{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7747386150504332\n",
      "5207\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(['House', 'Condominium', 'Townhouse', 'Apartment', 'Other', 'Loft',\n",
       "       'Guest suite', 'Guesthouse', 'In-law', 'Boutique hotel', 'Boat',\n",
       "       'Hostel', 'Bed & Breakfast', 'Dorm', 'Timeshare',\n",
       "       'Serviced apartment', 'Vacation home', 'Bungalow', 'Villa'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import random\n",
    "print(random.uniform(0, 1))\n",
    "df = pd.read_csv('Airbnb_data/listings_SanFan.csv')\n",
    "print(len(df))\n",
    "len(df['last_scraped'].unique())\n",
    "# s = pd.Series(np.random.randn())\n",
    "# a = pd.DataFrame(np.random.randn(0,1), columns=list('Occupancy Rate'))\n",
    "# df['Occupancy Rate'] = random.uniform(0, 1)\n",
    "df.head()\n",
    "df['property_type'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>accommodates</th>\n",
       "      <th>host_response_time</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>beds</th>\n",
       "      <th>security_deposit</th>\n",
       "      <th>cleaning_fee</th>\n",
       "      <th>guests_included</th>\n",
       "      <th>extra_people</th>\n",
       "      <th>...</th>\n",
       "      <th>review_scores_checkin</th>\n",
       "      <th>review_scores_communication</th>\n",
       "      <th>review_scores_location</th>\n",
       "      <th>review_scores_value</th>\n",
       "      <th>house_rules</th>\n",
       "      <th>amenities</th>\n",
       "      <th>bed_type</th>\n",
       "      <th>room_type</th>\n",
       "      <th>cancellation_policy</th>\n",
       "      <th>property_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>$56.00</td>\n",
       "      <td>4</td>\n",
       "      <td>within an hour</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>$100.00</td>\n",
       "      <td>$30.00</td>\n",
       "      <td>2</td>\n",
       "      <td>$15.00</td>\n",
       "      <td>...</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>- Private bathroom is on the first floor (slig...</td>\n",
       "      <td>{TV,\"Cable TV\",\"Wireless Internet\",\"Air condit...</td>\n",
       "      <td>Real Bed</td>\n",
       "      <td>Private room</td>\n",
       "      <td>moderate</td>\n",
       "      <td>House</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>$36.00</td>\n",
       "      <td>2</td>\n",
       "      <td>within an hour</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>$100.00</td>\n",
       "      <td>$35.00</td>\n",
       "      <td>1</td>\n",
       "      <td>$20.00</td>\n",
       "      <td>...</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1. NO SMOKING! This rule applies for inside ou...</td>\n",
       "      <td>{TV,\"Wireless Internet\",\"Air conditioning\",Kit...</td>\n",
       "      <td>Real Bed</td>\n",
       "      <td>Private room</td>\n",
       "      <td>strict</td>\n",
       "      <td>Condominium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>$80.00</td>\n",
       "      <td>6</td>\n",
       "      <td>within an hour</td>\n",
       "      <td>1.5</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>$150.00</td>\n",
       "      <td>$65.00</td>\n",
       "      <td>4</td>\n",
       "      <td>$15.00</td>\n",
       "      <td>...</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>I don't ask much, just to please respect the s...</td>\n",
       "      <td>{TV,\"Cable TV\",Internet,\"Wireless Internet\",\"A...</td>\n",
       "      <td>Real Bed</td>\n",
       "      <td>Private room</td>\n",
       "      <td>moderate</td>\n",
       "      <td>Townhouse</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>$80.00</td>\n",
       "      <td>2</td>\n",
       "      <td>within a few hours</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>$100.00</td>\n",
       "      <td>$10.00</td>\n",
       "      <td>2</td>\n",
       "      <td>$10.00</td>\n",
       "      <td>...</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Guests should treat my condo as their home wit...</td>\n",
       "      <td>{TV,\"Cable TV\",Internet,\"Wireless Internet\",\"A...</td>\n",
       "      <td>Real Bed</td>\n",
       "      <td>Private room</td>\n",
       "      <td>flexible</td>\n",
       "      <td>Condominium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>$20.00</td>\n",
       "      <td>3</td>\n",
       "      <td>within an hour</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>$100.00</td>\n",
       "      <td>$50.00</td>\n",
       "      <td>1</td>\n",
       "      <td>$10.00</td>\n",
       "      <td>...</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>If using the kitchen please clean up after you...</td>\n",
       "      <td>{TV,\"Cable TV\",Internet,\"Wireless Internet\",\"A...</td>\n",
       "      <td>Real Bed</td>\n",
       "      <td>Private room</td>\n",
       "      <td>strict</td>\n",
       "      <td>Apartment</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    price  accommodates  host_response_time  bathrooms  bedrooms  beds  \\\n",
       "0  $56.00             4      within an hour        1.0       1.0   2.0   \n",
       "1  $36.00             2      within an hour        1.0       1.0   1.0   \n",
       "2  $80.00             6      within an hour        1.5       2.0   3.0   \n",
       "3  $80.00             2  within a few hours        1.0       1.0   1.0   \n",
       "4  $20.00             3      within an hour        1.0       1.0   1.0   \n",
       "\n",
       "  security_deposit cleaning_fee  guests_included extra_people      ...        \\\n",
       "0          $100.00       $30.00                2       $15.00      ...         \n",
       "1          $100.00       $35.00                1       $20.00      ...         \n",
       "2          $150.00       $65.00                4       $15.00      ...         \n",
       "3          $100.00       $10.00                2       $10.00      ...         \n",
       "4          $100.00       $50.00                1       $10.00      ...         \n",
       "\n",
       "   review_scores_checkin  review_scores_communication  review_scores_location  \\\n",
       "0                   10.0                         10.0                    10.0   \n",
       "1                   10.0                         10.0                    10.0   \n",
       "2                   10.0                         10.0                    10.0   \n",
       "3                   10.0                         10.0                    10.0   \n",
       "4                   10.0                         10.0                    10.0   \n",
       "\n",
       "   review_scores_value                                        house_rules  \\\n",
       "0                 10.0  - Private bathroom is on the first floor (slig...   \n",
       "1                 10.0  1. NO SMOKING! This rule applies for inside ou...   \n",
       "2                 10.0  I don't ask much, just to please respect the s...   \n",
       "3                 10.0  Guests should treat my condo as their home wit...   \n",
       "4                  9.0  If using the kitchen please clean up after you...   \n",
       "\n",
       "                                           amenities  bed_type     room_type  \\\n",
       "0  {TV,\"Cable TV\",\"Wireless Internet\",\"Air condit...  Real Bed  Private room   \n",
       "1  {TV,\"Wireless Internet\",\"Air conditioning\",Kit...  Real Bed  Private room   \n",
       "2  {TV,\"Cable TV\",Internet,\"Wireless Internet\",\"A...  Real Bed  Private room   \n",
       "3  {TV,\"Cable TV\",Internet,\"Wireless Internet\",\"A...  Real Bed  Private room   \n",
       "4  {TV,\"Cable TV\",Internet,\"Wireless Internet\",\"A...  Real Bed  Private room   \n",
       "\n",
       "   cancellation_policy  property_type  \n",
       "0             moderate          House  \n",
       "1               strict    Condominium  \n",
       "2             moderate      Townhouse  \n",
       "3             flexible    Condominium  \n",
       "4               strict      Apartment  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from scipy.stats import norm\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from scipy import stats\n",
    "import warnings\n",
    "%matplotlib inline\n",
    "df = pd.read_csv('Airbnb_data/listings_SanFan.csv')\n",
    "\n",
    "# u'zipcode',u'location_price',,u'instant_bookable', u'host_is_superhost',u'host_response_rate',\n",
    "selected_features = [u'price',u'accommodates',u'host_response_time',\n",
    "       u'bathrooms', u'bedrooms', u'beds',u'security_deposit', u'cleaning_fee', u'guests_included',\n",
    "       u'extra_people', u'minimum_nights', u'maximum_nights',u'guests_included',  \n",
    "       u'availability_365',\n",
    "       u'number_of_reviews', u'review_scores_rating',u'review_scores_cleanliness', u'review_scores_checkin',\n",
    "       u'review_scores_communication', u'review_scores_location',\n",
    "       u'review_scores_value', u'house_rules',u'amenities','bed_type', 'room_type', 'cancellation_policy', 'property_type']\n",
    "df = df.loc[:, selected_features]\n",
    "df = df.apply(lambda x:x.fillna(x.value_counts().index[0]))\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df['review_scores_rating'].describe()\n",
    "# df['zipcode'].unique()\n",
    "# def clean_zipcode(row):\n",
    "#     return row[:7]\n",
    "# df['zipcode'] = df.apply(clean_zipcode)\n",
    "# df['zipcode'].unique()\n",
    "\n",
    "# pd.to_numeric(df['zipcode'], errors = 'coerce')\n",
    "# df['zipcode'][df['zipcode'] == '60660-1448']\n",
    "# df['zipcode'] = df['zipcode'].apply(lambda x: str(x)[:6])\n",
    "# df['zipcode'].unique()\n",
    "# df['zipcode'].astype(float)\n",
    "# sns.distplot(df['zipcode'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# FEES and PRICES\n",
    "df['price'] = df['price'].str.replace(\"\\$|,\", \"\").astype(float)\n",
    "df['security_deposit'] = df['security_deposit'].str.replace(\"\\$|,\", \"\").astype(float)\n",
    "df['cleaning_fee'] = df['cleaning_fee'].str.replace(\"\\$|,\", \"\").astype(float)\n",
    "df['extra_people'] = df['extra_people'].str.replace(\"\\$|,\", \"\").astype(float)\n",
    "df['availability'] = df['availability_365'] / 365"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove rows that have 'NaN'in key features\n",
    "# solve NaN cells in unimportant attributes\n",
    "\n",
    "# remove_criteria = df['price'].isnull() | df['zipcode'].isnull()\n",
    "# df = df[-remove_criteria]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # APARTMENT TYPES\n",
    "# enc = LabelEncoder()\n",
    "# room_type = enc.fit( df['room_type'].values )\n",
    "# df['room_type'] = room_type.transform(df['room_type'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# HOUSE RULES\n",
    "house_rules = df['house_rules'].str.lower()\n",
    "\n",
    "smoking = house_rules.str.contains(\"smoke|smoking\", na= False)\n",
    "df.loc[:, 'smoking'] = - smoking # False: No smoking allowed\n",
    "\n",
    "pet = house_rules.str.contains(\"pet\", na=False)\n",
    "df.loc[:, 'pet'] = - pet\n",
    "\n",
    "party = house_rules.str.contains(\"party|parties\", na=False)\n",
    "df.loc[:, 'party'] = - party\n",
    "\n",
    "guest = house_rules.str.contains(\"guest|guests\", na=False)\n",
    "df.loc[:, 'guest'] = - guest\n",
    "\n",
    "df = df.drop(['house_rules'], axis = 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "import re\n",
    "# AMENITIES\n",
    "amenities = list(df['amenities'])\n",
    "total = ','.join(amenities)\n",
    "total = total.replace(\"{\", \"\").replace(\"}\",\"\").replace(\"\\\"\", \"\").split(\",\")\n",
    "amenity_items = list(set(total))\n",
    "amenity_items = list(filter(None, amenity_items))\n",
    "for item in amenity_items:\n",
    "    if re.match(r'translation',item):\n",
    "        amenity_items.remove(item)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:10: FutureWarning: set_value is deprecated and will be removed in a future release. Please use .at[] or .iat[] accessors instead\n",
      "  # Remove the CWD from sys.path while we load stuff.\n"
     ]
    }
   ],
   "source": [
    "# Turn Amenities into OneHotEncoder\n",
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "\n",
    "amenities = list(df['amenities'])\n",
    "new_table = pd.DataFrame(index = df.reset_index().values[:,0], columns = amenity_items).fillna(0)\n",
    "\n",
    "for i in range(len(amenities)):\n",
    "    for item in amenity_items:\n",
    "        if item in amenities[i]:\n",
    "            new_table.set_value(i, item, 1)\n",
    "\n",
    "sum_table = np.array(new_table.sum())\n",
    "ind = (-sum_table).argsort()[:60]\n",
    "common_amenities = list(new_table.sum().iloc[ind].index)\n",
    "df = df.drop(['amenities'], axis = 1)\n",
    "df = pd.concat([df, new_table[common_amenities]], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:10: FutureWarning: set_value is deprecated and will be removed in a future release. Please use .at[] or .iat[] accessors instead\n",
      "  # Remove the CWD from sys.path while we load stuff.\n"
     ]
    }
   ],
   "source": [
    "columns = ['bed_type', 'room_type', 'cancellation_policy', 'property_type', 'host_response_time']\n",
    "for column in columns:  \n",
    "    unique_values = list(df[column].unique())\n",
    "    column_list = list(df[column])\n",
    "    new_table = pd.DataFrame(index = df.reset_index().values[:,0], columns = unique_values).fillna(0)\n",
    "    \n",
    "    for i in range(len( column_list )):\n",
    "        for item in unique_values:\n",
    "            if item in column_list[i]:\n",
    "                new_table.set_value(i, item, 1)  \n",
    "    df = pd.concat( [df, new_table], axis = 1)\n",
    "    df = df.drop([column], axis = 1)         \n",
    "# df.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>accommodates</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>beds</th>\n",
       "      <th>security_deposit</th>\n",
       "      <th>cleaning_fee</th>\n",
       "      <th>guests_included</th>\n",
       "      <th>extra_people</th>\n",
       "      <th>minimum_nights</th>\n",
       "      <th>...</th>\n",
       "      <th>Dorm</th>\n",
       "      <th>Timeshare</th>\n",
       "      <th>Serviced apartment</th>\n",
       "      <th>Vacation home</th>\n",
       "      <th>Bungalow</th>\n",
       "      <th>Villa</th>\n",
       "      <th>within an hour</th>\n",
       "      <th>within a few hours</th>\n",
       "      <th>within a day</th>\n",
       "      <th>a few days or more</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56.0</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>2</td>\n",
       "      <td>15.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>36.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>20.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>80.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>4</td>\n",
       "      <td>15.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>80.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>20.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>1</td>\n",
       "      <td>10.0</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 121 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   price  accommodates  bathrooms  bedrooms  beds  security_deposit  \\\n",
       "0   56.0             4        1.0       1.0   2.0             100.0   \n",
       "1   36.0             2        1.0       1.0   1.0             100.0   \n",
       "2   80.0             6        1.5       2.0   3.0             150.0   \n",
       "3   80.0             2        1.0       1.0   1.0             100.0   \n",
       "4   20.0             3        1.0       1.0   1.0             100.0   \n",
       "\n",
       "   cleaning_fee  guests_included  extra_people  minimum_nights  \\\n",
       "0          30.0                2          15.0               2   \n",
       "1          35.0                1          20.0               1   \n",
       "2          65.0                4          15.0               1   \n",
       "3          10.0                2          10.0               2   \n",
       "4          50.0                1          10.0               3   \n",
       "\n",
       "          ...          Dorm  Timeshare  Serviced apartment  Vacation home  \\\n",
       "0         ...             0          0                   0              0   \n",
       "1         ...             0          0                   0              0   \n",
       "2         ...             0          0                   0              0   \n",
       "3         ...             0          0                   0              0   \n",
       "4         ...             0          0                   0              0   \n",
       "\n",
