Skip to content

Instantly share code, notes, and snippets.

@takuti

takuti/20181016-mbed-connect.ipynb Secret

Last active October 17, 2018 17:05
Show Gist options
  • Star 1 You must be signed in to star a gist
  • Fork 1 You must be signed in to fork a gist
  • Save takuti/6edab4a504368a0e58bb81a3158a1e07 to your computer and use it in GitHub Desktop.
Save takuti/6edab4a504368a0e58bb81a3158a1e07 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
20181016-mbed-connect.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Mbed Connect Workshop: Introduction to Data Science\n",
"\n",
"- This notebook is available online at: **https://goo.gl/u6e3EN**\n",
"- WiFi SSID: **ArmTechCon2018** (no password required)\n",
"\n",
"## Hello!\n",
"\n",
"- **Takuya Kitazawa**, Data Science Engineer at [Arm Treasure Data](https://treasuredata.com)\n",
" - https://takuti.me\n",
" - [@takuti](https://twitter.com/takuti)\n",
" - kitazawa [at] treasure-data.com\n",
"\n",
"### Key takeaway points\n",
"\n",
"1. Understand what Arm Treasure Data (TD) Customer Data Platform (CDP) is\n",
" - Data management layer of the Arm Pelion IoT platform\n",
"2. Learn how machine learning (ML) and data science works\n",
" - Capture characteristics of historical data, and predict unseen result\n",
"3. See every single steps of real-world data science workflow on TD\n",
" - IoT-ish sample scenario based on real-life environmental data from City of Chicago\n",
" \n",
"### Slides \n",
"\n",
"https://treasure-data.app.box.com/v/20181016-mbed-hivemall (Link will be expired on Oct 26)\n",
"\n",
"- [Apache Hivemall](https://github.com/apache/incubator-hivemall)\n",
" - [Documentation](http://hivemall.incubator.apache.org/userguide/)\n",
" - [Step-by-step ML tutorial](http://hivemall.incubator.apache.org/userguide/supervised_learning/tutorial.html)\n",
"- [Digdag](https://www.digdag.io/) (Treasure Workflow)\n",
" - [TD ML workflow examples](https://github.com/treasure-data/workflow-examples/tree/master/machine-learning)\n",
"\n",
"## Real-world ML workflow demo and hands-on\n",
"\n",
"1. Access and login to TD console: https://console.treasuredata.com/users/sign_in\n",
"2. Create your working database: `mbed_XX` (`XX` is your ID)\n",
"\n",
"### Step 1: Data ingestion\n",
"\n",
"Scenario: City of Chicago had conducted [energy benchmarking](https://www.cityofchicago.org/city/en/progs/env/building-energy-benchmarking---transparency.html), and they have its resulting data.\n",
"\n",
"1. Get [City of Chicago energy benchmarking data](https://data.cityofchicago.org/Environment-Sustainable-Development/Chicago-Energy-Benchmarking/xq83-jr8c)\n",
"2. Go to [connection catalog](https://console.treasuredata.com/app/integrations/catalog) on TD console and upload the dump CSV file to your own database\n",
" - Database: `mbed_XX`, Table: `energy_benchmarking`\n",
" - Advanced setting: Rename \"# of Buildings\" to \"Num of Buildings\" so that column names do not contain irregular characters\n",
"\n",
"### Step 2: Define your problem\n",
"\n",
"The city gives rating to each building based on measured energy efficiency, and [encourages participants to improve the efficiency](https://www.cityofchicago.org/city/en/depts/mayor/supp_info/chicago-energy-benchmarking/Chicago_Energy_Benchmarking_Beyond_Benchmarking.html); it's a ***post*** action.\n",
"\n",
"What can we do ***before*** actually consuming tons of energy resources and receiving certain evaluation results? -- One possibility is to **predict future energy consumption** (electricity use, in our case), and take environmental-friendly actions as soon as possible."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Step 3: Understand your data and determin how to approach to the problem\n",
"\n",
"#### With Presto ad-hoc queries\n",
"\n",
"Go to your database page on the console, and hit `QUERY` button to work on your data.\n",
"\n",
"Query language is almost same as the standard SQL. See [Presto aggregation functions documentation](https://prestodb.io/docs/current/functions/aggregate.html), for example.\n",
"\n",
"- [What each column means](https://data.cityofchicago.org/Environment-Sustainable-Development/Chicago-Energy-Benchmarking/xq83-jr8c)\n",
"- Total number of records\n",
" ```sql\n",
" select count(1) from energy_benchmarking\n",
" ```\n",
"- Benchmarking time period and frequency\n",
" ```sql\n",
" select data_year, count(1) from energy_benchmarking group by 1\n",
" ```\n",
"- Distribution of different `community_area` and `primary_property_type`\n",
" ```sql\n",
" select community_area, count(1) from energy_benchmarking group by 1\n",
" ```\n",
" ```sql\n",
" select primary_property_type, count(1) from energy_benchmarking group by 1\n",
" ```\n",
"- Max and min values in `num_of_buildings` and `electricity_use__kbtu_` columns\n",
" ```sql\n",
" select \n",
" min(num_of_buildings), max(num_of_buildings),\n",
" min(electricity_use__kbtu_), max(electricity_use__kbtu_)\n",
" from\n",
" energy_benchmarking\n",
" ```\n",
"- Missing value rate in each `*_use__kbtu_*` (kBtu; thousand British thermal units) column\n",
" ```sql\n",
" select \n",
" sum(if(electricity_use__kbtu_ is null, 1, 0)) / cast(count(1) as double)\n",
" from\n",
" energy_benchmarking\n",
" ```\n",
"\n",
"Anything else?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### By using [pandas-td](https://github.com/treasure-data/pandas-td) and Jupyter notebook\n",
"\n",
"```sh\n",
"pip install pandas-td\n",
"```\n",
"\n",
"Load basic Python data science packages:"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"%matplotlib inline\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import pandas as pd"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Setup connection between TD and your notebook:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import os\n",
"import pandas_td as td\n",
"con = td.connect(apikey=os.environ['TD_API_KEY'], endpoint=os.environ['TD_API_SERVER'])\n",
"presto = td.create_engine('presto:chicago_smart_green', con=con)\n",
"hive = td.create_engine('hive:chicago_smart_green', con=con)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Read the benchmarking table:"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>data_year</th>\n",
" <th>id</th>\n",
" <th>property_name</th>\n",
" <th>address</th>\n",
" <th>zip_code</th>\n",
" <th>community_area</th>\n",
" <th>primary_property_type</th>\n",
" <th>gross_floor_area___buildings__sq_ft_</th>\n",
" <th>year_built</th>\n",
" <th>num_of_buildings</th>\n",
" <th>...</th>\n",
" <th>site_eui__kbtu_sq_ft_</th>\n",
" <th>source_eui__kbtu_sq_ft_</th>\n",
" <th>weather_normalized_site_eui__kbtu_sq_ft_</th>\n",
" <th>weather_normalized_source_eui__kbtu_sq_ft_</th>\n",
" <th>total_ghg_emissions__metric_tons_co2e_</th>\n",
" <th>ghg_intensity__kg_co2e_sq_ft_</th>\n",
" <th>latitude</th>\n",
" <th>longitude</th>\n",
" <th>location</th>\n",
" <th>time</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>2014</td>\n",
" <td>157767</td>\n",
" <td>Roosevelt Hi-CPS</td>\n",
" <td>3436 W WILSON AVE</td>\n",
" <td>60625</td>\n",
" <td>ALBANY PARK</td>\n",
" <td>K-12 School</td>\n",
" <td>319900</td>\n",
" <td>1927</td>\n",
" <td>1</td>\n",
" <td>...</td>\n",
" <td>80.0</td>\n",
" <td>129.0</td>\n",
" <td>75.0</td>\n",
" <td>122.0</td>\n",
" <td>2379.0</td>\n",
" <td>7.44</td>\n",
" <td>41.965013</td>\n",
" <td>-87.714513</td>\n",
" <td>(41.96501325, -87.71451268)</td>\n",
" <td>1539397863</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>2014</td>\n",
" <td>100256</td>\n",
" <td>Curie Metrop-CPS</td>\n",
" <td>4975 S Archer</td>\n",
" <td>60632</td>\n",
" <td>ARCHER HEIGHTS</td>\n",
" <td>K-12 School</td>\n",
" <td>447330</td>\n",
" <td>1990</td>\n",
" <td>1</td>\n",
" <td>...</td>\n",
" <td>96.0</td>\n",
" <td>268.0</td>\n",
" <td>93.0</td>\n",
" <td>261.0</td>\n",
" <td>7573.0</td>\n",
" <td>16.93</td>\n",
" <td>41.802759</td>\n",
" <td>-87.722671</td>\n",
" <td>(41.802759, -87.722671)</td>\n",
" <td>1539397863</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>2014</td>\n",
" <td>101551</td>\n",
" <td>METROPOLITIAN PIER AND EXPOSITION AUTHORITY</td>\n",
" <td>301 Cermak Road</td>\n",
" <td>60616</td>\n",
" <td>ARMOUR SQUARE</td>\n",
" <td>Convention Center</td>\n",
" <td>9245333</td>\n",
" <td>1971</td>\n",
" <td>8</td>\n",
" <td>...</td>\n",
" <td>90.0</td>\n",
" <td>204.0</td>\n",
" <td>86.0</td>\n",
" <td>200.0</td>\n",
" <td>115833.0</td>\n",
" <td>12.53</td>\n",
" <td>41.852840</td>\n",
" <td>-87.634931</td>\n",
" <td>(41.85284, -87.634931)</td>\n",
" <td>1539397863</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>2014</td>\n",
" <td>100396</td>\n",
" <td>St. Rita of Cascia High School</td>\n",
" <td>7740 South Western Avenue</td>\n",
" <td>60620</td>\n",
" <td>ASHBURN</td>\n",
" <td>K-12 School</td>\n",
" <td>250000</td>\n",
" <td>1960</td>\n",
" <td>1</td>\n",
" <td>...</td>\n",
" <td>84.0</td>\n",
" <td>139.0</td>\n",
" <td>80.0</td>\n",
" <td>135.0</td>\n",
" <td>2010.0</td>\n",
" <td>8.04</td>\n",
" <td>41.752003</td>\n",
" <td>-87.684637</td>\n",
" <td>(41.7520028, -87.68463714)</td>\n",
" <td>1539397863</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>2014</td>\n",
" <td>250052</td>\n",
" <td>Steinmetz Ac-CPS</td>\n",
" <td>3030 N MOBILE AVE</td>\n",
" <td>60634</td>\n",
" <td>BELMONT CRAGIN</td>\n",
" <td>K-12 School</td>\n",
" <td>323400</td>\n",
" <td>1934</td>\n",
" <td>1</td>\n",
" <td>...</td>\n",
" <td>85.0</td>\n",
" <td>142.0</td>\n",
" <td>80.0</td>\n",
" <td>136.0</td>\n",
" <td>2664.0</td>\n",
" <td>8.24</td>\n",
" <td>41.935603</td>\n",
" <td>-87.784340</td>\n",
" <td>(41.93560335, -87.7843396)</td>\n",
" <td>1539397863</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows × 26 columns</p>\n",
"</div>"
],
"text/plain": [
" data_year id property_name \\\n",
"0 2014 157767 Roosevelt Hi-CPS \n",
"1 2014 100256 Curie Metrop-CPS \n",
"2 2014 101551 METROPOLITIAN PIER AND EXPOSITION AUTHORITY \n",
"3 2014 100396 St. Rita of Cascia High School \n",
"4 2014 250052 Steinmetz Ac-CPS \n",
"\n",
" address zip_code community_area primary_property_type \\\n",
"0 3436 W WILSON AVE 60625 ALBANY PARK K-12 School \n",
"1 4975 S Archer 60632 ARCHER HEIGHTS K-12 School \n",
"2 301 Cermak Road 60616 ARMOUR SQUARE Convention Center \n",
"3 7740 South Western Avenue 60620 ASHBURN K-12 School \n",
"4 3030 N MOBILE AVE 60634 BELMONT CRAGIN K-12 School \n",
"\n",
" gross_floor_area___buildings__sq_ft_ year_built num_of_buildings \\\n",
"0 319900 1927 1 \n",
"1 447330 1990 1 \n",
"2 9245333 1971 8 \n",
"3 250000 1960 1 \n",
"4 323400 1934 1 \n",
"\n",
" ... site_eui__kbtu_sq_ft_ source_eui__kbtu_sq_ft_ \\\n",
"0 ... 80.0 129.0 \n",
"1 ... 96.0 268.0 \n",
"2 ... 90.0 204.0 \n",
"3 ... 84.0 139.0 \n",
"4 ... 85.0 142.0 \n",
"\n",
" weather_normalized_site_eui__kbtu_sq_ft_ \\\n",
"0 75.0 \n",
"1 93.0 \n",
"2 86.0 \n",
"3 80.0 \n",
"4 80.0 \n",
"\n",
" weather_normalized_source_eui__kbtu_sq_ft_ \\\n",
"0 122.0 \n",
"1 261.0 \n",
"2 200.0 \n",
"3 135.0 \n",
"4 136.0 \n",
"\n",
" total_ghg_emissions__metric_tons_co2e_ ghg_intensity__kg_co2e_sq_ft_ \\\n",
"0 2379.0 7.44 \n",
"1 7573.0 16.93 \n",
"2 115833.0 12.53 \n",
"3 2010.0 8.04 \n",
"4 2664.0 8.24 \n",
"\n",
" latitude longitude location time \n",
"0 41.965013 -87.714513 (41.96501325, -87.71451268) 1539397863 \n",
"1 41.802759 -87.722671 (41.802759, -87.722671) 1539397863 \n",
"2 41.852840 -87.634931 (41.85284, -87.634931) 1539397863 \n",
"3 41.752003 -87.684637 (41.7520028, -87.68463714) 1539397863 \n",
"4 41.935603 -87.784340 (41.93560335, -87.7843396) 1539397863 \n",
"\n",
"[5 rows x 26 columns]"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = td.read_td_table('energy_benchmarking', presto)\n",
"df.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Find columns having high missing value rate to avoid using uninformative attributes:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"district_steam_use__kbtu_ 0.954198\n",
"all_other_fuel_use__kbtu_ 0.942748\n",
"district_chilled_water_use__kbtu_ 0.881679\n",
"natural_gas_use__kbtu_ 0.339695\n",
"energy_star_score 0.232824\n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.118321\n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.118321\n",
"ghg_intensity__kg_co2e_sq_ft_ 0.080153\n",
"total_ghg_emissions__metric_tons_co2e_ 0.080153\n",
"site_eui__kbtu_sq_ft_ 0.068702\n",
"source_eui__kbtu_sq_ft_ 0.068702\n",
"electricity_use__kbtu_ 0.064885\n",
"primary_property_type 0.000000\n",
"property_name 0.000000\n",
"address 0.000000\n",
"id 0.000000\n",
"zip_code 0.000000\n",
"community_area 0.000000\n",
"time 0.000000\n",
"gross_floor_area___buildings__sq_ft_ 0.000000\n",
"year_built 0.000000\n",
"num_of_buildings 0.000000\n",
"location 0.000000\n",
"latitude 0.000000\n",
"longitude 0.000000\n",
"data_year 0.000000\n",
"dtype: float64"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"(df.isnull().sum() / df.shape[0]).sort_values(ascending=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Look into uniquness of values in each attribute to find out ID-ish (i.e., almost records take different value) or meaningless (i.e., everyone takes same value) factors:"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# of total records: 262\n",
"(# of unique values, Ratio of unique elements against all records, Column name)\n",
"( 262, 1.00000, natural_gas_use__kbtu_)\n",
"( 262, 1.00000, electricity_use__kbtu_)\n",
"( 262, 1.00000, district_steam_use__kbtu_)\n",
"( 262, 1.00000, district_chilled_water_use__kbtu_)\n",
"( 262, 1.00000, all_other_fuel_use__kbtu_)\n",
"( 261, 0.99618, id)\n",
"( 260, 0.99237, longitude)\n",
"( 260, 0.99237, latitude)\n",
"( 259, 0.98855, total_ghg_emissions__metric_tons_co2e_)\n",
"( 259, 0.98855, gross_floor_area___buildings__sq_ft_)\n",
"( 244, 0.93130, ghg_intensity__kg_co2e_sq_ft_)\n",
"( 190, 0.72519, weather_normalized_source_eui__kbtu_sq_ft_)\n",
"( 176, 0.67176, source_eui__kbtu_sq_ft_)\n",
"( 142, 0.54198, weather_normalized_site_eui__kbtu_sq_ft_)\n",
"( 131, 0.50000, site_eui__kbtu_sq_ft_)\n",
"( 129, 0.49237, energy_star_score)\n",
"( 94, 0.35878, year_built)\n",
"( 13, 0.04962, num_of_buildings)\n",
"( 3, 0.01145, data_year)\n",
"( 1, 0.00382, time)\n"
]
}
],
"source": [
"# uniquness in each column; high cardinaliry or cardinality=1 => useless\n",
"n = df.shape[0]\n",
"print('# of total records: ' + str(n))\n",
"\n",
"numeric_columns = [c for c in df.columns if df[c].dtype == float or df[c].dtype == int]\n",
"\n",
"print('(# of unique values, Ratio of unique elements against all records, Column name)')\n",
"for cardinality, column in sorted([(len(set(df[c])), c) for c in numeric_columns])[::-1]:\n",
" print('(%5d, %.5f, %s)' % (cardinality, cardinality / n, column))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Columns having reasonably small uniquness are worthy to be used as a categorical feature.\n",
"\n",
"Compute correlattion between numerical attributes:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>data_year</th>\n",
" <th>id</th>\n",
" <th>gross_floor_area___buildings__sq_ft_</th>\n",
" <th>year_built</th>\n",
" <th>num_of_buildings</th>\n",
" <th>energy_star_score</th>\n",
" <th>electricity_use__kbtu_</th>\n",
" <th>natural_gas_use__kbtu_</th>\n",
" <th>district_steam_use__kbtu_</th>\n",
" <th>district_chilled_water_use__kbtu_</th>\n",
" <th>all_other_fuel_use__kbtu_</th>\n",
" <th>site_eui__kbtu_sq_ft_</th>\n",
" <th>source_eui__kbtu_sq_ft_</th>\n",
" <th>weather_normalized_site_eui__kbtu_sq_ft_</th>\n",
" <th>weather_normalized_source_eui__kbtu_sq_ft_</th>\n",
" <th>total_ghg_emissions__metric_tons_co2e_</th>\n",
" <th>ghg_intensity__kg_co2e_sq_ft_</th>\n",
" <th>latitude</th>\n",
" <th>longitude</th>\n",
" <th>time</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>data_year</th>\n",
" <td>1.000000</td>\n",
" <td>0.232191</td>\n",
" <td>-0.215156</td>\n",
" <td>-0.100139</td>\n",
" <td>-0.031533</td>\n",
" <td>-0.106661</td>\n",
" <td>-0.064666</td>\n",
" <td>-0.018920</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>-0.005264</td>\n",
" <td>-0.039869</td>\n",
" <td>0.001192</td>\n",
" <td>-0.036829</td>\n",
" <td>-0.029515</td>\n",
" <td>-0.046587</td>\n",
" <td>0.020637</td>\n",
" <td>-0.081256</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>id</th>\n",
" <td>0.232191</td>\n",
" <td>1.000000</td>\n",
" <td>-0.187433</td>\n",
" <td>0.017634</td>\n",
" <td>-0.049704</td>\n",
" <td>-0.317108</td>\n",
" <td>-0.133163</td>\n",
" <td>-0.089715</td>\n",
" <td>-0.223803</td>\n",
" <td>-0.138594</td>\n",
" <td>-0.078548</td>\n",
" <td>0.120588</td>\n",
" <td>0.084036</td>\n",
" <td>0.128526</td>\n",
" <td>0.089546</td>\n",
" <td>-0.120194</td>\n",
" <td>0.082777</td>\n",
" <td>-0.210518</td>\n",
" <td>-0.116963</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>gross_floor_area___buildings__sq_ft_</th>\n",
" <td>-0.215156</td>\n",
" <td>-0.187433</td>\n",
" <td>1.000000</td>\n",
" <td>0.103160</td>\n",
" <td>0.299070</td>\n",
" <td>0.073972</td>\n",
" <td>0.874597</td>\n",
" <td>0.585381</td>\n",
" <td>0.899679</td>\n",
" <td>0.686560</td>\n",
" <td>0.282483</td>\n",
" <td>-0.008782</td>\n",
" <td>0.010984</td>\n",
" <td>-0.030422</td>\n",
" <td>-0.005838</td>\n",
" <td>0.862229</td>\n",
" <td>0.025442</td>\n",
" <td>0.011751</td>\n",
" <td>0.120967</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>year_built</th>\n",
" <td>-0.100139</td>\n",
