{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# Big Data Derby 2022\n## Introduction\nHorse races are often decided by inches. Owners, trainers, jockeys and bettors are constantly looking for an edge on the competition. There are an innumerable set of factors that can play a role in how a horse is prepared for, and then runs, a race. Decisions are made at each step in the process and stakeholders use a combination of information, intuition and experience to make those decisions. Racing technology in the form of GPS tracking has provided a much larger and more detailed data source that can be turned into valuable information. The Kaggle Big Data Derby has afforded data scientist the chance to explore a slice of this rich new data. \n<div class=\"alert alert-block alert-info\"><b>The GPS and fine-grained timing data are the magic in the Derby data set.</b></div>\n\nIn this notebook I will primarily explore the tracking data and consider:\n- The role of pace and running styles in determining race outcomes.\n- How tracking data can identify in-race impedance.\n- Discovering potential issues with course consitency using tracking data.\n","metadata":{}},{"cell_type":"markdown","source":"## Background\nThe value of information that can be produced is dependent on the characteristics of the data. It’s important to analyze what we have, as well as what we don’t. For example, in the competition data set we have:\n-  Fine-grained GPS coordinates for each racer at quarter-second intervals. This Trakus data is much more detailed than that previously available.\n-  Data for a range of tracks, distances, race types/grades and conditions.\n-  Jockey data\n-  Odds, weights, finishing positions\n\n\nWhat is not provided is just as important when deciding how we should proceed. The obvious omission is almost everything about the key participant, the horse. We don’t have any past performance, breeding, training, trainer or owner data. The excluded data is widely used by all stakeholders to make decisions and is available in many forms. We can’t assume that the Derby data can paint the whole portrait of a horse and a race, but perhaps it can shine a new light on a part of the scene. Subsequent analysis can layer on the past performance data to paint a fuller picture.\n\n**The GPS and fine-grained timing data are the magic in the Derby data set. Almost everything else has been available and analyzed for years.**\n\nNYRA tracks have used Trakus technology for some time and they provide some of the derived data to bettors on their websites. You can find for each horse their times at each point of call, distance covered, average and peak speeds, distance from the rail and lengths ahead of other horses. This represents a significant upgrade from what is available in standard Equibase files, for example. The Derby data set provides an even more detailed view of horse tracking, and it is in a form that enables more analysis.\n\n## Objectives\nSo, what are the possible dimensions for analysis that are uniquely enabled by the data? In this notebook I will primarily explore the role of pace and running styles in determining race outcomes. “Pace makes the race” is an old saying that underlines the importance of how and when racers use their speed and the effects of pace on individual horses. Handicapping services classify the running styles of horses based on past performances and these styles reflect the pace that a horse will typically use. The number and combinations of horses with different styles can determine the race’s pace and ultimate outcomes. At least that’s what traditional experience and knowledge suggests! But what does the data say?\n\nHypotheses:\n1. Horses can be classified based on how fast they run at in the early stages of a race and how close they keep to the leader. A class is represented as a running style.\n1. The running styles of race entrants can be a predictor of race outcomes.\n1. Factors such as the track, surface, racing conditions and distance can help or hurt a given style.\n1. Some combinations of the number of horses in a race with a given style can help or hurt horses with other styles.\n\nWe'll consider these hypotheses and try to transform the raw data into information that potentially lead us in the right directions. We'll also consider additional use cases for the tracking data that may be interesting to decision makers.\n\nWe won't necessarily come up with definitive ready-to-go answers answers; instead the objective is to identify useful data and methods, plus ideas for future research. \n\n## Data Features\nThe primary data features to be created include the instantaneous and average velocity, acceleration, and relative distance from the leader of each horse. Specifically:\n-  Distance Traveled: the distance traveled during one time interval, calculated from the lat and long numbers from consecutive intervals\n-  Total Distance Traveled: cumulative sum of distance\n-  Velocity: distance/time\n-  Average Velocity: a moving average of velocity over several consecutive intervals. Used to smooth the data\n-  Acceleration: change in velocity/change in time\n-  Distance from leader: relative distance (measured in feet) to the horse assumed to be in the lead. While it’s possible the GPS data can be used to derive the exact relative position of each horse, I use a simplifying assumption that total distance traveled can be used. The Derby data set lacks key GPS coordinates for the rail or finish line.\n-  Racing Styles/Classes: learned from ordered sets of velocity and distance values, combined with other variables\n\n## Notes about the Notebook\nThis notebook and its associated utility script contain all the code to import and process the competition data, generate new features, and create models for analysis. Some of the steps can take very long times, so be default the script loads pre-processed files that have the features already generated. See the code blocks below for setting variables that control script operation.\n\nThe script requires two packages installed over and above what is installed with the competition kernel: tslearn and dataframe_image.\n\nYou should be able to run all cells to completion.\n\nIn the sections below, we use some unsupervised machine learning algorithms that are non-deterministic. Results from one run to another can vary. Some comments in the text may therefore be inconsistent with what you see in some charts.","metadata":{}},{"cell_type":"code","source":"!pip install tslearn\n!pip install dataframe_image","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:03.056894Z","iopub.execute_input":"2022-11-09T21:38:03.057478Z","iopub.status.idle":"2022-11-09T21:38:27.480958Z","shell.execute_reply.started":"2022-11-09T21:38:03.057440Z","shell.execute_reply":"2022-11-09T21:38:27.479788Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\nimport math\nfrom datetime import datetime\nimport seaborn as sns\nfrom math import atan\nimport matplotlib.pyplot as plt\nimport geopy.distance\nfrom IPython.display import display\nfrom tslearn.clustering import TimeSeriesKMeans\nfrom matplotlib import rcParams\nimport dataframe_image as dfi\nfrom sklearn.cluster import KMeans\nfrom sklearn.linear_model import LinearRegression\n\n# import the lcoal utility functions that power the bulk of the analysis below\nfrom derby_utils import *\n\n# set some default display options for dataframes\npd.set_option('display.max_rows', None)\npd.set_option('display.max_columns', None)\npd.set_option('display.width', 1000)\npd.set_option('display.colheader_justify', 'center')\npd.set_option('display.precision', 2)\n\ndroot = \"../input/big-data-derby-2022/\"\nTRAKUS_PATH = droot + \"nyra_tracking_table.csv\"\nTRAKUS_START_TABLE_PATH = droot + \"nyra_start_table.csv\"\nTRAKUS_RACE_TABLE_PATH = droot + \"nyra_race_table.csv\"\n\n# pre-processed cached versions for fast loading\nSMALL_DATA_PATH=\"small_df.csv\"\nPREPOPULATED_NYRA_DATA = \"../input/nyra-enhanced/nyra.csv\"\n\n# file that maps distance IDs to distances in yds\nDISTANCE_ID_MAP_PATH=\"../input/distids-to-distydscsv/distids_to_distyds.csv\"\n\n# sample race data\nSAMPLE_RACE = \"../input/past-races-for-horses-in-sample-race/sample_race_aqu_20191220_6_brief.csv\"","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:27.484925Z","iopub.execute_input":"2022-11-09T21:38:27.485475Z","iopub.status.idle":"2022-11-09T21:38:27.500165Z","shell.execute_reply.started":"2022-11-09T21:38:27.485420Z","shell.execute_reply":"2022-11-09T21:38:27.498848Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<div class=\"alert alert-block alert-info\">The following notebook code cell has variable assignments that control how data is retrieved and processed in the rest of the script. The default is to fully process all the competition data files. If you prefer, you can skip the looooong pre-processing steps and read in a cached file by setting <i>use_pregen_data = True</i> .</div>","metadata":{}},{"cell_type":"code","source":"# set variables that control whether we regen all or partial data, or read previously saved data from files.\nuse_pregen_data = True  # set to true if you want to skip raw data process and simply read in pre-processed data\n\n# if using pre-processed data, you can use the whole data set or a speedier random subset\nget_all_data_from_saved_file = True\nget_subset_data_from_saved_file = False # smaller set primarily for dev\n\n# during dev, you can save aside the freshly processed data for re-use later\nsave_generated_small_df = False\nsave_generated_nyra = False\n\nuse_small_set_for_dev = False # set to True to use a small data set for development, for production set to False","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:27.503073Z","iopub.execute_input":"2022-11-09T21:38:27.503496Z","iopub.status.idle":"2022-11-09T21:38:27.517361Z","shell.execute_reply.started":"2022-11-09T21:38:27.503460Z","shell.execute_reply":"2022-11-09T21:38:27.515999Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Data Import and Cleansing\nThe competition data is in three data files that are merged. We then clean up the data, looking for observations that are out of range or incomplete. We can remove records for horses that are clearly debilitated or stopped racing. ","metadata":{}},{"cell_type":"code","source":"if not use_pregen_data:\n    # read in the tracking table\n    dateparse = lambda x: datetime.strptime(x, '%Y-%m-%d')\n    cols = ['track_id', 'race_date', 'race_number', 'program_number', 'trakus_index', 'latitude', 'longitude']\n\n    nyra = pd.read_csv(TRAKUS_PATH,sep=',',header=0,\n                      parse_dates=['race_date'], date_parser=dateparse, dtype={\"program_number\":str},\n                      names=cols)\n\n    nyra.sort_values(by=['track_id', 'race_date', 'race_number', 'program_number', 'trakus_index'],inplace=True)\n    nyra.set_index(['track_id', 'race_date', 'race_number', 'program_number'])\n\n    # remove trailing spaces in the horse's program number\n    nyra['program_number'] = nyra['program_number'].str.rstrip()\n\n\n    # read in start table to get the official_pos\n    dateparse = lambda x: datetime.strptime(x, '%Y-%m-%d')\n    cols = ['track_id', 'race_date', 'race_number', 'program_number', 'weight','jockey', 'odds','official_pos']\n\n    tdf = pd.read_csv(TRAKUS_START_TABLE_PATH,sep=',',header=0,dtype={\"program_number\":str},\n                      parse_dates=['race_date'], date_parser=dateparse,\n                      names=cols)\n\n    tdf['program_number'] = tdf['program_number'].str.rstrip()\n\n    tdf.reset_index().set_index(['track_id','race_date','race_number','program_number'],inplace=True)\n\n\n    # merge the tracking and start dfs\n    nyra =  pd.merge(nyra,tdf, on = ['track_id','race_date','race_number','program_number'],how = 'left')\n\n    # read in race table. It gives us information about the race length\n    dateparse = lambda x: datetime.strptime(x, '%Y-%m-%d')\n    nyra_races_df = pd.read_csv(TRAKUS_RACE_TABLE_PATH,sep=',',\n                      parse_dates=['race_date'], date_parser=dateparse)\n    nyra_races_df['track_condition'] = nyra_races_df['track_condition'].str.rstrip()\n    nyra = pd.merge(nyra,nyra_races_df)\n\n    # read in distance_id to distance_yds map table. This is a data file that was researched to determine the actual \n    # yardage assocatied with each race distance_id\n    dist_df = pd.read_csv(DISTANCE_ID_MAP_PATH)\n    dist_df.set_index('distance_id',inplace=True)\n    nyra = nyra.merge(dist_df[['distance_yds']], on = ['distance_id'],how = 'left')\n\n","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:27.519722Z","iopub.execute_input":"2022-11-09T21:38:27.520096Z","iopub.status.idle":"2022-11-09T21:38:27.539957Z","shell.execute_reply.started":"2022-11-09T21:38:27.520062Z","shell.execute_reply":"2022-11-09T21:38:27.538543Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Feature Generation\nOur nyra dataframe has the merged data from the competition data files and we have cleaned it up. Now we calculate the new data fields we will need.","metadata":{}},{"cell_type":"code","source":"if not use_pregen_data:\n    # Create an easy to use race_id for each race\n    nyra['race_id'] = (nyra.groupby(['track_id','race_date','race_number']).cumcount()==0).astype(int)\n    nyra['race_id'] = nyra['race_id'].cumsum()\n\n    # concatenate df to self to get consecutive coordinates on same row\n    # we do this to calculate point to point velocity and distance\n    data_next = nyra[['program_number','trakus_index','latitude','longitude']].groupby('program_number').shift(-1)\n    data_next.columns = [f'{c}2' for c in data_next.columns]\n    nyra = pd.concat([nyra, data_next], axis=1)\n\n    # calculate the distance traveled \n    nyra['dtraveled'] = nyra.apply(lambda x: calc_d2(x), axis=1)\n\n    # make sure that we account for change to new horse.\n    nyra['dtraveled'] = np.where((nyra['trakus_index2'] < nyra[\"trakus_index\"] ),float(\"NaN\"),nyra[\"dtraveled\"])\n\n    # clean up any too large distances. Replace too high values with a median value\n    med_dtraveled = nyra['dtraveled'].median()\n    nyra['dtraveled'] = np.where((nyra['dtraveled'] > 25 ),med_dtraveled,nyra[\"dtraveled\"])\n\n    # calculate velocity\n    nyra['velocity'] = nyra.apply(lambda x: calc_velocity(x), axis=1)\n\n    # clean up rows where the horse changes\n    nyra['velocity'] = np.where((nyra['trakus_index2'] < nyra[\"trakus_index\"] ),float(\"NaN\"),nyra[\"velocity\"])\n\n    # concat shifted lat long data so we can calculate velocity and distance traveled\n    data_vnext = nyra[['program_number','trakus_index','velocity']].groupby('program_number').shift(-1)\n    data_vnext.columns = [f'{c}3' for c in data_vnext.columns]\n    nyra = pd.concat([nyra, data_vnext], axis=1)\n\n    # number of horses in each race\n    nyra['num_entries'] = nyra.groupby(['track_id','race_date','race_number'])['program_number'].transform('nunique')\n\n    # Calculate acceration between points\n    nyra['acceleration'] = nyra.apply(lambda x: calc_acceleration(x), axis=1)\n    #nyra['angle'] = nyra.apply(lambda x: calc_angle(x), axis=1)\n\n    # Save running totals for distance and time\n    nyra['cum_dtraveled'] = nyra.groupby(['track_id','race_date','race_number','program_number'])['dtraveled'].transform(pd.Series.cumsum)\n    nyra['cum_time'] = nyra['trakus_index'] * .25\n\n    # calculate the moving average for velocity\n    moving_window = 4 # this is a tunable parameter that should be experimented with to get a reasonable level of smoothing\n    nyra['moving_velocity'] = nyra.groupby(['track_id','race_date','race_number','program_number'])['velocity']\\\n        .transform(lambda x: x.rolling(moving_window, 1).mean())\n\n    if 'max_cum_dtraveled' not in nyra.columns:\n        nyra['max_cum_dtraveled'] = nyra.groupby(['track_id','race_date','race_number','trakus_index'])['cum_dtraveled'].transform('max')\n\n    if 'dist_behind' not in nyra.columns:\n        nyra['dist_behind'] = nyra.apply(lambda x: calc_dist_behind(x), axis=1)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:27.544374Z","iopub.execute_input":"2022-11-09T21:38:27.544852Z","iopub.status.idle":"2022-11-09T21:38:27.566556Z","shell.execute_reply.started":"2022-11-09T21:38:27.544813Z","shell.execute_reply":"2022-11-09T21:38:27.564421Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<div class=\"alert alert-block alert-info\">In the analysis below, the \"distance behind leader\" feature (in feet) depends on simplifying assumption that the horse in the lead has gone the farthest at any point in time. This assumption is probably valid for the early part of the race. By the end of the race horses that have raced wide for considerable times may have traveled farther than leader. In the analysis below we focus on early stages. Moreover, we don't consider actual race position, so we suggest that the feature is reasonable. Assumptions are dangerous, of course, so further analysis is in order when true position data is available</div>","metadata":{}},{"cell_type":"code","source":"# DEVELOPMENT ONLY - OPTIONAL\nif save_generated_nyra and not use_pregen_data:\n    nyra.to_csv(PREPOPULATED_NYRA_DATA)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:27.568628Z","iopub.execute_input":"2022-11-09T21:38:27.569119Z","iopub.status.idle":"2022-11-09T21:38:27.585932Z","shell.execute_reply.started":"2022-11-09T21:38:27.569070Z","shell.execute_reply":"2022-11-09T21:38:27.584656Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# DEVELOPMENT ONLY - OPTIONAL\n#save a copy\n#df = nyra.copy(deep=True)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:27.587524Z","iopub.execute_input":"2022-11-09T21:38:27.587912Z","iopub.status.idle":"2022-11-09T21:38:27.602011Z","shell.execute_reply.started":"2022-11-09T21:38:27.587877Z","shell.execute_reply":"2022-11-09T21:38:27.600389Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**SAVE TIME AND READ IN PREVIOUSLY SAVED CLEAN DF**\nBelow is the code cell that loads the pre-processed data file. The variables that determine control are defined early in this script.","metadata":{}},{"cell_type":"code","source":"# Potentially load pregenerated data\nif use_pregen_data:\n    if get_all_data_from_saved_file:\n        # GET ALL NYRA DATA\n        dateparse = lambda x: datetime.strptime(x, '%Y-%m-%d')\n        nyra = pd.read_csv(PREPOPULATED_NYRA_DATA,parse_dates=['race_date'], date_parser=dateparse,dtype={\"program_number\":str})\n        \n    elif get_subset_data_from_saved_file:\n        # Read in a smallish subset of complete nyra data, about 600 races\n        dateparse = lambda x: datetime.strptime(x, '%Y-%m-%d')\n        nyra = pd.read_csv(SMALL_DATA_PATH,parse_dates=['race_date'], date_parser=dateparse,dtype={\"program_number\":str})\n        # For speed during development, work with a subset of the nyra data\n        small_df = nyra\n    \nif not use_small_set_for_dev:\n    small_df = nyra\nelse:\n    # For speed during development, work with a subset of the nyra data\n    small_df = nyra[(nyra.race_id % 3) == 0]\n    \n# We actually use the small_df dataframe in much of the code below\n\n# some extra cleanup to remove outlier horses\n# We remove up to 24 horses that seem disabled and out of range\ndef drop_slow_horses(small_df,cum_time=40,velocity=35):\n    too_slow_horses = small_df[(small_df.cum_time == cum_time) & \n                            (small_df.moving_velocity < velocity)][['track_id',\n                                    'race_date','race_number','program_number']]\n    small_df = remove_bad_horses(small_df,too_slow_horses)\n    return small_df\n\nsmall_df = drop_slow_horses(small_df,cum_time=40,velocity=35)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:38:27.603362Z","iopub.execute_input":"2022-11-09T21:38:27.604501Z","iopub.status.idle":"2022-11-09T21:40:45.584916Z","shell.execute_reply.started":"2022-11-09T21:38:27.604457Z","shell.execute_reply":"2022-11-09T21:40:45.582889Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# find distance behind leader  \n# for each race and index, find the max(total_dtraveled) = leader_dist\n# for each race, horse and index, calc diff between leader_dist and total_dtraveled ---> that is the feature\nsmall_df = add_req_flds(small_df)\n\npd.options.mode.chained_assignment = None\nsmall_df.sort_values(by=['track_id','race_date','race_number','program_number','trakus_index'], inplace=True)\npd.options.mode.chained_assignment = 'warn'","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:40:45.586657Z","iopub.execute_input":"2022-11-09T21:40:45.588003Z","iopub.status.idle":"2022-11-09T21:42:43.959728Z","shell.execute_reply.started":"2022-11-09T21:40:45.587935Z","shell.execute_reply":"2022-11-09T21:42:43.958200Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# DEVELOPMENT ONLY - OPTIONAL\nif save_generated_small_df and not use_pregen_data:\n    small_df.to_csv(SMALL_DATA_PATH)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:42:43.961855Z","iopub.execute_input":"2022-11-09T21:42:43.962662Z","iopub.status.idle":"2022-11-09T21:42:43.968301Z","shell.execute_reply.started":"2022-11-09T21:42:43.962624Z","shell.execute_reply":"2022-11-09T21:42:43.967047Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Descriptive Analytics\nIn this section we present some charts that break down the types and distributions of data in the races set. Some observations:\n- A few distances (e.g. 1320 and 1760) are the most popular. Some are run so infrequently that it may be hard to generate valid statistics.\n- Most races are run on Fast or Firm courses.\n- Dirt races are most common, followed by Turf.\n- Claiming races are the most frequent type.","metadata":{}},{"cell_type":"code","source":"plot_descriptive_data(small_df)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:42:43.969748Z","iopub.execute_input":"2022-11-09T21:42:43.970120Z","iopub.status.idle":"2022-11-09T21:42:48.104831Z","shell.execute_reply.started":"2022-11-09T21:42:43.970088Z","shell.execute_reply":"2022-11-09T21:42:48.103508Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Below are a couple charts which show average velocity during races for each track (all courses, all distances, all conditions). These charts show subtle differences across the tracks and distances. For our analysis, this global view is less interesting, as we are concerned with most customized combinations of parameters.","metadata":{}},{"cell_type":"code","source":"dist=1320\ninterval=2\n\nraces = small_df[(small_df.distance_yds == dist) & (small_df.trakus_index % interval == 0) & (small_df.velocity < 65)]\nplot_avg_velocity(races,title='Avg Velocity over 1320 yds (6 furlongs) - All Courses',dist=dist,interval=interval)\n\ndist=1760\nraces = small_df[(small_df.distance_yds == dist) & (small_df.trakus_index % interval == 0)]\nplot_avg_velocity(races,title='Avg Velocity over 1760 yds (1 mile) - All Courses',dist=dist,interval=interval)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:42:48.106563Z","iopub.execute_input":"2022-11-09T21:42:48.107299Z","iopub.status.idle":"2022-11-09T21:42:49.542383Z","shell.execute_reply.started":"2022-11-09T21:42:48.107260Z","shell.execute_reply":"2022-11-09T21:42:49.541493Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"If we drill down a little deeper by filtering the races we analyze, you continue to see differences in the charts. Below we see a chart that shows the difference between the Dirt and Turf courses at Belmont over 1 mile.","metadata":{}},{"cell_type":"code","source":"dist=1760\ninterval=2\n\nraces = small_df[(small_df.track_id =='BEL') \n                 &(small_df.distance_yds == dist) \n                 & (small_df.trakus_index % interval == 0)] #[['course_type','cum_time','moving_velocity']]\n\nplot_avg_velocity(races,title='Avg Velocity - 1760 yds (1 mile) - Belmont Courses',keycol='course_type',\n                  cols=['cum_time','course_type'],dist=dist,interval=interval)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:42:49.543991Z","iopub.execute_input":"2022-11-09T21:42:49.544615Z","iopub.status.idle":"2022-11-09T21:42:50.354220Z","shell.execute_reply.started":"2022-11-09T21:42:49.544582Z","shell.execute_reply":"2022-11-09T21:42:50.352840Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Continuing to drill down, we can profile velocity over time for specific courses, distances and conditions. Below is a chart of 1 mile Dirt races at Belmont under the various track conditions.","metadata":{}},{"cell_type":"code","source":"dist=1760\ninterval=8\n\nraces = small_df[(small_df.track_id =='BEL') \n                 &(small_df.distance_yds == dist) & (small_df.course_type=='D')\n                 & (small_df.trakus_index % interval == 0)] #[['course_type','cum_time','moving_velocity']]\n\nplot_avg_velocity(races,title='Avg Velocity - 1760 yds (1 mile) - Belmont - Dirt - Conditions',keycol='track_condition',\n                  cols=['cum_time','track_condition'],dist=dist,interval=interval)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:42:50.359478Z","iopub.execute_input":"2022-11-09T21:42:50.359844Z","iopub.status.idle":"2022-11-09T21:42:51.408464Z","shell.execute_reply.started":"2022-11-09T21:42:50.359814Z","shell.execute_reply":"2022-11-09T21:42:51.407027Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Running Styles: Learning Classes\nWith the data we have generated we're in a position to use machine learning algorithms to identify the different possible running styles. For this analysis we will work with two features: the velocity of the horse and the distance behind the leader, at selected points in the race. This becomes a time series of two variables for each horse in each race.