       "   Bungalow  Villa  within an hour  within a few hours  within a day  \\\n",
       "0         0      0               1                   0             0   \n",
       "1         0      0               1                   0             0   \n",
       "2         0      0               1                   0             0   \n",
       "3         0      0               0                   1             0   \n",
       "4         0      0               1                   0             0   \n",
       "\n",
       "   a few days or more  \n",
       "0                   0  \n",
       "1                   0  \n",
       "2                   0  \n",
       "3                   0  \n",
       "4                   0  \n",
       "\n",
       "[5 rows x 121 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# df.dtypes\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import pysal\n",
    "# from pysal.cg.kdtree import KDTree    \n",
    "\n",
    "# # locations = np.array(df[['longitude','latitude']])\n",
    "# tree = KDTree(locations, distance_metric='Arc', radius=pysal.cg.RADIUS_EARTH_MILES)\n",
    "# current_point = (-87.68493431,41.97654639)\n",
    "# # get all points within 4 mile of 'current_point'\n",
    "# indices = tree.query_ball_point(current_point, 3)\n",
    "# print(len(indices))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.06955685 0.06390113 0.03772549 0.0358486  0.02982785 0.0243029\n",
      " 0.02028904 0.0195189  0.0178268  0.01724169 0.01602302 0.01572053\n",
      " 0.01439182 0.01386771 0.01359957 0.01262482 0.01250612 0.01209813\n",
      " 0.01165445 0.01149893 0.01054338 0.01051744 0.0100191  0.00990338\n",
      " 0.00971958 0.00954385 0.00937708 0.00927741 0.00915474 0.00902109\n",
      " 0.00890188 0.00879662 0.00874394 0.00865802 0.00854619 0.00849935\n",
      " 0.00844427 0.00841201 0.00839733 0.0083103  0.00828395 0.00819403\n",
      " 0.00815511 0.00809037 0.00806534 0.00787964 0.00784945 0.00775434\n",
      " 0.00767163 0.00752333]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/utils/validation.py:590: DataConversionWarning: Data with input dtype object was converted to float64 by the scale function.\n",
      "  warnings.warn(msg, DataConversionWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[ 0.26141767,  0.14568472, -0.6375105 , ..., -1.17693586,\n",
       "        -0.46384047,  0.8320414 ],\n",
       "       [-0.22720222,  0.27145878,  0.89655499, ..., -0.93653801,\n",
       "         0.27048126, -0.1362913 ],\n",
       "       [ 2.80365383, -2.91127643, -1.9686958 , ..., -1.95706777,\n",
       "         0.36587932, -1.71600123],\n",
       "       ...,\n",
       "       [-0.48487174,  0.6425372 , -0.01368156, ..., -0.5314829 ,\n",
       "         0.37168929,  0.14531443],\n",
       "       [-1.39981328,  1.16749192, -2.57837329, ...,  1.47844786,\n",
       "         2.62637268, -1.18643124],\n",
       "       [-0.66715214, -0.02972972,  1.72968258, ...,  0.25366963,\n",
       "         0.02767576,  0.42830624]])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn import preprocessing\n",
    "from sklearn.decomposition import PCA\n",
    "# df.drop(['longitude','latitude'], axis=1, inplace=True)\n",
    "X = np.array(df.drop(['price'], axis = 1))\n",
    "X_scaled = preprocessing.scale(X)\n",
    "y =df['price']\n",
    "pca = PCA(n_components=50)\n",
    "X_scaled = pca.fit_transform(X_scaled)\n",
    "print(pca.explained_variance_ratio_)\n",
    "plt.plot(np.cumsum(pca.explained_variance_ratio_))\n",
    "plt.title('Principal Component Analysis to reduce 121 to 50 variables')\n",
    "plt.xlabel('number of components')\n",
    "plt.ylabel('cumulative explained variance')\n",
    "X_scaled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 1 0 ... 1 1 1]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEWCAYAAAB42tAoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJzsnXeYHMXxsN+aDZclnU4ZFEGAiAIkEFkiChGEyZhsMJhgA8bYGPgwwWBsbBN+BkzOwYBJJguEkAABkkAoAAIJFFEOl8PuTn1/9Jy0d7e7t3e34UK/z7PPzs70TNfMznRNd1VXiapisVgsFks8nGwLYLFYLJb2jVUUFovFYkmIVRQWi8ViSYhVFBaLxWJJiFUUFovFYkmIVRQWi8ViSUiXVhQislhEDs22HOlGRG4QkaeyLQeAiFSIyLBsy2FpiIj0FZGpIlIuIv/ItjyNac/PqojMF5GxSZZVEdk2zrZzROSjlAqXIrq0okiEd2PWiUivRutne3/2kOxIFhsR+bmIzPQa4pUi8paI7J/C4w/xztvfluOoaqGq/pAqudKBiEwRkRrvWq4TkZdEpH/U9r1E5E0R2SQiG0TkcxE5t9ExhoqIKyL3tqDeuI1IBrgAWAd0U9UrsyRDqxCR/4rIA43WvSIi/8pE/aq6k6pOyURd2cIqisT8CJxW/0NEdgHysidObETkt8CdwK1AX2AQcC8wMZtyRdNWBZMFLlXVQmA7oAdwB4CI7ANMBj4EtgVKgIuAIxvtfxawEThVRHJSIVCar+Fg4GttxQzcdvDfXgKcICLjPHlOAXYHrk5npe3gvDOHqnbZD7AYONRb3gGjGE6N2nYdMCOq/N+BawEFhnjrcrz1S4HVwL+BPG9bMfA6sBbTaLwObB11vCnAzcDHQDnwLtDL25YLPAWsBzYBM4C+Mc6hO1ABnJTgPG8AnvKWxwLLE1yHvYCZQJl3Pv/01i/1zrvC++zjrf8F8I13fu8Ag6OOq5iH+Hvgx6h123rLjwH3AG945/8ZsE3U/ocDC4BSjOL7EDg/zjnmYJTlT97nTiAn+pyBK4E1wErg3ATXa0p0Pd45zPOWPwLuSeLeWoRRIKuBE5MoP9W7NpXe9T0lSu4/AKuAJ9NxT3n/Qwio8+o+NMnrGS1X/brfR13j44AJwHfABuCaKDkdTEO+yJPneaBn1PYzgSXetmuJukfjXL9zgIWYl6TVwPg45QYA1Y3q2h3TmwoA22BeBNZ7654GejR6Vv4AzAFqAT9Nn5/p3vVdCfwLCDZ6Jn4D/OAd/3bAiTqHj6LK7gBM8q7dAuDkqG0TgK+9/3gF8Lu0tpXpPHh7/9T/wcAemIbw6BjbFgAjAB+wDPPmFa0o7gReA3oCRcD/gL9420qAE4B8b9sLwCuNHupFmLfWPO/3bd62C71j5Xt174kZFmh8DuOBMOBPcJ43kLyimA6c6S0XAmO85SHeefuj9jsO83CO8B6Y64BPGj0Uk7xrkxe1LlpRbPAeLj/moXzO29YLo6yO97ZdhmnM4imKm4BPgT5Ab+AT4Oaocw57ZQLeQ1YFFMc51pT6ejw5JmMaw3wgAoxr5r46ANOIFAP/B7yW5P24+do0kvuvmIY7L133lPdf/LmF1zNarvp113vX+JcYZfaMJ+dOQA0wzDvG5d7xt/aOcT/wrLdtR4zCOtDb9k/v2HEVhbffO5jG9/Fmyk0Gfhn1+3bg397ytsBhXr29MQr8zkbPymxgIFvu6cVseX72BMZg7tkhmJeoyxv9xx9gnolBGCVaf6+dg6cogAJMe3Oud6w9vHPbydu+EjjAWy4G9khrW5nOg7f3j/cH34h5ExoXY9uhmMbvL5gGeZL3p6l3EwjmDTD6LXgfvLfnGPWNBDZG/Z4CXBf1+2LgbW/5F5iHc9dmzuF0YFUzZW4geUUx1bsmvRqVGUJTRfEWcF7UbwfTAA/2fitwcKPjNFYUD0VtmwB86y2fBUyP2ibegxNPUSwCJkT9PgJYHHXO1Y1kX4OnBGMca4p3Hpswb2tPYxqNrTz5d2jmej+E13h790MI6JPE/RhLUdQBuQn2Sck9RVNF0dz1bCBX1DX2eb+LvPPZO6rMLOA4b/kb4JCobf296+THKJvnorYVePU1pyiu8+o8rJly5wOTG91XB8YpexzwZaNn5Rfxnp8Y+18OvNzoPx4f9fti4H1v+Ry2KIpTgGmNjnU/8CdveSlG8Td5eUzHx9oo4FeYt+AP4mx/Evg55k98otG23pi3s1meYXMT8La3HhHJF5H7RWSJiJRhGuEeIuKLOsaqqOUqzFt8fb3vAM+JyE8i8jcRCcSQbz3QK4Xjpedh3ka/FZEZInJ0grKDgbuizn0D5sHbKqrMsmbqi3f+A6L3VfN0LE9wnAGYoYp6lnjr6lmvquE4dcXiN6raQ1W3UtXTVbV+qMfFNGoxEZE84CSMckFVp2Me6p8nqCsRa1W1Jur4mbinoPnr2UAuj/WqGvGWq73v1VHbq6NkGQy8HHXvfIPprfWl6X9fibnP4yIiw4HfYYYo/5HgvABeBPYRkQGYXosC07zj9BGR50RkhXd9n8L0KqOJe0+LyHYi8rqIrPL2v7WZ/Rtf13oGA3vXXx/vGp0O9PO2n4B5sVoiIh96trO0YRWFURSDROSOWBtVdQnGdjEBeKnR5nWYm38nr1Hpoard1RhBwYyJb495q+qGuSnBNKYJUdWQqt6oqjsC+wJHY96yGzMd06U/rrljelRilJsRxDQwvaPq/V5VT8MMOfwVeFFECjAPU2OWARdGnXsPVc1T1U+iTyVJuRqzEjMsUS+nRP+OwU+Yh6ueQd66lKGqVZjrfUKCYj8DugH3eo3FKozijPXfJVVto9+ZuKeg+evZ2v+1nmXAkY3unVxVXYH57wfWFxSRfMyQW0y8e+MhzDDwrzH3+B/ilVfVTRjbzckYBf6s9yICZvRAMb2ubsAZNL22ic79PuBbYLi3/zUx9h8YtRzvPl0GfNjo+hSq6kXeOcxQ1YmY5/QVjI0nbVhFYYxB44EDReS2OGXOwwyhVEavVFUXeBC4Q0T6AIjIViJyhFekCKNINolIT+BPyQolIuNEZBevIS/DdMsjjcupaimmq36PiBznvXEGRORIEflbjEN/B+SKyFHeW9d1mPHY+nrPEJHe3rlt8lZHMOPNLhA9B+LfwB9FZCdv3+4iclKy59gMbwC7eOfkxxiU+yUo/yxwnYj09lyar8e8Daaa3wPniMhVIlICICK7ichz3vazgUeAXTDDQiOB/YCRntdcIlbT8PrGIu33lEe6r+e/gVtEZLAnW28RqffSexE4WkT2F5Egxl6SqK26CPPWfqt3354H/F5EdkiwzzMYJXmCt1xPEcY+sklEtgKuauF5FWGubYVX/0UxylwlIsUiMhBje/tPjDKvA9uJyJne8xwQkdEiMkJEgiJyuoh0V9WQV1+8/zElWEXB5jeMw4AjReTmGNsXqerMOLv/AWPQ/dTrar6HeeMD84aTh+l5fIoZlkqWfpgHpgzTLf+QOA+qqv4T+C2m0V+LeRu5FPOm0bhsKWZc9CHM+HslDYd0xgPzRaQCuAvjBVbjvU3fAnzsdYXHqOrLmF7Hc965z6Opm2irUNV1mCGcv2GGHXbEeGPVxtnlz972OcBc4AtvXUrxeksHe58fRGQD8ADwptewHIIxfq6K+szC/PdnN3P4G4DHvet7cpwyGbmnSP/1vAvjBPKuiJRjzmVvAFWdj3kxeAbTu9hInGFHr7G9FWMrq/P2/xr4B/Cg19uIxWvAcGC1qn4Vtf5GjOG4FPOy0ngUoTl+h+mllGNeImMpgVcx9prZXh0PNy6gquUYr79TMT2OVWxxHgDjFbbYe+5+hen5pA3Z0uOyWNovIuJgGovTE9iTLBZLGrA9Cku7RUSOEJEeYias1Y/1fpplsSyWLodVFJb2zD4YN811wDEY18rqxLu0X0TkADFhQZp8si2bxZIIO/RksVgsloTYHoXFYrFYEtIpglr16tVLhwwZkm0xLBaLpUMxa9asdarau7lynUJRDBkyhJkz43mvWiwWiyUWIrKk+VJ26MlisVgszWAVhcVisVgSYhWFxWKxWBJiFYXFYrFYEmIVhcVi6RKoumh4KRpZm21ROhydwuvJYrFYEqG1H6OlV4NbCrhoYBekx52Ir2+2ResQ2B6FxWLp1Gh4CbrxYnBXY1K31EFoNrrhbGxkiuSwisJisXRqtOpZTOqNaCLgroLQl9kQqcNhh54sFkunRLUa3AoILwXCMUqI18vwyod/gLpZ4PSCnP1JnE21a2EVhcVi6VSoVqOlf4KaN6PWCk0ymGoY/LsYI3fZH6H6LVNOHJA86Pkk4t8mg5K3X+zQk8Vi6VTopt9Czf+AuqhPY1tEHuQdg/i3hppXoeZtjP2iGrQS3PXoxktQVfMJ/4iGYybZ6xLYHoXFYuk0uOE1UDuZpoqhHgHpAcG9IbA7bqQMyu+GJmlOFCIr0ZrXofz2Ld5SvkFI8b8Q/9D0nkg7I6s9ChF5RETWiMi8qHU3iMgKEZntfSZkU0aLxdKBqPuI+EoCs003Qu17UHYLrB0D7or4xUuvNUZvqoFaiCxEN5yOl567y5DtoafHgPEx1t+hqiO9z5sxtlsslnaORlahVU+jVc+ikTWZqbTmvSQLhoEqYhu566mNsV1N76P2w9ZI12HJ6tCTqk4VkSHZlMFisaQet/IpKP8rxogMcCva7Qac/BPSVqdqNdRNTeERXe/TuKIwuF1rdne2exTxuFRE5nhDU8WxCojIBSIyU0Rmrl3btf40i6U1aGSNebuvehqNrEpfPeGlnpKoxRiIa8xy2Q1prZfwj2Tm3VcgMDID9bQf2qOiuA/YBhgJrAT+EauQqj6gqqNUdVTv3s0maLJYujRu1X/RtYegZX9By/6Krj0Mt/KJ9FRW8zYQibPt3fTUCWb+Q8KhpBQRPAAJ7Jj+etoR7U5RqOpqVY2oqgs8COyVbZkslo6MRlZD2Q00ecMvvx0NJ5XgrIVEiG1QVuIqkBQgvj4Q3BcIpq0OEOhxZxqP3z5pd4pCRPpH/fwZMC9eWYvFkgQ1k9hiK4gmgta8lfr6cg4FYs1qFsg5JPX1RdfQ45+QcxDgS8PRHQgehON0vRnb2XaPfRaYDmwvIstF5DzgbyIyV0TmAOOAK7Ipo8XS8YkQ0yiLGsNsipHAcCj4BZCLaWJ8Zrnw14h/UMrra1C3U4hTfA/0ngxOGyLDymBwBoAUeL/zwemFdL8pNYJ2MKQzRE8cNWqUzpw5M9tiWCztEg0vQ9dNwAw9RZOLlLyIBLZLT72hb9CadwAHyT3SKJBE5d1NaMW/jY1DciH/50j+6Yi0rneg7gZ0zSFAZav2Rwoh9xRwchD/MMg9ApGc1h2rnSIis1R1VHPl7Mxsi6WTI/6BaOFvoOJujLFXgSAUnJs2JQEggRFIYERSZVWr0fUnQGQVmyO9lv8DDX2J9LijUdkQWvUfqH4BUMg7Hsn/OSINbRPi9EQLL4SKezF2mRaiFVD9CBRcCLkTEOm6zaXtUVgsXQQNL0Sr3wRcJPeIZhtxjfwEtZ+AUwg5YxHJTZtsbuXzUH4LZgZ0NDlIr1fNGz2YuEsbz4e6zzAxnAAEfMORXq8h0nA0XbUO3fRrqJ2OUUCtNaYL5BwL/q0R/2Cvd5HXymO1H5LtUVhFYbFYmuCW3wmVD4P4MIZwQYofQoJ7pKe+TVd6gfwa4wMckADkHgU5R8Cmi9miJKLI/TlOjxtiHl9D36BlN0FoVgqkdUCKkJIXEP+QFBwveySrKNqd15PFYskuWvsZVD4K1IJWmWiqWoFuvBDVxgmAUoRvELHdWiNAyMhR/QqUXU1MJQFQ8xwaxzgvgRFIwflAfoytDi1rCl3QUnTTb1qwT8fGKgqLxdIArX6epkNAABGo+zwtdUr+yV7vJREhcMsSbHchkiDAX85YCO4ORA8Z5UHBr6DkLWIrkQSEv8VddzIaWd+y/TogVlFYLJaGaALDrzb2nEoN4uuPFD8EzlZADvGbplhuvlE43eLXIT6k+EGk+y2QczDkHo0U34/k7A/ltxl3WmmhS214Lrrxgpbt0wHpumZ8i8XSAA0vR8tvgdop8QpAMH2BEiQ42sx/cH9CqydBxR006dlIDkh/cJc1PYBvKOLEDA23ZXfxQ97RSN7RALjVb0DpH9niFVVvk4HE4crriUB4IRpeiPi3TaJ8x8T2KCwWC+qWGffU2g9o6hnkB3Kg202IU5j4OJH1aM17aN1MTBSeliEiiG8rMxTlFNCwifKbeE49/wsyoNGehZBzKFrzAarJeTapRqDsJhq6ztaHH2mBk4/4IdK5A5PaHoXFYkGrX/KGnBo37j4zRFP4q80uqvFwK/4PKh4wHkqoySTX8/FWzcYWJx9KXkBLr4e6T6gP/yHdb0ScHtB3Cm7dHKh5B6r/A4Sh6kG0+mnwDYSezzar1IisTDzMtplCjDKJM4td66CTBwm0isJisUBoLjEN2JKDBMc0qyS09kOofAjjKeXZMbQa3fhL6PU2IrFiTSVGfFshPR/2egjSZI6EE9wVt/xWMzGuXsFpFYR/QCvuRrpdk7gCpzvJzauow8SuiqUoBAovQJzuSRyn42KHniwWC/h3wMRmaor6BuFuuhp31S64q3bEXXcibnhpwzKVT8XIO+2aNKLhBW0STcTXQEmoumjdDGNfCH1F015QKM6cjEbHdYqMUbvZaLMOcXsTvsE4hb9utq6OjlUUFosFyT8RJEjDKLMB8A0zeaNrXmJzatDwHFh3BG543ZaiWhrnyD7vjT81aHgRunas8TQqu464PQI3lntvU6T7XyBnP4ynVQxFKXmQf7oXkbaxQslDiq5MXvgOjFUUFosFcYqRkuchMBrTLAQgdwIUXgHujzH2iED5zVt+5hxB7B6JC4GdUyKjqotuOA/c1d4kwETB/mrRuuZnYYtTgFN8P9L7PaTkGSi6AaQYoxTyIO90pOh3SPe/QXCMtz4IOOD0Q0MLcEPfoFVP45b9Fa15J+6kv46MDeFhsVgaYGwCDiKCW3YrVD0Wu6DTG6fPx2YftwrdcBKEl2NsHQ4QhG434+RPTI1cdV+hG89pRkHUI5A7AadRQMGk6lEXdBNIYYNAgxpejq77GVBO0+GuABAy4ch9A5FkjOntABs91mKxtIoGYb0lQZph6bVl0cmHkv+iVa9A7WTw9TYhwlPpDaTlxE7AFLMwtHLGtIgD0rPpEcuuB8qI7TrrhTbRKgh/j5ZeDT3u7DQRZ+3Qk8ViSUCCuRB5R25e1MhatOwvUHkvuCsgsBv4kwsxnjSBkZB0rKkAaAXuxsvQmsm0deRE1fXcdJM5jgu1k9D1x6NuK3NhtDOsorBYLPFxl8bZ4N88C1rdTej6iSY/hLsKwt9D+S0mWmsKEacQiq5hS+Y8gDwvk10OW3obPiAE4XlQ+xZa+lu09Ko2KguhZc2lQngBWvrHNtTZfugc/SKLxdIi1N2AVj5p8jr4ByP5sZMYSXAUWvOGGVJpgB8Cu5hjVT0DbjkNXEi1Gqqfxa37DAK7IIW/TEmIC6fgNDSwI1r9DEQ2ILlHQN4xEJpjzic0v2l4D62CmkmoczvqrgDf1kjeqYh/YNL1igiacxjUvkvyOS0Uat/GLbsbKfp1q+aStBesMdti6WJoZBW6bqJnFI5K/uPfDim8HMk9ZEtZrUXXHQ2Rn9g8Dk8u5OyNU/wgAO6Gs6FueoIaHTNxr/jRtOWzgPrUp/sTd84DPkwj7wcCSPF9SM6+LTv++p9DZGlUHULzw1ECeSfhdP9z0nVlCpuPwmKxxEQr7gIto2FeB2+oZNNvcSv+tXmtSA5S8qKZS+D0M3kjCi9Fety7ZVffIBI3Ja6ZpV12Y4rPpBE1k0hs7K7vCYSBarT0Dy0ajhKnJ9LrTej+d/DvCOSAdIeCi7yot/FQqH4Zt+rFNttKsoVVFBZLV6N2KvGHT6qh4n7U3bR5jTjdcbpdg9NnKk7v93AKL0AksGV7wdk0P7sZCH/bqkCBSaMhkveKwuS2iMSzwcSro9qEJA9/hwlXsskkeXKaG8YKQdn16LrD0fDyltXZDrCKwmLpakj8nA1mewBC85I/nH9bpPhf4PQhXhgQUzCPFjXkLSVnbAt3iEAL84Br9SvgltJweKsGwl9g5lIkIgyRZejGXzboWagqbuVjuGv2w101AnfdRJNlsB1hFYXF0tXIOxszXh+PiAnn3QIk50Ck91QzNFPwS5oqjFzIOy2tBl3xbw0FlyZZ2geBHRFfCxMVhWYQO/ufQ9Kus5El6IYzcMv/YexFFf8H5XeAuxaT3+Ibo0zqZrdMtjRivZ4slq6Gbkqw0WdsDv7tW3xYEQf8W0PhFai7EapfM4mGtBZyD0eKrmi9zAlQrUOr/gs1Xn3B/aDu4zil80AccEqQHne3vDLfMMwwW6O83eKAb4QXADFOTu/NhI3CCX2FVj1lEkLROHNgjYmA2/ORlsuYBqyisFi6GlWPENdG4d8JKb6nTW/+In6k+61o4ZUQWWJCWvgSzPBuA6oRdMM53lBZfW6JXJMLQ8tpeJ65UHA2knMABPZsErY8GST/ZLTqEZODYjN+cPpD8aNQejnUzTDJjLQG08uIZ5ep844T51qHv2uxfOkiq0NPIvKIiKwRkXlR63qKyCQR+d77Tpzb0GKxJI2qJoj0Kji9Xmz5cEy8o/lKkOAeaVMSgMnIF/6ahlnqajzX37xGhWug8nHjqUQdbvn/4a4Zh7vmIDMMlMQsavH1Q4ofj+pZBCA4Bun5JI6vG07PR5De7yM9n4HeH4FvMAntNkDcISv/Ns3KkymybaN4DBjfaN3VwPuqOhx43/ttsVhSgIiAf3jsjf4dMitMCtDaaTEmA4KZ8xErvHktWvGQ6YVUPmDCjbgrofJRdMPPk0qjKsHdcHq/jfT+EOnzqVEOvqi4V74+SGAEjq8EKXkZiq6E4D7E9wyTGNtykXaU5yKrikJVpwIbGq2eCDzuLT8OHJdRoSyWTo4UXUfDt1wBcpFu1wJe/uzQd6ibujwSacPpRfPeRtG4EJoD4W9paBeoM8NktVOSPpL4Skzyo0RlnHycgrNxej4OBedhQo1E4+Ujj+5VOFuZ4b9gs/PgMka2exSx6KuqKwG87z6xConIBSIyU0Rmrl3buRObWyypRHLMUAnBg8AZAMEDkZKnILAHbun16Jp90Q2noGv2wS27PeuTxFQVrXkHd8PZuOtPwa18EvXSrUr+8ST24GqMH3w9GtkY6iuqQkNzUiJzLKTwEsjZH2NDKcQoCBczbBYV7NDdAIE90yZHa+iwxmxVfQB4AEwIjyyLY7F0KCS4G9LzwQbr3LK/Q/UrbDGyAlVPob7eSME5GZexHi27yWTYq0+1GvoGrXkVej6L+LaCHnehpb/DzAB3ie2+6iF5kHu08U5qkmAoD/ENSNNZgEgQKb4Pt+4rqPsIIqtNIMUmVKMV9yLdfpc2WVpKe+xRrBaR/gDe95osy2OxdHpUFaqfoqFRGKAaKh/KhkgAaHgpVL/YKB93DYQXeiE7QHLHIX0+RYofRkqeBt/OxGza/COQkheR/JNoGG22nlrUn5psfLFQVdyy22HDGVD5CFQ/T1zvs+r/orXTk7KZZIJmFYVEz9Xfsq5ls3FaxmvA2d7y2cCraazLYrEAEG7UGEfhJpp3kWbqZoLEGFrSKrRu2uafIgHjYRXYGel5D/i2BikwH4KQdzJS8griH4pIDhQ/SFNF4cLG88y8jMgqtHYaGl6SslPR6peg6ilM6I9YWfKiC69HN12Mrj3YKMssE3foSUTGAU8COSLyJXCBqi72Nr8LtDkMpIg8C4wFeonIcuBPwG3A8yJyHrAUOKmt9VgslsSIBFDfEIjEyI8d2Cmjsqi64K4x4/hOMbHfZ/3gxHa7FV9/6DUJQjMhsgYCu5lZ29FlwnNQgjTpQWmNNy9jLkgQNIQGRyE9/mWy+LWFqkdIOCzWGK00wRQ3XYz0er1tdbeRRDaKvwFHqOp8ETkRmCQiZ6rqp6QoYIuqnhZn0yFx1lsslhSiGgZ8iAjS7Xp040UYbyDFNNA5SNE1GZPHrZ4E5X/y8lu4kDMONJZXkx/JOzHucUQEgqPjbtfwTzQdZsOsC83G9LA8r6i6z9Hym5Huf2nBmcSgVT0zF8JL0fASxD+4bfW3gURDT0FVnQ+gqi9i3FQfF5GfkVxQE4vF0k7Rui9w1x2Lrt4JXT0St+xWCI423k/BsSaMR85hUHQVWnYD7urRuOvPQOu+SJ9MoTlQeiW46zDKKmTcVQPDTBhvyTe9DClCetyJ+Ae1ui4J7m6O1wSXpvks6qD6f55SbQPBfYnZ5EoR+LaNvQ3M0FvMuSKZI1GPIiQi/VR1FYDXszgEeB1oP1MGLRZLi9DwInTDuWwZBqmGqudQdw1OjzuRnvcD4Fb/D0qvZfObd+hzMyzT85G0+PhrxUM0jXlUZ8Jz9Hob0VLjjRXYiRim05aRMxZ8QyC8KKrOXFNfTNtBBAih6jMpVsMLzezswK5JhzuRosvR2ileox/CKIYg0v12JPdg3Ip7oeI+ml6DAPgbZh9Utxytet54T/kGIPlnIYGWx+dKlkSK4mqgL7Bqs3Cqy0XkICDZEI0Wi6WNqFsJNe+AuxoCu0FwnzbFYtLKB2kauK4Gat5HI6sRX1/jBVX+V5oOz9Sg5X9DSp5vdf1xiSwh5mCFBBB3dUqz44n4oeczaOVDUPMq4IO8E02cprqpTeXwDwN10Y2nQfgbQEAVAttD8SMmn3dzdfq2gl5voJWPmHr8g5CC8xHPBiT5Z6HV/wP3J8+xwA/4ke5/RaIM+upuRNcdB+5GzP/jmP16/APJPSw1F6gRcRWFqr4XZ30pcEtapLFYLA1wq16AsuvZ4kaZA4GdoedjxnunNYQWENMtU4ImkY+vrzGkuo2DJnikK1hdcDSEv6fJ0I+GIAX5thsjTj5S9Bso+s2WqsKHo+u/8AL6hQA/SBDpdjNa/jcv+GCUkg3NR8tvRbrfmlydvj5It9hRicQphF6vmGGuuo/A6Y/kn4L4hzQop5UPesNz9ZP0zKQ9Lb0Wcg5uoFRSRXucR2GxWABkPZPmAAAgAElEQVStnQpl19GwUa+F0By08rHWHziwMzFnM2st+IaaZckzIbtj4aQmaGBjpOB8z24Q3SzlQcG5iNNMsqVUyeAf4uXUOAcCoyHvFKTkVdObqXmVpj2xEFT/L3X1Sw6SfyJOjztxuv2hiZIAoOY9Gszk3kxdbK+1FGAVhcXSTtHSm4jtNxKC6pdbfVwp+GUMJZALecduDm4n4oP8c2kagTUPCi5pdd0J5fL1M0H0co8yMZz8w5HuNyCF6cljEV+OvjhFV+GUPI3T/U9bvI1ihf0AjO0ig/49TvfY6zXihQZJQ5XNFRCRJvMYYq2zWCwpxl2WYGPrGybxDzJhsAOjgSBITyi8EOl2c8NyhZdCwdleCtMck0K14Dwkd/9W1928bANxevwDp88nOL3eQPJ+ltaseC0ipteSQHDvjMoo+WfTVIH7jJHf1y8tdSbTo/hjkussFkszuNWv4649FHfVzrhrJxgvmHhISfxtece3SQ4J7GjemPvNw+n7KU7hJU3GtkUcnKLfIn1mQLcbAR9UPoyuORB3wy9MFrsuhHS73ss3Xh95NxekG9LthswKknsU5P8co+QLjSL3b9O6jH1JIvG6TCJyJDABOBn4T9SmbsCOqrpX2qRqIaNGjdKZM2dmWwyLJSFu1X+h7EYaehLlIsX/h+Qc1LR85VOe51Ejd0nf1kivt1pvzG4hGvoaXX8qDeX2Q2BnnHR4P7Vj1C1Fq16E8HwTOyr/RMTJTm41jawzM8h9fcC/Y6t6NSIyS1Wb9XVO5B77EzATOBaYFbW+HMjsoKHF0sFRVaj4J7HdTW+PqSgk/3RUq6J86wVyj4Vut7QqjWdr0cpHaWrEDUPoWzS8EEmDR1J7RZzuSOF52RYDwNiTfOMyUlci99ivgK9E5BlVjWVit1gsSVMH7vrYm+IEnhMRpPACtOBccEvB6d72iWatIbKcmJPQxG9CZXchRdFVSSYfxV4icgMw2CsvgKrqsHQKZuk6vPn9Au6b+TlrqyrZe6uBXDFmX4b06Gyp0oNmfFtjxPtpJgeCSAB86QzY3AzBfcwQR+NehYYgMCIrIlkySzL914eBfwL7A6OBUd63xdJm7p/5OVdNepv5a9ewprKSN75fwLHPPcWy0tJsi5ZSRAQKL6Gpt0ouUtS+R3Kl4AxwutHwvTIP8s9CnJ7ZEsuSQZJRFKWq+paqrlHV9fWftEuWYVSVr1at5JVvv+GbtTZXUiaoDoW4+/PpVIe3zMR1VakOhbhnxqdZlCw9SP5ZUHQlSDEg4PSBbjcjueOzLVpCxOmJlLwCeSeb1Kn+HZHuNyFF7ScDmyW9JDP09IGI3A68RJT7haqmL4xkhimrreGMl15g4YYN+Bwhosqe/Qfw0DE/I8ffYbPFtnsWb9qIE8MoG1Fl5k8rsiBRehERpOAsNP9MIIRIMNsiJY34+iDdb8i2GJYskUwruLf3He1CpcDBqRcns7y1cAG3TvuQFeXlW1Z60RJmrFjOPz/9mD/u39QbxZIaehcUEnJjp3qsDYe5etI77Ny3LwcPHcaAosyEcMgExo2x4ygJiyXuPIqOREvnUagqZ778Ap8sTzTzFYKOw5yLfkPQl/ogWxbDxW+8xgeLf6A2Ej83sCPCfgMH8a8jj6EoJzNzByyWrkCy8yiSCeHRV0QeFpG3vN87emlKOywPfTmzWSUBUOe6/P69tzMgUdflH4cfyeHbDCfo85ETRyG7qkxfvowrJ72VYeksFgskZ8x+DHgHqPfh+w64PF0CZYL7Zn6edNnXv1tAWW2slImWVJAXCHDX+KOYcf5FXDFmP/Li2ITCrsvUJYvZWG2S7Xy1aiXXTp7EFW+/wbuLvsftBD1ji6W9koyNopeqPi8ifwRQ1bCIxB8n6ABsqkm+4XdVefP775i4/QjyAlmY7NRFKMrJoW9hIU6CMAR+x2FTbQ0vfD2POz/7hLpIBFeVST8uYsxWA7lz/FFM/nERZbW17DtwEMOKretmV0cjK9Cql8DdiOQeCMEDMzqrvbOQjKKoFJESvHCVIjIG6NBO7n3yC1hTVZl0+T9PncKfp03h9kOP4Mjh6Us32NU5eMiwhD2DoOMj1+fnjk8/bmDTqAqF+HjZEvZ66D58YrzWVOG0nXfl/x04tv1EH7VkFK15H910BZvTmNa8BIGRUPxgdma4d2CSUa2/BV4DthGRj4EngF+nVao0c/yInVpUviocoioU4spJb7O8rEPryHZNUU4O9044NqatIs/v509jD+azFcvxO01v29pIhJpwmMpQiJpwmNpImP/Mn8uHSxZnQHJLe0O1Di39HSa2lheBSKug7suUJhrqKjSrKLz5EgcB+wIXAjup6px0C5ZOfjVqL3yteMusCYd58ev5aZDIUs9BQ4Yy45cXc+PYQ9hv4CC2Ke7JIUOH8dhxJzBx+xHkB/xJ9xCqwyH+M79D36qW1hKajYk21JhqtOa1TEvT4Ul2NtlewBCv/B4igqo+kTap0ky3nBz+efgErpz0FmE3RrCzBLz49VwuH7NvmiSzABQGg5y560jO3HVkk20HDh4S8/GPR10Ct1tLZ8ZP/OROdg5LS0nGPfZJ4O9sifVUH++pQ3PM9jsw5ezzuGLvljX6P1VUMPmHRWmSquV8s3YNb3y3gO/Wr8u2KBkh1x/g4WOPpygYpDAQpCAQJOg4Mee65AcCTNzeBq3rkgR28zLzNULykPyTMy9PB6fZCXci8g0mUVFG/Q9FZDEm90UECCeaFNLWxEW73/8vSmtrmy/okef3U5KXT10kQlFODseP2JGzd9uD/Ax6RVXW1fGL115i3prV+MQhrC57Ddia+4+e2CnDjpTW1LCppoatu3XD5zjUhsN8tGwJNaEw+w4cxGfLl/HbSW8RcZWQGyHfH2CvrbfmwaOPwxfDpmHp/GhoDrrhXMA1+aRRyDsR6Xa9dXDwSHbCXTKK4gXgN6q6MlXCJYOnKEaparOvym1VFPfN/Jx/NQpO11J65+fzt0PH07+oiO1K0h8S+g/vvc2r335LXVQIjByfjzN3Hck1B4xNe/2Zory2lqsmvc2UJT/iE4ccv48bDzqEY7bfoUnZ5WWlvPTN12yqqWbckGHsN2hwQndbS+dHtQZqJoOWQnAM4h+abZHaFanIcFdPL+BrEfmchkEBj22DfO2KC/cczfqqKp6a8yV1LbRZ1LO2qopzX3uJgOPQt7CQR489nm16Jsh53AZUlVcXNFQSYDx/np8/r1Mpil+//TqfLl/m2RoiVIdDXP3+O/QvKmLUgK02l1NVVlVUUF5XS14gQJ9m5mREXJeKujoKg0Hb4+jEiORC3oRsi9HhSUZR3JBuIeKgwLsiosD9qvpA9EYRuQC4AGDQoEFtqsgR4boDx3LpXnuzxwP3tulYIddleVkZhz31GACDunVn7JCh+B2HQ4Zuw5itB7a52+uqEopjpK2JtL5X1N5YWV7OZ5uVxBaqw2EemDWDPfsP4IPFP/Li1/OYv2Y1q6sqCUUiOCI8OvsLrtxnf87bfc8G+6oqD34xk3tmfEpNOExeIMBle+/LObvtbocjLJY4NKsoVPVDEenLlmRFn6tqJhI27KeqP4lIH2CSiHyrqlOj5HoAeADM0FMqKuyRm8fAom4sKy9LxeEAWFpWyhNzZiPAc/PmMm7oUO4af3SbhkR8jsPu/frzxaqGo4H1wfM6C2uqKgn4fDEDBi4rK+WayZP433ffUhVqmKk3okokHObvn0zjqOHb0a+waPO2x7/6krs++2TzMGOotpa/fzKNPL+fU3feNb0nZLF0UJLxejoZ+Bw4CTgZ+ExETky3YKr6k/e9BngZ46Kbdq7e/8BWzbFoDsVM3Ptg8Y98sPiHNh/vlkMOpzAY3Oztk+PzURTM4f8dmJlk65lgm+KeMd2X/Y7DdiW9eHXBN02URDSOOHy4+McG6/7v80+b2KKqw2Hu/nx6aoS2WDohyQw9XQuMru9FiEhv4D3gxXQJJSIFgKOq5d7y4cBN6aovmsO2Gc6Ovfvw/Yb11LTBuB2PqlCImz/8gEUb1uN3fIwdMrRVMYm2L+nF+2f+gqfmzuabdWvZrU8/Tt15V0ry85uUrQ2H+d933zJ58Q/0KSjk9J13Y3hJeuwnqaQwGOSS0Xtz74zPNjfuPhEKAgEGduve7BwYEQhEuc2GIxE21lTHLLumMvmQLhZLVyMZReE0GmpaT3KhP9pCX+Blb8zYDzyjqhmJ9+13HP5z4ik88dVs/vvNfBZuWB932k5rWVpWym0fTwPgto+mcs7IPbh6/wMBWjQk1buggCvG7JewTHUoxIkvPMuSTZuoCofwifD8/Ln8/bDxTOgAcasuGT2GoT2KuX/WDNZVVbHfwEFctve+fLjkRwKOk1BZuKocMnSbzb+nr1iGEHsaVo+c3NQLb7F0EpJRFG+LyDvAs97vU4A30ycSqOoPwG7prCMRuf4AF+w5mgv2HM3aygoOeuzhtBmJw+ry0JczeehL496b7w8Q9PvwiXD4NsO5Ysx+9IrRS0iWp+d+xY+bNm7uHdWP3//u3bc5eOgwcv3tPzjahOHbN1FqR267HbdMmxKzfJ7fjwJ3HjGB7rlbFMBP5eX4HYdQDOUyonfvVIpssXQqkon1dBVwP7ArpvF+QFX/kG7B2gu9Cwr5fweNi5snIdVUhUNsqqlhfXU1L8yfy8TnnqKyrq7Vx3vz+wUxh9BqImGumtRxkzIV5+Xx76MmUhgMbv7k+f2ctetIbh53KNN/cSGHbzO8wT679e0X0xU2x/HZGdwWSwKSbf0+wcyQdoEZ6ROnfXLqTruwuqKCB7+YQch1WxwfqrWEVVlXVcXL337NGTHiHiVDYYLUoe8sWsjaykp6FxS0VsSU8NWqlby18Dt8jsMx2+3ADr2Se7s/YPAQZpx/EZ+tWI6qsvfWWyfsIe3QqzcHDBrCtKWLNyvPoM9Hv8Iijtmu6QQ+i8ViSGZm9vnA9cBkTDjGg4CbVPWR9IuXHG2dmZ0s1aEQ329Yz9mvvNiikB9tpWdeHjN/eTEAaysruXHqZN77YRGCcOS2w/l/B46jOC9GXBtg0qKFXPTmazHzPBQEAtw1/mgOHjosrfIn4s9TP+DZeXOoCYdxRAj4fFy21z5cOCo9Tm5h1+Wx2V/w7Lw51EbCHLntdlw6ekyDISqLpauQyhAeC4B9VXW997sE+ERV240lNFOKop7VFRXc9vFUJv+4iNpwuNWzuVvCb/bah5N23IlT//s8qyvKCXv/W8Bx2Lpbd94945yYwyqqylHPPMG3MYIG5vkDvHjSqYzo3Sft8sdi7prVnPLic02GxgQzTDRx+xGcsOPOFAZttE+LJR0kqyiS8V5ajgnOV085sKy1gnUG+hYWcscRE5h94aU4GUqrePfn0znwsYdYUV62WUmAmQm+prKCKUt+jLmfiHDH+KOaJAPyOw7DiotjKomV5eX8Z/5cXlvwDRVtsI80x7uLvqcuhv1EgdmrV/HnaVM49IlHWF1RkTYZLBZL8yTTyq3ATLK7QUT+BHwKLBSR34rIb9MrXvtGgdoMhsyI1/erCoVYsC5+7MTtS3pxz4RjKcnLJ88fIOjzMXrAVjw68YQmZe+b8RkHP/EwN334AddOnsSYh//NR0uXpOgMGhJwfDgJ4ixFVFlXXcVfP54at4zFYkk/yRizF3mfel71votilO1SOCKM7NefL1dlNLBuExSYvnwpF4/eO26Zg4cO47Pzf8XS0k0UBnNiutx+tXoV/5rxqRcyY0vYjIveeJXPzr8o5WHUj95ue+6b+TlhEs+FmJyCmewWi6X1JBPr6cZMCNJRuXncoRzz7JMpn5TXUmb+tIKN1dVxjdpgFNuQHsVxt//363kx4yqJCFOXLGb8tsNj7NV6hhX35Or9D+CWaR8m9CSLlUPbYrFkjmRiPY0SkZdF5AsRmVP/yYRwHYEde/fhwj33alF6znRQG4kw7omHuX/m5wnjHzV3jFjeUaqatiG2s3fbgzdOO4tAnCGooM/HKTvtkpa6LRZLciRjo3gaeBQ4ATgm6mPxuGzvfdij/wBys5xZrqy2lr9+Mo0JTz9OTbjlyuKo4dvHHF4Kuy4HDBocc58V5WW8s+h75q5ZTWuTIA4vKeHBY35Gvt/fQOEGHIcxWw3kktFjWnVci8WSGpJxj/1IVffPkDytItPusbFwVfl42RJmr1pJjs/PgvVrmbt6NeuqKtmUwTkX9fzpwHGcPXKPFu2jqlz2zhtM/uGHzXGhAo6Paw8cy+m7NIyo4qryx/ff5bUF3xDw+XBdZUiPHjx+3IkxAxMmQ004xEdLlrBgwzqKc/PYvf8ARiQ5+c5isbScVM6jOAQ4DXifhhnuXmqrkKmiPSiKRHy7bi3HPvtkA7fWTLD/oMHcPf4oeuTGt1s0RlWZvnwZ7yz6nvxAgON32ClmpNkn58zmto8+bBCy2+847LP1IB4/rqk3lcViaX+kUlE8BewAzIfN7imqqr9os5Qpor0rCoBQJMKoB++lPI3zEmLRPSeHD885n24pjo566JOP8MPGjU3WBx0fn53/KzvT2WLpAKQyZ/ZuqmqtiW0k4PNxyagx3PZJZucElNbWcsSTj9G7oICxQ4Zy7sg9E3pGJUtlXWwbiIhQGarrkopCVW06VUunJBlj9qcismPaJekC/Pfb+Vmpd3VVJfPWruG+mZ9z5NOPs7E6dvKelnDI0G3wx/BU6pWfT//CrjXFZtpLn3HWtpdyuP9kTtnqAl6//91WG/YtlvZIMopif2C2iCzwXGPnWvfY1rEqy6EoIqqsqark95PeJhRjvkRLuGzvfeiZm0euz3RK/Y5Dnt/PXw89oku9VX/6+iz+etbdrPxhNShsWLmRf1/5BK/e03FDuFssjUnGRhHTL1JV0xPXoRV0BBsFwDHPPMn8dWuaL5gBdizpzc0HH8rIfv1b3bCX1dbw3Ly5fLZiGYO7F3PWbiMTTujrjPxy19+yeF7T0GfdSgp5cc0jXUppWjoeKTNmewfbDTjA+zlNVb9qo3wppaMoio+WLuGsV9KWarzFCGbW818OOYKJO9jEPa3h6MLTqa1q6qDg8/t4ecOj5BW23R5ksaSLlEWPFZHLMJPu+nifp0Tk120Xseux/6DBXLnPfvhEWpQbO10oUBOJcMW7bzLmwfuYtmRxtkXqcAzYpl/M9QU98skt6HoGfUvnJJmhpznAPqpa6f0uAKar6q4ZkC8pOkqPop71VVV8vGwJuT4ft0//iEUx3EyzQdBxuHHsoZyys3VyS5bP3pjFzaf8s0GvIic/hwv+dgbHXjw+i5JZLM2TynwUQnQoUbOc/dfhDkxJfj7Hbj+Cw7fdjnfPOJcL9xyd1B+Rbupcl1umTWmzobsrsfdRe3LN05ez1fD+iCP02ronl9x1rlUSlk5FMvMoHsXko3jZ+30c8HD6ROpaiAglefn4HAc3Q7m4E1ERquPCN17l74eNp2de60JxRCIRXr9/Eq/f9y51NSEOOnkfTvn9RAq6Zzc3d7rYd+Jo9p04OttiWCxpI1lj9h4YN1kBpqrql+kWrCV0tKGnxkxbupiL33iNylZGfU01fsdhcPcevHPGOa2ypfz51Dv49PVZ1FaZiC+BnAB9h/Tm/i9vJ5hr05paLO2FNg89ichoETkSQFW/UNW7VfUuYKCI7JlCWbs8+w0czODuPQgmyPaWScKuy7LSUm6Z+gGlNTUt2nfx/GVM/9/MzUoCIFQbYt2KDXz4/PRUi2qxWDJAopbpduCbGOu/9ralFREZ703yWygiV6e7vmziiPDsCadw8k67pDyLXGupcyM8Nfcr9n3kgRZ5Q337+UIcp2kvpKaihtlT5qVQQovFkikSKYoSVV3ceKWqLgSahhNNISLiA+4BjgR2BE7r7GFEinJyuGncoXx5wSWUpCAWUyoIuS7V4RAXv/la0vktSgYUx8yDHcgJ0G9on1SLaLFYMkAiRZGotUq3VXIvYKGq/qCqdcBzwMQ019kuCPh8PH/SaVlPghRNxFXeXbSQz5Yv48dNiV159zh0FwqLC5r0Knx+hyN/cXA6xbRYLGkikaJ4T0RukUYxCETkRmByesViKyA6LsJyb12XYGiPYj4467x2Y7OoiYS5/J03Oe+1lzjqmSc4+YVn2VQTO7Cgz+fjH1NuZNs9hhHICZCTH6T3wBJuffNaem2V1o6oxWJJE3G9nryJdQ9h3u5ne6t3A2YC56tq2iLcichJwBGqer73+0xgL1X9dVSZC4ALAAYNGrTnkiXtJvRUylhdUcEBjz1IuB24zUYTcBzGJJGgaN1PGwjVhOg3tI+NeWSxtEPa7PWkqpWqehpwGPCY9zlcVU9Np5LwWA4MjPq9NfBTI/keUNVRqjqqd+/OmS6zb2EhdxwxIb2zG+tfFFTBVfzLy7esi0PIdfl0+VLWV1UlLNdrQE/6D+trlYTF0sFpdmzDsxP8z/v8kAmhgBnAcBEZKiJB4FTgtQzV3a44avj2/Hr0GHzxGltVJNTyHkdwZRXd3l9BcHklUhUCEXAV8gIQaX5uTch1Oe+1l1rsPmuxWDoeSU24ywYiMgG4E/ABj6jqLfHKdvQJd8lQWVfHwo0b6J2fz9sLv+fP06YA0PPVJUQK/JQeMsA09knQ/YOf6PnmciTiggsadCjfqzfrThyK1ETwldUR7pUDzdhI/CLsN2gIj048vq2nZ7FYskAqU6FmBVV9E3gz23K0FwqCQXbrayKV/mL3PQn6fNz2v3fpPm0VkTwfpeMGGJXaDL6NtfR8YxlOeMsLgtS5FE1fQ/WQQipH9SZ3djm6vJJIkZ+abbo1UBg5i8vxr62hdkgR4d65TF++lA3VVa0O92GxWNo/cRWFiPRMtKOqbki9OJZkOX2X3Zj3+mxm+BwC5WFyllVQO7AAfIl7AQVfb4oZ0lEiSt9nFlE7eSU5K6tMDHIHwt2CrLxkBOHCAP3//S05P1WZISqBqh2LKT9/BKW1tVZRWCydmEQ9ilmY5iLWeIYCw9IikSUpRIRT9xnF18H3qa6N0O/Bb9GAj58uGUG4V27cYSj1ATFMGoJZn/NT1ZY/3AX/pjoG3vIVGhAkpA1uhvxvNhJ5fQn9z7dKwmLpzCTyehqqqsO878YfqyTaAbscOIJCLyKrvzJCYFMd/R75Dqc6gtRGGno0AYRc6vrkIQmM1Y3Vi3gfp5GSwFtXNHkFpw+8iA2rN6XilCwWSzsk2eixxcBwYHPKLlWdmka5WkRXMGbHY8XClZy30xVEQltySLhBh4rdehLqm8emffqAKnnLqiiYt5Giz9fitMJLqjm69+7Gkz/cQ57N6maxdBhSZswWkfOByzBzGWYDY4DpgI3H0A7Yatv+uJGGDb9T59JtxjoASl5fFmu3lFO6toxHr3uG8287kzfuf5dJT07FH/Az4ZeHcthZB+LzJWFpt1gs7ZJkvJ4uA0YDn6rqOBHZAbgxvWJZWkKfgb1YvWRttsXg5bve4q2HJhOqDREJG+X1w1eLmfnObK577oosS2exWFpLMsGEalS1BkBEclT1W2D79IplaQnn3nIaOfkNEwI5PgeJEe473dRU1m5WEgC11XVMe3E68z/5NuOyWCyW1JCMolguIj2AV4BJIvIqjcJpWLLLIT8/gMvvv5A+g3qBmB7GJXefy3Z7tg+fA9dV7r74oWyLYbFYWkmLZmaLyEFAd+BtL/x3u6ArG7Mbo6oNYisdmXsa4bpwFiUy+AI+Xlz9MIU9OmfebIulI9LmoICNDraHiPwG2BVY3p6UhKUhjQPw7TtxND5/9g3J/oCf8g3pjiVpsVjSQbOKQkSuBx7HZLXrBTwqItelWzBLarjkrnMp2aqYvELjtppXmIvTzOztdJCTF6CoZwEVmyozXrfFYmkbzQ49icg3wO5RBu084AtVHZEB+ZLCDj0lpq42xMcvf86Sr5eRV5THEzf8h7rq5FKbpoqttuvH6h+NZ9agEVtz1WOXsO3IoRmVwWKxNCSVQ0+LiZpoB+QAi1oplyULBHMCjDt1P8656VT6De6NPwtpVld8t4pwKEI4FOGHOUu4cuyf2LS2NONyWCyWlpOMoqgF5ovIYyLyKDAPqBCRu0Xk7vSKZ0k1W28/gEgk0nzBNFNbXcc7j36QbTEsFksSJKMoXgauAT4ApgDXAm9hggbOSptklrSwzW5DGL6nyWedTSKhCJ+98UVWZbBYLMnR7BiEqj6eCUEsmePWN67h3ssf4/2npxEOhSkqLqRsfXnG5Zj38Tf8OHcJQ3cZnPG6LRZL8sTtUYjI8973XBGZ0/iTOREtqSavMI8rH7qIN6qe5q3aZ7nq0YuzIoe68PoDk7JSt8ViSZ5EPYrLvO+jMyGIJfOICD6fjxfveD1rMrx2zzu889gUfvbrIzn7xlPwB9pt0kWLpcuSKB/Fyqgyq1V1iaouAdYQO5mRpYPy9ScLslp/bWUtz932ChO7n82y72x0GIulvZGMMfsFGuZEi3jrLJ2E/G4JMtRl8JWgrqaOC3e7kpU/rM5cpRaLpVmSURT+6JAd3nIwQXlLB+O4S8c3iT4bzAsy8ZLxXP/ClYkVSYoJ1Ya56+IHcN3UJ1eyWCytIxlFsVZEjq3/ISITgXXpE8mSaU774/GMPWU/AjkBCrrnE8wNMOaoPbjg72dxwPFj+NOLV2ZUnlnvzuHogjN4/5mPMlqvxWKJTTIhPLYBngYGYAYilgFnqerC9IuXHDaER2rYsGojy79byYBt+tJrq5LN61WVv539Lz584RNCtZmNRDts5GD+/v4NFBUXZrRei6UrkGwIj6TDjItIoVc+8w73zWAVRfpRVWa++xXvPz2NL9+bw4ZVmzJWtwj87tFLOPyssRmr02LpCqRMUYhIDnACMIQod1pVvamNMqYMqygyz3ezFnHJ6KszVp8v4GPixUcw67055OTlsPMBOzBy7M7sfsgu5ObnZEwOi6UzkUpF8TZQijmpRSMAABU2SURBVAnXsTlIkKr+o61CxqnvBuCXQH0S6GtU9c1E+1hFkR3efXwKt597T9bq9wd8BHICXPvs5ex91J5Zk8Ni6agkqyiSmd20taqOT4FMLeEOVf17huu0tJDDzx5LXlEuN52YlneGZqmPRnvzyf/kyR/vpbhP96zIYbF0dpLxevpERHZJuySWDsnIcTvjD2Q3g54CHz7/SVZlsFg6M8koiv2BWSKywIvzNDcDsZ4u9ep6RESK01yXpQ0UFRcyeKeBWZUhEopQXV6TVRksls5MMoriSGA4cDhwDCb20zFtqVRE3hOReTE+E4H7gG2AkcBKIOa4hohcICIzRWTm2rVrYxWxZIjfP3bp5vkX2cAf9DH6yJFZqdti6QrENWaLSDdVLRORnrG2q+qGtEpmZBgCvK6qOycqZ43Z2adsfTmTnvyQ5d/9RMSN8NYDkzNW9wEnjuH65zM7KdBi6Qykwpj9DKb3MAszDBwd9UeBYW2SMA4i0j8qIOHPMBn1LO2cbiVFnHD5lkDDa35cx6xJmYlG/8mrMzhr+KUM2WkgP7/meHbYa3hG6rVYugoJ3WNFRICBqro0YwKJPIkZdlJMvu4LoxRHTGyPov1RWVrJ1eNv4dvPvs9ovYFcPze+/AdGH2GHoiyW5ki2R5HQRqFGi7ycMqmSQFXPVNVdVHVXVT22OSVhaZ9Me+kzfpy7JOMB6UM1YW44/nZqq2szW7HF0olJxpj9qYiMTrsklk5DXU0d9172KLVVdaZfmOn6q+t46OqnM1+xxdJJSUZRjMMoi0UZdI+1dGB+nLsUcbKb2+q9p6ZmtX6LpTORzMzsI9MuhaVT0a2kiHAo0nzBNFJVVp3V+i2WzkTcHoWI5IrI5cBVwHhgRX06VC8lqsUSk/7D+jJ0l0H4/A1vr5z8IIGcDOXEVhO40GKxtJ1EQ0+PA6OAuZheRXYC+lg6JDe98nuG7TqYnPwcbzJekDOvP4lb37yWop7pzy0hPvhqytdpr8di6QokmnA3V1V38Zb9wOequkcmhUsW6x7bflny9TI2ri5l+B5DKfj/7d15eFT11cDx75nJbFnYBNkKYimKCCW2cQGpgrJpsYjSolYLr1ql2lotbbVY60qtdX1frW/dF7QqSlGsC4ILoFIwUBHQqiCoFARRlkCSmWTm9I97wzMkM5OETDIzyfk8zzyZ+c1dzs3k5sxdfufXvgCAaHWUGWfezpLnS6mONM8pKq/fy5mXT2Ddyg3s3FbGyHOOY8zk4fiDNoqvMTWaXGZcRFbEJ4bar7OJJYrcE6mMcOOP/49lL63AF/BRFa7i8KGH8q/XVzfqTqlQUYiKsiTXI4R9ltW1z4E8+P7tliyMcaUjUUSBPTUvgRBQ7j5XVW2XplibzBJF7tr6+TY2rf2CXv17UtixgF8O/T3r3l3fbOs7+6qJTL52UrMt35hc0uQOd6rqVdV27qNIVfPinmdNkjC57cBenSkeMZCPStfxo27ns2ndZqQhN23vp3kPv958CzemlWrGXdKYhtn62ZfMOON2yndVUFFWicaab13b/vM1c+58kYaOFW+MsURhssD8RxcSi9bNDsHCAMGC9I6HrTHlgd/9jYd+/0Ral2tMa2aJwmTcjm1lVEWq67RrVBl34WhCRUHy/I3of1FPp/BweZhZN89lw5oWq3VpTE6zRGEy7sgxxQQLg3XaVZWTf3oiT268l0vvuYCe/bqlXI6I1LnTKZlodZSLSi7n5YdabtwMY3KVJQqTcSVjBjNgyCH7nGYKFgQYPWUEvQ7tSX5RiDGTR3D6pePw5qUYn7uBSaJGVbiaOy++n53bdu1/8Ma0AS1UT8GY5DweD398YTqvPr6YBY8twh/0cdJ5JzJ0/L5Fizev30K0OnkHPY01/gK1J8/Lshf/xaifHN/oeY1pK1IOXJQrrB9F27D470u5ecpdVOyubNR8voAPfyCPPQkKBXp9XtodUMTBg3pzxuWncsQJg9IVrjFZLy0DFxmTTYb+oISufbrg9aU4/ZTA2X+YyE3z/0Agv26P7GhVlO1f7GDF/Pe46gd/Yt4j1s/CmNosUZic4c3zcsebN3DMKSXOhesGmnnt03y9ZScTp52CP+jH58/D4637px8uj/DXXz2S8vSWMW2RJQqTUwra5fObBy/CH/Q1eJ7qSDU3/Og2CjsUcM/KWzjvxh+T3y6UcNqqyiqevnUurz3xJuXJakgZ08ZYojA5p6BdPr+4+/zUd0DVEqmM8PBVT3FAj46cftk4evbrnnC6cEWEx26YzR1T72FS95/yzrx30xW2MTnLEoXJSWMmj2DqbZMTnkJKxpvnYcPqzwE44/JTCeQn7vUd3hOmoqySyvIw1028xY4sTJtnicLkrHEXjqKwQ0GDp6+OVNOpWwcAhk04minXTSJYECC/KJR0jG/xCEtfWJGWeI3JVZYoTM7K8+Ux7YGfNWxav5f+R/WjKlLN/EcXUvrKSib88mSe2foAty26Lmk/ClUlUhlJZ9jG5BzrcGdy2mFH9yPPn0d1glpRHq+HQL6f6kiUAUMPoVO3Dlw4eBrePC8iQn77ELe9cR19B/dh5NnHsejpJVTuCe+zjFh1jCPHFrfU5hiTleyIwuS0jl07cNRJR+AL7HsXVCA/wIwXp3P7out5dO2djDrneJbMLSVSWUXF7krKyyr4atN2rjntZgCKRwxk2GlH7y0j4vF6CIT8nHfjWXTq1rHFt8uYbJKRIwoR+SFwDXAYcJSqlsa99zvgPCAKXKKq8zIRo8kdV8z8BbecezdLnl+ON8+DL+DjojumUDJq8N5p5t49r87RgsaUjR9vZvP6LXQ/uCu/ffjnjJkygkWz/0kg5Gf0T47n4EEHtfTmGJN1MnXqaTVwGnBPfKOIDADOAA4HegALROQQVbUeUCapUGGIq2ZNY/eOPez6qoyuB3Wpc+tsZXk44bwej4dwuXMNQkQoHjGQ4hEDmz1mY3JJRk49qeoHqvphgrfGA0+qalhV1wNrgaNaNjqTqwo7FNCjb7eE/SuG/2howk56wYIAvQ/r2RLhGZOzsu0aRU/g87jXG902Y5rk9MvG0f2bXfdeg8jzewnkB7hi5i/weLJtNzAmuzTbqScRWQAkGmnmSlV9LtlsCdoSlrcVkQuACwB69+69XzGatiO/KMTdy//MwqfeZvmClXQ9qAsDj+1PuDzC5k+20P2bXfdOG4vF+GTlp6gqfYv7WCIxbV6zJQpVHbkfs20EesW9/gawKcny7wXuBafM+H6sy7Qx/oCPUT85nqHjS7hy3J+Yffs/8OZ5qY5UM3T8kVwx8xI+Wv4J155+M+W7KkAgmB/kD89MY+Cx/TMdvjEZk21fleYCZ4hIQEQOBvoByzIck2ll7ph6Lx+VriVcHqF8VwWRyiqWzC1l5nVPc8Xo6/lq03YqdldSUVbJ9i07+N1JM9j1dVmmwzYmYzKSKERkgohsBIYAL4jIPABVXQPMAt4HXgYutjueTDpFwlW8NWcZVeF9O+iFKyI8e9fLxGKxOvNoLMbCp95uqRCNyToZuT1WVecAc5K8NwOY0bIRmbaiOlJNLMmQqZGKCLFo3e8l4YoIO760cbVN25Vtp56MaVb5RSF69e9Rp93jEQYee2idHt4AwfwA3z5+QEuEZ0xWskRh2pzL7p1KsCBAns85oPYHfRR2LOCy+6ZyxImD9t5CC04/i0HfO4xvH9e4RPHV5u3cd/ljXDL0Sv485S4+ee/TtG6DMS1JVHP/hqGSkhItLS2tf0JjXF9s2Mqzd73EZ+9vZMCQQxg3dTQdurQnGo0y/9FFvPTAAjQGY889gTFThjdqkKQvNmzlopLLqdxdSVWkGo9H8AV9XD37Nxw5xgoMmuwhIstVtaTe6SxRGJNeN579v7zx5Ft1roUc2Lszj62/u1HjfRvTnBqaKOzUkzFptnz+ewkvmG/fupPtW3ZkICJjmsYShTH1iEajfLD0Y/697GOiCe6Kqq2oY5JR92JKqDCY5uiMaX42cJExKaxa/AHXTrxl7yh3gVCAq2f/OmVP7dMvG8dfpz1KOK5irS+QxzGnlBAqDDV7zMakmx1RGJPErq/LmP79P7Lzy11UlDk9tXds3cn0k2awe8eepPN9/4JRnHz+ifiDPgra5+MP+Rk4rD/T7m/YsK3GZBs7ojAmiYWzlqAJe2orC2e9zfcvGJVwPhHhojv+h7OuPI31qz6j60Fd6NE3UX1MY3KDJQpjkti1rYxIZVWd9ki4ip3b6q/91KFLe444YVBzhGZMi7JTT8YkMXj4AAIhf512f9BH8YjDMxCRMZlhicKYJA4/tn/CntrfHT2Yw445JIORGdOy7NSTMUmICFfP/jWvPf4mLz34KiLC2HNP4ISzhtXpNBeuCDP/0UUsef4dOh7YnlMuGsuhJX0zFLkx6WU9s41posryMJcMmc6mdVsIl4cRj+AP+Lj4znM56dwTMx2eMUlZz2xjWshL9y9g07ov9vab0JgSrojwl0seomJPZYajM6bpLFEY00SLZy8lXB6p056X5+HDZWszEJEx6WWJwpgmKupUmLA9GlMK2ue3cDTGpJ8lCmOaaPzFYwnmB/ZpExE6devAt444OENRGZM+liiMaaLvjPw2Z115Gv6gj/x2IUJFQQ7s3ZkZL0y3kuKmVbC7noxJk53bdvH+ko8o6lTIgCGH4PHY9zCT3Rp615P1ozAmTdp3bseQU+rd54zJOfaVxxhjTEqWKIwxxqRkicIYY0xKliiMMcakZInCGGNMSq3i9lgR+RL4tAVW1RnY1gLraSm2PdnNtif75fo2HaSqXeqbqFUkipYiIqUNuec4V9j2ZDfbnuzXGrcpETv1ZIwxJiVLFMYYY1KyRNE492Y6gDSz7clutj3ZrzVuUx12jcIYY0xKdkRhjDEmJUsUxhhjUrJE0Qgico2I/EdE3nUfJ2c6pv0hImNF5EMRWSsiV2Q6nnQQkQ0issr9XHKu5ryIPCgiW0VkdVxbJxGZLyIfuz87ZjLGxkiyPTm7/4hILxF5XUQ+EJE1IvJLtz1nP6PGsETReLerarH7eDHTwTSWiHiBvwAnAQOAM0VkQGajSpsR7ueSi/e1PwyMrdV2BfCqqvYDXnVf54qHqbs9kLv7TzUwTVUPA44BLnb3m1z+jBrMEkXbcxSwVlU/UdUI8CQwPsMxtXmqugj4ulbzeOAR9/kjwKktGlQTJNmenKWqm1V1hfu8DPgA6EkOf0aNYYmi8X4uIu+5h9a5eJjZE/g87vVGty3XKfCKiCwXkQsyHUyadFXVzeD8owIOzHA86ZDr+w8i0gc4AlhK6/yM6rBEUYuILBCR1Qke44H/B/oCxcBm4NaMBrt/Eg3i3BrukT5WVb+Dc0rtYhE5LtMBmTpyfv8RkUJgNnCpqu7KdDwtxYZCrUVVRzZkOhG5D/hHM4fTHDYCveJefwPYlKFY0kZVN7k/t4rIHJxTbIsyG1WTbRGR7qq6WUS6A1szHVBTqOqWmue5uP+IiA8nSTyuqn93m1vVZ5SMHVE0gvuHUGMCsDrZtFnsHaCfiBwsIn7gDGBuhmNqEhEpEJGimufAaHLzs6ltLjDZfT4ZeC6DsTRZLu8/IiLAA8AHqnpb3Fut6jNKxnpmN4KIzMQ5bFZgA3BhzfnJXOLelngH4AUeVNUZGQ6pSUTkm8Ac92Ue8Ldc2yYReQIYjlO2egtwNfAsMAvoDXwG/FBVc+ICcZLtGU6O7j8iMgxYDKwCYm7zdJzrFDn5GTWGJQpjjDEp2aknY4wxKVmiMMYYk5IlCmOMMSlZojDGGJOSJQpjjDEpWaIwGSciUbea6GoReVpE8pNM96KIdNiP5fcQkWeaEN8GEemcoL1QRO4RkXVuRdFFInL0/q4nG4hIcbKqriJygFtBdbeI3NXSsZnMsURhskGFW010IBABpsa/KQ6Pqp6sqjsau3BV3aSqE9MVbJz7cQrf9VPVw4EpOP0GclkxkKz8dyVwFfDrlgvHZANLFCbbLAa+JSJ93Nr/dwMrgF413+zj3rvP/Sb/ioiEAETkW269rpUiskJE+rrTr3bfnyIiz4nIy+6YHFfXrFhEnnWLCq6pr7CgiPQFjgZ+r6oxALci7wvu+7+KqxN2qdvWR0T+LSL3u+2Pi8hIEXnLHc/gKHe6a0Rkpoi85rb/1G0XEbnZnXeViExy24eLyBsi8oy7/MfdnsSIyHdFZKG7XfNqeke7098kIstE5CMR+Z7bU/86YJJ7hDcpfptVdY+qvomTMExboqr2sEdGH8Bu92ceTgmEnwF9cHrAHhM33Qacb+x9cMYHKHbbZwFnu8+XAhPc50Eg351+tds2Bacg3QFACKeMRIn7Xif3Z037AfHrrRXzD4A5Sbbnuzg9eAuAQmANTrXRmrgH4XxJWw48iFOocTzwrDv/NcBKN47OONV+ewCnA/NxetR3xekJ3B2nx/NOnLpdHmAJMAzwAW8DXdzlTsLpiQ/wBnCr+/xkYEHc7+euej6veqexR+t6WFFAkw1CIvKu+3wxTk2dHsCnqvrPJPOsV9WaeZYDfdx6Tz1VdQ6AqlYCuF+u481X1a/c9/6O80+1FLhERCa40/QC+gFf7cf2DMNJInvi1vE9nLpA61V1ldu+BmfQGxWRVTiJpMZzqloBVIjI6zhFDocBT6hqFKcY3ULgSGAXsExVN7rLfddd1g5gIDDf/R14cZJkjZrCdstrrduYfViiMNmgQlWL4xvcf2x7UswTjnsexfn2naiEeiK169aoiAwHRgJDVLVcRN7AOSJJZg0w2L12Eqv1Xqo44uOOxb2Ose/+WCfGRiw36i5LgDWqOqSeeWqmNyYhu0ZhWg11xgfYKCKnAohIIMkdVKPEGes4hDMi2VtAe2C7myT64wx3mWpd63COQq6Nux7QT5xxSxYBp4pIvjjVbCfgHCk1xngRCYrIATinlt5xlztJRLwi0gU4DliWYhkfAl1EZIgbn09EDq9nvWVAUSNjNa2cJQrT2pyDcwrpPZzz890STPMmMBN4F5itqqXAy0CeO9/1QLJTXvHOd5e/1j11dB+wSZ0hMx/G+Se+FLhfVf/VyO1YBrzgxnG9OuNtzAHew7l+8RrwW1X9ItkC1BnqdiJwk4isdLd3aD3rfR0YkOhiNji3CgO3AVNEZKO0nvHWTQpWPda0KSIyBefi9c8zHUsyInINzgX+WzIdizFgRxTGGGPqYUcUxhhjUrIjCmOMMSlZojDGGJOSJQpjjDEpWaIwxhiTkiUKY4wxKf0XrEA5yV+tt9AAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "y_pred = KMeans(n_clusters=3, random_state=170).fit_predict(X_scaled)\n",
    "print(y_pred)\n",
    "plt.scatter(X_scaled[:, 0], X_scaled[:, 1], c=y_pred)\n",
    "plt.title(\"kMeans Clustering on PCA_transformed X variables\")\n",
    "plt.xlabel('Principal Component 1')\n",
    "plt.ylabel('Principal Component 2')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([   0,    1,    3, ..., 5204, 5205, 5206]),)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# group1 = y_pred.index(y_pred == 0)\n",
    "y_pred\n",
    "ind1= np.where(y_pred == 1)\n",
    "ind1\n",
    "# group1 = X_scaled[ind1,:]\n",
    "# group1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:8: DeprecationWarning: \n",
      ".ix is deprecated. Please use\n",
      ".loc for label based indexing or\n",
      ".iloc for positional indexing\n",
      "\n",
      "See the documentation here:\n",
      "http://pandas.pydata.org/pandas-docs/stable/indexing.html#ix-indexer-is-deprecated\n",
      "  \n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:10: DeprecationWarning: \n",
      ".ix is deprecated. Please use\n",
      ".loc for label based indexing or\n",
      ".iloc for positional indexing\n",
      "\n",
      "See the documentation here:\n",
      "http://pandas.pydata.org/pandas-docs/stable/indexing.html#ix-indexer-is-deprecated\n",
      "  # Remove the CWD from sys.path while we load stuff.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.348 (+/-0.219) for {'n_estimators': 50}\n",
      "0.358 (+/-0.188) for {'n_estimators': 60}\n",
      "0.351 (+/-0.238) for {'n_estimators': 70}\n",
      "0.335 (+/-0.218) for {'n_estimators': 80}\n",
      "0.335 (+/-0.230) for {'n_estimators': 90}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.preprocessing import Imputer\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "cluster_data=df.ix[:,2:]\n",
    "X=cluster_data\n",
    "y=df.ix[:,'price']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)\n",
    "\n",
    "tunedParameters=[{'n_estimators':range(50,100,10)}]\n",
    "clf=GridSearchCV(RandomForestRegressor(n_jobs = -1, criterion='mse'), param_grid = tunedParameters,cv=10)\n",
    "\n",
    "clf.fit(X, y)\n",
    "\n",
    "means = clf.cv_results_['mean_test_score']\n",
    "stds = clf.cv_results_['std_test_score']\n",
    "for mean, std, params in zip(means, stds, clf.cv_results_['params']):\n",
    "    print(\"%0.3f (+/-%0.03f) for %r\" % (mean, std * 2, params))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:2: DeprecationWarning: \n",
      ".ix is deprecated. Please use\n",
      ".loc for label based indexing or\n",
      ".iloc for positional indexing\n",
      "\n",
      "See the documentation here:\n",
      "http://pandas.pydata.org/pandas-docs/stable/indexing.html#ix-indexer-is-deprecated\n",
      "  \n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:460: UserWarning: Some inputs do not have OOB scores. This probably means too few trees were used to compute any reliable oob estimates.\n",
      "  warn(\"Some inputs do not have OOB scores. \"\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n",
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:465: RuntimeWarning: invalid value encountered in true_divide\n",
      "  predictions[k].sum(axis=1)[:, np.newaxis])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tjhuynh/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:40: DeprecationWarning: \n",
      ".ix is deprecated. Please use\n",
      ".loc for label based indexing or\n",
      ".iloc for positional indexing\n",
      "\n",
      "See the documentation here:\n",
      "http://pandas.pydata.org/pandas-docs/stable/indexing.html#ix-indexer-is-deprecated\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
       "           max_features='auto', max_leaf_nodes=None,\n",
       "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "           min_samples_leaf=1, min_samples_split=2,\n",
       "           min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=1,\n",
       "           oob_score=False, random_state=None, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X=cluster_data\n",
    "y=df.ix[:,'price']\n",
    "from collections import OrderedDict\n",
    "from sklearn.ensemble import RandomForestClassifier, ExtraTreesClassifier\n",
    "ensemble_clfs = [(\"RandomForestClassifier, max_features=None\",\n",
    "        RandomForestClassifier(warm_start=True, max_features=None,\n",
    "                               oob_score=True,\n",
    "                               random_state=0))]\n",
    "\n",
    "# Map a classifier name to a list of (<n_estimators>, <error rate>) pairs.\n",
    "error_rate = OrderedDict((label, []) for label, _ in ensemble_clfs)\n",
    "\n",
    "# Range of `n_estimators` values to explore.\n",
    "min_estimators = 10\n",
    "max_estimators = 100\n",
    "\n",
    "for label, clf in ensemble_clfs:\n",
    "    for i in range(min_estimators, max_estimators + 1):\n",
    "        clf.set_params(n_estimators=i)\n",
    "        clf.fit(X, y)\n",
    "\n",
    "        # Record the OOB error for each `n_estimators=i` setting.\n",
    "        oob_error = 1 - clf.oob_score_\n",
    "        error_rate[label].append((i, oob_error))\n",
    "\n",
    "# Generate the \"OOB error rate\" vs. \"n_estimators\" plot.\n",
    "for label, clf_err in error_rate.items():\n",
    "    xs, ys = zip(*clf_err)\n",
    "    plt.plot(xs, ys, label=label)\n",
    "\n",
    "plt.xlim(min_estimators, max_estimators)\n",
    "plt.xlabel(\"n_estimators\")\n",
    "plt.ylabel(\"OOB error rate\")\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.show()\n",
    "\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "X=cluster_data\n",
    "y=df.ix[:,'price']\n",
    "tunedParameters = [{'n_estimators':100}]\n",
    "\n",
    "clf2 = RandomForestRegressor(n_jobs = 1, criterion='mse', n_estimators=100)\n",
    "#Fit Model\n",
    "clf2.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "FeatImp = pd.DataFrame({'feature': list(X.columns), 'importance': list(clf2.feature_importances_)})\n",
    "FeatImp = FeatImp.sort_values('importance', ascending = False)\n",
    "#Set Index To Field You want to Sort Bar Chart By\n",
    "FeatImp = FeatImp.set_index('feature')\n",
    "FeatImp.head(100)\n",
    "FeatImp.to_csv('feature_imp.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['bedrooms', 'bathrooms', 'cleaning_fee', 'Boat', 'extra_people',\n",
       "       'security_deposit', 'minimum_nights', 'beds', 'availability',\n",
       "       'availability_365', 'Entire home/apt', 'number_of_reviews',\n",
       "       'Hair dryer', 'within a day', 'maximum_nights', 'Elevator in building',\n",
       "       'Kitchen', 'First aid kit', 'guests_included', 'Suitable for events',\n",
       "       'Laptop friendly workspace', 'Breakfast', 'Lock on bedroom door',\n",
       "       '24-hour check-in', 'Iron', 'within an hour', 'review_scores_rating',\n",
       "       'Buzzer/wireless intercom', 'Shampoo', 'Indoor fireplace'],\n",
       "      dtype='object', name='feature')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "FeatImp.index[0:30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.12157143, 0.09678738, 0.08556375, 0.08517269, 0.0843931 ,\n",
       "       0.07127343, 0.04658852, 0.04306044, 0.04182416, 0.03365694,\n",
       "       0.03096154, 0.01717346, 0.01413761, 0.0112159 , 0.0105321 ,\n",
       "       0.01033861, 0.00834966, 0.00833806, 0.00756614, 0.00715891,\n",
       "       0.00688507, 0.00571139, 0.0056662 , 0.00546585, 0.00524451,\n",
       "       0.00502909, 0.0044724 , 0.00440612, 0.00421355, 0.00419603])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "FeatImp['importance'].values[0:30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.figure()\n",
    "\n",
    "plt.title(\"Feature Importance\")\n",
    "y_pos = np.arange(len(FeatImp.index[0:30]))\n",
    "plt.bar(y_pos,FeatImp['importance'].values[0:30])\n",
    "plt.xticks(y_pos, FeatImp.index[0:30],rotation='vertical')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "54.4"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1224x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig=plt.figure(figsize=(17,10))\n",
    "# df['price'] = df.price.str.replace(\"\\$|,\", \"\").astype(float)\n",
    "# df['price'].hist()\n",
    "sns.distplot(df['price'])\n",
    "# plt.show()\n",
    "mean_price = df.price.iloc[:5].mean()\n",
    "mean_price\n",
    "# df['host_acceptance_rate'].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1ab7541ef0>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['review_scores_rating'].fillna(0, inplace=True)\n",
    "df['review_scores_rating'].dropna(axis=0, inplace=True)\n",
    "sns.distplot(df['review_scores_rating'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1ab78fa470>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Correlation Plot\n",
    "var = 'review_scores_rating'\n",
    "data = pd.concat([df['price'], df[var]], axis=1)\n",
    "data.plot.scatter(x=var, y='price')\n",
    "data.plot.scatter(x=var, y='price', ylim=(0,1500))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABmEAAADmCAYAAAAtBiGAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJzs3VlsY4l+JvbvLDzctHGXSElctJdKVSVVV5XauUCAwOP7EMCZuZj4IUaSt2AAG3cm3gAHPZMAjjOGB54k4yQD5DXLxDOOndix4SAzmcwdx7f7tlRVXdVVJWqhJFILtS8UKZLn8Jw8VM5pUiIlklLdVnd/P6Dh69bZSOm+nO9+/79gGAaIiIiIiIiIiIiIiIjodolf9wMQERERERERERERERF9GzGEISIiIiIiIiIiIiIi+gAYwhAREREREREREREREX0ADGGIiIiIiIiIiIiIiIg+AIYwREREREREREREREREHwBDGCIiIiIiIiIiIiIiog9Avubnxk/lKegbTdd1CIIAQRC+7kehu4l/GERERERERERERPSddF0IQ9SQYRioVCo4Pz+HYRhwuVyQJIlhDBERERERERERERERAMEwriy7sAlDl5jhi6ZpMAwD5XIZuq4DAGw2GxwOB2SZ+R5ZmMoRERERERERERHRdxJDGGqJrutQVbVmBFmpVIJhGBAEwQpjZFmG0+mEJElf8xPTHcAQhoiIiIiIiIiIiL6TxK/7AeibwTAMqKqKZDKJ/f39ujtgBEGAJEkQRRGapuH09BT5fB6VSuVremoiIiIiIiIiIiIioq8PZ0bRlQzDgKZp0DQNAKwRZFftfTHDGDO4UVUViqLA4XBAFJn7EREREREREREREdF3A0MYquvi3hez+SIIAqpH2BmGgaOjI9jtdrjd7pprVIcx5XIZ5XKZYQwRERERERERERERfWcwhKEahmFA13VommbtfWkUmJyeniKZTEKWZZyfn6OzsxOxWAx2u73muOowplQqoVwuw+FwQFEUhjFERERERERERERE9K0lVLca6rjyh/Ttous6VFW1wpd6I8eWl5fhdDpxeHiI8/NzjI2NweFwQNd1HB4eYm1tDR6PB9FoFDabre59DMPA27dvkUgk0NXVBbvdfuV4M/rG4y+XiIiIiIiIiIiIvpPYhKFLe18aBTCVSsUKX8bHxxEMBiEIAkqlEgRBQCAQgN/vRzabxYsXLxAIBDAwMABZrv0zEwQBqqpCFEUUi0WUSiWrGcMwhoiIiIiIiIiIiIi+LRjCfIc1G74YhoGtrS2sra3B4XBgdHQUoVCo7jUFQUBfXx9CoRC2trbw/Plz9Pb2IhKJQJKkmmtWjykrFAooFosMY4iIiIiIiIiIiIjoW4MLOb6DzPClVCpB0zRr70u94OPw8BCfffYZTk9P8eTJE/T09DQMaqqJooj+/n48fvwYhmFgfn4em5ub0HXdOsa8jiAIVlvm/PwcuVwO5XL50jWJiIiIiIiIiIiIiL5J2IT5DjEMA4Zh4ODgAB0dHVb4Uk8+n0cymYQgCJiamoLb7W7rnpIkIRqNIhKJIJ1OY25uDoODg9bemWrms+i6jnw+D1mWYbfbYbPZ2IwhIiIiIiIiIiIiom8chjDfEbquQ1VV6LqOL7/8Eh9//HHdYKNcLmNlZQUnJycYHR2F1+ut+bkgCG01VGRZRiKRQH9/P9bX13F2doaDgwNrr0w1s5Wj6zoKhQIkSYLT6YQkSQxjiIiIiIiIiIiIiOgbgyHMt1y9vS/V/9ek6zrS6TQ2NzcRj8cxPj7+QQIPRVEwMjKC4+NjHBwcYGNjA4lEAh6Pp+Y4cz+NYRjQdR25XA42mw0Oh8MaXUZEREREREREREREdJfxbfa3lGEYqFQqUFUVwFehRr3jdnd3sbKyglAohNnZWUiS1PC67TZh6l1nbGwMpVIJqVQK6+vrSCQS6OrqunSc+c/GxgYMw8DAwIDVjCEiIiIiIiIiIiIiuqsYwnzLmOGLpmkwDKNh+AIAJycnSCaTcLlcePz4Mex2+0/1WQVBgMvlwv3795HL5ZBKpSCKIhKJxKUdNGb4IwgCNE3D6ekpFEWBw+FgGENEREREREREREREdxJDmG+R6r0vgiBYi+7rHffq1SuUSiVMTEygs7Oz6Xvc1ogyM1AxdXZ24uHDhzg+PkYymYTD4UA8HofT6aw5R5IkSJIEwzCgqipUVbXCmEafl4iIiIiIiIiIiIjo68AQ5ltA13VomoZKpVIzvusiTdOwurqKQqGAsbEx+P3+tkKV2xhH1khPTw+mp6dxeHiIN2/eoLOzE7FYDHa73QqXgPdhkBnGlMtllMtlhjFEREREREREREREdKcwhPkGMwwDmqZB0zQAV+992dzcxPr6OgYGBuB2uxEIBH7aj3tJowBIEAT4fD54vV7s7e3h1atX8Hg8EAShphljHmuGMaVSCeVyGQ6HA4qiMIwhIiIiIiIiIiIioq8VQ5hvoFb2vuzv72NpaQlerxdPnz6FzWbDxsZG2/c2d7P8NAiCgGAwCL/fj52dHaysrKBYLMLr9UKW5UvHyrIMwzBQLBZRKpVgt9tht9tvbYQaEREREREREREREVErGMJ8g5jhy9bWFrq7u2G32xu2Pc7OzpBMJiFJEh4+fAiXy3XpWrcVTuTzeUiSdKmlcltEUURfXx/y+TwqlQqeP3+Ovr4+hMNhSJJUc2x1M+b8/BylUslqxjCMISIiIiIiIiIiIqKfJoYw3xC6rkNVVei6jp2dHbhcLjgcjkvHlctlLC8vI5fLYXR0FB6Pp+712g1hqpswxWIRS0tLyOfz0DQNfr8fg4ODl1oqtykQCGB4eBiZTAbz8/Po7+9Hb2/vpTCquhnz6aef4tGjRwxjiIiIiIiIiIiIiOiniiHMHWcYBlRVRaVSAfA+XBBFEbqu1xxXqVSQTqextbWFRCKBiYmJhmGDKIo3Gimm6zqWl5exu7uL4eFhdHV1oVKpYHd3F8+fP0dvby8ikcillspN6bpuNV1isRgikQjS6TTm5uYQjUYRDAYvfWZBEKzvrroZY7PZGMYQERERERERERER0QfFEOaOMgwDmqZB0zQAqNn7IkmSFSwYhoFsNotUKoW+vj7Mzs5eG360u9fFMAycnJxgf38f8Xgcs7OzEEURpVIJoigiEomgt7cXGxsbV7ZU2mUYRs21bDYbhoaGMDAwgLW1NWQyGcRiMfh8vksBi3meruvI5/OQZRl2u51hDBERERERERERERF9MAxh7hhz74umadbIsHqBgq7rOD4+RjKZRGdnJ548eQJFUZq6R70mzXXMewmCgGg0ilgsVvc4SZIQjUYRDoeRTqcxPz+PWCwGv99/47DDbMJcpCgKRkdHcX5+jrW1NaTTacTjcXg8nkvniKIIQRCg6zoKhYK1y0aSJIYxRERERERERERERHSrGMLcEYZhWHtfzPClUYNE13WsrKxAlmVMTk6io6OjpXu10oQ5Pz/H4uIiNE3D5OQkjo6OmjrXbKn09/fXBCNer7elZ612sQlzkdPpxMTEBPL5PFZXV7G+vo5oNFp3X4z5Hei6jlwuB5vNBofD8UH32RARERERERERERHRdwvfON8BZvhitjYaBQ2qqiKVSiGbzSISiWB0dLSt+zUTwmiahtXVVezt7WFkZASBQADA+0ZMKy0au92OsbExFAoFrK6uIp1OI5FItPXcZjh1Hbfbjfv37yOXy2F5eRn5fB75fB5ut7vmODOMEQQBmqbh7OwMsixbzRgiIiIiIiIiIiIioptgCPM1umrvSzVd17GxsYFMJoPBwUHEYjHYbLa273vVODLDMLC1tYW1tTUMDAxYe19uyuVyYXJyErlcDqlUCufn53WDkavout7Ss3R2dmJiYgJv3rxBMpmEw+FAPB6H0+msOU4QBEiSZP0+Tk9PoSgKHA4HwxgiIiIiIiIiIiIiahtDmK9Bs+GLYRjY39/H0tISAoEAnj17BlmWkclkWt7pUq1RE+bw8BCLi4vo6enB06dP6wY9rYwyq6ezsxMPHz7Ej3/8YywsLMDtdiMWi8HhcFx7brNNmGq6rsNut2NychKHh4d48+YNurq6EI1GYbfba46tDmP29/eRz+etZ7uNIIqIiIiIiIiIiIiIvlsYwvwUGYaBSqWCjY0NdHR0oKOjo2GokMvlkEwmoSgKpqena9oboihCVdW2n+NiE6ZQKCCZTMIwDExNTbXUTmmXLMuYmZnB/v4+Xr9+DY/Hg2g0emXDp50QplKpQJIkCIIAn88Hr9eLvb09vHr1Cl6vF4ODg5fuaY4n0zQN5XIZ5XLZasYwjCEiIiIiIiIiIiKiZjGE+SkwDAOGYVh7X05PTyHLMjo7Oy8dWyqVsLS0hEKhgLGxMXR3d186RpIkFIvFtp/HbLNomoaVlRUcHh5idHQUPp+vqXNviyAICAQC8Pv9yGazePHiBYLBIAYGBuqOAWt1HFm9cwRBQDAYvPaelUoFsixbzZhSqYRyuQyHwwFFURjGEBEREREREREREdG1GMJ8YLquW+GLOXZMlmVUKpWa4yqVCtbW1pDNZjE8PIxgMNgw8Lhqp0uzstks9vb2MDg4iNHR0ZbClZuMI6tHEAT09fUhFAphc3MT8/PzCIfDCIfDNWFHu+PI6gUmoigiHA6jt7fXumdfXx8ikQhEUbQaNObzybIMwzBQLBZRKpVgt9tht9tvNZQiIiIiIiIiIiIiom8XhjAfyFV7X2RZtv69YRjY3t7G6uoqIpEIPv7442tbFpIktR3CHBwcYG9vDz6fz9oxc1eIooiBgQH09fUhk8lgbm4Og4ODCIVCEAThVpowV91zY2MDc3Nz6O/vh6ZpNSPggNqdMefn5yiVSlYzhmEMEREREREREREREV10d97Af0tcFb6YJElCpVLB4eEhFhcX0d3djSdPnkBRlKbuYTY1WpHP55FMJiGKIvx+P/r7+9sKYMxRZh+SLMuIx+OIRCJYX1/HxsYGYrGY1SZqRbPBjSzLiMViiEQiSKfT2N7eRm9vb91GUnUzplAooFgsMowhIiIiIiIiIiIioksYwtwSwzBQqVSgaZo1NqvRC/lKpYJMJgOXy4WpqSm43e6W7tXKODJVVbGysoLj42OMjo7C6/UimUzeeJzZT4OiKBgZGUGxWMTq6irOz89xcnKCnp6epq9RPVasGTabDUNDQ1BVFYVCAfPz84jH4/B6vQ3DGF3Xa5oxNpuNYQwRERERERERERERMYS5DRf3vjRqXpiByN7eHrq6uvDw4cO27mc2aa57po2NDWQyGcRiMYyNjVnBwE3aLBfPNQMlm82G3t7eDxI+OBwOTExM4OTkBJlMBuvr6xgaGkJHR8e157YzwswUi8Vgs9mwurqK9fV1JBKJugGQeX1d15FOp+FyueD3+xnGEBEREREREREREX3HMYS5AV3XoWkaKpWK1Xyp99LdfDm/ubmJWCyGYDCIbDbb9n2va8Ls7e1haWkJgUCg7t6X2xgpZhgGdnZ2sLKygmAwiNPTU2xubiKRSMDr9d7o2o1IkoSpqSmcnp5iaWkJiqIgkUhc2t1Srd0QxmzQOJ1O3Lt3D/l8HqlUygpjOjs7L50jiiJOT08hSRIKhYJ1viRJDGOIiIiIiIiIiIiIvoMYwrShmb0v5nG7u7tWUDE7OwtJknB6etryTpdqkiTVDWHOzs6QTCYhyzKmp6cbhhOtjDO7SBAElEolfP7553C5XPjoo48AvA87yuWyFVQMDQ2hq6urrXtcp6urC9PT0zg8PMSbN2/Q2dmJWCwGu91+6Vhd19vafXNxjJnb7bYCoJWVFWtvzcVRcpVKBYqiWN9xLpeDzWaDw+Fo6zmIiIiIiIiIiIiI6JuLb4VbYO59WV1dRSgUunIR+8nJCRYXF+FwODAzMwOHw2H9rJlxYlcRRbHm/HK5jOXlZeRyOYyNjV27M6XdJkypVEI6nUahUMD09LQVspRKJQCA0+nE5OQkcrkcUqkUJElCIpGAy+Wquc5NWzgmr9cLj8eDvb09vHr1Cl6vF9FotCbs0HW9pZ0wpka7ZLq6uvDo0SMcHR0hmUzC6XQiHo9bv19N0yDLck0zStM0nJ2dQZZlqxlDRERERERERERERN9+DGGaYIYvmqbBMAwcHh7C5/PVbV4Ui0UsLi6iVCphbGysbhtElmWrRdMOswlTPeYsHo9jYmKiqbFXrTZhdF3H+vo6tra24Pf70dPTc2XLpbOzEw8fPsTR0RHevXuHjo6Ohk2VmxIEAcFgEH6/H9lsFs+fP0coFEJ/f78Vdt1kHFkjHo8HPT09ODg4wJdffomuri7EYjFUKpWaEEgQBEiSZLWnTk9PoSgKHA4HwxgiIiIiIiIiIiKibzmGMNfQdR2qqkLXdQiCAFEUYbPZoKpqzXGapmF1dRV7e3sYHh5GIBBoGIjctAkDvG+ffPrppwiFQtaYs2Y124QxDAN7e3tYXl627nNwcICTk5Om7uPxeDAzM2M1VXw+HwYHByGK4q3vSBFFEeFwGKFQCJubm5ifn0ckEmk7hGlml4wgCPD7/fD5fNjd3cXLly+hqmrd77Y6jFFVFaqqWmFMO89HRERERERERERERHcfQ5gGzJflZlhSvfelusliGAY2Nzexvr6O/v5+zM7OXvtS/SYhTC6Xw8LCAjRNw7Nnz2rGnDVLEIRrmzBnZ2dYWFiAoiiXxqm1ei+zqbK9vW01VT4USZIwODiIcDiMdDqNnZ0dKIqCnp6eloKfVo8NhUIIBAL48Y9/jC+++KKmjXPxWDOMKZfLKJfLDGOIiIiIiIiIiIiIvqUYwlxgjo0yQ5bq8MVkNmEODg6wuLgIr9eLp0+fwmazNXWPdlogpVIJS0tLKBQKGBsbw5s3b9oORi7ulKlWvV9mfHwc3d3dN352856RSAS9vb1Ip9M4OzvD9vY2ent7b70VA7wPyhKJBAqFAs7OzjA/P494PA6v19vU/drZWyOKohVabW1tYW5uDpFIBOFw+FLAUh3GlEolHB0doaOjA52dnQxjiIiIiIiIiIiIiL4lGML8/y7ufakXvpgqlQpSqRTcbjcePnx4afH8bdJ1HWtra9je3sbQ0BBCodCNQ4t648h0XUcmk8HGxsa1+2XaCShMkiRhYGAABwcHLYUj7d5TEATEYjGIoohUKoV0Oo1EInEpXLpN5mfs6+tDJpPB3NwcBgYG6gZOgiBAlmVsb28jEAgAAOx2O+x2+wcJp4iIiIiIiIiIiIjop+c7H8LUC18aNRHMlsjBwQG8Xi8mJyc/6HPt7OxgZWUFfX19Le99uYooijXjyPb397G4uIhAIHCr92nEMAyIooiRkREUi8WmwhFzJ0+rKpUKJEmCw+HA5OQkzs7OkEqlIAgCEokE3G73rd3rYlAkyzLi8Tj6+/uxvr6Oubk5RKPRuvuCNE2zgpfz83OUSiU4HA4oisIwhoiIiIiIiIiIiOgb6jsdwui6DlVVrZfujcIXXdexvr6Ora0tJBIJ+P1+HB8f3/j+Zuhz0cnJCZLJJFwuFz766CPY7fYb36ua2YTJ5/NYWFiALMuYnp6G0+ls+tzbeAYAcDgcuHfv3rXhiBnctErX9ZrzOjo68ODBAxwfHyOZTMLpdCIej9eMdtN1va0gStO0uufZbDYMDw+jVCphbW0N6XT6UvtHVVXYbDarGWMYBgqFAorFIsMYIiIiIiIiIiIiom+o72QI08zeF/O4em2U4+NjqKp6o2eQJAmVSgWy/NWvoFgsYmlpCcViERMTE+js7Lz2c7Tb2Njd3cXe3h7Gxsbg8XhavsZN1Hvu6nBkYWEBbrcb8XjcCqDa/awXQxhTT08PpqencXBwgC+//BLd3d2IRqNQFMVqz7RK07Sa3+dFdrsdY2NjOD8/x+rqak375+K5Zhij63pNM8YMaoiIiIiIiIiIiIjo7vtOhTDmEvTd3V1rJFSjF9pmU6Kjo+NSG0WWZSvAaZcsy1YIU6lUsLa2hp2dHQwPD9cdV3VRvRDnOoZhYGNjA6lUCh0dHZiZmWn5hf5tN2Eu6unpwczMDPb39/Hq1St4vV4MDg42DFOuc9V5giDA7/fD5/NhZ2cHL1++RCAQgN/vbyuEafb34XQ6rfbP6uqq1ciq95zmv9N1Hfl8HrIsw263M4whIiIiIiIiIiIi+gb4ToQw1XtfNE3D6uoqQqFQ3WPPz8+xuLgIVVVx7969um0Um812K00YVVVxeHiIVCqFSCSC2dnZpoOGi3tdrnN4eIhkMgmfz4fx8XEcHx9/bS/xrwtxBEFAIBCAz+dDNpvFixcv4PP52r7Xdd+pIAjo7e1FMBjE1tYWvvzyS9hstpaDn+uaMBd1dHRgamoKp6enePHiBb788kskEgm4XK5Lx4qiCEEQoOs6CoUCJEmC0+mEJEkMY4iIiIiIiIiIiIjuqG91CGMYBnRdh6Zp1t4X8+X6RZqmIZVK4eDgwGqjNHIbIUylUsGrV6/Q09ODJ0+eQFGUls5vNoQpFApIJpMAgIcPH8LlcuHw8LClAKdavSbM+fk5BEFoaqcM0PxoMVEUEQ6HEQqFsLq6ipOTE2xvb6O3t/eDBA+iKKK/vx8OhwNra2uYm5vDwMBA0/drNYQxdXZ2wuVyIRwO4927d3C73YjFYjV7aoCvxuaZf9e5XA42mw0Oh6Ot+xIRERERERERERHRh/WtfXNrjngyw5dGo8d0Xcfm5ibS6TQGBwfx7Nmza9sPoii2PZLLbNrkcjmMjIwgEom0dR1zHFkj1aHS6OhoTZPktkaKqaqK5eVlHB0dQdd1BAIBDA4OtjXK6yqSJKGvrw/n5+fI5/OYm5tDLBaD3+//IGGMIAjweDwYHBzE+vp60/drN4Qxz/N6vfB4PDg4OMDr16/R09Nj7am5+HzmP+l0GpIkIRKJWM0YIiIiIiIiIiIiIrobvnUhjGEY1tgxAA3DF8MwsL+/j+XlZfh8Pjx79uyDtgnMMWh7e3sYGRmB0+lsuf1SrVETxjAMbG1tYW1tDQMDA5idnb30+VsdZVbNDHAymQzS6TRisRji8TgqlQqy2Szm5+cRiUTQ19fXMMxqtglz8RxJkjA8PIxisYjV1VVkMhkkEgn09PS09VkaqVQqkCQJNpsNw8PDKJVKWF1dRTqdRiKRgMfjqXteuyGMqqqw2WwA6u+p8fv9GBwcvHRtQRCgqiocDgc0TcPp6SkURYHD4WAYQ0RERERERERERHQHfGtCmGbDF/Nnc3NzUBQFjx49anqMVrvPZYYi/f391t6XXC53ZZPlOvWaMEdHR0gmk+jp6cHTp0+tF/sX3aQJk8vlkM1mIUmSFVyVy2UYhoGBgQH09fUhnU5jfn6+YXuknRCmej+Lw+HAxMQE8vk8UqmUFY50dHS09Zku0jStJsSw2+0YHx9HoVCoud/FfUGaptXd53Kd6hDGVL2nZnt7G8+fP0coFEJ/f3/Ns5XLZdjtdkiSBMMwoKoqVFW1wphWdtoQERERERERERER0e36xocwhmGgUqlA0zTr5X6jF/ylUgnLy8s4Pz/HxMQE+vr62r6vKIpWY6IRMxTp7u6+FIpIkmQFRu3e32yzmCPONE3D/fv3rw0j2glhisUikskkzs/P4fP5MDY2Vvc4WZaRSCQQiURuta1SL7hxu92YmprCyckJFhcX4XQ6EY/HL+1SaVWj36vL5cL9+/dxenqKlZUV67Oawct1fw+NXNWgEUURkUgEvb292NzcxNzcHCKRCMLhMERRvNSiMcOYcrlsBTR2u51hDBEREREREREREdHX4Bsdwlzc+9LoRXOlUsHa2hqy2SyGhoYgCMKNX9TLsnypMWEqFApYXFyEruuYmpqC2+2+dMxthDDmTpbd3V2MjIwgEAg0fW6z48jM725nZwcjIyOw2WzY2tq69jyzPZLP57GysoJ0Oo2hoSG43e4bN2Eu6u7uxvT0tLVLxePxIBqNQpKktnbGVCoV2O32hj/v6urCo0ePcHR0hHfv3sHtdiMej9/KOLJGJEnC4OAgwuEwMpkM5ubmMDg4iFKpVHdnjBnGFItFlEolOBwOKIrCMIaIiIiIiIiIiIjop+gbGcLoug5N01CpVGqWlF9kGAa2t7exurqKcDiMjz/+GKIo4uzsDOVy+UbPYLPZoKpqzct6TdOwsrKCw8NDjI6OwufzNTxflmWUSqW27m0YBgqFArLZLGKxmDXirFnNNGEMw8DOzg5WVlYQiUSsexwfH7fUonG73Xjw4AGOj4+xsLAAt9uNYDDY9PnVz3NVoFK9SyWbzeLFixd1R6E1o1KpNPV9ejwezMzMYH9/H69evUKlUkFvb2/L91NVtenwRpZlxONxRCIRpNNpnJ2d4fDwsO5nFQQBsizXhDFmM6ad74WIiIiIiIiIiIiIWvONCmEMw0A+n4coileGL8BXo8C6urrw5MmTmraAoig3DmHMJoz5XBsbG0in0xgcHMTo6Oi1L7nr7XRpxsnJCRYWFqDrOhKJBAYGBlq+xnVNmFwuh4WFBTidTnz00Uc1QVO7+2R6enqswGJxcRGCILTUHGm2PSMIAvr6+hAMBq2QYnNzE319fU0HVa2MFRMEAYFAAH6/Hz/5yU+wsLCAUCiEgYGBpq+hqmrLzSxFUTA8PIz9/X0cHBwgnU4jHo/D6/XWfUazGXN+fl7TjGEYQ0RERERERERERPThfCNCGHPvi6qqeP36NcbHxxsuQK8eBdZoP4qiKCgUCjd6JrMJc3BwgMXFRXi9XmtRfTOqQ5xmlEolLC4uolQq4d69ezg8PGz7BXqjIKVcLmN5eRm5XA7j4+Po7u5u6/pX3TcQCMBut2NxcRHPnz9HX18fIpHItQHJVePI6pEkCX19fTg5OUGxWMT8/DxisVhT7ZhKpdLyWDGzdfLgwQNks9lLu1uuomnatePI6tF1HbIsY3x8HIVCAaurq1hfX0cikaj7u6tuxpyenuL4+BiDg4MMY4iIiIiIiIiIiIg+kDsdwpjhi6ZpVhPCbrejXC5fCmFUVcXKygqOjo6uHQVms9lu3ITRdR1LS0twOp14+PBhw1CokWabMJVKBevr68hmsxgeHkYgEIAgCDg+Pm56r8tFF0MYXdexsbGBTCaDeDyOiYny7j/VAAAgAElEQVSJD/pSXhAEuN1ujIyMWPtNotEogsFgw/u2u0dGlmUMDQ2hv78fa2trSKfTSCQS8Hg8Dc9rdhxZo/sNDAygr68P6XTa2t0SCoUaPn8zO2GuO8/lcmFychJnZ2dIpVIAgEQiUTeEFAQBqqri8PAQgUDAasbYbDaGMURERERERERERES36M6GMLquQ1VV6LoOQRCsl+J2u71ml4qu68hkMtjY2EA0GsXY2Ni1L5JvMo7MDHt2dnYQDAYxMTHR1nWua8JU72QJh8OX9r5IktT2Z6geR2Y2eXw+X1NNnnbHkVUzA5Xq/Sarq6vY2NhoGJC02oQBaseK2e12jI2NoVAoIJVKIZ1OY2hoqG5I0U4TxmT+7cmyjEQigf7+fqyvr2NjY8MaF3bx77OVnTDVyuVyzZg9AOjo6MCDBw9wcnKCpaUlKIqCeDx+KSQ098NIkgRd15HP5yHLMhRFYTOGiIiIiIiIiIiI6JbcuRBG13Vomma1RC7ufVEUBaVSCYZhYG9vD8vLywgEAi2NAjNHibX6XGZbJBqNoqurC8VisaVrVLuqCXN6eopkMll3J4tJFMW2dsqYI90WFhYgiiJkWW6pyXMbL+cvtloURbk2IGm3CXMxuHG5XLh//z5OT0+xtLQEu92OeDwOp9NpHdPKTpjrKIqCkZERnJ+fW+PChoaGasaFtTuO7KoGTXd3N6anp3F4eIh3797B7XYjHo9bf0vlctn6z+aOJV3X8cknn+Bnf/Zn8f3vf7+NT0tERERERERERERE1e5MCGMYBjRNs9ohF8MXk91ux/7+Pubm5uBwODAzM9PWUvNWWiR7e3tYWlqqCXv29/eRy+Vaum+1ek2YcrmMpaUl5PN5jI+Po6urq+H51W2WZp2enuJv/M1fQCqzjUqlglgkhD/93/6o5VFqN23CNFIvIEkkEnA4HLcWwpi6urrw6NEjHB4e4s2bN+ju7kY0GoWiKG2NI7vuO3E6nbh3717dcWHttHyA+k2Yi7xeLzweD/b39/Hq1Sv09PQgGo2iVCrB7XZbx5n/fctmswiFQi0/CxERERERERERERFddidCGMMwrHZLo/AFAIrFIjY2NnB6eorHjx9fGVJcRRTFpoKEs7MzJJNJyLKM6enpmrbEdePEmnkGM0TRdR3r6+vY2trC0NAQ7t27d23g0OxOGZNhGPi7f+8/RSYvYPSv/zIqFR3bn/0Zfud3/wH+/m//521/jnZcF6iYAcnBwQFev34Nj8cDm83WclvkunBDEAT4fD54vV7s7Ozg5cuXCAQCMAzjRqPPrnJxXJjdbm97t0+zu2QEQUAgEIDf77c+pyAINW0cUzabRTgcbut5iIiIiIiIiIiIiKjWnQhhAFz54lvTNKytrWF3dxcDAwMA0HYA04xyuYzl5WXkcjmMjo7W3VHSzkizauZuld3dXSwvLyMUCmF2drbpMVitNGFOT0+xsLCAd8kleGKTkCQZuq6iY2AMC4tLLT/3Rfl8HgBqmhVXaabVIggC/H4/fD4ftre3kUql0NPTg97e3qYDkmYbJoIgoLe3F8FgEJubmygUCtjY2EA4HG76XpqmtbTXpbu72wqadnd3sbi4iFgsdm2zpVq5XG6pxVT9OT///HMsLi4iHA4jEolYf3f7+/vw+/1NX5OIiIiIiIiIiIiIGmt9BtIHdLGdYhgGNjc38dlnn8Fms2F2dhbhcLjthfTVzB0Y1XRdx9raGj7//HP09PTg6dOndQMY4OZNmLOzM5yfnyObzWJmZgZDQ0Mt7SExF6pfpVwu482bN1hYWMD4+Dj+jZ+Zxcnaa+h6BYZewenaazyaut/ys5u/p3K5jLdv3yKZTOLt27dYXl6+UTBVjyAIVlAAAHNzc8hms001mVrd7SKKIgYGBuByuVAulzE3N4ednZ2m7qVpWst7ZMw2SldXF7q7u/Hy5UukUqmm/66aGUdWjyiKkCQJMzMzMAwDc3Nz2NzcRKVSaasFRERERERERERERET13YkmTL0RZAcHB1hcXITH48HTp09rxi7dxk4Ss8lit9thGAb29vawvLyMYDDYVCOl3SZMdcvGbrfjwYMHbT2/KIoNx5Hpuo50Oo3NzU0kEglrvNmv/+qv4OUXr/DiD/8hNE3DR9MP8Ou/9qst39swDGQyGaTTacTjcQwNDaFSqWB3dxcvXrxAX18fIpFIw5f57ex3EQQBwWAQHo8Ha2tr2NjYQCKRgNfrbXhOu7tWBEFAIpFAf3+/da94PH7lvSqVSktNGJOqqlAUBaFQCIFAANvb23j+/Dl6e3trGiqNzm11RJtJ13UoioJoNIpIJIJkMokf/OAHcLvdbX9vRERERERERERERFTrToQw1fL5PJLJJARBwMOHDxuOW2rnRX41RVFQLpdRLpeRTCahKApmZmbgcDiaOr/ZvTImXdeRyWSsF/oTExP48Y9/3PbnaDSObH9/H4uLi3XDJJfLhT/6Z3+AtbU1zM/P4wc/+EHLL9tzuRwODg7gdDrx7NkzyLKMcrlsNVZCoRAymQzm5uYQjUYRDAYvfb52QjQzGLDZbBgZGcH5+TlSqRTS6TSGhobQ2dnZ8JxW72M+r6IoGB0dRaFQwOrq6pX3anUcWb3zRFFEJBJBb28vNjY2MD8/j/7+/oYj2Nptwlz8/mVZxuTkJH7/938fv/RLv4TZ2Vn81m/9Fr7//e+3fG0iIiIiIiIiIiIi+sqdCWEqlQoWFxdxenqKsbGxhmPAgK9GgbXbAgDev/BeXFxEpVLB2NhY3SXlt8UMRgKBgBVcAO9HirXboDDPNRUKBSwsLEAURUxPT8PpdNY9TxAExONxbG9vtxRQFItFLC4uolgsoru7G2NjYw2fKxaLIRwOY3V1FRsbGxgaGkJPT8+l52jFxbDK6XRicnISuVwOKysrsNlsSCQSNZ9b1/WWv9t6I8xcLtele8Xj8ZqAsN0Qpl6bRZIkRKNRhMNhpNNpzM/PIxqNIhAI1HwH7Y4Oa9Sg0TQNP/dzP4ff+I3fwMLCQsvXJSIiIiIiIiIiIqJadyaEMfdjTExMXPuC3m63o1QqtRXC6LqO9fV17O3toa+vD2NjYzdq1FzFbPU0CkbMMKmdl/dmE0bTNPzFX/wFvvzyS3zve9/D9773vdt6fABffV/b29sYHh5GR0cHksnktecpioKxsTEUCgWsrKxYLRK3291W+6fROZ2dndaC+zdv3qC7uxuxWAw2mw2VSqXlkOKqPTLmvQ4PD/H27Vt0dXUhFotBUZRbDWFMNpsNQ0ND1li0TCZTMxat3bF85XIZdrv90r/f2tpCOBxGf38/+vv727o2EREREREREREREX3lzoQwNpsNfX19Tb2cN0eJtcIwDOzu7mJ5eRl9fX1IJBJ1d9G0wtzLcvGlvaqqWFlZwfHx8ZWtnottllbvfX5+jl/79d/AH/zhH6ErnMA/+sf/PX7jV/4O/s7f/mFb17xob28PS0tLCIVCmJ2dhSiKKBQKLV3D5XJhamoKx8fHWFhYQEdHB7q6ulp+lutGi/l8Pni9Xuzs7ODFixcIhULQNO3a3T4XXRXCmLxeLzweD3Z3d/Hy5Uv4/X4IgtBwdN5VVFW9Nryx2+1WoLW6uor19XXE4/G297aUSqW6Y8yy2SzDFyIiIiIiIiIiIqJbdGdCmFZeKJtNmGadnp5iYWEBLpcLH330Eex2O7LZLPL5fDuParHZbDUv+g3DwMbGBtLpNKLR6LUtG7MJ06qTkxMsLCxgd3cX/+QP/ilG/vrfhtLRjfLZCX7nd/8BfuHf/ZsIh8Ntfy5ztJkkSZf25LQbWvX09GBmZgZ7e3tYXl6Gw+FoKvAwNdOeEQQBvb29CAQC2NzcRDabhSAI8Hq9TT93s88kCAJCoRACgQC2trawurqKQCCAQCDQ0t+ypmlN7yGqHou2vLyMYrGIfD4Pt9vd9P2Axk2Y7e1tPHv2rKVrEREREREREREREVFjdyaEaeXlfrMhTLFYxNLSEorFIsbHx2saGIqi4OjoqK1nNcmyDFVVYbfbcXh4iGQyCa/XW7P35SqtNmFKpZK1l+XevXt49+4d3D1+KB3v99koHd1wdfuQzWbbCmE0TUMqlcLBwQHGxsassVcXtTsGSxAEBINB6LqOnZ0da/F8Mw2o65ow1SRJwuDgIPL5PEqlEubm5hCPx+Hz+a69TyvBEPA+POzv70culwMAzM3NYXBwEKFQqKm/6avGkTXS2dmJoaEhpFIpJJNJOJ1OxOPxpsOcUqlUd2fQ9vY2IpFIS89CRERERERERERERI3dmRCmFXa7HWdnZw1/XqlUsLa2hp2dHQwNDSEYDF56Ia4oClRVvdFz2Gw2nJ2dYWlpCQDw8OHDlkZSybLcVAhj7mXZ2trC8PCw9XkikQjUwimOM0n0DIzhOJOEWjjF8PBwU/c32yWGYSCbzSKVSmFgYADPnj1rGHjcxv4cURTR3d2N/v5+rK+vNxWStLNHBgAGBwdhs9mQSqWQyWQwNDR05Ti0VkMYk67riEajUBQF6+vr2NjYsPa3XPXczYwja3ReZ2cnEokEDg4O8Pr1a/T09Fj7cK5SLpfR3d196d9vb2/fqEFFRERERERERERERLXuTAjTahOm3k6Y6jAhEolYe0zqsdlsLe+VqaZpGk5PT7G/v4/JyUn4fL6WryFJ0rXjyC7uZakOCNxuN/7gf/4f8e/9+/8hNv6fAlxOJ/6X/+l/aGrnihm+nJ2dWaPanjx5UndXyEXtNmGqzxcEAbIsY2hoCJFIxApJhoeH0dnZ2fCcVpjtGYfDgXv37uHs7AwrKyuQJAmJRKJuYNZuCFOpVCDLMhRFwcjICM7Pz5FKpZBOp68MfjRNa7kJA7wPUhRFgSAI8Pv98Pl81j6cQCCAgYGBhuFOo3Fk+Xy+7ndPRERERERERERERO25MyGMqZmX7YqiXBpHdnx8jGQyic7OzqbCBJvN1lYTxjAMbG1tYW1tDU6nE4ODg20FMMDV48jy+TwWFhYgy/KlvSzVfuZnfgappSSOj4/R09PT0j6St2/fIp/PY2Jioqng5rZc/B2bIUkul8PKygpsNhsSiUTNyKxWxpGZKpVKzTkdHR14+PAhjo6O8O7dO3R2diIWi9X8rbQbwmiaVhN6OJ1Oa39LKpVqGPy0M47MPK/6uc19OMFgEFtbW3j+/DnC4TDC4fCl761UKl3678dNgzUiIiIiIiIiIiIiuuzOhDCCIDTddFAUxWqxFItFLC4uolwuY3JyEh0dHU1dQxTFll88Hx0dIZlMoru7G0+fPkU2m722yXIVc6dMNU3TsLKygqOjI4yNjcHj8Vx7HVEUG+5vucgwDGxsbODs7Az9/f2YnJxsqWFyG+PIGuns7MSjR49wcHCAN2/eoKenB9FoFDabre0mTL1AxePxYGZmBru7u3j58iUCgQAGBwetUKyd8WCNwpvOzs6a4KejowOxWMxqorQTLgHv2yz1/tbNHTW9vb3IZDJ1d9TU+4zHx8dN/a0RERERERERERERUfPuTAgDfDUiq9njlpaWsLe3h5GREQQCgQ/2XMViEclkEpqm4f79+9bLb5vNhkKh0PZ1JUlCsVgE8D4c2dzcxPr6OqLRKEZHR68NHczvodlwwgyRPB4Puru76+7KacZtjSNrxOfzwev1IpvN4sWLF+jt7W0rrLjqHEEQEAqFEAgEsLm5ifn5eUQiEWiaVndU13Wu+0xm8LO3t4dXr17B6/UiGo22fB9TuVy+skEjyzLi8TgikQjW19eRyWSsHTX1cB8MERERERERERER0e1r/X+C/wE1E8KYYUU+n4fNZsPs7GzbAYwoig3HgQHvGwPLy8vWaKfHjx/XtA/qNVlaIcsyNE3D8fExPvvsM5ydneHp06fo7+9vKhwRRRHFYhG/8qu/hsF4AuP3pvCHf/iHl44rFot49eoVUqkUpqamMDY2BkmS2gpTLj6X2dxZX1+/8rus1mzQ1tfXh8ePH0PXdZycnGBvb6+lZ24muBFFEQMDA5iZmUGpVMLm5ibOzs4+yHguQRAQDAbx+PFjOBwOzM/PQ1VV6Lre8rUujiNrxNxRMzU1hb29Pbx48aLucdvb24hEIi0/BxERERERERERERE1dudCmKscHR3hs88+w+npKXp6etDX19fWKCdTo70whmFge3sbn376KWRZbhj02Gy2G40jq1Qq2NnZwfLyMu7fv4/x8fGW9oOIooj/5JO/i3/65/8Cvn/zP4B8/6/hb/3yD/GjH/0IwPsQYnV1Fc+fP0dvby9mZmbgdrsBvP+u23n5D7z/fszdOJ999hkcDgckScL8/Dy2t7ebDlmaIUkSYrEY3G43crkcnj9/jqOjo6afs9m/D1mWkUgk4PP5kMvl8OLFC5ycnDR9n1YaRaIoIhKJ4OHDhxAEAXNzc01/b6ZWd8k4HA5MTEwgGo1CVVV88cUXyOVy1s/ZhCEiIiIiIiIiIiK6fXduHFk9hUIBi4uL0HUdU1NTcLvdeP36NUqlUlujo0zmbpnqpfcnJydYWFhAZ2cnnjx5cmXboN0mjK7rWFtbw+bmJhwOBx4/ftzWWDBJkvC//8mfwv/xL8DRE4SjJ4j86Cz+jz/7c0xMTGBpaQmhUAjPnj27tK+knZ04Jk3T8Pnnn6OjowNPnjwB8D5QCofDWFtbw/z8PBKJRMPRV+3sdwGA0dFRFItFrKysIJPJYGhoyAqVblMikYAoilhZWQGAa+/T7l4XXdfR2dmJ8fFxrK+vY25uDvF4HD6f79rvp5WAqZokSQgGgwiFQlheXoaiKIjFYshms5iammr5ekRERERERERERETU2J0OYTRNQyqVwsHBAUZGRuD3+62f2e12lEqlG91PURQrRCmVSlhcXESxWMS9e/fQ2dl57fmtNmEMw8De3h6Wl5fR29uL6elpLC0ttb3sXhRFuN1ulM+O4fT2AQD0wgnyZzlsbm5iZmamJmCq1uz+nWqqqmJpaQn5fB5Pnz5FV1cXgPf7SYD338fIyAjOz8+toGR4ePhWghIzuHG5XJiamsLJyQmSySRcLhfi8fiNwrhqlUoFkiTB5XLhwYMHOD4+xsLCAtxud8P7aJp2adF9MzRNg81mg81mw/DwMIrFIlKpFNLpNIaGhtDd3X0bH6lGqVSCoijo6urC9PQ0Dg8P8cknn+Ddu3dWoEZEREREREREREREt+NOhjDVS+oHBwcxOzt7KahQFOXGIYzNZrNefG9vb2N4eLilZfWtNGHOzs6wsLAARVGscKRcLje9R6WaYRh4/fo15ubm8Et/6z/CJ3/vP8P56FNUznPQdpbxwx/+dxgbG7vyGq2MI6v+fQwMDODs7MwKYOpxOp24f/++FZSYAYbZKmq3xVH9e+nu7sb09DT29vbwxRdfIBAIYGBgoK0wpJoZwph6enowMzOD/f19fPHFF/D7/RgcHKy5T7shzMWRYg6HA/fu3cPZ2RlSqRQEQUAikbgUYlUqlbbH8JXL5Zogyev14vd+7/fw8z//8/jN3/xN/NVf/RU++eQTuFyutq5PRERERERERERERF+5UyEM8H7vy+LiIrxeL54+fdpw74Xdbsf5+Xnb9zEMA8ViEevr64hGo/j4449bfrHdzEgvVVWxvLyMk5MTjI+Po6enx/qZJEkt75QxDAP/1T/6b/Dn//L/hWF3o3K6j1//1f8YycVF9Ibu44c//Cfo7e29lWcHvhrP1tXVhadPn0IQBGxvb1s/r1Qq0DTNCnSqA4zqoOTly5cIBoMYGBhoexzZReaie7/fj62tLTx//hyRSORGu4IuhjDmfQKBAHw+H7a3tzE/P49wOIxIJAJRFG8UwtQ7r6Ojw2rhJJNJOJ1OJBIJKzwpl8tXjsm7SqlUQkdHR82/E0URhUIBn332Gf74j/+47WsTERERERERERERUa07FcIIggBN0/DgwYNrR1jZ7XYcHx+3dZ9cLoeFhQUAQF9fHxKJRFvXuYphGNjY2EA6nUYsFsP4+Pil4EEUxabbKKaFhQX8+f/9rzH1/V9E/ryA08M9/On/+af4v/7sT+B0Opu+znVNmHK5jKWlJRQKhZrxbGZopOs6NE2zQgtRFFGpVKDruvX/m/cxg5LNzU3Mz8/D7XbD4/G09LmvIooi+vv70dvbi/X1dczPzyMej7e186ZeCFN9n0gkglAohEwmg7m5OUSjUUiSdKNxZI309PRgenoa+/v7ePXqFTweD6LRKFRVbTsoudiEMamqCpfLhV/8xV9s67pEREREREREREREdNmdC2GCwWBTx7azE8YMFvL5PMbGxqDrek2r47YcHR1hYWEBXq8Xz549a/iCvp02yNHREexdXhSKRaiqCl8ojB27C6qqthTCNGrCGIaBTCaDTCaDRCKBe/fu1TynYRjQdd3aAyPLshW4mM2eSqVSE86Y9xsYGEBvby9ev36N9fV1OBwOeL3elr+DRmRZxtDQECKRCFKpFM7Pz3F6enrl6LR6n/+634ssy4jH44hEIlhdXcXh4WFNw6lZqqpeaqVcZLZw/H4/tre38eLFC3R0dNyoCXPxXE3TGgZPRERERERERERERNS+OxXCtKKVEEbXdaTTaWxubtYEC/l83goT2mU2QCRJQrFYRDKZbLrN0yrDMOBwOHCQWYEnPoUOjx8byS8wEA5ZTZVmCYJwKYQxl9B7PJ664ZE5eswwDKRSKSQSiZqxX6IoQlEUqyWjaRoEQagJY2w2G/x+PwBga2sLmUwGw8PDt/pdORwOjI6OIp/PY3V1FbIsI5FItBRSNUNRFIyNjWF1ddXaTTM0NHRtsGJqNI6sHkEQEA6HEQqF8PbtW2SzWbhcLvT29rY0eq3e6LS9vT2EQqGmr0FEREREREREREREzblzIUy9cKAeSZKaWmq/t7eHpaUlBINBzM7O1vwv/m02G1RVvdHz2mw2FItFZLNZ7OzsYGRkBIFA4EbXrOfo6AjJZBIejwf/5e/8Fn77d/8hlg4OER/sx3/xO7/dcqumehxZqVTC4uIiSqUS7t+/fylEqB49JooihoeHkU6n8Zd/+ZcYGhpCf39/3TDGDG3MpoX53Zth0v3793FycoJkMgm32414PH5r+0h0XYfNZsPDhw9xeHiIN2/eoLu7G7FY7MoRYO20kyRJwuDgIFwuF5aWlmC325FIJOBwOK4877pxZI3u1dXVBZ/Ph0KhgPn5ecRiMfj9/qaf/eJx29vbiEQiLT0HEREREREREREREV3vzoUw5qLz614oX/fzs7MzLCwsQFEUzMzM1H0hftMQxjAMqKqK58+fY2BgALOzs20thL9qBFapVEIymUS5XK4JSP7kf/0DLC8vw+l0tvUC3dxHs76+jo2NDQwPDyMYDNY8h67rqFQqKJfLKJfLcLlceLvwDvOpt5Bddui5EpxbTmxtbWFoaOhSECBJEgRBsMaTnZycYGlxCdubWYyMDyMUCqG7uxvT09PY29vDy5cvEQwGMTAwcOPxWLquW78Lr9cLj8eDbDaLFy9eoLe391JwdBOapsHlcqGrqwuPHj3CwcEBXr9+be1waRS0qKracggDvB+r19XVhXA4jFKphFQqZY2Pu2osWqN9N9vb2wiHwy0/BxERERERERERERFd7c6FMK2oHgVmUlUVy8vLOD09xdjY2JUvpdtpPZhyuRwWFhagqirGxsaa3mVzkSzLqFQql0ZEmQHJ1taW1a6pfl5BEOB0Oq02S6uKxSK2t7fR19d3qSEEvH9hr6oqjo6O8K/++b9GKV9CqVLEWaeOB//WLGSbDWcnOaTerOFv/LV/G6lUCul0GiMjIzU7WERRhCiKSKfT+G//4T9G5UiArul4Hf0S3b/cjf7+fmsXkN/vx+bmJubn5639Me3+jqpDGPP76uvrQzAYxMbGBubm5jA4OIhQKGTdQ9f1tu5XPeJLEAT4/X74fD5rh0uj0KeVcWQXzzPDG7vdjomJCeTzeet3kEgk6o5Eq7cPBmAThoiIiIiIiIiIiOhDuZ0qwC1qpZ1QvRfG3Pvyk5/8BN3d3Xj69GnTy9KbGX9mKpfLePv2Ld6+fYvR0dGal/jtMJfZV9vf38enn34KXdcxOzt7qaFiMkOoVhSLRXzxxRc4Pj7G4OAgRkZGagIYXdet5ouu6/hX//xH6EQPRsLj6EA3VtMZ7O3vYeHtAnb2d5E7L0CSJExOTmJ0dBRLS0t48+YNisWidc0vX3+J//rv/z52vjiCdqjDJXSgtFPBj/7Fj2pCJFEUMTAwgOnpaZydnWF+fh6Hh4ctfT6TOTrtIkmSEI1G8ejRI5yenmJ+fh5HR0fWOe00cOrtWTF3uDx+/Bi6rmNubg7ZbLbmb03X9bbuVy6XL4UpbrcbU1NTiEajWFxcxNu3b2t+B+Z5drv90vUYwhARERERERERERF9GN/oJoyiKCiXyygUClhaWoLf76+7UP4q5m6Z687RdR0bGxvIZDKIx+OYmJiAIAg4PDy80Uiz6hCmUChgYWEBoihienr62kXy5kixZpjNmu3tbQwPD6Orq6vmRb45ekzTNIiiCFmWkc/nUc6r8IQ9AICgP4SDl1l89i8/RzjUj/3dHRh7h9D/nffP0NnZiZmZGezv7+Ply5cIBALwer14+eNXkIoKnJILXYoH5XIBnm4Pcrkz637mP8D7MXEjIyM4Pz/HysoKMplMy2HTdQGHoigYHR1FoVCwGiTtjkEzd97UI0kSYrEYIpH/j707jW4rv888/70X+w4CIEAS4r5KpBaqREnlpezYKccer3E6tuM4TjydScbpTtJ9upOcTNInmThbz5z0xDNxFk8yWdy2YzuuuGOXy1Ulu1zlci2SKFFUqbiT4AaQBEES+3K3eaECWhQXkSrbRc/8P+foVB0CxL24gN7cR8/vFyUWi7G0tERHRweBQODQx6nab4xZdbxbdSSa3++v7cHZLbwBWFlZESGMIAiCIAiCIAiCIAiCINs30g4AACAASURBVAjC98GRC2EO0yqRJImxsTEcDgenT5/G6XQe+njVvTD7hTCpVIrJyUmCweCOkMdsNu9oshyG2WymUqkwNTXF+vo6vb29B75BbzKZDhTCrK+vMzk5SSQS4cKFC5hMJmKxWO13q6PHqudTDUPsdjuSWaJYKuCwOzGbTJg2VUoTSZaWMgQdfiK+DuLxOB0dHbfHb10bQZLg7ANnqVQqPPfccywtxHE4HGTsWUxWKBVKrGSXeWjw/VitVlRVRdM0DMPYFmY4HA4GBgbY3NxkdHSUiYkJ2tvbdw0S7nb3OLK9OJ1OBgYGSKfTTE5Ooqoq5XJ518bIXnZrwtztzmCpGvrc7yg52P/vyZ0j0ap7cKptqr3GkYmdMIIgCIIgCIIgCIIgCIIgCN97P5QhjKqqzMzMkEwmCYfDHD9+/L6PV23T7NY6KRaLjI+PYxjGniGPxWIhn8/f17ENw6BUKnHz5k3a2tq4cOHCocax3WscWfX8gR3NmmqLplKp1EZ33dlGgdvv7Q1vfR3feeJZZN3M+NQYy0txmlbNNDY0YK43YTgNtra2eOaZZ/jMX32WOqkeZINvf+MZPvSxD+ByuZBtEI7UY5LNxJcXKZjzvO9d7+YNb3gDsiwjSRKapqFpWq3Bcud5eDwe3G43dXV1jIyMEA6H79laOWgIU+Xz+ejp6WF2dpbR0VGCwSAtLS0HalUdpElV5XA46O/vZ3Nzk5s3b3Lr1i06Ojru2XqqMgzjwOPz7tyDs7y8zPz8PJFIZMe1SaVShEKhA73ma0mSJC+gGoZReK3PRRAEQRAEQRAEQRAEQRAE4SCOXAhTZRjGjkDGMAyWlpZYWFigtbWVvr4+MpnMqzpOdUzTnTRNY3Z2lmQySU9Pz743qO+3CZPNZhkfH0dRFLq7u++ribBXE0bXdebm5lhdXd31/HVdxzAMstks4XAYs9mMJEmkUilUVSUQCNQaEy0tLbz3Q+/hc//180y8NIlb9aJLOuvxFLGFBdI3ktwcv0ExXSG3VqKh205LpI2rL73IU5ee4lf+/a9w4sQJLj32LbJGmo6mVt77E++hv7+/9vlWw5/qaLjqn2oYYxgGsiwTDocJhUIsLy8zPDxMc3MzDQ0NuwZ3hw1hqr/jdrvp7OwkHo9z7do1otEojY2N93ytw+4Fstvt1NXV0dDQwK1bt/D5fLWxYfs5TOBTZTKZaGlpIZPJoGkaV69epa2tjfr6eiRJql3fo0aSJMkwDEOSpE7g54GLwG8AlyVJOgdMG4ax9ZqepCAIgiAIgiAIgiAIgiAIwj6OXAgjSdKuN7Q3NjaYmJggEAjURoJtbW3tCFAOy2q11kZxGYbBysoKs7OzHDt2jIsXL97z5nR1nNlBKYrC1NQU2WyWvr4+UqnUoW/gV93dhDEMg6985RFefP4SoXATH/7wv94RwFRHjwWDQbLZLCMjI3R1dTFy7QaxlxcwyWZsXjPveO/b8Xq9ACwtLTF+dZK2YAdldDZzKZaSC3gDXlxWD42WNm5t3KTZ08bi3CKSKhEOhfH5/JjNZsLhMB/+2Q+hqiq5XI6pqSnGx8fp7OxE0zS++8xzxBcTBEJ1vOHNr8fr9dZGlN25p6b6nqvhSywWY3h4eNcdK9UQ5zDuDH6OHTtGQ0MDCwsLDA8P09bWRigUuu/P6m7VvS7BYJBAIMDq6irXr18nEolw7NixPc99r70uB6GqKj09PRiGQSwWY35+nlQqhd1ufzVv5ftJBjTgd4FvAu8CvK889p+BPwS+WQ1rXpMzFARBEARBEARBEARBEARB2MfR++fvbG8VFAoFRkZGiMVinDp1it7e3loTwGazUS6XX9WxquPIMpkMV65cYWNjg6GhIVpbWw/UDjhoE8YwDBYXF7l8+TI+n4/z58/j8/le1U6Z6kgxuH2dPvnJP+H6i5/mp96nc6Z3hv/zT3+bjY0NgNrosUqlgiRJ2Gw2jh8/zvHjx3n22Wd54dJl2iOddEW7sZQcvPjdyxiGgaZpLMWWiYQi2GwOylIevQjFdIXp2WleunmT0dEb6IbOwloMdAnJIrFRXmfw3Jkd18rv93Pu3Dnq6uoYHh7ms3//WRLjSZqcLRQTCo/+89dRVRWr1YrFYkGSpNr1ubP1U92x0t/fTzwe58aNG9vGwt1PE+bu4MZsNtPR0cGpU6dYX19nZGRkR/PqMOPB7nTnHiJJkmhoaOCBBx5AkiSGh4eJx+O7vm41vLkf1d+12Wz09vbS0tLCpz71qdr1O4KqF6DXMIy/A8aB+Cs/swBrACKAEQRBEARBEARBEARBEAThqDpyTRi4fVNaURRisdi+I8G+FyGMJEnE43HW1tbo6+urtT8O6iBNmK2tLcbHx6mrq6u1eKpMJtOhmjR3qo7vmpqaYn19ndjsVX7nP54mFLy9u2Y1OcG1a9d46KGHantfzGbztnDC7XbT2trG2vgm8/PzIMNSfJG5b8/wzFPP0NjQREUvEwrWkygl0FIakxu3SFbW6A+dwVANVhZWsddZeN1bHuTGxHWKbj8f+Nn3c/bs2V3PO5/Pc/WFYeamY1x78ToP9A3hd9fRUN/A1PIEGxsbNDQ0YDKZ2NraYmpqiqamJjRNwzCMbUGJw+FgYGCAdDrNxMQELpeL9vb270kIU1UNrHK5HDMzM5hMJjo7O3E4HPd1HNg9TKmODWtsbGR+fp6rV6/S3t5OMBisBZOvpglz94i/SCTCb/7mb/I3f/M3/Nqv/Ro//uM/zsc//vH7eu3vB8MwqqnbqCRJHwJOAPlXRpFZgaXX7OQEQRAEQRAEQRAEQRAEQRAO4EiGMIqi8OKLL9LS0rLvSLDqvpD7oes6CwsLLCws4HA4OHfu3H2NmtqvyVIul5mYmKBSqTAwMIDb7d7190ul0qGPaxgGqVSKVCpFMBjkwoULfOXLNu68HIZObaTX3eHLnYLBAJLVIBIM8+SlS6wvboIBqiyjOcHl9DI8fgU9B2vpVVxhJ35XL17Dj2IoFO15iuY8zScb+YXf+hgNDQ37nvcTjz6JmjI4ET3JlHWW9UQKv7+OZDJJzshisVgolUpMTk4CcObMGaxW67Z9MXe/H5/Px+DgIMlkkpGREaxWK+Fw+FDX9F77VtxuN6dPn2ZjY6O2x6WxsfHQO1rg9miwvRotFouFrq4uSqUSs7OzLC4u0tnZidfrve8QZq+wKB6Pc/bsWX7jN37jVQea30e/x+2dMCrwb4Eu4K8Mw9h8Tc9KEARBEARBEARBEARBEAThHo5kCGOz2Th//vyBbzbf/S/872V9fZ3JyUnq6+sZHBxkenr6Ve1luTsI0nWd+fl54vE43d3dtQXouzGZTIceR5bL5RgfH8diseDxeGhtbQXgoTe/n0//w2d4x1tCJFbz3Biz8h/ffaYWWFTHrrnd7tp4NMMwaG5u5tybB3n8q09QyBfw+JwkZtdYSC8xuzzDj/7YWymnFd765h8ln8vy9W98g9TmBvXREBVVodHfQNmf5oMf+QBO5+0WTqFQQJblHftGisUiqdVNjkf7ATg7OMhzLzxPOB/CsOi4gg4mJiYA6O3tJRgMbrvW1fZP9U91hwvcbjWFw2FCoRCjo6PEYrHaqK+DfL6apmGz2e75vEAgQF1dHaurq7z00ktIknToHTSKouwayt3Jbrdz4sSJbQ0cm82Gz+c78HGq9gpvVlZWiEajSJJ0ZHfDGIaxIEnS3wHfBZqBfzAM4+Zre1aCIAiCIAiCIAiCIAiCIAj3diRDGEmSDtwuqDZRDrIno1AoMD4+jizLDA4O4nA4UBTlvseB7aYa8EQiES5evHjPG/PVUOEgVFVlZmaGzc3N2ui0F198sfb429