" <td>0.017634</td>\n",
" <td>0.103160</td>\n",
" <td>1.000000</td>\n",
" <td>-0.034150</td>\n",
" <td>-0.167416</td>\n",
" <td>0.204463</td>\n",
" <td>0.014824</td>\n",
" <td>0.327664</td>\n",
" <td>0.294497</td>\n",
" <td>-0.787941</td>\n",
" <td>0.130847</td>\n",
" <td>0.282366</td>\n",
" <td>0.153242</td>\n",
" <td>0.292059</td>\n",
" <td>0.159934</td>\n",
" <td>0.307693</td>\n",
" <td>-0.115725</td>\n",
" <td>0.001669</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>num_of_buildings</th>\n",
" <td>-0.031533</td>\n",
" <td>-0.049704</td>\n",
" <td>0.299070</td>\n",
" <td>-0.034150</td>\n",
" <td>1.000000</td>\n",
" <td>-0.137488</td>\n",
" <td>0.212108</td>\n",
" <td>0.583969</td>\n",
" <td>0.864088</td>\n",
" <td>NaN</td>\n",
" <td>0.995715</td>\n",
" <td>0.138065</td>\n",
" <td>0.076069</td>\n",
" <td>0.109415</td>\n",
" <td>0.058035</td>\n",
" <td>0.315519</td>\n",
" <td>0.060828</td>\n",
" <td>0.019664</td>\n",
" <td>-0.179843</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>energy_star_score</th>\n",
" <td>-0.106661</td>\n",
" <td>-0.317108</td>\n",
" <td>0.073972</td>\n",
" <td>-0.167416</td>\n",
" <td>-0.137488</td>\n",
" <td>1.000000</td>\n",
" <td>-0.142036</td>\n",
" <td>-0.363684</td>\n",
" <td>-0.570782</td>\n",
" <td>0.007535</td>\n",
" <td>-0.887757</td>\n",
" <td>-0.659919</td>\n",
" <td>-0.648452</td>\n",
" <td>-0.664027</td>\n",
" <td>-0.658261</td>\n",
" <td>-0.217121</td>\n",
" <td>-0.628839</td>\n",
" <td>0.220335</td>\n",
" <td>0.008902</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>electricity_use__kbtu_</th>\n",
" <td>-0.064666</td>\n",
" <td>-0.133163</td>\n",
" <td>0.874597</td>\n",
" <td>0.204463</td>\n",
" <td>0.212108</td>\n",
" <td>-0.142036</td>\n",
" <td>1.000000</td>\n",
" <td>0.631519</td>\n",
" <td>0.952175</td>\n",
" <td>0.632375</td>\n",
" <td>0.694753</td>\n",
" <td>0.195503</td>\n",
" <td>0.285464</td>\n",
" <td>0.167008</td>\n",
" <td>0.258731</td>\n",
" <td>0.973492</td>\n",
" <td>0.302830</td>\n",
" <td>0.011699</td>\n",
" <td>0.107704</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>natural_gas_use__kbtu_</th>\n",
" <td>-0.018920</td>\n",
" <td>-0.089715</td>\n",
" <td>0.585381</td>\n",
" <td>0.014824</td>\n",
" <td>0.583969</td>\n",
" <td>-0.363684</td>\n",
" <td>0.631519</td>\n",
" <td>1.000000</td>\n",
" <td>1.000000</td>\n",
" <td>0.436563</td>\n",
" <td>0.951079</td>\n",
" <td>0.587372</td>\n",
" <td>0.498371</td>\n",
" <td>0.531869</td>\n",
" <td>0.446757</td>\n",
" <td>0.768373</td>\n",
" <td>0.479233</td>\n",
" <td>0.010892</td>\n",
" <td>0.004915</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>district_steam_use__kbtu_</th>\n",
" <td>NaN</td>\n",
" <td>-0.223803</td>\n",
" <td>0.899679</td>\n",
" <td>0.327664</td>\n",
" <td>0.864088</td>\n",
" <td>-0.570782</td>\n",
" <td>0.952175</td>\n",
" <td>1.000000</td>\n",
" <td>1.000000</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.404894</td>\n",
" <td>0.365514</td>\n",
" <td>0.404723</td>\n",
" <td>0.363628</td>\n",
" <td>0.959672</td>\n",
" <td>0.355816</td>\n",
" <td>0.297109</td>\n",
" <td>-0.744476</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>district_chilled_water_use__kbtu_</th>\n",
" <td>NaN</td>\n",
" <td>-0.138594</td>\n",
" <td>0.686560</td>\n",
" <td>0.294497</td>\n",
" <td>NaN</td>\n",
" <td>0.007535</td>\n",
" <td>0.632375</td>\n",
" <td>0.436563</td>\n",
" <td>NaN</td>\n",
" <td>1.000000</td>\n",
" <td>1.000000</td>\n",
" <td>0.029963</td>\n",
" <td>0.086752</td>\n",
" <td>-0.038395</td>\n",
" <td>0.038044</td>\n",
" <td>0.874011</td>\n",
" <td>0.132290</td>\n",
" <td>-0.232749</td>\n",
" <td>0.326126</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>all_other_fuel_use__kbtu_</th>\n",
" <td>NaN</td>\n",
" <td>-0.078548</td>\n",
" <td>0.282483</td>\n",
" <td>-0.787941</td>\n",
" <td>0.995715</td>\n",
" <td>-0.887757</td>\n",
" <td>0.694753</td>\n",
" <td>0.951079</td>\n",
" <td>NaN</td>\n",
" <td>1.000000</td>\n",
" <td>1.000000</td>\n",
" <td>0.834494</td>\n",
" <td>0.716880</td>\n",
" <td>0.914262</td>\n",
" <td>0.916482</td>\n",
" <td>0.847323</td>\n",
" <td>0.680197</td>\n",
" <td>0.468089</td>\n",
" <td>0.350674</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>site_eui__kbtu_sq_ft_</th>\n",
" <td>-0.005264</td>\n",
" <td>0.120588</td>\n",
" <td>-0.008782</td>\n",
" <td>0.130847</td>\n",
" <td>0.138065</td>\n",
" <td>-0.659919</td>\n",
" <td>0.195503</td>\n",
" <td>0.587372</td>\n",
" <td>0.404894</td>\n",
" <td>0.029963</td>\n",
" <td>0.834494</td>\n",
" <td>1.000000</td>\n",
" <td>0.931679</td>\n",
" <td>0.999454</td>\n",
" <td>0.933073</td>\n",
" <td>0.316579</td>\n",
" <td>0.904224</td>\n",
" <td>-0.126103</td>\n",
" <td>0.095082</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>source_eui__kbtu_sq_ft_</th>\n",
" <td>-0.039869</td>\n",
" <td>0.084036</td>\n",
" <td>0.010984</td>\n",
" <td>0.282366</td>\n",
" <td>0.076069</td>\n",
" <td>-0.648452</td>\n",
" <td>0.285464</td>\n",
" <td>0.498371</td>\n",
" <td>0.365514</td>\n",
" <td>0.086752</td>\n",
" <td>0.716880</td>\n",
" <td>0.931679</td>\n",
" <td>1.000000</td>\n",
" <td>0.936193</td>\n",
" <td>0.999612</td>\n",
" <td>0.355438</td>\n",
" <td>0.997439</td>\n",
" <td>-0.094817</td>\n",
" <td>0.110036</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>weather_normalized_site_eui__kbtu_sq_ft_</th>\n",
" <td>0.001192</td>\n",
" <td>0.128526</td>\n",
" <td>-0.030422</td>\n",
" <td>0.153242</td>\n",
" <td>0.109415</td>\n",
" <td>-0.664027</td>\n",
" <td>0.167008</td>\n",
" <td>0.531869</td>\n",
" <td>0.404723</td>\n",
" <td>-0.038395</td>\n",
" <td>0.914262</td>\n",
" <td>0.999454</td>\n",
" <td>0.936193</td>\n",
" <td>1.000000</td>\n",
" <td>0.936855</td>\n",
" <td>0.280278</td>\n",
" <td>0.910003</td>\n",
" <td>-0.136343</td>\n",
" <td>0.091182</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>weather_normalized_source_eui__kbtu_sq_ft_</th>\n",
" <td>-0.036829</td>\n",
" <td>0.089546</td>\n",
" <td>-0.005838</td>\n",
" <td>0.292059</td>\n",
" <td>0.058035</td>\n",
" <td>-0.658261</td>\n",
" <td>0.258731</td>\n",
" <td>0.446757</td>\n",
" <td>0.363628</td>\n",
" <td>0.038044</td>\n",
" <td>0.916482</td>\n",
" <td>0.933073</td>\n",
" <td>0.999612</td>\n",
" <td>0.936855</td>\n",
" <td>1.000000</td>\n",
" <td>0.326385</td>\n",
" <td>0.997047</td>\n",
" <td>-0.100365</td>\n",
" <td>0.107097</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>total_ghg_emissions__metric_tons_co2e_</th>\n",
" <td>-0.029515</td>\n",
" <td>-0.120194</td>\n",
" <td>0.862229</td>\n",
" <td>0.159934</td>\n",
" <td>0.315519</td>\n",
" <td>-0.217121</td>\n",
" <td>0.973492</td>\n",
" <td>0.768373</td>\n",
" <td>0.959672</td>\n",
" <td>0.874011</td>\n",
" <td>0.847323</td>\n",
" <td>0.316579</td>\n",
" <td>0.355438</td>\n",
" <td>0.280278</td>\n",
" <td>0.326385</td>\n",
" <td>1.000000</td>\n",
" <td>0.357144</td>\n",
" <td>0.001091</td>\n",
" <td>0.091176</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>ghg_intensity__kg_co2e_sq_ft_</th>\n",
" <td>-0.046587</td>\n",
" <td>0.082777</td>\n",
" <td>0.025442</td>\n",
" <td>0.307693</td>\n",
" <td>0.060828</td>\n",
" <td>-0.628839</td>\n",
" <td>0.302830</td>\n",
" <td>0.479233</td>\n",
" <td>0.355816</td>\n",
" <td>0.132290</td>\n",
" <td>0.680197</td>\n",
" <td>0.904224</td>\n",
" <td>0.997439</td>\n",
" <td>0.910003</td>\n",
" <td>0.997047</td>\n",
" <td>0.357144</td>\n",
" <td>1.000000</td>\n",
" <td>-0.089224</td>\n",
" <td>0.114764</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>latitude</th>\n",
" <td>0.020637</td>\n",
" <td>-0.210518</td>\n",
" <td>0.011751</td>\n",
" <td>-0.115725</td>\n",
" <td>0.019664</td>\n",
" <td>0.220335</td>\n",
" <td>0.011699</td>\n",
" <td>0.010892</td>\n",
" <td>0.297109</td>\n",
" <td>-0.232749</td>\n",
" <td>0.468089</td>\n",
" <td>-0.126103</td>\n",
" <td>-0.094817</td>\n",
" <td>-0.136343</td>\n",
" <td>-0.100365</td>\n",
" <td>0.001091</td>\n",
" <td>-0.089224</td>\n",
" <td>1.000000</td>\n",
" <td>-0.483734</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>longitude</th>\n",
" <td>-0.081256</td>\n",
" <td>-0.116963</td>\n",
" <td>0.120967</td>\n",
" <td>0.001669</td>\n",
" <td>-0.179843</td>\n",
" <td>0.008902</td>\n",
" <td>0.107704</td>\n",
" <td>0.004915</td>\n",
" <td>-0.744476</td>\n",
" <td>0.326126</td>\n",
" <td>0.350674</td>\n",
" <td>0.095082</td>\n",
" <td>0.110036</td>\n",
" <td>0.091182</td>\n",
" <td>0.107097</td>\n",
" <td>0.091176</td>\n",
" <td>0.114764</td>\n",
" <td>-0.483734</td>\n",
" <td>1.000000</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>time</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" data_year id \\\n",
"data_year 1.000000 0.232191 \n",
"id 0.232191 1.000000 \n",
"gross_floor_area___buildings__sq_ft_ -0.215156 -0.187433 \n",
"year_built -0.100139 0.017634 \n",
"num_of_buildings -0.031533 -0.049704 \n",
"energy_star_score -0.106661 -0.317108 \n",
"electricity_use__kbtu_ -0.064666 -0.133163 \n",
"natural_gas_use__kbtu_ -0.018920 -0.089715 \n",
"district_steam_use__kbtu_ NaN -0.223803 \n",
"district_chilled_water_use__kbtu_ NaN -0.138594 \n",
"all_other_fuel_use__kbtu_ NaN -0.078548 \n",
"site_eui__kbtu_sq_ft_ -0.005264 0.120588 \n",
"source_eui__kbtu_sq_ft_ -0.039869 0.084036 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.001192 0.128526 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ -0.036829 0.089546 \n",
"total_ghg_emissions__metric_tons_co2e_ -0.029515 -0.120194 \n",
"ghg_intensity__kg_co2e_sq_ft_ -0.046587 0.082777 \n",
"latitude 0.020637 -0.210518 \n",
"longitude -0.081256 -0.116963 \n",
"time NaN NaN \n",
"\n",
" gross_floor_area___buildings__sq_ft_ \\\n",
"data_year -0.215156 \n",
"id -0.187433 \n",
"gross_floor_area___buildings__sq_ft_ 1.000000 \n",
"year_built 0.103160 \n",
"num_of_buildings 0.299070 \n",
"energy_star_score 0.073972 \n",
"electricity_use__kbtu_ 0.874597 \n",
"natural_gas_use__kbtu_ 0.585381 \n",
"district_steam_use__kbtu_ 0.899679 \n",
"district_chilled_water_use__kbtu_ 0.686560 \n",
"all_other_fuel_use__kbtu_ 0.282483 \n",
"site_eui__kbtu_sq_ft_ -0.008782 \n",
"source_eui__kbtu_sq_ft_ 0.010984 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ -0.030422 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ -0.005838 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.862229 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.025442 \n",
"latitude 0.011751 \n",
"longitude 0.120967 \n",
"time NaN \n",
"\n",
" year_built num_of_buildings \\\n",
"data_year -0.100139 -0.031533 \n",
"id 0.017634 -0.049704 \n",
"gross_floor_area___buildings__sq_ft_ 0.103160 0.299070 \n",
"year_built 1.000000 -0.034150 \n",
"num_of_buildings -0.034150 1.000000 \n",
"energy_star_score -0.167416 -0.137488 \n",
"electricity_use__kbtu_ 0.204463 0.212108 \n",
"natural_gas_use__kbtu_ 0.014824 0.583969 \n",
"district_steam_use__kbtu_ 0.327664 0.864088 \n",
"district_chilled_water_use__kbtu_ 0.294497 NaN \n",
"all_other_fuel_use__kbtu_ -0.787941 0.995715 \n",
"site_eui__kbtu_sq_ft_ 0.130847 0.138065 \n",
"source_eui__kbtu_sq_ft_ 0.282366 0.076069 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.153242 0.109415 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.292059 0.058035 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.159934 0.315519 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.307693 0.060828 \n",
"latitude -0.115725 0.019664 \n",
"longitude 0.001669 -0.179843 \n",
"time NaN NaN \n",
"\n",
" energy_star_score \\\n",
"data_year -0.106661 \n",
"id -0.317108 \n",
"gross_floor_area___buildings__sq_ft_ 0.073972 \n",
"year_built -0.167416 \n",
"num_of_buildings -0.137488 \n",
"energy_star_score 1.000000 \n",
"electricity_use__kbtu_ -0.142036 \n",
"natural_gas_use__kbtu_ -0.363684 \n",
"district_steam_use__kbtu_ -0.570782 \n",
"district_chilled_water_use__kbtu_ 0.007535 \n",
"all_other_fuel_use__kbtu_ -0.887757 \n",
"site_eui__kbtu_sq_ft_ -0.659919 \n",
"source_eui__kbtu_sq_ft_ -0.648452 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ -0.664027 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ -0.658261 \n",
"total_ghg_emissions__metric_tons_co2e_ -0.217121 \n",
"ghg_intensity__kg_co2e_sq_ft_ -0.628839 \n",
"latitude 0.220335 \n",
"longitude 0.008902 \n",
"time NaN \n",
"\n",
" electricity_use__kbtu_ \\\n",
"data_year -0.064666 \n",
"id -0.133163 \n",
"gross_floor_area___buildings__sq_ft_ 0.874597 \n",
"year_built 0.204463 \n",
"num_of_buildings 0.212108 \n",
"energy_star_score -0.142036 \n",
"electricity_use__kbtu_ 1.000000 \n",
"natural_gas_use__kbtu_ 0.631519 \n",
"district_steam_use__kbtu_ 0.952175 \n",
"district_chilled_water_use__kbtu_ 0.632375 \n",
"all_other_fuel_use__kbtu_ 0.694753 \n",
"site_eui__kbtu_sq_ft_ 0.195503 \n",
"source_eui__kbtu_sq_ft_ 0.285464 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.167008 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.258731 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.973492 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.302830 \n",
"latitude 0.011699 \n",
"longitude 0.107704 \n",
"time NaN \n",
"\n",
" natural_gas_use__kbtu_ \\\n",
"data_year -0.018920 \n",
"id -0.089715 \n",
"gross_floor_area___buildings__sq_ft_ 0.585381 \n",
"year_built 0.014824 \n",
"num_of_buildings 0.583969 \n",
"energy_star_score -0.363684 \n",
"electricity_use__kbtu_ 0.631519 \n",
"natural_gas_use__kbtu_ 1.000000 \n",
"district_steam_use__kbtu_ 1.000000 \n",
"district_chilled_water_use__kbtu_ 0.436563 \n",
"all_other_fuel_use__kbtu_ 0.951079 \n",
"site_eui__kbtu_sq_ft_ 0.587372 \n",
"source_eui__kbtu_sq_ft_ 0.498371 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.531869 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.446757 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.768373 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.479233 \n",
"latitude 0.010892 \n",
"longitude 0.004915 \n",
"time NaN \n",
"\n",
" district_steam_use__kbtu_ \\\n",
"data_year NaN \n",
"id -0.223803 \n",
"gross_floor_area___buildings__sq_ft_ 0.899679 \n",
"year_built 0.327664 \n",
"num_of_buildings 0.864088 \n",
"energy_star_score -0.570782 \n",
"electricity_use__kbtu_ 0.952175 \n",
"natural_gas_use__kbtu_ 1.000000 \n",
"district_steam_use__kbtu_ 1.000000 \n",
"district_chilled_water_use__kbtu_ NaN \n",
"all_other_fuel_use__kbtu_ NaN \n",
"site_eui__kbtu_sq_ft_ 0.404894 \n",
"source_eui__kbtu_sq_ft_ 0.365514 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.404723 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.363628 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.959672 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.355816 \n",
"latitude 0.297109 \n",
"longitude -0.744476 \n",
"time NaN \n",
"\n",
" district_chilled_water_use__kbtu_ \\\n",
"data_year NaN \n",
"id -0.138594 \n",
"gross_floor_area___buildings__sq_ft_ 0.686560 \n",
"year_built 0.294497 \n",
"num_of_buildings NaN \n",
"energy_star_score 0.007535 \n",
"electricity_use__kbtu_ 0.632375 \n",
"natural_gas_use__kbtu_ 0.436563 \n",
"district_steam_use__kbtu_ NaN \n",
"district_chilled_water_use__kbtu_ 1.000000 \n",
"all_other_fuel_use__kbtu_ 1.000000 \n",
"site_eui__kbtu_sq_ft_ 0.029963 \n",
"source_eui__kbtu_sq_ft_ 0.086752 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ -0.038395 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.038044 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.874011 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.132290 \n",
"latitude -0.232749 \n",
"longitude 0.326126 \n",
"time NaN \n",
"\n",
" all_other_fuel_use__kbtu_ \\\n",
"data_year NaN \n",
"id -0.078548 \n",
"gross_floor_area___buildings__sq_ft_ 0.282483 \n",
"year_built -0.787941 \n",
"num_of_buildings 0.995715 \n",
"energy_star_score -0.887757 \n",
"electricity_use__kbtu_ 0.694753 \n",
"natural_gas_use__kbtu_ 0.951079 \n",
"district_steam_use__kbtu_ NaN \n",
"district_chilled_water_use__kbtu_ 1.000000 \n",
"all_other_fuel_use__kbtu_ 1.000000 \n",
"site_eui__kbtu_sq_ft_ 0.834494 \n",
"source_eui__kbtu_sq_ft_ 0.716880 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.914262 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.916482 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.847323 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.680197 \n",
"latitude 0.468089 \n",
"longitude 0.350674 \n",
"time NaN \n",
"\n",
" site_eui__kbtu_sq_ft_ \\\n",
"data_year -0.005264 \n",
"id 0.120588 \n",
"gross_floor_area___buildings__sq_ft_ -0.008782 \n",
"year_built 0.130847 \n",
"num_of_buildings 0.138065 \n",
"energy_star_score -0.659919 \n",
"electricity_use__kbtu_ 0.195503 \n",
"natural_gas_use__kbtu_ 0.587372 \n",
"district_steam_use__kbtu_ 0.404894 \n",
"district_chilled_water_use__kbtu_ 0.029963 \n",
"all_other_fuel_use__kbtu_ 0.834494 \n",
"site_eui__kbtu_sq_ft_ 1.000000 \n",
"source_eui__kbtu_sq_ft_ 0.931679 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.999454 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.933073 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.316579 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.904224 \n",
"latitude -0.126103 \n",
"longitude 0.095082 \n",
"time NaN \n",
"\n",
" source_eui__kbtu_sq_ft_ \\\n",
"data_year -0.039869 \n",
"id 0.084036 \n",
"gross_floor_area___buildings__sq_ft_ 0.010984 \n",
"year_built 0.282366 \n",
"num_of_buildings 0.076069 \n",
"energy_star_score -0.648452 \n",
"electricity_use__kbtu_ 0.285464 \n",
"natural_gas_use__kbtu_ 0.498371 \n",