\n\n<table>\n  <tr>\n    <th>Time</th>\n    <th>Velocity</th>\n    <th>Distance Behind (ft)</th>\n  </tr>\n  <tr>\n    <td>2.5</td>\n    <td>13.2</td>\n    <td>6.1</td>\n  </tr>\n  <tr>\n    <td>5</td>\n    <td>26</td>\n    <td>4.3</td>\n  </tr>\n  <tr>\n    <td>...</td>\n    <td> </td>\n    <td> </td>\n  </tr>\n  <tr>\n    <td>50</td>\n    <td>42</td>\n    <td>14.6</td>\n  </tr>\n</table>\n\nThis is an unsupervised machine learning problem as we are discovering unknown classes without any labeled data. There are numerous algorithms for unsupervised clustering, including multiple variants in the K-means family.\n\n<div class=\"alert alert-block alert-info\"><i>K-means clustering aims to partition data into k clusters in a way that data points in the same cluster are similar and data points in the different clusters are farther apart. Similarity of two points is determined by the distance between them. There are many methods to measure the distance. Euclidean distance (minkowski distance with p=2) is one of most commonly used distance measurements.   <a href=\"https://towardsdatascience.com/k-means-clustering-explained-4528df86a120\">towardsdatascience</a></i></div>\n\nFortunately, there are many python resources available for K-means cluster learning, we use a package named tslearn. It provides support for multivariate time-series analysis.\n\nIn the code below, we create a subset of our overall data. This subset contains just the rows in the dataframe that correspond to the points in time for our time series. But which timeslots should we use as features? Initially we just use educated guesses. We want to model changes in velocity and position during the early stages of races during which the horses are accelerating, and some time when they are slowing down. During the feature generation process above, we generated features for eight timeslots, i.e., observations at 2.5, 5, 10, 20, 40, 45 and 50 seconds from the start of the race. \n\nBy default, we only use the first 5 pairs of observations. We're not sure this is the optimal set, but the effort and time to do a thorough study are beyond the scope of our competition project. The results we obtain suggest the features are reasonable.\n\nThe features are z-normalized values of the raw features. It may be interesting to use other pre-processed derivative features in the future. Further, more technical, analysis is also recommended to look at parameters such as the choice of distance measures. In our experiments we use the Euclidean distance option for the tslearn K-means algorithm.\n\nWe're not learning how to predict now, so we won't divide the data into training and test sets.","metadata":{}},{"cell_type":"code","source":"\ntimsdf, hdata_df = add_features(small_df)\nX_train = hdata_df\n\n# Cluster using SOME columns THIS IS IMPORTANT TUNABLE PARAMETER\nall_feat_cols = ['f1','f2','f3','f4','f5','f6','f7','f8','f1a','f2a','f3a','f4a','f5a','f6a','f7a','f8a']\nfive_intervals_feat_cols = ['f1','f2','f3','f4','f5','f1a','f2a','f3a','f4a','f5a']\nfeat_cols = five_intervals_feat_cols # change this line to change features used\n","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:42:51.410249Z","iopub.execute_input":"2022-11-09T21:42:51.411232Z","iopub.status.idle":"2022-11-09T21:42:52.543545Z","shell.execute_reply.started":"2022-11-09T21:42:51.411181Z","shell.execute_reply":"2022-11-09T21:42:52.542498Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Number of Running Styles/Clusters\nRace analytics providers identify running styles for horses based on the track and distance. For example, as decribed on the Brisnet web site, \n\n> *Brisnet sorts running styles into four different categories: “E” (Early), “E/P” (Early/Presser), “P” (Presser), and “S” (Sustain or Closer). Basically, “E” types are pacesetters, “E/P” types can set the pace but generally sit just off the lead, “P” types rally from midpack, and “S” types are deep closers coming from well off the pace.* \n\nThe K-means algorithm takes as a parameter the number of clusters that will be created. Based on the work of handicapping services, an educated guess would be in the range of 4-7. We can analyze our data to identify a reasonable number of clusters for the K-means algorithm. Below we use the \"elbow method\" to find a good number of clusters for our experiments. Per Wikipedia:\n\n >In cluster analysis, the elbow method is a heuristic used in determining the number of clusters in a data set. The method consists of plotting the explained variation as a function of the number of clusters and picking the elbow of the curve as the number of clusters to use... Using the \"elbow\" or \"knee of a curve\" as a cutoff point is a common heuristic in mathematical optimization to choose a point where diminishing returns are no longer worth the additional cost. In clustering, this means one should choose a number of clusters so that adding another cluster doesn't give much better modeling of the data.","metadata":{}},{"cell_type":"code","source":"# find the elbow\nfind_elbow(X_train,feat_cols)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:42:52.544811Z","iopub.execute_input":"2022-11-09T21:42:52.545919Z","iopub.status.idle":"2022-11-09T21:43:08.606462Z","shell.execute_reply.started":"2022-11-09T21:42:52.545875Z","shell.execute_reply":"2022-11-09T21:43:08.604934Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"The line chart above shows the elbow is in the 3-5 range. Let's try some values by running the algorithm.","metadata":{}},{"cell_type":"code","source":"\ntimsdf = plot_clusters_for_ks(X_train,timsdf,hdata_df,feat_cols)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:43:08.608483Z","iopub.execute_input":"2022-11-09T21:43:08.609011Z","iopub.status.idle":"2022-11-09T21:43:14.421643Z","shell.execute_reply.started":"2022-11-09T21:43:08.608963Z","shell.execute_reply":"2022-11-09T21:43:14.420458Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_cnts_per_ks(timsdf)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:43:14.423345Z","iopub.execute_input":"2022-11-09T21:43:14.424345Z","iopub.status.idle":"2022-11-09T21:43:15.185590Z","shell.execute_reply.started":"2022-11-09T21:43:14.424302Z","shell.execute_reply":"2022-11-09T21:43:15.184319Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"The above charts demonstrate that the K-means algorithm is capable of separating horses into distinct clusters, and that the horses in each cluster run different shape races. These charts do not tell us if these different shapes matter in terms of winning races. That is what we do now.","metadata":{}},{"cell_type":"markdown","source":"### Analyzing Finish Position by Running Style/Cluster\nTo analyze the relationship among running styles and final positions (finish order) we need to consider how many horses in each race have each running style. Obviously, if all horses have style 1 then, that is the winning style. We determine the percentage of entries with each style and from that the random probability of a horse with that style winning (i.e., we assume each horse has an equal chance).\n\nBelow are sample charts that describe the output of running the clustering algorithm for a given set of races. We'll use these charts to analyze combinations of tracks, courses, distances, and conditions.\n![allclusters_sample_annotated.png](attachment:allclusters_sample_annotated.png)\n","metadata":{},"attachments":{"allclusters_sample_annotated.png":{"image/png":"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"}}},{"cell_type":"code","source":"# Cluster all races, all tracks\n# DEV ONLY\n# nyra_copy = nyra.copy(deep=True)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:43:15.186906Z","iopub.execute_input":"2022-11-09T21:43:15.187584Z","iopub.status.idle":"2022-11-09T21:43:15.193149Z","shell.execute_reply.started":"2022-11-09T21:43:15.187549Z","shell.execute_reply":"2022-11-09T21:43:15.191866Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#nyra = nyra_copy\n#nyra","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:43:15.195180Z","iopub.execute_input":"2022-11-09T21:43:15.195715Z","iopub.status.idle":"2022-11-09T21:43:15.206692Z","shell.execute_reply.started":"2022-11-09T21:43:15.195683Z","shell.execute_reply":"2022-11-09T21:43:15.205358Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# select races on\nall_tracks_clusters,all_tracks_cluster_relative_df = analyze_styles_for_races(nyra,\n                        num_clusters=5,set_description='All Tracks - All Races')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:43:15.208350Z","iopub.execute_input":"2022-11-09T21:43:15.209663Z","iopub.status.idle":"2022-11-09T21:46:14.171460Z","shell.execute_reply.started":"2022-11-09T21:43:15.209625Z","shell.execute_reply":"2022-11-09T21:46:14.169514Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Comments\nThe bar chart above suggests that horses in one or more clusters win a disproportionate number of races. Of course, there are innumerable factors that go into a horse race result, but this analysis of running styles implies that a horse may gain an advantage if:\n1. The pace characteristics associated with the cluster/style can be translated to an actual plan\n1. The horse is capable of executing that plan\n\nThe edge is seen over many races, and that adopted strategy may be wrong for any specific race.\n\nThe choice of features, especially the 'Distance behind Leader', can insert a bias into the training data. If we used distance behind right before the finish line, it's likely that would be a dominant feauture. Further analyis is required to determine if the features are biased towards leaders.","metadata":{}},{"cell_type":"markdown","source":"### Variable Models\nThe analysis above lumped together all races, at all tracks, distances and conditions. This one-size-fits-all model appears to show disproportionate returns for some clusters. It seems possible, perhaps even probable, that more granular, customized models can be developed for each combination of track, surface, distance and condition. The number of dimensions makes this a large problem. Just in this data set there are over 300 combinations of the track, course type, distance, race type and conditions. We'll consider a few variations below to see if we find interesting information.\n\nIn a custom model we learn clusters using just the races that satisfy the parameter filters we define.\n\n#### Track and Distance\nLet's consider models for a couple of the Dirt course types at Belmont and Aqueduct at two distances: 1320 (considered a \"sprint\") and 1760 (a \"route\").\n","metadata":{}},{"cell_type":"code","source":"# PICK A VALUE FOR NUMBER OF CLUSTERS\n# Note, when we use KMeansConstrained, the number may be larger or smaller\nnum_clusters = 5\n\n# now get the subset of races from a track/course/distance\nbelmont_sprint_df = nyra[(nyra.track_id == 'BEL') & (nyra.course_type =='D') & (nyra.distance_yds == 1320)]\nbelmont_sprint_df = drop_slow_horses(belmont_sprint_df,cum_time=40,velocity=35)\n\n# select races on\nt_df,r_df = analyze_styles_for_races(belmont_sprint_df,num_clusters=num_clusters,set_description='Belmont - Dirt - 1320 yds')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:46:14.174478Z","iopub.execute_input":"2022-11-09T21:46:14.174919Z","iopub.status.idle":"2022-11-09T21:46:18.400900Z","shell.execute_reply.started":"2022-11-09T21:46:14.174886Z","shell.execute_reply":"2022-11-09T21:46:18.399460Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"aqu_sprint_df = nyra[(nyra.track_id == 'AQU') & (nyra.course_type =='D') & (nyra.distance_yds == 1320)]\n\n# select races on\nt_df,r_df = analyze_styles_for_races(aqu_sprint_df,num_clusters=num_clusters,set_description='Aqueduct - Dirt - 1320 yds')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:46:18.403106Z","iopub.execute_input":"2022-11-09T21:46:18.403980Z","iopub.status.idle":"2022-11-09T21:46:24.804128Z","shell.execute_reply.started":"2022-11-09T21:46:18.403909Z","shell.execute_reply":"2022-11-09T21:46:24.802589Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"belmont_route_df = nyra[(nyra.track_id == 'BEL') & (nyra.course_type =='D') & (nyra.distance_yds == 1760)]\n\n# select races on\nt_df,r_df = analyze_styles_for_races(belmont_route_df,num_clusters=num_clusters,set_description='Belmont - Dirt - 1760 yds')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:46:24.806791Z","iopub.execute_input":"2022-11-09T21:46:24.807184Z","iopub.status.idle":"2022-11-09T21:46:27.785365Z","shell.execute_reply.started":"2022-11-09T21:46:24.807151Z","shell.execute_reply":"2022-11-09T21:46:27.784200Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"a_route_df = nyra[(nyra.track_id == 'AQU') & (nyra.course_type =='D') & (nyra.distance_yds == 1760)]\n\n# select races on\nt_df,r_df = analyze_styles_for_races(a_route_df,num_clusters=num_clusters,set_description='Aqueduct - Dirt - 1760 yds')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:46:27.787422Z","iopub.execute_input":"2022-11-09T21:46:27.787804Z","iopub.status.idle":"2022-11-09T21:46:32.814415Z","shell.execute_reply.started":"2022-11-09T21:46:27.787769Z","shell.execute_reply":"2022-11-09T21:46:32.812879Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Running Styles on Off Tracks\nThe above charts describe clusters for all conditions. The competition data enables us to build models for specific conditions. Below are example from Aqueduct and Saratoga for different conditions. Again, the quickest frontrunner appears to win most often.","metadata":{}},{"cell_type":"code","source":"am = nyra[(nyra.track_id == 'AQU') & (nyra.course_type =='D') & (nyra.distance_yds == 1760) & (nyra.track_condition == 'MY')]\n\n# select races on\nt_df,r_df = analyze_styles_for_races(am,num_clusters=num_clusters,set_description='Aqueduct - Dirt - 1760 yds - MY')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:46:32.816070Z","iopub.execute_input":"2022-11-09T21:46:32.816521Z","iopub.status.idle":"2022-11-09T21:46:35.199155Z","shell.execute_reply.started":"2022-11-09T21:46:32.816485Z","shell.execute_reply":"2022-11-09T21:46:35.197524Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sm = nyra[(nyra.track_id == 'SAR') & (nyra.course_type =='T') & (nyra.distance_yds == 1210) & (nyra.track_condition == 'FM')]\n\n# select races on\nt_df,r_df = analyze_styles_for_races(sm,num_clusters=num_clusters,set_description='Saratoga - Turf - 1210 yds - FM')\n","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:46:35.201291Z","iopub.execute_input":"2022-11-09T21:46:35.201714Z","iopub.status.idle":"2022-11-09T21:46:38.937462Z","shell.execute_reply.started":"2022-11-09T21:46:35.201678Z","shell.execute_reply":"2022-11-09T21:46:38.936123Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Running Styles and Race Types\nWe've looked at some of the combinations of tracks, courses, conditions and distances. We can also consider custom cluster models that include the race type as a feature. Below are a couple examples of cluster results for Stakes and Claiming races at the three tracks on Dirt.","metadata":{}},{"cell_type":"code","source":"sm = small_df[ (small_df.course_type =='D') & (small_df.distance_yds == 1320) \n          & (small_df.race_type == 'STK')]\ntoo_slow_horses = sm[(sm.cum_time == 30) & (sm.moving_velocity < 40)][['track_id','race_date','race_number','program_number']]\nsm = remove_bad_horses(small_df,too_slow_horses)\n\nt_df,r_df = analyze_styles_for_races(sm,num_clusters=num_clusters,set_description='All Courses - Dirt - 1760 yds - Stakes')\n\nsm = small_df[ (small_df.course_type =='D') & (small_df.distance_yds == 1320) \n          & (small_df.race_type == 'CLM')]\n\nt_df,r_df = analyze_styles_for_races(sm,num_clusters=num_clusters,set_description='All Courses - Dirt - 1760 yds - Claiming')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:46:38.939005Z","iopub.execute_input":"2022-11-09T21:46:38.939630Z","iopub.status.idle":"2022-11-09T21:48:00.253899Z","shell.execute_reply.started":"2022-11-09T21:46:38.939598Z","shell.execute_reply":"2022-11-09T21:48:00.252447Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"It's clear that there are differences in the two models. Whether these differences are consequential is a question for further research.","metadata":{}},{"cell_type":"markdown","source":"#### Comments\nHistoric past performance data has consistently shown that front runners win the most races. The charts above appear to confirm that since all show that the fastest front runners win a disproportionate number of races. However, in some models there are two different styles/clusters with different pace profiles that outperform. This data provides more granular information on just how fast horses with a given style run at various points.","metadata":{}},{"cell_type":"markdown","source":"### Making a Plan: Characterizing Clusters\nIf you are a trainer or jockey, knowing that a style may give your horse an edge is pretty useless if you can't understand what that styple implies in terms of a pace strategy. Let's try to understand more about the clusters we discover. We used the features 'moving_velocity' and 'distance_behind_leader' to train the algorithm, so we want to look at the resulting clusters to determine:\n- how fast the horse is going at select intervals\n- how close the horse is to the lead\n\nIn the table below we see how fast horses with a given style accelerate over 20 seconds. The quickest horse has the disproportionate number of wins. To emulate it, you can plan to accelerate at the same rate and stay within that number of feet of the lead.\n\n![clusterp.png](attachment:1335d30f-ee4c-4851-9d65-9661cd19a96a.png)\n\nThe data often suggest the obvious, to win it pays to be fastest earliest and stay close to the lead. But analysis of the cluster data may be able to give more specific parameters of what it means to be fast and close.","metadata":{},"attachments":{"1335d30f-ee4c-4851-9d65-9661cd19a96a.png":{"image/png":"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"}}},{"cell_type":"code","source":"gb = t_df[t_df.trakus_index.isin([10,20,40,80])].groupby(['cluster','cum_time']) \\\n    .agg({'moving_velocity':'mean','dist_behind':'mean'})\ngb_unstacked = gb.unstack(level = -1)\n\npd.set_option('display.precision', 2)\n\ndisplay(gb_unstacked)\n#dfi.export(gb_unstacked, 'cluster_profile.png')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:00.255283Z","iopub.execute_input":"2022-11-09T21:48:00.255627Z","iopub.status.idle":"2022-11-09T21:48:00.297388Z","shell.execute_reply.started":"2022-11-09T21:48:00.255597Z","shell.execute_reply":"2022-11-09T21:48:00.295461Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Comments\nThe cluster profiles described above use specific velocity values as atttributes. The speeds of front runners will vary from race to race, so intuitively the velocity values are not hard and fast rules. Depending on the class of the horses in the race, the top speed may be higher or lower. The idea of using speed differentials among horse in the race should be explored, so that the plan should be more like \"try to go faster than X by 20 seconds\", where X is relative to this race and horses.","metadata":{}},{"cell_type":"markdown","source":"### The Company You Keep\nThe cluster-based information suggests that over a large number of races, horses with a given style have a greater or lesser chance of winning a race. But this is just one factor among many when you are considering an individual race. Trends that hold over the large population may not apply to a single race where, for example, a 'slow but steady' horse is up against three or more 'early speed' horses. In other words, the number and combinations of horses with various styles may be important. What does the data say?\n\nLet's consider the company that winning horses keep in races. We'll first get a set of races from Belmont and generate clusters.","metadata":{}},{"cell_type":"code","source":"#clust_col = 'cluster'\n\n# get an example sample of races and clusters\nsm = nyra[(nyra.track_id == 'BEL') & (nyra.course_type =='D') & (nyra.distance_yds == 1320) \n          & (nyra.track_condition == 'FT')]\n","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:00.299905Z","iopub.execute_input":"2022-11-09T21:48:00.300529Z","iopub.status.idle":"2022-11-09T21:48:01.454675Z","shell.execute_reply.started":"2022-11-09T21:48:00.300479Z","shell.execute_reply":"2022-11-09T21:48:01.451527Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# select races on\nt_df,r_df = analyze_styles_for_races(sm,num_clusters=num_clusters,set_description='Belmont - Dirt - 1760 yds - FT')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:01.456882Z","iopub.execute_input":"2022-11-09T21:48:01.457938Z","iopub.status.idle":"2022-11-09T21:48:03.821921Z","shell.execute_reply.started":"2022-11-09T21:48:01.457875Z","shell.execute_reply":"2022-11-09T21:48:03.820456Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"In the chart below we look at the average number of other horses in a race with a given style. ","metadata":{}},{"cell_type":"code","source":"testdf = t_df[(t_df.cum_time == 2.5)][['race_id','program_number','trakus_index','cum_time','num_entries','official_pos','cluster','cluster_count']]\n\nall_races,winners = get_all_races_and_winners(testdf)\n\n# get non-winners\n#losers = all_races[(all_races.official_pos > 1)][['race_id','cluster','cluster_count']]\n\n#all_races.groupby([\"cluster\"]).agg(\n#        races=pd.NamedAgg(column=\"race_id\", aggfunc=\"count\"),\n#        horses=pd.NamedAgg(column=\"cluster_count\", aggfunc=\"sum\") )\n\n#all_races\n\navgdf = get_avg_all_races(all_races)\nprint('\\nAverage number of horses with given style per race:')\n\n#pd.set_option('display.max_rows', None)\n#pd.set_option('display.max_columns', None)\n#pd.set_option('display.width', 1000)\n#pd.set_option('display.colheader_justify', 'center')\n#pd.set_option('display.precision', 2)\n\ndisplay(avgdf)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:03.823400Z","iopub.execute_input":"2022-11-09T21:48:03.823765Z","iopub.status.idle":"2022-11-09T21:48:03.871617Z","shell.execute_reply.started":"2022-11-09T21:48:03.823733Z","shell.execute_reply":"2022-11-09T21:48:03.869416Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Now let's look at what winning horses face in their races compared to the average race. You can see that the winners face different combinations of styles than the overall averages.","metadata":{}},{"cell_type":"code","source":"all_wavg_df = get_winner_competitor_avgs(all_races)\nall_wavg_df","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:03.872896Z","iopub.execute_input":"2022-11-09T21:48:03.873307Z","iopub.status.idle":"2022-11-09T21:48:03.964440Z","shell.execute_reply.started":"2022-11-09T21:48:03.873273Z","shell.execute_reply":"2022-11-09T21:48:03.962653Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Finally, we look at the effect of having a certain number of entries with associated styles in a winner's race. This is the most detailed level and there are many combinations. Below we show some sample entry counts.","metadata":{}},{"cell_type":"code","source":"for num_racers in 6,8,10:\n    num_ent_df = all_races[all_races.num_entries == num_racers]\n    if len(num_ent_df) > 0:\n        num_ent_avg_df = get_winner_competitor_avgs(num_ent_df)\n        wavg_df = add_all_races_cnt_col(num_ent_avg_df,avgdf)\n        print(\"\\nNumber of Entries:\",num_racers)\n        display(wavg_df)","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:03.967681Z","iopub.execute_input":"2022-11-09T21:48:03.968278Z","iopub.status.idle":"2022-11-09T21:48:04.256110Z","shell.execute_reply.started":"2022-11-09T21:48:03.968234Z","shell.execute_reply":"2022-11-09T21:48:04.254770Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Comments\nThe above analysis suggests that it may be possible to build models to measure the effect of having combinations of racing styles in a race with N entries. With such models, owners and trainers may be able to select races to enter, trainers and jockeys may select the best pace strategy, and bettors can make more informed wagers. Further research is necessary to build such models and measure their validity and significance.","metadata":{}},{"cell_type":"markdown","source":"### Stylish Jockeys\nHorses can have a running style and each style demands that a jockey guide the horse a certain way. Intuitively, jockeys bring to the race their own unique set of skills that may fit one running style better than others. Above we have shown that running styles can be learned. Can we do the same and discern which jockeys align with those styles? What does the data say?\n\nHere is a portion of the following chart. In the right hand column you can find the win percentage for each jockey, for each racing style. Here the racing styles are for all tracks, all races.\n\n![jockeys_annotated.png](attachment:f95289b6-df92-48a8-819d-7f800966c41e.png)\n\nThe complete data is shown below.","metadata":{},"attachments":{"f95289b6-df92-48a8-819d-7f800966c41e.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABQAAAALQCAIAAABAH0oBAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAADsMAAA7DAcdvqGQAAP+lSURBVHhe7N0FWBRbHwbwsbAbEbuwFRO7MLEDuzFRwRZQEcUuDAwUDMRERSVFERAB6e7u7t1le3fumWBdETCuisD/9/jdjzk7Mzu9591zZhbDAQAAAAAAAACAGgACMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAAAAAAACAGgECMAAAgN9JUJyTlZlRxBGK6YLvEfGLmYVMNldED/8RIhZPKPruEomEPK6Q/vvnCdlFuRnZeUX8H13z7xHx2czCQhZH8H9mKOQW5WVk5jF4P719xUIeu6iQUcz79U3yc8T8YkZRIYv7w4fO1wQ8geB/baoKiJi87y+VSCjk//xm/mvEAjbavMxf3bw0ATsnOzOjkP3zcxELiplo93J+2/lRMXT4FhcWMVn8X90jPI6gEnamkJNHbN/i37uViHUR/9Ici/kCNCk9UJoYXaIYhQx0ytIFAFQREIABAAD8TtFvVsyeMtzgfSqfLvietDcbxjedoGGU8KMT/DRmzLuLg2775FZYARQX+9w4ffPKsxR6+Odl2p9RGz5f82IYiy74v1JsNKY0b7tI/3MRXfArst3Obxk9TuOqdx5d8MOYgXe3dWsybNW9cLrgT8uy29qnm/w8o1DWzwcPdsrr3TsffYrm0MO/kTDr85vzw0zDK84PgiLnU4dumznk0MP/nkKnvYN7tJ5q4MegC35JkvXqBVOGn7JP/uktneOydZJCk4UnbLJ+a7YrT77PrdVNOo9eYJ1GF/yMoih74606TgX04F+U8n7LwinDT7yJLqYL/jdelueDrbusEwX08I8T5nrcHmVo51tID3+DHWOhPaBJ/1mX/P7KLgXgt4EADAAA4Hdi50aEBPmnFvBEIk5ebHxCXBGrODMmOjo6LimNS4+DCFi5iagwIbso+uVqJWyw2uU4KgDzM1MS0AtIAVUgZGaGhMbGxpOpSMzLjYyMjkln/WiFi1eYYHu0S7uW2EYTz4h8NpWsitOpd0hIySRrhUJ2fujTXRMa1VVacdUjPbe4rPYMcXFeZFxsdH4x8c7C4tzomPiEXGqF2AVZkQnJOYV5qdH+oTEpbDQGNzc1ITojr6AwMzOeeKcctiTeC/IS4sh3j4lLrLiSm/RKTQlrOPegQ0RWKjFBchE5E2Z6bFx0ahGP2gQCZkZsXEwq9VIZ+EUp0QGBMamMkhFYGcnE7kAKeHQRiZMVS5bGpxRS68/wu7muJaYw/1YEGmDmpYSjdcvikO/KzkyhRk5lkltUUJgaHReTwaTq2GJeUVpUXGwGo6J2IVFeZhI5i+jckq2Q8WZNu5YNJl8JZgp4hWkxcfFZ5AzFfGZWQmxSKj334oIMajp08FBFnPwUG52xnbBOGw3fx+QVE0skyIsnFzAuIZdebyEnPz06KSk5Ly8nOjI2JZX1Q41WIk5OwCPNxm3kWh6wiEoopPa3AG1SUlJGPrk5+Mxsr5srB9VpNG2XWUBWQTmdGUTcgrS4hPis/PzUqKi4+Gy09QXFOdScoqNTsvO/TCbMzSi9cRD+NyuFCDm55IjR0bFpORUeTwXv1LvJ1hlz3i05lphPTDw5FzEnPy0sLjGVQ20PESs3PTI2Ob3Mc4DAzY0IC/JPyi9ZSU42dSTEJaVInd5ovKJMsjw6LZcO3Fkf1gyTxaYdep2N3rUwJSE8OCaLSb6NmInOLCQ2IZlNDKOLQ0p4fHI6fYgLi7KTI+NSs+izt2z8XMnGoY6TPI+rC7Bm/aa+QgFYwMhOjo5PzeYQcxBxslLjY9NyqTUWM9IT6fMhuoickl8UZ318chNZucXXPRKzyU0qZtP7KTkjj94wfEZmckxCZk5OempcZExGATnr7xFy8tJjopLSmMSWF7JzU2Oik7Op003ATEsKT8lhsPLiwoP9k3LLW1kxj5EdH5OaXlBUSC5TfHqe1HZHq0btjhTq5BOxsj4bLWsn22jicZvwDEa5HSQEzGz6qhudUUCuspCV5nVvQt/W2OxDFj4ZGQnRcbHpTKpZWswvSiMGWSJ21LOdnbCO409703uKmUPNJiGbQe0FAP5NEIABAAD8TsH3hnVvK6P1JqmY4XmqVft2zTdsUx+MIU26jrnuTNb8UIU34fXeOW2IwjFLd+6aPbL+WPVr8ajGxk1xNxrdpQkxOoYtvfgpB9WimD63esvWa9TxiBtHnOdxYl6tOv3nGgZ9SQAVC7nZo209an51Gpz5yECZI8B03yCqBC3SNfdUJp7vel5VnirCZOdovcugJ5YmCrmj2LZNw32WBWj50600a9Vv03WtJarL45mvjkxsoLjmpr25zlSZXrP3BApxUZDe3H61Ri9ctmzoYPLdZx20jCWqyNx8N6MVsg3Id8JqN9t5JzLvqxT6leQ3W8Y3aTln/sZRSp2J8buvP/85hyP20Fdq12jgsQ/pRBWZGXBhYss2Cur22eXUbdPf7prSpMUsHccsNCAuTLLbO5ycGbL4+OsIKheLWDHWR8fXJksbd1v3ICQXLRXL/9aGNtiAFY+TBEzvvfN61W8+5eiDWJGAkfL+8Kiu5LiYgvq1ABRi8hx29ZBtNsnAk8g64myHI0Mbtht0xLXcpkZ+VoT5wlHtqHnM2ncvspDYCllW6zvLNVMximQwQ0zmNGnbcetrYk9w4p7vUJAdrGKZiYuKMz31V42gpsOajFa755fNF4cYTO9Zhy4befxdRlG2p9H8JrWIwdpNZ990LSQ2VIHP5TW1unZTmDdzIoZ1UNnn8SPtoCKm66mm9N6SadnZOBjlxdyPBhu7U0XyQxaaBWdx8NTXuyY1o4qwLusu+ZbZcChmBdya36pdq4mq87pjtTqMPxWUGnZzy3h6MkxhhubbRC6xE/mZYU/mKrWlimfvfxBdSOwjdqbLtXmlVkrMZ0Y+PzSbHBFRGK/7Lr78bx0KHDT7dGo4RH3bmsZ1idGbqV5zZQkEkU80mzfuMfZGCJFY+DF3dgyrpzBNr9xm4ugHI/u0ldn2LI6IqpxYW/1J1Kav3Ubh+Ps4KnyKmTk++uvGkOVY31lbHROJ2JjjtH5kO2zOCcd8bqqFeo8mzdstvYvehlfk93B7Q3LFsMZdjtvmCvEi7xvLmjQbNNssiVgZrt+F1QoyA1ZdDv8qYUvjZThdndeInEfdFvOM3NA+wQs8ry3GWg9Ssc5DZ9LDPaNrdZm15zMTjczw2Ta9a/P5xzwLUGBLsNg7tOTUx5aedEhn8aMeLCK3D9JUYec7Nq8o0vjAFKqg3fAVryPJHh5JTzaOb9p2wtQp/TrXxvrvevxDLbasKFP1XrV6q1yJQgPZL7VnNqjVa/0ZT7S0Qt9zkzo2XnLBzctCdXAHma13g4hFLQM70nxfn4YDhy5fs5Q8dhoPW2Tsm01udmaa0/GJPamzuMemy5+y2cI8223yjckCNOZis8SyLptiXn7os51TZOnRBszf75zGxHPsZwyRbJg1O6Z1bS8/7oJ7LjF+rvsxpSbyijq+XF68xb7uWDflc77owOXmR9zbOrUpOUFz5U0vogrLuSQBUPkgAAMAAPidAk36t2+O7bFILGZ4n25RH1XH5HsOHTqoT/tmWMveYx5FowoUO/jWthFYvcbt+wwdOrRf55YNMWz6XuMUFNYezR7YuW2fEROUlZXHDmzXcNzeD8lEEnO93Lt9I9l5R28eno/VVph6zY9sJ/ohca+WTu4lU69urbZ9Rk99GlKQ6KytVBdr1nkgeutBvTs2q99J5ZxzeqT5oUnyzWrXatKm+9QDN72Iat43BCE3hsnJjjsTwOTlexkooVpe467b3bh4uu3B0XX6LnoamvhacxTWeaqGPwrAgYdVFTGsfqsO/YYOHdhNRgbDVC98zhXjYSad2jVp1n3I+EnKkyZNHDNF5fTH8vvMJr/RmNEG5Ty57sTS9myDNqaKwYfMAv+Tg9rIDD73CVWR2X6XxzRt1m2rZU55lc00m+3j6tSfsu8Deh+u89HJ3eo3bd97ELH2nZthtYdpfkjhCZlBx6Y2q9W444ChQ4cM6NoMq99t/Rsmzg26vVFeZsDSy0/u757cqll7tYvOxDwCDGb0ke/YdzTaR8oj+zauv+SKa6aA8WFbt5bNxt6MEOGi3I+HBtaXHXig3PzLS7PaOKmTTEti4wzt3xFtyfF67zMEeKGdWkkADr49s0Fr+U2vyAAca67etcWAGbZMnOV2ekWb2s3aDx2P3nziuDEjNC69zRNmWuyb3aOJDFa/bfcBW0w9vR+odWvZqf/oScrKk0b3V2jd5bBzkRjFnhvr0DvJ1O/Sb+iCLSfKDqmlidhBppOHdqpdt169DsNmLrZPZIY939q7Vq3WPQajRR/YQ65Rw/6r7/ql+d7YNly2EVa7Rfue804+iywzu4hZQbfnEekEHXwj52wydo34eHPF0KHDR44epzx2UOeWWK+5Bz3ZuDDr9brxHeq36khunA5o/EnHnTO5ma/WdWjeUWqldF0YIl6S9YEBWINm/YmdMWnCuFHL918IKjeIFTru7NOuDiYj31MR7efe7RrUbtXxuF9xcdSTTU1kFMaYJaNjPMpsy0Cs2+T95XeTjrw3qGtzbNOzeC7OiTw5vQXWoH2/IeRx00qmVf+tz/OFODf69o7xtRo07kG8T68WdbFu87RQyOO4qY1qj83UevLwwvTucrLdD1slcHExw3lfu1ZteykRKzZucPfOXTc/yhMyfC4sadhIcZZtNjpaAs8sV6jdd/6l8PLOek6E4VI5mQbteqOdMlhBtlat5nMfZfOKfG4ukQRgs51KWIfpO93JAOy9ZXLHxnOPeBfjBXYqA+XqtO87krjiKI8ZPXa3ZWRu/LvDSj2aoX0u22vKqtvuUU6X5tRr3HXoRDTG2CFduw5bYEgk1tTHWyajsNe0bY+BQ1UvWEX+SP5Fl7LQu1s6tVRc+qZQKPbXX9kP7d4RmwzicWHwWeV2DWeddWbF201VaIGpGQeUF4AjXur0x7A6Mm16oQNkQLdWDbHei7Q+84V87ytz+jduKq9AbPUeLTCs49ZHocneV+cMk6tXt07zzkMm675FJ9m3+NFG6mOxxm36jiI2wvgxo+brPw0TMYMOrBrXtGFdrGXHfkPOP7u1tWf7pmOv+wpxcb7rqUENWvbf58zF+XEv9hIB+GIwjufZ6s3r3Fx+GHFqjh/arsWEVad8i8r9NgaAygUBGAAAwO8UdEexUyts3ysiAJ9oWr9F86Fm8aj2n2a+W65WxyH6PqgC6nFhacu6rZR17Ynsl+9wdHILbLTGnQwxx3ru4OZt5h01d/H09LQ3UG3Xos72x9FEowXD95oi0V5Yq9HQZcbBZdXiKhB8rUsH2TqHHFEVVRhjPB+rLz/unD/R4ihOdzqk1KzlkAXvWTjb/bhKozqTNK3Kz6Min2v95eXH3/fL9DJWaYS1btWl/agbXpnOF6bU6j7JJE2U83LHaKzbjJ0BRAuw7qLeWFMVPdt8VIP32jSyQx2FnTaJPDzp2egWMr3Vrtq7e/r4B37nZuHkN+oT6jdTUDMLIar+8XeX9amPzT7nlM8IPzaiXZMlD2JY/ChT5eatumraZVJTlCHNVmOiTKPpWi4sYa75EsUWTUcctkkn0rLI/9zUdo2wbY9CE+2ON2nWorumPVHnZgXfVZuwcINJDM4NNtmu0LylbI/uchg2eqNBAjGVIOjSvF7YwNXHrdA+8nxxok+Hxgq6Npk8huPuvrIt1d5lcvO9rwxs2GbY4U/lRQJhpMnyQVjjGYftiNuSme771s8ec+RVJAtnvN0gaQE2nt1YrsMWsgWYG/9SoydKMrYMnBNgsLx18x4zb9i5e3oGJaONS/M7P7kr1vOgTRqOF1ks7dJaZsk5SzSKp+XZVTKyHVZYF+JFfjdWYnU7jVT/9HM3dopZH/UayLVvfzkI/Z336ZQyVr/nwrvRxEvciKcbFRrKT9nmI8LzrdSVajdRPUtGrDKJWYFGs2o3aj3k0Kevbp9lJgfYm59dNVJGZoymdaE4/f7SAViTWXrviZVjuu1ZN3u0vmVU5LN1Cm3qS69Ux5W2OYIMh0NDm7bsedjc1dPTLzKl4ttyCz5o9Gwr03GFeRxxAiW/WN+8aesuF/zEnKgH22RbKGq68fBM8wNDaneddtiv/DgXdX9oj1bYthdxjPT3O5o0adpvzztyg7J8r81u1aT7pBex7ACTpS1q95ypE0LEn1SzY2vGrNVHe5LvvmlcN6xrj56NGjTCpl3wJPdekaO+XJsW4w89+YxW7N2tzX2adhl0JwUvdr+8ommb0SfRuZR8e02vOgNVL8SU1/7L9zzVSa5Js2XPUYLH8bjXeybPXXI1IL/Q5+bSLy3Au0dhnWfuJluAmX7bpndrNv+oNwtnuq7r2aTlDK2HTujtvXJK4ho35PbsRm07bnvLIBbw3FxMro/6XTdiAW9vGS0jP13DS4jnvtgyvl6dblvvBZXbLl0WVqjRmvZNJm7yS001UpvQoFHTxl1VdZwToq7N7YWp6DoU4rlvJvdsjW00CSwvAEe+2NddpqnS1sfE0wrYwUarutbvvOhcdLz/XpU2tbptexhOrEX0w02qY9fecs8XoyNzUzu5pvPvRZbTpxoXpDzeMqKO/Bhtc1cPT89YqtM3Ic122sA22KobYegSzHLZ2U+u5VqLdC7L9/rIRrL9dFzQAvJjnhMBeOrVKHGxu96cBnJKC+87oe3kfGfxSFmZWcc/ED1kAPgHQQAGAADwO0kFYI/jTdrIyxl4ETUvjtfN9bU6DTnlR3QZ3j0e6zpjexDV+TfbZv2YWsO3GGXzUq/OHdQQxVwpfU465RG5ix96ZywabKGw3vannwjlf6Vz+9Z1Dr5nikUF9vsbtuna2yyOfqnI7swCrMfU24k4x11/RqM6E3a8SqdfKkPm271928upmb29t7Z3t9EHD27v2X/4GVPj9Z06DNjmxcMzzLdLBeBZPbDR242TiYXPsJk+SLaHxssEVFVOfbdx1eyRPWTItVtjaO9ewX2byW/Wj6zdd/2FWKpiHnJ1fE+sxy7zJL7A99iItk1VH4XFv1Zv37aXun1meZVbSQA+6M7nBuya2A5TOWtf8jisOPN+3dtOOOvw2WRZ23Zt1luVelpQkd/dXb0xjOzh2nLIOmN/4quIQueNc/qQC//F4hshReIixz0928pufx0TarayqfxgPfdy2xCLvPZO6YxN3POgdGrPsPzSBdp4dhO5DtstiX7bnOgn29qhJEP0Ny8Of3Nx3eThbRsR7ztg/kmbSOouZO+zk7pgCtqWSQI8xlihE92JnibT/lwAr8jv8jKsjeJ0+598JpKw8INugzbt2l0KIMKH6bq6rQdMsimp12ca7xyDDVr+Og8vsFQfXrvJolMuX0J5KWKW/42ZRKyypqfmZMe52btYmVzVnjBuSFfZOi0m73vL5oZpTe6IKR94TKz5FyE3u3cs6cpKkWl3ypMnzPV6tGfquD5tiJ3UcujW+4GR2WX1ciXlO+zo3a7hlkfx5BjsCNNZjdp0Wfe6GOeGmm1v1XLYEcc4z8uzW3Qdd6iib2aIANwa224ek+9+o4d8m6aG6MgnsYMfrMc6jT3vGPvumDLWYepmr1IttlmO65V7oeWsW6sOVnu+vmsmcW4kP1/Xluo1W6JRxyXvinHW5yuqzdrMOO8R73xsROM+i85H0W/zrcw3a7rINZp1J+rrJJr3VRdoFIC7ztrjSaxY4eetU9o2m3/MoxDHGQGHNVdP7N+KfOfJB+9ZpRWgk40beHNmI7kO6naFOD/VSo/o7SFtyPLrGSLm602jajVfcN693B1eJnHOu8Ozm/VYdPqpyf5Zk2et2DZrwri5Ry5dWjxRvv820wQBnmYx6TsB+JFm5+aD5lvT92kkXt4yAJugY+30aNsorOWKG/6lryfJL9XQbppnElLuBhRl2hgfnjamN9mLv930TcaBCeSzEVKspgyQxVZeJRuji97tGdC29dpXMdGP1rRuq7jvE3Ed5kU/34MC8HTDeF7WM61Z5Nb5YuC268EVfycDQGWBAAwAAOB3+joAy8rLnftMVLhZn6+tpgNwrv3B6bWaK654RT1w2fv2PIVaI3cYZwnz7qsoNm7Ya8KiNWrIBmTdqideBSJcVBCku4QMYjId11xPIif7YSK/y53by9Y56k4MRF2bU6dlj30OdN08+f6GQfV6TXmSibM+HZ3WsM6Uve8r6F0tznHaPUS2vvIk5eadlU2cPYzV+8oNHzeqS+fBG7xRzTr5SakAPFL9VjxRkUx9M02xtYLGC/K2SVLi4z071i2booA1bLvRqswO14TkN1vH1as/eh/ZUo4LXA4qtcWG6rxOF+Jc7+PD2jeYtW3bgqbyA3faZpeffyUtwJ9YeNazOb3q9l1/N4Z6he1wWEG23jxD9+An6o2atlI+5EcF7ezQVzZv36ey2UF3d3TGmiotUNu5fEzT9sN22qBQxvU8MLMn1mbgxKUl+2jjSv03UUVCPN9Rc0DbZgu3qc9s1F5J17Oc+jvC9j+u0hNTWHgxmExiomSPD89tg1JZwhxbSQtw8C2V+m06LLYmnj9b+On0SEx26OzXkgd0iz1fXVRTWzS2c8NGw/eSDZp+Zyd1xnofdSxAi/9wdIfmtftNWbquZAk1990PZRT6EgF44DQbsp3wxwnyHQ7Vl+vQ4Sax1QqddUfWkR9+lt5S3KCTc7rKKK18z8RzX28ZVqvpskv+5AtloQIw0axNfscizHMzmNsQU1hvQwQT91Pz6tceufstSxB7ekYPrNfiK0SzGxor6fOHF3ZBqX63R3duWeerldpt7F2yjbPdH6iprZw1sK1MZxV98j7sshQ47OjZtn7v/Z/IEXjepxXR6alpg45cbtij7Z1l+8zZvHxy+14TtYLL/06mJABvexnHSLVRb9qitdpDeoPmOhweU7fjyJs+mS7nZtVuPGDJw0SynBntZfHSJThLlOe6cUI3rG3fzdqa09s2777YNAQl1gIrjRatGnQeu7BkxdQ09x9wQwca6/P51e06DV60ceGYNoPnGcSUG9/wwvcaci0bD9z4kfpihxn32tbGNj4v3+u6pAt0kumuEZicyn5PYoToO0t6YS1VT5M3tJKKHE/qbFGbN7xObUzFOJgnFIQazWzUttNOFxTRC+yPjccayw6bI1nANXqGj9IFBRabR9VqNvfMp/K7jJRJnG97fHmH9l2GD+g9fo7OW/e7G8eMHTikR8eeg/Y9jEL7PMF84ncCsPmerpj8xMOfyMFit8PT5ZtO13X3tds6oUH9kbrvqN3B8Pho89QjqkCAp1tskG/TTPXxdy+bmXbG+9VWz+zeCuu58rQ3OgYyUQBuja03iSVfznHUUpRvvGDb9lmN5IcddCYPIm60ORGApxnGCgtstFWwRu37zF1Nb6cNa3Yb24aXfxkAoDJBAAYAAPA7fROAz0oH4ONeYpyX/Hr/yGaY/MhNl588uac7ZWBLDJt84E4aLg7cM6R1m8m7jV7ZkV49d01mojqoKPza/NoYpjhp7oTeck177H/yUx3rRN4X2sg2qzVX+56FV0raxysDmmF1xu81fvLkyf0zW0fLYF0XHosV4nnvD81ojPWZtvvGB+/oonICpbjgg86A+iiHNx901jcvz+fWHPR3rQ4jd7ihep44sdwA/JoIwJov4zn8VLeHZqaPn754Y2tr9fjofKx+izH3Spqjv5H8RmNacwzrN2e/odmTJ8eX9GyMtd12L5CI0Wzvo0PRZsOwet21PpBNaeWhA/ABp3xcGHtuYddaDUevO3cfrf3dvWNksKa9r3/OZkS/Gks8Amex/iNioywfgtUdtN4vXxhqskEOG7jaIlec9ni5fP0uC68HCvC8t+r96vect9fUhtxHli+fhaQyiAUQ5Toc6E09XUnpuEcF3yPgOe7Hp7bBag1arkfsBT2VFlijMRfds/mF7zd2IQJwNLM49sFyrE4zWbXTjx/e1l6iiKGjZa5tAV4c8dnW5J6Z+as3VnZ2d7Umta/VfvX1AA6O+xwf3QVrM3fv1Yd+UW8PD21ed+bh+5bE8tna2FhbpaClKfAxWIrJDpxm/bMBOM9eu1aTFi1WnnxlE5QaZXlYoV7dVnP00A55cvvQ4r5Y3SFbbqYR336sU6pTV2npEWOXIPI54N8oFYCLI+9u6YDJ9lUzMH9oembB8LZYwynabwtwpvexSbJY7SGrjpqgt9Cd0RxrOunK53j3M6Ob1VGRWinrVB7Oz45wMjIxffbiFSo2v7RyWKOGI9Rty3qEG6HQUbOXfF2s/aJDxo8ePzq3rj9Wv+GUJ2S64UQ+0iLuxcbq9FM5Qh5g5aK6QG99kcDmJViM61gb67/p6uMnTx5d3zezJdZ8zK7AIjzjg/aoFphs/w0X0Brc2D5SFms181a4kOW9eZQ8Nv+cs4ARrzG+OTZqn3u2CI97PLZxwyGaN6gVs7axsQ+i2lQZrte3Ec/Jw2TGrTCIKz//Ei2V83vXwbBZ+43Q+13dPgnDui5wiivyNVpGBGArFIDTX+hMrIt1m7TtxpMH5zcodcKwjqsvejLwgkDLZw/MHj9/ZW1ja2uo0bIR1unkRxafF3JjGta81eT9pnZOEZ9Mt8nXb7/iMjriEOs3dp98U4XoTZ+o/VIAxvE8R4NVxIo1nrTlPpOVfWEV0cJcb6z2K7KTd8yz7wXgVzp9MUyu+1w9dBYb6czpgzUYvvZJISPHeEe/Blj3mZuvEVthR6taMm12Pkvk4Dkv1jRvVX/opvNPX4cTDxX8BjstwOHJHdMnL15b2ti9vDyzb+MmU/e9zcXxjNdj+shiY9aeuvMhMocrKvqs04/shdBkwFF36peROFFkAJ58KRxdXG6s7tam95TLFuR2snn9wj046ad6hwPwF0EABgAA8Dv5Xe+EaknqT+NYRZ90sFr16ui5EAGY6XJ+Pta4q/ZHok9ckd9r7QH0U0cbt+zStCk2TM0gToRzY2/rLCQCJqXVwD1vE9j5QSYqHbDa49f5ZYmTnq1FVceWahejfrwjtDjZUpV42AyicjeSleZzXmUA9cxZpMGUvTddiCosI9hsPdX/usfCk+V23BOnW6oT49TZ4cTFeWnvdOTRLLoseU/MQRh3f01frPmoDT4oAPvtGdsS6736SgxR40w2H9EJa632JI7LCTNdM1aqf27Xmbs/S9o1v5H4fPUwVNVt05l6zi4affQlR7oZudjz9Ohm9Wpj3fc4ZVHNkeUgA3D9qfvJR1vnu9zYOaTkEbeY3LD5F1zTiL1T5G+0ckTJc4yxWrMP2KKQxvK5sqQOJj/teiQa4e356U1adVz1KI5b5Htr/0iqCzdBTvGyI9G3G22cLGfd3nXr1GukeNKr4oYfYcqLaxvb0U9XxrDOY264pKENlW2h2qQONupCWLGQFfFyWm/65ZZ9+3fD6itMeFOEFzibHugiWU6sTscZlxziiZ2VYL6tB3XgLL/x2fP5gfn0Q5SRVuPXfEDbuMDz5EysVscRFlS75A8TFYfeU6aeV117vXUWM9pBZ2w3chCpJbfgxGPygVE5bhdVupCH1XD1h4ll7REx0/viGKxOw5XmqcQgL8vdeBX9LG2s1WjlYe1btJ1zkmg4T3l2Ra2tZAN3GWfklinAxdHP1WeSeZDUavw65wxxcZzzxeFd6IecIw167jfxLrdLbv7bDfIN0YbsVvJw364z1UuOPk7IsyNoYRoMmHiswvbfkgC86XEs2uys4NurR345nEdsPWlNrhsrwO6iUpeS/Vu/w447H1Bgyn+v2lsGG7vfIgvHU19tHIw1H3vZNQfPcN21SKpT/dC1pkHU6Vf0+bZ6BwxrNGL5ZfJBAOVjRz3dMqM9PQMMm7j+GdrJBZ8vzEDH1mhzdChzEmyPTaYPCVn5gW3aYI1nH/nMxNNs9BeWPBMdaTJkjhXxK97i5Lf7UMhE6ne85pficW/ygC9r2W79VTeUqVNMF6OlnniEfLr6z8m00lFBM+qz9EwMujA66UxHF8wuG69RD/mKfDCoNYYtveZbTkM+O/KF9sC6crI9OremLgv12m24akUsBNPnrs5EyXK2UdxuE0XMo9DnLP18d+ygE/F9YmlFYVYnVOQll0SsZZ+DT4nu4XiB/94Z1JHfXdchXYRz3I8OrlenTsMBhz3oy29xuNmWFljzIce8cZyf4Gi8pIekO3vtdhuvf6z40gRA5YEADAAA4HdK+3zmiLbm68A8Hjfh7X6dQwcdiN83wnnxLg80j5x7S92AiItSPz/X10BuW1g43r+z3+DJx1yysiSIu3tBj3gBsY8lqoTZroab92rrfiZCpjgn5MWO7QeOnPL5mXYXdszLw/s1NXbuf+BL9pLM8bxBvYHexXsJkipawcdXNzU1NTSuPnPJKquhhMSNd9q/68D+Q85ERV+Y4/lCc6fehQ9k4BHl+ZobaB678TJFhIvTbIz0NS8+c8sl2pILgq6f0dF/HkD9Rkz8Gz3tncS7ax7Qe0E+TKk8+cFPDPYcev7qo6npJWKC8++iqIYXQq69hlxrma677Srq/4xw3fZOqdd27mFP+tlPBe63z+8l5qahYR0mFfTZgab7ycXSPnWZfOgWzk1xe3xQ8+Qdd7LFMv+Dqf62S08CiTo1w/fuxQPkLDSuffzSzCNItljWuEXzgUe8KriHtET6q7unqVkYvIun8nJRyCO9g/tvfEwnthM//sM+tDf2nDSwDfJ/e/rg2RtUyyDbz/Y2NZ2Gxk1b75KYUBxz44IeGl/jhTdxaCS81tUihnZqHbGIJSv97ETHe5o6p6+H/Nw9m4g4N8hsD5qZlt6LcGKDseMdL5Jvr3H69htJ/BFlvjG9SLylsU1QQVm7RMxL+WS096Du0yA6PYgLIuwuEFOcf/AuIy/M4oL6+Qe+1NKlvTQ5Rb6DxqX3iSUbUxxrUXqlUGG2nzk1pobGubtEtiwXO/L5Sb19113cnY4e3KehsffQi+gv+aTI4/YqrH6viYepfV++7Jcjesm22W2ZRB073CCzA7uIN9c+aRAqdThxQ97fJRdK48g9O+qkZ0Q9unhY86ZtCLnPUq3OHNr6wDWL2FTZnw5SMzmgf1bq6cc5dvrzsYYD5xnGVtT+SxHGvDx0gNg4+w/r+1EHSqLLXU2dszeDyANfkOxqfox43cjBIeTl0zNat2zjyQ2b7XLjBH0+aN7xIR84gLYqO85hH7oY7Dt61puYPNP3CTWKxllTS/J0wPN8HxloHrj9PvYHjvVS2MnOZoc191x+QR7RhYHPDXZpXn/jQ/1qULbf1eM6mo8/p5bTesqOfLqrm8yQaSefvHtKLNAho1eRku3ODHxwWYtcTg0Lf/JRd4RMDz1ttOr7D39IoH9ZuTRhquvDo9R0GsefuaaVHBcZjob6xI656hhL7LS012pNWjTqf+hjyVnHz/K3OKZ5zNAhiZyvOOXjE/rifeyRW3K5V1EAKh0EYAAAAKDqibgytZVMp112aRXl3+jPLw+rjeyMDVt9LOL7IeL/4vscH9SssaKu689nAlDJ0q20FGTajdKtsP9zceATE+3FHVrXHqX/qfwviX6bKLM1nWQUpp+P//OHbtUhiHqyuwvWdeqlULrgL/I9NaZZw56HXKibrQGowiAAAwAAAFLYCc9undr2jR0vfXL+REWcm+n1nH4LaWeMXyaXdwdduueJQ5vGymI9Zl3wpdrNcX66jwU95RdbVQdQ3T7bLjlvR4/4hwgTnQ7tWazUDOu/5AH94BtBbqjdOXpJpOjceBNMN0ZXT4yIWxeO0CsrobHrwNsIdsVt9b9HUfCb2zr0u0o59yY4p+x+xIKCSLuL66YP6Fy3h8adZLr/c2GYjdFBelKJzTO7k4cTNmDni/jiP7kyvCyfV6dXTlDo1ETxyLMMarm5ydYmZ+glkfbcK6Ua3mrKDHtyRGsHvYolNHUv2gQEPNraHGuhdMKLHvNHCQtDbXdplp7ltqOXzaIqvmGBIEp21d23ZkRzrK/qzVDyln8AqjQIwAAAAIAUTvJr08sHvqFtHZBX8Z2Iv4ab7WdJv4W0K2ZW5XWCxDP9DE7p6R4/a/2l+yU/M8CanvILLR09/VPnz5+/Ze2Z+ofb64QprqeOHjxy0vBdUknQFuRFOBjSSyLl+F27sC8duashZrTpjTP0ykpoHzrqGM35GwGYEWb34Dj9rlIM34aXc/gKCmMcbqDlu/KU7txLKIp6f/8EPamE1qGjp8+i4+m8RxzVXfeP4WUH2FzR1j12+5W95FHN3NR3D6/SSyLN0i+9GgZgVqTFWX0tehVL6Jy+8T4iM+aj8QWT597l//R32YRFkQ6HdUrP8sC5W89/oBe3KNXz9LHDR04Y2CdWw40NaiAIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwAAAAAAAAAAAagQIwOBfJswszI0t4NFD5RIVBGUxeSJ66F8j4uYkBHz2D01jiemSXyEqzox0efvRK67wZ+ci5hckefj4+qWz6AIgDW2eBM93jp/Csvl0yY9jpfv5+ngkFfD/z579P3h5EUE+n6MyGAK64KeICjLzsvM49BAAAAAAQE0AARj8uwqCHy/WPbnbl00Pl02YYmu8p/01Py49/M9hxz3c1hlrN/J8EF3wS3jp7nc2zlqv/TzqZ8MWL9NmD9a4tdwF73/1K4JKJWZEvt4/f+Xmm94//QWByOeCXKvG2B7rzO9+R/OHpLyc0qcxNveSWy5d8BOY0c/U15wzdi2ghwEAAAAAagIIwOCfJOakhlltndIVazV4uaFjcEyRkCjlhPl+siV5JVNppSjh/dW1Mg3q111+9s37hAJyrG+IWIkBrrY+fkEBAT62r12iMsS8vNDP76g52fpKt8xyUt/Z26FCh4/u+XQRwkoOpsb9HJYgiTrsMD9XqtTWM4ldQSplxz/W6Fmr74QLjglhxNgOftF5RHGSn4vVu6CEIiqWMhM8nK0cQtNYZa8DLmJnx3o6ewQkFVFvxUkLcybf3NY3OpssofCSvV3ekuV+mdSceZm2B2q1bNfxSiAa4Ka4WVnaOgTnknPhZkR8JMd1CkpgoGFBlqfre2v/mDyqQZGd88nZwSIwsbxlqhbE/KIUfxd3z+hcqgVYVBjvRm4TW/fAeOmGVXFOsOd78gUP+ugT+V7qKN+ylpZ9LtrSeX7WtpZWnmnUAcJI/EyO6uAVmoMGxYWh3k5Wn0NSqK9pBEV+7o6vvSJzfu5bG3ZiMLXPP4anU1OmvFJRbIktMvRARysz0cnxvW14FhftWnFxevjbD86hWeRoQmY8tTTEOkURC8jNCLE8NKYTVk9ptYFNcDLZq6DQx/0DOc7bRCYxFY7zCxM9HZydg2KjvWxsPziFZVfn4wAAAAAANQQEYPBPEsbf3D4Wo3WZrxMoxDnB7y8PlK9FFTWdsfN6aCZb5HuhYW2qBDnxsew2PG74pWUDsS5denRs2wTDJu65/ei6emfJZJ0WGXpnkamFmWh2RJUuxRrtNnHKQcXC3Oj3O1W6UYX1+80zCyDSa3HCh2tjO8tQpVi9kRueB+ZKonEp7PhnewfJKPSYMnlsV3L0TmOPOkbz8xz3DKmFDdv5IoVIWQWWm/rWxQbpO6SW0xGXE22uKY+1HqHvgcIKI9Ht7LwB5MwwrLXSijs+iWQu5sR/urJAli7vsPT0+0Q2WoMcO63asp173o4S5jhpz26JYRM2P4rCRYwkj/MLB9Ejd5t0+XOSiB+yd3gbrPtGswgiOLG9dDq2wFrssUj7uZxWtYhy3c8NwRq332iJYh83J+zelkn0fq3bZdL5D3TqE2T4Pd7aqyn1QhOVnU8i89D+DrzSuYOszBEXZmHY9Y09MKzvzKs+TDSTcLPtU+mZNO9zwCqMhyddW9wHaznjmAvR3iqIuzm2M9ZgySW/ImLeP0RcFON4aqoCNVNs0MLz7okMHM98M3NQa2zxNe9CHA+5LNu8LrbjdQZaYGHs4+1Y4zYbrfJwfn7Ym52TW9ATYnKDDZwSi9NerFKqQ5c0U70ekFPo/eqAXMnxPGW3TQrRMbrIz3B87fp1e48a1ADDGg47H1idjwMAAAAA1BAQgME/SZTrZn58ZC9ZrGmHgbN079jmcJKfzVNo0qTb8AWrkXlDUX4dc8AuKfqT9urRDRrWr6s4a4PGu5iyMyg38vq2cagG3773tMWrr9vYOF0+v3fZyvVbNHbuXqci3wjretA6SSgWRN1dWrc2NmDyitWrV6oMksOwlnu8C0RRj7Yp1pUbu3oPsmVeP4U+Cy+6csRpN2b2w2QHzty0CxVrbFyleutdfHk9aNnxLw8OReG5scLEZatXq07u3wqTn7bTgS3wPzSxeW2Vgw4ovBTabRzYov6k025Z5d3MyYm12N8D66x8IRwXxj/dOUxGRnb4XGJbzB0qV6/eQM3HoSJO4b3N/RrLdB07D63C0jEoCMup2WTyRYXvD8rIdeyibWiyeTRWf7iafgiKxYL01zrK8q16LNlGrMGKQR0GT9jpVIDHnFLuWmvIQYsUtNk+a/XuULvz8bepv3SDaVUhyvO8PKa+XK+djlxRnsvlqS2w+n2nExtWdUKvZljH+Wcd0Y4ttD4ytn2dtopT0R5cNaUbhtWbd8O7UBBh1L2TbP2t5x/smV0X66WyyY1o7WV6Ga3qUl921ka0YXetn9h3QN/5TzPwrMfrh9TrsPp8sAjHY65N6y3TZouxfzG1CD+A5We6XrF+444jieN/7tDGGDZA/WmamPFuzuCSABx2o2uHltg+60wiAMc/39ugbbed77l4geup+Vh9uTEriUNVU33T1uO3vfOLwp8cm9OtdS2ZToNV9ph+TvE6O6Zru4HT1XYSR/kshfrD1N6k8ETFEXdm1qldt1a7UQvV9um9iocGYAAAAABUfRCAwT8r6+HusViv+afi0N/Fjkemt63TacuDAPIlYdD+OZ1qt91nlSjG427Kt2/T+Iw3+UKZUADeOARrMWmvJYoJXyTYPDPWW9+1dSNM80lEMT/xplLDVs3Gm6cSrxUHvdi8W/dMYGLaq8OjsLaD1C68QEx0Z/bFuszVCeNzHTWH1u0wYv3VR89evHBPJ/sTl4cd/3R370ZtlI44EL2M8fhnaj2xtrP2BfBEHgcnNeuw9GoAzg3dN1S+3pSzTjnlZk0iACtg3WbcCMkLNFzauW6HeQaBZGMxP/Dm7A71em+6FR9kPXNAk8ZTLwaRDXVxL44cUD/rmMrh5n3Qa9qiYe1O8i2xOr1G3CO2Jy7OdtafL9O876STpmjFHp2br9QSU9ayYQnTTo/p1nLGeReWMPHWwG51ux+xT//5h0NVJUQAHttAvr+WY3qC5d5+WIvhe95Tx0mq9dYBTdtM2+PPLnbePK5Z7QFaVmlEea7j2UObjlgE5HNDbvXq2BTr3F4Ww9rIn/QmD4PiiHvqrRp27Kd1E21Y8+tbZnTGBi69kYkXPVo6uMnA7c9yxCybeUPqtV1nHPjj+bfY78LqLlibhRd9iCFBmOmZzZpG9sni7LdzpQJwt46t6AAsSnyp1VC+u+Y7Fs4Jv72hQ63WM46YPn3xwimQ3PuETJuNo2q3XHYpHP2d9GBU7ybys3bdRKO8uLN3Upumdbe+TmczIkym1m3WYcLFaGoSAAAAAICqDwIw+Ffx42+pj8R6zNL156Ha+otlSvVbrbkVRrfxikNPD+qMqRn7MbkhBvLyso30HAvLbZ9CAXjVIGzAUn3y9kecl/3B/M65Z6aXdA5sXKzcpmkDbI95JIvtfaqhvFzT675fzUYQ9HBvX7pjaAmltSZJguLgZ9cOrBlfjywZvET7WUBGefcBs+Mfaig0HTTdMp0cLPTUW9AEG73ZKg/nuB8c00Re7a6P17V5Cq3Hn3Misks5qADcQ+WWV5LDUWUUw3eH0CMLozTndG+seuHj26tzemP9Dr1O+zpF8zLf6TWph9XCsAZN2rdU1H1H9MIVJz/XnU8u/BedV53x5+PJx8Z0bT3NwD3w+dpBjfvqWaZV7/xLB+B2A3XeJgTfWSkjP2KOC/lFBZJxRV0JG7XVMS35xfqhWOs1RoGlnscWcK1H2wZoyzVqKlun3ZaHycQeKfC+uZHanhLNJ6i9ZeGFd5cOaTJgt3Wwk+7Ulu3WXQ6o+OFuX8n5cGxOvdrjNG2JJmYpKRYqki7QVADWts0WoeM2+tHGhvLddtjmohMpwfPVzs3zBjQnFqVFz42m9u5Ez+tE89XDa7dYeNKLieP+p8b3bkguqkSnS545+cEmU5vI99hh99UXRwAAAAAAVRkEYPCvQgF46wis14LT8cRQwJHZPer01HwZRb6Gx5xa2rVey50W0Wxh4EU5+TaNz/pRL5QFBeDVg7C+qnpE11+cHXJhclusxTrTFDTAeDque1Ns++OoYl6s4YCGzZvMvBlLTpPvf/fEVSPnsIiXB5Uw2d6LDlyhXTrz0PJjBvV4KVz0ycbsykX1cS2wJio3PHPLjq/sePM9vbFWw3fZkz84k/ZGfUCt1tN3OKMcyvq4Z1JbhYkTZnbo3EXljEu5/Z8RKgB3nWYYxUw0WdutXreFxnS7XLTZou71u627EeVnMWNAwxYjT3uRz/XlRd6+dk7nXSSDnf1Bt37jZm0X7rh5al3fOu3mHUlAmTbng978Ok16jNh1hl6xK+evW7nGc3Fx+NEJCu2HTps5TLbR4NN2idW6/zNCBeC2/fZ/zM7+cHg0JjvyqDt1r2ue0y6lVi2Ud3qzGA4bxjWp03e3RST5QtoTkwsHX3pkFPvd6tG+WZ1xq64b7h1Tp/koNS+ULNkRd7c0kZHvsVGP3q5XLl59ZhOEQnW6ycoR7XqNnTG+a6se2254/szPDzG9z67qgMnPPf+W/Dqi2On5Nd07tmGcpHdzh0juAb7WsX1LbKMN8QQrrsPuLvXbdN5hk0WMTsoNNL9icGj+0Ob1+y+4m4qW5sWqYVjL5ZeJMyrp3shejWUnrtW7QC3x5asGp1zjGPkBt6c0adt9y5tfeMY0AAAAAMC/CQIw+FfxY65uHYs16zF2w/HHDvH5wXemdajbWnHGjkOI+rT6GNZ3s0VisUjsdbpZi8Z1x606dOpVaNm/kft1AM62nTuiHdZdeeU+Xa3dsxUa1MYwHYs4Ac4Lujodw7DuM7ehN9BQHdIAa7DmZWxR6F217vU6zaYD8IWz1199TMbFGc9unt534KD+uQsGVy5qTper03xmBQH45cEhGNZYdvjSfYcO7Vo+tj3WYvyGR+nE2IwPh2a1JxrcmqpedanwIbtUAO4y+WK4iBNqsqFXrUadZmwltsXWqd0aYl03mvgKirOvrOpRF+usvGoPKl83AcNaDr4TwuDnO+jUke2s8DAZL3a/otSyTpeddgW4KPXZ7hFt5BU1qAB86dypZ+8DyKf/ssNOj+1Tl1ioQWccq/f9vwQ6APfe5cTmpdvpD29Qq8motcSG3aU6vA3WctoRKxQus58eGN4Skx06QwO9sH1OXax+b13rFHbwre4dZeud8uKK4qxnd63daLFJCo4XuV9ZIN+o/ToqAF+6cPb+S4cc4hsTfsqDlRObERu2u8a9n7j/FxEXut1e1h2T6TRovfahQ/tXda5ft82iKwGcbMd5KACrkgE41XRcq0ZYR5UtWtr7V41ojtVt3XXPh0JhXuibc5p7dPROnLl4xeDI4hHyTRXJAJz6eNlImbq9J67b/8QtzHnfiA7tRq/UJQPw5ctnz1hEoGOBFXRTuTEEYAAAAABUKxCAwb9KVOh6e1N3Ii5gSpteFOLFri+OdGlMDmNYrRHLDn6MI/KaON1+E4qXhNmmcWUGYE7YxfldMPnp+8mbLjnJL08ptaGm6LVg80qlJlhPHUvi12vyws6rT6ZeQCZo30tBhfw0v+crRrSmC7Hmg09aRQsFWdY3t7X68vzpnnPOu5bbBbo45oF6F6xp04ED+1DP4m05YNczr0KqEZnx8dj0brWxpvMMXSro/4xQAbjzxJNe6G1yQ20PjOtMzgzD6veafuZ9KBmxGEEWRyY3osuxoWvuR7DQhsy21MBqNWxyxh/HuSkftg9u3mDMjoACcV6YvdZE+vHWGNagl6apFxXhOZEnJw9sgqY/b1v98y8VgOu37alujQ6n4mSPC5InY2MtBx0wd08i+81zYt9fW9a+5MHJvVRO+GRycdz/bJMGGKZhnSfCWeHnVbpgCvMcUvDiFO/LS4bRo2JY6yXn35IPVcaF6ffXT2uFYQo7bwT8VP5FhFl+FnuV5Ol5thy8+lVIPo5nPR/buRY27bwr+luYaKVBnwnYwBmzh2ENZDfb5AkLoq2PTO1Cl2NYo267zMgu0IzAmwv6k899VtKyiE55d3ez5MDBes6/6pErEDH9Lg3F6rVc+Vz6d7YAAAAAAKo0CMDg3yXKCXV5euvWLbN3ngnkI5YZrvYv0DBiHUr8GBGFm+FuanLrlompSzkPYhYWhDq+uGX+3of+eVu8wPntc3IuYYXCgiDLp9c+hNIvFUXev2uMXjF59DLgy42PeWHOxOiIlWtQSc9Vhvu7l3TpLY+U8h4BjQgYMW7P7lha+4eEuBEjm77zTqW7UCPxjxcMwRpPP+WaVWH+RaH37eGejXtOM42gcnZRlNsT8r1vvfVJlIrerHDbZ3eI4tvGAcTP06LNyE7yvG368FEwef+oIM3f/O7DZ9Hkfa6MmM/EBkYeO4V9aebLdZ6j1BZTPP428TvLVC2Iw25PbN6h72lXquMzPyPAgtomrxwDpbspC5Kcre5Tr3gSP0GEZAc9emh62zOJHI0RY3v/3p1AstOxMCf4NTWqqfXnJMld1OygXXO7Y102Gfn8xP2/XxQEOz+m5uoQmEDuGma83QvTW+9DiDSO5Hx+jJbQ+NHjyIy0gHuPn7oSP4NFnEgB9NLcMrV1Sy3ZqYVeTuSa2vsnEJNn2b0yI8cxuRNMdqLHeTmhVncfPHFJ+Jne2gAAAAAA/zQIwABUJsa7vcPr15tywoF4dFG5GH5PViwY36ehXN+pb5Lpsj9HGLx3bDtMUcs2udrf/1scYrV11YxhrZu17WlMPWD8j0q9urRvrS7Lr/n9bPsvAAAAAAD4PSAAg2pCzIq3PDJnemkLdp2yIZ529RfE229ctZB+WwnVtTs/JNEjlMKOMtLfptS1frtRW16GM+n8m/Rh6xpVelqJGUpdehA9Uxv1XHsv+c9mp3jL4+pT+tXCRq6461tU7dt/OXEmC8g+v02mnQr8s486znK7rTN/SLN6/abp2hNd6wEAAAAAQGWAAAyqCTEvP+LDQ9PSHtt+jCTuePwL8mNev3hCv63EUwvbuHKSFT/X28nq4TML56gsqgMroSDO8uVTelqJxw+f2ltbW9u6eJB9Vf+k/AhnqydPXr0OySIfiFXNCRmJXnZW1tbvnYhnPv1RrCSf9+aPXrzyTsiqqK0fAAAAAAD8URCAAQAAAAAAAADUCBCAAQAAAAAAAADUCBCAAQAAAAAAAADUCBCAAQAAAAAAADWOT2iK0TMPXUP7DUeeL9n7cNCiy/Dvn/o3X8MU7ZpdZyzRbnLyiuXyf88PlEAABgAAAAAAANQUkfHZKPSOXX1j/eFnl+47PbH2dvII9wqIyQP/mICweLRrrBwD0W7SOPkK7TIUhlESpnfkr4IADCqZWCzOycnJBgAAAAAA1Rqq8qGKH10FrAxpWUUoQS3eY4ZCb1JKBh2zQBWRkZWNwjBKwhuOPA+KSqd36s+DAAwqWVFR0ZAhQ/oAAAAAAIBqDVX5UMWPrgL+deZvA1W23kEJio5ToMpy8ghfeeDxlYeu9K79SRCAQSUTiURJSUmJAAAAAACgWkNVPlTxo6uAf9e5u856197S+QlUCzcef9p1xvIXbgyGAAwqmVgsFgAAAAAAgBqgUrpAo5hkbO5GxyZQjVg5Bq7SesJgcek9/WMgAINKhq6DxQAAAAAAoAb4+wH43F1nSL/VmP2nkG3HX9E7+8dAAAaVDAIwAAAAAEAN8ZcDsPnbQOj5XO0Zm7udu+tM7/IfAAEYVDIIwAAAAAAANcTfDMCR8dmL95jRIQlUaxonX/34zyNBAAaVDAIwAAAAAEAN8TcD8K4zlvDM5xrCKyBmyd6H9I7/HgjAoJJBAAYAAAAAqCH+WgCG5t+a5scbgSEAg0oGARgAAAAAoIb4awEYmn9rGlefyNXaT+ndXyEIwKCSfS8As1hsHj0mn8NisejiYq6QLORz0ChkAZv75XflhFzJeOCvQXuH3isEEY9N7i6pMlREj/pHEIcK9aEq5KIBuhT9xeGTpV/7wwvzu6FtKb0aIuk1rJpKrZH42zUizmixEO0n4kBiSx1bBDjHQTmkjxQ+Osu/HCjoOOKQPxYpQJ8ldBkApZDXGhFRt6Chg4UrEKEPFx59MKHPGqkfXBHz0WcdWV5dECv8pT5FfFaWWj/igiz1s6v0+VTqmo6qZ2QdoCx/JwBz+YKxq29kZGXT2QjUDCpb76RlFdEHQfkgAINKVnEAZrEFYl6Ou/2jKzcsPGJzeTwucSXmoUtzccyzB7cffYrNY3K4bA5PIBQywm2uX0FuP3gRyyQyEPir2FweNzPE4fktYic8fO+bWsjncTkcQXbUkwdGqOiahW9yPovD+UO5Ex0EvIJkL5vbVx9aBqZziCOFhIpzY2yf3SOWSuKq4UvfpD+4ML8ZqkSgagU73cvuLrn4rz3CCwViQRWudVFrxMvwtr52VPfwYd07dr7ZQhTrJWvEZvNEhRGPTT/7JxWweTwem5ER/PKM/hFd0pnLJn6pfO6XGioAFL4YZ4Tdv3JG97DukZMX3DN4fB6PPk5QaBHyCmN8PM1ehOagaxNVCoAU9DnCF7Linc18YnOZxcQHBAq7fLGAmRzl+9glnsXjoMsuW4gLc93t7h89fPjwkav2ASlc8itJehZVHnlxLgx4eec0ujYf0bcJS2cLiC+06ZeJE0kk4rMTnE4QF+8jF28/i2XjInQ+Edf0ZE+rS0eI7XL5zeckDgrPZW+XvxOA3fwTthx9QaciUGMcMbR7bh9EHwTlgwAMKllFARjFX3QddrqzsEs9rM4kXet4ot0HFXGSze9qqzTEsH4H7BMKRWIBl5Mb8HT39FYYpYXaGe9UrqAkAoG/giPCRcEmqwfXInZBvUV69qnE1+h43qetXZqSu2XU+U8pDIGQ/1VLnhBlni9Nt1LIL49ZVHsN2ZCDshHxzQc6ZMg2wWKUkMiXSHw2MWbap2Mz62Cdplz355GfxwQejse9mTug5Ngo0fvoh0Sm1AwI6Ijhf1VENmJLuiB8QTQOUFFNesHFPPINiSm+aq38/80DXL6Imeb5SHNwG2rhmyquvOeRwRbzq0iA/wabwxex091uH5gxbszMtevXr1Xfssbc6XMul65msTlCvND/iGoLTN3QI4uL49z8mLd608bNW7Zy9epVq1ap7zvukihAdVE4xYEUDleUFWm2f8eoWUuWr1u7Ys6otZtfBGRxOFx0ohAXGTE73kFryIhOo8xiiJo5AF8hWz1FUY93LVSsr2edVMQRcKgWzQLfW/OUZ/Tf68RAnxKoDlIca3F384x5M9etX79o1uRFt58FFPCF1aTCQXzg8dJfHD00ZcbclevWLxujqmpgH5LPJvriENhsPre4MNRRd9Xy4fPXrlunukhlls4JnwwhuqYnud4+On/KXNV169epzpo898Zdj3xByXRf+zsB+MStD/ctPtOpCNQYVo6Bu85Y0gdB+SAAg0pWbgBG4QelCIbbtYVdm6A6f9elp+0TxbiIneh6Z99UrGnD+rXqNpx0wjmJgYJLUZzVju4Y1m3OrsP7ty8c3qxO+/kmYbyyL7zgD0G7SxhmtnWiLJnRum02+pyDqhM57zUV5IlMXKvedEOPDCLLJLkd19c7Qjh73SSsCMVJfn6My9Ozxx88c35n94x4weCFczwDF/L5WaFPrp898sgxPI/N57ETXK9cPGPiFJdfLBCL+PmBj/WPopGPGRg5pxCfpuL0z+eWtqk1cIFJEA8XlgRgMZ5gs2JkJ0xh3BJNbd3DB3WQg4dMXBPyGIWR708YXDht/+7tEd1TFw0DU7mpgW+I9yfcfGqVysUFaOHi7O4ZHHlk6/Dh8eOzxCtvAhKZ6OhCdSMxnu1tcFYflR0/c8Uvi+gLR+RyXkG0sxEx5pEj1x/ZZ6BozkexmFqen0eE7CL/Wyt7Y80VZu3Q1VWf1b1x3QHb7wcxq1o/7hLENirwNto0dfDiCwHkVxxF7w6r7z5o5J+Po2okmy3EGcH7z01TbNpA+55/Dg/nFyR+MlY+6UOMSoNOrOBrbDaXL0x/vrCh0pqbIRziGMnzvThnpI59TD6bx0VnEZ8ZE2oyYULbSUNmPo+DAAy+wuZwOLkhL49orx85tfuwprJn7dJYAhG3KOnzy51bVk4Y2bbP6AnH3NE1Vyjmpb3YtV5l3ZnP5LUr8eQC1V2nXqfx0DWJnldVxuYKxZkOh6ZN3WXgVEysX5bZ8pEaJu/iiqnPVPS6oDDa8sRgbOrlSHIDZH+6tnPegruJOMPr+KqRI5ddySJK8bQbm5ep6TyNpacr5e8EYLgBuGb6wWdBQwAGlaycAEy29nGSLU+MGTm8g1K/vvXl5+q/S0LjF9juGDN79vorF7V6dZQdqfs+iYnjnAzHMz2ayU8844dG4IXdXSDbpu/iVylEix49O/DnUQF40+hmzXr17ivbXGGDgXO+GPcxGNSpdz/FHk2bN5lq6JVRlOZyUrkbGZFJ7VaeDMvH8UzHU4uaYM1aKLRrSTYgY/IT9S1iWDge/3Ja5zqYss6bFFSj5bqdbtO4luIR1ywWP8/v9toJ9clxkRGal4mxM8sNwEPbYEuuSOcnhF+U8mYPVgfDerQnQnujRaZvzdT6NKTmiLQef+JpEkrsWc7bxrbDWnbq0pFaNqzD2HP2ccVC9H6eJzaPp8oQ5Q1nfVJRBi6Mtted2psuxGQ6r71sHZIv+vU++VwRzo001RyFyc+/QKyC2PP0ZNlaU/c9jUdrWSV7+qNDRRz+fN/BLfOsstF+KOaK8UK3dYuOrr7px0EbkI/nBhww2GR8+1Df7ueMPTL5OCcv7pXW2O1aq9eu26CmefCCHzpm+PTMACCxizkcbtLnq1dcPRMKUS1dyC5KdDwmt9jYO5PBF4q5RQEeZppnVh/bcXLQ1CdxuBACMPiCyL+ceDPd9fs1NM2NznTtcdk6iSnCi1JDH+hM233u6Fk9vfWjDrswiE+OJMudC7dM0/fj4WIeH8ej7ysvObTPKo4nrg4JmLgYv9s2fPM2Pc88HBegFUx9PVd5zxW7OKZIgD5u2Gx+cUGy94dzN/xTuWgAxxlR967s7bv