/xLux2J88MX8blquPf/4f31ca4LS0t8U9PfBXFApVsEUnR8HY1UcznMT9d4WMf+Ajv++B7+eTYn7E2sYGa0ylQJp6M8anpP8ft9NLbfpxQMEhvby9XlOe5tTlCwBPEZbXxUz/zQZxOJ4qi8Pi3LjG1uoChGwx2HudNr38jsiyzvLzM5MQU37nyLFfHRnE73Zxp6af7ZAdnfvR2U6hQKOB2u8nn86RSKbxe77bPVpblWhijququYYwsy3g8HhobG8lkMgwPD9PR0UEgENj3+h4mSKmGO3a7nenpaYaHh2lpaSESiRwo8FFV9cDf8WoDZ3Nzk1u3blEqlfD7/YdqxJTL5V2fn0gkuHDhwoFf5wdFkiTJMAxDkqSTwC8YhvHLwOwrjwUkSTptGMaRXGQjCIIgCIIgCIIgCIIgCIJQdWRDmIOq7oXZL4RRVZXZ2VlSqRQ9PT3b2hX7jRM7jHw+z8TExLaA5yAOcnzDMFhZWWF2dpaWlhZ6enpq16jawtE0DUVReP0b3shDb3rzttFTqqryyJOPEjzXhb8+yM2bNxn59vO06QqF1TyFbJ4/W/sLTp8+yfTcDJUtjXJeIV3YIKumUVAoZSp85h8+T3tPFLvNjtPuoqE7wsnBk7z14bfQ33+72fLi8FVmlRS9D19g9qVxPvvoP7OyFOfMmTN8+2vPsBxfJec3kw+Crd7D10a/yYd+7F34fD4qlQqnTp3C6XRiGAbxeJyrV68SjUY5duzYtvcky3JtRFk1jKk2omRZRtd1rFYr3d3dFItFZmZmWFxcpKurC5fLtet1PmybBW63nvx+P62trcRiMZaWlg4U+Oy2E+Ze6urq8Pv9eL1eRkZGqK+vp6Wl5UDnXKlUdm35JBIJmpqaDnUePyASYADdQARAkiSPYRhZ4EeBDwPvkyTJZBjGwVJMQRAEQRAEQRAEQRAEQRCEH7AjGcIchtVqpVwu7zrayTAMEokEc3NzNDc3c+HChR17Me53DFmVpmmUy2Vu3LixY3zWQdyrCZPNZhkfH8fpdDI0NLRrm6FSqaBpWq0lcvd7LBQKlCQVf/3tczNMEvOTMRYemSDoimCvszO9UmRxfIG+8ABXFl9kJZegqOdx4SNKF2ljnc38GrmRTaJ1LUQCDSxlVokGWrh14xa9vb2YzWaWkwm8DQH+7Fd/n9nnJjDLVuafnKS9/2ne9cb3MJVYZPANryeVSRFpr8fhMZPaWCcUepBwOFz7PCRJIhqN0tDQQCwW4/Lly3R2dhIKhbZ9ZiaTqXYNFUXBMAxMJtO2HSgOh4OBgQHS6TQTExO4XC7a29t3XMv7CWGqjRaLxbIj8Ons7Nxz5Nj9hDDV32tsbOTYsWMsLy8zPDxMNBqlsbFxz50/sH8TJhqNHvo8foAiwCbAKwEMQD2w+JqdkSAIgiAIgiAIgiAIgiAIwgEdyRDmsE2YSqWy4+fpdJrx8XE8Hs+e4cWdDrtXxjAMVldXmZmZwWKxcObMmdo+lIP8brWZsFcTRlEUpqenyWQy9PX17dgDous6qqqi6zorKys0NDTsGSA4nU6smkw6tYkvWMf6cpKNsQRnvEME3PUklhdIptIMPNB/e29MKUPIaCTNBi585NhC0zUsWGkgStjShFE2sCoOxibGKKlFLrz+Aq2trQS9fv7gd/531p9fp9naTVkvkVxJUlJLvP5UGkPRmH95goqso1FEWc/S/bpzRCKRXc/dZDLR2dlJNBplZmaGhYUFuru78Xq9O55X3c2iadqu3wmfz8fg4CDJZJKRkRHC4TDNzc216/ZqQpiqauCTyWSYnJzE4XDQ0dGxo4WiKMqBx5HtdjxJkmhubqaxsZGFhQWGh4dpa2vbEVJVVSqVHdcMbu8X8ng8hz6P7zfDMPRX/vcrwJskSXoEeArwAseBr1af+hqcniAIgiAIgiAIgiAIgiAIwoEcyRCm6iDBSHUcWVW5XGZqaopisciJEycOdIPZbDajadqBb4pX2ykOh4OhoSHGxsbQdf3evwjceuklnnzisxh6EV9dG+9570e2NWE2NjZYXl5mc3OTjo4O+vr6dlyDausD4OzZs8zNzRGPx+np6dl1abvZbOb9D7+TL196lBWbzPJ3Rmm1Rynlc5RUOxZMrGwkia8tYbXZcEguNEOlQgVI48CNW7ehoZA0EgSUEJVSmZJRJOD2sUaKJx/7Jh/7n36Wl6+/xOyzk7RqvciqTJ0tiKaVUbQSN6eus7m8wdJ318jqeWSbQX9HH6Oml2hvb6elpWXP62a32+nv7yebzTI5OYndbqezs3PbMnlZllFVlenpaTRNw2azoSjKtn0xkiQRDocJhUK1JklzczMNDQ33HcLs1mjxer0MDg6SSqUYHR0lGAzS0tJS+47pun7oY1Xd+X0wm810dHQQjUaZm5urNXDu/h7s1oQ5bPD4WjAMIyFJ0v8CvAMIA83AZwzDePyVxw/2F08QBEEQBEEQBEEQBEEQBOE1cCRDGEmSDnxz2Gazkcvl0HWd+fl54vE4XV1d20Zb3YvFYqFSqdwzhNmrnWKxWGqhyH6SySRPffNv+egHmgkF3Vy5FuNr//JZevqG0HWdf/zyF/n26GXsDgf90Q7OnTuHJEnkcjnW1tawWq3U19fXRm1Vz7evr49cLsfk5CRWq5Wurq5t4QRAS0sLH//wx0in0/jzJv5p7HPojiyjy2OQs2HxSayujrK2tcp6ScKBDzt2dHRUKuiYCdHIBmtMbbyMAZhtMl2eDuoD9SxOLvPII4/wd3/+GewmG1ZsWLBRKhcpGQV8fhc2rxV7wsGZ3gEy6zlMVpkzAydprGviia9d4ud/6X+85zX0eDycPXuW9fX12l6U1tZWTCYTiUSC+fl5Ojo6CIfDGIaBoii1dsydYYwsy7XwJRaLMTw8fN8hzF77fyRJIhQKEQgESCQSXLt2jaampvvewVLd/7Mbm81GX18f+XyemZkZADo7O2v7b3bbCbO1tYXf77+vc/l+kyTpR4DnuN16UYEnABMwA5gkSZJFACMIgiAIgiAIgiAIgiAIwlF3JEMYuH0De7+bzlU2m410Os0LL7xAJBLh4sWLh76RbrVaqVQqe44TMwyDpaUlFhYWaGtr29FO2Wuk2N0SiQSdrRZCwdt7QobOtvHtZ19C7RrkkUce4Tuzo7zhI+/C6XIxPfwSl55+igdOneHzj30F3WOlkitwuqmLH3vrwzveo9vtZnBwsBZORCKRHUvb7XY7kiSxtHyTd75LxSblGb6R48XnFDo6JU73g6I7uHK9wujVEgVyWGU7ZizYTHZ0m4xeNnAqTvJyFqfZy0Z5jX/1lp/gz774f3Hp2ncoyzKaU2IxO41Dd5EzsqhymZ9+/0+iFnS6X3+c5Oo6W6s5KOusrKzS2dpNfjN/4BBEkiTq6+sJBoMsLy/z4osvAhAIBBgaGqqFU5IkYbPZ0DQNVVVRVbU2tqwaxty5y+XKlSvcunWLrq6uWnhxLwdpUMmyTDQaJRKJsLCwwNWrV1FV9dBNlLtHn+3G5XJx6tQptra2GB8fr+2/qVQqOxo7Kysr9x0I/QC8HngW+BDQBmSAAlAGdOCPXvmZIAiCIAiCIAiCIAiCIAjCkXWkQxhd1/e9SZ3P5xkfHyefz/Pggw/uaH8clNVq3bPJUr2Z7ff7uXDhwq43wQ/ahHG73STWyqiqhtlsYjmxRUmBUqlEWVfpfmAA5ys3/+vboizfWGb50mPUnWqlLhLCMAxeeuYa/UtLtLa27nj9ajjhdDqJx+NcuXKF9vb2ba2gubk56v1b/NxPPcjEzSlaj20wNjnL8Q4vLpuMnXoeOg6JiRzFrJWkdRXZasXi9yNZDVyuJrREDl/OR9TVSjqd5InnHydXKdMcaae4OYXbHGCtPINFs+DEydCDDyFVTOSVDH/7xb/HXnGRS2cwOUzYHFaezD3B+XecPXR4ZhgGpVIJs9mMy+Uim82STqcJBoPbnlcNXqp7dAzDwGQybTuew+HA4XDQ1tbGxMRELby41y6hgwQjVdXRYZFIhOvXr3P9+vVdR4ftpVKp3PN8qvx+f60xNDo6SqlUqjWoquLxONFo9ECv94NmGMbvA0iSNAJ8HvAB9UAAiADF1+7sBEEQBEEQBEEQBEEQBEEQDuZIhzB7UVWVmZkZNjY26O3tZXx8/L4DGPjv48juVC6XmZiYoFKpMDAwgNvt3vf3D9KE6ejoYKzpjfzt576D1w03bq7z0Ft+BpfLRai+nstP/zfGEzE2llZhq8Tb+oZYy21xvL6f3FaG1YU4yfV11tfXaW1tZWNjgyce/wqZ9BrHmnt5wxvfypOPXWJ+agFkuPDQEKlUiqWlJXp6evB4PMiyTLFY4QtfHubpp6fQdZ1UukCmYiEoOTEbZgplhbIM5ZCGocjkQiXkhgpeUx0up5cNvYi7YkG1lqlvCyC5NYrLefLJHKFKkOVMjIpaIequ4/U/8joeftvDXHr+EqPJl+kcOsHYlRtYvRZkk0TeyFIq5zg7NHiozyyZTDIzM0M0GuX8+fNIkkSxWGRqaoqFhQV6enq2tVlkWa79qY4nq7ZY7gwmfD4fg4ODJJNJRkZGCIfDNDc37xkQHSaEufNcfD4fHR0dzMzMIMsyHR0dezaxqhRF2XX/zF6qoVwgEOCFF17YNg5NlmVWVlaObAgDIEmSC/iIYRjveq3PRRAEQRAEQRAEQRAEQRAE4X78UIUwhmGwvLzM/Pw8LS0t9PT0fE8Wi1utVsrlMsB97ZYxm83k8/kDHevht72Lp5/2sLm5yS/+mwdpbm7m+eefJ1BXx8vfHWZ6cQHThoScheXQLR7+4Nu4/I2nufYvL6CuqSjFInIsT2trK//Hf/ltuo/N0tpgZ2l+hP/tu9+hyXGcgeYzqKrKi9+8yo9/9D00NzczOTmJw+EgNhfjS/90g57ODR5+k4MnLinkMgaPf2eTi4MVzEaey6Maenc3znoXuuqhsLiG7rOQNfJsGpuUyklyW2UiJ4L850/+MZ/+808jZU0YXgNUcGpOik4zP/c//yySIZFIJEisr+Ku89Ha3kM2maHOiGAqlbnwunPkSB84RCuVSkxMTCBJEoODg9v2nDgcjtoorpdffhm3201nZ+e29kg1iDGZTKiqWhtVdmeQIkkS4XCYUCjE8vIyw8PDtf0xd38X7ieEURSl1t45deoUm5ubvPzyy3i9Xtra2vZsuxymCXP3Obrdbvr7+1lcXOTq1asEAgHi8TinTp069Ov9AFmAlyRJer1hGN99rU9GEARBEARBEARBEARBEAThsH5oQpjqWDCfz8f58+d3NAIOu1/jThaLhVwux/r6OpOTk4feLXOQJoyu6ywuLrK0tMTJkyeJRCIApNNpkskkf/3X/w8vT01hrtjwORswTBVSS5v81R//BXaHnToa8JnrCLjqGbl0iz/kDzFrL/Detx9Dkg2sN5f553+5RaXVwUsv3UQtazi8duLxOO3t7Zw5c4Ynn3ySb3zpSdw2K2+64OJP/jRPSnJj6oywmS3x+ONFDEXHCNnxNpjRKeMJhFGTeSo3VrC3BjFnVKzzUKLC2OTLrK2uoaoaViwsL84DoFCmYlEYjV3nbO85VjcTlOQcgbYWkusJHH4XyxOzhD1+8koOT9RFKBQ60PVLJBJ0d3fvGDl2J7/fz7lz51hdXeXatWs0NDTsaLPIsozVakXXdRRFqY0pu3NklyzLtfAlFosxPDxMR0cHgUCg9jrVPTOHcXejpa6ujgceeIDV1dXaPp9jx47teN3DNmGqquGN2Wymvb2daDTKn/7pn/L5z3++9j08SiRJkozbC6F6gF8FfkSSpCvAKrd3wtwwDOPx1/IcBUEQBEEQBEEQBEEQBEEQDuLIhzClUonJycl9x4JZrVYqlcq2VsRh6LpOPB6nUCgwODiIw+E41O+bzeZ9d8Jsbm4yPj5OKBSqhTv5fJ5//MqXmVpd4ImvPs7m+Boutx99s0iOFRx2CW8og15WSa1lsUseVLnCVj7Fmhpn7PMv8MGfcIBRh8VkJeTPYOjrrG18nhtXLXiNVlRJwdfswuPx8Oi/PEp8cYVirkQuX+IfPp9ntWjDc64Zs82KaqjkzMswWcDIFCmlM9isTgqZJMZGGdeyCWk5gwUbFpwECVFcz/P4F75FbDbGSnYZL3X4CbLJOl4lxJXLV3j3B95Jr7uD1sUo//zkY2ypebbiKcp6jnDUTbg3wNvf+WP7hgvpdJqJiQmCwSBDQ0MHCj0kSaKhoYH6+noWFxe5fPky7e3tRCKRbWGdLMtUKhXGxsaor6+vhSp3HsNisdDd3U2xWGRmZobFxUW6urpwuVwYhrFtnNlBqKq64/1WzzccDrO0tLRr+6ZSqew7Fm8v5XJ5W4PGarXy67/+6zz//PM89thjfOMb3+ALX/jCPceh/aC8EsBgGMZlSZLeBXwceC+QBVYAA3hckiSTYRjaa3emgiAIgiAIgiAIgiAIgiAI+zvSIczS0hJLS0t0dXVRX1+/Z9PFZrNRLpcPHcJomsbs7Cxra2s4HA7OnDlzX+e6VxOmGiApisKpU6dqO0pKpRK/+we/z/X0PBWbhN5Uh7FVQF0q4Cw7UUw5zg7qDJ12s140+G+PpcnNGdRpYea1Kdz40BSJxcUyn/vSHO2tFq7fLGOzGrz9Yh1yKc/T31pEsWn8zV/8PY9++RvkzGUMGXJrGUxZKKBiC7vxWm83SgpSFo87hMI6WkUhd3UB+UQTusOExe9CT6k05aOkSaGjESDMupTAJXvYWNtCR6VEkTIl6qQQZcpkN5L0D5zgsa8+TiFe5mLkPJdefJKG0220nujCpEFsY2nPz01RFKanpykUCvT392/b8XJQJpOJtrY2mpqamJ2dZXFxke7ubvx+P6qqMjs7Szqd5sSJEzidTnRdr40puzuMcTgcDAwM1EIhl8uFruuHPqf9Gi2yLNPS0kJjY+OO9s39jiPbK6BMp9N8/etfZ3p6+sgEMACSJMmGYeiSJD0MfBQYAx4FOoFbwD8CiABGEARBEARBEARBEARBEISj7siGMHB7TFM0Gr1n86EawhyUYRisrq4yMzPDsWPHGBoa4vr16/d9nnc3Ye7cK9Pd3U04fiJbhAAAIABJREFUHK499tRTT/F7v/0JJuYW8F3oQq+TkSQDW8hHeTyO1WxDs63T12vF7pII2U08cNrCU3MacWI4cNJIO4VShrm5l2g+pnBrosTmls67fsRDyShg9RTINVswhdwYm7Bk3sTX2YJuA3vESWF4GWfZiyVnQ98qYfLY0TMVpKUyHnxssoZFsmDdkLDJToqVHHlnnlR+FRUFGZkMG9i8FtSKgtPhwJAMTJhwyV5kSSKrpbF5LKRSKbLxPMdb+ykWi0RCzUzdnMSyIlMulblR2uBHH3wTg4OD2z6flZUVYrEYbW1t9PX1verdP1arlb6+PvL5PJOTk6iqSqVSobW1le7u7trrV3fGaJqGpmnouo7JZNrWdvH5fAwODpJMJkkkEsRisR3jzvajKMo9A6U72zfV8MgwjPsaR1Yul3c9nqIoWK1WTpw4cejX/D6rftjvAyaAPwGcQAT4S8AG/F01rHltTlEQBEEQBEEQBEEQBEEQBOHejmwII0nSgUcvWa3WA4cw2WyW8fFxHA4HQ0NDWK1WDMNA0+7/H9WbzeZaEyaVSjExMbHrXpkvfvFL/MUffprJqRh6RKbsUXC1N6KsZClNrmEyZEpKnmwlx+SsGwkdVZVILEtkjE38BDFkjXzkJpKsk0vLTE2rrKxrnOxx0Ba1c21E4fLNEk5/hFI6i24Ca9iHZDNh0qCCjurQMdQKlUKJ1JUkhqJgwYwFCxISXgLYVCf6pkRZTlNnrUcqKqxKi2CAjo7iqOPCqddx/cYIXe29bG5usjQdp6QXkJFQ7BU+9SefxGw2I0u3AwyTycTaWpzCfJ5Sukh2M8Nmdp0/+sQf84k/+j26uroolUqMj4/jdDo5d+7cfYUO+6kGKpIkYTKZKBQKO8aDVYMYk8lUa8VUmzHVMEaSJEKhEC6XC5PJtOv4sL3sNo5sLw6Hg/7+fjKZDCMjI8zNzdHV1XWo1lelUtm2xwaovZ8jynjlvyvAdcMwikARSL2yG2b+rucJgiAIgiAIgiAIgiAIgiAcSUc2hDkMm81GsVjc9znV0VaZTIa+vj58Pl/tsVfbsqg2J0ZGRjAMY9e9Mvl8nq/946MU10vofhk1XSQ9PEtpdROjrFGYi4OkY/KZiZ4JMa6rXLmUwhyuJ+uOIvVlWVtY4/wJnZ98pw+rxcZff2ELzBJveYuLmTmFX/2dBFLBy0rGiu5cxRL1oKxkcdo81AXCpNfjmIN26i8OkJ9foVTJY80q2NZNOHUnEhIlCmyQxKgY2CpO/M4QXqsfk64h2TVOnTyJWtTJqmkWF5Y41tbEmy++hbe98e383aN/g2FSCEfC/Mqv/gq9vb3kcjlkDywmFnC7PKxvrKBldVKZNayKHTdexl6c5C8+9Ze8451vx+Vy0d/fT11d3av6TO5mGAaLi4u1dlIwGMQwDOLxOFevXiUajXLs2LFtjRdZlrFarbURZaqq1sIbWZZrYUo1fLl7fNhe9htHthev14vdbiccDjM6OkogEKC1tRWz+d5/he/eCQOwtrZGJBI51Dm8Bs4B/0mSpKe5PYasHggBz0qSdAxIAGIkmSAIgiAIgiAIgiAIgiAIR9aRDmEkSeKVHd37stlsbG1t7fqYYRgsLy8Ti8Vob2/fd7SVYRiHDmR0XWdubo5CoUBfXx+hUGjH47quk8/nUVSVjdwGHrsPk+Enub5GRd1EByztdSjpIla/m62ilYLDT8EhYyrquM4do+GkzOpXL/P2hw1e96CZ2YUKHq/E+z8UIdhWR/96mc2/XGR6toTus6CmC8iaHVvBgjFdYHXjBnK7F4cniGUdSq0+5IUS0oaOWTYT0Y9hwsw6K2io2HCgUCZemMVT9KGbK4S8Yeyyg662PuLpJfLlDG6Tl7WNVZobWhg6fY6P/tJPb7u573a7+fDHPsSz3/4u2UyW8LF6EutJ7AU7XlsdM6WXMStWnvnn51ldWqOjs53Lz1/hQz/9QZqamg71Wewlk8kwPj5OIBBgaGio1gCRJIloNFoLUC5fvkxHR8eO/UPVMEbTtFoYU23JVEOQO8eHzczMsLi4SFdX155jwO6n4SNJEvX19QSDQRKJBNeuXaOpqYmmpqZt4dHddtslk0gkvmfX93vtjhFj/wn4DLfHkDUBzUAS+CWgDXgDsPoanKIgCIIgCIIgCIIgCIIgCMKBHOkQpto2uFcwstdOmK2tLcbHx/H7/Vy8eHHf1oDFYqntyDioZDLJ1NQUDQ0NuFwugsHgtsd1XUdRFAzDwO/3097XyugzL6MXJExmK1annfrzJ1mbnqa4vI5tQyKw6cVQDYrmNNgUsskUxYkMpqAVXVd56rkSiysahqZjdciYAx7WEgW0bIljrTZCURObGTfD03ascyoRtRnV0NhYWaFMjlxiDSmvo1gU3NEGrN12CgsrLK3MYcKEQgU3fsI0ISMzzxSzxhhtchcFNYfdYsditmA2mwj4g6zFk6ylVimUcxzraaS+vn7HdQoEArzn/e8GYGZmlsz8c6wWFtnQ1jAMaLF3sWIsYFqzg9eGJkl8+lP/N+/9V++hr69vR6vooFRVZWZmhmw2y4kTJ/Ycb2cymejs7CQajdYClO7ubrxe747nSZJUG09WKpV2jPRyOBwMDAyQTqeZmJjA5XLR3t6+7Xt1Z3hzUHeGkbIsE41GiUQiLC4uMjw8TGtr647wqKq61+ZOiUSCaDR6qHP4QTMMYxQYvfvn0u03aX9lTJkgCIIgCIIgCIIgCIIgCMKRdaRDmIO6O4Qpl8tMTk5SLpcZGBg40G6Zw4QwhUKB8fFxTCYTZ8+exW63k0wmaze7q+OrNE2r7RaRZZmf+dhHuPHcTWauLqJSwu73UU5lyE4t4Cy4COj1kNMp6Tl0FFRy1DtacHkDKKhs2NcomyQeepObr3w1zcsTFZ55fI1QxIo7YOeFqxkKSQOXfRNKJVSTi4wpjV5S0NHQV8o4cVGmiNPpxtjMkK2s4Ja9SJJB0GigSB4DnSIFvPhx42WLdeKVBVojrbjqXLy8dBN7wMrZvrPgU2k700x7ZxtD58/t28gAeOgtb2B5Is5KcplMMU3AEqJiKeIyu+hs6MHQVY6393Nz8Roul4vR0VGCwSBtbW2HCi6SySTT09M0NzfT09NzoIaT3W6nv7+fbDbL5OQkdrudzs5O7HZ77TnVfTH5fJ7Z2VkaGxtRFGXbvhgAn8/H4OAgyWSSkZERwuEwzc3NmEym+9rHslt7xmw2097eTlNTE7FYjKWlJTo7O7eN2tur3XWUmzD3YtxOpEQAIwiCIAiCIAiCIAiCIAjCkXekQ5h73dCvqt7Y1nWd+fl54vE4XV1dhMPhA48Xs1qtVCqVXcdHVWmaxuzsLOvr6/T29m7b+2E2m2u7N1RVrf3szvfQ3d3Nv/7Vn+PXfv43UcsGxXWFwtgETtlLUA+jo2LoECFKiQI5kx1PJIwl5MaSLFKouBibVRidlYi0uijcsvHNpXqYKbN5I04g00ATDZQ3i1ity0hvqkdzOsjdWkabrmDFjBsfFqxQkChTxIcft+ZDwoSEjI6OhEyGDSQk0mzixIPX8LOeW8XcaNDRfQyX3Y0jbOHXfv4/HKpR0dHZwUYpRXOkla2NNBllk87Wdkyyj83cJp2tbei6hmbohEIhurq6WF5e5sqVK7S0tNDU1LTvZ1oqlZiYmECWZc6ePXuoBfZVHo+Hs2fPsr6+zsjICPX19bX9K3e2a/r7+3E6nbXArRquVD9zSZIIh8OEQiGWl5cZHh6mubn5QCP27rbbSLEqm81Gb28v+XyemZkZJEmis7Ozdm67hVeJRII3velNhz4PQRAEQRAEQRAEQRAEQRAE4eCOdAhzUJIkoaoqL7zwAuFwmIsXLx66aVBtwuzGMAzW1taYnp4mGo1y4cKFHQGR2WymVCpta77sdp7vft+7+eq/fI2vfOkryEkzFmyYdTMFclQoEaUNG3YKpjwmhx2pYuAMBMgWV1DTJShZeOobeVKShKWzHRwWTC4rDtVC8domuWKaCmVMionSrSSO/+EEwbedZDV3GX3FIEiEZebwUYedJnJkKFHAgZtN9yalsIpi19BVlWR8AmtOxk+IAgUaTRE2N7c4d/EBHA4Hx44dIxaLIUnSPVsV1fDi2Wef5VTbIKcePgPA1ZGrTKzdwuqyU1IzyA6Jl+ZHOflgfy3kqi6+n5ub48qVK3R1de1YfG8YBouLi8Tjcbq7u3eMhjusO/evVEOguro6NjY2aGlp2dausVqttfaTqqq1tkz1OyDLcu09xGIxisUiGxsbO97Dfg6yR8blcnHq1Ck2NzcZGxvD4/FQX1+/axC1srLyQ9uEEQRBEARBEARBEARBEARB+GFxpEOYg7RYCoUCExMTKIrC2bNnDzR6bDfVJszd8vk8Y2Nj2Gw2zp07t+OGdvXmu8/nY3x8nO7u7n1vrsuyzL/9d/+G4Reus764gYqCjzpc+NhgjTQbgIRsNqO7FLLlDaQVG6nsMoVimuBGgDl9nZzdRF2jGZvPgzXopTCawDCbWCNOHfUELA2UDY2tb05jPdOAo6EOaSXLPJNIyFixYcGKnxBx5kjb09gvRnH31KMpFeL/9Xm8RTdmLGTYwISJ5FaSr35phu8+9gKGWcNldhMJNIJL45d/65d4+9vfvuP9VgOs2dlZWlpa6OjoYGuiUHu8u6ubhoEAv/jLv8DLL7/M2soa4YYwAwMD217HYrHQ09NDoVBgamqqtrfF6XSSyWQYHx8nEAgwNDR06ABuP7IsEw6HSaVSpFIpTCYTDodjx3dTlmWsViuaptWaMYZhbDsXi8VCe3s7W1tbxONxFhcX6erq2rd9VbVfE+ZudXV1nD17lrW1tdp39+4RaD8MO2EEQRAEQRAEQRAEQRAEQRB+2P3QhjDV0WDJZJLe3l7g4OPLdmO1WikW//uaiWpzY2Njg76+Purq6rY9X9f12ggqgGg0SiAQYHJystbG2GsU1pkzZ/jFf/fzfOLX/xCP5seOEw0FF27WWMaBi4pSIafl8LQ0kFY3yG4msWVN5EgjS0W0lAVlKY3nWCP50WU8yzacpiAleQuv4adCBXPFTsWlsPX8LBaLjRBhFEqYXglWbDix4UChTDmg4+kKkLu+RP7KPH7VT4gGHDgpkCNPDk2pECCMVtBAk3HY/LS1dVNRy/yXT3ySBx98cNs+kmpAZrVaeeCBB7BarUiSxNNf/w5rG6vYLDZm16Z520++BbPZzKlTp+DU/p+T0+nk9OnTbGxscPPmzVrQceLEifsO4PZiGAbLy8u1sKS+vp5iscjU1BQLCwt0d3fvOKbJZEKSpNp3o7onqPrdVBQFu93OwMAA6XSaiYkJXC4X7e3t+4YsB2nC3EmSJCKRCKqqkkqlaqPQGhoakCSJVCpFKBS6vwsjCIIgCIIgCIIgCIIgCIIgHMj9pxY/QHfu0DAMg5WVFV544QUsFgsXL14kGAzu2WQ5qOo4MsMwSCQSvPjiizgcDi5evLgjgNE0DUVRau0Ci8WCLMu4XC4GBweJRCJcv36dWCyGruu7Hu/DH/4wncc7UFHQuR3kyJiQkFhhkbgeo5LKUJ5OUZ5MYs9aKVMgby/SecyJ1TAo3VgjeWmE4vgKTt2Dw+JGM1RMkgVJMmEUdcobGUwVCSmh4sCFhIyETIE8KgorLJJhC03WyTwxjn45hU21Y8eOBSsWrDjwUqZAMahiPd+IcdJDxX17ZJmmabQ0tqCVNL71rW+xubmJruvMzc1x8+ZN2tra6O/vrwUMTU1NfPTjH0GqV8k5NvmxD7yVB1/34KE/r2rbxOPxoKpq7bjfK7lcjuHhYfL5PENDQ9TX1wPgcDg4deoU7e3tjI2NMTY2tuN7J8syFosFq9Va21ekKEqtNVXd0eLz+RgcHKSuro6RkRFisVgt1LvbYZowd1IUhUgkwuDgIPl8nuHhYZ599lkMw3hVoaUgCIIgCIIgCIIgCIIgCIJwb9I9loQffoP491ipVAJu/8v+bDbL+Pg4DodjR9Nkenoaj8dDJBK5r+NkMhmmp6fRNA2Hw0FPT8+Om97Vm+iapu3Y+3E3TdOYn59nbW1tzx0lL730Eu9867ux5pw4cJFmgzJl7NjQ0CmSx44DgAI5zHU6Ns1M2ONCNUqsxk24PUEUv4E35cJRcaKoRXKkkTCBWUKzg1ooUtYLBIlgx0mFEilWMWNFoULJnsPs8GBJQ0APUyCLgUEbvdhwkCfNjH0c1+s6sClmKktblLJZWK/wvgs/SUkukLIk+K0/+E2KxSKVSoVoNEpXV9f3/EZ/qVRifHwcs9lc+4xUVSUWi5FKpejs7HxVDQ9N02qv1dfXh9fr3fO5hmGwurpKLBajoaGB5ubmXUehVb83uq6zublJOp2mu7t7x3OWl5dJJBLbGitVExMTNDY27ns+u5mcnKS+vr4WJOZyOT72sY9x8+ZNvv71r3P69OlDvd59uvdcQUEQBEEQBEEQBEEQBEEQhP8PMv3u7/7ufo/v++APQrVFMDk5ydLSEr29vbS0tNTaBFX5fL62m+WwqqPHkskkp06doqWlZdvN9DuXrhuGgdlsro2d2ossy9TV1REKhYjFYiQSCXw+37aRUuFwmIfe8kaeevYSsxvT6Og4cOLAi4yMjk7OlqfkKnP6nIOPfsBPJWsitWFlK2CnrJQo57IYhkGmkgJVx8CgTBELVmTDTFAL4dZ9bLJGkQIyMlukyMkZXGf9uHtc+Mw21CUdnxGgjnoMdLKkUShTIMcacfJ1FbKxJeSZMv6sD49Rh6IX2ZBXKJiz1PkDPHvpOZ5++hkamiLY7XYURcHn831PghjDMFhcXGRqaor29nba2tpqn5EsywQCAYLBIAsLC8TjcTwez6GbI9URZz6fj+PHj2O32/d9viRJuN1uGhsbyWQyTExMYLFYcLlc274bkiRhMpmoVCrMzc3h8/lwOp21x6r/9fl8hMNh1tbWmJubw26343DcDuESiQT19fWHGkm22+9ZrVaGhoa4evUqly5dIpVK8eCDh28iHdL/+v0+gCAIgiAIgiAIgiAIgiAIwlF05EOYbDbL1atXaWhooK+vb88b4+VymXw+v2vjZC+GYRCPx7l16xbhcJhSqURXV9e251RDoOruEbPZvG/4cjez2UwkEsFqtTI2Nka5XN4WTLjdbhqbG8mV88xuLaE5DSqlIlbDRtFcQHHrhG12fvLhAFq+wtXrOUw+jdc95CHS42HL42JzMo6syzgMB2ZuBw8uvFix4DDcKJTZJAkY5NiiQBH/m3vBZKW4kqOUzGEqWfERREOlQA4JCRs2su4c2WABeUPBV/LRQDM23UFZydMUjhLq9mOWLNgLHmxZF9ayk5GbI/hCPnr6upmYmMBsNuN2uw913e6UyWQYHR3FarXS39+/5yJ7i8VCJBLB4XAwMTFBNpvF6/Xu2k65k6IoTExMkEqlGBgYIBQKHepcZVnG7/cTiURIJBLEYjFcLlftu1rdLTM5OUlHRweRSARN02pj9u4MqUwmE8FgkLq6OhYWFkgkEng8HlZXV4lGo4cOtBYXF3f83tjYGFtbW3zuc5/j5MmT9zXm7JBECCMIgiAIgiAIgiAIgiAIwv8vme/9lNeWy+Xi/Pnz97xRfNidMJlMhrGxMTweD+fPn8disbC0tFR7/DCjxw4iEAgwNDTE0tISly9fprW1la2tLYrFIm6vm9aHTlHvyiLpEqWxVTIvr6K12LCt6OSzCl94NMnWqoZkM3jjeQv9PSaCATPFhQSpqA/zsooJEy48gE6KFeoIUaZAkmUqIQld1ZAVEy63FyZy+B0RZN3Ncn4chTwpVggSQaGMFz9rvjUcD7YQ2AhCIYcr58TQDBw4MdCZX51lfXUVB040KtR5QvzE0E9TzOVYnogjv1vm3LlzzMzMsLS0hNvtplQqEY1Gqa+vZ3x8nHQ6TUNDA21tbTuuWbWhlM1mOXHiBG63+0DX2u/3c+7cOVZWVhgeHqapqYnm5uYdn191nNjc3Bzt7e1EIpH7Dorg9newr6+PfD7P5OQkJpOJaDTK3NwcHo+HoaGhWoNLkiQ0Tav9MZvN287P4XAwMDBAOp1mYmKCQqFQ20F0GNXXvlMikSAajdaaPIIgCIIgCIIgCIIgCIIgCML3x5EPYWRZ3nETeTc2m41yuXzP5ymKwtTUFLlcjuPHj+/YsaHrOpqmoaoqkiTtuDn+asiyTHNzM3B7x4fdbufkyZN89pEvotXZae/sYC42h601yObSMraUCbliwn2iDbXBS4kZQmcaKfY6eOTrMRbn8pRNdsxGhbJLIVPOUlSLKJTRUFlnFRWFsqmM7VgYp8+Py15HLpFAW8mjhGW0QgVPIkCGdbJsUTAVULQyFbsCTQ484Xpyy4tYbDbsJTdFLUOFMiWK5MgSoYkwzWgozGXH+db4Nzg5cBLp/2XvzgPkqMv8j7+r+r7n7Om5z56ZJJOTDCAgiIh4ga6CuvjDA9Hdn/sTr1UUVwRZZcH7wpVFRcVV0VXZFfECMWKIExNyz0zPfd/T0/dZVb8/4vRmyEwuooI8L/4iVV31rSP/1CfP8ygKhmFgsVhob2/nbTe8jZ/98OcYmoLVYeaKV15BibkMp8lNQo9y5bUv48KLLizcq9nZWQYGBqirq6O1tfW0wxFFUaisrMTv9zMyMkJXVxcVFRX89pHfEurux+vzsGn7RhoaGti+fftpt/k6EZfLxebNmzly5Aj79++nvLycpqamFe/xcqhnMplWhDEmk2nF++bz+di6dStPPPEE+/btw+/3rzl75qnWmve0HMIIIYQQQgghhBBCCCGE+PN6xocwp/rx/WQhjGEYjI+PMzo6SmNjI+vWrVv12Ol0uhD8nO2h8rFYjJ6eHrxeLxdddBHJZJLu7m4mRsboi46xYdN2KotrePRnP0HLZCBjwlFRRtX6jYTnxvFsqMHaUEYstsRAd4aS7Zvwum3k9AyLjx5Gy2SxYKIEPwaQJEaWDOYKNy6PlXp1kXJ3kgFPmgXVTfHGZoycxtDIo9SXmXAXmRlKKOiWIrJ5BRIZzCkVt7cUbSnOojaDjyJiRMiSxoETL6VkSeOjFC/FjC8Os15pZyo6wdjYGLt+v4uvfuUejuwNUYIfl+okklji4e//gve/7UMEAgEy2TT/84OH6Dy3E13X6enpwWw2c84556yogBobG+O//+unxCJR2jpaecVVrzhpeGIymWhqaqKqqoo7P34X4b44zTXNzAzO8Nj8Dj58+4VnNYABiEQi9PT04Pf7ueSSS5ienuaPf/wj1dXV1NTUrHivjg1jliuvnhrGGIaB1Wpl27ZtTExMsGfPHmprawkEAif8+7FaFQzA9PQ055133lm9ZiGEEEIIIYQQQgghhBDH+5sJYVRVXfNf/kciEbq7uykqKuK888477sP0cuuxoqIijhw5Qmtr65pzR85ELpdjcHCQaDRKe3s7Ho8HOFrlsH37doaGhnj0y48znLZiNploSpcyn8lhYGBSjq41uRgmrxjMPtTNSGgGZ0sFLm8JupYnFQ2jzOdRNRUHLqpoBAySxBimh0Q0ynpTmje9rgTFYmM0bOHe++bIJOPk42mqytP8w1UBRpfi2JQy5l1lzIcWye2aI9I1gCdQSSwdIZ1fIsYiTtxYvW5S1Xlms2HyE3EW03NEWMBpt9HT14NDcfB/H3snqqEwPjdGBTWUUE5Wz2DCDJpKb28PLpcLt9uNoUF/fz/RaJRgMHjcbJ/FxUX+/XP3oEatPPHk7/n+t37Affd+k6987W6qqqpO+gxUVSU8u0RteTPpVJbK8ipmkpNMTEzg8/nOynNebp8Wj8fp6OgovEPV1dUEAgGGh4fp6uqiqamJ8vLyFe+2qqpYrdZCFdbyzBhFUcjlclgslkIl1fKx9uzZQ1NTEyUlJauuJ5PJrNrGb3Jy8pTumRBCCCGEEEIIIYQQQoin5xkfwpyu5Q/XANlsllAoRCqVoqOj47j5F8e2HlNVlWAwyNLSEocOHaK8vJz6+vrTnsHx1LVMT08zPDxMfX39qm21FEXhvPPOY/svdmFdcmDoCh1tlzA1Osr01CTpxQhTBw+Sc+XI7JrBFrdiwUo+nCS8tx8jmSc9G8avVWLDTowlYoTxUkKCOComPKY8l57rZX2blZmZOPGYjk3Nk1SSLI0MsK3RhrVRIdyVwVfrZKYvQqZ7Dqtqwj5hxjQaI2D4mSFLhDBRe4TicytxuC0oeYW010zsyVkMdJJLaYaWBtnacC65ZJ5EPIENB36qceIhT54YS2RI4PA6iMYiHAjtwxw4Wu3R2dm56j0fGhrCiKk8uvPX2BNuGsztTB4Y4UPv+zD3fuueE1az5PN5BgcHiSfjeP1ufJ5iYvEYE8MTzM/Pn9Gslaean5+nr6+P2traVZ+zyWSiubmZ6upqBgYGGBsbIxgMHtcOz2QyFVqU5XI5DMMgnU6vuD6LxUIwGCSVShWO1dLSclxwmM1msdlsx611enpa2pEJIYQQQgghhBBCCCHEX8AzPoQ5nVkgZrOZfD6P2WxmbGyMsbExmpubVx24vvyRe/l3y62fSkpK6OzsZGxsjN27d9Pc3Ex5eflprzsej9Pb24vT6TzpzJH6+npe+YYr+cVPfoWW1ZjTp/j05+/ita99LclshMzhBPkSE864GYqsqF4VPZMjM7CATyvGqwVwKE5UxYRJNzPJCBEWyKOhKVmet83O//wixqZ2C9m8wVhfhkzETPz3A5hr7MxlU8TiOQJFNp7cNUF2wktT5VZCo7/FAmhG9k+xSRSNLGUlNWQiUcwhHVNCIRmbJUMcCzYMDFTMhIa7yahp8uRw4CRDGpWjQUeKOO2bW6naUkZf72FKGkt44YsvJZPJkM/nVw1EzGYzUwuT5GM6TZ52dEPDZrIz3DPE9PR0YdbOU83NzdHf309tbS3/+K6388P7foxryUNKS3Lpy59PbW1toTq4XLrqAAAgAElEQVTF7/ef9uyZbDZLb28vuq6zbdu2VUOPY9ntdjZs2EAsFiMUCmG322lubsZut6/Yz2QyoSgK+XyexcVFzGYzuq6vaGXmcDjo6OggEonQ29uLy+WisbGxUP2yViVMLBYrVGMJIYQQQgghhBBCCCGE+PNR1mrh9Scn3PiXsFwJoCjKST+QLw9BHxkZobS09Lhh6PC/rcc0TSvM41hr9ks6naavrw9N02hra8PhcJx0vctVF5FIhLa2tuMqHU4kGo2SSqUoLi5mcXGRC867kOw8LDGDmyJ0F9hLiyizVKIoJsIL4yhhjSLKcKkecmoOTcszYQyhADklzeuvhRdc7OPuexeIGzYUt525iRSxCQuaolH8gja0mSgVqQVKPXkOdidxZBpw2rxMRvpRAA9FGOgkiBFhCXuFG5dWjCUMupYnT5YkMcqoREElwiImzJgUExbDio6OGx927CSIM6tO8NgfHiESiVBXV1dojTU3N8fg4CAejwdVVXG5XIVwZXR0lDtuv4MnfrqHDe5tGOgU+4s5nNrDAz//TyoqKlbcy0wmQ29vLwBtbW2FcGR8fJzJyUlcLhfr1q1DVVWy2SwDAwMkEgmCweAptSczDIOpqSlGRkZobm7G7/ef8nM+9hjz8/MMDAwUKq+OfV9TqRQ9PT3YbDYaGhoK7+mx82KOPdbc3BzDw8P4/X5qa2uZmJjAarUSCARW7HfxxRezb9++017v03B6yZYQQgghhBBCCCGEEEL8jXhWVMKcSnVCJpMhGo2STCbZtGnTca2Zntp67Njql7XY7XY2btzI4uIiBw4coKysjIaGhlUrNQzDYGZmhqGhIWprawkGg6ddVeH1eguhTSAQ4F/vuJ0b3/Zu7LjxUsKcMkOVWoHbXYqKCXPGxGy4nwgLpPUkVuxEWCRDiiwZPvERD/V1dvQSL3lzFNfWIEUBJxMPT5GdmiIXT5EdiWAuchKPNDK2d4R8VsdODi2dophSllggThQVFRMWLFY7eBzk+2P4acCJmzw5huklTpxiyjBhIkuaBlM7Jk1l3pgmQYwYYTJKije+/Q3ous4555yzolLD7/eTSqW47eaPMTUyg9Vm4bIrXwh2lV19BzGKDKiDyYVhKn3VjCT7uPCK560IYAzDYGJiotCi66lVTDU1NdTU1Kz4M6vVyrp164jH44RCIaxWKy0tLcdVpyxLpVJ0d3fjcDhOWuV0IoqiUF5eTmlpKRMTE+zevZu6ujoqKysLYVFra2th5ouu6+RyOTRNK7RQW36HFUXB7/dTVlbGxMQEe/bswWq1HlchtLS0RHFx8RmtVwghhBBCCCGEEEIIIcTpMd16660n2n7CjX8pmqYBq7cm03WdkZERent78fl8VFZWHjfU/dj5GmazudDq6VQ5HA6qqqoKLcZsNtuKkCeRSHDw4EHy+TwdHR0UFxefdgCzmg0bNhDqD3HgyD4cOEnpcbzWUuwuL6qhEJ+bJ5FeJE+eCAsYhoYdJ6X4SRAl41AxGxqh7gS7DhvEa+oZeGAA85yCraIIc5GD1MQCeS2LEcmip7OU61WYsRAljIoJExYqqGWWcZJEcWpOzIs6SWJ4KKKMKqzYyJAiTxYrNpLE0dHYUrednJ6jobKJOFHspRYue+Wl3HrbrWvO2/nkHZ+i9w8D+JUajJiJh37z34SYo2hjkLzXzlxqienBfsK5GV7wqov4l1s+XAhB4vE4Bw8eRFEUOjo6Trvl1nLViKqqdHd3k8lk8Pl8haBj+V0bGBigpaWF2trapz1LBo6+1z6fj0AgwMTEBIcPH0ZVVbZs2bLiPVMUBbPZjKIoGIZR+HuxvO3YY/n9fsbGxlhaWsLhcBSquIaHhzl8+DCvec1rnva6T8Ntf8mTCSGEEEIIIYQQQgghxDPFM74SBkBVVfL5/HHBxsLCAr29vfj9fs4//3ymp6fJZrOF7afTeuxU1lBfX08gEKCvr4/x8XGCwSBTU1OEw2Ha2tpOqY3V6Z7zzk/9Gw6nnQe+9UPIG8zMD6LHslgMK4nUAgoqaVJ/ikMasGBhgRkMj5k+/3r2/WoQI5Em6/Rg6hrGMW/CdX4NvsZacnqGqf4jJHqmWYyHceacaOQBFTMWFpgmQ4Z5prDjxIIVjTxO3NhxMscki8zSxHry5DChkiACGJR4SxmLD7J9WycTk5NYPSau+LuX8KGPfBCn07nmNe/9w16chg+31UO1t5bDi/sx7E7cNi/9w32YszYsWSuOvIcffedBrr7marZs2cLQ0BALCwu0t7efVgu4pzq2OmV8fJyuri7q6+txuVz09vZSWlrKueeee8bv0Vp0XWdsbIxEIkFHRwfT09McOHCAYDCI2+1ese9yiLj8fhuGgclkWhEIWSyWQkXP6OhooTJocnLyuEogIYQQQgghhBBCCCGEEH8ez4pKGF3X0XW9EMKk02kOHTrE0tISHR0dVFRUFGZ7xGIxSkpKyOfz5HI5FEUpfKA+G9UpZrOZ8vJyUqkUhw4dKlQsnChYOBPLA98XFhZ4/d+/ntYNQZ58Yj+zsWlMORVTTqGIMhy4iLKAhoaCQpQl5pRJ8pVmFNWEOVBKcjpOcXoJvxEDcxbVX43d6yFn5EhFlzCG45gykHJnSZZnSVRkyKQTFOdKKKIUHQ0LVuw4yJGlmDKa2IAZMzmyjBACFFLEUbFQW1XDW66+gXllGlelA2+Nk5tuez9vfPN1J2zd1dPTw3e+9l1Sc1kWYvPMpqZYiM9i8Tnxt9Qx2j/I/I4jlKRKafa0Mxed4dEdv8JkNdHU1MT69evXbCF2upYrSsrLywmFQoyMjNDU1ER9ff1ZeY+OFYlEOHjwIC6Xi/Xr1+N2u6moqMDhcNDb20s0GsXn860IWRRFKYSKy4GMpmkr2veNjY3R2NhYONYDDzzA/fffT1tbGxdffPFZvYaTkEoYIYQQQgghhBBCCCHEc9KzIoSBoy3FDMNgaGiIvr4+GhoaaGlpWfFRX9M05ubmKC4uRtf1M2o9djKJRIJDhw5hGAZbtmzBMIxVW5SdKcMwmJycpKenh+rqalpaWrBarTQ3N3PvvfeSjByd/WLDgRUbObIkiJMiSabCIO6OY9laQekLN6DarcS7J2Bxkfe92c3LL/XR3mJl394ZlPJqZodDqDYLzoYKUqYk+UwGR2M5ZredkkwZzrgDEybMWKmglmqaAIUw85QRIEcWBYiyiBMXFeYa7HYblXUB7DY7geZy/vlD7+W1f/9avF4v//PgT9nx2O+Ix2PHhRmGYfDxW+8gPpVgdnGGheQsk0tjKJqJ6NwsiyPjTD52EHNCIa5FGEz04NVKIG5hYnyMpvWNBFuDT/v+H2txcZHDhw9TVVVFS0sLY2NjzM7O4vV6z3gOzLE0TaO/v5+pqSnWr1+P3+9fcU/sdjtVVVVomkZ3dzf5fB6Px7OiCmc5ZDy2ZdpyGHNs1YvdbmfLli08/vjjPPjgg8Tj8ac1z+Y0SQgjhBBCCCGEEEIIIYR4TnpWtCMDmJ+fp7+/n0AgwPnnn39cOyhd1zGZTEQiERYWFvD7/We1ZZSmaYWWV21tbRQVFQFQV1dHRUVFoUVZW1vbGVfFxGIxenp68Hq9dHZ2Yjb/7+PZuXMnmYU8tbQABimSRAmTIo6Ghrm9lFw6S34hiTK+RCwziG9bI/aKIpZGZjErJkbH8zhsJkptKQ4+8jss9T78HUEy9gyBcwNM3f84uSfnUTTIa6XYqCBNAgduVFQUFJy4mUMjRZI0KcLMYcPBVZf+HYO9Q6RiGXp7e3AFbHz5/V+irKyMVCrFZ/7tswzuH2V8fIxkNknnC7bxqc99slC5ks/n2du1lybrOsqragiN9hJTw2yvvYCl1CJ/PPx7Go02SqlgkTnmmcarFFPprcaSVHng2z/k8hdfflZmtORyOUKhENlsli1bthTWuGXLFhYWFjhw4AAlJSU0NjaueEanY3FxkVAoRHV1NcFgcM2gUFEUAoEA5eXljI2N0dXVVahuOfY3qqpitVrRdZ1cLkcmk0FVVXRdL/w9UFUVu93ON77xDfbv308qlTrrFVxCCCGEEEIIIYQQQggh/tezIoRRFIVkMsm2bduOazd17NwXk8nEtm3bGBgYYHp6mra2tsJA8qdjbm6OgYEBqqqq6OzsPC7csdlsdHR0EA6HOXjwIKWlpTQ2Np5yIJDP5xkYGCAajdLe3r7qQPnB0CDltkpiiTgGCmYszCjjmCpdFF+2DXO5i/CObhwVReiZPKQ1EnvGUMwKlhIft/1HhAuubCU+H2NsErSFNJamIqKRWcxpC+neOfRwFn+2AhWFMHMUUUKOLFHCVNNEkjhxIqRI0MNeVFRsOLBgZeeOXayv24Dd60DJ5HHb3NhsNgBCoRDjoUnC00tsLjoXVTGz77d/4IcP/Bf/541vACCVSmG3OLBqDnRTCqfJiYqCyaHgsrqwzdkptpRhUSzYMnas2DBMBj6fl1Q+QTqVLgRxZ8owDGZmZhgaGlo16AAoLS2lpKSEiYkJdu/eTW1tLdXV1adcbZXL5ejr6yOTybB58+ZTfj9NJhMNDQ1UVVUxODjI2NgYwWCwEAYuU1WVVCpFT08PVVVV5PP5FfNipqenaWho4JJLLjml8wohhBBCCCGEEEIIIYQ4c8+aEKauru648EPTNHK5HHB0VouqqpjNZjo6OlhcXOTAgQP4/X7q6+vPqCpm+WO2xWJh69athVBhLcXFxXR2djI+Ps7u3btpamqivLx8zQ/0hmEwOzvL4OAgdXV1tLa2rrlvcWkJpZXFxGMpFKsJVA1zqRd7VQmpnaPgs4BmkN47hclnI19qIzk5j2ECk91M3upg3955yi8MsngggXmdnWwkgTJjITkWxRf2UBWvQVOy+I0acuSYZIQ4UQx0EsQAgwwZLFjR0SmiFBMW4kSIaUscGN5HZXUl5bVluKxeotEoHo8HwzCIxqL41BKsZhu6rlNq99NzsLdwfQ6Hg5b2ZvIjeWLxKFmyaLYcqslE7/hhwCChxXFYXDgsDmL5MOWechS7wURkhHMbtzI9PU11dfUZPet0Ok13dzdWq/WkbboURaGmpoZAIMDQ0BBdXV0Eg0FKSkpOeI7Z2VkGBgZoaGggEAicUZs8q9VKe3s7iUSCUCiEyWSipaUFp9NJPp+nv7+fRCLBpk2bsNvtx4WU09PTVFZWnvZ5hRBCCCGEEEIIIYQQQpw+xTCME20/4ca/FMMwSKfThaHjx35YXh5OvtqHd13XGR0dZXp6mmAwSGlp6SmdT9M0hoeHmZ+fp7W1leLi4tNecyaToa+vj1wut2qLsmQySU9PD3a7vTD35UTm5+d58cVXEJ/TQdMJq/Po6FQZ9fgcfmKZRabzw7RkO4gqi0yaRrBuC2Ar9WAPz6FH4uSjSYrK7YzOKuB04GoKYFJV7Kqb+J4RGmeaSBDDhIk5JlgyhSnSS1ANE2kSeClhnik8+HCrPpJ6AgWFJmUD08Ywi8wSqKzgws2XkHRH+OxXPo3VaiUajfLP734/o13TtBatJ56NYfOb2XBZkHe//12Fa+zq2s39X/0O0fk43f2HWYws4DL7UJImGhsb6Rk9jJFSSelxnOU2Av5KVMXEK1/3Ct70ljfxwQ/czBOP7sLusPH6t7yWG99140mDDsMwGB8fZ2Ji4rTekWOlUin6+vrQdZ1gMHjcbKBMJkNvby+KotDW1nbSZ306FhcX6e/vx2azkUwmV63MWZ4TE4vFuOCCCxgdHT1r5z9FZ28okxBCCCGEEEIIIYQQQjyLPGsqYRRFwTAMcrlcIXxZrn5Zi6qqhaqD3t5eJiYmaG1tPa6l2bGWZ89UVlau2nrsVC23KFtaWuLQoUOFGSIAw8PDLCws0Nraelw7qbWUlZVR01LN75N7yESTkDYoMoopVsrQsznchgu34cFucmA2/Cxqs9irS6lOTfG6N7mxRfP85KEsfVEnjuZi0lNhFIuKIxgg1T9LmgQh9mNgkCNDVslRpdVRRiVLzKOjscQ8PkpxWV14jCLsuotFZlCAIlMZiq6SUdIogTzX/N1r+NJX7uaPvQcxW81s2BhEy2mMjvRRUlmKr8nLa99wzYpr7OzczuO//R17HzvA9tbzmE3O4Klyoi5ZeP7GF3JJ7AUMTvQxzzTf+u43yefzWK1WFEXhzn+7kwOPHOG8kktIpJJ8+3PfpaSkhOveeN2a9zQej9Pd3U1RURGdnZ1n3MrM4XCwadMmwuEwhw8fxufz0dTUhNlsZmpqipGREVpaWigvLz+j45+Ix+PB5XIRjUYxDANd1zEM47h5Mb///e/54Ac/yJVXXnncdiGEEEIIIYQQQgghhBB/Hs+KEAagp6eHmpoaLBbLScOXp7Lb7WzevJn5+Xn27dtHZWUltbW1K46RSqXo7e1FVdUVw9ifruUP/OPj4zzxxBMYhkFtbS3bt28/7YDn2muv5be/24mRzWHHiR0HbnyoukKEBXJkMTTIkyev5LBG4lz1EgfBZjvOuIUjE1Zs563DODKONW7B6IuRTOqYqzzos2lKqUTHIMoiqmHCQzHzTFFMGQFqWWKeBWZJaypV9nrmc7OkSRE1FjEZJhSnwXVvewPvfu+7ePLJJ9k3P0jD65+PxW6l/4kjvOBll/K8zvPIZDI0Njbi9XpXXN/o6CgDB4Y5P/h8otEojZVBQrGD1G2qZf/AHhwWF3l7ln/6x3esmHMC8MRju2j0BPE4vHgcXuK5Jn764EOcd/55hUBkma7rDA0NsbCwQHt7+3HrOFPL7eimpqbo6upa8WfHnv9smZubo7+/n4aGBtavX4+u6wwPD9PV1UV9fT0VFRUkEgluueUWRkZGePDBB6mrqzvr6xBCCCGEEEIIIYQQQgixujMr8/grePzxx3nxi1/ME088ccb/ir+srIzOzk7y+Ty7d+8mHA4XPsjv37+f2trawiyNsymTybC4uIjb7cbn87G4uEgqlTrt41xzzTU4zZajwQsKNhyMGSEWmGaWcZLEmWeSUXrRVQN9MUcuaaDFs9z7jSiZpAPTjIZ7wkqdpZUqcyNKb4Lwr4/gNYrJk8OOnSJKUVCYYxIdjRICaGiYMGPChNlqIcoSenEWw5FjyT1LtjLOxVefz9v+4QbMZjOhwX5KNzZSWVuN2+PB1VTBrn17CAaDbN68edXgIxKJkIymSKWSVFT4KS0uxaSY+Yf/93Ze9faXc8FrzuHGD/8TF1xwwXG/9ZX4iGVihf9P6ylagi24XC52797N+Pg4hmGwtLREV1cXJpOJ7du3n7UA5lj5fB5VVfH5fESjUZaWls7q8bPZLAcPHmRqaopt27ZRWVmJoiiYTCaam5vZsmULDz74IBdeeCEXX3wx559/Pg8//LAEMEIIIYQQQgghhBBCCPEX9qyYCbOsv7+fG2+8keLiYj7+8Y/j9/vP+FjJZJKDBw8W5mg0NTWdceuxtaw1k2ZpaYlQKERxcTFNTU2n1Qbr//3fd/Kb+x9nQh+hghrsOEmTZIFpNHKomNExaFY3kPClsJdMcuWLzPzoR1lyyQqSRRnKw+WkcnF0i44JMxPJfjwUUUMjbnzEiTLGAAmiOHFRRRN2HAAM0UNtayXr16/n4ssu5jXXvLoQcDQ1NRUCrF/88pd878AjbLzqEhRFYWjPYXz9CV522YtpamrC7/cXwjTDMJiYmCAUCvFf//ljAmotZT4/o7NDFAVdfODD7z9p8LZv3z7+6c3vxJn0oSsGekmGb//gPmpqagoD62dmZrDb7WzcuPG4GT1nw7HtzZafazqdpr+/n2w2S2trK263+2mdY2ZmhsHBQZqamqioqFh1n6WlJW6++WampqZIpVI0NTXxhS984c8SOJ0i6X0mhBBCCCGEEEIIIYR4TnpWhTBwNNj40Y9+xMc+9jHe+ta3cv3115/2LI90Ok0oFELXdcrKyhgbG6OmpoaampqzNisjHA4TCoUoLy+nvr7+uDUuD4QfHx8/LpQ4kR//14/5t3/5JMnpHBPZIcyY0dDJksaCFRNmFBRqaMaDjxQJxi2HyRsqDfl1jCghig0/YFBLkIya4oi+GxWVVjaTJsUS87gpIkmUJHFsOHDiJkmMlDXOoYEDlJSUrLo+wzAIh8OkUinuvf8+hlLzmBxWbItZPvLuD1BaWkp/fz+pVIrW1lZUVaW7uxuPx0NzczMzMzN8++v3Mzs1R7C9mWvfdC0+n++U7vnAwACPPPIIJpOJK6+8shDSzc7OMjAwQCAQIBKJoCgKra2tOByOUzruySy3AZufn1+zvVkkEiEUCuF2u2lubsZqtZ7WOTKZDD09PZhMJtra2rBYLKvu9/Of/5xbb72Vm266iWuvvRaAX//611x66aV/lpZop0hCGCGEEEIIIYQQQgghxHPSsy6EWRaLxfjoRz/Krl27+OQnP8nWrVtP+ptjK1NaWlooKysDQNM0BgcHCYfDtLW1nfJH/9Vks1n6+vrIZrO0tbWdtOJief9MJkNbWxsul+uE+2uaxt1f+gr/+c3vkkll2XReBzfc8Fa++pPv4N3WyENf+BbzB0dw4MSOkxxZEpYEvvZ61MNJ8noWA40iyqigFh2Nbp7Ejp1SKsmSopwqcuRQUJhhjDzZP/0/NLTX8cvf/WLV69I0jXu/+jV2ProLBYVgRxNXvOIKTCYTwWBwRRVIJBLhwIED6LpOR0dHoUoIjlZ7dHd3Y7FY2Lp16xlXrSwHF6qq0tbWVgg+FhYW6O/vp7S0lIaGhqcVTkQiEXp6evD7/dTX15+wmsowDGZmZhgaGqKyspK6urqTVl8ZhsHU1BQjIyMEg8HCO/tUCwsL3HTTTeRyOb70pS+tWSXzVyIhjBBCCCGEEEIIIYQQ4jnpWRvCLNu/fz833ngj69ev55ZbblkzQFlcXKSvr4/y8nIaGhpW/fgdj8fp7e3F6XTS0tKyZrXBapZbao2NjZ1WZcuySCRCb28vxcXFNDY2rhoM5PN5BgYGiEajBINBvF4vqqoyNTXFh77wCTa95aX0P7GfJx/dyYEHH4OcDoDD5MVZVkJyPkxOS+LCi4JKI+1YsXOY3ZRTSZYMSWJU04SCSpIYMcLkyZMigU8twe62sfmiDdz7zf84Lhz59a8f4ftf/iFb6s5FVVQOjT3Jea/Yxhvf8sYV+4XDYXp7e6moqMBmszEyMkJNTQ3V1dUMDQ1x522fxJ52k9dz+Bpc/MttHz5pOLXWs2hpaaG8vLywLRaL8dvHfks0EqOi0o/T6aS+vr4wV+VUaZpWeBbr1q07rfVpmsbo6CgzMzM0NTVRXl6+6rnT6TTd3d3YbDZaW1vXDIsefPBB7rjjDm655RauvvrqU17HX5CEMEIIIYQQQgghhBBCiOekZ30IA0crXL7xjW/w+c9/nve+971cffXVhZAlk8kQCoXQNI22traTtqAyDIPp6WmGh4dP+eN8NBqlt7cXn89HU1PTGVdWrBXkGIbB3NwcAwMD1NbWUl1dvWJNuq5z+6f+jSlfnor2Bub6RimZ1Ti3YyvX/v0bKKKUgLMJTdUYih/CwEBFxYYNK3aSxHHhJkAdEZbQyWPHQZ4cHoqYYYIiyvCbKnHYnYyb+rnq7S/lo7fdAsD09DTz8/P86IEf0fWrvRR5StnctI1sLk2uIs6tn/goALlcjr6+PtLpNOvWrSs8C03TGBoaYmFhgZ/++CHUaQfVZbUAHBjew5VvewlXXHEFe/fu5b57vkUsGmP7edt4y9vewtLSEoODg7jdbjZt2kQqlaKnp6fQ9uvYZ5FIJPjoh24lOZ7DaXExn5/mLTe+kUAgUAi2ioqKTvqcFhcXCYVCVFdXP60WdplMhoGBgUJrNo/Hs+I9GB8fp7W1dc3Wb7Ozs7zvfe/DbrfzhS98YUU10TOMhDBCCCGEEEIIIYQQQojnpL+JEGbZ/Pw8N910EyMjI9xxxx385je/Ydu2bQSDwRXVEKcil8sxODhILBajra2t8IH8WMsD3+PxOO3t7U976PqybDZLf38/6XSa+vp6xsbGMJvNtLa2rjlLJB6P8/0f/ZDBiVEaKmt43auvxuv1smXTVlIDGl6KGKWfEn8WI68wE86iKjZKdD8qJmYYx44dHY08OVTMBKillAqG6KaR9bgtHnyeIkaS/ZSd6+LBh3/Cr3/9a+7/6ndJRJL8bucOfEYpGnnipiVecN5lXPaaS7jhH95aaMPV2NhIRUXFqsFFKpXinW9/F+XZauoq6zGZzfSN93D+1VvpPLeTW99/O63eDbgdHronDlEUdDE9MotL95DSklS1+XnJK17C+vXrV62I2rFjB9/9/A/ZXLcdgEhiiUl1kM//++dIJBKEQiHMZjPBYBC73X7c75dDpEwmw7p161bd50zEYjFCoRB2u53q6moGBgZwu920tLSsOu/IMAx+8IMf8JnPfIZ//dd/5aqrrjor6/gzkhBGCCGEEEIIIYQQQgjxnGS69dZbT7T9hBufaZxOJ1dddRWJRILrr7+efD7P9ddff0YVAiaTibKyMjweDz09PcTjcYqKilBVtTDb4/Dhw1RUVNDa2orNZjtr12EymSgtLSUajdLX14fb7WbdunUnbI9mtVrZtnkLlz3/Es7ZsrWwnopABQ/++EEWlQUa2jO8+x3FXPVyD74Shb6wQq7dw+LEKF58lFFJDS14KSXMHBbFih0ncWLoaHiUIqw2M6O5AYJbm7no4ov49O2fY3NZJw8++mP82RrqCOJXqsnoaUILh/jMFz9dmJGzceNGfD7fmpUjFouFbD7Dvr37IKMQTUSYyU/wqtddxejoKINdY9T5GzCpJopdpfzw4e9zcfBFBHzVWHI2+ob6uOQlz6ehoWHV4/f399O7ewC/LwCAYejMpCZ5xatejtVqpbKyEpPJxJEjR8hkMvh8vkJF1ezsLIcPH6aysvK0W9WdjM1mIxAIMD8/T19fH8XFxbS2tq4awExNTXHDDTewsLDA9773PTZv3rFmp6MAACAASURBVHzW1vFndNtfewFCCCGEEEIIIYQQQgjx13DmE8mfoX7yk5/ws5/9jJ07d/LQQw/x8pe/nFtvvZXLL7/8jNpGeTwetm/fzuTkJLt376aqqoq5uTlcLhfbt28/qx/jl4XDYUKhEH6/nxe84AVMTU2xe/fuE1aRrGX9+vVUX9RIz5NHuGC7k4oiB4oBz+u08fAjC8w9OYldcaIbGgmipEmSIo6CwqwxSYYMCgYzjBHVF7AmbZh9Ch+4+f0sLS1hxYbD5iSVSlKDF5NiwlAMrIaVVDLNhduez/ot7XzrO9867l5lMhn+579/ysjACHVNdVx51St41atfRSaT4fFHf09ey/O8y8+luLiYSCRC2khhGAaKohBJLAGg5wwiqSXKSstYyJeRSCTWvBcbNmzge44HGJsbwe3wMDDXy2XXXLJin7KyMkpKShgfH6erq4uamhoWFhYwmUycc845a1YiPR3xeJzu7m6Ki4u5+OKLmZycpKura8XzNgyD+++/n7vvvpu77rqLK6644qyvQwghhBBCCCGEEEIIIcTZ9TfVjgyOzhdRVbUQVIyNjfGe97wHTdO46667qK6uPuPj9vf3Mz09jd1up6Oj47SGsZ+KbDZbqBppa2tbMfh+uUVZKpWira3tlFuffepLn2PH1GF+8+UfcOnzrFx+sRMDg7HRHA/8JI09WouXIhaYJUYYE2Z0dAwMLFjJkMRHKVkyJImBAuWVZdxz71epr6/nXz/8cZoc67nnJ1/GnnJRQzMZkgxwhBLKKaKMBfMMRUEXj/9hR6G6Q9d17vrEJxnaM4bfFWA2MUPDtipu+vBNheqT5eseGBggEonw85/+gsnuGWyqg7A2j2oHR8zL5pZtRBMR+mNH+Ninb6WqqmrN+zE4OMj3vv19YtEYnRd0ctUrr1x1ho9hGIyNjTEwMIDNZmPDhg2rtjh7OnRdZ2RkhLm5Odrb2/F6vSuue3BwkO9+97ts2rSJ73znOzQ2NnLXXXet2hrvGU7akQkhhBBCCCGEEEIIIZ6T/uZCmNUYhsHDDz/MzTffzOte9zre8Y53nFYFy9zcHAMDA1RVVVFbW0s0GqW3t5eSkhIaGxtXbRt1uuubmppiZGSEpqYm/H7/mtUukUiEUCiEz+ejqalp1QDhWB//zF3815OPMfCjXVgrHTgbysjGUqTnYjiryjHNG+gTCYr0EqzYmGGcBDEaaMOGnWnGiBIGDNx4CdDACD2YsaCoKhbP0QqRZDTJrj/+ATt2DAy8lFBPK5qSw6Y4OGTexWN/fJTm5mbgaFutD73zXzi36qJCpUfXxON84ou3rxqixGIxDh8+zMjICKlUipqaGjZv3sy3vv5tDuw9iK/Iy1vfcf1Zac+VSqXo7u7G4XAQDAZJpVKFmS0tLS1npfVcLBaju7ubsrIyGhoaVgRPywzD4M477+TrX/8669at4+tf/zq1tbVP+9x/BRLCCCGEEEIIIYQQQgghnpOO//L7N0hRFF72spfxxBNPkEwmueyyy9i5cycnCaBIpVLs27eP6elptm7dSl1dHYqi4PP56OzsxGazsXv3bubm5s54bfF4nD179hCLxejs7DxpuzGfz8f27dtxOp3s3r2b6enpE17HC86/iLn9A6ildkqu2IqluARr3oU1bUXvjmDBStqZxk8tLrxUUo8JM0ssMM04BjoWLLgpwoSFUXopo5KNnM9WLsIWdbNnzx/5+Gdv5803XEelu5YypRI7DmyKAx0dHQ0DVg0anmqta3G73dTV1VFZWUlzczOtra2Ulpbyvpvey33f+zpf+Ornn3YAYxgGo6Oj7N+/n8bGRtatW4fZbMbj8bBt2zbKy8t58sknGRoaQtO0MzqHruv09/fT09PD+vXraWpqWvW+DA0NcdVVVxUCv/e9733cfvvtT+v6hBBCCCGEEEIIIYQQQvxlPScqYZ6qp6eHG2+8kaqqKj72sY9RVla2Yvtym6iZmRlaW1spKSlZ81iZTIa+vj7y+TxtbW04HI5TWoOmaQwODhIOh49rRXWqcrkc/f39JJPJVVuU5XI5+vr6eNN1b2YgPEnRpiZ8Ew6KrRVkF2KMjh/AW17J/NwompGnhjoSJLBiQ0dnnklMWKighhL8pEgwyBFa6KAYP2azmXljmgEOcflVL0RVzBzp6sU662IqO4YLL2bMLCnz1G2p5tEdvy4EDqFQiFs+9FEWxyNsrNtMIp+gdmuAD/3LB48LJZYrU+x2O8FgEFVVGR4eZn5+npaWFkpLS0/73j3V8lyWoqIimpqa1qxu0nWd0dFRpqenaWxsPGHV0lNFIhF6enoIBAKFQG+1499zzz185zvf4Ytf/CIXXHDB07quZwiphBFCCCGEEEIIIYQQQjwnPSdDGDj6sfv73/8+d9xxB//4j//Iddddh8lkYnFxkb6+Pvx+P/X19adUvQEUfldeXr5me6lly+3NqqurqampOeWP+GtZrpbwer00NzdjMpmYmZlhaGiIhoYGdu3axRvf/lacFg+NxjrMqgUjrxGOTzPBEFasGBjkyFFPC8VUkCXDCL34KCFAHU7cJIgRYj9lBKhTWjGbzAxqR5gyRigrKSef0YjnophNFshDSktgs9q46IUX8vX7vl6YoXPo0CE+8S93UqyVMzUzxYw2wZvffh3Xvek67HY7cLQqpa+vj97eXsxmM+effz7FxcVkMhl+9atfMTs9R0NTPSUlJSiKQmtr6ykHYMfSdZ2hoSEWFhZYt27dKc9byWQy9Pf3k06naW1tPeHvNE1jYGCAWCzGunXrVsz6OVZfXx/vfOc7ed7znsdtt91WuBd/AySEEUIIIYQQQgghhBBCPCc9Z0OYZUtLS3zkIx9h9+7deDweXvSiF3HDDTec8Qf95SqJYDB4XIVGOp2mt7cXVVVpbW09K7NFlhmGweTkJMPDw6iqitfrpbW1FYvFgq7rlPnKsRsu6tQ23HhJajHmmCBOhHa2AQq97KOYMiqpI06UUfoo5+h8lmoaSRFniBBxwngoAhQSRMiRxUsJKirF+EmTIOWNcPsnP8Y111xz3HV+7CO3kwxpVJcenW/SPX6IC6/Zzhuue0PhWr7y5X/nZw/8HK+9CMWl88+3vBev18stH76FpZEYDWUthPPzXPl/XsoVL72CAwcOsOePe0nH0wTXBXnV370Sq9V6wnu2XJlSUVFBXV3dKQdux4pGo4RCIZxOJy0tLcedc3FxkVAodMLATdM0vvjFL/LjH/+Yu+++m87OztNexzOchDBCCCGEEEIIIYQQQojnpBNPdX8O8Hg8rF+/nl/+8pcATExMkM/nz+hYqqrS0NBAIBCgt7eXiYkJWltbsVqtjI2NMTU1tWo4czYYhkE2m0VVVRwOB+l0mkwmg8ViQVVV2lraiYeyzGqjpCgmQYwl5mmgHTc+DHQC1DLFCDYc6IpBzsgSJYwFK0f4IxlSuClivXUbh/K7aQw048zZWZhbwI2PcipxU8QS8yxGVW78h3czOjrKzTffvGKtmXQGi/l/q0EsJguZTBY4Gkg88sgjPPzAL7i45XIcdgfh+CK33Xw7uVSOcF8cm83GYHaAizdeyne/+X1mZ+Z4+Kc/x55043cHOPS7Hob6B/nAzR9YM/To7+8nFovR0dFRqNA5E16vl3POOYeZmRn27NlDVVUVtbW16LpOX18fqVSKzZs3rxnqHTlyhBtvvJEXvehFPP7442c1mBNCCCGEEEIIIYQQQgjx1/WcD2E0TSMWi7F3717sdjv33HMPl19+OTfddBOvfOUrz6g6wm63s3nzZubn59mzZw+6rlNZWUlnZ+eas0aejqWlJXp7eykvL+e8885DVVWi0Sjd3d14vV6cTif9ff3U04YTNzoGCaJ/akGWLhzHjIkkcQY5gmocvW6NPAli6GhU0UAZAab1EerKG7GpDsLxMDoaOjrKMf85cFBKgM984nO8613vwuVyYRgGO3bsQCPPnsGddDY+j7yuMadNcsFF1xeqRlKpFIGSKhz2o8FFkauYgccHeH7wMpyOMMWOUvrChxieHqK/ZwB3zsfCQIQacwllAT+Vtmoe/OFP6Q2FcFgdnHvhufz9ta/HarWysLBAX18fNTU1tLa2Pu1WcACKohAIBCgvL2dkZISdO3diGAZNTU20t7eveo5cLsdnP/tZfv7zn/PVr36VzZs3P+11CCGEEEIIIYQQQgghhHhmMd16660n2n7CjX8LTCYTF154IVarFVVV6ezs5NWvfjV333039913H9u3b6ekpOS0j5vL5RgfH0fXdYqLiwmHw7jd7jNqc3aic/T29jI3N8eGDRtWDIm32WxUVVWRzWZ5+9vfzsTwFHny2HESYYEwc+TJkSRBhjRRFplgGDBw4SFLiioaCLKJOlpIk2COCWaZxFPjwkgarLNvpYQAmVSGMLMkSZAjQ5IEMcIEqGWRObY/bxuNjY187Z6v8YP/+DHqoo25+CyTyRHqOqp4yzvehNlsZmFhgY6ODkpLS3noZz/DoxZht9oZnOljbHGI81svYnZujnwmT1JLMBodwGF1srV5O5Mzk1RYqlmMLzKRGqI31EtkJMHE6CTTQ3PMx2Zxe90sLi6ycePGwiyZs0nTNGZmZlAUBYfDQSwWw+v1HteibP/+/bzhDW9g/fr1fO1rX6OqquqsruMZ6La/9gKEEEIIIYQQQgghhBDir+E5PxNmLYZh8Nhjj/H+97+fl770pbznPe85pUHphmEwPT3N8PBwoTWZoigkk0l6e3uxWCyFFmVPZ20zMzMMDQ2tOMda/GV+3KkS6mklTQIFE0N0Y6BjxkKcCBo6Ljx4KSZGmDQp7Dix46SBNnJk6eFJmlrreelVL+Whb/ySOmsLhmGQTCfYvbQDBRUTZjQ0gmzEAAY4yO/3/o6xsTG+eOfdXFR7GRbz0Tk1XZO/458/9h5yudxx19HV1cUXP3U32VSGmsZqSsqK6X9ihIaiIEd6D3MkvI+UniAX0bBabZitJjLxLDmyWLHRYumgvqKRmcQEC5YZbOUmvvXAfVRVVZ318AVgdnaWgYEBGhsbqaioQFEUIpEIoVAIj8eD3+/H5XJx1113sWPHDu655x42bNhw1tfxDCUzYYQQQgghhBBCCCGEEM9Jz/l2ZGtRFIVLL72UnTt38ulPf5rLLruM22+/nUsvvXTNj/iJRIKenh6cTifbt2/HYrEUtjmdTrZs2cLc3Bx79uyhpqZmzUHtJ5JMJunp6cFutx93jjVpkCWNGQtF+EkTByBNiqMTSI62EDNjJkuaPHmqacSEmQgLhNj/p9ZlGSqKqzj4q17SuTRWn5mt27YxONfPwccd1BpBLLqdOSbo4yAqCm/9f9fT1taGy+Uin8sTjUTx+XwoikIqmWZmZobLL7/8uFDq3HPP5Zvf2046nS7MuLn33+9l1++7KN7ioeiID9ewj1K1ilw2Q09yPyXWcrweN4lYCjgaVrlNPkbTAxS7KhkdHcVut5/VmTzZbJaenh4UReGcc85ZcR0+n4/t27dz6NAhnv/852O323nzm9/Mjh07/ixt6YQQQgghhBBCCCGEEEI8s0glzCkaHh7mXe96F1arlTvvvJNA4P+3d5+BVZXp2sf/e6dX0hsJkBBS6BBgVBCceREGUSFHRkVQZN7hIMehjOigoAMKSAmoDIoIAtbhjGKBAYQXRXHEFnoxPaEESN/phey91/uBQw5IqIIBcv2+wX7WWvezEr7si/t+Qho+s9lsHDp0iOLiYmJiYvDx8bngvWw2G9nZ2VgsFmJjY2nRosVFn2+32zl8+DAFBQXExMTg6+t7ybX373cn+3ccxBNvPPGhHAulFGFgEEQYReTjhTfhRFFCEVbqCSEcD7ywUEQWBznJSX57+x109u6Ft1sLkjO/46fsg8REx3K0PBv7CUe6BdxCeVk5peUW9hjf8tLSJEaMGIHJZMJutzPtr89yYl8RXmYfiioLcIt0ZMnyVy+pw+hMhYWF3NHzd3Rz6Y0TLpSVlpFTnYZ/Gz+6xybw9Y6vcC53w9c9gHKjhDLPQqbP+xsDBt5JRkYGdrudmJgY3N3dL+u5Zzqz46lt27YEBQU1uq6mpoYXX3yRXbt20aFDB3744QeWL19O165dr/jZNyB1woiIiIiIiIiIiEizdPmnzjdTbdq04dNPP2XEiBEkJiby+uuvY7Va2bVrF8nJyTg6OtKjR4+LBjBw6hyadu3a0aFDBzIzM0lJSaG+vv6860tLS0lOTgagZ8+elxXAACxe+necXZ2ooZrj5FBMHk640Inf0IY4bFhxwwsTDlRSChgUU8BP7OQERzDjQK+ePbFabXi4emIymegZfSvt2rbjdw/04dH/HIXhYMdut+PdwhuTjx0XP1dKqyo4cuQIAGazmRGPPkS9XzWH7Km0viWUPzx0HxaLhYsEgedwcnLCycWJsnoLTo5OeHp5Um2qJMg/iECvYIL8gzjheojy0HxcYk3Me3UO9w65Bzc3Nzp37kzr1q05cOAAGRkZWK3Wy3o2QG1tLXv37qWkpIQePXqcN4D57rvvuPPOO2nTpg1ffPEFr776KmvWrDkrwBMRERERERERERGRm5c6Ya5AVVUVU6ZMYePGjURERPDRRx9dcVfFmee7tGrV6qwzS+rr68nIyKC2tpa4uLhf1LmRk5PDtGem8a9/rccRZwIIwR0PKiijmDy88aMFfpgx0wJ/rJykgGMEEQ5AZK9wKl0ttKgPICaoPWXVFg7Vp7P4zUV4eXlx3z3DKE2txrAb5Abk0/6BfgQFBFL9YzYvT59DTk4O856fj69DEIbZRsv2ITw/ewa5ublUVVURGxuLl5cXACUlJRw9ehQ/Pz8iIiIa3c+c2XN594338bL7UGOrxuZ5krahMTjbXLB7nuSxyf9Jly5dCAgIaHT0l2EY5Obmkpube0nn6py+5vjx4xw9epR27dqdd6xZVVUVzz//PKmpqSxfvpzIyMjL+VHdjNQJIyIiIiIiIiIiIs2SQpgrsHbtWmbMmMGf/vQnPvroI6KiopgxYwZ+fn5XfE+r1UpWVhYVFRXExMRQVVXFoUOHLjkguFQlJSXMnT2XL7/YRnFREZ5uXhSVF3Cy0oYjToQQgTuelFCIO564mtzo07c3bl6u5JBK5+6d2PnDTnz9fBk38TE6deoEnOrWefftd/lww6cEJvagQ98eFBQWcmDjt7jvKyKoRRC+lUFE+EdiGAa7c3/g4ScfZPDgwVRUVJCWloanpycVFRXMez4JZ6sbNfYq/jDqP3hoxEPn7MNut7Nx40Z+/P5H/AP86da9G0VFRWRnZ1NTWUN8x3gGDx6Mm5vbBd9HfX39We/9fKPhampqSElJwd3dnejoaBwdGz9Oadu2bUydOpVx48YxZsyYq/Zzu8HpJYiIiIiIiIiIiEizpBDmCuTm5uLr64uHhwd2u513332XBQsWMH78eB566CHM5iuf8lZYWMiBAwdwcXGhW7duFw0RrkRaWhp//csUftqbQnVtNSFhweQX5lFVUYU/obQmFguFVFJK9za/oUfPBHIKs/Dp4Mbs+bMueO/Jzz1DVfcArJ5OZKcfojotD6d/F1Bcms/dne4jwCcQgNRjB/jdo70Z+fBI4H+7TP7vyD8R796VsIBw6qwn2VXwLQvemEdUVNQ5z7Lb7eTk5FBcXEx8fDwfr/mEde9uINA5lPL6UkI6+DNv4VycnJwu+k4qKytJS0vD1dWV6OhoXFxcGurKzc3l2LFjxMbGnncUXHl5Oc899xzHjh3jjTfeOG8HTzOlEEZERERERERERESaJZ0JcwXCw8Px8PAATp11MmrUKLZt28auXbu45557+Omnny77nqcDhezsbLp27Urr1q3Zs2cPeXl5l31myvkYhkFOTg4j73+EY9uLaFkWQ9u6zhTkFFNXYcVkNmP2gDzPHOo8Kqh1q6LY/Rh78pMpcy/ksfFjL/qMu383gOx133Hw3zswjlVStyOX37S+DV9XX/Yd2YnNbqOqrhILhXTo2KHhOpPJhLe3N84OLni7+FBSUoLZMOHu4ElhYeE5zykrKyM5ORkHBwd69OiBq6srH63+mO6ht9I2OIauLXtyLCWPAwcOXNK78fT0pHv37gQGBrJ7924OHz5MRUUFO3fupKam5oJn8WzZsoWBAwfSu3dvNmzYoABGRERERERERERERABofKaSXDY/Pz+WLl3KDz/8wPjx4+nduzdTpkxpCGsuxGKxkJ6eTlBQED179sRsNuPr60tQUBAZGRkcO3aMuLi4S7rX+VRVVZGSksI332yn7EQFIbY2GIYdMDBhJpBQnO0u5FUdJSG+J93a9GJf3g56DOhKjx496N2793lHdZ1WV1eHn58f9/cZxGtvLsPZ8OWOoD4EtwihdUgkLhFmth//HGcXJ/7vpNF069btrOs9PT0JCA3AYikmtEU4eUUnKKzKIzAwsGGNzWYjMzOTyspKOnXq1HBOTl1dHYbdwNHh1K+0yWTCweyI1Wq95HdkMpkICgrC19eX/fv3k5mZSXR0NK1bt250vcViYerUqZSVlbFp0yZCQ0Mv+VkiIiIiIiIiIiIicvPTOLJrwGq1smTJElasWMG0adMYPHhwo2eD1NfXk5GRQW1tLXFxcQ2Bws+VlZWRlpaGn58fkZGRjR40fz5njuyKjY1lyd+XsGrxO/hWheJLIAY2fmIXrrgTThRVlJPFT/h6+OEf5sOaf31IdXU15eXlxMbG4u3tDUBKSgpJLy7kxLETdOjcnkf++DBVVVVER0cTGBjI+n+tZ8UrbxPhFkXNySoqPS0sWvoy/v7+mM3m856VkpOTw/PTXqC8uBIcDB4ZM5KQ0BBCQ0Px8PAgMzOT8PBwWrZsec495sycw/4vU2nt25aSqiJqfct5ddlivLy8Lvl9VVZWkpKSgp+fH6GhoWRlZWG1WomJiTkrBNu4cSMzZ87kmWee4YEHHtDZLxemlyMiIiIiIiIiIiLNkkKYa+j48eNMnjyZyspKkpKSaNWqFXBqLFheXh6HDh0iMjKS4ODgi36Jf+bZJFFRUQQFBV30+RaLhbS0NEJCQmjVqhVms5m3Vr3N/L8lUV5SiQ8B1FCFJ95YKKItHaihisMOqQzr/jClFBPW3Z95L81tODPF3d0df39/Hnt0HGH2KPzdAziYuw/n1vD2P95qOH/FMAy++mob3277Fk9vT4Y9cB8tW7a8pPdms9koLi7G29sbV1dXamtr2bNnDzU1NcTFxZ2346S2tpZ3336Pfbv2ERIWwh//c/Qld6f8/HyZM4Mbi8VCRkYGu3fv5s4772TWrFkYhsGrr756VpeOnJdCGBEREREREREREWmWFMJcY4Zh8PnnnzNlyhQSExPp378/P/74I3369CE6OvqSDo0/U11dHRkZGdTX1xMXF4ebm9s5a87ssImPjz9rza5du+h/+wC88SOAEMBEJWVYKCCYcIrII6ZNLL/vci9Wm5Vvi75g4xfrG/aSl5fHpk2b+HTFBjr5JVBXV4eHhwfJlq95/5N38fHxueS9FBQUsHnzZnYk7yCqbRQPPvggwcHBZ63Jz88nOzubyMhIWrRoQUZGBoZhEBMT0+jer0R5eTkpKSkEBQXRunVrzOZzj0oyDIPJkyfzwQcfMGLECBYtWtToOmmUQhgRERERERERERFplvQt8jVmMpm488472bZtG1999RX/8R//gc1mIy4u7rIDGAAXFxc6duxImzZt2LdvH9nZ2djtduB/Q5IdO3bg6+tLt27dzgkqFi9ajC9BOOJIMfkcJwcw8CWQfHKpoYo72g8AoLiykMCggLP2EhoaSvv27SkozSc9I4OC/AJKKywYZuOyQpGDBw8yZGAiM8bN5pMV65n77AIG/5+7OXbsGHAqbNq7dy+FhYUkJCQQEhKCm5sbnTt3JiIign379pGVlYXNZrvsd3iazWYjIyODtLQ0OnbsSGRkZKPBSn5+PiNHjqSqqoqdO3fi6OjIE088ccXPFREREREREREREZHmQZ0wv5IhQ4Zw2223ce+99zJ58mR8fHyYPXv2OZ0fl8Nut3PkyBHy8vJo1aoV+fn5ODs7ExMTc96AZ+jdiez58gAd6cVBfsSDFoQQQSVlnOAIZZRgYMcRRwygc49ObPn8/+Hk5ITNZiMrK4v9+/cz/ennodQBN7sHxaYCxk8dx1+e+Msl1/7I8Ef5aX02fkYQjjiRbuyl1lzNyHHDGT/xzxw5coR27doREBDQ6PV2u52jR49y4sQJoqKiCAwMvKxzWUpLS0lNTSUsLIyIiIhGrzUMg3/+858sWrSI2bNnc/fddzd8ZrPZLutsnmZOnTAiIiIiIiIiIiLSLDnMmDHjQp9f8EO5dMOGDaNv374EBgYyfPhwAMaNG4dhGHTt2vWKRluZTCZatGhBXV0d2dnZODk5ERcXh4uLy3mvsVhK2br1C6qMCqxYcceLGiopoYCWRFHEcXwJpB1dCCSEjONpHMjYT9++fdm/fz9+fn5s2/o1TvketG3ZDh+/Fvh4+FNhLyWhZwJeXl6XFIa8kvQKTvkeeJpb4Ghyop6TWCjAzcOFvnf0pVOnTnh6el5w7z4+PgQGBpKbm0tubi7e3t44Oztf8Lk2m4309HTy8/Pp2LEjAQEBjdZ7/PhxxowZQ2lpKatXr6Zz585nfa5RZJfl+aYuQERERERERERERKQp6JvkX8mZnSlms5lhw4axfft2Dh8+zMCBA9m9e/dl37O8vJzk5GQA+vbtS2RkJHv27OHQoUMNI8p+7tE/juL+R4ZR61VOGcUcIo2jZGPFShYHMeNABNG0MPnhSxBhtOazdZ9x9OhRunbtSkREBDarDUezI+G+rYgJaU+IbygB/gFUV1ezY8cOysrKLlp7dFw0peZCrPZ6ao0aLEYhNsPOoHsGERcXh6Oj4yW9AxcXFzp06EBUVBQHDx4kPT0dq9Xa6Nri4mKSk5Px9PSke/fuqGh+RQAAGntJREFUuLu7n7PGMAzefvtthg0bxoQJE1i+fDktWrS4pFpERERERERERERERM6kcWTXgb179zJhwgTat2/P3/72t4t+6X96LFh5eTlxcXFndYzY7XYOHTpEYWEhMTEx+Pr6nnO9YRhs2rSJ/3z0MeoqT+KIE174cJJaLBQRT3cCTWFgQA6pFLgcJq84r6FjZPfu3Twz8VnaOMdgNpnJqUll2txn6NOnD5WVlaSlpeHm5kZ0dPR5O1Py8vIYcMfvKcotwWSYMGHQuU8nNm7ecMXv0TAMjh07xtGjR2nTpg0hISGYTCbq6+vJyMigrq6O+Ph4XF1dG73+yJEjTJw4kejoaObNm3fBThy5LBpHJiIiIiIiIiIiIs2SQpjrhN1uZ+XKlSxatIjJkyczbNiwRkdeFRUVkZGRQUREBC1btjzv6K+amhpSU1NxcnKiXbt254wouyXhVkpSKzlJLXF0p546rNSTwi4MDFoSiZV68snlTxNGM2fOnLOuT05O5oP3P8Rms5N4/xBuv/32hs8Mw6CgoIDs7Ozz1mm329m3bx/vvfMe9XVWfnNbLx548IGrcs5KfX09WVlZVFZWEhgYyPHjx88KZX7ObrezYsUK3nrrLV555RX69ev3i2uQsyiEERERERERERERkWZJIcx1prCwkClTpnDkyBEWLFhATEwMAHV1daSnp2O324mNjT1vN8eZDMOgsLCQrKwswsPDCQ8PbwghIkPbElARQZlRRLwpgZNGLdWmSnJdMqk0l3Gyuh6T2cTDo0fwyqJXLuvQ+9OsVis5OTlYLBZiY2MbOnxKS0tJS0sjODiYVq1aXZPzVU6ePMnBgwcpKysjICCAmJiYRrtysrOzmTBhAl27dmXWrFmNjiiTX0whjIiIiIiIiIiIiDRLCmGuQ4ZhsH37dp544gl++9vf4uHhgYeHB8OGDSMwMPCy72ez2cjJyaGkpKQhDLmlx62Up9Rw0nSSYCMCF9woNRfSorMraz/7BG9v76u2n6qqKtLS0nB2dsZsNlNTU0N8fPw1Czzy8/PJzs4mKiqKoKAg8vPzycnJOSuIstlsvPHGG6xevZrXXnuNW2655ZrUIoBCGBEREREREREREWmmFMJcxw4cOEBiYiJubm48/fTTDBky5Io6Uk6rqqoiNTUVNzc3amtrGTJwKPZqE/WcxDDZ6dK9M6+vWEK7du2u4i5OKSoqIiUlBcMwiIyMPKsr52qpq6sjNTUVBwcHYmNjcXJyavjsdFfO66+/Trdu3Xjvvffo06cP06dPv6SuIvlFFMKIiIiIiIiIiIhIs6QQ5jpVU1PDkCFDmDVrFqGhofzlL3/BZrMxf/58WrZsecX3NQyjoTOkrq6O9999H2cnF4aPfJA+ffpcxR2cUl9fT3p6OvX19cTFxeHk5ER2djYWi4WYmBh8fHx+8TMMw+DEiRMcPnyY6Ojo83YLWa1WZs2axT/+8Q86d+7MihUrCA0N/cXPl4tSCCMiIiIiIiIiIiLNkkKYG4RhGHz22WdMnTqVBx54gP/6r/86q9PjctTX15OWlobFYsHZ2Zn27dvj5eV11estKCg4ayzYmZ0vp0eUubi40K5du0bPa7kUtbW1pKSk4OLiQkxMDI6Ojo2uO3jwIBMmTGDgwIFMnTqVLVu2sGHDBpYsWXJFz5XLohBGREREREREREREmiWFMDeYmpoaXnzxRT777DPmzp3LrbfeelljvU6fl9KmTRtCQkKorKwkLS0NLy8v2rZte94Q43LU1taSlpaGg4MDMTEx5w1Yzgxqzjyv5VIYhsGxY8fIzc0lJiYGPz+/RtfV19ezcOFCtmzZwhtvvEHnzp2veF9yxRTCiIiIiIiIiIiISLOkEOYGlZqayoQJEwgLC+OFF14gICDggusvFIycOc7rdDhzJee1nBmMREdHX7Sm02w2Gzk5OZSUlFzSiLLq6mpSUlLw9PQkOjoaBweHRtft3r2bJ554gqFDh/LUU09dlYBJrohCGBEREREREREREWmWFMLcwOx2O//93//NnDlzGDduHA8//PA5gcSZwUi7du3w9/c/7/3q6+vJyMigpqaGuLg4PDw8LrmW08GIh4cH0dHRVxR4XGxEmWEYHDlyhBMnThAXF3fesKauro65c+fy7bffsmzZMuLj4y+7luvZ6X+zVxKUNZEbplARERERERERERGRq0khzE2gtLSU5557jj179rBgwQI6deoEnAo1UlJS8PLyumDHyM+VlZWRlpaGr68vUVFRF7zOMAwOHz5MXl7eBYORS2UYBoWFhWRlZdGyZUvCw8Mxm81UVlaSkpKCr68vkZGR560pOTmZyZMn89BDDzFx4sRL3vONwmazNeypqqrqsoKyJqQQRkRERERERERERJolhTA3kZ07dzJhwgS6deuG3W4nNjaW+++/nxYtWlz2vQzDIDc3l9zcXNq2bUtgYOA5nRcVFRWkpqY2hDVms/lqbaVhRFlxcTFeXl5UVlYSFxeHt7d3o+tramqYNWsWe/fuZdmyZURHR1+1Wq5Hs2fPZseOHSQmJtK1a9fr/awbhTAiIiIiIiIiIiLSLF29b81vMGlpaaSmpjZ1GVdVQkICSUlJrF+/nu+//x4fHx+8vLyu6F4mk4mIiAgSEhIoKChgz549VFdXA6fGoGVmZpKSkkJcXBzR0dFXNYABcHBwIDg4GLvdjsViwc3NDRcXl0bXbt++nf79+xMdHc3nn39+UwUwhmHw9ttvc+LECQCsVitjx46lqKiIqVOn8o9//IMVK1Zgt9ubuFIRERERERERERER+blmeVK53W7nlltu4fPPP2/qUq6q8vJy5s2bx5YtW/D29ubJJ59k9erVJCUl0bZt2yu6p7OzMx07dsRisbB//368vLwoKysjLCyMnj17ntMdU1dXR2lpKQEBAVc8Csxut5OdnY3FYqFTp054eHhQWFjIrl27zhlRNmPGDDIzM/nkk09o06bNFT3vema323nzzTepqKhgzJgx1NXV4eTkxLRp05g6dSoODg48+eSTVz0EExEREREREREREZFfrll+c/vyyy/zwAMPkJCQAJzqNrgZOgm8vb1Zu3Yt7dq1Izg4mHfeeYenn36a0aNHM2fOHGpra6/43l5eXnh7e1NSUoJhGHh4eJwTwGzatInf3vY7hg68j7sH3kNGRsZlP6esrIzk5GQcHR3p0aMHnp6emEwmgoKC6NWrF/X19YwaNYrFixczYMAAunfvzubNm2+6AOb0mEAHBwcef/xxPvvsM/bu3Yu3tzfp6enExsaSkJDAhg0biIiI4LvvvqO0tLSJqxYRERERERERERGRMzW7M2GKi4tp3749+fn5ABw8eJAOHToAp7oObsaOgpMnT7Jw4UI++OADZs6cyW9/+9tzApQLKSoqIjMzk4iICMLCwqirqyM9PR3DMIiNjcXV1ZUjR47w4JCH6OTWC28XH46U5VDhX8D6zf+6pHdqs9nIysqivLyc+Pj48x44X15ezl//+le++uorevbsyWuvvUZISMgl7+VG8/TTT5OTk0NaWhre3t6sWbOGjRs3snHjRpYsWUJAQADTp09n586dLFu2jLCwsKYuuTE6E0ZERERERERERESapZsvcbiI+++/n/nz5wOQkZHBY489xtixYykvL78pAxg4NVLsmWee4ZNPPmH58uU8+uij5OXlXfS6kydPcuDAAY4dO0bXrl1p2bIlJpMJV1dXOnfuTHh4OHv27OHQoUNkZmbiZfbB28UHgFYtIikuKKG8vPyiz7FYLCQnJ+Pm5kZCQsJ5A5jNmzczcOBA+vfvz+HDhxk9ejRvvPHG5b2MG8jGjRv59ttv+ec//8mXX34JwPr16+nfvz+dO3dm8ODB9OvXj5SUlOs5gBERERERERERERFptm7O1OE8vvzyS8xmM6NGjaKmpoaKigq2bt1KfHw8TzzxRMO66upqPv744180vut61KZNGz799FNGjBhBYmIir7/+Olar9Zx1hmGQn5/Pzp07CQwMpEuXLri6up6zzt/fn169emG32yksLKTsZAknbXUAlNQU4ezmjJeX13nrsVqtpKSkkJOTQ5cuXYiIiGi0Q8disTB27FjeffddNm/ezCOPPILJZOLuu+9m+vTpv+CNXJ+WLl0KgNlsJiYmhqqqKnx9fXnuueeYM2cOBQUFPPvss6xbt46XX36ZDz74gLCwMGw2WxNXLiIiIiIiIiIiIiJnajYhTH19PXfffTdvv/0233zzDePGjeOpp55i8ODBREREkJaWht1u57333mPkyJEcOHAAV1dXDMPgIiPbbigmk4mhQ4fy7bffUlBQwJ133klycnLDHmtra9m7dy9FRUX06NGD4ODgC97PbDYTFRXFvffeyx139eObwi3sKfuONOtuZs+fiYODQ6PXFRUVkZycjLe3N926dcPNza3RdevXr+euu+7innvu4aOPPrrpRo/9/Cyiuro6Zs+ezfvvv09kZCQlJSUNXUt33nknTk5OLFy4kLKyMoKDg+nevTtwapzb+d61iIiIiIiIiIiIiDQNx6Yu4NdSVlbGSy+9RFhYGKtWraJVq1a89dZb7N69m1tvvZUxY8ZgNptJTU1l8+bNtGzZEjgVWixatIi2bdty9913N/Eurh4PDw/mzZvHgQMHGD9+PFFRUYSFheHk5MTo0aPx9/e/rPu5ubkxd94chiZ+z/79++nevXtDQHCm+vp60tPTqa+vp1u3bo122MCpkObJJ5/EwcGBzz//nMDAwCva5/WqtrYWV1dXzGYzNpuN7du306tXL1xdXVm5ciXjx49n165ddOrUiTlz5nD77bdTWFjIbbfdxsiRI2nRosVZ91MAIyIiIiIiIiIiInL9MV2ky+PmaQE5w+OPP05ubi6ffPIJH374IdOnT2fr1q34+fkxZMgQBg0aRHl5OdHR0SQmJmK327HZbDg6OmK1WvH29m7qLVxV6enpJCYmAvDYY48xevToX3Q+js1mIycnh5KSEmJjYxsCg4KCArKysoiMjCQ4OLjR0WOGYfDxxx+TlJTECy+8wNChQ6+4juvV3r17KSsro2/fvuTk5HDo0CEmTZrEypUr6d69OyaTifHjx1NdXc2KFStYs2YNa9euxcXFhZdffvmCI96uU+f+oEVERERERERERESagWYZwtTV1fHCCy+Ql5fHqlWreOmll5g0aRLz589n69atbNq0CYDy8nKGDBnCjBkzaNWqFT/88APu7u4MGjQIJyenJt7F1WG1WklMTGTatGnExMQwdepU0tLSSEpKon379r/o3lVVVaSlpeHs7IzVasXBwYHY2FicnZ0bXZ+Xl8cTTzyBj48PL7/8Mr6+vr/o+debr7/+mpCQEOrr65k5cyalpaUUFxeTnJzMtGnTKCoqYtasWQQGBrJjxw7uvfdekpKSGDFiBDU1NQ0j2+x2+y8KyZqAQhgRERERERERERFplprNOLIzubi4MHv2bNLT02nVqhWTJk3ixIkTLF26lLVr1zasW716NSaTiX79+nHbbbdhsVhYuXIlTk5ODWeoNNbNcSNxdHTkX//6V8Ofly5dyg8//MD48ePp3bs3U6ZMwcPD44ru7e7uTmhoKBkZGZhMJqKiohoNrwzDYPXq1SxevJi5c+cyaNCgK97P9cpisWCxWOjZsyd2u53i4mIKCgp4//33AXj22WcZNmwYH330EY899hjl5eXcfvvtHDhwADj1Ows3ZAAjIiIiIiIiIiIi0mw1y06Yxnz44YfMnz+f5ORkAI4ePcqQIUNYvHgxXbt2ZeLEiezevZvg4GCWLVtGeHh4E1d8bVmtVpYsWcKKFSuYNm0agwcPvqzAqba2ltTUVJycnIiJicFkMpGVlUV5eTlxcXENI7WOHTvGpEmTCA8PJykp6aYb9WYYxlnvbc2aNWRnZ/PQQw/x4osvkpCQwO9//3tatmzJ1q1bWbVqFUePHqWmpoaVK1fSoUOHJqz+qrmxk0oRERERERERERGRK6QQ5n9UVVUxceJEHBwcSEpK4plnnsFsNrN48WJee+010tLSmDt3LuXl5WzevBknJyfuu+++m75D4fjx40yePJmKigqSkpJo3br1BdcbhsHx48c5cuQIMTEx+Pv7n/V5RUUF33//PWvWrKFbt2688847LFiwgP79+1/LbTQJm82Gg4MDAPv27aNz586sXLmSf//738yYMYOjR4+ydOlSxowZQ79+/aipqaGuro7k5GT69evXMLbtJvjdUggjIiIiIiIiIiIizZJCmJ85ceIE+fn5DBo0iJ9++omcnByWLl1Kv379GDFiBLW1tcybN4/Kykr+/Oc/XzSUuBkYhsGWLVuYMmUKiYmJjB8/viF8OlNNTQ0pKSm4u7sTHR2No2Pj0+5ycnKYOHEi+/btY86cOfzxj3+84ce6nU9paSnr1q3jpZdeYsGCBfTu3ZukpCQqKyuZN28ec+fOpaCggF27dhEZGcmyZcsawpczQ5wb3M35wxURERERERERERG5iBv6v9dfC6GhoXTt2pV169bh6+vL7t27cXV1ZcSIEQB8/PHHWCwW+vfvT+vWrbFYLDz44IOUlpY2nBNjs9macgtXnclkYsCAAXz33XcA9O/fn6+//rphv4ZhcPToUfbu3UtkZCRxcXGNBjB2u5033niDhx9+mOnTp5OWlsbOnTvZtGnTr7qfX8uWLVtISEggMzOTwMBAlixZQl1dHYmJiVRUVPDhhx8yZcoUBgwYwMCBA1m5cmVDAAPcLAGMiIiIiIiIiIiISLOlTpiLSEpK4vXXX2fbtm04Ojryyiuv0KpVK0aPHo27uzvjxo2jtLSU1atXA2d3L2RkZNCuXbumLP+ayMzMZMKECfj4+PDII4/w+eefM3LkSNq2bXve4CArK4vx48fTo0cPZs6ciZub269c9bX1864Vu93OrFmziIuL4/777+fgwYO88847ODo6Mnv2bN577z0+/vhjZs+eTXx8/Hnvc5NQJ4yIiIiIiIiIiIg0S+qEuYinnnqKV199lZ07dwKnwoQ77rgDd3d3/v3vf5OcnMycOXMoKCjg6aefZujQobz77ruUlZWxaNEili1bdtN1xkRHR7N27VocHBx4+OGHcXBwOG8AY7PZePXVVxk9ejTz5s1jwYIFN10AY7fbG/b+/fffU1NTg9ls5vDhw2zduhWAtm3b0r17d7Zs2cL27du59957GTt27FkBjGEYN2MAIyIiIiIiIiIiItJsKYS5BHfddRdDhw7FMAwOHjzIrl27OHbsGGPHjuXxxx+nsLCQ+fPnk52dzYwZM9iwYQPTpk3Dw8ODDh064ODgwEU6jm44kyZNIioqipSUFOrr6xkwYAC7d+8+a01qaiqDBg2ivLyc7du385vf/KaJqr22zGYzBQUF3H///YwZM4YRI0Zw4MABJk6cyIkTJ9i2bRuurq4EBwcTHR3N22+/jZOTEwMHDgRo+N24Wc/FEREREREREREREWmuNI7sMv3000+89dZbbN68Gbvdzo8//shbb71FWVkZw4cPp3Xr1ixdupQ333yTxx9/nNGjR591vd1ux2y+8bOvn4/N2rt3LxMmTKB9+/ZMnTqVVatWsWHDBpYuXUq3bt2asNJrb+fOnTz55JOMGDGCP/3pT0yaNIkWLVpwzz33cPDgQf72t78xduxY1qxZw/Dhw0lPT2fu3Ln4+fk1l+ClWWxSRERERERERERE5Odu/DTgV9a+fXvmz5/Piy++yJtvvombmxvfffcdLi4utG7dmoqKCjIzM7ntttsYNWoUhw8fZt26daxbtw7gpghg4NxD47t06cKXX35JQkICXbp0wWq18s0339x0AcyZo+VOB5hWq5XMzExqa2sBGDduHAUFBezevZtRo0axcuVKvLy82LRpE7fccgt5eXm4uLg0lwBGREREREREREREpNlSJ8wvZLfbmTRpErGxsTz++OMsWrSI9PR0HnjgAUJCQhgyZAgLFy5k6dKlODk5sWrVKry9vZu67GuqoqICLy+vpi7jqrNarTg6OgIwf/58KisriY+PZ/jw4bz11lt88sknLF++nKCgID788EM+/vhjRo8ezYABAygvL+e1117jvffeY/Hixfzud79r4t38qpQ2iYiIiIiIiIiISLN0c7RlNCGz2cyYMWP44IMPGD58OAsXLiQmJoY+ffrw2GOPkZuby8GDB/n000/p3r07ubm5TV3yNXezBTAzZszg5MmTODo6UlVVxZ///OeGLp+ZM2eyatUqunbtSnx8PAsXLgTgD3/4A3fccQddunQBTgVTrVu35sCBA80tgBERERERERERERFpthTCXAWdOnVi27ZtjB07lpEjR3LXXXfxxRdfUFpaSkVFBYcOHaJv376888477Nixo6nLlcvk6+vL8OHDG/6cmZnJ4sWLSUxMZNGiRfz444/Y7XaGDRvGDz/8wNq1awEYO3YswcHBGIZBy5YteeihhzSCTERERERERERERKQZcZgxY8aFPr/gh3K2Nm3a0KdPH4KCgjh58iSHDx/m3nvvZfDgwXTs2JHq6mqefvrppi5TLtMtt9xCUlISgYGBdO7cmS+++AI/Pz+io6OJjo5m+fLlANx33334+/vTo0ePs7qBFLzwfFMXICIiIiIiIiIiItIU1Alzlbm4uGAYBoGBgRw9epSHH36Y3NxcevXqxd///vemLk+u0Pr16xk/fjwAMTEx7Nq1i+Tk5IY/h4eHA3DPPfcQFhbGRc5aEhEREREREREREZFmwKQvi68tk8n0PNADWGgYxtamrkeunMlkmgjEABOBqcAdQD3gCgwzDKOw6aoTERERERERERERkeuNQphfgclk8gWqDMM42dS1yJUznZortg+YYBjGlyaTqRvgZxjGF6c/N/QPSkRERERERERERET+h0IYkctgMplaAZmGYTj/7O8dDMOwNVFZIiIiIiIiIiIiInId0pkwIpfBMIwjnBovd7oz5vTfK4ARERERERERERERkbOoE0ZEREREREREREREROQaUCeMiIiIiIiIiIiIiIjINaAQRkRERERERERERERE5Br4/4TYuNZ6/4pPAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#cluster by locations\n",
    "import numpy as np\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "def plot_3D_clusters(X, k):\n",
    "    \"\"\"Plot 3 attributes in dataset to explore natural clusters within data\"\"\"\n",
    "    estimators = {'k_means_3': KMeans(n_clusters=k)}\n",
    "    fignum = 1\n",
    "    for name, est in estimators.items():\n",
    "        fig = plt.figure(fignum, figsize=(4, 3))\n",
    "        plt.clf()\n",
    "        ax = Axes3D(fig, rect=[0, 0, .95, 1], elev=48, azim=134)\n",
    "        plt.cla()\n",
    "        est.fit(X)\n",
    "        labels = est.labels_\n",
    "        # Change the 2nd column in X[:,_] to choose attributes for plotting\n",
    "        ax.scatter(X[:,0], X[:,1], X[:,2], c=labels.astype(np.float),edgecolor='k')\n",
    "        ax.w_xaxis.set_ticklabels([])\n",
    "        ax.w_yaxis.set_ticklabels([])\n",
    "        ax.w_zaxis.set_ticklabels([])\n",
    "        ax.set_xlabel('latitude')\n",
    "        ax.set_ylabel('longitude')\n",
    "        ax.set_zlabel('price')\n",
    "        fignum = fignum + 1\n",
    "    plt.show()\n",
    "# change the n\n",
    "df = pd.read_csv('Airbnb_data/listings_SanFan.csv')\n",
    "df['price'] = df['price'].str.replace(\"\\$|,\", \"\").astype(float)\n",
    "X = np.array(df[['latitude', 'longitude', 'price']])\n",
    "plot_3D_clusters(X, 4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Transformation\n",
    "# right skewed: log transform\n",
    "# left skewed: power transform\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "# Create KNN model: 5 closest neighbors\n",
    "knn = KNeighborsRegressor(algorithm='brute', n_neighbors = 5)\n",
    "cols = ['accommodates', 'bedrooms', 'bathrooms', 'beds']\n",
    "knn.fit(X_train[cols], y_train['price'])\n",
    "features_predictions = knn.predict(norm_test_df[cols])\n",
    "features_mse = mean_squared_error(norm_test_df['price'], features_predictions)\n",
    "features_rmse = features_mse ** (1/2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}