"district_steam_use__kbtu_ 0.365514 \n",
"district_chilled_water_use__kbtu_ 0.086752 \n",
"all_other_fuel_use__kbtu_ 0.716880 \n",
"site_eui__kbtu_sq_ft_ 0.931679 \n",
"source_eui__kbtu_sq_ft_ 1.000000 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.936193 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.999612 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.355438 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.997439 \n",
"latitude -0.094817 \n",
"longitude 0.110036 \n",
"time NaN \n",
"\n",
" weather_normalized_site_eui__kbtu_sq_ft_ \\\n",
"data_year 0.001192 \n",
"id 0.128526 \n",
"gross_floor_area___buildings__sq_ft_ -0.030422 \n",
"year_built 0.153242 \n",
"num_of_buildings 0.109415 \n",
"energy_star_score -0.664027 \n",
"electricity_use__kbtu_ 0.167008 \n",
"natural_gas_use__kbtu_ 0.531869 \n",
"district_steam_use__kbtu_ 0.404723 \n",
"district_chilled_water_use__kbtu_ -0.038395 \n",
"all_other_fuel_use__kbtu_ 0.914262 \n",
"site_eui__kbtu_sq_ft_ 0.999454 \n",
"source_eui__kbtu_sq_ft_ 0.936193 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 1.000000 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.936855 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.280278 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.910003 \n",
"latitude -0.136343 \n",
"longitude 0.091182 \n",
"time NaN \n",
"\n",
" weather_normalized_source_eui__kbtu_sq_ft_ \\\n",
"data_year -0.036829 \n",
"id 0.089546 \n",
"gross_floor_area___buildings__sq_ft_ -0.005838 \n",
"year_built 0.292059 \n",
"num_of_buildings 0.058035 \n",
"energy_star_score -0.658261 \n",
"electricity_use__kbtu_ 0.258731 \n",
"natural_gas_use__kbtu_ 0.446757 \n",
"district_steam_use__kbtu_ 0.363628 \n",
"district_chilled_water_use__kbtu_ 0.038044 \n",
"all_other_fuel_use__kbtu_ 0.916482 \n",
"site_eui__kbtu_sq_ft_ 0.933073 \n",
"source_eui__kbtu_sq_ft_ 0.999612 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.936855 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 1.000000 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.326385 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.997047 \n",
"latitude -0.100365 \n",
"longitude 0.107097 \n",
"time NaN \n",
"\n",
" total_ghg_emissions__metric_tons_co2e_ \\\n",
"data_year -0.029515 \n",
"id -0.120194 \n",
"gross_floor_area___buildings__sq_ft_ 0.862229 \n",
"year_built 0.159934 \n",
"num_of_buildings 0.315519 \n",
"energy_star_score -0.217121 \n",
"electricity_use__kbtu_ 0.973492 \n",
"natural_gas_use__kbtu_ 0.768373 \n",
"district_steam_use__kbtu_ 0.959672 \n",
"district_chilled_water_use__kbtu_ 0.874011 \n",
"all_other_fuel_use__kbtu_ 0.847323 \n",
"site_eui__kbtu_sq_ft_ 0.316579 \n",
"source_eui__kbtu_sq_ft_ 0.355438 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.280278 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.326385 \n",
"total_ghg_emissions__metric_tons_co2e_ 1.000000 \n",
"ghg_intensity__kg_co2e_sq_ft_ 0.357144 \n",
"latitude 0.001091 \n",
"longitude 0.091176 \n",
"time NaN \n",
"\n",
" ghg_intensity__kg_co2e_sq_ft_ \\\n",
"data_year -0.046587 \n",
"id 0.082777 \n",
"gross_floor_area___buildings__sq_ft_ 0.025442 \n",
"year_built 0.307693 \n",
"num_of_buildings 0.060828 \n",
"energy_star_score -0.628839 \n",
"electricity_use__kbtu_ 0.302830 \n",
"natural_gas_use__kbtu_ 0.479233 \n",
"district_steam_use__kbtu_ 0.355816 \n",
"district_chilled_water_use__kbtu_ 0.132290 \n",
"all_other_fuel_use__kbtu_ 0.680197 \n",
"site_eui__kbtu_sq_ft_ 0.904224 \n",
"source_eui__kbtu_sq_ft_ 0.997439 \n",
"weather_normalized_site_eui__kbtu_sq_ft_ 0.910003 \n",
"weather_normalized_source_eui__kbtu_sq_ft_ 0.997047 \n",
"total_ghg_emissions__metric_tons_co2e_ 0.357144 \n",
"ghg_intensity__kg_co2e_sq_ft_ 1.000000 \n",
"latitude -0.089224 \n",
"longitude 0.114764 \n",
"time NaN \n",
"\n",
" latitude longitude time \n",
"data_year 0.020637 -0.081256 NaN \n",
"id -0.210518 -0.116963 NaN \n",
"gross_floor_area___buildings__sq_ft_ 0.011751 0.120967 NaN \n",
"year_built -0.115725 0.001669 NaN \n",
"num_of_buildings 0.019664 -0.179843 NaN \n",
"energy_star_score 0.220335 0.008902 NaN \n",
"electricity_use__kbtu_ 0.011699 0.107704 NaN \n",
"natural_gas_use__kbtu_ 0.010892 0.004915 NaN \n",
"district_steam_use__kbtu_ 0.297109 -0.744476 NaN \n",
"district_chilled_water_use__kbtu_ -0.232749 0.326126 NaN \n",
"all_other_fuel_use__kbtu_ 0.468089 0.350674 NaN \n",
"site_eui__kbtu_sq_ft_ -0.126103 0.095082 NaN \n",
"source_eui__kbtu_sq_ft_ -0.094817 0.110036 NaN \n",
"weather_normalized_site_eui__kbtu_sq_ft_ -0.136343 0.091182 NaN \n",
"weather_normalized_source_eui__kbtu_sq_ft_ -0.100365 0.107097 NaN \n",
"total_ghg_emissions__metric_tons_co2e_ 0.001091 0.091176 NaN \n",
"ghg_intensity__kg_co2e_sq_ft_ -0.089224 0.114764 NaN \n",
"latitude 1.000000 -0.483734 NaN \n",
"longitude -0.483734 1.000000 NaN \n",
"time NaN NaN NaN "
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[numeric_columns].corr()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's see which columns are strongly correlated to `electricity_use__kbtu_`; using those columns as a feature would be beneficial to make accurate prediction.\n",
"\n",
"See histogram of values in numerical columns:"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA5AAAAQBCAYAAABc0WVMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJzs3Xm4HFWd//H3BxIhLLJFryHEBAXR\nAAqYQRy3K+gMRh5xRkUYQKIgoqI4g+OA409xG7fBDRwxCoZNEHAhCi6oXBGVnUBYXCIEkxAIa0LA\nhej398c5N9Tt20vdrau783k9z31uddWpqm+frlNVp+rUKUUEZmZmZmZmZq1sVHUAZmZmZmZm1h1c\ngTQzMzMzM7NSXIE0MzMzMzOzUlyBNDMzMzMzs1JcgTQzMzMzM7NSXIE0MzMzMzOzUlyBNDNrQNIC\nSR+rOg4zMzOzTrFBViAlzZN0ZeFzSNppnNfRlhNPSVMkfU/SakkXjnFZQ/KlkzXLX5/0W7tJGpB0\nVNVxWHcZ3FdJeomk347jcn8g6YjxWl6LdTXd30paK+kZtWkl9UtaPsp1jnpeG67ZOdBEnB/1ol4o\ny9aYy8hwG2QFcrxVXPF6PdAHbBcRb6gohq6yoRZ2s0GSJlUdgz0hIn4REbu0SifpJEnnlFjeqyLi\nzBLLWyrpFWXjHI2I2CIi7pjIdUyksnm+Ieqmi87t0s1l2RdCx18vlxFXILuApI2bTJ4J/C4i1rUr\nHrNeJWlPSTdIekTSN4FN8/htJH1f0n2SHsrDO+RpHwdeApya77acmsd/QdIySWskXS/pJS3W/TRJ\nj0narjBur7zOyfnzWyTdnmP4kaSZhbQN15dPVi6SdI6kNcC8ccs06xhKfFzvIL5YY6Phsmydrqc3\nTkknSPpDPhm8TdK/jGFZW0k6K5/M3SXpA5I2kvQc4DTghfnk8eHCbNtIuiSv/2pJzyws79mSLpP0\noKTfSjqoMG2BpC9LulTSo8DLG8T0YeCDwBvzuo+svaolaVa+4zap8D1Ol7RS0orc5KJZBbV2nUOW\nl8etv2olaSdJP1dqUnt/Pglv+Z3HQtKWki6X9EVJyqOn5nU9kuOZmdNekafflPPsjfWuEMl3KTc4\nkp4EfBc4G9gWuBB4XZ68EfB10gWbpwN/Ak4FiIj/Bn4BHJvvthyb57kW2CMv6xvAhZI2bbT+iLgH\nGACK5eJw4PyIeFzSgcD7gX8FnpLXeV4hbav1HQhcBGwNnFsqU2xcNblAMaRJpqT/yvvnR/K+cj9J\n+5N+/8H9/U057YCkj0v6JfAY8AzV3EmQ9NZ84WHwWLiXpLNJ2/L38vLe1yL2F0v6laSH84WKeYXJ\nzY51pfalkraX9K18jL1T0rsL06bk4+JDkm4D/qHE8t4s6XuFz79X4TGP/B32yMN1L740yfOGx9F8\nPPmlpM9JegA4qUmMrY7X8yTdkfP1TkmHFtI2vJg0Fvl3XiapvzB6bo7jfkmfUZNznzrbXk/egenW\nsixpU6ULiQ/ksnytpD41vhDa7Fz11ZJuzOVmmaSTCtMGt+U352kPSTpG0j9Iujmv+9QS+ewy0qki\nomf/gDcA25NO/t4IPApMI119v7KQLoCdWizrLOBiYEtgFvA74Mg8bcjy8rgFwAPA3sAk0gnb+Xna\n5sAy4M152p7A/cDswryrgRfl2DdtEtdJwDlNPs/K329S/vwd4Cs5hqcC1wBva/Q96qxvyPLyuAHg\nqDx8HvDfg3EDLy7znUfx2y4APgZsl7/Dx2qmPQK8FNgE+EKz37vB79dym/Bfb/3l7eVuQIVxvypu\nW4XxewAPFT6vLwNNlv8Q8LwWad4I/DIPbwzcA+ydP/9gcJ+TP29EOsmY2Wp9eb9wRdV5vCH/AU8C\n7gL+HZhMevzg8bwf6weW53S75H3l9vnzLOCZhd/xnJrlDgB/BHbN+9bJNfvkNwArSJUuATsNbjPA\nUuAVJWKfmfeph+TlbwfskactoMGxLk9fvy8d3G/n4eJ33gi4nnRB9EnAM4A7gH/O0z9JumCyLTAD\nuGVw3iYxPwN4OC97+5z3ywvTHgI2yp8Py99pEnB8LnebNsnzVsfRdcC78vKmNIlxyLIpHF/zstcA\nu+Rp04Bd8/CBwBLgOTntB4BfjWHbjLxd7E/a9vaumXZ5zvunk859jip819pj5wCFfWG9NN3+R3eX\n5bcB3wM2Ix1jng88ucFv1+pctR/YnVTGngvcC7y2Zls+jXQu+E/An0kXaZ8KTAdWAS9rEe+QfMJl\npGP+evoOZERcGBF3R8TfI+KbwO9JB7kRyVcWDwZOjIhHImIpcDLp7kAz34mIayI1Lz2XdNIJcACw\nNCK+HhHrIuJG4FukncOgiyPilzn2P4805gbfow+YC7wnIh6NiFXA5/J3Gy+Pk042to+IP0fE4FWV\nMt95pLYHfg5cGBEfqJl2SURcERF/IVVoXyhpxhjWZb1ve2BF5D16dheApM0kfUWp9cEa4ApgazW5\ney/pvfnq5+p85XErYGqLGC4GZkvaEXglsDoirsnTZgJfyFduHwYeJJ1ETC+5vmXlssEmyD6kE8LP\nR8TjEXER6a5xrb+RLnzNljQ5IpZGxB9aLHtBRNya962P10w7Cvh0RFwbyZKIuGuEsf8b8JOIOC/H\n/kBELCpMb3SsK+sfgKdExEci4q+Rnpn8Kk8cmw4CPh4RD0bEMuCLrRaYl/FIjuWlwI+AuyU9G3gZ\n8IuI+HtOe07+Tusi4mRS/td9jq3kcfTuiDglL+9PI8yLor8Du0maEhErI+LWPP4Y4BMRcXvO8/8B\n9hjjHZY3kCrFryrscwZ9Kuf9H4HPky4kbMi6uSw/TrpYslNE/C0iro+INQ3SNj1vi4iBiFicz1Nv\nJt1AeFnNMj6azwV/TLqJc15ErIqIFaSLQnuOMP5aLiMV6ekKpKQ3SVpUOOHajdYncPVMJe0sigX1\nLvKJWxP3FIYfA7bIwzOBFwzGlWM7FHhaIf1EnOzNJH2PlYX1foV0NWi8vI90UnuNpFslvaWw7lbf\neaReDUwhXeGqtT7/ImIt6WR7+zGsy3rfSmC6tL4ZNKSriZDuSuwCvCAinkw6IYW0rUO6ArmeUhO4\n95FOfLeJiK1JrQqKyx4mXyy6gHRH5HBSc9pBy0h3ObYu/E2JiF+VXN+QGK3tGl6gKIqIJcB7SFfe\nV0k6X1KrfVez48UMoNVJayutltHoWFfWTGD7muPD+0kdxEHKu+J3LHvS/HPSXZKX5uEB0gnuy/Jn\nYMQXe8ocR8d8/I6IR0ktEo7J67okV34HY2h4MWmU3gNcEBG31JlWm/cb+rG0m8vy2aSLKedLulvS\np5Wfsa+j6XmbpBcoPT50n6TVpG21ttzcWxj+U53PI91XrOcyUq2erUDmqwxfBY4l9VC6NanZS9MT\nuAbu54k7a4OeTmpKACM/MVsG/LzmRHCLiHh7Ic1oT/YeJTVNGFRbKf0LMLWw3idHxK4jXD6N1hER\n90TEWyNie1JTif9Tev6lzHceqa8CPwQulbR5zbT1dxslbUFqWnB3k++0WSH9WCq11r1+TWp69m5J\nkyX9K0+0WNiSdLB7WNK2wIdq5r2X1CyOQvp1wH3AJEkfBJ5cMo6zSE1aXsPQCuRpwImSdoX1z2EN\n3sEfy/qsPZpdoBgiIr4RES8mHXMC+NTgpAbLbna8WAY8s8G0sseZZssYD8uAO2uOD1tGxNw8fSWF\nfToN8q2OwQrkS/Lwz6mpQJa4+FKbR2WOo2Xztdnxmoj4UUS8ktQ07zekY95gDHUvJpVcbz1vAF4r\n6bg602rzfvBYWu97Nv1OPaJry3K+Y/rhiJgN/CPpLuObGiyj1XnbN4CFwIyI2Ip0jBrNOXYzLiMd\nqmcrkKS20UE6oULSm0l3IEcsIv5GuivwcaUOW2YC/wEMPth7L7CDUiccZXwfeJakw/OJ6mSlB4uf\nM5r4aiwCXirp6ZK2Ak4sfI+VwI+BkyU9OT/k+0xJtU0OGoqI+0gV58MkbZzvMBY7THiDcu+UpGdM\ngtTEYKK+87HAb0kPj08pjJ+r9KDzk4CPAldFavoEw0/2bwJ2lbSHUqcjJ40xJutCEfFXUgc180hX\nK98IfDtP/jzpbvf9wFWkCxdFXwBer/Sw/hdJV3h/SHoW4i7Ssx+l7kpExC9JZeaGYvOkiPgO6eTj\n/NyM9hbgVXnyqNdnbdPsAsV6knaRtK+kTUi/459I2wOkfdcsjax3xq8B75X0fCU7FZpx1e4LGzkX\neIWkgyRNkrSdcgc04+Qa4BGlDkem5GPLbpIGO8u5gHTxZJt8fHlXyeX+nNQJ3ZSIWE5qMrc/qQnf\njTlNq4svQ/J8PI6jBQ2P10odmxyYL47+BVjLE9tBs4tJo3U3sB9wnKTaC7v/mfN+BnAcMNg5Xr1z\nn0XAvyo1+98JOHKMcXWiri3Lkl4uaXelxy/WkG6OFGMqLqPVeduWwIMR8WdJe5Oauo83l5EO1bMV\nyIi4jfSc4q9JP+DuwC/HsMh3ka4a3AFcSbryckae9jPgVuAeSfeXiO0R0gPFB5M2yHtIJ4abjCG+\nwWVfRtpwbyZ1SvD9miRvIj0AfhupgncR6crNSLwV+E9Sxwm7kjoaGfQPwNWS1pKuTB0XEXdM1HfO\nTUiOBpYDF+uJXie/QbpL9CDpIfHDCrOdBJyp1LThoIj4HfAR4Cek52S7rzcsGxcRcV1E7Jnvfrwx\n/30g0rPU/fnq67Mi4isRofxsBRHx6zx+m4h4d6RnS96S70xMi4hPR8SsiPhJyVCWMfTu42B8Z0fE\n7nm5MyLiLXl80/VFxEkRcVjt8qx9WlygKNqE1GnM/aT95FN54qRpsBfRByTdUHK9FwIfJ+0THyF1\nYrFtnvwJ4AN5X/jeJsv4I+m5v+Nz7IuA55VZf8kY/0a6E7IHcCfpu3+N1JQU4MOkCyN3kipvw8pG\ng+X+jnRS+Yv8eQ3pGP7LvE5offGlXp6Px3G01fF6I9KF6rtJef4y4O15vmYXk0Yt/877ASdo6PsA\nL87xLQIuAU7P4+ud+3wO+CvpvOtMerDH524uy6S7XReRKo+3ky6yDJanIRdCS5y3vQP4iKRHSB1g\nXVDme4yEy0jnUsRoW0qamdl4y3ddLiM1C3qk6njMzMzMivyCWzOzNpL0A9IzWbX+h9RRz2tJd+5d\neTQzM7OO4zuQBZJuZWhHOYPeFhGV3WJud1xKnQr8oN60iBh1j1lN1tfo+91HemF6rbb9Hp26TZiZ\njSelF3B/pc6ku2JkHa21jaSnk5qR1jM7NzerlKTTGPoIxaBzIuKYcV5Xo2P3FNLzd8NMxDG9nnaf\nV2zIuq0su4wk3VZGXIE0MzMzMzOzUrqyCevUqVNj1qxZdac9+uijbL557RsdquWYyunEmGBscV1/\n/fX3R0S9u6jjKvf8dRbpvWkBzI+IL0g6idTp0X056fsj4tI8z4mk3r/+Brw7In7UbB3Nyh305u83\n0Rzb6LSKrV3lrh26qdw5lvo6KRaYuHh6vdx12u/YTDfFCt0Vb6fFWlm5i4iu+3v+858fjVx++eUN\np1XFMZXTiTFFjC0u4LpoQ5kg9QC4Vx7ektSj4GxSj7PvrZN+Nun1JZsAO5JeTrxxs3U0K3djzaeJ\n1KlxRTi20WoVW7vKXTv+uqncOZb6OimWiImLp9fLXaf9js10U6wR3RVvp8VaVbnr2dd4mG1IImJl\nRNyQhx8hdc89vcksBwLnR8RfIuJOYAl13mNlZmZmZlbkCqRZj5E0C9gTuDqPOlbSzZLOkLRNHjed\noe86W07zCqeZmZmZWXc+A2lm9UnaAvgW8J6IWCPpy8BHSc9FfhQ4GXjLCJZ3NHA0QF9fHwMDAw3T\nrl27tun0qnRqXODYRquTYzMzM+t1rkCa9QhJk0mVx3Mj4tsAEXFvYfpXge/njyuAGYXZd8jjhoiI\n+cB8gDlz5kR/f3/D9Q8MDNBselU6NS5wbKPVybGVIWlj4DpgRUQcUHU8Zt1K0qbAFaTn+ScBF0XE\nh2rSbELqZO75wAPAGyNiaZtDNespbsJq1gMkCTgduD0iPlsYP62Q7F+AW/LwQuBgSZtI2hHYGbim\nXfGabeCOIz2nbGZj8xdg34h4HrAHsL+kfWrSHAk8FBE7AZ8DPtXmGM16jiuQZr3hRcDhwL6SFuW/\nucCnJS2WdDPwcuDfASLiVuAC0ku4fwi8MyL+VlHsZhsMSTsArwa+VnUsZt0ud0S5Nn+cnP9qX3B+\nIHBmHr4I2C9fdDWzUeq5JqyLV6xm3gmXDBu/9JOvriAas/aIiCuBegfES5vM83Hg4+MVQ72y53Jn\nNszngfeRXrdT10iePV714GpOOffiIeN2n75VqUAWr1g9bFzZeevppGdTHUtjnRbPWOUm4dcDOwFf\nioira5Ks7zQuItZJWg1sB9xfs5ym5W688m28y1093fYbd1O83RTrROq5CqSZmVknknQAsCoirpfU\n3yjdSJ49PuXcizl58dBD+dJDG6cvqnuxteS89XTSs6mOpbFOi2escuuZPSRtDXxH0m4RcUur+eos\np2m5G698G+9yV0+3/cbdFG83xTqR3ITVzMysPV4EvEbSUuB8UpPzc6oNyaw3RMTDwOXA/jWT1nca\nJ2kSsBWpMx0zGyVXIM3MzNogIk6MiB0iYhZwMPCziDis4rDMupakp+Q7j0iaArwS+E1NsoXAEXn4\n9aRyV/ucpJmNgJuwmpmZmVk3mgacmZ+D3Ai4ICK+L+kjwHURsZDUQ/nZkpYAD5Iu3pjZGLgCaWZm\n1mYRMQAMVByGWVeLiJuBPeuM/2Bh+M/AG9oZl1mvcxNWMzMzMzMzK8UVSDMzMzMzMyulbRVISWdI\nWiXplsK4z0j6jaSbJX1n8EFoMzMzMzMz6zztvAO5gOFdK18G7BYRzwV+B5zYxnjMzMzMzMxsBNpW\ngYyIK0i9XxXH/Tgi1uWPVwE7tCseMzMzMzMzG5lO6oX1LcA3G02UdDRwNEBfXx8DAwN10/VNgeN3\nXzdsfKP07bB27dpK11+PYyqvU+MyMzMzM2u3jqhASvpvYB1wbqM0ETEfmA8wZ86c6O/vr5vulHMv\n5uTFw7/W0kPrp2+HgYEBGsVbFcdUXqfGZWZmZmbWbpVXICXNAw4A9ouIqDgcMzMzMzMza6DSCqSk\n/YH3AS+LiMeqjMXMzMzMzMyaa+drPM4Dfg3sImm5pCOBU4EtgcskLZJ0WrviMTMzMzMzs5Fp2x3I\niDikzujT27V+MzMzMzMzG5t2vgfSzMzMzMzMupgrkGZmZmZmZlaKK5BmZmZmZmZWiiuQZmZmZmZm\nVoorkGZmZmZmZlaKK5BmZmZm1nUkzZB0uaTbJN0q6bg6afolrc6vi1sk6YNVxGrWS1yBNOsBjQ6i\nkraVdJmk3+f/2+TxkvRFSUsk3Sxpr2q/gZmZ2YitA46PiNnAPsA7Jc2uk+4XEbFH/vtIe0M06z2u\nQJr1hkYH0ROAn0bEzsBP82eAVwE757+jgS+3P2QzM7PRi4iVEXFDHn4EuB2YXm1UZr1vUtUBmNnY\nRcRKYGUefkTS4EH0QKA/JzsTGAD+K48/KyICuErS1pKm5eWYmZl1FUmzgD2Bq+tMfqGkm4C7gfdG\nxK115j+adEGVvr4+BgYGhkxfu3btsHGjcfzu64aNG4/lFo1XrO3STfF2U6wTyRVIsx5TcxDtK1QK\n7wH68vB0YFlhtuV53JAKZKsDalHflOEHxk7YyXbyzt6xjU4nx2Zm7SdpC+BbwHsiYk3N5BuAmRGx\nVtJc4Luk1jdDRMR8YD7AnDlzor+/f8j0gYEBaseNxrwTLhk2bumhY19u0XjF2i7dFG83xTqRXIE0\n6yG1B1FJ66dFREiKkSyv1QG16JRzL+bkxUN3KeN9UByNTt7ZO7bR6eTYzKy9JE0mHffOjYhv104v\nVigj4lJJ/ydpakTc3844zXqJn4E06xENDqL3SpqWp08DVuXxK4AZhdl3yOPMzMy6gtJV0tOB2yPi\nsw3SPC2nQ9LepHPfB9oXpVnvcQXSrAc0OYguBI7Iw0cAFxfGvyn3xroPsNrPP5qZWZd5EXA4sG/h\nNR1zJR0j6Zic5vXALfkZyC8CB+fn/81slNyE1aw3DB5EF0talMe9H/gkcIGkI4G7gIPytEuBucAS\n4DHgze0N18zMbGwi4kpALdKcCpzanojMNgxtq0BKOgM4AFgVEbvlcdsC3wRmAUuBgyLioXbFZNYr\nWhxE96uTPoB3TmhQZmZmZtZz2tmEdQGwf824Ru+oMzMzMzMzsw7TtgpkRFwBPFgz+kDSu+nI/1/b\nrnjMzMzMzMxsZKruRKfRO+rMzMx6iqRNJV0j6SZJt0r6cNUxmZmZjVTHdKLT6h11ZV9oXu9l5lDt\nC8078aXXjqm8To3LzLrOX4B98wvNJwNXSvpBRFxVdWBmZmZlVV2BvFfStIhYWfOOumHKvtC83svM\nodoXmnfiS68dU3mdGpeZdZfcedXa/HFy/vPrBMzMrKtUXYEcfEfdJxn6jjozM7OeI2lj4HpgJ+BL\nEXF1nTSlWtxA/VY39dIvXrF62Ljjdx++vJG0tqhdZt+Ualv7FHVSy5FOigU6Lx4z6z7tfI3HeUA/\nMFXScuBDNH5HnZmZWc+JiL8Be0jaGviOpN0i4paaNKVa3ED9Vjf1WtzMO+GSUvGNpLVO7TKP330d\nB3VIa41OajnSSbFA58VjZt2nbRXIiDikwaRh76gzMzPrZRHxsKTLSa+3uqVVejMzs05RdS+sZmZm\nGwRJT8l3HpE0BXgl8JtqozIzMxuZqp+BNDMz21BMA87Mz0FuBFwQEd+vOCYzM7MRcQXSzMysDSLi\nZmDPquMwMzMbCzdhNTMzMzMzs1JcgTQzMzMzM7NSXIE0MzMzMzOzUlyBNDMzMzMzs1JcgTQzMzOz\nriNphqTLJd0m6VZJx9VJI0lflLRE0s2S9qoiVrNe4l5YzczMzKwbrQOOj4gbJG0JXC/psoi4rZDm\nVcDO+e8FwJfzfzMbJd+BNDMzM7OuExErI+KGPPwIcDswvSbZgcBZkVwFbC1pWptDNespvgNpZmZm\nZl1N0izSe1avrpk0HVhW+Lw8j1tZM//RwNEAfX19DAwMDFnI2rVrh40rWrxi9bBxu0/fati443df\nN2xcs+WORqtYO003xdtNsU4kVyDNzMzMrGtJ2gL4FvCeiFgzmmVExHxgPsCcOXOiv79/yPSBgQFq\nxxXNO+GSYeOWHjo8fdl0Y9Eq1k7TTfF2U6wTyU1YzczMzKwrSZpMqjyeGxHfrpNkBTCj8HmHPM7M\nRskVSDMzMzPrOpIEnA7cHhGfbZBsIfCm3BvrPsDqiFjZIK2ZleAKpFkPkHSGpFWSbimMO0nSCkmL\n8t/cwrQTc5fmv5X0z9VEbWZmNiYvAg4H9i0e6yQdI+mYnOZS4A5gCfBV4B0VxWrWMzriGUhJ/w4c\nBQSwGHhzRPy52qjMusoC4FTgrJrxn4uI/y2OkDQbOBjYFdge+ImkZ0XE39oRqJmZ2XiIiCsBtUgT\nwDvbE5HZhqHyO5CSpgPvBuZExG7AxqSTWzMrKSKuAB4smfxA4PyI+EtE3Em6Krv3hAVnZmZmZj2j\nI+5AkuKYIulxYDPg7orjMesVx0p6E3Ad6WXLD5G6L7+qkGawS/NhWnVrXtQ3ZXj35J3Q1XUnd7nt\n2Eank2MzMzPrdZVXICNihaT/Bf4I/An4cUT8uDZd2RPZeiexUO2JbCee7Dim8jo1rhK+DHyU1DT8\no8DJwFtGsoBW3ZoXnXLuxZy8eOguZby7Jh+NTu5y27GNTifHZmZm1usqr0BK2obUpG5H4GHgQkmH\nRcQ5xXRlT2TrncRCtSeynXiy45jK69S4WomIeweHJX0V+H7+6C7NzczMzGxUKn8GEngFcGdE3BcR\njwPfBv6x4pjMup6kaYWP/wIM9tC6EDhY0iaSdgR2Bq5pd3xmZmZm1n0qvwNJarq6j6TNSE1Y9yM9\nr2VmJUk6D+gHpkpaDnwI6Je0B6kJ61LgbQARcaukC4DbgHXAO90Dq5mZmZmVUXkFMiKulnQRcAPp\nZPZGclNVMysnIg6pM/r0Juk/Dnx84iIyMzMzs15UeQUSICI+RLpjYmZmZmZmZh2qE56BNDMzMzMz\nsy7gCqSZmZmZmZmV4gqkmZmZmZmZleIKpJmZmZmZmZXiCqSZmZmZmZmV4gqkmZmZmZmZleIKpJmZ\nmZmZmZXiCqSZmZmZmZmV4gqkmZlZG0iaIelySbdJulXScVXHZNbNJJ0haZWkWxpM75e0WtKi/PfB\ndsdo1osmVR2AmZnZBmIdcHxE3CBpS+B6SZdFxG1VB2bWpRYApwJnNUnzi4g4oD3hmG0YfAfSzMys\nDSJiZUTckIcfAW4HplcblVn3iogrgAerjsNsQ+M7kGZmZm0maRawJ3B1nWlHA0cD9PX1MTAw0HA5\nfVPg+N3XDRlXL31tmkaaravVMvumlJ9/8YrVw8btPn2rMacdtHbtWgYGBkY173gbjKVTdFo8bfBC\nSTcBdwPvjYhb6yVqVe5a5Vu9Mla2LI7379Ftv3E3xdtNsU4kVyDNzMzaSNIWwLeA90TEmtrpETEf\nmA8wZ86c6O/vb7isU869mJMXDz2ULz10ePp5J1xSKrZ68zZSu8zjd1/HQU1ibRVPo3WPJO2ggYEB\n+vv7RzXveBuMpVN0WjwT7AZgZkSslTQX+C6wc72Ercpdq3wru621Y5vstt+4m+LtplgnkpuwmpmZ\ntYmkyaTK47kR8e2q4zHrZRGxJiLW5uFLgcmSplYcllnX64gKpKStJV0k6TeSbpf0wqpjMjMzG0+S\nBJwO3B4Rn606HrNeJ+lpudwhaW/See8D1UZl1v06pQnrF4AfRsTrJT0J2KzqgMzMzMbZi4DDgcWS\nFuVx7893RsxshCSdB/QDUyUtBz4ETAaIiNOA1wNvl7QO+BNwcEREReGa9YzKK5CStgJeCswDiIi/\nAn+tMiYzM7PxFhFXAqo6DrNeERGHtJh+Kuk1H2Y2jiqvQAI7AvcBX5f0POB64LiIeLSYqGyvdPV6\npIPx7+FqJDqxxybHVF6nxmVmZmZm1m6dUIGcBOwFvCsirpb0BeAE4P8VE5Xtla5ej3TQ/l7Xijqx\nxybHVF6nxmVmZmZm1m6d0ImVmDZiAAAgAElEQVTOcmB5RAy+C+siUoXSzMzMzMzMOkjlFciIuAdY\nJmmXPGo/4LYKQzIzMzMzM7M6Kq9AZu8CzpV0M7AH8D8Vx2PWVSSdIWmVpFsK47aVdJmk3+f/2+Tx\nkvRFSUsk3SzJd/zNzMzMrJSOqEBGxKKImBMRz42I10bEQ1XHZNZlFgD714w7AfhpROwM/DR/BngV\nsHP+Oxr4cptiNDMzM7Mu1xEVSDMbm4i4AniwZvSBwJl5+EzgtYXxZ0VyFbC1pGntidTMzMzMulkn\n9MJqZhOjLyJW5uF7gL48PB1YVki3PI9bSY2yr8+B+q/Q6YTXn3Tya1gc2+h0cmxmZma9zhVIsw1A\nRISkGMV8pV6fA/VfoVPl63MGdfJrWBzb6HRybGZmZr3OTVjNete9g01T8/9VefwKYEYh3Q55nJmZ\nmZlZU65AmvWuhcARefgI4OLC+Dfl3lj3AVYXmrqamZmZmTXkJqxmPUDSeUA/MFXScuBDwCeBCyQd\nCdwFHJSTXwrMBZYAjwFvbnvAZmZmZtaVXIE06wERcUiDSfvVSRvAOyc2IjMzMzPrRW7CamZmZmZm\nZqW4AmlmZmZmZmaluAJpZmZmZmZmpbgCaWZmZmZdR9IZklZJuqXBdEn6oqQlkm6WtFe7YzTrRa5A\nmpmZmVk3WgDs32T6q4Cd89/RwJfbEJNZz3MF0szMzMy6TkRcATzYJMmBwFmRXAVsLWlae6Iz611+\njYeZmZmZ9aLpwLLC5+V53MrahJKOJt2lpK+vj4GBgSHT165du37c4hWrh63o+N2Hr7x2GSndumHj\nTjn34rrB7z59q2Hj6q27Nl0x1pHOO9K0I1lmI43i7UQTGet45GW7dEwFUtLGwHXAiog