3PZsZaWtt8cQ3WyTgoA8tYeCRUYc2HnXNKfNa/XcC8Grtp64+kXQqAjVGTELqZLXb9EFQPgjAoJKVE4DRVZib8EK7Z5fOC6/df6w+q37T6XpEABYxY5ycI9JZeMJjpR6thh58l8LCRayw++saycmvexorQFfjOMsDAxp0G3bOiy2qsl1EqyIqAK8fWq/typ1Hlk/soLztdnRR2v0FfQavP3J0Zc/OLSZedk/Ly/B7dOLkhUvXTMyML+ye3baB3MAzXiI819VwQ2cMk+s5a8/5K7qrx3TA6o3fZox2Z5qN6iA5bP5x21SUjLiel3t1ajPxrHd2ftzrHd3rNVh00Njs0YPrOstGNuiw4XmyOM/r/DK5MgKwLdEC3HXEbLXtOzV3bEc0n3tksHmcjPe6zerK1MembjxjZPokJCLczkjvpIHhrXsPDQ8uGVi36YgVT5NxvMBt/4L+GNZ2+Jz9py/rre8sXwcbrf02nlkcemNqw1rY2NXHLhpc0F05GMNqTXycUBz3Rq9f/dpNJ6qfvXr1it4G5c5YgxE7jBOLxb8YVlEVQ5z1SX9JO0xpw9Mo4pyJeKg6GOu28pxvtlhMPOSnyiFagPN9jQ7MmbnLOI28DDBc9CY1m7biclAhGuDH3dq//o5LgPudUb2O3vTMFuGsNMcTPXpP3rBB64juvm3Lx2/ccDEgm3wCHgBSqEfuEfcG8EQiYeirhQOWnHJLLEK19OLEgId6+/Xcwy31tPpPfgwBGHwrNyKeeHJOoeOFbt3OWSUxeCJGfkFKAopwXH/jI0tHHKICcOjD7dpbVr6OQ/mQuNul4MOuQTpaRgH51aJXAfrATH6tulhv+70IPi7icrg8tqvBgKUXzIIzhVSPI/KJhCi+orUlbjxiBT06MXHkrlfJXCrTirisYlRc+EF3xdzNum8z+dKPE5P4OwEYpSCUhehUVDWh6sqsWbPogSrC1ta2UaNGUVFR9HBlGLToMn0QlA8CMKhkZQVg4oHOrFjzA4N791v0PBFnOWqOq9Vkzhl39Dkk5qLrM/o/UaDR4G6thh16jwKwsMjHcHozWdmVDyJ56Lob/Xp3X6zj8L2OeWIBBOC/hwrAakNrt1xx8f61BeN7jzj47O31hf0aLTr18PamQR2bjLvwKY3sU4X2X6y1utqy6YNb1u008uBHIQrABitbYx3nnvEgXg25vaAbNnjd2QgOnm4rFYB53oa9OstNOu+Rlu50VqFh0ybKK/foHNTWXDlFAWs7YMX7PKbvxXIC8FgFukGWtsYojMETZttrN2jaRm6NfT61VOR/8WTHe1tXzRnRRabBmHUmKHEWuO9X6YYC52U/4nawiJPjejTuoWERVRB2f1zLlrWm3wkjp0v3vHb3kZl/eIDZlva15IYf+cgmSvFsF+2J7RqNXPkqToQLf+loRHUMcbLN4dn1aw/dYEa8lzDknuogrOmC4x9TxFW11x2HJ+ZmOt7cPUVpiMrixapbdA9tXdF788FL/igAJ5jqH7toF5jNCzEZ2PWokUcWH2fnR7iZ3P0QTd6NLYo115kzavGNCJaAuLcTgK+x2VyBGI+33z5z6CTNT7FEX43irCCrYwcM3jLYCU90DkAABmXiCIRCIa8wxfZst27nrZKKuMQVhoeOH3GG6/XDy0oCcMjjHaf3rndKJ5o2OXyc535xvP5F48Dc6hKAE98s3XdW2zIWfdagc4mHB5oNX3fjcUgWHYBpLOKrWV6Sw6lJI6ZuuZbGxqkXiSeE4UzfeztmDF1laJlSXsfwvxOAUQqi89DPQ8kT1RS+DZ+GhoaoXFFRMTX1b0Tr8gIwtXgS/1RIhgAMwA8pIwCj2gsn3fr80pZYvU79F6itXTSmu2ytuu0GKh955pLKEgh4HB5fEHiLDMDvkpm4iBX1eFsTuXZrn8SQLcBv9vVv3EPJ0J8nrJLNY1WVJAA3Vb3o7GysOqDdmFkLBnWQG6v/xvPproHtGo6/7JGWn+Z6cdNmtaN3Xl85r7dlensUgHVdiAB8YXkrTHHZ3RDimIh9sXpkrSFrzkVy6QC86OTbdOK7DUcthfYtlC96pKbYHm8q6f9M6aJ0OSAz4PLysgKwzQqlDtigBbuvoIh6/y5y/3NUHpdbnG6j3aBNu3aX3ZlsgUjESPE0O7Bq+86zD22unt41d0DD+qPX3oshA/CUrthYdSOilzUe+2rSgNa9NJ9HF3oade/QEjtulcLi4jw+/XFeGHBnbfPWisrWaag2hAIzN+HqtlHYoCUPQtHm+aWwykazzvE8tbIrNnz9E7oFeNEgrN86g8A84qZjerSqhXyyqogV+dn68oVz5y7e++DwQnPT1UMPozgJFgbnbzokEVE36p5i9zOmwYUiXEg9m4x40irx5VicjYlWr23WOULo4gFKIW/CxxPt1af3GbnxjEcmXyQSMlODbE/rGbkSX79lvtLTVZz+KhNV7qvPY3vB78JGmbdAKgATFxgi6H0VgMMe7dizcRnRHUdAfFmf/XbjwAO7bwUUVJcAnPpm8cxDW4xCiWc4o8oWw+lE/6VnH34dgFHwF/PTnE5PGNZn5Tm7dC7aFMSDITno5GP539s5feC4PQ+8swRE/C3zNKsqAbhUkEtPT+/YseOIESP+hQD8z7YMQwAG4IeUEYDRFbgozHD7BCrYSBmqbhhchON86QBM3gOc5WYwuKncEC0nVG0u8Lys3KTtwA02WTgf7gH+iyQBuMmCU97xgaen9Goi0wDDeu1/ERpjubtP20aTrronhTzfIFur3jiDcLTrs6z3t6jVYcRhSQDuv+RWAPFte+TTFcNqDVlLBOA0m/mD22DDdawz0AR+p4a2a1hn8iXf9Ez3SyOatmq96YKFrZ29na2NjZWVo2tkborXhWXl3AMsh8075SD1w3BCvpBblIYCsKy8/GmXIi6OF8faHBmE1em5yRa9zHc8Pb0eprSeDsBTu2IjN1+PIGo/UU8m9mvVS+NFdGG8+cLWzWv33PYmHRWLkqwOL1632cg51OnkTKxx39WPEoi34YY/29inSZtJG98RrbW/djiiKpYg1nzfhDrNph92ZBMPjNIa01hm1qE3SYKq+2tfqK701aOyg/Wm7d502Dc/7dasYQMGDJkwY/oUZaXeTRp07zda52Vg3CeLTeqWgYUsLvEM8YKwFyYaihc9eQIePAYafIUjELP9jTdPmT/34HWfXPLzRcRM+3RjSbOWg6bMmDZFeUz/Ll0atxy+7vDzKDZ8QwpK+aEAnG6/f776ZO3PLJSABTgeeH3SZr0jTqmCqvp15FeIbxg/ao1dt+2QSyb6XOML8NgnynP0bjknc8QCLnXB5QiEvLxIC60po5bsvOuQiapdfOIFNCkvy+Ly+nHjZxi8Ds5HWbj8DVIlAvCMGTNUVFS0tLToIjLaoeh79uxZCMAVgAAMwA8pIwCzOVxWXmyQ+2vE0uatjcm+qb2x+kNXnLCJTGWwuVwOGYCNBnVtMfTguySGGBcyU531lGSwln0nTJ88dmDHBs0UNrxIEMIzsP4qMgA/WD8EazRX/2Muw0N7XGcMwzpvuRuYk2WzXaFNo4lXPsV73ZpXr0HdjkOVVaZMHNy1BVar0yg9NzGe++n8shZYv8VGVAB+snwoNnjN2XAOzv6kNapb3XodBoyZPGNcn9ZYbazOjMv+OYUZLqdGNqg/QHnRkqXLli5ZrLpx51G/IjzH89wSWWzAfOPSAXj50DaYymGLRJYkMRIVHUaajZZMKzm5Uy5F6CM8P+Tpnp5Y/Za9x8ycpjyie5v6WKOJG03jUN5y2zelCzZi0zUqAD+e0LeFwnbz6EJ+ttU22cZYo24jJ0+bNmlYx4ZYkw2PUhkZrjcWdsBqdxg8GX18jlPs1KRu13mn7AuIH1Kk3vqnoXoFJ/zJtjFN6rfqM27GjDF9WtZrPlHnZTS3yuZfNl8kTPlkaHDi8Kt4tE15kZdWzxyuedmzWMgId7F/9sjM9L7p/fsn1ndovXADOu0z0/3uLmoweK9dMqo0CdKdjbZM2nI1nCEq+YFXABCiu6Y47v2mKfMXHX0TR9WvUTVbwC5MDXE0f/bwvunDx3cvbpszu33fnaYfQ7PYPOhAAL5GB+AzXbueKxWArx1aqnQQBWChQCQsdDq5YfHafZbE447Z3ppT1u286JgrrCDuVSUcAc7wN5w1ftfh58lo/fghl6ZN0XrgmoI+bvh8kYDL5Qq5RUkfrisrLTn4DH0IkWeZSCgUisX5nsY7F4+Zfukj8bxSVCwSEr9VR8/3a1UiAKOEaWhoKJ11USHKw9KFWVlZKCRTTTSS1EcVXrt2rWPHjqhcRkbGwcFBUm5iYkLM6+ugWOZ8kB8PwOnp6d27dzc2Nm7RogX1jj+1bJKXqPElMy9vwUrx9/dH74vGQTPU19f/7nqVV462CVWI/J9vGSAAgyqgjABM9qQRUJ0eCenPN47AZKaf+ER83HBYLOKn6gUBN/q1bzxQ621ikVDE5/FYSe63tvanzprGCnrmyQU8qB3/XUQADr2/egBWb6buuxyc92H/yFZYqzUmvnmiglcbOjerPfKie0ZWnN3ekeTDlGu32qh/fsuEOp2VL0bw893PqTbGei644U8E4IjHiwdi/VecCinCcVac2Y5BLcn9On6z+mo5+WbjT7pnFYu4aa43dvQiy5GWc8+8S2fjmZ9PL2yG9Z59K/CrABxvtXhgC2yqzsuEUgE41WpfraYtW55wLmQLxEJWnNNpVSK0Y/U6TdI5s3t230ajNltn4CyPvRPaYUPUUORCB2Dko9EKjTtvfRKehw7G1A9GW0oeGy0z5ZhVCgMXCfgF4eZa81tTpY17TrniFJfL5v+fw5GDFrYoydlgvmJtcqYKsw+8jWZW4R+eZKM1Ks70eKWzSKXP6NGjhiqMXLD/aWChUMQlbrgrEXGnb4eT94hOHyJBVpTFSfUeg5VGjRo6atT4k/diGXy4wx9IQ58LwgKfExNbYY3aK44cO3b0SCUlpbGTpt33zOV+qWynvzx8sI+yRQZ0gQZlIANwsvWp9u3PvEmUDsAuV7UXDdZyZuACDocnFuR5We9eMK3/qFGjBo+at/25aypHUG3ux2BzhEKmr+npGROGoevtsO7Ldj7yTWUL+awkuxOLjrwMyGcVBd5fLothsgPGThw7agQ6zZRmb9LxS8m1OjS9VgOZ1kPHTx47iihVmnfIzCsD1dC+/fCrKgEY5To5OTkqIkr+lgRgKsVJAq0klFLlkmyJysscXxKAqXJJ5pQO2D8VgFGmLZU8f3bZJPNE5Wjabwulw6qE9GjSi1He5KXKJesrvbX/JwjAoAooKwBLY3PYBSlhvi6ugTGZhRz6I4bNZucl+Xi4+URnFrHQABmJORmhnz4i7t4B2Rx4Quzfh3ZVfkqYj4trUEwmg8MvSArw/OgVk15YzGWlh3u6f/JLyGWyecVZYW5oL7l5BWSz2BmRzu4ekdnFrJyEELePXiFJecSvURSkhvh89AmNz2dxOHwBIznYm9yx8TnZqT6+Hn6x2YxiLl8gYGeGfHIhXvjkGZBYJBbyOIyc+BD3j95ByWg2JZURNMPCjBAf94+BxKJwJH1m0VHDYmREf3J3d4/LRjVhDpfHYWYn+hMz9A6JLyrOjfN39AxILuCwc6MDPD76hicWoOVBC+fv5eYZkZrPJBYOZ6WhNUOTuLh5RucKiJ+J4PCFfHZeoh8xo48ffUJimQLy5lX6fX8Ji7gxnluQGOZBzjQ4Pp3FF/DYVbgCz2LzhAJWaqS3jaWlpZWtd2RSsUAovUZoP+UmenvEJuUwiok+H3xhQcx7Oys08jtnj3Q2nODgG2wOIy3U7fNHh3d21uhIQceKlbXt24i0IuLaRLzO5hQXpEZHefmnFn65EgDwBfG5UJQR7+ERn0HULagyDoeRnRgT4hudzSIOJHTxEgkYsaGfrYljzCksJY9HZLxq830KeXtKcZK/+1viemsbnF7A4fO5xcz8BD8Hn5gMBpOZHefr4Oz03o7YAMRpZvXuk1d6XmFSuN+HD44OtsTJhwqtrN57R6YXcsp6VmFVCcBUVKN6QaOcRmU26YAqrbxgXCroflteKvhRbbmSgCoJitJQOfltOA3Nk0qekpmX8t1lKzN8Sl6lBst7i1LTSqYqb/Ly1pcaQbrP+S+DAAyqgO8FYHQx5hBPNcGJ+09YXz5iOMQTeXGpNjA2RyC5ovI51eazqCr5elcRv5CA9hCPqBiQxTj5dTqn5HFROJ9LPKyVKEalZIu/mKhFoCnJ0UXE4zOIW0XJ+SBCPgqcCNG1HZWzuZI5EQcCMS6bmiE5G3qZ0FIR1RWiFOUrqWIC+YBPYobob+KtOJL2RyGPyycXgshkHLKUrOIQC0e9TgygySRLR45LzpXso/ClIVP01XH7y9AK84jfniAJyWWhX6miiO1NbmMStbfpl2jEOS4gbuRnEfuN3lkktPoAlOGrG8sp0rfCkOc4GqVKf3kE/jD0OYIOG6lbVoiPFj66+hIfM3QB+lQouXiJvq6aVAvExbnk00ZEXoPJS7CIGpB6UULMZUt9QpUod9tUlQCM/qBSXFhYmCQ3SgdglOWorr+UXw7A0jOhUKNVEIBLlX+bTn9q2UqFVUqpwvICcKnRypundAAub32pcdCgdN/sXwABGFQB3w3AAAAAAACgeqhCAZiKZOvWrZNkOUkApoKcJBD+rhZgab8cgH922cpchvISLDUoUWpayVTlTV7B+kqgaf9PBoYADKoACMAAAAAAADVEFQrA1N+Y1HOhygzAVLqrOGSiv0vlaqqcGv8Hgy7lpwLwjyxbqWVA80fjUBNKF1IzoQYlqNGorsvU+1LzLG/yCtZX4kdCcgUgAIMqAAIwAAAAAEANUbUCcKnWSOku0OhvsgMv/fTj7wZgKhOi8VHJ+/fvO3ToQJVTk1CzQiTjfxt0Kd+WU3OWvCnyy8uGSGYuXVhm+qVQuReNU2q9ypu8zPVFC0MPk6TX5WdBAAZVAARgAAAAAIAa4t8PwKBKgwAMqgAIwAAAAAAANQQEYPBHQQAGVQAEYAAAAACAGgICMPijIACDKgACMADSWAAAAED1BQEY/FEQgEEVAAEYAGkcAAAAoPqCAAz+KAjAoAqAAAyABJvNpk8MAAAAAPwqCMA1FgRgUAVAAAZAAgIwAAAA8P9BAK6xIACDKgACMAASEIABAACA/w8CcI0FARhUARCAAZCAAAwAAAD8f385AG/fvn3WrFnf/v3/ZWVlqaiomJiY0MN/wO9dYAl/f/8WLVpgpD8x//JAAAZVAARgACQgAAMAAAD/X9UKwFTKLXOqHw/A0oHzpwLzzy4wtUjUGzVq1CgqKop+QUp6enrHjh21tLTQ39SCVbxI1PgyMjIODg500a+CAAyqAAjAAEhAAAYAAAD+vyodgKXn8IMBWDpwoklUVVWpXPojC/OzC4zGV1RUTE1NLfW3NBR65eTkqDRLrQK1bOUxNDTs2bOnZJL/AwIwqAIgAAMgAQEYAAAA+P+qVgAuRXoOPxiApQOntB9ZmJ9a4FJvVN77SgdyFG7LayimUDMxNjaGAAxqCgjAAEhAAAYAAAD+v18IwCinSRozpfMbYmtrS0U4Ko5+2/u34gD8I3OmppKePyIpuXbtGpoKlZTXSZiarfT7fjsr9BbSrbWSpZJe4PJWUEKywNQg9b5l5nM0/wrmIw0tANog5WXpnwUBGFQBEIABkIAADAAAAPx/vxCApQMYinmqqqqrVq2iXpJOp5KsKB1rpTOk9N+U785Z+o9Sf1NvKsm96CXpECsNzZkKnNKdjaVnhUaQzIeaLTWmZJwKVlACFUoHWun5lIJmK51+qTG/jcpo43To0AGNAwEY1CAQgAGQ+G4AZrwzPb77mk00gx6uUvgZ4S5n3kTz6EEcZ7paGKxesWLFyl133sWI6EI8zeGi9sZVqBhZtfqxbw4fFfKzfIw2r1tJlW7dd8IzkxoZgB9SHG5OHj4rdx83ihHShQB8V2HQ47XoyFm599SNOMlFCi/48Og0cZFafcnWl0mXVSeMzzf1NxPny1pT3/SyzhdBjtvFtcQIGnpXI4hLNIXp9uoisbnQ9rrjnPBle1WKXwjA0lkOxbxr16717t0bBTNJciuV0NLT07t37y7JpWXGV8p354zKy5tDqdxYqgH2W2j+0rHz21lRiyG9LpJxKlhBiR8MwGieKDwnJiaiV6kUXWZbcXmL9H9AAAZVAARgACQqDMBMp8enVLo2w2pPOvQ+gy6rQor8DMcPHdVN06mYGmaHmxtvn7FO8+zZs/rbVVQuXLEjV0oYc1tzu8a2g6gYOXfeKbpQiAvZifaH5q7QOnmGLDU0fhJRQM4EgO/jZno+XzNz1dLD6NjR3bNyxf79Dqn0SwBUgJPq9mSFyqqV+mfP6u1euVJH930aUZxldeXEtDmbtNHxtF1t5u7b1tEscvTqgh97Y8fBOav2ovU7sXr1DP3X/tlfIi6Bn+dpuXzCxs1ohKP71i7bvfNtElHMi7Q4uG7cpHV6qPyI+rRJ+qdskyrzy6ZfCMAISncoCqJURj1ECqU4KvdKmiglj1mWqDi+SlQ8ZzRCeXOgIuKPB2CEWk4qVZZaGLQYVByV/IEKJeNUsIISpRagzFhLzYcqpEZA7/XixYtvlxzNTbIYaCoIwKCmgAAMgES5AZgR8eL0wg6dW9Vt3LhJR9VTzll0eZXAz3G/N2/e7CHj+05QVjrsRq6iOM1KS3PycsNQcgzO893L9hy5m4IqS1lvZp22ciski0uI2ZmfTcZcosYF4Oewo54f691IzYJqqGP73Ty4ft7tKHIAgAqwQsx0ezXdbMUhBvIdL22dPuteCo7H35ijOmzO3XhynM/7Z6zcc9mjiByoHlJsds5ecOBBCDnAfbdLace998lccogkKgp7vWfCxEOBZCrmBd85vHq6SSKO57zW2TN0/nl3KiwXe9t4OQTmVmYj8K8FYCqRBgUFzZ49G6UyFM9QMqSyK/VqeQmtvPgqUfGckfLm8AsBWLpZtdTCUHHUyMhIep6ScX4kgpYap8xJUKH07x5Rb4qydJkNxVTMlia9wL8AAjCoAiAAAyBRbgCOe35sz/x1D6zMN0/t2kRFr0oFYB4j2/7aduMXZs9uX1Abov2JXMW8jyc2bZqoT9WxUOXSbvkinY2PknE83Ehda9PwafMnT548ZbqJT74YvSpmpb7eNVlt6yRlVLpg7V63XGoqAH4AvyA+wtopg67BcyNvnts5XNeVGgKgfLzc2HCbj1lUoIs1P71pymabQhwvCPkc9imcavTNe7Nu0ZZDj8LJjFw9sJ22jNDccSZEQA0y3q8ap3HVPuVLAubEvTqn2PGCD12S/9H49JxFZlFJlju2qag++Icuzr8WgKnouG7dOkkgHEuishz1apnxrLz4KlHxnJFSc5C0i/5gAJbO0micCu4ZRiUoZErPRPLWFaygRKlxpBdbghpH8r6SAFzxz/yWytLfLvkPggAMqgAIwABIlBuAWRkpmWmo6hWnp9q9/tQjVSoACwVCZiFx42/C6wsrB2tRATjl7bGTWxdZlfREFfmfm3RU+0oEznNcOn727C3nnjx9+vSm5toJKvfcc/giptvZHn0WHDZApY+u6SxYO3e3K9wDDH5F3ifDI2OVL/lUpxY78GcxQ97ojxvRf+ycY+ZZXxo0BTE39NaOG9B/4UnriHIu21VTrt3S9Wf2Pic7NSPigNuDVl0zj5E8d0LMinxzvtvylwl072ZO1MMrxyZc/uBvfNJgwUk7S6M945Dxe669reyPqV8LwAhKkiiqSSde6axIlRDNlCTJS9I5sMxMiFQ8Z+mpqBZUNDIqocb8bgBG0BzIhfoqZ0rPiipBc5AeRKTfurwVlCYJtEh5GbXUfKjlR1ugggwMARjUIBCAAZD43kOwin215ndrMK1qBeASxWHm5yQBOPmtvsHOJR+yyVeQEOOZp85cCWaLCoI9Q2PodMLxOjVr6Lan/gW8/CQPj1j6tt+8jwYbp6kYBH95mBYAP4Tl93iH4vC5F95Xx8cWgT+Fl5fga2llY35G66a6vlfJ7RnCwlAfZ0ubN6YbVt25bpNKdFWpHrLfLt9vcNhWcpt8xBOlDcbPY6UCcJSdQe/d1hl0EzE//vnti8pXP8a8PLxAbuyKnS8+WCI3tOaN1zr+hritpdL8cgCu9krlzH8TytjfPn/rB0EABlUABGAAJL4bgP2qSwBOf6e/X232nWjyFfSaq+Y47U1XgqVuNCP5X1dYbeKY9VXWFWd8eHS07x7HatXqAv44ZsDTXYpDRh63qk7dVcFfFHFp+5rBao5fP6AAT783beSOnS8SK/mJx79Pof2KqYfUb0bTK8T9pKe4yuBl3JfnfBVHvTjbbfid4JI+0n4mp7YqG34MNTu5d+iBN0n087JSn80evXXb3ZLZVAYIwOUxLOvHjf41aCG/vWH4B0EABlUABGAAJGpOAOYGXVPbOXqLI92sEGa8cJ/eMT+O0G3HAoNLb0secp1js7/veWv/+E/mc2aaerGoRpbiiGdGO0bdiag+TS7gTxNmut7cOHr8rju+SdUmp4A/jZfl+kh73mmXkv4C8bfVdUateB/qa3jq8N7XJV2Ec9+q9juwzTSyMps6fyuR/7npa7brfsyjBuPMxi2/+Cig4MuJw093NlrQbYc11X9HlGpx/ej4Qx6sDHu9i1t3fCrZWuxPK0fs3XApqBK3CwTgMlE9k385W1YJEIBBFQABGACJ73eB3je7Yx3lQ05V8Q7Y4tCnp5f23+dCrSI/7unhbatXG4cwmcycTztn7dx+wRNVseLvTBo+f8N93xxUnGSrO3btGZc0Ni/Neld7RY3HYQwmMzfc8uKmCdoWWXTzAwDfIS4Kf3a4d++Vl5yTGAIhp5jJLGZzhPD9CfgOMT/R+cKOvpMeB6KrETP6pf6WIfNOebI5TKcDC7bO13bNRaWpTuozF6iesUn7+neCqjR+ysvlyvv33iWuzXnOJyaonnodWiTGxQJuMZdPnDjCvJA7BwdMuOuJRsj0faC3caqOCx8XF7tdVt89b9enDGJzhZtuGLps89WAyvyBKAjANRYEYFAFQAAGQOK7AThAd1n/1vNOfJLcO1uFFEdYXNowRte9ZBV52cFXdixU6NevX58RK3VsIouIpgIxP8f76aIpo/ug4l6dd9/2ySMrltyCwMeHe6Kyfn2HjZlxx7kQ2vHADxIW+d9d3rZB0w59Bg0aQBxC/fpNW7/fPZ9+GYDyCYpinunO7kEcNAMnzTN4n0b+irmoMNRq9xIl4iLVe6zaybfxjGr2dZwg2+XW0qmKaP36dt180TGZja7Nwsz3x2YeehnIIK7TImaK9TFVYrv0GTl58WMvJrkBREUZfqc1VInt0k9h4baLfmmsSr1QQwCusSAAgyoAAjAAEt8LwGI+syAnK5/Jq4r5TyzgsApzmTyppjcOMycxAUktKJaqQQryM1OJ0sQksq5FE7NSkojS5PRsePwV+AkiLjMnNSMjlTx8SCkZ2USVHoDvE3OyyItUUkau1M3jAlY+eZFKSCviVsvOBJz8rGRytQvoK7ZYyMpNyWFwRSWry80hz6iUzFzp+M9lZBGlCYk5jEq/TkMArrEgAIMqAAIwABLfC8AAAAAA+D4IwDUWBGBQBUAABkACAjAAAADw/0EArrEgAIMqAAIwABIQgAEAAID/788FYOpByiYmJvRwtbB9+/ZZs2Z9+/dv8Ze3GARgUAVAAAZAAgIwAAAA8P9BAP4p/z8AU5ulzAkhAANQGgRgACQgAAMAAAD/3z8VgH9vm+pvb6FFfnsAlp4JBGAASoMADIAEBGAAAADg/4MA/FP+fwAuBQIwABWBAAyABARgAAAA4P/70wH42rVrHTt2xDBMRkbGwcFBUi6Jeba2to0aNQoLC0OFaDTKt8GyvLlJXqImRLOKioqSLkFKzc3Q0FBRUTE1NRX9nZ6ejmaopaVFvUQtTKk5UCXUCN8NwD8yc2rCbxeygnX8EyAAgyoAAjAAEhCAAQAAgP/JySt2xLJrdB763ag4J0lxKBlKZ8tSAZhKmBW0qZaaGxqTyplUuXQo/e7c/P395eTkqPmgd1dVVV21ahX1knQ0lUwrnWml51nm/L87c+k/Sv1d3jpSr/52EIBBFQABGAAJFIAFAAAAQPVF1//+AAaL+9DKT2XrHY2Tr1x9Iuk89LuVCrpUcygaLFX+UwH426mkJ0ck74L+Lm9u1KyohlkUbq9du9a7d280B8lbSIdYBM2ze/fuklBaZnaV+O7MUXl5MylvHanB3w4CMKgCIAADII0FAAAAVF+o4kdXAX+f+NS8E7c+TFa7fcLoXVhUEp2E/oxSce7fCcAIiqboJTRPVVVVNC0aE02Ccm+HDh3QIPqjRYsWVM9kiW/nWd78K545GqG8mVSwZf4ECMCgCoAADAAAAABQQ/zeAPzWNXLDkefzdty/b/E5IyubzkB/Uqk4R6VKNFhBzCsvUiLlTfVrAZiKo0FBQbNnz05NTUUzQWNSwZV6VboFWFp52VVaxTNHypsJBGAASoMADAAAAABQQ/yWAJxTwLrz0ktl65295yydPMLp6PNXUHHu23uA0d+S1EflVeny8u56LS8cUnOQzpPS9+tWPLd169ZRE6LIOpZEzZ96VTJPadJ5VfpvaRXPHCk1E8lClreO1OBvBwEYVAEQgAEAAAAAaoj/GYAj47N1De1R9D1358Of7u1cJhTn5s2bZ2xsTHUnln6mMZVaUSEKeO/fv5f0DUZZkRr522BZQTiUzA2RTrwVzA1BgRy9JJ14pdMmVULOkiB56UcCMFLxzKUnlF7ICtbxT4AADKoACMAAAAAAADXELwfgt66Rq7WfLt5j9sTa++/0dgZVEQRgUAVAAAYAAAAAqCF+NgDnFLCMnnlMVru995zln3u2M6g2IACDKgACMAAAAABADfHjATgoKl3X0B5F38umzjEJZdzyCsC3IACDKgACMAA/iF2CHq4G6BVC6AISXYTQBRS6rDqtP/h76IMHDh9QsW+OEPq4+aGLVPVBr10F60e/LjUGPfwFXf6N7wZgLl9g6RS2WvvpygOPn1h707EGgB8DARhUARCAAfgRbK5ARNcZxAIehy6t0tAqCUuqQSIBt2SdOPySFcWFfC5dhWIT608VVpfVB38Rm8sX0oePSHJQAVAKus7wRbjUEcLm8AQl113xl3IOT0hfjqQvXdUHsRno1f5yEZZGbJeSC7KIvCCz2By+5HpOEwl4ZZ9rFQTgnALW1UduKlvvaBtYQ29n8GsgAIMqAAIwAN/FFeGsuBfnNSdOmDBh4lK9J345OC6o2rV4tErFieaXdo4eNXLkyFELDt32ysLFfDZXiKd/2r92Hlk65pJjQiFPzOPxBNyiGIv9k1Dh6NkrrnsXifmcatvyAn4zDh/nFcW9uzhmNDp+JixVNwvh4iL4DgV8DWVfdJ1hpTpfXP3ULaGgGF1iuCJRcZrng83EgTNy0qqdTyP56LrLFuB4hts1rbnkRWrhqWd+ecSli55NNcAR4sKYlwfWTyHWb/Tx14G5XHTBpV9E2ByBoDgn5LHmGGKE2at3O6WgrCvM9L1/VJXYUkTp6LHjJ4wfv/u6ZSxDJOTSE35RZgD2CU3RMrCdrHb7+iMX6O0M/g8IwKAKgAAMQMXYfJE40+u6hmJD8kcLMKy+3CTtJ35ZIpF0paRKQauE57le3aw6dqb+Qzs7u0cn5008uP9uJFfED7u3YtSCdUdM3tja3tNSG7fXzDurWMCIeqQzaejO4y9sLR9c2LFr4krbaAGPy4EIDL6HzRNxCqKtjo0aNnjTBVs7u5f39I5fWH8vkF2tIgv439hcPp+XZX9k0oA2mLZFbCEHFwuZiS7Xd4wdoX3J2s7u1U2tlXvm7vqUIxYVRxjt2jBq7i5jW1tbsxMzJ+odfxFdLBJ8G/OqJK4Qz/XQWrR38U7DV3Z2z/Q2jdlu9C62QCikG4LZXBE3P+610QSlfWcs7eweXt2ntmHxnUA2XpwR5mn/imBpZ//0xLhufbuvuOPHFIrKaAUuFYDfukbO1zBdf/jZi7e+dIIB4H+AAAyqAAjAAFSII8DxdKdTs1phvWadtg/0frhieDNs2LZ74cW4mEePU8UQrb8R97fuGLLExJ+6DrBiAiMCIvMEeNLzGZ1X7H4aTZYmmC7spf8qJD078pmagsoJjzyiMN3x1Fr5aY8jxXweNAKD7+DwxLycgAfHFDpqm+fyyKMq2lJ/z6bVb1PwMhqmQE3EZnO4/Czns3OmDm6/dOaS1vLnX8UVossuI8HGYFmfxVfi0N9I1OvDu5asfpeHp75au2zxvEsBZGnCrcmbdA7ZJOK4oDr0KuAIxIKwOyvmrtj3NIxcP57z0WHrDZ8H54lERAJmc/gCVqrbPbXBcy8Gs4gRGEF39LeOO/mZjQsl966Ioq9sWDJR09iXKcbL7AQtHYC5fMGGI89nbDGxcgyk4wsA/w8EYFAFQAAGoGJsDrswKz7cxzcqiYHjRW7q0zsPXH3JL1dYVduwiA7dwUbbd0xe8CxFLGTm5eblswS4GBfxhXjuR50xSltPWcUWcYrjbQ+NVr3yPjaXzcyICUzMZ7A5vMI45+sHRw418GQKeNACDL6Hw8NZGR4Pj3db/TxVKBQRdy1mOJzSWTrmlBcXF0I/aICwiQSc/Mr0wXu79+E2lxS6nXsRlc8V8liFGclRwcmF6ArMR0kw6ul27VFz3mTiTC+dVSsmbnsYy+Fw0uw3j9+376p3vlhURb+O/BpPjGdbrx65c9MZf/Rpw+fyxdm2m4etP/ciIl8kRFmWwxVzc0IenejZw8C9kLyzXpzheE1v/sSTXsVkP2niC9sYs60rl6sZerPRHMq+SJdqAUaCotJ3nbFU2XrnvsVn+IFf8D9BAAZVAARgACrGYhVz+GR1oTjsssaMbmOW6r0Nz+dyOVU1AKI6Vp7XxePH52g9uHV+fb++/fr2Xn7AMDhbIOJzOMzMWFuTGZNG9evfX2HPJYvA9GJUy0TVMLT2mX531vbtOWDQhNO2eazq+OQZ8Ptx+KLiVK872/uN1nkVmV9YmB39Um+e7KDJZ1DtHAIwkGDkFnJxXFDkckmh65nnUfkcIZd4IpaQeOYeV4iLUh0v7l8+Xds+Tyjis7PiXa9uWaLUv3//iRu1Tf1T8jjoelwtvo9DF+ek14s3n9xjHi1CJwhxX7T37cGrrjwMziT7MqMAzMoOenGm+7qnUYxiLocnwguCTS/pDtd2ZBBt4Gj6fPdD8zfP1nmbWcGDKr4NwJS0rKITtz6gGAw/evQ/bd++fdasWfRAzQMBGFQBEIABqBi7mFmUl5ORlV/IKMhMS/S01Jg7st9Gg4+pAnHV7HSH6khFoSYHpjYdvHifTWB0dHTEu7OLR6zXuBrGEaZYr529acdlmwBU+vGx1tTRx82Ds9hiIdH6wMxPj4uM8nO4fmHHnGMOudAEDH4EhyvgZoY9PTytk0KfPn2mrl5w+sCpgxNPfuZAAAZfoKjH4/MY2R8ufgnARDER8YSZ7rfUpyhOP22fLcJxYYKl7oqNq/Za+UdHRXu9VJuvpn7tQxq/6j6SQRoZgJdsP7XvZYyYCsBC3ztDVxs+Cs6SCsAWZ3psNo8pogNwqNnVo0o6TkwUgPk4XvBx//xN8zQ+5OF4+W3i5QVgSk4By+iZB4rBRwztwqKS6EwDfsZPBWB/f/8WLVpQjxhRVFRMTa3yXz1AAAZVAARgACrC4YtFGc4X1vVoM2H7k3DinIm8PacHJqt69jPxDXtVrHOhOlaB79Vj2hMPOOSjahBRE0p7s2L4jtU3Y2M/LOy/TcM4kkOsqLAo3GRVz1XXHZKyi9LTi5gcoraF44X+94/MU9zvWCDkQwIG38Uq5vD4fGFxemxkRFRcRlaA9RmDzTMfxVSTPqvgd0FZl/t1AGZzuAIRN+2jscbMwcq61vHFxO+2iaPNNdcvnWvoS/0KkDjCbHa/XdvuhzHx6nBAoYtzptWKsXs2XQhk4mI+yr95b3cOWnX2WXhuSQDm5AQ/OtG971WPwmKyC3Smy3W9pWOPurNwgQDHc132Ldg6f9f7CvPvdwIwhcsXPLcPQjF47zlL+D0kaT8Sbn88AGdlZc2bN8/BwQH9nZ6e3rFjRy0tLeqlqgsCMKgCIAADUCHiNykSbI5ObIB1nrHTMSHN88aqflitsTvvRjGqaI2LI8BFcbYnj29Y9jSCrgalvFqutH21UVzch0X91TXvRJABWMSKMFnVR+3Wu0Bb43HD9C3Ti8lxs91v7ps34pALBGDwA9jsYkZBVnpMInX04OlOJ7YtnHHGk41XixY78NuUDsDE1yZFKe8vrh+kvPBFMFeS2GKe71y3ZJ6hD10Qbja3/67tD6pJAEYXZ77/1fmq6/e/jiFXT+hloLTq0mO/bPQ5JBQJ+HyBgJHobLSizwqTKPKUYoU/OLNnzGFHhpiP8m+e25F5O2bsdMipMP/+UACWeOsauVr76frDz+ApWZTfG4BL+eUJ/ykQgEEVAAEYgIpxBGJRqovBhgEN61J9lGo3bKZ68lkUs+o2YfFEoqLAG7sPTltl+jEkJCTQ6qTqiI27bkRwhKm2a5WXLj5p7hYSEvTOYP2QQXrPI/KZ+QHGm4ernX7tHRziaW2otmzM4qchAngKNPg+Dk/EyQp+ZjB61Jmn/iEh7g/1lsyftNI6QSisBmkF/E5UAH53pkPb408j83m4mJ/r92jz+NYd9t1z9w8KDvD18fYLiclmC5OsDy9cNmuTGbpIhTgazR+3aP0Vpwxh9egCTV6c095tnrF77TELH3TC3N41au1Fy7BcgZCVGRsUk5bL4oq52eGPzg5TOvEAnVEfzU/tWDfzvHsRLiA6i4cYr1uzZP6dcOIGYnqGZfmpAEzxCU2hnpL1xNq75jwly9bWlvrURxQVFRMTE1VUVOhhDENJFeVV6U7LhoaG1KB0js3KypJM1ahRo6ioKKr8W1RfaBMTE3q4yoIADKoACMAAfBdHIBIkOpzaPKZH9+7de6zQvx9eKKrazQ1snkiQF3Hr8Np+/ZF+qw/eDCOeLsPhCQSR1mpzJxLF/fqsuPkpmSEQCcXcoqS35wYPIgpHTVn0LIDL59O/SglAxdhcAS8v1uHq8r7E4TN4xvIbHvni6vKjreD3YXO4HEau+91pk2/ZxRUJRLyC6Fcn+/Xq2XuA4qCBA4jrVP8Ri3Y9DOaLRYI4e/1N46mLlNrJxzGF1ao/PVcoZvqbbpo/nFy/vffc09giMZeRYKU7U/upZyZDIBQJilLdrq0hNkm/cXPWW0aIBDx2MVckKvK5pHly7eHPuWJRhWfYLwRgSlpWka6hPfWUrKSUDDruVFMojsrJyVGdk6VJh1uUkGVkZKhxqKBLdWCWjEMVSsaXJGRqkCJ9D3A1SL8IBGBQBUAABuAHsDkENgVV1KrsI6AliDViF7MYRUVFDAaLXD9ildhcLqeYxSSLmcVoTclmXrTOHDaTgUoZDCYLjQONv+DHkacMizx8mCwi6MDRA34ECykuRhceCSY6gMiozEbF1EWKvDjRE1QTbGL96BOGeAw/dblFl2UmsT2IEb4+o0o9Avv7V+dfDsAU6ilZk9VuV++nZJV3R+63rbvUONKBWTJOqRSN5tm9e/dvQzWFekfJzKsuCMCgCoAADMCPQdUsLqn6VLfKXiVJKZ1+KajGRatutU3wFxA19pLDB44fUC629NdrXy46NMmFSuoiVT2vR1+uzdJbQ+rkkZxRX1+80XQ/sEX+ZwCm1ISnZFGJFMMwSTMvIh2AEUmjrnTrrnQAlrTuSlTQzGtra1txN+kqAQIwqAIgAAMAAAAA1BC/JQBL1ISnZEl3dS4VgKmQbGRkpKKiIkm25bUAfxcEYAD+EgjAAAAAAAA1xO8NwJTq/ZSsUt2bS93Hi0qwrx9wJQnApe4B/haa88SJE6m5fXfkqgICMKgCIAADAAAAANQQfyIAU6rTU7KkHwGNSFp3Jb2aJUmVGlM6uEq3ElOxlpwH4dsGXio/U6pB+kUgAIMqAAIwAAAAAEAN8ecCMIV6Sta8Hfer/ZOiKT/bz7nagwAMqgAIwAAAAAAANcSfDsCU+RqmAWHxdCSq1sr8caMaKyMre8Sya/RBUD4IwKCSQQAGAAAAAKgh/k4A3nb81VuXYDoVVV/Sv4QEkICw+PkapvRBUD4IwKCSQQAGQBrxy5MAAABANfV3ArCuof0Ta286FYEaw8kjfMOR5/RBUD4IwKCSQQAGQBofAAAAqL7+TgC2dArTNrCmUxGoMS6bOhs986APgvJBAAaVDAIwABJsNps+MQAAAADwq3IKWJPVbtOpCNQYKw88DopKpw+C8kEABpUMAjAAEhCAAQAAgN9itfZTV59IOhiBGiAsKkll6x1691cIAjCoZBCAAZCAAAwAAAD8Fg+t/E4YvaOzEagBjM3dzt11pnd/hSAAg0oGARgACQjAAAAAwO+yZO9Dr4AYOh6Bai0mIVVl6x0uX0Dv+wpBAAaVDAIwABIQgAEAAIDfxckrVuPkKzohgWrthNG7h1Z+9I7/HgjAoJJBAAZAAgIwAAAA8BtpGdi+tPejQxKoppw8wrcdf/WDzb8IBGBQySAAAyABARgAAAD4jVAoUtM1h6dhVWNhUUmrtJ7kFLDoXf4DIACDSgYBGAAJCMAAAADA74WiEQpIcDNwtRSTkLr+8LPI+Gx6Z/8YCMCgkkEABkDiRwKwWCQQCIQiMT1Y1YmFPLTWbDZXSBd8RYguEPSfJcTVZ93Bn8HncohDisMV0QUkMZc60PhflQJQCnGF+eoSIyQPp3K6VorFolKjVw9iAY86i8q8MBNE5OtsDu+r7SIW0ueZ8J/bKGlZRSgDQ1/oasbVJ3LxHrOfTb8IBGBQySAAAyCB6g30iVEOcbKN9oKe3RYffZX8oze6/MPEglSn01vG9e2HTNp13T1TIF1n4ie91VU2ehtb9CWwsGLsTffPfBFFDwJQmpgb+HTuxBHoiBoweKiBYw6fLBUWBptrK/RBpUNnrbgVwKgGJw/4E8SMqNfak+66pXHpYWG+72PtXsQVSll1k2NyqTQoDLmzXlPfwD2HHq42uMEPN84bRqx2rz33vHK+CcEiZuQb/d7E6/3GzV5vE0ONIOanOxnsGE8WT9x2zi39nzvPuHyBloGttoF1TEIqnZ9AlZWUknHuzgc1XfOf6vksAQEYVDIIwABIfCcAi1Ms9qq0xjBsoua9OKpmX5Vxwy6prxu1wsAtJiYm6NmqiQe1TMLItRIWF+TGvT4yqm8jbOejOPKjTcgtSo9xPjVbtntXhbshRAkA3xCG3p8yfNkOM/eomJjI94bqozTM44QiTpTFzp5Dxp63QwdalJvJ5fOLrrjzoRsBkCYWsPJyIh9qDFJojunZZpDRTZzr+2BvzxF7H4XExAS+PrV386KLflKZjhfzSm1EbfnF+u65dEn1IMp5v2nG/i0XPqDzJfyp9ti1l+2ipQOGmBFtcbj30O23PdAIgfZXD2yadd5bhIvz3p5cpjp1/atA4oJuvmGmtsa1EO6/eJ5ZOoWpbL1zwuhdWFQSnaVAlYKir7G5G9qJd156/fhTr0qBAAwqGQRgACQqDMDc2Jf7ZndsVK9ePWzGXrP4qh+A0223rV06+0o4OZDzaJa67kH7TPQnz+v8osn9um7buLljxwsPY5joMsGNMN3Ya4JSn00r93Ub/DCUnAKA0pJezh6id80tnxwQZTvoK6jbRqcHmZ/tNvCaN1mI49lOF06pL7RMowcBIBQ46k+e0L/Ljg072nU2tKVagBlBjy+OHHOb+sKN8emK5tzJ16jLFcIJfnTz5qaJCksPnXDOosuqBXHI9fnLNx96l0ENht0ctdbweWjRlyTLSbA2HDnsqgf1ESRKfHJVe9SpYJzlefbQ5tU3A8hSHPc5p3L44NWQYnrwH4NS00MrP5SgFu8xu2zq7OQRDmH4H4dCL9pNKPeuP/xs7Oob5+46M1h0R41fAwEYVDIIwABIVBCAC/xPzx8iu3Kd3orZitiYbdUhAPOibhzYqbLphntkZGTIm60LzuibRRLf5Ra5fnCPjE5h+99V7H76XiQDx4W8tPd2Hvk5cT73TnVXvA8twKBsQh6DKSjpSM8Mu7VywA776IwQG4MBK82SqVI8793xbYsmnA2iBwEg5Du9c0+MTy5wv96zu4FlMlmzFgu4XEaxSMQtTI/0fqq/b/OSm2H0wcX3NLty3dTC8uqKzQcOfyC+t6s2it6tnqitfjWSvvWE/fHAkLUGr+K/JFmxkMctYpXcmVLkZ7RvyeRzviIxn8PhsEt6S+fZbR2/Z9OtiH/9cyoyPtvomceGI89RGB606DL8+2f/odCLdtP5ex99QlPonff/QAAGlQwCMAAS5QVgcV7ACdVhrRfqh6XGn1rSBxutaZ5Ov1SVcTLj/Ux2rRvXv39/5Y3aZoHJOTyiWEw/5SrwjmK3k/ciCskBsogRdF8fAjD4AeKi4Hs6iko3HIvEYk7Ma/2hwzdfdw4ICAj3unt4fvtR0y4G0yMCQBBTgY792bBnt4t0AKYVBphr9e/Xf8q8004xmWTfeW7Up73njQ3CC9PuqK7ZU80CcI7d8l0XD1qn0oN42MPhareexzDowa+IiwNv7dk2WtUm5atHy3FzXE7NXbXvuH0G/X0BAP8cCMCgkkEABkCi7AAsZPhdXN0b6zP3sktYmKXa+E6Y4tLTzon5ki/bqyZ+0st9cxev0PMt5HK5eV77Fy+ef9wiSeqeMT8TqQBMygu4BwEYfF9R0P2d/YeOPvcxlzpH+NmeJqsHDFRUVJyzde3NEyanlS+W9NQE4Ash0+3qNwEYFwn56BoV9+LAQeWFj7P4wsRPJ248uRiBXkk2XrhBW/+z1DWq6suxW7bl3AGLkh4TePDdIWtumJcVgFmhdw5M6z1Z7W0SXUDiZrifXzRqyqozbtXr1mhQzUAABpUMAjAAEmUH4Fy344tlsVI6rLzmSd3qWFUlvdq7bd28+wn0YLHj0kH71t0M+5LqIQCDX8BLDbh7eOiosWcc00t9Q4Q+bcQ4nvnO4OyqRVY//aMZoAb4KgCLuHlp0QEZkiycYrLr4PBlziE+uvrq48+9cHJ2dDTRGD1t8aoLLvE5/+TTnn4Jy2HDqD3bzpdciVkfNo/cctkmiUMNlhDm+t3XXqo4c51dIl1Cykp0Or18/Nw1Jz5Wuydjg2oGAjCoZBCAAZAoOwAXhT4+uXTsuLFjx4wePUKxfcsGWJO2CgtOW0Yw6RGqpizHYxvnzzzlQtWzRL63Fg7Zs/9V0peudBCAwc8SFLpf2z5hmvaX9iuEz0iJcfGij6Nkm4NrZy65G0kNASBNEoCJmzH42c631w9bcJm+47DI8+yCDbOOuAS/N9IcPXrUqBHIgC4tZOXb995j7JL9VR/gKi3u2Yb5K7Sfx5ADQjc9pS3GNnHSn0xiQX7o7V2Lpqx5QI1TghHveGjJ2PlHTKv2V7OgZoAADCoZBGAAJCp4CBZNGKyp3Abrv8Hs9zwGolIxfIw01FUWnzQ1Nze/e3DKuFUHzAKlmxm8r3dtpWsUWkAPIrm+Rjqtu94KpAcB+Iowx2pv+zp9Zhw0ev7yxXN0WL149SYqR8iIfWs0UWnfaVTw8KS6iuqiA75F9BQASBMyPp6Ta3XyZSJ1IWJE2xhPHr3/LDpyzK/vX7Fh9p43X194wy5PWbJDy7GatXayA/VXaS/WuEScMDqbJu43I34XWcxO8bX1is8XikSZn45Px+qO33/nlQV5mplbO3nkFvPDjdd06dVh5InHr6mzz9zGM5p6rAMA/x4IwKCSQQAGQOL7AZgbd+/4lnn7jZyzSnXvrJo4oQ9PLldGJk/VM3WQirqE2Pc7Nz//kCL1OxqM+A9Pt+z88FWfOwBKcEPu7ti4YJbKtCmTiYNKWXn6nIUvQogurMwYq31TiJJZG7XeSLcOAyBFxAm3Vt/82itH8vDiwvDXu6kjZ/NBm28ePZhkc/Ly3UfB1eoeYFKWvf6OOcRqTzX4EEt+GyDMcbm0+uzbKI6An2B3bPa8OTOnl5xmykt3nQjLZrnePzpz1qyZkrNPeeUpi1D4qgn8oyAAg0oGARgAie8HYAAAAAAA8D9AAAaVDAIwABIQgAEAAAAA/igIwKCSQQAGQAICMAAAAADAHwUBGFQyCMAASEAABgAAAAD4oyAAg0oGARgACQjAAAAAAAB/FARgUMkgAAMgAQEYAAAAAOCPggAMKhkEYAAkIAADAAAAAPxREIBBJYMADIAEBGAAAAAAgD8KAjCoZBCAAZCAAAwAAAAA8EdBAAaVDAIwANK4AAAAQPWFKn50FRCASgIBGFQyCMAASGMBAAAA1RcEYFDpIACDSgYBGAAAAACghoAADCodBGBQySAAAwAAAADUEBCAQaWDAAwqGQRgAAAAAIAaAgIwqHQQgEElgwAMAAAAAFBDQAAGlQ4CMKhkEIABAAAAAGoICMCg0kEABpUMAjAAAAAAQA0BARhUOgjAoJJVFIDZHJ5AhEYQi4V8HpsuLC7m8IQiqpgg4HG+vERMIhSLhEKuVBn4ezh8gZDeMTT+V/unDMReRrtM8J3RKkbO5CtCdGDQL/4UNvc3LM4fwUYblzruBdKnQ5XwZf+IhFL7hc3jU8Wir05j4gwni9G4pVaUzeHyhdxf2rOgRmGjA6Xk8EF/k8eTBDrBq9opBCoJWacoOWgkVx50SNFXY9F3P+KqJDZPQF+Fy1k/yUmFtkvp+hb6rOLz+BXVwtB0dBUQgEoCARhUMnQdpK+IpbC5Ag4jJy0ukpCclc/mU1dZDrsoOzEuiiyOjIyKSswqYBazySstm8NmM3MSI6Pj4nOKqCLwF6GtX5CeFE/vGlpGAfFZWF4NAe0lRl56UmRMfHI+41d3GZqOmZ+ZQr8hLTE9l0PUgOlxfgybU1yUnRzlHxwRm8NARxNd/A9AAZLLzE6NJdctOTufzfu1fF8JUGbl8YrzYgP9kcDwmIxiIZ9Ydg6fV5ydHkcWhyZlFXCpE5wrELKzYiOI4sDYHHQd4JYcPGhGXC6rID0+u5AqAKBsHC6fx8iMz8pjsNhsNqswM5E4ykgBgcFhYZFxcWn59McGAOUhLjmoThEbFICOnKDI+Cy2kPjihPhkKcpICCWPp/DU3CL62lVtoMzPzkmMDiZWMCg5n0lsBvolEhtd0nlFaeRmCQiNii/gSrU5oKobj5WZnZ2SjWph5W0WCMCg0kEABpWsnADM5gkEhcEv1Ke3wwhDVx50jC/k8jhcgSjnnc6Ebg3IYqTW5CvuaUwB+vzhCETCvAiL7YNk5PuNeRImhq/4/zIejme6aasqydSpRe8cwqLDTwKyUQTmCURod4uI6MNiEZ+RxOcf+h8r0e7wbJlOY5bZJ+JiHqpZSD4XiVZBNC6rmEOMKxYIBCUvkPlJgivEWUkOF9bI1Ktbh35TpP/MHZYRhRzya2gWmyv68nFLpmJWMZeP/haRf7PIKg35IjvZ8eY2pRlqhz9liNBnOjreyGIELQ6aE/2efxebzeNzcv3ub1SWJddtxLpTbqlcoaAqVOCJ+MtKC3hycF73vv379+s3apLeo+BidCqLROy4z2c2zu3ef0D/3sMW7TL3yUXVSjaPXRDy+MiUMf379e/fbe7+p+GpKChz0GyIvS9Mcrqut3zIHV+pWAyANDa6HuB4UZzdlZU9z76Jy+eJ+YVR1ucGoGMPGaA4oHeXNi1bDFt5O5wrFnx1IQHgK0S/leJk77v7lDv3RteuARNmn30VU8zjotIMVzP1ecP69B8woNeIFYdsgvL51ai+wRYIuXHWl+dPGti3/4A+ndces4wgPsIl68dB2yU33v7yjk69+vfv31tp/DxjhwI22eDL4RMftAXBD5dtOLTjUaQYF5ZzikEABpUOAjCoZGUGYDYfF2Z7nFjZH6vVsFPvvgrtm9fHlNQv+eajCQpc9o7s2kS268ChI0YoDRk0aKOZXwaTLyC6UjKinu4YhPJBix7DH4VAAP7biAD8ae/Mvli9ZnLde/Xt26dPn75dW2NYrUFbH/unM7JjfP19ozMLUbWCU8zMSw338vOLyxcVJ9poTcVkhy6wSURHQ3FGYqQ3LSo1i8XjcTjM3ISQAO/QtPT0kKAAVO4bk8PicL7kHzIAvzu1GMPqNuzQg3zbvj07NmtYp+U49Qe+TL4I5aqCmIggaq7e3sn5DC6Xy8yM9/T3i84sIpaHzSrMSPTyDQzKyGVkxge5eQWEZxZxeCJOQUoM8ZaE6JRcNp/PrYQMjFaQF/PmgIqcTOO23fv27dqmUT35uSdsE9BZwqVH+VehmhKvMML64KQeI4zd81EUjrU+d3DMtOshApwXcm+/2rLVxsE8Hi/LRVNFffNJx0Icz7HaPryX2kmrBBaf43tnSvdthu7JLJRUCjIiPtgYbxrXtHvH3sa+vOrZ7RD8T0Rzb06sz2eLo6u6dmvYQOd5ciFHyGWzmIwCpJDJ4THDX+lPHT96vX2S6J8/e0Cl4gjExckfr6l3HKH7PoPHY4Y90TmgPP1pMs4Lv71+zLT5B92zURhOcdg4ZZuGoWuuUMSvFpckNl/MiXiwcMKu3SZBRTxe9sfj4+cceeKdzsep9ePwRazUz7dWtxq0901sMY+X4WOmpzZ5j1WGGJ1QBWnxoe/uaszrJ9N23v6XcRCAwT8MAjCoZGUGYK5QVBSD6jCD5+y4HIfjua/VBtRuMveEczYaP+LJ+L4tOy099tLJx9crIpsrFPC5HC5PxM/1N947pKGsXGtMrs+ox6EQgP82IgC77p/aFRu15WYUC8dFQhGeaLdsQsu62EB9S8+nmj1bNR53+mM2Fxcz4l/t6VtXVlHPncdPfXtoBiY/Ysn7DGFeqIXahK5kIycyeMV5lxSeGC9yOTi8c51uy1cubdGQKK89Xts6FoWpkjtByQD8/ugcTG7IHCJFC4U4zoowOjC2Sd3Gs/ZaJYvYqe/ODO1Un5wpsvDwuySmsOCzQe/mspPPu+eh5eZkfLw0E2vZdfG7qAybo1OwtqNW2mWiUrdby0Y0pifrPVPbJryQK/jrN6ByhLgw/rWOcv2GE3a9ycLxVIvtw+s3nHfUNkWI/+NNWGwOj8/I8Hx6cK7+2xQ2j2hNT3e/eH7+uGepwoDLq7eP3PguB53TqDjg5rydB7VccoWhxuu0rpoFFRAXB85n3U4D9Z6E5osEhX6Gq3pPnzhq3zbdboomvhweBGBQGpvomZn+5sAU5aEDdQ8c6NT94rP4XCIAowORy+UJ0DUp1dlwwWTlDR/ScSGPngqAsrC5IlFhxOvTi9oteZJOXI7Y/tf0Ngy9GIRnO57T191nES3GxeizRuiuM+bQofN+ebioOhxS6OOG7X5MedVWHYcUtNIiIR5+f+yiU3fd0nhiousVhyPk5EVZ3Z654LE/k+yXw4x+cEVHcZ89B8fzAu7tndtj0qb1Kyds2KPxKAoCMPiHQQAGlazMAEwg6s4isYiVHup2Q2Oe8krtJ4FEpTjfWXtMz7p0JsF6b7v2MYmFghYvw11/UYPe4/bdOL4Ba9NzBLQA/33SATiSiSoGxOOk8BjzSQqtsRWG3iGvdyi0aj7GMLCYz0iw3tlLps0QbTcmzk22O4gC8MjlLplsn7sH54wYNU55msqsCQPkG2MjtxqHsHDe50Mju9XBasv1HDJijFKP5li9+qOvemSxePSDqr4E4MFzbBLQ+6KPaBSBg04t6Y8pLL4ampv6SmflpJFjJkybMXPa8C6N62ArjP1yiou9z/dr3nbiGT+WmJvteXEa1qyjinW2INnysDLWZsRqRy4eob9qGNa0/dBJs+fMnjl10vil+g8DsoSlOmD/eXwxnudzdk1PbMi6x5HEORP+YMFATHHDlZB8XMynR/qXEacy0T+O2C2M4Kcaq4epuWZl2R/du3bm7WAU4tlctIYfdozep3bOKx9dEgQCQXFa0HvrF6c2DJ9y3zaRKRAVFSY5f4rGORk+JnpdB9z2Jm6BhtMbfI1NyHD5FJKbX8j+aNir28lHsVQAJl7kiwX53mYH1g1b8i4dZRU4fEDF2FwhJzfSzkh14a4LppaWFvePaRls1/fMEgp5XOKKhj4H0GeeKOza/LG79pvFsYkW0KqPJ8bT3iyZqr3JMLiYuCmJxy900B2w7Oyj0Gyh5Kxhc7h84g4lNtoA4mQHgzXT5lzx54pFaeERIQFhOB53Z4Xa5i1mEIDBvwwCMKhkZQZg8r5P6vX0Dxd2TBy9fKfho8DoZA6OB90e06kOJtt75KRp08YNaFOntuKx9ymMzKCjy5spqt4OzU98sBWT7TX6FQpCInbl3LFZU0kC8MjN18OL0GcnqkHwhXnv9/dv32DIkQ8JsVY7+8m2WmOTUJDseKZXgzYjdJ1RTGYn2RIBWGmRLdplJF5skOura9vHdsPaqF74lIULPQ+P6FK7sfLlz/+xd9cBUaSPG8D3usO40Cv17Fbs7u5OUOzAADtQVBC7G1sxUDFAERUTBBSU7u6O3Z3Ynd35vTO7rHCHcd/feUM8nz/unHdnl53dd959n3ln3kknq7ta1K/0zScrr8XLC3q2hQJw/+tR4mF4IWpFOczo+NGnXVfciNa/bqan261Tqwb+/t2HRuvvJCg5pfv2Jt/93Hmvf0aK574eH1Q0Gn4zmWdjHVcJI8AT7ir5+P0Tmn70c7tlJ+49engvMF44AZ9nxSuG/0tkT9DGO6/q//mHRqangkivgQs4NryJ7Lsh6x8kaEv6ELCOkEtolYbX5Dw7MmvsyKFHwsnXc32V7exhznFCPRHGuL339rZav8kzjefVpNZke+2fMLhD+5bL9pzwihPPmhcu/+fZpKcHViEAwxtQJJyoVVSO644iAZjsRmzCw10WPTpues7wKtQeeBvh2hh5VozX+c2WYzp06NB58NBlF+76JimEyy8EKi2vSfc+ML3VWMvDPhll5ZgKCcCxV0cu2LjkaoRwMJnsTtoXJ4yM954NSHsVgHXHmhhOyzMxNxaPHt1l8uMU0nALH5kwt4Y26OCYSQjAUMIhAIPEigvANMPI04I8bp658dg3SwjCKbcmNP+gSv8V7tnqWJeN5hMX2j/LIMX0g7W/V/m6wepLt7eO/eqbit2Wn7p8ZueMLrKvq9Q0t3sQlKrARDn/peICsJpPdZlQ72dZ5/XuyflJzmY1f/xh2gX/pweGfv1z8zUPM8lvoDLWeZkQgEfcSdJmB3s5Xbl2eu3KWT07NPr5W9nv43Y8Tue5pytb/v5hg0XXovJJOgo/a/zLz9+vvBr7hgCsYDQ8F2Fn2lpWsc/aR8lsotfdy06Ox6abjO5hVOPrzz5st/FuHM1rsp7sbFTht36bXF129Pm8ausNT4U8HnVlVTcSgMfeTOPV8ZcPrhzYsaZ4EnTVLhN3PQlPUVD/dfKiSABOub922A+ylqbnwoRdJuT08Kay38Zs9EolAbgU1HD94azcl3bke+0y5WwoQ5bCrq60nDL4YoQwiE1reMrdqsPyBTYemcIGatUMqTnCUx4sa9Z15cWXGYyaFXpWdIIHAjC8BUWzjCK7aACmOZ6Nf3DIoluHrT4KLSoPvBXNaph0z5Oz2zcZ4hwvtEaZ93fObNRr5qkYWsMK7ZM24+mhGa0aD1x2zk/4uSsjVYoE4DjH4eOs5p4OVfMahgRgxmN3o9HbTvmnvgrACqV4QYsq9uaSUc0bD1v2NJWsqntI2NU0gQjAUPIhAIPEigvAJL5QEefnt5HJao06LYwL5jhN61BZ1nbOuaiM5Ef3rl3yiaeE5yofWPzxy6ffzN22qqN4QnRhX/1m7prz2tYX3gNdAO7+h6y92YkU4fsRyF1W1a74wa8W56PlanWC08wGP3zd33hi1y+qtlntnsVyBQG4StsJj2OTzkytK/ui83pf8rycQ+OayioM2WYIwPUWXAnP5Xk25OS4qj8VE4At+8t+bjH0gThMK4i7Ma15xa86zryanP14acsvZPVWuOSRYvdpjX76uJmVa6yC55mMp9tbVPi60YB+HT77tb2VZy75SVcWBOAbYp9HEHn9yPK5k/vVqCCrPmr741SV9j8OnYyWp4KPzGop+334Hj/hDflu71VF1mXBqSi65M+CRdBqNUcludvM6T2431z7UJbXCFdjJrpYGk/usNKbIT0nFc/HXxkxbNHkkxG50c4OPt7huSTxMiyrTr+54Jehm51istUqBgEY3kkxAZhWaekEt31LenbZ+kKpZnH/I3gbilFzmaEXrIf/ufhBHi+0Pjwfe2jB8tqdTgk/DVTg/V1TOrccv/q8P/kpUZeds83Iz032zSktZs6wepYjno7Dp1wf2X72tusRecJ9EcR1hPu3M/Igh+Vj+7eduJykX77QpTgIwFBKIACDxIoLwEITyiRct+j1jeyHplM22u1f0fXPz2W1Jh4MVuU+nFrtQ9nvfcw2Hzx6dOmozz+SVZ9j/8Rl/4IZk0wmTTY1GTe4TU3ZZ99W7jHD+lpIPi4D/i/pAnCf2rIancav32tnd+jAoSO75rb9SvZ5bWvnCKWGVyviry+o/aFwfOL7Nuse55CAo+UVYgCu2mHyfU/3ZX0+l1VvP3vz8QO2sxv+8qWs6sT9nlk85/EuAXjdUNk3vzU222h35OCBI3ZH1k9q+6Psm16rr+UyYbad/5R9037y2t3H9i0d9NXnn8jab30UL1ykTGV7n+jys/B+KnWw8cgT7oSkD8BtJ7imsHFuDvu27dxvd+LspSt2lp3//OTbXmuc40nq/G8Pqwin2mV57hr+u+zHZiYb7ezWGzep8MGfpgee55aCYQeKIck15cXJGb27TtkQIg7r6rDxly1mdR1h6yXeZyrn2MyRc5bZxfOM+zSjnmO2u5Ovmki6Nqdtu3n2z9IpjADDu/p7AKZYrTrFZ59Vp3rWj2gO+RfeTrhoQ5FwZ8+ERt2X+AvnrPB89A3LrkP7W/oquUS3fdN7dJh+8qlYLtCwlHDLvtKPYjXauKtmvfovOOApttd5l6e3mHHQKSyf9NbIIqcSmvS88LvLRg8etuSWcCqejprWbT8CMJQSCMAgsWIDsHBXVirT/cL6pr/o57uqUHfcgXsRCo0mO8Ru6XDDXYCrdR13JTBP9+skUoXtGyP74Id656PJS9O4Bvi/RAJwyv1ZXQ2zOOtUajbAxjUqW6nmVGpaHnNraa0PPvz0m+bW7pksy6o0vCL66ry2ss/qDbodnnB1jf5Gt7IKDXp2q/XN9/XnnAxh1B4Wtb+W/TztfKgQgIMODyQRduHFaDmt0Qdg8iIxzmsH6p5p8F2dyXtdwhRaPuXO2uE/625N/GWt7r1bfyqrOvqQZxqt1WrYvMjrxt/KZJ82tvHMUQmXNCkizpsbyT6pP+RWCh1yybJnRfF5oh+aW573y2E0wq2D/1MUq9bkRrrsHldDPyP1J9X673CNzteq9QfkSyzSiVRRaQ/3tCVfa98Z1laWK5cvX7Zs9dYj92LUvDz4/MEZPceYWFpaLjTu3mvjDpckFcdxCXfNzYeNmD7P0nL1nD4j+xo7PEvXzXdGkwAc92iH+be/7PJAAIbXEgNwlvPGH75beTQ8gzQTNNmzc0OvrZxYoefZcF6FAAzvglKpqYxAx80jOk4wXr3GctnUoV17j7B8nMun31zT/qMffmgybY3tulXLBNscHkVnizP9lX4K0tBSkfvnLuk1fNoyS0uLoaN7r3XwTlVyTPpzhy3nPKLyGXnoldkNZLJ6Y5atX7tKaNKXbdx34mWacMN20nnT8Fq/nf1HjB13JITnNQjAUFIhAIPEig/ASoV4GWeW+9VD5oJ19m5RCvE8G5Ky8oPttlqJxRZOoXJWuDBFTLoUo6JyYh+cMV9lvd0rQZykEf5DJOxkh1+12yJ+NQaXvNNU4tFhsT+RG3J66OffVmy22j1LuJMNzaoU6QHXD5qv2X4sRKFVRrqdXi886bDzy5Q418NWi47cCs+lY2/uWGdhe+VFSr5KRSd5nlprucLpZbpwNa7u7wovEnTnpPjnDHafu5/BCVWG0WpSPY5uXy4UXn6WnuS3d9PyTddepuezKpZV5sY9PLR8zQaHwGyaEt9O6gunfeaWO0/6Zah5Xh5x64Suqpmbb7zoFacQrzAW/+p/SaFQCm8m6vbJdeJbOXTTK0PFc/oj7iUYiSJUXqr35fnzFsyfbmpirDN1ie3VYIWK7OAZDy5tnzxx4kQTc7tbQWQHF+Yv0/LygL1W5iYTJxpPmuoQkMtqtLodXNi/MyIfO63f9CRaWeg+0ABFkT1ZkRfkZrPBxTs1V7honyTinHgfl2ObnYJyMDEEvCOFcHBVnRt1b+NkY9IcTV666VgkzWsVUbfsd5jNnTtritBIiZba3QrLVJWVM87En5uMRwc2zCKbbTLlhGeMXKPlqPykx0cWHbgbkp2XF3lv71zz+bOnTtJtvrHxfKvtXilq8QYJFPlhTXh88uz5817JKtXrPhIEYJAcAjBI7DUBmCC9e6ZQGymcYCN0g0k0LnQipXAW7KsUQEIW+cXSlZf4cFD2KEgmFT/9IoSJM8hXR9HCN5N1b2m9r79svvp+Js0I+ZV8S7RwcRWhohSU/p+ERqX7t5ZkIv0Iv5oWVyepVrdg+OLFFynm51Q4KU18VLwBrUgjDDoLVOKbEuby0D1RN5uy8Er62kXejkJRsKAjBDHhFaXwlzejYYT3WgqQN1n0Q9QhX6vwURa/gxd+QpEdXD/1inAUAuCNKGFH13Ik7eoqFcWohcpGdmvd4wDvoNAvgkg8z1ffySiizJwCLSq6kRrx15K05MJvrzgRVnGfgPjR6J4sDF8QwiHa130kCMAgOQRgkNjrAzCUGZRwQ4n8kBtzTfs0/ErWaOwp/0yWKROniwEAAMA/gQAMkkMABokhAJcHFE0p8yLcNq6zXLd5/+1IhYqhcR0eAABA+YMADJJDAAaJIQCXFwUnGwtnRumLAAAAoHxBAAbJIQCDxBCAAQAAAMoJBGCQHAIwSAwBGAAAAKCcQAAGySEAg8QQgAEAAADKCQRgkBwCMEgMARgAAACgnEAABskhAIPEEIABAAAAygkEYJAcAjBIDAEYwICiKA0AAEDZpe//AUgHARgkhgAMYEACsH7HAAAAAID3AAEYJIYADGCAAAwAAADwXiEAg8QQgAEMEIABAAAA3isEYJAYAjCAAQIwAAAAwHuFAAwSQwAGMEAABgAAAHivEIBBYgjAAAYIwAAAAADvFQIwSAwBGMAAARgAAADgvUIABokhAAMYIAADAAAAvFcIwCAxBGAAg7cH4MQQTzffqBxWv1hacOle1y4T1+55xKr1ZTxPJQTeuUJKr9wOiM/TlxFcvKebsPZVz5RX6/J8po8TKXR0eRjB6EsA/hlNbkaETxqOMsE/xqQnh/rEGJopTVyo11Wx8XILSaX1hWWINuXFE6G9vXItMLP4zVOnim36FZdHz3P1RYQiIdBV/FjuBMTL9WUAJRECMEgMARjA4M0BODn00bb+jb/+qOuKu6n6otJAnR/7aOe0gQ07EG06dVq+434OR4pzw12PTunfsY1Q2r7vpNNuUeK2K+Pub1/UpVNbUtxwkOluvzhaS5oJKsH70IxhjduTdTu2X2h7JwsZGP4x5YtD000HtboQrV8GeEd01LUZ/Tu1Xe+lW8zysZ/Tr3UDoZVq1anf5LNu4fmkmSo75D4nbft0EdrmdrXHmNn7pQmtcCFaNilg52SxTW/TuXP/7ediFUJpToTrwSn9O4hteoe+k87fiyqDhwagrEAABokhAAMYvDYAs5lBd3f0afWLTPbBx7+NsLmfpi8v+TRU6BWzxjV63kgUlmjfk+u799wcyqijjk9r33bw+ghxpRjbCdOnrnfOIz3Li2Ma1pt12EcY4s68MfyX0TueZLCqjPvmzat1PSeuHH1lfacaSx4VGnQAeDMtkxp4a/9O8271v2/esMe1BH0xwLvQKNI91g2tVqVF723PhWWF98ZZLYbOtUsRHw3YtODo3huJGnGhTMgP2Nu/i4X1zWxhIfpg/+7Lz3lnCgct9Uj8dd88qcFgpyRhKfn21lk9JztTvCpi/4wOHQevjxFXirCdMMXU6k5m4ZN4AEoQBGCQGAIwgMFrA3Cc47Ix9esZmy3oY/Rbhf5rS1EA5vJD7u6dfypAfzpcvs/OfQNaX4xngy6fOrjpfoaulA/b1Gvt/JUvNXz05dUHrt1NFMcbtM+W/NraxilWQaW4H5h//KXYH+OzH24fW3X4NTFPA7wDrSLtyabR7adON7Gct7FRl0vx+nKAt9PkR3me3GY2e/byqd03PiMFGVfnjDLpv7PMtkAqj+Xtp85b4y0M6hKpjj0Hrj3ikf4qAbPxt3Z2q7rioe58cG2a8yGbTtOvZykD7E8f3PHgVZvece08m+dK/SJACYMADBJDAAYweG0ATnv+wOOen5aPXzuixmc9VpeiAFwU7W8/e2h7C+8s/bKOJmyvcb/ZSy4mFvSxtIkP7fZsnNq7zYQ7vvmGjheb4n3Geums4aOHronDqXXwzrQKKjskmlSZvDubltfrhAAM746J9j07e83JFz52VuO72PiS1sp71VzbUWuf+Nw/ZU1suuGTUIZGf4nsm6MHrZlzLEp/1rPq6eZGY7dfiHx1Ra8i7JLtH91OB+sHdxX+R23NWqzwKHKWtCZwr3HvWYuuGdp0gBIGARgkhgAMYPC2SbCUz5cMrv55z1IagNWpD21HDJgw62bRd5/kttWk9Xhr+7BX286GOiw3m2g8avOerffCC64/4+TRtzaMnzhl6sRjFy76lNZDACCheCfrFQjA8O7Y1FD72dssApTsy92rxokBmAvYsnxS67GLFtiuGk/0HjbY5OK9kDc33KVK+q0xS3esdhHPbxaEnm815YhDpGH+L60i7Nb2uoudUvUBWB3jcGRL2zUehQ4DxLltHdd6/NYzYThQCSUWAjBIDAEYwOCtAdin9AbgdA/bYa27j9nglqkvEKXe2WFiZDRui1NEcaMo/tZtWllcCNCfi6endN83vcWAg9G4uAz+IQRg+Ac0TGLQzcXH14lRMMrOZnKPXZHkX0G2Cwb/OmrrXd1pLNz9pc07TV99P7PoNFGlWNbNUSPXm53WXcpLNtB7V6OxOy9EGEaAtYrQy5v+6H8uXD+4SwUd32LRYoV7QQuecGfHOCMjky1OMWVrZBzKGARgkBgCMIBBWQ3Aiohbs0e1mThr96PC5z4zz66sG965o/l+l4K+Fp/34sC2B3eDCq4bUz5Y+uuoXS4RLx5v3OqbrO9OcbHX9k76c/adMjTmAv8NBGD4B+i4S9ZNZW2N58+ZOXPysHbNa1VtNWf/hQiPw8t3memneiJYj3F1l049FFhmDsixjxa0nmm2zlc/epvlPLDXsv0PUlS6RYKJubG1bRWrp/pmOvP24Y1dRh9PEo4AsO5X1g3t3HHxfhfMNAclHAIwSAwBGMCgTAZgOunp7lEdR1tdyNEXiFQxN7fP7NF9+c1gfYFI7reyW4eRWx/oTreLPzOmyYAN9yLTAi4Prd9lY4B492Mu4vTCbm3nXEzECDD8QwjA8A+oc4Jc19muWbmUmG/SvV393zpbnLwen5pwetlok2XndI2w6u7iZl2mWz4sOyPAfNbTDX16Wux8LLa3GWdHd7A4/bjgfGeBloq5u3ZYUxNdM537ZM/8/iMupPO8JtJlx9Qe3dcUbdMBSiYEYJAYAjCAwdsCsMJrbrfvZa0tXHU34CgN1DmPttX9QiZrNWia8dhRggkWm90i0h9t6yX7/JOqfY1Nx40Wi+fuuRqaTyJwwN4Ng3oPHkRK+jcda7r2aZzwkSiib8yb3KLvyFGjRgzt3cbC6lIiLi6Dfyz2yqoFVZrax+oXAd5Z6N5VI1vZBoj/znvqeHHUYJOhQrvVbdhQ08sPwsvW+SgZLlusevXuTzZvWOtR4w8+iZFzPJfleXjevnvhCo5EXWXE0+VDhnUiKwwe3Hvg2p1+2bwm/Z5Vb9mXn1TtX6hNvym06QAlEQIwSAwBGMDgbQFYnfn83qXTri9SSk/845QxXoePHDu6d+dWvZ0nHANS8qJ8XA4fPnJw1zZ96dYjN72SdFuf6nLBTijettsrXSwQ5b44uFtY78DJqwllZ7AF/kuKBH8fx1uxRa8pB3g7NjPY99Et/4J7/PBU4GMHoTnauuPWS91d28oWLsL14j5h+3Y9jNNlWI08zNXuln8yo78UhY64vos8vu3w+ZupwrI2L/Sp2KbvfNWmO/vo23SAEgcBGCSGAAxg8LYADAAAAAD/LwjAIDEEYAADBGAAAACA9woBGCSGAAxggAAMAAAA8F4hAIPEEIABDBCAAQAAAN4rBGCQGAIwgAECMAAAAMB7hQAMEkMABjBAAAYAAAB4rxCAQWIIwAAGCMAAAAAA7xUCMEgMARjAAAEYAAAA4L1CAAaJIQADGJAATPYIAACAskrf/wOQDgIwSIw0hfq+PwCIGRgAAKCsQgYGySEAg8QQgAEAAADKCQRgkBwCMEgMARgAAACgnEAABskhAIPEEIABAAAAygkEYJAcAjBIDAEYAAAAoJxAAAbJIQCDxBCAAQAAAMoJBGCQHAIwSAwBGAAAAKCcQAAGySEAg8QQgAEAAADKCQRgkBwCMEgMARjg7RQKitHo9xmy16hohUKhf6g0IG+WVr3q8WhZZaF3T7ZNRcp4NVN4o8QyjqUL1iy8/SpKVwbwLv5S/QTCHqR/FOAVilGLDSwlNjyvabiE5lhfRGhVVBmrTGRzCjW3Glb3YRRCFmn21eeiLro3CU/nCsqL/2QQgEFyCMAgMQRggLcg3S01r87xf3B5p+C4i18qTXpdQj+jNCB9SLWWo1J9T65dtXLlqvX7z3hmkQRS0JVktDyb6ubx7G5gCssyumhLq3lN8tMzXgGxWXKWpihWzanzAi/arFq5as3Wg/dTtMjA8M4ollVmRXufWWe5SmfrwXOB6SqW1j8OIBIaGg0d++CUd3hqnpK0RTSn5RRJz4+LDdeGg+ee5fA8ycA0x3OZ7jePr1m5cuXqXS4vEhhew/w1I5ZeCqWK53NfXLIj7e3K1VZOQcmUmoTgQttHPieVMtP3jJWwwpZD5yMpXmM4VCmkX21OiMu9e/eC0gxN+l8gAIPkEIBBYgjAAG/BqLTKJLfDI2p+KRNV7bjwrG9SvppEQ/0aJRnFsHRe3MPjM/sM6T5hksn4wX27WVgcj8/Xkp6ROIxCe55a8Gez8fMuhPI8RwmdL57PfLx1SrOPZh54mqTgOTWjTPd2WDRgeLexJhOGD+42e8a+8AwNXSq2HiRHqzhlss958y7thk6YMF4w33LXs2S1CgEYDIRUS0RfWjay8YfLL4fnUFq1msmNundkWu/BPY0nmYwb1LfbkmVnEhQcy1ORl49O6z2or8mkScP6dRt26PyLHBVXfNIrdRQUy7PJDmtWdO89cJzJpNHthg/f5hKQTXGsfvsomuw62b535vUZ1m2iicnwYX36LVv/LFmlFhtkhZLleWXw0em/1eo9+2I4LyTjYiAAg+QQgEFiCMAAbyScC5ztvrVfJdnHv/ebu2qFactfP5a1nX86VBx2KOkoWqVRJj/YN/qP+uMf5Qi7vPr5yVWD+1g8yOE5Zfx9hx3Tx3ZvWee7JmPXOkWT3pJaEXN7z4aZvYc0avzZ1xb7fVNJRmazgi5Orl5zqGOs8PzYG1Z9qk64mqBRMRQiMLwNxWoViT7O+3vuCBGqjx5OgYZCaJrJCXZcv2JK+961m3317QbHmDzS7FIJrjuGV2s86alcqDKs55FlQwYu92L49MvzJ/Ux2eihFopjNwwZPt/aMYnl1WXhiArFcNrUOyt69pi/zU0pbF/aqTGt5x65HaXkOXH7KFbNZgZeX9a/6aS76eIKD3bM7T/waIRWK4yCk/ib7Xdr37D6X9Qft+J6lFb/rL9CAAbJIQCDxBCAAd6IYlVsWpDLMas1J1wSyC6TsLRP5S87WDjFsa/pW5QoFM0weWmBD8/vdInKZ8VLw5JcV23u1vp4OM+nuh+3mTls9cndy3pPX2DhECY0CBlO28znTZ595bhtoxrWu58m0DxHZcU8vbDtRlimODgcc92mz6/G1xCA4Z3Qai471ufMghbTl5hOnjR58iIbuxDSs2cRgEGPIq2UKs5+7VTzWXMdDtvWqLHtUng2q1Hnpfg9OL/LNVqhIrGQtLzOi2x7dndMo+5YDJ3e08qH5bUsaZLCj3cducLiehSrLQsJmNHyubdntZg2y1K4UEVNNjDRcWDXhTtvRsk1aobsToyGzQq+aN2iypoHmTSt1vDqBKe9Vp2H7Q9lNSqVhs/2sj+0adea2b1nLDS/EIoADCUWAjBIDAEY4M0UCgUtRj9eHXdx26yu7QZNPeYeq+Y4pvBlWSUWCRo06ReRd08JwwPq2Js7JvTsvymA9Kcy4lLSsiiSaY8PmzhtwcVwLa9hssOiM8lqfMiJFtVWbvdIpHgVrZsAi80Mub5+1IDeHfqY7s2keFzCCe+A7Dp06guHxb/W6DF3zurVq+ZNGdpxzoJDYbnCILB+HSj3KCozJFo4Q4Xx2FGrms3FsGyao0m+E2KaSik0XKqoa5vH9h66P14Ze2Hu0unjHKOEJo0E45y785ssW3LgRTavYfWvVoqxWj7ecfgIy9nHQlSkOaYZlnq8reGoLaf8UzlhNiya0SrTX57b8Hv/k/75SoZRaflMn8M2C5ovuZsnNNJhBw5uNLd397M3njxt3oUwBGAosRCAQWIIwABvplBQrG5KTir+zsVDK6wmz5hvcfp2WJZaXYouOxNPjuMyfOymte08bb1nnlZDK2mWZdWkG/ni4JCJ0xcKAZijKZVKxapV6meHjaqt2qELwLqBZHlm1KMT1hs3rJm7d4fNtWClqoxcdAfvE0WzVE6Cr9vhk4+F8yd4XuG9e3qfrlPORLNCfx5Ah1ZxarVKnn53a81qG8UArGtfxIZLneZ10LRNl1m2PkqeDzo728Z8kluyEI1pFc+6b+1otfXwy8yyEoBjr46ysF16LVKYkoFiWP7lqRYm+84GpBUEYHm639VNtRdcihEmCmM0fG7w2X3WrZY/UPGqWHv7VTNOX87g4o4OnThtPgIwlGAIwCAxBGCAN6FYTp367KzN+DErDnmI11xF7Oz1q6zKmD3Ps4UbcOhXK9mEiVV4TY7viWmtGg1ZejYoT3e9mIDmePXLQgGYFJHMwhYJwOQzIJ0xFaO78Ubukz2z2na0eS7nSscsYCAp4fiRrrOtphQKhufz/Q5uXmS01E35mvl5oJwi7Q6TXzQAiw0Xl+l9xLRFo6ErzocKFwNrg87MWThl9LkwcQRYw6ffmtJo8YKDL3LKSgBOvDqi74rpBwJpskE0q8p3W99glO3pVwGYSvc7s/63jnY+JACzai2f7nXAakbLVb7ZSQ4OmxdcdabJhxR1cuSUOUtuxJOPq9ir7RGAQXIIwCAxBGCANxF67wqfw1Ory2RNZx4kCVhx17JzRVndSdt8M0pJACaboKKi3Xb27Nh84eabQiey0KjAWwJwEs1z8qTH541Nr73IUTAqkoRzA+03j685+0YmJ9ytE+ANKIajMoKdTpvMcYlgGVY4kpLuvn3nvI4H/LnScvgI/ht/C8AMabiU4be3dO9gZLHdNUxBGi7hLsHJLosGz+y21IMsq8niy71dplmudktUl4n6RBK94sGS9iazVjxM5XmNSs1H2ncdYHnwfjytVTPC1dKcKjv82pZuv1jcShavTaFjLu9e03GWc2qMw5xR1b5s1K1vr+7dWtf9ocovvzWac/hmjJzj/j5ZIwIwSA4BGCSGAAzwZqSTwYZenNv+qw++/rlVj97t61X+RNZg6jb3TE1puBsuRavUytRnp+a3bTLrwKNc3V6v4dQFl/CKAfjA4AnTFlwoGoAPNf9DuAaY5rVshr/jlDbNF92OJ50mbcqjTZNbtl3pnMkg/8LbkACsyY9wsu31Wev1XsJFisrQCxuN+5gfi6VxBjQUoQ/AW/78w+ZCWDaj1WiUiR5H57ZpOveIR77YcGk0HKfhuVy3DaYjjC2upZEiytusu8m8rfcyuTJyb3Jazef77u7Xcf7Ki/Fk+1QB23t2X3LycQLDcyqVRs0yrEqdF/XAdnLdPheEedXpsCsbpnQ1vZqhUUS9cDt74fSJ46fPnrKZ3Lx9l95mZ3wjs5XF3QoYARgkhwAMEkMABngLmlEzebH3tw1p8ol4H+A/+8++FpTOsEwpCIDCpKG5MVdW/fSpTFanZbcObVu1JDqNWXjan9YIV/cKAfjFvv5jTM3OF1wwJgZg7wONqizb6p5A8WoVq86Le2g9s2bTlm3aGLVt09rKzjtDqcElwPAOKIbsJ3Ev7JZOrNa0TZs2Tdt1HLjbIZ7iMPwLRekCsOum36psOB+ao+bVeeEXllT6RPZB3dbdO7QV2q2WXcYvOR9MHsnyurFgSM8GQn1qM2j2xceJwr1xy0iFomiOkz8/YdO7k5HQ3tYYPe/M80Syvyjibq4fttrhWZpcq+GoGN/dpv3rtG7Tplm7Dn23nUvIY1lGrZuqQhB5fIjxDPPrQoTGKdBQMiEAg8QQgAHeRqFk1FomJzbo6QOBf3QqpdawVGm4kQtF0cr87Lgg17tubrecrl8jrl+/7uzmEZpG6QI8RVNZcX6BwWFJOXRBH5KiqMy4Z08jYjLylDRZYEjaz4m4fZM8/7rLfc8UmhfuyAHwDoRZyNUsnR588wapfTfuuvtmsLyqVMygDv8tilLI06I9n0Yl5Sho0nBlxga63nG7d8tJaLfEhuu+Z3iakmY16vzIQA+hPl13C0rIYoVjeWWmQpGfG45Xxvm63xLaW2f/5BxapWKU8uwYnzvPIpJzFRTNsCwlj/dyFj6AO4+fpzFFhr8LmvTQQk36XyAAg+QQgEFiCMAA70AhzIZV0GfghO5WaelvkfcpTmJVlHDPTP0WCAFFq9FwRRO9cOcnTpjmWl9W+DWEkWOAd0b2Ht2tuERaUtP0DwAURQlX+YoXaJA6QgtzDhSlFdte0mZxBfVJQ1qpUtMavyOxTS7YPmG6fWEDhZsRiAu6Rl34oPTYv9xRzNCkFzv6K0AABskhAIPEEIABAAAAygkEYJAcAjBIDAEYAAAAoJxAAAbJIQCDxBCAAQAAAMoJBGCQHAIwSAwBGAAAAKCcQAAGySEAg8QQgAEAAADKCQRgkBwCMEgMARgAAACgnEAABskhAIPEEIABClMAAACUXQjAIDkEYJAYAjBAYTQAAEDZhQAMkkMABokhAAMYUBSl3zEAAAAA4D1AAAaJIQADGCAAAwAAALxXCMAgMQRgAAMEYAAAAID3CgEYJIYADGCAAAwAAADwXiEAg8QQgAEMEIABAAAA3isEYJAYAjCAAQIwAAAAwHuFAAwSQwAGMEAABgAAAHivEIBBYgjAAAYIwAAAAADvFQIwSAwBGMAAARgAAADgvUIABokhAAMYvD0AU8GXdy8w23bRO5PTl5QeqpTghxuvhrP6RZ5Pvnd88YSxxPqzDzP1ZTzPhe1bM28cKR1vcik0X1/I8/Gus0xJ6fi5yy5H6YsA/k7uu8F8ulB9psx2SdKX8Tyb4rbVRKhV4xbYHojQ6ksB/kId4Wj9MDRLpV/ktRl+Z42FimO6asfDHH0hr4m8tl9spMYtOusery8sS/I99ltNE7bP+MTz5OJ+a9jwK5OExnucmeWuEMOnled2fpfxeFIsmL/vaphc/wBASYMADBJDAAYweFsApvyPzav3qUxmNPVwhKHPUUrk+ezu2LxNdTM3pbhIR121mdqxy7CFtoSN+ZjTN5ySSbEiePesOT0mWliR0uUjJ/Tbeiea0vKqsGsz+jfpOGmDre0m60Xr9s6/EqsRXwWgiDQf6wlTek1faU0q1cLhxsMOe2aQmqJJf3xmcLseo9aQWrVqwYhhcxdejtA/AaCQkCPGXavJVt5I1DWvXG6wi3Wf8WOFirNo0qClU7eHkQfogDMrFvYdaCY0UitndO+768jDdHH1skIVuW/O8gHjzcn2rZ8wobeVo296kZ8bVZaP4+R+44cvIytYLjIes2DerTihnIo+vWNl54lCseDgDY9EnNIEJRUCMEgMARjA4M0BmAo6NbXelzKi36JT0aUnAKsy3I8NGtS/Wcd6nbq2XPlE3MTcZ3tn9m8185wQekk/88WuGQvW7glklO6Tfu89bvNLtVicf2WW0bwrL1MzfI+2/Wbc2iCFWJrtvn+q6ZQy1uWEf0XSxd7f9J13IVFc4GJOTG622C1FkRXkOK9Vp9WPdYdeVNGB9x2fp4n/BtBT+h+fbty7zqBWPb+rvO1miti8UuFXd3dqPvuGrlUOvbhyytDFIQzzeNXITl2X3dSdyqI4OXOI+ZYLia9ObCn1Epzm9R+y+GSAuMDcnt9yzjHXeEZcEjGxN2zqfTbhbJa4pAo8vtq4z54Q8s9Ep+3HdtmUxQFxKIMQgEFiCMAABm8KwOk3Ngyr1r31gHYt/5R1nlOKAjCbn+6yZ/Zhh1PnD22Z3GzpI3ET464vt5ozzLXgDGfGy7S/pcmRCCr54d2I0DT9ttHeG6qNtbsbGX1v92+jriboCnk24tIRkwZrvcpQjxP+JXnRLrej4/J0p2yqku6u+NXkgl+M+ymzFkNOI/LCG2Rc3WdzZO8hL8ddNWvuvJ4g5D11iuv2qX+MctYFPV4dumvVolbm7jlJL1/6eafoDtLxGq9ZzVcu3B5Ueg5Ivg3lNr2V2ZyNAfoNzHcd32HuLhfxE9FR58WG3LiXrDuexLNRJ3bMa7zkEfkowu+dWNG254ie3bp16z5+0/kQNNFQgiEAg8QQgAEMXh+A0y+t7fp1w55nb3lYjmooaz3zdIy+e1LycWpOnit0hWIct4xrukQXgOmoCysHDZxsGyJejZnluqjTF10n7Q8rslGxNwY1nL7paSpFx1+Z0bS7zf0MoZSLOmP2R32jZU9ocSWA4ml8z3ZvMP9ETHKi28npTXff9T5r1oHoPnHppVj9KgAF2Ixc4bIK7fO9tatvuRYvNFhU/I1Dxk32PSu4BDbp+opN4/tdSdUvipIuTx4xYeVJP2XZuaw88+aoSRvNL4onNRPaF4eajN9zIeLVdAxF5HodsWrbYZNnHs/n3d0ybmDdbpvtzp07d3Td4B4zLexe6lMyQImDAAwSQwAGMHhNANYkX1nf5btGvQ8Ek1S4elBNWZs55xJKXYdLGXRhkyEA81p5mLP1yI5/GrVu3XrIkjVLZrZatn5v4KvuUvztKd3r9V7xJF4YWdHkRp6y6FWjaavWrfuO6bfecum8FssfIQDD6wWeH9auweht4Qpek+e+q0+Flt1Wnrx5jbBbPWFKj/nnMZEaFEMjf7KrVvWt+gCc4HRsSrNTulOBidS71jsXDrmmu2xDkOqyYnCXrosv6S/OKCPSb41ZtG2ls+46AiLEvqXp4YuRxQVgKvCyeVOj3hucxAfVWZEvgwKjdEcxVS+ODG4+Z7F9NGZrgJIJARgkhgAMYFB8AM71P2Ba5wPZZz/WadGyRb0fv/1M9lXlav2trgS95ph8CVU0AAtyk0Lu371z586TEH+n2XN3Lr8UqR9rib9t2qNu90WHg+SvTixUhj92I+s+9Hrw2P6seevDwehYwWsEXRjatuYQ6ytxQoxh0+4dGl192elofUpJdbYyHtB7d5huCaAQrnAAZhKcd0+ov+JeQbyNOW6+qvPQG+KJKDyf5rpiaJeOE/Y8Syljo5y5LmN7rJi5P1zfwDKPLBuP33Yp6m8hnwpxXNKsWYtVlwsdtywseEOvtas2++JAJZRMCMAgMQRgAIPiA3Cmh83E6rIPhNmvXvl94j6vbP0KpcNfRoA13KsxbK3Xom7mU9e/FPpKar+Dxl2GTNx2PfbVRWdajeZV3E2/tmVSGzNDrxSgsPzHm4e1HTHnxIN0fZXRZPtcsa6/3K2gwuS7b1s5rt16X/0iwCuGACw0PtrspwdX1mh9VH+sJMtjw4YZfU5GknrFRByY37v9iHnOUTll7zicxndTr4mzVz3QX/ocdarDmK1nXhTdUG3mM7sZ7drN2f/01bU4GU93blw2z75gfvXse1Nm2Fo5YgQYSigEYJAYAjCAQfEBmKOzU2PCdV5cMW5fRdZgpO3jxFy6dHUtlIHnbEY1sHio28SMp9s3Lp5x7IVcLs94smrIoK4rz0WpeE3STZO2ffsuvxCcRbMMpSCPsmotk/n4WO9eR++nkcXQG1Zjjdrte4FxBfg7LuR8/8Z9xm5zjc1XsbRSLlcoVJxGHnl7nXGntXfiSf2RR1xZOaNpPyuPQrPaAuhx8sc7qv26yTFOrB6a3KDrq4e3W30nlVScEPuVS7oPv5il0aa7rOjfr8PIYwGJCpWKJo2UklFpytCtpVUJl8Z0XWR+NIBsddb99Z2GWzsG5ml5rZoRNpSsoIi6alW/9khbl5g8FSfuZhSt1qoSLizt17eX2eMU8rw0N/Mh86ZufppfauaqgPIGARgkhgAMYPD6SbAKaENXj2zydXeLyyn6gtJDGXJ5u2m7Ve66TVQrEr13zRz6Z/369etW7zl9k7t4p8kwu/EtK1WqUrN+40YNyCP16zdsvMstRUNWdtk8tk5DYeUWbRddDBNOUAT4C9rXtlft7yr/Vrdh44Zi/WnaqsMJH4rXsvJgh3VD6whFjToP23QzRVEwsRFAIZzi6f4mDXY5F8x5rGVTn+ye/UddoeJ0Gbz7Tpqa10afWtnnm0qVf63fpInQIhEDrM6FlK7rUd5Cnf7w4Kgejcmm1as2beu9eIrsLlyq69q+KxwDabUq1N7kl6++rlqnSVN9M91tnNmDDJ5nc0OvrprQWvi06rYZu/R2WB6Gf6HEQgAGiSEAAxi8QwBW5edkpGbl0aWvA69V04rcTDn7aqiEkWfExhBxGXL9gC6bl5GRmpQYLxbrZCv1k6pkiqWxSWk4+RmKp6Vz0tJTkhLjCupPbFx8XsGJEoo0sSguJettOxmUX1qVMiNDyRS6PoNX58bpKk6mruKo8nMzk5NILRPrkyAxM58ta1GPzk6LF7YtLkffYms5RWZCBmm/tVqVIjMxJaVQMx2fnKbU/SJxVE6iWJSUoy8BKJkQgEFiCMAABm8PwAAAAADw/4AADBJDAAYwQAAGAAAAeK8QgEFiCMAABgjAAAAAAO8VAjBIDAEYwAABGAAAAOC9QgAGiSEAAxggAAMAAAC8VwjAIDEEYAADBGAAAACA9woBGCSGAAxggAAMAAAA8F4hAIPEEIABDBCAAQAAAN4rBGCQGAIwgAEJwGoAAICyS9//A5AOAjBIDAEYoDAFAABA2UU6fvouIIBEEIBBYgjAAAAAAOUEAjBIDgEYJIYADAAAAFBOIACD5BCAQWIIwAAAAADlBAIwSA4BGCSGAAwAAABQTiAAg+QQgEFiCMAAAAAA5QQCMEgOARgkhgAMAAAAUE4gAIPkEIBBYm8LwFQB/bJIX1S0sMBrH4D/2Pv8IsrnlyxstUC/WEr97e2Tgldl+k0sRP9AWdl8+I8UqSj6qvOKvhzgjfTVpRD9AzpluB7pN/cNu4r+8eLXeE2xHgIwSA4BGCT2xgBMs1xBK6nlWFpXSLGcvoznSWGRRpYizxAfYGmFvggkoVAyauG7E7+2f/u7oBjxPvpqRr9cPhTaGzTqovW+1KBolZr7yxdHsxpOTeqJeHtIiuE0+o0soFGJez750gseMbQFAK9BMSoNr2bJbkJaH5oVG4zChFqF3wh4M7EWFaXlGMpQcWhWq1GTWqYoc1WJbPmrXxummF8bsoKhI6b9e4NMq7QaDff6TwYBGCSHAAwSe20ApkiXJdPz2NoRXYiuY9ad9EnntSpGxbMBpxaM6yWUdjM/F5RFc/rmmXSdOUXqi+Mzuw2eOMs1imOLa7XhP0GxnOLlqQVju43dfTsih6Qb8lVQdIHCR4ap4otflf/lODJFs7Q83eNw36HDh57yySU9kL8eZy7mmYX+iOjVA/oCUqQvEVYu5i0JZeSBgseK/M3C6+uL/nUU6WukeR9ePVLcG8z3X4uleE3pyv8UxWp4KtHd+fT0g8/ySKeREj5qFc/n+O04dGrzzXCO4zSaxPtLpg1pI2rbtl2Hzl26dVl0/kUWw3PyqCuLu7Zt06b9gHH7nuVpVUW/BIACpI1Qs0ziox0m5x7F5zAcmxvhsqsdqTpirWrXvmPnzn2mLj4bTOkPrQD8HUWrVPkpnifm6OqNUHM6dOrcZdwCK/cUjoRgse3KfLZhz5n992M4TlWWWiOa47nIy0sm9xA3e/3Vl5kMaXD1DxIk96vkCQ92dWgnrDBg4sL7ibyG1T9IHlZrNdHOazZt23Mv9nWfDAIwSA4BGCRWfAAmfWVNhvep5e0rfSrT+fwHk82OSSo+3WPR4IYf6AtlPzWeYx+UqVCxtDByRGd62Y2vJJN9UqX+CX/d0X+QBK3mFfdXG1WQVZp67EWmhlfTNMOqC6hYRh8VSVdVpdKXqlWqV8WMofjVyiKKUTP5SdcWyj794rNVd7KVDAlC+ocE5Jms4ZkFr0e9KtPRlVNMofdU8FcKvwB5BZbWjz6SQpVI9wCpcLo/W2R98VUKv51/B6XSaFKf2S3uWOFjfb3/vtXiMy8ytJpSVMVJn0qdHXBhYfVfGze0uJerVZNPivQgFZHHl/f75qfes+1DeNKHYrJjn9xzvnTp0uXLjo5Xjy/p9oWsz9p7cXJ55Lnl3Zqbrb1ww/H45tnzO4+/GUE+bERg+BuKIftj6t31nZpUlpmfj81lNKwiI8qX1CniqrPzWVtTo7oVWmx6nMeV1vMo4L9A0bQiNyHwgVhxrtxwvnzQrF+rz+rMtk9gNCryC8fnBe8y6/rFL0NX3ojmea7sHEthOD7Lc+lw8+Fmuy45OdmvmtJ+zkHXyBwh9QsP0wxHZ0ff3NW1zcQFh685OZ3caT5pysijLxT6nyPho0lyXdH3i4rdFlyJfN0ngwAMkkMABokVG4Bp0jNOe7R+YCVZte7rrz17fnVDv1qyzzrNd0yLe2LapIKs3dwDt71e+J4yr/r1D732PU1Vchqthkl/dmzUD0JE+LFum7OBWgRg6QgB+NGGzn/Iqpmd9ssi37KK8jnTv1OrxkTbrktvxSlI8FVpuOyIu9vHNm0iFDceNG25R7KGY5Ssls/02TJ3uFjaZN5h1wSKJ8U6YgBOdl7+ZeWffljvVjQA0yqtNv253Yp+4jO7zN3imqDhtRra/9S84W3EF2vWrFnz5s2aWt8IzdbwqkyPox3bNifl7XoMcgik1ayK0yji7hyaL6xL1m46xmLLyyytRpPuZTOzX/OZG7bYTjLpKLyK5VnfNEqjJtmXyY9yWtyptbB+m65DzvmR5P63Men/J6FDkXLfpn8lWe0+VtefPbuypufvsp/GbnuSwfOlYgiLZjVUVviVDU1atm/Ytd3gAW1WPsgl34s87LzNxEZG3Tt1qNN/ysLzoVrSVVJQ5NvXY19s69do2pr7CbQmJ+DMpJq9rZ6SqsTzyfesjX/ueTZU+69/0lC6UcLhqIxHG4f0avrrqD7DK/1ieyEmiyb9dvrV2ZqJ1y0nt+mz86FCpSk7kQXeE/3FNoI8r72LW/SZfymRlGhzAuxWj2rYome39tX7z7a6ESW0XfqnlHa0WssFHR03cIy5faC44ex9S6NJexz8szQaIQHTjIbNCr6woe7PCx3Txd0q3//oupkd1rtTPOlykV8rNtzp1K7RRlV7zVp2JeJ1nwwCMEgOARgkVmwAJp3g/OAjU1t9UH/UwVCykvLpjC4/fdpz2YPchJsDGn8vG7T1cTYp9t1fp+q3n1hciJTzPB3vZFm/SfWuM6YOlv1Qu9WZAARgCRUKwGcCcnk+0H5gs190g5fEp3+07GEfoGWz/E+Y//ytvlD24Wc/D7VwiCHByH/3nN7ffvahrvjzip1WXQ0mP766r/O1AZhWkaD68tSyzhW/1D1R9tm3fZceCaZ57bMtfWrpX01n6mm/9Bz/k+3r/6AvkMn+aDXgUoQmP8LZqlulz/Vl5J22mLnZn+Zz3aZ0/032wZfffP3RR2L555XMTz/LUfFMwg2L1tUKzlKQyX5sMtD6XgZrGB/+V5C9QR6wZ2oTWZ2xdkHCPhNwsH8NWbs5h8MVhc87K7loVpWdHOtit+nOi/vX9qwbb7TUjdQJbcJzN8djB59GPNgzctqMGaeCX3WVhBMD1AFnetaZvvRutJznWHlmQtiLmKx8ilHlRT/Yt7x1s62ecvVfJgCA8k68QiHh6vETN51vh13fWevPjWcjM4UALD5KsgwTc2vTogbtdjygtdrSsOdASUCxWm2m93HLEe2mX0zkeLWa08R43Lx80s476t7mPpPMlr4+5pVC5Ocm/caE1vOm2fjm87yKUWnTnacaTdrkEJKtEXclilbKsxNjnwek5dGUcHQ203f/auMWy+7m82qW45mQq3v3bD1yxHbUjLkLLoQhAEOJhQAMEis2AJOOjCI3OSbUPyohT82rQ+xnt/qqQp9VzqlaznN9rzqff/7dD1V/+/33X3/8/ONPPx56wDcrL+3G2po1q5vYeT0/PENWuVarcyHCqKP+5eA/IpwprBKuvTYE4Orz7INysh9ZNP7+I5nx7rsvIiIC7u+bKvvo60YW91NSHtsOk33z56SjT/xf+vj6+ASERaWzfPYzy8HVPm1qvPd2YGDg8zPzG/z65chNt1Jo8lss/oniAzCl5rVpXntGV6nQsPcBN/LMx+fGtav+Xe/lbolaJis2IjQsNOTmpubVv5J92GLJ1fCMxKszanzxySfzjj4ND/Zz3jT+o4q/dT8TqJJnxocFBAQFBPj7OO+d1vLjz9uZOiTzlPvcQTVkspZTNtwLCL+/p0ONr2VGy2/E5Kme7230w+eyLlbXfUMjAp6cm9dI9lXNDmcDWeEUaeFd/RvIZ6lNcrUc/N0HLUzPhgi7TNDJ4U1lP46wdk/W8urS0e+S5ytycxmez/OxWzOqOQnAWo7Jz8mXUxqejzw1dtLk6YUCsHASa8aDpW0GLj7nk8Fp1eSjpIUZWZSpPnYm9Ws3bNLJ2ilboWbKSpcT/k35GTk0z2uo+7tqV99wxhCASYrRZPqfsBrVfvb5TF4YqwJ4F4yap2NvbljUuPueF5zYRlFUXna+nCZNUuD+/uNnLilrATjOccTUDQsvhGvIVpHfdLX3oabjdp72T+UM19zQrEqYkUVolqmIy+vMjboc8qHUKo7XJnus3bF94Z2omDNjjafNRwCGEgwBGCRWbAAW6Cd75kLOL+35w5etx6/zTGI5FWmAE29adq399ccfffjRhPHjK1f5ddDO+wFXbf78veLAnd5JcV47xskq1Gh+xFtBfqb0LwX/DYU8OzM9LStHLvQZlI9tOv/xQfX59sEx9y3r//h562W3U1jhG1WlPVreplKV+vuC5En39kyqWOnXP2v/+ccfncYtuJvEcbwm+vycvpVln3/306/VqlX746dvP5TJ6s495peh5YUjGq8LwIyGV0RcWNNf9sFnX/30G3nmbz9/+QnJrBN3+uYIf5VPdDHtW/fzL7+cuOl+slKj9LCpUqnSZ2bOqSrymFqeGR0XH5uVR1MkfKnyIp13W7RoXLvqtx9+12PW+QQhAPf8VdZszvEQ4YS4kKOta37fYLFDVKb//oY/fvfJzAsxSuEvaHM8d4/7+Mc/zZ3ThBnY/q3aJwTgRJfVA7/+0Mj0TDD5O5rAE8ObyCoN2/A4qdQEYOHsVFbDa1KfHrLUBWC1OGsY6W1pAo+NmWRaKACTr0CT7rqo2eA1Di9I/hW/dZoRZlKTZydFBoc+c92zZe5AqzuZ/+ZhBigrSI+dVbGKbNcdhQMwpdJq0n3Prh7ZfL5LptCv168N8Gbkt4yKum47v2HrQ6G63yBCmPeQtF0qn939xs8qgwF45Gxri0viVpHdSfPcrvmE3Wf8014FYIIS0q8i1nnr6FotBh58wnIaNadJ9N55cMecu5E8H3NsmMlM8ytR5PcKARhKJgRgkNhrAzDNchwdfHFl/18rdpi0wUc45VmYtDMvJyMpIjQqLjk9Mz3cflr1nyqYHbt1YHQVmUz24ZfffffNV59/Ivvgw4++b9R9k7vQy9a/GrxvrFab7r10ZPNfBq+6n0y+LM5jXec/ZDXmnwvJ8jvS8Jfv26x2TWWEb5xNd1/TrnKVWtu9KZWa43KiHu+d27pShUrff1u59Zil3lnp1xaQAPzJF998X7FSxYqVf/y5SpWqc/c/TlRqhSHgVwH45x83PhEGEHmthiD/pyLPW/aXyT7+7JsK5IkVK5Fn/tzSdJtbkpbLe7RlYMOPP/+6zbJL4Qqyqlr+yPqnyhU/W3g1iSKLGhWVm56ZraBSXpya3aRqu9F7XuSnvjy7oMPnn7SfdiFRCMA9fpW1nXc8TEi6wRfa1K7UiATg7IRzY36p/PWcS7oArMn13m/88U81Fzin/psBWOiR5L3cPrmBrOGEE8IIMB9oN7COrNWM/SH5peMUaD1KVTQACx8Qzf0tAAsXbGa6WLQcYXXyZYbQ5aKEIysZcQm5ctLpFL7oXN/jqwc1WnQvhys6AxqAgKJZ5i8BmGK1XMbz0+tHtpjvmqHG5Ffwrhi1Vhl91WZhw/aHwwquw9EhbVcZDcCp18e2Xzh1y0u5MMrLcpm35jUZb3s+OPNVABYmWtfmRbrsMKnTstvehxmMWsVpNSlPN+xYMvCgc0xyctztjb2Hj5tywCtHIUyYrXtWYQjAIDkEYJDYawIwQ35rwp3W9an0UY2ukx184lISYqMT0/LUymc7Bzas1GT+cc/YzOgzxr9/V6nH7gdep2Y0rPHzzz9WrlSxwndffib74KOPKzXtvdUDAfg/xPJ85vNlg2vIvui49PzLpKQXx3rW/eaTgZvdUljFA7P6FT75eOoRj7CkpCjPo3M++uTbOjMdE+Q5aXGRsXGxEeGhwd5Oq4fVkf3SbV1YQsSGYX980sxk/52g4IAXPj7enj4Bsak5wnnx5M/oA/CKryr9UGmJQ3BkTFx0VCQRm5JFJXnuHvVjhUZ9Dz0IDvLz8Xn+zOtlYEQWq4m+NqlzNZlM1nvxUY+o5MT42JTM7IhzJr9/+tHng3c8iE6Ieu6wsnXFWn1tb3qcmfu7rKLRQsfEtISH1uPryz7pPPeSeAo0CcCtzY6FCOk5yL51rYoNFzlE56qf7atX+cuPem+6ExqfFOV7dYnRh19Vb33S/189BVroaGkTbq/t/c2HTcfse5aU5LFnRJ0PK4zY6JZaegaABe8WgEn+1WS5LWwycu0J7wyNlqVpFZMXf3tv62ZrryWLBxr4dPf9FoNarXiIAAzF+XsAFu4qkBl40np44wXOWWrUGnhXwrlM0a7WS+q2OxKoFW/nZ1BGAzCt5lW+uwYPn7TIMUJsbjmvbS3Hbz/rk87xHKcRZl6gGTWbHXbZalSrHmuuxwqHkEXy8POrJtT4tWb9BvXr1a1e5Zvvvq/4y4StlyLkmmLu2IcADJJDAAaJFRuAGfLTEuW8ol8lcWYh2QcfilMY9Zp/MkrF+e7sUUs3E5FM9mWFapZXorIpVp6dmZmZnUdlxT6xHS77rEqDw8/e511ZoRi0WqMKcVjc96eCe/XIvqhguudRsjDfzLOjPQpNOPXZz406nPCj8+MebR/6a8XPyPf78Scff/TBJ5+1GLPUM49TvNw6tes3X3z+0SeiSr/03uIUlqM/+i4G4KSbKz8vqAN6PzXvcy1O4X98YZvvv/rsY/GJX1T4ebTlqQRe6Tqm1Y/61XQ6b7sfk+J1pEXN7/UFMlnlel0vBGWHOq5p/Zm4/OlXFapU/f7Tb40mHI7glE9mdKwoazLjcLAQgANPN/n18z/N7CPzeFXc5TlNqn4iPkNQsU6P1TdT2X/5Bj3CvJwJD7aaNhRO6hZ8+EW9qfvdkzVF7s1Y4gkBOMV93/LB9RbeLRyAAw4PHTth0rEgfQBmNZzfwSZ/LrRxiqR5Nek50Sybn/z80DSjyTaO3v4Bnjd2Tx7dbsS5ADVmgYbi6ALwrS2//bzmRIQYgIU5CaLvrJv3S6uj/pg8HN4ZabPUKd7HlvWusfhOlnDvcX25QAjAz7Z3HzFlgUN4GQrASiWr0STdntZ7gfHay88CAtwPzW9jvPVaUKaaU6RG+kUkZdMcleS5Z1KFyv3WODx94efn+/zZsxdBoWl5yvzczNT0lOTUrNzMh5t7DB1jesSXYsmnVsz+hgAMkkMABokVF4CFCX2TPY4s6FarVu06tWvV/LOGyHjdpUgFr1W67xzTpa5QUMvyfFAWLczFIEy+pOK0vDo35OzSOq16jbgRyRc9XAvvnzDPTJrHmum9xS/sz76rLvqlc+IYHkvlPrXratRAKG/U0swxPJtRqVUcm+rvaNVfKCSaDp1tH0SRYKTScqme60z76Mtnbrsak8eJM2AJf4NW0fmpj/e0al6vdkHFELQaMOF6rFarSfHYs6CzvmiSpXMEy2lS7y+Y2L12nTp1DDVp9EGPRFarTnmwz6hxHbJcy6jzGpd4lZbXZAResBogrNfDxNrl0YmlgxuMWvUgJ8/fxrRTnZFC/dNoNOHXR/ZqNWCjU0yucPOt7JBLc4waCq/asFXP495yEtfew5EXWq1Rx97ZMLWt8IdqjrU6EZyrKU2nPwuEaJvue37X/KFbnuZpdJ1Jsl1c2KWFy5avdozQaITxbJrV5LhvGzRv5yU/wzVnFMsxeXG3NjVt0qBBg/ptug87/4JRqcS5jQD+iqIZWpHz+Hjv7vuvxWbRwk3IOXVWmMu+Bf1sHqQZ7jYO8DakK0LFP7bb2mPY5VD1X1oc0naxAaemWazZfDtG13aVGQynlfuemDq4hdDe1jU/5p5EabRMfsz1VX2XnvfJysv1PzOjYe1a9Ro2btyQrEJ0Gzv7drxWoybtMumJkY5d+MW5Sy1tXGJxH2AosRCAQWLFjgALFAryH6owXbkw/YS+gHR0SGfmb70ZdG8kI9yGpPDXUzAKr/+m9MiCOERa8AUWEJ8rrl3kVcR/iy9ThP5hPX0Z6esWfWZBZ1dfJNKXFH1PtPCWxH8YCoTHFbqKKKyv+99f6DaygPiS+kf+Zfqt0f+dVxtWRpHvQv8vHfGLkefn5+Xl58sVlK4CAfwjqDTwvylvNYcSfsgUYnsrV4o/PEKpQk4aX/H3UCH8jyy+Itf/Tr4jBGCQHAIwSOy1AVhME0UU9PlfPVJcCBAfLOPpoCQTwpn+2ynyLVAktOgVLtcFTVHh9V+9Cln7719moRcr8Pfa8aro1d/QKXjJ4t6TfmXxN1/4t+5FhELD+xNzruEJ4qJOcW/1X2TYjkJ/vLTRpfeib18oKVIkrENW1C8VKPw564sAXkfYuYvUoWJqHsDbCT8pr6s4f2u7yg6yaQXNbcEGih+EuPCqMdb726fwlk8GARgkhwAMEnttAAYAAACAsgUBGCSHAAwSQwAGAAAAKCcQgEFyCMAgMQRgAAAAgHICARgkhwAMEkMABgAAACgnEIBBcgjAIDEEYAAAAIByAgEYJIcADBJDAAYAAAAoJxCAQXIIwCAxBGCAwhQAAABlFwIwSA4BGCSGAAxQmAoAAKDsQgAGySEAg8QQgAEMKIrS7xgAAAAA8B4gAIPEEIABDBCAAQAAAN4rBGCQGAIwgAECMAAAAMB7hQAMEkMABjBAAAYAAAB4rxCAQWIIwAAGCMAAAAAA7xUCMEgMARjAAAEYAAAA4L1CAAaJIQADGCAAAwAAALxXCMAgMQRgAAMEYAAAAID3CgEYJIYADGDwDgFYy6k4jVa/UPqoi753jYpsMkWpit8ijtPoHtCoxdV0aIZVqVVqjbgKwLtS02L9YVRqfQHAu+DEilO03mjU+upE0OxfmrWyQKtmxQ2kOX3B62gLfo+0Gv1TRDRLqNVcmftgoKxAAAaJIQADGJB+g37HKI5WnftsnWmHesbbn2bqi0oTLuGO/c5+tp4Fm8gpIg+tnFC3fv36dUetOBSs/EtPifM71afjbocQhueVz462bdGErFi/QYOGDerXqVWz08jdfir9igBvxSQ/2DqytlCF6nUeMe1mDPO2bj0AoeWyn59dKlacrsOn3ovXVxsq8PDSEXXqCcVEX6tz/vm6B8oKxv/0lEFGwsbVXnjMK+N1e0uOn8ORJUOvx4gLuX7HV/QRP5D6DRo2alDrzz+rG2+7moA9DUomBGCQGAIwgMFrA7CWozOfWc/uU+GzD2UVB6y7l6ovLx00rCI95umRGZV/rNtw0SPdJjLBu+YZNxux9VFERITb7lGTzNfdiCs0qKt6fr5Xg69kc44HKcnzqZzoqEiyYmRkpP/t3SPa/tL2kA+NoQV4N9rc4BPLG7Sec9yP1CGf6+uXzhl/OED/GMBraTOfnzSv1cr8TEBExEtHa/Npw7b6CMPA2e6W+3dMOeefQMTHxsSn59Jl6YQUTYbr1N6Lpm+5S3aX4HNL2xvvuBmu0D+mp1XTuYlBN1d0qfhnXaMLkWIZR+ekxZGnRERERUV4HRncuEO7hTficZwSSigEYJAYAjCAwWsDcNoD20kNvm9s1KLu7z/+NHjd/TR9eWmgpZOuranevGOTYaNmjG6x4rG4iVzEabMVvae7ZImrhG4dMWHm8lvp4gLBPnPot2vZ2J9+2nraP1dfppPqtmZMyx7H/TCsAO+Kibu7Z0yT6dcM1UuVn5HP6v8N8Fr5fme3tm53SHewJP/RTrOB3faEkhYt6tyOA1abH6ekETllLuFpA/YOHjNtxe0U3WLQ/jbGuy8G5hU+4KjOeb5/7J+d2zY2HrGkTvsL0fpiA8bLauSIvpY3EtFOQ4mFAAwSQwAGMHhtAI51trPbdDwq3X/54Opf9FxdqgIwI8/3dbkWn53ge2HbhKZL9CPAHEMxCkajykqJC729aeTUpTZ39R0uXv3yxIA9F+647GtbZ+OxF4UCsDbd58ji+u2OvcD1v/DO2FS3g5MbLriZlpcYSkQkZNL6RwDeSKtmmHylRsPkJod6n7OymDZyfxDJgfnPzq6b8lu9jsIp0HXam1q7xOSXpcvK825P6Lx05q5QfStLPVjczHjblWilblGgVcjjbjt7Z2eH3dmxtlabc1H6cj0u6MDUiVMWXorDSTpQgiEAg8QQgAEM3nAKNKclvQmV37Ih1T8vZQGY15LdnPxPFeqweZwhAOuoog5bjm1Qr9mkxadDk3PF4QJVvNPapcfuhuWEnjX6c0PhAEwn3toxrNlC56JjwgBvRKfcOzapweqD57dOaEA07DTY5mp8Poam4F3lvriwpEH9Bt0H2bhFpKq0fMa9dRNmTVrrkZ6fn59wb3rvQcM33kguOwPBGTfHzN+6/EaifpEPOt1i8sGLEYWvcibtuRhuMx7sWPW3AJzvtnDIrEHWPjjOBCUaAjBIDAEYwODNk2DxvNJnyeDSF4D1lEEXNv01AJNkr1YxbK67RV+zaeu9aV6d7rppg4NXvIbnE880rbnFPpzRr8nT8W5bBzdafFd31jTAu1FlPtzW/dvm3Tc/zGCI2BuWM5r2sXyM7jm8Kw2nIjUnymHx8q5DTsf+5fyTrJsTGy8xPxlWZgaBM26Onr5p8eV4/SLvf7TZxH0XigTgAslu2/8WgOV3zEbNHLz6JS4ygJINARgkhgAMYFB+AjCbHRcXHZpd0JdUPZnSymLc3rB0DwvjBSu3XnFzc7tnZ1GjiumSwx6JutE6JtZ1x5CmS+5n4bw6+CeYFLcTM2rveqFf5OVPt68e33adj34R4DU0TFZSxItkwyG4+CPzlzUZeD0xNz4xMUauL+R531W9rK1PhRlWK+0Ud03bLJi1OUh/koTi7rTW03c4xRV3xKiYAJx7d86omUOtfJF/oYRDAAaJIQADGJSfAJzpuXPOsH7Lb+u6SRr/Y2Obz1t4JTLEadmkzq1btWxBNKj2xWe/1Gg0zyFYuPpMnX5n38Dqi5wx/gv/jFYR7rq37eStL/Q9+JQbayf077s3XLcE8Dps2v2Dk5oP3ZmgW8zztB06ue8Kn3CvtZP699v4WHf0jnm0c8SUTXt9S+Od6V4j6tzkweOWOujndn6ypuX0w05RhS4BfuXvAVjuvnKYeZ+V3si/UNIhAIPEEIABDN4WgBXeZt0rytosulO6boOko/A/bTW4hpmbrieljb+wzrr7wLUnHRwcTm4c0nP6QjvvorfaCD5UvfIaO39doSY36PCC75sefv6X8w8B3kqd+fLKql6zbY+QuuZwYPm4Kb3MHf42cS3A3+SHOx3q2nbxZqHi7F88zrTfgquk6VV47po+rc9o2zOk9Oja3l1stjkZLpgtE5Qv1o5bOtJsJ9m+syumdbY4+SSR4bVUos8t7+hMVaFTcBJdbSx+bHAyQr9I2ung/eNNBw27YDh/GqCkQgAGiSEAAxi8LQAzkcc3TBm26sxf7g1UOjDxjy9tmXc80DA2QIdc2DSui2CUjb3f384gjHM1m3rhbowuL6tzw25umXbYMwMnQMP/Qh52foFY1/pMXumI7jm8q5zgy/O6ihVn6rIbSfpCXvny6NqRYn0av/NGaBkc7kx1WTu7v7B93bbeiRTPneAyHm4bb+scJC80f1yW39XT0y0eFszfz/OqsDO2R1fuL4ufCJQ1CMAgMQRgAIO3BWAAAAAA+H9BAAaJIQADGCAAAwAAALxXCMAgMQRgAAMEYAAAAID3CgEYJIYADGCAAAwAAADwXiEAg8QQgAEMEIABAAAA3isEYJAYAjCAAQIwAAAAwHuFAAwSQwAGMEAABgAAAHivEIBBYgjAAAYIwAAAAADvFQIwSAwBGMAAARgAAADgvUIABokhAAMUxgAAAJRdpOOn7wICSAQBGCSGAAxQmAIAAKDsQgAGySEAg8QQgAEAAADKCQRgkBwCMEgMARgAAACgnEAABskhAIPEEIABAAAAygkEYJAcAjBIDAEYAAAAoJxAAAbJIQCDxBCAAQAAAMoJBGCQHAIwSAwBGAAAAKCcQAAGySEAg8QQgAHeDcWqOLK/EGoVS+kLSxOKZlhOReuXCLJFanGDNGqWNmwRxag5jVisVTGvSmlWV6jRcK9KAf4BoQaqUH3g/+9Va6xRFWq8ygyKLWiF37R5FFVohyJNtErfcBdQF2rBiyCP6buAABJBAAaJkXZQ3yICwOuQUKiislLiQ0UJ6dkUyxZKkqUAxTC0Mi8zKilbQYl9IlrFKtOTo176EoFxaTmMWugrkQ1l8mPCg16Q0hd+SXnC0yiKBBdGnh7oL6zrHxyRTSPEwD9F+uoMlZudHJORp6uBAP8b0nZRqYm6tutFSGJmvq7xKjNolqMyYsPF9tYvPltOmt9iNo8EXobOz06OFnYoiqbzsxKiAoWnCF68DAwOCopOTM0h7f3fn4wADJJDAAaJIQADvI1waD038t78Yc1kgg86TLL2TKQ5dWnpx9MqDdnXM70PbjA3WuyapVbRNM1pqCiPjVMG1mjQsEEdo2HzLzzLVHEszSpzHu0Z1MGoVv0GDWr/NnvHo6gctUqtlsf5Hp1VrVa9Bg3qGrXpsds5M0/I0wDvhmI5UgOZoMtrl03sci6E50rlKRRQElA0p1WG3loyulO1ug0bNqzTdvTia8/ShCOSZaROUWqOibqxY3CXRvUaNKz7u/HaayHpqr+cdEQxwg6lCr+xccX4dqf9tRqtRpPidWhJ7/pEvfoNGzeu88P3333Vb825SIVWw+if9QoCMEgOARgkhgAM8GaUSqvJ8LKe3OKjL775vXa9Wr9/88VHVYdaOUWzPPf3nkVJQ9GUIi8t/uHNk8uGfl25dhOLu9katUqjof2PLpo8esJhf5Zl0x6a9Zk5bcO9XC2f52Zh1NR0w/UoBUvH3lhqNGDr7eg8Jttr24A/2u1wy6TYzIAr1qatZ1xP0Lzu7DqAwiiKVuZnJQW6XNwyqtHXdRp0uhjGaxCA4X/DaLTMyz1z2g8avdo7nWWYeOc5IzatPRMm15SC1vgdkJ8bOuTk0E7zFxzxy2PZ9AfrOg5Ybe+drOJV+n1GGNGVZycHuTruGt/8m1q1Wp0J1B1RUuTn5+bk5OTm0azC9+joln0GrnFL1mgLX/VSAAEYJIcADBJDAAZ4MxXPZ3ja9K8iq9ZvXSDLq15a9vlT9vNIW/cULa8u6cOgNKtRZkSeXfxHm96dTGaYj2u26E4WR/o+WU+3jJ/desrtDNIGkHbgxf5B85YvepDBp13p89uoufZhYusQcWxQ7bWXg9Jz026uq/7D/GvpQmGe/6HVvdodC+BJAEaKgbcRrh7Iiru7fmDtAX27zTRe3aDTuTCMAMP/hsRDLummxbDJ3VY9Z8VGiuBopUKhX6G0ozmecl/bdfyMZXcSyKZpOD74ePth1kefJLH6JCvsUDlJjzcNrdW3Zycz01V12pwO0qqFHUq8VIUhv1d8rMOcPoMG2dxP5V/zE4UADJJDAAaJIQADvBGt0nCJHvvndR1msTOQJr2tJzu6VPm40/xTMVQpGAGmaDonI+PlQw85n+13zmZUY3PXTOF86MhrKxdO7HvIn+PVFKPls+7OaWsxeYsvxcdcWTxs6cz1xx2uXbU7uGzMZkf/dKVaHuO4dWxvK6tzjtcuHFm/ZcXsUz45KqYMTj0D/zpSA+WpadEPnkTzbPS1dUvrtD+LAAz/I1bLZz9aMXnlQPO7ASFPLl+4cPGyZ3gqreVVVNnIwGQDk66O7LF06m5/Ja9lhYtv7qxqONr2TGA6pztvguxQioyMmPuPI3k2/s6W5TVbngzUBWABxWo18ntrJoyauuJm8uviLwIwlAAIwCAxBGCAt6JolVrsMKjin+zsXvePqqO3uSZSWnUpOedOmMGZ57N8T20wBOCwa6tsZw9zjiN9LGHQgfPe23vt+s3eeVp50NObe4YM6tuhQ3ujRRuPXYvMohiGzY67c3HvyjZtO3To1qvX5gv3vTIUGuRfeEcUJV6zSIdfWbMYARj+dyQfZnqsWzqnx/Q1Fssnd+zYsV3DgeMW3nqWRJeRa4DJBsZeHblg45KrEcLQNsWw2hcnjIz3ng1I0wdgEdmhyC8SE+1iu7RIAKbUvCbhwqTBpgM3vaT4Yi7+1UMABskhAIPEEIAB3kqhZEhqVMc/Jun3lx9arb4WlM9rVVTpGXIQxgUyfAoF4PCrKy2nDL4YQTZDSWt4yt2qwwrzrX45UTvbtRpmdjVKmGIl331LzzqjNj0ITwo6a/xDm2XCiALPp7vbTm7Vdtn1dFqFCAzvjCRgZdhlBGD4/yD5MOvZxlndKvZYdCZSPAc61d60Ya9pVo9S1aRF1q9VipENjHMcPs5q7ulQkmUZYep0j92NRm875Z9aOAATtJqnom4VDcAk/3KJ54zHmA609lPymtdfoIMADJJDAAaJIQADvIWCYoU5kx/vGdCiYd2+2x+mibsO6Y2UnpPuigRgIdwmulgaT+6w0pshnSwVz8dfGTFs8VT75JirnepbLrsQpybBmFFpsl0X/jpq6+WnpzZVrXPwca6SVZPnykMddoyvNuNaJqdGAoZ3hQAM/3+Mls97smGhjemGENKKMQqSgbPujKpjMWmvT65wMK/UIxuYfXNKi5kzrJ7l8LyaNM4p10e2n73tekTeX27AXkwAJvlXFXN+4uhJQzYL479vmKACARgkhwAMEkMABngzitGo5b72c5pXkMnqDbDYdfbU0QP7ztx5Hp/DlJqz7ooEYDWn0bLxly1mdR1h6yUMB/M5x2aOnLPMLp5n78+o03fCRs88sXVIcjLr0GGRg1fYw4NdKo3cG6wUCrVRl1YP7Th8X7BcXWZuPALvHwIw/P9RrIZPeWxrPmbcRicSEAnNY6tWLacucIxmhMmwSj2ygdq4q2a9+i844KkWti/v8vQWMw46heWTplfYXLLn6I67/j0AUyrSZrssHDqv5xovBa990wU6CMAgOQRgkBgCMMAb0RzPxlyy6PqVeA/gAh/3WXEnXpx/tFSgWF6b7n1kZd/fZzhnqFUMy/H5wecPzug5xsTS0nKhcfdeG3e4JKk5Ljvo4Jw5nUfONLe0XD2nT+cOI+y9MlmWDrppNmRsz+lLLS0XTxvVbsCIdS/SeFWZuOsI/EcYDa8MObd49o8Nj4XgNkjwP1LQKi2bdHXH3n6D5plbrlmzwmRMr/67TnlkMhxbBvKvuIE8Fbl/7pJew6cts7S0GDq691oH71Qlx6Q/d9hyzj08S86II8FCAI64vsasUq3DfrwYgBktz3htHGbSfej5KJ5Xv2kXQwAGySEAg8QQgAHeiFap8mPvOuwyMy/M5uzd8ExV6bkVLq1ic2O9XE5sdgzKFfpPCmHmq4wHl7ZPnjhxoom53a0gBc+zSiXL8/LA/VbmJhMnGk+auvNxYj7LqVVqTs1E3pgzzZiUTrWwvh4jzLqqf2WAd0GzKmWy782rG3d5J6vKRlYBKSgUJOhpkl1OWZNGaqLx5L03nudqeY4pM3dCUijVPJ/x6MCGWaS9NZlywjNGrtFyVH7S4yOLDrgGZuSz4g3oKEalTA28c916u2eCihF2KPJTlRd29eiVPRfC89TF3f33FQRgkBwCMEgMARjgzUjHSpgC66+0wjXApafPpVBQjHADYOF+IfoCYcTAQE3rtkahoMUz73SEW0+ScvJA4VKkX/jnSC3SVSKOVDV9GcD/QGjMxIZIR1OG7gMsIltT6CdHQ5pbsQkWJm94dQq0uEOphCbcEP7J/3Q3R37rTaEQgEFyCMAgMQRgAAAAgHICARgkhwAMEkMABgAAACgnEIBBcgjAIDEEYAAAAIByAgEYJIcADBJDAAYAAAAoJxCAQXIIwCAxBGAAAACAcgIBGCSHAAwSQwAGAAAAKCcQgEFyCMAgMQRgAAAAgHICARgkhwAMEkMABgAAACgnEIBBcgjAIDEEYAADiqI0AAAAZZe+/wcgHQRgkBgCMIABCcD6HQMAAAAA3gMEYJAYAjCAAQIwAAAAwHuFAAwSQwAGMEAABgAAAHivEIBBYgjAAAYIwAAAAADvFQIwSAwBGMAAARgAAADgvUIABokhAAMYIAADAAAAvFcIwCAxBGAAAwRgAAAAgPcKARgkhgAMYIAADAAAAPBeIQCDxBCAAQzeFIA1GTEB9wVe4emsvqyUUeenPovK4fRLZJOSI59dcyTuBCbk6csIyt/j7lWh2CNWpS96hc5O8XyeotQvAbyeIsY7KUel0S8JNHmZkb5pOMoE70LLZkS4Cw2R6Jqzy+17T5+/zBAbJWWs312x7fKMzHrVopUhKX7uN4XtuxGUyeiLiqPIivRJ/XtzrMkKj4qMSKf1iwAlDwIwSAwBGMDgtQGYywu7adbxV5ngo6bjdj+Oz1PrHyo1Mu+vnm1cZ+F9fW9JmxN433JU/yYdO3Zs177fpNNuUcK2c3m+55a3b9O6FSnuOHHD1gepRTaUjTtvMqJq/SOh+mWA18jz3ru4Z2Wb21mvjhYpXxyaYTq49YVo/TLAm3B5L8+bCw0R0blb+3q/VZRVGbDCV8nnBN/YPKJ7sxbCA5MWLHOPjn9TRiyF5D4nbft0bUs2r32dMWbn/NJorf6RItjAE/OmDGhyJky/XIBLdbE0aj9q4bUEfQFAyYMADBJDAAYweF0ApkNOjK8tk33ftO+Qod3+rPThl71WX08stktSEmkUcc9PHLAx6fXNr41arHiiGxWQ++xeOHrsvBs5wkL0hpFTZ1rdJr3I4B1GbetOORsoDqokOg2caHE1KEsfgbVMttuM0RUa/Wl0MlxXAvB3bKqXw9ktEwdWrfNVpY2384QRYCY18Nb+nebd6n/fvGEPdMvhn6OeHR03uP3cW5k8E3bUtEvL4bZe4qGVRHvTeWt33E4pQxE4P2Bv/y4W1jezhYXog/26Lz/nnVl0mJvNCHE9tGdprybfN63V7lLRI0pMwvMT45p88sfINa4p+iKAkgcBGCSGAAxg8LoAnBt4aaPxuA2nhSPtaav6V/6wjcX1lMKndpZkakXSpZWNhpiOW265ztRo+SNxExVem43ntJ3vpT/HOf7csNmrFz2gsq936rF96/1cXSmf/3hhtRXnnmcLm6rJS3xsvanXhqUbarU8hQAMr5V+efHAMYP6HdxsXaPW3pvZpOdOpz/ZNLr91OkmlvM2NupyKV6/IsA7yvc+Pat3i7m30sm/o87MWDxvomu+7hFe/mjoKNvF12JKzQHJt1F5LG8/dd4ab4VuMdWx58C1RzzSCydgOst7+/h2k00nrF9kW7fd+cIBmI29c3rX/qXGXWYuW3YNexqUXAjAIDEEYACDt02Clfvo0q5pPQcNt3XyKz2XAatoJj5U6AnFOm4d13SJLgAnulitNB1kX9BzYjyXtVsxd090etCiUcaLToToOlv5nvO6/dp/56Mc0rfkst33Lje+FeR1aEOdFsf/es4dQAGtPDA8S/jHoz11/tzqlKnmeYrKDommeT7vzqbl9TohAMM/w0Rd3DGm18SLaeJS+qP1FsPH7Hmka4LZZzYdK42ZfTa66BBpKZZ9c/SgNXOORekTverp5kZjt1+IlOsWBVqKzgmOJA250mP3qlptzkXpy4lMx5Nb15556OswZcZs86vY06DkQgAGiSEAAxi8LQAnOexYMnrO5OV7jj/yyyhtFwErgy5sMgTg+FtW2+ePcssQHyECj/a32bw7SMPHHpwypsPQSYtXrly5at+OIf1+G3LQQ87xVOCxNauPJimznx5YX9MIARjehsu/vbP2n9ucMgsdKop3sl6BAAz/EB1wwnJYtVmeBXM90aFnlpp06jxaaKR2HFxv3O8nk/X20aXnkORbpN8as3THapck/SIfer7VlCMOkYWmKTRIdtteJABn3ndetejMsUQu88yoSbMtEIChBEMABokhAAMYvCYAK6MfnN9oc/pOtJh547b1rCqrOfV0xJvDcolTJAAn3Fq7ZuoQhzjxEZ5XPVvTfuX8LS+E64NznhxaN3fa1KlTp11yd91S1+rq89S0KIddOxwzhTVDT22p3fpSqvBPgNdTIwDDv4L2O24ztL0zagcAAH/JSURBVPocr0JDoLzC4/KemUIjter8rUtzxx3cdiXy7zPWl1JZN0eNXG92Oka/yHnvajR254WIwptfoEgAVmXec9217uYF4dTpzIvjZ1msvi/O8ABQIiEAg8QQgAEMXheAX+wb/4Ps0w6rrwun2fnt6vuLrLrxNn/9NVqlRZEAnOduYzKz/RJf/Wl2qVeGT14xyzHhL9c1R+/t0nnf7ZAQl2U1qvU0nTV75szJQzq0+PaHHvM239AdDQAoHgIw/CuUgXZWg2ot9Hxdcxtl3WnhYmufgkuCSz/20YLWM83W+ervYZTlPLDXsv0PUooL+IUDsCbJcbZpi4q9p5vNmjlzTJcadRo06W1zzTOjtMxVAeUNAjBIDAEYwOB1p0Cz4eemNfrgo58bTjQ3n9bnzy9lv463fVTaelxFAjCf/chm3sDRqx6K3aqMPSbGsy0v52r43OdbVzvd1g0+xJ8d36DX2jtR+RlB17ZtslyxbOnSpfMn9Orw/c8Dl++/E19mrrmD9wEBGP4NbIzzBrPq4++I55/oZD4/eeasvb/YAKUdn96t5dJdnsWdIFxaZT3d0Kenxc7H4p6TcXZ0B4vTj1OLPd5YJADnBjywX2u9ejlpppdO61enUbNWw3bc9slCAIaSCQEYJIYADGDw2muAVekvry5tq7sPsOz7lsNOeot3zS1VFP6n1w6qPvee/ko6LvnxtUUDB3cbNWrU0L49Rxxy8BX6kIqQ0+ZmnfsMJKWDm7YfaHE/TFF0dlX/Q+t+qHvGcH0aQPHUeU62P1bacCWjUACOvbJqQZWm9rH6RYC3Snffs7ROzd0vCuW43BDHg1N7DexOGqlBLVv1WXLyuWEugzIiw2WLVa/e/ckGDms9avzBJzFyjueyPA/P23cvXFHoyGPibRvzH+qd+Puk/KlnBk2caX5bnIwOoERCAAaJIQADGLxxEqy8YPdLJwT3ApNK41F1dU5MwKPrz1Nf9Z9UYd43dmwl7O4HCPcX0aEDnE7vJoXbdjtH6M/Ce4VJD315+UZEWRpvgfdCo0oKvnI5KIEptK8oEvx9HG/FlrJrB0BKVEqg7wWnuL/UGTrkyRWh6dq2y8UvriyOcXIRrhf3CW3zrodxunONNPIwV7tb/smFdygqJcj3inP0309Gyou4/+Tps3jsaVByIQCDxBCAAQzeGIABAAAA4P8LARgkhgAMYIAADAAAAPBeIQCDxBCAAQwQgAEAAADeKwRgkBgCMIABAjAAAADAe4UADBJDAAYwQAAGAAAAeK8QgEFiCMAABgjAAAAAAO8VAjBIDAEYwAABGAAAAOC9QgAGiSEAAxggAAMAAAC8VwjAIDEEYAADEoDJHgEAAFBW6ft/ANJBAAaJkaZQ3/cHADEDAwAAlFXIwCA5BGCQGAIwAAAAQDmBAAySQwAGiSEAAwAAAJQTCMAgOQRgkBgCMAAAAEA5gQAMkkMABokhAAMAAACUEwjAIDkEYJAYAjAAAABAOYEADJJDAAaJIQADAAAAlBMIwCA5BGCQGAIwAAAAQDmBAAySQwAGiSEAA7wrhUJJq3U7DscIS6UJee8qjvR6VJRuUaGgVbptEalpUiI+UoDRkPaBYynyzAIKJcWSUmHrAYpHvapWGkZX2XRIjWM5Un3UTCnbdeC/QzGkhdWqaH2zI7Q4nL42kXJWX6qgGH0RUfra4rchmyO0vnqaIm2wiCzSrCHCallK93jR5/Gqv7XpBgjAIDkEYJAYAjDAuyGdC06bH3nfYc9O+zsvU+UsS+sfKfkUJLiqlelRvqfcIvNo4X2TlMImeV7fuXrlypWrt131iKIN3UuCIuvnhpw94eEbl0OxpDcqotWsPDnk+bEzL5JpuqAQ4BWG4zXJbvs2W65atdrWIVjBqlhdCBb2Hp7Ojgi8ez4ghWVKz64D/xnS6qg1dOyDU97hqXlKod6QVotOD721jrRSqzYcPPcsh7RaSgUJedpsr5vHLUnbtdb+SWCuhldTZScDK4TGOffFJTubVaRttnIKSqbUJAQbtk88lMkxWb5nrFatXGW1+7h7Bgm7QiNPdr4sn3sn1wht+o6rHnE0Cc+GNr0IBGCQHAIwSAwBGOBdkA68lmeiHFa0/E4mazJpn3+u0BcrBRRK3WgJHXxqdJ+ef8y8mcmpVSoNlfjo4PrB3QcON5lkMrx/t4E7jntmqYTxXoKiOT7Xd/Xw72Uzdz9No3ld0mc0nCrDZ0fndhVq7PbRqAyxGECHolWK2EebTYy7Dhk5wXh0t7FDLT1CFBTNUOK4lCb5xtKePTs1ORUkDmoBvKIgbQ4RfWnZyMYfLr8cnkNpVCo1kx/+aM+oka1HTDIZN7hv9yXLziQoOVaV/fT0noF9Bw80MTHpOXTQkrP3ExTioHFZoKBYnk12WLOie++B40wmjW43fPg2l4BsSt82C2PkLC1PfHjGrM/Q7sYm44cO6L5g/vGYXPJx8VS8ywmzPoN6kTZ9ZL9uA/cdfZqtNjyvCARgkBwCMEgMARjg7cSD8mzM5flGlWRE7/l2QbnCiGmJRzIJnZfqc2PBbJOu7X+sZ2S05G62hvR9st03j23dfPTOJLEZSN03bYLpiuORSp6jlRTF8fn+izb1bPzN50uP+abrAjDDsUxa9MV+Q35tW8PoiB+HAAxFUBTNqvI91hnVaLPuTopQqyLtR1XquftpGqVlskLv7Z0xY2ijhlW71Ot6MRwBGIqgaSYn2HHdiinte9du9tW3GxxjchmtlpOH3tg7oOUIuwShOrHuBxYN7b0qUMNHHurfp2lHG0+hlHKe1nvsvJ1PsoUTgfWvVppRDKdNvbOiZ4/529yUwgamnRrTeu6R21Fi2yw26ByV7Hl09B91xt9JJ49zAScW964+zZV8ABnXLGf3GbHyMS08L2GX8cjpSy7GM7rn/QUCMEgOARgkhgAM8DYKWs1rct0PmDZuV7t5nTpVZZ1mlpYATLNcflr0+dXd51ot22y73riphWsmp1GrUvxuXb904lk6xzHCCYUvV3dZMW3xY9KhUtEqPvPlkm1TDh9cUa/GpsNPU0gAZmiOpuICrs6yHrdt9bqaRkdeqlkEYCiMIn1zOtvPccdlz8As8Ur5iEujf+i9xzOT5bLiH9iajrRet3yTxfr6He3D+OKHpaB8oiiaUcXZr52ycMbsi4dsa9TYdik8W8Xz8hjnTdN/7X0xVsuypEZlPd+4am6r1c/zEu+duXXkZjwrVDLF7Rl9x8/fVWYCMKPlc2/PbjFt1uqnWSTdsio+0XFAV/OdN6PkGjUjBGCGyU8LeWy/1TkqTwixKv8TK/r+OcMtm89yWzJhapdFTxleQ57Gx1zoM2GxmX0oW1wCRgAGySEAg8QQgAHeTIi/2hyXXf0q12hoe/i6xYhGslampWYEmKLycnLjI1NJZyr0gs3IxuaumWoVpdRNgKVhFErSs8x7tHrsoBmrr6WyGrWaV0UfWmRy5KGvu13b2mv2e5AArOZUTGa089p58+4GP967pmbzQwjAUBzdBFiadM/Ds4xHtm3ae4V/vHAZZ3Z2engCeSjNef3i2u3PIgDDX1BUZkhUDqk7jMeOWtVsLoblCIk37NKusQ03ueXTDM1yvDrCfuGaQb3so4U6RuJx2Mldy8d2bdFr+dlH2eRRuvirXUsZVsvHOw4fYTn7WIiKtM80w1KPtzUcteWUfyqnO29CIVwDLMRXOuXlpdWj+3XvNHzu0VyyHOG4cOWkkedCea1KuOZA8XhRu6XmNk+FQwO61y4EARgkhwAMEkMABngjEhXV6S67+v/SsI2tB8/Hrx1cU9Zu9rl40kFTlYqJV0gGFiZS1WT4nNowSheA9dlVmDWFV/gfn9u/+fgt1xIYniP9yMiTVmu33nyZxgYcaVRtzQGPFIbXMnkxbhvNN1xIz01+uMuyRjMEYHgNiqJpKt3v6taNa1ct3GNrcdo7KU/JsqyadLiVYZfXLK6DAAzFoNWcWq2Sp9/dWrPaRn0ADr98aFKTY8/VJAALk0PH3lq3ZcEghyiS6BiVKjfGyX7fBlvrLVPnXb7wJJlSl6JJCV+PBODYq6MsbJdei+R5jqYYln95qoXJvrMBafoALKJols1PC7lzyNpmveWsvXu3303iY50sNs0efyteCLw0x2ue7eprtWmzVwYCMJRICMAgMQRggDcQbvuT7Xd0WtMvZBXq9hhnOrFf3SpfyyrXbDNt753QTLW6lHTkKVb7lwCsUNIakn4Dj8/v16j97ONeyWpeo9ZQIec2b95/J044uzD8WOMaG08GKHheHnd/6+qdN9NIofLZwQ01W5yKIIEaARj+SrhLmIZjhbRCZN1f1abhwiuhuRTLUAyHAAxvQkIdk184AGeHX94xpsGa29mMOALMhh03W9mjz4U4sXKpWf2dkAI3tu9kOu9SFF0m7s5GAnDi1RH9Vkw/ECjM4UyzKrnbhgajNp02BGCFeB2w+AGI51vwKc4bTDsNtIuKdl2y1HjY8UAtrxZGgHPvz267eO5GzxwEYCiREIBBYgjAAG9Ac7w27vbyQRWEua8K+3agzZ144Ri9fsWS7W8BmHSQ2DTnnZM7duxle9kvg2RhFa3SsgHbejZv1Kh55z69e3RvVffrL2o06LD8ku9z+xlVf2rcpUePnj06GdWp/sU3jYfNtXuRR9I/YgzoURTDqCKc5kw5d9Uvk9GqOY7KCT4z9afRh59lyVmGRgCGNysSgIVrgJVxrlvn/dbhVJhGpSItVKrH+rWzu+z1S/U4vHen9d0ErVZNszyfeWdY7YUme3xyea1K/0qlmHDu8v0l7UxmrXiYyvMalZqPPNe1/+oD9+MorZoRzq9QU1mRrmeNZ9wKpWhGOLcny2vf2nHNVvqk+KyfOLPT7AdyEo5JcfDxXjOXLbkRoyZ5WP/iryAAg+QQgEFiCMAAbyDMOZKbHOjpcoW46ux8dsvAZj/JavWcd8ojNkvOMKVkGFQXgE+uH9loIQnAak7D5/gemjemXa+ND3TjKVqthuNUyiQ/t5vnTp04dvTY0aNWxr9UGjLZytE/NSX0yZWLp44dO3H66M7l4wZV/n3KLqdniQqcBA2FCLNAZ3qt7tF0/N5HqaR7rY11WtWuzuRDL1KUJP/qR4AtF9VudwYBGIqhD8Bb/vzD5kJYNq3VaOjYhyemdO22yVeY1zj1ho15jz67o9WqsBMm3caMtAkiKY9X+q0Y3qfLMruXeVzZqFS0ms/32dW34/xVDsLk1+rAHT27LT75WLhARaXWqFWsWkUleJwc36rVavdMsgIX67xyfOuu1g8UnPLJrvkjRs5wFKb2V71YOdjEdOXttGLzLwIwSA8BGCSGAAzwRgqFeAmtHvNyYa/fZEbTTseypeUaYIEYgJ8fXzu0zjyXTNL1oeMure3xwWefVmrWqXvHtq0EQ1bZeYl3rykQYlf3l/XH/PK0vEa4i6tOnte+tb81PhHO8xoGGQaKoGgVkxt5ZF2HNm1btG3bqnmtmRsuRWSqWVo4UiIE4FCHlQurtTwVigAMf6cLwK62v1VZfy40m+aEC33ZlMCL5uOqNyfVyahj9zWH3HMZtUpLRTofHderlVHbtm2bthsww+5BbL6aKyvtEcVoOPmz49a9Ohm1atu2xZ+jzU4/S6Q4lSLu5vphlg7P0hRaTp0ZcXPl5OpN27Rta9S+Q6dNp32zaTWrUee+dF45slc98rkYtekz6eztaPK8Yg9TIgCD5BCAQWIIwADviqLp/PQQX4/7z4LishVit760IG9dnpkQ6e8VmiqnaVqRGR38zNX1zu0bVx0dHa8KbnsFJeUahrTJfzNjPT0iYjPylYbtpGhlXlp0uIdXXBaFs5+hGDRJJ0lP7t4k1eqqs2tkrkbN6oMJqYHKnKSIsKfPE3JK1a4D/xmKUshTo596RCXmyIUmiFQnFZ2XEuAkVqeH3jEK8TpfWs0zaWE+rkLbde1uYHyOSiguC3NAixTiwaI4n8fiXnTDLymHfAysUp4V/dzVOzw5V04zLMsoM0KcrwmN9+2Hz9NY3QfAcmp5bIjHdaH4zsuoDIZXvWZubARgkBwCMEgMARjgnSkUwpmchEboWJSqHpc4dYpaK3aUxBtpvBrV1hMuMSu8TcKdkjgViS+vysiLMMLlZWTrAYqnu8OWqOhAL6l1LEdqD0uXmawC/zbSwAjzOxU0MEKLI1z/q6PWx1xSynIFCU7zl3arLCC7ijBtum77xDaYbKAwIaO4ICxRwrRgBQr2KKFhVxd8WlruDZ8LAjBIDgEYJIYADAAAAFBOIACD5BCAQWIIwAAAAADlBAIwSA4BGCSGAAwAAABQTiAAg+QQgEFiCMAAAAAA5QQCMEgOARgkhgAMAAAAUE4gAIPkEIBBYgjAAAAAAOUEAjBIDgEYJEbaQQUAAAAAlAMIwCA5BGCQGGkHaQAAAAAoBxCAQXIIwAAAAAAAAFAuIAADAAAAAABAuYAADAAAAAAAAOUCAjAAAAAAAACUCwjAAAAAAAAAUC4gAAMAAAAAAEC5gAAMAAAAAAAA5QICMAAAAAAAAJQLCMAAAAAAAABQLiAAQ0mmiXM7s3quwMr+YSKnL01/sG+9uVg67+DzHK1Q4nV80QKxhDAzM5szb62NY1TB+iAFbdyd3WvN1l72TVPpS/4BZeRZm3lj1554kPG/f4lc6s0z20hlWH/ucTytL+N5KsRx3coF210jlBp9yXujzfJ23GZmffRpmr7g/4lLuHPSUqziB275yPWF5Yj82Z6p48eOHTdjsc1Lhb4M4G/kvhvMp48dO3b8lNkuSfoyrTzq/lYTUn1E89fs8c/RPwLwOpw8xNlGX2fGjp0waeq02VZ7z0YkxjrvnqsvHDveePrMGdMPXPPJeu+/KP+pfI/9VtPI9o0zPvE8ubifYW3mk23GZAWTxTa30/VlRMzdY/PHCc8zt7sdU7Y+EihjEIChxKIiXQ+MrPG1TKdCnaVn3fN5XhG8Z1bHT/WFss+bD3YMy6ODDnfVFxT4ZfalZDS+UtI+XdP8O9n3004FK/Ul/wCb5ulsf/iae6j8f/4Stcl2U9uLdaGR+ZkQRl/KZzvNrPyprMHqu5nvv3okHJ1RT1al1/4g/fL/hzLq5p7Rf3wpbpDso9/77bwXS+kfKg80lP9tky5TJ1vb2q5fNmXk7KnXotT6hwAKSfOxnjCl1/SVpKbYLBxuPOywZzrpvaszXjos7D9+ha3OgdPXYnAMBd5GQ8V7n9fXmc1bV45qX0f2i8nRpMxU72v7dKWbtm6cWe2rn3+eeOBlvnAsvoxQRe6bs3zAeHOyhesnTOht5eibXvQ4tiLGedfojqYzyArLZw63mGATQDpnPBPsaG3Rt89cS1K8ZlafPtu3O/9LR38B/n0IwFBSyQP2jv9D9lPzOfvs7feadfxV9l3PJXfy1HGHO/z6YYU+83eePnd6Wa+KH8l+3+iWlB3v4XDenjhvf2DZ8NaVai6x/0t7Df817bNNXavLapg7RLJ8+v3FZtPGnvTOEg4kZz0/N3nWPDtfBc9lPzu71GTlavur146MGTN27FRru6e54pOpWMe9a8x2ODzNFmOqwn/T0lljBOM2PU4V19DSUS6bppnN3mV/es2YCWZLz4SwYnmBXJ89w+r8Urthyz9++9Zo3onwgjHgHFeL2lU+7bT5YVqCx/ZZs2Y5BYkVRZsbdG352FlLzwaTBcXLmzvFv0bek5Wdo/gTrgg+st5s7O5Dx1cvXzxm2K7bMWo+7uLS6ePEtSZNO+yre+PCwLPHybFjx4xZtHXrncPW3T6oP/xYqO4Rue9pm+m6l919+y9v923yX+4f+ausauvZ++3td89qUVlWfdrRwP/hyEIpRcXdWdG1w0J33YccfXn9uI47A3CAC/4m6WLvb/rOu5AoLnAxJyY3tXBN1Krzg+8c7HwoViwF+F8kOG7p2s/ELvLV6USCnFOLR3Q3s/PJ0i+XCQlO8/oPWXwyQFxgbs9vOeeYa7zhIDLZsbKeHx/9W5slL8SR4fjLK0dVm+im4BXea027dpxzUfcJUZeXjJi7+kxM0c8LoMRAAIYSSh1zcnJj2c/9d0UIS4EL+vz0QdfF9+SaYNsOv8hqrn0o5AffZfUqf1B17d20V4deE50tOrUauws9HcnpA7DFpSgVH7mv6rcy2ZTz8ULYjL84T/b5dxOuZPGqhMsWdWQyWaUfKn4kjmx+XG3MFnchb+a6z271rayR6YE4LU8F2E7v/qH4uKB2p0FXgkj0Y57v7EIWv6ggnCNQecixhCJnaeU9Xj+w4gdGi+zsrUfV+bDODPtQ/XipGIA/6bT1SXr6M8umn37edG+AMJSY/3RnT/JGRp9PUcfdte1WUfxTogq9rS+S7rTi/ri2lWVffy2ekdBw8ZUHzvP6/CKuIKrSwfppNnkdRdDlOXUKnl29Ru0fvmk54VgYeSDv+anFHSroH5B9/FuTGee9Ut55EFMTe9yslaz6mCPi3hCyr9+vsm4W55LL0JDDG6nT7+1sWtX6gXgAhXzILy5u69V9c+A/O4YA5UFetMvt6Lg8XUVRJd1dUWXc2XCGygo7N7PN+Ok9unXr1n2smY1f+Tl2BP8KJubajgntRh6OKXpgPe3WrBHDZ+5+Kgx/lh2U2/RWZnM2ir+MRL7r+A5zd7kkvErAGio9yutWUJ74A8T4HFndrfE6X5ZnU/z8fb0SCz4iv+VtV06x9UEAhhIKARhKKI0yzt/9lndQJunKZLtv7Pb9N+3ML6doeUXcuTmNa9dq3L57z+7Nfv2607zzofreDunvJF5fWb/ab4vvl6eTQ0uqwgE42q7+HxVl866IP42JV5d9XaXaDKds0j91Xm1EYmftMVuvu1zdNrXDJ7Kag1c9JyspvBb1qCbrMO90uibTwbTWB7KWU7c4uBC7pn76oey3mddSaWXgkYGyj776vuuy866PvGKLXhOb+3TdoB8+aDLjRiYdsXfwb7KaZmdCdL/DugDcYbN7Hqf22971x4rNj4ZqNXL/3b2/+q7WxOvZvCYv7sW9a8537t67//Dy5nGNZN/2mH9PwWueTu9fXSZrMmb5aSeXyBR5duxD19u3bt91e3D/1pbJX3z9zR/7Alg64d7K3z/7sOKfy0663Lxxcu7AOrLPjExOxfB86lGz1p99UH/MhnPCVuw1rvGN7KeZx8TTxt5Frof1yMoftJ5+JV5Yirkwqrnst/HbXpatbtfrsSl3dtYYcTZAoc81cc5HVjeycMNuDm+i8T3bvf4cuyiKo8Idl/9WZ+TGQ+fOnTu2aXZv0/GWvnn6tQDeKttr17r23ay9i8ZfVeD6YabjLO8Ihz7LksyboyZtNL8Yp1/UvjjUZPyeCxF//7nJenFiaofmLdsOsb79t5ibfn3+hPFme3zl5eUwLZQ6CMBQ0uU+PTi51jcNe869F6sQTnpUekxs/4duHI2ouvmhOA0Woc1+YTeu9s9j9uHqwJJA+2xz8QE46doKQwC+urSB7OcWKx8JX2H2fcuOn//ad64nQ75kMQB3Mr8QF+Q4vO6nVYYeCNSnnchNg377usqaR+nZLw/1/azir2PtU3QPFJbjbtu/8gctZp0XHgva1a/6B3XnnIoQf6J1Abi97SPya8682NH+px86HPfNeHaw108/Nlj/4NV1gSqvU2t7dzGqXvHDCgMWO+cIAbhbVVnTxboIWkDrf2D8sJ5GNT/+uvKPB3zy055sayv7rv70U7oeUfRJs4ayBuPOZXCpJ8Y1+bRa3zU++q2IsOte47O6Zpei3/HoeOa9dYM//7Dl9IvRwlLk2VHNZN+P3ORV1jper8Om3N9dZ+6lSKUuAHNJd89sbrb0AYYW4PUCzw9r12D01hDhtHmtMjnKwztWN+6rjbmwZPyA0Xbi2RQAb5f5aNfyNo22hRXNcqzf0m6mpkvvlLnTCdJvjVm0baWz7joCIsS+penhi5F/C8BaOjPs4bUbV09Y7l4/77Rw9VCB7Ls2Y7u0n3HuuW6QGKAkQgCGEi3L8/CUOt/X7zH9VrxuiC/z0Zz+v3zZc6HdfR8fn3MrqlWsXWPL3VQx8cq99/Wv8nPt/boLV+C/R4VtXzCkw7y9upFJ3y3dqsuqmzuIAbgeCcDzrwmX72pDjo746qffZzjlkADsuKSBrHqbLT7CkY0sD5uB3/zWd56XIQB3ND+fGe7Wr/G3Nc3PJ+iPamRdN61e5ZsVrinZvof6fvXTH3OdM3QPFJL7eOuEajLZtz/WatK8uVGj6t99IZN9b3L4kXCZVp4uAG98KEwCq3yxqeuvVZtu2Lulz0+/NdgsXtGkjLxu1a/l6Nn7Hvm7Xd08tuknX/a2uJUrBODOVWU91lzT5+0op4mDerTceP7hw0f3tpl++f2PP+9/npt836bxx1Wbr3qquzw12X5pR1n9UefStFlXJreUNZm0peDU/GzXAUaVfpthH/GOM/EwQTsn15EZmTqIB+Wj7Yc3lhnNOBBVXs4BVqW4bK/RbL9Hnq4SUKGXdkxpuuQxAjC8RtCFoW1rDrG+ElfowkUDLvz47lWtLL30iwBvlv1g5+rWTXeGF5xqpkP7rukxY/zy+2VvQv5cl7E9VszcH66fZYF5ZNl4/LZLUa/9sZI/2bGwb5ctQfr1s+9tnNCt9dAdj+Ix0RyUZAjAUHJlPTs6tc53tTubOEWLt6zQanltwNaOv8tkw/Y/Ezo2Ecfr//S5rM8eP2FcTRmwt1+VKrWtvIvr8sB/Qe47v3cV2bf9d7jnarW5LqObV5L1sn2QoeUjDtf59TtZn8NBCjUbc2LE159VrDLrZp4+AP/ReqOXEGwyHq/v93WRANxh/tksOtamR7WPv5907HmOlgg8ZVTt+0+77g/Ny3t5oM+XP/4+U0zVhanjzy/r/82ri4b1vjLe6ZZLwtM9fQDOEg5NK3x29/taVrFCpQq/NN4cIIxP08GnZ/0qq9BhjRd5POHIrGayT3stdckTAnCXqrIuqxwTxUPaMWc6Vvlc1utAIHlO3kXjL76v+MM+Pzov0H6C7MMfjAa7pGg1ipCTc+t/8XGziacTeG3omhF/fvLTkN3uGcJWBNuNqPvph/2sH6TpuwxvlXZ7Tc/PP+80/0amVpviONvo06+G29wvLwPAvCbX98TIatMvROsSf8b9E+tazrmWjcEFKEb+483D2o6Yc+JBesHepQy/dqjXoPPCtfiCbPft2+b1PV9whifAG2V6bl9tZLQ3pGj+ZZ6vGjhr3MJHZTDkaXw39Zo4e9UD/cReUac6jNl65kXOqx8rOs7FrveAswXzEOY/27d6bMvVwhElTYLdisFtB5jeCP5/3MEQ4D+BAAwllDbLe5dpLTG5VKxeq1aNatWqTbK9mSV/ublvzS++qPDTb6SgSoWPP/uq2zb3DNIOqxOuzvmqwh/jrmDafclomJfHp7b9/vNvf/i1WrVfK3/y6cdDN91LJV9Ozo3pdSvIPvzu59/Jt/bjlzLZtz/PFAJw4mWL2rKqRhs8xQD8aG2vz37uOUc8BdpzIYmbbWYfS+K5uPMDG1T+9NsfyWuSJ3//wRffLbwcy2lZn93dPq5QZZrjX06BVkUcm2P0gazZ+F2Pwgs4ren4h+yTQdaPs7Xcw0XVK8tab3iQKY4mUrGXZ31MKtl3Dba/FAsUfifn15Z9/Fml36rXqPnHjxU+k1XsYyFcGuwxpcMPsg7LLyeIsSvsZPPa38i++/H3atVr1qj6zUeff1N5i6dKyyTeWNz6ow9k31etVr161YpfyWRfGI0/Gk6qZ8hZsw4VP/um8i/iVnz3yaf1Zp30zX3VpXgLTbLr6iG/fPZlRfL8qt9//smvI7fef+f0XPpp5LHnLRu3PeCWI5dnBV+1ndpx/u1yE//hH+BCzvdv3GfsNtfYfBVLK+VyhULFcRlehyf/2GL5rUS5XJ7kedjStN8GV/0F5QBvpghxsBpTc8HDv4z0xhwbazxz8qkil8SUFaqES2O6LjI/GkD2l6z76zsNt3YMzNPyWjWjZFSclrTHES6rBjZfdD2arJD5wt5sXNtBJ8Nobc69TaP7tex35FmsQqWiFXK5khZWByiREIChhMp5dnh6k6+/+OrrLz//7KMPPxD0W3wuXstTGS4rO/z5pVD00WfdrK9HiB0Zbb6/3aDKNRotcEa/WEpM0t05Axt8TL6dDz9uPH3/8zSV8POnpaPPT6n906cffVih0rwd282q1G20wCWHVyXfWNXy63pdtj0Tsmemx6ZhP9UZulgY3Fc+WzGw3tc9Fp1KEI5tZD4/3qXeT2Il+KTTRic/hbC60vfAoMrV687/y40GucSzS/t+/3W7+RcLzwQesHVwoy8qTTjklZ//fK1RzQo9tj3JFvMur0q5Yvb1L7Ua7Hqp6w9rVWmeu6c1/uLTjz6o32ex/aX141r+MnjtMzXzYmH/Wl/3XXcjSYydasWTg0a1K33y0Qff9V53xXHrwFp12h30Iduqzni02eTLTz784PuWrafOmNqrYhvjUyQA8zyb9/LQpPaVxa34qN+KY/5y5p8EWK0qP9xhXv9fhE/2k06zd77MVZevjgWd7mY7vm6D+vXrGbXrY/cgv+h0NAAC2te2V+3vKv9Wt2HjhqSu1K/ftFWHEz4UzynTHuyf93sdUlKnbc9xDj5l6aat8F6lPz24wajTub9cMk4/Xzly+YQ1XmV0Jj51+sODo3o0JvtLvWrTtt6Lp8jPI5fqurbviksv84WfSnX2SzuLP4Qdqm6broPPeArXoSVdtxr+XaWKVes3adpQ2Pnq1++19LSvePoeQMmDAAwllIaR56SlpBaWnU/rQgqTm5mmK8o1dIM1KkVmWnpG3j9KFfAeKPOzdF9ONlXou9AoMtNJWXoGo+Ho9AzdF0W+5MzUjCyF+C2SbzAnLSMnX4jMJO5lZ6Rm5Qs/u+JjuVkZuhfN1eVWYRVldjFfuFYtz81MTs2Rs4V7uBo6Nys1KYtkTo1KTt5IloIteJqWpci7ylDp/xJBXlncgmw5SwKyIis5M5fRatX52emk6NWfU2fptkhO3pE6LyMjJV9/TapaKTw7PTeXopR5qZk5VMFb5pS5wjOI/CLv7l1p6Xzdp5BLlcv4x2bGxxDxyem4ARIUS0vnpKWnJCXGxQo1hYiNi8+jdTubOi9OLEpM0x/8AngHHKuQp2UZfjD0tGxutjxbPBZbRtHZaWJ7G5ej/7XScorMhIx8RqP/8eLy48XdLDE1S/cxqBW5WUlJyYm6/UyQkI4eGZRYCMAAAAAAAABQLiAAAwAAAAAAQLmAAAwAAAAAAADlAgIwAAAAAAAAlAsIwAAAAAAAAFAuIAADAAAAAABAuYAADAAAAAAAAOUCAjAAAAAAAACUCwjAAAAliBoAAKDs0v/aAUgHARgAoKTQarUKAACAsov80ul/8wAkggAMAFBSkG6BEgAAoOxCAAbJIQADAJQUCMAAAFC2IQCD5BCAAQBKCgRgAAAo2xCAQXIIwAAAJQUCMAAAlG0IwCA5BGAAgJICARgAAMo2BGCQHAIwAEBJgQAMAABlGwIwSA4BGACgpHiHAEwR+n+WFWR7im6SsI0F9EWvlLnNh/+Ivkah/sA/o683f6k5+rIyW530m/eGDdQ//prP5Y0fDAIwSA4BGACgpHhbAKZYjvQbtGqGUij0RaUezWo4NcfShi1i1IX7RmRb9eUERbNqDa+i9YsA74jUHGHXEQi7j74U4G0oRsXpK46WNbQ8FMNpdPVJy6noslefyFYXtMKaYvcXoSkuWKPQ50J2M42+kFO/4XNBAAbJIQADAJQUbwjAFK3meWXYoVXjepsf8kzmOFWp73VRFKXi+Ry/HYdObb4ZLm4RxWqU/ifNRnVt20bUvsOOO9HZFMfQFK3SaPMTva9vmnI9iKLpsjvyAv8yIaxQab4nZrUjNarLmDlngllejeoDb0WreFVOmLNtO9IctR06bfHTdF7DKJWslldEHNswXWik2i044BRHi8VlBs3xXOTlJZN7kO1r22791ZeZjLbIUUdarVZmBNrP6yCsMGDiwvuJwgdAkV+oTK/jawaLjffszfZRSpKN9U/5CwRgkBwCMABASfHaAEy6XPmRJ21M637/qeyLbstvxPA8V9rHQYX0q4g8vrzfNz/1nm0fImwRpdLKvaxnL7XaePT6zeuOjpevXPGNycxXMsLAAp0ZeGR8jdp//rbPm1GxZXDUBd4Dilap5Imee+Z3aL1oi9ONKweWT1zQf86DJK26LEUW+PdRrIbOCHCw7dB28oozN26c3DR78szJJwLVGo085KDF7HZDlhxyuuG0dX5P0/WHPVM0mtJ/RFKH4fgsz6XDzYeb7brk5GS/akr7OQddI3M4Tj8QTDEaJif66sHOLc2tHZ2cTu5YOGnKyKN+rFab5Xlobr+WPc12X3dyun5kxbBdh46G5mq54nY0BGCQHAIwAEBJUWwAptUaLslj39w23//x208/Vvq+ysA1t2K1pTsAM5wmP+y8zcRGRt07dajTf8rC86HCFrEcH3auxwZ7x3CF/hPhtWqWVSmyA+znNe3c/I9BXUx+a2z3nGERgOFdUIyay49y3T2y1uBdYSypT5qgE/PG/TnqujhmBfA6NKtl0zwPr6rz+/I7lNASpdzePL1P190hPOe/dZBJu3HnY8T26cmS7qPmbriTwpeNSzNotZYLOjpu4Bhz+0Bh+3j2vqXRpD0O/lkajZCAKVqlViS6HzNtOmCLv9hK5/vZrZ3ecYu/NuvB6hmDuixzyhWfxwUcHTtrlfXtWL7o6LEOAjBIDgEYAKCkKD4AczwbaG+zZsaK25635/b+48vuy0t9AGY1moTnbo7HDj6NeLBn5LQZM04Fky1i1Xzuww3TZwyr2bxt40YNG9Rfezkgk9ZyTE5m8Jn9R70eubjsXVe90WFvCgEY3g1FUfKc1PjQl3G5SoZhc4IuWi7o2PVQIFNmRuzg/aBoZX5GfKxPYEY+w6joBLddy0x6LbmXw6uzokJig2JzKYph8333DR01d82lEIYvdqSz1GG1fPqNCa3nTbPxzSeZnlFp052nGk3a5BCSrRHGgGlGy2QEnFlfu8ZW91zx+mhtyr09lkN67QoMd1m0ZupoxzjyO6bS8Hza7eGdFk7e8UxR3HnQCMAgOQRgAICSotgATPrwyty01PQshle8WDLoj8+6LSv1AZjEkrycfDlF+kmRp8ZOmjxdDMBqDR99ZXjHYWPXnn3y4oXPzc3TWzZfdzEonVLTisx8sm7Ws8OrqjU8hAAM/wTFqMnOlR95Z9uQejWbtBu272Ee+6YZegBEFM2SVodNfXraokHtGm16r70ao+RYmlEJE2DlvLCZP6R+jbqj11wKyldp2LJRn0gAjnMcMXXDwgvhGuGiFIZVex9qOm7naf9UTthGEoAV6S8dNtYwORcmXpui4XP8T2xf1WqNF5XiuHbo4IFz78Tk5eXl+p+d1uqzjjN3++chAEOJhAAMAFBSFB+ACdIPIT0TPuPJwgF/fF4GAjBB0TRNtkkTeGzMJFNdAKYpWp6XFJOUkkcLU4lq8v33j6038YBbYr6a4Hg6weMAAjD8UxTNcByrzE2PDw8N83S0sjIbs91drlGXhVNW4T0iFUfFqWh5ZnJscJif2975VgOnXovR8BpWpWYV2bGRYSHhXucn9l6z8mhgnoZUsdJPDMAjZ1tbXIoQW2SG1Ty3az5h9xn/tFcB2O/yxj+nXYjI0wfgwFO71rRcfp/h2cQnm6b3+aN2vXr1elisXTOru5XN9mfZCMBQIiEAAwCUFK8NwAKWBGD3shOABTT3lwCcn52SocynGIamyH9YymN7jYkHXWKz1SqGUSEAwz9HKeV5melxiaS3rru1S+od6wWDO6z3osvELgTvC0Up8rMzUhKyFDTLCqcQsEE7ZsxoNMwxgcnLSk/JpVjhoBzPJ5zr22TO1KMBcpKL9U8txUgATr0+tv3CqVteynmtimG5zFvzmoy3PR+cWRCA6Qz/M+tr1Nv1NFcpngKd+nCv5aj2lu7CFcFqeW5qRGhISGRybrDD6NnrljuEcnwxR5oQgEFyCMAAACVF+Q7AvEabeHV0i3kWx0PEWWfU8sDD0xrZOr5IV2hYmkYAhn9MvGTx5ZnNLVpteqQQTivg+djrljOH9rB9ThfXMQcQCZe35sbc2je13fRTCUL85flcn61jZ3eYctf3yYZ5kybs89aVagNODKg/f87J4LIRgGk1r/LdNXj4pEWOEeL2cV7bWo7fftYnneM5TqNWqdTq/Nj7B8bWHXskTCmsoAg+uXFhu5Vu+Vo6KyExJUUuFJLd7MrklnMn7X6ZX+zHggAMkkMABgAoKd4agB/P6fmjrJ2FU0zZCcABh4eOnTDpWBDZIjXHRVwe22n4sPUXngQG+LnsmNlq2r6HsTkqlTD7CgnA8U/2LP3pz32eCMDwjihGrcoKv71lotHMXXf9AwIe2a8dNrrHlBtRwknRAK9FqzRUvMchy6atbC/6BQQ8d9o606TNmH1+nDrfc/3E+b0mH38SEBBwz25Y5zGmu91SOE2ZmAWa/M5oNEm3p/VeYLz28rOAAPdD89sYb70WlKnmFKmRfhFJmQpGy6QHn7E1arn+pG9AwIML1nNM+m5+SvF80p1Nq6cMtXIgH8vjY2O7DRpudT2RK24OaARgKAEQgAEASoo3BmBGo8nysZnRtZHx1ocJmjJxBaNwh6ewSwuXLV/tGCFuEc1y6tAbkwd2rt+gQYP6dS0dAjNpfQ+KZjV0so/9jm4DzvuJ50iLrwDwNrRKS2X4nV9ZV6xUXYZMvxmuwhzQ8FYUq1HnRt3eMbIeqTgNmvebcMg7R8MxDAl1kbetpnQQ61M3832PktRl6rbSZAPlviemDm4hNsLmx9yTqP9r7z7AorgWNgDvn5trEnNzk5tibhJzNdbYY++9YcNewY6iiA0LKiIqqNgVCxbELioqTSxgR0VAQJpI7x2k7c7O9v/MzrAuVWOF2e99fB6Zs7OzO7tnz84358xZhZIuSnBfO8Tc6UlmkUyukBWmPtw3lXlZmvcYPsMtUiGTUKTtzg0/tGYq82o1bzZltX34S2UFl/+qIQDDJ4cADABQXVQZgHUDk21FwuKiwsLComKRWEwhqMC7opgZ10RcpRKKxDTzi6YAr6WuOUJ1xSkSqhsnpuaQQtIuvWqkyCK7Ol9QzH4Ki8gOsvvHtsKkWS4WCtkVKDFNXhf1Cszrwr4A6teFe7XUL12lLwsCMHxyCMAAANXFawMwOaYgx2D8OtxiDjFL7ZF6J9XK7ShFMcdavNp9+DjIETtXqfj18YEPjMl6JRXnVc3RbqR4WZ80O/hq/5jQ+2pntV8Xroh4dbcqP2YIwPDJIQADAFQX6AEGAAB+QwCGTw4BGACgukAABgAAfkMAhk8OARgAoLpAAAYAAH5DAIZPDgEYAKC6QAAGAAB+QwCGTw4BGACgukAABgAAfkMAhk8OARgAoLpAAAYAAH5DAIZPDgEYAKC6IIcFQgAAAP5CAIZPDgEYAKC6IIcFUgAAAP5CAIZPDgEYAAAAAAAAdAICMAAAAAAAAOgEBGAAAAAAAADQCQjAAAAAAAAAoBMQgAEAAAAAAEAnIAADAAAAAACATkAABgAAAAAAAJ2AAAwAAAAAAAA6AQEYAKCGUMoZCiW3yCOKUjsll0nEDFrBFZQmlyt5+BLA+6ZQlq4nMpqpVBKpnFsGeCNsxaGlMm65NFLNuL/4RimXsntecTv8irKiFrl0mw5Q3SAAAwBUf0pRlMPysf/7vW7d37sa7/PJlPPl4EKR93i19T5LjwRuWSlPvDp7RK9mLVq0aN50vUt0UemDrwJ/hx199S/Fc4sAFaJeuK/u53A/U8wt0xn3d036k6lUvUfN80p43RE9gJpSnHhrz9imzUnN6T/e6F4qzZWryTLvLLA8uON2GrfML+Igx2nD2qrb4VXnQvIr+8jkBZ45smritRRukaHIe7Da+oClRzK3DFAdIQADAFR3isx7myb/7zOB4B+f//Oz//viX63mHfHN5W6rsRS0sCDVb5txzy9/GrLcjTtYkgTv1Gs3deXxx9HR0X5HF3We7eCb8eqgk3rhOP/ber+1dHjBFQCUppSJXuZGnjdr1eSbf6x2zZaqy16Gn9jcrueeq6RSPXXdsmTOyN3B6AaG15Km3Nk+pmWnhaeCSM0JuGKzZ9WMqylMElSIi/OTfSwM2v/zl7Gbb2eya/OJIuvatP4rTO1ukx0PPrm064z9dxIp7jaOUiYuTAlxXdHtP43a9LiSpC5T0MUFqY+3GXf+8qeRy93S1WUA1RMCMABAdZf3ZO/kP37qN88poaDg4YI+3ws6LHZKquGH8JLYs+vHNOgwWl+/g+GiZZcS2dIMl6H1py9zYfsTMs6Pb7LJLbKA7e2WFkTd2NfaeMKSRh1PR6tLAMoqumc1qHeLegtmzful3h5X9ckTEmPsJzWffSlbvYIs5vhW837WjysezwqgQaddt5vSZIELd65R8uLgZosh20OUKkXkoRXDGnaeMH5EqykrbG7yL+gpn+0aMmG+9T32I6MK2tt+qr1HjFB72JEk9/GesY369mg7edTKVv2uqE9g0nFOq8c06Dhaf0QHQ7Nll7R7hQGqGwRgAIDqTiGlivMLKFpSlBfnMmNIS71V55Jr+gF8blSY/737uUXRpwyN55lc4IZAy0L3j+tjtva0f3R0tP+VtYOWnA7IUvcAK4rDPfcOPe52a/+mph1PRKlXBigr/+4Nn7i4pGIf+z8bbndOY+pOcejpTQ3/OvqcXUGV/2CP1dR+W0O5RYBKUGnXjs5ovf1uSe6L3D3LpO14L6GqMDLI7+GjPGG4/YhZiy08+TcEOu/q2K7mS4/GcuOei7yXtJq+73pqyQUFhKKwIOGGp3/eywi3zeua9bqkDsDZ0eGkTc8rinYwNDY1uYAh0FCdIQADANQQyji7RcOaDDTc4v08s1A9uLMGU3IHV0nnphrN1QRgJZ0W5XN8+MAeLVq0aLJin3tQnki9njjV54jZunvpWU8O2TRufxwBGCqmZKfekT888GeDbWwALgpx2d50kWcOU05IXjju3dTTOoBbBKiMLPnerukdOtu4hISEhD+7uXtm45YDJnjlqEqmfIo+oj99IR8DcLrbKNNd1l4Z3KIq5HibmQ4eySJukUFeAvWrkHFt85qSAKxp009MNTJBAIbqDQEYAKCakxVlJoRHJGaJFAoJnee/Wq/pf4ZZeWfy4zLGuDOGrwKwLPbU0F4Gi86FFojF4rTb1n1bLT4e8lIqy/M7u87Wt5gEm9Bjto07OeHQCqqioB/sfxWAn13a2sjIveRKTfHzozvX99jozy0CVKE4/NL65i1btWrV1WD+xh1rD8w30JxJUameHxrBzwCc4aY/e7uVZseUgUdaTjvsnqQdgEskX92kCcAl4hwNEYChukMABgCo5gp8d076UdB0ioN62Gac46jGgp8nbQ8qVN9Y05UKwBkX+zSzWHWx5MDrpafpH4YHvcIeHR45bO3Zm/fv3fXYv3jGr01WXQpMKOR6GwDKKRWAhWFnNjVrsj+MvUn18tae9WOG7sEgAnhTSgXzmz5U6L6dS/vsDOYKCd4G4MJrk9uZmR2I5trY/Osz2889cCtdwi6WggAMNRQCMABAdZf/xG7Uj4Ifu0w5++iR24ah9QRfDF59OZMfCbBUAM67MaXZ1PmHuXBCB28b1cH0jG+oh/Xgvl06d+rYsWPbZvV+++LrJkNnHw6r6WPA4cPRBGDmqkVZ1pMTRp1G7WerGB1qbzV/8L4I9QJAFRSi7MSwJ4nsgGehv/3oGd0X+uarl9R4G4BV0SenDDe08mDndpbdsWg798StZPVkDGUhAEMNhQAMAFDtFYWdter/Yy0B64c/Fh+/lcVdh1bTxTiOnmI443Qcu5TnZzFh1rAlu087OzsfWTKqQx/7u+pevBJFT+zW/tT0WAy3CFARBX1n568/bDybwg7aFCZedezffbktqVSO20xnLZh1IVZdDlAVaZrvucXdZ6w8SSqO/dIJI/vNv1BqZuOwvX3HzTZz5eNsx8VPV49bMXmZHdnzkytn9VrpFJAlUcmFyU+vByTmy7S+exJdLJf+1vYcN4s/K8Z+9JQZM06XKgOoZhCAAQBqAur5OdsZwxhrT93J4wp5IOPuTrv9B32yuEWScZ+uM57Qh+jb39wzrsxlZ8IX15zmLLnNwz4XeI+U0vCr84wuP8zTDNosinY268fUqokLNgYVc4UAryFNum/fX90aGSxeH16mNUq+vnH34VMBPGqNtaW5r5kzhNn1/nb3k9XjbSRpt7YbbL0ZK9YafJTjf+HU/FWPXl0YTWR477Q7eNCnVBlANYMADAAAAAAAADoBARgAAAAAAAB0AgIwAAAAAAAA6AQEYAAAAAAAANAJCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIBOQAAGAAAAAAAAnYAADAAAAAAAADoBARgAoLpQKpU0AAAAf5FvOu47D+ATQQAGAKguyGGBEAAAgL8QgOGTQwAGAKguyGGBCAAAgL8QgOGTQwAGAKguEIABAIDfEIDhk0MABgCoLhCAAQCA3z5aAM7JF7rdidjmeG+W5UXyr82Y3fhXzf+x75T1oVsXb4SkZRVyb+QHgAAMAFBdIAADAAC/fYQAHBCeMn/jlX4zD5vv9Dhy4eEd3+fkXx5Ue+w7dfzyY0u7a3rGDiNNT9zxi+Xe1PcKARgAoLpAAAYAAH77oAE4Latw8Ra3GRbnr98P5UIV1FjBEfGmNldmWV6MT83j3uD3BAEYAKC6QAAGAAB++3ABOCQq3WDlOffbz7j8BLxwx/c5eVtvP4nh3ub3AQEYAKC6eH0ApmipXKmUS2mKK+AXSiyTK8iroEUmEVNiiaJ0qUIuEXN3AagA80EhNYdbUi9zNQdVB94QaXhk6lqjIZdJtSqVSERWkElLF/EDJSlpiaXaH6NXSj5Q5BMl0/o2oiTMF5S6nGm5udJyyO3cd9575Xo7fIbF+ZiEVC42AY8kpWSY2lwhbzH3Zr8zBGAAgOqCHBZwBwgVomgJlZ+RFPUiPjWnUEjx7aiLosSF2TGRYUGc4JDwyBfh8Rl5Bdmpoc+CX5VGvIiJTsnl3wsA74mYlkqKMuOz8oqEIlJJyOdGQhekhz5jqk9ETGYROWTn1gSoBEUJi3JTothWhwgODQuPjIlLyCmiuJaH1CtRbmp6enqepogfxBI5lZMYHcrsdkjyy2IxXfqEK/OBkhSmhTBtcnB4VHw+LVevIJZIqJz0OHVbHRyRlF1Aa2djbR8iAIdEpZP0m5GVzQUm4COzrW7XHkRyb/m7QQAGAKguqgzAlFSlLH56zKBD7Vq9TR2fF6iUEu4WnqDlihRv0yn9m6u1aNWy8S9f1vrF0OFJ3APHnh3bMIUtW7Vs1vDXH2o3GmwbUCjnY88LvBuKlpMPUmHctT1TGtu6xuWK5ORYXZod6bp1yn//aNGyRZv+I7Z4hebRfB1DAe8HJZbKitIfHpmpboxIa9Sqef3ffvqi6YhF97NUCub8CWmP5Xk3N/QbaWJxM1GpkvFmXAElk9NxHrtH9mnVrEXLP/83bb1bZLZUKtF8XsRiiSw3/sbuBb83adGiRdOOPfWPeOdTUrFYrhQneNnM6FOvacuWLZt2HLXE6WGmqOIO5PcegNOyCg1WnkPfL+9lZGXPsDgfEpXOvfHvAAEYAKC6qCIAi2UqlSjowKT2tQUCQT/TI+H8C8CM4sLCgvyCgsIiUXGu37ZRzZquPR+RK5ZRRQX5+QVFIrEw4cGh8T1a6196riCHZMgwoI2iKGFObMDjy1YG9f/46stVF5MLJCollfbw0Pgf2g4+/ZSmJXmRF+ys9CydM6VspxVA5YTFhfmk4SkU0eLU+3ZThjUbdSiEUjBZl7TH0kzfY2NbflZ3gtWdFP4EYEqqFEeeHN1r8ZKjIYUSSfa9jT2HW57zT5eqpOrPi1iqEKY+PmT4fRsz11iRRJIRcGrdzH5LPLJUKmn4iTW9hoxY7ZMhoemU2+aTtiw/HlakkFcw3OK9B+DFW9xw3a+OiElI1TN2oKXkkOidIAADAFQXlQRgoYiWqeRZD3ZP6PPr9//+9mtBf1OHCH4GYEpMk5QiU6jE8YfGd25tcz68QKKUiClSKlWpFC9DL8zs13nklVgSf9H9C6VQzMjMdNcV/fu2a7V2xYrfG+w4H/9SppJk+h5Z92u3ExEyqZzpHU6/vXX77P7HIxUKKXdHgIpRTHskUahUhc/OLxnTcuKRcKk66VIkDibftt++Z8HQ1gar1l6P500AFstV1KP1fQ2MV3mnkA+LQq56frz7mE3HHqZJlFIm+Ivl4rwo98NDRp0NKmbOyqqKo0/uWdVmlY9c+GjttBm9Fj+kma8yBkm+QiG33TLebwB+GJQw18qZi0egA6ztb552D+Te/reFAAwAUF1UGICFFK1QyaKum3ZsUmfO4j0z9dsIOs/mawBmULSUzr+xpmP7TSdC8ko6HkS0TF4c5bV1ZvNBl6KkVc2vAjqKYmTcfxCW+7KAumfX5A+bM7F5JADnBl3Z3G7EFv9C9kMWf2HZyIG9rJ9QKjlPOu3gQxLLlXSs6+7ZQ4btCZcwOVcokqpUOS4HNm896XZ9/6Tpi1Z68icAS5SqNNfxA8yN7EJF5BtGLJEWeK9tOdH2THi2XFEyDJoS00wTLKIk5Dsr2Xvn1IGjDsXkh+2dZ9Fn4Y2IF48vXbhw4dLDiDSRSiWjKszA7zcAG5o7+QS84LIR6ICYhNR+Mw9zb//bQgAGAKguKgrAlFihpCOvLejZ6vfZR9LzE6zHNBF0M3FKIgdh0srOr9doFC2XJJwb/9fMvZ7RxUoFO1CVksglOREXrPq1XXsnT0rTyL9QIYqWSGVSKt9rtzoA59JKpSQ/5uqeXu30F9idO+/kfHb98lFN+vTb7IcADK9HSRTKgpDTFlPrT7iawVz8KySpT5l/193U8vC+6MK4/fqTF/IsACe6jl+yZaVrjIp8QMjHSRl8ov20/WfDsl4FYPZcEy1XqugEjxUTJ/aZ+ThPVRSyf+2i7rMtzS2Me/bs2b3VkLGmLg9TqA9+DXBaVqGesQMXjEBnzLA4HxDOjFJ4awjAAADVRfkATIllSnHSjdV6vwr+GLDqhMflnSP++lnQeMACR5+43GI+BkESdWVxZya1M9p2LbpIyV0/Jmbyb7jruj5t193NpJF/oXKUWEILX5YEYLGclsgVVHbYJbP+fXr37jN1pelR6xPWvbb7SZQy1CKomjr/PjtpbVB/wrUMlYxmTs+J8x+4W9leOswcfMccHDljhdWDLJVKIebF6UgSgJNcxk7ZYHr6hYzkfRKA6cd2rSbuPBWa+SoAC5nXRaWSJl5bOaFd6zGrfDPJS/Ey9NDiof/uY3IwmmK+y3LclrQdNNniRppMVcHn7D0G4Is3QiztrnGpCHTG/jP39555yFWCt4IADABQXZQPwGK5SpV003zYN4Iy/jV8k3cyc46eW5EvKFomiT85udO8nS4xRcxVZAyxREbnhDqv79fR6n6mSIz8C5UrG4DFFHMNwSvpLptXj5jgks7O5QtQOZLzFHkhjlaTGxp6ZqmkNNMeS1+cnj1neOO5m+z32+1eNqp5+14D5p14HJMnktA8aIxpperltdkd5hlvCMhXkcQvVWW4j+9ustM9plBRMm0cJZHK6OII59WTh3WduppJv1KShwuf7jffNN4qhHzEaKFEpSp6OLX9MoONjwvIzeq7aXuPAXjlTk/n60+5VAQ6wyfghaG5E1cJ3goCMABAdVFBD7BEJs0Jd7VfPp2YMWvWlOHNf/tG8FOTHvMP3Y7Ok8pe/TgFP5D8q0p3n9zQyMYprKAkopBC6cv4Kzt6/77aIwv5F6pUOgDTCpk4LzHg7sHjUSLmIyYOODpvWK8ZTrEybnA9QKVouarwhbvVovr6V+JJ/qVEtExWGOdxfN34SWNG6o8aPbJPy5/rNmzcde2ZR2lCOQnINR7T5Z3kunDQsCX2T9Sz7BZentvB+NDVqCLy9UQWyU5Kabow+taqiSPHrLyew6zCyQrYtdZgouXFl+xi4K6e3QxMTkXRFZ2lfY8BeJblxTu+z7lUBDojIipJz9iBqwRvBQEYAKC6KB+ARSKhkJK8OliQPlvQ6wdB8+knE8nRiYx/1wCLpUrxw031Wy7d6k32kLuujpYqitOe7J79dedDT4opCY2fP4LKqQNwnueWn76zOBadQ6sUkvy460eGdZliaGFlZWY4ouuE2TuiClXMlLYAVRFLVZKke3vMWta3eSBUSJl2hzS52uMJnu/qO9Zk+e0M0nbT/GiNhWSnqdiDpisHj52zap3V8tETB6939s8Uyensp87bnR7HFdPFL66YtBAImk1aZb1+7ZrVq1at2nLgeOhLlSLD64j98GGLFltZWa2dZTB4yFZ7nyxaXtEn7T0GYJKCSBbiUhHokjZjdnOV4K0gAAMAVBcVBWAtYpksP+aS/cYlO889Si2S8vEQnqKlBeFXtp6++iThpZTrUWF+iqQo+8UdR+tLgRlCSoz8C1WgKLFYWBhxZ9PGG34ZBZRETMtIBk68azvD0MBg6ty1u26nq1QyER/nj4P3i7RGosywm65WO/xTpeUGOFMSGR1/69AJZ9fQLNJW8aZVEopkKlX2/YMbjZkPzMzjvgnFCqWcKkp9cGTZQe/nLwsLY2/tM1mycN6saYacRVY7nqRL5eQbLOPBhc1TDQwMDKdvP/84R1nZeYH3GIBJCuLyEOgYBGAAAJ54TQBW9waza1bxE4s1nJCSqndQov37GUKRmJl1hS9TrcKHRklJdVHKxRQTdElFEktfHXArJXz97MB7pqk58opyHCliB+dIxUJe1SiyM1rd3Aop+RSR3WPmo1CpJ8IqdU09R8a8BuR+WsGWmTO7kpcFARjeHQIwAABPvC4AAwAA1GwIwPDuEIABAHgCARgAAPgNARjeHQIwAABPIAADAAC/1dwA7OnpWbt27aioKG75k/pwTyYrK0tPT+/o0aPccrWEAAwAwBMIwAAAwG+fMACbmJgMHTqUW/j7qs6cZOPcD/ULBLVq1fL29uZuqBKbNt/iWSEAc5XgrSAAAwBUFwjAAADAb/wLwOnp6XXr1tXesp2dHYnBlWVI7afBjwBc5oV9x9f5TSAAAwDwBAIwAADwG/8CMNls69atU1NTuWW1CgtZ7yUfIgBzleCtIAADAFQXCMAAAMBv1SEAp6enN2jQ4MiRI9999x07XJlNfewAZu1gGRQURNYhhWS1DRs2lM+cbPdv+bhIAiq7Ze3HYrevwT4ZzbPSPBZr5cqV6i29UtmTqezJV1ZOnhtbSJRP6ey99u3bR/aLrKA9nLv8BrVLiMGDB2svsvtV4dMo/xaoH+FNIQADAPAEAjAAAPBbNQnAJN2VCZCaEKtZjS2v8C4abCgtH4BJeZ06ddgAXOaOZTpIyywSmvDMLatV9mS0ywmytQrL7ezs2KyreWJseXnsHTVPgNyx6g2Sv6vYo8ruVdnr+YYQgAEA+EMBAADAX9y33fvwjgG4slG+lWXFCkcdv2EA1l6hirhIsOtX2P1b4ZMp86w0D1dmfVLeoEEDzfMpv30NNrJqnvBrN0j+rmKPqn4a5V+3N4QADAAAAAAAuuU9BmCS07SHH7MBuEy2rDAAVxbkNCuXX6HqAEwWNT2r2ip7MmXKNQ9XZo9Y7NNg1yGLFY49riIAV7bBKvaosntV9rq9IQRgAAAAAADQLe8rALMhTbP4t3qAiTLxj6XJseWTXhVxkTxEZRfEVvZkqgjA2utXqMKHKxOANS9OFRusYo8quxcCMAAAAAAAwN/wIQIwe5N2dmVHC7PrVBiA2ZvKJEBNsCyf9Mit2n28ZZ5VZYOTK3sybLnm0TUbL3PxbYUqTKfsHV97DbC28nukWazsXuVflr8FARgAAAAAAHTL+wrABIl57OhcEvw2bNigyW9s1CTlJAR6eXn99ttv5QMwwW6Q3QKhnQbLP5Zmm+yT0TwrT63JmYnyobGyJ6P96NoPzYZPtpxgc2yZRykfQcm99PX1NdNWa3cRV7hBUl5mj8osVngvBGAAAAAAAIC/4e8GYOANBGAAAAAAANAtCMA6CwEYAAAAAAB0CwKwzkIABgAAAAAA3YIArLMQgIHPxBnRT+4y/GIyxVyZSpT2zP+BuvTu3Xv3Q9OKZerigsgH90nJ/SeBaRJ1AXx08qLYEN8HYQn5Uq7gb1GKUl743wuMytG81dqkKYFPmDeYeY99HmdX/QjFubkpaULyhzDR/46Lp3/c33lGSjo7Jviu77OoQrZmVRd0WhT7aQhPzVNyZTVGcaynm4uLy00f/wKuhFCkxwYwpS7e4SmFXBkhT/a750FKXa/GFCm4MiI35IZ65ev3n+RxRQCVESUGpObRcm5JpSqMuurKVCrvxyFFXBHAG6Nz0qOCE8tUHTomKS224i+smi4j5NE1prn1iMiluaLSuCbd1csnUKtNp9Ke31Y3048j0z/O9ycCsM5CAAa+khcmPd46qg07Zdw/O0w4GZCmDjH5ngvafsuWMjrselqgVBXHHDZtyZX8a4Td7cRi9Tbg4yr0M+35k2DQ6tu5XMHfE71tSgNBy2kXErjlVwr8b+0a/A33BjOmbPG4++qcSGni5Kvms9dZXUwjf2fd3Lpk5NTtnjF/o0LIk5xMOgp+7Db9YT5X8unJCxN8dgxvwe58vSHLvRKLXx3aV3NKSVbkqTWzGnbq2bNbn76D91yMV78Zyvzwu+smDGvTkxR3Hzrj9J04Sl1c4LPPok/vLqS4U8OZq69FFzCHUcrcF4eN+zfqyKzcq0dPK4fgzBqz//DxiZ4dWz3w+w0eaeo6paAywh1WTm/UmVSfzn37jtl/OVmkXg3gzYjj3IyH9e5q48ctM5Qxt7d31Fuy5moyV8AfxYEnbfX6diWfl+5NJy10CskSlzrlWtKkM+1xlz59hu9ySmRONysLYrx3TR3cti0p7jlt/uIHL2Iqzs7vFQKwzkIABp6SJl1c2vOrL35sO3jUqMHt634taD5lkx9pZKWPTDrU/755n4kzjIxmTzOYbOOdIhImOI+rVeufLXuNGjVqeK9Wv/y7wSyXDG478BEVBSwd8L+vRqy791KlojN8b3leDg2P9PLyunz2aTqtevn8pvuVy4SL583nrzr8hCnPmML7IUHP9i5pJ2g/+3IidwtHkuu8odHXAkH9DoP0yTs8auTQzvVIDKw90zmH+VYWpYd7u9+6E/w89J6z2033mx67hzcXCBr0W3XaLzorL+O5n/e90JRipaoo+t5NV+aR1K64ukVmlv5WLyFPdl7cU/B73zm+ReTAJz3wnusN/8AQ34dXL7oFxOa9Vef2u5IkOJt0+erLOn/pjRo18K+fan3Ty8ItqXr1T1dKlumzd/DnXWyeMYk144adUdeJF7JUKirYbunEyYs81CcZ4m3GG83bcIN8vgv9Nvfose6IPxNc5GHb+va19k6kZeIYe4Mm/W0fqU95FPq5bzNxeIZxHlABWfZTF6dds8bUa/Llvza45TAfV2mK17Z+/+yxI1K9QpK77fxh891z1AsAr6cQZj/eOLr+Lx0G7wxUFygLwx+6HVg/ukfjLxsb7LiXrS7kj6Kw/cP6LNt0jXyNk7b50ND+q538c7VOOJY06cHqL8MM7z0mg2e7FqhkiaeX6nXQs3jMhGFVhtvipWvWeyarT0F9SAjAOgsBGPhJkXLJuJ3gs24rnzAZxW/RwG8F3U0vk1Sb4tyz+be/6a9wvHL1qmeouqVV5bqvFvzwS/2T6hOxWZeNBD+3MvLCMLePryjAbGC92iOt7pF4m+k2oOHXgratWjN9lt8sdnTdNLOkj14gqDVg7d0s5huVTvfbPqKZuqxu/fGD9Bp+02N+mQBMha5vW6/Wr20WeSSU5NVU97GdfhJ81WeVX6FKnuBm3ljwVe0/WzUQCL746b/Ne2l+iK/nGrdbJwxbCGqN2BMsUQVb1KvF3cBa7pJaYZolAXhJL8H/+s31I3krzXlqx//7rEGDxv8md647/8zzD/59XpE0F/M+n/1nmM0T8rf84bpe33w+fJ1nDTmEp9MeupqseJCqXhA/dVzav/fBOJUkaPu0BV0X+3HvQLLTGBNLs7sU9WRRK6N1hyO5joPYsz2Gb3cPjbq/a0g7y0cY1gGv9dLdYuzEYUMP79rUoJHdlXTmlAmVePeKyaqHmewKikQHO4u2a+6zSwCvoSiKe3Jy50ITk9VG/bf4q4vyA+zXTOgybcsG097Gq62uMWONeET6eHV3o0VW/uzRlSrTZeCI9UcfZ79KwFyTnsIuKVKdDlu2tghQvbyxdI3xJHd1bCZov8nTtyw+E1XhWeb3CAFYZyEAAz/Js++dsDaydQxkDnrTLo9r+G09wwMxEpXEz6x7k39wCUbww6hVbonFsoJHW3/+v39+PX7V0aOH91qMay345s9uh55rXT0IH4d2AM6+NrLdbwLBf9sMnTJrnpv/E/dFC41nG5uarVqxcGQHwbc/D76UrZKmP9zS5wfBVz/2NzCaOVW/+f9qC/7Tb1npAKyIONaxUe1vTc4laqfVfO/5v33/4887w1TSZG+bDgLBV/9qMnD6it07j59zmP/X7//87Oc/By44dPN54MlpHQQ/TjwQLFKl3bS1MJlnumKm3k+1SZj9oZtjYH6FdaRUAM5wNR30vUDwc4v+E2ZbOfmmfYQhXeWJgrcZ1hd0mnNFfYon4cL4toLms+xeVDIEvHpS5ERc22Q5b8qkeUsfkiOr9BvrLWbpn4vnbqWfrOq2ZtGBpLi7w4ds2nA1nS1VFt6y/nP6wavejiYtN3tHBV7etmnTJtsD5/xxdgsqIQyPVl+AEXC4WYOtzuwQaG3KBJctI/vPdS4zzASgYnR80FkTq5PBgQ4bDPpsZgPwy+SM7FxyaJJ6fOR00zWePAvAL69N1Lda4BjHJVep77ZWk3ddiK3s/GPqjV2je8+8kKoSBm+3HD/O9jr7HSl7vq//L6ON9kd+6GtVPloAzsrK0tPTO3r0KLcMpX381wcBGPgu56n9hCYN63bf7ZUoUykjD3b8r0DQpL+h6aJFc4Y1+lzwx8rryeKCkBMzepVcGFy3Qat/Nuyy+jGGSH50pQKwp36bnwSNza9mcTeqZd45uGv9tN6C//ze5EysPNNrc3fBVy2nbVV3D8pub9D/4R+dTUsHYPrR+o6/C9qtcU/Vzp501LFRdX5vuC+CBOAbls0F//5r+gXuil1hgPWobz7vtfAO84WdcGRcGxKA9/uXnJam7h5c8PM/BM0G7/DPrWQMcZkAPK9nrX+2mH8pibv1U8i5tWFErc86zL2gzosxZya0Ffx7rK1vjZoKSpzy2MHAcMa8mVaup65nyXK9Nu5aMuGOphc7/NiwzbsOJT6/O3LFgf0le6ZUBp38a/6pG/eOGzeZMNN6u8VMAwODsSP7zZ7vEMC3YYfwPinpB/v/bLCtXAB+GXZs4oAxUzbFfpITWVDjSDJfnDPZuSxMJHlmt3ZKSQAu8fzQiOkLLfgWgLOvTzLfbXlDs1cvzneafdQ5VmuawleKX5yZOnjkuHVRzNlYSbSbzaxe3cebWVhY7DhkM1v/t8nrDrz40AdiNT0Am5iYcIeuakOHDuVuqJbYF6HCJ4kADPBekfQ77s+6dVqtcw9mu7synjo7bHfwZdvm0D1//PrVL5Ze6pY5++Z2qzWrtm7beMh85NeNOu4N/dAnHqEc7QCc5anf6kfB1L2P2O/NrLADdnvXnDtpt8B4xqC2gv/Ua3k+io67tPT3/2s0aH+cehVV6kHT9oIOZa4Blofat2/4xU+LnVO131CRz/LGP/zy391MD/A1i2b/V6/HFn/uhHX2vTVD//V5tzmuTKqOOzr+LyYAB6jTcX7AlmnfCATfdJ3okVwq/SqKYh4c3XP4bALzZBWpl836Cur2meVLjp4zXOd0/fybkZYPP2XaLA6wnfyboPNcF7YH+OKEdoLG0/dE1MiJfMLtVvRpax2c5LN545zRziWnFaQBVt3XLtufGHN32ChbmxvcFfxK4d1tf04/dO3Okan1RhgcYS85U+Vcs+3bacTh+BpyDTR8AoqKAnBBmOOE3s0HrziahO8GeAMKOjXi2orjG9WHG3EOm2cO2PtCfUMJfgbgvGsTxlsvPF0yF6Xcf2+ryXsuVDCPJBV1duqAFv0XH4zXjM6i/T0Pz59jZGS06ozbxaVGR2yPR1R4ndF79EEDMEmnmrD34QJwNQ+92soE4I/w+lQBARh4S57uf8SgXcvmffb5stcPEoWRZ0/b774Tq07Dkvtz69YR/LruRlpywHHrvQ5+6rP6OedHf1Pnz0me1WcGX91RPgBP2Hlf/RMJqWeHffUPQYcz6n676/ME/6n759k4ecYNmy6Cr1vP3qMufnnDvPtPn3dbUOYa4KKApS1++UeTftYPNd2F+Q8XDf1N8M9/j72cq5KleK758/9+77bhEfdFm31v9ZDan/deeIepJElHx6p7gAOLSVI+uqzvV58Jes9wLDdrpyTeZdEPgtpNlzxlljIumzQTNBq09hmJWOmuc7p9/q+haz7pRCfKlPPLegh+Hr0rlFkKtB3ww//prXYp1bVebSmE8b7nbK8nlnS55XnabGnW0zXs+a7Z87uvDOLOWmReGTvTwvRaUdHjOc3mWznEcOE28WL/AdZOT566WzWyDSzpSZAluh6Z0cT01ie5GhtqhHIBWJLxaK/BQP2Jm06m4NoYeDPipEub/hJ0nbZ4wbx5M8d0a9f4107z7E6+4K6NJfgZgCUPlnSet3BjENvnQFrsEYNWHbyXUTrIynL9D84YrD92vUNCJWci0/cNXL7A6sP/lAIC8CeEAAzwAdApnrZ6PwoE/2nVz3TN2pXLzMyOejwrLgoxb/K54Nt2o+abmZlN6SMQCDpveZgtyrhh9LNA8PsgE7MlRsP/+vaL/y31TP/QUy9AeRUF4AfqAPx409ff/kvQa+KSZUuMR7f+WvCv35s7J6mkqfc3dv1W8PV/9eeTd25Ei1/I+z2gzDXAKhWVbL/4N4HgH38Nm7OYvO1mS01G/8WMFRq+JiJPpZIneZQOwJne5iO+E/zaauRc+6sPHh2a0fGzHyfah0iLQ1Y3q/MZuVu3iSYrzJcvW2F+wi9dwtUSRVHUBaPWgi+/1ZtjZmY6pVe9L37otfI20x2coh2AiyOO7bI0c7wX9+oQ6OMQR52a0fKzb/4YZGxmNntQ/dq12i69EFszxvgrRC/cl/VpOu4e+1uRydesuw0ctztFKPK1XTRi4tr76nctZ9/0aSbrrpBVcm+v6N5/w7nn7Mr2IzqvvRldJEzxnj1u9P4Qdo+zrm7p3VH/aGXHXQBlArAs99klw8ETlx6PUt8I8EZk+RFeG22tLMyJxdP7d2v+e2+zY86xr0be8DMAq/J8bfQGLtvjo25vc85O7LHstE+mdnMrK3zuPmvI+AX2EVwBq+CZ8/kzxwLUTXr+WbPh7RdseFhy7dGH84ECMBvnmAMNNRLz2JJ9+/bVrctMtFmrVi1vb+/yK9euXTsqKkpTWOH62soH4PT09AYNGhw5cuS7775j7/V3t6+9fpmAyhZqNlKGnZ1d69atU1NTyd/kaZAtr1y5kr3J09OTvRf7hLW3RlT9+nwgCMDAT0XBhw0asJ+sEi2nH0kgbfOtw4vrcCWCDjPWB6gnJxSFHZ/0P67w81l73dJrRjjgm0Jf487/FvRb4U2STIZL398/Ewyxua3+BswJ3jj5c/b9+VVv/tyhv335oz4TdMUpPtYDG6rL//Vdu+YtfhS0me5UMjOShjLD7ZR5B/VanIEL7U/HsOMA4i4t+a/gXy1WsVGK5OUkN7P+PzErNZq1/vzOmS0Fnw/f+0yRcbVH/a/U9y3RZVdAseY0CYnAV426fc3d9MNfNh7spB+JF6e0EAh6LvFm5pBNd2r3vUDQ3uI6N6HsxyPNfuZm3uVX9tl913H2xbDc0ifkqzGFKNF39dgxPcdNmDBh5NA+81acUs/eLk/3cVs+YmQ/Ujp6yMBxh52D1IPl5RnOa1f2GzKcFA9ra2B0LiiL2dHixKuWS6cPJYUTRo3oa7TwZBB+xgYqpxDf3vHf7zecTWXaCFHYycnkO6PewKkzDSYyVWiC6Ya9IZhIDf6OF/vXju9kG84tscL29Bk7a6kLNx0yj+Tc2L5h0OBh5KMypvMEg0MPE4rlKnnekyOLDtyNo+WyWOeZvwkEdfsZzDRkP1Dz124PzFcVR3udNNbTVzfp3ToNmH/0ATeb4Qf1kXuANbmO3MRmRbZc0/OpuUtl66vXekX7IVhs8iwTdN98+2y5ZpuknNy3fGGFGTgoKKhOnTrsBkniHTt2rIGBAXuT5nE1f5T5+w339z1CAAZ+otJD754/dkKbd0A8m1dy7111VpecDHt1epFOe3DmFCk773o9C52/n4g019/r0pnbwZkks4hTbrs5nbgXkcmNfaWTLp0/feLEmcuB2Qoqweu88/kIdlRxYdwT5s109vF5Hhzgferqo8QKp5sUhV+9zLzBxKmzF+M0Z+EVxcmBLiecrz3L0gxtpKKDbjHr3Xz6PPmF77VT5+9H5SvF6XfcL6rvzjp1+m70y1KdiPKiyBvsba53/dRhjBAmP7526sytwEzmSJrsk4fTiduaffq4Cp8/uqR+erfDS1/BXAOIE9z37iD2nrwSxo2sI6RR/h67mWKHu2FaQ8zFYa6nmbV32vllv4r58ljPfTtJ6W6Hi+q54QEqpZRnRrpcDkukmEaBzgi5esjxmP1epvqoHb1wNREj6OFvkOQ+D3pwPbT0ebeCqNs+fkEpNXIyhteQx3hdPMB8VvbeT2LPFSmKo7wcrodnShXy3Ijrh48fs7fbpf40EYfPubLDomRR/u7qJn2Pp3/Ux7nU/iMHYE0Q1fSIsosamk7UN1yfPAR7YptF1mcDsOaOZbx2+9ohVqPMo1f2EOw22V5f8kD79u1r2rQpuZf2Y1UdgMs/H3bxQ0AABgAAAAAA3VIdAjDJnN999x0XYQWCvxuANQ/BKp9O/9b2K3ygMoVVZGySe8nzIRsfO3YsWZ88PbIaeQK//fYbe3cEYAAAAAAAgE/jkwdgNp1qyt+iB7jqAPx3t0/Wf+seYILNuiEhIcOGDSOPQu5Inh6bitkVEIABAAAAAAA+jQ8dgNm0Sf5+kwDMBssPF4DfZPtsuXZGJeuwd9Qu1OxXGezdp0+fzq5MHr27muaBtJ/wm7w+7OKHgAAMAAAAAAC65YMGYDZ8CrRmOa4w4NnZ2bGDk2vVqrVhw4b3G4CJv7t9dgvsXTQb1y6sLP2y2Idjt8w+ivYz137Cb/j6fCAIwAAAAAAAoFs+aACG6gwBGAAAAAAAdAsCsM5CAAYAAAAAAB2Sky/sN/Mwl4dAxyAAAwAAAACADrE+dOvEFV8uD4GOQQAGAOAPJQAAAH9x33bv5kV89rilp7gwBLoHARgAgCfIkQEFAADAX+8lA483O+0XHMOFIdA9CMAAADxBDgtEAAAA/PXuAfjC9WeWdte4JAQ6CQEYAIAnEIABAIDf3jEAFwnpkaYnYhIq/SVb0AUIwAAAPIEADAAA/PaOAXjrsbtHLjzkYpCuMjExGTp0KLegkxCAAQB4AgEYAAD47V0CMOa+Yv2tABwUFPTdd98J1Fq3bp2ayofOcwRgAACeQAAGAAB+e5cAbGju5BPwgstAPPUm4fbNA3BWVpa+vr63tzf5Oz09vW7duitXrmRvqtEQgAEAeAIBGAAA+O2tA/C1B5FmW924AMRf7zcAl/HWd6xuEIABAHgCARgAAPjt7QIwLZXpGTvwbO4rT09PdmQy0bp168TERD09PW5ZICBJleRV7UHLdnZ27KJ2js3KytLcq3bt2lFRUWx5eexY6KNHj3LLNRkCMAAAT1QdgIVCSsKuJxULhUKutPoTiihaxj5xQkaeO1MoFIqlXBGDKWbLxa/WVcloityd+ae9spyuOTsPnx6pWGJp6QNu5hPE3QpQKdJ2SeRcnXlVa0hLTHNFDKWE4ldtIntDK7idU6kUEnUjXB7zMsjJemXbY+bu6hetpFEv7+0CMP/mviJxtE6dOuzgZG3a4ZYk5Fq1arHrsEGXHcCsWYct1KyvScjsIkv7GmB+pF8CARgAgCcqD8DkWINJgEV+10/uv/QgNk8iIdGwJiDHSAqFjEq8u2m9paXVzqNnYimVQiwSkb2RpPl52FlZkuLd7r7xYpWSFolIwqfjLzvsJsXr1tvcji+gZTJaLJHKxOl+O7ZYrbO02rb/ZGQxOebitg/wGpREInqZ4H/OZv061q7D5yOypRJSCwEqJ6QkSoWkIOqmNWmOrPaevJREM20Xc4pOXuB78+xGpvFaf+Dms3RSTlccEmsgIdM4FwRfcbRdR3bb2jMig5KREFxm90jIlSnpwsTnd5xC0yXiV99GTPpV5kfevHvnTkRWZd9TbxGAX8RnG6w8l5GVzaUfXqjsitzyvbvsOtqBWbNOmRRNttmgQYPyoZrFPqJm4zUaAjAAAE9UHICFQiHT8ysKu+to3O43gaD7cvd4lTpFVnsUJaVFBRF31hlOaj9y2tSxY/T01tj6ZchlCnGqz2GbUf2Hj542bdqYIf3095zwy1PIZEWJbhvNew4ePXbqNEP9DhMNHR4miqQKOvPhuQW9W/aeaDh13Lghg8ysHicIa8oZAPjExFK5KD3w/LLeXUdOmTJlMrHQco9/uozUTIBKUWKpRJQdeG35xPFdxk6dOnqUnv5Gu6BsuVSqksS72hr26tR7LGm8Rg3pNXT3Cd8ssYwfZ1SYYUaS9EvrLfoPGj5p2rTxXceO23Uz/CUll2jaW6FQxA7TyfW2Gj64+58Oz1SaW0n8JV9VkY7G/2sy2ORidGXfU28RgOdvvHL9figXfXiETaQCgUDTzUtoB2BC06mr3burHYA1vbsaVXTzenp6Vj1MuqZAAAYA4IkKAzBFS6UvE+4eMWnZ8D+Czz7/su6oDdcTlSp5DTjcomi5vCDazbqtYMCuF8yguMxb+2Z3HXQoXl7st2NK53YT96Sq9ztzv5HhrDWnUpT0vbm/dhhufjuHKZXFHRrbZcnFoMyCzNvWbX6adDmJKc0POLy0++DTUUopTSECw+tQEqUwNdDz4MDdkUz14WAINFSNopXyl8EnV7X6fPiRNKbKJFzZNL33qOOJKlWexxKjNmPWXCxUV6UYx60XT95MKJZJ+RCASZOtzLxlMXDA4p23RczuZZ6c2MX06M04keYbh6Jp4cvY+4dNFoxv89evPRt1OxtREoCFzBCel6HXD4xp/lWzyWvc4yr7nvq7AfiOX6ypzRUu9/CU9lDnMgGYDcn29vZ6enqaZFtZD/BrIQCzEIABAKqLCgOwWKGSh59aZtip56qttuO6/f7vQWtqTACmpKL8ZD/vLXZBaeqLwjJv7zPurmcfLS94ftP98omAbLmcphUq5TPLPmvmmD8upiOO2d32Cs2l5VKJWFIcsK3uuL03YjMz79q0/n7cGXUAfum/Z2GXwWcQgOGNiGXyl4mBZ5Z2MDY3mjVr1uyVW469IEf2EgRgqAollopyox942R4MUZ+OUyVe2Tyr79hzacrkkzMmjtM/rE7FJUhrzI8KRStVBTdNOsyZb+mbR3ZLIlWlugzva7bnWlyxQqYeM0HRUjon9dH2OWOt15nvXm3TtPOpcKWMDcCkLc/3dzq8de86k8HGS80uvHgvAZiXc1+VV2Z4c5nreEmJoPQEV5oAXOYa4PLIlnv37s1u7bUr1yAIwAAAPFFJD7CESgvxDw5MUMmjLEbX/6LvqpoSgJkuAUo9H4qyMOb+UcPRA/rpzbcOFqkU5MCKUNBCkUylKnxgOVnf2NItg5tZRk6LKFKcfmd6u9HL3cLypBJh+INDC1YNmGgwddq0iXaOF8PSKFrrsjOASoilKnFm8KUVdRsMWDDfwmLNwlmje5ouPRJdyHQCc+sAVICZAIuZCkqRF3pjv6F+7376K3ZFy1XCeysXbpm44cbVcxunTJo02XCPq3++gmnKeFGdJEpVssvYcetMHCOlZKfE5MvHZ2fLCdtPhWbKmdmwCIoSvszPjkoirXXe3R3mjTqeZAOwUKxQSaIPHd6y7OyjkHPTZs5ZdCHqvQTgPad99p+5z4UeftGeAprQ9O5qRjVrkiq7pnZw1e4lZmOtehuM8h28bH5m8SP9EgjAAAA8UWEAVk+qKVMwh2I5D5cMr/dlvxoUgNUoWkrlJwVc2bhp40az1Y4WB4IzFcwFc0KKZGNh6HHTYe0Mtrul0CVTWzHpN9vPakKHjoYugXkShaQg8q7l8o1bzNbZ2GxcM3/tpo17QjJVZAuIwPAalFhC5acE3j50/EGK+jMm9N87V6+v0ZkEUq9Qf+A1xDTTEfzoovXmjTaLV57YfCop7cWOpfqNBxkt2Gtns2nTeqMpA4fsOv4wRyznxRBoEoATXScsszV3i2X6tSlaonp2qsP0A2fDskoCMEFRYhn5QqLirtuWBGBm4mdxgpOThfHpS9nypGOjp85Z/D4CcHxq3rilp3g299Vb+LvjnHUBAjAAAE9UHIA5EhKAHy2tcQGYkpA4q5TR7K+JSML2LR3SaLpntkomJUdQwvDji4e26m5y3C9dxuyRkKxP1st5Yjnur9bjll2PL1Ko5KLUu3u7fTfOLlh9TZoi7uwq/Z6j94UXy7TnHgWoEPPjYezBtowSCmmVqijk0LZl7c3viGrGPHLw6YiZRksppdUdwSqR73oj/W7rrz7cu3pqM6OjQUXqWqV8ZNO5g9HiK3FiXkxNTz4rqa7jhq6Zax8uVikkYom0+I5NiwlbT5cKwAyxTDsAy+TKwoSzF7Ytdb0qJq9K/MnxsxesvJpMXp8Kr7Z/8wC8eIub++1nXOLRYRX+uJGOQwAGAOAJngVgSiyRFGUGum6dbe+bJRbTMqVKluy8dl2j9g6xZGezPPfM7NlzkO3lkBxygCmlmKwiUyliLpjrj+w7b9PtVAl5QWQKOj/ugvWPDQ8+KBBJmV+dLH7hvGvKH8buuXIZAjBUjaLlVM5zzzMzF96MldA083vU2Y9271nU82CoXEmqHECFSNslzU/2uWhr7PC0QEozvyMtizs0f22H0eevnTXfYboimGRjCdMo53qPb7psxv6gAhKVuTvXYLRCJby7stv0+WvuZ6pUCqlMFevUd5il/d0kSikrNetCqQBM4iwVddp4/B+1Ww8YNnhg/y7N6vzy2/9amx69nlgsr+DEwBsGYF2Y++pNaP8SEmggAAMA8MTrA/CSYf+r1de8xkyCRUulBZHXrcb80fdCJNOLUuB3bHmbXjMdU+ni4COLJ3UbtOWeemYrsucKmUQqV2X4rho1dpCp4zNKXaxSqpQKcXbIqWl1OpvfUI9hzfLZNrtz1zVXc8VS5F94DRKAFUUxV7cM/KKLjT/TZUdFXbCdrmd2LFGMEdBQBUoik+YFnl82qvFQj2Sm4cm5vWd+60Gmzrn56c7Lx86avT1cxhSnnZ7RUm+OtU8mT06okFxbFLh3SM/Fa52Z9lYWvntgvxUnfZgLVKQy0khrGl02AG9Z2bDDiXClXCYpzokKvHXG6aTjsROnT9jMaNet96AFpwNjXooq+sG6NwnAtFQ20vRERFQSF3cASkMABgDgidcGYJ+Fg/4r6LHcM6GmDIGmJBKpKDloz9xhf3bs2rVzh27dF++8lSpVpLisH/B/X9T6oW2v/j27dmKMsjwZnh1/anIjwZffN+3Qo1cPdXHXbvtvJRbJXsZcsjZu2LZT164d2rRvO32Ld45ISWMAK7weRZOKkhR0dOXU+m27du3atnvPEXsvJlEKdP9C1SiJVFYY82jz9IHNunTt2qlDj77m9vfTxUq5goq57zFrzMj2Xbt17dxshOEKr8AMsYQvs/JRtEJeHHB806Be7Zn2tuHEhacDUim5VJh0zXrMOueArCKpOgUzATjW02bZ760dw5hrgEnBq1Abd3zUNGMzd+bMwVsPgT7o9Hj3ibtc1qkcErLOQgAGAOCJKgMwJRYXZ78IfnQvMCqjQCyuKfGPoiUSqjj1iYeri4uLx+1HscUqlUyYG/88wMvL+6a61JVx0+95Zn5OjP+Tu143PTXFbm4hCXlCiVwmz4++cZWUunp4P4ouUPLhcjv4KIQisVQmEWdFXHUjlcrN+2FgtkQlpXgxZy98UCQCS4T5iY/cmdbo6n3/RJFKKRGpL9QojAv1YRopF/eAmAxKrpRQ/JgFmhAy81mJkgJ9rqnb25C0fLFUKhEV58U/9fKPTi8QitmR0JRYVJARF+Xrn/ySK+GIxVRe4rOwiMjU/Mq+p14bgHPyhXrGDlXPfZWUkmG+04OkIN7/QhJUCAEYAIAnqgzAzCWytJw5bpDTNexoi2KuveSQ6EqevFiqnhNLm0ImpphL7cqQS9Q/sUmxk2ipyTF4Ff4OUn3EzJxrLCascDcAVIXUEzEziR9HRtoudTFpwJjpCNSUzPTPPKtQZP803bkKKc1Mz8CcDlCwCyWY37iTkWa53A+KMe27UqEgTXdl31OvDcCvnfvKJ+DFuKWnXG+HG5o7kb+5UtAZMQmp/WYe5qrLW0EABgCoLqoOwAAAADVd1QH4YVDCXCtnLuhUZKvDLZJ707IKycqYJlo3+QXHjDc7zVaYt4MADABQXSAAAwAAv1URgGmpjASb4Ih4LuiURsqnrDjrcMmPW1ulsj/v+yaXCgPPnPPwX2t3g6sEbwUBGACgukAABgAAfqsiAJ92D7S2v8mlnNJI5iHZOCQqnVtV7UV89rilp7g1QGeY2ly54xfLVYK3ggAMAFBdIAADAAC/VRaAK5v7KiklY66V81q7G7T01SXZGv1mHsY8WDqF1BDyphcJaa4GvBUEYACA6gIBGAAA+K2yALxyp+elG4Fcyilx/X7oSNMTD4MSuJXKqaLTGHjpyIWHW4/d5d7+t4UADABQXSAAAwAAv1UYgAPCU2ZYnOcijlpGVjZJtvM3XsnJF3IrVYSWygxWnvMLjuHuBrxGaoWesUPVVeJNIAADAFQX5LBACAAAwF/lA3D5EEv+Hrf01Gn3QG6NKpHwPNfKuerfDQZ+WLfv+oXrz7g3/h0gAAMAVBfksEAMAADAX+UDcJlhzPvP3Cd5OD41j7v5DZBQRKIRd3/gqfcy+JmFAAwAAAAAAJ9ATr5wpOmJpJQMknBiElJnWJzfc9qnwvmuqmZ96BYuBuYxkn4t9l7n3ux3hgAMAAAAAACfwFq7G+c8/EnCuXQjkCThgPAU7oa/77R7oKnNFTZLA29kZGWv23f9ffX9shCAAQAAAADgYwuJSp+y4iyJrOY7PVbu9HzH37Yhbj+J0TN22H3iLmIwP5y44kve0Pdy3a82BGAAAAAAAPjYxpudPn758UjTE663w7mid0ZStP15334zD5ttdTvn4Y8Jomuc4Ih45+tPzXd6kDfR+tCtd5/zuTwEYAAAAAAA+KguXH/WaeI+Q3OnD5FwaKnsus+LtXY3SMZuM2Y3/tWgfyNNT6zc6el2J+JDVAwWAjAAAAAAAHw8RUJaz9jB4ZIftwzwESEAAwAAAADAR5WWVcj9BfBxIQADAAAAAACATkAABgAAAAAAAJ2AAAwAUBMUPrmy35Sx/syjdAVXWFNIoq/MMJg8eYrpuu2RUq5MpSr2ubzTcDIpXuxwM6bcLhUFb7X2epJKVhdHesyZOZWsyKxrON1o7qL1m11iZNx6AOVJ41w333ueL+EWS4ie3/bavD8Ywy7hDVGp97dOm0Janllrd9/P5wpVinh3+yWkRZs8eZtLUBFXyCdFjw9umMO0t9NOPE2Xc4XlFUU/dd4RkMUtEcUPr+xQv1xTzBzuJtS07ynQJQjAAADVnijWfVv/OgLWv5sZ7r8T/6GmRnzvpHmBLjOHGoxbY2u7Yfn0iUuXeiQxxeLnF46YDJ6+0JYUm+jpbd9zLUO9Oqfg8fz+XwvmHQmlSHxOfrJn5zayoq3tNtvV0zvXE9SZeyGt8qMy0HUxJwz61xesvJJR+idFqVgP0/rN6vW4mMYVAFSFSr7nMLr3EAMr0vSsmDlq5Zy9EaRGiSPPWRj2GDBhJdMibV460cnrbraSuwcvSGMPLFg93MCM7J61oeHgDS5B2a9OW76iSL08v33H9s1OR3EFkheXV0/v0Wf6OnJHy3kD+2zY5JmEZhqqKQRgAIDqrsBvx5BvBP9pv+DwOSfboc2/EnRaci6xhpxdpxM9Njf7wvBsHrNABTgu6ttrX4xKlXV15cJ+k+zYH34UX1wycanlsRTNUVaB76yd47t/+83GMxFluleSPGx6d1tyMrNs1x4AQxx+zGT6oGYjO/X/ps42tyytAKwsig4/1LZ/03Hth7uWOtcCUDFp0s2dE/877hLb8Rt+2GZ6xwPxKpHvpml6vVa45ahLJY82GS7ddixc0zlc86VcXTRs1IqTYeoF+ubijgscvZJLnUpSCqPcLIeM7de4R8sxjbs7x7OlmVdWLW03ctsjthkX+V/1836Wi05gqKYQgAEAqjtxZviDy1ceRzK9vsnmw37+ZZiNb0EN6XOQFSZGetxOp9QLaTfs5nY3cM1TFT+0MTLqvYE9xlKpXl6bNGbV7NNxbHdBto/R3nXnz29p23jHieACdRGLSvTc067FIodsbhmgjDz3/bZHDhx+emVfk0a7L6Wx1Y5QFAY/PjnLbu0B881/Db6UzpUCVEGe8WDfrDq9dj9nFvLvbLYY3Od8rir94vx1ZjN9xOpVVKri+6P6Wi08X/4ajpqKujO308IFW8K4a0yKvAx6mO69kaJ9LkmYH+aw0eqi8/Ezm7c07eoUxxQqkpzmzNMbezJXvQpAdYcADABQUyQ7rJ7QssuE1Tfi1P2pNYkk3ntr986tO/db7phKonvq9fU2xmPcU7lbFUFbe1uZ7whm8oo8Yu+KVReCE0POdGls46gVgGX5T06bdhu5LYFXww3hvZLmFKrPovjbN2uw1VkTgMWpz85arD4WlXpto0WrgQjA8EYUVHaMm92KMZ27d+87eryV15NIkUpFRx5dPHis6UG23zPjytx2tYYucuLPWN/caxNmbDG7qL5QhVAGH25jsO9CjPZQHJmcfllM/hc+slvbuAsbgKlnu9ftGGlzze3g0u5Ez6X7rmtdGwxQ3SAAAwDUFMXPn9xycpw/dcKYLU4RNeYiYDVZUUqwi6v7hV1WB2evDSrKv22zc/GEW5qu3LAjQzZt2R0qUamiDq/ffdI3hVbFnWzTYNPxEM1xlyw34Khxy3HnkrllgMoo6Qf7/2ywrSQA09mhbpYWZ8IVqqwr6yxaDbqSqS4GqJIiK9Bpk+HEQ2ddXVzczm1aY2W0xo00P4qCkAvm+t2adOzSpcsEy03mM9us2XkmpvTl5jVY9vVJy3daXCs5OamKPNdx1pGLsRVN9ZV+Z5cmAEviTq3Qr9N9ysKL3i7E/hX6Pc2t3VJxDTBUUwjAAADVXGHgsRW9O8+wvau+6izl6PA/BHUm73sVDWuSpJOrprWcdef5E9tVs4Ydi+ZKRT6LeqyaeyCGSji5addR7xRmeGHE0dYNbByflfQAy7L9T85tNfF8ImZ/htdRlARgpiJJ0kNdVq1wVA8vSL20dg16gOGNSDJv7Tf8c/w+ri+06PHmUTP7mPiopxDPSQy7fZN4GB3qOnWO3ZYbybwZAp1/Y/KANfMORnM7RD9Y13rKzktxFZ1w1Q7A1HN7y8Vtl19J5GZySHUa2nXufEf+DA0HnkEABgCo5mQJV1a0EwiaTd0WT468PJZ1qS1ov+BIbI2YB0peEHx5zbANXuyMMeSQ6fJqi9Z6XrGJR+Ys6jr3Dhfinx8ZvWz9hpDCzBNjuzZt2KR1u7/atGn6+5e1/luv2YIL6r5uRf4Txzktp5zDrKLweiUBOJ35jBT6H5n43feN27Unlap5vZ/rfPlN8/GLz3ET9wBUhk7x2DWpsdktEbecdHSheevhrrkqpVaok9yb02n50n2R/JmVTxFkO2iqydp73GU28ad6TNxxJuhlRUFWOwArElwtdhgv8CkJytQDg07LZu0KQYMN1RMCMABAdafMuGU16pd/1Kr9yx9/1P2+9j8Fvc0da8gRl1KW9dR+YdMujk+KhEJhkteuRa0HrPKi5PKE82vmGU49Gk5Kc30WDV1ksv2JRCnNTYoNexYc+JQ4v6b+z8aWZ4OzRcwhlCjhnHXDxnaPK/o5DoDSFPR9u4a/215QD4GWi3ITwsLUlSrg6pYZ0+u33/YgNqtkDiOASijpKA9L/b8GOIWSRkoYc3mj0V/DbR4L5bl3Nm5YtczpOSnMuLNo0NChW9yS+TQuRZLsPLHv8mWOTNucf8+m15hNLmGFSpVSLqFomUJ7Boa02ztW1+9wNla9ICv22T1v8cglPlnMyxV5cna7CUZ7gmrWlTqgQxCAAQCqPaW0KPriwmF1PxMI/u8fPedvDcyjas7IMoUoxd1mTMPmRMvuAza6xhWrj6EkWaG7TUY3IqV/dpps7hFZUKarIP5c11a7L0Szl9bJ8oKOzWs+3y0RE2DB6ynoR4fattjtwk0+/kretS2buoy6xqPfrIEPSEHnhjpZDFO3XY079Z967hnTGywrir9vPWMY03Q1rTd08aHAl3w7LSfLumc/fkBrsoPN6httu5VMkbZZnum1fsiaS8+KtNrpjPv7NrTuU/IzSCpFYXrgpgVj/2Rerkaj5m9/mirEAGiorhCAAQBqBpm48CVDJK15KVBekJFGZOTka88VI6UK0pni7GJJ+QMlOV1YKJbI2X1VKmR0cSEtQ/6FNyJjao+0VHcVQyGlqMJiKQ7L4Y1JX6obqfScAq3eTIlIXZqeUUTzc4yvpOhlJrPbGcUl+yen8rPyRaVaatIqU4VFpYpkVB5zt7T0fCY1A1RbCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIBOQAAGAAAAAAAAnYAADAAAAAAAADoBARgAAAAAAAB0AgIwAAAAAAAA6AQEYAAAAAAAANAJCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIBOQAAGAAAAAAAAnYAADAAAAAAAADoBARgAAAAAAAB0AgIwAAAAAAAA6AQEYAAAAAAAANAJCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIBOQAAGAAAAAAAAnYAADAAAAAAAADoBARgAAAAAAAB0AgIwAAAAAAAA6AQEYAAAAAAAANAJCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIBOQAAGAAAAAAAAnYAADAAAAAAAADoBARgAAAAAAAB0AgIwAAAAAAAA6AQEYAAAAAAAANAJCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIBOQAAGAAAAAAAAnYAADAAAAAAAADoBARgAAAAAAAB0AgIwAAAAAAAA6AQEYAAAAAAAANAJCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIBOQAAGAAAAAAAAnYAADAAAAAAAADoBARgAAAAAAAB0AgIwAAAAAAAA6AQEYAAAAAAAANAJCMAAAAAAAACgExCAAQAAAAAAQCcgAAMAAAAAAIAOUKn+HwCdQ//3W1ufAAAAAElFTkSuQmCC"}}},{"cell_type":"code","source":"#groupby jockey,horse to get tot_races_by_jockey\nraces = all_tracks_clusters[(all_tracks_clusters.trakus_index == 10)][['race_id','official_pos','jockey','cluster']]\ntot_r_by_j = races.groupby(['jockey']).agg(tot_races_by_jockey=('race_id','count'))\ntot_r_by_j.head()\n\nraces = pd.merge(races,tot_r_by_j,on=['jockey'])\nraces.head()\n\n#groupby jockey,horse,cluster to get tot_races_by_jockey_cluster\ntot_r_by_j_c = races.groupby(['jockey','cluster']).agg(tot_races_by_jockey_cluster=('race_id','count'))\nraces = pd.merge(races,tot_r_by_j_c,on=['cluster','jockey'])\n\n#groupby jockey,cluster where official_pos == 1 cnt/tot_races_by_jockey_cluster\ntot_w_by_j_c = races[(races.official_pos == 1)].groupby(['jockey','cluster']).agg(tot_wins_by_jockey_cluster=('race_id','count'))\nraces = pd.merge(races,tot_w_by_j_c,on=['cluster','jockey'])\n\n#races.reset_index(inplace=True)\njdata = races[['jockey','cluster','tot_wins_by_jockey_cluster','tot_races_by_jockey_cluster','tot_races_by_jockey']].drop_duplicates()\n\nif isinstance(jdata.index, pd.MultiIndex):\n    jdata.reset_index(inplace=True)\n\njdata['wi_pct_style'] = jdata.tot_wins_by_jockey_cluster / jdata.tot_races_by_jockey_cluster\n#jdata['wi_pct_overall'] = jdata.tot_wins_by_jockey_cluster / jdata.tot_races_by_jockey_cluster\n#jdata.set_index(['jockey','tot_races_by_jockey','cluster'],inplace=True)\n#jdata.sort_index()\n#groupby jockey,cluster where official_pos < 4 cnt/tot_races_by_jockey_cluster\n\nif isinstance(jdata.index, pd.MultiIndex):\n    jdata.reset_index(inplace=True)\n    \njdata = jdata.reset_index() \\\n       .sort_values(['tot_races_by_jockey','jockey','cluster'], ascending=[False,True,True]) \\\n       .set_index(['tot_races_by_jockey','jockey','cluster'])\n\ndisplay(jdata.head(30))\n#shortdf=jdata.head(20)\n#dfi.export(shortdf, 'jockeys.png')\n","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:04.258535Z","iopub.execute_input":"2022-11-09T21:48:04.259755Z","iopub.status.idle":"2022-11-09T21:48:04.359587Z","shell.execute_reply.started":"2022-11-09T21:48:04.259712Z","shell.execute_reply":"2022-11-09T21:48:04.358343Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Comments\nUsing win percentages for jockeys is questionable because of the selection bias. The best trainers want to work with the best jockeys. The best jockeys often get the best horses. Further research is needed to explore the correct way to related jockeys to racing styles. The curse of dimensionality raises its ugly head since jockeys can perform differently at each track, course, condition, distance, etc. Still, this seems like a big data problem that can be tackled by machine learning methods.","metadata":{}},{"cell_type":"markdown","source":"## Velocity is good. Is Endurance better?\nThe fastest horse wins, right? Often, winning is all about decelerating more slowly than your competitors. Almost invariably, horses are constantly slowing down after hitting their peak velocity about 20 seconds into a race. Intuitively, horses that can stay at or near their peak velocities for the longest period will win more often than horses that decelerate quickly after reaching their peak. But what does the data say?\n\nThe competition data enables us to look at velocity, acceleration and deceleration rates. We can measure the interval that starts when a horse reaches peak velocity to when it drops below some increment below that value. \n\nIn the following chart, we see that for almost all distances, the race winner's peak interval is longer than any of its close competitors. **Note that this analysis is independent of the actual velocity of the horse.** The winner just has to have a peak interval longer than its rivals (though it is assumed the winner's peak velocity is at least close to the others').\n\nThis data suggests that training to maximize peak interval length is very important.","metadata":{}},{"cell_type":"code","source":"unstacked = get_peak_intervals(small_df,within_pct=0.05)\ndisplay(unstacked)","metadata":{"scrolled":true,"execution":{"iopub.status.busy":"2022-11-09T21:48:04.361353Z","iopub.execute_input":"2022-11-09T21:48:04.361812Z","iopub.status.idle":"2022-11-09T21:48:29.955285Z","shell.execute_reply.started":"2022-11-09T21:48:04.361777Z","shell.execute_reply":"2022-11-09T21:48:29.953867Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Enhanced Past Performance Analysis\nPast performance analysis is a fundamental practice by anyone that wants to understand a given horse's capabilities, especially bettors who want to predict the outcome of a future race. There is a wealth of historic data available for each horse's previous races. Many years of data are available from Equibase and other providers. Professional handicappers use this data to build complex models and to generate value-added attributes and predictions that they market to bettors. Historically, the in-race performance data includes a series of observations made by track staff. These observations are made at fixed intervals (points of call) depending on the race length. \n\nThere are typicall 4-5 points of call and the observations include:\n- fractional time of the leader\n- lengths behind the leader\n- race position\n\nWith this data you can derive fractional times for non-leaders, but that is always an estimate. In many ways, this whole process of recording and storing observations is \"pre-computer\". What happens between the points is virtually a black box.\n\nThe Trakus data fast forwards us to the \"big data\" era and provides so much more information. In terms of past perforance analyis, you now can look at hundreds of points of calls and get precise observations about distances traveled, velocity, positions, etc. It's a brave new world for past performance data.\n\nThe competition sponsors chose not to include horse names in the data set, providing program names only. But with some manual data collection using Equibase sites we can find the horses names for a given race, and then we can use those names to find other races that include those horses. Fortunately, horses commonly run on all the tracks in the competition data set. With this scraped information we can then create \"past performance\" data sets from the competition data.\n\nWe did just that, starting with the 6th race at Aqueduct on 12/20/2019 -- a claiming race for three year olds and upward which have never run three races. The horses included 'PRIORITIZE','FAMILY BIZ','SARGEANT DRIVE','INSCOM','ALLURED','VARIANT PERCEPTION' and 'RUCKSACK'. We then searched online and found the other races these horses completed in order to get there program numbers. Once we have the races and program numbers, we connected them to the competition data set to retrieve all the available past performances.\n\nIn the analysis below we look this race, and drill down on the horse named 'FAMILY BIZ'.","metadata":{}},{"cell_type":"code","source":"# read in the file that maps race+program IDs to the real horses\ndateparse = lambda x: datetime.strptime(x, '%Y%m%d')\nsample_df = pd.read_csv(SAMPLE_RACE,sep=',',dtype={\"program_number\":str},\n                      parse_dates=['race_date'], date_parser=dateparse)\n\n# use only the race and horse identifiers to find all the races in 2019 for the horses in the race\nsh_df  = sample_df[['track_id','race_date','race_number','program_number']]\n\n#cols = ['track_id','race_date','race_number','program_number']\ndf3 = pd.merge(nyra, sh_df, on=['track_id','race_date','race_number','program_number'], how=\"outer\", indicator=True)\ndf4 = df3.loc[df3[\"_merge\"] == \"both\"].drop(\"_merge\", axis=1)\ndf4 = pd.merge(df4,sample_df, on=['track_id','race_date','race_number','program_number'])\n\n# df4 contains all the races for all the sample horses\n# from this set, get the data just for our sample race\nsample_race_data = df4[(df4.track_id == 'AQU') & (df4.race_date == '20191220') & (df4.race_number == 6)]\n\n# get the data for one horse in the sample race\nhname = 'FAMILY BIZ'\nfamily_biz = df4[(df4.horse_name == hname)]","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:29.957352Z","iopub.execute_input":"2022-11-09T21:48:29.958412Z","iopub.status.idle":"2022-11-09T21:48:44.480228Z","shell.execute_reply.started":"2022-11-09T21:48:29.958291Z","shell.execute_reply":"2022-11-09T21:48:44.478119Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Let's look first at the speeds each horse ran over the race. Incidentally, the finish order was \"PRIORITIZE\" (by 1 1/2 lengths), \"INSCOM\" and \"FAMILY BIZ. The winner finished in 99 seconds","metadata":{}},{"cell_type":"code","source":"plot_horse_velocity(sample_race_data,keycol='horse_name',cols=['cum_time','horse_name'],\n                    suptitle='Trakus Data: Aqueduct - 2019-12-20 - Race 6')","metadata":{"scrolled":true,"execution":{"iopub.status.busy":"2022-11-09T21:48:44.481865Z","iopub.execute_input":"2022-11-09T21:48:44.482297Z","iopub.status.idle":"2022-11-09T21:48:45.070964Z","shell.execute_reply.started":"2022-11-09T21:48:44.482262Z","shell.execute_reply":"2022-11-09T21:48:45.069967Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Looking at this one chart, you may not find it extremely useful for predicting the future. You can see that at the 99 second mark that PRIORITIZE was not the fasted horse. But the data can be used to precisely describe what happens when.\n\nLet's compare that chart with what would typically be available with traditional data sets, i.e., 4-5 points of calls. We can simulate that with the competition data by extract the approximate points for each horse. The chart belows paints a vivid picture.","metadata":{}},{"cell_type":"code","source":"split_ints=[98,195,296,300,396]\nsample_race_old_data = df4[(df4.track_id == 'AQU') & (df4.race_date == '20191220') \n                           & (df4.race_number == 6) & (df4.trakus_index.isin(split_ints) )]\n#sample_race_old_data.head()\nplot_horse_velocity(sample_race_old_data,keycol='horse_name',cols=['cum_time','horse_name'],\n                   suptitle='Without Trakus: Aqueduct - 2019-12-20 - Race 6')","metadata":{"scrolled":true,"execution":{"iopub.status.busy":"2022-11-09T21:48:45.072227Z","iopub.execute_input":"2022-11-09T21:48:45.073064Z","iopub.status.idle":"2022-11-09T21:48:45.926760Z","shell.execute_reply.started":"2022-11-09T21:48:45.073004Z","shell.execute_reply":"2022-11-09T21:48:45.924821Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"A picture is worth a thousand words. With the less complete data we don't get to see what is really happening during most of the race.\n\n### Our Feature Horse\nNow let's take a look at our feature horse, \"FAMILY BIZ\". Below are charts of past performances by our horse in 2019. We can see that races are at multiple distances. We have not done anything to account for this in our analysis; a more thorough study should.\n\nAgain, the differences between the charts are obvious. Machine learning loves big data such as the Trakus data.","metadata":{}},{"cell_type":"code","source":"\nplot_old_vs_new_race(family_biz,split_ints=split_ints)\n","metadata":{"scrolled":true,"execution":{"iopub.status.busy":"2022-11-09T21:48:45.928502Z","iopub.execute_input":"2022-11-09T21:48:45.928891Z","iopub.status.idle":"2022-11-09T21:48:46.935884Z","shell.execute_reply.started":"2022-11-09T21:48:45.928862Z","shell.execute_reply":"2022-11-09T21:48:46.934304Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"As you can see, the Trakus enhanced data gives a much more complete and detailed view of the races. We can use machine learning to calculate the regression lines for every horse in the race. The regression line is a line that optimally describes the behavior of a set of data, i.e. it best fits the trend of the data. We will need to use polynomial regression to fit the velocity times series of the racehorse. A key input into learning algorithms is the _degrees_ of the polynomial expression. Some experimentation suggest that we use a 5th degree polynomial.\n\nThe result of the polynomial regression for our selected horse is shown below.","metadata":{}},{"cell_type":"markdown","source":"family_biz_old_data = family_biz[(family_biz.trakus_index.isin(split_ints) )]\nplot_one_horse_regression(family_biz_old_data,degrees=2)","metadata":{}},{"cell_type":"code","source":"# plot the regression line for a sample horse\nplot_one_horse_regression(family_biz,degrees=5)\n","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:46.942414Z","iopub.execute_input":"2022-11-09T21:48:46.943976Z","iopub.status.idle":"2022-11-09T21:48:47.332794Z","shell.execute_reply.started":"2022-11-09T21:48:46.943875Z","shell.execute_reply":"2022-11-09T21:48:47.331367Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Likewise we can calculate the past performance regression model for every horse in the race. The chart below shows the lines for our example race. Note, the model can go a bit wild predicting intervals we don't have data for (liek after the race). We've excluded the selected 12-20 race from our model.","metadata":{}},{"cell_type":"code","source":"horsenames=['PRIORITIZE','FAMILY BIZ','SARGEANT DRIVE','INSCOM','ALLURED','VARIANT PERCEPTION','RUCKSACK']\n\npast_perf = df4[(df4.race_date < '20191220')]\nplot_regresssion_lines(past_perf,horsenames,degrees=5)\n  ","metadata":{"scrolled":true,"execution":{"iopub.status.busy":"2022-11-09T21:48:47.333998Z","iopub.execute_input":"2022-11-09T21:48:47.334335Z","iopub.status.idle":"2022-11-09T21:48:47.865923Z","shell.execute_reply.started":"2022-11-09T21:48:47.334305Z","shell.execute_reply":"2022-11-09T21:48:47.864500Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"It's just one race and one chart, but it these line appear to predict the top two finishers. Further research is suggested.","metadata":{}},{"cell_type":"markdown","source":"\n### Enhanced Past Performance Models and Running Styles\nWe've shown that we can build running style models for sets of races using multiple filters/parameters. We've shown a correlation between running styles and race winners. Now we've shown how you can build rich past performance regression models for each horse. Combining all this data creates opportunities to research:\n- The use of the enhanced past performance regression models to map horses to running styles\n- The use of the enhanced past performance regression models to predict the outcome of races\n- Use horses' past performance regression models to build clusters of similar horses. Perhaps a K Nearest Neighbors alogoritm would be a good choice.\n\nThese are just a few of the possibilities. We look forward to a future competition where Trakus past performance data is provided","metadata":{}},{"cell_type":"markdown","source":"## Get Out of My Way: Impedance \nChanging gears, let's consider another way to use the tracking data. When horses are caught in traffic they sometimes have to pull up quickly, and then must accelerate to regain position. The traffic could be in front, or perhaps they are bumped by another horse. With the granular coordinate data we can calculate moving averages for velocity and acceleration. Sudden and sustained drops in acceleration may indicate impedance. By identifying impedance we can analyze for its effect on a horse's chance to win.","metadata":{}},{"cell_type":"code","source":"# calculate the moving accelerations\nsmall_df['moving_acc'] = small_df.groupby(['track_id','race_date','race_number','program_number'])['acceleration'].transform(lambda x: x.rolling(8, 1).mean())\n\nsome_horses = small_df[(small_df.trakus_index > 40) & (small_df.moving_acc < -10)][['track_id','race_date',\n                            'race_number','program_number']].drop_duplicates().head(10)\n\nimpeded = pd.merge(small_df,some_horses,on=(['track_id','race_date',\n                            'race_number','program_number']))\n\n\nfig, ax = plt.subplots(figsize=(15,7))\n\nfor key, grp in impeded.groupby(['track_id','race_date',\n                            'race_number','program_number']):\n    ax = grp.plot(ax=ax, kind='line', x='cum_time', y='moving_velocity')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:47.867423Z","iopub.execute_input":"2022-11-09T21:48:47.868730Z","iopub.status.idle":"2022-11-09T21:48:57.846548Z","shell.execute_reply.started":"2022-11-09T21:48:47.868685Z","shell.execute_reply":"2022-11-09T21:48:57.845040Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Comments\n\nI'm not sure there is much to do with this impedance detection by itself. Basically, it's not good to get bumped or impeded! It may be able to identify subtle impedance that is not easily quantified by observers.\n\nThat said, if you also bring in horse path data and build traffic models, you may be able to predict the potential for impedance in races with horses that have different running styles. If you also use the path data in combination with wind data, you can measure the relative importance of drafting. Then you could test models that compare the risk of traffic impedance versus the risk of wind impedance. Perhaps the model could suggest something like \"if the wind is higher than 10 mph, draft whenever possible, even though the risk of traffic impedance increases.\"\n\n## Course Consistency\nThe competition tracking data can give us perspectives on the nature of the course surfaces. Ideally, you might hope that the consistency of the surface (dirt,turf,etc.) would be uniform throughout the race. This is not necessarily true. If you, as a track manager, strive for consistency, then you can use the tracking data to identify portions of the track where average velocities change. For example, if horses' average velocity dips between 1000 and 1500 feet, perhaps the dirt/sand mixture needs tuning. \n\nConsider the chart below for 1M races on Turf at Aqueduct and Belmont. Intuitively, the smoothness of the Belmont line seems like the goal. The Belmont line shows a couple noticeable dips at specific places on the track (as measured from the start).\n\nNote: the charts don't take into account turns. Belmont with its larger turns probably promotes a smoother journey.","metadata":{}},{"cell_type":"code","source":"dist=1760\ninterval=2\n\n\npoverd_df = nyra[ (nyra.distance_yds == dist) & (nyra.course_type == 'T' )\n          & (nyra.trakus_index % interval == 0) \n          & (nyra.velocity < 65)][['track_id','cum_dtraveled','moving_velocity']]\n    \nplot_v_over_dist(poverd_df,ax_title='Distance: 1760 yds - Turf')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:57.848690Z","iopub.execute_input":"2022-11-09T21:48:57.849749Z","iopub.status.idle":"2022-11-09T21:48:58.870890Z","shell.execute_reply.started":"2022-11-09T21:48:57.849709Z","shell.execute_reply":"2022-11-09T21:48:58.869387Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"dist=1760\npoverd_df = nyra[ (nyra.distance_yds == dist) & (nyra.course_type == 'D' ) & (nyra.track_condition == 'MY' )\n          & (nyra.trakus_index % interval == 0)\n         & (nyra.cum_dtraveled > 0)][['track_id','cum_dtraveled','moving_velocity']]\n\nplot_v_over_dist(poverd_df,ax_title='Distance: 1760 yds - Dirt - Muddy')","metadata":{"execution":{"iopub.status.busy":"2022-11-09T21:48:58.872246Z","iopub.execute_input":"2022-11-09T21:48:58.872594Z","iopub.status.idle":"2022-11-09T21:49:00.201555Z","shell.execute_reply.started":"2022-11-09T21:48:58.872565Z","shell.execute_reply":"2022-11-09T21:49:00.200014Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"The above chart again shows that Belmont is seemingly more consistent throughout the race, while a section of Aqueduct (the stretch???) is less so.\n\n### Comments\nThe Trakus data presents a much more detailed view of how horses run over each distance. Before it was available, analysts had just a few points of call and the intervals between those points were black boxes.\n\nGiven the availability of horse path data, you could extend the analysis above to specific paths on each course. Thus, you may be able to show that the rail is a poor path on the Turf course at Saratoga, but only between the two points around the first turn. Such analysis may be even more significant when the conditions are poor.\n\n## Summary & Future Opportunities\nWe set out to explore how the detailed tracking data in the competition sets can enable new types of analysis and refine existing types. The analysis of running styles suggests that multivariate custom models of classes can be developed using machine learning. Further research is necessary to consider:\n- Can the generated models be used to predict future race outcomes?\n- How could the models be used to generate features for more complex prediction processes?\n- What are the appropriate combinations of variables (track, course, race type, distance, conditions,...) to build models for?\n- How to best use a running style model to devise a plan for a given race?\n- How can you use past performance data to assign horses to specific running style models?\n- How do the custom models and styles compare to styles developed by handicap data providers such as Brisnet?\n- What are best clustering algorithms to use? Which algorithm options and parameters should be used?\n- How can the betting public use any new style classes?\n\nCombining the competition data with horse path data, that includes information on track geometry, distance of the rail, etc., (plus throw in precipitation, wind and temperature data) enables further research into:\n- Importance of drafting under certain circumstances\n- The effect of post position on the success of any running styles\n- Track and path consistency\n\nTaken all together, the detailed sets of data could support research into large-scale models of how races are run. We're sure stakeholders have models now, but those models are inherently limited by having data for a very limited set of in-race observations. \n\n### A New Experience???\nImagine for a moment that you have a real-time system that monitors each horse in the race (i.e. Trakus). Now imagine if that system could calculate odds for each horse at each interval. The odds would be updated based on complex models that use both the horses' past performance data and current race status. Imagine if bettors could place multiple bets **during** the race based on these dynamic odds, at least up to the point where the system decides that results are locked in. The race becomes an interactive experience! Online betting sites already support in-game wagers in other sports. Maybe long distance racing will become more popular","metadata":{}}]}