtUnwAmEyrlj5VZ7p44vQqMJ/xImJDkPR+4CbgPmEU4\nEfSK/eF3J/Stfw+YBPwC6IlzzjnnaukkQkPfX/IOxAUya0936tqTZGZWqstEyfTOTtIDwF/M7Oq8\nY6lnjfr5O+ecc0uTxGcIo6ltb0alDytsSPVUt2i4PuadkaTtgG2A/fOOxTnnnHONTULAZcCvunql\nvN54V5acSbqWMI73SWb2Ud7xOOecc67hHUJ4QvkFeQfiFuddWVyn4p+/c845134S/YH/AvuZ8XTe\n8dSDeqpbeIu5c84551zX8RvgRq+U1yfvY+6cc8451wVI7ArsRvmhj13Oat5iLql7fHLkHfH9ypLu\nl/SqpPsk9a11DM4555xzXZnEcsAfgB+YMSfveFxpS6Mry4mEpz1mndlPBe43s42AB+N755xzzjlX\nOz8FxplxR96BuPJqWjGPT63ci/DUx6xT/X7AtfH1tcABtYzBOeecc64rk9gCOJrQWOrqWK1bzC8A\n/o/CY+QBBprZlPh6CjCwxjE455xzznVJEt0IT9s+w4xJecfjWlazirmkfYCpZjaWQmv5YiyM1Vif\n4zW2kaTTJN1VlPZambRvdmA9zZI+08L8YZIWSpoTpzckHdve9TnnnHOuUzsWWEDoveDqXC1HZfkC\nsJ+kvYBewEqSRgNTJK1mZpMlrQ5MLVeApBHJ2yYza6phvB01BviJssEww7YtAwyW1M3MmmPa+sAj\nHVxXa2NtPm5mQwAkDQYekfSkmY1boiBpGTNb0MF4nHPOOVdnJNYERgJDzBbrvdClSRoKDM05jJJq\n1mJuZqeb2dpmth5wEPCQmR0O3A4cGbMdCfythTJGJFNTrWKtkmeAHsDg+H5n4GHg1aK0N+KPkj6S\nrpI0UdK7ks6W1A1A0gaSxkj6QNI0SdfH9KxC/3xsDf9GmVgWVdxjZfwlYJNYxqDY6v5tSW8DDyg4\nQ9J4SVMkXStppaL8wyRNkPS+pGMlbSfp35JmSrq4SvvQOeecc9VzMfB7M17KO5B6YmZNaR0z73hS\nS3Mc86zLyi+BmyR9BxgPtLtbRz0xs/mS/gXsAowFhgCPAhPj6+fi3zFxkWuAyYQW9BWBO4F3gCuA\ns4F7zGwXST2BbeM6hkhqBrYyszcriUvS9sBGhB8OqSGEyroBRxF+JA0FpgHXAZcARyT5twc2iNt3\nJ3AXsCvQExgr6WYz6+iVAOecc85VgcQBwGbAIXnH4iq3VJ78aWZjzGy/+HqGmX3ZzDYys93N7IOl\nEcNSMoZQ4QXYidBl5dEkbWdgjKSBwJ7AD83sEzObBlxIuLIAMB8YJGlNM5tvZv9sYxw7xpbs2cCT\nwHVm9npRnhFx3XOBQ4Hfmtl4M/sIOA04KGvBj86OsdwPzAH+YmbTzWxi3Mat2xijc84552pAYiVC\na/kxZszNOx5XuYZ78qdGagQwvMSskTZ8ycsVZfKXzFuBR4DjJPUDVjGzNyRNA66NaZvHPOsSur1M\nkhb1OukGTIivTyG0mj8laSah0jyqDXE8aWY7A0haFbhe0rlmdnqS553k9erA28n7CYRjIx0xZ0ry\n+pMS71dsQ3zOOeecq51zgHvMFl2ld51Ew1XMY4V6RK3yt+JJoA/wXeBxADObLWkiYfzQiWb2tqT5\nwDygv5ktcTNGHE7yaABJXyT0Ax9TafeVorKmSrqVcFd2WjFPR8OZCAxK3q9DuIN7Snxd0araGptz\nzjnnqktiR+BrhMZA18ksla4sXYWZfULoy30yi4+88lhMGxPzTQLuA86X1FtSN0nrS8pGUvlGfDgT\nwAeESm9WgZ9C6JdeEUn9gQOB/7aQ7Xrgh/FGzxWBc4EbSv1oaGlVbcjrnHPOuSqT6EEYs/yHZszM\nOx7Xdl4xr74xwCqEynjmUWAAi1fWjyDcOPkiMAO4GVgtztsWeFLSHODvwAlmNj7OG0HoGjNT0tdL\nrN+Az2fjmMfypwDHF+VJXQ2MjvG9CXzcSv5SvMXcOeecy9ePCd1Rb8o7ENc+Cs/4qT/ZcOCVpruu\nwT9/55xzbkkSGwJPAJ8zW+y+MdeKeqpbeIu5c84551wnJiHgMuAcr5R3bl4xd84555zr3I4A+hGG\nSHSdWMONyuKcc84511VIrAL8CtjLjAV5x+M6xlvMnXPOOec6r98CfzLjubwDcR3nLebOOeecc52Q\nxFcITxXfIu9YXHV4i7lzzjnnXCcjsTzwB+D7ZnyUdzyuOrxi7pxzzjnX+ZwJ/MuMu/MOxFWPd2Vx\nzjnnnOtEJD4LfBvYMu9YXHV5i7lbjKRrJJ2ddxzOOeecW5JEd+BK4DQzpuQdj6sur5hXiaTxkqZI\nWj5J+19JD+cZV0skDZP0aFGyxck555xz9ec44GPg6rwDcdXnFfPq6gacmHcQVVAXj6V1zjnnXIHE\nOoS+5ceYeSNaI/KKefUYcB7wY0l9imdK2kTS/ZLel/SypG/E9PUkzUzyXSlpSvJ+tKQT4+uVJY2S\n9J6kGZJui+n9JN0paWpMv0PSmkkZwyS9IWm2pDclHSJpE8Ld3J+XNEfSjCx7Udz7SBonaaakxyV5\nfzbnnHNuKZMQcClwkRmv5B2Pqw2vmFfXM0AT8OM0MXZvuR/4E7AKcBDwe0mbmNlbwGxJW8fsQ4A5\nseKcvW+Kr0cDvYDNgFWB87NVAFcB68TpE+CSuO4VgIuAPcxsJeDzwDgzexk4BnjCzHqb2cqxrEW/\nwGNMVwHfBVYGLgdul9SznfvHOeecc+3zNeAzhKd8ugblFfPqMsIlpuMlDUjS9wHeMrNrzazZzMYB\ntwLfjPPHAEMlrRbL+Cuwi6T1gJXM7HlJqwN7AMea2SwzW2BmjwKY2Qwzu83M5prZh8C5wC7J+puB\nLSUtZ2ZTzOzFmF6uy0pWOT8auNzMnrbgOmAesGN7d5Bzzjnn2kaiL6GR7Wgz5ucdj6udxquYSyOQ\nLE4jyswvl15+uQqZ2QvAncCpFCq46wI7xO4gM2PXlUOAgXH+GGAo4eldj8T3uxBayx+JedYGZpjZ\nrCVD1/KSLo83oM6Ky/eRJDP7CPgWcCwwMXZ52bjCzVkX+FFR3GsBq1e8Q5xzzjnXUb8Abjfj8bwD\ncbXVeOOYm40ARrQyv+3Ltc1w4Dngt/H9O8AYM9u9TP4xwG+AdwndVh4j9P+eG+dlZawsqU+JyvmP\ngI2A7c1sqqTBcf0CzMzuA+6TtCxwDmGYpSG0PvrKBOAcMzu39U12zjnnXLVJ7ATsB2yedyyu9hqv\nxbwOmNkbwI2EEVqM0IK+kaTDJPWI03ZZP3Ize51QCT+MUIGfA0wl9CcbE/NMAu4m9E3vG8vYOa5y\nRUK/8lmSVib8MABA0qqS9o99zT8FPgIWxtlTgLUk9UjCF4UuLlcCx0raXsEKkvaWtGL19pZzzjnn\nSpFYFrgCONGMD/KOx9WeV8xr5yxgeYDY73t3wk2f7wGTCJel0psom4DpZvZe8h5Cy3fmcELl+mVC\npTobmvFCYDlgOvBPQgU+aw3vBvwwrvd9QneZ78V5DwIvAJMlTY1pi8YxN7NnCTd+XgLMAF4Djmjb\nbnDOOedcO50CvA7ckncgbumQWX0Ogxm6R9sSNyeWS3ddg3/+zjnnugKJjYHHga3NeCfveBpZPdUt\nvMXcOeecc66OSHQjdGE5yyvlXYtXzJ1zzjnn6stRhC6ql+YdiFu6Gm9UFuecc865TkpiIOE+tK+Y\nLRqswXUR3mLunHPOOVc/LgRGmfF83oG4pc8r5s4555xzdUBiT2AHYGTesdQLjVS/vGNYmnxUFtep\n+OfvnHOuEUmsAPwXOMaM+/KOJ28aqUGEHyg7ARvbcFtQs3XVUd3CW8ydc8455/I3Enisq1fKNVID\nNVIXA88C44Gta1kprzd+86dzzjnnXI4ktiE8RHCLvGPJk0bqUOB3wHXApjbcpraySMPxirlD0l3A\n9WY2WtIw4DtmtnOc1wxsYGZv5hmjc84514gklgGuBE4xY1re8eTsSUIL+YS8A8mLV8xrQFITsBWw\nmpnNryD/UGC0ma1dg1j+ACxnZkcWpX8W+FeMca9qr9c555xzFTkB+IDQStyl2XB7I+8Y8uZ9zKtM\n0iBge2AqsF+uwQTXAP8jafmi9MOBO8zsg6UfknPOOeckBgGnA8eaUZ+jcVSZRqq7RurQeHOnK+IV\n8+o7AngAGA0Ut1LvJekFSbMlvSvp5FhhvhtYQ9KcOG81SctKulDSe3G6QFLPWM7QZPkpkibGLihL\nMLMngfeAryVxdAcOJv46l9Qk6TutbZikvSWNlTRL0gRJw9uzg5xzzrmuTkLAZcD5ZryWdzy1ppGS\nRmpfYBxwHLBiziHVJa+YV98RwI3ATcBXJa2azLsKONrMVgI2Bx42s4+BPYCJZtbbzFYys8nATwkt\n75+N0/bAGUlZA4GVgDWA7wCXSupTJqbrYlyZLwM9gLvie4tTaz4EDjOzPsDewPck7V/Bcs4555xb\n3LeAtYDz8g6k1jRSnwUeB84hXCH4og23/+YbVX3yinkVSdoJWBO43cxeA14EDkmyzAc2l7SSmc0y\ns7HZoiWKOwQ4y8ymm9l0wjBKhyfzP43zF5rZ3YRK88ZlQvsTsIukNeL7I4A/m1mbHvVrZmPM7IX4\n+j/ADcAubSnDOeec6+okVgYuAI42o9V70TozjVQ34C+EngRb23C7w4bX6UN06kDDVczV1DRCTU0j\nqvW+jY4E7jOzOfH9zSzeneVrwF7A+Nh9ZMcWyloDeDt5PyGmZd43s+bk/ceUuSxkZhOAR4DDJa0I\n7E87bjKRtIOkhyVNlfQBcAzQv63lOOecc13cr4FbzHgi70BqzYZbM/BVG26X2fC2NQh2RQ03KosN\nHTqimu8rJWk54JtAN0mTYvKyQF9JW5nZv83sGeCA2Mf7eEJ3l3Uo3Y1kIjAIeCm+Xyemtde1wE+A\nycBbSWt9W/yFML7oV81svqQLgAEdiMk555zrUiR2Ab5K6NLaJdhwezfvGDqLhmsxz9EBwAJgUwr9\nwjcFHgWOkNRD0qGS+sQuJHOA7JfjFKC/pJWS8q4HzpA0QNIA4EzCZaD2uoVQuR9BGKmlPVYEZsZK\n+faE7jZ+Oco555yrgEQv4HLgeDNm5x2Pqz9eMa+eI4CrzexdM5sapynAJRT6mR8GvCVpFnA0cCiA\nmb1MqIi/KWmGpNWAnwPPAP+O0zMxLdOmCnG8yfQWQh/4P7eUtajs9PX3gbMkzQZ+RrjJ1TnnnHOV\nOQ140Yy/5R2Iq0+yOu1/L8nMbImbIsulu67BP3/nnHOdkcRmhPu9PmvGe3nHU0saqWVtuM3LO45K\n1VPdoqYt5pJ6SfqXpHGSXpT0i5g+Io7DPTZOe9QyDuecc865vEh0A64Ahjd6pTy6QSP1rbyD6Ixq\nevOnmc2V9CUz+1jSMsBjcUhBA843s/NruX7nnHPOuTrwXaA78Ie8A6k1jdSBhHvsvLtOO9R8VJbY\ntxmgJ+GgnBnf18UlA+ecc865WpFYg3CP2K5mNPRwgRqpPsDFwCGdqStLPan5zZ+SukkaRxh55OHs\nATXA8ZKel3SVpL61jsM555xzdU5aBqk30gol5nVvIb13HS93EXCFoRfrPM4OL7fKR/xmwEfcbyN4\ntubra1A1r5ibWbOZDSY8dnaIpKHAZcB6wGBgEvDbUsvGvujZNLTWsTrnnHMuJ7HLK6Fe8FCJHJu0\nkD4xTnW1nMR+hLrOz+s5zmosd8aX+MabF/K/U3/D15dSnO0maWhax6xm2R21VEdlkfQz4BMzOy9J\nGwTcYWZbFuX1UVncEvzzd865BhUevncAcCv1OmRcG0j0Bl4AhplVt2JZjzRSywMb2/B2PcAwV/VU\nt6hpxTw+GGeBmX0Qn4x5LzASeMHMJsc8PwS2M7NDipb1irlbgn/+zjnnOgOJC4E+ZhyVdyyuZfVU\nt6j1zZ+rA9dK6kboNjPazB6UdJ2kwYTRWd4CjqlxHM4555xzS4XE9sBBwOZ5x+I6F3/AUB2I/edH\nm9na8f144Dtm9mCecdWjRvz8nXPONQ6JHsDTwHlm/CnveFzr6qluUfObP7sKSeMl7VaUNkzSo+0o\nzuLknHPONS5pZaSfI9VFpahKfghMBf6cdyC1ppEN9bnVBa+YV49Xpp1zzrm2e6kRbvYEkFgfOAU4\n1qyx6wQaqVWBxzRSvfKOpZF4xXwpkdQs6TPJ+2sknZ1nTM4551yuzGZg1hAtyxIiDAf9azPezDue\npeB84HEbbnPzDqSR1PzJn11MWy7peAu7c8451zgOBVYFLsg7kFrTSH0V+AKwZWt5Xdt4xbx6BPxN\n0oIkrSeUePrV4ss455xzrhOTGACcB+xrxqd5x1NLGqkVCFcGvmfD7aO842k0DVcxb1JTVVqhh9rQ\ntlaaDdjfzBY9REDSkcD/ViMe55xzriFImwFbYHZT3qFU0XnADWY8nXcgS8Fw4AkbbvfmHUgjariK\neTsq1LUkCt1VPgaWT+atDryz1CNyzjnn8hKea3IFDTRiicSuwJeALfKOZSl5FPhX3kE0Kr/5s/ay\nHwrjgEMldZe0BzAkx5icc865PBxJqHtcnncg1SCxHGFbfmDGnLzjWRpsuN1hw21q3nE0Kq+Y11Z6\ng+eJwL7ATOAQ4LYSeZ1zzrlG9lfgYMya8w6kSs4AxplxR96BuMbgT/50nYp//s455+qBxJbAQ8BW\nZkzKOx7XfvVUt/AWc+ecc865NpDoDlwJnOGVcldNXjF3zjnnnGubY4EFhMp5Q9NIfU0j9d284+gq\nvGLunHPOudqRzkL6ct5hVIvEWsBI4GgzGqWvfEkaqX7A74D/5h1LV9FwwyU655xzrq5cDzTSKB4X\nA5ea8WLegSwFvwT+bsPtibwD6Sq8Yu6cc8652jF7Ke8QqkXiQGBT4OC8Y6k1jdTOwN7A5nnH0pX4\nqCyuU/HP3znnXB40+NpV+KTfi2z6t9+w9aj/xGSz4XbPEnlHqhvw1RLFdJb8PYCXgFNsuN1aYrmG\nUk91C28xd84555xrzeY3P8jAf3enzztfIjzpE8INoEtUbIHuwAkl0jtLfoCbWfKZK67GvMXcdSr+\n+TvnXJ2TROiH/WvMJuQdTjVIfJ51x9zBhndvZ/f/8q2843HVVU91C6+Y50TSOsALwEpWrx9CHWqU\nz9855xqWdBjwI2A7zBbkHU5HSfQEngXOMeOGvONx1VdPdQsfLrFKJI2XtFul+c1sgpn1rlWlXNII\nSaNrUbZzzjlXkrQMMBw4uhEq5dGPgXeAG/MOxDU+72NePRanhiCpu5ktzDsO55xznYjZAqTPYTY7\n71CqQWJD4GRgW7PG+R/v6pe3mNeQglMlvS5puqQbJfWL8wZJapbULb5fWdIoSe9JmiHptqScfSSN\nkzRT0uOStkzm/UTSu5JmS3pZ0q6S9gBOA74laY6ksTFvH0lXSZoYlzk7Wf+wWPb5kqYTWjycc865\ntmmcSrmAy4FzzRifcziui/CKeW2dAOwHDAFWB2YCl5bJOxroBWwGrAqcDyBpa+Aq4LvAyoSTxO2S\nekjaGDgO2NbMVgJ2B8ab2T3AucANsbvM1nEd1wDzgfWBrWP+/01i2B54I67/3I5uvHPOOddp7Xv0\n5QwdviHhyZfOLRVeMa+tY4AzzGyimX1KeITv17NW6oyk1YE9gGPNbJaZLTCzR+Pso4HLzexpC64D\n5gGfJwxztCywuaQesd/6m1mxccrWMRDYE/ihmX1iZtOAC4GDklAmmtmlZtZsZnOrvC+cc841Kql7\n3iFUk3b76SZs/Lfv0HPOT8xolL7yrhNouD7mTU0aQeluGCOHDrURFeYvmbcdBgG3SWpO0hYAA4vy\nrQ3MMLNZJcpYFzhC0vFJWg9gdTN7RNJJwAhC5fxe4GQzm1SmnB7ApDCSFRB+mKVDWb1T0VY555xz\nGWkt4Bakz2PW3Gr+zmDNp/7Bu194zu49/y95h+K6loarmMcK9Yha5W+jCcBRZvZE8QxJg5K37wAr\nS+pTonI+ATjHzEp2LTGz64HrJfUmdHP5FXAES96I+g6hpb2/lT9x+o0tzjnn2sbsXaT9G6VSrr2O\nP5VNXlmHcUfumHcsruvxrizV1VNSr2wC/gicG8csR9IqkvYrXii2cN8N/F5S39h/fEicfSVwrKTt\n482kK0jaW9KKkjaKN3suS6h0zwWykVQmA4MUm8fjOu4DzpfUW1I3Sesn63HOOdfopN5IbyBNRlry\nKqnUs4X0yWWXM5tck3iXMu36s/5sfPvZvLLvWfbQ2dPyjqfRqKnpKTU1rVut943IK+bVdRfwcTL1\nBW4H7pM0G3iCcINlJm2hPhz4FHgZmEJ8dK6ZPUu48fMSYAbwGqFFHEL/8l8A04BJwADCaCwQHqUL\n8L6kZ+LrI4CewIuxrJuB1ZJYvMXcOecamdkcYCgwGNiuRI5PW0gf3MJyjeGZY8/g+SPH2j8uPTvv\nUBqFmprS3hl7A+9V8X3D8Sd/uk7FP3/nnKtQeNjPQMwauiJTLRKDCVeWtzRjSt7xNAo1Nd0HnGJD\nh47LO5Zy6qlu4S3mzjnnXGPaCzg97yA6A4nuhK6jp3mlvOoOAZ7PO4jOwlvMXafin79zzlUoG4Kr\nXv/R1xGJE4EDgS/5Ez67nnqqW3jF3HUq/vk751wJ0o7ABMwm5h1KZyOxDvAc8EUzXsk7nkagpiYB\nZwCX2NChM/OOpzX1VLfwrizOOedcZyVtgvR3ws386+cdTmejL40Q+x/1EGs8Pcor5VXVjTDIxId5\nB9LZNNw45s4551wXshowBvgW/sTmtltx8q9Z57G1mLX2iEYebGZps6FDFwKX5h1HZ+Qt5s4551xn\nZfk4RZQAACAASURBVNaE2fleKW877XbGumz895N5ba+T7OGzPso7nkahpqYN8o6hM/OKuXPOOdcZ\nSKcibZx3GA1jtbF3MWmbl+zui/6QdyiNQk1N/YDRamrqmXcsnZVXzJ1zzrnO4XlgTt5BNALteeL3\nWP25TZi89d55x9JI4o2eX7ChQ+en6U3bnndwU/cHh+YTVefiFfNOSNJ4Sbu1c9lBkpolVf2zlzRM\n0qPVLtc55xxgdrePutJxEssycbvTefHr59mDP38773gajQ0duthwf03fOXxtTvvFaA66Yd28YupM\nvGJeRZJ2kvRPSR9Iel/SY5K2rcGqLE7OOecaidQd6XCkx5CWzTucBnUK/z5sLHdffGregTQKNTVt\nrKamm0vO3OD1u3h9g9eG/vnKa5dyWJ2Sj8pSJZJWAu4EjgFuApYFdgbm5RlXSuHxzM455+rXXcCK\nwGmY1c3/j0YhsTFwIrC1P0ioql4HRhYnNqnpCxy43dr0mzk4h5g6JW8xr56NADOzGy2Ya2b3m9l/\nJI2QNDrLWNydRFKTpLNiC/tsSfdK6p/kP1zS25KmS1rs8coKTpX0epx/o6R+Rev5tqS3gQcoammX\ndJSkF+N635B0dDJvqKR3JZ0saYqkiZKGJfP7S7pd0ixJ/8LH0HX/z959x0lZXX8c/3xZilQFBUVE\nURQLYO8afRA0aOwak/xijb2hYowtuIwlGjVGjT02NDHGXhJjiXIVe++KMbGLqIiKBUU4vz/uszC7\nzLKzOzP7zC7n/XrNi3nafc7ssrtn7tx7rnOuVPsCm2L2UNaBtDcSHYDLgJPNeC/reNoTS5LZliQv\n5+8LCp2By7h15wOSK659O5vI2h5PzMtnMjBb0tWSRtclx6li3pX/Atgb6Ad0Bn4NIGk14CLgl8DS\nwOLAMnnXjQG2BzYD+gPTmb926GbAKsCPgYYrW00FfmJmvYB9gD9KWivv+JJAr/Te+wIXSlo0PXYh\n8A2xju6v0uu9B8I554ohLTHfPrP3qdYludu+fYCueH3tslEIAxTC1o0cPgZ4m7j4lSuSJ+ZlYmYz\ngE2JiemfgY8l3S6pH/Mnw/NdDlxlZm9arEV7A1D3sc+uwJ1m9rCZfQ+MA+bkXXsg8Fsz+9DMZhE/\nStq1weTO8Wb2rRX4WNTM7jKzt9LnDwH3Eofg1JkFnGxms83sX8RVvFaWVAPsDJyUtv0KMKGI1+qc\nc07qD9xHBSbiu/lp1AlDWffiPwL7mzE763jakaUo8Gl5UBgCHAUcmljibzSboWK/ECQtIukJSc+n\nQyVOT/f3kXSfpDck3StpsTLfd7wkSx/jGzne2P5GryuGmb1uZvuY2UBgGLGX+VyK60X+KO/5t8Qx\nhqRtvJ93j2+AaXnnDgJulTRd0nTgVeAHYk93nUY/spO0taTH08mq04FtiL3ydaaZWf4bgW/S2PoS\n5yjkt/3uAl+hc865yGwKsC71f7+6ShnwxD8ZfN8rZryQdSjtiSXJM5YkF+TvCwoCLgFOTSzxqjfN\nVLHEPO35HWFmawKrAyMkbQocB9xnZkOA+9Ptct53vJkpfYxv5Hhj+xu9rgVxTCb2IA8Dvga65R1e\nqhlNfQgMrNuQ1I36ifO7wGgz65336Gbxl/7ccAo1rDjj/2bgTKCfmfUmTjwqptf7E+IbgGXz9i3b\nyLnOObfwkvohDZ1vv5n33LYC/eSwcfR5cwAfD90261jaC4XQTSEULihxxLkXMeKBZYA/tW5U7UNF\nP0JLe3chjpmuIY5/3p6YsJL+u2MlY2gtklZOJ0kOSLcHEseNPwY8D2wmaWA6Pvv4Qk000vTNwLaS\nNpHUGTiZ+t+3S4DfSVo2vW9fSdsXGXbn9PEpMEfS1sBWxVxo8Q/KLcB4SV3TsfB74WPMnXMuknoh\nnQK8BozKOpyFkbYY15eV7ziJydv91h44ZVrTV7gi7Quc2HBn2G/31UjCAazy+kmJJf7GswUqWj4v\nHef8LHH80cVm9oqkJc1sanrKVOoPuWjLZgAbAGPT4TmfA3cCx5jZV5L+DrxI7Gk+E2j4zt0aPDeA\n9Gt2KHAd0B04h/rDR84jJvX3Sloa+Bi4HrijQLv17mVmMySNYV55xzuB2xcQV0OHAVcRh+G8BlwJ\nJAs43znnFiY9iZPy18bMP9IvgrYYtyQ13w3lqUM+5ItBP8x3QlI7gI7f1sy3v7Hz/+/ZW/h0yHt2\n1wW/r0jAC68LiHlDfau+9g9eGv5MctEN17d+SO2DWmPyd9pLfA+xp/iWdMhE3bHPzKxPgWuM+jUx\ng5kFSWZmPsFwIeXff+eca5+0xUmLMvyvH9NhVgcmTPyA6YPnT7T33Xhpen44f6fihPs/LHj+Hlv1\n4YP1Nrb7T3u9IkG7ucIRO53Axo/muOfHSydXXvNJ3X5JBwPXmdkXGYZXj6SE+h2JtdWSW7RKYg4g\naRxxUuN+QGJmHynOSp9oZqsUOL9gAuaJ2cLNv//OOdc+abefPkj3qSvaVQ8NyDoWV5hCOAR43ZLk\ngfz94Vd79uXH93zIoxuPT8679bR618RP5m81s6qtHV9NuUUlq7IsUVdxRVJXYEvgOeIQi73S0/YC\nbqtUDM4551yrkbojDc86jLZIW/16e5af+CPe26TYOVIuG08D/5tv7xMbnMrT677WMCkHMLPzqzkp\nrzYV6zFX/OU0gZj8dwCuNbOzJPUhjmlellh4fjcz+7zA9d5j7ubj33/nXNWSNgR+gdkRSn9ZZR1S\nW6AR42tY5bbpfLTmP+3Wq3+RdTyueYLCWobdfQAHHPYmb44CDmpr//erKbdotaEszeWJuSvEv//O\nuWonaSNgUTO7O+tY2gJ1nHkQa105lh4fr2oTx3sljyqkENYGXrUkmZm/PyjUzGHOE7uze+cpTOlC\nXATxRk/MW66iVVmcc865hVAO6A14Yt4EiaVhkVN4+pARviJnVdufuKr5sw32H96BDjOmMvV44Gkz\nm38CrmsW7zF3bYp//51z1UzSKsTkxYiL7D2ZcUhVTeJGYLIZv806FlccSZ3MbFZQWBZ4BtgkseSN\nrOMqRTXlFhVdYMg555xr16T+xJLAdU4COgFdqV/y1zUgsT2wJjDfhEFXHRRCvTwxXUzw/rD33mKf\nK++i92cXt/WkvNp4Yu6cc861hCTgCuIqiKSLvO1MHCYqYISkIdkFWL0kehIXqTnQjG+zjsfNL03K\nH1UIy+Ttfg3YgcWnnc0WD6zIDrf7wk1l5om5c8451zK7AQOBP6XbBxET8jodiCsku4Z22uM+Vrvx\nJTMeaPpkl4kRI7pyww0HWJK8X7fLzGzir3ZflE0eOZLHNxyTXDXh6yxDbI88MXfOOeda5htgP8xm\nAZjZSWbWBfgq3e5sZmOyDLAaafRRezP43vVY6vkjso6lPZO0g6RJ6eOYRo7Pv79bt11UU/Mo8F8u\nvniz+Roe8sZdvDHkteSCmy6rTOQLN0/My0zS3pJekvS1pCmSLlI6/lDSeEmzJM2QNF3SI4p1b+uu\nm50e+0LSi5J2yvbVOOeca5TZnZg9kXUYbYlG1HZlxbsvZvL2l9j9p72ZdTzthaQ9JK3UYPeTwAnp\no9Bijk8W3H/BBZty8MGPApuZ2QX5h8LhuxzKim+uzH9W2ro8kbuGPDEvI0lHA2cARwO9gA2B5YD7\nJHVKT/ubmfUE+gIPA7fkNfFIemwx4ti76yT1bq34nXPOuYpa/D838d2iXzNjgA/xKa9OQL0a42Y2\nxcwmpY//NLwgPf4fhSCFMGzugRVWOMYuuODXZlZvUmdQ6MLMRWp5dOM/JFdc6yt5Vogn5mUiqRcw\nHjjMzO41s9lm9g5xDOIgYHdi+SwBpLU+rwGWSldDJe+YAX8BugCDW/FlOOeccxWhLY8dweB7tubt\nzXexieOrs1ZzG2VmV5aw7P3iwPkKoQbAkqSxWuTHcfn+j3LO0ce28D6uCL7AUPlsDCxC/R5wzOxr\nSXcBo4C57z4ldQH2Bt41s8/i5P65x2qAfYDPgckVj9w551xxpF2ASZh9nHUobYmEGLjDadTM+rvd\n9/sHs45nYacQ1gWmWpK8Z0nyKbDFgs4PCqsAhwNrJpb4m6oKaneJucR4ALP4b6nbzbAE8KmZzSlw\nbAqwNjEx303StsD3wEtA/jjyDSVNB7oDPwBbmtmMZsbhnHOuclYDHs86iDbol7y3cVfe23iPrANx\nAGwKvAg02cseFDoAlwG5xOZVaHGV4St/lomk0cCdQJeGybmkCcRhKv8FBpvZngWu3xvY18x+JKk7\nsTZuNzPbvuLBtyHV+v13zrk6kmYAPfx3VSSxBPAysJ0ZT2UdT3uR5g03mdlXTZ4bQl9ge0uSK5p7\nn6CwH7A/sHFiyexmB9oGVFNu4WPMy+cx4Dtgl/ydknoAo4F76nY11ZCZfQ0cDGwuafMyx+mcc861\nprOB6z0pLx9JWwPjgEKf0hfyPdBfITQr+Qz7/3I4Q18+E9i/vSbl1abdDWXJipl9ISkH/EnSl8AD\nwADgImJP+Q3Aic1ob7qky4DjAB+P55xzrs2RGAmMAIY1da4rjqSuxNziQDP7ptHzQjgauMeS5GVL\nki+AU5t9s9Ve/QfLvjs5GXfYiy0O2DWL95iXkZmdRawXejbwJfA/YiWW0RYXoLD0UfDyAsfOJS7p\nvHplInbOOdckaQck78hqJm1xUm9+fNSNdJ4xxgyfL1U+M4H/M7N7mzjvZWIRiRYJR+5YS/8p/Xlr\n+W1a2oZrPh9jXkHp+K/fAxuZ2f8yDqddaEvff+dcOyBtB5wDrI7Zt8Vd4mPMAfSznR+h6/Rl7eqJ\nA7OOpRxO2ySs8l0XDtliIs8BTF+Mrs+uzWojH+CZath+ZTUGPL0uo/a6hgklv9i+H3fhgsMu4JFN\njk/Ou/WskturctWUW3gPQAWZ2dWSfgA2IPaeO+ecayuknsCFwN7FJuUu0uixO7PGQxvxzP5rZR1L\nqYJCX+D4jcQ+r63Kh0BPgC7fscji0xhKrKSW+fbKk6np+wkzgM1KftEHX7we03s/tzAk5dXGe8xd\nm+Lff+dcq4krNv8Ys38077KFu8dcI2o7seqtn/Hh2rfZbVe32fKIQaEXMBY4DLgeOC2xZEq2UVVe\nCOpFLGixeZLYp1nH0xqqKbfwMebOOedcIWazmpuUO2Cxt/7CnI5z+GLQ3lmH0hJBYZGgMBb4D7DC\n5CFsNmIi47NMyiUNknRha9wrSexLYPWFJSmvNp6YO+ecc64sJJZhZu/t+N/I/7OJ49tUeb2g0DGt\n2f0GcTjIyMSSPQ+6lBWAU7KNjk+Am1vrZklibep71574UBbXpvj33zlX7RbmoSwStwIvmlGbdSzF\nSle23JWYfH8AnJBYUm91V4UgS3wp+vaqmnILT8xdm+Lff+dcRUkrAD/D7PSWN7FwJuYSOwGnA2ua\nMTPreJoSFERcAPA04kI9xwP/TswT8IVNNeUWXpXFOeecm+cb4JWsg2hrJHoB5wO7t5GkfBPim4i+\nwG+BWxom5ArhIGBWS5axb2tC0BDgvz6EJXuemLdBkq4G3jOzcVnH4pxz7YrZR8AdWYfRBv0OuNus\nuleqDgprEHvIhwPjgWsTS35o5PTbgEVaKbT5SNoMWNHMrqzkfUJQH+IK41sDz1fyXq5pnpiXiaS3\ngX7AbGAW8ChwkJm9X4HbLWgFUeecc67VaNTxe7PM9j/j/Y2GZB1LY4LCisDJwBbEnvJdEku+W9A1\nliQftUZshUhaBLgMOK4VbncWcFOSmCflVcAT8/IxYFsze0BSF+Ai4E/ATs1pRFJHM2vs3Xu9U1sQ\no3POOVc2GnFSd4bffAndpl1i7200Pet4GgoKA4BxxMmd5wEHJpbMWNA1CmFT4B1LkvdaIcTGHAi8\nama3VfImISgBtgKGVvI+rnheLrECzOw7Ylmj1QAkBUn71h2XtLekSXnbcyQdIuk/wGRJm0t6X9JY\nSVMlfShp78buJ2lbSc9Lmi7pEUnD0/3HSLqpwbnnSzq3vK/YOeeaQboPactG9lv6aOx4613Xzkj0\nl7hQ4jsJK/kx8HFj0ENf8XXf6cxY+qisX1++oLB4UDgTeAmYAaycWHJKU0l5ak1guYoG2LSLgX2b\nPKsEIWgR4FLgsLR2eZsg8SuJ9bOOo1K8x7y8BCCpG/Az4spZUNzQkx2A9YBvgY2AJYFewNLEd7M3\nSbrVzL6od0NpLeAKYFvgaWAP4A5JQ4C/ALWSFjWzLyR1TOMaXeoLdc65JkkJ0A+zGxoc2aqRKxrb\nn9V17YJEb+BY4ADgamAZoOWLx4zrPJgOs/4ArA3KMWWtq23i+KoYXhkUegBHpo+bgeGJJR80pw1L\nkgsqEVuzYjD7Hvi+wrfZG3g5Sez2Ct+n3N4FPs86iErxHvPyEXCbpOnE/zAjgbObcf3pZvZ52tsO\ncZz6yWY228z+BXwFrJx3ft0vwQOAS83sKYuuAb4DNjSzKcAk4KfpuaOBT8zsuZa8QOecK4q0FNLd\nwJXEeTf1mRmFavXW7W/qeGtd18ZJdJc4gbiC5RLAGmaMNeOTBq+8WQ9qZi2CeBCxko2fc7ld+kwx\nwy8rKih0CQpjiK91NWCjxJIDm5OUK4RuFQuwOl0G7JN1EE2RWF7iCil2fprxbzPeyDquSml3PebK\nqSy/XK222fUsDdghHWMuYEfgQUmrFXl9w7Fs08xsTt72N0CPAtctB+wp6fC8fZ2IPe0AE4CDgMuB\n3YFri4zHOedaajpwA/AXYs+fa0USnYmdNicSO2c2MWNyudq3WnsZeLlc7ZUiKNQQPynOEYetjE4s\neaGFzd2uEI6zJHmmbAE2g9Ji2q11vySxOUBVDmHJS8KNmB9dkz5v99pdYt6ChLr8McQfrFslXQps\nCnwNdM87ZalCl7Xwdu8Cp5nZ7xo5fjtwkaRhwE+AX7fwPs45V5z4yV9FS7y5+UnUAL8kJqmTgZ+Y\n8WyL28upN9DFai2z6iSNSRcH2pm4WuenwC8TSx4usdmdiH+vs3KppL+b2f0ZxlAtriAOx33AjB+g\nustwllO7S8wzVjfGXMD2wGLAq8S6oDtLuhwYQJzQUcovOtXdC/gz8U3Av4GngG5AAjxoZl+Z2beS\nbgauA56oUPlG59zCSvo5sAKNdw64Ckt7F3cATiX2gO5dSj1x5dQdOBw4mjg2vareZAWFUcSShx2J\nMd5djtU6LUm+KrWNEuWAzzKOITMSi+QtTpUDFsp8xRPz8rpT0mxi7/fbwF5m9pqkPxIndk4FXiC+\nCxyZd12hXygL+iUzdzKpmT0jaX/gAmAl4uTRSdR/dzmB+Gag6seSOefanJA+XAYktiAu7tOVWPP6\nny39yF85dQb2Z94QmE2t1so2BKZUQWED4msdSCyBeGNiyZwFX7VgCqEDsY736ZYkLZ8QWwZm1qxJ\nqi0RghZLEqu6iZNplZUccZEjzHgn24iyo2qd75IOtZpvWEpj+13jJA0EXgeWNLOsewRK4t9/5zIU\nFz35rj1OlCwnSTOAHpX8XSWxHjFJXYGYpF5vRouTVOXUCXiR2Kl0otVai4fAlFtQGEb8NGAd4iJB\nVyeWzCpH22li/gvgekuSdr0cfQjaHjgB2ChJsv8Zllga+MiMOemnPt3MshlKVE25hSfm7ZykDsA5\nxD8S+2UdT6n8++9cBmKp1X2Ak4CfYvZ4xhFVtUom5hKrEpPUDYnjq68wozxJak7LW629VY62yiEo\nLE/sRR0NnAFclFgyc8FXuUJCUE/gFWDPJLGQcTgASNwFHG9GSyfrljGW6sktfChLOyapO3H4zFt4\n7XLnXMsdT5y7sitmT2Qcy0JJYjlgPHHNirOAPcz4ppz3qJakPCgsBfyW2JP9J2DFxJKyVw9RCGtb\nkmT2yYCkTsAtwEGtMIzlVOC+LJNyiZ7AIDNeSndtW8qnPO2V95i7NsW//85lQOqEWVl6ZRcG5e4x\nlxhHXDDnYuBss5YvrqKcegJjrdZy5YgtKKxFnChaLl2IHUkTgNMTSz5paUOSepnVX9FS0irADkyc\neCnwd2A7S5Lv0/2/SU+bbGa/L3RdI/tbet25xEIOo0stk6icBgOHAYs2PNavC0usviijJn3KLd/N\nacaiRTMX7cpHa6zKoIeeLcv2ZyssxZcDl2HQg0+X8lrzjLPa8ryhqabcwnvMnXPOLZgn5ZlJJ3fu\nBww1K6maV51TgCWVk6y2tGQwKHQB/kZMcMs1Wc+A4xNL3m1pA5IGEYfArCtpeIM1Qb4AXrIk+Rz4\ncYP9deUWC32dv4C5Pb0N97f0uhuAO0tJypVTf+Icg92ASxreS8BPl+GkN2Zw7XdzaPrTrrc3G8TA\nR9+h5gdjVtdufLXUNEhLbpa63ed/8VE+Zf3EqFp4j7lrU/z771wrkJYH3vZJni1Trh5zia7ECZlj\nzbiz5LhyWh+4AxhqtTat1PaCwnhgjcSSnUptq1zScsVPAP8C/tCwxxxAIciS0ssrZkk59SH21O8P\nXAWcYbVWsKpMCFoZeKOYCZ8S1wEnmfFmOeOtdtWUW3hi7toU//471wqkfwLHYPZq1qG0RWVMzE8D\nVjJjt5JjilVXngLOslr7a6ntBYVVgYeAtRJLqqretKQODXrJ5x0LoW4IydZtMTlPa8wfARwF3Aqc\nbLUtX59EYhCwrBkPlSfCtqmacosmh7JI2hR43sy+krQHsBZwnpkttDUmnXOundvWe8uzJTEcOABY\nvUxNHkUsBnBdqQ0FhQ7AZcD4LJNySd2AdcxsUv7+xpLy1GTg4GpIyiUtA0y1IoaKKacuzKsx/yCw\nidXaG2UIoz8wDBbuxLyaFDPG/GJgdUlrAGOBy4FrgM0rGdiCSMr8B8o559otT8ozJVFDXNX5RDOm\nlKnZAcDBpY4rT+1HzB8uKUNbpVgC2Iu4GFJR0oS8KqrPEMuPTgKubewE5VQD7E6syPMasI3V2nMt\nvaFEB2It87PNmGnGY8BjLW3PlV+TQ1kkPWdma0mqBT4ws8slPWtmazfZeFzY5hqgH3FCx2Vmdr6k\n8cQf7LrZ1seb2d0Nrq2ajxWcc865YpU6lEXiUODnwObVVk4uKPQnjnvfIrGk0KTGqqQQBgEbWJL8\nPetYACT1Iy45/wGwQsMJoMpJwI7EMofTgeOt1op6AxKCapLEGl0sSeIw4K9mTG9p/O1NNeWcHYo4\nZ4akE4jv2P4hqQboVGT7s4CjzGwocTGEQyWtSkzSzzGztdLH3QtsxTnnXGVJfbMOwYHEMsTe0QOq\nLSlPnQdc1ppJuaLtJY0soZkewCLliqkMxgKzib3+W+cfUE4jgceBWuAY4EfFJuWpK0LQL+a2J7aR\n+FXdthkXeFJevYoZyrIb8H/Ar8zsI0nLEhc3aJKZfURaPigdo/4a8eM0iFV8nHPOZU0aDtyDtCJm\n7bIEWRvyJ+AiM17LOpCGgsK2xHlme7XWPSWtTKw60hMY09J2LEleBl4uV1ylkNQDOJR5bxROAe5S\nTusBpwPLEUsg3mC1CxwvP58QNBIYQaxpXudNKNuQKFdhxfSYH2Vm59RNrjCzd4kTBZolrSu6FvFd\nIMDhkl6QdIWkxZrbnnPOuTKIn4L+GRjvSXm2JHYCViUmZ1UlKPQALgQOSiz5thVvPQ24CFjTzCY2\n92KF0EshdC1/WCU5AKiLyYC19UtNJFZZuQFYzWrt+hYk5V2BS6dN63/EiBF2R7rSJma8YUaLx6W7\n1lVMj/lWwLEN9m3NvJWumpS+O7wJOCLtOb8YODk9fArwB2DfYttzzjlXNosBjxIn9ruMSPQCzgd2\nN2Nmye3ltBHwrdXa8yUHF50CTEwsub/Je8chq5+Y1a+rne5fPN18vZHj9a5Ln/+lhLh/ASxNHBZS\nEuXUEVif4jo1GzeYJfiGK5nC/nRgDj9iJp/wH4awjdVaKW96xgHP7LLLh7cBbwNflRSny0Sjibmk\ng4FDgMGS8seS9QQeKfYGkjoBNwN/MbPbAMzs47zjl0PhhRPSSaJ1gpmFYu/rnHOuCGbTiONdXbZ+\nB9xtxoOlNqScuhILL4wFSk7Mg8J6xAR37qflkroSc4Q3zez2BpfsAdwD872WPYDN0ucnNnK80HUt\nZklyqUIo1yrnZxAnZJa2Ause6b/jge58yghWtlr7opQmd9vtnJ07dDjziIMO+s1gALPSv+/tmaQE\nSDIOo6BGq7JIWhToTfyPeCzzxoTPMCtuxbB0Ba4JwDQzOypvf38zm5I+PwpYz8z+r8G1VTND1jnn\nnCtWc6uySGxE7MAaWo5JecrpNGCI1dpPS20rKNQtTHR2YslfACStD9wCPAmcaGZVNx6+3JTT2sTV\nRIdb7bzOxZLajKWfnzeztUpta9ddz11u9uyO291662EXlCG0hU415ZxFrfyZVmJZkrwe9nSseVPX\nbUosWv8icRwVxPqZvwDWTPe9BRxoZlMbXFs1XyTnnHOuWM1JzCU6A88Ap5pRcik/5TQceABY3Wqt\n5Al/QeEYYEvgx4nFRXkk9QZWMrMnS22/UhTCMcDLliT/KrmtOITlCeBPVmtXl9re3HbLmJi70lRT\nzlnMyp+HE8dmfUws7VNneFPXmtnDFB6LVfIPinPOuRaKn4gOw6zoYYmuYn4NvEuc9FcS5VS3Iudv\ny5SUL2/Ysd/wzQY/sZ/M7cUzs+nE3vJq9g+gpOEhecakbU0oU3tlIzEA6GbGf7KOxZVHMeOujgRW\nLnb4inPOuaq3IvATmjFfyJWfxErEceDrmFGOFTlXAT4lVtkpSVAQcPGxHPvyUzy1sWH/LTm6VmRJ\nUpbhNcppEPGT/o3KtGpqua1NrORzZtaBuPIoJjF/F/iy0oE455xrJWbPEIdPuIxICLgU+J0Z75Sj\nTau1V4HtytEWcchp/2lMG038xLxNUAg/Al60JCm5tzxdffMi4ByrtarskTbjThopoOHapmIS87eA\niZL+CXyf7jMzO6dyYTnnnHPt2l7AosQSiVVB0lJm9lFQWJxYxniH/9p/29rCNNsBnwPlWJl0Mal5\nUAAAIABJREFUN2AgcHYZ2iqrq64a2nv55V/9G7Bzkvj6A+1Jk5M/80oW1p0oYmKeq2BcVTUQ3znn\nnCtWU5M/JfoSV6Hc2oxnWzW4gvFoGeAkYCdg6EQmngHMSCw5ItvIsqOcegOvALtYrT1WkXuUMPlz\n1Ki/Pr300v/7+pprxm1egdAWOtWUczbZY25m41shDuecc5UkLQd8h1lpNZhdOZwDXFsNSXmqlrjC\n5soTmbgmMAoYmm1IzaMQOlmSzCpjk2cCt1QqKS9FCNp47NgeA26//ZANso7Fld+C6pifZ2ZHSCo0\ndsnMbPuKBlZF716cc65Ni2tK3A/cjtl5WYfT3i2ox1xiK+LY8mFmfF3yvXL6EfBIc5dvrx9T/IMb\nFLoSyxuPTSxpM+OWFUIHYq31HS1J3iu5vZw2B/4KrGa1VrE5di3pMQ9BnYFngZOTxEqu5OOiaso5\nF7Ss7DXpv39o5OGcc65t2BPoBVyYdSALM4luwCXAIcUk5YrWLbB/KUmTlNNGxDKLvfL2v5E+JjV2\nXcP9Nq+H7kTg+baUlANYkswBtixTUr4I8Y3T4ZVMylvq4ovP+uvrr6/3CXBj1rG4ymh0KIvFWfuY\nWZDUBRhCHGc+2czK+XGRc865ynoH2BezH7IOZCFXCzxu1vRaHpI2Ak4HlpK0gVm9JdunsRj7ATcB\nR1mtfT53P2ybPi/0vZ5GnHQ6n6AwDDgQWL2oV1JlLEk+K1NTxwOvWa3dWqb2yurLLxef+tBDO//x\noIOerMbSja4Mipn8mRCL6teVc1oW2MvMHqxoYFX0sYJzzjlXrEJDWSTWBO4FhpsxtdGL57VxJjAZ\nmGAF3lAppxOATYGflFpfOyh0AB4GJiSWXFpKW61JIXQCcsApliTfltxeTqsRVytf02rt/VLba/J+\nvvJn1aimnLOYconnAFuZ2WQASUOA64lF7Z1zzjm3ABI1xEV/ji8mKQcws9802l5OdQsTrVumRW8O\nBOZQhoWJWlkN8DYws9SG0lVTLwVqWyMpby6J5YB3y7QQlatixSTmHeuScgAze0NSMdc555xzDg4D\nvgaubHhAUg/gYOBsa+oj7Prt/c5q7e1SAwsKA4CTgSSxpMUTSLNgSTITuKxMze1HzIkuKVN7ZZMu\nRnUlcBBQlQsdufIpJsF+RtLlwF+INcx/CTxd0aicc86VRtob+DdWfb1/CxOJZYFxwCaN9HZ+S+z5\n7cS8RfyaMrZM4UFc4OjixJJXythmRSkEAUMsSSY3eXIx7eXUHzgN2MJqbXY52iwnM0xilPeWLxyK\nScwPIr47H5NuTyIuUeucc656LQr4RP0MpT2dFwLnmVEwiTSz2cAZzWm3XMljUNgRGEbscGtLlgPO\nVwijLUnKkayeB1xmtVaO1ULLKgRtCSxtZhOyjsW1jkbLJUpaUtJ5wC3AEsA+Zrazmf3RzL5rtQid\nc841n9l5mBU1ntlVzC7ACsDvJXWWdJikcVkHBRAUegF/Ag5ILCl5jHZrsiR5GyhLUq6ctgPWAk4t\nta1yGzNmTJ+xY/99+5tvrl51ZRtd5SxogaF7iENWJhHLL/Uws71bLbAqmiHrnHPOFauuKgvYh8Bu\noLeAR4DXgBPN7LlMAwSCwvlAt8SS/bKOJSvKqSfwCrCP1dr9rX7/JqqyhKAzH398m9WPO+6fo1s5\ntIVONeWcCxrKspSZnZg+v1tS5r9InHPOuTbkDjMeSYe07Fq3PkhzKaduQHertU/KEVRQ2AD4KTC0\nHO21FoWwKbCnJckBZWryFOCBLJLypoSgtYC9NtzwruFZx+Ja14ISc0nqU/ccqMnbxszKVczfOecq\nJy5Hf9zcbbPTCx5vbH/bus6AczFrU0MTsibRHTgAWKQ8LXbunM7jPB7mrqzZ0qS8M3G17Y7A/qVG\nFhQ6Ecsijk2sbIvytJYngLK8OVFO6wE/J46xryp77ZXr1LfvfjeNHj3h+FGjvv8463hc61pQYt6L\n+X+R1G0bcdycc85VO5EuWQ4FqxoI6NlOrptBrEftmudkYB3gsdKbmlEDszsCmPF5U2c3RjnVECdl\n5ohDYAqu2NkCRwMfENcjaVMsSWZB4Um0zaGc6t6c/Npq7dOSAyuz779fZMVHH91++qefDrhq1Kis\no3GtrcmVP7NSTeN9nHPOtU8S6wD/BIaZUXKSJqkb8BnQpaV/w5TTFsQyhp8Dx1utTSo1LoCgsCLw\nOLBeYslb5WizNSiE1YAlLEkeKkt7OR0DbAn8uEwLNLUsDl/5s2pUU87pCwU555xbKEl0JPacHlOO\npBzAzL6RNAvoUkozwLHAXeVKHIOCiIvnnN6WkvJUX2DpcjSknFYgfm3XzzIpLySdi9DVjG+yjsVl\nxxNz51z7I60O/Bezr7MOxVW1I4BpxAX0qobV2sQKNLsH0IdYs7tNsSR5sBztKCcBFwNnWq39rxxt\nltl6xKFLW2cdiMtOo3XMnXOuDduDOGbYuYIklidOzjwoqxUVldNqacm+igoKfYGzgP0TS36o9P3K\nRSH0UgjlzFP+D1gS+GMZ2ywbM54Eds06Dpetov7DS/qRpH3S530lLV/ZsJxzrgRmx2BWlvGorv1J\nhwxcBJxtxn9Lb09LSjqo6PNzGqScrgYCsHqp9y/CH4C/JJa0qDJMhn5DGSrRACinxYlfh/2t1qpu\nRdwQtH8I6muGf8q3kGsyMZc0nvjDcXy6qzNV9rGfc8451ww/AwYQE7VyOBcY1NRJymlJ5XQ+scLZ\ne8BKVmuPlCmGgoLCKGAzoLaS96mQk4Ary9TW2cDfrdaeKlN7ZbPrrucecvPNY84Equ4Ng2t9xYwx\n34m4XO0zAGb2gVT5j96cc865cpPoQxzKsKNZ6YmQpFHA+sC+CzwvpwHAi8SOrVWt1ipenzoodCNO\n+DwkseSrSt+v3CxJ5lCG8p9plZuRVOGCShIddtzxorEPPPDzPySJtbi8pms/ihnK8p2Zzf3BkNS9\ngvE451zLSKuki+04tyBnAjea8USZ2psEjDazBVbSsFr7gJiQH9EaSXlqHPBUYsldrXS/slAIJymE\ndcvSVk5dgUuBQ63WZpSjzXLq04d+yyzzn1f23POU07KOxVWHYnrMb5R0KbCYpAOAXwGXVzYs55xr\nBmlF4GFgODAl42hcAQqhE2CWxMmHCmEJ4PNW3d520zWg41bA0ILHXzruOz57ohsAy+7eh0/CF3z7\n/mzgWzaf2Lng+ePpBnypq/ZYNe98ped0rHf+5hPnrrRZ6df7h3XCZmvDvoo/E4XPH7vG9zzX28yY\n0eLttz+DI1bsyoBd+/A43/NFB2PYDV+XtP3ztV/ks8dm6LJzlmdWVzHshjjuevJ23Zq9vXy3MXT+\n5nmrtTvTFV47mDED5q74mtk2QPfuNX1h9mFJUl2lG112mkzMzewsSVsRV5QbAowzs/sqHplzzhUj\n9pJfApyOmSfl1evfwIHA6+n2rcB+zFvJsbLbc7iNfjOX5q0eh8XEkrvmO1+d/kUsoQiLb9ibTyd9\nCcwmjnPetOjza+jMbAAGt9rry9sOCjV9BvGPN1fk9/v/J5maf1zif+kQnlv5stO9xHrrv232ttmt\nLHvuu2zx5y2BGqbe05Ol5nxPPwF8zdR7erd4++U0sV6ha1fm1HSAdELkCv9u/vZ3Pb+n8zfD06/B\nL4Fl09dTDdt88smiXybJtPdwLtXkyp+SjgauN7MPWiekufetmlWYnHNVTOoCHAWcjVmbKQXX3qVl\n7ha3JPkk3V7UkuSLzOIRJwOrmVW+HJ2kGUCPrP6GBYXDiWX3RiSW1BujLRGAQ8x4taXtK6cRwO+A\nbsCJwD+rbbGetsBX/qwe1ZRzFjOUpSdwr6TpwPXAjWY2tYlrnHOudZh9B5yRdRhuPtsTF0o5ECDj\npHwocDCwhnIaSEwmT7Za+7Bl7elQYLqZXVfGMMsiKAwkVmDZtGFSntoZmN6StpXTusSEfAVixZTr\nrdZKnpzpnJunycmfZjbezIYChwL9gYck3V/xyJxzzrUpCmFthVDX63QHUHRt70qR6ABcxpIv/p7x\n+jXwPPAZlFQv+g7g0XLEV05BQcCFwPmJJa8XOseMz5q7oJJyWlU53QTcDtxCnMR6nSflzpVfMT3m\ndT4GPiIuX9y3MuE455xri9KEfDyxZ/qDtNRd9hZ9ewwb/XEgG5x/PPA3YKjV2kelNGlm1TomeGdg\nReCn+TvTNyc54oJKRX9yoZyWI/a+b0usA76n1davPpMuOLiOmd1UYuzOOYpIzCUdAuwG9ANuBPYz\nsxaPTXPOuZJJ3Yi/lybQ1EQZVzEKYTBxHPmTliRGHL5SNSSWpu/XJzHs+kmIo6zW/pd1TJUSFBYF\nzgd+lljyXYPDHYD3gaJqmSunfsThPrsDFwNDrLbRGtsnAj+SdLP5z+IChaAewAZJYj7qwDWqmB7z\nZYEjzez5SgfjnHNF6gMs40l55lYkrqD5ZNaBNOI8Phl6oZ01dVzWgbSC04E7E0sebnjAjB+ItbwX\nSDktCvwaOAT4K7Ca1TY+p0xSX2KlkTnAKMArti3YycASgCfmrlGNJuaSepnZl8BZgEnqk3/czD4r\nfKVzzlWY2fvAqVmHsbBRCJ2BscBZliSzLUnuyTqmOsqpA9DDau1LAIntgTWAPcrSvnQKcJOZvVDE\n6c0ZJlqyoLAJsAMFVraUWNlsbknFgtJFeA4DjgHuAtaxWnu7iFsfmf7bDTgFT8wbFYLWJb6JGZZ1\nLK66LeiXx9+AnwDPQMGJIstXJCLnnHPVahaxd7QbUBWrKConAaOJ1UL+BZwg0RO4ANjTjJkl30Pa\nHNib2FFVjEnET3UqLih0Bi4DjkgsqTfcRKIvcJnEyLTXvB7l1Im4aOA44qceidUWN1Q1XQV8DLBI\numu4pHXM7JmWv5r2KQR1BP4MHJMk9knW8bjq1mhibmY/Sf8d1GrROOecqyoKYWegxpLkxnQc+ZlZ\nx1RHOW1CHMLRl7hoyy3poVOB+8wIZbrVacDh6afITTKzrcp032L8BvgvcPP8cfCJRNKwCkv66cLP\niEMr3gF2tlpr7nCk/YAeedvdgGOJcz9cfUcSC2dcm3UgrvoVs8DQ/WY2sql9ZQ+sioq9O+eqgDQQ\nGIDZ41mHsjBRCKsDZknyUtax1El7ya8DNiZWgrnWauPiUhLrE8v6DTWjLEMuJS1qZpnUYVdOQ4gJ\nb/+Gx7rP7N59lQ9W2fCVga88MrPzzOZ8MjCI+InHCVbbsomIkjoAIva0r03a0Wdms1vSXnuV9pY/\nBfw0SezN/GO+wFD1qKacs9HEXFJX4jvgiUCSd6gXcLeZrVLRwKroi+Scy5gkYrL1NGYnZx1Oe6YQ\nehErcG1vyXzVPaqGckqAx63W5iakEp2Ap4EzzPhbVrGVQ7oQ0knAjsC5wHP1js8RJ9xywhlTek95\n9MqRV95R7+LXtx/E6zv8jB33/X0jzX8JPFKO1TolPQWs63+vGxeCapJk/jcsnphXj2rKOReUmB8J\nHAEsDeSvjjYDuMzMLqhoYFX0RXLOZUzahTi5bK10pU9XRmkNctXVHlcI6wLPpENX2gyJY4kdSds0\ndxGdaqGclgCOJ45pvww402ptvpU6g8KviDXjN0wsqZf0SdQAQ8x4reLxemLeYp6YV49qyjmLGcoy\nxszOb6V48u9bNV8k51zGpEHAYnjZ1opQCCcD71mS/DnrWFpKYjDwBLCeGW+V3p5qyjEsQ9JXZtaj\nkWMJcLSZbaecehIr3hwO/B041WptSqHrgsKSwEvAViMY0Zm4+E8/4BtiwYYxZvZtqbEXo2FiLukE\nM/tdEdc1+nVZWHhiXj2qKedssqSTmZ0vaRiwGvNmX2Nm11QyMOecm8uKKt3mmkEh9LIkqZvMeD7Q\n2AIyVUnSMDN7OT5HxIVwzihTUt4deFTSRmb1V7psgSZ6v+ignI4CjgPuBdYvYiGkPwJXjWDEFOKb\nkZ+Z2RMSg2CvfeGankCTibmkjmY2X7WWYkkqlEMcT6yQ05Sq/kRDcfgcvmiSa20dmjpB0njgT8TS\nUyOIM/KranU355xzxVMISwMPK4QOAJYkn1qStDhBa22SaoDbJa2W7volsTLLueVo38y+BkaXISmf\nS9JZkl6S9KKk3ZRTR0axDcswkqv5NafyNeNRXVIu6W1J4yU9k16zMkBQ2BrYAMgBhwJXm9kT6W36\nwoR3zOxjSetLelTSs5IekTQkbXdvSXdIuh/4t6S/SdomL86rJe0sqUMa85OSXpB0QHo8kTRJ0u3A\nKw1e4xlAV0nPSbo23XebpKclvSxp/wJflyXSOLeW1FfSTek9n5S0cXrO5mmbz6WvZ76edklj06/v\nS5KOqIsnXb287pzxko5Onx+T99rGp/sGSZosaQLxE4llmvltBiAEDQ5BXVpyrXOY2QIfwMtADfBC\nur0k8O+mrkvPHUicPPpK2s6YdH8f4kIEbxB7CBYrcK0Vcw9/+MMf/vBH0w8mTuzPxIk987a7ZB1T\ns+Ifz+8Zz05mBrEk3xzgerAlwD4CWzfrGAvGHedl7Zz+rRNdWZJOfMJY3uTnPIeYQZzLJeBRYOP0\nureAQ9PnBwN/nsjE7hOZ+NZEJm6V7r8Z2K6R+/YEatLno4iLI0Ecu/5e3d9d4uTSq9PnnYF3gS7A\nAcCJ6f4uxMoig4hj+L8ClkuPPZX/9xqY0SCO3um/XYnJbu+8r0s/4HFgZLrvOmCT9PmywKvp8zuA\njdLn3epeV9491gFeTO/RPc031kwfIe+8V4gr1W4FXJru6wDcCfwofX2zgfVb+v2eOJFFJk7kjYkT\nGV3E/w0Dnsv6/6g/qivnbLLHHPjW4ji7HyQtCnxMTLiLMQs4ysyGAhsCh0palfiR3X1mNoS4NO1x\nRbbnnFtYSAchrZd1GO3IicAmdRvVXHGlIeW0HrAXMCkdYnAKMZHdAV69CLjOjKezjLEJmwLXMZ4f\ncyz/YjBiIhP4O0diPGZmH1rMDp4nJod16uqyP5vuHw88nFhyb945kughzTc0dTHgJkkvAecQh6PW\nudfM6oYu3Q2MkNQZ2Bp40OIE662APSU9R0ye+wArptc8aWbvFPnaj5D0PPAYMXdYKd3fmfj3/xiz\nuSUbRwEXpPe8HeiZDit6BPijpMOJiX3Dsf+bAreY2bcWP+24BfiRxTkp/ST1l7QGMN3MPkhf21bp\nfZ4BVs57be+YNbume74TgZeSxO4uoQ23ECtm2eCnJPUmrlr1NPA18V19k8zsI+Cj9PlXkl4jvlvd\nHtg8PW0CEPDk3DlX31vA1KyDaKsUQndgHUuSh9Jdh7e1Kiswd3XKPwO/tlr7VOM1ktjDDKgGzhkN\nly9dlnulM8DK0VY9i9OfddiG2EP6W15nB+aVP8x/gzSb+n+X647N7k73xYA9qb+k+yvE3uKVgE7A\nGXnHTgHuN7OdJC0H9RZbmjtEx8xmSgrAj4mfROSXmTzMzO7LfynphNWvm3jF+eeOBDZM7zOReXPV\nZhFzitHElVIhvtnawMy+b9DU7yX9g7ga+SOSfmxmk/OOW3rt3Fszbwz7jcCuwFLA9XnnnG5mlzWI\nd1Cxr62QEDQMOIjYU+9cizTZY25mh5jZdDO7hPgucy8z26e5N0r/w69FnKiypJnV/cGdShwe45xz\n85jdg9m7WYfRhvUFdq/baItJeepI4ie1f023xxOHVhh06ghXLQLqVOpNJK1PgdUzS2ozp9XpSFcS\nRvIks3iF1RnPQ8BmxIV5iqoC0Z3uNQMZuDLwm8SS/CXdLwD2go6TiL3iSNpJUj/imiN1pY6b+pv9\nd+BXxOEcdT299wCH1E3wlDREUrciwp2VNym0F7GXeqakVYifnNex9J6rSPpNuu9eYEzdCZLWTP8d\nbGavmNmZxKEzKze45yRgR0ld0x72HZmX7P8d+AUxOb8x77X9Kj0XSQMk9S3itTUqBHUglrc8KUns\ng1Lacgu3RnvMJa1DI7OmJa1tZs8We5N0osbNwBFmNiOd7AzEQT1pySDnnFuoKIQ1gE8sST5MtzcB\n3ipheyfgUUuSqZYkbxPHCZcep7Q8sWcW4G0ze7rA8cUb2d/y6wbQj605kWVY22rN0uEW1wAPA8ex\n2JBPWPrlmxnG/ymnqXxAP6awAgC9+Jgh1K9u8gH9+JoeBfevwW8YwO3KaddmfnkKEbAjsxkJzGQ4\nA7mZU7iRZ4l/V4+xOEFzVZqoThIUdBRH7X4DN/yQvva50jZ+DrPPBvWTmAM8SEyuzwQmSPot8M+8\n+1iBe95LXC7+NptXpeVy4vCZZ9PhQx8DOzVyfb7LgBclPQPsCxwk6VVgMnE4S174ZpJ+Adwh6Uti\nUn6hpBeI+cmDwCHE4TAjiPMKXgb+1eDr8Jykq4lvdgD+bGYvpMdeTXOQ9+s6BM3svvRr/1iaj8wg\nvolt6rUtyLbptZe28HrngAUPZfkDC/4POqKYG0jqREzKrzWz29LdUyUtZWYfSepP/IEvdO34vM1g\nZqGYezrn2iBpbeDThayXfDfgNub1bG5GHL7Q0u1BwGuUMASokeEcKwA/T58/APON516B2ItZaH/L\nr/uERZjE3+31WKkkHeLwZ6007A+8+QqsMXkKI+hL3d+jHvRlCQYD0I2PoUEC3oO+dKJnwf3r8y1L\nMyAv3lI9x2VcwA/80WrtO2r5DfCb/BPM7EFi8lm3fXje8+WDwkbA6SMZudRIRq6XWP1PPSSOBnva\njM0K3P9x6vcsj0vbnUAcQpofxw/A4g32GXG89IkN2q0Xc0Nmdhz1h6Zu08h5vdJ/vycOZ6kz39ff\nzMY03FfgnD8Sy0gWOrZ6gX3nE8uENjTfuUW6E3gwSWxOC693rSgdZpVkHEZBTS4wVFLj8a3oBGCa\nmR2Vt//MdN/vJR1HnB1+XINrzaqk2LtzroKkwcDpxAlce2L274wjqhiF0B/YxJLkpqxjKURST+Ah\nYgWMmU2dnwWNqO3EwBumc+3r3YEDzKwqF0WSdBBxsaAjrJn/p4PCcOA0YA1iWcRrEpu/nKXEBsD7\nZmQ2dEK+8meLyRcYqhrVlHM2OflT0l4U6Dm34hYY2oT48dCL6exniIsPnAHcIGlf4G1ir5FzbuG0\nJLHqxD7EigrtWQfmVaWoOulQwx2rNSkHYLG3/sKcDlXfK5nOy7qkOdcEhcHERHwU8e/kbokljX4v\nzHiisWPOubapmKos6zEvMe8KbEH8I9pkYm5mD9P4BNNRxQTonGvnzB6lyEpPbZFCOB64xpLkA0uS\nD4ifDmQu/USzt5l9lr+/GWXwWp1GHb8u69z1U+7bfG94dUKTF7QRQaE/cajJbsThFQcnlsxo7HyJ\n9YEXzajeN1DOuRZpMjE3s8PytyUtRpzl7JxzzSMdCTyA2YtZh9KK3iROWqsa6US63wGvEifotQ3L\nPnIHb414kOduur/pk6tfUOhDHHe+P3AVsEpiyadFXLovcB7x++cyEoJ6J4lNzzoO174U02Pe0DfA\n8uUOxDm3UHgV+LzJs9owhbABsK0lSZxslyQ3NnFJq5K0FHAhcCr16zpnTjl1JcZ2mNXaN/WObXfQ\nGazwweK8lWxPLMPXZgWFHsARwFHExXDWSCx5v9jrzTiwUrG54oSg9YG/haCVk8TmG//vXEsVM8b8\nzrzNDsTVw26oWETOufbL7N6mT2p7FILy6oS/QVxCvCql1bCGVmQhndKdCPScLynf4rcDWfuOX/Pq\nLkfaxJNnSKe0ycQ8KNQtdX8CccGfjRNL3ij2eokaMxqueulaWQhzF70a50m5K7diesz/kPf8B+Jy\nte9VKB7nXFsndQB+SqxJvFV7n9CpEATcqxAOtCT5nyXJdOIiKJmTtCzQ18yeyd9fjUm5choGHEih\ncnX9n7+Lj9Z4w+760wWtHlgZBIUaYiGE8cRPjbZOLHm+OW1I1ABPSmxj5iviZmwsMIX6q6Q6VxbF\njDEPAJJ61Z0vqU/DCUPOOZe6GRgAnNCek3KF0MWS5DtLElMIBwNvZR1TAUOB5YBnmjoxS8rNXTXx\nt1ZrU+od2/rIAxj29Go8u9/gbKJruaAQFxqKw4Y+A/ZMLJm04KsKM2O2xLaelGcrBA0GjgHWS5Lq\ne4Pr2r5ihrIcSCzf9B3zJjAZpKurOedcfYcDH1CFvbLlohB+Tly+/FAAS5I3s42oMDP7V9NnVYUD\niX9f6tUk14iTujP8rvN5ffvz7f5T384kshYKCiOJE2y7EBO5fzVcIKi5zJjS9Fmuws4FzkgSq8Y3\n4q4daHKBIUlvAhuaWTEzxcummoq9O9euSVsBB2K2S4H9dQvh3NfI8fmvq0IKYRBwpyXJ8DJtLwLM\nsST5vmwxSo8Re7gBNjazlwsc37+R/Y1eVw0klptIOAr41RP0qbmTpTufysvfAjxBn5o7WGqRU/Ty\ntx2sw5z841dtdm3Xx5f5T80S1908+zRemXv+TdD5adboeBRHzVySvec0bC/77f6dT+WVt4Fxo9hs\n+mw6HGDGLunXYivgwJZuV5tqXWAoBPUgTq49FOgGhCSx7QucNwK4vVATjZw/GHinHGPLfYGh6lFN\nOWcxY8z/B3xb6UCcc61AqgEWx+zjvL0PAI8UOPsBYJn0eaEJZ41dVxXSsd8rWJL8F3gX2DjvcEnb\nljS+6EsJtmTeug+FhgBtSeHfxU1dlymJft2Zdfds6FEDK/fgh5nD+bwLscIXPfihZnW+7NLBOszd\nHs7nXXb/0f0bfLDhbbcu9+DRW6/OFy/kn/89FywHPHcFV/zmd+z+l4btZb/9RRfg48SSObNFR+Dh\nvC9Jw5+b5m67BQhBHYDDiBNsHwBGAB9Q+HcYwCTm/Z7LV/D8JLH/liFM5xpVTI/52sDVwGNAXe+Q\nmdmYigZWRe9enGs3pDHAKGz+nqD2Ju3lvhzYMq9iSlVI14Pob2avZR1LJUisAkw3Y2pQWMTgBcGx\niSW3FXX9iPFipX9OY9qQB+3mv+40f/saALwPHGBmf56/BddaqrHHPASdDNycJPZC1rFGj8mcAAAg\nAElEQVQsiPeYV49qyjmL6TG/DPg3/8/eeYdJVSV9+P0NOaiAggEDuiIYUBEjpgsqBsy6xjWtK+a8\nhmXVmTFn19X1MyuGNWfXrHPMrosJzIqimAADgqKS6vvj3MFmpmemZ6Z7ugfqfZ5+uPeEOnV76O66\ndetUwVhiDKD4vRKo4zitBWkZ4DRgo2Kr0hJYkoxXCCVnlKesCWxDLC4zP7Ir8CrwBDBS8E6uRjkA\nC0+4hnbTO/BLj90LpaAz/5IkdlqxdXCcppKLYd7GzI4ruCaO4xSaXsApmL1fbEVaihI1yquzXYUi\nq5E3JLoDm5txF4AZZwIEhVWBQ4k3IrnJ2mzkKgx66ADe2mdve/yivMXwO/MXIUjACh5a4sxvlDU8\nhEclHSxpSUk9ql8F18xxnPxi9hpmVxZbjUKjEDZUCPP9dZYYZcDaEnMfBQeF6hSI5YklX2YOVqXW\nVaWyh1Mt/erDfD74NXv8opKqSuqUDiFofeBp4J4Q1KbY+jhOPsnFY74XMXTl5Brty+dfHcdxnGYz\nmpgz2ikgEocAT5vxkRnfASfVGDKCGPo4z02SKtUOuA44q5bM4YePpN+Hy/DmfusVSG2nFROCViPm\nhB9ETON8Y5KYV0J15ityKTDUpwX0cBzHyQuWJL8BJbepMt2wONXMphVblzwxmToyVwSFpYAzgCGJ\nJXNqdP+VuHHzjsxGDT21J2s8VMkH21XYM2dMLoTCTuslBB0GlAPnAXskiRUiM5LjFJ1cCgztR5bN\nnmZ2U0E0chwnf0hrANMw+6TYqhQahdAfWNSSpORSy0kScCtwOzU8yK0FiVWBfc2iZ9yMe+oZfilw\nVWLJvDnXK7UicDywjpXXSAnW6+2H+b7vF/aff9XypDsOcB9wS5LY1GIr4jiFJJdQlnX43TDvBAwF\nXgfcMHec0mdNYBqxHsH8zlJAb0oz5/MBQBdqVLZsZXxGzLJSL0Fhe2ANYJ/MdlWqOqzlHCuft2qi\ntjx+D9Z4cW1e+8uAPOrrzEckiXnVU2eBIJdQliMyz9P8u3fUMdxxnFLCbFSxVWgpLEmeKbYO9TCV\nWLWzVcXDStwBnG3GW2b8RNxwVydBYSHgcmC/xGoVYeoFfEv0pv++xpDydqzy+HV8OPxGe/rsd/Oo\nvtPChKAR/F6FNpNrkqR2Rdo6xncELk4S+6AAKjpOyZOLx7wm0/GNn47jlAgKoSsw3ZJascwlg5nd\nXWwdckWinRkz09PTgI8bMf1M4OnEkqqaHVZuE4E9as3oPu5OZnWcwY/L/aUp+jolxTfEJ+s1md6I\n8Qb8lk+lHKc1kUuM+UMZp2XAKsCdBdPIcRyncfwN+By4qtiKtHYktgT2I2bjwoycvZZBYV1gd7J7\nTLOvt/nJ6zPo0R1448CtrKqiJHPOO/MSgtYBNk4Su7hmX5LYg42R1djxjrMgkIvH/KKM41nAeDP7\nokD6OI7TXKR1gLcx+6XYqrQQpwGey7iJSPQ0ozoLShVNiNEPCu2I8fPHD6kY0p9Kjgb2tfK6M2do\nSIVY4fkH+GTzp+2J8xuMXXeKSwhamfhEZH2gorjaOM78S52GuaS+wOJphbrM9o0kdTDzaluOU3JI\niwP/ATYHxhRZmxbBkmQ2daTtKyaSdgOeMrOSzameFgR6WGJ3M8abMQNoSrXNY6d1nPbzDiftsBfR\nY14Oc8NhstP16wvpMmkRPtlsxyas57QQIWg5oiE+HLgA2CdJrK7QFMdxmkl9lT//QdywVJOpaZ/j\nOKXHJcANmM33RrlCGKkQ1iq2HtlI0yMOoAQ9+RILSawIYIYBg80Y31R5B2908JCf2/98xqEHHbqi\nyR4D+lm5jbLyuje6arNT+tD/gWP4aPhRVnX6z01d22kRDibmne+bJHaBG+WOU1jqC2VZ3LL8uJvZ\nGEm++dNxSo1oDD4GtJqNhs3kVWIKv5LDzAw4tdh61MFWQH9iASDMmv60IShoyy5bXvBi/xef+XLR\nL/9o5fZTThOXfP0Rvhr0nj36j6uburbTMiSJjSy2Do6zIFGfx7xbPX0d862I4zjNxMwwuwlbMDxa\nliRPWZJ8V2w9Sh2JNhK7pGErmHGXWTTK88DePX7u0WbYmGHb5WqUa+ujD2eJN1fi64Fb50kHJw+E\noPbF1sFxnPoN89GSRtRslHQQ8FrhVHIcx6kbhbCuQmhKqtcFFQOGAd2bI0SVWkiV6lp9HhQWAy4E\nRiSWzMpJxpDTFmKl/1zMB9tdbM+cOaE5+jj5IQS1C0EHA5+EoP7F1mcBpOT2xzjFpb4ft2OA+yTt\nze+G+CCgA7BToRVzHMepiUIQcCwxReL44mozL2lc+Y7Ag8UuJCQxGOhoxjNmzCHGCTdNVqU6AocA\nJwNH8Xu63AuB2xJL/pezsJ7vP8C0Jb/npyVPaqo+Tn4IQWXE9JanEz9LOyWJvV9UpRY89gYmFlsJ\np7So0zA3s28kDQaGAKsRvS4Pm1kpV9dznAULqS1wInARZvN9UQ5LEgP2LNb6qlQZMetI9dPGGVZu\n76XHuwGnAP9RpdoS47hrkjn+d7nJ6QthbMbQ8liq/ofl2/HG/ksztPwDK7f3JDoAK5jxXi7jid/t\n7TL07tAofX4fvzYxHeVbwDArj/uOgsJmxN+G3HOWDzthOAOfTnjtoPXzlbO8b1+WyNYeQt3XmyS1\nr3cBHN8fuJ1YyOfgJPHf9WJgZv8utg5O6VHv4+B0A9Mz6ctxnNKjA/HHtSkp7loVCqGsBKp7nkP0\nclXHtn8FbC2pOzEjzi5mNkOV6g3clGX+V8DWEGO/gTlmGL90X4XJK4+i+inA7LbtmDC4N/BSOn5J\n4Dxge4CGxpvxXI11F2tInzrGfwnsaeX2UnVHUOhELOZ0RGJJg3HlGnrqkiz64c0MfGooH2w/yp46\n99WG5jTEfvux1KhRsMkm7AZZ4+Wber0LyviviCkQH0gS88JOjlNCyEr0MynJzEzF1sNxnOKjEDoC\nLwObWJJMK4oOlVoTeAIYkJaX/71P6gBsZmaP5CxPHAosbta6irUEhbOAvoklu9U3TkNOW4jun97I\nio/vyMQBn/PVoAPtyfOb7eQJQfriC57YZx82B0aY2TXNlek0HUn/A9b232unNVNKNqdvoHIcp+Sx\nJPlVIWxXLKM8ZQKwa02jHMBiGFHORnnKtcDC+VCspQgKA4ARwOp1jdGQ8g4sPOEK1npiP6Ys9y1v\n7bObPX7RPXlUY5e2bVk2j/Icx3FKhvqysjiO45QMliRfFHX9cvvOyq1miEjT5RkzzWg16R6DQhvg\nGuDviSVf1+zXkAppu4MvZPVbp7LEW7vw7i5H2HUvLlFtlKebY5unQ1A34NL33uPE5spynFJB0mxJ\nb0h6W9Kbko6r/rxIGiTp0gbm55aqVNpR0hxJ/eoZEyQNatwV5A9Ja0jKmkq1vr4a4/aXdFn+tWsZ\n3GPuOK2NGM+8LWY3F1uVQqMQOgFHAheWQHx5XpDYEfjGjFeKrUsjOQSYSfT0z0VDKkTnb09ilXAq\nmiM+3vJ0pvc8O8sGz50lfWlmzb3uk04/ndHNlOE4pcR0MxsIIKkn8G/i07QKM3uNhlNU5xqTvCfw\ncPpvRT2yihnjPJCYAfDRRvZlUpox2hlIamtmWdPMusfccVof5wPrFVuJFqIT8FspGuWSVpZ0XBOm\n/pa+Wg1BYWniD/mIxH7/W2irYw5kpYe/p+8jp/P5hlfyzm4L2X/+dVZNozz1/p1BTM3XZJLEpiSJ\n3dIcGY5TypjZZGK42BEAkhJJD6XHXSXdIGmMpLckzZO6WtJikl7K5lWW1JX4u3EEMU1mdXsnSbdL\nelfSvcTvXEk6WNL5GePmeqFTj/7Y9HV0xph9U73elDQqbdtO0iuSXpf0pKReafu6qa6vS3pR0kqS\n2hO/I3ZPnyD8MUN2zb7dJK1TU0aW6x6ejllU0rD0+DVJd0rqko45N/33LUkXZJHRQ9L9af/LkgZI\nKpP0qaRFMsZ9JKln+rpb0qvpa3DaXyHpZkkvAKOy/f2B6mKBpfciTQrjL3/5K+MFGxp8YbBw0XVZ\nwF/A0sA2xdajJV5VVN1XRVXF3GsfdvyOHLDR1xy31Cx22P8GktM6NPBeJcBPwC9A/zy8972JXrGD\niv3eLOgv4H/+e93s93BalrYfgJ7pZ+ehtO084OKMMd2q5wO9gFeIm9CzrbE3cGV6/BywVnp8HHBt\nejyA+FRsLWKmn48y5j8CDCZ6rMcQDfguwNvAmsTUqR8APdLx3TN1TI//AlyYHi8EtEmPNwfuTo/3\nA/5ZxzXM01ePjP2By4g1d54DFkmv51mgUzrmJOBUoAfwPnMTEdb+bU1lnZoeDwHeSI//AeyfHq8H\nPJEe/xvYMD1eFng3Pa5IPy/1fl96KIvjtC7eBIZjNrXYihQShdAG6GVJ7VjmFtOhUt2Bm4kbPn+t\n2W9mXwA5x71LLAt8YbHgT6shKOwErAzsoS1O2pQlX7uBNcYsx7hhDzF+yH72zOk/5iDmDOKP+Cxi\nXvS9Cqiy48yvbEaGt9vMpqSH7YGngcPM7Pk65u5JTOkKcFd6/jqwMXBpKm+spDHp8beSPpG0HvAx\n8Yb6pdRDfq+Z/QKQetk3Jt4o32lm36fzf0jXWkbSncASqZ6fpu3dgJskrZjOrbZHlb6yUbOvLhkA\nQ4l1GLYws58kbQusAryUhu+3J6aX/RH4Nb2WnYihPjXZENg5va6q1PveFbiD+H12I7BHeg7xJmHl\njG01C6XeeSMWoKv3iamHsjhOa8LsZ8zeKrYaLcAg4Ioi63AuMCGbUd5EziJ+wbcagsIiwGX39Png\nvCF7XTiaQddU8WuP8Ywesazdc8uOuRjlktYA1klP2wK7SFqqkHo7TmtG0grAbIthLbW6s7TNBEYD\nW9UhrwfR03udpE+BE4A/Zg6pQ5XbiYXTdgbuTdusxvi6jqu5jOjlXp1Ygbhj2n4G8LSZDQC2I3rg\nG6Jm7HhdMgwYB3QFMje6PmlmA9PXqmZ2kMUqzeum/dsCj9WxdrZrewVYUdJiwA78/h4JWC9jrWXM\n7Oe0b3pDF+mGueM4JYclyavALsVaX5XaiPgl/bc8it3XjLq8WSXJhIW+++cp21zZ7vLd/3Ydc9r+\nxv8O7W933jnUnjnzy0aIOZBoOFTzG3BALhNDUFkIuiIELdkYvR2ntZJu/rySaNDW5Eng8Iyx3dJD\nA/4M9JeULWPRrsBNZtbHzJY3s2WB8ZI2JoZ67JXKW415U6HeB+xI9K7fnrY9D+yYxqZ3SfufIxai\n/GN6E4BikgKIm1i/So/3z5Cd2Z75fTCVGKKSjWk1+uqSIeCz6uuWtArwX2BDSX9I9esiqW96DdXv\n43HAGlnWfZ4YCoSkBJhsZj+lsS/3EZ9EvJvxlOAJ4Ki5ykTnRM54KIvjOCVJsTZ8puXorwaOsvK5\nj4qrfwSPA4JZ40uYmxU2U0BQ6AJUPzJuFtPa/9z25sH39nn8kHsW/uXLDT5i9Ijtmlqx08yOAo6S\nZMC3ZtazEdMPID6OntSUtR2nldBJ0htAO2K4101mdnHal5kl5UzgX5LGArOJMcv3E+OjTdKewIOS\npprZlRny9yA+AczknrT9eOAGSe8C78HvGY/MbEravrKZjU7b3pB0I1D9fXCNpU9xJZ0FPCtpNjFM\n5s+pjndJ+oFovC+XzjsfGCXpFOA/GddYBZycvh9nm9ldGTpn9p1TjwxL35MPJO1NDN3ZlnhjcJti\nQTiAvxON/QfS8+eBY6lNBXC9pLeAn4mx7tXcQYwbz2w7ivh3eotoZz8LHJahW7145U/HKXWk/YBf\nMbujwbGtHIWwPbCqJck5RdOhUqcRQ2l2tPL4BamY5eAm4CGg3Mwm5CRLCLgbOM6MzwqkMgBB4Xxg\nBeIPVpMY1/XHjpUbP1Dx1aoPD+0ycdXJvT/a9Jh3X664veGZDdNYwzwELQ6MBbZIknnDtyT1Jsb3\ne+XPIiOv/OnMB5SSzekec8cpfV5iwQk7e5640aiYfAbcUG2Up4wBNjGz9xojyAyTOAf4PJ8K1iQo\nDCR6gwYkltSqTNoQGlLejoW/uJw/PPlnpvaewpg/7fPj4xf/O++KNo5/ANfXNModx3HmZ9xj7jiO\nk4GkMtJHw8XWJRfSipyvAP+XWHJ9Y+ZqSIXo8s3ZrPDMsczsPIPxm4y0R/95eSH0bIzHPARtQ4yx\nHZAkVmuzlHvMSwf3mDvzA6Vkcy4oXjjHcUoYhbCGQqhVHKJI3ExMd9VkJCSxh0S7POlUH0cQc4Tf\n0JhJ2ubI41j53qksH47hk83P4/0dFymUUd4EOgMHZTPKHcdx5mc8lMVxnFJgVWK2jg+LrQhwDPBt\nM2V0Iub2va/56tRNUFiWWCRjcGJJTh5+bXXsfizz0sX0m7gQ47a4imm9j7GqitmF1LOxJIndXWwd\nHMdxikFBQ1kkXQ8MByaleSaRVEGs/lSdn/NvZlYrb2QpPVZwnBZH+ivwGGZvF1uV+RlJvc2sMan/\nSoagIOJm1FcSS85saLyGnTCc3q9ezaIfLsG4LW9jynIHWVXlL4XXNF2/aVlZ6pLloSwlgoeyOPMD\npWRzFjqU5QZqJ7w3YknZ6sTrdSVzd5wFE2kd4K/AN8VWpdAohI4NjyrAulK/tBrdSzpO26lSe+ZP\nNp3zJasBdgWWJ6YMqxNtcfJg7bPlR6x540NM7f0Gr/+ll913459a0ih3HMdxcqOghnlaGvaHLF0l\ncVfiOCXKhcBfMWtuOEVr4CqFMLwI654JvM4Q1mERLqP5oSsASCwLvJCmSSwYQaE7sYz2QYklM7Lq\nstnIVbTX9m8w6OoXmLHQ14w+eDm759/b2jNnfFdI3RzHcZymU6wY8yMl7UtMZH+82e9FPBzH4Y/8\nHuo1v3MokNWwLCRm9kcAVeoi4HkrtyfzI5fPJTYudDEhYrGQ+xNLXqrZoaGnLEPP9//NOs9syOcb\nvcXoQwbYU2e/U2B9mkUI6kPMV+5hKY7jLNAUPF2ipD7AQxkx5r343eg4A1jSzA7MMq9k4n0cx2md\npOWWt6lRPS72VWoQ8AiwmpVbTjdCWuPm9Vjh6ePQnKJ9N/3hl3aLDZ3SZf1bev34+C9tbOY8nW1/\nWZw+YWO+Xmsc3wzcz548t5bhXizqijEPQSL+HZ5NEqtZnbAuWR5jXiJ4jLkzP1BKNmeLe8zNbG5p\nZUnXEjcvZSXdKFpNMLNQOM0cx2kpFMKpwM2WJOMLvRSwqaR7zWxu5hFVqi1wDXBiLka5xDIwp5y9\nbt+Hrt9M5Kcl533KN7t9B2Z0WYROPxS0dHyZoRVmtu372NKTJ/7Sfs4fag2Y03YGb+6/gz1xQZ3f\nqyXI7kBv4KJiK+I4zoKBpARIiqxGVlrcMJe0pJl9nZ7uRCy5nBUzq2gRpRyn2EiilRS0yRPjaIFw\nHTP7iZjnuyaLAc8CN9U3X6In8DdgPzY6L7Di4+Mpm726ldtvNcb1Agaa8XieVM9KUDgNGATsmGt6\nxFImBPUALgF2TJIa3n/HcZwCkTp6Q/W5pPKiKVODghrmkm4DNgUWkzQBKAcSSWsSs7N8ChxcSB0c\np+SROgMBaSjRkJzvsSTJa7n3tFrnrkBHM6vX2AawcvsGOLZueSwMHEc06m9n0NXrs/nIKmDPmkY5\ngBmToOBGeX/gKGDg/GCUp5wP3J0k9t9iK+I4jlMKFNQwN7NsKcgaVTLaceZ7zKYj7bkgGOUKYXVg\nrCX5MywlLUcs5DMbOLF5sugIHAacRDS01zXjE1UefAXwiJXb8zXG9wI6mfFZc9ZtiKBQBlwNVCaW\nTCjkWi1FCOoAdCPeADmO4zh45U/HKQ3MxhVbhUKjENoBFwO7Ad/nUfTXQAVxk3mTDH6JtsD+wGnA\n68BmZrwNoEoNBnYkVietyQbAmkBlU9ZtBH8GOgBXFHidFiNJ7DfiUw7HcRwnxQ1zx3FaBEuSmcDm\nzZWjdPv8XLlmM4AHmyaL6hCYM4Avgd3MeGVuf6XaEz3Vx1i51arJYMYDwANNWTtXgsISwNnA5okl\nsxsa7ziO47ReCl3503EcB4WQlzRUkkYAf2/S3EodqEr1iHKQxFbEWgonEGPJN8s0ylNOAMYD86Rb\nLHQBoRr8A7gusWRMC67pOI7jFAE3zB2nGEirETdHz/cohJ7ACwohH0/o7iUaqo3ToVIJMdxllsRg\n4m78S4CziHHkT9YsCqRKrUTcIHq4ldcKkSmXOKDR2jeSoLANsDZweqHXchzHcYqPG+aO09LEDCLX\nkJGqaX7GkmQysL8lyazGzJPUXdI8nmkz+9YauUlWleoIXMVHW19Ahd0K3AbcCAww455sVTpVKQFX\nAWdauWXb2HkJhQ9h6UqMKT8kseSXQq7VUrRtS7sQ1KvYejiO45Qqbpg7TsuzPvAb0Thv9SjSLkt7\ne0mdJXVmyJDxdfRnm7ewpFOAj4A1mq3gj8tcwGcbdeLWR0YSc5evZMYNZsyS6JwZlpJxvj/Qleue\nvzZbvxk/muV1A2s2KoFnE0ueKvA6LUbnziwMuGHuOI5TB26YO05LY/YSsDlmc4qtSp44mWwp7zp3\nPoOysh+Ab4Fjssw7pY7244FVgPXN7M2mKqV205fSdiPeZ2bHw3n00tuBvkRjN7OQzWSYJ158Micu\n1gs4DxjBhI0m1uqn8PHlQWEQsDfxvWjVhKCyELQHwIwZ/JIk9naxdXIcxylVVKrFBtPECy25wcpx\nnCagWCDpV6txo6EQ+gA7WpI0Oia8efrQg3izcCDLvPAV36/4b/tpiXNynl+pW4GvrNxOKJiS9RAU\n2gKvApckltxcDB3yRQjaDLgImDFkCOsA35pZz+bKldQb+AIYYWbzxZOn1oqk/wFr+++105opJZvT\n0yU6jpMzkvoDH2Ya4WY2PdtYS5LxNGGjZlOR6Er0wB8D3A2sbp9v9GWjZFRqK2AwsFr+NcyZo4l5\n3m8pog75ogfxKcX9wPzyhMhxHKdguGHuOC2B1AmYgVmrzEMtaXmigbUlsDHwYZ1jQ+gAdLAkmdoi\nug05bSFmdzgf/r4jUEUs+vNxtk2d9cqpVBfiZstDrdx+LoCqDRIUlgf+BqyfWP6qoxaLJLGMNJMl\n4YxyHMcpaTzG3HFahoOBkcVWohnsCXwC9DWzOo3ylC2J4QsFRUPK22mHA69l4A0/sOJjO9H1623N\n2MuMj4CrJTZrpMhy4GUrt8cKoG6DBAURbwwuTCz5uBg6NJUQ9IcQ5I4ex3GcZuJfpI7TMlwKdCy2\nEk3FzM7OeWySPKgQHi2ULhpSIbp+cy4DnjqGGQv/wns7H2mPXvp/XD/PsEriptPcZFZqILAfMCC/\n2jaKPYDetMBNTb4IQb2B04BdgM2At4qrkeM4TuvGPeaO0xKYGWYln4taUpmko7KlMWwMliQzGx7V\neCQ2B/uM5Z49knFbnskH23e3Ry/9v1rrG1+Y8WtOMivVhpi68mQrt0n51jkXgkIP4GLgoMQK897l\nkxC0aAi6ABgD/ACslCTmRrnjOE4zcY+54zhzMbM5kroCCwPfNWauQtgXmGNJkvdNixLrAucAyzD6\nkJGg26yqola8vsS+wKNmTG6E+COBacSiQ8XifOCuxJL/FlGHnAhBqwDPAXcCA5LEviqySk5xcQef\n4+QRN8wdp1BI7TAree9nTRoTtlKD58hzuI7EqsCZ/F6W/kabtuRMqMg2VsAyNCL7hyq1HDGf+mAr\nL07u2KCQAMOAVYuxfhN4H1g7SWx8sRVxSoJxQP9iK+E48wuex9xxCkEsJf8EcAFmTxRbnZqkpe63\nBzY0sxOLrU9NtMXJG7Hk6zfy4gk9+GSLs4ArzMhrKJAqJeAh4BUrtzPzKTtXgkJHYlz2CYklDxZD\nh5ZCkuF5zB3HKUFKyeZ0j7njFIZ9iTmcnwFQpbYADswy7ikrt2trNhZyvKSOdOI1OtCbjRmjSt3e\nZPkzO1YxZcx63LVMdyZ1/G1u++q3LM7A61eoNX78phN5tvyTWu2Z48tmtmfQO4vzyWbPsdyzm9q4\nLRrMRS6xmFntzZ5BYT3gEKBTzb7D1zt8mbsG37XaTZfdND1UhNtr9rcQvYGxpWiUh6AVgcFJYjcV\nWxfHcZwFBTfMHacwbAiMwGxWev4pschKTcbVMb9g483sV/XXmewGtKmV6zs3+bPbidcP3Igxe5/G\n0HY/stqPN/JMx8/n9s/stDjfr1jbMP+120Ri2sV5mXe88dnGV9vTZ79bhy7zINEeqJLY2IwpAEFh\nVeAsYgjMRcDX81xMz0+7XLv5tRfsX7X/RR1mdWgo/WMhMaAo6RnrIwQJuJb4RMFxHMdpITyUxXHy\njCq1ipVbTkZlSyCprf1+g9BMWYidvziaTrNP5NblxgN/M+PZfMhupl5tzZiVFuipBLYibqj8V2JJ\nrRAYVeoqYLaV22EtrGqrIAT9mfikYYMkyU9RLA9lcRynVCklm9N3UztOHlGlNgf+o0q1L7YuAJI2\nAe7IjyyGAC/x4mIj6DHjamDDUjDKAaoIiwWFy4HRwHigb2LJhXUY5RsDw4kVNp0ahKDFgXOBg/Jl\nlDvzN5IWlfRG+vpa0hfp8TRJlzdD7vqSXkllvSupPG2vkHR8/q4gv0jqI2lslvblJO2Zw/xEkj+t\nWkDxUBbHyROqVGfgSuBwK7cZxdYn5SVirukmo7Zz1ubcsQ+zeL9fmdjx70zseJv9s+8c/pknDZuq\nl1i0PbP/72FeGNcORgA3ASsnltSZi1yV6gBcDRxl5fZjS+nayrgEuN7zkju5YmbfAQMBUuN5mpld\nnAfRo4BdzWxsumG9OvtLsx71Syozs5yzN+WR5YG9gNuKsHZeyeeTWGde3GPuOHki+ZRLgP9ZuT1S\njPUlLSJpkcw2M5tlZlOaJK/dnJUl7mZ22QM82/NGtv56ZTNuNcs9HWGhCAqdH+W5g8/k7WHtsF7A\nwMSSY+szylNOBj4E7iu8lq2PENSF6LA5vdi6OK0awbye39TLPUrSc5LGS9pZ0ivrQLMAACAASURB\nVIWSxkh6VFI2R2FP4BsAi7yX0beKpCpJ4yQdOXdh6U+S/pt62a+UVJa2/5Su9yYwWNI9afsOkqZL\naiupo6Rxafuaqbf+LUn3SurWQPugtO1NoK4QuXOBjVPdjk496M9Jei19bVDrjZTWkfS6pOXTNYKk\n0ZIek7REOuYoSe+k69cy+tPruiF9r1+XlKTtL0taJWNckLSWpC6Srk/fx9clbZ/27y/pQUlPA0/W\ncY1Oc4kFCUvvFVUrvh7+8lcur2WPYa2XezPjiK3p29JrEzOOnABMAvZovjxbll0mPM3RH0wHOxGs\nc7Hf3+pXFVXtq6g6rIqqr6qouquKqv45X1cFK1PBt1SwTLGvY0F8Eb2ck/Mkq3cq76BiX5e/av1t\nyoHj0+MEeCg9riDWOmgDrA5MB7ZM++4Fdsgi61Tg+7R/BNAhQ9aLQDtgUeDbVO7KwINAm3TcFcA+\n6fEcovcd4s3nuPT4QuC/wGBgU+DWtH0MsHF6XAlckkP7Runx+cDYLNezafX7kZ53yrimvsD/Mt+3\nVKfRwNLptb4ELJqO2R24Lj3+EmiXHi+cZd3jgWvT437AZ0AH4BigIm1fEng/PT4b2Ds97gZ8AHQG\n9gcmAN2K/f+sAP9vrdg6VL88lMVx8sDn3ThtwwM5bHalfVSE5bcENgASs6ZvOtXQSX2o6nUMsA+h\n53Xs8OWfzfgsb1o2g6DQBtgTqJxOm0/+Sd+Kx2yJq3Odr0qVAVcBlVZuEwqlp+M4dWLAo2Y2W9Lb\nQJmZPZ72jQX61JpgdoakW4kFuPYifgcMSWU9bLGA23eSJgFLAJsBg4DRMfKFTqQed2A2cE8qd1bq\nae8PrANcDGxCNO6fT588LmJmz6dzRwF3SVq4jvbq8S+k7TcDW2d5D2puLmwPXC5pjVS/vhl9KxO/\ns7Yws28krUYsQvZUem1tgOqqu2OAf0u6n+zZuTaEGHxoZh9I+ixd605ivY0KYDfgrnT8MGA7SX9N\nzzsAyxLf9yetiU9hndxww9xx8sM+c8r4uRgLm1ldX8Y5IbEI7eacwHWfnsx7C13PN51WsW87TITa\n2Q5bmqAgYDti6sOfgAOHs/Ek4KBGijqQ+ONyRX41dBynEcwAMLM5kjKrIs+hDnvEzD4BrpR0DTBZ\nUo9MWSmzM+aPMrORWUT9aqlrNOU5YBtgJvA00cguA/6aZW5d2Toa216TY4GvzWwfSW2AXzP6viZ+\nZ60FPJLKfMfMBmeRM5x4Y7Ed8HdJA8xqbdyupZOZfSXpO0kDiIb5wRndO5vN62iStB4U53duQcJj\nzB0nD1i5TbPywm8mkrSlpGXyIuvPn3bXCj9VAh8xs2xpLuy3mn3daYQZE/Mhv7mkpepfBM4ERgKD\nE0uCGe+acWyuclSpJYmPZg+ycs8yMp/wPbATHufammh0KjpJwzNOVwJmAVPqkGVEA3tXST3T+T0k\nLVuH+OeJoRwvmdm3xJCYlczsHTP7EfhB0kbp2H2AYGZT62j/EZgiacO0fe861pwKLJRxvjC/e/T3\nJXrBq5kCbAucI2lTYjhJT0nrp9fWTtIq6abYZc0sEPfQLAJ0yXKte6fzViJ6vz9I++4ATiKGwLyd\ntj0OHFU9WdLA6sM6rsvJI+4xd5ymEr8Ql8TsqwbH5o8Vge+IcX5NQkLAX9i9zQVs+c0ErlwxMePd\nGEpYfA5IDth16e+WPrljv45Lvdf7vZvuHHznszPbzpzD7HZbcmqZaPtbY2+ADgOutXJrVnaa+ZUQ\ntBUwOkmsVuXUUsXMfqEZT4mcgmMZ/2Y7htqZVbJlWvmTpIuJ8eiziHHPcxRz4tcab2bvSToFeCLd\n9DmT+Pn/PMv4V4FeRM85wFvA4hn9+xE99Z2JhdcOaKD9AOD6VLcn6rieMcDsdIPoDcQnePdI2pdY\naOyneS/HJknaFng0lb8r8M80dKYtMYPSh8DNaZuAS9MbiEyuAP5P0pj0fdwvDQMCuBu4lHk3fJ8B\n/CMdX0YsCrc9tf+GTgHwAkOO01SkXYm5sNemVD9IGSiEMmAjhiR9gJNpM2d/m1X2apHVmssew/bY\n+v3e71/9+WKfL9VxRscPvun2zRdWlvG+/rjM4vzSvQdLjHmvHjHZ+JboLZ+eV4XnA0JQH+Lmsg2S\npLD7I5THAkOO4zj5pJRsTveYO04TuHaQljoQLhXsXgijPH3ceAQxu8HMhsbnSAd+KzuZjrPX4tc2\n29isstfzJLdZ7LX5Xut/vOTHoz4Y+EHftcet/djqn62+5k3P3PRdzXESZUA3M74vgprzHSFIRE/a\nxYU2yh3HcZzccMPccZrAtWtxQbvZvLvfm3N34ecNSecRNyteTIw5bLJhrhDWB36xJHnLkuQXiW+B\nW8woulH+p83+1P/TXp/+e+zaY9dc89M1X9p+9PbDbn765jqzwFjMn+5Gef7YDViGmC7OcRzHKQHc\nMHecRqJKbcjSJP9dmlX3K8wSjwDnmVk+jNBliZuIkNgC2BhYLQ9ym8w+Q/fpPWGxCbe+vs7rm6z2\n+Wpjtxu93Rq3PnVrrfLV1UhsA8w24/G6xjiNIwR1J8an7pIkJVOlttFIWhR4Kj1dgpidYzIxDnZd\ny1NlQkn7AxcAXxBT3F1iZtfWM74PMV/1AElrE3NpH53DOosAe5nZ/zUwbq78HC/BcZxWghvmjtMI\nVKn2xJLuR1t583O5Kg1sy2wzs2ebLC+E7sDhwFmWJGZJcmdch87AlcBhZsVJd7Xv0H27f9X9q1Gj\n1x49fMVvVvx0+GvDN73tydueb3gmP+AbjvLNlsDdSWIvF1uR5mCFKwVfayngNjM7Ks348Y6kB8xs\ncg46jibG8edCd+JmxXoN89aAvGS74zQJT5foOI3jROJO/HuaK0hSd+A1Se2ardXvTAN+pPZn+zTg\nv2Y8mse1cuKwwYd13HrHrW96ZOAjkyctMmntLd/ccqfRN41eMUejHDNeNuOVQuu5IJEkdjvQoAe3\nFSJJm6Ulz8dIuk5S+7RjvGJZ+NfSvn6SyiR9KGmxdEyZpI9ST3wt2QCpMT4O6CPpRkm7ZCz+U61J\n85al3zTV7Y201HnXGsPPBf6Q9p+XlkZ/KkPn7bPIXyGVNUjSHxTL249WLPXeLx3zR0ljJb0pKeuN\nv6QL0jFjJO2Wtt0maZuMMTdK2jl9ny6Q9KpiGfgRGdf6vKQHgHfq+iM5jlM37jF3nBz517raeOBa\nHPfGkqxh5c3f8GlmP0jaobmbOxXCMcDbliRPWZLMAi6bp1+sAfwZaNHH3vsP2b/Nd12/u3j0uqMP\nXWT6Ij8l7yYH3f3o3TfkOl9iOeAbM34roJoLLElS+pmEmkBHYhq6oWb2saRRwKHEdHAGTDazQZIO\nBf5qZgdJuoWY4/lSYHPgzdQTnxVJKxCrb31Mbin/MjkeOMzMXk7T7dX8v30SsKqZVT8FaAPsZGbT\n0puHl4kl56t16QfcRkx/N1bS08DB6bWvR9zcuxmxtP0wM/tasXplzWvaBVgDWB3oCfxP0nPEHNe7\nAY+kNzhDiUVo/gJMMbN1JXUAXpD0RCpuYHoNJVE12HFaG26YO06OHDKahTf8nIvX/KbxJd3TnLrd\na/7gm+WlPPwLwJfZ16UNcA3wt5YqHLT/kP01pfOUU99Y9Y2T28xpM2eDDzcYee8j9zZlg+FRxBzD\nD+RZRWf+pQ3wiZl9nJ6PIoZ2XZqe35v++zqwc3p8AzEn+qXEG9hsN48Cdk8Ly/wGjEhvrBur34vA\nJYpl5u81s5qf25oCy4gFZjYmVsdcSlKvtK9XqvdOZvZ+6n3fgFgivnp++4x1R0m6k9/fg0w2BP6d\nhtVNSr3q6xDzZ1+aGuVbA8+a2W+ShgEDFFPGQiyUsyIxR/arbpQ7TtNxw9xxcqTNHPvPmvCfxsxR\n/IUcBpwDvEKMH20WCmEF4HJgeBpHXl/86uHE4hzXN3fdXNhl610OHdNvzLnTO0zvMOiTQf/o8VOP\nv91YdWNTPbPZSmM7TkOoxnHm/79qD/XcEu5mNkHSRElDicbonllkGnC7mR1Vo30WadhYevPdvubE\neYSYnSfpYWIJ9RclbWlmH9QzZW9gMWAtM5st6VPiUwGIm7o/I27ofj/VY0q1t73GuodKWjdd9zVJ\ng7JsLq/1vqVGeCDuSdiN6J2v5ggzm6fyqqQEL9nuOM3CDXPHKSxLE9PRVdKMuHSF0BaYY0kyB/gU\nOM6SpF6DV2IZ4iPsjcwKu3lyty132/3dpd+9bNLKk3qsPW7t63tO7Xn4qKpRzQrRKbTOCxIhqAxY\nM0ms6GkyC8xsYuz3H8xsHLFkei6bqa8FbgFG1dyMnSKylyMfDwwC7iJWRqx3v0iq1zvEzaPrAP34\nvTQ6xD0iNUu2T0qN8iHAchl9M4he/8cl/WRmt0n6VNKuZnZ36hQYYGZj0nVfBV6VtDXxeynTMH8e\nODgN/VmUaOwfn/bdARyUXmd1IqrHgcMkVZnZLMW6C1/Ud+2O4+SGG+bOAk9Q6EHc1LkX8VF43qii\nCsMQuhS4NCg0Sc4DC9G9469MDzPDb1VpW6BuWQbcT9sebbGZXZj9TChQPbPPFvuszVnJHT0mrv7f\nNuuNG3TvoE8G/fmmMOphoM8o+AhA4lngL2Y5n78AjDDj3cJovUDyF2C/ELTRfBpbXs0vxNLld0lq\nSyy7fmXaV7McfOb5Q8QQlrr2QNRVivwa4AHFEuu1SqpnOT46NbDnAG/DvJuxzew7SS9KGktMm3o+\n8JBiafTRwHvzDrfpiiXbn5Q0jehh/z/FsvTtiB7uMcD5kvoSby6eMrMxNda9T9IGxLL0BpxgZpPS\n7ieAm4H7M7KsXAv0AV5PbwAmATvV8z45jpMjKtVK4iqh8qjO/ElQ6ErMTHEs0Zt9CdFjNZfl2py3\n6i+z154wic2mtqRuv7WHx7aixw4PRq/Wk5uzyCbP8WOHHDNOH8/q23xJ57+ey5gt+zA9X5VD53LN\nZtcs8eDaDx7zc4efh3f6eFjo+tyJZ0+cMHQ0gMTiwPdmsTBSE84XM+PbfOu8oBKCliQaXEOTxN4u\nlh6SDPjWzHoWS4e6UMw1fpGZbVpsXRzHaXlKyeZ0w9xZ4AgKHYARwEggAKclltQqSX7Gpup5wBt8\n+Vk39tvwc7utZn826trk2VgUwjpAhSXJ8EbPFd2Iqcp2M+PF5uhRS3alFgVOJm6SuxY4z8rzUgjJ\nKRAh6E7gwySxU4qpR6ka5pJOBg4hFvZ5qdj6OI7T8pSSzemGubPAEBTaAH8CKoB3gb8nlrxZ1/j7\n++vVrjNZbPNxtkKua0jaEDjezHZucHDNuSGsBnxgSTIzPW9jSTK70XIUi5OYcWhj59Yps1ILAccQ\nnzDcBZxBhXUAxnsseOkSgrYDLgZWTxL7pZi6lKph7jiOU0o2p8eYO/M9QUHE+Mczge+AfRNL6i1u\no0qttdsAVlp9Iutv3oi1zOxFSf9toqonEWNKxwI00SjfiLgJbdUm6jCvvEp1JHoTTwaeBta38piK\nThXcT7zJqfPmxikeIUjEzb8HF9sodxzHcXLDPebOfE1Q2Bw4m5jGbCTwaGINZDOpVFtiasPLrdxu\nLJRuCmEJYCVLkufyIk90AN4ATjVrXmXS9D3YFygnbh77u5XPu2FMQu4tL21CUOcksenF1gPcY+44\nTulSSjZnQT3mkq4n5k2dZGYD0rYexPRLyxFTTe1mZlMKqYez4BEU1iXmDl+W6DW8M7FkTo7TjwSm\nEouTFJLexMIeeTHMiZllPiZ7AZGcUKUE7AKcQcy0sJeVW9Y4dTfKS59SMcodx3Gc3CioxzytVvYT\ncFOGYX4+0WtyvqSTiBvlTs4yt2TuXpzWQ1BYlRiysg5wOnBDYkmjspKoUv2AmVZun+Q0XuoEzDCz\nekNPFIKA44ArLUnyWoRDoh+xut9AMxpfmTQa5FsQny6UEZ8uPG7l835BpJVEHwL2N2NSLUGOUwfu\nMXccp1QpJZuzoB5zM3teUp8azdsD1SmpRhGzYtQyzB2nMQSFPsQiPlsT47T3SixpXFxtzMf7V4Nr\naNxTnAuAb4g3BHViSWIKYSbQlTxWx5MQcBVwehON8vWJTxeWAk4B7rFyy/p0wYzZEqcAk5uhsuM4\njuM4WSjG5s/FzWxiejwRWLwIOjjzCUFhCeDvxOJA/wL6Jpb82ERxbYjFMXJ+/C9pfWL1vaybLRXC\npkA/S5KrASxJ/tlE3erjAKAz8fpzRpUaQLyZGEi8qRll5XMLiNSJGfN79chWSwhaFPgxSRr+OzqO\n4zilR1GzspiZpY83F1gUwiLAwZYk5/t57udVQ7gaOMHg0NfX4p1Br7NyYskkhbAIIZzYJPlmsxTC\nNcAxFr3uDc6nW7dBDB/+tN1yyw9Z+2OZ6oIZSWlxnnOBLczIKYvLnter2+0T+CewZTp3dyu3X0PQ\nCiHoz1mmfDpkiN0A7AncXr1OCFqBmM+81vgksetqNvr4Fhk/DLiamGPecRzHaWUUwzCfKGkJM/tG\n0pJQd5yqpIqM02BmodDKFYlf/Ty38+0foNPAN9gQOAF46JX12WTkOQy1JJmUy/xa5+9ULEv3tYaq\ncsiFGeEbuc+/775biJ72rP2WJOOAcRSOS4DrzXiroYFpBchTZsxhV2Kp8r5WbpkVTedQ+1oBZgAL\nE+P2b89xfDZ8fOHH30rhNy07juO0aiQlQFJkNbJS8HSJaYz5QzU2f35nZuelFde6+eZPpyGCQnvg\nQGKGlReBUxNL3m+qPFVqcWI8dXUIzFlWbr/lQ9eWQmJr4HJggFnd4TchqDsxY8uIV7/n4ZPHMtxg\ngJXb1y2lq+P45k/HcUqVUrI5ywopXNJtwEtAP0kTJB1A+thd0ofA0PR8gUMh7KgQtiu2HqVOUGgT\nFP4EvAfsCGyfWPLHphrlqlQ3VepMYuXP2ZuMZw2r4EuroFGZW4qNRBfgCuCQBozyPwEfAot9N4O1\nThpLP4ORjTHKJbo2X2PHcRzHcRqi0FlZ9qyjqzHFFOdXGp09Y0Eirda5LTF930/AgYklIQ+idyBm\nH1nLyu0zYrjUGphdlasASWsC75kV1cNeCTxvxpMNjHsX2DBJ7ENV6nBgJo2IP5ZYmRizvHGTNXUc\nx3EcJye88qdTcgSFhGiQL0TMuPJQQ9U6m4TUD3gBGIjZF7lP09XAFWZWlFL0EmsBjxBDWHJKW6hK\n9QbeBDaxcnuvket1rs8r7zi54KEsjuOUKqVkcxY1K8uCiELoC3xhSSNzbC8ABIVBRIO8L3AacFti\nSU6ZRmqiSpUBZQ2k//sE2KoxRjmAmY1oik75QKItcA1wUrVRHoJEfBLwXJLY93VMvQy4orFGOYAb\n5Y7jOI7TMrhh3vL8mbh58eFiK1IqBIX+xBLwg4l5ta9LLKkrE0W9pBUstwXOSl931DnYbCbwWlPW\nKSJHAlOAmwBC0FDizUwn4kbWWoa5KrUTsEranxMSg4ChZlyQB50dx3Ecx8kBD2VxikZQWBYoJ1aD\nvRC4LLGkyd5ZVWpTYgXLuSEwNUvKt2Yk+gCjgfWrqtSdaJAvT8xUc0eS1K7WqUotDLwD/MnK7dlG\nrLUk0M+MkAfVHcdDWRzHKVlKyeZ0j3kLoRBkSQHipFshQaEnMBLYF7iSWK1zSlPlqVLdiPm1VyIN\ngbFya1IITJ1rSP3M7IN8ymzc+oiYheWiqioZcC/x6cL1SWL1ZZQ5G3isMUY5gBlfA55O0XEcx3Fa\nEDfMW457FEKFJcmYYitSLMJeIxZn7ICR9O6+LwtNe5DtH9yarR97J0ns51pjg5YglpmvycQs439c\ndWHuH9mfqqU6MRNYLgRlHy9tA/whVHFXpvyZM9tz//2HLfXSS9v/8OabQ2rE/+/XXyq7Zaed9tmn\nV6+lfsrsefXVrb6tPR522eUfiy266NedarY3Y/xQYBngwiSxmSFohQYMclSpDYCdgVXrGzfPHLE0\nYGZ8mescx3Ecx3HygxvmLceJwKfFVqJYhGfK9mKPrjfyx7tmsNC072k7e2NiCr5DgcezTKkglhev\nSa3xVm4WgtYkvscNjX+T6AmeK/+33zp1nDq1e4+NN76v7ZtvJpOBDMN5FvDWUsOGrWt77PHs/TWF\nv/324G/nHR9ZccU3Fhs4sKqWod2M8TOAfc1ivvUcjPL2xDSHx1q5/VDf2BpsAixOrCjqOI7jOE4L\n4jHmTosQdjzmDN7vvyMf9F+zGZlWegJrWrk1lLs7N3kx7eDZZIbAGLPnHaMuxA2X51mpfliyoEqN\nBDYEtp2f4uyd1ovHmDuOU6qUks3phnmBUQgHAbdbkkwrti7FIigsDowFhiWWNDr3d7qB8XjgCOAq\nK7eRzdFHoh8xC8xGxDjta81oUhaYUkSV6gu8DAyycvsspzlCZpTml4EzX+CGueM4pUop2ZweylJA\nFEJbYDliPMR8T5pPeyvgxSSxqRldlwA3NNYoV6U6AYcRQ1QeA9a2cmtyOJDEMsQsMDsAFwEHmFEr\nvr01k6aLvAo4K1ejPOUiiZfNuKtAqjmO4ziO0wBumBcQS5JZwCnF1qMlCEEbEVMVLgr8kZiij6Cw\nNbAe8JcmiL0a6AoMtXJ7p6m6TdAyx/6Nc9aHP21ONFpXMqMxcdetiX2BhYkFhRrDmUC9ceuO4ziO\n4xQWN8wLhEJY1JLku2LrUWjSTZdnETN/lAO3JElMVRgUuhBT/B3cxPzkf7Fy+62pukksvCEvnPEA\n04/8kJVuAVY145uG56kbsJWZ3d7UtYtBGoN/PrB1AxVPa2FWuzCR4ziO4zgtS1mxFZgfUQg9gGcV\nQvti61JIQtBKwKPEMJN+SWKjqo3ylArghcSSJ5oiv6lGuURHieOAj07ntO2msvD5r9q6++ZilKf0\nJKYmbG1cDNxs5fZ6rhMk9pFYroA6OY7jOI6TI775s0AohHaWJPN9aEAI6pgk9mutdoW1iEb7aokl\nk+uar0ptSDTgD7By+6I5uki0BfYneu5fA04x9CXwE1Z/esHWjio1jBims5qV184LX+c8cRjwgOct\ndwpNuvlzipl1L7YujuM4mZSSzemGuZN3gkJb4BXgssSSUdnGqFJrEENgViMa5jc3tVqnRBmwKzFO\n+gtgpBmvNEVWa0SV6gy8DRxu5fZosfVxnGxIegH40sx2L7YujuM4mZSSzekx5nlEISwPnGRJckix\ndckXIagrcCzQMUns7zlOOxL4EbipZocqtSxwLrGS5dnALs0IWanOAnMWMBs4HHhqAUz7Vw680hij\nXGJJM74uoE6OMw9mtlGxdXAcxyl13GOeRxRCZ2AdS5Jni61LNSGoF3Bnlq7JSWJ/rG/85GlderVt\n91v/NyYuNOvG/608bcK/XhxbS8rCE9qxx05zS74vObVH2f/dd3TXI7e85YcJD95We/zib3VizRt7\n8eT0MuZs8wPsMGXeAUf2gZe7xeNTx2fv33xK2t4DaAecAlod2BSgDCpmm4XMWZIqgSrL3r4pYMCJ\nZva/WjqXABLnAv8x4/n0/EzgcSo0DXiCy9+9lW9Xvrdmf7bz9IbmMeBAM5oVPuQ4juM4rZ1Ssjnd\nY55HLEmmA0UzykNQuyyl2qcSQ0VqUpeXeipQ8d20ru3PfqPbwz9/tVr46PO+DzKz0wzg3drDrT0T\nBq8CIIPznt9ixFvdpnwyYfqyj2QdP3ENePwS4OI+0GkKUMPwXr4PKDXMO43P3j933kzgv2bMkngf\nqGoHGgfnIu2CWWbc9J1Atlj3O4Gq9LikMpNItDObm8LwRiAzy8/tdPt0MvAwcDLfrvwqMHGe/jrO\nzTCJvc34tmDKO47jOI7TaNxjngcUwqLA8pYko4ulQwhaAxgFrJUkNqe58rTrno/S/dO1+XirXlZV\nkdN/kqCwOzFv+6DEkrmVNCWVAb3NbEJz9coJaQXMPmmRtQqExNrAmWZsVeeYSh0N7EjM816aH2TH\ncRzHKXFKyeb0dIn5oT+wXbEWD0FtgGuAy/NilIsV+aVHwvhNd26EUd6dWOFzRKZRnjICuKG5euVM\nKzXKJRZNw0wAXgf2rHNsjNU/FTjYjXKntSFpR0lzJPXLaOsjqXb4W9Pk3yjpE0lvSHpN0vrNkFUh\n6fh86JXKW0PS1jmM219SYwuFOY7TyvFQljxgSfIi8GIRVTgc+AW4vrmCUsPwSh751ynV8ck5ch5w\nb2LJy/PK01LAGUDSXN0WAP4NnAy8YcYcyF6dVJUS8C/gH1ZuH7agfo6TL/YkhmHtSfZQu+ZiwF/N\n7F5JWxBTia7R0CRJArB5HyXn+8Z3IDCImE62Pkr+hltSW7PGFTNzHKd+3GPeDBRCW4VQ1EcfIWgZ\n4DTg4Hx4y4F9iJsqL81ZB4VNgK2BkVm6uwGnmNk7edBtviIthNQvo2m4GW/kMHUXYAVilU/HaVVI\n6gqsBxwBZE2dmHrPn0u93a9J2iBtTyQFSXdJek/SLfUtlf77PLCipC6SnkrljZG0fcZaH0gaBYwF\nlpH097TteSDTq/8HSY9KGp3q109SG0mfpP3dJM2WtFF6/pykP2TMbw+cDuyeevN3k7SOpJckvS7p\nRUkrZXk/hqdjFpU0LD1+TdKdkrqkY86V9I6ktyRdkEVGD0n3p/0vSxogqUzSp5IWyRj3kaSe6etu\nSa+mr8Fpf4Wkm9P0l1nT4TqO03TcY948jgA6AecUUYfLgH8mib3fXEESPYELiAZiTl6QoNABuBo4\nMrFkas1+M3uXrJtG84jUG7gW2BZrWi70IrEhsD1wNEAu77kq1Y1407SblVvNkCHHaQ3sADxmZp9L\nmvz/7d13vBT19f/x15smIAgqiCIaRAUVFUGDFR2KYC8xRo2xl2iMPcZYL9de4i/ma4sascUWu8ZY\nkYnGioUqGsUWC4gGCzYQzu+PmYvLZXfv7sxcdoDzfDwubJk9e3bv/eye+cxnPh9JA8wWWq12OrCt\nmX0vaW2io0k/je/bCFgP+Bh4RtKWZlbuiOXOwASio4q7m9lXkroAzwEP3FcLpwAAIABJREFUxNus\nBexnZi9K2phoh6Ef0axPrwAN5w9dA/zazN6StClwpZkNjYv49Yh2mF8GtpY0FuhhZlMbEjGz2ZLO\nADY2s2MAJHUEBpnZXEnDiKaR/TnxjoWk3YmmrN0+zuc0YKiZfSvpZOAESVcAu5nZOvFjlivyPtQD\nL5vZbpIGAzeZWX9J9wO7AzfEr+kdM5sh6VbgT2b2jKTViWZyWi+OtQ6wlVmyqW6dc6V5YZ7OFcCy\nNc7hEuDFDGPdYkY1J7GeAkwJLLgvoxySuAwYm/eiPB4mtAvwTzPmmDEaGF1lmAuAB6yubCHiXJ7t\nQ3Q+CsCd8fXGhXkb4HJJ/YjWKFi74L4XzewjAEnjgJ4sPJRQwMWSTgc+AQ4BWgLnSxoEzAO6S1op\n3v49M2v4HB0E3GNm3wHfSXogfq5lgS2AO+MRLw15QtQrvzWwBlFHzWFEM3QVm35V/NibD9FRxZsk\nrUU0fKXwe3kIsAnRTsosSTsRFcfPxjm0AZ4lWjfiO0nXEQ0R+keR590S+BmAmY2Je987AHcQHXW9\nAdg7vg4wDFi34LV2jN8DAx7woty55uGFeQoWBHNYaDq/RSsIrJpx4CVp6GnHsuouw/hw04UOo5YS\nKlyXaHx7/yxySCT6MusF/LJmOVRnCFERUvUMNarXVkS9f32b2ta5PJK0AjAYWF+SERXLBpzUaNPj\ngY/NbD9JLYHvCu4rLAjnUvx7bP4Y84LnPhDoAgyIe6ffAdrGd3/d6LGFhXPD5RbATDMr9nn3FPAb\nYBWiIvckovNqniqRW6GzgdFmtruknwBhwXZTiYr9PkQ98QCPm9lCn3eSBgJDiXrbfxtfXmizIrc9\nTzTUpwvR0YyzCrbd1GzBI3Nxof5NkTjOuQz4GPMEFIZ7Kww3rXUeWdGQM1Zkw1supt+NfzFjViWP\nCRW2IDqsOzKwYIFFahTNOrB+c+S6ELO3gE2IerdyR6KfxE4QzR9uxrFmiYryhiFDx1id1XRn0LkU\nfk40hKKnma1hZqsD78S92IWWA6bFl/cnKuCr1bgIXQ74JC7KBwM/KfG4p4DdJLWNh5nE7de+inP9\nOUQnisY9+hAdtdwCmBv3JI8Hfk3xwvwroGOjvD6KLx/UKP/3iN+zeKjMC8CWDePW43Hza8c92Z3N\n7GHgBIqf6Po0sG/8uACYYWaz4hNd7yU6ivGamTWcdP4YcMz8ZH58rc65ZuSFeTJfEn24Lhm6Tn6Q\nmb0+toeuPKvpjec7lKin6i9F7luLaAziopHhrAASy0rsmdV1oveoXQapnQy8CdzT1IbO5djeREVg\nobvj240fe5OvBA6Ih6r0gQU6DBr3OJeavaTx7bcAm0iaQHSS+5Ri25rZq0TDOcYD/2TBoYL7AofE\neU0iniY37lV+n6j3GaKCvIOZFZv+cQywXsPJn0QncZ8v6RV+PILQkJOZ2Rvx894JdAAOBG6TNJ5o\nGEsfokL/wfi2p4mOODQ2Etg43uY84ICC++6In+OOgtuOid+v8ZImE+1oLPR+Oeey5QsMLeU04sQ9\n6HfTnbxyWD974ryK5hAOFa5CdDLVkMCCTOYdzguJFYHTzaIvtrTXM8mpXusA/wb6W90iWqTJOeec\nW0rkqeb0wrwKCsNewLsWBFlMS5hIGOoU4OMgsBvSxtLgutasc+//mDbgHrv3hgOafkScg8K/A28G\nFpyWNodEpNZAy6yGr0isC5xkxsFZxMuS6tWCqIftLqszX2zEOeecy1ieak4fylKdc4nOkK+JMFQf\n4ETgyUwCTttoJN92+Z7Pe1ZckIYKdyI62fOcTHJIZlfg0gzjvU106DyPDiI6QS2v+TnnnHMuI95j\nXgWFoSwIavKGhaFEVJDfFwRW8eI/pUisBrwKbGVGRXOghwo7AJOBgwMLFpjmL57/doqZLTSXebOQ\n2mH27SJ5rhpRvboRLXgyzOpsQq3zcc4555ZEeao5vce8CrUqymMHEp34c3naQPF82pcDl1ValMfO\nBsYUKcq7APcDFU+1mFoGRblEd4mtskinmVwKjPKi3DnnnFs6+DzmTVAYtiBarOEwC4IPa5FDGGol\nooVlRgRBJovo7E5URP+i4hwU/pRoIZBi0yBeAtxmZtUsTJQHaxCtJvjvWifSmOq1AzCQaGEU55xz\nzi0FvMe8CfGJnqfw4zyzNUkDOC4IbFzaQBKdgP8DDjejopXbQoWtgWuB3wUWfFpkkweAM9LmVlZ0\nwmemzHjGbP4KhLmhenUgGlN+hNWZL+ThnHPOLSV8jPlSRhv+7T4m7T3d5rX6ddNbR0KFJwHbAiMC\nq9FwHukm4Eksg9loxDLAHDNqNrtOOarXJUBXq7P9a52Lc845t6TLU83pQ1lKUBiKaBGK2ywI5tQ6\nnyxou+N/zYjbdqbr670qnVQlVNiLaHGbgTUsyrcFtiZa8joLxxIdLbogo3iZUb02Bn5F8SFDzjnn\nnFuCeWFeWgeiaQFvrXUiWdDgM5dlg4f+zOu7Xmqjz3mvkseECgVcBVwUWPB282ZY1iDgN5jNanLL\nylxCNAVhrqherYiGDJ1kdTaj1vk455xzbtHyoSxLiFBhJ6I5zo8l2qlYwDyshcloYZoXTcpSEQFj\nga0C+/GogaSWwJnARWb2derkHQCq14nA9sC2VpfThumcc84tYfJUc/rJn0UoDDvW8vnDUK3CUFeH\noVZocluF7eIx4G8CqwMbAW0Kf7Yb8uiO2/5uL9t569FbCrVpfH+Zn9bAZoVFeawVMANYrOYRlzhc\nYvNa51GM6tWT6CTjI7wod84555ZOPpSlEYXhmsCtCsPNajhv+dHA2sDMUhvEM6UcTDQbyotAEFjw\nWuPtJESn665l6oj7vgnPfTaL5MzsezKYT70GpgLTap1EY6pXw5ChS6zO3qp1Ps4555yrDS/MG7Eg\nmKowDGq4wmdP4DRg8yBYuOc0VNgC2As4C3gX+FlgwYtlQh7IfddPY5uz98o82eYk9QH+Q4ZjrcwY\n3fRWNbEX0AP4Y60Tcc4551zt+BjzHAlDCXgI+HcQ2HkL3BediLkDcC7wPXBq4xU4G5NYiWhJ9+3M\neLV5sm4GkoBHgF9j9m76cPQF3jDjh7SxsqZ6rQBMBna3Onu+1vk455xzS5s81Zw+xjymMOynMGze\nRXKa9gtgNRr1nIYKBwFPAxcB9UTjvivp/f0TcGMWRbmkTpKOTBunItHe4nZZFOWxU4C+GcXK2kXA\nXV6UO+ecc65mQ1kkvQt8CcwF5pjZwFrlEvuI2i/N3g44PAhsNkCosD9wHrAOUAfcElgwt5JAEtsB\nmwMbZJTbeUQngy4a2Q5h+VVWsbKkegXACPK70+Ccc865RahmQ1kkvQNsbGb/K3F/bg4rLGqhwt5E\nY8i3IRq6cm1gwfeVPl7bnNOFVw55mVmrHG7Go2nzkfRT4H6gr5mVPCE1byRkRi7HaqlebYHxwMlW\nZ/fVOh/nnHNuaZWnmrPWQ1lq/iYoDFdVGPasdR4AocIeocJrgWeBCcBagQWXV1OUA7DKy/9k58O/\nz6Ioj40Htm32olzqnl0oWgJPSXTLKmbGTgUme1HunHPOuQa1nJXFgCckzQWuNrNra5THZkAv4OIa\nPT+hwi5E46APAq4BegcWFD2S0BSNOHFv+j2zMS8fmtUQFsxsNtEJis1H2gq4BWktzBrPm141M+ZK\nHGTG9Ayyy5TqtR5wJNGc884555xzQG2HsqxiZh9L6go8DhxtZk8X3J+bwwrNJVTYETgBOAa4Azg7\nsODjpPE0uK416939OR8OvM3uG3VoVnk2O2kZ4FXgTMzuqnU6zUn1agE8BdxqdXZlrfNxzjnnlnZ5\nqjlr1mNuZh/H/8+QdC8wkGjmkfkkjSy4GppZmNXzKwxbWBDMyypeNUKFbYEjWePt0xj44uvcsffA\nwIKpqQMvP/Xv/NB2Nl+sfliaMJJ6A8PNbFEtItQZuBO4O22geAjLkcDVZqTueW8GhxENIftLrRNx\nzjnnlkaSAiCocRpF1aTHXFJ7oKWZfSVpWeAxoN7MHivYpln3XhSG5wOvWRDc3FzP0ViosBVwAFBH\ni7mv8uDOPWj/7aggsCvSxtZ6d/dlpyMnMu7A7eyxix5r+hFlYkVHMYab2S1p81rUJJYjOgpxlhk1\n2fEqRfXqTjRef7DV2aRa5+Occ8457zEH6AbcG60jQyvglsKifBE5f1E9Ubxa5x7A2cDHwF6MHrYB\n0Zjy1D2nEoI9LmTGuhfYjPVSv49mNgNY7IpyADO+BEbWOo8S/gxc7UW5c84554rxlT+bUbxa5wii\nKQ+NaCaOxxkzeGWintOhQWAT0z6PxC+AM4EBZsyu/HHqAZwB/NvMFtmRg+YQ7Zywohmf1jqXYlSv\nXYBLgA2tzr6tdT7OOeeci+Sp5qz1dImLnMLwVwrDtZr7eUKFWwAhcClR7/xPAwseCyyw+LZrMyrK\nl4/jHVZlUT6MaErGmcBDafOomtQNaVCGEfsA98QFeq6oXh2By4Ffe1HunHPOuVJqOV1irSwDlRew\n1QoVbgicA/QD6oGbAgt+mH9/qOWIVtA8J6OnvBC414znqnzc88D6ZvZRRnlUaw1gCxqd8JuUGa9L\nDMnpgkLnAKOtzp6sdSLOOeecyy8fypKRUOGaRKt1DgUuAP4SWPBdcz6nur4+nE/XGQX0NeOLkttJ\nbQEzs+oWKnKpqV4DgQeAvlZnn9U6H+ecc84tKE8151IzlEVhuGpzxA0Vdg8VXgW8ALwOrB1YcGmz\nF+WDz+zIPjs/yKZ//lO5ojx2GbBjc+ZTKxKbS5xV6zyKUb1aA9cCJ3pR7pxzzrmmLBVDWRSGAm5X\nGB5gQfB2FjFDhSsAJwOHAqOAPoEFi6746vr6/cxa5TN7/thLKtj6t0twb/nrwNxaJ1HC8cA04NZa\nJ+Kcc865/Fvih7LEM6P0n9OKdq1/yGz88WCioutuotU6P8gobkU0/KQd6X/9g7x82EB74vyX5t8e\nzT+5iZmNldgMGGsWFa15uH7ZRsu2/vtG33R6uif/44Wj+zLgr1No/W0013ja6/nTCbgZGGh1lsnO\noHPOOeeyl6ehLEt0YX7CLuEJwx/jwLbf0x74JKPUAN4iKsjfrGTjMNRwYGwQ2My0T6zBI1vS54HP\nmL7Bo3bvjXvNv11/OhTOOgA+7wpsDnYPsL0Z30X3M6ZW108bqrWuXGbfV9jwlg5t5vLmJx34jOkb\n9KXr5NdoMS/6A0x7PZ+utTq7vtZJOOecc640L8wrkOZNChUOBM6ftSx9x23EX7Z6hrMCC2rSsxqG\n6gW8CAwMgvQ9p9rlsOvp8cLPmbL7Cjamfv6S89L9z8N9j8INZ5vZD+ViLCqqVwfg+FZzOWH4VD7q\nP439zhltr9Q6L+ecc865Bl6YVyDJm3TuluHWq3zMlb3eoTPRDCnXBxbMaepxzSUMJeARYHQQ2EVp\n40msQtv/TaT/qP147ncTgU3MuC91ohlTvZYBjgBOAUYDdVZnb9U2K+ecc865heWpMF8iTv4MFfYE\n6reA7cOAZzp9wS/3+CzIw0IuvwS6AX/KKN6lfDf3ep47aUM4+g1YptkXSqqG6tVq74lc2mZd9pjd\nipeA4VZnE2qdl3POOefc4mCx7jEPFXZ7eQD39RvPeq3m8mfgksCCpqYOXCTCUCsCk4BdgsDGpo0n\nsRNwI3zRHzrvA1yal5lWVC8BPwPO6TqLH3b+D3+87n67sdZ5Oeecc841xXvMUwoVdgZOAo5Y6ROe\nuHNPDr76tmBKrfNqZEfg9oyK8g7AFcBV0Ok7M7swdXYZiAvyYcB5QEvg+BkdePS6+3O6t+ecc845\nl2OLVY95qLD9Oz05d6VPOHLZb7gFOCuw4L0apdikMJSCINkbLKkNHHMhbNYP9hoPLZY348CMU0zs\nuO10yFM9uXBCNz6f24JTgLutzvI6daFzzjnnXFF56jFfLArz3/wibP/zuzighXHGPPHsnXvy16vu\nCB6pdY7NQUKgbsCzsNrb9Dj0DYaEh3L7/ava9x0/rXl+9Vp/wEfcN60DPUdM5W+rfcGv68fkY0iN\nc84551y18lSY53ooS6iwJbDPkBW5+quOTOj0JbsMmRe8NKTWiTUTiRbAGPjml9B+T4KDJtH3tpn8\nd8vra12Uq169gHpg+Detue6QV7nwrCctF+P5nXPOOeeWBLkuzIFxwKxZHdjrwHeDf9Q6meYindXR\n7MyvzJgncSi0+8jMPtQe+z7I98t9y5c9jqxVbsMOUK/RvTgR2Au4DDhqyuX2Za3ycc4555xbUuW9\nMD8NePDAd4PcjLeR2A7Y0owz4ut7Av3GjNGZwPqDB1sfoJ8ZpxfeX/r6NVfB+gOBjQHMeBNA2548\niAGP7Mgrhw61MSObfP11g9X6w+W47Pke7DelCx/Ma8ECj+k5k57vd+K9Yre3MATwbmfebXx/p+6s\n3W4OV3zbmnWszmo+lMY555xzbkm1WIwxzxOJdsDaZkyIr3cG2o0Zo92BvXfc8fNdvvmmUzszPi68\nv/T1v68EU74yq5s/77oGjxRrPvIJn68x1u66bYey+UQzo+y6+ufcuPx3aKv3Oe9vG/LAF21Z4ETM\n45+j56j+vF/s9pbzosL82o15r/H9xz3HnD89YlOTvFfOOeecc3mXp5rTC/MMhKFWJRp2EwSBTS61\nnaSfANPN7Lty8dRi7i/Z8G/nsfzb69qY+pILJaleQ4imKmzX51Mu3HsSt40ck9NfqHPOOedcDuWp\n5vTCvEISmwJzzXip8X1hqLuByUFgZxZ/rLoBpwK/AnY2s2fLPE/DwkS7mvFisW122lc7PtSb44A1\ngDOAO3yqQuecc8656uWp5sz7GPM86QosVPyGoXYD1gf2LfPYXeP/1zWzT5p4nouBvxcrylWvddv8\nwAWdu7PDsrM57us2XGN1NqfC/J1zzjnnXI55j3kKYSgBrwDHBYH9K208iSHADUBfM76af3u9fgKM\nJFpN9OKVZnHF9Ivtm7TP55xzzjm3tMtTzdmi1gnknUTnaNGfhcWrem5VWJRLOlhSpwTP0w64Gjiq\noSg/dajWW/sY3UFU/H8ArG11drEX5c4555xzSx4fytK0i4CHgXuL3RkE9nWjmzoBnYHqFt/Zff+H\neGfwNHv1oAeP2UHdP+zIHWN+ypYbf8Rzb63IelZn05Mk75xzzjnnFg8+lKXJPGgFzDNbeHx5Zs8x\n4ne7sdEN9zBl1y3ZeNQgGScNfodPf/oRh1zweOkTRZ1zzjnnXDp5qTnBC/PEJG1NtBrmby3Fm6jB\nI1vS596ZiCmsPKEH8DxwhtXZa1nl6pxzzjnnistTzbnED2WRJOBI4B5GMgLoNv/OR9iYWXTgXV5m\nFlea2bQfH8clrL1FD4Y+9wYrM6vh9mM+pe89V9EXWAGoAwQY0gbnDeLA04aywJCT3aaw8hUPMav7\nLGYBkzD75/znqGt50Fqrr3L29M7TOnzVdu6XwG5WZ2Ob551wzjnnnHN5tsQX5mZmkn4AWgLLA13m\n3/kDXfiBDsyjS3x/oUdpM/EoRBeg7fZG9z90YpcWG9BhpfHcffrTbGFmswu2X6bVXFYE5hYGkbFi\nq3m0BdoCHSFe2XPZT0ay7oqnftzla4aOX/Wq+x9+76jMX7xzzjnnnFtsLHFDWVSvdkBvq7PxWeQR\nhloNqMPYteu/eLH9+xy4xiibkTSetj/2aH7y1Lm0+boNbw+9hK+7nW5jRubzl+Ccc845t4TzoSzN\nQIGW4QuuYRZ7si83AUckiiO2Bp4xY24Yqg/wLHANonffkTYzcX4jfrcPqz3zf6z7YWemDr+OL1c9\n2sbU++JAzjnnnHMOWAIKc9WrBXP4Ba9xA635htU52kbadYliRTOwHP0K/ddA4zoxhsuAdYIgRQ/5\ntiePYNUXr2XDKavy1vC7mLrtwTbmrMZTLDrnnHPOuaXcYj+URfW6HlifcVzGTG62MIMXJP0EaInZ\n24lDDPvDQFYefxOrju3N1G0f439r7WdPnp24wHfOOeecc9nzoSwpSepkZg0L+Pwe+NTuTVaQv3mM\n2ny4G/vP/qFNh+HDv78UALP3Euc25Mw+dJtwC5s8NYB3gud56YheNvqcd5PGc84555xzS4fFrsdc\nUn/gYjMblib+Oweq5ewV+ePMjTlqxg8rTd7riLeW/fbbjn3NSDTuW0F9d7pOvoVeo7fhg80nMX2D\n/eyJ8zM5AdU555xzzjUP7zGvkuq1FnAKcJKZvSpphzTxwlDbtdmRUa1n0WnFFzij2xefXPTttx27\nJSnKNeTMTiw/9SYGPr4T0zZ8j1cOHWyPX/ivNPk555xzzrmlT64Lcw3SxqzP8azMdsClwPcAjeYP\nT2L4vDac2OVf3L7GDdEhA7uBaU09aIHcBtctQ6f3r6b/Y79iZq8ZjN9/T3v0j/ekzMs555xzzi2l\ncj2URe2wDlvx+L77cfReqzETmBME9nnjbcNQbYBORcJE20vrAm9gNm/B56ALcJAZF1ec1+CRLenw\n0R9Z84mj+GbFWby/5Un2yKWJZoFxzjnnnHO15UNZKnTr9XyxcjcGAM/EN70A7FRk0wHAg0VufwFp\nZ+Ai4DhgaqP7WwOfVpKLBo8U7WecSt8xp2EtjP/scAbfdrnIFwdyzjnnnHNZyHWPeV72XrT9cYez\n+tMX0faLdrw99P+YtcrvvSB3zjnnnFv85armXKIKc2lFYDBmd5XfjHbAcmZML7vdiBP3oMfzV9D5\nvS68NfxGvlztNzam/vuqcnLOOeecc7nlhXkFqnqTpJZEs7YcB9wB/JYyL0xiBLCHGYcXvX/bkwfT\n/aVRdJu4Om+NuI+ZaxxoY876qvpX4Zxzzjnn8swL8wpU/SZJpwJ3YvZmZZsjMxZ48Rp2Sn+6TbiZ\nHs+vx9vDnuSz3vvZk2d/XF3mzjnnnHNuceGFeQUW5Zukoaf1outrt9LzXwN5d5uxzFhvXxt97luL\n4rmdc84551zteGEOSGqYm7wl8Fczu7DR/cXfJKkPcDBmJ1f/nBwGtDLjKgANOaMbK755M70eH8aH\nm05hWr/97YnzX07yepxzzjnn3OInT4V5i1o8qaIx4ZcD2wHrAfsommu8Ev8FxlT5fEF88V7gYQ0+\ns6N+tv+dbHzNR7Sf0ZtXDhluf/tn30qL8oJ4mchzvDznlnW8POeWdbw855Z1vDznlvd4ec4t63h5\nzi3reHnOLe/x8pxb1vHynFtzxMuLmhTmwEDgLTN718zmALcDuxbdUlpwD8bsG8weqfL5gujfui/Y\n5ZDT6X/9TJafujUTfrW33Ti6pz1+0ROJ4mUnz/GyjJX3eFnGynu8LGPlPV6WsZa2eFnGynu8LGPl\nPV6WsZa2eFnGynu8LGMtDvFyoVYLDK1K1PPd4ANg04W2ku4DQqIhLyn0W13bnvwXNrj7IGYv9y1T\nfna0Pfznq9LFdM4555xzLju1KswrGti+9v6rD327VffN5x209enMXaYdrb+Jpiyc17JVVdc3+bAr\nvf8xh6kjzuGbruf44kDOOeeccy5vanLyp6TNgJFmtl18/RRgXuEJoJK8eHbOOeecc80uLyd/1qow\nbwW8AQwFPgJeBPYxsymLPBnnnHPOOedyoCZDWczsB0m/BR4lmi7xOi/KnXPOOefc0iy3Cww555xz\nzjm3NKnVdIklSRolabqkiRnEWk3SGEmTJU2SdEzKeG0lvSBpnKTXJJ2fQY4tJb0q6cEMYr0raUIc\n78UM4nWWdJekKfHr3SxFrD5xXg0/X6T5fUg6Jf69TpR0q6RlksaK4x0bx5ok6dgEj1/o71bSCpIe\nl/QfSY9J6pwi1p7x650raUAGuV0c/17HS7pHUqeU8c6OY42TNFrSakljFdx3oqR5klZImdtISR8U\n/O1tlyZefPvR8fs3SdKFpR5fQW63F+T1jqRX0+QmaaCkF+N4YyX9NGW8fpKeiz9XHpDUscJYRT97\nU7SJUvEStYsy8apuF2ViJW0TZb+3qm0XZfKrul2Uyy1hmyiV2x1J2kWZeInaRZl4VbcLlagfUrSJ\nUvGStolS8ZK0iVKxkraJsrVXtW0i98wsVz/AIKA/MDGDWCsDG8WXOxCNa183Zcz28f+tgOeBrVLG\nOwG4BXggg9f7DrBChr+LG4GDC15vp4zitgA+BlZL+PiewNvAMvH1O4ADUuSzPjARaEs0tOpxYM0q\nYyz0dwtcBPw+vnwycEGKWOsAvYkW1xqQQW7bAi3iyxdUmluZeB0LLh9NtJpvoljx7asBj1T7N10i\ntzrghIR/G8XiDY7/RlrH17umea0F9/8ROD1lbiEwIr68PTAmZbyxwKD48kHAWRXGKvrZm6JNlIqX\nqF2UiVd1uygTK2mbKPm9laRdlMmv6nZRJlbSNtHkd3Q17aJMfonaRZl4SdvFQvVD0jZRJl6a74pi\n8RJ9V5SIlahNlIqXtE3k/Sd3PeZm9jQwM6NY08xsXHx5FjAF6J4y5jfxxTZERdz/ksaS1APYAfgr\nkNXZwJnEifeKB5nZKIjOCzCzL7KIDQwDpprZf5vcsrgvgTlAe0UnErcHPkyRzzrAC2b2nZnNBf4F\n/KyaACX+bnch2rkh/n+3pLHM7HUz+081OTUR73EzmxdffQHokTLeVwVXOwCfJo0V+3/A7yvNqYJ4\nidpFiXhHAudbtDgaZjYjZW5IEvAL4LaUuX0MNPRodaaKdlEi3trx7QBPAHtUGKvYZ++qJG8TRT/L\nk7aLMvGqbhdlYiVtE+W+t6puF2V+F1BluygT6wiStYmy39HVtosy+SVqF2XiJW0XjeuHmSRsEyXi\n/S/ld0WxeIm+K0rEStQmSsWLryf6rsiz3BXmzUVST6LeoBdSxmkhaRwwnWiv+7UU4f4EnATMa2rD\nChnwhKSXJB2WMtYawAxJ10t6RdK1ktpnkCPA3sCtSR9sZv8DLgHeJ5rV53Mzq3b11kKTgEHxIcX2\nwI5UUaiW0c3MpseXpwPdMojZHA4G/pk2iKRzJb0PHEDUs5I0zq7T8udDAAAIxUlEQVTAB2Y2IW1O\nBY6OD6FeV+mh4jLWBraW9LykUNImGeQ3CJhuZlNTxvkDcEn8e7gYOCVlvMnx7wNgT6Leqao0+uxN\n3Say+iyvIF7V7aJxrLRtojBeFu2iIN7z8U2J20Wj19qblG2ixO8hcbto9FpTt4tG+SVqF0Xqh8mk\naBMZ1yOVxKu4TZSKlbRNFIvXTN8VNbdUFOaSOgB3AcfGe72Jmdk8M9uIqHDbWlKQMKedgE/M7FWy\n6y3f0sz6Ex2qO0rSoBSxWgEDgCvNbADwNdGHWyqS2gA7A3emiLEmcBzRkJbuQAdJ+yaNZ2avAxcC\njwEPA6+S3c5Sw3MYFS6stShJOg2YbWaJd5QamNlpZrY6cAPRTmeSfNoDpxIdZp9/c8rUriLa0dyI\nqOfskpTxWgHLm9lmRDvWf08ZD2AfUuysFrgOOCb+PRwPjEoZ72DgN5JeIurhml3Ng+PP3ruJPnsL\ne8sStYksP8vLxUvSLorFStMmCuMRfR6lahdF8kvcLhrF+oqUbaLM7zVRuygSL1W7KPJ6E7WLIvXD\n4Eb3V9UmsqpHKolXbZsoFStpmygSbweiHawsvytyYYkvzCW1Jvpi+JuZ3ZdVXIuGdTwEJO0t2wLY\nRdI7RIfphki6KWVOH8f/zwDuBQamCPcB0Z7o2Pj6XUSFelrbAy9XeqizhE2AZ83sMzP7AbiH6P1M\nzMxGmdkmZrYN8DnRWMK0pktaGUDSKsAnGcTMjKQDiYZSJd6pKeFWoOKTDhtZk2iHa3zcNnoAL0ta\nKWkyZvaJxYiGjaVpFxC1jXvi2GOBeZJWTBosHo61O9G5EmkNNLN748t3kfK1mtkbZjbCzDYBbgcq\n7rks+Oy9ueCzN3GbyPqzvFS8JO2igtyqahNF4qVqF8XyS9ouSrzWxG2izO8hUbsoES9xuyjx3iVu\nF/HjG+qHjcngeyKDeqRsvDTfFWVyS/Q9URBvANGOZWbfFXmxRBfm8fi064DXzOzSDOJ1aTjcJ6kd\n0UkRFc+iUMjMTjWz1cxsDaKhHU+a2f4pcmuv+MxwScsCw4lOaEzEzKYB/5XUO75pGDA5abwC+1DF\nONoSXgc2k9Qu/h0PA9Iewlsp/n91oi+DLHovHyA6XEf8f1Y7hql7BRTNwHASsKuZfZdBvLULru5K\n8nYx0cy6mdkacdv4gOgEpsQ7NfGXXYPdSdEuYvcBQ+LYvYE2ZvZZinjDgClm9lHKvADekrRNfHkI\nkGisaQNJXeP/WwCnE/WyVvK4Up+9idpEhZ/lFbeLUvGStIsysRK1iWLx0rSLMvlV3S7K/B4StYkm\nfq9Vt4sy8RK1izLvXdXtokz9kLRNVFKPVNMmisZL2CZKxVqrYLNq2kSxeM9l/V2RG5aDM1ALf4iK\nto+A74H/AgeliLUV0SHAcUR/AK8C26WItwHwShxvAnBSRq95G1LOykK05zgu/pkEnJJBXv2Izj4f\nT9QbkmpWFmBZopM9OmaQ2++JdhQmEp0w0zplvKfieOOAwQke3/B3O7vh7xZYgejEoP8QDZPpnDDW\nwUQnBP0X+BaYBjycIreDgTeB9wraxZUp490V/y7GEfUurVRlrKLtnWj2nWpmZSmW201xex1P9KXX\nLUG8+fkBrYGb49f7MhCkea3A9cDhGf3NbUI0BnYc8BzQP+V7dwzR0aM3gPOqiFX0szdFmygWb/uk\n7aJMvKrbRZlYSdtE0XhJ20WZ/KpuF2V+r0nbRMnv6CTtosxrTdQuysSrul1Qon5I0SZKxds9YZso\nFS9JmygVK2mbaLL2qqZN5P3HFxhyzjnnnHMuB5booSzOOeecc84tLrwwd84555xzLge8MHfOOeec\ncy4HvDB3zjnnnHMuB7wwd84555xzLge8MHfOOeeccy4HvDB3zrkUJM2TdHPB9VaSZkh6sInH9ZO0\nfQXxm1x6XtKBkuZK2qDgtknxglmpVZKDc8659Lwwd865dL4G+kpqG1/flmgVuqYWiehPtMx1Uypd\nbOID4LQEj6tE4liSWmaYh3POLdG8MHfOufT+CewYX96HaBVNAUgaKOlZSa9IekZSb0ltgLOAvSS9\nKmlPSR0kXS9pgqTxknZvCC7pHEnjJD0naaUiz2/AP4h2EHo3vrOwx1vSzyVdH1++QdKVcdypkgJJ\nN0p6rWGbgsf9v7gX/glJXeLb1pT0sKSXJD0lqU9B3L9Ieh64MOmb6pxzSxsvzJ1zLr07gL0lLUO0\nfPQLBfdNAQaZ2QCgjmgJ79nAGcDtZtbfzO6Mr880sw3NrB8wJn78ssBzZrYR8BRwWIkc5gEXAacW\nuc9KXIZoCfDNgeOBB+IYfYENJG1YkMNYM1sf+Ff8OgCuAY42s02Ak4ArC+J2BzY3s9+VyNc551wj\nrWqdgHPOLe7MbKKknkS95Q81urszcJOktYiK4obPXcU/DYYCexXE/Dy+ONvMGmK+TDRUppRbgdPi\nXCpKHWgYCz8JmGZmkwEkTQZ6AhOIiv474u3+BtwjaVlgC+BOaf7LaFMQ904zy3I4jXPOLfG8MHfO\nuWw8APwR2AboWnD72cBoM9td0k+AsEwMFbltTsHleZT53DazuZIuAf7Q+K6Cy+0a3Te7IPb3FTyX\n4ngtiHr4+5dI55tSeTrnnCvOh7I451w2RgEjG3qcCywHfBRfPqjg9i+BjgXXHweOargiqXMVz11Y\n0N8ADGPBnYPpktaR1ALYnepP5mwB7Blf/iXwtJl9Bbwj6edxvioY+uKccy4BL8ydcy4dAzCzD83s\n8oLbGorfi4DzJb0CtCy4fQywXsPJn8A5wPKSJkoaBwSF8YvEbZxDQx5zgD+zYGH+B6KTQ5/hx52E\nBfIvcrnQ18BASRPjvM6Kb98XOCTOdxKwSwWxnHPOlSAfAuicc84551zteY+5c84555xzOeCFuXPO\nOeeccznghblzzjnnnHM54IW5c84555xzOeCFuXPOOeeccznghblzzjnnnHM54IW5c84555xzOeCF\nuXPOOeeccznw/wGF6zZXZ/BGmwAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# define figure size\n", "FIG_SIZE = (12, 8)\n", "\n", "# create line styles to cycle\n", "lines = [\"-\",\"--\",\"-.\",\":\"]\n", "linecycler = cycle(lines)\n", "\n", "# calculate limits\n", "max_played = max(played for _team, played, _points in btm_team_summary)\n", "max_points = max(points for _team, _played, points in btm_team_summary)\n", "\n", "# plot\n", "fig = plt.figure(figsize=FIG_SIZE)\n", " \n", "for team, _played, _points in btm_team_summary:\n", " team_cum_points_list = team_points[team]\n", " team_match_list = range(1, len(team_cum_points_list)+1)\n", " plt.plot(team_match_list, team_cum_points_list, next(linecycler), label=team)\n", " \n", " # if there is a key event for this team then annotate\n", " if team in key_event_dict:\n", " # get match number and event description\n", " event_match, event_desc = key_event_dict[team]\n", " # calculate position of annotation\n", " x, y = team_match_list[event_match], team_cum_points_list[event_match] \n", " # annotate with arrow below event\n", " plt.annotate(event_desc, xy=(x,y), xytext=(x, y-10),\n", " arrowprops=dict(facecolor='black', width=.5, \n", " shrink=.05, headwidth=4, frac=.05))\n", " \n", "plt.xticks(range(1, max_played+1))\n", "plt.yticks(range(0, max_points, 5))\n", "plt.xlabel('Match Number')\n", "plt.ylabel('Cumulative Points')\n", "plt.xlim(1, max_played+1)\n", "plt.ylim(0, max_points+1)\n", "plt.legend(loc='upper left')\n", "plt.title('Cumulative Points vs Match Number For Bottom 9 Teams \\\n", " \\n(at {}, with key events)'.format(DATA_FILE_DATE),\n", " fontsize=16, fontweight='bold')\n", "plt.show()\n", "#fig.savefig('PvsPTS_btm9_27Apr2015.png', bbox_inches='tight')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "QPR and Burnley look doomed.\n", "\n", "Note the rapid improvement achieved by Pulis and Pardew. John Carver appears to have flat-lined." ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "## Finally" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Hopefully you've got an insight in to just how powerful iPython Notebook, python, pandas and matplotlib can be." ] } ], "metadata": { "kernelspec": { "display_name": "Python 2", "language": "python", "name": "python2" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 2 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython2", "version": "2.7.8" } }, "nbformat": 4, "nbformat_minor": 0 }