4oOp4zMzM\nzGzDEBHzgfkAc+bMif7+/iHTBwYGGBw374RLSi1z6aH9w8aVnXck89emK8Y60nlHmnYky2ykUbyd\naCJjHY+8bJdOasJ6HHB71UGYmZmZWU9YAcwofN4hjzOzMeiICqSkHYBXA1+rOhYzMzMz6wkLgTfl\n3lj3AVZHxLDmq2Y2Mp3ShPXzwPuALRslaNU2fVDflPrty6tsW92JbbsdU3mdGpeZmdmGTNJ5QD8w\nVdJy4EPAZICIOA24FJgLLAEeA95cTaRmvaXyCqSkA4BVEXG9pP5G6Vq1TR90yrkXc/Li4V+ryjbE\nndi22zGV16lxmZmZbcgi4pAW0wN4Z5vCMdtgdEIT1hcBr5G0FDgf2FfSOdWGZGZmZmZmZrUqr0BG\nxIkRsUNEzAIOBn4WEYdVHJaZmZmZmZnVqLwCaWZmZmZmZt2hoyqQETHgd0CamVmvknSGpFWSbqk6\nFjMzs9HoqAqkmZlZj1sA7F91EGZmZqPlCqSZmVmbRMQVwINVx2FmZjZarkCamZmZmZlZKZW/B9LM\nzMyeIOlo4GiAvr4+BgYGGqbtmwLH775uyLhTzr14WLrjdy+37nrz7j59q7ppa9fbN6X+/PXnHT6u\n0fesXQ+0Xs9gLPXWM5LvWM/iFatLzT+YrlW+jGTd42Ht2rVNtyko/x3NbMPkCqSZmVkHiYj5wHyA\nOXPmRH9/f8O0p5x7MScvnthD+dJD669/3gmXDPl8/O7rxhRL2fWUMdJYGq27bDz15h9M1yqWkax7\nPAwMDNBsm4Ly39HMNkxuwmpmZmZmZmaluAJp1uMkLZW0WNIiSdflcdtKukzS7/P/baqO02xDIOk8\n4NfALpKWSzqy6pjMzMxGwhVIsw3DyyNij4iYkz+fAPw0InYGfpo/m9kEi4hDImJaREyOiB0i4vSq\nYzIzMxsJVyDNNkwHAmfm4TOB11YYi5mZmZl1CXeiY9b7AvixpAC+kjvo6IuIlXn6PUBfvRnH2htk\nq57+2qFMj4NVcWyj08mxmZmZ9TpXIM1634sjYoWkpwKXSfpNcWJERK5cDjPW3iA7ode+Mj0OVsWx\njU4nx2ZmZtbr3ITVrMdFxIr8fxXwHWBv4F5J0wDy/1XVRWhmZmZm3cIVSLMeJmlzSVsODgP/BNwC\nLASOyMmOAMq9/dvMzKyDSNpf0m8lLZE0rEM4SfMk3Zd7Il8k6agq4jTrJW7Catbb+oDvSIJU3r8R\nET+UdC1wQX6FwF3AQRXGaGZmNmKSNga+BLwSWA5cK2lhRNxWk/SbEXFs2wM061GVVyAlzQDOIp3o\nBjA/Ir5QbVRmvSEi7gCeV2f8A8B+7Y/IzMxs3OwNLMnHOiSdT+plvLYCaWbjqPIKJLAOOD4ibshN\n7a6XdFmdq0dmZmZmZoOmA8sKn5cDL6iT7nWSXgr8Dvj3iFhWm6BVr+PF3p9rexxvpF5v0WXnHcn8\nzWId6bwjTTuSZTbSTT1rT2Ss45GX7VJ5BTK/SmBlHn5E0u2kHYIrkGZmZmY2Ft8DzouIv0h6G+nd\nx/vWJmrV63ix9+d5J1xSasX1eiIvO+9I5q9N16in6jLzjjTtSJbZSDf1rD2RsY5HXrZL5RXIIkmz\ngD2Bq+tMK/U+unrvooNqa/CdeGXFMZXXqXGZmZlt4FYAMwqfd8jj1suPbAz6GvDpNsRl1tM6pgIp\naQvgW8B7ImJN7fSy76Or9y46qLYG34lXVhxTeZ0al5mZ2QbuWmBnSTuSKo4HA/9WTCBpWm7tBvAa\n4Pb2hmjWezqiAilpMqnyeG5EfLvqeMzMzMyss0XEOknHAj8CNgbOiIhbJX0EuC4iFgLvlvQaUp8b\nDwLzKgvYrEdUXoFUer/A6cDtEfHZquMxMzMzs+4QEZcCl9aM+2Bh+ETgxHbHZdbLNqo6AOBFwOHA\nvoWXvM6tOigzMzMzMzMbqvI7kBFxJaCq4zAzMzMzM7PmOuEOpJmZmZmZmXUBVyDNzMzMzMysFFcg\nzczMzMzMrBRXIM3MzMzMzKwUVyDNzMzMzMysFFcgzczMzMzMrJTKX+NRpVknXDJs3NJPvrqCSKwb\neHsxa7965W7B/ptXEImZmZmB70CamZmZmZlZSa5AmpmZmZmZWSmuQJqZmZmZmVkprkCamZmZmZlZ\nKa5AmpmZmZmZWSmuQJqZmZmZmVkprkCamZmZmZlZKR1RgZS0v6TfSloi6YSq4zHbELjcmbWfy53Z\n+GpVpiRtIumbefrVkma1P0qz3lJ5BVLSxsCXgFcBs4FDJM2uNiqz3uZyZ9Z+Lndm46tkmToSeCgi\ndgI+B3yqvVGa9Z7KK5DA3sCSiLgjIv4KnA8cWHFMZr3O5c6s/VzuzMZXmTJ1IHBmHr4I2E+S2hij\nWc9RRFQbgPR6YP+IOCp/Phx4QUQcW5PuaODo/HEX4LcNFjkVuH+Cwh0tx1ROJ8YEY4trZkQ8ZTyD\nGQ8TUO6gN3+/iebYRqdVbC531XAs9XVSLDBx8bS93JUpU5JuyWmW589/yGnur1lWq3LXab9jM90U\nK3RXvJ0WayXHu0ntXuFoRcR8YH6rdJKui4g5bQipNMdUTifGBJ0bVzuULXfQufnUqXGBYxutTo5t\nPHRruXMs9XVSLNB58XSKVuWum/Ktm2KF7oq3m2KdSJ3QhHUFMKPweYc8zswmjsudWfu53JmNrzJl\nan0aSZOArYAH2hKdWY/qhArktcDOknaU9CTgYGBhxTGZ9TqXO7P2c7kzG19lytRC4Ig8/HrgZ1H1\n81tmXa7yJqwRsU7SscCPgI2BMyLi1jEsslSznzZzTOV0YkzQuXGN2gSUO+jcfOrUuMCxjVYnx9bQ\nBlDuHEt9nRQLdF48o9aoTEn6CHBdRCwETgfOlrQEeJBUyRyNbsq3booVuivebop1wlTeiY6ZmZmZ\nmZl1h05owmpmZmZmZmZdwBVIMzMzMzMzK6VrK5CS9pf0W0lLJJ1QZ/omkr6Zp18taVYHxDRP0n2S\nFuW/o9oQ0xmSVuX3INWbLklfzDHfLGmvDoipX9LqQj59cILjmSHpckm3SbpV0nF10rQ9n9qpUR5I\n2lbSZZJ+n/9vk8c/W9KvJf1F0nvrLG9jSTdK+n6D9ZUunxXEVrqcjmdskpZKWpzXeV2D9ZXaDiuI\nq3SZHefYtpZ0kaTfSLpd0gtHm2edTB12vCsRz3/k3/dmST+VNLOqWArpXicpJE1Y9/tlYpF0UGHb\n/0ZVsUh6ei6HN+bfae5ExdILym5nVVOL86tO0uhY0IkkbSrpGkk35Vg/XHVMlYuIrvsjPSj9B+AZ\nwJOAm4DZNWneAZyWhw8GvtkBMc0DTm1zXr0U2Au4pcH0ucAPAAH7AFd3QEz9wPfbmEfTgL3y8JbA\n7+r8dm3PpzZvJ3XzAPg0cEIefwLwqTz8VOAfgI8D762zvP8AvtHodxxJ+awgttLldDxjA5YCU1us\nr9R2WEFcpcvsOMd2JnBUHn4SsPVo86xT/+iw413JeF4ObJaH3z5R8ZSJpbCdXQFcBcypMF92Bm4E\ntsmfn1phLPOBt+fh2cDSidyOu/mv7HbWCX+0OL/qpL9Gx4Kq42oQq4At8vBk4Gpgn6rjqvKvW+9A\n7g0siYg7IuKvwPnAgTVpDiSdXABcBOwnSRXH1HYRcQWp17FGDgTOiuQqYGtJ0yqOqa0iYmVE3JCH\nHwFuB6bXJGt7PrVTkzwolqMzgdfmNKsi4lrg8dplSdoBeDXwtSarLF0+K4ittPGMraRS22EFcZU2\nXrFJ2op0snR6TvfXiHi4ziq7vex22vGuZTwRcXlEPJY/XkV6N18lsWQfBT4F/HmC4igby1uBL0XE\nQ5C27QpjCeDJeXgr4O4JiqUXdOT5XT2ddn7VTMlzr46Qjx9r88fJ+W+D7oW0WyuQ04Flhc/LGb7R\nrU8TEeuA1cB2FccE8LrcXOQiSTPqTG+3snG32wtzU4EfSNq1XSvNTb/2JF1dKurUfBp3NXnQFxEr\n86R7gL4Si/g88D7g703SjKp8tik2GEU5HYfYAvixpOslHd0gzYi3wzbFBaMos2OMbUfgPuDruRne\n1yRtXiddt5fdTjvejTQ/jyTdAa4kltxkeUZEXDJBMZSOBXgW8CxJv5R0laT9K4zlJOAwScuBS4F3\nTVAsvaDb9yEdr8m5V8dQevxlEbAKuCwiOjbWdujWCmS3+h4wKyKeC1zGE1eMbagbgJkR8TzgFOC7\n7VippC2AbwHviYg17Vhnp2mWBxERtLjiJukAYFVEXN/FsY24nI41tuzFEbEX8CrgnZJeWmKeTolr\nxGV2HGKbRGqq9eWI2BN4lNT01TqEpMOAOcBnKlr/RsBngeOrWH8dk0jNWPuBQ4CvStq6olgOARZE\nxA6kZt5n5/wya6tuOfeKiL9FxB6kFhV7S9qt6piq1K07ixVA8a7ADnlc3TSSJpGaaDxQZUwR8UBE\n/CV//Brw/AmMp6wyedlWEbFmsKlARFwKTJY0dSLXKWkyaQd2bkR8u06Sjsun8dYgD+4dbO6X/7dq\ncvUi4DWSlpKa+ewr6Zw66UZUPtsZ20jL6TjFRkSsyP9XAd8hNZuqVXo7bGdcIy2z4xTbcmB54Srw\nRaQKZa1uL7uddrwrlZ+SXgH8N/CaQnlqdyxbArsBA7nc7wMsnKCOdMrky3JgYUQ8HhF3kp752rmi\nWI4ELgCIiF8DmwITepztYt2+D+lYJc69Ok5+VOJyYKJaEHSFbq1AXgvsLGlHSU8idRqwsCbNQuCI\nPPx64Gf5qnZlMdU8d/MaUnvvqi0E3qRkH2B1oRlZJSQ9bfD5HUl7k7bTCav853WdDtweEZ9tkKzj\n8mk8NcmDYjk6Ari42XIi4sSI2CEiZpHKwM8i4rA6SUuXz3bHNpJyOl6xSdpc0paDw8A/AfV60Su1\nHbY7rpGU2XH8Pe8BlknaJY/aD7itTtJuL7uddrwrc6zbE/gKqfI4Uc/5tYwlIlZHxNSImJXL/VU5\nprq9CU9kLNl3SXcfyRdYngXcUVEsfySVGSQ9h1SBvG8CYukFZfLTRqjkuVdHkPSUwdYCkqYArwR+\nU21UFYsO6MlnNH+kJhe/I/WM9d953EdIBwdIO8MLgSXANcAzOiCmTwC3knrwuhx4dhtiOg9YSeqA\nYjnpquMxwDF5uoAv5ZgXM0E91I0wpmML+XQV8I8THM+LSc3lbgYW5b+5VedTO/+a5MF2wE+B3wM/\nAbbN6Z+Wf7s1wMN5+Mk1y+yn0DPnaMtnBbGVLqfjFRupd7+b8t+t5P1HnmfE22EFcZUus+P5ewJ7\nANflZX2XJ3q37KmyS4cd70rE8xPg3sLvu7CqWGrSDkzk718iX0RqUntb3hYPrjCW2cAvc5ldBPxT\nu7bnbvyrl5+d+Eed86uqY2oSa91jQdVxNYj1uaQelG8mXUT9YNUxVf2nnDFmZmZmZmZmTXVrE1Yz\nMzMzMzNrM1cgzczMzMzMrBRXIM3MzMzMzKwUVyDNzMzMzMysFFcgzeqQdIakVZLqvUqhNu3nJC3K\nf7+T9HA7YjQzMxurER7vni7pckk3SrpZ0tx2xGjWa7q93LkXVrM6JL0UWAucFRG7jWC+dwF7RsRb\nJiw4MzOzcTKS452k+cCNEfFlSbOBSyO9Y9PMRqDby53vQJrVERFXAA8Wx0l6pqQfSrpe0i8kPbvO\nrIeQ3sNkZmbW8UZ4vAvSO2IBtgLubmOoZj2j28vdpKoDMOsi80kvJ/+9pBcA/wfsOzhR0kxgR+Bn\nFcVnZmY2Hhod704Cfpxb22wOvKK6EM16TteUO1cgzUqQtAXwj8CFkgZHb1KT7GDgooj4WztjMzMz\nGy8tjneHAAsi4mRJLwTOlrRbRPy9glDNeka3lTtXIM3K2Qh4OCL2aJLmYOCdbYrHzMxsIjQ73h0J\n7A8QEb+WtCkwFVjVxvjMelFXlTs/A2lWQkSsAe6U9AYAJc8bnJ7bqW8D/LqiEM3MzMasxfHuj8B+\nefxzgE2B+yoJ1KyHdFu5cwXSrA5J55Eqg7tIWi7pSOBQ4EhJNwG3AgcWZjkYOD/crbGZmXWRER7v\njgfemsefB8zzcc9s5Lq93Pk1HmZmZmZmZlaK70CamZmZmZlZKa5AmpmZmZmZWSmuQLYg6SRJ57Rp\nXe+X9LUS6U6T9P/aEVO7SQpJO410mplZr5LUL2n5BCx3raRnjPdyx0LSrZL627SuBZI+1o519QJJ\nb5d0b95utqs6nk5Wm1c+fylH0oCko0Y57/pzaEmzcp7XfdtE8dxe0tPz77Tx6CPvfJL+RdKy/F33\nHOvyXIFsg7IH/4j4n4hoWXAi4piI+OhIlt3rJM2TdGXVcZiNhLdbq1JEbBERd5RJ264T4IjYNSIG\n8jrbdgG3SpL2kXSZpAcl3SfpQknTqo6rSNJk4LPAP+Xt5gFXiuqrl1dVx7QhKHsOXWe+P+bfqdff\n4f2/wLH5u94oaamkV4x2Ya5AdohGV0nMrH26rRxWGW+35ZVZB9sGmA/MAmYCjwBfrzKgOvpIrw64\ntepARqPN+6u25NV4fqdev/tmQNq3jNs26QpkJml7Sd/KV//ulPTuBun2kfQrSQ9LuqnY1EbStpK+\nLuluSQ9J+q6kzYEfANvn28Zr87pOknSRpHMkrQHm1V5tlfTiwrqWSZqXxy+Q9LEmy36s2LxE0l75\ne01u8v1r1z3k9n++U3KHpEdy/hxaSPsWSbfn7/wjSTNH/APUj+nF+Xv3F0bPzXHcL+kzkjZSeifO\nacALcx48nOcf0hTCd3u6W6MymrfdCySdlbfPWyXNaTVfYd7acjhF0pl5e75d0vsG7/JL+k9J36qJ\n64uSvtAi9mHlp8l2+2pJN0pak7f/kwrLGSyXR0r6I/CzJuvcNH+vB/I+5FpJfXnasH1VYb63Slqi\ndDdkoaTtC9NC0jsl/R74fR73bD1x9+S3kg5qlhfWWN5X35i3kwslfVOFJpaSjpe0StJKSW8ujN9O\n0vfyNnNtPj603NepcAdJ6bjyJUmX5PVfLemZedoVeZab8rb6xjz+AEmL8vb1K0nPLSx7qaT3SrpZ\n0ur8XTbN06ZK+n6e70FJv5C0UWG+V0jaH3g/8Ma8zpskvUHS9TXf4T8kXTzKLC8uZ0tJl+fyrDHk\n6a6F8nCvpPfn8ZtI+nwuc3fn4U0AIuIHEXFhRKyJiMeAU4EXFZa5iaT/lfTHvMzTJE1pEUezPN5T\n0g35d/6mpPPVpCmvpGcBv80fH5b0s0bbRLs1KjPKrbMk/Zeke8gV8kb7t/ybfy6XrzWSFkvaLU+b\nK+m2vI4Vkt7bJJ5heVUnzVZKx6v7JN0l6QOF32aj/PmuHMtZkrbK00rv/3P6CyXdk8vfFZJ2LUxb\nIOnLki6V9Cjw8mbbmaRt8vZ0n9Ix4/uSdij9Q43MMyVdk3+HiyVtm2MY1uJOhbtoatJiQdKOkn6e\nf8PLgKmFabXnuwOSPirplzn9jyUV078p/z4PSPp/NTHsLem6HPu9kj47mgxQ8+P3kO8i6dRG3zun\n30TSWmBjUnn9g6SzgacD38vl930jDjIiNvg/UkX6euCDwJOAZwB3AP8MnASck9NNBx4A5uZ5Xpk/\nPyVPvwT4Julq4mTgZXl8P7C8Zp0nAY8Dr83LmlKzrsGrkIfkZW0H7JGnLQA+1mTZlwJvL3z+HHBK\nizxYv+78eRYQwCRgc2ANsEueNg3YNQ8fCCwBnpPTfgD41Rh+iwB2AvYHlgF710y7HNiWtOH/Djgq\nT5sHXFmzrIHB6Y3S+K87/kqU0T/ncrkx8Angqlbz5en1yuEngZ/ncrwDcPNgGcvb/qPA1vnzJGAV\n8PwmsTcrP/W2235g9xzPc4F7gdfmaYPl8qy83ClN1vs24HvAZjlfng88OU9rtK/aF7gf2AvYBDgF\nuKKwzAAuy2VwSo5hGfDmnBd75vlnV73NdNtf3j7vAo7Lv8m/An8FPpa3iXXAR/K0ucBjwDZ53vPz\n32bA7PybtNzX5d9zpzy8gHQ82zv/lueS3m07LG3+vGfe9l+Qt68jgKXAJnn6UuAaYPu8vdwOHJOn\nfYJ08WRy/nsJT7xWbCnwikL5LB6XNgEeBJ5TGHcj8LpR5vmCnL/b5Vg/Vpg24jwFtgRWkt7Ztmn+\n/II87SPAVcBTgacAvwI+2mA57yHvw/LnzwELcz5uSSrXn2gRS908Lmxn/57Hv560D/xYi+XNytvA\npEbbRIeWmU/l7WYKTfZvpGPJ9cDWOZ+eA0zL01YCL8nD2wB7jSWvSPvvi/NvOYt0LnNknvYW0jnV\nM4AtgG8DZ9cst+X+v7CsLfN3/TywqGbbX026ULFR3l4bbmekMvI6UnnYErgQ+O4E/KYDwApgt/wd\nv8UT58X9DD/fXUqd/UXtb0B63+Jnc168lHR+3SjtAPAH4Fl5uxkAPpmnzQbWAi8mbX//Syo/ryis\n5/A8vAWwzyjzodnxu+F3abHM2n34+rwbVYxVFfxO+iMdAP9YM+5E0hWr4gb5X4MFuZDuR6QD5zTg\n7+QDek2aehv9SRROzOps/CcC32kQ7wKaVyDfCPwyD28M3EOhItZgmevXnT+vL1CkQvwwaecxpWa+\nH5B3fPnzRqQTm5mj/C0if/e7gN3qTNu/8PkdwE/z8DxcgezZvxJl9CeF8bOBP7WaLw/XK4frK5j5\n81HFMpa3+bfm4QOA21rE3qz8tNwmSQf+z+XhwXL5jBJ59hbSSepza8Y321edDny68HkL0sFxVv4c\nwL6F6W8EflGzjK8AH6p6m+m2P9KJwApyRSqPu5InTob/xNAT0lXAPqR9/OPkCxR52sfK7OsYXoH8\nWmHaXOA39dLmz1+mpgJEuvPysjy8FDisMO3TwGl5+COkE+hhlQ+aVCAL6/14Ht4VeIhcaR1Fni8A\nzgBuAf6zMH5UeUq64Htjg2l/AOYWPv8zsLROuueSKsmDFRaRLlo9s5DmhcCdLWKpm8d5O7u7Zjv7\nFd1ZgWxVZv4KbFqY1nD/Rqpc/i6XqY1q1vNH0gn9k0vG1TCv8rb1VwoX2fKyB/LwT4F3FKbtkmOc\nxAj2/3Vi2jrPu1Vh2z+rMH1E2xmwB/DQBPymA+TKWv48O+fXxoyyAkm64bAO2Lww3zfqpS3E8IFC\n2ncAP8zDHwTOK0zbLMc3GMMVwIeBqWPMh0bH76bfpcUyx7UC6SasyUxSM9CHB/9ITWf66qR7Q026\nF5NOyGYAD0bEQyNY77Im02aQDjijcTEwW9KOpLukqyPimlEui4h4lHSieAywUqmJ07Pz5JnAFwr5\n8SBpRzR9tOsjXX29ICJuqTOtmGd3ka5uW+9rVUbvKaR9DNg0N0cpU7Zry+H2NeNqp58JHJaHDwPO\nbhZ4i/IzjKQXKDWlu0/S6jzf1JpkzfYdg84mXeA6X6nJ3KeVmrE321dtTypXg7GvJd2VKpbn4rpn\nAi+oyd9DgaeViM+G2h5YEfnInhXz+oGIWFf4/BjpBPgppJOkZttsWbXlaIsmaWcCx9f89jMYuk9u\ntLzPkO6y/FipafcJI4jxTODfJAk4nHSs+MsI5q/1atJdhtMK40abp82O20PKFnWOX0rNiX8AHBcR\nvyjEshlwfSGff5jHN9Moj+ttZ3cNm7s7tCoz90XEn2vS192/RcTPSE2HvwSskjRf0pNz0teRLqjc\nlZsOvnAMMU8l3S2t3RYG97H1tpNJND9mDSNpY0mfzM0V15AqC4Prr7ecptuZpM0kfSU33VxDqiht\nrYl5drL2PG8yw4+BI7E9qbL7aM1ym2m07xpyfhCpyXmxk6QjSXcuf5ObnR4wypgbHb9H810mhCuQ\nyTLSVZatC39bRsTcOunOrkm3eUR8Mk/bVtLWdZYfdcY1Gz+4rmeWiH3YMvIO8wLSye3htDjBzR4l\n7TwGDTkBjIgfRcQrSZXl3wBfLcT5tpo8mRIRvyqxzkbeALxW0nF1ps0oDD+ddCUV6udl0+9kXaVs\nGR3NfLXbzkpS09VBM2qmfxd4rtLzMQeQmvo11aT81Ntuv0FqRjQjIrYindiqdpEl1vl4RHw4ImYD\n/5hjfRPN91V3kyoGACg9Z70d6Sp/vXUvA35ek79bRMTbW8Vnw6wEpueK0aDaba+e+0hXpJttsxNh\nGelOYPG33ywizms1Yyku+qcAACAASURBVEQ8EhHHR8QzgNcA/yFpv3pJ68x7FemK/0uAf6Pc8a2Z\nr5JOlC/N2zuMPk+XkZoe1jOkbDH0+IVS3wE/Id3VLX6n+0l3n3ct5PNWEdGsct8sj+ttZ08v8d06\nUasyU7v9NN2/RcQXI+L5pLtezwL+M4+/NiIOJDU//i7p/Gq07ifdUazdFgb3sfW2k3WkRxkafa96\n/o30iNErgP/P3r3H2VXX9/5/vSXciYSLTmNCCa0Ui6RcGhEP1o6ASgEJPYfmBBETxaa1gqBpNVLP\nAS94YltEihWMogQbCRjhFwSlpsiU0iNRApEAQUEIkJALAoEMesTA5/fH9zu4M9l7Zs/M3mvty/v5\neMxj9v6utfb6rLXXd6/1Xet72ZP0lA22PZdUfs5wx9lc0tPQN0bEK0lPfwd/XqMMvs77TY5vm2u6\nXHgd7kYKpONkr4r8PfC5o7HN9YFSG9GX+xyJiAcj4jTSsfI5YMmg9dZliPN3I7elnuOoJhcgkx8B\nW5QaW++a79wcIukNg+b7V+Cdkt6R59lFqVHv5IhYT7pz+CWlxsY7ShrIYBuBfZQbQtdpEXCcpBmS\nxik16D+syny1PvsqUvW4k6nvBLsSeIvSeDh7kqr5ASCpR9L0fMD+mlT/+6U8+XLg48qNs5Uah/9F\n/ZtZ1RPAscA5kgZfiP5d3r/7kdo9XJPTNwKTJe00aJv+e75z9lrSnSFrT/Xm0UYsdy3pmN5L0iTg\nrMqJ+QbNElJB70cR8dhQAQyTf6odt+NJTwj/n6QjSRcCIybprZKm5pPsc6ST8EvD/FZdDbxX0mFK\nHXx8FlgeEWtqrOZG4A8knZE/Z0dJb1DqIMhG5ofAi8BZ+Td/Oqk94pAidT1/HXBB/q17HelCo9E2\nsm3h6CvAXys9MZek3ZU6gBo/3Acpdb7z2nzh/yxpu1+qMutGYIpyByMVriI9LfpNRDSiY7SzSNVv\nvyNp1zHs0xuBiZLOVeq4YrykN+ZpVwOfkPQqpQ45/jfpmoL8O/MD4IsRUfkklIh4ibSvL5b06oH5\nJb1jqECG2Mc/JBVIPpTz63+njuOshsHHRNFGmmdq/r7l3603Kj3leZ7Urv4lSTspdXq2Z0T8hvRb\nWu1YrUs+tq4FLszHx/7AR8jHQo7xw0odpeyRY7xmUO2DeownnW+eIhW6PjtMXMMdZ+NJBczNSp3a\nnD/CeEbi3ZIOlrQbqSr2krzffkaqXXRi/p4+QWoHOKSIeBS4E/hk/j7fDLxzlLEtIZUD/ls+b19A\nRSFa0rslvSrvz805ecTHyxDn70Zuy5jyrwuQvJyhTyLV6X6EdKfjq6S7NpXzPU66o3Me6Q7l46Q7\nVAP78QzSl/wAqX3KuXm5B0g/Cg8rVQ0Yttplvig9gXTX52lSYejQKvNV/eyI+C/SQXtXPuCGW98y\nUmHsHlJD8hsrJr+C9AP3RI7lT4EP5OWuJ91lWaxUreFe4M+GW18d8TxGKkTO07aDyi7N8a0kdQRy\nRU7/Aal74g2SfpHTLibdqd5IqvY07JMia0315tEGLfcpYG2e/99JJ4zBVeQWkjq6qefmTM38Q/Xj\n9m+AT0naQrrIHO3d7t/JsT9H6sDkPyrirfVb9e/A/yJ1XLCeVAtiZq0VRMQW4O15nidI1X4GOq2w\nEYiIF0idgJxJuvB4N+l3uJ7qmWeRjukNpO/46jqXG4kLgIX5PDMjIu4E/pJUkHuGVF1ydp2fdSAp\nb/WTCgFfiohbq8z3rfz/KUl3VaR/g9TJRkPGiMxVIOeQ8v1Spd5iR7xPc354G+mCbgOpp+K35smf\nIV343QOsAu7KaZDaWf8eqcA60KN6f8VHf4y0f+/I59l/Jz0NGkrVfVxxnM0m/R79T1JheTQuoOKY\nGOVnjNpI88wwv2+vJBWgniFVCXyKVA0Y0u/lmrzv/5pUTX8sziYVUh8mtdn8JqktLvn/N0hVRB8h\nFWTPHsU6riJtxzrgflIHTsMZ6jj7Aqmq9y/yZ908ipjq9Q1SG80NpM59PgQQEc+Szo9fJW3X86Q8\nW493kfpEeJpU+L1qNIFFxH2k72Mx6RjqJ51DB46544H7cv69BJgZEb8axaqGOn83ZFtIHW19Iuff\nmj0L1zLQ65l1IKXuo78ZEV8tOxazdqX0FHxmRPxpRdrvkgpfvxMRz5UWnHU0SctJHc98fYTLfY50\nbM5qTmTlUqo2tonUG+aDBa2zY/eppCtJnZN8ouxYxmq0ecZsNPJT4s3AgRHxSEkxXEDqHOfdw83b\nSH4C2aGUqugdwW+reJpZHSRNlHS00nhcB5FqAVxfMX3gieJiFx6tkST9qaTfydXxZpF65Bz2Tr/S\nWJx/lKuSHkl6InP9cMu1sQ8AP25m4bEL92lbGm2eMRstSe/MVdt3Jw3jsYrfdlLUNVyA7ECSFpKq\nHpybq9QMpF9eWT2m4u/y2p826hj+pMa6XqyR3j/8pzY9tsJisJa2E2koii2kKqZLgS/By50uPEeq\nprZNG5Bax5SkP2lWoEptc6qt875mrdOa6iDgJ6Q72nOBU3Ob1eGMJ1VDfJ500/AiUlXMjvutk7SG\n1P597qD0+2ps6yM10oerhtjy+1TSeTVi+V7Rn9foWEZgtHlmTErc3soYWuL3f4hzX63rvVrpTTtX\nNth0UpONJ0hVxWfGENU5h/ieav02jer7G8vxMJplXYXVzMzMzMzM6uInkGZmZmZmZlaXcWUHMBr7\n7rtvTJkypeq0559/nt13H/GQKx3B2956275ixYpfREQ94xS1vFr5rlX3/QDHNzbtGF835Lt20urH\nUKN0+3Z2S75rl+/ZcTZWq8ZZVr5rywLklClTuPPOO6tO6+vro7e3t9iAWoS3vbfsMLYjadghVNpF\nrXzXqvt+gOMbm3aMrxvyXTtp9WOoUbp9O7sl37XL9+w4G6tV4ywr37kKq5mZmZmZmdXFBUgzMzMz\nMzOriwuQZmZmZmZmVpe2bANpZmbWyvJ4hVuAF4GtETFN0t6kMQWnkAaenhERz5QVo5mZ2Wj4CaSZ\nmVlzvDUiDouIafn9POCWiDgQuCW/NzMzaysuQJqZmRVjOrAwv14InFJiLGZmZqPScVVYV617ltnz\nbtoufc38E0uIxsyse0zxb2+lAL4vKYAvR8QCoCci1ufpG4CeagtKmgPMAejp6aGvr68pAa5a9+x2\naVMn7dnw9fT39zdtG1qJt7O7+ffPuknHFSDNzMxawJsjYp2kVwPLJD1QOTEiIhcut5MLmwsApk2b\nFs0ae6zqzdbTG7+uVh0/rdG8nWbWLQqtwippgqQlkh6QtFrSmyTtLWmZpAfz/72KjMnMzKzRImJd\n/r8JuB44EtgoaSJA/r+pvAjNzMxGp+g2kJcAN0fE64BDgdW4UwEzM+sgknaXNH7gNfB24F7gBmBW\nnm0WsLScCM3MzEavsCqskvYE3gLMBoiIF4AXJE0HevNsC4E+4GNFxWVmZtZgPcD1kiCdZ78ZETdL\n+jFwraQzgUeBGSXGaGZmNipFtoE8AHgS+LqkQ4EVwDk0uFOBnl1h7tSt26V3Q4Pvbm7Y3s3bbmat\nJSIeJtWyGZz+FHBs8RGZmZk1TpEFyHHAEcDZEbFc0iUMqq7aiE4FLl20lItWbb9ZzegYoNV0c8P2\nbt52MzMzM7OiFNkGci2wNiKW5/dLSAVKdypgZmZmZmbWBgorQEbEBuBxSQflpGOB+3GnAmZmZmZm\nZm2h6HEgzwYWSdoJeBh4L6kQ604FzMzMzGzUJH0YeD8QwCrSdeZEYDGwD6n/jTNyR45mNkqFFiAj\nYiUwrcokdypgZmZmZqMiaRLwIeDgiPiVpGuBmcAJwMURsVjS5cCZwGUlhmrW9ooeB9LMzMzMrBnG\nAbtKGgfsBqwHjiH1uwFpuLhTSorNrGMUXYW1NFPm3bRd2pr5J5YQiZmZmZk1UkSsk/RPwGPAr4Dv\nk6qsbo6IgfHd1gKTqi1f73BxtYYNa7Uh5NpleDPH2Z66pgBp1o0k7QdcRRpfNYAFEXGJpAuAvySN\nzQpwXkR8t5wozczMxkbSXsB00rjjm4FvAcfXu3y9w8XVGjZsdrUHFSUOIdcuw5s5zvbkAqRZZ9sK\nzI2IuySNB1ZIWpanXRwR/1RibGZmZo1yHPBIRDwJIOk64GhggqRx+SnkZGBdiTGadQS3gTTrYBGx\nPiLuyq+3AKupUX3HzMysjT0GHCVpN0nit8PF3QqcmufxcHFmDeAnkGZdQtIU4HBgOemu7FmS3gPc\nSXpK+UyVZYZtE9Lq7QIc39iMJL4y2gC1+v4zs2JExHJJS4C7SLVv7iZVSb0JWCzpMzntivKiNOsM\nLkCadQFJewDfBs6NiOckXQZ8mtQu8tPARcD7Bi9XT5uQVm8X4PjGZiTxldEGqNX3n5kVJyLOB84f\nlPwwcGQJ4Zh1LBcgzTqcpB1JhcdFEXEdQERsrJj+FeDGksIzsxbiHsvNzGw4bgNp1sFyO5ArgNUR\n8fmK9IkVs/05cG/RsZmZmZlZ+/ETSLPOdjRwBrBK0sqcdh5wmqTDSFVY1wB/VU54ZmZmZtZOXIA0\n62ARcTugKpM85qOZmZmZjZirsJqZmZmZmVldXIA0MzMzMzOzurgAaWZmZmZmZnVxAdLMzMzMzMzq\nUmgnOpLWAFuAF4GtETFN0t7ANcAUUm+QMyLimSLjMjMzMzMzs+GV8QTyrRFxWERMy+/nAbdExIHA\nLfm9mZlZW5O0g6S7Jd2Y3x8gabmkhyRdI2mnsmM0MzMbqVaowjodWJhfLwROKTEWMzOzRjkHWF3x\n/nPAxRHxWuAZ4MxSojIzMxuDoseBDOD7kgL4ckQsAHoiYn2evgHoqbagpDnAHICenh76+vqqrqBn\nV5g7dWtdwdT6jHbV39/fcdtUr27edjNrPZImAycCFwIfkSTgGOBdeZaFwAXAZaUEaGZmNkpFFyDf\nHBHrJL0aWCbpgcqJERG5cLmdXNhcADBt2rTo7e2tuoJLFy3lolX1bdaa06t/Rrvq6+uj1n7pdN28\n7WbWkr4AfBQYn9/vA2yOiIE7nGuBSdUWrPeG6VgVdbO1W27weTvNrFsUWoCMiHX5/yZJ1wNHAhsl\nTYyI9ZImApuKjMnMzKyRJJ0EbIqIFZJ6R7p8vTdMx2r2vJvqmm+sN1u75Qaft9PMukVhbSAl7S5p\n/MBr4O3AvcANwKw82yxgaVExmZmZNcHRwMm55/HFpKqrlwATJA3cuJ0MrCsnPDMzs9ErshOdHuB2\nST8BfgTcFBE3A/OBt0l6EDguvzczM2tLEfHxiJgcEVOAmcAPIuJ04Fbg1Dybb5iamVlbKqwKa0Q8\nDBxaJf0p4Nii4jAzMyvJx4DFkj4D3A1cUXI8ZmZmI1Z0JzpmZmZdIyL6gL78+mFS238zM7O21Qrj\nQJqZmZmZmVkbcAHSzMzMzMzM6uIqrGZmNmJT6hwCwsys0/j3z7qdn0CamZmZmZlZXVyANOtgkvaT\ndKuk+yXdJ+mcnL63pGWSHsz/9yo7VjMzMzNrfS5AmnW2rcDciDgYOAr4oKSDgXnALRFxIHBLfm9m\nZmZmNiQXIM06WESsj4i78ustwGpgEjAdWJhnWwicUk6EZmZmZtZO3ImOWZeQNAU4HFgO9ETE+jxp\nA9BTY5k5wByAnp4e+vr6tpunv7+/anqrcHxjUyu+uVO31rV8s7et1fefmZlZp3EB0qwLSNoD+DZw\nbkQ8J+nlaRERkqLachGxAFgAMG3atOjt7d1unr6+PqqltwrHNza14ptdZy+Ea07fftlGavX9Z2bF\nkTQB+CpwCBDA+4CfAtcAU4A1wIyIeKakEM06gquwmnU4STuSCo+LIuK6nLxR0sQ8fSKwqaz4zMzM\nGuQS4OaIeB1wKKnZhtv8mzWYC5BmHUzpUeMVwOqI+HzFpBuAWfn1LGBp0bGZmZk1iqQ9gbeQznlE\nxAsRsRm3+TdrOFdhNetsRwNnAKskrcxp5wHzgWslnQk8CswoKT4zM7NGOAB4Evi6pEOBFcA5NLDN\nP6R213OnvlhXQGW2z26X9uGOsz25AGnWwSLidkA1Jh9bZCzWvqbU2d7RzKxE44AjgLMjYrmkSxhU\nXXWsbf4hFQovuv35ugJqdhvwobRL+3DH2Z5chdXMzMzM2t1aYG1ELM/vl5AKlG7zb9ZghRcgJe0g\n6W5JN+b3B0haLukhSddI2qnomMzMzMysfUXEBuBxSQflpGOB+3Gbf7OGK+MJ5DmkXrEGfA64OCJe\nCzwDnFlCTGZmZmbW3s4GFkm6BzgM+Cypzf/bJD0IHJffm9kYFNoGUtJk4ETgQuAjuYfIY4B35VkW\nAhcAlxUZl5mZmZm1t4hYCUyrMslt/s0aqOhOdL4AfBQYn9/vA2yOiK35/VpgUrUF6+0dq2dXmDt1\na9Vpg3Vab0rd3ENUN2+7mZmZmVlRCitASjoJ2BQRKyT1jnT5envHunTRUi5aVd9mldk7VjN0cw9R\n3bztZqNRrWfVNfNPLCESK4J70jUzs0Yp8gnk0cDJkk4AdgFeCVwCTJA0Lj+FnAysKzAmMzMzMzMz\nq1NhnehExMcjYnJETAFmAj+IiNOBW4FT82zuHcvMzMzMzKxFtcI4kB8jdajzEKlN5BUlx2NmZjZq\nknaR9CNJP5F0n6RP5nQPW2VmZm2v6E50AIiIPqAvv34YOLKMONwGyMzMmuDXwDER0S9pR+B2Sd8D\nPkIatmqxpMtJw1a513EzM2srrfAE0szMrGNE0p/f7pj/gjRs1ZKcvhA4pYTwzMzMxqSUJ5BmZmad\nTNIOwArgtcC/AD+nwcNWjUS9w1tVM9b1d8swS95OM+sWLkCamZk1WES8CBwmaQJwPfC6ESxb17BV\nIzF7DMN4jHXIq24ZZsnbaWbdwlVYzczMmiQiNpN6G38TediqPMnDVpmZWVtyAdLMzKyBJL0qP3lE\n0q7A24DVeNgqMzPrAK7CamZmQPWeqVPbOZ8qRmgisDC3g3wFcG1E3CjpfmCxpM8Ad+Nhq8zMrA35\nqsDMzKyBIuIe4PAq6aUNW2VmZtYorsJqZmZmZmZmdXEB0qyDSfqapE2S7q1Iu0DSOkkr898JZcZo\nZmZmZu3DBUizznYlcHyV9Isj4rD8992CYzIzMzOzNuUCpFkHi4jbgKfLjsPMzMzMOoM70THrTmdJ\neg9wJzA3Ip6pNpOkOcAcgJ6eHvr6+rabp7+/v2p6q3B81aXeVYfXs2v981bT7G1r9e/XzMys07gA\nadZ9LgM+DUT+fxHwvmozRsQCYAHAtGnTore3d7t5+vr6qJbeKhxfdbOrDNlRzdypW7lo1ehPFWtO\n7x31svVo9e/XzMys07gKq1mXiYiNEfFiRLwEfAUPK2BmZmZmdXIB0qzLSJpY8fbPgXtrzWtmZmZm\nVqmwKqySdgFuA3bO610SEedLOgBYDOwDrADOiIgXiorLrJNJuhroBfaVtBY4H+iVdBipCusa4K9K\nC9DMzMzM2kqRbSB/DRwTEf2SdgRul/Q94COkIQUWS7ocOJPURsvMxigiTquSfEXhgZiZmZlZRyis\nCmsk/fntjvkvgGOAJTl9IXBKUTGZmZmZmZlZ/QptAylpB0krgU3AMuDnwOaIGOgjfi0wqciYzMzM\nzMzMrD6FDuMRES8Ch0maAFwPvK7eZesZjw5af8yyZurm8dC6edvNzJppSo0hX9bMP7HgSMyGJ2kH\n0hjH6yLiJPe1YdZ4pYwDGRGbJd0KvAmYIGlcfgo5GVhXY5lhx6MDuHTR0pYes6yZunk8tG7edjMz\nM3vZOcBq4JX5/edwXxtmDVVkL6yvAn6TC4+7Am8jZepbgVNJd4dmAUuLiqmaandafZfVzMzMrLVJ\nmgycCFwIfESSSH1tvCvPshC4ABcgzcakyCeQE4GFuWrBK4BrI+JGSfcDiyV9Brgb9xBpZmZmZiP3\nBeCjwPj8fh/q7Guj3qZS/f39zJ36Yl3BlNm0pl2a9jjO9lRYATIi7gEOr5L+MHBkUXGYmZmZWWeR\ndBKwKSJWSOod6fL1NpXq6+vjotufr+szy2wW1S5NexxneyqlDaSZmZmZWQMdDZws6QRgF1IbyEuo\ns68NM6tfocN4mJmZmZk1WkR8PCImR8QUYCbwg4g4nd/2tQEt0NeGWSdwAdLMzMzMOtXHSB3qPERq\nE+m+NszGyFVYzcysUO7t2syaKSL6gL782n1tmDWYn0CamZk1kKT9JN0q6X5J90k6J6fvLWmZpAfz\n/73KjtXMzGyk/ATSzMyssbYCcyPiLknjgRWSlgGzgVsiYr6kecA8UvU6M+tArm1hncpPIM3MzBoo\nItZHxF359RZgNWnsuemkgczJ/08pJ0IzM7PR8xNIMzOzJpE0hTQG8nKgJyLW50kbgJ4ay9Q1oHkt\nq9Y9u13a3Kkj+oi61BtXtwzA7e00s27hAqSZWYerVo3Kmk/SHsC3gXMj4jlJL0+LiJAU1Zard0Dz\nWmYX9H3XO0h6twzA7e00s27hKqxmZmYNJmlHUuFxUURcl5M3SpqYp08ENpUVn5mZ2Wi5AGlmZtZA\nSo8arwBWR8TnKybdQBrIHDyguZmZtSkXIM06mKSvSdok6d6KNA8lYNZcRwNnAMdIWpn/TgDmA2+T\n9CBwXH5vZmbWVtwG0qyzXQl8EbiqIm0eHkrArGki4nZANSYfW2QsZtZaPLSHdQI/gTTrYBFxG/D0\noGQPJWBmZmZmo+InkGbdp66hBKC+4QRavUt3xwdzp24d9bI9u45t+UsXbd/Mr9qQEqPdB63+/ZqZ\nmXUaFyDNuthQQwnk6cMOJ9DqXbo7vrEN6zB36lYuWtX8U0W9Q0IM1urfr5mZWacprAqrpP0k3Srp\nfkn3STonp7tDD7NieSgBMzMzMxuVIttAbgXmRsTBwFHAByUdzG879DgQuCW/N7Pm8VACZmZmZjYq\nhRUgI2J9RNyVX28BVgOTcIceZk0j6Wrgh8BBktZKOhMPJWBmZmZmo1RKG0hJU4DDgeXU2aFHPZ15\nwNg7fKimXTpo6ObOJLp524cSEafVmOShBDpUtS7izczMzBql8AKkpD2AbwPnRsRz0m+HyhqqQ496\nOvOA1ONfozt8GG3nDkXr5s4kunnbzczMzMyKUug4kJJ2JBUeF0XEdTnZHXqYmZmZmZm1gcKeQCo9\narwCWB0Rn6+YNNChx3zcoYeZmVlbqFZdes38E0uIxMzMilRkFdajgTOAVZJW5rTzSAXHa3PnHo8C\nMwqMyczMzMzMzOpUWAEyIm4HVGOyO/QwMzMzMzNrcYW2gTQzMzMzM7P25QKkmZmZmZmZ1cUFSDMz\nMzMzM6uLC5BmZmZmZmZWFxcgzczMzMzMrC4uQJqZmZlZW5O0n6RbJd0v6T5J5+T0vSUtk/Rg/r9X\n2bGatTsXIM3MzMys3W0F5kbEwcBRwAclHQzMA26JiAOBW/J7MxsDFyDNzMzMrK1FxPqIuCu/3gKs\nBiYB04GFebaFwCnlRGjWOcaVHUA7mDLvprrnXTP/xCZGYmZmZmZDkTQFOBxYDvRExPo8aQPQU2OZ\nOcAcgJ6eHvr6+qp+dn9/P3OnvtjQeGutayz6+/ub8rmN5jjbkwuQZmZmDSTpa8BJwKaIOCSn7Q1c\nA0wB1gAzIuKZsa5rJDc4zbqBpD2AbwPnRsRzkl6eFhEhKaotFxELgAUA06ZNi97e3qqf39fXx0W3\nP9/QmNecXn1dY9HX10etbWgljrM9uQBpZjZItYty1y5orloFoTbd71cCXwSuqkgbaIc1X9K8/P5j\nJcRm1rEk7UgqPC6KiOty8kZJEyNivaSJwKbyIjTrDG4DaWZm1kARcRvw9KBkt8MyayKlR41XAKsj\n4vMVk24AZuXXs4ClRcdm1mn8BNKsS0laA2wBXgS2RsS0ciMy62h1tcOC+ttiAcydurWBIY7dpYu2\nvzY/YM8duqLtULe0kWrh7TwaOANYJWllTjsPmA9cK+lM4FFgRknxmXUMFyDNuttbI+IXZQdh1k2G\naoeVp9fVFgtgdhu0gbzy+N27ou1Qt7SRatXtjIjbAdWYfGyRsZh1OldhNTMza76Nuf0VbodlZmbt\nrLAnkEX2SmdmdQng+/lJyJfzU49t1FOVbqA606p1z243beqkPRsd84iNprpVtWqBzaqyNTi+avsR\n6t+Xja7S2LNrudUkh9vvLVydbrCBdljzcTssMzNrY0VWYb0S90pn1kreHBHrJL0aWCbpgdz5x8vq\nqUo3UJ2pWlW6ZnRNPlKjqW5V5LYMjq9WlcR619/oKo1zp27lolXltXYYbrtbsTqdpKuBXmBfSWuB\n83E7LDOrocN6obYuUNhVQUTclgd2rTSddJKF1CtdHy5AmhUiItbl/5skXQ8cCdw29FJmNpyIOK3G\nJLfDMjOztld2JzoN75Wu1atbNVMbVeVquG7e9tGQtDvwiojYkl+/HfhUyWGZmZmZWYsruwD5skb1\nSnfpoqUtXd2qmVqxKldRunnbR6kHuD4Nm8U44JsRcXO5IZmZmZlZqyu7ALlR0sSIWO9e6cyKExEP\nA4eWHYeZmZmZtZeyC5Dulc7MrA7VOllwBwtmZmZWtMLGgcy90v0QOEjS2twT3XzgbZIeBI7L783M\nzMzMzKwFFdkLa1f0SuenBGZmZmY2Vr6mtFZVdhVWMzMzsxHxhbWZWXlcgCyAT3RmZmZmZtYJXIA0\ns5ZU7cZLLa12Q6aom0Yj2UftyjfgzMzMWosLkGZmZtY0q9Y9y+xBNwJGchOgG26UmJm1k8J6YTUz\nMzMzM7P25gKkmZmZmZmZ1cVVWM3MzKwjuQ2tmVnj+QmkmZmZmZmZ1cVPIM2saeq9+9+MTjIGPnPu\n1K0vd+DRjHWPZfkp827aJj4zMzOzVucCZEnGUq2m1gXrlcfvPqaYzMzMylLmzRwzM6ufq7CamZmZ\nmZlZXfwEsoW4sb+ZmZmZjUS160fXSrNmcgHSzMzMClVmdVPfrB2e95GZDcUFSLM6+YTauuq9GB1r\nhzdj4fZZZmbWAxU8AAAAIABJREFUamqdm3x9Y0NxG0gzMzMzMzOrS0s8gZR0PHAJsAPw1YiYX3JI\nZh3P+c6seM537aXRPaa34lOdbqgd0Un5rt7va9W6Z7cbIqoZx1+98bTisW+jV3oBUtIOwL8AbwPW\nAj+WdENE3F9uZO2n2o9FNc7EQ+uSk6nznVnBnO/Miud8Z9Z4pRcggSOBhyLiYQBJi4HpgDO2WfM4\n35kVz/nOrHjOd9YW2qXWAoAiotwApFOB4yPi/fn9GcAbI+KsQfPNAebktwcBP63xkfsCv2hSuK3O\n29569o+IV5UdxGANznetuu8HOL6xacf4uiHftZNWP4Yapdu3s1vyXbt8z46zsVo1zlLyXSs8gaxL\nRCwAFgw3n6Q7I2JaASG1HG97d257M9WT71p93zu+sXF8xav3fNcuOvE7qsbb2d467TrTcTZWu8RZ\nlFbohXUdsF/F+8k5zcyax/nOrHjOd2bFc74za7BWKED+GDhQ0gGSdgJmAjeUHJNZp3O+Myue851Z\n8ZzvzBqs9CqsEbFV0lnAv5G6V/5aRNw3ho/smGo/o+Btt7o0ON+1+r53fGPj+BqkCee7dtE239EY\neTtbUBdfZzrOxmqXOAtReic6ZmZmZmZm1h5aoQqrmZmZmZmZtQEXIM3MzMzMzKwuHVOAlHS8pJ9K\nekjSvLLjKYqk/STdKul+SfdJOqfsmIomaQdJd0u6sexYukmr5blaeUHSBZLWSVqZ/04oMcY1klbl\nOO7MaXtLWibpwfx/r5JiO6hiH62U9Jykc8vef5K+JmmTpHsr0qruMyX/nI/JeyQdUWSsNmQ+bInj\nvNEGn39yRy3L8zF4Te60pa1JmiBpiaQHJK2W9KZO/D6HO6dJ2jl/pw/l73hKxbSP5/SfSnpHC8T6\nkZwH75F0i6T9K6a9WPF73tTOhOqIc7akJyvieX/FtFn5+HpQ0qyS47y4IsafSdpcMa2w/dlSIqLt\n/0iNon8O/B6wE/AT4OCy4ypo2ycCR+TX44Gfdcu2V+yDjwDfBG4sO5Zu+WvFPFcrLwAXAH9b9j7L\nca0B9h2U9g/AvPx6HvC5FohzB2ADsH/Z+w94C3AEcO9w+ww4AfgeIOAoYHnZ+7Lb/obIhy13nDdo\ne7c5/wDXAjPz68uBD5QdYwO2cSHw/vx6J2BCp32f9ZzTgL8BLs+vZwLX5NcH5/l3Bg7In7NDybG+\nFdgtv/7AQKz5fX8L7dPZwBerLLs38HD+v1d+vVdZcQ6a/2xSR0yF7s9W++uUJ5BHAg9FxMMR8QKw\nGJheckyFiIj1EXFXfr0FWA1MKjeq4kiaDJwIfLXsWLpMy+W5Ns4L00kXaOT/p5QYy4BjgZ9HxKNl\nBxIRtwFPD0qutc+mA1dFcgcwQdLEYiI1GDIftuJxPiaDzz+SBBwDLMmztP12StqTdBPnCoCIeCEi\nNtN532c957TKbV4CHJu/8+nA4oj4dUQ8AjyUP6+0WCPi1oj4ZX57B2nsy6KN5TrhHcCyiHg6Ip4B\nlgHHt0icpwFXNymWttEpBchJwOMV79fSHheODZWrUxwOLC83kkJ9Afgo8FLZgXSZls5zVfLCWbkq\nz9dKrmoVwPclrZA0J6f1RMT6/HoD0FNOaNuYybYnyFbZfwNq7bOWPi67zaB82IrH+VgNPv/sA2yO\niK35fSccfwcATwJfz1V1vyppdzrv+6znt+PlefJ3/CzpOy/6d2ek6zuTVDNjwC6S7pR0h6RmFvzr\njfN/5PPLEkn7jXDZRqh7Xbkq8AHADyqSi9qfLaVTCpBdT9IewLeBcyPiubLjKYKkk4BNEbGi7Fis\ndVTJC5cBvw8cBqwHLioxvDdHxBHAnwEflPSWyomR6sOUOrZSbrN1MvCtnNRK+287rbDPbHtDnZM6\n4TvrovPPOFIV8ssi4nDgeVKV1Zd1wvfZqSS9G5gG/GNF8v4RMQ14F/AFSb9fSnDJd4ApEfFHpKeM\nC4eZv2wzgSUR8WJFWivtz8J0SgFyHbBfxfvJOa0rSNqRdKJeFBHXlR1PgY4GTpa0hlTl4BhJ/1pu\nSF2jJfNctbwQERsj4sWIeAn4Cs2tWjSkiFiX/28Crs+xbByoZpn/byorvuzPgLsiYiO01v6rUGuf\nteRx2W1qnJNa7Tgfq+3OP8AlpGrT4/I8nXD8rQXWRsRAbY4lpAJlp32f9fx2vDxP/o73BJ6qc9lG\nqmt9ko4D/h44OSJ+PZBecR56GOgj1RIoJc6IeKoitq8Cf1zvskXGWWFw7Zwi92dL6ZQC5I+BA3Pv\nZzuRvuCu6Akp17+/AlgdEZ8vO54iRcTHI2JyREwhfec/iIh3lxxWt2i5PFcrLwxqA/fnwL2Dly2C\npN0ljR94Dbw9x3IDMNDD3CxgaRnxVdimfUer7L9Bau2zG4D3KDkKeLaimp0VYIhzUqsd52NS4/xz\nOnArcGqerRO2cwPwuKSDctKxwP102PdJfee0ym0+lfSdR06fqdRL6wHAgcCPyoxV0uHAl0mFx00V\n6XtJ2jm/3pd0I+T+EuOsPL+cTGozDfBvwNtzvHuRzpf/VlacOdbXkTr0+WFFWpH7s6WMG36W1hcR\nWyWdRTq4diD1jnRfyWEV5WjgDGCVpJU57byI+G6JMVmHa9E8VzUvAKdJOoxUxWoN8FflhEcPcH26\nvmYc8M2IuFnSj4FrJZ0JPArMKCm+gYLt29h2H/1DmftP0tVAL7CvpLXA+cB8qu+z75J6Yn0I+CXw\n3iJjNaB2Pqz1nXWajwGLJX0GuJvc+UybOxtYlC+uHyblq1fQQd9nrXOapE8Bd0bEDaTv8huSHiJ1\n7DUzL3ufpGtJBYetwAcHVXEsI9Z/BPYAvpXPOY9FxMnAHwJflvQS6TucHxFNKfDUGeeHJJ1M2m9P\nk3plJSKelvRpUuEO4FMRMbgztSLjhPR9L843DQYUtj9bjbbdD2ZmZmZmZmbVdUoVVjMzMzMzM2sy\nFyDNzMzMzMysLi5AmpmZmZmZWV1cgDQzMzMzM7O6uABpVoWkr0naJGnYIQsk/a6kWyXdLekeSScU\nEaOZmZmZtZ8RXmdeLGll/vuZpM1FxDhkTO6F1Wx7kt4C9ANXRcQhw8y7ALg7Ii6TdDDw3Tw2mJmZ\nmZnZNkZynTloubOBwyPifU0Lrg5+AmlWRUTcRhqT6GWSfl/SzZJWSPrPPKgspPHxXplf7wk8UWCo\nZmZmZtZGRnidWek04OpCghzCuLIDMGsjC4C/jogHJb0R+BJwDHAB8P18V2h34LjyQjQzMzOzNlTr\nOhMASfsDBwA/KCm+l7kAaVYHSXsA/w34lqSB5J3z/9OAKyPiIklvAr4h6ZCIeKmEUM3MzMysjQxz\nnTlgJrAkIl4sMrZqXIA0q88rgM0RcViVaWcCxwNExA8l7QLsC2wqMD4zMzMza09DXWcOmAl8sKB4\nhuQ2kGZ1iIjngEck/QWAkkPz5MeAY3P6HwK7AE+WEqiZmZmZtZVhrjPJ7SH3An5YUojbcAHSrApJ\nV5My6UGS1ko6EzgdOFPST4D7gOl59rnAX+b0q4HZ4e6NzczMzKyKEV5nQnr6uLhVri89jIeZmZmZ\nmZnVxU8gzczMzMzMrC4uQJqZmZmZmVldXIA0MzMzMzOzurgA2aUk3Sept0p6r6S1JYRk1rIkrZF0\n3CiW+xNJP21gHM6f1pFGm8dGuI5+Sb/XwM8LSa9t1OeZNVsR+awdSbpS0mdGOq2buQDZpSLi9RHR\nV3YcZp1k8AVlRPxnRBxUMd0nb7OSRMQeEfEw+KLQWo+kPknvLzsOG5luvZHkAqSZmZmZWRuTNK7s\nGKx7uADZpQaehEjaNd+JfUbS/cAbyo7NrFVJOlLSDyVtlrRe0hcl7ZSn3ZZn+0muKvc/K6ucSvoG\n8LvAd/L0j1arklr5lHK4/CnpNZK+LelJSY9I+lCz94FZM0naWdIXJD2R/74gaec8rTePlzZX0qac\nB99bsew+kr4j6TlJP5b0GUm3V0wPSa+VNIc03tpHc178TuX0ivm3eUop6e/yOp+Q9L4qcf+TpMck\nbZR0uaRdm7enrFXl3/C/lXSPpGclXSNpF0l7Sbox/14/k19PzstcCPwJ8MV8TH5R0pR8TI6r+OyX\nn1JKmi3pvyRdLOkp4AJJvy/pB5KekvQLSYskTRhh/LtKWphjXJ3PVWsrps+T9HNJWyTdL+nPK6a9\nVtJ/5O3+haRrhlnXcNtY8/MkvU7SMklPS/qppBkj2c4hYhov6VZJ/yxJOXnfvK4tOZ7987zVzvuz\nK3938nwd95TSBUg7H/j9/PcOYFa54Zi1tBeBDwP7Am8CjgX+BiAi3pLnOTRXldvmxBkRZwCPAe/M\n0/+hjvXVzJ+SXgF8B/gJMCnHcq6kd4x+88xK9/fAUcBhwKHAkcAnKqb/DrAn6Zg/E/gXSXvlaf8C\nPJ/nmUWN81lELAAWAf+Q8+I7hwtK0vHA3wJvAw4EBldFnw/8QY77tTm+/z3c51rHmgEcDxwA/BEw\nm3TN/XVgf9LNxF8BXwSIiL8H/hM4Kx+TZ9W5njcCDwM9wIWAgP8DvAb4Q2A/4IIRxn4+MAX4PdLx\n/u5B039OKuzuCXwS+FdJE/O0TwPfB/YCJgOXjnDdg1X9PEm7A8uAbwKvBmYCX5J08FhWJmkf4Bbg\nvyLiQxERedLpOZZ9gZWk349hz/udzAVImwFcGBFPR8TjwD+XHZBZq4qIFRFxR0RsjYg1wJeBP23i\nKofKn28AXhURn4qIF3Lbrq+QTqRm7ep04FMRsSkiniRdoJ5RMf03efpvIuK7QD9wkKQdgP8BnB8R\nv4yI+4GFDYxrBvD1iLg3Ip6n4qI8P6WYA3w459UtwGdxXuxm/xwRT0TE06QbfYdFxFMR8e18fG4h\nFfjGev54IiIuzeekX0XEQxGxLCJ+nfPP50exjhnAZyPimYhYy6Drwoj4Vt62l3KB6UHSjR5I+XN/\n4DUR8f8iYpsncaNQ6/NOAtZExNfztt8NfBv4izGs6zXAfwDfiohPDJp2U0TcFhG/Jt3kepOk/caw\nrrbnAqS9Bni84v2jZQVi1uok/UGudrRB0nOki8R9m7jKofLn/sBrlKrTbpa0GTiPdCfarF29hm2P\n80dz2oCnImJrxftfAnsArwLGsW1+qXzdiLhq5cVXAbsBKyry4s053brThorXvwT2kLSbpC9LejSf\nP24DJuSbH6O1zTEuqUfSYknr8jr+lZGfowYf64PX8R5JKyuO9UMq1vFR0lPQHyn19r9NVe9RqPV5\n+wNvHHT+O51U+2C0TgR2BS6vMu3lfRAR/cDTbPu71HVcgLT1pCoOA363rEDM2sBlwAPAgRHxSlKB\nTUMvso0Y9P550oUnAPlCovKic6j8+TjwSERMqPgbHxEnjCAes1bzBOnicMDv5rThPAlsJVVzGzDU\nE4LBeRHShf5uFe8rL0aHyou/IFVHfH1FXtwzIvaoI27rHnOBg4A35vPHQPXHgXNItfMD1D4mqy3z\n2Zw2Na/j3YzsHAXpWK+aj3Lbv68AZwH7RMQE4N6BdUTEhoj4y4h4DfBXpGqlQ7X9G3Ibh/i8x4H/\nGHT+2yMiPjDCba30FdKNn+/mKrKVKvfBHsDe1P5dGnxeH0uhtmW5AGnXAh/PjbsnA2eXHZBZCxsP\nPAf0S3odMPhktZHUbqSWwdN/Buwi6URJO5Laeu1cMX2o/PkjYIukj+VOD3aQdIgkd4Rl7exq4BOS\nXiVpX1I7wn8dbqGIeBG4jtSRyG45f75niEWq5dWVwLtyXjqebav+XQvMlnSwpN1I7cQG1v0S6eLz\nYkmvBpA0ye2RbZDxpBsNmyXtTcUxlG1zTOYqqOuAd+dj8n2k9vDDraMfeFbSJODvRhFn5XlnEqmw\nOGB3UgH1SQClTqwOGZgo6S/yuQrgmTzvS7VWNNw2DvF5NwJ/IOkMSTvmvzdI+sNRbG+ls4Cfkjq7\nq+wE6wRJb1bqNO/TwB25WQls/1vyE+D1kg6TtAsjb4PaFlyAtE+SquI8Qmqo/I1ywzFraX8LvAvY\nQrpgHNxg/gJgYa5SU61HuP9DujjeLOlvI+JZUic8XyWdRJ8HKntlrZk/8wXzSaROOx4hPQX5Kqlj\nA7N29RngTuAeYBVwV06rx1mk438DKa9cDfy6xrxXAAfnvPj/5bRzgHcCA9XhBtKJiO8BXwB+ADyU\n/1f6WE6/I1cd/HfS0yazAV8gVZH8BXAH6WlXpUuAU5V6Px1od/iXpELgU8Drgf87zDo+CRwBPAvc\nRLqpMlKfIp2HHiEdx0vI+Si3Lb4I+CGp4DQV+K+KZd8ALJfUD9wAnDMw9uoQhtrGqp+X25C+ndTO\n+AlSnv8c296AHbHcac4c0vYvzQVASJ31nE+quvrHbNux0AVUnPcj4mekffjvpPahY20H2pL02w6G\nzMzMzDqDpM8BvxMR7l3cbJQkfQCYGRHN7DDO2oyfQJqZmVnbUxoX7o+UHEka5uP6suMyayeSJko6\nWtIrJB1EarvpfGTbcAHSzMzMOsF4UpW950nVyy8ClpYakVkLkvQ9pYHvB/+dB+xEGqJqC6mq9lLg\nS2NY15/UWFd/Y7Zmu/XdV2N9j9RIP70ZcYwwtsJiaBRXYTUzMzMzM7O6+AmkmZmZmZmZ1WVcoz9Q\n0n7AVaTBrANYEBGX5C6LrwGmAGuAGRHxjCSRep46gTQG0+yIuGuodey7774xZcqUqtOef/55dt99\n8PAt7auTtqcbt2XFihW/iIiOGEx6qHzXaO18rDj2clTG7nzXWO18XNTD29cY3ZTvOvmY6eRtg87b\nvtLyXUQ09A+YCByRX48njXN2MPAPwLycPg/4XH59AvA90iCkRwHLh1vHH//xH0ctt956a81p7aiT\ntqcbtwW4Mxqcx8r6GyrfNVo7HyuOvRyVsTvfNVY7Hxf18PY1Rjflu04+Zjp52yI6b/vKyncNr8Ia\nEesjP0GMNE7LamASMB1YmGdbCJySX08Hrsr74Q5ggqSJjY7LzMzMzMzMxqbhVVgrSZoCHA4sB3oi\nYn2etIFUxRVS4fLxisXW5rT1FWlImkMa3JOenh76+vqqrrO/v7/mtHbUSdvjbTEzMzMza29NK0BK\n2gP4NnBuRDyXmjomERGSRtT9a0QsABYATJs2LXp7e6vO19fXR61p7aiTtsfbYmZmZmbW3prSC6uk\nHUmFx0URcV1O3jhQNTX/35TT1wH7VSw+OaeZWZ0kfU3SJkn3VqT9o6QHJN0j6XpJEyqmfVzSQ5J+\nKukd5URtZmZmZu2m4QXI3KvqFcDqiPh8xaQbgFn59Sx+O7jvDcB7lBwFPFtR1dXM6nMlcPygtGXA\nIRHxR6TOrD4OIOlgYCbw+rzMlyTtUFyoZmZmZtaumlGF9WjgDGCVpJU57TxgPnCtpDOBR4EZedp3\nST2xPkQaxuO9Y1n5qnXPMnveTdulr5l/4lg+1qylRcRtuc1xZdr3K97eAZyaX08HFkfEr4FHJD0E\nHAn8sIBQrQRT/Jto1nacb1tXtWtNfzfWTRpegIyI20lDclRzbJX5A/hgo+Mws228jzQOK6ROqu6o\nmDbQcdV26u28qtHauZOiVox97tSt26VVi7EVY69XO8duZmbWTpraC6uZlU/S3wNbgUUjXbbezqsa\nrZ07KWrF2KvWyji9d7u0Voy9Xu0cu5mZWTtxAdKsg0maDZwEHJuf9oM7rjIzMzOzUWpKL6xmVj5J\nxwMfBU6OiF9WTLoBmClpZ0kHAAcCPyojRjMzMzNrL34CadYBJF0N9AL7SloLnE/qdXVnYFkeh/WO\niPjriLhP0rXA/aSqrR+MiBfLidzMzMzM2okLkGYdICJOq5J8xRDzXwhc2LyIzLqbpA8D7wcCWEXq\nYXwisBjYB1gBnBERL5QWpJmZ2Si4CquZmVkDSZoEfAiYFhGHADuQxl79HHBxRLwWeAY4s7wozczM\nRscFSDMzs8YbB+wqaRywG7AeOAZYkqcvBE4pKTYzM7NRcxVWMzOzBoqIdZL+CXgM+BXwfVKV1c0R\nMTAoZ8uNv1pLp4+x2arbV+/4rcNp1e0zs/blAqSZmVkDSdoLmA4cAGwGvgUcX+/yZY2/Wkunj7HZ\nqttX7/itw2nV7TOz9uUqrGZmZo11HPBIRDwZEb8BrgOOBibkKq3g8VfNzKxNuQBpZmbWWI8BR0na\nTWkMnWNJw+bcCpya55kFLC0pPjMzs1FzAdLMzKyBImI5qbOcu0hDeLyCVCX1Y8BHJD1EGsqj5lA7\nZmZmrcptIM3MzBosIs4Hzh+U/DBwZAnhmJmZNYyfQJqZmZmZmVldXIA0MzMzMzOzurgAaWZmZmZm\nZnVxG0gzMzOzkkypMt6jmVkr8xNIMzMzM2t7kiZIWiLpAUmrJb1J0t6Slkl6MP/fq+w4zdqdC5Bm\nZmZm1gkuAW6OiNcBhwKrgXnALRFxIHBLfm9mY+ACpJmZmZm1NUl7Am8hj68aES9ExGZgOrAwz7YQ\nOKWcCM06h9tAmpmZmVm7OwB4Evi6pEOBFcA5QE9ErM/zbAB6qi0saQ4wB6Cnp4e+vr6aK+rZFeZO\n3bpN2lDzt5P+/v6O2ZZqOn37iuICpFkHkPQ14CRgU0QcktP2Bq4BpgBrgBkR8Ywkkar5nAD8Epgd\nEXeVEbeZmVmDjAOOAM6OiOWSLmFQddWICElRbeGIWAAsAJg2bVr09vbWXNGli5Zy0aptL6HXnF57\n/nbS19fHUNve7jp9+4riKqxmneFK4PhBabXaffwZcGD+mwNcVlCMZmZmzbIWWBsRy/P7JaQC5UZJ\nEwHy/00lxWfWMVyANOsAEXEb8PSg5FrtPqYDV0VyBzBh4ORqZmbWjiJiA/C4pINy0rHA/cANwKyc\nNgtYWkJ4Zh3FVVjNOletdh+TgMcr5lub09YzyEjahDRSO7dRaMXYB7fVgertdVox9nq1c+xm1jBn\nA4sk7QQ8DLyX9LDkWklnAo8CM0qMz6wjuABp1gWGavcxzHJ1twlppHZuo9CKsc+uMlB5tfY6rRh7\nvdo5djNrjIhYCUyrMunYomMx62SuwmrWuWq1+1gH7Fcx3+ScZmZmZmY2JBcgzTpXrXYfNwDvUXIU\n8GxFVVczMzMzs5pchdWsA0i6GugF9pW0FjgfmE/1dh/fJQ3h8RBpGI/3Fh6wmZmZmbUlFyDNOkBE\nnFZj0nbtPiIigA82NyIzMzMz60QNr8Iq6WuSNkm6tyLtAknrJK3MfydUTPu4pIck/VTSOxodj5mZ\nmZmZmTVGM9pAXsn2A5oDXBwRh+W/7wJIOhiYCbw+L/MlSTs0ISYzMzMzMzMbo4YXIGsMaF7LdGBx\nRPw6Ih4htck6stExmZmZmZmZ2dgV2QbyLEnvAe4E5kbEM6TBy++omGdgQPPt1Dugec+u9Q+a3Q46\naXBsb4uZmZmZWXsrqgB5GfBpIPL/i4D3jeQD6h3Q/NJFS7lo1fabVW3Q7HbQSYNje1vMzMzMzNpb\nIeNARsTGiHgxIl4CvsJvq6l6QHMzMzMzM7M2UUgBUtLEird/Dgz00HoDMFPSzpIOAA4EflRETGZm\nZmZmZjYyDa/CWmNA815Jh5GqsK4B/gogIu6TdC1wP7AV+GBEvNjomMzMzMzMmmXKvJu2S1sz/8QS\nIjFrvoYXIGsMaH7FEPNfCFzY6DjMzMzMzMyssQqpwmpmZtZNJE2QtETSA5JWS3qTpL0lLZP0YP6/\nV9lxmpmZjZQLkGZmZo13CXBzRLwOOBRYDcwDbomIA4Fb8nszM7O24gKkmZlZA0naE3gLuflGRLwQ\nEZuB6cDCPNtC4JRyIjQzMxu9osaBNDMz6xYHAE8CX5d0KLACOAfoiYj1eZ4NQE+1hSXNAeYA9PT0\n0NfX1/SAh9Lf3196DM1U9vbNnbq17nlHE2fZ22dmnccFSDMzs8YaBxwBnB0RyyVdwqDqqhERkqLa\nwhGxAFgAMG3atOjt7W1yuEPr6+uj7Biaqeztm12l985a1pzeO+LPL3v7zKzzuAqrmZlZY60F1kbE\n8vx+CalAuXFgXOT8f1NJ8ZmZmY2aC5BmZmYNFBEbgMclHZSTjiWNd3wDMCunzQKWlhCemZnZmLgK\nq1mHk/Rh4P1AAKuA9wITgcXAPqT2WWdExAulBWnWec4GFknaCXiYlO9eAVwr6UzgUWBGifGZmZmN\niguQZh1M0iTgQ8DBEfErSdcCM4ETgIsjYrGky4EzgctKDNWso0TESmBalUnHFh2LmZlZI7kKq1nn\nGwfsKmkcsBuwHjiG1C4LPJyAmZmZmdXJTyDNOlhErJP0T8BjwK+A75OqrG6OiIG+49cCk6otX9Zw\nAu3c7Xwrxl5tmIBqMbZi7PVq59jNzMzaiQuQZh1M0l6kwcsPADYD3wKOr3f5soYTaOdu51sx9mrD\nBFQbDqAVY69XO8duZmbWTlyF1ayzHQc8EhFPRsRvgOuAo4H/n717j5esqu+8//kqKIoXQEwP0sQm\nkTjBEFE7iKPJHO8IRnQeQnAYhEhCLpJoxFEw80RjdKbNBBU10aAYMEGBeAlE8ILIidEoURC5amix\nDbQNHRWVNo+X1t/zx94Hqg9Vp+ucU6du5/N+vepVtde+1G/tqn1Ze6+91h5tlVaAtcDmUQUoSZKk\nyWEBUppu/wYcmuT+ScLd3QlcDhzVTmN3ApIkSeqLBUhpirUdmb8PuIqmC4970VRJfQXw0iQbabry\nOGtkQUqSJGli+AykNOWq6lXAq+Yl3wwcMoJwJEmSNMG8AylJkiRJ6osFSEmSJElSXyxASpIkSZL6\nYgFSkiRJktQXC5CSJEmaCknuneQLST7UDu+f5IokG5Ocn+Q+o45RmnQWICVJkjQtXgzc2DH8euCN\nVfUI4A7gxJFEJU0RC5CSJEmaeEnWAkcA72yHAzyFpj9kgHOA544mOml6WICUJEnSNHgT8HLgJ+3w\nQ4BvV9X2dvhWYN9RBCZNk11GHYAkSZK0HEmeDWytqiuTzCxh/pOAkwDWrFnD7Oxsz2nX3A9OOWh7\nz/FzFloBoXKqAAAgAElEQVTGuNq2bdtExt2vac/fsFiAlCRJ0qR7IvCcJIcDuwEPAs4A9kiyS3sX\nci2wudvMVXUmcCbA+vXra2ZmpucXveXcCzn92p2fQm86tvcyxtXs7CwL5X3STXv+hsUqrJIkSZpo\nVXVaVa2tqnXAMcAnqupY4HLgqHay44ELRxSiNDUsQEqSJGlavQJ4aZKNNM9EnjXieKSJZxVWSZIk\nTY2qmgVm2883A4eMMh5p2gz8DmSSdyXZmuS6jrS9klya5Kb2fc82PUne3Hbuek2Sxw46HkmSJEnS\nYKxEFdazgcPmpZ0KXFZVBwCXtcMAzwIOaF8nAW9bgXgkSZIkSQMw8AJkVX0S+Na85CNpOm+FHTtx\nPRJ4dzU+S9NS1j6DjkmSJEmStHzDegZyTVVtaT/fBqxpP+8L3NIx3VwHr1uYp9/+eXr1zTOpfb5M\nU3815mU0kuwBvBP4BaCAFwJfBs4H1gGbgKOr6o4RhShJkqQJMfRGdKqqktQS5uurf55effNMYl88\nMF391ZiXkTkD+EhVHZXkPsD9gVfSVCvfkORUmmrlrxhlkJIkSRp/w+rG4/a5qqnt+9Y2fTOwX8d0\nPTt4lbR4SR4M/Apts+VV9cOq+ja9q5VLkiRJPQ3rDuRFNJ23bmDHTlwvAk5Och7weOA7HVVdJS3f\n/sC/A3+d5NHAlcCL6V2tfAf9Vh0ftEmqIjzfOMbeb7X+cYy9X5McuyRJk2TgBcgk7wVmgL2T3Aq8\niqbgeEGSE4GvAUe3k18CHA5sBP4D+I1BxyOtcrsAjwV+v6quSHIGd7eCDCxcrbzfquODNmFVhHcw\njrGfcOrF90jrVq1/HGPv1zjGnuTewOeBzVX17CT7A+fRdGZ+JXBcVf1wlDFKkrRYAy9AVtXze4x6\napdpC3jRoGOQdJdbgVur6op2+H00Bcjbk+xTVVvmVSuXNDgvBm4EHtQOvx54Y1Wdl+TtwInYfZUk\nacIM6xlISSNQVbcBtyR5ZJv0VOAG7q5WDjtWK5c0AEnWAkfQtIBMkgBPobmIAz57LEmaUENvhVXS\n0P0+cG7bAuvNNFXF70X3auWSBuNNwMuBB7bDDwG+XVVzD6TOdVt1D6N69riXaX++dNT56/aMci9L\niXPU+ZM0fSxASlOuqq4G1ncZdY9q5ZKWL8mzga1VdWWSmcXOP6pnj3sZx+dLB2nU+ev2jHIvS+mS\nbNT5kzR9LEBKkjRYTwSek+RwYDeaZyDPAPZIskt7F9JuqyRJE8lnICVJGqCqOq2q1lbVOuAY4BNV\ndSxwOXBUO5nPHkuSJpIFSEmShuMVwEuTbKR5JvKsEccjSdKiWYVVkqQVUlWzwGz7+WbgkFHGI0nS\ncnkHUpIkSZLUFwuQkiRJkqS+WICUJEmSJPXFAqQkSZIkqS8WICVJkiRJfbEAKUmSJEnqiwVISZIk\nSVJfLEBKkiRJkvpiAVKSJEmS1BcLkJIkSZKkvliAlCRJkiT1xQKkJEmSJKkvFiClVSDJvZN8IcmH\n2uH9k1yRZGOS85PcZ9QxSpIkafxZgJRWhxcDN3YMvx54Y1U9ArgDOHEkUUmSJGmiWICUplyStcAR\nwDvb4QBPAd7XTnIO8NzRRCdJkqRJssuoA5C04t4EvBx4YDv8EODbVbW9Hb4V2LfbjElOAk4CWLNm\nDbOzsysbaWvbtm1D+65BG8fYTzlo+z3SusU4jrH3a5JjlyRpkliAlKZYkmcDW6vqyiQzi52/qs4E\nzgRYv359zcwsehFLMjs7y7C+a9BGHfu6Uy/uknrPXf2mY2fukTbq2JdjkmOXJGmSWICUptsTgeck\nORzYDXgQcAawR5Jd2ruQa4HNI4xRkiRJE8JnIKUpVlWnVdXaqloHHAN8oqqOBS4HjmonOx64cEQh\nSpK0bEn2S3J5khuSXJ/kxW36XkkuTXJT+77nqGOVJp0FSGl1egXw0iQbaZ6JPGvE8UiStBzbgVOq\n6kDgUOBFSQ4ETgUuq6oDgMvaYUnLYBVWaZWoqllgtv18M3DIKOORJGlQqmoLsKX9fGeSG2kaiDsS\nmGknO4fmOPiKEYQoTQ3vQEqSJGlqJFkHPAa4AljTFi4BbgPWjCgsaWp4B1KSJElTIckDgPcDL6mq\n7zZdHzeqqpJUj/n67rZqzf26d4803yR2LTTtXSJNe/6GZagFyCSbgDuBHwPbq2p9kr2A84F1wCbg\n6Kq6Y5hxSZIkabIl2ZWm8HhuVX2gTb49yT5VtSXJPsDWbvMuptuqt5x7Iadfu/NT6G7dJY27ae8S\nadrzNyyjqML65Ko6uKrWt8M+3CxJkqQlS3Or8Szgxqp6Q8eoi2haGwdbHZcGYhyegTyS5qFm2vfn\njjAWSZIkTZ4nAscBT0lydfs6HNgAPD3JTcDT2mFJyzDsZyAL+Fhb//yv2uoCfT3c3G/d9F710ie1\nvvM01dU2L5JWgyT7Ae+mOZ4VcGZVneEjG9LKqapPAekx+qnDjEWadsMuQD6pqjYn+Sng0iRf6hy5\n0MPN/dZN71UvfRLrocN01dU2L5JWibn+6K5K8kDgyiSXAifQPLKxIcmpNI9s2J2AJGmiDLUKa1Vt\nbt+3Ah+k6Yfu9vahZhZ6uFmSpElQVVuq6qr2851AZ390PrIhSZpoQ7sDmWR34F5t5667A88AXsPd\nDzdvwIebJUlTZCn90S2mO4FhmPYq+6POXz/dQcxZSpyjzp+k6TPMKqxrgA+2/fHsArynqj6S5HPA\nBUlOBL4GHD3EmCRJWhFL7Y9uMd0JDMO0V9kfdf5OOPXivqddyuM4o87farauy2+7acMRI4hEGqyh\nFSCr6mbg0V3Sv4kPN0uSpshy+qOTJGmcDbsRHUnSGOh2Zfzsw3YfQSTTp4/+6HxkQ5I0sSxASpI0\nWHP90V2b5Oo27ZU0BUcf2ZAkTTQLkJIkDZD90Wml+EydpHEw1G48JEmSJEmTyzuQkiRJQ9DtDqIk\nTRrvQEpTLMl+SS5PckOS65O8uE3fK8mlSW5q3/ccdaySJEkafxYgpem2HTilqg4EDgVelORA4FTg\nsqo6ALisHZYkSZIWZAFSmmJVtaWqrmo/3wncCOwLHAmc0052DvDc0UQoSZKkSeIzkNIqkWQd8Bjg\nCmBNVW1pR90GrOkxz0nASQBr1qxhdnZ2xeME2LZt29C+a9BGHfspB21f8ryjjn05Jjl2SZImiQVI\naRVI8gDg/cBLquq7TT/njaqqJNVtvqo6EzgTYP369TUzMzOEaGF2dpZhfdegjTr2E5bRSMfZh+3u\nepckSQuyCqs05ZLsSlN4PLeqPtAm355kn3b8PsDWUcUnSZKkyWEBUppiaW41ngXcWFVv6Bh1EXB8\n+/l44MJhxyZJkqTJYxVWabo9ETgOuDbJ1W3aK4ENwAVJTgS+Bhw9ovgkSZI0QSxASlOsqj4FpMfo\npw4zFkmSJE0+C5BSn9Z1NE5yykHbOeHUi9m04YgRRiRJkiQNlwVISZIkacys69KqtheuNQ5sREeS\nJEmS1BfvQEqSJElD4F1FTQMLkJKkseXJliRJ48UC5ATqdkIFnlRJkiRJWlkWICVpGeZf0DnloO3M\njCYUrULeoV1Zk7B+JyFGLazXjQFpXNmIjiRJkiSpL96BlCRJWgbvAmpY/K9pHHgHUpIkSZLUF+9A\nSpIkDdgon2vr/O5TDtrOCade7F0qSQNjAVKSJE28YVXtm9QGT2zBXdKgWICUpD5M6knjqPW73jyJ\nlST14rOf48UCpCRJmkr9nnR6gUiS+mcBUpIGbFKvlF67+TucMKGxS5IWr1tfxj4zq50ZiwJkksOA\nM4B7A++sqg2D/g6rUUk7GsZ2p8Ubxzshw7iLM475XgnTtN2N8plDj9WDsxrW7zRtd9I4GHkBMsm9\ngb8Ang7cCnwuyUVVdcNoI5teq+FgoYWN23bX6wrofMv9n/rf1yiN23YnrQardbvzAs94WEzjVZO0\nLkdegAQOATZW1c0ASc4DjgQmcsMetzudq+WqvhZt4NvdMHZ8K/F/tuVGDdFYHe/6/U/2uqCznGVq\nvE3SiWwfxmq7k6ZBqmq0ASRHAYdV1W+2w8cBj6+qk+dNdxJwUjv4SODLPRa5N/CNFQp3FKYpP6sx\nLw+vqoeudDCLtQLb3aBN8n/F2EejM3a3u8Ga5P9FP8zfYKym7W6a/zPTnDeYvvyNZLsbhzuQfamq\nM4EzdzZdks9X1fohhDQU05Qf8zJ5+t3uBm2S16+xj8Ykxz7fqLa7XqZp3XZj/gSL2+6meZ1Oc95g\n+vM3LPcadQDAZmC/juG1bZqkleN2Jw2f2500fG530oCNQwHyc8ABSfZPch/gGOCiEcckTTu3O2n4\n3O6k4XO7kwZs5FVYq2p7kpOBj9I0r/yuqrp+GYscm2o/AzJN+TEvY2IFtrtBm+T1a+yjMfaxT8B2\n18vYr9tlMn9TbIW2u2lep9OcN5j+/A3FyBvRkSRJkiRNhnGowipJkiRJmgAWICVJkiRJfRmrAmSS\ndyXZmuS6jrS9klya5Kb2fc82PUnenGRjkmuSPLZjnuPb6W9KcnxH+uOSXNvO8+YkWeg7ViAvr06y\nOcnV7evwjnGntXF9OckzO9IPa9M2Jjm1I33/JFe06ee3D4aT5L7t8MZ2/LoB5GW/JJcnuSHJ9Ule\n3KZP3G+zQF4m8reZVEnuneQLST7UDp/crpdKsnePeZ7c8ftcneT7SZ473MiXFns73Z+1/7kbO//j\nw7KMuF+f5Lr29evDi3iHGObHfm677V2XZl+7a4/5uu5vdE9JDk7y2Xbb+nySQ9r0/9mxzV2X5MdJ\n9uoy/1OTXNVO96kkjxh+LnobQP7+qWO6ryf5++HnorsB5C1JXpfkX9v90x8MPxeToddxfxqkx/nR\nNJl/LNEyVNXYvIBfAR4LXNeR9mfAqe3nU4HXt58PBz4MBDgUuKJN3wu4uX3fs/28ZzvuX9pp0877\nrIW+YwXy8mrgZV2mPRD4InBfYH/gKzQPet+7/fwzwH3aaQ5s57kAOKb9/Hbgd9vPvwe8vf18DHD+\nAPKyD/DY9vMDgX9tY56432aBvEzkbzOpL+ClwHuAD7XDjwHWAZuAvfuYfy/gW8D9JyF24L8An+74\n73wGmJmAuI8ALqVpcG13mtYMHzQG6/zwdl8R4L1z21iX/0jX/Y2vruv4Yx373cOB2S7T/CrwiR7z\n/yvw8+3n3wPOHnWeBpm/edO9H3jBqPM0wN/uN4B3A/dqh39q1HkaxxcLHPen4UWP86NRxzXgPO5w\nLPG19NdY3YGsqk/SnBR2OhI4p/18DvDcjvR3V+OzwB5J9gGeCVxaVd+qqjtoTn4Oa8c9qKo+W82/\n6N3zltXtOwadl16OBM6rqh9U1VeBjcAh7WtjVd1cVT8EzgOObO9cPAV4X5eYO/PyPuCpy73TUVVb\nquqq9vOdwI3Avkzgb7NAXnoZ699mEiVZS1MweedcWlV9oao2LWIxRwEfrqr/GHB4C1pG7AXsRnPS\ncV9gV+D2FQrzHpYR94HAJ6tqe1V9D7gGOGzFAu2iR+yXtPuXorn4tLbLrF33N8OIeUIV8KD284OB\nr3eZ5vk0Bfalzj9Ky80fAEkeRLOPH5s7kCw/b78LvKaqfgJQVVsHHuF06HrcH3FMA7OE86OJ0u1Y\noqUbqwJkD2uqakv7+TZgTft5X+CWjulubdMWSr+1S/pC37ESTk5TrfNdubs65mLz8hDg21W1fV76\nDstqx3+nnX4g0lS7fAxwBRP+28zLC0z4bzNB3gS8HPjJMpZxDDs50VshS4q9qj4DXA5saV8fraob\nBx9eT0td51+kuchz/7aa65PZsUPuYegZe1t19TjgI13m67WNqruXAP83yS3AnwOndY5Mcn+aAvj7\ne8z/m8AlSW6l+U02rGCsS7Hc/M15LnBZVX13RaJcmuXm7WeBX2+rv344yQErGu3kWjX7lC7nR9Ng\nEOceak1CAfIu7dXmFe13ZIW/4200O+qDaU4iT1+h71kRSR5AcwB6yfyD56T9Nl3yMtG/zaRI8mxg\na1VduYxl7AMcRNOn19AsJ/b2ebCfp7lTti/wlCS/POAQe333kuOuqo8BlwD/TFNg/wzw48FG2Fsf\nsf8lzR3SfxpWTJMsycc7nmftfB1JcxfqD6tqP+APgbPmzf6rwKerqlfNmj8EDq+qtcBfA29YqXz0\nssL5m7PTu5QrYYXzdl/g+1W1HngH8K6VyofG30LnepNqEOce2tEuow6gD7cn2aeqtrQnjnNVKzaz\n45XwtW3aZmBmXvpsm762y/QLfcdAVdVdVdaSvAOYe4i3V17okf5Nmmqhu7R3sjqnn1vWrUl2oanO\n8s3lxt5e6X8/cG5VfaBNnsjfplteJvm3mTBPBJ6TppGi3YAHJfnbqvofi1jG0cAHq+pHKxJhb8uJ\n/XnAZ6tqG0CSDwNPAIZR8FnWOq+q1wGvA0jyHprnYoalZ+xJXgU8FPjtHvP22t+sWlX1tF7jkrwb\nmGs04++4ZzWvnnf9kzwUeHRVzd2tOJ/ud4VX1Erlr2MZe9NUY3zeMsJckhXO263A3HH9gzQXAHRP\nC50PTIUe53rTYBDnHuowCXcgLwLmWs87HriwI/0FbethhwLfaas6fhR4RpI922qIz6CpLrYF+G6S\nQ9vnzl4wb1ndvmOg2gLQnOcBcy20XgQck6aVzv2BA2ie6/kccECaVj3vQ3MQuKi9E3c5zXNg82Pu\nzMtRNA/NL+uuXbu+zgJurKrOq8oT99v0ysuk/jaTpqpOq6q1VbWOZp19Ygk78JHcAVhm7P8G/Nck\nu7QH6P9K83zJiltO3GlarHtI+/kXgV+kabBjKHrFnuQ3aZ5xfP7cc1tddN3fDCXwyfR1mv8lNM/4\n3TQ3IsmD23G99r93AA9O8nPt8NMZ0v97EZaTvzlH0TS+8f0ViXDplpu3v6epnk477TAvEk2Srsf9\nEcc0MAuc6028AZ17qFONQUs+cy+ak8ItwI9oroidSPOM2GU0O8SPA3u10wb4C5oWsa4F1ncs54U0\njZ1sBH6jI309TcHgK8BbgbTpXb9jBfLyN22s19DsdPbpmP6P2ri+TNuaWpt+OM3O/CvAH3Wk/wxN\nQWYjzRXH+7bpu7XDG9vxPzOAvDyJpuroNcDV7evwSfxtFsjLRP42k/yiuTs016rmH7TbyXaak6F3\ndvwv3tkxzzqaK773mqTYaVrv+yuak+obgDdMSNy7tfHeAHwWOHhM1vn2drub24b/uMf/pev+xlfX\n9fsk4Eqa516vAB7XMe4EmsbE5s9zCfCw9vPz2n3oF2nu9I7V/m25+WuHZ4HDRp2XFfjt9gAubn+/\nz9DcTR55vsbxRY/j/jS86HF+NOq4ViCfdx1LfC39NXeSLkmSJEnSgiahCqskSZIkaQxYgJQkSZIk\n9cUCpCRJkiSpLxYgJUmSJEl9sQApSZIkSUOS5F1Jtia5ro9pfzrJ5Um+kOSatj/LkbIAKUmSJEnD\nczZwWJ/T/i/ggqp6DE0/ln+5UkH1ywKkJEmSJA1JVX0S+FZnWpKfTfKRJFcm+ack/3lucuBB7ecH\n0/TdPFK7jDoASZIkSVrlzgR+p6puSvJ4mjuNTwFeDXwsye8DuwNPG12IDQuQkiRJkjQiSR4A/Bfg\n75LMJd+3fX8+cHZVnZ7kCcDfJPmFqvrJCEIFLEBKkiRJ0ijdC/h2VR3cZdyJtM9LVtVnkuwG7A1s\nHWJ8O/AZSEmSJEkakar6LvDVJL8GkMaj29H/Bjy1Tf95YDfg30cSaCtVNcrvlyRJkqRVI8l7gRma\nO4m3A68CPgG8DdgH2BU4r6pek+RA4B3AA2ga1Hl5VX1sFHHPsQApSZIkSeqLVVglSZIkSX2xAClJ\nkiRJ6osFSEmSJElSXyxATqkk90vyD0m+k+TvFpjuhCSfWuJ3/HSSbUnu3Q7PJvnNHtOuS1JJdmmH\nP5zk+KV8r1aPJK9M8s5Rx9Gp/c//zBLnPTvJaxc7bpz1u6+RVtpyjisLHQuXc5wctSSvTfKNJLeN\nOhZNr/nng5p+9gM5vY4C1gAPqartK/EFVfVvNC1CLWXeZw04HE2hqvrfc5+TrAO+Cuy6Uv/pPmNa\n0n9+uZIUcEBVbRzF9y9gh31NkhOA36yqJ402LK02ozquJJkF/raqxu1i108DpwAPr6qt47IP1XRI\nsolmX//x5ZwPajJ5B3J6PRz4Vw8SklaY+5pVyrsNY++ngW9W1cg6G5c0nSxALkOSTUleluSatvrW\n+Ul261bdpa2++Yj289lJ/rKtbrMtyaeT/Kckb0pyR5IvJXlMH9//82210W8nuT7Jc9r0PwH+GPj1\ndvkn7nxReWubhy8leeq8PD6tY/jVSf62/bxDtdR5C7x3kj9vq87cDBwxb/xd1V3n1lc7/R1Jvprk\nWR3T7p/kk0nuTPLxJH/REcNuSf42yTfb9fC5JGt2tu40fpK8Isnm9nf+cpKndv7fgE+2799u/9dP\naOd7YZIb2//OR5M8vI/v+s9JLk3yrfa7ju4Yt0NV7Pnbc+e2vMz8PjDJ5UnenCRt8t5tXHcm+ce5\nvCSZy/sX27z/+s72Mwt87+FJbmi/Y3OSl3WM+59JtiT5erteF1xel33Ni4C3A09oh7+9+DWjQemx\nTd23PdZ8vX29Kcl92+n7OXa9LcklSb4HPDlNFebTk3ytPYZ8Ksn92ukPTfLP7b75i0lm+oj5wUnO\nav+Hm9NUwZx7TKJzf9Dt0Yiej1EsYd393zYvD7476Z7HySSvA34ZeGv7n3/r/Lj6jS3JI9rt/jtp\njp3nd4x7evu932m/4x8XWl6a4/alwMPauM6mxz5UWqwkf0NzgeIf2v/Sy7tsi69tt/9taR5zeEiS\nc5N8N8252rqO5fU8Jms8WYBcvqOBw4D9gV8ETljEfP+LpgPRHwCfAa5qh98HvGGhmZPsCvwD8DHg\np4DfB85N8siqehXwv4Hzq+oBVXXWTmJ5PPCV9rtfBXwgyV595qOX3wKeDTwGWE9TzW1nMXy5jeHP\ngLOSu06q3wP8C/AQ4NXAcR3zHQ88GNivHf87wP+3zNg1ZEkeCZwM/FJVPRB4JrBp3mS/0r7v0f6v\nP5PkSOCVwH8DHgr8E/DenXzX7jQnVu+h2XaOAf4yTUe9Q5HkIcBlwKer6g/q7g55jwX+lGY7uBo4\nF6Cq5vL+6Dbv589f5iKcBfx2u55/gabjYpIcBrwMeDpwAPC0nktoddnX/AXNNviZdniPZcSpZVhg\nm/oj4FDgYODRwCE0x6J+/XfgdcADgU8Bfw48DvgvwF7Ay4GfJNkXuBh4bZv+MuD9SR66k+WfDWwH\nHkFz/HgGMJBCYT+S3CvJO2iO58+oqu+0o7oeJ6vqj2j2Oye3//mTl/H1f0pzTN8TWAu8pY1pb+AD\n3H3O8BXgiQstqKo+DjwL+Hob1wl02YcuI1atYlV1HPBvwK+2j3Vc0GWyY2jO1/YFfpbmPPevafYH\nN9JsR2NxTNbiWYBcvjdX1der6ls0BbqD+5zvg1V1ZVV9H/gg8P2qendV/Rg4n+bAuZBDaeqbb6iq\nH1bVJ4APAc9fQh62Am+qqh+1J6ZfZt4dwyU4ul3mLe26+T87mf5rVfWONv/nAPsAa9I8w/FLwB+3\n+fwUcFHHfD+iKTg+oqp+3K7T7y4zdg3fj4H7Agcm2bWqNlXVV/qY73eA/1NVN7ZVKP83cHAWvgv5\nbGBTVf11VW2vqi8A7wd+bbmZ6NPDgH8E/q6q5p+4X1xVn6yqH9Cc6D8hyX4D/v4f0aznB1XVHVV1\nVZt+NPDXVXVdVX2P5mKNJlevbepY4DVVtbWq/h34E3a8KLczF1bVp6vqJ8APgRcCL66qze0++J/b\n/+//AC6pqkuq6idVdSnweeDwXgtOU3vkcOAlVfW9turlG2lOKIdhV5oLUHvRnBj/R8e4lThOzvcj\nmirhD6uq77fHO2jWyfVV9b6q+hHwJsBGcTTu/rqqvtJehPkw8JX2ecntwN9x93nuqI/JWgILkMvX\nuRP/D/p/iPj2js//X5fhnS3nYcAt7UF8ztdorvQs1uaOOyBzy3nYEpbT6WHALfOWuZC71mPHQfsB\n7XK+Ne9A3rncvwE+CpzXVsf6s/burCZI2zDMS2gKLVuTnJekn//gw4Ez2ipy3wa+BYSFt4OHA4+f\nm6ed71jgPy0rE/07ArgfTVXP+e76b1fVNpr8LHdbnO//oTkh/VpbDW6uGttit1mNsQW2qYex42+7\n2P19539kb2A3mjti8z0c+LV529mTaC4O9vJwmkLclo55/ormrsQwPAI4EviTqvrhvHErcZyc7+U0\n+69/SfNYygvb9B22zTaOW7rML42Tfs9zR31M1hJYgFwZ3wPuPzeQZCU2gq8D+yXp/A1/Gti8hGXt\n21FddG45X28/75AX+t+gt9BUK+1c5lJsAfZK0hnDXcttrwb/SVUdSFOF6tnAC5b4XRqhqnpP23Ln\nw4ECXj9/ki6z3UJTHXOPjtf9quqfF/iqW4B/nDfPA6rqd9vxS/3P9+sdwEeAS9qqO53u+m8neQDN\nnZCv092S9jNV9bmqOpLmpPzvubvq0aC22W6/k0agxzb19XZ4Ts/9fY//VOfv+w3g+zTV0+a7Bfib\nedvZ7lW1YYGQb6F5pGPvjnkeVFWP6hYfg982bwR+A/hwWwW400LHyfn/+e+174uKtapuq6rfqqqH\nAb9NU43vEczbNts4llIzwW1TgzSo/9POjskaQxYgV8YXgUclOTjJbqxMVbAraO54vjzJrmkaJ/hV\n4LwlLOungD9ol/NrwM8Dl7TjrgaOacf18yzjnAvaZa5Nsidw6hLioqq+RlPt6dVJ7tPeLfnVufFJ\nnpzkoDSNLHyXpgrQT7ovTeMqySOTPCVNYx7fp7k6Of93/Pc2rbMPxrcDpyV5VLucB7f/4YV8CPi5\nJMe1/+tdk/xSkp9vx18N/Lck929P3nbWCNVSnExTBe4f0jY40jo8yZOS3IfmeajPVtXcnYbb2THv\niz0TDOoAACAASURBVN7PtNvQsUke3FaF+y53r+cLgBOSHNhesHnVEvN2O7C2zYNGZIFt6r3A/0ry\n0PbZuj8G5hqmWdR/qq0B8y7gDUkelqbxtCe03/m3wK8meWabvluSmSRrF1jeFppnAE9P8qD2ecSf\nTfJf20muBn4lTZ9zDwZOW9raWTBP76V5rvrjSToLxgsdJ3fYNtuqwZuB/9Hm/YV0L2TvIMmvdayf\nO2hO0H9C8yzpo5L8tzSNlPwBSys8d9uHSks1/5i0VDs7JmsMWYBcAVX1r8BrgI8DN9E0NDDo7/gh\nTUHqWTRXgf8SeEFVfWkJi7uCptGMb9A0jnBUVX2zHff/0hz47qB5VuY9fS7zHTRVS79I0zjQB5YQ\n15xjgScA36RpkOF8mqvU0BxE30dzInwjzbNlf7OM79Jo3BfYQPMfvI3mZG2Hk8O2GvPrgE+31VwO\nraoP0txVOS/Jd4HraLaJnqrqTpqGOY6huYNwW7uM+7aTvJHm2a7baZ7HPXcQGZwXQwEnAbcCF7Yn\n69BsX6+iqbr6OJrnyOa8GjinzfvRy9jPHAdsatfX79BsX1TVh2merfoEsLF9X4pPANcDtyX5xhKX\noeXrtU29luai3DXAtTT759fCko9dL2uX8zma/+3rgXu1Fz7mGrn6d5q7DP+TnZ93vAC4D3ADzXHn\nfbTVXtvnKM9vY7+S5sRz4KrqHJr18Inc3VLkQsfJM4Cj0rQE/eY27bdo8vtN4FHAQrUi5vwScEWS\nbTTP+r+4qm6uqm/QPA+2oV3eAcCnl5Cve+xDF7sMqcP/obkY9W36v7lwD30ckzWGsmOVfmn8pWna\n/EvVtAApaYUkKeCA9nk6SWMiySzwt1X1zlHHImn18Q6kxl5bleFn2+pMh9Fc1f77UcclSZIkrTYW\nIMdY+5zHth6vvhu4SPL2Hsvo1grkOPpPwCywDXgz8LttM89SV0l+ude2M8DvuL7Hd3y1R/qxg/ru\nZcS2pBiWs7xBx6LJt8Bx7ZcHtPyex7xRHwsHHcNyljcO60PSZLIKqyRJkiSpL96BlCRJkiT1ZZdR\nB7AUe++9d61bt26HtO9973vsvvv8LtWmwzTnDaY7f1deeeU3quqho45jELptd6vJNP9Pl2Kc14fb\n3eKN2+9pPAsbx3i+9KUvrZrtbtzW/5xxjQuMbal2FtvIjndVNbAXTce2l9M0v309TRPU0DQ/v5mm\nD6ergcM75jmNpsn4LwPP7Od7Hve4x9V8l19++T3SpsU0561quvMHfL4GuI2N8tVtu1tNpvl/uhTj\nvD7c7hZv3H5P41nYOMazmra7cVv/c8Y1ripjW6qdxTaq7W7QdyC3A6dU1VVJHghcmeTSdtwbq+rP\nOydOciBNvy+PAh5G03Hvz1XVjwcclyRJkiRpmQb6DGRVbamqq9rPd9J07L7vArMcCZxXVT+oqq/S\n3Ik8ZJAxSZIkSZIGY8WegUyyDngMcAXwRODkJC8APk9zl/IOmsLlZztmu5UeBc4kJwEnAaxZs4bZ\n2dkdxm/btu0eadNimvMG058/SatPkk3AncCPge1VtT7JXsD5wDpgE3B0eyyUJGlirEgBMskDgPcD\nL6mq7yZ5G/CnQLXvpwMvXMwyq+pM4EyA9evX18zMzA7jZ2dnmZ82LaY5bzD9+ZO0aj25qr7RMXwq\ncFlVbUhyajv8itGEJknS0gy8G48ku9IUHs+tqg8AVNXtVfXjqvoJ8A7urqa6mabhnTlr2zRJkqbN\nkcA57edzgOeOMBZJkpZkoHcgkwQ4C7ixqt7Qkb5PVW1pB58HXNd+vgh4T5I30DSicwDwL8uJYd2p\nF3dN37ThiOUsVtKAdNtG3T41hQr4WJIC/qqtRbOm41h4G7Cm24w7e2RjUK7d/J27Pq+5H7zl3As5\naN8Hr8h3Lda4PdpgPAvbtm3bqEPQmLt283c4Yd7x32P/5Bp0FdYnAscB1ya5uk17JfD8JAfTHFA3\nAb8NUFXXJ7mAptuP7cCLbIFVkjQFnlRVm5P8FHBpki91jqyqaguX97CzRzYGpfNk7pSDtnP6tbuw\n6diV+a7FGrdHG4xnYeNUmJW08gZagKyqTwHpMuqSBeZ5HfC6QcYhSdIoVdXm9n1rkg/SPLpx+1yN\nnCT7AFtHGqQkSUsw8GcgJUlazZLs3vaFTJLdgWfQPLpxEXB8O9nxwIWjiVCSpKWzAClJ0mCtAT6V\n5Is0z/VfXFUfATYAT09yE/C0dljSIiTZL8nlSW5Icn2SF7fpr06yOcnV7evwjnlOS7IxyZeTPHN0\n0UvTYcX6gZQkaTWqqpuBR3dJ/ybw1OFHJE2V7TT9iV/V3um/Msml7bg3VtWfd06c5EDgGOBRNA02\nfjzJz9nmhrR03oGUplySTUmuba/Ifr5N2yvJpUluat/3HHWckiTtTFVtqaqr2s93AjcC+y4wy5HA\neVX1g6r6KrCRu7uTk7QE3oGUVgc7NJckTZUk64DHAFfQ9ARwcpIXAJ+nuUt5B03h8rMds91KlwLn\nYrrPGbduVOaMa1zQdBV0ykHbd0gbl1jHeb2Na2wWIKXV6Uhgpv18DjCLBUhJ0oRI8gDg/cBLquq7\nSd4G/ClNl3F/CpwOvLDf5S2m+5xx60ZlzrjGBU0/s6dfu2Oxw26Ddm5cY7MAKU2/serQfP4VSGgO\nLPONS4fmc8b1KuCouD4kjUqSXWkKj+dW1QcAqur2jvHvAD7UDm4G9uuYfW2bJmmJLEBK02+sOjTv\n7Lx8IeNyZXLOuF4FHBXXh6RRSBLgLODGqnpDR/o+HRdGn0fTdQ403ee8J8kbaBrROYCmdWRJS2QB\nUppydmguSZoiTwSOA65NcnWb9krg+UkOpql1swn4bYCquj7JBcANNC24vsgWWKXlsQApTbG2E/N7\nVdWdHR2av4a7OzTfgB2aS5ImRFV9CkiXUZcsMM/rgNetWFDSKmMBUppua4APNjV+2AV4T1V9JMnn\ngAuSnAh8DTh6hDFKkiRpQliAlKaYHZpLkiRpkO416gAkSZIkSZPBAqQkSZIkqS8WICVJkiRJffEZ\nSEmSBMC6Lv20btpwxAgikSSNK+9ASpIkSZL6YgFSkiRJktQXC5CSJEmSpL5YgJQkSZIk9cUCpCRJ\nkiSpLxYgJUmSJEl9GXgBMsl+SS5PckOS65O8uE3fK8mlSW5q3/ds05PkzUk2JrkmyWMHHZMkSZIk\naflW4g7kduCUqjoQOBR4UZIDgVOBy6rqAOCydhjgWcAB7esk4G0rEJMkSZIkaZkGXoCsqi1VdVX7\n+U7gRmBf4EjgnHayc4Dntp+PBN5djc8CeyTZZ9BxSZI0TEnuneQLST7UDu+f5Iq2xs35Se4z6hgl\nSVqsXVZy4UnWAY8BrgDWVNWWdtRtwJr2877ALR2z3dqmbelII8lJNHcoWbNmDbOzszt817Zt25id\nneWUg7Z3jWX+9JNkLm/TatrzJ2nVejHNRdQHtcOvB95YVecleTtwIta6kSRNmBUrQCZ5APB+4CVV\n9d0kd42rqkpSi1leVZ0JnAmwfv36mpmZ2WH87OwsMzMznHDqxV3n33TsTNf0STCXt2k17fmTtPok\nWQscAbwOeGmag+BTgP/eTnIO8GosQEqSJsyKFCCT7EpTeDy3qj7QJt+eZJ+q2tJWUd3apm8G9uuY\nfW2bJknSpHoT8HLgge3wQ4BvV9VcNZm52jb3sLMaN4PSWWNnzf0Yqxo841YzxXgWtm3btlGHIGmI\nBl6AbK+yngXcWFVv6Bh1EXA8sKF9v7Aj/eQk5wGPB77TUdVVkqSJkuTZwNaqujLJzGLn31mNm0Hp\nrLFzykHbOf3a7qcEo6jBM241U4xnYeNUmJW08lbiDuQTgeOAa5Nc3aa9kqbgeEGSE4GvAUe34y4B\nDgc2Av8B/MYKxCRJ0rA8EXhOksOB3WiegTyDppG4Xdq7kNa2kSRNpIEXIKvqU0B6jH5ql+kLeNGg\n45AkaRSq6jTgNID2DuTLqurYJH8HHAWcx441cST1Kcl+wLtpGmMs4MyqOiPJXsD5wDpgE3B0Vd3R\n1ow7g+ZmxX8AJ8z1FiBpaVaiH0hJY8buBKSx8AqaBnU20jwTedaI45Emkf2NSyNmAVJaHea6E5gz\n153AI4A7aLoTkDRgVTVbVc9uP99cVYdU1SOq6teq6gejjk+aNPY3Lo3eivYDKWn07E5A0roeXVxJ\nk2yY/Y13GrdWcOeMa1zQvaXncYl1nNfbuMZmAVKafmPVnUCvrgLmG7cd5rjuxEfF9SFplIbd33in\ncWsFd864xgXwlnMvvEdLz+PSR/s4r7dxjc0CpDTFxrE7gRP6vBMyLgeWOeO6Ex8V14ekUbG/cWm0\nfAZSmm5z3Qlsomn58Sl0dCfQTuPBVJI0Efrobxzu2d/4C9I4FPsbl5bNAqQ0xarqtKpaW1XrgGOA\nT1TVscDlNN0JgN0JSJImx1x/409JcnX7Opymv/GnJ7kJeFo7DE1/4zfT9Df+DuD3RhCzNFWswiqt\nTq8AzkvyWuAL2J2AJGkC2N+4NHoWIKVVoqpmgdn2883AIaOMR5IkSZPHKqySJEmSpL5YgJQkSZIk\n9cUCpCRJkiSpLxYgJUmSJEl9sQApSZIkSeqLBUhJkiRJUl/sxkPSWFp36sX3SNu04YgRRCJJkqQ5\n3oGUJEmSJPXFAqQkSZIkqS8WICVJkiRJfbEAKUmSJEnqi43oSFox3RrCkSRJ0uQa+B3IJO9KsjXJ\ndR1pr06yOcnV7evwjnGnJdmY5MtJnjnoeCRJkiRJg7ESVVjPBg7rkv7Gqjq4fV0CkORA4BjgUe08\nf5nk3isQkyRJkiRpmQZegKyqTwLf6nPyI4HzquoHVfVVYCNwyKBjkiRpWJLsluRfknwxyfVJ/qRN\n3z/JFW2tm/OT3GfUsUqStFjDbETn5CTXtFVc92zT9gVu6Zjm1jZNkqRJ9QPgKVX1aOBg4LAkhwKv\np6mN8wjgDuDEEcYoSdKSDKsRnbcBfwpU+3468MLFLCDJScBJAGvWrGF2dnaH8du2bWN2dpZTDtre\ndf7500+SubxNq2nPn6TVpaoK2NYO7tq+CngK8N/b9HOAV9McHyVJmhhDKUBW1e1zn5O8A/hQO7gZ\n2K9j0rVtWrdlnAmcCbB+/fqamZnZYfzs7CwzMzOc0KPVx03HznRNnwRzeZtW054/SatP+zz/lcAj\ngL8AvgJ8u6rmrnL2rHGzswumS9Hr4uqcNffrPc0oLvCN24VF41nYtm3bdj6RpKkxlAJkkn2qaks7\n+DxgroXWi4D3JHkD8DDgAOBfhhGTJEkrpap+DBycZA/gg8B/XsS8C14wXYpeF1fnnHLQdk6/tvsp\nwSguwI7bhUXjWdgwC7NJ3gU8G9haVb/Qpr0a+C3g39vJXtnRYONpNNXFfwz8QVV9dGjBSlNq4AXI\nJO8FZoC9k9wKvAqYSXIwTRWeTcBvA1TV9UkuAG4AtgMvag+6kgYgyW7AJ4H70mzv76uqVyXZHzgP\neAjNXZLjquqHo4tUmk5V9e0klwNPAPZIskt7F7JnjRtJCzobeCvw7nnpb6yqP+9MmNfa/8OAjyf5\nOc81peVZiVZYn19V+1TVrlW1tqrOqqrjquqgqvrFqnpOx91Iqup1VfWzVfXIqvrwoOORVjkb85CG\nLMlD2zuPJLkf8HTgRuBy4Kh2suOBC0cToTS5bO1fGr1htsIqaciq0asxj/e16ecAzx1BeNK02ge4\nPMk1wOeAS6vqQ8ArgJcm2Uhz9/+sEcYoTRtb+5eGZFitsEoakeU05iFp8arqGuAxXdJvxrsf0kpY\n8db+O41bI0ZzxjUu6N5Q17jEOs7rbVxjswApTbnlNOax3NYgd9by42KNcic6rjvxUXF9SBoXw2jt\nv9O4NWI0Z1zjAnjLuRfeo6GucekhYZzX27jGZgFSWiWW0pjHcluD3FnLj4s1yoPNuO7ER8X1IWlc\n2Nq/NFwWIKUpluShwI/awuNcYx6v5+7GPM7DxjwkSRPC1v6l0bMAKU23fYBz2ucg7wVcUFUfSnID\ncF6S1wJfwMY8JEkToKqe3yW55zGsql4HvG7lIpJWHwuQ0hSzMQ9JkiQNkt14SJIkSZL6YgFSkiRJ\nktQXC5CSJEmSpL5YgJQkSZIk9cUCpCRJkiSpLxYgJUmSJEl9sQApSZIkSeqLBUhJkiRJUl8sQEqS\nJEmS+mIBUpIkSZLUl11GHYAkSRqcdadePOoQJElTzDuQkiRJkqS+WICUJEmSJPXFAqQkSZIkqS8W\nICVJkiRJfRl4ATLJu5JsTXJdR9peSS5NclP7vmebniRvTrIxyTVJHjvoeCRJkiRJg7ESdyDPBg6b\nl3YqcFlVHQBc1g4DPAs4oH2dBLxtBeKRJGlokuyX5PIkNyS5PsmL2/SuF1MlSZokAy9AVtUngW/N\nSz4SOKf9fA7w3I70d1fjs8AeSfYZdEySJA3RduCUqjoQOBR4UZID6X0xVZKkiTGsfiDXVNWW9vNt\nwJr2877ALR3T3dqmbWGeJCfR3KVkzZo1zM7O7jB+27ZtzM7OcspB27sGMH/6STKXt2k17fmTtLq0\nx7st7ec7k9xIc2w7EphpJzsHmAVeMYIQJUlasmEVIO9SVZWkljDfmcCZAOvXr6+ZmZkdxs/OzjIz\nM8MJPTpQ3nTsTNf0STCXt2k17fmTtHolWQc8BriC3hdTJfUpybuAZwNbq+oX2rS9gPOBdcAm4Oiq\nuiNJgDOAw4H/AE6oqqtGEbc0TYZVgLw9yT5VtaWtorq1Td8M7Ncx3do2TdIAJNkPeDfNiWoBZ1bV\nGb0OtqOKU5pGSR4AvB94SVV9tzmXbSx0MXVnNW52pldNnIWsuV/v+UZRQ2TcaqYYz8K2bds2zK87\nG3grzbFtzlz18A1JTm2HX8GObW08nqatjccPM1hpGg2rAHkRcDywoX2/sCP95CTn0WzQ3+m4Oitp\n+eaexboqyQOBK5NcCpxA94OtpAFIsitN4fHcqvpAm9zrYuoOdlbjZmd61cRZyCkHbef0a7ufEoyi\nBs+41UwxnoUNszBbVZ9s7+x36lU9/K62NoDPJtljbhscTrTSdBp4ATLJe2k24r2T3Aq8iqbgeEGS\nE4GvAUe3k19CU61gI03Vgt8YdDzSauazWNLwtdXmzgJurKo3dIzqdTFV0vKseFsbncbtDvCccY0L\nutdyGJdYx3m9jWtsAy9AVtXze4x6apdpC3jRoGOQdE9LeRZrFFXpFvKWc7ufbx+074MH+j3djOtO\nfFRcHwt6InAccG2Sq9u0V9L7YupYW9erbYENRww5EmnnVqqtjU7jdgd4zrjGBc3xe34th3Fpn2Sc\n19u4xjb0RnQkDd9Sn8UaRVW6pRjGQWhcd+Kj4vrorao+BaTH6HtcTJW0bLa1IQ3RwPuBlDReFnoW\nqx3f81ksSZImwFz1cLhnWxsvSONQbGtDGggLkNIU6+NZLPBZLEnShGjb2vgM8Mgkt7ZVwjcAT09y\nE/C0dhiatjZupmlr4x3A740gZGnqWIVVmm5T9SyWJGl1s60NafQsQEpTzGexJEmSNEhWYZUkSZIk\n9cUCpCRJkiSpLxYgJUmSJEl9sQApSZIkSeqLBUhJkiRJUl8sQEqSJEmS+mIBUpIkSZLUFwuQkiRJ\nkqS+WICUJEmSJPXFAqQkSZIkqS8WICVJkiRJfbEAKUmSJEnqyy6jDkCSJE2edadefI+0TRuOGEEk\nkqRh8g6kJEmSJKkvFiAlSZIkSX2xCqskSROqWzXSUbJaqyRNP+9ASpIkSZL6MtQCZJJNSa5NcnWS\nz7dpeyW5NMlN7fuew4xJkqRBSvKuJFuTXNeR5rFOkjQVRnEH8slVdXBVrW+HTwUuq6oDgMvaYUmS\nJtXZwGHz0jzWSZKmwjhUYT0SOKf9fA7w3BHGIk0V74RIw1dVnwS+NS/ZY520wqzpJg3HsBvRKeBj\nSQr4q6o6E1hTVVva8bcBa7rNmOQk4CSANWvWMDs7u8P4bdu2MTs7yykHbe/6xfOnnyRzeZtW056/\nETsbeCvw7o60uTshG5Kc2g6/YgSxSatJX8c6Scv25Kr6RsewxzxpwIZdgHxSVW1O8lPApUm+1Dmy\nqqotXN5DW9g8E2D9+vU1MzOzw/jZ2VlmZmY4oUeLdJuOnemaPgnm8jatpj1/o1RVn0yybl7ykcBM\n+/kcYBYPptLQLHSsg51fMO3U66LpYq253+CWNd9SLhCO24VF41nYtm3bRh3CQjzmSQM21AJkVW1u\n37cm+SBwCHB7kn2qakuSfYCtw4xJWoX6vhOymBPZblbqhHS+YZxIjdsJ26i5Phat72Pdzi6Ydup1\n0XSxTjloO6dfuzKnBEu5gDtuFxaNZ2FjtC9YsZpuncZ1/zeucUH3i1TjEus4r7dxjW1oBcgkuwP3\nqqo728/PAF4DXAQcD2xo3y8cVkzSarezOyGLOZHtZlAntzszjBoG43bCNmquj0XzWCetvBWr6dZp\nXPd/4xoXwFvOvfAeF6nGpXbgOK+3cY1tmHcg1wAfTDL3ve+pqo8k+RxwQZITga8BRw8xJmk18q6/\ntIKSvJemytzeSW4FXkVTcPRYNyDrulyc2rThiBFEonFiTTdpOIZWgKyqm4FHd0n/JvDUYcUhyTsh\n0kqqquf3GOWxTloh1nSThmfYjehIGiLvhEiSVglruklDYgFSmmLeCZEkrQbWdJOG516jDkCSJEmS\nNBksQEqSJEmS+mIBUpIkSZLUF5+BlKQhsNsBSZI0DbwDKUmSJEnqiwVISZIkSVJfrMIqaSC6VdEc\nt++2yqgkSdLyeAdSkiRJktQX70BKWjWW05DNtZu/wwk2hCNJklY570BKkiRJkvpiAVKSJEmS1BcL\nkJIkSZKkvliAlCRJkiT1xQKkJEmSJKkvtsIqSZJWzCj7abWPWEkaPO9ASpIkSZL6YgFSkiRJktQX\nq7BK0jL0W0VuWN89bVXxVkMe1ej8rU85aDsnDHHb6rUdr9b/mtudpIV4B1KSJEmS1JdVcwfSq2mS\nJEmStDxjUYBMchhwBnBv4J1VtWHEIUlTz+2uMcqLSytR/dUWL8eb293KWcntqbNKbb//30m5cD3K\navjD4nYnDdbIC5BJ7g38BfB04Fbgc0kuqqobVvq7+925L2bnOo4HB2m+UW530mrldicNn9udNHgj\nL0AChwAbq+pmgCTnAUcCI9mwh3VHoN+C6ijvhFgY3tGUraOx2u6kVcLtTho+tztpwFJVow0gOQo4\nrKp+sx0+Dnh8VZ08b7qTgJPawUcCX563qL2Bb6xwuKMyzXmD6c7fw6vqoaMOYr4BbneryTT/T5di\nnNeH293ijdvvaTwLG8d4dl9F2924rf854xoXGNtS7Sy2kRzvxuEOZF+q6kzgzF7jk3y+qtYPMaSh\nmea8wfTnb5LtbLtbTfyf7sj1sXJGsd2N2+9pPAsb03jWjTqO5VjMdjdu63/OuMYFxrZU4xrbOHTj\nsRnYr2N4bZsmaeW43UnD53YnDZ/bnTRg41CA/BxwQJL9k9wHOAa4aMQxSdPO7U4aPrc7afjc7qQB\nG3kV1qranuRk4KM0zSu/q6quX8Kiprma3TTnDaY/f2NngNvdauL/dEeuj0Ua8+1u3H5P41mY8fRp\nhba7cc3vuMYFxrZUYxnbyBvRkSRJkiRNhnGowipJkiRJmgAWICVJkiRJfZn4AmSSw5J8OcnGJKeO\nOp7lSrJfksuT3JDk+iQvbtP3SnJpkpva9z1HHetSJbl3ki8k+VA7vH+SK9rf8Pz2IXdpaBa73aXx\n5vY/e02Sx442Byuj3201yX3b4Y3t+HWjjFv3NI7/8XH7fyXZI8n7knwpyY1JnjCq9ZPkD9vf6bok\n702y2//P3r3HyVHV+f9/vUmAhICEm2NMkAR1USQKmB/igjqCIgIK7rIuLCIRXNY7Sr7Kxd0VL7vi\nKiqrLsiKGl0gIIKwuCqIjIiXIGAg4SYRIkkMhFuAQQWCn98f50xS6fTMdM/0dFX3vJ+PRz+6u66f\nqjqnqk7VqVPtXj+Svi5ptaQlhW5Nrw9Jx+Th75J0TCtiGwuD5ZGaYXolPSppUf78a6HfmJ2PNhjb\nhwtxLZH0jKRtc79lkhbnfje0OLZJkq6XdHOO7eN1hhk0jUo6JXe/U9IbSojtxLxeb5F0taSdCv2e\nKazTlja81GBscyU9UIjhnYV+5eariOjYD+lh6N8BOwObATcDu5Yd1yiXaRqwZ/69FfBbYFfgP4CT\nc/eTgc+UHesolvFE4Hzgivz/IuCI/Pts4N1lx+jP+Po0m++Ag4AfAAL2BhaWvQxjtF4ayqvAe4Cz\n8+8jgAvLjt2fjbZl5dJ41dIXMB94Z/69GTC1jPUDTAfuASYX1svcdq8f4NXAnsCSQrem1gewLXB3\n/t4m/96m7PwwyPLWzSM1w/QOpNea7mN6PtpIbDXDvwn4SeH/MmD7MVpvArbMvzcFFgJ71wxTN42S\n9kE3A5sDs/I6nNDm2F4LbJF/v7uYf4D+MUxvjcQ2F/hynXFLz1edfgdyL2BpRNwdEU8BC4BDS45p\nVCJiVUTclH8/DtxOOpgcSjq4kb8PKyfC0ZE0AzgY+Fr+L2A/4OI8SMcum3WuEeS7Q4FvRfIrYKqk\naW0Oe0w1mVeL6+liYP88vFVE1dJ41dKXpK1JBaZzASLiqYhYQ3n7gInAZEkTgS2AVbR5/UTEtcDD\nNZ2bXR9vAK6KiIcj4hHgKuDA0cY2FobII40Y0/PREcR2JHBBq+Y/TGwREf3576b5U9tC52Bp9FBg\nQUQ8GRH3AEtJ67JtsUXENRHxx/z3V6T3hI65BtfbYErPV51egJwOLC/8X0Hjmb3y8i3+PUhXJXoi\nYlXudR/QU1JYo/VF4CPAX/L/7YA1EbE2/++qbWidp8F819X7nqyZvLpufeT+j+bhrYIqksarlr5m\nAQ8A31CqVvs1SVMoYf1ExErgc8C9pILjo8CNVCP/Nbs+OnJfWZNHar0yVzv8gaSX5G5tW85heQ/n\n0gAAIABJREFUYkPSFqTCxHcLnQO4UtKNko4fg5gmSFoErCYVbGpjGyyNjvl6ayC2ouNId9IHTJJ0\ng6RfSWr5zY0GY/vbXL32Ykk75m6l56tOL0B2LUlbkjL/ByPisWK/SPevO+79K5IOAVZHxI1lx2JW\nTzfmu5FwXu1eVUjjFU1fE0nVNc+KiD2AJ0hVNNdp4/rZhnRnZhbwXGAKFbxr1637xKHyCHATsFNE\nvAz4EvC9CsU24E3AzyOiePd434jYE3gj8F5Jr25lXBHxTETsTrp7t5ek3Vo5/dFoNDZJbwPmAJ8t\ndN4pIuYA/wB8UdLz2xzb/wIzI+KlpLuM82unUZZOL0CuBHYs/J+Ru3U0SZuSdhDnRcQlufP9A9Vj\n8vfqsuIbhX2AN0taRqresR9wJqmqy8Q8TFdsQ+s8Tea7rtz3FDSbV9etj9x/a+ChdgZsw6tQGq9i\n+loBrCjcAbiYVKAsY/28DrgnIh6IiKeBS0jrrAr5r9n10VH7ykHyyDoR8dhAtcOI+D9gU0nb04bl\nHC62giOoqb6a72oTEauBS2lhNdGa+awBrmHjCx6DpdG2pY8hYkPS64CPAm+OiCcL4wyst7uBPtKd\n37bFFhEPFeL5GvDy/Lv0fNXpBchfAy9UaplsM1KmaWkrSe2W64SfC9weEZ8v9LocGGhl6RjgsnbH\nNloRcUpEzIiImaRt9ZOIOIqUaQ7Pg3XksllnG0G+uxx4u5K9gUcL1bo63gjyanE9HZ6H77o7E52s\nSmm8iukrIu4DlkvaJXfaH7iNcvYB9wJ7S9oib7eBWKqQ/5pdHz8CDpC0Tb6zekDuVjlD5JHiMM8Z\neL5U0l6k8+iHGOPz0UZiy8NtDbyGwnmUpCmSthr4TdoGS+pPYUSx7SBpav49GXg9cEfNYIOl0cuB\nI5RaaZ0FvBC4vp2xSdoD+Cqp8Li60H0bSZvn39uTLuLc1ubYis9Vv5n07CtUIV9FG1vsGYsPqeWv\n35Jabvpo2fG0YHn2JVUJuQVYlD8HkeqKXw3cBfwY2LbsWEe5nL2sb3lvZ9IOYynwHWDzsuPzZ3x9\nms13pNbTvpL3O4uBOWUvwxium2HzKjAp/1+a++9cdtz+bLQdK5nGq5S+gN2BG/I6+h6pdcNS1g/w\ncdLJ5BLg26RWKtu6fkh3sVYBT5Pu0B43kvUBHJtjWwq8o+y8MII88i7gXXmY9wG3kloO/RXw14Xx\nx+x8tJHY8nBzSY3SFMfdOcd7c4691bG9FPhNjm0J8K+5+ydIhbIh0yjpzt/vgDuBN5YQ24+B+wvr\n9fLc/a9zWr45fx9XQmyfLqS3a4AXFcYvNV8pB2FmZmZmZmY2pE6vwmpmZmZmZmZt4gKkmZmZmZmZ\nNcQFSDMzMzMzM2uIC5BmZmZmZmbWEBcgzczMzMysUiT9naRbJf1F0pwhhlsmabGkRZJuKHT/pKRb\ncvcrJT03d99G0qW53/WSdsvdJ+X/N+f5frwwrf0k3SRpiaT5A+9klfThPP1Fud8zkrbN/T6Up7NE\n0gWSJuXukvRvkn4r6XZJHxhmPWwm6Rt5GW+W1DuK1doSLkCamZmZmVlpJPVK+mZN5yXA3wDXNjCJ\n10bE7hFRLGh+NiJeGhG7A1cA/5q7nwosioiXAm8HzszdnwT2i4iXkV7rc6CkvSVtAswHjoiI3YDf\nk99rGRGfzfPdHTgF+GlEPCxpOvAB0ittdgMmkN4PCul1KzuSXsvxYmDBMMv2j3les0nvizwjx1Qa\nFyDNzMzMzKxSIuL2iLhzFOM/Vvg7hfQuTYBdgZ/kYe4AZkrqiaQ/D7Np/gTp/adPRcRvc7+rgL+t\nM8sjSe9PHTARmJzvVm4B/CF3fzfwiYj4S45hNYCkKZK+nu+C/kbSoXXiXQ2sAQa9I9sOLkCamZmZ\nmVmnCuBKSTdKOr7YI1cVXQ4cxfo7kDeT7mwiaS9gJ2BG/j9B0iJgNXBVRCwEHgQmFqrRHk66g1ic\nzxbAgcB3ASJiJfA54F5gFfBoRFyZB38+8PeSbpD0A0kvzN0/CvwkIvYCXgt8VtKUHO+bJU2UNAt4\nee38280FSDMzMzMzaztJC3OB7WukQtLA84RvaGIy+0bEnsAbgfdKevVAj4j4aETsCJwHvC93Ph2Y\nmuf7fuA3wDN5+GdyddQZwF6SdouIIFU//YKk64HHB4YveBPw84h4OC/XNsChwCzgucAUSW/Lw24O\n/DlXt/1v4Ou5+wHAyTmuPmAS8LzcfwVwA/BF4Bd15t9WE8ucuZmZmZmZjU8R8QpIz0ACcyNi7gim\nsTJ/r5Z0KbAXGz83eR7wf8DHctXWd+T5CrgHuLtmmmskXUO6q7gkIn4JvCqPcwDwVzXTP4INq6++\nDrgnIh7I41wC/DXwP6TC4CV5uEuBb+TfAv52kGq7Hxr4IekXwG/rDNM2vgNpZmZmZmYdJz83uNXA\nb9JdvCX5/wsLgx4K3JG7T5W0We7+TuDaiHhM0g6SpuZhJpMarBkY59n5e3PgJODsQgxbA68BLivM\n715gb0lb5ELq/sDtud/3SFVUyeMNFAZ/BLw/D4+kPfL3FnnZkPR6YG1E3DaC1dUyvgNpZmZmZmaV\nIuktwJeAHYDvS1oUEW/Ir+P4WkQcBPQAl+Yy10Tg/Ij4YZ7E6ZJ2Af5Cajn1Xbn7i4H5kgK4FTgu\nd5+Wu08g3WS7KCKuyP0+LOmQ3P2siPhJIdS3AFdGxBMDHSJioaSLgZuAtaRqsucMxAWcJ+lDQD+p\nEAvwSVIV1VtyK6v3AIcAzwZ+JOkvwErg6KZXZospVes1MzMzMzMzG5qrsJqZmZmZmVlDXIA0MzMz\nMzOzhpRegJT0TUmfKnsaY0nSqySN+EWokp4nqT/XyR63JJ0t6V/KjqMbSJor6bqy4yiTpF5JKwr/\nb82twLVyHqPaN41FTKMh6ShJVw4/5KDjh6QXNNuvyiTtI+kuSX+WVGqjBmXrhDzVrUZ7niCpT9I7\nm+1XZZJ6JF0r6XFJZ5Qdj21I0jJJrxvhuOvOB2v3O9YeDRUgm9nIo0kQ3SoifhYRu4xi/HsjYsuI\nKPWdL2Ol0cJMRLwrIj45ivl03EFQ0rMlXSDpD5IelfRzSa9ochoz88m5G80aQkS8JCL6yo6jqBiT\npNMk/U/J8ZwXEQe0e75Vu+BRk6c+AXyZ1DjDw524nxkrVcxT3aqs84SKH1+OJ70A/lkRMa8K+1Br\njaHOB10OaY/S70CaNWIc333dEvg18HJgW2A+qSWyLUuNqqBdJw4VPUEx24nUil/LOE9Vn9ddR9gJ\nuC3cWmQpnEe6XEQM+QG+TWr+9k+kpmY/AryZdMBcA/QBLx5s2Nz9O8B9wKOkF3u+pDD9bwKfaiCO\njwCrgD+QmrsN4AWFaXwF+D7wOLAQeH5h3AOAO/P8/wv4KfDOBuZ5LOmdLY+Q3s2yU6FfAO8B7srz\n/CTwfOAXwGPARcBmedheYEVh3JNIzfA+nuPaP3ffC7ghj38/8PncfWae38T8/7nA5cDDwFLgHwvT\nPi3P+1t5+rcCc4abd7OfPJ/vkF6I+jiwmPRS1VOA1cBy4IDC8FsD5+ZtuBL4FDCB1JTyn4FncppZ\nU9imZ5Fe+voE6YWsG6QV0jt9FuX19TvgwCHi/bc8jz/n+Xw5d/9rUgHt0fz914Vx+vJ2/XlexiuB\n7XO/SXnZHyLlg18DPcOss21JL4v9Q05T38vd3wHcnLflwzm+/y2MtxzYHXgRcFVOC8uAtxbS6XJS\nM9Fr8/RPK4x/bx6nP39eCcwFrgM+l2O5B3jjcNsr95ub18kX8vIPmn8bmM9wafnivJ4fI+X702gu\n3b2DlIcfJ70k+J8K/XrZMF8uA16Xf68prK8n8vqbmfsdQkp3a0j5/aWFaexBarL7ceBCYMFQ6yeP\nsz1wRZ7ew8DPgE2KMZFeZPwU8HSO6ebhttNI9m3U7GsK+eCdxe05kn1GYb85sN/eN2+v3kK/D+Tt\n9CDwWdJFzsH2Eeviqo0tb/diHrorpxuR0u3AtH6bt+HDeZiB/e/1pBdC9+f/y6mfp/6Sv5/MMS4v\ndPsLcLnzVOXy1ItJaWcN6fj45nppvV56zzG/N6eVe3K3l5D2yw+Tjtun5u6bACeTjk0PkY7L2zaQ\nR/bO62AN6bjQW299Frbn/wyWd5vMm+uWnfQqg1uADxf6fZqULx4jvetu25q8UDy+rIurmdjy+r47\nb+t7gKNy9wmk9P5g7v/e4aZHOl94mrTf7M9pbKN96Hj9AB8GvlvT7T+BMxl6X/V84Cc5TT8InAdM\nrUmjJ+X08+Qw22gZaT9zG2lf9g1gUr28V8h/xfP+T+XfveT9DoOUQ/wZgzTUYEJbt9MiHVieIL1c\nc1NSwW4p6wtL64YtjH8ssBWwOen9JosK/dYlgiHmfyCpAPoSYAvSga42IT1EKoBNzAl6Qe63PWmH\n9ze53wl5BzJkAZJUOFlKOthMBP4Z+EVNQr4MeFaO60ngamDnnPluA46pk7h3IR2Qn5v/zyQXdoFf\nAkfn31sCexeGKRYgryUVhCeRChYPAPvlfqeRTmQOIu10Pw38arh5N51w1s/nDXn9fIu0w/9oThf/\nSD7A5uEvBb4KTCG9z+Z68okH9XcU3yQV6vYhHYgnseEOY6/c//W5/3TgRcPE3MeGJwfbknZaR+dl\nODL/364w/O9IaX5y/n967vdPwP+S0uME0h3CZw0z/++TToC2yevoNbn7UaQd3svzNnkMeDL32znH\nNCVvu3/N6/1VpJ33+0jp9O05LfwL6UTsfuCweumnsM6fzttpAvBuUsFTDW6vtcD783qbPMQyDzef\n4dLy08BheRtPpvl0dzDpgCfSy3r/COxZmy8H23fl7v+e49yUdDK7GnhFXp5j8nibA5uR3jP1oTzs\n4Tn+4fZvnya9kHjT/HlVYf2si4mak7LhttNI9m2DpJU+WlyAJO3TlwN71fS7hpQvn0cq3A06X4Yu\nQO5MOgHfhFSg+j2wIqebxbnfFNIJ0gfzevghaT/+cmA3Ulq8KU/jpQySp1hfyJ+bt/edpHToPFWx\nPJV/LwVOzdPej1RY2WW4NFVIo1eR0uhk0nnNKmBeXt9bAa/Iw54A/AqYkZflq8AFw8Q9nXQuc1De\nPq/P/3eotz4ZgwIkMIuU946v6beSlC+mAN8dar6MoACZp/tYYVtMI99sIFUPvwPYMa/7axpZVja+\n6LxBXOP5k9fvE+TCHynvrybt/4baV70gp8vNSe9mvBb4YmG6y0jnIDsyxH6sMOySwnb9OevP8eYy\nggJkvXzizxiloQYT2rqNQTpBvajQb5O8Y+ltZMMBU3Mi2Lo2EQwxzteBTxf+v6BOQvpaof9BwB35\n99uBXxb6iXTiMlwB8gfAcTXL+UfWX6kPYJ9C/xuBkwr/zxjIVGxYgHxBzqSvAzatmee1wMfJd7kK\n3Wey/mRlR9LV860K/T8NfDP/Pg34caHfrsCfhpt30wknzeeqwv83ka72DFyl2irHPJX0ktcnKexM\nSIW1a/LvudQvQH6rTreBHcZXgS80GXMfG54cHA1cXzPML4G5heH/udDvPcAP8+9jqblSPsy8p5EK\nidvU6Xcu6cC5J3AEKb3/BdifdLX/cuDv8/wWA6cU1sFdg6TTcwfWD4MXIJcW/m+Rh3lOg9vr3gaX\ne6j5NJKWrx1puhsknu8BJ9Tmy/x/GRtf/Pr73H3gBO4s4JM1w9xJOpF+NYUT+dzvFwy/f/sE6WLU\nC+r0WxcTG5+UDbmdhpjfoPu2QdJKH60tQJ5CKhTsVqffgYX/7wGuHmy+DH+yv5z1eeoc0knQMaSL\nkT/L3X+Wh51AKphcCHwsd/v3mul9kTp5ig0LkEtZfyLuPFWxPEUqSN5HvhuZu11AvrvcQJoKcmG8\nsA1/M0gMt1Oo4UM6BjzN0IWok4Bv13T7EesvRm+wPml9AfLzeR5H1ul3euH/rqS7eRPqzZeRFyDX\nAH9LTcGDdMfrXYX/BzSyrLgAOdw2/wG5hgLpDu1tNHlcIV2M+k3h/zLg2Abnv6xmux4E/C7/3iDv\n5W6BC5CV+YzkGciBq7kARMRfSAfq6fUGljRB0umSfifpsbxhId0ZbGaeywv/l9cZ5r7C7z+S7uBt\nNG6k1NVIa007AWdKWiNpoBqM2HA57y/8/lOd/xs9pxYRS0lXvE8DVktaIOm5ufdxpLtdd0j6taRD\n6sT1XODhiHi80O33NXHVrotJkiYOM++RqF3eB2P9A/x/yt9bktblpsCqwvr8KunK1lDqbecBO5Lu\nDo7GBmk5G25dDmzTb5MO7AtyAzf/IWnTYeJ9OCIeGSSOO0g7wVeT7mT/kXRH4DWkKtfPJ1Vt+ivg\npLwOjwK2I6XTxyU9TbqLMRl4G8PnsXXLFhF/zD8b3V5DbZtG59NIWq43n0bTHZLeKOlXkh7Oy3EQ\nDe57JO1BahzlLRHxQO68EzBvYL3kae6Yl+W5wMq8jykuz3A+Syp4XCnpbkknNxIfI89XjezbxtIH\nSRchl9TpV9zevyet05H6Kevz1E9JJ8GT8nRnki6y/HVhHUwk3V17Th5fwG6SHpD0KOkuiPNUZ+ep\n5wLL83lLcXrNpP3i+hvqOLQTcGlhmW4nFe57hpj2TsDf1ayLfUmFz3Y4inRD4OI6/Wrz5qY0dx43\npIh4gnRx4V2kfPJ9SS/KvWvPARtJAza8+aRzBfL3txlmX5Vbtl0gaWU+p/8fNk4HzezLWrnPtzZq\ntABZ3Hn/gZTAAJAk0k50ZZ1hAf6BVGXqdaSqnTMHRm0izlWkaiADdhzpuDneGYMPvs5y0i37qYXP\n5Ij4RRPzrisizo+IfUnrMYDP5O53RcSRpIz6GeBiSVNqRv8DsK2krQrdnsf69T+ieY+x5aQrWtsX\n1uWzIuIlA2ENFu4w03x+k3HUTm+DtJw1tC4j4umI+HhE7Ep6jvIQ0t3uwSwnbbepdfr9gXQC20u6\nQn496UR3F1IB8hekA+v9pKuCA+twoIGdfyLdWf4IsEVEiHRVfyCPDbUeB4t1qO01kmnW00haHvF8\nJG1Oqmr1OdLzqVNJz9QOu++R9GzSnZX3RsRvCr2WA/9Ws1/YIiIuIO1rpud9THF5hhQRj0fEvIjY\nmfR8+YmS9q83aM3/RrZTPUPt257Iw2xRGP45G09iVP4OOEzSCXX6FfftzyOlEaifDp5g6DgHCpCv\nyr9/SspPzwbeApxIqgb/JVL1qbWk58Xfncd/Fylf7hgRW5OqRDaSp+r1c56qRp76A7CjpOK5T3H9\nDJemYMP1t5xUXbqe5aRnU4vLNSkihjq+LCfdgSyOMyUiTm8ivtE4jfRoxPl1Gq6rzZtP52FHkjfr\niogfRcTrSQXmO4D/zr1W1Zn/SLQij3WT7wEvlbQb6RzmPIbfV/07aT3OjohnkQqetfm/mfU82D5/\ngzQkqZm07u3cBo0WIO9n/U7yIuBgSfvnOy7zSIntF3WGhVT95UlSPf4tSImvWRcB75D0YklbkKrR\nNur7wGxJh+UWod5LYzuzs4FTJL0EQNLWkv6u2cBrSdpF0n75QPxn0tXdv+R+b5O0Q746uiaPUrxS\nSkQsJ63rT0uaJOmlpDuXwzZNPdS8x1JErCI1QHOGpGdJ2kTS8yW9Jg9yPzBD0mZNTPZcUprYP09v\neuFq5WBq0+b/AX8l6R8kTZT096SqOVcMN3NJr5U0Ox9kHyMdTAddl3kd/AD4L0nbSNpU0qtz7wtI\nz3TuT7p7+D5SwfBVpDuMp5J26k8BR+VxN5X0/+VYTyFVFX4Y2FzSKaQLNwMeyLENdqJTL9ahtldL\njCYtN2gz0nMaDwBrJb2RVPVpSHk/cTGpqtNFNb3/G3iXpFcomSLp4HzC/ktSIeQDefv8DWm7Dje/\nQyS9IJ8kP0q6S1EvLd0PzBw4+R3Fdhp035bvCq0E3qZUe+RYmr9QM5w/kNL6CZLeXdPvwzl/7Eh6\nhuzC3L3ePmIR8DeStlB6f+RxNdP6KfBa0kWXFaRqqweTCpCLSSdPk0nPdW1Ceu7nPyXtIWlX0vPt\nayPiz5L2ovE8VbufcZ6qTp5aSKrd8ZE8vV5Sld0FedTh0lStK4Bpkj4oaXNJW2n9a5bOBv5N0k45\nph0kHTrM9P4HeJOkN+T8N0npHXcDF70XAUfk2OeQngltpadJF3imAN/ShgXtt0naNZ+DfQK4ON+l\nrpcXFgGvVno35dakY9SQlO5sHap00fxJUjXqgf3gRaQ0MEPSNqTGiUZig33oeBcRfybly/NJj/Pc\n28C+aivStnlU0nRSYzyj8d68XbclPXc9sM+/GXiJpN0lTSJd3GjURvtga71GM9GngX9WupX9JtIV\nhy+Rrj69CXhTRDxVO6yk/0d6IP/3pJOS20gPlTclIn5Aah3qGlK1lIFpPNnAuA+Sdoj/QSrE7kpq\naW/IcSPiUtLduQVKt+mXAG9sNvY6NgdOJ627+0gnMwM71wOBWyX1k1rCOiIi/lRnGkeS7uT+gXTS\n87GI+PEo5z3W3k46+Rhobeti1lfL+QmpNbz7JD3YyMQi4nrS84FfIJ0g/JSN7ybWOhM4XNIjkv4z\nIh4iXXWbR0obHwEOyWlmOM/Jy/AYqWrST0nVP4ZyNOkAfQfpjuEH87L8mFRI3Jx0Ne75pDR7N2l9\nHUJq7GF7UitlT5G24WdIJ8WfIe3Qv0Fat+8mHXDJ0/8jqRXan+d8uXcDyzfU9mqlkablYeVqfB8g\nrYtHSAWAyxsYdQap8P5BpRdzD3yeFxE3kBoV+XKe5lLSsxrkfeDf5P8Pk+4aX9LA/F4I/Ji0DX8J\n/FdEXFNnuO/k74ck3ZR/N72dGti3/SPppOAhUgNho651USeGe0mFyJO14TsTLyM9T76IdPHv3Ny9\n3j7iC6S8cD+pKtZ5NfP4LWmd/iz/f4y033smL9utpAZRJpPS3+tIz0z+jPR8zaXA8yQ9Tmq8qm6e\nIl01f3Fh1uv2MzWL7TxVcp7K03sTKb0/SGps6O0RcUceb8g0VSuvj9fnad5Heib9tbn3maR1c2VO\nQ78iNRQ01PSWk2psnUoqmC0n5cWBc7V/IR0fHiG1l3D+MOuhaYV13gN8vVDY+jYpX9xHqiHzgTz8\nRseXiLiKVBC4hZSfh70oS1rGE1lfI+c1pGMZpIsMPyIVKm6isTRQT7196Hg3H5jNhucvQ+2rPk7a\nTz5K2kePdFsMOJ9UYL2bVB38U7Bu//0JUj6+i9TydKNqyyE2BgZabesokl5MOunZPCLWNjnuJqRn\nII8a5CTNzMzGOUlzSQ2q7Ft2LGa2nqSZpJaCN232HNA2JOl5pIvaz8kX2cwa0jG38SW9JVcR2YZ0\n9fx/G91x5OogU5Wqbp5Kqq/d9J1QMzMzM7NOl2+onEh67Z0Lj9aUyhQgJZ1aU7Vl4PODPMhAQyG/\nI1VBqn12ZiivzOMNVLk9LCL+JOnsQeZ5diuXreok/WCQ9RCDdD+17JgHM0i8/ZJeNR7mX4jjVYPF\nMgbzcj4aRgP7t1bPb0y3yRDp65l2pbsRxDaiGEYzvZGO6zw1vHbnqVaRdNQgcd/awnkMdhwaLH+2\n7fjU6mPkaKZXleN1GZSeNX2MVAX7Y2M0j+cNsY5H2hCSVURHVmE1MzMzMzOz9qvMHUgzMzMzs5FQ\narX2ekk3S7pV0sdz91mSFkpaKulCNdfiu5nV0ZF3ILfffvuYOXPmRt2feOIJpkypfW1id/EydpYb\nb7zxwYjYoew4WmGwfNcOVUkTVYkDHMtQcTjfjVxVtuUAxzO0KsVTdr6TJGBKRPQrvWbuOtLrgE4E\nLomIBbka+M0RcdZQ0yrzeNeIKm33VuvWZRur5Sot30VEx31e/vKXRz3XXHNN3e7dxMvYWYAbogJ5\nphWfwfJdO1QlTVQljgjHUs9AHM53I1eVbTnA8QytSvFUKd+R3jt+E+nVKQ8CE3P3VwI/Gm78Mo93\njajSdm+1bl22sVqusvLdxLaXWM2srSRNBb5GemF6AMcCd5Le0zUTWAa8NSJq31tnZmbWMSRNIL17\n8gXAV0gNKK6J9a32rwCmDzLu8cDxAD09PfT19Y15vCPV399f6fhGo1uXrduWywVIs+53JvDDiDg8\nP/uxBel1NldHxOmSTgZOBk4qM0gzM7PRiIhngN3zhdNLgRc1Me45wDkAc+bMid7e3jGJsRX6+vqo\ncnyj0a3L1m3L5UZ0zLqYpK2BVwPnAkTEUxGxBjgUmJ8Hmw8cVk6EZt1J0jJJiyUtknRD7ratpKsk\n3ZW/tyk7TrNulI9z15CqrE6VNHDDZAawsrTAzLqE70CadbdZwAPANyS9jFS15wSgJyJW5WHuA3rq\njVyVKj1VqfpRlTjAsVQ5joLXRsSDhf8n4zv/ZmNC0g7A0xGxRtJk0jsOP0MqSB4OLACOAS4rL0qz\n7uACpFl3mwjsCbw/IhZKOpN00rpORISkus0xV6VKT1WqflQlDnAsVY5jCIcCvfn3fKAPFyDNWmUa\nMD8/B7kJcFFEXCHpNmCBpE8BvyHXyDGzkeu6AuTMk79ft/uy0w9ucyRmlbACWBERC/P/i0kFyPsl\nTYuIVZKmAavbFVC9POr8aV0ogCvzxZmv5osxlbrzv3jloxt1m7X1hErdxa3aXWXHU10RcQuwR53u\ndwN7tT8iqwqXDVqv6wqQZrZeRNwnabmkXSLiTmB/4Lb8OQY4HVfpMRsL+0bESknPBq6SdEexZxXu\n/M+tc1L1zQOnVOoubtXuKjseMzMXIM3Gg/cD5+UWWO8G3kGu3iPpOOD3wFtLjM+s60TEyvy9WtKl\npDsgpd35NzMzaxUXIM26XEQsAubU6bV/u2MxGw8kTQE2iYjH8+8DgE8Al+M7/2Zm1uFcgDQzM2ut\nHuBSSZCOs+dHxA8l/Rrf+Tczsw7nAqSZmVkL5UY7Xlan+0P4zr+ZmXW4TcoOwMzMzMzMzDqDC5Bm\nZmZmZmbWkLYWICV9SNKtkpZIukDSJEmzJC2UtFTShbmlSDMzMzMzM6uYthUgJU0HPgDJBoIXAAAf\nD0lEQVTMiYjdgAnAEcBngC9ExAuAR4Dj2hWTmZmZmZmZNa7dVVgnApMlTQS2AFYB+wEX5/7zgcPa\nHJOZmZmZmZk1oG2tsEbESkmfA+4F/gRcCdwIrImItXmwFcD0euNLOh44HqCnp4e+vr6Nhunv72fe\n7Gfqzr/e8J2ov7+/a5ZlMONhGc3MzMzMOlHbCpCStgEOBWYBa4DvAAc2On5EnAOcAzBnzpzo7e3d\naJi+vj7OuO6JuuMvO2rj4TtRX18f9Za9m4yHZTQzMzMz60TtrML6OuCeiHggIp4GLgH2AabmKq0A\nM4CVbYzJzMzMzMzMGtTOAuS9wN6StpAk0suUbwOuAQ7PwxwDXNbGmMzMzMzMzKxBbStARsRCUmM5\nNwGL87zPAU4CTpS0FNgOOLddMZmZmZmZmVnj2vYMJEBEfAz4WE3nu4G92hmHmZmZmZmZNa/dr/Ew\nMzMzMzOzDuUCpJmZmZl1NEk7SrpG0m2SbpV0Qu5+mqSVkhblz0Flx2rW6dpahdXMzMzMbAysBeZF\nxE2StgJulHRV7veFiPhcibGZdRUXIM26nKRlwOPAM8DaiJgjaVvgQmAmsAx4a0Q8UlaMZmZmoxER\nq4BV+ffjkm4HppcblVl3cgHSbHx4bUQ8WPh/MnB1RJwu6eT8/6RyQjMzM2sdSTOBPYCFpHeOv0/S\n24EbSHcpN7pgKul44HiAnp4e+vr62hVu0/r7+ysd32iMxbLNm722bvd681m88tGNus2evvWoY+i2\nbeYCpNn4dCjQm3/PB/pwAdJs3Fu88lHmnvz9DbotO/3gkqIxa56kLYHvAh+MiMcknQV8Eoj8fQZw\nbO14EXEO6fVyzJkzJ3p7e9sWc7P6+vqocnyjMRbLVrtPG7DsqI3nU2/YesM1q9u2mQuQZt0vgCsl\nBfDVfJDsydV9AO4DeuqNOBZXZOtdCRxuulW5cleVOMCxVDkOMyuHpE1JhcfzIuISgIi4v9D/v4Er\nSgrPrGu4AGnW/faNiJWSng1cJemOYs+IiFy43MhYXJEdydW9qly5q0oc4FiqHIeZtZ8kAecCt0fE\n5wvdpxUumL4FWFJGfGbdxAVIsy4XESvz92pJlwJ7AfcPHFQlTQNWlxqkmZnZ6OwDHA0slrQodzsV\nOFLS7qTaOMuAfyonPLPu4QKkWReTNAXYJLdINwU4APgEcDlwDHB6/r6svCjNzMxGJyKuA1Sn1/+1\nOxazbucCpFl36wEuTTV7mAicHxE/lPRr4CJJxwG/B95aYoxmXUnSBFKrjysj4hBJs4AFwHbAjcDR\nEfFUmTGamZk1ywVIsy4WEXcDL6vT/SFg//ZHZDaunADcDjwr//8M6YXmCySdDRwHnFVWcGZmZiOx\nSdkBmJmZdRtJM4CDga/l/wL2Ay7Og8wHDisnOjMzs5FzAdLMzKz1vgh8BPhL/r8dsCYiBt5jswKY\nXkZgZmZmo+EqrGZmZi0k6RBgdUTcKKl3BOO3/P2r9dR7J2vP5I27l/luzaq929PxmJm5AGlmZtZq\n+wBvlnQQMIn0DOSZwFRJE/NdyBnAynojj8X7V+up907WebPXcsbiDU8NhntP61iq2rs9HY+Zmauw\nmpmZtVREnBIRMyJiJnAE8JOIOAq4Bjg8D+bX55iZWUdyAdLMzKw9TgJOlLSU9EzkuSXHY2Zm1jRX\nYTUzMxsjEdEH9OXfdwN7lRmPmZnZaPkOpJmZmZmZmTXEdyDNzMzMzKzjzazTOJi1nu9AmpmZmZmZ\nWUNcgDQzMzMzM7OGuABpZmZmZmZmDRk3z0DWqxO97PSDS4jEzMzMzMysM/kOpJmZmZmZmTWkrQVI\nSVMlXSzpDkm3S3qlpG0lXSXprvy9TTtjMjMzMzMzs8a0+w7kmcAPI+JFwMuA24GTgasj4oXA1fm/\nmZmZmVlDJO0o6RpJt0m6VdIJubtvVJi1WNsKkJK2Bl4NnAsQEU9FxBrgUGB+Hmw+cFi7YjIzMzOz\nrrAWmBcRuwJ7A++VtCu+UWHWcu1sRGcW8ADwDUkvA24ETgB6ImJVHuY+oKfeyJKOB44H6Onpoa+v\nb6Nh+vv7mTf7mYYDqjeNquvv7+/IuJsxHpax3SRNAG4AVkbEIZJmAQuA7Uh58eiIeKrMGM3MzEYq\nn0uuyr8fl3Q7MJ10o6I3DzYf6ANOKiFEs67RzgLkRGBP4P0RsVDSmdRcBYqIkBT1Ro6Ic4BzAObM\nmRO9vb0bDdPX18cZ1z3RcEDLjtp4GlXX19dHvWXvJuNhGUtwAqnK+LPy/88AX4iIBZLOBo4Dzior\nODMzs1aRNBPYA1hIC29UVEU3X2gf7bLNm7224WHrzafe+K1Y1922zdpZgFwBrIiIhfn/xaQC5P2S\npkXEKknTgNVtjMms60maARwM/BtwoiQB+wH/kAeZD5yGC5BmZtbhJG0JfBf4YEQ8lg55yWhvVFRF\nN19oH+2yza3z2r7B1LuRVG/8Vtxw6rZt1rYCZETcJ2m5pF0i4k5gf+C2/DkGOD1/X9aumMzGiS8C\nHwG2yv+3A9ZExMBlthWkaj4bGYsrsvWu7n3pvI2z/ezpW6/7XZUrd1WJAxxLleMws3JI2pRUeDwv\nIi7JnX2jwqzF2nkHEuD9wHmSNgPuBt5BasjnIknHAb8H3trmmMy6lqRDgNURcaOk3mbHH4srso1e\nHSxe8avKlbuqxAGOpcpxmFn75do15wK3R8TnC70uxzcqzFqqrQXIiFgEzKnTa/92xmE2juwDvFnS\nQcAk0jOQZwJTJU3MdyFnACtLjNHMzGy09gGOBhZLWpS7nUoqOPpGhVkLtfsOpJm1UUScApwCkO9A\n/r+IOErSd4DDSS2x+oqsmZl1tIi4DtAgvX2jwqyF2vYeSDOrlJNIDeosJT0TeW7J8ZiZmZlZB/Ad\nSLNxIiL6SO+/IiLuBvYqMx4zMzMz6zy+A2lmZmZmZmYNcQHSzMzMzMzMGuICpJmZWQtJmiTpekk3\nS7pV0sdz91mSFkpaKunC/EorMzOzjuJnIM3MzFrrSWC/iOjPLza/TtIPgBOBL0TEAklnA8cBZ5UZ\nqJmZtd/MOu/EXnb6wSVEMjK+A2lmZtZCkfTnv5vmTwD7ARfn7vOBw0oIz8zMbFRcgDQzM2sxSRPy\ny8xXA1cBvwPWRMTaPMgKYHpZ8ZmZmY2Uq7CamZm1WEQ8A+wuaSpwKfCiRseVdDxwPEBPTw99fX2j\njmfxykc36jZv9sbD9UyGebPXbtCtFfMfqf7+/lLnX8vxmJm5AGlmZjZmImKNpGuAVwJTJU3MdyFn\nACsHGecc4ByAOXPmRG9v76jjmFvneZt65s1eyxmLNzw1WHbU6Oc/Un19fbRi+VvF8ZiZuQBpZmbW\nUpJ2AJ7OhcfJwOuBzwDXAIcDC4BjgMvKi9LMzBrR6Q3ejAUXIM3MzFprGjBf0gRSWwMXRcQVkm4D\nFkj6FPAb4NwygzQzMxsJFyDNzMxaKCJuAfao0/1uYK/2R2RmZtY6boXVzMzMzMzMGuI7kGY2Zuo9\nN2BmZmZmncsFSDMzMzMzG5Ibk7EBrsJqZmZmZmZmDXEB0szMzMw6mqSvS1otaUmh22mSVkpalD8H\nlRmjWbdwAdKsi0maJOl6STdLulXSx3P3WZIWSloq6UJJm5Udq5mZ2Sh8EziwTvcvRMTu+fN/bY7J\nrCu5AGnW3Z4E9ouIlwG7AwdK2pv0UvMvRMQLgEeA40qM0czMbFQi4lrg4bLjMBsP3IiOWReLiAD6\n899N8yeA/YB/yN3nA6cBZ7U7PjMzszH2PklvB24A5kXEI/UGknQ8cDxAT08PfX197YuwSf39/aXE\nN2/22o26tTqOZpZt8cpHN+o2b3bj86o3n3rL2Oi4Q6ldrnasy7HkAqRZl5M0AbgReAHwFeB3wJqI\nGNh7rQCmDzLuqA6oje6I6/nSeZet+90zOf2fPX3rEU+vFco6aNfjWKobh5lVxlnAJ0kXTj8JnAEc\nW2/AiDgHOAdgzpw50dvb26YQm9fX10cZ8c2t1wrrUa2No5llqxdPM+rF3ug0m13u2uVqx7ocSy5A\nmnW5iHgG2F3SVOBS4EVNjDuqA+pod+4D5s1eyxmLJ5a+cy3roF2PY6luHGZWDRFx/8BvSf8NXFFi\nOGZdw89Amo0TEbEGuAZ4JTBV0sAFpBnAytICMzMzGwOSphX+vgVYMtiwZtY4FyDNupikHfKdRyRN\nBl4P3E4qSB6eBzsGuKz+FMzMzKpP0gXAL4FdJK2QdBzwH5IWS7oFeC3woVKDNOsSba/Cmp/HugFY\nGRGHSJoFLAC2Iz2ndXREPNXuuMy61DRgfs53mwAXRcQVkm4DFkj6FPAb4NwygzQzMxuNiDiyTmcf\n28zGQBnPQJ5AugPyrPx/4HUCCySdTXqdgFuDNGuBiLgF2KNO97uBvdofkZmZmZl1srZWYZU0AzgY\n+Fr+L9LrBC7Og8wHDmtnTGZmZmZmZtaYdt+B/CLwEWCr/H87Wvg6gf7+fubNfqbhYDqxuffx0Ez9\neFhGMzMzM7NO1LYCpKRDgNURcaOk3mbHb+R1An19fZxx3RMNT7PsVwKMxHhopn48LKOZ2ViZ2aLX\n55iZmdXTzjuQ+wBvlnQQMIn0DOSZ5NcJ5LuQfp2AmZmZmZlVVr0LdctOP3jQ4ebNXtuyd2NXQdue\ngYyIUyJiRkTMBI4AfhIRR+HXCZiZmZmZmXWEKrwH8iTgRElLSc9EusllMzMzMzOzCirjNR5ERB/Q\nl3/7dQJmZmZmZmYdoAp3IM3MzMzMzKwDuABpZmbWQpJ2lHSNpNsk3SrphNx9W0lXSborf29Tdqxm\nZmbNKqUKq5mZWRdbC8yLiJskbQXcKOkqYC5wdUScLulk4GRSOwBmZi3VaCuhZiPhO5BmZmYtFBGr\nIuKm/Ptx4HZgOnAoMD8PNh84rJwIzczMRs53IM3MzMaIpJnAHsBCoCciVuVe9wE9g4xzPHA8QE9P\nD319fU3Nc97stSMLFuiZvPH4zc6/lfr7+0udfy3HY2bmAqSZmdmYkLQl8F3ggxHxmKR1/SIiJEW9\n8SLiHOAcgDlz5kRvb29T8x3Ny6rnzV7LGYs3PDVYdlRz82+lvr4+ml3+seR4zMxchdXMzKzlJG1K\nKjyeFxGX5M73S5qW+08DVpcVn5mZ2Ui5AGlmZtZCSrcazwVuj4jPF3pdDhyTfx8DXNbu2MzMzEbL\nVVjNzMxaax/gaGCxpEW526nA6cBFko4Dfg+8taT4zMzMRswFSDMzsxaKiOsADdJ7/3bGYjZeSPo6\ncAiwOiJ2y922BS4EZgLLgLdGxCNlxWjWLVyANOtiknYEvkVq7TGAcyLizE49qNZ7rxX43VZmZsY3\ngS+TjnkDTsbvXjVrOT8DadbdBl5oviuwN/BeSbuy/qD6QuDq/N/MzKwjRcS1wMM1nf3uVbMx4DuQ\nZl0sv3NuVf79uKTiC81782DzgT58VdbMzLpLQ+9ehdG/f7WdGnn/Z733wTazTItXPlpnmhsP1+r1\n1My7TUfzzluoH/topjnU9Oq9Y7eR8avKBUizcaLTXmheNNyOt1073Sq9tNuxVDeObuOq49YNhnr3\nau4/qvevtlMj7/+s9z7YZt7p2uj7ZFv9nthm3m06mnfeQv3YRzPNoaZX7x27jYxfVS5Amo0DnfhC\n86Lhdrzt2ulW6aXdjqW6cZhZZdwvaVpErPK7V81ax89AmnU5v9DczMzGKb971WwMuABp1sX8QnMz\nMxsPJF0A/BLYRdKK/L7V04HXS7oLeF3+b2aj5CqsZt3NLzQ3M7OuFxFHDtLL717tAoM9i91N6i1j\nVZ81H9cFyE7aUGYj4Ream5mZmVkrjesCpJmZmY2ML8KamY1PfgbSzMzMzMzMGuICpJmZmZmZmTXE\nVVjNzMzMzGxcGQ8N84wV34E0MzMzMzOzhrgAaWZmZmZmZg1xFdYablXOzMzMzMysvrbdgZS0o6Rr\nJN0m6VZJJ+Tu20q6StJd+XubdsVkZmZmZmZmjWtnFda1wLyI2BXYG3ivpF2Bk4GrI+KFwNX5v5mZ\nmZmZmVVM26qwRsQqYFX+/bik24HpwKFAbx5sPtAHnNSuuMzMzKw1/BiIWXU5f1qrlPIMpKSZwB7A\nQqAnFy4B7gN6BhnneOB4gJ6eHvr6+jYapr+/n3mzn2l5vPXmVZb+/v5KxTMWxsMympmZmZl1orYX\nICVtCXwX+GBEPCZpXb+ICElRb7yIOAc4B2DOnDnR29u70TB9fX2ccd0TLY952VEbz6ssfX191Fv2\nbjIeltHMzMzMrBO19TUekjYlFR7Pi4hLcuf7JU3L/acBq9sZk5mZWStJ+rqk1ZKWFLq5wTgzM+sK\n7WyFVcC5wO0R8flCr8uBY/LvY4DL2hWTmZnZGPgmcGBNNzcYZ2ZmXaGdVVj3AY4GFktalLudCpwO\nXCTpOOD3wFvbGJOZmVlLRcS1+Vn/IjcYZ2Zdxw3zjE/tbIX1OkCD9N6/XXGYmZmVoKEG46CxRuMG\nLF756Ebd5s0eRZCTYd7stSOfQB2jaRStao2qOR4zs5JaYTWz9pD0deAQYHVE7Ja7bQtcCMwElgFv\njYhHyorRbLwZqsG43H/YRuMGzK1z9X805s1eyxmLW3tqMJqG6KrWqJrj6UySlgGPA88AayNiTrkR\nmXW2tjaiY2Zt9038LJZZFbjBOLNyvTYidnfh0Wz0XIA062IRcS3wcE3nQ0nPYJG/D2trUGbjkxuM\nMzOzruAqrA3wA8LWZcbkWax6WvUs1XDPZbXrGaAqPW/kWKobh6QLSA3mbC9pBfAx3GCcWZkCuDJX\nHf9qria+gdEe79pl8cpH6ZkMXzpv/TWo2dO33mi4Ro+/xelsOP7I4oOxee651c9mj4Wh4h7p8+VV\nTYcuQJqNY618FqueVj2fNdxzWaN5xqoZVXreyLFUN46IOHKQXm4wzqwc+0bESknPBq6SdEeuobPO\naI937TL35O9vdEysdwxs9fPRzRiL557LXJ5GDbUdRvp8ebvOb5rlAqTZ+HO/pGkRscrPYplZJ3LN\nIGtGRKzM36slXQrsBVw79FhmNhg/A2k2/vhZLDMzGxckTZG01cBv4ABgSblRmXU234E062J+FsvM\nzMa5HuBSSZDOe8+PiB+WG5JZZ3MBcoTqVZ8BV6Gxamnns1iD5QkzM7OyRMTdwMvKjmMooz2n7Lbj\nb7ctTzdyFVYzMzMzMzNriAuQZmZmZmZm1hBXYTWzjtcJLTJ2QoxmZmZmw/EdSDMzMzMzM2uI70Ca\nmZmZmZmNwnhq/Md3IM3MzMzMzKwhLkCamZmZmZlZQ1yF1czGjdE0ZDPz5O8zb/Za5tZMww3hmJmZ\n2XjiO5BmZmZmZmbWEN+BbAM3329mZmZmzTS0Mp4aZbHO4juQZmZmZmZm1hDfgTQzM7Mx02gtnHrD\nzZu9lt6S4jEzs/p8B9LMzMzMzMwa4gKkmZmZmZmZNcRVWM3MzKytxqJxkNFMc7Bxx0PVVlfpNbNm\nuQDZYm4xy2x8KTPPd8KJXyfEaGbWjEb3++N1Xzfa555dPKk+V2E1MzMzMzOzhlSiiC/pQOBMYALw\ntYg4veSQzLpet+e7Rq8Ql3mHbDR3Lwe7Stupd/yaWRedsDyD6fZ8NxaqdJe/mVZhOzUvQmfHXo/z\nnVlrlV6AlDQB+ArwemAF8GtJl0fEbeVG1n6NHCSHOniNpspAPWWdRLei2faxOPh10wHV+c6s/Zzv\nzNrP+c6s9apQhXUvYGlE3B0RTwELgENLjsms2znfmbWf851Z+znfmbWYIqLcAKTDgQMj4p35/9HA\nKyLifTXDHQ8cn//uAtxZZ3LbAw+OYbhV4GXsLDtFxA5lB1GrxfmuHaqSJqoSBziWegbicL4buaps\nywGOZ2hVisf5rn2qtN1brVuXbayWq5R8V3oV1kZFxDnAOUMNI+mGiJjTppBK4WW0dmok37VDVdJE\nVeIAx1LlOEarzHxXtXXoeIZWtXg6WVWOd43o5u3ercvWbctVhSqsK4EdC/9n5G5mNnac78zaz/nO\nrP2c78xarAoFyF8DL5Q0S9JmwBHA5SXHZNbtnO/M2s/5zqz9nO/MWqz0KqwRsVbS+4AfkZpX/npE\n3DrCyXVE1YNR8jLaqLU437VDVdJEVeIAx1JPVeKoq0PyXdXWoeMZWtXiqZwOyXfN6ubt3q3L1lXL\nVXojOmZmZmZmZtYZqlCF1czMzMzMzDqAC5BmZmZmZmbWkK4oQEo6UNKdkpZKOrnseBohaZmkxZIW\nSbohd9tW0lWS7srf2+TukvSfeflukbRnYTrH5OHvknRMofvL8/SX5nHVhmX6uqTVkpYUuo35Mg02\nD+sMrUo3LYplR0nXSLpN0q2STigjHkmTJF0v6eYcx8dz91mSFub5XZgbhEDS5vn/0tx/ZiviqIlp\ngqTfSLqizFhate8cj6qSvuvEVYm0lecxVdLFku6QdLukV5a5fiR9KG+rJZIuyPuG0taPjR3VPxa+\nTNIv8z7vfyU9q9DvlLyt75T0hkL3Sp0TN7NckmZK+lPevy+SdHZhnLaf1w6llftTDXLeW2kR0dEf\n0gPRvwN2BjYDbgZ2LTuuBuJeBmxf0+0/gJPz75OBz+TfBwE/AATsDSzM3bcF7s7f2+Tf2+R+1+dh\nlcd9YxuW6dXAnsCSdi7TYPPwpzM+rUg3LYxlGrBn/r0V8Ftg13bHk6e3Zf69KbAwT/8i4Ijc/Wzg\n3fn3e4Cz8+8jgAvHYDudCJwPXJH/lxILLdh3jtdPVdJ3VdNWnu584J3592bA1BL3R9OBe4DJhfUy\nt8z148/Yfah/LPw18Jr8+1jgk/n3rqTz3c2BWaTz4AlU8Jy4yeWaWRyuZjptP68dZrlasj9liPPe\nKn9KD6AFG/CVwI8K/08BTik7rgbiXsbGJ0F3AtPy72nAnfn3V4Eja4cDjgS+Wuj+1dxtGnBHofsG\nw43xcm2Q+duxTIPNw5/O+Yw23YxhXJcBry8zHmAL4CbgFcCDwMTcfd2+j9S64Cvz74l5OLUwhhnA\n1cB+wBX5AFhWLKPed7Y7fVf1U5H0XaW0tTWpwKaa7qWsH1IBcjnpxHJiXj9vKGv9+DP2HzY+Fj46\nsA1J77K8Lf/e4Fx3YNtT0XPiJpZrg+EKw5d2XtvEMo5of1q7LLXDVfXTDVVYB3awA1bkblUXwJWS\nbpR0fO7WExGr8u/7gJ78e7BlHKr7ijrdy9COZRpsHta5mk03LZerf+1BuvvX9nhytb5FwGrgKtJV\n5TURsbbOvNbFkfs/CmzXijiyLwIfAf6S/29XYiyt2HeOe2Wn74Iqpa1ZwAPAN3KV2q9JmkJJ6yci\nVgKfA+4FVpGW90bKWz/WfrcCh+bff0cqbEHz51BVM9hyAczK+e+nkl6Vu1XpvHYjo9yfdso220A3\nFCA71b4RsSfwRuC9kl5d7BnpMkSUEtkYaccydeN6G+/K2KaStgS+C3wwIh4rI56IeCYidifdodkL\neNFYz7MeSYcAqyPixjLmX8e423e2WhXSd46jamlrIqmq3VkRsQfwBKkK2jptXj/bkE6yZwHPBaYA\nB7Zj3lYZxwLvkXQjqZrkUyXH0yqDLdcq4Hk5/50InF987rOKqrI/bbduKECuZMMrFzNyt0rLVxaJ\niNXApaQTxPslTQPI36vz4IMt41DdZ9TpXoZ2LNNg87DO1Wy6aRlJm5IOBudFxCVlxxMRa4BrSFWT\npkqaWGde6+LI/bcGHmpRCPsAb5a0DFhAqmp4ZkmxtGrfOW5VLH1XKm2RrvyviIiF+f/FpAJlWevn\ndcA9EfFARDwNXEJaZ2WtH2uziLgjIg6IiJcDF5BqokDz51CVMthyRcSTEfFQ/n1j7v5XVOu8dp0W\n7U87YpvV6oYC5K+BF+ZWyTYjPTh+eckxDUnSFElbDfwGDgCWkOI+Jg92DKk+Nbn723MLTnsDj+bb\n4z8CDpC0Tb5SeQCp7vsq4DFJe+dWqt5emFa7tWOZBpuHda5m001L5LR1LnB7RHy+rHgk7SBpav49\nmfRcxe2kguThg8QxEN/hwE/ylc9Ri4hTImJGRMwk7V9/EhFHlRFLC/ed41JV0veAKqWtHM99wHJJ\nu+RO+wO3UV76uhfYW9IWedsNxFPK+rH2k/Ts/L0J8M+kRpMgbesjlFrenQW8kNTITEecEw+2XPnY\nNyH/3pm0XHdX7LyWHF+r9qd1z3vbshCjUcaDl63+kFo2+i3pSsVHy46ngXh3JrWMdTOpHvhHc/ft\nSI0J3AX8GNg2dxfwlbx8i4E5hWkdCyzNn3cUus8hnVj9DvgybXiQnnQVaRXwNOlK7nHtWKbB5uFP\nZ3xalW5aFMu+pOomtwCL8uegdscDvBT4TY5jCfCvufvOpJOEpcB3gM1z90n5/9Lcf+cx2la9rG8p\ns+2x0MJ953j8VCV9VzFtFeLYHbghr6PvkVpFLG39AB8H7sj7gW+TWt0sdT/gz9h8qH8sPIF0fvtb\n4HQK53LAR3Pau5NCi6RU7Jy4meUC/jbv2xeRGo97U2E6bT+vHWa5WrY/ZZDz3ip/BjaYmZmZmZmZ\n2ZC6oQqrmZmZmZmZtYELkGZmZmZmZtYQFyDNzMzMzMysIS5AmpmZmZmZWUNcgDQzMzMzM7OGuABp\nZmZmZmZmDXEB0szM7P/fKBgFo2AUjIJRMAqIAgABeSG3KFnFYgAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 1080x1296 with 20 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"df[numeric_columns].hist(bins=30)\n",
"fig = plt.gcf()\n",
"fig.set_size_inches((15, 18))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In order to effectively design feature representation and preprocess data, it is a good idea to find:\n",
"- similarly distributed coolumns,\n",
"- anomalous values in each column,\n",
"- similarity with famous probability distributions e.g., uniform, gaussian.\n",
"\n",
"See value distribution of categorical columns as well:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAfEAAAGMCAYAAADKjptiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSCQSiQlK6sQTiUQikZig/B+TPmFU0poJYAAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 576x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAfEAAAHBCAYAAAB5SILQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"<Figure size 576x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"for c in ['community_area', 'primary_property_type']:\n",
" (df[c].value_counts(normalize=True) * 100).plot(kind='bar', figsize=(6,4))\n",
" plt.title(c)\n",
" plt.ylabel('% of samples')\n",
"\n",
" ax = plt.gca()\n",
" ax.yaxis.grid(True)\n",
"\n",
" fig = plt.gcf()\n",
" fig.set_size_inches((8, 4))\n",
" \n",
" plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We can expect ML model and prediction results would be biased against buildings located in `LOOP` area or used as `Office`."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Step 4: Cleanse data and create feature vectors based on the above observations\n",
"\n",
"Convert potentially useful columns into categorical/quantitative features, and concatenate them to single feature \"vector\" (i.e., array):\n",
"\n",
"```sql\n",
"select\n",
" id,\n",
" array_concat(\n",
" categorical_features(\n",
" array('Chicago community area', 'Primary use of property'),\n",
" community_area, primary_property_type\n",
" ),\n",
" quantitative_features(\n",
" array('Total interior floor space', 'Building age', 'Number of buildings'),\n",
" gross_floor_area___buildings__sq_ft_, data_year - year_built, num_of_buildings\n",
" )\n",
" ) as features,\n",
" electricity_use__kbtu_ as annual_electricity_consumption\n",
"from\n",
" energy_benchmarking\n",
"where\n",
" electricity_use__kbtu_ is not null\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Step 5 and 6: Build ML model and evaluation\n",
"\n",
"See workflow definition `mbed_XX`:\n",
"\n",
"```yml\n",
"_export:\n",
" !include : config.yml\n",
" td:\n",
" database: ${database}\n",
" engine: hive\n",
"\n",
"+vectorize:\n",
" td>: queries/vectorize.sql\n",
" create_table: vectors\n",
"\n",
"+shuffle:\n",
" td>: queries/shuffle.sql\n",
" create_table: samples\n",
" engine: presto\n",
"\n",
"+train:\n",
" td>: queries/train.sql\n",
" create_table: model\n",
"\n",
"+evaluate:\n",
" td>: queries/evaluate.sql\n",
" store_last_results: true\n",
"\n",
"+show_accuracy:\n",
" echo>: \"RMSE: ${td.last_results.rmse}, MAE: ${td.last_results.mae}\"\n",
"```\n",
"\n",
"#### `+shuffle`\n",
"\n",
"Add `random_flag` to the output from Step 4 for the sake of 8:2 train and test splitting:\n",
"\n",
"```sql\n",
"select\n",
" *,\n",
" abs(from_big_endian_64(xxhash64(to_utf8(cast(id as varchar)))) % 100) as random_flag\n",
"from\n",
" vectors\n",
"```\n",
"\n",
"#### `+train`\n",
"\n",
"On the 80% training data, build linear regressor that predicts annual electricity consumption at a given building:\n",
"\n",
"```sql\n",
"select\n",
" feature,\n",
" avg(weight) as weight\n",
"from (\n",
" select\n",
" train_regressor(\n",
" features,\n",
" annual_electricity_consumption,\n",
" '-loss_function squared -optimizer AdaGrad -regularization l1' -- hyper-parameters\n",
" ) as (feature, weight)\n",
" from (\n",
" select features, annual_electricity_consumption\n",
" from samples\n",
" where random_flag < 80\n",
" CLUSTER BY rand(1) -- random shuffling\n",
" ) t1\n",
") t2\n",
"group by\n",
" feature\n",
"```\n",
"\n",
"#### `+evaluate`\n",
"\n",
"Predict electricity usage at the 20% test buildings, and compare the predicted values with actual consumption:\n",
"\n",
"```sql\n",
"with features_exploded as (\n",
" select\n",
" id,\n",
" extract_feature(fv) as feature,\n",
" extract_weight(fv) as value\n",
" from (\n",
" select * from samples where random_flag >= 80\n",
" ) t1\n",
" LATERAL VIEW explode(features) t2 as fv\n",
"),\n",
"prediction as (\n",
" select\n",
" t1.annual_electricity_consumption,\n",
" t2.predicted_electricity_consumption\n",
" from\n",
" samples t1\n",
" join (\n",
" select\n",
" t1.id,\n",
" sum(p1.weight * t1.value) as predicted_electricity_consumption\n",
" from\n",
" features_exploded t1\n",
" LEFT OUTER JOIN model p1 ON (t1.feature = p1.feature)\n",
" group by\n",
" t1.id\n",
" ) t2\n",
" on t1.id = t2.id\n",
")\n",
"select\n",
" mae(predicted_electricity_consumption, annual_electricity_consumption) as mae,\n",
" rmse(predicted_electricity_consumption, annual_electricity_consumption) as rmse\n",
"from\n",
" prediction\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Read model table and visually check if the model clearly captures characteristics of data:"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>feature</th>\n",
" <th>weight</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>Building age</td>\n",
" <td>2.275155</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>Primary use of property#Office</td>\n",
" <td>1.899472</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>Chicago community area#LOOP</td>\n",
" <td>1.690781</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>Total interior floor space</td>\n",
" <td>1.464106</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>Number of buildings</td>\n",
" <td>1.421706</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" feature weight\n",
"0 Building age 2.275155\n",
"1 Primary use of property#Office 1.899472\n",
"2 Chicago community area#LOOP 1.690781\n",
"3 Total interior floor space 1.464106\n",
"4 Number of buildings 1.421706"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_model = td.read_td(\n",
" 'select feature, weight from takuti.model order by abs(weight) desc', presto)\n",
"df_model.head()"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAABLAAAAH2CAYAAABgGHUZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nNoIC+AAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 864x576 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"n = 10\n",
"\n",
"df_model_top = df_model.head(n)\n",
"x = range(n)\n",
"plt.barh(x, df_model_top['weight'].values, align='center')\n",
"plt.xlabel('Model weight', fontsize=14)\n",
"plt.yticks(x, df_model_top['feature'].values, fontsize=16)\n",
"\n",
"plt.title('Top-%d important features to predict electricity usage' % n, fontsize=16)\n",
"\n",
"ax = plt.gca()\n",
"ax.xaxis.grid(True)\n",
"ax.invert_yaxis()\n",
"\n",
"fig = plt.gcf()\n",
"fig.set_size_inches((12, 8))\n",
" \n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Step 7: Predict unknown \"NULL\" energy consumption\n",
"\n",
"Even though electricity usage of some buildings is still not benchmarked, we can now predict their potential energy consumption by using the ML model:\n",
"\n",
"```sql\n",
"with null_samples as (\n",
" select\n",
" id,\n",
" array_concat(\n",
" categorical_features(\n",
" array('Chicago community area', 'Primary use of property'),\n",
" community_area, primary_property_type\n",
" ),\n",
" quantitative_features(\n",
" array('Total interior floor space', 'Building age', 'Number of buildings'),\n",
" gross_floor_area___buildings__sq_ft_, data_year - year_built, num_of_buildings\n",
" )\n",
" ) as features\n",
" from\n",
" energy_benchmarking\n",
" where\n",
" electricity_use__kbtu_ is null\n",
"),\n",
"features_exploded as (\n",
" select \n",
" id,\n",
" extract_feature(fv) as feature,\n",
" extract_weight(fv) as value\n",
" from null_samples t1\n",
" LATERAL VIEW explode(features) t2 as fv\n",
")\n",
"select\n",
" t1.id,\n",
" sum(p1.weight * t1.value) as predicted_electricity_consumption\n",
"from\n",
" features_exploded t1\n",
" LEFT OUTER JOIN model p1 ON (t1.feature = p1.feature)\n",
"group by\n",
" t1.id\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"## What's next?\n",
"\n",
"Play with the other relevant data stored in `chicago_smart_green` database:\n",
"\n",
"- `community_areas`\n",
" - Definition of geographical boundaries of community areas.\n",
"- `environmental_comiplaints`\n",
" - List of complaints reported to the local government.\n",
"- `sensors_historical`\n",
" - Environmental monitoring results from many sensors deployed in city.\n",
" \n",
"For example, you can find corresponding community area to lat/lon geopraphical point posted by the sensors:\n",
"\n",
"```sql\n",
"with data_streams as (\n",
" select\n",
" data_stream_id, latitude, longitude\n",
" from \n",
" sensors_historical\n",
" group by \n",
" 1, 2, 3\n",
")\n",
"select\n",
" t1.*,\n",
" t2.community\n",
"from \n",
" data_streams t1 \n",
"cross join \n",
" community_areas t2\n",
"where\n",
" ST_Contains(\n",
" ST_GeometryFromText(t2.the_geom),\n",
" ST_Point(t1.longitude, t1.latitude)\n",
" )\n",
"```\n",
"\n",
"Eventually, it is possible to enrich the original `energy_menchmarking` data with sensors, and we may get a new finding like \"higher temperature area has more energy consumption.\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"anaconda-cloud": {},
"kernelspec": {
"display_name": "Python [conda root]",
"language": "python",
"name": "conda-root-py"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.5.2"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment