ml_lesson_3.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "datapath = ['../data']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Question\n",
    "* Import the data using Pandas and examine the shape. There are 79 feature columns plus the predictor, the sale price (SalePrice).\n",
    "* There are three different types: integers (int64), floats (float64), and strings (object, categoricals). Examine how many there are of each data type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1379, 80)\n",
      "object     43\n",
      "float64    21\n",
      "int64      16\n",
      "dtype: int64\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1stFlrSF</th>\n",
       "      <td>856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2ndFlrSF</th>\n",
       "      <td>854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3SsnPorch</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alley</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BedroomAbvGr</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BldgType</th>\n",
       "      <td>1Fam</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtCond</th>\n",
       "      <td>TA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtExposure</th>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <td>706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtFinType1</th>\n",
       "      <td>GLQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtFinType2</th>\n",
       "      <td>Unf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtFullBath</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtHalfBath</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtQual</th>\n",
       "      <td>Gd</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtUnfSF</th>\n",
       "      <td>150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CentralAir</th>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Condition1</th>\n",
       "      <td>Norm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Condition2</th>\n",
       "      <td>Norm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Electrical</th>\n",
       "      <td>SBrkr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EnclosedPorch</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExterCond</th>\n",
       "      <td>TA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExterQual</th>\n",
       "      <td>Gd</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exterior1st</th>\n",
       "      <td>VinylSd</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exterior2nd</th>\n",
       "      <td>VinylSd</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fence</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FireplaceQu</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fireplaces</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Foundation</th>\n",
       "      <td>PConc</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FullBath</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LotShape</th>\n",
       "      <td>Reg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LowQualFinSF</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSSubClass</th>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSZoning</th>\n",
       "      <td>RL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MasVnrArea</th>\n",
       "      <td>196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MasVnrType</th>\n",
       "      <td>BrkFace</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MiscFeature</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MiscVal</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MoSold</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood</th>\n",
       "      <td>CollgCr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OpenPorchSF</th>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OverallCond</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OverallQual</th>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PavedDrive</th>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PoolArea</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PoolQC</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RoofMatl</th>\n",
       "      <td>CompShg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RoofStyle</th>\n",
       "      <td>Gable</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SaleCondition</th>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SaleType</th>\n",
       "      <td>WD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ScreenPorch</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Street</th>\n",
       "      <td>Pave</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TotRmsAbvGrd</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <td>856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Utilities</th>\n",
       "      <td>AllPub</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WoodDeckSF</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YearBuilt</th>\n",
       "      <td>2003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <td>2003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YrSold</th>\n",
       "      <td>2008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SalePrice</th>\n",
       "      <td>208500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>80 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     0\n",
       "1stFlrSF           856\n",
       "2ndFlrSF           854\n",
       "3SsnPorch            0\n",
       "Alley             None\n",
       "BedroomAbvGr         3\n",
       "BldgType          1Fam\n",
       "BsmtCond            TA\n",
       "BsmtExposure        No\n",
       "BsmtFinSF1         706\n",
       "BsmtFinSF2           0\n",
       "BsmtFinType1       GLQ\n",
       "BsmtFinType2       Unf\n",
       "BsmtFullBath         1\n",
       "BsmtHalfBath         0\n",
       "BsmtQual            Gd\n",
       "BsmtUnfSF          150\n",
       "CentralAir           Y\n",
       "Condition1        Norm\n",
       "Condition2        Norm\n",
       "Electrical       SBrkr\n",
       "EnclosedPorch        0\n",
       "ExterCond           TA\n",
       "ExterQual           Gd\n",
       "Exterior1st    VinylSd\n",
       "Exterior2nd    VinylSd\n",
       "Fence             None\n",
       "FireplaceQu       None\n",
       "Fireplaces           0\n",
       "Foundation       PConc\n",
       "FullBath             2\n",
       "...                ...\n",
       "LotShape           Reg\n",
       "LowQualFinSF         0\n",
       "MSSubClass          60\n",
       "MSZoning            RL\n",
       "MasVnrArea         196\n",
       "MasVnrType     BrkFace\n",
       "MiscFeature       None\n",
       "MiscVal              0\n",
       "MoSold               2\n",
       "Neighborhood   CollgCr\n",
       "OpenPorchSF         61\n",
       "OverallCond          5\n",
       "OverallQual          7\n",
       "PavedDrive           Y\n",
       "PoolArea             0\n",
       "PoolQC            None\n",
       "RoofMatl       CompShg\n",
       "RoofStyle        Gable\n",
       "SaleCondition   Normal\n",
       "SaleType            WD\n",
       "ScreenPorch          0\n",
       "Street            Pave\n",
       "TotRmsAbvGrd         8\n",
       "TotalBsmtSF        856\n",
       "Utilities       AllPub\n",
       "WoodDeckSF           0\n",
       "YearBuilt         2003\n",
       "YearRemodAdd      2003\n",
       "YrSold            2008\n",
       "SalePrice       208500\n",
       "\n",
       "[80 rows x 1 columns]"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "fileName = ['Ames_Housing_Sales.csv']\n",
    "filePath = os.sep.join(datapath+fileName)\n",
    "df = pd.read_csv(filePath)\n",
    "print(df.shape)\n",
    "print(df.dtypes.value_counts())\n",
    "df.head(1).T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Question 2\n",
    "* As discussed in the lecture, a significant challenge, particularly when dealing with data that have many columns, is ensuring each column gets encoded correctly.\n",
    "\n",
    "* This is particularly true with data columns that are ordered categoricals (ordinals) vs unordered categoricals. Unordered categoricals should be one-hot encoded, however this can significantly increase the number of features and creates features that are highly correlated with each other.\n",
    "\n",
    "* Determine how many total features would be present, relative to what currently exists, if all string (object) features are one-hot encoded. Recall that the total number of one-hot encoded columns is n-1, where n is the number of categories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Neighborhood     25\n",
      "Exterior2nd      16\n",
      "Exterior1st      14\n",
      "SaleType          9\n",
      "Condition1        9\n",
      "HouseStyle        8\n",
      "RoofMatl          8\n",
      "Condition2        8\n",
      "Functional        7\n",
      "BsmtFinType2      7\n",
      "FireplaceQu       6\n",
      "Foundation        6\n",
      "RoofStyle         6\n",
      "SaleCondition     6\n",
      "BsmtFinType1      6\n",
      "Heating           6\n",
      "GarageType        6\n",
      "BsmtExposure      5\n",
      "Electrical        5\n",
      "Fence             5\n",
      "GarageCond        5\n",
      "GarageQual        5\n",
      "BsmtQual          5\n",
      "LotConfig         5\n",
      "MSZoning          5\n",
      "MiscFeature       5\n",
      "BldgType          5\n",
      "HeatingQC         5\n",
      "KitchenQual       4\n",
      "LandContour       4\n",
      "BsmtCond          4\n",
      "LotShape          4\n",
      "MasVnrType        4\n",
      "ExterQual         4\n",
      "ExterCond         4\n",
      "PoolQC            4\n",
      "Alley             3\n",
      "GarageFinish      3\n",
      "LandSlope         3\n",
      "PavedDrive        3\n",
      "CentralAir        2\n",
      "Street            2\n",
      "Utilities         2\n",
      "dtype: int64\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "215"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "mask = df.dtypes == np.object\n",
    "categoricalcolumn = df.columns[mask]\n",
    "categoricalcolumn.shape\n",
    "\n",
    "num_ohc_cols = (df[categoricalcolumn]\n",
    " .apply(lambda x : x.nunique())\n",
    " .sort_values(ascending=False))\n",
    "print(num_ohc_cols)\n",
    "\n",
    "small_ohc_cols = num_ohc_cols.loc[num_ohc_cols>1]\n",
    "\n",
    "small_ohc_cols -= 1\n",
    "#print(small_ohc_cols)\n",
    "#print(num_ohc_cols.index)\n",
    "small_ohc_cols.sum()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Question 3\n",
    "Let's create a new data set where all of the above categorical features will be one-hot encoded. We can fit this data and see how it affects the results.\n",
    "\n",
    "Used the dataframe .copy() method to create a completely separate copy of the dataframe for one-hot encoding\n",
    "On this new dataframe, one-hot encode each of the appropriate columns and add it back to the dataframe. Be sure to drop the original column.\n",
    "For the data that are not one-hot encoded, drop the columns that are string categoricals."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Neighborhood_0', 'Neighborhood_1', 'Neighborhood_2', 'Neighborhood_3', 'Neighborhood_4', 'Neighborhood_5', 'Neighborhood_6', 'Neighborhood_7', 'Neighborhood_8', 'Neighborhood_9', 'Neighborhood_10', 'Neighborhood_11', 'Neighborhood_12', 'Neighborhood_13', 'Neighborhood_14', 'Neighborhood_15', 'Neighborhood_16', 'Neighborhood_17', 'Neighborhood_18', 'Neighborhood_19', 'Neighborhood_20', 'Neighborhood_21', 'Neighborhood_22', 'Neighborhood_23', 'Neighborhood_24']\n",
      "['Exterior2nd_0', 'Exterior2nd_1', 'Exterior2nd_2', 'Exterior2nd_3', 'Exterior2nd_4', 'Exterior2nd_5', 'Exterior2nd_6', 'Exterior2nd_7', 'Exterior2nd_8', 'Exterior2nd_9', 'Exterior2nd_10', 'Exterior2nd_11', 'Exterior2nd_12', 'Exterior2nd_13', 'Exterior2nd_14', 'Exterior2nd_15']\n",
      "['Exterior1st_0', 'Exterior1st_1', 'Exterior1st_2', 'Exterior1st_3', 'Exterior1st_4', 'Exterior1st_5', 'Exterior1st_6', 'Exterior1st_7', 'Exterior1st_8', 'Exterior1st_9', 'Exterior1st_10', 'Exterior1st_11', 'Exterior1st_12', 'Exterior1st_13']\n",
      "['SaleType_0', 'SaleType_1', 'SaleType_2', 'SaleType_3', 'SaleType_4', 'SaleType_5', 'SaleType_6', 'SaleType_7', 'SaleType_8']\n",
      "['Condition1_0', 'Condition1_1', 'Condition1_2', 'Condition1_3', 'Condition1_4', 'Condition1_5', 'Condition1_6', 'Condition1_7', 'Condition1_8']\n",
      "['HouseStyle_0', 'HouseStyle_1', 'HouseStyle_2', 'HouseStyle_3', 'HouseStyle_4', 'HouseStyle_5', 'HouseStyle_6', 'HouseStyle_7']\n",
      "['RoofMatl_0', 'RoofMatl_1', 'RoofMatl_2', 'RoofMatl_3', 'RoofMatl_4', 'RoofMatl_5', 'RoofMatl_6', 'RoofMatl_7']\n",
      "['Condition2_0', 'Condition2_1', 'Condition2_2', 'Condition2_3', 'Condition2_4', 'Condition2_5', 'Condition2_6', 'Condition2_7']\n",
      "['Functional_0', 'Functional_1', 'Functional_2', 'Functional_3', 'Functional_4', 'Functional_5', 'Functional_6']\n",
      "['BsmtFinType2_0', 'BsmtFinType2_1', 'BsmtFinType2_2', 'BsmtFinType2_3', 'BsmtFinType2_4', 'BsmtFinType2_5', 'BsmtFinType2_6']\n",
      "['FireplaceQu_0', 'FireplaceQu_1', 'FireplaceQu_2', 'FireplaceQu_3', 'FireplaceQu_4', 'FireplaceQu_5']\n",
      "['Foundation_0', 'Foundation_1', 'Foundation_2', 'Foundation_3', 'Foundation_4', 'Foundation_5']\n",
      "['RoofStyle_0', 'RoofStyle_1', 'RoofStyle_2', 'RoofStyle_3', 'RoofStyle_4', 'RoofStyle_5']\n",
      "['SaleCondition_0', 'SaleCondition_1', 'SaleCondition_2', 'SaleCondition_3', 'SaleCondition_4', 'SaleCondition_5']\n",
      "['BsmtFinType1_0', 'BsmtFinType1_1', 'BsmtFinType1_2', 'BsmtFinType1_3', 'BsmtFinType1_4', 'BsmtFinType1_5']\n",
      "['Heating_0', 'Heating_1', 'Heating_2', 'Heating_3', 'Heating_4', 'Heating_5']\n",
      "['GarageType_0', 'GarageType_1', 'GarageType_2', 'GarageType_3', 'GarageType_4', 'GarageType_5']\n",
      "['BsmtExposure_0', 'BsmtExposure_1', 'BsmtExposure_2', 'BsmtExposure_3', 'BsmtExposure_4']\n",
      "['Electrical_0', 'Electrical_1', 'Electrical_2', 'Electrical_3', 'Electrical_4']\n",
      "['Fence_0', 'Fence_1', 'Fence_2', 'Fence_3', 'Fence_4']\n",
      "['GarageCond_0', 'GarageCond_1', 'GarageCond_2', 'GarageCond_3', 'GarageCond_4']\n",
      "['GarageQual_0', 'GarageQual_1', 'GarageQual_2', 'GarageQual_3', 'GarageQual_4']\n",
      "['BsmtQual_0', 'BsmtQual_1', 'BsmtQual_2', 'BsmtQual_3', 'BsmtQual_4']\n",
      "['LotConfig_0', 'LotConfig_1', 'LotConfig_2', 'LotConfig_3', 'LotConfig_4']\n",
      "['MSZoning_0', 'MSZoning_1', 'MSZoning_2', 'MSZoning_3', 'MSZoning_4']\n",
      "['MiscFeature_0', 'MiscFeature_1', 'MiscFeature_2', 'MiscFeature_3', 'MiscFeature_4']\n",
      "['BldgType_0', 'BldgType_1', 'BldgType_2', 'BldgType_3', 'BldgType_4']\n",
      "['HeatingQC_0', 'HeatingQC_1', 'HeatingQC_2', 'HeatingQC_3', 'HeatingQC_4']\n",
      "['KitchenQual_0', 'KitchenQual_1', 'KitchenQual_2', 'KitchenQual_3']\n",
      "['LandContour_0', 'LandContour_1', 'LandContour_2', 'LandContour_3']\n",
      "['BsmtCond_0', 'BsmtCond_1', 'BsmtCond_2', 'BsmtCond_3']\n",
      "['LotShape_0', 'LotShape_1', 'LotShape_2', 'LotShape_3']\n",
      "['MasVnrType_0', 'MasVnrType_1', 'MasVnrType_2', 'MasVnrType_3']\n",
      "['ExterQual_0', 'ExterQual_1', 'ExterQual_2', 'ExterQual_3']\n",
      "['ExterCond_0', 'ExterCond_1', 'ExterCond_2', 'ExterCond_3']\n",
      "['PoolQC_0', 'PoolQC_1', 'PoolQC_2', 'PoolQC_3']\n",
      "['Alley_0', 'Alley_1', 'Alley_2']\n",
      "['GarageFinish_0', 'GarageFinish_1', 'GarageFinish_2']\n",
      "['LandSlope_0', 'LandSlope_1', 'LandSlope_2']\n",
      "['PavedDrive_0', 'PavedDrive_1', 'PavedDrive_2']\n",
      "['CentralAir_0', 'CentralAir_1']\n",
      "['Street_0', 'Street_1']\n",
      "['Utilities_0', 'Utilities_1']\n",
      "215\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder,LabelEncoder\n",
    "df_copy = df.copy()\n",
    "# print(df_copy.shape)\n",
    "le = LabelEncoder()\n",
    "ohc = OneHotEncoder()\n",
    "\n",
    "for col in num_ohc_cols.index:\n",
    "    dat = le.fit_transform(df_copy[col]).astype(np.int)   # Transform labeled data into integer\n",
    "    df_copy = df_copy.drop(col,axis=1)                    # drop  the axis \n",
    "    #print(dat)\n",
    "    new_dat = ohc.fit_transform(dat.reshape(-1,1))          #   Reshape into sparse matric\n",
    "    #print(\"dat shape:\",dat.reshape(-1,1))\n",
    "    n_cols = new_dat.shape[1]                               #   column size\n",
    "    \n",
    "    col_names = ['_'.join([col,str(x)]) for x in range(n_cols)]       #  Rename column  \n",
    "    #print(col_names)\n",
    "    new_df = pd.DataFrame(new_dat.toarray(),index=df_copy.index,columns=col_names)   # Convert to  the data Frame\n",
    "\n",
    "    df_copy = pd.concat([df_copy,new_df],axis=1)        # Concatenate the data into each other\n",
    "\n",
    "\n",
    "print(df_copy.shape[1]-df.shape[1])\n",
    "df  =df.drop(num_ohc_cols.index,axis=1)\n",
    "#df_copy.head(1).T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Question 4\n",
    "* Create train and test splits of both data sets. To ensure the data gets split the same way, use the same random_state in each of the two splits.\n",
    "For each data set, fit a basic linear regression model on the training data.\n",
    "* Calculate the mean squared error on both the train and test sets for the respective models. Which model produces smaller error on the test data and why?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/maniram/anaconda/lib/python3.6/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Data_mse_error</th>\n",
       "      <th>OHC_mse_error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Test_mse</th>\n",
       "      <td>1.876566e+09</td>\n",
       "      <td>7.758777e+17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Train_mse</th>\n",
       "      <td>9.679326e+08</td>\n",
       "      <td>3.812915e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Data_mse_error  OHC_mse_error\n",
       "Test_mse     1.876566e+09   7.758777e+17\n",
       "Train_mse    9.679326e+08   3.812915e+08"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error,r2_score\n",
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "label_col = 'SalePrice'\n",
    "\n",
    "# Normal original data \n",
    "feature_col = [col for col in df.columns if col!=label_col]\n",
    "x = df[feature_col]\n",
    "y = df[label_col]\n",
    "x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=.3,random_state=20)\n",
    "\n",
    "\n",
    "\n",
    "# One hot column datasets\n",
    "feature_ohc_col = [col for col in df_copy.columns if col!=label_col]\n",
    "x_ohc = df_copy[feature_ohc_col]\n",
    "y_ohc = df_copy[label_col]\n",
    "x_ohc_train,x_ohc_test,y_ohc_train,y_ohc_test = train_test_split(x_ohc,y_ohc,test_size=.3,random_state=20)\n",
    "\n",
    "errordf = list()\n",
    "r2scoredf = list()\n",
    "scoredf = list()\n",
    "\n",
    "LR = LinearRegression()\n",
    "(x_train.index  == x_ohc_train.index).all\n",
    "\n",
    "lrmodel = LR.fit(x_train,y_train)\n",
    "y_train_predict = lrmodel.predict(x_train)\n",
    "y_test_predict = lrmodel.predict(x_test)\n",
    "\n",
    "errordf.append(pd.Series({'Train_mse':mean_squared_error(y_train,y_train_predict),\n",
    "                 'Test_mse':mean_squared_error(y_test,y_test_predict)},name='Data_mse_error'))\n",
    "r2scoredf.append(pd.Series({'Train_r2score':r2_score(y_train,y_train_predict),\n",
    "                 'Test_r2score':r2_score(y_test,y_test_predict)},name='Data_r2score'))\n",
    "\n",
    "scoredf.append(pd.Series({'Train_score':lrmodel.score(x_train,y_train),\n",
    "                 'Test_score':lrmodel.score(x_test,y_test)},name='Data_score'))\n",
    "\n",
    "\n",
    "lrmodel1 = LR.fit(x_ohc_train,y_ohc_train)\n",
    "y_ohc_train_predict = lrmodel.predict(x_ohc_train)\n",
    "y_ohc_test_predict = lrmodel.predict(x_ohc_test)\n",
    "               \n",
    "errordf.append(pd.Series({'Train_mse':mean_squared_error(y_ohc_train,y_ohc_train_predict),\n",
    "                 'Test_mse':mean_squared_error(y_ohc_test,y_ohc_test_predict)},name='OHC_mse_error'))\n",
    "\n",
    "r2scoredf.append(pd.Series({'Train_r2score':r2_score(y_ohc_train,y_ohc_train_predict),\n",
    "                 'Test_r2score':r2_score(y_ohc_test,y_ohc_test_predict)},name='Data_ohc_r2score'))\n",
    "\n",
    "scoredf.append(pd.Series({'Train_score':lrmodel.score(x_ohc_train,y_ohc_train),\n",
    "                 'Test_score':lrmodel.score(x_ohc_test,y_ohc_test)},name='Data_ohc_score'))\n",
    "\n",
    "errordf = pd.concat(errordf,axis=1)\n",
    "errordf\n",
    "r2score = pd.concat(r2scoredf,axis=1)\n",
    "score = pd.concat(scoredf,axis=1)\n",
    "errordf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### OneHotEncoder MSE error is so high. it is overfitting the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "#    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Question 5\n",
    "For each of the data sets (one-hot encoded and not encoded):\n",
    "\n",
    "* Scale the all the non-hot encoded values using one of the following: StandardScaler, MinMaxScaler, MaxAbsScaler.\n",
    "* Compare the error calculated on the test sets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Mute the setting wtih a copy warnings\n",
    "pd.options.mode.chained_assignment = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ohc_training_maxabs_scalar_scaling           2.256695e+09\n",
      "ohc_training_minmaxscalar_scaling            2.295065e+09\n",
      "ohc_training_standard_scalar_scaling         2.042922e+09\n",
      "original_training_maxabs_scalar_scaling      7.583137e+27\n",
      "original_training_minmaxscalar_scaling       1.217606e+28\n",
      "original_training_standard_scalar_scaling    1.867713e+09\n",
      "____________________________________________________________________________________________________\n",
      "ohc_training_maxabs_scalar_scaling   :   2256694636.35\n",
      "ohc_training_minmaxscalar_scaling   :   2295064842.01\n",
      "ohc_training_standard_scalar_scaling   :   2042922054.11\n",
      "original_training_maxabs_scalar_scaling   :   7.58313682154e+27\n",
      "original_training_minmaxscalar_scaling   :   1.21760567115e+28\n",
      "original_training_standard_scalar_scaling   :   1867712755.89\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler,StandardScaler,MaxAbsScaler\n",
    "\n",
    "training_stes={\n",
    "    'original_training':(x_train,x_test,y_train,y_test),\n",
    "    'ohc_training':(x_ohc_train,x_ohc_test,y_ohc_train,y_ohc_test)\n",
    "}\n",
    "\n",
    "scalars ={\n",
    "    'minmaxscalar':MinMaxScaler(),\n",
    "    'standard_scalar':StandardScaler(),\n",
    "    'maxabs_scalar':MaxAbsScaler()\n",
    "}\n",
    "\n",
    "errors ={}\n",
    "mask = x_train.dtypes == np.float\n",
    "float_columns = x_train.columns[mask]\n",
    "LR = LinearRegression()\n",
    "\n",
    "for train_encoder,(xtrain,xtest,ytrain,ytest) in training_stes.items():\n",
    "    for scalar_encoder,scalar_instance in scalars.items(): \n",
    "        trainingset = xtrain.copy()\n",
    "        testset = xtest.copy()\n",
    "        \n",
    "        trainingset[float_columns] = scalar_instance.fit_transform(trainingset[float_columns])\n",
    "        testset[float_columns] = scalar_instance.fit_transform(testset[float_columns])\n",
    "        \n",
    "        LR.fit(trainingset,ytrain)\n",
    "        predict_ytest =  LR.predict(testset)\n",
    "        \n",
    "        key = train_encoder+'_'+scalar_encoder+'_'+'scaling'\n",
    "        errors[key] = mean_squared_error(predict_ytest,ytest)\n",
    "\n",
    "        \n",
    "errors = pd.Series(errors)        \n",
    "print(errors.to_string())\n",
    "print(\"_\"*100)\n",
    "for key,val in errors.items():\n",
    "    print(key,'  :  ',val,end='\\n')\n",
    "        "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Question 6\n",
    "Plot predictions vs actual for one of the models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0,0.5,'Predicted value'),\n",
       " Text(0.5,0,'Ground truth'),\n",
       " Text(0.5,1,'train - Predict block')]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6UAAAN7CAYAAAC9FTwZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3XmcVfWd5/93YSFUQSkFYtx4CIJr\nRwhu49JKBJ2ggqYjJq2oSNKTSrTjkrgkGtywxUTRNNohiaE1BIxBp2dEpR0jKD0zOsnP0da4EGRT\nQDHBogq1iqWo+v1hc8eSxYqBOoDP5+NRjwd1vuec+7nXR4wvzrn3lrW0tLQEAAAACtCh6AEAAAD4\n9BKlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAKwQ3jrrbcyffr0v/g85513Xi677LIt\nMNHm/fa3v82BBx7Y6ueQQw7Jcccdl+985ztZtmzZVnncAw88ML/61a+SJHfeeWeOO+64Nh87b968\nPPHEE5tc/5d/+ZcceOCBWb169Sb32dKvb3v98wJg6ykvegAA2BKuvPLK7Lbbbjn99NP/ovPceeed\n2WmnnbbQVB9v6tSp2XfffZMkzc3NeeONN3LNNdfkv/yX/5Lp06enrKxsqz32V7/61YwcObLN+3/9\n61/PqaeempNOOmmrzQTAp48rpQDwId26dUtVVVW7PV51dXV69uyZnj175jOf+UyOPPLIfOtb38rc\nuXPzhz/8Yas+dpcuXdK9e/et+hgA8HFEKQDbvfPOOy+/+93vMmPGjBx44IFJku9+97u58MILU1NT\nk4EDB+aWW25J8sEtpn/zN3+TAQMG5NBDD82XvvSlzJ49u9W51t8Ouv4W23/7t3/LGWeckf79+2fY\nsGF58MEHt+rzWX+lduedd86SJUty4IEH5ic/+Un++q//Oscff3yWLl2alpaWTJo0KSeddFJprqlT\np6alpaV0nvnz5+drX/taBg4cmBNPPDEzZsxo9TgfvX23sbExt9xyS0444YQMGDAgZ555Zv73//7f\nSZLBgwdn6dKlufvuuzN48ODNzv/QQw9l8ODB6d+/fy644IK8/vrrm9x33rx5ueiii3L00Udn4MCB\n+bu/+7vMmTOn1T6PPfZYvvSlL6V///4ZNGhQ7rjjjqxbt26Dc7W0tOT666/PkUcemRdeeGGzMwKw\n7RClAGz37rzzzgwcODBDhgzJ//pf/6u0febMmTn00EPz0EMPZeTIkZk5c2bGjBmTs88+O48++mju\nv//+dO/ePVdeeWVWrVq1yfOPGzcuV1xxRf7lX/4lBx98cK699tosXrx4iz+PdevW5ZVXXsnEiRNz\nyCGHpHfv3qW1adOmZdKkSbnrrruy995754477sgvfvGLXHXVVXn00Udz4YUX5q677srtt9+eJHn3\n3XczatSoJMmvfvWr3HrrrZk4ceJmH/873/lOHnvssVx//fV5+OGHc8wxx6SmpiavvfZaHnzwweyx\nxx4599xzPzbK77nnnowbNy7Tpk1Lx44dc8455+T999/fYL+lS5fmb//2b9PU1JRJkyblvvvuS8eO\nHTNy5MjS6/vEE0/k0ksvzZAhQ/LQQw/lxhtvzP33358JEya0OldLS0tuuOGGPPbYY5k8eXIGDBjw\nsa83ANsG7ykFYLvXrVu3dOzYMZ06dUrPnj1L2ysrK3PRRReV3pf59ttvZ+zYsfnSl76UJNlnn30y\nevTofPWrX82yZctaReCHfetb38pf//VfJ0muuuqqTJ8+PS+88EJ69er1F89+5plnluZb/wFB/+k/\n/afceOON6dDh//3d8dlnn126CtzQ0JB77703Y8eOzcknn5wk6dWrV+rq6nLLLbfk7//+7/Poo4+m\nvr4+t956a+kW3VtuuaX03D9q0aJFmTlzZv7pn/6pdCX08ssvT1lZWd57773sv//+2WmnnVJRUfGx\nt/z+4Ac/SP/+/ZMkt912WwYNGpSHHnoo55xzTqv91kfoHXfckcrKyiTJHXfckZNOOim/+MUv8v3v\nfz8///nPM2TIkFx00UVJkj59+uSGG27I8uXLS+dZH6RPPPFEpkyZkn79+n3cyw7ANkSUArDD6tWr\nV6sPCjriiCPSo0ePTJw4MQsXLswbb7xRulV0Y7eDrrfffvuV/rz+/aZr167d6L7XXnttHn744dLv\nw4cPz4033rjJc991112luO3YsWN69OiRTp06bbDfh4N53rx5Wb16da699tpcf/31pe3r1q3L6tWr\ns2TJksydOze9evVqFZB/9Vd/lc6dO290jvWvw+c+97lW27/zne9scvaN2XnnnXPooYeWft91113T\nu3fvjb4/9g9/+EMOOeSQUpAmSefOndO/f//S/n/4wx/yzW9+s9VxQ4cObfX7zJkzs3bt2uy5557Z\nY489/qx5ASieKAVgh/XRAJsxY0auuOKKDB06NP3798/f/M3fpKGhIRdeeOFmz9OxY8cNtn34vZsf\ndskll+RrX/ta6feuXbtu9tx77rln6dN3N+fDobr+sX/4wx/moIMO2ug5NzVjefnG/69/Y8/xkygr\nK9vgE4Obm5s3G/0f1dzcnJ133rnNc/Xo0SMTJkzIN77xjdx88825+eab/7yhASiU95QC8KkxceLE\nDB8+POPHj8+oUaNyzDHHZOnSpVv0MXr06JF999239NOjR48tev7kgyu3HTt2zNKlS1s91gsvvJAf\n/ehHSZKDDz44ixcvzttvv106bv78+Xnvvfc2es71t7y++OKLrbafddZZ+dnPftbm2VavXp158+aV\nfl++fHkWLVqUgw8+eIN9DzzwwLz88stpaGgobVu1alVeeuml0jx9+/bN73//+1bH3XvvvTnjjDNK\nvw8cODD9+/fP97///fzX//pfW31wFQDbPlEKwA6hS5cuWbp06WYjc6+99sq///u/58UXX8zixYsz\nbdq0/OM//mOSZM2aNe016l+sqqoqf/u3f5u77rorDz74YBYvXpx//dd/zQ033JCuXbtm5513zmmn\nnZY99tgjl112WV566aW8+OKLueqqq1q9T/XD9t133wwdOjT/8A//kP/5P/9nXn/99fzwhz/M3Llz\nc+KJJyb54DV+/fXXW4XuR5WVleWyyy7Ls88+m1deeSWXXXZZevTo0Soi1zvnnHPS1NSUb3/723nl\nlVfy6quv5tvf/nbef//90vtPv/GNb+Q3v/lN7r777rz++uuZNWtWJk6cmCFDhmxwvvXfoXrNNdek\nvr7+k7y0ABRAlAKwQzj33HOzdOnSnHrqqZuMpjFjxmTvvffOBRdckDPPPDP/7b/9t9xyyy3p3Lnz\ndvcVIt/73vfy9a9/PT/5yU9yyimn5NZbb815552Xa6+9NskHH/I0efLkdOvWLeedd14uvPDCnHHG\nGZv9kKJ/+Id/yAknnJCrrroqZ5xxRp577rncfffd2X///ZMkX/3qV/N//s//yemnn56mpqaNnqNr\n164ZOXJkLr300px99tmlOTZ2G/Pee++dKVOmZO3atRk5cmRGjhyZpqam/OpXvyrd0jxo0KD84Ac/\nyEMPPZTTTjstN910U0aNGpW///u/3+jjX3fddVmzZs1m38cLwLalrGVTb4oBAACArcyVUgAAAAoj\nSgEAACiMKAUAAKAwvqd0C1n/EfY9e/bMTjvtVPQ4AAAA7WrdunX505/+lM9+9rMbfFf45ojSLeSl\nl17KyJEjix4DAACgUFOnTs0RRxzR5v1F6RbSs2fPJB/8A9hjjz0KngYAAKB9LVu2LCNHjiy1UVuJ\n0i1k/S27e+yxR/bZZ5+CpwEAACjGn/t2Rh90BAAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBh\nRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIUR\npQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaU\nAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCls41bUNWb+otqsqGssehQAANji\nyoseANi4+Ytq8+isuVmybGVamlvSoUOH7L1HVU4bfED69u5e9HgAALBFuFIK26D5i2pzz7TnU1vX\nmMrOHdOlcudUdC5PbV1j7pn2fOYvqi16RAAA2CJEKWyDHp01N+XlG/+fZ3l5h8x4cm47TwQAAFuH\nKIVtzIq6xix5a+Vm91n81krvMQUAYIcgSmEbU1vXmJaWls3u09LckhX1ohQAgO2fKIVtTPduFSkr\nK9vsPmUdylK9a0U7TQQAAFuPKIVtTHW3iuyz5y6b3afXnrukupsoBQBg+ydKYRt02uAD0tTUvNG1\npqbmnHriAe08EQAAbB2iFLZBfXt3z+gvD0yP6oo0rFqb9xvWpGHV2vSorsjoLw/0PaUAAOwwyose\nANi4vr2751ujj86KusasqG9M9a4VbtkFAGCHI0phG1fdTYwCALDjcvsuAAAAhRGlAAAAFEaUAgAA\nUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABA\nYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACF\nEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRG\nlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhR\nCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAUpt2idPr06Rk4cGCrn4MO\nOihjxoxJfX19Lrroohx++OH5/Oc/nwceeKB03Jo1a3L11VfnqKOOyrHHHpuJEyeW1lpaWjJ+/Pgc\nffTROfLII3PTTTdl3bp1pfV77703xx9/fA477LBcfvnlaWhoKK098sgjGTJkSAYOHJiamposX768\nfV4IAAAAStotSk8//fQ8//zzpZ8f//jH2W233XLRRRdlzJgxqayszNNPP50JEybktttuy5w5c5Ik\nd9xxR958883MnDkz9913Xx544IHMmjUrSTJ16tQ89dRTmT59embMmJHnnnsu9913X5LkySefzKRJ\nkzJ58uTMnj079fX1mTBhQpJkzpw5ue6663L77bfnmWeeyW677ZYbbrihvV4KAAAA/kMht+++//77\nueqqq3L99denqqoqTzzxRC6++OJ06tQp/fv3z7Bhw0pXS6dPn56amppUVVWld+/eOffcczNt2rQk\nyUMPPZRRo0Zl9913T8+ePVNTU9NqbcSIEenTp0+qqqpyySWX5MEHH8y6devy8MMPZ8iQIRkwYEA6\nd+6cyy+/PDNnzsw777xTxMsBAADwqVVIlP785z/PAQcckJNOOimvv/56ysvL06tXr9J6nz598tpr\nr6W+vj7Lly9Pv379NlhLkgULFmywNm/evLS0tGx07d13383bb7+9wVp1dXWqqqqyYMGCNs2/YsWK\nLFy4sNXP4sWLP/HrAQAA8GlV3t4P+P7772fKlCm5++67kyQNDQ3p3Llzq306d+6cVatWpbGxMUlS\nUVGxwVqSNDY2tjq2oqIizc3NWbNmzUbX1h/z0bX16+sf7+NMmTIld911V1ufMgAAAJvQ7lH6xBNP\nZK+99srnPve5JB/E4PrIXG/VqlWprKwsheOqVavStWvXVmvJB4G6evXq0nGNjY0pLy9Pp06dNrqW\nJF26dGkVth9eX3/ej3Puuedm2LBhrbYtW7YsF1xwQZuOBwAA4APtHqVPPvlkTjnllNLv++67b5qa\nmvLmm29mr732SpIsXLgw/fr1S7du3dKjR48sXLgwu+22W2mtb9++SZK+fftm4cKFGTBgQGltv/32\nK619+HbchQsXpqqqKrvvvnvpuPVqa2tTX19fOu/Hqa6uTnV1dattHTt2/HNfCgAAgE+9dn9P6Qsv\nvFC6SpokXbt2zZAhQzJ+/Pg0NjbmxRdfzCOPPJLhw4cn+eBTe++8887U1dVl0aJFmTJlSs4444zS\n2qRJk7Js2bIsX748P/3pT1ut/frXv85rr72W9957LxMmTMjw4cPToUOHDBs2LI8//nieffbZrF69\nOrfffntOOOGEDUITAACAratdr5SuW7cuy5YtS8+ePVttHzt2bK677roMGjQolZWVueKKK0pXPy+9\n9NLcfPPNOeWUU1JWVpbzzz+/dKX1nHPOyfLlyzNixIisXbs2w4cPz+jRo5MkgwcPzpIlS1JTU5OV\nK1dm0KBBufLKK5MkBx98cMaOHZtrrrkmf/rTn3LEEUdk3Lhx7fhKAAAAkCRlLS0tLUUPsSNYsmRJ\nhgwZkpkzZ2afffYpehwAAIB29UmbqJCvhAEAAIBElAIAAFAgUQoAAEBhRCkAAACFEaUAAAAURpQC\nAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoA\nAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAA\nAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAA\nFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQ\nGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBh\nRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIUR\npQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaU\nAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEK\nAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkA\nAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhWnXKF22bFlqampy2GGH5YQTTsjkyZOTJPX19bno\nooty+OGH5/Of/3weeOCB0jFr1qzJ1VdfnaOOOirHHntsJk6cWFpraWnJ+PHjc/TRR+fII4/MTTfd\nlHXr1pXW77333hx//PE57LDDcvnll6ehoaG09sgjj2TIkCEZOHBgampqsnz58nZ4BQAAAPiwdovS\nlpaWXHjhhdlvv/3y29/+NpMmTcpdd92V5557LmPGjEllZWWefvrpTJgwIbfddlvmzJmTJLnjjjvy\n5ptvZubMmbnvvvvywAMPZNasWUmSqVOn5qmnnsr06dMzY8aMPPfcc7nvvvuSJE8++WQmTZqUyZMn\nZ/bs2amvr8+ECROSJHPmzMl1112X22+/Pc8880x222233HDDDe31UgAAAPAf2i1KX3jhhfzxj3/M\n5Zdfno4dO2b//ffP/fffn8985jN54okncvHFF6dTp07p379/hg0bVrpaOn369NTU1KSqqiq9e/fO\nueeem2nTpiVJHnrooYwaNSq77757evbsmZqamlZrI0aMSJ8+fVJVVZVLLrkkDz74YNatW5eHH344\nQ4YMyYABA9K5c+dcfvnlmTlzZt555532ejkAAABIO0bpyy+/nP333z+33nprjjvuuHzhC1/ICy+8\nkPr6+pSXl6dXr16lffv06ZPXXnst9fX1Wb58efr167fBWpIsWLBgg7V58+alpaVlo2vvvvtu3n77\n7Q3WqqurU1VVlQULFrTpuaxYsSILFy5s9bN48eJP/NoAAAB8WpW31wPV19fnt7/9bY4++ug8+eST\neemll/J3f/d3+dnPfpbOnTu32rdz585ZtWpVGhsbkyQVFRUbrCVJY2Njq2MrKirS3NycNWvWbHRt\n/TEfXVu/vv7xPs6UKVNy1113/RnPHgAAgI1ptyjdeeeds+uuu6ampiZJcthhh+ULX/hCJkyYUIrM\n9VatWpXKyspSOK5atSpdu3ZttZZ8EKirV68uHdfY2Jjy8vJ06tRpo2tJ0qVLl1Zh++H19ef9OOee\ne26GDRvWatuyZctywQUXtOl4AAAAPtBut+/26dMnjY2NaWpqKm1bt25dDjnkkDQ1NeXNN98sbV+4\ncGH69euXbt26pUePHlm4cGGrtb59+yZJ+vbtu8HafvvtV1r78O24CxcuTFVVVXbfffcNjqutrU19\nfX3pvB+nuro6ffr0afXz4duPAQAAaJt2i9Ljjjsuu+yyS8aPH5+mpqY899xz+c1vfpOhQ4dmyJAh\nGT9+fBobG/Piiy/mkUceyfDhw5Mkp59+eu68887U1dVl0aJFmTJlSs4444zS2qRJk7Js2bIsX748\nP/3pT1ut/frXv85rr72W9957LxMmTMjw4cPToUOHDBs2LI8//nieffbZrF69OrfffntOOOGEVFdX\nt9fLAQAAQNrx9t3OnTvnl7/8ZW688cYce+yx6dq1a77//e/nc5/7XMaOHZvrrrsugwYNSmVlZa64\n4ooMGDAgSXLppZfm5ptvzimnnJKysrKcf/75OeWUU5Ik55xzTpYvX54RI0Zk7dq1GT58eEaPHp0k\nGTx4cJYsWZKampqsXLkygwYNypVXXpkkOfjggzN27Nhcc801+dOf/pQjjjgi48aNa6+XAgAAgP9Q\n1tLS0lL0EDuCJUuWZMiQIZk5c2b22WefoscBAABoV5+0idrt9l0AAAD4KFEKAABAYUQpAAAAhRGl\nAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQC\nAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoA\nAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAA\nAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAA\nFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQ\nGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBh\nRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIUR\npQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaU\nAgAAUBhRCgAAQGFEKQAAAIURpQAAABRGlAIAAFAYUQoAAEBhRCkAAACFEaUAAAAURpQCAABQGFEK\nAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAAAIURpQAAABSmXaP05z//eT772c9m4MCB\npZ9nn3029fX1ueiii3L44Yfn85//fB544IHSMWvWrMnVV1+do446Kscee2wmTpxYWmtpacn48eNz\n9NFH58gjj8xNN92UdevWldbvvffeHH/88TnssMNy+eWXp6GhobT2yCOPZMiQIRk4cGBqamqyfPny\n9nkRAAAAKGnXKH311Vdz2WWX5fnnny/9HHHEERkzZkwqKyvz9NNPZ8KECbntttsyZ86cJMkdd9yR\nN998MzNnzsx9992XBx54ILNmzUqSTJ06NU899VSmT5+eGTNm5Lnnnst9992XJHnyySczadKkTJ48\nObNnz059fX0mTJiQJJkzZ06uu+663H777XnmmWey22675YYbbmjPlwIAAIAUEKUHH3xwq23vv/9+\nnnjiiVx88cXp1KlT+vfvn2HDhpWulk6fPj01NTWpqqpK7969c+6552batGlJkoceeiijRo3K7rvv\nnp49e6ampqbV2ogRI9KnT59UVVXlkksuyYMPPph169bl4YcfzpAhQzJgwIB07tw5l19+eWbOnJl3\n3nmnTc9jxYoVWbhwYaufxYsXb8FXCgAA4NOhvL0eqLGxMYsWLcrkyZNzxRVXZJdddsnXvva1HHLI\nISkvL0+vXr1K+/bp0yePP/546uvrs3z58vTr16/V2tSpU5MkCxYs2GBt3rx5aWlpyYIFC3LyySe3\nWnv33Xfz9ttvZ8GCBRk4cGBprbq6OlVVVVmwYEF69Ojxsc9lypQpueuuu/6i1wMAAIB2jNLly5fn\nsMMOy9lnn50JEybkxRdfzDe+8Y2MHj06nTt3brVv586ds2rVqjQ2NiZJKioqNlhLPgjdDx9bUVGR\n5ubmrFmzZqNr64/56Nr69fWP93HOPffcDBs2rNW2ZcuW5YILLmjT8QAAAHyg3aK0V69emTJlSun3\nI444ImeccUaeffbZUmSut2rVqlRWVpbCcdWqVenatWurteSDQF29enXpuMbGxpSXl6dTp04bXUuS\nLl26tArbD6+vP+/Hqa6uTnV1dattHTt2bNOxAAAA/D/t9p7Sl19+OT/72c9abVu9enX23HPPNDU1\n5c033yxtX7hwYfr165du3bqlR48eWbhwYau1vn37Jkn69u27wdp+++1XWluwYEGrtaqqquy+++4b\nHFdbW5v6+vrSeQEAAGgf7RallZWVueuuu/LYY4+lubk5zzzzTB599NGMHDkyQ4YMyfjx49PY2JgX\nX3wxjzzySIYPH54kOf3003PnnXemrq4uixYtypQpU3LGGWeU1iZNmpRly5Zl+fLl+elPf9pq7de/\n/nVee+21vPfee5kwYUKGDx+eDh06ZNiwYXn88cfz7LPPZvXq1bn99ttzwgknbHD1EwAAgK2r3W7f\n7dOnT370ox/ljjvuyHe/+935cpV4AAAgAElEQVR85jOfybhx4/JXf/VXGTt2bK677roMGjQolZWV\nueKKKzJgwIAkyaWXXpqbb745p5xySsrKynL++efnlFNOSZKcc845Wb58eUaMGJG1a9dm+PDhGT16\ndJJk8ODBWbJkSWpqarJy5coMGjQoV155ZZLk4IMPztixY3PNNdfkT3/6U4444oiMGzeuvV4KAAAA\n/kNZS0tLS9FD7AiWLFmSIUOGZObMmdlnn32KHgcAAKBdfdImatfvKQUAAIAPE6UAAAAURpQCAABQ\nGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAbEdW1DVm/qLarKhrLHqULaK86AEAAAD4ePMX1ebRWXOz\nZNnKtDS3pEOHDtl7j6qcNviA9O3dvejxPjFXSgEAALZx8xfV5p5pz6e2rjGVnTumS+XOqehcntq6\nxtwz7fnMX1Rb9IifmCgFAADYxj06a27Kyzeeb+XlHTLjybntPNGWI0oBAAC2YSvqGrPkrZWb3Wfx\nWyu32/eYilIAAIBtWG1dY1paWja7T0tzS1bUi1IAAAC2sO7dKlJWVrbZfco6lKV614p2mmjLEqUA\nAADbsOpuFdlnz102u0+vPXdJdTdRCgAAwFZw2uAD0tTUvNG1pqbmnHriAe080ZYjSgEAALZxfXt3\nz+gvD0yP6oo0rFqb9xvWpGHV2vSorsjoLw/crr+ntLzoAQAAAPh4fXt3z7dGH50VdY1ZUd+Y6l0r\ntttbdj9MlAIAAGxHqrvtGDG6ntt3AQAAKIwoBQAAoDCiFAAAgMKIUgAAAAojSgEAACiMKAUAAKAw\nohQAAIDCiFIAAAAKI0oBAAAojCgFAACgMKIUAACAwohSAAAACiNKAQAAKIwoBQAAoDCiFAAAgMKI\nUgAAAAojSgEAYBu2oq4x8xfVZkVdY9GjwFZRXvQAAADAhuYvqs2js+ZmybKVaWluSYcOHbL3HlU5\nbfAB6du7e9HjwRbjSikAAGxj5i+qzT3Tnk9tXWMqO3dMl8qdU9G5PLV1jbln2vOZv6i26BFhixGl\nAACwjXl01tyUl2/8P9XLyztkxpNz23ki2HpEKQAAbENW1DVmyVsrN7vP4rdWeo8pOwxRCgAA25Da\nusa0tLRsdp+W5pasqBel7BhEKQAAbEO6d6tIWVnZZvcp61CW6l0r2mki2LpEKQAAbEOqu1Vknz13\n2ew+vfbcJdXdRCk7BlEKAADbmNMGH5CmpuaNrjU1NefUEw9o54lg6xGlAACwjenbu3tGf3lgelRX\npGHV2rzfsCYNq9amR3VFRn95oO8pZYdSXvQAAADAhvr27p5vjT46K+oas6K+MdW7Vrhllx2SKAUA\ngG1YdTcxyo7N7bsAAAAURpQCAABQGFEKAABAYUQpAAAAhRGlAAAAFEaUAgAAUBhRCgAAQGFEKQAA\nAIURpQAAABRGlAIAAFAYUQoAAEBh2hyl77zzTn7yk5/kqquuyjvvvJMZM2Zk7ty5W3M2AAAAdnBt\nitJXXnklQ4cOzVNPPZVHH300DQ0NeeaZZ3LWWWflmWee2dozAgAAsINqU5SOGzcu559/fu6///50\n7NgxSTJ27Nicd955ue2227bqgAAAAOy42hSlL7/8ck4//fQNtn/lK1/J/Pnzt/hQAAAAfDq0KUp3\n3XXXLF26dIPtL730Urp3777FhwIAAODToU1RevbZZ2fMmDGZMWNGkuTVV1/N5MmTc/311+crX/nK\nVh0QAACAHVd5W3b6+te/ni5duuTWW29NY2NjLr744uy222755je/mVGjRm3tGQEAANhBtSlKk2Tk\nyJEZOXJkGhoa0tzcnK5du27NuQAAAPgUaFOUzp49e7PrgwYN2iLDAAAA8OnSpiitqanZ6PZOnTpl\njz32EKUAAAB8Im2K0jlz5rT6fd26dXnjjTdy44035owzztgqgwEAALDja9On737UTjvtlD59+uSq\nq67KhAkTtvRMAAAAfEp8oihdr6GhIStWrNhSswAAAPAp06bbd3/4wx9usO3999/PzJkzc/zxx2/x\noQAAAPh0aFOU/v73v2/1e1lZWTp27JgRI0bkq1/96lYZDACAbcOKusbU1jWme7eKVHer2GbOBewY\n2hSlv/zlL7f2HAAAbGPmL6rNo7PmZsmylWlpbkmHDh2y9x5VOW3wAenbu3th5wJ2LJuM0qlTp7b5\nJCNHjtwiwwAAsG2Yv6g290x7PuXlHVLZuWNpe21dY+6Z9nxGf3lgm2NyS54L2PFsMkonTZrUphOU\nlZWJUgCAHcyjs+amvHzjn4lZXt4hM56cm2+NPrrdzwXseDYZpbNmzWrPOQAA2EasqGvMkrdWprKi\n4yb3WfzWyqyoa/zY94VuyXMBO6Y2vac0SZYuXZoFCxakubk5SdLS0pI1a9bklVdeyaWXXrrVBgQA\noH3V1jWmpaVls/u0NLdkRf3Hh+SWPBewY2pTlE6ZMiU333xzmpubU1ZWVvoXS1lZWQYMGLBVBwQA\noH1171aRsrKyze5T1qEs1bt+fERuyXMBO6aN39z/Ef/8z/+cb37zm3nxxRfTo0ePzJ49O4888kgO\nOuignHzyyVt7RgAA2lF1t4rss+cum92n1567tOnK5pY8F7BjalOU/vGPf8wXv/jF7Lzzzjn44IPz\nwgsvpF+/fvne976XadOmbe0ZAQBoZ6cNPiBNTc0bXWtqas6pJx5QyLmAHU+borRbt2559913kyR9\n+vTJq6++miTZe++98/bbb2+96QAAKETf3t0z+ssD06O6Ig2r1ub9hjVpWLU2Paor/uyvcNmS5wJ2\nPG16T+mJJ56Ya6+9NjfddFOOPvro3HTTTTn22GPz+OOPZ88999zaMwIAUIC+vbvnW6OPzoq6xg8+\niGjXik98m+2WPBewY2lTlH73u9/NuHHjMmfOnHzxi1/M448/nlGjRqWysjLjx4/f2jMCAFCg6m5b\nLiC35LmAHUOborSpqSk33XRT6fdbbrklY8aMSadOnVJe3uZvlQEAAIBW2vSe0uOOOy7f+MY3MmPG\njKxatSpJ0qVLF0EKAADAX6RNUfqLX/wivXr1yg9+8IMcc8wxufzyyzN79uysW7dua88HAADADqxN\nUXr44YfnmmuuyezZs3P33XenW7duue6663Lcccflhhtu2NozAgAAsINqU5R+2BFHHJGhQ4dm6NCh\nWbNmTX7zm99sjbkAAAD4FGjTm0JbWlryu9/9Lv/jf/yP/OY3v0ljY2NOOumk/OM//mOOO+64rT0j\nAAAAO6g2Relxxx2X9957L8cff3yuvvrqDB48OJ06ddraswEAALCDa1OUXnbZZRk6dGiqqqq29jwA\nAAB8irQpSs8666ytPQcAAACfQn/2Bx39pZYvX55jjjkmTz75ZJJkyZIlGTVqVAYOHJgvfOELpe1J\nUl9fn4suuiiHH354Pv/5z+eBBx4ora1ZsyZXX311jjrqqBx77LGZOHFiaa2lpSXjx4/P0UcfnSOP\nPDI33XRTq6+vuffee3P88cfnsMMOy+WXX56GhoZ2eOYAAAB8VLtH6TXXXJO6urrS75dcckn69++f\n3/3ud7n66qvzne98J7W1tUmSMWPGpLKyMk8//XQmTJiQ2267LXPmzEmS3HHHHXnzzTczc+bM3Hff\nfXnggQcya9asJMnUqVPz1FNPZfr06ZkxY0aee+653HfffUmSJ598MpMmTcrkyZMze/bs1NfXZ8KE\nCe38KgAAAJC0c5T+6le/SkVFRfbcc88kyfz58zN37txcdNFF6dixYwYNGpSjjjoq//2///e8//77\neeKJJ3LxxRenU6dO6d+/f4YNG1a6Wjp9+vTU1NSkqqoqvXv3zrnnnptp06YlSR566KGMGjUqu+++\ne3r27JmamppWayNGjEifPn1SVVWVSy65JA8++GCrK6kfZ8WKFVm4cGGrn8WLF2/hVwsAAGDHt8n3\nlM6bN6/NJ+nXr9/H7rNo0aLcc889mTZtWr70pS8lSRYsWJC99947nTt3Lu3Xp0+fvPbaa3n99ddT\nXl6eXr16tVp7/PHHU19fn+XLl7d63D59+mTq1Kml8350bd68eWlpacmCBQty8sknt1p799138/bb\nb2evvfZq0/OdMmVK7rrrrjbtCwAAwKZtMkqHDRuWsrKytLS0pKysrLS9paUlSVpte/XVVzf7IE1N\nTbniiityzTXXpFu3bqXtDQ0NqaioaLVv586ds2rVqjQ0NLSK1Q+vNTY2JkmrY9evJUljY2OrYysq\nKtLc3Jw1a9ZsdG39MW117rnnZtiwYa22LVu2LBdccEGbzwEAAMBmonTmzJmlP//bv/1bfvGLX+R7\n3/teDj300HTs2DEvvfRSbrnllpxzzjkf+yA//vGPc/DBB2fQoEGttldUVJRCcr1Vq1alsrJys2vr\no3LVqlXp2rVrq7Xkg0BdvXp16bjGxsaUl5enU6dOG11Lki5dunzs81ivuro61dXVrbZ17NixzccD\nAADwgU2+p3Tvvfcu/dx9990ZN25cBg0alO7du6eqqirHHHNMbrzxxjbdxjpjxow8+uijOeKII3LE\nEUfkzTffzLe//e0sXLgwS5cuzZo1a0r7Lly4MP369cu+++6bpqamvPnmmxusdevWLT169MjChQtb\nrfXt2zdJ0rdv3w3W9ttvv9LaggULWq1VVVVl9913b8vrBQAAwBbUpg86WrlyZXbaaacNtq9du3aD\nq5kb89hjj+X//t//m2effTbPPvts9tprr9x+++2pqalJv3798qMf/Shr1qzJ7Nmz89vf/jZDhw5N\n165dM2TIkIwfPz6NjY158cUX88gjj2T48OFJktNPPz133nln6urqsmjRokyZMiVnnHFGaW3SpElZ\ntmxZli9fnp/+9Ket1n7961/ntddey3vvvZcJEyZk+PDh6dCh3T+IGAAA4FNvk7fvftjJJ5+cq6++\nOt/97ndz0EEHpaWlJS+88EJ+8IMflGLvk7rzzjtz7bXX5phjjsluu+2W22+/vfTpvGPHjs11112X\nQYMGpbKyMldccUUGDBiQJLn00ktz880355RTTklZWVnOP//8nHLKKUmSc845J8uXL8+IESOydu3a\nDB8+PKNHj06SDB48OEuWLElNTU1WrlyZQYMG5corr/yLngMAAACfTFnL+k8u2ozGxsaMGTMm//qv\n/5rm5uYkSXl5ec4888xcc8013k+ZZMmSJRkyZEhmzpyZffbZp+hxAAAA2tUnbaI2XSmtqKjIbbfd\nluuvv770Xs399tvvz/pwIAAAAPioNr+R8p133smUKVMyderU7LXXXpk9e3bmzp27NWcDAABgB9em\nKH3llVcydOjQPPXUU3nkkUfS0NCQp59+OmeddVaeeeaZrT0jAAAAO6g2Rem4ceNy/vnn5/777y+9\nf/Smm27Keeedl9tuu22rDggAAMCOq01R+vLLL+f000/fYPtXvvKVzJ8/f4sPBQAAwKdDm6J01113\nzdKlSzfY/tJLL6V79+5bfCgAAAA+HdoUpWeffXbGjBmTGTNmJEleffXVTJ48Oddff32+8pWvbNUB\nAQAA2HG16Sthvv71r6dLly659dZb09jYmIsvvji77bZbvvnNb+aCCy7YyiMCAACwo2pTlL755ps5\n++yzM3LkyDQ0NKS5uTldu3ZNU1NTXnzxxfTv339rzwkAAMAOqE237w4ZMiR1dXVJksrKynTt2jVJ\n8sYbb+S8887betMBAACwQ9vkldL77rsvd955Z5KkpaUlp556asrKylrt09DQkP3333/rTggAAMAO\na5NR+uUvfzmVlZVpbm7O1VdfnQsvvDBVVVWl9bKyslRWVuboo49ul0EBALYlK+oaU1vXmO7dKlLd\nraLocQC2W5uM0vLy8nzxi19Mkuyzzz4ZOHBgVq9eXbp1d968eenXr1/7TAkAbNd2pICbv6g2j86a\nmyXLVqaluSUdOnTI3ntU5bTBB6Rvb1+VB/DnatMHHe2+++45/fTTc+KJJ+bKK69Mkpx//vnp2bNn\nfvzjH2fvvffeqkMCANunHS3g5i+qzT3Tnk95eYdUdu5Y2l5b15h7pj2f0V8euF0+L4AitemDjsaO\nHZuDDz44NTU1pW2PP/549ttvv9x4441bbTgAYPu1PuBq6xpT2bljulTunCSZt6g2E3/5/2X+otqC\nJ/zzPTprbsrLN/6fT+XlHTLjybntPBHA9q9NV0qfe+65PPTQQ9l1111L27p27ZqLL744I0aM2GrD\nAQDbrw8HXG1dY15b+E5Wvrc6LS0ffDbFosUr8oOr//N2c2VxRV1jlry1MpUVHTe5z+K3VmZFXeN2\nf4syQHtq05XSXXfdNfPmzdtg+xtvvJGKCv/SBQBaWx9wyQdB+u8vL0vjqqZ0LN8pO3fcKR3LO+St\nP72fH0/+3XZzxbS2rjEtLS2b3aeluSUr6hvbaSKAHUObrpSeddZZ+f73v5+LLroon/3sZ5Mkr7zy\nSn784x/nzDPP3KoDAgDbnw8H3GsL30mHDmUb7NPS0pKmpubMeHJuvjV62/80/+7dKjb4eryPKutQ\nlupd/YU9wJ+jTVF64YUXZt26dfmnf/qn1NZ+8LeZPXr0yKhRo/K1r31tqw4IAGx/1gdc46qmrHx3\ndTp23GmDfcrKytK5c/l2c8trdbeK7LPnLqmt2/SV0F577rLNPw+AbU2borSsrCwXX3xxLr744tTW\n1mbnnXcufTUMAMBHrQ+4eYtqs6kbXnfp2ikVnTvm/YY1WVG/7Udpkpw2+IDSp+9+VFNTc0498YAC\npgLYvm0ySqdOnZoRI0akU6dOmTp16mZPMnLkyC0+GACwfTtt8AGZ+Mv/Lxu74bW5uTn796lOsn3d\n8tq3d/eM/vLAzHhybha/9cHX3JR1KEuvPXfJqSdun19zA1C0TUbppEmTcuqpp6ZTp06ZNGnSJk9Q\nVlYmSgGADfTt3T3fPO/ILFq8Im/96f20tLSkrKwsu3TtlP37VKd7t8ok298tr317d8+3Rh+dFXWN\nH1zh3bViu5ofYFuzySidNWvWRv8MANBWfXt3zw+u/s/58eTfpampOZ07l6ei8//7SpXt+ZbX6m5i\nFGBL2GSUNja2/ePMfS0MALApfXt3z4XnH1W65fX9hjVueQWgZJNROnDgwI/92PP1Xn311S02EACw\n/VlR15jausZ038TVQ7e8ArApm4zSyZMnl/780ksv5Z//+Z9z4YUX5tBDD03Hjh3z+9//PhMnTsyo\nUaPaZVAAYNszf1FtHp01N0uWffChPx06dMjee1TltMEbvwLqllcAPmqTUXrUUUeV/nz99ddn3Lhx\nOf7440vbDjrooOyzzz4ZM2aMMAWAT6H5i2pLX49S+aH3idbWNeaeac9n9JcHujUXgI+14ZdsbcRb\nb72Vz3zmMxts32WXXVJbW7vFhwIAtn2Pzpq70e/rTJLy8g6Z8eTcdp4IgO1Rm6L0mGOOyY033piF\nCxeWts2ZMyfXXnttTjzxxK02HACwbVr4+or8/tU/pnFV0yb3WfzW/8/evYe3Ud/5o3/PaEaakS1b\nknOxHeciggMhXGIoJMulJUkvC6F029J2z25pm9NzDi1sl54t3f2VHh5Kr7tPKf01S0svPwqlQNvA\ncsotPWVJUloupUlJAgRCEsdO7DiOY0uybGtGmtv5Q9ZEsiX5Jt/fr+fp0yeakWYk63H79uf7/XwS\niMVH3ziRiIjmp6LLd3N985vfxD/90z/hmmuugc/nAwCk02lceeWVuOuuuyb1BomIiGjmyO4hfeOd\nU9j3Vie8XglVlV40RmoQHrJX1LGdTFMj7iElIqISRhVKw+EwHn30URw5cgRHjhwBAJxzzjmIRCKT\nenNEREQ0c+TuIQ1Xq/DKHsiSCE03se9AJ9auqc0LpoIoIFTNQEpERKWNavkukKmMvv322zh06BDW\nr1+Prq4udHd3T+a9ERER0QySu4dUVWRUBXzuMVEUcLilJ+/8pXVVrJISEdGIRhVK29ra8Ld/+7e4\n++678ZOf/AR9fX145JFHsHnzZrz11luTfY9EREQ0zWJxDe0nE3mPNUbCsG3b/XeiP+XuMTVNG9du\nWDWl90hERLPTqELpt771LVx55ZXYtWsXvF4vAOCee+7B1Vdfje985zuTeoNEREQ0/aJxDY7j5D0W\nDvqxdk0tVEWGYdpIpy3EejXUhNQZNQ4mFtfQ3Bpl0yUiohlqVHtK//rXv+I3v/kNRPFMhpUkCZ//\n/Ofx4Q9/eNJujoiIiGaGcFCFIAjQdBOabkBVZKiKhHDQj3VNfmi6gVivjq/cfBUiy0PTfbsAzjRl\nau9MwLEdiKKIJbUBbN64asYEZiIiGmUo9Xq96O3tHfZ4W1sbKioqyn5TRERENLNE4xqaj0XRebof\njgMIgpDXdVdVZDTUVc2oQJptyuRXZPfxaFzDA9v2zqhKLhHRfDeq5bvXX389vvGNb2D//v0AgGg0\nih07duDOO+/EddddN6k3SERERNMrG/CW1AbgEYVhXXejcW3G7SHNbco0lCSJ2L7r0BTfERERFTOq\nUPqlL30J69evx4033ghN0/CJT3wCt956KzZt2oR/+Zd/mex7JCIiommUDXjD9pAaFizbQcepxIyq\nPBZqyjRU28kE95gSEc0Qo1q+u3//fnzxi1/ErbfeiuPHj8OyLCxbtgx+v3+y74+IiIimUTbg+dXM\nEtjcPaS6bkJRJDhA3nzS6VaoKdNQju0g1qtxZA0R0QwwqkrpzTffjObmZvh8PjQ2NuLcc89lICUi\nIpoHigU8VZERGtxLmg14M0W2KVMpgiggVM1ASkQ0E4wqlC5duhQtLS2TfS9EREQ0w8zGgBcKqmio\nqyp5ztK6KlZJiYhmiFEt3125ciVuu+023HfffViyZAl8Pl/e8R/84AeTcnNEREQ0vbIBL1pi/+VM\nDHibN65yu+8ONdOaMs1HsbiGaFxDOKjOuO8OEU29UYVSURTxoQ99aLLvhYiIiGag2RjwVq4IY8vH\nm7B91yG0nczMKRVEAUvrqnDtBs4pnS6cHUtEhQjOSJ0AaFTa29uxadMm7NixAw0NDdN9O0RERGXV\n3BqdtQEvFtcyTY2qWZWbTrmzY4cyTXtGdXAmovEZbyYqWim1LAs//elP8dxzz8Hr9WLTpk3YsmUL\nZFku9hQiIiKao1auCOMLW9bPyoAX4hLRGWE0s2O/sGX9FN8VEc0ERRsd3XffffjZz36GCy+8EOef\nfz5++tOf4utf//pU3hsRERHNMKGgirOWhxnyaEw4O5aISilaKf3tb3+L7373u9i0aRMA4L3vfS8+\n97nP4Wtf+xo8Hs+U3SARERERzW6cHUtEpRStlHZ2duL88893/71u3TqYponu7u4puTEiIiIimhtm\n42ghIpo6RUOpaZqQpDOFVFEU4fV6kU6np+TGiIiIiGhu4OxYIiqlaCglIiIiIiqXzRtXwTTtgsdm\n6mghIpoaJeeUPvHEE/D7/e6/LcvCk08+iVAolHfeP/7jP07O3RERERHRnMDZsURUTNFQWl9fj1/9\n6ld5jy1YsABPPPFE3mOCIDCUEhEREdGIZvNoISKaPEVD6c6dO6fyPoiIiIhonuDsWCLKxT2lRERE\nRERENG0YSomIiIiIiGjaMJQSERERERHRtGEoJSIiIiIiomnDUEpERERERETThqGUiIiojGJxDc2t\nUcTi2nTfChER0axQdCQMERERjV5zaxTP7jyE9s4EHNuBKIpYUhvA5o2rsHJFuCzXiMU1ROMawhyn\nQUREcwhDKRERjRnDUb7m1ige2LYXkiTCr8ju49G4hge27cWWjzdNKJhOReAlIiKaLgylREQ0agxH\nhT278xAkqfCOGEkSsX3XIXxhy/pxvfZkB14iIqLpxj2lREQ0KtlwFI1r8CsyKvxeqIrkhqPm1uh0\n3+K0iMU1tJ9MlDyn7WRi3HtMRxN4iYiIZjOGUiIiGhWGo8KicQ2O45Q8x7EdxHrHHkonO/ASERHN\nBAylREQ0Ioaj4sJBFYIglDxHEAWEqse+93YyAy8REdFMwVBKREQjYjgqLhRU0VBXVfKcpXVV42oI\nNZmBl4iIaKZgKCUiIgCl52syHJW2eeMqmKZd8Jhp2rh2w6pxve5kBl4iIqKZgt13iYjmudF01M2G\no2iJ5bnzORytXBHGlo83YfuuQ2g7mfkcBVHA0roqXLthYp2JN29c5XbfHWoigZeIiGimYCglIprH\nxjJuZKaFo5k2K3XlijC+sGU9YnENsV4Noery3NdkBl4iIqKZgKGUiGgeG8t8zZkSjmb6rNTQJITk\nyQq8REREMwFDKRHRPJXtqOtX5aLnZDvqZgPQdIejsVR2RzLTKq2jMRmBl4iIaLoxlBIRzVNj6ag7\nNAhNVzgaS2W3mJleaSUiIppv2H2XiGiemm0ddcsxKzVbaY3GNfgVGRV+L1RFciutza3Rct82ERER\njYChlIhonirXuJFSo2TKqaUthmg8CU03ip4z0qzU0VRaiYiIaGpx+S4R0Tw2kY66U7UMNnudI61R\nvP7WKcheCVWVXjRGwggH/XnnlqrsjmcPLREREU0+VkqJiMZpqiqEkynbUbcmpCKpGxhIppHUDdSE\n1JJNg6ZqGWzudcJBFTVhP2RJhKab2HegE9F4EgCg6SaicQ011f6igXIse2iJiIho6rBSSkQ0RnOt\nUc5IHXULdaktR8Oh0Rh6ncZIDfYd6IQoChBFEfveOoUKVUaiPwXLcnDhahNbf/7ngj+L2baHloiI\naL5gKCUiGoNyjiSZaYZ21C0Wvq+8dFnBZbCabkLTDaiKXJZlsIWW24aDKtauqcXhlh50dQ+go6sP\ny+qDCAdVNEZCCAf9RX8W2T200RKV7dHsoSUiIqLy4vJdIpo3yrHcdr40yim1PPd//eq1vCWu0biG\nV/e248Xdx7Dn9Q68uPs4/rq/A/sOdE7oHoottw0HVaxrakCoWkXdokqsPW8x1jUtydtfWuxnsXnj\nKpimXfB6I+2hnYhi3628GXIAACAASURBVL25sASciIhoolgpJaI5r1zLbedTo5xS4TtQ6cWe/R1Y\nUpupOmaX08qSxz1HT1t4Zsc7WLaketyV41LLbc9UZb0IVisFzyn0s8juod2+6xDaTma+D4IoYGld\nFa7dUP7l18W+execsxhvvHNqziwBJyIimgiGUiKa08q53HYsjXKmMpQW2vM5ES3HYnjj4CmEgypU\nZXgAVxUZEDLB8HBLD0QxPzgapg2PR4AkeSa0t7TUcltNN+AAqK70FbxHoPjPYqQ9tOVS7Lt3pDWK\nbU8fwMUX1OZVd+fCEnAiIqLxYCglojmtnA15prtRztDwWe6GS9nXe/NgF/Yd6IS3xOiVyNIgemJJ\nJPpSkOVMhVTTDfTENOgpEwsXVODF3ceheCVc8a5luPiC+nG952Ija1RFhmM7aIyEij53pJ/F0D20\n5Vbsu3e4pQeSJOJwSwzrmvI/13I2iSIiIpotGEqJaM4q93Lb6WqUUyh8er0iojEdC2v8ZWm4lFvV\nCwVVeGVP3uiVtWvyq3rhkB9XXLIMLcdj0NMWND2N091JKIqE+tqAe09aysD9v34N1QFlXEG51HLb\n9777LHjE4q0RprNpUbHvnqabbpBP9KfcxlC55soScCIiotFioyMimrMmYy7lVDfKKdZwaM/+Drxx\n8FTBgDyehku5VT1VkVAV8LnHRDFT1cu1tK4K7163HJdcWI8rL12KSr8PyxqqsaS2Ki8kC4KAygrv\nhBpAZZfbfuXmq/D5Gy/FV26+Cl/Ysh6fuaFpxJ/FdDUSKvbdyy47BgDHcaDr5rBzOCuViIjmG1ZK\niWjOmozltlPdKKfQElBNN9xq2+GWHqxrahj2vLFU2wpV9XLngQLIq+plA1+2cnyisw+WZbvLeHNV\nDe75LEf1b+hy21I/i/NXLZ7WWbLFvnuqIiP7qCAIUJTh/zPMWalERDTfMJQS0Zw1Wcttp6pRTvEl\noGeqbZmwaEIdEm7G0nCpUFUvdx5ooj+FdNpCrFdHw5DwvXnjKtzzs1dQqB5t27a753OyGkAV+llk\nlzBP5yzZYt+9bBVa0003sA/FWalERDTfcPkuEc1pk7ncNhRUcdby8KQFiGJLQHOrbY4D6Lox7Jyx\nVNuKVfWy80CvvHQ5ms6vc5fN5ga6lSvC+OzfN0FVJBimjbRhwTBtqIqctw91sqt/uT+LmTJLtth3\nrzFSA9O0CzZpmsxZqURERDMVK6VENKdN9XLbciq1BDRbbRMEQJlgtW2kirKqSGioW4TI8sKdbi++\noB4fev+5ONGZgK6bUBRpWAVwqqp/M2mWbLHvXmMkjA9/YDXePHRq1n0niYiIJgNDKRHNeVO13Lbc\nSoXFxkgY+w50Di4Bzf9VPp5qW7HRK6N9vezz1eDwpcZ9/Wlc/75zxnQ/4zXTZsmW+u5tuCIy676T\nREREk4GhlIjmjcmeSzkZioXFcNCP889djIVhP5K6MeFq20QrykOfH40l0dIWhwAB9bWV+PHDe3DW\nshA+tnnNpFYCp3uWbDHFvnuz8TtJRERUbgylREQzWKmweONHLsLKFeGyVdsmWlHOPv+1Nzrw81/v\nxcrlYbR19KL5WByO4+DAodPY+VILvnLLVdhwRWTc91nKdM2SJSIiovGb0kZH27dvxzXXXIOmpiZs\n3rwZzz//PACgt7cXt9xyCy655BJcffXVeOyxx9znpNNp3H777bjssstw+eWX47777nOPOY6D733v\ne1i/fj0uvfRSfPOb34RlWe7xBx98EFdddRUuvvhi3HbbbUgmk+6xZ555Bps2bUJTUxNuuukmdHd3\nT8EnQEQ0dsXmdGYrjuVuuDTR13tx93GIHuDgkdPQdBOyJMIreyBLItKGhX+/709obo2W5V4LmepZ\nskRERDQxUxZKW1pacPvtt+Nb3/oW9u7di69+9av44he/iGg0ijvuuAN+vx8vv/wytm7dirvvvhsH\nDx4EAHz/+99HR0cHduzYgUcffRSPPfYYdu7cCQB45JFH8Ic//AFPPfUUtm/fjtdeew2PPvooAGDX\nrl24//778dBDD+GFF15Ab28vtm7dCgA4ePAg7rzzTtxzzz145ZVXsGDBAtx1111T9VEQEY3LZHf7\nLYdso6HDLVGIYuH/iekfSOPxZ9+atHvIVpdrQiqSuoGBZBpJ3UBNSJ2ScTBEREQ0NlO2fDcSieCl\nl15CRUUFBgYG0NXVhYqKCni9Xjz//PP4/e9/D5/PhwsvvBDXXXcdHnvsMdxxxx146qmncPfddyMQ\nCCAQCOCTn/wktm3bho0bN+LJJ5/Epz/9aSxatAgAcNNNN+Hee+/FjTfeiCeffBI33HADIpHMErFb\nb70Vn/nMZ/DlL38ZTz/9NDZt2oSLLroIAHDbbbfhiiuuQE9PD2pqaqbqIyEimnOicQ2abiDRl4Is\newqe4zhA8/HopHbAna3NrYiIiOajKd1TWlFRgba2Nrz//e+H4zj42te+huPHj0OSJCxdutQ9LxKJ\n4LnnnkNvby+6u7tx9tln5x175JFHAABHjx4dduzIkSNwHAdHjx7F+973vrxjfX19OHXqFI4ePYqm\npib3WCgUQiAQwNGjR0cVSmOxGOLxeN5jnZ2dY/9AiIjmmHBQhZ4yUar/rSAAPq80JR1w2UiIiIho\n5pvyRkd1dXV4/fXXsWfPHtx888347Gc/C0VR8s5RFAW6rkPTMo0qVFUddgwANE3Le66qqrBtG+l0\nuuCx7HOGHssez15vJA8//DDuvffeMbxrIpqPYnEN0biG8BQEo6m8VimhoIqVy8N469DpoudUVfqg\nqtKUd8AlIiKimWnKQ6kkZS75N3/zN3j/+9+PN9980w2ZWbquw+/3u8FR13VUVlbmHQMyATWVSrnP\n0zQNkiTB5/MVPAZkqrW5wTb3ePZ1R/LJT34S1113Xd5jnZ2d+MxnPjOq5xPR3NbcGsWzOw+hvTPT\nLVcURSypDWDzxrGPapnotVqOxdDSHkOkIYTI8tCw509GmP3Y5jXY+VIL0oY17JhtO2iM1LADLhER\nEbmmLJS+8MILeOCBB/Dggw+6jxmGgWXLluGPf/wjOjo6UF9fDyDTFOnss89GMBhETU0NWlpasGDB\nAvfYypUrAQArV65ES0uLuze0paUFZ511lnvs6NGj7rVaWloQCASwaNEi93lZ0WgUvb297uuOJBQK\nIRTK/z93siwXOZuI5pPm1qg7V9SvnPm9EI1reGDb3rI22il1ra/dswtdPQOI9epwHAeiKKChrgr/\n9Ol12HBFZFKD88oVYXzllqvw7/f9Cf0DaThOZsluVaUPjZEaVFX62AGXiIiIXFPWffe8887Dm2++\nid/+9rewbRsvvPACXnjhBXziE5/Apk2b8L3vfQ+apuH111/HM888gw9+8IMAgOuvvx7/+Z//iXg8\njtbWVjz88MP40Ic+5B67//770dnZie7ubvzkJz/JO/ab3/wGhw8fRn9/P7Zu3YoPfvCDEEUR1113\nHZ577jns2bMHqVQK99xzD9797ncPC5pERGP17M5DkKTCv1olScT2XYcm/Votx+N44c/HcPR4DH5V\nRoXfC1WR0RPTcMfdO/HIE/vxwLa9iMY1+JXscckNzuUY17Lhigh+9M3rcP17z8X55yzCxefX44LV\ni9EYCbMDLhEREeWZskrpwoUL8eMf/xjf/va38fWvfx0rVqzAD3/4Q6xcuRLf+MY3cOedd+I973kP\n/H4/vvzlL7vVzy9+8Yv49re/jWuuuQaCIOBTn/oUrrnmGgDAP/zDP6C7uxs33HADDMPABz/4QWzZ\nsgUAsHHjRrS3t+Omm25CIpHAe97zHvzrv/4rAGD16tX4xje+ga9+9as4ffo03vWud+E73/nOVH0U\nRDRHZceh+NXiKyfaTibK0nW21LX+sr8doigglTJhmBZk6UwXXEkS8d0fv4S/+9vVBV83G5y/sGX9\nhO4PyFRM/+2WK8vSAXem7JklIiKi8hMcxynVJJFGqb29HZs2bcKOHTvQ0NAw3bdDRNOguTWKHz+8\nGxV+b9FzBpJpfP7GS3HW8olVCotdKxbX8fj2A5AlDyzLRn1tAGrO0l7DtNB8LIZPXHc+QkFl6MsC\nAJK6ga/cfNWMCH9TuT+XiIiIJma8mWjKlu8SEc114aAKQRBKniOIQlm6zha7VjyhIfunRkHAsOW9\nhmHDcYDePn3Yc7Mc20Gsd3TdyEuJxTU0t2bmkY5Hds/sZC4zJiIiouk35d13iYjmqlBQRUNdFaIl\nQli5us4Wu1awSkU2q/p8Ut7SXQCQZRGCAFQHCldJgYkH53JVN0ezP7ccy4yJiIhoerFSSkRURps3\nroJp2gWPmaZd1q6zha4VCiqoCvhgO07B8CtLHtQuqCi6dBeYWHAuV3Uzu2e2lOz+XCIiIprdGEqJ\niMpo5YpMd9makIqkbmAgmUZSN1ATUsvedbbYtd535UosDPvzxsRkmaaNL3/uikkLzuXqPhyNaxip\n5UG5lhkTERHR9OLyXSKiMlu5IowvbFlflq6z473Wrpda8MOHXsXxEwk4jgNBELBsSRVu+dSZOaXb\ndx1C28nMEltBFLC0rgrXbhh/A6Fydh+eyv25RERENL0YSomIJkloksaXFBqPMvRaG66IYMMVEbQc\ni+HYiTiWLwkisvzMLObJCM5jqW6OdK2p3J9LRERE04uhlIholhhPA6HI8lBeGB2qnMG53NXNzRtX\n4YFtewsuBy73/lwiIiKaPtxTSkQ0C8yG8SjZ6mYpY6luTuX+XCIiIpo+rJQSEc0Cs2U8Srmrm1O5\nP5eIiIimByulREQz3GwajzJZ1c1QUMVZy8MMpERERHMQK6VERDNcORsITQVWN4mIiGgsGEqJiGa4\n2ToeZbK6DxMREdHcwuW7REQzXLkbCBERERHNJAylRESzwOaNq2CadsFj82E8Siyuobk1OiP2zRIR\nEVF5cfkuEdEskG0gtH3XIbSdzMwpFUQBS+uqcMW7lgHIBLe5Vi0dz2xWIiIiml0YSoloXovFNUTj\nGsJTvP9xPNcd2kAontDx4u7j+PXTb445sE3l+x7vtbKzWSVJhF+R3cezs1k5q5SIiGhuYCglonlp\nuipw5bhuKKgiGtfwxO/eHnNgm8r3PdFrzZbZrERERDQx3FNKRPNOtgIXjWvwKzIq/F6oiuQGuubW\n6Iy/7mgC22RefyQTvdZsms1KREREE8NQSkTzzngC3Uy67ngD21S+74leayyzWYmIiGh2Yyglonll\nuipwo71uy7HYiF1mxxPYpvJ9l+Nas3U2KxEREY0d95QS0bwylkA3ngZAxZr6ZK+r6QY03YCqyFBz\n9oJ2nErg9be7cOp0PxbVVCCVthCsUvB3HzgXF19Qn3eNYoFN0033tYcGtsl+37nKca3sbNZoieDK\n2axERERzA0MpEc0rk1WBK9XUJxxU8eahLry6tx2WDTiOA0EQUFXpxaIaP1ra4mg5HoNlO9B1EynD\nhKrIUHwSnv/TUWy6KoLP3HCmcdHQwBaNazjc0oNEfwqOAwiCgLqFFYjmjIiZyspjua61eeMqt/vu\nUPNhNisREdF8wVBKRHPGaEaPTEYFrtjokiOtPfj87c8gHFTR2h5HV3c/RFFEOKTCr8joiWs48E4X\nBEGAx+OB6MlUDwVBwEDSwIKwH7Ik4pW/tsOxkddRNxvYEv0p7DvQCVEUIEseAIBt26ivDeR14Z3K\nymO5rlVqNuu1GzinlIiIaK5gKCWiWW+so0fKXYEr1NQnGk8OhkURr7/diUULKrGwpgKdXf3o7OpH\n7aJKRGMaHACJ/hSqK32AkKlyGqYFTTeR6E8hUOGFfTIBOA48HgF3fWkjgDOB7d++/Rws24Fp2YPV\nVx8aIyGEg34AyBubMpWVx3Jda+hs1lD11M6TJSIiosnHRkdENCvE4lrBBkDjGT2SDXQ1IRVJ3cBA\nMo2kbqAmpBad71nqvgo19TncEoUoijBMG719KRimBVWRUbuoEl7Zg87T/UjqBkzThmM7CFWrsG0H\nhmmhfyAN23ZgWw4AAaIgoG8gjef+eBSvvdHhXiMcVHHW8jCuvHQpLr2wHldeuhTrmpa4gRTIbyhU\nzvc9knJfKzT4XhlIiYiI5h5WSoloRhupCjqa0SPZSmGuclXgCjX10XQDib4UZNkD07TgOJnqoCx5\noCoyGupkd2mrqshoP9kLB87gc828/Zi2ZcPjESBJIhwHePK5t93GR9lr+1VvXtOkXEMbCk1l5ZFV\nTiIiIhoNhlIimrGK7dXMVkE/cs1qtJ9MwK8WDmTAmUphqTA0QqPYkgo19dF0Aw4Aw7RgmDYcxxkW\nnBWfBMuyoSoSvF4PFJ8M207CtGyIOa8nekT4fB7IkgeGaSOWSLnvZyINhUIl9t2W21Rei4iIiGYf\nhlIimrFGqoI++dzBcY8eGes+1GKyTX1OdCbccSx6ysTJrj5Ypg3HAdKGhVOnB9wGRwAgyx5UVXph\nWg6W1VfDcTDsvUoeEYIA996rKn3wyR73/XBsChEREc0F3FNKRDNSsb2aeef06kilrZLnFKoUjmcf\navaehu5rbW6NojuaxB9fPY49r5/E8386ih0vtsCyMkFXFAU01FXDMCx0dmX2kQKZgHneqoUwTRsX\nnVeLtWtq0VBbBcd2MvtJbRuBSi9qF1XCr8iwbRuNkdCw97N54yqYpl3wfjk2hYiIiGYDVkqJaEYq\ntFdzKJ/XA1Up/WusUKVwrPtQi1VVLzhnMXa8dBSSJOLSi+pxuKUHB7v6YJo2DMOG7BVRvygwOHPU\ng56YhtM9A1i8oBKNZ4XRGKnBhz9wHt48dAptJxO4/F1LYZgWemI6Kiu8qPDLEAQBqiK7HXVrQvlL\nYTk2hYiIiGY7hlIimpGG7pfUdMNdHptt6iOIAv7uA6vxxO/edkOmppvuebIkDqsUZiuwo92HWmpf\n63d++Cece/YChAf3d164uhbRmAYIgCR54PEIqFBlJPpT8HhELF5YAckj4rb/63JcedlyN1xuuCLi\nNgP63z50AZ743dswTAu6bkJRJPf9Fqt8sqEQERERzWYMpUQ0I2X3Sx5p7cHhligS/Wk4jjM4i9OL\nxkim0njxBfWoDih46L/24cW/tCHWm2kWFAqqePe65cNedzQV2Nx9qMWqqtrgmJPDLT1Y19TgPuaR\nRHhlDwDAMG1cuHoxAAG6bkAZXIa75pxFw0JjbjOg6oDiVj5t20FSN0ZV+WRDISIiIpqNGEqJaMa6\n4JzF2Pb0AUiSCDknGGq6idfe6MSHP3Ce+9jpaBKCkAmtik+CaTrYs78DPTENX/o/L3fD3Fg61paq\nqmY77Cb6U9B0E+pgRTP3lR3Hga6bCAVVd5lxUjcKdsPNxconERERzSdsdEREM9Yb75zCxRfUQVUk\nGKaFtGHBMC2oioSLL6jDm4dOAQAefHwv3jx4CqblIFDpg1/1QpZEaLqJNw+ewi/+a6/7mtkKbCnZ\nfailqqrZAOo4gD7YvEhVJFQFfO45giBAGbLndSzdcENBFWctDzOQEhER0ZzGSikRzUjZKmU4qGJd\nUwM03XSXwGarjm0nE2g5FsPLu9sgimf+xmaYFkzTHqywevDSnra8WaWbN65y94kOlbtvs1RVVVVk\nVAV8SPSnoOTsNW2M1GDfgU6IooCqSp+7H3Toa0+mWFxDNK4hzOW8RERENAswlBLRjDS0SplZHpv/\nK8uxHbzyWhtORwcQqPTBNG30xLTBMTEOAAGyJMAre/D6W514z+URAKPvWDvSHNDGSBjvHO3Ju69w\nUMXaNbV452g3GuoDGEimp6wbbrlmrxIRERFNJYZSIpqRRtr7GY0nceDQabR3JNDVncTpqAZdN+BX\nZciSB4ZpQ9MNWJYN07Kx9YFXsf/gKTegjXbfZqmqalWlgv/x+avckS7ZcNsYCePW/309wkF1yvaE\nluoS/MC2vdjy8SYGUyIiIpqRGEqJaEYqVaWMxpPYd6ATFX4vljUEoSoSevtScBygfyANxeeBnrIg\nCAIEQYAsedDbpxcMaCN1rB1NVTV3pMvQADpVy2fHOnuViIiIaKZgKCWiGatYlfJwSxSAgMZIDVRF\nQt3iAHpiGkQxE0L7BtLwyplfb47jIFitQEuZbpfcsQa00VRVp3Mcy1hnrxIRERHNJOy+S0QzVrZK\nWRNSkRycCxrt1ZBOW1i7phbhwYDVuKIGfr8MURRg2TZM04Zl2xBFARUVXixeWJnXJTcb0MZqqrvh\nxuIamlujI97rWGavEhEREc00rJQS0Yw2tEoZ79Xxq6feQIXf655Tu6gS9YsDSPSl0DeQhmXZ8CsS\nqgIKQkEVfkWGYVpul9xsQJvuqmGxLrljbVg0ltmrRERERDMNQykRzQrZ5bGxuDYsgKmKhMULKlBV\n6YOmGzjW5mD50mDeOJbMeJbMr7zpDmilQieAMTcsGqlLMDC2+ahEREREU4nLd4loVskGsKEaIzWw\nbQeqIiMUUvMCqW07aIzUuP+ezoCW7ZIbjWvwKzIq/F6oiuSGzgcfL9zpFzjTsKiQzRtXwTTtgsem\naj4qERER0XgwlBJRWY12H+REFApg2fmgPp8HF65eDD1lwjAtqIqUt/90ugNaqS65hmnj5T1tJZ9f\nbD9sof23Sd1ATUjlOBgiIiKa0bh8l4jKYqz7ICei2JiW3Pmg+9/qxKt7T6CnNwnHdpDUjbwxLlmF\n9nUW2+s5USN1ydV0A339KWi6kVfpzVVqP+xoZ6/S7DJZ30ciIqKZgqGUiCYsuyR1LPsgJ2qkAHb1\n5RFcfXnx+aGFQrTP64EDB6m0NSnBeqQuuaoiAw6g62bRUDqa/bDTOZ6Gymcq/9BDREQ0nbh8l4gm\nrNSS1FL7IMthpDEtucdjcQ2vvdGBJ39/EPf9cnfevk5NN/CHV1rxwiut0HVj2F7P5tbohO91pC65\nqiIhUKlAUYr/vZANi+aHkfYel+P7SERENFOwUkpEEzLSklTgzD7I6Wwu9ODje/Hynjb09afQeXoA\nHo+IZfVVuOi8zH7Twy09EEUBgIDDLTGsa/K7z88G6y9sWT+h+xhNl9wrL1sKyypcTZ3u/bA0dUbz\nh56Jfh+JiIhmClZKiWhCRlqSCpzZBzkdmluj+N7PXsYLr7TCshzIsgeO48CxHbS2xfHKX9tx8lQf\nEn0p9zmJwX2duYo1GBqrkbrkfuqja9mwaJ7L/qGnlHJ9H4mIiGYCVkqJaEJGWpIKTO9c0Gd3HkLL\n8RhEMfM3OMOwgcEMLQgCYr0aDjZ3IzdWO44zbF9nqQZDY1GoSVMqbSFUreAj16x2QycbFs1fY/lD\nD78XREQ0F7BSSkQTUmxuaK5y7IMc66iZ7P7Rv77egVPdAzAGq5OyLAI5GTqVNtE/YMCyzlQvBUEY\ntq+znME626Tp7z94PlRFAgQgqRn4zdMHsPXnf3b3C460X5bmppn+hx4iIqJyY6WUiCZs88ZVbvfd\noSa6D3KsHUiz5795sAtvHDyFw609SKdtVPgzzWJqQip8XilvCa0DB4rvTFW0qtI3rPttuRsMNbdG\n8cTv3oYkiQjnhIvJ7FhMs8No9h6z4RUREc0lrJQS0YRll6SWex9koQ6kgIMjrT340UN/GdaBNHv+\nkdYoDjZ3Q/QI8Hkzf3tLagY03UBnVz8q/DLsnOWRsiTi3JULYNsObNtGYySU97pjCdajreiOt2Px\nWCvGNDuNtPeYDa+IiGguYaWUiEYlFtcQjWsIF5mBOdLc0PHIDW7ReBKHW6JI9KfhOA4EQcCx9ufw\nH7e/3w292fOznXRF0QNVkaFpBmwH0HQTVZU+DCQN1C6qRCyuwTEciKIAn0/ChstXwIEDPWVhIJmG\nIApYWleFazfkV2ULfRZjqeiOp2MxZ1bOL4X2Hhf7PhIREc12DKVEVNJYw1CoSGgdq9zgFo0nse9A\nJ0RRhJxTXTx5uh/3/XI3Pn/jpQgHVbSfTEAQBCT6UpBlDwAgHFIxoKWRTBowLQeWbcNK27BMG7bt\noDqgYGldFQRBQE3Ij80bVyEcVAsG62KfxQXnLMaOl45CkkT4c5b9FluK23I8hlivBkEQMntKC8ht\nZJOtAI/29WlumIw/9BAREc1EXL5LREUVWj6rKpIbhoYuny2n3A6kh1uibvfcXI4DGIaFx599C3sP\nnISmZ5bo5vYt9SsyltZXI1ilQACQTlvQdRNd3f1YUleFq/9mOc5btQihasV9X9G4NqzBUKnP4js/\n+hMS/XrB95G7FLe5NYqtP/8zHvyvfdj/Vide3H0cr+5tL7h3MLeRzXiX+tLcwIZXREQ01zGUElFR\n0xmGsh1INd3ImyGaS0+ZOHDoNP7rd2/hV0++ib/sO4G3Dp2GnjLzzvMrMiLLQjj37IW4ZsPZWLEs\niI9cuwYbL48gHPSP+L5icQ2/eHyf28E3l6abGBhI43BLrOh7aTuZwGtvdLihNlytoibkhyyJ0HQT\n+w50Dgum2UY2nFlJREREcx1DKREVNN1hKNuBdGjlM0vTDfT26TAte3BZq4RwUIVp2ehN6EjqxrDn\nhIMqFi+sRFWFD6Fqpei1s+8rW9m88/u78P/94XDBymb2/hL9KWgFrglkluI++dzBvIDfGAnDtjMh\nVxQFHG7pcY/lNrIZy8xKIiIiotmIoZSICpoJYWjzxlWQJQ8KTWzsjiaxIJSpcmbnijZGamDbDhaE\n/eiJ5t9XtqtuX38aZy0PFXjFMxzbwf63Ot3KJhwHkuQpWNlUFRkCAMdxoOtmwddLGRZi8fzlveGg\nH2vX1EJVZBimjZ5YEtG4NqxjMWdWEhER0VzHRkdEVNB0hqGWYzG0tMcQaQjh5k9dhmPtcZw83Q/H\nAQQhEwSrA4o7SzQ7V1RVZKxdU4vDLT3QUxZs24aeshCo8KLxrAVojNTginctw6+fenPE9/Xnve1u\nZTMbPLOylc11TQ1QFQlVAR8S/WkoRZoWhap8SGrDA2s46Me6Jj803UA0rmHLx9ai6YL6/OdyZiUR\nERHNcQylRFTQBgn8nwAAIABJREFUdIShXS+14N5fvIq2jgQcJzOqpaGuCh/529XYe6AThmFBUWRo\nuoE9r3cAwLC5ouGginVNDeiJafjoNauxfEkQDpy8zqUv7j5e8n3VBFX0xDR3ZIuqyKgK+KDlVEIz\ny3VNqIMV2neOdLshOZdp2vjINatLBmFVkREKAiuWFq7gbt64yu2+W+j1ObOSiIiIZjMu3yWiojZv\nXAWzQHMfoPxhaNdLLbjj7p1uGMx0t5XRE9Nw3y/3oGlNLc6OhOHAgeNkrp+tjKqKF9G4lhcaVVXC\nRefVIrI8NKxz6Ujv69KLlgxWL5PuPtHcPaBApvOvPnisqtKH/3HLVagJqUjqBgaSaSR1w12Ke/EF\n9Wioqyr5/ksF/OzMymKvz3EwRERENJuxUkpERWXD0PZdh9B2MjObUxAFLK2rwrUbCs8pHa97f/Fq\nyU6//+/v38bjP/l7d2bjtmcOoONUH9461IVorwbJ44Ese1BV6UVjpAaNkeIjNEq9r/NXLcar+9rx\n6t4TkCQRgiAMvmZ4cGlwDIn+FEzTggOgJqS6n8WGKyJFZ0pOtNrJmZVEREQ0VzGUElFJUxGGWo7F\n0NaRgCyLMM1MN11Z8uSdc/xEAvveOIlAwAdRECBLIp587iAc23H3mnq9EgxTRd8bJ/HhD6we8/tq\naYvh/l/vRWWFFzUh1a28ZpsbrV1Ti3VNmSqq4vPgi5+9fNhnEQoW/nzKFfCLvT4RERHRbMVQSkSj\nMplh6OU9bejs6oOT6WMLQIDP60FNSIU6uIe0s6sP/8/dO5HUDXRHk2g/2QvJI8J2AFkSIUkiUmkL\nPVENV17agDcPncKGKyKjel/RuIZfPrEfTz53EJpuQhAEeDwCNM1EhT+zT1QURRxuiWFdkx+y5MGn\nPto05s+D1U4iIiKi4RhKiWhaNbdG8creNli2A1kSgcE+t6Zpo7OrH8FqBfFeHam0ja7uAfj9MvoG\n0jAtB5ZlobLCC0EQEA6qqKzwQpY86OrR3FmjI4W+5tYoHti2F4ZpQddNeOXcCq0Dx3FgWjYcB0hq\nBlRFwg3XrpnQ0mVWO4mIiIjOYCglomFicQ3RuIbwFISnZ3cewoKwP9PddsjYFEEQcKKzD5V+LyA4\n8PvlwfBouMcHNAMVioxEX8odT5PoT0HTjEw1coT7f3bnIUiSiES/jqFTWTPNliRcuLoWum7AdoCP\nbV6Ds5azsRARERFRuTCUEpGruTWKZ3ceQntnZs+jKIpYUhvA5o3lbWqUFYtraD+ZgF+VcdlFS7Dr\n5VaI4pmJoJZtI5U2IUseLAj5AQCGYUMQBFiWjbTtwLYzlUxhQAAEYGFNBSRPZinvSDNUc68/dBZp\nVqI/BSBT3UzqxqTMZSUiIiKazzgShmiOisU1NLdGESsxjzNXdhlrNK7Br8hulTAa1/DAtr1obo2W\n/R6jcQ2Ok6lPRpaFsOHyFVBVCSnDhKYb0FMmJI+IC85dgOoqBQAgyyIcx4ZlOYCTWeybCZMOkpqB\njs4+6CkTK5eHRqyS5l4/O4t0qNzRL+Wey0pERERErJQSzTnjrXZml7EWIkkitu86hC9sWV/Wew0H\nVQjCmfpkdZWCFQ1BdHUPQE+ZkGUPonENdYsDONISAwDIkgeG6UCWRaTTFhwnUz21B8OlIJhIGxYu\nu6hhzNdvjISx70AnRPHM5yAIgKLIZZ/LSkREREQZrJQSzSHjrXZml7GWkm0cNBYtx2LY+dJRtByL\nFTweCqpoqKsCkKla7jvQCU03Eaj0YWFNBYJVCir8MlqOx+HxZMKjYVqQJBGiIECSPHCcTDMiURAy\nXXNFEQvDfjyz450Rq7u51weAcNCPtWtqoSoyDNNG2rCg+mQ01AWw5eNNk7KEmYiIiGi+Y6WUaA4Z\nb7UzdxlrMY7tINaruecPbYKU2xxp34FO3PuLV9HWkcgERlFAQ10V/unT64aNadm8cRUe2LYXh1t6\n8vaTAoBt27jsoiVoaYu7/zYMG7JHhKfCi2ivBp9PggBAVSRUVnixsKYCHk9m3uloqrvZ62c/t3DQ\nj3VNfmi6gf6BND779xfj4gvqS74GEREREY0fQynRHJHbtKeYYmNShi5jLaQ3kcK2Zw4g2qvBsR2k\n0haCVQouubAex9rj7nLh9o4EXt3XjoU1Fe69GKaF9pMJ/I//+G/8+7+9Ly+YrlwRxkeuWY1XXmuD\nYdpwHAeCIKCq0ofGSAjhoB/VVQreOnQaS+oCaDkeg2HZ8MkeVFX6UKF6Eaj0uuNgMtezoSjSqMbC\nrFwRxpaPN2H7rkNoO5l5D4IoYGldFa7dMDkNnoiIiIjoDIZSojliLNXOoSEtu4w1WmR5bjSu4Vh7\nHPW1Aei6icMtPUj0p5DUTfz8N3tx1vIQ1p5Xm6mSvt0J24Y7Y3QgaSCVtgA4AAT833f9Dv/107/P\nC3vVAQUXn18HURSg6yYURYKqnAnX4aCKpvNr8Q8fuhDVVT784vF9aG3rxdtHThcM4VWVPqiKjIFk\nuuhYmNzK7soVYXxhy3rE4lrm/GrOESUiIiKaKgylRHPEaKqdgigUHWkydBlrroPN3Tj37Bp336co\nCpAlD/r6BiB5RHR09gEOcNayIBJ9KciSB6Zl4/iJXlQHFHjEMz1yT/UM4M7v78S//B+Xu8tis/eu\nKnJeGB167yuWBhEKqvjUR9fivl/uxpHW4fdq2zYaI6Gi73ekRlAMo0RERERTi42OiOaIoU17Cik1\n0iS7jLUmlJnHOZBMI6kbUH0yli+pRjjoz9v3aZgWUmkTAJBKW7BsGwcOnXartZpuwrYdWLbtnp/o\nT6GvP4W9b5zEv377v7H1539Gc2sUoaCKcFBFNJ6ENjh+pdS9r1wRxudvvBR1iyrdhkSGaUNVZKxd\nU4tw0F/w/U7H2BsiIiIiKo2VUqI5pFS1czQjTQotY43GNfzP+19Bx6k+RGMa1Ow+UcPOrMgFADgw\nTTsTQi0HKceEYVoQIMC2Hdi2hf6BNITBDrmVFQo03cCJzj7c87NXEA4pONU1gNcPnoLHI6Kq0ovG\nSNgNl4XufeWKMP7j9vfjRw/9BaZpAwLgOI5baS30nOkYe0NEREREpTGUEs0h5WraExrsrNvcGsVj\nzx7Aq3tPwLQsdJ4egF+REQ6pkGUxuyIXgADTtBGNa0gZFszB6qUoZEJpKm1BEATYjgPZI0KSBDgO\ncOp0P44ej6HC78W6piVQFMndr/qXfR24YPUiXHju4qL3vnJFGNdtOgc//MWrOJ7T6XdpXRVu+fS6\nvOdMpBEUEREREU0ehlKiOaZcTXuyS10lSURNSEWiPw2PmIRp2ujs6kftokpIkoj0YODsjiZhOw4a\nFgdwrKMX2V2kA0kDppXpqmtZDqoDXhw/kYAkZRKtV/Yg0Z+CphsIB1Wsa2qAppvQdQP1i6tKVi6b\nW6PY8dJRnH/uYqxcUYP44MiaYLWKHS8dxbIl1W4wLcfYGyIiIiIqP4ZSojkqNMFAlbvUtTFSg30H\nOuHzSjBNG5Zl4+ixGGRJQP+AAUEAJMmDQIUXC2oq4JFEHGmNQgBgmCZSaRuKz4NQtQ+q4gWQCazH\nT/RiSW0AjoPBoBuAqkgAHDhwcKKrdOUye4/RuOZWWB0HgyNlvPB4BNz1pY0ARtcIKt6n5429GdoE\niYiIiIjKj6GUaB7LHYuSG/yGLnUNB1WsXVMLQQDeOdqDZNKAAweBCj9qgn60n+qDaVrQ0xb6+nXU\nL67C2vNq8fbhbhw/GUdffxrhoAqPmAm5juMgUKmgq3sAR4/F4PVm5ou+8U4XNN2AX5Hh80kwTQv/\n8/5X8KmPrh0WCrP3qKfMvI7AWZpu4rk/HsWH3n8uLr6gfhRjb5KZkLy4Cv6cDsDZJkhbPt7EYEpE\nREQ0CRhKieahkcaiFFrqGg6q2HB5BIZp40RnAqm0hUClD5Ztozrgw4KQH5IkIlDhw7qmJQAAn0/C\n6dgANM3MNCOSAJ9XQk1IhWnaGEgagABUVnoBAeiOJiEKApJJA7WLKiHLHugpq2AozN5jbkfgoQQB\nePK5t93RMyXH3hzpwblnLyj4OmyCRERERDR5OBKGaJ5pbo3iRw/9BUdaeyAAeWNRfvTQX7DzpaMQ\nB7vkZmm6gWg8iVivDgFA7cJK+GQPBAHukt7evlTm3JTpjnVRfB7ULQpgxdIg6hYFsGxJEA11VVAV\nGb19qUw4dBxUVfoQjWkQB68pCAJ6YhqqKn1QFckNhbnCQRWplIXE4HULEQQBsUQKscHqaNGxN4qE\n5UuCCJdY7pxtgkRERERE5cVKKdE80twaxb9++zl0nh6A4zju3stFNX509SSR6E9j34FOvOuiJWg+\nFkWl34uungEk+tNwHAeplIlTPQOwLRsVfi9C1ZkQl+hLuQ2QasIqdN2EqshQFRkCgPrFATRGwjjc\nEkOiPwXDsKDpBiorvUinTCi+TCjOLu8FAD1lYll9tfvvoZ1xQ0EVwWoFpVoXVVX64JM9mYZPOTNO\nC429+fHDu0t+dtkmSGx8RERERFReDKVE80RzaxT3/XI3Orv6Ictn9l5G4xoOvNOF2sUB+BUZetoE\nAFT6Zex8qQX1tQFIkgeWZUOSPOjvT8N2HCyoqXBfoyakorOrH4IgIN6rQ1Eyv1pURUaF34vGSM1g\nZ10/NN1AZ1c/AMCvyrDtTKy0LAeADUCAz+tBOKhA8Z35FVUoFP7dB87Ff/+xueD7tW0bjZEQBFFw\nw3OuoY2gRmqCVOx1iIiIiGhiGEqJ5olndx5C2rCGVRZ7YhpEUcTpngEsDGeCpq4b6OpJIhxS0dIW\nhyAIEATAcYBUykSg0gtNM4DBkKYqMmoXVaInpiFtWEhqJhwAS+uq8JVbrsKOl46618ucG8CR1ihs\n28baNbVQFRl6yhzs4itCljwwTAtKTsOhQqHw4gvq8d53n4U9+08Odt7NVn99aIyEEA76URMauQvx\nSE2QMPheWCUlIiIiKj+GUqJ5INupNrucNsswLQwk00ilLRimhXTKAkRg31udaDuZQE80Cct2oCoS\nUikr0/xIAOJ9KegpE4FKH6oqfQAyYbN+sQcrGoL46DWrcdF5tW6IW7akGtt3HcLh1iiSyTT8fi/q\nFlWivjaAcNAPIFNt1XTTvbfsftKsYqHwMzc0wbEz70XXTSiKBHUwzJqmjWs3rCr5uWS7D5dqgjTS\n6xARERHR+DGUEs0D2U61flVGVcDnhr/+gTQGkgZEUYAwGFclj4jDLT3oPD0Aj0eA4wCGYcHr9UAQ\nBdiGA1kS3S68Tm1mhEq2Oqkocl4gzXIcAIMzROEA561ahO5Y0j3eGAkPjnYRYdsOGiM17rFSoTDb\nvGj7rkNoO5mAbTtI6gaW1lXhinctAwB3L2o2hPb26Xhx9/Fh3Yc3XXEW3jx0Cm0nM48LooCldVW4\ndgPnlBIRERFNFoZSolms2JzRobKdalNpDcvqq/H2kW6IooC+/hRyS6eW7cAxbAxoBgScCZIAkE5n\ngqkoCrBtB5IkoqG2CpUVXqy/uMGtTg5dLtvcGnUrkJnutjn7OCHAdmzoKQs+r4Rzzl6AWFxDuNoP\nn9fjhsuRQuHQ5kXxRCZ0/vrpN+HYDnr7UuiJJVET9MN2bLx1qBvVVT53ryuQCe47XjqKLR9vQjio\nuk2QuGSXiIiIaHIxlBLNQtk5o0dao9B0A6oi4+wVYWzeuArhoJoXVLPnHjjcBU03IQgCJI+AtGEi\nlbYgCgJsx4HX64HP54FtZyqTDgAMNiESBBGO4yCdtiB5MlVSryxAkkSYVuYcTTfRP5DG9e87J+9e\nn915KG9JrKYb7j0vrMns+fzkhy/KC4G5nXHHEgpDg+/9id+9DUkS4VdkRONJHDxyenDfbKYy6/V6\noOkm9h3oxNo1tW4wzZ1HyjBKRERENDUYSolmmebWKO752Ss4ejw62NwnsyR2/1ud2PbMAaxuXIBg\nQIEoivB6RURjOiorZCwMV+DQ0R6oqgxBEGAYNhaE/FgYFtAdS2LRggqc6EwgmTRgGBYcxy2SIm1Y\nwOC/s4HVtGzoKRO2buKPrx6H3y8h0hDEb54+gBd3H3cDcvvJBPxqJhwebom642Wy42ga6qsAAGct\nP1MJHdoZdyyGhuDDLVGIg6NmLNtGx6k+LF8SBACIooDDLT1Y19Tgnj909AwRERERTS6GUqJZ5heP\n78MbB09BFAXIUma0i6YbaO/IjFkRRWDj5WcBAHa+dBStbb1wHAeWbcO0bFiWA78iIRT0I57QsWbV\nIkSWhfDaGx2IxjUIEGBZjtttFzjz3wAAAfAIGOyQawOOg8vW1ud1xo3GNTywbS82XhGB4ziIxpPu\nflE5r2pq4o23u7D/rU5cfXnEfXy0y5KHyjZ08quy+7kk+lLuCBzDsJFKmTBM272PRH8Kmm66TZU4\nj5SIiIhoag1vMzmJ9uzZg4997GO45JJL8N73vhe//vWvAQC9vb245ZZbcMkll+Dqq6/GY4895j4n\nnU7j9ttvx2WXXYbLL78c9913n3vMcRx873vfw/r163HppZfim9/8JizLco8/+OCDuOqqq3DxxRfj\ntttuQzJ5pqnKM888g02bNqGpqQk33XQTuru7p+ATIJqYWFzDS7uPQxTzZ2r2xLTBsS0C2jv6oOkG\nOk4lcOhoD3r7dAxoBryyBL/iRaDCBweZxFm7oBLL6qvQE0vC4xGh+CTIsghBLD6zUxQFiKIIVZFw\nunsADgrP75QkEbv3n4AgCHnVyqE8HgGv7j0BIFMF3vrzP+M7P/oTfvzwbvz7j17E1p//Gc2t0VF9\nPtmGTlmabrjVXsO0YVk2bAcwzTO/JxwnMwIni/NIiYiIiKbWlIXS3t5e3Hzzzbjxxhuxe/du/OAH\nP8A999yDl19+GXfccQf8fj9efvllbN26FXfffTcOHjwIAPj+97+Pjo4O7NixA48++igee+wx7Ny5\nEwDwyCOP4A9/+AOeeuopbN++Ha+99hoeffRRAMCuXbtw//3346GHHsILL7yA3t5ebN26FQBw8OBB\n3HnnnbjnnnvwyiuvYMGCBbjrrrum6qMgGreW47FMc6IchmkjlT4zSkVPmYj36njjYBcGkpmwZVo2\nLPtMWJM8IrSUCY8HeOvwaVi2DcO0IIoCTNPOq5IOZZoORFFAfzI9uHzYhpYT6nL1xDWoioREX6rg\ncSAz+qWnN4nX3ujAA9v2IhrX4FdkVPi9UBXJrbqOJpiGg2qmu++g7PzT9pMJHD8RR1fPADTNwOme\npHvPgoC8eaicR0pEREQ0taYslHZ0dOA973kPrr/+eoiiiDVr1mDdunV47bXX8Pzzz+Of//mf4fP5\ncOGFF+K6665zq6VPPfUUbrrpJgQCAaxYsQKf/OQnsW3bNgDAk08+iU9/+tNYtGgRFi5ciJtuuinv\n2A033IBIJIJAIIBbb70Vjz/+OCzLwtNPP41NmzbhoosugqIouO2227Bjxw709PRM1cdBND4C3G65\nhmlB0428Kh8AmJaFPftP4FBLD/SUCT1tIp22kE7nn5dKmUjqJhaE/TAMK1NJtB04jgPHAYoVSyVJ\nGDwuwOMREOvV0dnVV/Bcx3ZwTmRBXiDOlR394tgOfvv7twvOCM1cM9OAaCShoIqGuir335puojeh\nwzRteEQRHlFEdZUC07LR2dUPTTfy5qFyHikRERHR1JuyULp69Wp897vfdf/d29uLPXv2AAAkScLS\npUvdY5FIBIcPH0Zvby+6u7tx9tlnDzsGAEePHh127MiRI3Acp+Cxvr4+nDp1atixUCiEQCCAo0eP\njuq9xGIxtLS05P2nra1tjJ8IUWmxuIbm1ihicc19LLI0BI9HHKz8JdBxqg+dp/vRN5AeDKlp9PWn\n0d7ZB103YTsOLNOBbdvQdAtGzrLVtGmjJzqA5mNRaGkLAwMppHQDHlEYbESUfz+CAEgewCt7IMsi\nKit88HhESB4RrSfiiMaTwyqmKcNC/eIAzonUQFUkGKaFtJG5D1WR3M63qbSFeKJ4NRU404BoJJs3\nroJp2gCAwy09qAn7YQ+WfR3HQd2iStQuqoTX68GpngE01FUhqRuoCanY8vEmziMlIiIimmLT0uio\nr68Pn/vc59xq6UMPPZR3XFEU6LoOTcv8H1BVVYcdAwBN06AointMVVXYto10Ol3wWPY5Q49lj2ev\nN5KHH34Y99577xjeMdHoZUe4tHcm4NgORFHEktoANm/MVPA8ooC0YcEjZsqmHhEQBRO9fSkYhgXF\nJ2XmieoCPKIIy7Zh2wIcONA0E3LAA8O0MJBMY1l9NQQIiPfqEEURoijCsgGPKGaW8MIe7O6b6fCb\n2U8qDC6RdSBJHqTSFo60xKBpFmTZg6pKL6oDPnSc6oMgCPDJHpw41QdZFnH+qkVQfBIURXarkwAQ\nqlaQ1AovAc4abQOi/5+9ew2O67zvPP99nnNOn3MajUZ3gyQA3kSKomTrYpOOHMlxHEfyVrZs2U5l\nxvFmPRnvKPFuMpl1TWarZrKZqam8cFU8U7uZSSUVZ1K7We9mUrUzXlc2mbGcihPLs/FVtiXRlihL\npCiQIkiAJNDdaDT63J9nX5zuJgCCulDUhdL/88IKgUajGxZc+fF/O3SgxYMfP8oXH3qKlU6E62p2\ntKpEcU41cHAcjVKKWw5Ms3NHyK/9/R/nwL6GtOwKIYQQQrxOXvNQevbsWX71V3+Vffv28bu/+7uc\nOnVqHDJH4jimWq2Og2Mcx9RqtU2fgzKgJsnl6koURbiui+/7234OYGJiYlOw3fj50fO+mF/8xV/k\nwx/+8KaPLS0t8Q/+wT94SV8vxNWcOt3m8194fHxjc2Q0V6k0vOuu3cPKYkyS5FgLvu8wiMpTK7WJ\nCo7WeK4zrBhqlC43z0I5gxonOZM1n7mZSc4t9dBKUQ09oiRHKSgKi+tqHEdT5GUwdYZ9FaOZTWMs\nrqPwXI1WCqXKBULHT1yisxoxEVbYMzvJk89cZM/sJPNnO/zo2WWO3DG7KZDmueHvfPDt/If/9OQL\n/mxezgKiQwdafOyB21lYWkUrRRC4hIE3bHfOx39eH6RM1X0JpEIIIYQQr6PXNJQeP36cT33qU3z0\nox/lN37jN9Bac9NNN5HnOefPn2f37t0AzM/Pc8stt9BoNJienmZ+fp4dO3aMP3fo0CEADh06xPz8\nPO985zvHn7v55pvHn9vYjjs/P8/k5CS7du0af91Iu91mdXV1/Lwvptls0mw2N33M87yrPFqIl27r\njc2Nstzw7UfPcuSOWd515xzPn1+l3Y3I8gKFwhpLf1htLIzBrzgUhSHXZnwX1PccrDWkaVlpXTi/\nSq+fAAq/4hL6LmlWkBcGYwwoRRA4GGuxRg3nTS2e51ANPaZbVRRwaWWdKMlpdyL6gxTPdfB9h8ma\nTxTnzJ/tcnBfk4srA546cYmjd86itGLfXJ0P3Xcrhw60+Mb3nqf9Au25L3cBUasREgbe+DwMlIuP\nwg1hXzbtCiGEEEK8/l6zmdLl5WU+9alP8eCDD/Kbv/mb4/MQtVqND3zgA/zO7/wOURTxwx/+kC99\n6Ut85CMfAeCjH/0ov//7v0+32+X06dP86Z/+KT/7sz87/twf//Efs7S0xPLyMn/0R3+06XP/8T/+\nR06ePEm/3+f3fu/3+MhHPoLWmg9/+MN85Stf4fvf/z5JkvBv/s2/4ad+6qeuCJpCvJZGNza30+5G\nfOexs5yYX+E7jy3w5IlLABy5fZb3vfsmfuwdu9m/p0G95jO7c5LdM5NMN0K0VsPKaMEgymmvxgyi\nDMcplxR1ezFr6yn9QTmTWg09PFfjV1wqnsN0I6Q1VWV6qkqj7nPzvga3HGjx0f/qVmZ31qgOQ95t\nh3ayqzXBzI4JJkKPes3HGMYzrForLq6sc8/RPdx+204+8bPv4Dd/7X18+sF7xzOcG2dBt7qWBURb\nlx5tRzbtCiGEEEK8/l6zSukXv/hF2u02f/iHf7jp1ugnP/lJPvOZz/Bbv/VbvP/976darfJP/+k/\nHVc/f/3Xf53f/u3f5oMf/CBKKT75yU/ywQ9+EIBPfOITLC8v87GPfYwsy/jIRz7Cgw8+CMD999/P\nwsICv/Irv0Kv1+P9738//+yf/TOgXLr0mc98hn/xL/4Fly5d4u677+azn/3sa/WjEGJbW29sbvz4\nseNLFMagVTnf6bmaKM7H7bDNqWAYJh2CwKW/nnB2sYextpw9dTV5btBaMYgzaqGHX/EojEVFOVop\nBlHGzfub5IWh24tp1ANWOgO6azEToYe15eKinTtqWMv4/qcxhn276xx/5iKOU7bylix5bvBcB4Be\nPyGKc3zP2bZldjQL+uWvneDsYjlPu7Wa+nI9cP+t43borWTTrhBCCCHEG4Oy2/1/weJlW1hY4AMf\n+ABf/epX2bt37+v9csQNqNON+OwffH1TuynAI48vEMXlHdLnz3WZm5kcBz2AMHC55+heHnl8gfVB\nCsBTJ5dJhqdglCrbbvOinAFVSpEXhpkdE3iuQ28txlgw1rKjWWX/ninCwGX/7ikef3KJMPQ4cvsM\ncZLz/LDdN00LkrRgz2ydwwebgOL7PzyPUvD8udXhgiXL/j318WtNs4J3v2M3fuDym7/2vhesUHa6\nUbnUaCp8xZXMU6fb1zXoCiGEEEKI7V1rJnpdtu8KIa40ajfdOFcZxTm9tQTPK4Pd3t31yyXKoVEF\n8uD+JgpFf5Dwg6cujBcbjf7eyXMVxgLDP6/1E6bqIZ5XVl1RiijOSNKcd7x9hpPzKwSBwztv30UQ\nuFgs73j7DKCI44z5hS7vunMOgE43Jk5ywsDFr7jkeTnTujE8KwVB4LF3bvJFg2az8crD6MihAy0+\n/eC91zXoCiGEEEKI60dCqRCvUKcb0e5GtK5DkNrabhrF2aY22SN3zAJwcr5Dr5+UFdC8IPQd/odP\n/AQAv/vH3yLwXYrCYGyZQbXWJGlOUVjKYy7QH2SkmcFxFArIC8NE4LJ/dwMoK5sH9zU4Od+m10/H\ny5LqtQpgLD1HAAAgAElEQVSHD7aY3VHj7Pke5y/06PUTOt2IS0V53zTLCmZ2Tm16b/Waj+fq161l\n9noGXSGEEEIIcf1IKBXiGr3QPdFrbQu9Yq7SlrOPU5MBhw82aTXKs0X3HK2Oz5tYBf/4l98zDlz/\n06fey8PfPI2xliQp0FpRGDOukI5CrlLguhprLWHoMTnhM90K+Xs/dxd75+p87k++y4nnVtBa422Y\nyYzinGPHl5jbNYnrOUB5L7XZCLi0PMD3XaYmK/i+Q5ab4cZeeP+9N/HJv3tEWmaFEEIIIcQmEkqF\nuAYvdk/0wY8ffUXBdGO76Re+dJw4ya943Oi8yXRzcwXw4E1Nbr6pybkLa2hdLh3Kc4N2NMaYcSi1\npqyGKspW3mpQbt599Inz3HXbDPPPd8dbsrfSWvP4k0v83AffxuEDrXFAjpKcs8O5U89xOHLHLpp1\nn5/9mbfzrrt2X9PPQwghhBBCvLlJKBXiGrzQPVHX1Xz5ayf49IP3vqLvMWo3/fkH7njZG2Q/9d/+\nGH/7yBkUCmMtxli0Ks/A5MXliqkxBtdx0BpW1xKStODPB0+zdGmds4urNKfCTXc9R7LcEKeXg/Io\nIDeB3TOTRHFOZzXin/zSezh4k5xaEkIIIYQQV/ea3SkV4s3ihe6Jjpxd7NHZsLDolRi19E43QwZx\nxvogZRBnTDfDq1Zk77x1F++9ez+OqzCFwRhbnodx9fisDAC2DKZpZnBdTeC79NYS4jilNlFh8UKf\nKM6ueP40LZia9MvqaJzT7kbjDcFQbgSemvSxW7cyCSGEEEIIsYVUSoV4ma52T3Qja2y56fVV2iA7\nqoC2rvL8rUbInbft4pYDLb7/w/M88+wlKhWXvLC0swGuU/59VMXT5LlFWUuaFmSVAoAgqBD6HnMz\nDmv9ZDh7WobZes3ntpun+c5j5zh+4iJRkg8/N1qCNE2rEaK0ojkli4WEEEIIIcQLk1AqxMvUaoSo\ncalxe69WIGt3o5e0XGnjeZmf+alDVAOPtfWE80trm860gMJYg6MVWiuiOGeq7hNUHOqTPlGc47qa\nu+/aA1iCwCMMXNrdiCTNyItg0/ONliAduWOWwwdbsu1WCCGEEEK8KGnfFeJlGgW+F7Jvrn5dA1mn\nG/HwN5/jc3/yXdrdiGrgMVGtjAPi57/wOKdOtzd9zU++ez+XVgZEcc7tt+5E67K6qjXkeUFRFCRJ\nTp6X7b2FMeR5Qb3mEwQehw+2ysVIFsDSbISEQfn3WE+fWuaeo3swxlzxWrVWPPPc8lVPv3S6EadO\nt69be7MQQgghhLixSaVUiGuw9Z7oRldbPvRitt477XQjjh1f4pFjC7RXIx79wXnitBjfCR2dh4HN\ny5U2nqpJ0oxnnlvGGmhNhZw518VacBwoCsbnWvLcUBhLvVahGlaI4pQw8DhyxyzHT17CAuuDFKUV\n01NVbtozxZ7ZOlP1YNPN1LKF12fv7skrWotfjRM6QgghhBDixiehVIhrcMU9UWNRWrFvrs6H7nt5\nIWsU1p4902YwSMlyQ389xXMcTo9CpFb0BymtRripRXZjMD272OOxJ87zZ3/5o/GpGtWsUg09lFL0\n11OmF6sUuaXdjYbnYiwUFu0oFNBbS+iuRnz/h4vjGdH33r2Pf/LLP1HOyE6FtLsR/+5PvwdAq1Hd\ndDM1CFzCwGN9kG6aqX01T+gIIYQQQogbm4RSIa7R1uVDzanwZbfsnjrd5nf+t28x/3yHXj8lilMu\nLa/j+2Wwq1Qc8tyQZgVRlLO6FjOzs0Y18Dg53+Geo5dDqTWWv/jK02S54ez5Vc4trQ2XEJUVzDjK\nOHlqGZRGKSiGrbdKQ56Xbb2uo0nSgornkOWG5XbEuaU+7W60KTRunakdnYQZf37LTO1DD58gyw29\nfjJ87OX/6bleJ3SEEEIIIcSNSUKpEK/Q6J7otfg/v/g4Tz59Aa01nqu5sJbiOA7rg5TlzgBHayoV\nB2staVZwcXmdNCnYt3eKLDdEcTYOg88vrvLcfIckL2h3IpRS+BWH6Wb52p490yY3FmuLy6HSghlu\n1VUoXFezuhYzf9agtaJZD3j62WV+47e/wr/+5z/DoQOtTUuUrmbjTO1jT5znL/7qaeJ0+y29cPmE\njixGEkIIIYR465FFR0K8TubPdPibv32OwpTnZbK8IEnLW5/99QRT2PESotE/07RguTvgxKllojgl\njnPa3QEPf/M5vvnI85y70OP5hVUGgxRjDElacPb8KguLPYy1gGKb3UTjcy9BxcHRmh3NKjftaVCf\nDPBczeKldT73J98dL1N64P5byfNtnojNM7WnTrf53//vx4iSHM91qHgOnqvHLcijYDs6oSOEEEII\nId56pFIqxGtsNEP66A8XmV/oUnE1lYrLRLWseBbGkmYGFJjh/VAUWCzGgs0tq2spz86vMLerxlo/\n5dzSGjM7ayxd6mOspTCWQSfC0QrlKLKkQGmFKSxKlTOqoIaLjiyFKVtuo6Rgulkh2NCKC+VCpDw3\n4zbblzpT+9DDJ5isVdjugI7WipPzK9xzdK/cNBVCCCGEeAuTUCoEV26+fbWeb+PCn8B3cJRCa02e\nG1Y6EUVRYIdhUSmFMWVoNMZih4VJO3yuKC74L988zW2HdzA1GTBZ81luDzCFIctN+fXW4qKwlNVI\npcp2XWvL7bdKqeHzG6y1gMVzy1bijZRSBIG7qc32xWZqO92IhcUe1dAb3zzdqtdPiOL8up/QEUII\nIYQQNw4JpeIt7XqfKXmx53vo4RPjMzKNqXA4L1p+reto4iTHc8oqplLlJ6wFay6HUa3Lj01MeOS5\n5eLygF3T5cKjHa0qS8v98cyoMXbjy8NxFHo412nt5c8prfA9B5RisuZf8b7qNX/brbpw9Znadjca\nf4/DB1scO76E1pvDrrWwvp5e0wkdIYQQQgjx5iAzpeIta1S1bHcjqoHHRLVCGLjjMyWj+cnr9XyP\nPXGehcXe+PFh4LJ/z9Q4uBXGMNo/pLUqFw85CotlY7S0ppz/9BwNqpxFXemU85hB4OJ7Dkoxbs01\nxuI4ivqkj1IKR2tajZAw9HDdcpFS4Ls0pgImwgq1icqm92WM4fDBJnDlVt0X0mqE43DcalQ5cscs\nYeCRDbcJZ7khDF1+6RfkHIwQQgghxFuZVErFW9bGquVW13Km5MWe7y++8qNN1ckozjmwt8GllQEX\nV9ZJ0hxrLH7FReuymlkUFlNseTIFjlIkaYE1FtfRGGvJ8oI8N0xUKwyiDEtZKZ2a9MvXZSHLCowt\nY+7kRGUYWDXv/bF97NxR5W+/cwZQpFm5oTcMHPbMNMYbfl9Om+3WLb3b3TTdO1fnXXftfsk/YyGE\nEEII8eYjoVS8JW2cd7yal3Om5KU8X6eXgIEkjXjqxCXaqxFFYcahrRp6YBV75yaZqHqcW1zj7NLq\npucYnwdVakM7r6I1FZKmBZWKg+tqahMVojjH6HI2dUezOvxYxsLSGrM7aihVBsfbD++g1aiS54Z/\n+1sf5EtffYbl9np55zQuOHm6w7NnukyEHr/5j973oj+LjR64/9bxDO3I6Kbpxi29QgghhBDirUtC\nqXhL2jjveDWjMyUvJZS2u1FZAUyycejayvccLi6v87ePnCFOcrRWDKJsWJF00Uqxe6bGT//EzURx\nytnzqygUbGjeVYDjarRS5JmhNlG24AaBy837mlxqD3DdcnFSNfTYv6fO4QM7mN1VIwzKX/dBlBEG\nDp1eUrb6asV0M9y0Nfdf/eHXiZJ8fCpmdFf0q998jv17pl5yu+1L3dIrhBBCCCHeuiSUirekjfOO\nV/NS5ydPnW7z/zx0nEceP4frlhttyxDXotUo21WjOOPs+R6PPbFY3uNUCjtsp/U8TbFuCAMPA3RW\nI558+gJxUlY+J2s+3dVoHE2tLYOd42iU1kxOVOj0Yk493wYUO1tVrIWjd86xe2Zy02vNc8Ov/L27\nOXSgddWtuU88c4Ejt8/SWY1RlAuZRoEWeNltzS+2pVcIIYQQQry1SSgVb0lb5x2381LmJzeeeJlu\nhuOzJ1Gc8+1HzxIGHoWBKE6Zf76LpVxwlGWWNC8wxpAkBj2cCz23tMbSxXWKwlAU5Q0Yv+Iw3QxZ\nXUsxw7MunufAcIlRrVrhp9+zmzDwxrOaa+spO1tVBnG2qTr53rv3A2W7McDWYvFjT5znL/7qaeK0\nGJ+l2Riw4eW1NW/9mUsYFUIIIYQQW0koFW9Z2807jrzUeceNy4327Z7i8ScWqfgOWW64sDyg4jns\nnatzrhPjupokLYiTgtpEBd84tLO8DJqUAbHfTyiMRWtFkhRoXd4qDYMKruuwHmXkWUE19IiijOlG\nyN3v3D0OjKO24TDwmG6G/OLPvZPOakS3F/ON7z3Pf/jPT9LuDDh9dhVr4eabGrQaVfbMTnLXbTN8\n8ctPESU5Fc8Zv8cozjl2fIkjd8zSalRfVluzEEIIIYQQL0ZCqXjLeqXzjqPlRnGScXK+zcXlAb21\nAXFqGEQpnudSFAZrLRPVCllekKTl9ttON8J1HYqiPPeSpgVKFbiuxvUcHK1xtMFSzoBqrdBaMRF6\nFBWHHc2Qnqv5r3/60KYW4Y3zrGeH52eshT/7yx/hupo4zjjx3Mr4Xugzp1Y4codHuxvx2c99nQN7\nG2zX1Ky15uR8h3uOVl/WWRghhBBCCCFejIRS8Zb2SuYd292IdnfAI48vsHhxeNJl2A5bnmZRzMzW\nObC3SRTn9PrljGaSljdeHCxaawpjsLZcZ+Q6Gj2Mhc7wjItfcej2YrRWYMEq8Acpd962C4BHHl+g\n10+vaLf1Ky6d1WhTNffkfHscSKHc3HtyfoV3vH2W9fWUs+d71Cf9cRvyRr1+QhRnL+ssjBBCCCGE\nEC9GQqkQXNu8Y6sR8ugPz3PmXA+tFM4w7I2qo/31lAsrfRr1AM/V+BUXVDL++qIwKAWO1mTGoBXl\n8zgKpRRTVZ80K6urWitq1QoA9ZrPT7/nAN89do5L7QET1QrehhbkKM75/g/Os2/PFL1ewrPzbVDl\n3dPeWlLOo27Q6yfjRUq9fsIdt+7k6WeXyxC8weg9yRkXIYQQQghxPUkoFeIVWFjqo4dbfI21WGux\n2HHVcm0tJQhc6pM+cZJzcblPpeKQZQatFXlellYdrWg2gnJh0lRIe7WsjGa5oT4ZYIxhZ2uCauhx\n9zvnaDWqxGlOtm6YGIZVgCjOWOlEJGnB4sU1fu1fPsTCYo/pVhVF2XK8a8fEppM1o+puOddqCX2X\nI3fMcnJ+hV4/GZ+FqQYev/wL75IzLkIIIYQQ4rqSUCrENfrBU0tgLUVhyAtDnhdlG64FY0Epi7GW\nZ+dXOHxwmuVhVTNJCxxHMxF6rPVTlIJKxcF1NP31FFMYtFL0BjlJWhD4Lo5W7JiucvvhHbQaVTqr\nMVqVd06zvMBzHaI4Y+liH6UURWHIMjve4nvhUp8drSp5YVi62B/eLS2DqVLl2Zf6pE+vX4boMPC4\n5+heojgnjjOCwGPv3CTvumv36/xTF0IIIYQQbzYSSoW4Rqtr5abcMvgVmPKCy7jyaC0UueVvv3uG\nD/zkzbzrzjmW2wMcRzM16YOCvXN1BlHGwmKP3tqwKgm4roNfcah4DqYo2L+7yU++ez/tbsQjjy9w\nYblPZzWiKCwXLvap1wM63QhjLb7nYLEkacHSxT7rUYYxlkGU4Toav+Kw0onYO1eG0nrNJwxcDh+c\n5plnlzdVUcPAJQzcl7yNWAghhBBCiJdLQqkQ1+DU6TaPPXme9UEKwPCkKMaC1qCG1VILRFHGl796\nkjvftot9u+vkuRlXS/PcsLqWYIzFDBceGavL6mtuKYzF0YrFi32+9q15BlHGRLVCGHg4jsZxFDO7\namSpoRpWaEwFADx3ps1gkOLVQyZCb1iBtXj+sBprLFlucLTi8MFpoAyn//M/eh9PnrhwTduIhRBC\nCCGEuBYSSsUNqdONaHcjWtewoOiVOnW6zee/8DhZbggDd7ypdlggxRjGZ1W0VkzVy5nQPDckiUFr\nmKz5ACws9oiilPUoheHsZpyUz+e5Gq31+LFnz/coCsPsLoXrarQqz8R4rkOel2dodk6PzsPkuI6D\noxWOdqhNVFgfZIClGrjl97CW2w7tIAhcppvhOHje996D17SNWAghhBBCiGshoVTcUE6dbvPQwydY\nWCoreVprmlMB9xzZy5E7Zl+TADU6sdLrJ+yZrTN/tkNeXG7bBcptt0DF0xhjcR2H8xd63LS3ieOA\n62qWVwasdNbpDzKw5dbdMPCIkxxjLUUBoa9RSmGMJYrLe6WnznSoBi5JVqBRLCyWZ1zSrCDLC6Cc\nKQ0nKmRZgXY0nutQrynmZiYByDLDH3zmAep1f9vgeS3biIUQQgghhLgWEkrFDWNUoXRdTTXwaHcj\nTs5fpNdP+PO/epp3vH2GO2/bxQP3v3qtpp1uxMJij2roEQYejXrAzfubPHu6Q5qV90dHy45cR1EU\nll4/wdEKv+Kyvp7S68fsnJ7AWMMgytBKYYcXXcqbpbbc6KsgSgr27QlZ6UZkaU6aFligXquMA+ZK\nJ+Li8jrGGOI4x3EUjmZYGQUUuFoxWfPH86K1qqJe97n5JmnJFUIIIYQQry/94g8R4o1hVKEEaHcj\njh1fIopzPNch8F3Onu/R7kZ8/guPc+p0+1V5De1uhB2WRMPhqZepesjOVpUwcKl4DgpwnOEWXmPH\nM6SrawnnlnpkWTmA6rnO8NxKWQkdPX6jojCA4qbdU4S+x2TNp17z2TldnnUJA4+9c3Vu2jvF3K46\nbz+8A9dxCMMKSpXtw1opCmPJCsMgLpceHbqpRXNKKqFCCCGEEOL1J6FU3BBGFcqRk/MraK02PabX\nT4jiDNfVfPlrJ16V19FqhKjhXdJ2N2I9yjiz0CUrDEVhyYcbj4wBi0XpMmg6Thk8e/2ENDO4ribL\nzfi5tFYU5vKfobx7WhjLwmKXtfVk+P7ysh3Y3fyr67kOrUbI3K5Jpuo+ge9RDT2UAhSEoYejFWv9\nhCN3zHLX23dJe64QQgghhHhDkPZdcUPYWKGM4pzeWoLnOZseY60ljnPCwOPsYo9ON7ouwWvrUqW9\nc3WePd3m2PEltFbsmZ1kpROR5wWd1bhceDQ87WKHlc8kLU/GFMayuhZz4dI6E1UPpRXZsO0XC1le\nUBiLorxx6mjF3rkp/IpLGHpEUUacQJ4bPPfy+zfGsnt3jdW1hHuO7mX/7ikee3KRSsUBFHle4A4f\n77lazrsIIYQQQog3DAml4oawsUIZxRl2m8copQiC8l9pa2y5PfYVhNLtlirtmZ3krttm+NJXT4wr\ntWULrcfUpI81XfrrMVkBjqth2JKrFVhVtuhaDOuDlEsr6+RpGUjNcPOu1grXcUiSjKDiEgQugV++\nJ8/V6GoFz9VcWhkwu6uGUuUpl8MHp4njHL9SBs+5mUne47ucnF+h1y/vn8ZxeU7mwx+4Tc67CCGE\nEEKINwwJpeKGMKpQtrsRYeChtnlMfcMiH6XVK5qZ3LpUaaTdjfjSV5+hORmysrpOpxvhOBrPc6lN\n+DQbIXlhcPICrTVpmmOxZTvvMEnnWVktLVt8y4+POpHz3BD4Gs9zSLOCiqc5c24VRyvywjC7s0Zz\nKiSKM+66bYbGVEA4DOKh77HSHYxfa6sRcvjgNE+duEQcl68zTnIeObbA/j1TEkyFEEIIIcQbgsyU\nihvGA/ffSj68DVqf9Dd9zhjD4YPN8Z/3zdXHVdJON+LU6TadbvSSv9fGpUpbLbfXefz4IlFc4DgO\ng0HGpZU+a+sJa+sJ/fWUJC2Ik5wsLwMpjCqhDM/FKBS2vDPqlSF6dEYminOSpABlcV0HawwVz2F2\nZ41ONyKKMxxHE/jOOJDmueFjD9zO3rn6+HWOlkHlhWGy5lMNPaabVeIkf1WXQQkhhBBCCPFySKVU\n3DAOHWjx4MeP8uWvnWDf3BQ/fPoCjqOH7atNWo0qUAa0D913K489cZ4//6sf0V1L8D1n3H77Yidj\nNp592ardHXByvs16lLKTCZSCtfUUpRSL62vl7KiF8WDpkB22546qpYUpSDNbVnyVwtEKz3VQGqIo\nR6lyWVKU5GityPKEJC3YOV2l10+phi6d1QgLHD7Y4kP3le/pgftvHVd4ty6D2hjcR8ugPv3gva/s\nvxQhhBBCCCFeIQml4oZy6ECLTz94L51hFfC7P1hgpRthjWUQZ+ybq3PnrTP8X188xl9//TmUKmdN\n67UKhw+2xidjHvz40asG041LlbY6Od9GaYVSijjOWF1LUEqR5QWdbkyWFy/4+h2thlt67Tio6uGs\nbJob8tyMg6ui/DyUC5Dy3KAU+BWH5lSA77sopdj4UkfB/YsPPcVKJ8J19fD9bw7uwHVdBiWEEEII\nIcS1klAqbkjNRsh97z3Ife89SKcblUuNpsJx6Hz0ifPjpT9QtsQeO77E227ZSeC7fPGhp/iNf/ST\n2z73xqVKG52/0OPpU8vkuSXNCs6e7xElGYHvMogysmxYJb0aW1YoLQW+5xKZHO2UodJYi7V2U8Ac\nLSfSWo//PIgzWlMhhw+02DNbtupuDdqHDrT42AO3s7C0ih4ufwqDK6u+12MZlBBCCCGEEK+UhFJx\nw2sOT7UA/Ps/+wFZbq44GRPFGSudiLOLPeZ2TfKdfAGL5ecfuOOKiunGpUoj7e6Ax59YJE0K8ryg\nKAzacUjTgkGUlVt1XyCQKkBphl8LCTnWQp6Xvb5aw3YrhYtiOIdKucXXYimMwfc3/+pubcdtNULC\nwNu2BXn8ml7hMighhBBCCCGuB1l0JN40RrOgW0/GRHHG0sU+eW4ohi2wrqtZvLB21YU/o6VKI8eO\nL7Hcjej0Ijq9mCQzpFlZ6dRabdoGvE2RFUs5Izrq7i22dPkawxVVVkUZSCsVB89z0I7CcTTV0OPs\n+dUrvseoHRcuB+sXsnEZlBBCCCGEEK8XCaXiTWM0C7r1ZMxKJxq345bVSTO+aTqqMG41ms2cbobM\nn+3w7On2+Os8z6UoDHGcDyundnOgvErF1NFlyFSqvDnqOtsdttmsrMDacjaWcv505/QEvX5CFOeb\nHztsxx3ZGqw3Gi2DEkIIIYQQ4vUmoVS8abQaIUlSEMXZeIYyyw1Jejm8jaqkG2+abqwwbjRaqrRr\nusbsrkl2z9Spht6wBbdMno5z5a/QdpVSKKuh1pSB1PcdKhvai7elynxbDBOvUtBqBNRr/njedNPD\nt7TjbgzWgzhjfZAyiDOmm+ELLnoSQgghhBDitSQzpeJN4dTpNg89fILjJy8O73zmdHsxtYnKpsf5\nvouj1aabpi+08KfTjUiSnNB3WRuk9NfTcjHRcL5TocanXkb/vNpo6ejj1lqStNj0wNHlFqUVrqvI\ns+GsqaOpT1SYqFbIcsPczGT5OAXBluVF27XjHjrQ4hd/7p3Mn+2gUBzY15CWXSGEEEII8YYioVTc\n8E6dbo9vc95x606OHV+iUvNxXc3F5XXiOMdxFVpp9u+Z4sgdM5tOo6z2E554+iIKxcGbmpueu92N\n8H0Hx9WsrAwoTNlKqx0FoybhwoIGR+vhAiQNForCbLuNNy/KwLl1067WCr+iqVUrrK1n5HlBo+5z\nYF+LViNk13TI/NkuwLDSe/nXd7t23FFQX1jqYY19yXdahRBCCCGEeC1JKBU3vIcePoHrlm20rUaV\nI3fMcnK+Q6+fgCpbeH3fY7pZZaJ6ubo4/3yb7/7gPINBxuNPLKK1Yu9cnf/xv7uH+957cPh85XkY\nBeSFKSukG9LkaFbVr7i4jiZOC7CGvNg8Z6rYWCllPCOq9PDPgOMotNa4rsPOaY8oynnf3fu45ebp\ncavxVD3gmWdX2Dc3xfogRWnFvrk6H7pvc9DcGNSrGyqqL+VOqxBCCCGEEK8lCaXihjbauLvx9Emr\nUeXwQVjtReyanqBe86mGHn7FHd8rnW6GPPrDRZRSzO6qjUPfSifiX/6vD/MZ7ue+9x6k2QhpNULW\n+mUAVFgKMwyVw0KpHvbe1ms+cxMeK52IOMmJ4mzTlt1Rey+UIdQM/+m5erhASTNVD9gzM0mrETI1\nGXD0zllWutE4gB4+OM0//qX30GqE49us27XjbgzqW209HyOEEEIIIcTrSUKpuKGNNu5udXK+TaVS\n/uvtOJrDB6a5uNKn10+wFr75vQXqkz7TzRDXdYjiDNd1yq24ruYP/uSRcbX0niN7+f3Pf4c0yXEc\nB0txufXWWiq+hy0M080Q33c5t7gGlJtyrbZYM6ySbtPKa2158qVe87HA++6+if17pwgDl+lmyKcf\nvJdON9o2gF5tNnS7oL7VaLmTzJcKIYQQQojXm4RScUMbtdduFMUZvbUEb7jdVimY3VXj4P4GUZxz\n4VKfU6c7TE36rHQikjTHGIsx5TmZndNVnj+XceyJRSYnffrrCVlelkddV2OsLVtvlUI7ismJCsYY\nXNfBWIvjlrOlowyq9XDz7obX6DoKY+14E3A19KjVfHbtqBIG7qYZ0WZj+2ro1VwtqG/0QsudhBBC\nCCGEeC1JKBU3tGYjZO9cnfaGky5RnG0KgBuXAoWBS5zkpFnOuaUejqOJ4py8KO95rkcZK50Bvu/y\nr/7wG+zbXedb3ztLnlvCwGWi6pFmBUlSYIbBL0pydrUm+O8/8S7++D88Rl4Y0qQMutaAVeC4ajiP\nCpWKZrpRpeI57GiFTE2GRElOluVYYLoZXjEj+nJsF9S32no+RgghhBBCiNeLhFJxw3vg/lvHS30A\nwsAb7cXFGMvhg9NAWUE8Ob/CxeUB64OULHPIC4vn6vG90cIYkrQgTnJ830FrRVYY/IpDfz2lNlEh\n9D1C36MwZnwa5t679zC7cxKtNTubVQZRShTlRGmOKcoq7GiZ0Wgp0o5Wlff82D5ajZAozgh8h1//\n5Z94xdXL7YL6VtudjxFCCCGEEOL1sP0mFCFuIIcOtHjw40eZboYM4gxjLEHgEgYuR+6YpdUIaXcj\njpDED+cAACAASURBVB1fIopzgsDFczXFcD1umhmKYaU0z03ZlqsV5xbXhrOmmolqhdpEhSQtKIyl\nMOWaojDwmN1ZY3ZnjW987wzWWHbumMD3PaamQhqTPmHg4DgaSzlDWqtWOLCvMQ6kAJ7r8Mm/e/S6\nBcUH7r+VPDfbfm678zFCCCGEEEK8XqRSKt4UDh1o8ekH72X+TIfTC10+fP9t/Ke/eZosL4jijJPz\nKxTGkqQ5WWaY2VFj4cIaWimUKs+9KF222AJMTlTorEYopclzw0TVI05ywsBj90xt/H3jJOfipXWe\nemaZ80s9umsxeWHJc0MUZ6RZTl5YFBbPVezd3eD2wzuZ3VHDrzgM4mzbky7X4+fx4MeP8uWvneDs\nYnmn9GrnY4QQQgghhHg9SSgVbwqnTrd56OETLCz1aHcGnD67yiDOsNbSW0s5f6FHxdNUqxXqNZ80\nN9SqFeKknCe1xqJUed6lXqtQqbhkueHYk4tcag/I84KiMKRpQbsbkeeGXj9hMCi39h57apEsMziu\notdLcBxNELgUhUVri1IKay133TbDO2+fYa2f8KH7buXIHbOvWhvtKKhfbXuvEEIIIYQQbwQSSsUN\n79Tp9nimNI4zTjy3gtYapWBtPSUvDIHv4jiandMTVAOPtX5Kby2hORUAkGYFM9MTrPYTHF1u2G13\nIyZrPjunqyxd7OMPT8ysr2dM1X06qzGVikNtooJSmihOy/ulw7ulvbUE19VYa3G0Jgw91tYTACZr\nPk+euDA+O/Nqernbe4UQQgghhHgtyUypuOE99PCJ8ZKjk/NtkrRgYbHH8+d6nD3XY+niGlGSY62l\nM1z+s3O6iuMoojjHcx1C32VqKsCvuFhbVk0Dv5w9rQYes7tqeJ4myQocR3HhUh8U1CYqeK6Do8vz\nMElaoJSiMIaiKLfvKsoqaZYbOt0BUZwBl2+FCiGEEEII8VYmoVTc0DrdiIXFHlCegrmw3GfpYn+4\n5MdirMUCCuivp6wNUrK8KGdDZ+tkeUGS5jjDULt7pkarEVIUlulmSJYXDKIMz9XsaE2wd67Orp1V\nHEczEXp4rjN+LY4uK65RnDGIyrbgrCgYrQL2Kw4XLq2zdHENuHwrVAghhBBCiLcyad8VN7R2N8KO\n7oXGGZ1uPL7ROVpaBGUgjOKcOMpZW08ZDDLSNKcaeCilaNZDDu1vMTtTYxClfPN7Z1m81CeOM7RS\nVEOP6UYV11E42sFx9fj5C2NJ04w0LauodrT01pbbdrPMMFX38VwHYwznlvoc3N+SW6FCCCGEEEIg\noVTc4FqNcBxClVKkWTGuXmpdts1aC47jUJvQrK7FLC6uUfEdtNZUqx4/875DNKZC1tdTfuEjd3J2\nsceX/voERW7QWuM4Cq01hbVcXF5nulnFGot2NL21mKwwJEk+XmaklMJ1NFopPM8BLGlaEPoevu/S\n6yecv7DGof0tmfUUQgghhBBveRJKxQ2t2QjZO1cfV0wrFQdrGJ6CKc+/AKyuxRhjKYqCidDDWIvv\nKvbtnmJuZhKAMHD586/8iO88dg7XdQCFN/w+SZqzsBiTJAX9QQbGkOaWNCuAshUXVc6QoiyuAq/i\nkKZ5+fUU5IVhslZhfZCSZuXHf+//+A4P3C8nWoQQQgghxFuXzJSKG1anG3HqdJuffPd+8twQBh7N\nqZA0L+ivpxhj8dyy0pllZSgsT7RofK881zKIMtoblg39l2+dYbUXM90Mh8uJCnr9hM5qTLeXECc5\ncZxhUaRpgTWWorAYC9ZQtu8O23YVCs/TZSXXWvK8oN/PcLTmx4/sYffMJO1uxOe/8DinTreveF+y\nBEkIIYQQQrwVSKVU3HA23iS1pgyZlYom8F2aUwEXlvvlWRYLExMVBoMU66jhwiNNlORMVCtMN0PC\nwOPk/Ar3HN1LFGesdAYANKYCGlMBz59bpTCWPDcoyhZh39f014fnX1BoylDqOArH0eO7pJM1H+zl\nNt9q6OG6mtpEhd0z9fH7cV3Nl792gg/dd+sV72vP7KRUUoUQQgghxJuahFJxQ9l4k7QalM21UZyx\n2ovxXIdf/m+O8r/80bep7vUAS5zkLF3s47kOaV6QpTmVisvMzonx7OloxvPpZ5dZbg/QWtHrp0Rx\nRjX0AMWqidHDoJumOVorHK3xvLLZIEkL/IqLUhAnOdYYClMQVDwKY1Ba4VdcjLWEgUcU54TB5V+/\nJ350kdNnu0zW/PH7AsaV1Ac/flSCqRBCCCGEeFOSUCpuKBtvkra7A07Ot+n10/GCoePPXOTWgy16\n/ZRePykXHQGep9m1c4J2JyJJcvLcjEPpIM459uQiXsWhWvXAQpRkJGlOlhUEgYvWCq0UhTU4WlMU\nZvw9FeBohTUGqxTWWFxX42hFkuWYwuJ7Lp6naTZCXEcTx9mmUHrqTJs7b9u17XseVVI//eC9r/aP\nVwghhBBCiNechFJxwxjdJK2GHu3ugGPHl9Ba47mXR6M7vZiVbsR7fmwfYeDSXU04dnyRMCyrj8Wk\n4dx6uul5u6sRu2fqZQVUKQZJRr+fkuVlhdMMwyeAVopgwiPvJRRFQZKUC4uMMZhhWC0MFKZgrZ8x\nMeHRbAQM4pxmI6QaeGR5QbChGhrF5ZmaILj6r+PZxR6dbiTbeoUQQgghxJuOLDoSN4yNN0lPzrfR\n+sp/fR2nbOs9Ob9CGHjMzdRoNUOiOGNhscfFlQGup7m4MuD5c10utfskSYG15RKk2kQFV2sma355\n4sVYkiTHFAalYN9cnYrn4DqKYniP1ALWKpRSaKd8TX5FjW+T1icDKq7D0sU+gzijXvM3VUmjOGNy\nokK4IahuZY2lsyqLj4QQQgghxJuPhFJxwxjdJI3ijN5asu1jlILbbtlBtxcTxWUVc9d0jcUL/XJZ\nkYLdM5PsnK5igQsX1/FczTOnVnjubIeLy+vkhSHNClxX43kOnufQnAq55UCLHdMT+BUXxykrtFop\nisKgNWWrsLF4rqLieTi6/HynGzHdDFHASjvi8MHpTa+54jncfFPzBd+70ormlFRJhRBCCCHEm4+E\nUvGGtN1ZlNFN0ijOsFf5unrNZ/fMJHfcupPQdxjEGQuLq+ycrhKGLjtaVbLccH5pDUdrZnbV6KxG\nJGmBMZb+esogyshygzUWYwyuq5mq+xy6qYXraKpVj8JY6pM+1aqHX3GpVMrw6jiqbNOteigFhTGs\nRxkAB/Y1mN05AQrWBymDOGO6GfIP//67uettMy/489g3V5fWXSGEEEII8aYkM6XiDWW7cy8bz6I8\ncP+tfO5Pvova5muNseMqZKtZ5R//8nvorsZ8dvXrNKcCwsCjsxrzjUfOMN0sFw6dObdKlGSUx10U\nxlgcVxH6HtpzKIwhzw2nF7qkWUGjHrKzVWW6HuK4mvZqxPkLayjKhUR5bvArLkVh2LVjAtd1iOKU\nI7fPMTdTY32Q8omffQdTdZ/mVDgOmg/cf+t4q/BWeW740H23vno/dCGEEEIIIV5HEkrFG8Z2517g\nyrMov/bJH+fMQpfFS32sLVt26zWfwwenaQ1D3qiy2O5GTE36RHHGtx89y5lzq6y0IyyQ5wWWcg7V\nFBYznFfNc0Pulq2+aVrQaoRorZhuTvBT9+wH4G+/c4Z3374HheLr3zszngd9/lwXKF+T6w6XMAUV\nGlN++XGtOLCvcUXV89CBFg9+/Chf/toJzi6WgVxpxb65Oh+6T+6UCiGEEEKINy8JpeINY+O5l602\nnkU5dKDFv/7nP8Mf/vvvDU+2eJsWB631E37qnpvodCNajZBuL+bY8aVyXjQ3aK3IhoG0yA1WKzxX\nk2QF1oK1kKQ5rta4jkJrRTX0yjulQBh4NKYCTs6vcM/RvczsmBjPr/oVt6yW+u745Ey52KgMrS/U\nhnvoQItPP3gvnW5EZzXaVEkVQgghhBDizUpmSsUbwujcywsZnUWBMsD9w7//bm452MJiWR+knLvQ\n48mnL7DWT/nLr53kX33uG/z7P/sBz59bpbMaj7fjWizGWBTl5tzC2OFMaRlIAfLckmYFSVbQH6RU\nQw9rLfEwfB4+2BovUzp8cBpjyi+cboYYY8dh0hjD4YPN4XO+tDbcZiPk5ptaEkiFEEIIIcRbglRK\nxRvCxnMvVzM6izIKa6PK4vyZDt9+9Czfeuwsd25ZGHRuqcf5i3366ymB7+JohR7eHLXD/7DAdkOq\n1pZ/axNUHLqrMa6jxrdEW43qeJlSFGfM7aqxdKlPrVbhPXv3stKJsMpyy95pgsBjuhlKG64QQggh\nhBDbkFAq3hBG515GojgjijPCwBu3vm49i7JxKdKjPzhPnBbUaxUOH2zRalSJ4ozFi+USIs/VuK4m\nSXOCYZC0FhxXkedlGB5/d3X5H0HgkWWWMFAM4mLTLVGlFUHgQS+mPukzu2uSuV017j26l3fePgsg\nbbhCCCGEEEK8CAml4lXX6Ua0h/OdVwtno3Mvz55e4eR8m14/xVqLUmocNEebdU+dbrO6FvNnf/kj\nXFejUMRJjuc5RHHOtx9dIAxcCgNZVnBpZZ1BlHHLjglct0qc5OX90F6CMWYcRi1lEFUKtFIorbAG\nMgrywhAG7jgot7sDnj+3yp6ZOq2pEIZhOU5y/uYbz7Fv9xSHDkgLrhBCCCGEEC9GQql41bzYeZet\n7rpthi/85+OXt9YORXHOt763QGEtn/3c17HG8viTS3ieHgdVO35sxsXldSoVhz2zdTxXM1GtsD7I\nePZ0m2roladbjC1Dr7YoBY6rwVqspdzCq8p50CgxOI6iMAYLxHFOGHg8/ewKb7tlx7bve+NSJiGE\nEEIIIcQLk0VH4lUxOu/S7kZUA4+JaoUwcMfnXU6dbl/xNU88c4F33TVHGLhkeUGaFeWWXGtRGp47\n3aUalFtwy/benGPHl0iSfFztXOlEKKVIkpwsLwCYqHplqLSQpAWO1gS+i3YUWVpWRj1Hgyo37Sql\nxouL/IqLQjGIMs6d79JdiwkDl5v2NMbnZ0aiOKfdjYjifNNSJiGEEEIIIcTVSaVUvCpe6nmXkdH2\n3VYj5J6je4ninDjOCAKPH/5oCaUUvX4ynjUdVUa1Vjx/fpX6pE+vn5CkOY7WWFtuu/Vch/VBxnSz\nilLQXY2J0xxrLEVmcD19uX3X2GEFtfyzUuA4iqBSYWKiglIwM13jYx+6g3/3p98bv/Z2N+Lk/Aq9\nfjK8m6oIKg7Hji9x33sPvgo/XSGEEEIIId48pFIqrruXe94Frty+GwbueB6zt5YAjE+yhIG3aVlu\nr5+wf/cUaVqMP6ZUGX6z3BAnOXO7ahzc1+S2Qzu4+645du2o0myEBBUXpSDLC5RSbFwAbC3EcYbn\nKbSCuV01Hj1+vpw3HS5lancjjh1fIopzPNeh4jl4riZOC7701We2rQgLIYQQQgghLpNQKq67jQEz\nijPa3QFRnG16zOi8y8jW7bsjG6uiSpUnWcLAoz7pX34uC4Hv8q4753AdTWFMOSMKuI5i13R1vDXX\n8xziJEdpXd4mtYY8L++WFoW54vtbC/1BRmMqIAw81tYSur2IvXN1AE7Or1AYSxRnZPnlr6/XfCZr\nPl/+2olr+AkKIYQQQgjx1iHtu+K6azVCur2YJ5+5uGmLbug77JmdZHbX5BXnXUbbd9tb5jA3VkXr\nNX8cLg8fbHHs+BJaa5QqT7c0GyFvu2UHq72EWw+2mNk1CcA3vntmw/M5tLsx7U5EUZRzpkpD4FfI\nsowktSgYtx77fllJXelE+P7o10XxwP238tt/8HWefnaZfBxmFX7FoTkV8I637wIuV4RlC68QQggh\nhBDbk0qpuO7a3Ygz51aHLa2aojBcuNTnmefafPWbp/nrrz/Hc6fbVwTQB+6/lTzfXK0MA5f6pI8x\nhsMHm+OPtxpVjtwxW94x9T2MMQzijB8/socjd85yYH+TMHDHXw/lNt09M5NcWO4zGKRkWUGaFhhj\nyfICY8qTMBawWLRWOFphrSXLDZdWBtRrAQf2NQCIowzU8KypBUW5ybdU/h9bK8JCCCGEEEKIzaRS\nKq67hx4+wdsO7Sg346Y555bWsLYMea6jWe0l7J6t8/kvPM7f+eDbmZoMaDVCDh1o8eDHj/Llr53g\n7GJ5RkZpxY8f2cOl9oBWo7rp+4RBhZv3N/n5B25n/54pmlPlHdRTp9t88aGnOHWmje+77Jub4sxC\nl9tu2UVvLaHXS3AcjVdxsGmBsqpcclTYcatwUVisNawPhut5SRlEKQf3TdHuRjz08An27q6ze2ES\nVLlUqTxl4wBlW+89R/deUREWQgghhBBCbCahVFxXG7foHtzX5K/+v5OsR1lZflTliZWZHRPESc7J\n+RW+/fhZ3nXH3KYbpp9+8F463YjOarQpaI7Caqcb8dyZLpnJmZ2u8f/+1dMc+v/bu/P4qKt7/+Pv\n2ZKZyUImAWRfDLK4AGGJIKAFerHI4gOLtnVBofXGKurVq/fn8vB2obfa20J9CNW6FS9Sq8C1YsX6\ncwH5uVeUpeCCZMFECJCdJDOZ7fz+GGbIkBAShExCXs/Hg4fke2Ym3+QwgbfnnM9noEezpg2NBcby\n6noZSWXldRqanaUfXX6Bdu4+oOde+qesVovCxshht8ntcuhwrV9+fzCuyJHtyGOMJIuJVPl1JjvU\nv3e6fv/UBzpUXqdB/TOUnpYcK3LUWKRScFD9e6ezdRcAAABoAaEUp1S0yFG0TYrDYVOGw65wOCyr\n1Sqb1aKySq8+/KRYmR63AsGwrFZLXA/ThVflKHtQZlyYyx6UqVsXTtDb7xdq2VMfKBgKKRSSSkoP\n65sDtdpdUKZ/bP1G6enJstuscS1avt5fow8+LdH180erd/c0HTxUJ19DUA5HJEjabVYdqqhT2ETO\nkyYlRa6HwkbGGCU57HLYberhceqb0sOqrD6kQ+V1Kik9LJs1UowpxZ0c930wRqqr8+uyqUPb7XsP\nAAAAdEaEUpxS0Sq6kaq0kfOhNqtFNuvRlcSGhqBq6/3K9Ljj2rxIzfcwlaT8ogpt2Lhbf3npnyre\nXyOLJVJUKMvjksvpkNcX1AefFivT45Ir2SGr1RJbvfT6AvpiT7Vuue9vcjjs8gdCMkay2y1KTrLH\nHmvCIckSqdgri0VOm1VJSZE2L8FgSHW+oIIhI6fToVA4cvLUarVJskSq9x4JsRaLRS6XXYt+GAnX\nAAAAAI6PUIpvrbLKq4oqrzIzIlttMzNcqjncIPsxW1qlyOpj9AxmtDeo0xn/x/DYirX5RRVauWar\nAsGwSstqZbdF6nMFg2GVHqxVr56psttt8gdCKthbqeFDuh8Ji5FAWryvWnX1ARkj+QMhpaYkqbbO\nr2DQyGYLKyPFGfvc1Ycb1Ktnquq9gbgtuTU+v87qnipJctitR4JqWA67TSnuJLmcDo0c0VM+X1BO\np139eqdrzAV9Tu03GgAAADgDEUpx0qKrlyWlkaJE0XOh/XqnKRQ2cibblJxkj6uoGwqH5Uq2yZhI\nqMzMONpDNCpasTYaSjds3C273aqD5XVq8IeU1CgsWiwWlVd6leVxKRw2amgIyucLypEaeUx5pVc+\nX1AWi0XGRI62hsORw6P+QEgN/qAa/CFZJGWkOzXinO7qlpasXV8eks8flM0aWZF1JtmUknJ0i25m\nhktW69G+qjW1DZIirW2CwTDbdgEAAIBWSkhLmB07dmjy5Mmxj6urq3XLLbdo7Nix+s53vqO1a9fG\nxvx+v+677z7l5ubqoosu0mOPPRYbM8Zo6dKlmjBhgsaPH69f/epXCoVCsfFnnnlGU6ZM0ZgxY3TX\nXXepvr4+NvbKK69o+vTpysnJUV5ensrKyk7zV31mia5eVlR55XY6jqwWRs6Fbt1ZqoF9MuRyOtQt\nLVnBYEjhcFgOh1W9ekRWNS2WSPGgxm1eoqIVayurvPp0xz7tKaqIjUXbt0S3BktSgz+k6BZaWRSp\nTCQpEAzL6wsoGIp8bLFEVjlra/2RFdpku5KSbHI7Hcro5pQxRv16pclht0XOlR45YxoKGTnsVjns\nR98uPbunaPyovnI57QoEQ/L7g6qs9inL44qdiQUAAABwYu26UmqM0f/+7//qoYceks12dLXrgQce\nkNvt1vvvv68vv/xSN954oy644AINHz5cv//977Vv3z699dZbKi8v16JFizRs2DBNmzZNf/7zn/X2\n22/r5ZdflsViUV5enp577jldd9112rRpk55++mmtWrVK3bt315133qlHHnlE99xzj7744gv97Gc/\n05/+9CcNGzZMS5Ys0S9+8QstX768Pb8dnVp09bI5aanJCoWrdGFOX3l9AZUerNU3pTXyNkQq3Npt\nNjmddo0f1adJmxdJcibb9OyL21VSGtnGu/2zUjmTHQoGg/J6A7G2LXabVS6n/ciKpZHL6VC9NyBn\ncmTlNXhMeLXZrAqFjVJTHWpoCCsYCkd6khqjrAyXauv82rbrgAb2y1CvHqkqPVgri8WiUCisYMjI\n6wvI5XQoHDY6Z3CWMjNcujCnn7y+oCqrvbr35ikaPLBpyAYAAABwfO26UvrHP/5Rq1at0k033RS7\nVldXpzfffFO33XabkpOTNXLkSM2ePTu2Wvryyy8rLy9PaWlpGjRokK699lqtWbNGkrR+/Xpdf/31\n6tmzp3r06KG8vLy4sfnz52vw4MFKS0vT7bffrnXr1ikUCulvf/ubpk+frlGjRsnpdOquu+6KhV6c\nWLTty7G8vqAqqrzy+oLyeCIhz+V0aPAAjybnDtTk8QM1fmQfTZs8WFMvGtRsID1YXqfyCl9sBTYz\nw61g0GhvSZVKSmuVkuKQRZLVElkZjZwNjfQIzfK4leVxxc6C2u022ayRP+LGGCUfqaqbnORQelqy\nuqU75UqyyuGwyWK1qKbWr/JKrwqLKxUIhJSV6ZbbZVfP7ilyuew6WF4nl9Ou0ef1UmajysAup10X\nDO9JIAUAAABOQruulH7/+9/XTTfdpH/84x+xa3v37pXdblf//v1j1wYPHqzXX39d1dXVKisr05Ah\nQ+LG/vznP0uSCgoKmozt2bNHxhgVFBToX/7lX+LGDh8+rAMHDqigoEA5OTmxMY/Ho7S0NBUUFCgr\nK+uEX0dlZaWqqqrirpWWlrbhO9G5Rdu+NP64cQsWi8UiZ5JNi36Qo4pqr4r3R86cWqwWDRmcGTtv\nGe07Gh3r3ztd4bCJO6vpctrlbQjIYrHIokjQTEmxyOcLKhiKrIL6AyGlpybr7AGZumhsP61at121\n3kAsiPqTIxV0jTFyJR/9Ix8Oh+UPGBkFVXqwVnVe/5HrRsGQ0cQxfdX7rDR5fUEdOHhYu746pJEj\nesl1TGEmzpACAAAAJ69dQ2nPnj2bXKuvr5fT6Yy75nQ65fP55PV6JUkul6vJmCR5vd6457pcrkjQ\n8PubHYs+59ix6Hj0853I6tWrtWLFilY99kwUbfsiRQLptl2lcS1YJMnnD+nTnft184JcZWa4IoWL\nurnieo/eunCCKqu8sTFJevAP78jtOlr4yHukXUx9fSSYKmzUq0eqDtc2qN4bUChklOVx6TsTBmv+\nrHOVPShTE8b017pXdyl/b6WGDPLon18cUs3hBmWkJ6v0UK0CgZCsNqtq6/yR7bxJNslIFlkiZ1Il\nVdf49EV+mXqflSaX065BAzyyO6xyJdtU7wvEBenLpg7lDCkAAABwkhJefdflcsVCZpTP55Pb7Y4F\nR5/Pp9TU1LgxKRJQGxoaYs/zer2y2+1KTk5udkySUlJS4oJt4/Ho657Itddeq9mzZ8ddKy0t1Q03\n3NCq53d2ngyX+vVOj62QNl7ZjEpPTVZaanKs52jjMHrsazVu/dJ4BVaKtHRxJtvVq2eqyiu9qveG\nJBn17J4qV7Jd6anJMgprQk6/WDDMHpSp/3PzlFjgrarx6U9rPtXrmwsiq7lhI6vNKn8gpG6pSXLY\nbUf6jkbOqUa3/BbvqzkSiiNvk0yPW7f/eKIkNRuyAQAAALRdwkPpwIEDFQwGtW/fPvXpE+nrWFhY\nqCFDhigjI0NZWVkqLCxU9+7dY2PZ2dmSpOzsbBUWFmrUqFGxsbPPPjs2VlBQEPs8hYWFSktLU8+e\nPWPPi6qoqFB1dXXsdU/E4/HI44k/P+hwOI7z6DPTrGlD9dizH6vmcIMcjvh+pOFwOFZV99ieoy1p\nvAIb5XJGzpC6nA716x0pZJRzXi9VH27Q9s9L9fmeBoXCRl/ve02D+mVo8fUXauqkwZKOBt78ogpl\npLk073sj9PU3Vfoyv1yhcFgFeyvlawjJbg/JYbfJZrPEbc1t8AdVVe2Tyxn5HyL9e6fHvg7CKAAA\nAHBqJKQlTGOpqamaPn26li5dKq/Xqx07duiVV17RnDlzJElz587V8uXLVVVVpaKiIq1evVqXX355\nbOzpp59WaWmpysrK9Pjjj8eNvfDCC/rqq69UW1urRx55RHPmzJHVatXs2bP1+uuva8uWLWpoaNCy\nZct08cUXNwmaOL7sQZmaNW2onMl2BYJh+QMhBYJhuZyOI4WAIqvO0Z6jrRFdgW3M5bQrPe1of9As\nj1s1hxu0+cMi1db55bDblOpOUlpKssorvXrgdxu16b3CuNeIVgp2Oe0alt1dk3MHKNWdFKm8K6nO\nG5DdbtWAvt1iFXoDgdCRfqaRFVTOjQIAAACnR8JXSiVpyZIl+tnPfqZLLrlEbrdbd999d2z189/+\n7d/061//WjNnzpTFYtGCBQs0c+ZMSdLVV1+tsrIyzZ8/X4FAQHPmzNHChQslSdOmTVNJSYny8vJU\nU1OjSy65RP/xH/8hSRoxYoSWLFmi+++/X4cOHdK4ceP04IMPJuaL78RGn9dLY0f2kcUi+XxBOZ12\nuZzxK8bRnqOtNWvaUK1cszWu3cw5g7O0bVepvL6AwuGw/rG1RIFQWMZIVovUv8/RIGu3W/WHVR/F\nVkujlYIbn1PNzHBpcu5AleyviRVLOqtHigLBsGSRamsbItctFn2+p0xn9Uxp0nu0ssqriiqv50xc\nDwAAHylJREFUMjPYwgsAAAB8GxZz7CE+nJSSkhJNnz5db731lvr165fo22k3j/zpQ1VUHX8lNMvj\n0q0LJ7TpNfOLKppU5vV5g9q6c5/qvEF9kV+mcNjIyMhmschmsyozw6VePVNjvUpf+MOVGjzQo/yi\nCv1x9cdKcSc1+Twb3y/U3uIqhcNGmR6XKqt9sh7ZPhwMhdWnV6qmTjxbDrs1Fkrziyq0YeNulZRG\n7s1qtapvrzTNmkaxIwAAAHRtJ5uJEr59F53brGlDFQyGmx072S2v2YMydevCCbr35in66XXjde/N\nUzRkcKamT8lWn7NSFTZHPp+RQmEjfyCkQxX1Kt5XLa8v0gpm7zeRlj3NnVONGn1uL/XsniKHw6aq\nGp+MMQqFjex2q3r3TNWEnP5yOe2y2616ddNu5RdVaOWarbEeqinuJLmcdlVUebVyzVblF1W0+WsF\nAAAAujpCKb6V7EGZWnhVjrI8LtX7Aqqr96veF1CWx9Vky2tbeTJcOntg5Pkl+2skSQfL6hQORRb3\nLRZL7FcoFFZtXUClB2tlsVg0sG9G7DWOPacalZnh0sSx/XX+8B7K8rjVIytFfc5K1TmDMzVxbH9l\nNtqWW7y/Rms37IrbVtxYNLgCAAAAaJsOcaYUnVt0ZbNxz9FTec6yosorY4y8voAOVdbLbrMq3Mym\ncyOj6toGjTqvlwYPPFq0qrlzqlHpqcn6tx9P1F9f+0JWi+R0OuIq8EZ5vUEVVFeqz1lpx73PtlQa\nBgAAABDBSilOmejK5qkOZdEtuJXVXjX4Q0pLTWrSz1SKrJwGAyHNmnpO3PUTrebmnNdbLqddngxX\ns4FUkhoagkpOsjU7FtWWSsMAAAAAIlgpRYcX3YJbVeOVjJHLGSladLjOr2DIyBgjiySH3aruPVOV\nO7rpoeoTreb2653eYsGm7IGZKq+qb/E+21ppGAAAAAArpegkZk0bqrSUZCUnR/4/isuZJE+3yMqm\n3WZVtzSn3K7ICqrR8QtKH28190QFm+bPOve4Z1Oj+vdOZ+suAAAA0EaEUrSryiqv8osqVNnCqmRz\nsgdl6uYFuRp2dpaCobB8/qAO1/pls0bawaS4k2S1SL16punFv3/e5kq4rSnYdDoqDQMAAABdHdt3\n0S5ORX/P7EGZ+sOvZuvXf3hHmz8oks1qkdUaafdit1vl6ebU6HPPilXCbWt/1BNt8Y0G12N7qPbv\nna7LptKnFAAAADgZhFJ8K5VVXlVUeZWZcfyKu9H+nna7VW6nI3Y92t+zLa1jsgdl6pYF41VUXKnD\ntX4FQyHZbTZledw6Z7BHmRluSd+uEq6nha/ldFcaBgAAALoaQilOSltWPjds3H3C/p5tWdXslubU\nmPN7y2q1yOcLyum0y9Uo7EpHK+GersDYUnAFAAAA0HqcKUWbRVc+K6q8cjsdSnEnyeW0x1Y+G5/n\nrKzyqmR/TYuvF13VbK1oixiX03GkjYujyWPaqxLuyZ6RBQAAABDBSinabMPG3QoEQ6qp9cnldMSF\nwmNXPiuqvM32FG2sraua0RYxLbVwOd2VcE/FGVkAAAAArJSijT795z6t/79f6N2Pi7Vlx369+3Gx\nPtpaoopGPTwbr3xGVzVbcjKrmomshNuWlWIAAAAALSOUotXyiyr09PNb5W0IymG3Kslhk8NuldcX\n1LZdpbFgGl35lI6uarbkZFY1W9PC5XRpzRlZAAAAAK3D9l202oaNu5WakqTm1j2tVqu+KqzUhTnu\nJiufs6YNjVXfPda3WdVMRCXc6BlZt6vpOdaob1P5FwAAAOhqWClFq0TDmMtpV3pacrOPqaltkNcX\naLLyebpXNT0ZLp09MLNdQmBbzsgCAAAAODFWStEqjcPYOYOztG1XqazW+DVTY4xq6/zNrnwmqr9n\na/qotsXpOiMLAAAAdFWEUrRK4zCWmeHS6PN66avCctXUNsgYyWKR3E6HfvzDMS2ufLZXf8/TVR23\nI1T+BQAAAM4khFK0yrFhLDPDpQtz+snrC8rnC8jpdKhf7zSNuaBPgu/0aHVcu90qd6N2NdHquN92\ny/DpOiMLAAAAdEWcKUWrNdeGxeW0y5PhksNu7TBh7HRXx01k5V8AAADgTMNKKVotGsZe3bRbxfsj\n22ItVov6907XZVO/3bbYU6W9quMm6owsAAAAcKYhlKJNOmIYa1zMqC3VcU/FfbfXGVkAAADgTEUo\nxUnpCGGsuWJGnm5OVdX4lOJOOu7zqI4LAAAAdByEUnRKxytm5GsIau831XI57crMcDf7XKrjAgAA\nAB0HhY7QKbVUzGh4dnd9uae82TGq4wIAAAAdCyul6HROVMwoM8Olgf0y5HLaVV7l7ZAFmQAAAABE\nEEqREI2LE7V1K21rihl1S0/WlbPOk6ebq8MUZAIAAADQFKEU7aq54kR9e6Vp1rTWr2BmZrhksVha\nfEy0mFFHKMgEAAAA4Pg4U4p2Ey1OVFHlldvpUIo7SS6nXRVVXq1cs1X5RRWteh1Phkv9eqe3+BiK\nGQEAAACdA6EU7aal4kR2u1Wvbtrd6teaNW2ogsFws2MUMwIAAAA6D0JpF1ZZ5VV+UYUqq7zt8rlK\n9te0+Jji/TWtvpfsQZlaeFWOsjwu1fsCqqv3q94XUJbHpYVX5VDMCAAAAOgkOFPaBZ2Kc51t1Zri\nRCZsIkWJWrntNntQpm5dOEGVVV6KGQEAAACdFCulXcypOtfZVm0pTtRWngyXzh6YSSAFAAAAOiFC\naRdzKs91tgXFiQAAAAA0h1DahZzqc51tRXEiAAAAAMcilHYhbTnXeTpQnAgAAADAsSh01IWcznOd\nrUVxIgAAAACNEUq7kOi5zooWtue217lOTwZhFAAAAADbd7scznUCAAAA6EgIpV0M5zoBAAAAdCRs\n3+2CTuW5zsoqryqqvMpkOy4AAACAk0Ao7cK+zbnO/KIKbdi4WyWlNTJhI6vVqr690jRr2tDjrrYS\nYAEAAAAci1CKNssvqtDKNVtlt1vldjpi1yuqvFq5ZmuTbcAnE2ABAAAAdA2cKUWbbdi4W3Z78390\n7HarXt20O/ZxNMBWVHnldjqU4k6Sy2mPBdj8oor2um0AAAAAHRChFG1SWeVVyf6aFh9TvL9GlUfa\nzrQlwAIAAADoegilaJOKKq+MMS0+xoSNKqu9bQ6wAAAAALoeQinaJDPDJYvF0uJjLFaLPN1cbQqw\nAAAAALomCh2hTTwZLvXrna6KY1Y3vb6AvL6AXE6H+vdOj1XXbW2ABQAAANA1sVKKNps1baiCwbAk\nqaKqXh9tLdG7Hxdry479+n8ffa1D5fXKL6qIBdiWNA6wAAAAALoeQinaLHtQphZelaOwCevj7ftU\nfbhBklG3tGSNH9VHVqslVlm3cYA9VjAY1mVTh7bvzQMAAADoUNi+i5OSPShT3T0puvjCQfL5AnI6\nHXI5j/5xilbWvXXhBC28Kkevbtqt4v2RPqUWq0X9e6frsqn0KQUAAAC6OkIpTkq0sq7bFR9GG4tW\n1s0elKlbF05QZZVXldVeebq52LILAAAAQBKhFCepLZV1owHUk0EYBQAAABCPM6U4KW1pDQMAAAAA\nx0MoxUlpbWVdScovqlBlFb1IAQAAADTF9t0zXGWVVxVVXmWehq2zs6YN1co1W2W3N/1/G4fK6xUK\nh/Xgo+/IhI2sVqv69krTrGkUNwIAAABwFKH0DJVfVKENG3erpLTmtIXCaGuYYyvrupLtMjKyWa1y\nO48G1ooqr1au2aqFV+UQTAEAAABIIpSekfKLKmIrmG6nI3b9dITC5irrPvvi9uOeN23cKgYAAAAA\nOFN6BtqwcXezW2qlo6HwVPNkuHT2wEjQLdlf0+Jjo61iAAAAAIBQeoaJ9g9tyekMhW1pFQMAAAAA\nhNIzTKJDIa1iAAAAALQFofQMk+hQ2NpWMae6EjAAAACAzolQeobpCKFw1rShCgbDzY4Fg2FdNnXo\nafvcAAAAADoXQukZKNGhMNoqJsvjUr0voLp6v+p9AWV5XHGVfyurvMovqqDoEQAAANCF0RLmDHS8\n/qH9e6frsqmnrk/pie7h2FYx0dXZ9uihCgAAAKBzIJSeoVoKhe3JkxH/eduzhyoAAACAjo/tu2e4\naP/QjlJYKBE9VAEAAAB0XIRStJtE91AFAAAA0PEQStFuEt1DFQAAAEDHQyhFu0l0D1UAAAAAHQ+h\nFO2mI/RQBQAAANCxEErRrhLdQxUAAABAx0IoRbuK9lDN8rhU7wuort6vel9AWR4X7WAAAACALog+\npWh3HaWHKgAAAIDEI5QiYTwZhFEAAACgq2P7LgAAAAAgYQilAAAAAICEIZQCAAAAABKGUAoAAAAA\nSBhCKQAAAAAgYQilAAAAAICEIZQCAAAAABKGUAoAAAAASBhCKQAAAAAgYQilAAAAAICEIZQCAAAA\nABKGUAoAAAAASBhCKQAAAAAgYQilAAAAAICEIZQCAAAAABKGUAoAAAAASBhCKQAAAAAgYbpsKP3s\ns880f/58jR49Wpdffrm2bduW6FsCAAAAgC6nS4bShoYG3XTTTbriiiv08ccf67rrrtPixYvl9/sT\nfWsAAAAA0KV0yVD64Ycfymq16uqrr5bD4dD8+fPl8Xi0adOmVj2/srJShYWFcb+Ki4tP810DAAAA\nwJnHnugbSITCwkJlZ2fHXRs8eLC++uorXXrppSd8/urVq7VixYrTdXsAAAAA0GV0yVBaX18vl8sV\nd83pdMrn87Xq+ddee61mz54dd620tFQ33HDDqbpFAAAAAOgSumQodblcTQKoz+eT2+1u1fM9Ho88\nHk/cNYfDccruDwAAAAC6ii55pvTss89WYWFh3LXCwkINGTIkQXcEAAAAAF1TlwylEydOlN/v17PP\nPqtAIKB169aprKxMkydPTvStAQAAAECX0iVDaVJSkp588klt2LBBubm5Wr16tR577LFWb98FAAAA\nAJwaXfJMqSQNHz5czz//fKJvAwAAAAC6tC65UgoAAAAA6BgIpQAAAACAhOmy23dPtVAoJCnSrxQA\nAAAAuppoFopmo9YilJ4ihw4dkiRdc801Cb4TAAAAAEicQ4cOaeDAga1+vMUYY07j/XQZPp9PO3fu\nVI8ePWSz2RJ9O51ecXGxbrjhBj3zzDPq379/om8HbcDcdV7MXefF3HVezF3nxLx1Xszd6RUKhXTo\n0CGdf/75cjqdrX4eK6WniNPp1Lhx4xJ9G2eMQCAgSerVq5f69euX4LtBWzB3nRdz13kxd50Xc9c5\nMW+dF3N3+rVlhTSKQkcAAAAAgIQhlAIAAAAAEoZQCgAAAABIGNvPf/7znyf6JoDmOJ1O5ebmyuVy\nJfpW0EbMXefF3HVezF3nxdx1Tsxb58XcdTxU3wUAAAAAJAzbdwEAAAAACUMoBQAAAAAkDKEUAAAA\nAJAwhFIAAAAAQMIQSgEAAAAACUMoBQAAAAAkDKEUAAAAAJAwhFIAAAAAQMIQSnHK7dixQ5MnT459\nXF1drVtuuUVjx47Vd77zHa1duzY25vf7dd999yk3N1cXXXSRHnvssdiYMUZLly7VhAkTNH78eP3q\nV79SKBSKjT/zzDOaMmWKxowZo7vuukv19fWxsVdeeUXTp09XTk6O8vLyVFZWdpq/6s5ty5YtuvLK\nKzV27Fh997vf1fPPPy+JuesMXn31Vc2cOVM5OTmaNWuW3nzzTUnMXWdRVlamiRMnatOmTZKkkpIS\nXX/99crJydGll14auy4lZk7RvKeeekrnn3++cnJyYr+2bNnC+66DKy0tVV5ensaMGaOLL75Yq1at\nksTPy47u5Zdfjnuv5eTkaPjw4XrggQeYuzOJAU6RcDhs1q5da8aOHWtyc3Nj12+99VZz1113GZ/P\nZ7Zv325yc3PN559/bowx5qGHHjLXX3+9qampMYWFhWbq1KnmrbfeMsYY8+yzz5rZs2ebAwcOmIMH\nD5p58+aZVatWGWOM2bhxo5k8ebIpKCgwNTU15ic/+Yl58MEHjTHGfP7552bMmDFm27Ztxuv1mvvu\nu88sXry4nb8bnUdVVZUZP368Wb9+vQmFQmbnzp1m/Pjx5r333mPuOriCggIzatQo88knnxhjjHnv\nvffMeeedZ8rLy5m7TuJf//VfzfDhw83GjRuNMcZcccUV5ne/+53x+/3m7bffNjk5Oaa8vNwY0/4/\nS3F8d955p3nqqaeaXOd913GFw2Ezb94889BDDxm/3292795txo8fbz755BPmrZN5//33zaRJk8z+\n/fuZuzMIoRSnzKOPPmrmzJljnnzyyVgora2tNSNGjDBff/117HG//OUvzS9/+UtjjDEXXXSRef/9\n92NjTz/9tMnLyzPGGDN//nyzdu3a2Nhrr71mZs+ebYwx5vbbbzcPP/xwbOyf//ynGTt2rAkGg+a/\n//u/zd133x0bq6ioMCNGjDBlZWWn4avu/D777DNz1113xV1bvHixWb58OXPXCdTW1sb++9e//tXk\n5uaaw4cPM3edwHPPPWduv/12M3XqVLNx40azZ88ec/755xuv1xt7TF5ennn66acT8rMUxzdz5kzz\n3nvvxV3j77uObevWrWbSpElxf7bz8/NNSUkJ89aJ1NbWmilTppg33niD99wZhu27OGW+//3va/36\n9brgggti1/bu3Su73a7+/fvHrg0ePFhfffWVqqurVVZWpiFDhjQZk6SCgoImY3v27JExptmxw4cP\n68CBA03GPB6P0tLSVFBQcFq+7s5uxIgR+u1vfxv7uLq6Wlu2bJEk5q4TSElJUXFxscaNG6d77rlH\nd9xxh77++mvmroMrKirSypUr9fOf/zx2raCgQH379pXT6Yxdi85NIn6Wonler1dFRUVatWqVJk2a\npJkzZ2rdunX8fdfB7dq1S+ecc45++9vfatKkSbr00ku1fft2VVdXM2+dyFNPPaWhQ4fqu9/9Lu+5\nMwyhFKdMz549ZbFY4q7V19fH/QNLkpxOp3w+n7xeryTJ5XI1GZMif/E3fq7L5VI4HJbf7292LPqc\nY8ei49HPh+M7fPiwbrrpJp133nm68MILmbtOonfv3tqxY4dWrlyp3/zmN9q4cSNz14EFg0Hdfffd\nuv/++5WRkRG7Xl9fHzcv0tG5ScTPUjSvrKxMY8aM0Y9+9CNt2rRJS5Ys0UMPPaRNmzbxvuvAqqur\n9dFHH8nj8WjTpk168MEHtWTJEv6d0onU1dVp9erVWrx4sST+jXmmIZTitHK5XLEfAFE+n09utzv2\npm48Hh2TIj88GhoaYmNer1d2u13JycnNjkmRVaPGP3Qaj0dfF80rLi7WD3/4Q3Xr1k0rVqyQ2+1m\n7joJu90uh8OhiRMnasaMGdq5cydz14E9+uijGjFihC655JK46y39vEzEz1I0r3///lq9erUuueQS\nJSUlady4cbr88su1ZcsW3ncdWFJSkrp166a8vDwlJSVpzJgxuvTSS/XII48wb53Em2++qT59+mj0\n6NGS+DfmmYZQitNq4MCBCgaD2rdvX+xaYWGhhgwZooyMDGVlZamwsDBuLDs7W5KUnZ3dZOzss8+O\njTXeKlFYWKi0tDT17NmzyfMqKipUXV0de100tWvXLl111VWaPHmyHn30UTmdTuauE9i8ebNuuOGG\nuGuBQEADBgxg7jqwV199VRs2bNC4ceM0btw47du3T3feeacKCwv1zTffyO/3xx4bnbdEvB/RvF27\ndumJJ56Iu9bQ0KDevXvzvuvABg8eLK/Xq2AwGLsWCoV07rnnMm+dxKZNmzRz5szYx/w75cxCKMVp\nlZqaqunTp2vp0qXyer3asWOHXnnlFc2ZM0eSNHfuXC1fvlxVVVUqKirS6tWrdfnll8fGnn76aZWW\nlqqsrEyPP/543NgLL7ygr776SrW1tXrkkUc0Z84cWa1WzZ49W6+//rq2bNmihoYGLVu2TBdffLE8\nHk/Cvg8dWVlZmX7yk59o4cKFuvfee2W1Rn4sMHcd37nnnqudO3fqpZdeUjgc1ubNm7V582b94Ac/\nYO46sNdee02ffPKJtmzZoi1btqhPnz5atmyZ8vLyNGTIED388MPy+/3avHmzPvroI33ve99LyPsR\nzXO73VqxYoVee+01hcNhffDBB9qwYYOuueYa3ncd2KRJk5Senq6lS5cqGAzq008/1RtvvKHvfe97\nzFsnsX379tgqqcS/U844CSyyhDPUhx9+GNcSprKy0tx2221m/Pjx5pJLLomrdub1es0DDzxgJkyY\nYCZOnGgee+yx2FgwGDTLli0zkyZNMrm5uWbJkiVxVfP+53/+x0ydOtWMHTvW3Hnnnaa+vj42tmHD\nBjNjxgyTk5NjbrzxRqqiteCxxx4zQ4cONaNHj477tWzZMuauE/j444/NvHnzTE5Ojpk3b5754IMP\njDG87zqTaPVdY4wpKSkxixYtMmPGjDEzZsyIXTcmMXOK5r311ltm9uzZZtSoUWbGjBnm73//uzGG\n911HV1RUZBYtWmTGjx9vpk6datatW2eMYd46g2AwaIYPH2727NkTd525O3NYjDEm0cEYAAAAANA1\nsT8HAAAAAJAwhFIAAAAAQMIQSgEAAAAACUMoBQAAAAAkDKEUAAAAAJAwhFIAAAAAQMIQSgEAOAl+\nv19PPPGE5syZo9GjR2vs2LFasGCBNm/enOhba+Kee+7RbbfddtzxN954Q/v37z/p16+rq9PatWtj\nH1933XX6zW9+c9KvBwDoWgilAAC0UUNDgxYsWKDnnntOP/3pT/XKK6/o+eefV25urm655Rb9/e9/\nT/Qttto333yjxYsX6/Dhwyf9GitXrtRf/vKXU3hXAICuxJ7oGwAAoLN5/PHHtXfvXm3YsEGZmZmx\n6+ecc47C4bAefvhhzZw5M4F32HrGmA7xGgCArouVUgAA2iAcDmvNmjVauHBhXCCNWrRokZ577jlJ\nUklJiYYNG6ZHH300tooqSdu3b9d1112nnJwcTZkyRcuXL1c4HJYkvfjii7rwwgvjXrPx9tuPPvpI\nkyZN0osvvqhp06Zp3Lhx+ulPf6qKiorY4zdv3qxZs2Zp5MiRuuOOO+Tz+Y779UyfPl2SNGfOHL34\n4otavny58vLytHDhQo0bN04bNmxodjvusGHDtGnTJr344otasWKFdu3apWHDhsXGKyoqdPPNN2vk\nyJGaOnWq1q9f3+rvMQCgayGUAgDQBsXFxTp06JByc3ObHU9NTVVWVlbctXfffVdr1qzRnXfeqYKC\nAi1YsEDDhw/XunXr9J//+Z9avXq1nnjiiVbfQ1VVldauXavly5drxYoV2rZtm/74xz9KkvLz83Xz\nzTdr1qxZeumllzRo0KAWtxNHz4I+++yzuuyyyyRJb7/9tiZNmqTnn39ekyZNavFeLrvsMi1atEjD\nhw/Xu+++G7v+t7/9TZMmTdKGDRs0Y8YM3X///SovL2/11wgA6DrYvgsAQBtEVyQzMjJi14qLizV3\n7ty4x61bt07JycmSpAULFmjQoEGSpIceekgDBgzQ/fffL0nKzs7WoUOHtHTpUuXl5bXqHoLBoO65\n5x6dd955kqS5c+dq+/btsc87YsQI3XzzzZKk22+/Xe+8885xXyu62puRkSGn0ylJcrvd+vGPfyyL\nxXLCe3E6nXK73bLZbOrRo0fs+sUXX6xrrrlGknTLLbfomWee0e7duzVx4sRWfY0AgK6DUAoAQBt0\n69ZNklRTUxO71qtXL7300kuSpAMHDui6665TKBSKjfft2zf2+z179mjUqFFxrzlmzBjV1tbqwIED\nrb6PgQMHxn6fmpqqQCAgKbJSeu6558Y99oILLmjTKmXfvn1bFUhbMmDAgNjv09PTJUUKRAEAcCxC\nKQAAbTBgwAB5PB59+umnGjlypCTJ4XDEQqLNZmvynOiKafT3xwa+6HnScDjcbBgMBoNNrjkcjriP\nGxcbOrbwkN3etr/uk5KSWhxv7n6O1dz3gYJIAIDmcKYUAIA2sNvtuuqqq7Ry5UpVVVU1GS8tLW3x\n+dnZ2dq2bVvcta1btyo1NVU9evSQw+GQ1+uNC3AlJSWtvr9hw4Zpx44dcdd27dp13Me3ZkU0KSlJ\ndXV1sY+Li4vb/BoAABwPoRQAgDZavHixBgwYoCuvvFLr16/X3r17tXv3bq1YsUI33nijBg8erO7d\nuzf73GuuuUYlJSX6r//6LxUUFOiNN97QI488oquvvloOh0MXXHCBGhoa9OSTT6q4uFhPPPGEPvvs\ns1bf21VXXaWioiL97ne/U2FhoR5//HFt3br1uI93u92SpC+//DIueDZ2/vnn680339Qnn3yiL774\nQr/4xS/iVlPdbrfKy8ubhFUAAFqDUAoAQBslJSVp5cqVWrBggVatWqV58+bpBz/4gd555x39+7//\nu9avX99suxhJOuuss/TEE09o27Ztmjt3rn79619r4cKFuuOOOyRFzoree++9WrVqlebOnav8/Hxd\nf/31rb63/v3768knn9S7776ruXPn6uOPP9a8efOO+3iPx6MrrrhC9957r1544YVmH7No0SKNHTtW\nixYt0k033aS5c+eqV69esfEZM2YoOTlZs2fP1sGDB1t9rwAASJLFcMADAAAAAJAgrJQCAAAAABKG\nUAoAAAAASBhCKQAAAAAgYQilAAAAAICEIZQCAAAAABKGUAoAAAAASBhCKQAAAAAgYQilAAAAAICE\n+f9lSXK48cIP4QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f25f45ff080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "sns.set_context('talk')\n",
    "sns.set_style('ticks')\n",
    "sns.set_palette('dark')\n",
    "plt.rcParams['figure.figsize']=(15,15)\n",
    "\n",
    "ax = plt.axes()# Mute the setting wtih a copy warnings\n",
    "pd.options.mode.chained_assignment = None\n",
    "ax.scatter(y_test,y_test_predict,alpha=.5)\n",
    "\n",
    "ax.set(xlabel='Ground truth',ylabel='Predicted value',title='train - Predict block')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}